Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters 978-9811393709, 9811393702

Table of contents :
Preface......Page 5
Contents......Page 7
Experimental Methods......Page 9
1 Experimental Methods: Generation of Cold Gas-Phase Molecules, Molecular Ions, Their Clusters, Metal Clusters, and Laser Spectroscopy......Page 10
1.1 Generation of Cold Gas-Phase Neutral Molecules and Clusters......Page 12
1.2.1 Laser-Induced Fluorescence (LIF) and Resonant Two-Photon Ionization (R2PI) Spectroscopy......Page 14
1.2.2 Ultraviolet–Ultraviolet Hole-Burning (UV–UV HB) Spectroscopy......Page 15
1.3 Laser Spectroscopy to Measure the Vibrational Levels......Page 16
1.3.1 Dispersed Fluorescence Spectroscopy and Stimulated Emission Ion-Dip Spectroscopy......Page 17
1.3.2 Infrared-Ultraviolet (IR-UV) and Stimulated Raman-Ultraviolet (Raman-UV) Double-Resonance Spectroscopies......Page 18
1.3.3 Infrared-Vacuum Ultraviolet (IR-VUV) Spectroscopy......Page 19
1.3.4 Laser Spectroscopy to Measure the Vibrational Levels of the Electronically Excited State......Page 20
1.4.1 Electrospray Ionization with Ion Trap......Page 22
1.5.1 Laser Ablation Method......Page 24
1.5.2 Magnetron Sputtering Method......Page 26
1.6.1 UV- and IR-Photodissociation Spectroscopy......Page 28
1.6.2 Double-Resonance Spectroscopy for Molecular Cation......Page 29
1.6.3 Photoelectron and Photodetachment Spectroscopy......Page 30
1.7.1 IR-UV Pump–Probe Spectroscopy......Page 31
1.7.2 UV–UV and UV–VUV Pump–Probe Spectroscopy......Page 32
1.8 Conclusions......Page 33
References......Page 34
Spectroscopy and Characterization of Gas-Phase Functional Molecules and Clusters......Page 40
2 Microscopic Study on Molecular Recognition of Host–Guest Complexes Between Crown Ethers and Aromatic Molecules......Page 41
2.1 Introduction......Page 42
2.2 Experimental and Computational......Page 43
2.3 Phenol·3nCn (n = 4–8) Complexes......Page 44
2.4.1 Complex Structures and Induced Fitting......Page 48
2.4.2 Effect of Host–Guest Complex Formation on the Photophysics of 3nCn·Benzenediols Complexes: Cage Effect......Page 55
2.5 18C6·Tyramine and 18C6·Tyrosol Complexes......Page 58
References......Page 65
3 Chirality Effects in Jet-Cooled Cyclic Dipeptides......Page 69
3.1 Introduction......Page 70
3.2.1 Experimental Methods......Page 72
3.2.2 Theoretical Methods......Page 73
3.3.1 Nomenclature of the Studied Systems......Page 74
3.3.2 Symmetry Properties......Page 75
3.3.3 Theoretical Results......Page 76
3.3.4 Experimental Results......Page 79
3.3.5 Localization of the Electronic Transition......Page 83
3.3.6 Localization of the Charge in the Cation......Page 85
3.4 Conclusion and Perspectives......Page 87
References......Page 88
4 Hydrogen Bond Networks Formed by Several Dozens to Hundreds of Molecules in the Gas Phase......Page 94
4.1 Introduction......Page 95
4.2.1 Neutral Clusters......Page 97
4.2.2 Protonated Clusters......Page 98
4.3 IR Spectroscopy of Large-Sized Water and Protonated Water Clusters [31, 32]......Page 99
4.4 IR Spectroscopy of Large-Sized Methanol and Protonated Methanol Clusters [35]......Page 103
4.5 IR Spectroscopy of Large-Sized Ammonia and Protonated Ammonia Clusters [40]......Page 106
4.6 Conclusions......Page 110
References......Page 112
5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes......Page 117
5.1.1 Introduction......Page 118
5.1.2 DB18C6......Page 120
5.1.3 B15C5 and B18C6......Page 123
5.1.4 B12C4......Page 124
5.1.5 DB15C5......Page 126
5.1.6 DB21C7 and DB24C8......Page 133
5.2.1 Introduction......Page 139
5.2.2 K+·DB18C6·(H2O)n......Page 140
5.2.3 Rb+·DB18C6·(H2O)n and Rb+·DB18C6·(H2O)n......Page 145
5.2.4 M2+·B15C5·L and M2+·B18C6·L (M = Ca, Sr, Ba, Mn; L = H2O and CH3OH)......Page 148
References......Page 153
Spectroscopy and Characterization of Metal Clusters......Page 158
6 Metal Cation Coordination and Solvation Studied with Infrared Spectroscopy in the Gas Phase......Page 159
6.1 Introduction......Page 160
6.2 Metal-Carbonyl Complexes......Page 164
6.3 Metal–CO2 Complexes......Page 168
6.4 Metal–Water Complexes......Page 169
6.5 Metal–Acetylene Complexes......Page 175
6.6 Metal–Benzene Complexes......Page 179
6.7 Conclusions......Page 181
References......Page 182
7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages......Page 197
7.1 Introduction......Page 198
7.2.1 Dual-Laser Vaporization Nanocluster Source......Page 200
7.2.2 Spectroscopic Methods for Nanoclusters......Page 201
7.3.1 Mass Spectrometry for Si Cage Compounds......Page 202
7.3.2 Development of Intense Nanocluster Source with Magnetron Sputtering......Page 204
7.3.3 Immobilization and Characterization of Si-based BCSs on Solid Surface......Page 205
7.3.4 Dispersion Trapping of Si-Based BCSs in Liquid......Page 206
7.3.5 Structural Analysis of Si-based BCSs......Page 207
7.4.1 Mass Spectrometry of Al-based BCSs......Page 209
7.4.2 Anion Photoelectron Spectroscopy for Al-based BCSs......Page 212
7.4.3 Photoionization Spectroscopy for Al-based BCSs......Page 214
7.4.4 Theoretical Calculations for Al-Based BCSs......Page 215
7.4.5 Size Evolution of Al-Based Nanoclusters for Assembled Materials......Page 216
7.5 Conclusions......Page 219
References......Page 220
8 Characterization of Chemically Modified Gold/Silver Superatoms in the Gas Phase......Page 225
8.1.1 Metal Clusters as Novel Functional Units......Page 226
8.1.2 Naked Gold Clusters as Prototypical Superatoms......Page 228
8.1.3 Ligand-Protected Gold Clusters as Chemically Modified Superatoms......Page 229
8.2.1 Experimental Methods......Page 232
8.2.3 Characterization of Transient Clusters in Solution......Page 233
8.3.2 Fragmentations from Synthesized Clusters......Page 237
8.4 Ion-Mobility–Mass Spectrometry (IM MS)......Page 240
8.4.1 Experimental Methods......Page 241
8.4.3 Collision-Induced Isomerization......Page 242
8.5.1 Experimental Methods......Page 244
8.5.2 Electron Affinities of (M13)5+ (M = Au, Ag) Superatoms......Page 245
8.5.3 Thermionic Emission from (M13)5+ Superatoms......Page 246
8.6.1 Photodissociation Mass Spectrometry (PD MS)......Page 248
8.6.2 Surface-Induced Dissociation (SID)......Page 249
8.7 Summary......Page 250
References......Page 251
Dynamics of Vibrationally and Electronically Excited State Molecules, Ions, and Clusters......Page 256
9 Time-Resolved Study on Vibrational Energy Relaxation of Aromatic Molecules and Their Clusters in the Gas Phase......Page 257
9.1 Introduction......Page 258
9.2 VER from the OH, CH and NH Stretching Vibrations of Bare Molecules......Page 259
9.3 IVR of the OH Stretching Vibration of Hydrogen-Bonded Clusters of Phenol......Page 271
9.4 VER of the CH Vibration of Benzene Dimer and Trimer......Page 275
References......Page 283
10 Non-adiabatic Dynamics of Molecules Studied Using Vacuum-Ultraviolet Ultrafast Photoelectron Spectroscopy......Page 287
10.1 Introduction......Page 288
10.2 Experimental......Page 290
10.3 Furan......Page 292
10.4 Nitromethane......Page 298
References......Page 303
11 Femtosecond Time-Resolved Photoelectron Spectroscopy of Molecular Anions......Page 307
11.1 Introduction......Page 308
11.2 A Brief Overview of Past Anion TRPES Studies......Page 311
11.3 Application—Electron Attachment and Photodissociation Dynamics in I−·Nucleobase Clusters......Page 313
11.4 Methodologies for I−·Nucleobase Clusters......Page 316
11.5 I−·CH3NO2, I−·U, and I−·T Clusters......Page 317
11.5.1 I−·CH3NO2......Page 319
11.5.2 I−·Uracil and I−·Thymine Binary Complexes......Page 323
References......Page 329
12 Excited States Processes in Protonated Molecules Studied by Frequency-Domain Spectroscopy......Page 336
12.1 Overview......Page 337
12.2.2 Time-Resolved Experiments [7–9]......Page 338
12.3.2 Proton (H+) on the Substituent (Amino Group)......Page 339
12.4.1 Simple Aromatic Benzene/Pyridine/Pyrimidine......Page 341
12.4.2 Protonated Polycyclic Aromatic Hydrocarbons (Protonated PAH)......Page 343
12.4.3 Aromatic Amines, Aniline, Aminophenol......Page 345
12.4.4 Phenylethylamine, Tyramine [44]......Page 346
12.4.5 Tautomers and Electronic Spectroscopy of Protonated Amino PAHs......Page 347
12.4.6 Protonated Aromatic Amino Acids, Phenylalanine (PheH+), Tyrosine (TyrH+), Tryptophan (TrpH+)......Page 348
12.4.7 DNA Bases and Substituted Pyridine......Page 354
12.5 Conclusions......Page 359
References......Page 360
13 Time-Resolved Study on Photo-Initiated Isomerization of Clusters......Page 365
13.1 Introduction......Page 366
13.2 Experimental Strategies......Page 368
13.2.1 Nanosecond Static and Picosecond Time-Resolved REMPI-IR Spectroscopy......Page 369
13.2.2 Nanosecond Static EI-IR Spectroscopy......Page 372
13.3 Isomerization in Aromatic Clusters......Page 374
13.3.1 PhOH-Kr Dimer......Page 375
13.3.2 PhOH-Ar2 Trimer......Page 379
13.3.3 PhOH-Ar3 Tetramer......Page 382
13.3.4 PhOH-CH4 Dimer......Page 384
13.4 Concluding Remarks......Page 386
References......Page 388

Citation preview

Takayuki Ebata · Masaaki Fujii Editors

Physical Chemistry of Cold GasPhase Functional Molecules and Clusters

Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters

Takayuki Ebata Masaaki Fujii •

Editors

Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters

123

Editors Takayuki Ebata Hiroshima University Higashi-hiroshima, Japan

Masaaki Fujii Tokyo Institute of Technology Yokohama, Japan

ISBN 978-981-13-9370-9 ISBN 978-981-13-9371-6 https://doi.org/10.1007/978-981-13-9371-6

(eBook)

© Springer Nature Singapore Pte Ltd. 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface

A technique of generation of cold gas-phase molecules, molecular ions, and clusters has been extensively developed in past several decades. In addition, a variety of laser spectroscopic methods is now available to measure the vibrational and electronic spectra of the low-density cold gas-phase molecules. Also, there has been a great progress of quantum chemical calculation to determine the geometrical and electronic structure of molecules and clusters. By combining the advanced experimental techniques and theoretical/computational methods, we are now able to carry out detailed study on photophysics, photochemistry, and reaction dynamics of gas-phase molecules and clusters having variety of functionalities, whose structures are well defined. This book contains the recent experimental and theoretical studies on the structures and photochemistry of cold gas-phase large-size molecules, molecular clusters, host–guest complexes in the neutral and ionic forms, and metal clusters. The first chapter introduces experimental technique of generating cold gas-phase neutral and ionic molecules and their clusters. Special methods to vaporize nonvolatile molecules and generation of metallic clusters are also described. Then, laser-based spectroscopic techniques to measure the electronic and vibrational spectra for the low-density cold gas-phase species are described. The book also describes the ultrafast time-resolved spectroscopic technique on the reaction and energy relaxation of photoexcited species. After the general description of the principle of the experimental methods, specific topics are presented by the scientists who are working actively at the forefront of cluster science and laser spectroscopy. We hope this book will be useful for researchers studying in spectroscopy and dynamics of gas-phase molcules, and particularly for younger researchers and graduate students who are interested in the recent progress of the fundamental research on the functional molecules. We also hope the book is useful for the scientists who are working in the condensed phase to know how the gas-phase study can deal with the molecular interactions in detail.

v

vi

Preface

Finally, we would like to thank all the contributors to this book for presenting their up-to-date researches, and Mr. Shin’ichi Koizumi and Ms. Asami Komada for their assistance in editing this book. Higashi-hiroshima, Japan Yokohama, Japan Spring 2019

Takayuki Ebata Masaaki Fujii

Contents

Part I 1

Experimental Methods

Experimental Methods: Generation of Cold Gas-Phase Molecules, Molecular Ions, Their Clusters, Metal Clusters, and Laser Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Takayuki Ebata, Yoshiya Inokuchi and Atsushi Nakajima

3

Part II Spectroscopy and Characterization of Gas-Phase Functional Molecules and Clusters 2

Microscopic Study on Molecular Recognition of Host–Guest Complexes Between Crown Ethers and Aromatic Molecules . . . . . Takayuki Ebata

3

Chirality Effects in Jet-Cooled Cyclic Dipeptides . . . . . . . . . . . . . . Ariel Pérez-Mellor and Anne Zehnacker

4

Hydrogen Bond Networks Formed by Several Dozens to Hundreds of Molecules in the Gas Phase . . . . . . . . . . . . . . . . . . Asuka Fujii

5

35 63

89

Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Yoshiya Inokuchi

Part III

Spectroscopy and Characterization of Metal Clusters

6

Metal Cation Coordination and Solvation Studied with Infrared Spectroscopy in the Gas Phase . . . . . . . . . . . . . . . . . 157 Michael A. Duncan

7

Superatomic Nanoclusters Comprising Silicon or Aluminum Cages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Atsushi Nakajima

vii

viii

8

Contents

Characterization of Chemically Modified Gold/Silver Superatoms in the Gas Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Kiichirou Koyasu, Keisuke Hirata and Tatsuya Tsukuda

Part IV 9

Dynamics of Vibrationally and Electronically Excited State Molecules, Ions, and Clusters

Time-Resolved Study on Vibrational Energy Relaxation of Aromatic Molecules and Their Clusters in the Gas Phase . . . . . 257 Takayuki Ebata

10 Non-adiabatic Dynamics of Molecules Studied Using Vacuum-Ultraviolet Ultrafast Photoelectron Spectroscopy . . . . . . . 287 Shunsuke Adachi and Toshinori Suzuki 11 Femtosecond Time-Resolved Photoelectron Spectroscopy of Molecular Anions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Alice Kunin and Daniel M. Neumark 12 Excited States Processes in Protonated Molecules Studied by Frequency-Domain Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . 337 Jennifer Noble, Claude Dedonder-Lardeux and Christophe Jouvet 13 Time-Resolved Study on Photo-Initiated Isomerization of Clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 Masaaki Fujii and Otto Dopfer

Part I

Experimental Methods

Chapter 1

Experimental Methods: Generation of Cold Gas-Phase Molecules, Molecular Ions, Their Clusters, Metal Clusters, and Laser Spectroscopy Takayuki Ebata, Yoshiya Inokuchi and Atsushi Nakajima

Experimental technique for the generation of cold gas-phase neutral and ionic species and various laser spectroscopic methods will be described in this section

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_1

3

4

T. Ebata et al.

Abstract In this chapter, we describe the methods of generating cold neutral and ionic (cation and anion) molecules, their clusters, and metal clusters in the gasphase. First, a technique of supersonic free-jet or supersonic beam to generate cold neutral molecules and clusters is described. In addition, heating and laser ablation nozzles for the geneation of supersonic free-jet of nonvolatile molecules, such as high melting point and bio-related molecules, are introduced, while the methods of laser ablation and magnetron sputtering to generate metal clusters are also described. We then introduce various laser spectroscopic methods to measure the electronic and vibrational spectra for the jet-cooled molecules. Laser-induced fluorescence (LIF) and resonance-enhanced two-photon ionization (R2PI) spectroscopy is used to measure the electronic spectrum. UV–UV hole-burning (UV-UV HB) spectroscopy is used to discriminate the electronic transitions of different conformers and isomers. For the measurement of the vibrational spectrum of a specific molecule or cluster, we apply infrared-ultraviolet double-resonance (IR-UV DR) spectroscopy. If the molecule has no chromophore, a combination of IR and vacuum UV laser light (IRVUV) is used to obtain the vibrational spectrum. Second, we describe the generation methods of gas-phase cold ionic molecules and clusters. The gas-phase ions are generated by resonant-enhanced multi-photon ionization, electron impact, electron attachment, matrix-assisted laser disorption/ionization (MALDI), and electrospray ionization (ESI). Cooling of the ions is achieved by supersonic expansion or by the use of cryogenically cooled ion-trap. A time-of-flight (TOF) mass spectrometry or quadrupole mass filter is used for the mass selection, which is also applicable to obtain the single-sized metal clusters selectively. To obtain the electronic and vibrational spectra of the ionic species, we apply UV photodissociation (UVPD) and IR multiphoton dissociation (IRMPD), respectively. IR-UV DR spectroscopy is also used to measure the IR spectrum of a specific ion. In addition to the detection of the ions, a measurement of the photo-ejected electron, called photoelectron spectroscopy, is

Present Address: T. Ebata (B) Department of Applied Chemistry, National Chiao Tung University, Hsinchu, Taiwan e-mail: [email protected] T. Ebata · Y. Inokuchi Department of Chemistry, Graduate School of Science, Hiroshima University, Higashi-Hiroshima 739-8526, Japan e-mail: [email protected] A. Nakajima Faculty of Science and Technology, Department of Chemistry, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan e-mail: [email protected] Keio Institute of Pure and Applied Sciences (KiPAS), Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan

1 Microscopic Study on Molecular Recognition …

5

also described. Finally, we introduce pump–probe spectroscopy to investigate the dynamics of the vibrationally and electronically excited molecules and clusters. Keywords Supersonic jet · Laser ablation · Electrospray ionization · Cold ion-trap · Resonance two-photon ionization · IR-UV double-resonance spectroscopy · UV-UV hole-burning spectroscopy · Ultraviolet photodissociation spectroscopy · Metal cluster · Magnetron sputtering

1.1 Generation of Cold Gas-Phase Neutral Molecules and Clusters The cooling of the neutral gas-phase molecules is achieved by supersonic free-jet expansion method [1–14]. Figure 1.1 shows a schematic picture of the experimental setup combined with the supersonic free-jet. The mixture of gaseous sample and carrier gas, such as He, Ne, or Ar, at the total pressure of 3–5 atm, is expanded into vacuum through an orifice having a diameter of 0.5–1 mm. The expansion is operated by pulsed manner, normally at 10 Hz. The cooling of the molecules occurs during the adiabatic expansion in the vacuum. The collisions in the adiabatic expansion region lead to the very narrow velocity distribution, typically ~10 K in terms of the temperature. The “vibration (V)–translation (T)” and “rotation (R)–translation (T)” energy transfers in the expansion region also lead to the cooling of the vibrational and rotational degrees of freedom. Thus, most of the molecules are populated at the ground vibrational level (v = 0) and low-energy rotational levels. This distribution makes the vibrational and electronic spectra very simple. During the collisional cooling process, molecular clusters, which are bound by weak noncovalent interactions, Fig. 1.1 Experimental setup of supersonic beam, and mass-resolved R2PI spectroscopy. V: pulsed nozzle, F: sample folder, S: skimmer, R: repeller, E: extractor, D: detector (multi-channel plate)

6

T. Ebata et al.

are also formed. By skimming a supersonic free-jet with a skimmer, a molecular beam is obtained. For the liquid or solid samples, they are heated in the sample housing to obtain enough vapor pressure before the expansion. Figure 1.2 shows a typical heating pulse nozzle [15]. In this setup, a sample housing is attached to the head of a commercially available pulse nozzle (General valve, series 9). The solid sample is put in the sample housing at the top and is heated by a spiral heater. The housing is made of polyimide resin, which can be heated up to 200 °C. A carrier gas (He, Ne, or Ar) is filled from the opposite side of the nozzle. The mixture of the sample and carrier gas is expanded into vacuum through a ~1 mm aperture of the housing by applying square-shaped current to a solenoid of the nozzle with a voltage of 100–200 V and pulse-width of 80 μs. For the molecule which is hard to be vaporized, or decomposed by the heating, a laser ablation (laser desorption) method is used. The laser ablation technique has been described by several groups [16–25]. For laser ablation, we use fundamental (1.064 μm) output of a pulsed Nd3+ :YAG laser. Figure 1.3 shows the typical experimental setup of laser spectroscopy for supersonic beam with laser ablation [25]. The pulsed laser (normally nanosecond laser) with a power of 2–3 mJ/pulse is focused on the sample/matrix mixture with an f = 500 mm convex lens, where carbon black powder is used as the matrix. The heat of high-temperature plasma generated by the laser is immediately transferred to the samples and they rapidly vaporize before decomposition. The vaporized samples are mixed with carrier gas (Ar at a pressure of 15 atm) in the expansion region, and are cooled in a supersonic free-jet. In Fig. 1.3, we set a graphite disk near the aperture of the pulsed nozzle. The powder of the “sample/carbon black” mixture is rubbed on the graphite disk. The graphite disk is rotated by a stepping motor so that ablation laser irradiates to a new position of the

Fig. 1.2 a Schematic view of the high-temperature heating pulsed nozzle. b, c Photographs of the nozzle

1 Microscopic Study on Molecular Recognition …

7

Fig. 1.3 (Left) Setup of the laser ablation nozzle and R2PI laser spectroscopy. At the head of the pulsed nozzle, a channel nozzle is attached. H2 O/Ar mixture gas is expanded with sample vapor and the hydrogen-bonded cluster with H2 O is formed during the collisions in the channel nozzle. (Right) Photograph of the laser ablation nozzle with a skimmer

disk in each shot. In addition, a channel is put at the exit of the pulsed nozzle. This channel is used to generate clusters during the collision inside the channel.

1.2 Laser Spectroscopy to Measure the Electronic Transition 1.2.1 Laser-Induced Fluorescence (LIF) and Resonant Two-Photon Ionization (R2PI) Spectroscopy The electronic spectrum of the cold gas-phase molecules in the supersonic free-jet is obtained by laser-induced fluorescence (LIF) excitation and resonant two-photon ionization (R2PI) spectroscopy (Fig. 1.4a). In case of LIF spectroscopy, we measure the total fluorescence of the laser-excited molecules with a photomultiplier tube, and in R2PI spectroscopy, we measure ions generated by two-photon absorption. The ions are mass-separated by time-of-flight tube and detected by a multi-channel plate or channeltron. In the mass-selected R2PI spectroscopic measurement, in addtion to the electronic transition we can identify the species and clusters by their masses, which cannot be done in the LIF measurement. Another difference between the LIF and R2PI spectroscopy is the relative intensity of the vibronic bands. The intensity of the fluorescence, If , in LIF and that of the ion, Iion , in R2PI is expressed by,  2 If = C  μS0 →S1  · QY · Ilaser , and  2  2 2 , Iion = C  μS1 →Ion  μS0 →S1  · Ilaser

8

T. Ebata et al.

respectively. Here, μS0 →S1 and μS1 →ion are the transition moments from S0 to S1 and from S1 to the ionic state, respectively, QY is the fluorescence quantum yield, and Ilaser is the laser intensity. C  and C  are the instrument-dependent constants. By comparing the two equations, we notice that the fluorescence intensity is proportional to the fluorescence quantum yield of the vibronic bands of the S1 state. In general, the fluorescence quantum yield becomes smaller with the excess energy due to the competition of the nonradiative decay process, such as internal conversion or intersystem crossing. Thus, the LIF spectrum becomes weak at higher vibronic bands, as shown in the upper panel of Fig. 1.4b. On the other hand, such a factor is not involved in the R2PI spectrum, so the R2PI spectrum looks like absorption spectrum (lower panel of Fig. 1.4b). However, if the S1 decay rate is comparable to or faster than the ionization rate, the ionization signal becomes weak and the vibronic band becomes broad.

1.2.2 Ultraviolet–Ultraviolet Hole-Burning (UV–UV HB) Spectroscopy In case of flexible molecules and clusters, several conformers and isomers are expected to exist at nearly equal energies. If the barrier height between the conformers or isomers is high, the relative population does not reach to equilibrium during the expansion, and they coexist in the jet. Although the mass spectrometry is not able to discriminate them, they have different electronic transition energies. So, we apply ultraviolet–ultraviolet hole-burning (UV–UV HB) spectroscopy to discriminate them [26–28]. In the UV–UV HB spectroscopy, we use two tunable UV lasers, one for probe and the other for making a population hole (burn-laser, HB) in

Fig. 1.4 a Energy diagram and schemes (laser-induced fluorescence, LIF, and resonant two-photon ionization, R2PI) of spectroscopy of the electronic transition of jet-cooled molecules. b (Upper) LIF spectrum, (lower) R2PI spectrum. For more details, see the text

1 Microscopic Study on Molecular Recognition …

9

Fig. 1.5 a Scheme of UV–UV hole-burning spectroscopy with fluorescence detection. Here, probe laser monitors the population of isomer B. b (Upper) LIF spectrum, (lower) UV–UV hole-burning spectrum of isomer B

the vibrational ground state (Fig. 1.5a). The two lasers are irradiated to the supersonic jet or molecular beam synchronously. The probe UV laser monitors the ground state population of a specific species (conformer or isomer) in the supersonic jet or molecular beam by LIF or R2PI. Under this condition, the burn-laser pulse is introduced at ~10 mm upper stream of the supersonic jet (molecular beam). The power of the burn-laser is strong enough to make a population hole in the ground state (v = 0). The timing between the burn and probe lasers, t, is set so that the probe laser monitors the population at the same spatial area of the jet where the burn-laser creates the population hole. A distance of 10 mm corresponds to t ~ 5 μs. When the burn-laser creates a population hole by an intense laser excitation, the fluorescence or ion signal intensity monitored by the probve laser becomes weak. On the other hand, such the weaking (depletion) does not occur when the burn-laser makes a population hole of the other species since the probe laser monitors different species. Thus, the UV–UV HB spectrum is obtained as a fluorescence-dip or ion-dip spectrum (Fig. 1.5b). The reason for the different positioning between the burn and probe lasers is to eliminate the interference caused by the strong fluorescence or ion signals generated by the burn-laser. In case of the fluorescence detection, the aromatic molecules have the S1 lifetimes of a few nanoseconds to hundreds of nanoseconds in general, so that the fluorescence signals due to the burn can be easily separated. In the case of ion detection, we can remove the ions generated by the burn-laser by applying an electric field before the detection by the probe laser.

1.3 Laser Spectroscopy to Measure the Vibrational Levels Vibrational spectroscopy is essential to determine the structures of molecules and clusters. In this section, two types of vibrational spectroscopy are described. First is the dispersed fluorescence and stimulated emission pumping spectroscopies,

10

T. Ebata et al.

which are useful to observe the Franck–Condon (F–C) active vibrations. The other is infrared-ultraviolet double-resonance and stimulated Raman-ultraviolet doubleresonance spectroscopies, to observe the IR and Raman active vibrations, respectively.

1.3.1 Dispersed Fluorescence Spectroscopy and Stimulated Emission Ion-Dip Spectroscopy In most of the aromatic molecules, the S1 state is ππ* and the C–C bond becomes longer in S1 , so the vibrations involving the movements of carbon atoms are Franck–Condon active in the S1 –S0 transition. Thus, they appear strong in the LIF and dispersed fluorescence (DF) spectra (Fig. 1.6 left). The right upper panel of Fig. 1.6 shows the DF spectrum of jet-cooled phenol–H2 O cluster obtained by excting to the zero-point level of S1 [29]. Here, the spectral resolution of monochromator is set to 10 cm−1 . In the spectra, the bands 10a, 12, and 1 are the vibrations involving the movements of C atoms of the phenyl ring, and β and σ are the intermolecular bending and stretching modes, respectively. The intensity of each band is determined by the F–C factor between S1 and S0 . Similar spectrum can be obtained by stimulated emission spectroscopy with better spectral resolution and higher sensitivity. Figure 1.6 (left) shows a scheme, which is called stimulated emission ion-dip (SEID) spectroscopy. In this spectroscopy, two tunable lasers are used; one (hν 1 ) for the excitation to the S1 vibronic level and the other (hν 2 ) for the stimulated emission to the vibrational levels of S0 . The stimulated emission by hν 2 transfers the S1 state molecules to the vibrational levels (v ) of S0 , leading a depopulation (population hole) of the S1 state. In SEID spectroscopy, hν 2 works not only for stimulated emission but also for the ionization of the S1 state

Fig. 1.6 (left) Schemes of dispersed fluorescence (DF) spectroscopy and stimulated emission iondip (SEID) spectroscopy. (Right) DF spectrum and SEID spectra of phenol–H2 O (1:1) cluster from the zero-point level of S1 . DF was obtained with 10 cm−1 spectral resolution of a monochromator. SEID was obtained with a laser having 0.2 cm−1 spectral resolution

1 Microscopic Study on Molecular Recognition …

11

molecules. By scanning ν 2 frequency, we obtain the continuous ionization efficiency curve from S1 . When the stimulated emission to the vibrational level of S0 occurs, the molecules in S1 is dumped to S0 (v ) and the ion intensity decreases, leading to the depletion of the ion signal. Thus, we obtain the stimulate emission spectrum as an ion-dip spectrum. Since the stimulated emission induces the population transfer by an intense laser, even a weak band in the DF spectrum clearly appears in the SEID spectrum, as seen in the right lower panel of Fig. 1.6. Another chracteristic of SEID spectroscopy is that the molecules dumped to v by stimulated emission may relax to isoenergetic S0 vibrational leves by intramolecular vibrational energy redistribution (IVR). If the IVR at v occurs faster than that of the reabsoprtion to S1 by hν 2 , the moleucles at v will not be pumped to S1 again so that the dip becomes deeper. Thus, this specroscopy is suitable for high energy vibrational levels or the vibrations of clusters having larger IVR rate constants.

1.3.2 Infrared-Ultraviolet (IR-UV) and Stimulated Raman-Ultraviolet (Raman-UV) Double-Resonance Spectroscopies In general, the infrared and Raman transitions are weaker than the electronic transition, so the vibrational spectroscopy is difficult to apply to the low-density molecules in the supersonic jet. However, by using a double-resonance technique, the vibrational spectroscopy becomes a high-sensitive and species-selective method, which we call infrared-ultraviolet double-resonance (IR-UV DR) (Fig. 1.7a) [30–34] and stimulated Raman-ultraviolet double-resonance (Raman-UV DR) spectroscopies

Fig. 1.7 a Scheme of (left) IR-UV double-resonance spectroscopy and (right) stimulated RamanUV double-resonance spectroscopy combined with R2PI. b (Upper) IR-UV double-resonance spectrum, (lower) stimulated Raman-UV double-resonance spectrum. Owing to the difference of IR and Raman activity, the relative intensities are different between the two spectra

12

T. Ebata et al.

(Fig. 1.7b) [35–39]. In these spectroscopies, we monitor the population hole in the vibrational ground level of S0 created by the vibrational excitation, similar to UV–UV HB spectroscopy. In IR-UV DR spectroscopy, a tunable IR laser is obtained by difference frequency generation (DFG) between fundamental (1.064 μm) or second harmonics (532 nm) of Nd3+ :YAG laser and the dye laser pulase pumped by the same YAG laser, or by optical parametric oscillation (OPO). Typical spectral resolution is 0.5 cm−1 and power is a few millijoule for the tunable IR laser. In the simulated Raman pumping method, two laser pulses are used to induce stimulated Raman process; second harmonics (532 nm, hν 1 ) of YAG laser and a dye laser (hν 2 ) pumped by the same YAG laser. Different from UV–UV HB spectroscopy, the IR laser beam (or Raman pumping lasers) is introduced coaxially with the probe UV laser beam to the free jet (molecular beam). This is because the vibrational pumping laser does not generate ion or fluorescence. The timing between the IR or Raman pump laser and the probe UV laser is set at ~50 ns. When the IR laser or stimulated Raman pump creats the population hole at v = 0 at resonant condition, the fluorescence or ion intensity monitored by the probe laser is depleted. Thus, in both spectroscopic methods, we obtain the vibrational spectrum as a fluorescence-dip or ion-dip spectrum. The advantage of these spectroscopies is that they provide us with a conformer or isomer-specific IR or Raman spectrum because the probe laser monitors the specific species in the supersonic jet or molecular beam. By comparing the observed vibrational spectrum with the calculated one obtained by density functional theory (DFT) or ab initio calculations, we can determine the structure of the molecules as well as clusters. Among many vibrational modes, the information of the X–H (X = C, N, O, etc.) stretching vibration is very useful to investigate the hydrogen (H)-bonding network of the clusters.

1.3.3 Infrared-Vacuum Ultraviolet (IR-VUV) Spectroscopy IR-VUV spectroscopy was developed to obtain the IR spectrum of the molecule which does not have a chromophore. This spectroscopy utilizes the tunable IR laser and a VUV light source. For the VUV light source, frequency tripling of the third harmonics (355 nm) of Nd3+ :YAG laser is normally used. The VUV output with a wavelength of 118 nm is obtained by focusing the 355 nm laser light into the gas cell containing rare gas; Xe/Ar (1:10) mixture at a pressure of 200 torr. The cell has an MgF2 window or lens at the output. The input power of 355 nm is 10–20 mJ. The tunable IR and the 118 nm VUV laser lights are coaxially focused on the molecular beam, or sometimes a LiF prism is placed for the VUV output to separate intense 355 nm laser light. Two types of IR-VUV spectroscopy are shown in Fig. 1.8. Figure 1.8a (left) shows a scheme of IR photodissociation (IRPD)-VUV spectroscopy [40–43]. In this spectroscopy, a population hole at v = 0 created by the photodissociation (PD) of a tunable IR laser is measured as the depletion of the ion signal monitored by VUV laser light. Figure 1.8b shows an enhance-type spec-

1 Microscopic Study on Molecular Recognition …

13

Fig. 1.8 a Schemes of (left) IRPD-VUV spectroscopy and (right) IR-VUV spectroscopy. b (Upper) IRPD-VUV spectrum, (lower) IR-VUV spectrum

troscopy, IR-VUV [41]. In this scheme, the energy of the VUV light is lower than the ionization potential (IP0 ) of the molecule, but the “IR + VUV” energy exceeds IP0 . Thus, whenever the IR laser is resonant to the vibrational transition, we obtain the ion-enhancement corresponding to the vibrational bands of S0 . IRPD-VUV is suited for the molecular clusters, while IR-VUV is suited for the monomers.

1.3.4 Laser Spectroscopy to Measure the Vibrational Levels of the Electronically Excited State When the timing between the IR and UV lasers is reversed in the IR-UV DR scheme, we can observe the vibrational spectrum of the S1 state [33, 44–46]. This spectroscopy is called as UV-IR DR spectroscopy. As was described above, the S1 lifetime of aromatic molecules is a few tens of nanoseconds. Thus, in the UV-IR DR scheme, we first pump the molecule to the zero-point level of S1 and at a few nanoseconds delay time we introduce a tunable IR laser pulse to excite the S1 state molecules to higher vibronic levels. Since the fluorescence quantum yields of the vibronic levels are smaller than that of the zero-point level, we see a depletion of the fluorescence intensity when the molecules are excited to the vibronic level. So, by scanning the IR frequency while monitoring the total fluorescence, we observe the IR spectrum of the S1 state as a fluorescence depletion spectrum. Another method of obtaining the vibrational spectrum of the S1 state molecule is to detect the emission of the IVR relaxed vibrational levels or the fragment species in the case of clusters. In these methods, we need a monochromator. The emission from the IVR relaxed levels is very broad different from the sharp emission from a single vibronic level, so we fix the wavelength of the monochromator to detect only the broad emission. By

14

T. Ebata et al.

scanning the IR frequecny while monitoring the broad emission from the IVR relaxed levels, we obain IR spectrum. The spectrum is backgroud free so that the signal-tonoize ratio is better than the fluroescence-dip spectrum. Also, the emission of the fragments produced by vibrational predissociation appears at different wavelengths and we can also discriminate them from the emission of the parent clusters by the monochromator. Figure 1.9a shows the excitation scheme of UV-IR DR spectroscopy. Figure 1.9b shows the dispersed fluorescence spectrum of aniline from the zero-point level of S1 . Figure 1.9c shows the enlarged portion of DF with and without IR excitation to the NH2 stretching vibration. The broad band appeared beneath the sharp band at 34,000 cm−1 is due to the fluorescence from the IVR relaxed levels. Figure 1.9d shows the UV-IR double-resonance spectrum of aniline showing the vibrational spectrum of S1 in the 2950–3900 cm−1 region. In this measurement, the monochromator is set to detect only the broad fluorescence from the IVR relaxed levels. Figure 1.9e shows the IR-UV double-resonance spectrum in S0 for comparison.

Fig. 1.9 a Scheme of UV-IR DR spectroscopy for the vibrational spectroscopy of the S1 state. b DF spectrum of aniline from v = 0 of S1 . c Enlarged portion of DF with and without IR excitation. d UV-IR DR spectrum in the 2950–3900 cm−1 region of S1 . e IR-UV DR spectrum of S0 . The bands at ~3000 cm−1 are the CH stretches, and those at 3300–3500 are the stretch of NH2 group. In (d), the bands higher than 3600 cm−1 are the transition to S2 state [46]. Reprinted with permission from J. Phys. Chem. A 106, 11070–11074 (2002). Copyright 2002 American Chemical Society

1 Microscopic Study on Molecular Recognition …

15

1.4 Generation of Cold Gas-Phase Molecular Ions There are several methods to generate cold molecular ions in the gas phase. 1. Resonance-enhanced multi-photon ionization (REMPI) Resonance-enhanced multi-photon ionization (REMPI) is used not only to measure the S1 –S0 electronic transition of neutral molecules, such as R2PI, but also to generate the ions by state-selective manner. For example, when the molecules are ionized via the Rydberg state, Franck–Condon prefers the generation of the ions having the same vibrational quantum with that of the Rydberg state, because the ion core of the Rydberg state and the ion have almost same structure with each other. In addition, by using two-color scheme with the wavelength of the ionization laser close to the ionization threshold, we can generate the cold ions. We can also generate cold ions by using collisions in the expansion region. The collision with carrier rare gas during the expansion cools down the internal temperature of the ions generated by REMPI. 2. Electron Impact (Electron Ionization) In this case, we collide the energetic electrons to the gas-phase neutral molecule. Owing to the collision of high-energy electron to the molecule, an electron is removed from the molecule to generate a cation (positive ion). In the case of free-jet, the generated cation is cooled in the expansion region of the free jet. The high-energy collision sometimes causes an extensive fragmentation of the ions. 3. Electron Attachment In this case, an electron with thermal energy softly collides with gas-phase neutral molecule. Under the soft collision, an electron is attached to the neutral molecule and occupies the LUMO orbital to form negative ion. For the negative ion to be formed, the molecule should have a positive electron affinity. 4. Matrix-Assisted Laser Desorption Ionization (MALDI) [47–49] MALDI is a technique similar to the laser ablation described earlier, but in most of the case it produces protonated species. This method is used to investigate the structure of nonvolatile large-size molecules and biomolecules by mass spectrometry. 5. Electrospray Ionization (ESI) [50–52] ESI is a widely used method to generate not only positive but also negative ions. In the next section, this method will be described in more detail.

1.4.1 Electrospray Ionization with Ion Trap Electrospray ionization (ESI) is a powerful method for the extraction of ions in solutions to the gas phase [51, 53]. ESI was initially utilized extensively for mass spectrometry measurements of large (biomolecular) species [50, 54]. For spectroscopic purposes, some researchers tried to use the ESI for fluorescence spectroscopy

16

T. Ebata et al.

[55, 56]; these experiments have been performed for the plume produced by the ESI source in atmosphere. For gas-phase spectroscopy in vacuo, Wang et al. successfully performed photoelectron spectroscopy of multi-charged anions in the gas phase with an electrospray ion source in 1998 [57–59]. Figure 1.10 shows the ESI source for multiple charged anions developed by Wang et al. [57]. Gerlich et al. developed a cold ion trap to achieve interstellar conditions for ion–molecule reactions [60, 61]. Wang et al. reported photoelectron spectroscopy of anions cryogenically cooled in a quadrupole ion trap [59, 62]. One of the pioneering works for spectroscopy of cold ions was reported by Boyarkin et al. [63]. Figure 1.11 shows the tandem mass spectrometer build by Boyarkin et al. for ion spectroscopy [63]. It consists of a nanoelectrospray ion source, quadrupole mass filters, and a cryogenically cooled 22-pole ion trap. Mass-selected ions are cooled in the ion trap and irradiated by an output of a UV laser. The resulting fragment ions are mass-analyzed by the quadrupole mass spectrometer before being detected. UV spectra of cold ions are obtained by plotting the yield of photofragment ions as a function of the UV frequency. Figure 1.12 displays the temperature effect of UV spectra for protonated tyrosine (TyrH+ ). Under cold conditions, the UV spectrum of TyrH+ shows sharp, well-resolved features. This indicates that ion spectroscopy under cold gas-phase conditions is a powerful tool to obtain spectroscopic information of ions. Since then, a number of cooled linear and Paul ion traps have been built for ion spectroscopy [63–80].

Fig. 1.10 Electrospray ion source coupled with a quadrupole ion trap and a time-of-flight mass spectrometer [57]. Reprinted with permission from Rev. Sci. Instrum. 70, 1957–1966 (1999). Copyright 1999 AIP Publishing.

1 Microscopic Study on Molecular Recognition …

17

Fig. 1.11 A schematic drawing of a tandem mass spectrometer with a nanoelectrospray ion source and a cold, 22-pole ion trap [63]. Reprinted with permission from J. Am. Chem. Soc. 128, 2816–2817 (2006). Copyright 2006 American Chemical Society

1.5 Metal Cluster Generation Methods Laser ablation or magnetron sputtering is mainly used for generating metal clusters in vacuum. A feature common to these generation methods is that a cooling gas, known as a carrier gas, is used to generate metal clusters by the aggregation of metal atoms. Details of these methods are provided in the following two subsections.

1.5.1 Laser Ablation Method Laser ablation (also known as laser vaporization or laser evaporation) is a method used for generating metal clusters by irradiating a pulsed laser onto a metal target (rod or disk), as shown in Fig. 1.13. This method was developed independently by Smalley et al. [81], and Bondybey and English [82] in the early 1980s. Via a multiphoton process induced by laser focusing, the temperature of the plasma generated at the target surface reaches about 10,000 K, thus enabling the evaporation of metal elements with high melting points, which constitute a major feature available for all solid targets. The high-temperature atomic vapor forms metal clusters by colliding with the carrier gas of helium (He) to aggregate, and a metal cluster beam is generated along the He gas flow. Since a pulsed valve synchronized with the pulsed laser allows for a reduced amount of He carrier gas, it allows us to use rather compact pumping

18 Fig. 1.12 Ultraviolet photodissociation (UVPD) spectra of protonated tyrosine observed under uncooled and cooled conditions in the gas phase [63]. Reprinted with permission from J. Am. Chem. Soc. 128, 2816–2817 (2006). Copyright 2006 American Chemical Society

Fig. 1.13 Schematic of a laser ablation cluster source with a sample rod.

T. Ebata et al.

1 Microscopic Study on Molecular Recognition …

19

system for the vacuum chamber. Moreover, an Even-Lavie pulsed valve with a high stagnation pressure of 10 MPa (100 atm) [83] facilitates the efficient formation of metal clusters and leads to lower internal temperature of the metal clusters (vibration and rotation temperatures). A pulsed valve that reduces the required amount of carrier gas is widely used in laser ablation. Since atomic cations, A+ , and electrons coexist in the laser plasma during ablation, charged cluster ions of A+n and A− n can be generated as well as neutral clusters, An . When the target is a multi-element alloy, mixed alloy clusters can also be generated. Also, dual laser ablation can be used for generating alloy clusters; in this case, two lasers independently irradiate two different metal targets (see Chapter 7, Fig. 7.2). When irradiating a pulsed laser onto a rod or disk target for a long time, fluctuation in the metal cluster generation per laser pulse can be reduced by employing mechanical translation and rotation of the rod or eccentric rotation of the disk, for which a motor is installed outside of the system or within the vacuum chamber. Despite these refinements, large fluctuation of the beam intensity in each laser shot and relatively high internal temperature of the generated metal clusters are still to be the disadvantages of laser ablation. Implementing laser ablation in combination with a time-of-flight (TOF) mass spectrometer is advantageous to measure size distributions of metal clusters because clusters can be generated in a pulsed manner.

1.5.2 Magnetron Sputtering Method Sputtering is a thin film deposition technique involving vapor deposition in vacuum over a period of time; atoms are ejected from a metal surface (target) after being bombarded by high-energy species. Since the mid-twentieth century, along with the development of vacuum techniques, sputtering has been widely used for producing antireflective films for optical applications and surface coatings of cutting tools. Magnetron sputtering is an improved sputtering method, in which a target is efficiently sputtered by confining the high-energy species (plasma) with a magnetic field. In the early 1990s, Haberland et al. developed a metal cluster generation method using magnetron sputtering [84]. Since sputtering can provide a beam flux sufficient to form a thin film, magnetron sputtering becomes more attractive than laser ablation for fabricating nanomaterials. A schematic diagram of a magnetron sputtering apparatus is shown in Fig. 1.14. A metal disk target is mounted on the front of the magnetron head, and argon (Ar) gas is allowed to flow in the vicinity of the target. The aggregation cell cooled by liquid N2 surrounds the magnetron head to feed cooled He gas into the cell. To sputter, high-voltage direct current (DC) is applied between the anode and the cathode, and Ar+ cations ionized by the discharge are accelerated to the cathode, colliding with the target to eject metal atoms. A chemically unreactive gas, typically Ar (also krypton (Kr) and xenon (Xe) are used) is used as a sputtering gas to maintain the target composition during operation; the sputtering efficiency generally increases as the atomic mass of the sputtering gas increases. Although the efficiency of Xe is high, Ar gas is typically used due

20

T. Ebata et al.

Fig. 1.14 Schematic of the experimental setup of a magnetron sputtering apparatus. Instead of a magnetron sputtering with DC power supply, high-power impulse magnetron sputtering (HiPIMS) is shown with a scheme for time-resolved measurements, where the retarding voltage (−V ) was grounded in the gate window time T 1 –T 2 , after the sputtering trigger initiated at T 0 . Reprinted with permission from Ref. [86]. Copyright 2013 American Chemical Society

to its lower cost. The secondary electrons generated in the collisions of Ar+ cations with the target are confined to the front of the target while being moved helically (cyclotron movement) along a magnet field placed at the back of the target. Plasma is obtained due to continuous collisions with Ar molecular flow, and the neutral metal atoms and atomic ions generated by magnetron sputtering grow into metal clusters in the aggregation cell due to carrier gas He cooling. In the case of a magnetic target material, a thinner target is used because the magnetic field becomes weaker in front of the target. When the target is an insulator, sputtering is carried out with radio-frequency discharge instead of DC discharge. In order to observe cluster-size distributions in a mass spectrum, combining magnetron sputtering with a quadrupole mass spectrometer rather than a TOF mass spectrometer is advantageous, owing to duty factors of the continuous (cw) beam. In addition to conventional DC magnetron sputtering (DC-MSP), high-power impulse magnetron sputtering (HiPIMS) has been developed, in which sputtering is performed in a pulsed manner [85, 86]. Under the same time-averaged power as the DC-MSP method, the HiPIMS method can elevate the voltage within a specific time window, thereby increasing the amount of charged species (atomic ions and electrons) from the sputtering target. Since the proportion of ions is enhanced by pulsed sputtering, the generation density of cluster cations and anions of A+n and A− n that capture electrons can be magnified more efficiently than can be achieved using the DC-MSP method. Furthermore, when higher temperature is achieved by

1 Microscopic Study on Molecular Recognition …

21

pulsed energetic sputtering at the target surface, the rapid cooling in the growth cell exhibits a steeper temperature gradient, thereby increasing the proportion of a thermodynamically stable chemical species of magic clusters in the size distributions.

1.6 Electronic and Vibrational Spectroscopy for Molecular Ions Similar to the neutral molecules and clusters, we measure the electronic and vibrational spectra of the molecular ions to investigate the electronic and geometrical structures, as well as their photophysics. Since the direct absorption is difficult due to the low density of the generated ions, we measure the absorption spectrum as an action spectrum.

1.6.1 UV- and IR-Photodissociation Spectroscopy When the cations (M+ ) are excited to the electronic states by UV laser light, they relax to the electronic ground state by internal conversion (IC). After IC, the ions have large internal energy and decompose to daughter ions (F+ ) and neutral (N) species. Thus, by monitoring the daughter ions (F+ ) while scanning the UV laser frequency, we obtain the UV photodissociation (UVPD) spectrum [63, 87, 88], corresponding to the UV absorption spectrum (Fig. 1.15a). Similar dissociation spectroscopy is possible for IR dissociation spectroscopy. However, one photon of IR light is not large enough to dissociate the covalent bond, so multiple-photon dissociation (IRMPD) method is used (Fig. 1.15b). In this spectroscopy, the first IR absorption is resonant to the vibrational transition. The second and further absorption is not the resonant process, but a successive absorption of photons occurs if the input IR laser power is strong. For the IR laser source, high-power tunable IR laser is necessary. In the case of the

Fig. 1.15 a Scheme of UVPD spectroscopy for the electronic spectroscopy of the ions. b Scheme of IRMPD spectroscopy for the vibrational spectroscopy of ions. c Scheme of L-tagging IR spectroscopy (L = Ar, Ne, and H2 )

22

T. Ebata et al.

molecular clusters, the intermolecular binding energy is comparable to or smaller than the vibrational energy. Thus, after the vibrational excitation by IR laser, the input energy is redistributed in the cluster ion by IVR and the cluster ion dissociates (Fig. 1.15c). Thus, by monitoring the fragment ion (M+ ) while scanning the IR frequency, we can measure the IR spectrum of the cluster ion. If we choose rare gas atom or light molecule, such as H2 , as a ligand, the perturbation to the molecular ion is small so that the observed IR spectrum can be assumed to be that of the monomer ion. This technique is called tagging or messenger technique (Fig. 1.15c) [89].

1.6.2 Double-Resonance Spectroscopy for Molecular Cation In the UVPD and IRMPD spectroscopic methods, we cannot distinguish the conformer and isomer of the ions having the same mass. To discriminate those species, several double-resonance spectroscopic methods have been developed. We first describe UV–UV hole-burning (HB) spectroscopy [90, 91] combined with the cold quadrupole ion trap (QIT). In this spectroscopy, we use two lasers, burn (pump) and probe lasers, similar to that described in UV–UV HB spectroscopy for neutral species. In the case of ions, an intense burn (pump) laser creates a population hole for the ions in QIT by the UV photodissociation. The fragments produced by the burn-laser in the ion trap are removed by applying an auxiliary radio-frequency (RF) voltage to the trap. Then a probe laser is introduced to the ion trap to monitor the same ion depleted by the burn-laser. Thus, a depletion of the the probe laser monitored ion signal occurs when the burn-laser depletes the same ions. By scanning the wavelength of the burn-laser, conformation or isomer-specific UV spectrum of the ions is obtained (Fig. 1.16a) [90]. Another type of UV–UV HB spectroscopy is also reported by combining QIT and time-of-flight (TOF) tube [91]. In this method, an intense pump UV laser generates a population hole for the ions trapped in the QIT, and a probe UV laser monitors the population depletion for the ions at the field-free region of the TOF mass spectrometer [91].

Fig. 1.16 a Scheme of UV–UV HB spectroscopy for the electronic spectroscopy of ions. b Scheme of IR-UV DR spectroscopy for the vibrational spectroscopy of ions

1 Microscopic Study on Molecular Recognition …

23

We then describe IR-UV double-resonance spectroscopy. In this method, the UV probe laser monitor the depletion of the ioninduced by the IR laser excitation. We first fix the probe UV laser frequency to a sharp vibronic band of the specific ion in the ion trap and monitor the the ion intensity. Under this condition, tunable IR laser pulse is irradiated to the ions in the ion trap with a timing prior to the UV laser. When the IR laser excites the ions to the vibrational level, the excited ions are immediately relaxed by IVR or dissociation. The electronic transition of the IVR relaxed levels are very broad in contrast to the sharp transition of the cold ion, so the ion intensity monitored by the probe laser becomes weak. This depletion of ion intensity occurs only when the IR and UV lasers monitor the same species. Thus, by scanning the wavelength of the IR laser, a conformation or isomer-specific IR spectrum of the cold ions is obtained [92, 93].

1.6.3 Photoelectron and Photodetachment Spectroscopy In these spectroscopies, we measure the kinetic energy distribution of electrons ejected by the photoionization of neutral species or photodetachment of anionic species [94, 95]. The ionization potential (IP0 ) of aromatic molecules is located in the range of 7.0–9.5 eV. In photoelectron spectroscopy of neutral species, we ionize them by R2PI with UV laser light or directly by VUV ionization. On the other hand, the electron affinity is less than ~4 eV and UV light can directly photodetach the LUMO-occupied electron of the anions. M + hν → M + (v, J ) + e− M − + hν → M (v, J ) + e− The internal distribution (v, J) of the generated ions is determined by the Franck–Condon factors between the neutral and ionized states and characters of the ejected electrons under the restriction of total energy and angular momnetum conservation.   hν = IP0 + EM + (v,J ) + KE e−   hν = EA + EM (v,J ) + KE e− Here, IP0 and EA are the adiabatic ionization potential and electron affinity, respectively. Thus, by measuring the kinetic distribution of the ejected electrons, one can determine IP0 , EA, and the information of the structures of the ion and neutral states, as well as structural difference between them. Figure 1.17 shows the energy levels diagram and schemes of (a) R2PI photoelectron spectroscopy and (b) photodetachment spectroscopy. Right panel shows the photoelectron and photodetachment spectra. The ejected electrons are detected by magnetic bottle time-of-flight spectrometer [96], or imaging method [97, 98]. By combining time-resolved pump–probe tech-

24

T. Ebata et al.

Fig. 1.17 Left a Scheme of photoelectron spectroscopy. b Scheme of photodetachment spectroscopy. Right (upper) Photoelectron kinetic energy spectrum, (lower) photodetachment spectrum. Here, the horizontal axis is the binding energy (= hν UV - electron kinetic energy)

nique to these spectroscopic methods, we can investigate the dynamics of the excited state.

1.7 Time-Resolved Spectroscopy In Sects. 1.2 and 1.3, we described the generation of cold gas-phase neutral and ionic species, and the laser spectroscopy for the electronic and vibrational transitions. In addition to the static information such as geometric and electronic structures, the dynamics of the chemical reaction and relaxation is also of great interest and importance. For this purpose, we apply time-resolved study, called pump–probe spectroscopy. In this spectroscopy, the time evolution of the vibrationally or electronic excited molecules prepared by the first laser (hν 1 ), is probed by second laser (hν 2 ) with the detection of ions or electrons, or fragments.

1.7.1 IR-UV Pump–Probe Spectroscopy Figure 1.18 shows the excitation scheme of IR-UV pump–probe spectroscopy to investigate intramolecular vibrational redistribution (IVR) [99, 100]. In the figure, the vibrational level noted by “v” is the IR active vibration (normal mode). This level

1 Microscopic Study on Molecular Recognition …

25

Fig. 1.18 Energy-level diagram and scheme of time-resolved IR-UV pump–probe spectroscopy. V anh is the anharmonic coupling strength

is coupled with isoenergetic vibrational levels (called doorway states) by anharmonic coupling, and these levels are further coupled to the dense vibrational states (bath modes). The picosecond or femtosecond IR laser pulse (hν IR ) coherently excites these levels. After the IR excitation, each state evolves in time and the UV laser probes the time evolution of “v”, “doorway states”, and “dense bath states”. The electronic transitions from these states occur at different UV frequencies, so that we fix the probe laser (hν  UV , hν UV, hν UV ) at appropriate frequency to monitor each state by R2PI as a function of delay time (t). The time profile will exhibit quantum beat or exponential decay, depending on the number of coupled levels and anharmonic coupling strength (V anh ). The analyses of the transition frequencies and time scale of the decay or quantum beat give us the information of the doorway state and anharmonic coupling strength.

1.7.2 UV–UV and UV–VUV Pump–Probe Spectroscopy The chemical reaction and nonradiative decay dynamics of the electronic excite state is investigated by UV–UV and UV–VUV pump–probe spectroscopy. Especially, by using femtosecond pulse laser, we can prepare the wavefunction of v = 0 in S0 to the

26

T. Ebata et al.

Fig. 1.19 Scheme of ultrafast time-resolved pump–probe spectroscopy. Femtosecond laser pulse (hν 1 ) prepares wavepacket (wp) at Sm , and the wp evolves along the potential curve or transfers to other electronic state (Sn ) via conical intersection (CI). Second laser (hν 2 ) ionizes the wp at a delay time t. Depending on the position of the ionization along the potential curve, the kinetic energy (K.E.) of photoelectron is different. This situation is also held if the wp transfer to other electronic state

excited electronic state, called a wavepacket. The prepared wavepacket then starts to evolve in time along the excited state potential curve, or to the conical intersection. Such evolution is monitored by the second femtosecond UV or VUV pulase laser. The second laser ionizes the wavepacket of the excited state potential curve and we detect the ejected photoelectron [101, 102]. The photoelectron has a kinetic energy determined by the energy difference between the excited and ionized state potential curves, as shown in Fig. 1.19. Another probe method is that the probe laser excites the wavepacket to higher electronic excited state and we monitor the emission from the state or fragments generated from the state.

1.8 Conclusions In this chapter (Part I), we described the methods for generating cold molecules, molecular ions and molecular/metal clusters in the gas phase. Since we study many kinds of species in the gas phase, volatile and nonvolatile molecules (solid samples, metal or bio-related molecules), development of vaporization and cooling methods is essential. In addition, various laser spectroscopic techniques were described to

1 Microscopic Study on Molecular Recognition …

27

measure the electronic and vibrational spectra. The density of the generated cold gas-phase species is very low in most of the cases, and we need highly sensitive spectroscopic techniques to measure their spectra, either by detecting photons, ions, or electrons. Double-resonance technique such as UV–UV hole-burning, and IRUV double-resonance spectroscopy enables us to discriminate different isomers and measure the UV and IR spectra of each species. Although this chapter does not fully cover all the experimental methods, we hope the knowledge will be a useful guidance to go into the specific theme from Part II to Part IV.

References 1. Milne T.A., Greene F.T.: Molecular beams in high temperature chemistry. In: Eyring, L. (ed.) Advances in High Temperature Chemistry, pp. 107–150. Elsevier (1969) 2. Sinha, M.P., Schultz, A., Zare, R.N.: Internal state distribution of alkali dimers in supersonic nozzle beams. J. Chem. Phys. 58, 549–556 (1973) 3. Smalley, R.E., Wharton, L., Levy, D.H.: Molecular optical spectroscopy with supersonic beams and jets. Acc. Chem. Res. 10, 139–145 (1977) 4. Behlen, F.M., Mikami, N., Rice, S.A.: Dynamics of radiationless processes studied in pulsed supersonic free jets: some naphthalene lifetimes. Chem. Phys. Lett. 60, 364–367 (1979) 5. Levy, D.H., Wharton, L., Smalley, R.E.: Chemical and Biochemical Applications of Lasers. Academic Press, New York (1977) 6. Chance, K.V., Bowen, K.H., Winn, J.S., Klemperer, W.: Microwave and radio frequency spectrum of XeHCl. J. Chem. Phys. 70, 5157–5160 (1979) 7. Levy, D.H.: Laser spectroscopy of cold gas-phase molecules. Annu. Rev. Phys. Chem. 31, 197–225 (1980) 8. Hayes, J.M., Small, G.J.: Rotationally cooled laser-induced fluorescence/gas chromatography. Anal. Chem. 54, 1202–1204 (1982) 9. Dietz, T.G., Duncan, M.A., Liverman, M.G., Smalley, R.E.: Efficient multiphoton ionization of jet-cooled aniline. Chem. Phys. Lett. 70, 246–250 (1980) 10. Lubman, D.M., Kronick, M.N.: Mass spectrometry of aromatic molecules with resonanceenhanced multiphoton ionization. Anal. Chem. 54, 660–665 (1982) 11. Amirav, A., Even, U., Jortner, J.: Analytical applications of supersonic jet spectroscopy. Anal. Chem. 54, 1666–1673 (1982) 12. Langridge-Smith, P.R.R., Brumbaugh, D.V., Haynam, C.A., Levy, D.H.: Ultraviolet spectra of benzene clusters. J. Phys. Chem. 85, 3742–3746 (1981) 13. Vaida, V., Mcclelland, G.M.: Electronic absorption spectroscopy of cooled supersonic expansions: dynamics of the 1 B1u state of trans-butadiene. Chem. Phys. Lett. 71, 436–439 (1980) 14. Montero, S.: Temperature and entropy in supersonic free jets. Phys. Fluids 25, 056102 (2013) 15. Ebata, T.: Study on the structure and vibrational dynamics of functional molecules and molecular clusters by double resonance vibrational spectroscopy. Bull. Chem. Soc. Jpn. 82, 127–151 (2009) 16. Lubman, D.M.: Lasers and Mass Spectrometry. Oxford University Press, New York (1990) 17. Tembreull, R., Lubman, D.M.: Pulsed laser desorption with resonant two-photon ionization detection in supersonic beam mass spectrometry. Anal. Chem. 58, 1299–1303 (1986) 18. Zenobi, R., Zare, R.N.: Advances in Multiphoton Spectroscopy and Processes. World Scientific, Singapore (1991) 19. Land, D.P., Pettiette-Hall, C.L., Hemminger, J.C., Mciver, R.T.: Identification of molecular adsorbates by LITD/FTMS: a breakthrough for surface chemistry. Acc. Chem. Res. 24, 42–47 (1991)

28

T. Ebata et al.

20. Jones, A.C., Dale, M.J., Keenan, G.A., Langridge-Smith, P.R.R.: Photoionisation and photodissociation of laser-vaporised metallotetraphenylporphyrins. Chem. Phys. Lett. 219, 174–180 (1994) 21. Haefliger, O.P., Zenobi, R.: New sample preparation for quantitative laser desorption mass spectrometry and optical spectroscopy. Rev. Sci. Instrum. 69, 1828–1832 (1998) 22. Chin, W., Piuzzi, F., Dimicoli, I., Mons, M.: Probing the competition between secondary structures and local preferences in gas phase isolated peptide backbones. Phys. Chem. Chem. Phys. 8, 1033–1048 (2006) 23. Abo-Riziq, A., Grace, L., Crews, B., Callahan, M.P., Van Mourik, T., Vries, M.S.D.: Conformational structure of tyrosine, tyrosyl-glycine, and tyrosyl-glycyl-glycine by double resonance spectroscopy. J. Phys. Chem. A 115, 6077–6087 (2011) 24. Sohn, W.Y., Ishiuchi, S.-I., Miyazaki, M., Kang, J., Lee, S., Min, A., Choi, M.Y., Kang, H., Fujii, M.: Conformationally resolved spectra of acetaminophen by UV–UV hole burning and IR dip spectroscopy in the gas phase. Phys. Chem. Chem. Phys. 15, 957–964 (2013) 25. Kenjo, S., Iida, Y., Chaki, N., Kinoshita, S., Inokuchi, Y., Yamazaki, K., Ebata, T.: Laser spectroscopic study on sinapic acid and its hydrated complex in a cold gas phase molecular beam. Chem. Phys. 515, 381–386 (2018) 26. Lipert, R.J., Colson, S.D.: Low frequency vibrations in phenol-(H2 O)2 revealed by holeburning spectroscopy in a supersonic jet. Chem. Phys. Lett. 161, 303–307 (1989) 27. Ebata, T.: Population labeling spectroscopy. In: Field, R.W., Hirota, E., Maier, J.P., Tsuchiya, S. (eds.) Nonlinear Spectroscopy for Molecular Structure Determination. Blackwell Science (1998) 28. Dian, B.C., Longarte, A., Winter, P.R., Zwier, T.S.: The dynamics of conformational isomerization in flexible biomolecules. I. Hole-filling spectroscopy of N-acetyl tryptophan methyl amide and N-acetyl tryptophan amide. J. Chem. Phys. 120, 133–147 (2004) 29. Ebata, T., Furukawa, M., Suzuki, T., Ito, M.: Stimulated-emission ion-dip spectra of phenol–H2 O hydrogen-bonded complex: estimation of intramolecular vibrational redistribution rates of ground-state vibrational levels. J. Opt. Soc. Am. B 7, 1890–1897 (1990) 30. Page, R.H., Shen, Y.R., Lee, Y.T.: Local modes of benzene and benzene dimer, studied by infrared–ultraviolet double resonance in a supersonic beam. J. Chem. Phys. 88, 4621–4636 (1988) 31. Page, R.H., Shen, Y.R., Lee, Y.T.: Infrared–ultraviolet double resonance studies of benzene molecules in a supersonic beam. J. Chem. Phys. 88, 5362–5376 (1988) 32. Tanabe, S., Ebata, T., Fujii, M., Mikami, N.: Oh stretching vibrations of phenol-(H2 O)n (n = 1–3) complexes observed by IR-UV double-resonance spectroscopy. Chem. Phys. Lett. 215, 347–352 (1993) 33. Ebata, T., Fujii, A., Mikami, N.: Vibrational spectroscopy of small-sized hydrogen-bonded clusters and their ions. Int. Rev. Phys. Chem. 17, 331–361 (1998) 34. Pribble, R.N., Zwier, T.S.: Size-specific infrared spectra of benzene-(H2 O)n clusters (n = 1 through 7): Evidence for noncyclic (H2 O)n structures. Science 265, 75–79 (1994) 35. Hartland, G.V., Henson, B.F., Venturo, V.A., Hertz, R.A., Felker, P.M.: Applications of ionization-detected stimulated Raman spectroscopy in molecular-beam studies. J. Opt. Soc. Am. B 7, 1950–1959 (1990) 36. Henson, B.F., Hartland, G.V., Venturo, V.A., Felker, P.M.: Raman-vibronic double-resonance spectroscopy of benzene dimer isotopomers. J. Chem. Phys. 97, 2189–2208 (1992) 37. Ebata, T., Hamakado, M., Moriyama, S., Morioka, Y., Ito, M.: Raman depolarization ratios of jet-cooled benzene and its dimer by stimulated Raman-UV optical double-resonance spectroscopy. Chem. Phys. Lett. 199, 33–41 (1992) 38. Minejima, C., Ebata, T., Mikami, N.: C–H stretching vibrations of benzene and toluene in their S1 states observed by double resonance vibrational spectroscopy in supersonic jets. Phys. Chem. Chem. Phys. 4, 1537–1541 (2002) 39. Yamamoto, R., Ebata, T., Mikami, N.: Stimulated spectroscopic study on intermolecular vibrations of size-selected benzonitrile clusters. Eur. Phys. J. D 20, 403–408 (2002)

1 Microscopic Study on Molecular Recognition …

29

40. Matsuda, Y., Mori, M., Hachiya, M., Fujii, A., Mikami, N.: Infrared spectroscopy of sizeselected neutral clusters combined with vacuum-ultraviolet-photoionization mass spectrometry. Chem. Phys. Lett. 422, 378–381 (2006) 41. Matsuda, Y., Mikami, N., Fujii, A.: Vibrational spectroscopy of size-selected neutral and cationic clusters combined with vacuum-ultraviolet one-photon ionization detection. Phys. Chem. Chem. Phys. 11, 1279–1290 (2009) 42. Hu, Y., Guan, J., Bernstein, E.R.: Mass-selected IR-VUV (118 nm) spectroscopic studies of radicals, aliphatic molecules, and their clusters. Mass Spectrom. Rev. 32, 484–501 (2013) 43. Lee, Y.-F., Kelterer, A.-M., Matisz, G., Kunsági-Máté, S., Chung, C.-Y., Lee, Y.-P.: Infrared absorption of methanol-water clusters (CH3 OH)n (H2 O), n = 1–4, recorded with the VUVionization/IR-depletion technique. J. Chem. Phys. 146, 144308 (2017) 44. Ebata, T., Fujii, A., Mikami, N.: Structures of size-selected hydrogen-bonded phenol-(H2 O)n clusters in S0 , S1 and ion. Int. J. Mass Spectrom. Ion Processes 159, 111–124 (1996) 45. Matsumoto, Y., Ebata, T., Mikami, N.: Photofragment-detected IR spectroscopy (PFDIRS) for the oh stretching vibration of the hydrogen-bonded clusters in the S1 state application to 2-naphthol-B (B=H2 O and CH3 OH) clusters. J. Phys. Chem. A 105, 5727–5730 (2001) 46. Ebata, T., Minejima, C., Mikami, N.: A new electronic state of aniline observed in the transient IR absorption spectrum from S1 in a supersonic jet. J. Phys. Chem. A 106, 11070–11074 (2002) 47. Karas, M., Bachmann, D., Bahr, U., Hillenkamp, F.: Matrix-assisted ultraviolet laser desorption of non-volatile compounds. Int. J. Mass Spectrom. Ion Processes 78, 53–68 (1987) 48. Tanaka, K., Waki, H., Ido, Y., Akita, S., Yoshida, Y., Yoshida, T., Matsuo, T.: Protein and polymer analyses up to m/z 100 000 by laser ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2, 151–153 (1988) 49. Arakawa, R., Okuno, S., Wada, Y.: Recent developments in soft laser desorption/ionization mass spectrometry. J. Mass Spectrom. Soc. Jpn. 52, 33–38 (2004) 50. Fenn, J.B., Mann, M., Meng, C.K., Wong, S.F., Whitehouse, C.M.: Electrospray ionization for mass-spectrometry of large biomolecules. Science 246, 64–71 (1989) 51. Kebarle, P., Verkerk, U.H.: Electrospray: from ions in solution to ions in the gas phase, what we know now. Mass Spectrom. Rev. 28, 898–917 (2009) 52. Nemes, P., Goyal, S., Vertes, A.: Conformational and noncovalent complexation changes in proteins during electrospray ionization. Anal. Chem. 80, 387–395 (2007) 53. Fenn, J.B., Mann, M., Meng, C.K., Wong, S.F., Whitehouse, C.M.: Electrospray ionization–principles and practice. Mass Spectrom. Rev. 9, 37–70 (1990) 54. Yamashita M., Fenn J. B.: Electrospray ion source. Another variation on the free-jet theme. J. Phys. Chem. 88, 4451–4459 (1984) 55. Zhou, S.L., Edwards, A.G., Cook, K.D., Van Berkel, G.J.: Investigation of the electrospray plume by laser-induced fluorescence spectroscopy. Anal. Chem. 71, 769–776 (1999) 56. Chillier, X.F.D., Monnier, A., Bill, H., Gülaçar, F.O., Buchs, A., Mcluckey, S.A., Van Berkel, G.J.: A mass spectrometry and optical spectroscopy investigation of gas-phase ion formation in electrospray. Rapid Commun. Mass Spectrom. 10, 299–304 (1996) 57. Wang, L.-S., Ding, C.-F., Wang, X.-B., Barlow, S.E.: Photodetachment photoelectron spectroscopy of multiply charged anions using electrospray ionization. Rev. Sci. Instrum. 70, 1957–1966 (1999) 58. Wang, X.-B., Ding, C.-F., Wang, L.-S.: Photodetachment spectroscopy of a doubly charged anion: direct observation of the repulsive coulomb barrier. Phys. Rev. Lett. 81, 3351–3354 (1998) 59. Wang, L.S.: Perspective: electrospray photoelectron spectroscopy: from multiply-charged anions to ultracold anions. J. Chem. Phys. 143, 14 (2015) 60. Gerlich, D.: Experimental investigations of ion-molecule reactions relevant to interstellar chemistry. J. Chem. Soc. Faraday Trans. 89, 2199–2208 (1993) 61. Schlemmer, S., Kuhn, T., Lescop, E., Gerlich, D.: Laser excited N2 + in a 22-pole ion trap: experimental studies of rotational relaxation processes. Int. J. Mass Spectrom. 185–187, 589–602 (1999)

30

T. Ebata et al.

62. Wang, X.-B., Wang, L.-S.: Development of a low-temperature photoelectron spectroscopy instrument using an electrospray ion source and a cryogenically controlled ion trap. Rev. Sci. Instrum. 79, 073108 (2008) 63. Boyarkin, O.V., Mercier, S.R., Kamariotis, A., Rizzo, T.R.: Electronic spectroscopy of cold, protonated tryptophan and tyrosine. J. Am. Chem. Soc. 128, 2816–2817 (2006) 64. Rizzo, T.R., Stearns, J.A., Boyarkin, O.V.: Spectroscopic studies of cold, gas-phase biomolecular ions. Int. Rev. Phys. Chem. 28, 481–515 (2009) 65. Svendsen, A., Lorenz, U.J., Boyarkin, O.V., Rizzo, T.R.: A new tandem mass spectrometer for photofragment spectroscopy of cold, gas-phase molecular ions. Rev. Sci. Instrum. 81, 073107 (2010) 66. Voronina, L., Scutelnic, V., Masellis, C., Rizzo, T.R.: Can mutational analysis be used to assist structure determination of peptides? J. Am. Chem. Soc. 140, 2401–2404 (2018) 67. Pereverzev, A.Y., Kopysov, V.N., Boyarkin, O.V.: Peptide bond ultraviolet absorption enables vibrational cold-ion spectroscopy of nonaromatic peptides. J. Phys. Chem. Lett. 9, 5262–5266 (2018) 68. Harrilal, C.P., Deblase, A.F., Fischer, J.L., Lawler, J.T., Mcluckey, S.A., Zwier, T.S.: Infrared population transfer spectroscopy of cryo-cooled ions: quantitative tests of the effects of collisional cooling on the room temperature conformer populations. J. Phys. Chem. A 122, 2096–2107 (2018) 69. Noble, J.A., Broquier, M., Gregoire, G., Soorkia, S., Pino, G., Marceca, E., DedonderLardeux, C., Jouvet, C.: Tautomerism and electronic spectroscopy of protonated 1- and 2aminonaphthalene. Phys. Chem. Chem, Phys (2018) 70. Nieuwjaer, N., Desfrancois, C., Lecomte, F., Manil, B., Soorkia, S., Broquier, M., Gregoire, G.: Photodissociation spectroscopy of cold protonated synephrine: surprising differences between IR-UV hole-burning and IR photodissociation spectroscopy of the O–H and N–H modes. J. Phys. Chem. A 122, 3798–3804 (2018) 71. Ishiuchi, S., Sasaki, Y., Lisy, J.M., Fujii, M.: Ion-peptide interactions between alkali metal ions and a termini-protected dipeptide: modeling a portion of the selectivity filter in K+ channels. Phys. Chem. Chem. Phys. 21, 561–571 (2019) 72. Sekiguchi, T., Tamura, M., Oba, H., Carcarbal, P., Lozada-Garcia, R.R., Zehnacker-Rentien, A., Gregoire, G., Ishiuchi, S., Fujii, M.: Molecular recognition by a short partial peptide of the adrenergic receptor: a bottom-up approach. Angew. Chem. Int. Ed. 57, 5626–5629 (2018) 73. Nieto, P., Muller, D., Sheldrick, A., Gunther, A., Miyazaki, M., Dopfer, O.: Effect of alkali ions on optical properties of flavins: vibronic spectra of cryogenic M+ lumichrome ions (M = Li–Cs) in the gas phase. Phys. Chem. Chem. Phys. 20, 22148–22158 (2018) 74. Baek, J.Y., Choi, C.M., Eun, H.J., Park, K.S., Choi, M.C., Heo, J., Kim, N.J.: Ultraviolet photodissociation spectroscopy of cold, isolated adenine complexes with a potassium cation. Chem. Phys. Lett. 635, 163–167 (2015) 75. Catani, K.J., Muller, G., Jusko, P., Theulé, P., Bieske, E.J., Jouvet, C.: Electronic spectrum of the protonated diacetylene cation (H2 C4 H+ ). J. Chem. Phys. 147, 084302 (2017) 76. Yang, N., Duong, C.H., Kelleher, P.J., Johnson, M.A.: Unmasking rare, large-amplitude motions in D-2-tagged I- •(H2 O)2 isotopomers with two-color, infrared-infrared vibrational predissociation spectroscopy. J. Phys. Chem. Lett. 9, 3744–3750 (2018) 77. Bell, M.R., Cruzeiro, V.W.D., Cismesia, A.P., Tesler, L.F., Roitberg, A.E., Polfer, N.C.: Probing the structures of solvent-complexed ions formed in electrospray ionization using cryogenic infrared photodissociation spectroscopy. J. Phys. Chem. A 122, 7427–7436 (2018) 78. Salpina, J.Y., Haldys, V., Steinmetz, V., Leon, E., Yanez, M., Montero-Campillo, M.M.: Protonation of methyluracils in the gas phase: the particular case of 3-methyluracil. Int. J. Mass Spectrom. 429, 47–55 (2018) 79. Marsh, B.M., Duffy, E.M., Soukup, M.T., Zhou, J., Garand, E.: Intramolecular hydrogen bonding motifs in deprotonated glycine peptides by cryogenic ion infrared spectroscopy. J. Phys. Chem. A 118, 3906–3912 (2014) 80. Inokuchi, Y., Soga, K., Hirai, K., Kida, M., Morishima, F., Ebata, T.: Ultraviolet photodissociation spectroscopy of the cold K+ •Calix[4]arene complex in the gas phase. J. Phys. Chem. A 119, 8512–8518 (2015)

1 Microscopic Study on Molecular Recognition …

31

81. Powers, D.E., Hansen, S.G., Geusic, M.E., Pulu, A.C., Hopkins, J.B., Dietz, T.G., Duncan, M.A., Langridge-Smith, P.R.R., Smalley, R.E.: Supersonic metal cluster beams: laser photoionization studies of copper cluster (Cu2). J. Phys. Chem. 86, 2556–2560 (1982) 82. Bondybey, V.E., English, J.H.: Laser induced fluorescence of metal clusters produced by laser vaporization: gas phase spectrum of Pb2 . J. Chem. Phys. 74, 6978–6979 (1981) 83. Even, U., Jortner, J., Noy, D., Lavie, N., Cossart-Magos, C.: Cooling of large molecules below 1 K and He clusters formation. J. Chem. Phys. 112, 8068–8071 (2000) 84. Haberland, H., Karrais, M., Mall, M.: A new type of cluster and cluster ion source. Z. Phys. D At. Mol. Cluster 20, 413–415 (1991) 85. Tsunoyama, H., Zhang, C.-H., Akatsuka, H., Sekiya, H., Nagase, T., Nakajima, A.: Development of high-flux ion source for size-selected nanocluster ions based on high-power impulse magnetron sputtering. Chem. Lett. 42, 857–859 (2013) 86. Zhang, C.-H., Tsunoyama, H., Akatsuka, H., Sekiya, H., Nagase, T., Nakajima, A.: Advanced nanocluster ion source based on high-power impulse magnetron sputtering and time-resolved measurements of nanocluster formation. J. Phys. Chem. A 117, 10211–10217 (2013) 87. Inokuchi, Y., Haino, T., Sekiya, R., Morishima, F., Dedonder, C., Feraud, G., Jouvet, C., Ebata, T.: UV photodissociation spectroscopy of cryogenically cooled gas phase host-guest complex ions of crown ethers. Phys. Chem. Chem. Phys. 17, 25925–25934 (2015) 88. Féraud, G., Esteves-López, N., Dedonder-Lardeux, C., Jouvet, C.: UV spectroscopy of cold ions as a probe of the protonation site. Phys. Chem. Chem. Phys. 17, 25755–25760 (2015) 89. Kamrath, M.Z., Relph, R.A., Guasco, T.L., Leavitt, C.M., Johnson, M.A.: Vibrational predissociation spectroscopy of the H-2-tagged mono- and dicarboxylate anions of dodecanedioic acid. Int. J. Mass Spectrom. 300, 91–98 (2011) 90. Kang, H., Féraud, G., Dedonder-Lardeux, C., Jouvet, C.: New method for double-resonance spectroscopy in a cold quadrupole ion trap and its application to UV–UV hole-burning spectroscopy of protonated adenine dimer. J. Phys. Chem. Lett. 5, 2760–2764 (2014) 91. Feraud, G., Dedonder, C., Jouvet, C., Inokuchi, Y., Haino, T., Sekiya, R., Ebata, T.: Development of ultraviolet-ultraviolet hole-burning spectroscopy for cold gas-phase ions. J. Phys. Chem. Lett. 5, 1236–1240 (2014) 92. Stearns, J.A., Mercier, S., Seaiby, C., Guidi, M., Boyarkin, O.V., Rizzo, T.R.: Conformationspecific spectroscopy and photodissociation of cold, protonated tyrosine and phenylalanine. J. Am. Chem. Soc. 129, 11814–11820 (2007) 93. Inokuchi, Y., Boyarkin, O.V., Kusaka, R., Haino, T., Ebata, T., Rizzo, T.R.: UV and IR spectroscopic studies of cold alkali metal ion-crown ether complexes in the gas phase. J. Am. Chem. Soc. 133, 12256–12263 (2011) 94. Ervin K. M., Lineberger W. C.: In: Babcock LM (ed) Advances in gas phase ion chemistry, pp. 121–166. JAI, Greenwich (1992) 95. Osterwalder, A., Nee, M.J., Zhou, J., Neumark, D.M.: High resolution photodetachment spectroscopy of negative ions via slow photoelectron imaging. J. Chem. Phys. 121, 6317–6322 (2004) 96. Kruit, P., Read, F.H.: Magnetic field paralleliser for 2π electron-spectrometer and electronimage magnifier. J. Phys. E: Sci. Instrum. 16, 313–324 (1983) 97. Bordas, C., Paulig, F., Helm, H., Huestis, D.L.: Photoelectron imaging spectrometry: principle and inversion method. Rev. Sci. Instrum. 67, 2257–2268 (1996) 98. Eppink, A.T.J.B., Parker, D.H.: Velocity map imaging of ions and electrons using electrostatic lenses: application in photoelectron and photofragment ion imaging of molecular oxygen. Rev. Sci. Instrum. 68, 3477–3484 (1997) 99. Ebata, T., Kayano, M., Sato, S., Mikami, N.: Picosecond IR-UV pump-probe spectroscopy. IVR of oh stretching vibration of phenol and phenol dimer. J. Phys. Chem. A 105, 8623–8628 (2001) 100. Yamada, Y., Mikami, N., Ebata, T.: Real-time detection of doorway states in the intramolecular vibrational energy redistribution of the OH/OD stretch vibration of phenol. J. Chem. Phys. 121, 11530–11534 (2004)

32

T. Ebata et al.

101. Suzuki, T.: Femtosecond time-resolved photoelectron imaging. Annu. Rev. Phys. Chem. 57, 555–592 (2006) 102. Stavros, V.G., Verlet, J.R.R.: Gas-phase femtosecond particle spectroscopy: a bottom-up approach to nucleotide dynamics. Annu. Rev. Phys. Chem. 67, 211–232 (2016)

Part II

Spectroscopy and Characterization of Gas-Phase Functional Molecules and Clusters

Chapter 2

Microscopic Study on Molecular Recognition of Host–Guest Complexes Between Crown Ethers and Aromatic Molecules Takayuki Ebata

Laser spectroscopy and theoretical calculation revealed that 18-Crown-6 and benzenediols form host–guest complexes by “induced fitting manner”

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_2

35

36

T. Ebata

Abstract Crown ethers (CEs) are the typical molecules in the host–guest chemistry. In this chapter, laser spectroscopy and quantum chemical calculations of the cold gas-phase neutral host–guest complexes of crown ethers (CEs) are described. Here, we chose 3n-crown-n (3nCn, n = 4, 5, 6, 8) as the host and substituted benzenes as the guest. The cold gas-phase complexes are produced by a supersonic expansion technique. Electronic spectra of the complexes are observed by laser-induced fluorescence (LIF) spectroscopy and discrimination of different isomers is performed by ultraviolet–ultraviolet hole-burning (UV–UV HB) spectroscopy. Conformer specific vibrational spectra are measured by IR-UV double-resonance (IR-UV DR) spectroscopy. The structures of the complexes are determined by comparing the obtained IR spectra with those of the energy-optimized structures obtained by quantum chemical calculations. We will discuss how the host and guest species change their flexible structures to form a best-fitted stable complex via “key and lock” or “induced-fitting” mechanism, and what kinds of interactions are operated for the stabilization of the complexes. We also will show the effect of complexation on the photophysics of the guest species. Keywords Crown ethers · Host–guest complex · IR-UV double-resonance spectroscopy · Key and lock · Induced-fitting · Cage effect

2.1 Introduction Crown ethers (CEs) are the well-known macrocyclic molecules consisting of several oxyethylene (–CH2 –CH2 –O–) units [1, 2]. They have a cavity with the size determined by the number of oxyethylene unit and can include cation and neutral species which are fitted to the size of the cavity. In addition, because of the flexible nature, CEs can modify their structures to include the guest species with different shapes. CEs have been playing an important role in host–guest and supramolecular chemistry, and applications of CEs as molecular receptors, metal cation extraction agents, fluoroionophores, and phase transfer catalytic media have been described in a number of studies [3–13]. The structures of the complexes in the condensed phase are studied by UV, IR, NMR, and X-ray diffraction methods. In the gas phase, the complexes have been mostly studied for the ionic species, such as metal ion (M+ )·CE complexes, so that they are investigated with mass spectrometry [14–23], and ion mobility methods [24–27]. In this chapter, the UV and IR spectroscopic study on the host–guest complexes of CEs with neutral molecules will be presented. A difficulty of the UV and IR spectroscopic study of CEs is that their flexibility causes a broadening of the spectra. That is, the binding energy between the host and guest is T. Ebata (B) Department of Chemistry, Graduate School of Science, Hiroshima University, Higashi-hiroshima 739-8526, Japan e-mail: [email protected] Department of Applied Chemistry, National Chiao Tung University, Hsinchu 30010, Taiwan

2 Microscopic Study on Molecular Recognition of Host–Guest …

37

comparable to the thermal energy at room temperature. In addition, CEs can form various conformations and the barrier height between the different conformers is also similar to the room temperature thermal energy. Thus, when we measure the UV and IR spectra of the complexes at room temperature, the observed spectra become broad or complicated arising from the average of overall possible conformers and fluctuation at a given temperature. It is very difficult to obtain the useful information from the broadened spectra. However, by using a supersonic jet technique, we can generate cold gas-phase complexes of which the population is concentrated at v = 0. This leads to the spectra very simple. Thus, a combination of laser spectroscopic method with the supersonic expansion technique enables us to measure the sharp and well-resolved spectra of the complexes [28–39]. In this study, we measure the electronic and vibrational spectra of the host–guest complexes of CEs. The essential part of this study is that we need a chromophore in either the guest or host species for the detection of the complexes and discrimination of different isomers. The transition energy of the chromophore sensitively changes under different environment, such as different conformations and by the complex formation. We investigate the shift of the electronic transition frequency caused by the conformation change or complexation. There are two ways to put chromophores to the complex, either on the host CEs site or guest site. In the former case, we choose CE having benzene(s) chromophores, such as benzo-3n-crown-n (B3nCn) and dibenzo-3n-crown-n (DB3nCn). The study of spectroscopy and the determination of the structures of the complexes of B3nCn and DB3nCn have been already reviewed [28–40]. In this chapter, we focus on the latter system, that is, we choose benzene derivatives as a guest, namely, phenol [34], benzenediols [36, 37], tyramine and tyrosol, and 3nCn as a host. The electronic transitions of benzene derivatives appear in the UV region. We measure laser-induced fluorescence (LIF) spectra of the gas-phase complexes under jet-cooled condition. A discrimination of different conformers and isomers is carried out by UV–UV hole-burning spectroscopy. Then, the isomer-specific IR spectrum is measured by IR-UV double-resonance spectroscopy. The obtained electronic and IR spectra are analyzed by comparing those of the possible complexes obtained by high-level quantum chemical calculations. We will see how CEs and substituted benzenes change their structures to form best-fitted complexes via “key and lock” model or “induced-fit” model, and examine what kinds of interactions are operated between host and guest. In addition, effects of the complex formation to the electronic structure and photophysics of the guest species are discussed for some of the complexes.

2.2 Experimental and Computational The detail of the experimental methods is described in part I. The neutral complexes are generated by supersonic free jet. 3nCn and substituted benzenes are solids at room temperature. They are put at different sample housings and independently heated to vaporize. The supersonic free jet of the complexes is obtained by expanding the

38

T. Ebata

gaseous mixture of 3nCn/benzene derivatives diluted with 3–4 bar of He carrier gas into vacuum through a pulsed nozzle having 1 mm orifice. The S1 –S0 electronic spectra are measured by LIF spectroscopy. UV–UV hole-burning (HB) spectroscopy was carried out to discriminate vibronic bands belonging to different isomers. The IR spectrum of each species is measured by IR-UV double-resonance spectroscopy with fluorescence detection. The S1 lifetimes of the bare molecules and complexes are obtained by measuring the fluorescence decay curve or by picosecond pump–probe experiment. In the latter experiment, the two tunable picosecond UV laser pulses are used. They are generated by two optical parametric generation/optical parametric amplifier (OPG/OPA) systems (Ekspla PG401 SH) pumped by a mode-locked picosecond Nd3+ :YAG laser (Ekspla PL2143S). The two lasers (pump and probe) are introduced to a molecular beam chamber and crossed the molecular beam at right angles in a counter-propagated manner with each other. The first UV laser pumps the complex to the specific vibronic level of S1 and second (probe) laser ionizes the excited species, and the ion intensity is measured as a function of delay time. The ions are mass analyzed with a 50 cm time-of-flight tube and are detected by a channeltron. The decay-time constants are obtained by a deconvolution method. The spectral and the time resolution of the two UV pulses are 5 cm−1 and 12 ps, respectively. In addition to gas phase, we carry out absorption and fluorescence spectroscopic study for the 18C6·catechol complex in solution at room temperature. The UV absorption and fluorescence spectra of pure catechol and catechol/18C6 mixture are measured in cyclohexane solution with a commercial spectrophotometer. In addition, the fluorescence lifetime of catechol and 18C6/catechol mixture is also measured under the same condition. The complex structure is determined with an aid of theoretical calculation. Possible structures are calculated and their IR spectra are compared with the IR spectra measured by IR-UV DR spectroscopy. The theoretical calculation consists of molecular mechanic force field calculations and density functional theory (DFT) calculations. The initial structures are obtained by Monte Carlo simulation by mixed torsional search with low-mode sampling [41] in MacroModel V.9.1 [42] with the MMFF94s force field [43]. The geometries are optimized by the PRCG algorithm with a convergence threshold of 0.05 kJ/mol. For example, in case of 3nCn·phenol complex, 300–1000 isomers are obtained within 20 kJ/mol energy, and 165 isomers for 18C6·HQ, 46 isomers for 18C6·RE, and 193 isomers for 18C6·CA are obtained within 20 kJ mol/mol. All the isomers are then optimized by DFT calculation at the M05-2X/6-31+G* level with loose optimization criteria using Gaussian 09 program package [44].

2.3 Phenol·3nCn (n = 4–8) Complexes We first describe the simplest host–guest complexes of 3nCn·benzene derivatives (n = 4, 5, 6, and 8), by choosing phenol as the guest species. We discuss the size dependence on the complex formation between 3nCn (n = 4, 5, 6, and 8) (Scheme 1)

2 Microscopic Study on Molecular Recognition of Host–Guest …

39

and phenol, [3nCn·phenol] (Fig. 2.1). By changing n = 4–8, we investigate which size of 3nCn can form a uniquely stable complex with phenol, and examine how the host and guest species cooperatively change their structures via OH···O hydrogen (H)-bond, CH···π, and O···HC (aromatic) interactions. Figure 2.2 shows the LIF spectra (black curve) of (a) phenol·diethylether (DEE), (b) phenol·1,4-dioxane, and (c)–(f) 3nCn·phenol complexes in the band origin region of the S1 –S0 transition [34]. The 0,0 band of phenol is located at 36,349 cm−1 [45], and those of all the complexes are red-shifted. The redshift indicates that phenol is acting as an H-bond donor in these complexes. The UV–UV HB spectra (blue curves) obtained by probing major LIF bands tell us that a number of isomers for the phenol·ether (ether = DEE, DO, 12C4, 15C5, 18C6, and 24C8) complexes are 1, 1, 3, 2, 1, and 2, respectively. Thus, the number Fig. 2.1 Forming complex between phenol and 3nCn (n = 4–8)

Fig. 2.2 LIF (black) and UV–UV HB (blue and green) spectra of phenol·ethers complexes. Green spectra for 24C8 were obtained by fixing probe UV frequency to positions near bands A and B. Reprinted with permission from [34] J. Phys. Chem. Letters 3, 1414–1420 (2012). Copyright 2012 American Chemical Society

40

T. Ebata

of the complexes does not increase monotonically with the size of the ethers, and the appearance of only one isomer for the 18C6·phenol complex indicates the formation of a unique stable structure in this complex. Figure 2.3 shows IR-UV DR spectra of (a) phenol, (b) (phenol)2 , (c) phenol·H2 O, and (d)–(m) 3nCn·phenol complexes [34]. In the spectra, the vibrational bands in the 3300–3700 and 3020–3100 cm−1 regions are due to phenolic OH and CH stretching vibrations, respectively. The OH stretching bands of the phenol·ether complexes are redshifted by 100–300 cm−1 from that of bare phenol at 3657 cm−1 [45]. This means that the phenolic OH is H-bonded as a proton donor in the complexes as expected from the redshift of the electronic spectra. The low-frequency intermolecular vibrations are accompanied with the OH stretch in many complexes (Fig. 2.3d–m). The CH stretch bands of phenol site in the region 3020–3100 cm−1 are classified into four groups (marked by solid lines). We see that although the bands at ~3050 cm−1 are commonly observed for all the complexes, the intensities of the other bands of phenol·3nCn complexes are much weaker than those of phenol and the simple complexes in Fig. 2.3a–e. In addition, a new band emerges at ~3070 cm−1 (marked by arrows) in the phenol·3nCn complexes. The results indicate that the phenolic CH groups are also deeply involved in the complex formation with 3nCn. To find the reason for the unique conformation of 18C6·phenol, stable structures of 3nCn·phenol complexes and their energies are calculated. Figure 2.4 compares the optimized structures of three lowest isomers of 3nCn·phenol (n = 4–6) complexes obtained by DFT calculation at ωB97X-D/6-31++G** level. Blue dotted Fig. 2.3 IR-UV DR spectra of a phenol, b (phenol)2 , c phenol-H2 O, and d–m phenol–ether complexes. The gray spectra for 24C8 were obtained by fixing probe UV frequency near bands A and B. The red stick bars are the calculated IR spectra. Blue bars of phenol-diethylether (DEE) are the calculated low-frequency modes. Reprinted with permission from [34] J. Phys. Chem. Letters 3, 1414–1420 (2012). Copyright 2012 American Chemical Society

2 Microscopic Study on Molecular Recognition of Host–Guest …

41

Fig. 2.4 Three lowest energy isomers of a 12C4·phenol, b 15C5·phenol, c 18C6·phenol, and d 24C8·phenol complexes optimized at ωB97X-D/6-31++G** level. Lower panel of (c) shows the conformations of 18C6 part of 18C6·phenol; I, II, and III. Blue dotted lines are the O··HO (rO··H < 2.7 Å) and CH··π (rH··C < 3.0 Å) interactions. Reprinted with permission from [34] J. Phys. Chem. Letters 3, 1414–1420 (2012). Copyright 2012 American Chemical Society

lines represent the O··HO (r O ···H < 2.7 Å) and CH··π (r H ···C < 3.0 Å) interactions. In all complexes, phenolic OH is H-bonded to the ether O atom(s). In addition, CH groups of CEs are bound to phenyl ring by the CH··π interaction. Especially, in the 18C6·phenol and 24C8·phenol complexes (Fig. 2.4c and d), the crown CHs are bound to phenyl plane on both sides. We define the interaction energy E int in the complex by the following equation: E int (CE-phenol) = E CE-phenol − E CE − E phenol .

(2.1)

Here, E CE and E phenol are the energies of CE and phenol, respectively, with their geometries fixed to the same ones in the phenol·CE complexes, where zero-pointenergy (ZPE) correction is not performed. Table 2.1 lists the relative total energy of the complexes (E), and E int as well as for each of the complexes. As seen in the table, the energy difference between the lowest and second lowest conformers of 18C6·phenol is largest among the examined complexes, supporting unique nature of this complex. On the other hand, in the other complexes, the energy difference between the most, second as well as third stable isomers is rather small, suggesting the presence of multiple isomers even in jet-cooled condition. Another point is that the interaction energy between the host 3nCn and phenol does not exactly correlate

42

T. Ebata

Table 2.1 Relative total (E) energies and intermolecular interaction (E int ) energies of the most stable three isomers of 3nCn·phenol complexes optimized at ωB97X-D/6-31++G** level. E CE is the relative energy of the conformation of the crown part in each complex. All the values are in unit of kJ/mol 12C4 (n = 4) E

E int

15C5 (n = 5) E CE

E

E int

18C6 (n = 6) E CE

E

E int

24C8 (n = 8) E CE

E

E int

E CE

I

0

71.0

2.22

0

82.2

6.06

0

99.9

3.17

0

121

0

II

0.51

71.9

3.62

0.19

85.8

11.5

7.68

95.3

4.20

14.8

117

0.86

III

1.1

70.1

1.48

0.88

73.3

0

8.18

92.2

2.14

50.1

116

7.51

to the total energy as seen in 3nCn·phenol as seen in 12C4 complex. This is due to the fact that the conformation stability also contributes to the total energy. We examine the interaction between phenol and 18C6(I) conformer in 18C6·phenol complex more detail. Figure 2.4 picks up the structures of the 18C6 part in the 18C6 (I, II and III)·phenol complexes. In 18C6(I), four oxygen atoms, O(1), O(4), O(10), and O(13), are directed inside of the cavity, so that the phenol and 18C6 interact by four oxygen atoms through the bifurcated [O(1)··HO and O(4)··HO] H-bonding, CH··π, O(10)··HC(aromatic), and O(13)··HC(aromatic) interactions. On the other hand, in 18C6(II), O(4) is directed outside of the cavity and H(3) prevents phenol from forming the bifurcated O··HO H-bond. This situation is also the same in 18C6(III). Thus, only the pair of phenol and 18C6(I) is best matched with each other, resulting in largest E int of the 18C6(I)·phenol. Since the most stable conformation of bare 18C6 is quite different from that in the 18C6(I)·phenol complex, we can say that this complex is formed by “induced-fitting” mechanism to lead “key and lock” relationship. Such “key and lock” relationship is found in other systems and we will discuss it later.

2.4 18C6·Benzenediols Complexes 2.4.1 Complex Structures and Induced Fitting We change the guest from phenol to benzenediol and produce complexes with 18C6. Benzenediol has three structural isomers depending on the substitution positions (para, meta and ortho) of two OH groups, called hydroquinone (HQ), resorcinol (RE), and catechol (CA), as shown in Fig. 2.5. In addition, each species has several conformers coming from the relative orientation of two OH groups. Thus, the interaction between benzenediol and 18C6 will change for different conformers of benzenediol, which is the interest of this study. Figure 2.6 shows the LIF spectra of jet-cooled (a) CA, (d) RE, and (g) CA and (b–f) the 1:1 complexes with 18C6 in the band origin region of S1 –S0 transition [36, 37]. Similar to 3nCn·phenol, in all complexes, the 0,0 band is redshifted from the

2 Microscopic Study on Molecular Recognition of Host–Guest …

43

Fig. 2.5 Three structural isomers of benzenediols and 18C6. Reproduced from [37] with permission from the Royal Society of Chemistry Fig. 2.6 LIF spectrum of jet-cooled a HQ, d RE, and g CA monomer. LIF spectra of b 18C6·HQ, e 18C6·RE, and h 18C6·CA complexes. UV–UV HB spectrum of c 18C6·HQ by monitoring band AHQ , f 18C6·RE by monitoring band ARE , and i 18C6·CA by monitoring bands ACA , BCA , and CCA , respectively. Taken from Ref. [36, 37]. Reproduced from [37] with permission from the Royal Society of Chemistry

44

T. Ebata

monomer. The LIF spectra of the jet-cooled monomers are already reported, and there exist two conformers in HQ [46] and RE [47], and single conformer in CA [48, 49]. Although RE has three conformers, energy of RE(III) is 8.86 kJ/mol higher in energy than RE(I) at ωB97X-D/6-31++G** level of calculation, so that the electronic transition of RE(III) appears only under warmer jet condition. A measurement of the UV–UV HB spectra gives the number of the isomers of the complexes coexisting in the supersonic free jet. Thus, we identify single isomer for 18C6·HQ and 18C6·RE, and three isomers for 18C6·CA under jet-cooled condition. IR spectra measured by IR-UV DR spectroscopy are shown in Fig. 2.7. In the spectra of HQ and RE, free OH stretching bands appear at 3658–3662 cm−1 . An appearance of only one band in spite of two OH groups means that the two OH stretching bands have the same frequency. On the other hand, the IR spectrum of CA shows two OH bands at 3611 and 3673 cm−1 . In CA, its two OH groups are H-bonded with each other as shown in Fig. 2.4. This intramolecular H-bonding weakens the bond strength of the donor OH while strengthens the acceptor OH. Fig. 2.7 a–c IR-UV DR spectra of trans-HQ, cis-HQ, and 18C6·HQ. d, e IR-UV DR spectra of RE(I) and RE(II). f IR-UV DR spectrum of 18C6·RE. g–i IR-UV DR spectra of CA·18C6. j Calculated IR spectra of trans-HQ, cis-HQ, and 18C6·HQ(I). k Calculated IR spectra of RE(I), RE(II), and 18C6·RE(I) and 18C6·RE(III). Lower pannel of (g–i) Calculated IR spectra of 18C6·CA(A1), 18C6·CA(E1), and 18C6·CA(E4). Arrow at the bottom are the position of OH stretch bands of CA. The calculated IR frequencies are scaled by 0.9346 for CA and its complex, 0.9325 for RE and its complex, and 0.9325 for CA and its complex, in order to reproduce the observed OH stretching vibration of each monomer. Reproduced from [37] with permission from the Royal Society of Chemistry

2 Microscopic Study on Molecular Recognition of Host–Guest …

45

The complexation with 18C6 drastically changes the IR spectra of benzenediols. The IR spectrum of 18C6·HQ (Fig. 2.7c) by monitoring band AHQ shows two bands 3444 and 3657 cm−1 . The former band is accompanied with low-frequency vibrational bands. It is clear that the former band is the H-bonded OH and the latter is free OH. The IR spectrum of 18C6·RE (Fig. 2.7f) measured by monitoring band ARE shows three bands in the 3450–3500 cm−1 region. The positions of the two bands indicate that two OH groups of RE are H-bonded. The appearance of three bands may be due to that one of them is the low-frequency vibration accompanied with the 3645 cm−1 band. The IR spectrum of 18C6·CA shows different IR spectra for different probed bands, CCA , ACA , and BCA . In band CCA monitored IR spectrum (Fig. 2.7g), two bands at 3281 and 3507 cm−1 emerge. The former band is broadened due to strong H-bonding. In band ACA monitored one (Fig. 2.7h), two bands at 3385 and 3407 cm−1 emerge, and in BCA (Fig. 2.7i) single band emerges at 3424 cm−1 . Most probable structures of the complexes and their IR spectra are calculated at ωB97X-D/6-31++G** level. Figure 2.8 shows (a) the most stable and (b) second stable structures of the 18C6·HQ complex within the energy of 10 kJ mol−1 . They have a similar structure, where the complexes are stabilized by bifurcated OH··O H-bonding and CH··π interactions. The calculated IR spectrum of Fig. 2.7j well reproduces the observed IR spectrum. Very interestingly, HQ prefers the cis-conformation in the complex in spite of that trans-form is more stable than cis-form by 0.23 kl/mol. The structure shown in Fig. 2.8c shows the lowest energy isomer in which HQ has a trans-form. This isomer is higher than 18C6·cis-HQ-I isomer by 23.2 kJ/mol. The reason for such high energy of the complex is partly due to that cis-HQ has a dipole moment (DM), while trans-HQ does not. The calculated DM of bare cisHQ is 2.795 D, while that of trans-HQ is zero. On the other hand, 18C6 part in

Fig. 2.8 (left) a Most stable and b second stable isomers of 18C6·HQ complex. c 18C6·trans-HQ isomer. It should be noted that most of the stable isomers have cis-HQ conformer. (right) Cooperative effect in the complex formation between cis-HQ and 18C6 (see text)

46

T. Ebata

the 18C6·cis-HQ-I isomer has 0.532 D. Thus, 18C6·cis-HQ-I gains an additional stabilization due to dipole–dipole interaction. Actually, calculated DM of 18C6·cisHQ-I is as large as 4.662 D, Fig. 2.8 (right), which is much larger than that of the sum of the individual DMs. Thus, the complexation leads to cooperative effect. As will be discussed later, both HQ and 18C6 change their conformations to form best-fitted structure of the complex. The calculated IR spectrum of 18C6·cis-HQ-I (Fig. 2.7j) well reproduces the observed IR spectrum of band AHQ (Fig. 2.7c). Thus, from the viewpoint of energetics and agreement of the IR spectrum, the structure of species AHQ is determined to 18C6·cis-HQ-I. Figure 2.9 shows four lowest energy isomers of the 18C6·RE complex. 18C6·RE has many isomers with similar energies, and RE takes conformer III (Fig. 2.4) in all the complexes. In bare form, conformer III is 8.86 kJ/mol higher than conformer I. The reason that RE takes this conformation is that RE can bind to 18C6 via two H-bonds. Thus, despite the energetically unfavorable conformation, RE prefers conformer III to obtain more stabilization energy. The calculated IR spectrum of 18C6·RE(III)-I (Fig. 2.7f) well reproduces the observed one (Fig. 2.7k) for the redshift and a split of the H-bonded OH stretch bands. Thus, RE modifies its most stable conformation (I) to unstable one (III) to form best-fitted structure 18C6·RE(III)-I. The stabilization energy will be discussed later. Figure 2.10 shows the most probable structures of the 18C6·CA complexes which reproduce the observed IR spectra in Fig. 2.7g–i. In 18C6·CA-A1, CA keeps the intramolecular H-bond, but the acceptor OH is H-bonded to the oxygen atom of 18C6. This leads to large redshift of this OH stretch band as seen in the IR spectrum Fig. 2.9 a–d Structures of the lowest energy isomers of 18C6·RE(III) within the energy of 5 kJ/mol at the level of ωB97X-D/6-31++G** calculation. Reproduced from [37] with permission from the Royal Society of Chemistry

2 Microscopic Study on Molecular Recognition of Host–Guest …

47

Fig. 2.10 a Lowest energy isomer of 18C6·CA. b, c Structures of the lowest energy isomers of 18C6·CA within the energy of 5 kJ/mol at the level of ωB97X-D/6-31++G** calculation. Reproduced from [37] with permission from the Royal Society of Chemistry

in Fig. 2.7g, showing a good agreement between the observed and calculated ones. On the other hand, 18C6·CA-E1 and 18C6·CA-E4 have very similar structure with each other. In these complexes, the intramolecular H-bond is broken and the two OH groups are independently H-bonded to the oxygen atoms of 18C6. This rearrangement of the H-bonding leads to the two OH stretching vibrations to have similar frequencies at ~3400 cm−1 . The calculated IR spectra of 18C6·CA-E1 and 18C6·CA-E4 well reproduce the observed IR spectra of Fig. 2.7h and i, respectively. Of course such the structure is not stable in bare form. In addition, 18C6 also changes its structure from the lowest energy bare form to higher energy conformation for the best fitting. Similar to 18C6·phenol complex, this type of complex formation cannot be described by a simple “key and lock” model but should be described by “induced-fit” theory. “Induced-fit” theory was first proposed by D. E. Koshland Jr. to describe the enzyme interaction [50]. As was described in the introduction, CEs have two characteristics which are the key factors for the recognition of the guest molecule. One is the size recognition, that is, different sizes of CE incorporate different sizes of the guest species, especially the cations. The other is the shape recognition. This comes from the flexibility of CEs. Since CEs are built by flexible “–CH2 –CH2 –O–” units, they can adjust their shapes to include the guest species having different structures. These characteristics cooperatively work as the molecular recognition. In the case of benzenediols and 18C6, both the guest and host adjust each structure to form best-fitted complex. Here, we will discuss the energetics of the “induced fitting” occurring in the host–guest complex formation of 18C6·benzenediol. Similar to the case of 3nCn·phenol complexes, we define the interaction energy E int by the following equation: E int = Binding energy + (E CE + E benzenediol )

(2.2)

48

T. Ebata

Here, E CE is the distortion energy of bare 18C6 from the lowest energy conformation to that in the complex. E benzenediol is also the distortion energy in the benzenediol part. Figure 2.11 (a) shows the schematic energy diagram for the 18C6·CAE1 complex formation, and (b) for the 18C6·CA-A1 complex formation. In case of 18C6·CA-E1, 18C6 and CA take 17.3 kJ/mol higher energy conformations in total. However, they obtain the interaction energy as large as 93.2 kJ/mol as a result of the complex formation, and the binding energy of 75. 8 kJ/mol. In case of 18C6·CA-A1, the conformation distortion energy is 12. 5 kJ/mol, slightly less than 18C6·CA-A1, but they obtain the interaction energy of 92. 1 kJ/mol, and corresponding binding energy of 79. 7 kJ/mol. Such the large binding energies come from that the host 18C6 and guest CA are bound via multiple interactions, such as the OH··O H-bonding, CH··π, and dipole–dipole interactions. Similar results are obtained for the 18C6·HQ and 18C6·RE complexes. These results strongly indicate that the 18C6·benzenediol host–guest complex formation is described as the “induced-fit” model. Figure 2.12 shows the cartoon picture of the “induced-fit” model for the complexes.

Fig. 2.11 Energy level of 18C6 and catechol for forming a 18C6·CA-E1 isomer and b 18C6·CAA1 isomer. In each complex, lowest energy conformers in bare forms are shown in the left, that of the distorted ones for forming the complex in the middle, and that of the complexes in the right. E CE and E CA represent the destabilization energies due to the distortion of conformation to form the stable complex. E int represents the interaction energy given in Eq. (2.2). Reproduced from [37] with permission from the Royal Society of Chemistry

2 Microscopic Study on Molecular Recognition of Host–Guest …

49

Fig. 2.12 Induced fit between crown-ether and benzenediols. (left) lowest energy conformer of catechol (CA), resorcinol (RE), and hydroquinone (HQ). (middle) 18-crown-6 (18C6). (right) Stable structures of 18C6·benzenediols. Both 18C6 and benzenediols change their stable conformation to form best-fitted complexes

2.4.2 Effect of Host–Guest Complex Formation on the Photophysics of 3nCn·Benzenediols Complexes: Cage Effect Here, we show that the complex formation with CE sometimes changes the photophysics of guest species drastically, by showing example of 3nCn·benzenediols complexes, which we call “cage effect”. Figure 2.13 shows the decay curves of jet-cooled bare (a, b) HQ, (d, e) RE and (g) CA at the zero-point level of S1 . HQ and RE have similar S1 lifetime (a few nanoseconds), while that of CA is as short as 8.0 ps. The unusual short lifetime of CA is described by the fast internal conversion (IC) to nearby repulsive 1 πσ* state, followed by the release of the H atom [51]. The complexation with 18C6 drastically lengthens this short lifetime as shown in Fig. 2.13h and i, that is, the S1 lifetime of 18C6·CA becomes 10.3 ns. This is an increase by a factor of ~1000. In this complex, intramolecular H-bond of CA is broken and each OH group is independently H-bonded to the oxygen atoms of 18C6. This intermolecular H-bond works as a barrier for the H atom elimination of CA after UV excitation to S1 . Thus, 18C6 is working as a “cage” not to release the H atom from CA. We call this effect as “cage effect”. In reality, TD-DFT calculation showed that the complexation with 18C6 lifts the 1 πσ* state of bare CA to higher energy so that IC is prohibited at low energy [37].

50

T. Ebata

Fig. 2.13 Decay curves of a, b HQ, d, e RE and g CA monomers at the zero-point level of S1 , measured by picosecond pump–probe spectroscopy (red). Lifetimes of c 18C6·HQ, f 18C6·RE, and h, i 18C6·CA complexes. f, h, and i are fluorescence decay curves (red). Solid curves are convoluted decay curves. Reproduced from [37] with permission from the Royal Society of Chemistry

The lengthening of the S1 lifetime upon the complex formation with 18C6 leads to a drastic enhancement of fluorescent quantum yield (QD) of CA. We examined this effect in solution at room temperature. Figure 2.14 (a) shows the UV absorption and (b) shows fluorescence spectra of CA with and without the addition of 18C6 in cyclohexane solution. Addition of 18C6 leads to the UV absorption spectrum of CA slightly broad but its effect is small (Fig. 2.14a). On the other hand, the fluorescence spectrum exhibits drastic enhancement by the addition of 18C6 (Fig. 2.14b). Since 18C6 does not have a chromophore, this enhancement is attributed to the increase of the fluorescence QY of CA. We measured qualitatively the enhancement to obtain the equilibrium constant of the reaction, 18C6 + CA  18C6·CA.

(2.3)

The equilibrium constant K of this reaction is expressed as K =

[18C6·CA] , [CA][18C6]

(2.4)

where [18C6·CA] is the concentration of the 18C6·CA complex under the equilibrium condition. By employing the formation probability (α) of the 1:1 complex, Eq. (2.4) can be rewritten as

2 Microscopic Study on Molecular Recognition of Host–Guest …

51

Fig. 2.14 a Absorption spectrum of catechol and 1:1 mixture of catechol/18C6 in cyclohexane solution at room temperature. The concentration of catechol is 5.3 × 10−4 mol/L. b Fluorescence spectrum of catechol at different mixing ratios of 18C6. The excitation wavelength is fixed at 280 nm (35,714 cm−1 ), and catechol concentration is fixed at [catechol]0 = 1.0 × 10−4 mol/L. c Plots of normalized fluorescence intensity of catechol versus [18C6], where the catechol concentration is fixed at [catechol]0 = 1.0 × 10−4 mol/L (see text)

K =

α[18C6]0 , ([CA]0 − α[18C6]0 ){(1 − α)[18C6]0 }

(2.5)

where [18C6]0 and [CA]0 are the initial concentrations of 18C6 and CA, respectively. In the experiment, the fluorescence of CA monomer and 18C6·CA complex cannot be separated, and we measure the total fluorescence intensity (F) of the sum of CA(F 1 ) and 18C6·CA (F 2 ) as a function of the [18C6]0 /[catechol]0 ratio as shown in Fig. 2.14c. We define the fluorescence detection efficiency of the apparatus as A, and fluorescence QY of CA and 18C6·CA as φ CA and φ 18C6 ·CA , respectively. Then, F 1 and F 2 can be written as F1 = A[CA]φCA = A([CA]0 − α[18C6]0 )φCA ,

(2.6)

F2 = A[18C6·CA]φ18C6·CA = A(α[18C6]0 )φ18C6·CA .

(2.7)

In Fig. 2.14c, we plotted the normalized fluorescence intensity R given by the following equation:

52

T. Ebata

R=

F1 + F2 ([CA]0 − α[18C6]0 )φC A + α[18C6]0 φ18C6·CA = F1 ([18C6]0 = 0) [CA]0 φCA

(2.8)

as a function of [18C6]0 /[CA]0 . Since φCA  φ18C6·CA , Eq. (2.8) can be simplified to R =1+

αφ18C6·CA [18C6]0 . φCA [CA]0

(2.9)

Hence, R is in proportional to [18C6]0 /[CA]0 . In Fig. 2.14c, we plot R versus [18C6]0 /[CA]0 and the slope, in the range of 0 < [18C6]0 /[CA]0 < 2.0, is obtained to be αφ18C6·CA = 5.5. φcatechol

(2.10)

The fluorescence QYs of CA and 18C6CA are not known. However, we can assume that the ratio φ 18C6 ·CA /φ catechol is roughly equal to the inverse ratio of the fluorescence lifetimes of catechol and 18C6·CA in solution. The fluorescence lifetime of the 18C6·catechol complex was experimentally obtained to be 1.9 ns by another experiment [37]. In addition, we further assume that the fluorescence lifetime of CA monomer in cyclohexane is the same as that of gas phase (8 ps). Using these assumptions, we determine α to be 2.0 × 10−2 . Since the [CA]0  α[18C6]0 condition is satisfied, Eq. (2.5) can be simplified as K =

α . [catechol]0

(2.11)

From [catechol]0 = 1.0 × 10−4 mol/L and α = 2.0 × 10−2 , K is determined to be 2.0 × 102 L/mol (log K = 2.30). This value is compared to other reactions involving 18C6, such as 18C6 + Mn+  18C6·Mn+ . For example, log K = 2.34 for M+ = Li+ in acetonitrile solution at 300 K [52], 2.31 for Na+ in methanol at 298 K [53], 2.42 for Hg2+ in water at 298 K [11], and 2.44 for M3+ =Nd3+ in methanol at 298 K [54]. For molecular cation, logK = 2.37 for PhN2 + in methanol at 298 K [55]. Thus, we can conclude that 18C6 efficiently captures catechol even in solution.

2.5 18C6·Tyramine and 18C6·Tyrosol Complexes We further investigate the host–guest complexes of 18C6 by changing more flexible molecules, tyramine (TA) and tyrosol (TS), as a guest. TA and TS have side chain of –CH2 –CH2 –NH2 and –CH2 –CH2 –OH at para-position of phenol, respectively, and many conformations are possible as shown in Figs. 2.15 and 2.16.

2 Microscopic Study on Molecular Recognition of Host–Guest …

53

Fig. 2.15 Possible conformers of tyramine (TA)

Fig. 2.16 Possible conformers of tyrosol (TA)

The analyses of the conformational landscape of bare TA [55–57] and TS [58] have been done based on the laser spectroscopic study and theoretical calculations [56–59]. In both species, most stable conformer has the structure in which NH or alcoholic OH is directing to the benzene ring by NH··π or OH··π H-bonding. Figure 2.17 shows LIF spectra of jet-cooled (a) TA and the 18C6·TA complex and (b) TS and 18C6·TS complex in the (0,0) band region of the S1 –S0 transition. Also shown are the UV–UV HB spectra for 18C6·TS complex bands. Laser spectroscopic studies of TA in bare form are already reported [56–58]. In Fig. 2.17a, the

54

T. Ebata

Fig. 2.17 a LIF spectra of TA and 18C6·TA complex. Bands a–g are the (0,0) bands of different conformers of TA, and bands A, B are those of 18C6·TA. b (upper) LIF spectra of TS and 18C6·TS complex and (lower) UV–UV HB spectra (black) of 18C6·TS obtained by monitoring bands I and II, respectively. Bands 1–3 are the (0,0) bands of different conformers of TS, and bands I, II are different isomers of 18C6·TS

bands labeled a–g are the (0,0) bands of different conformers of TA. TA monomer has nine possible conformations, as shown in Fig. 2.15, and seven conformers are observed under jet-cooled condition and are analyzed. The bands A and B in the 35,000–35,150 cm−1 regions newly emerge by addition of 18CA, so that they are attributed to the 18C6·TA complex. In Fig. 2.17b, the bands labeled 1–3 are the (0,0) bands of different conformers of TS. Among nine possible conformations, three conformers’ bands appear in the 35,500–35,700 cm−1 region. The conformational assignment for these bands is also reported [58]. Bands I and II in the 34,800–34,900 cm−1 region are due to 18C6·TA complex. The UV–UV HB spectra indicate that they are different isomers. Figure 2.18 (left) (a)–(f) shows the IR spectra of the TA monomer obtained by monitoring bands (a)–(g) in Fig. 2.17a. All the spectra show the free OH stretch band at 3658 or 3659 cm−1 . Unfortunately, the NH2 stretch bands are not observed probably due to its weak absorption intensity. Figure 2.18 (left) (g) and (h) shows the IR spectra of 18C6·TA obtained by monitoring the complex bands (A) and (B), respectively. In these spectra, the OH band appeared at 3439 and 3458 cm−1 is assigned to the H-bonded OH band of TA. Since the magnitudes of the redshift for these bands are similar to those of 18C6·phenol and 18C6·benzenediols, the complexes have the structure that phenolic OH is H-bonded to the oxygen atom of 18C6. The NH2 stretching band is not observed in either spectrum due to the same reason for the monomer IR spectra. Figure 2.18 (right) (i) and (j) shows the IR spectra of TS monomer by monitoring bands 1 and 2 of Fig. 2.17b, respectively. Both spectra show two OH bands at 3652 and 3665 cm−1 . In Fig. 2.16, TS-Gg’t and TS-Gg’c have the OH··π H-bond between

2 Microscopic Study on Molecular Recognition of Host–Guest …

55

Fig. 2.18 (left) IR-UV DR spectra of a–f TA monomer, g species A and f B of 18C6·TA complex. m Calculated IR spectrum of 18C6·TA-I shown in Fig. 2.19. (right) IR-UV DR spectra of i, j TS monomer, and k species I and l II of 18C6·TS complex. n Calculated IR spectra of six stable isomers of 18C6·TS in Fig. 2.20. Frequencies of the calculated IR spectra are scaled so that the OH stretch IR bands of bare TA and TS are equal to the observed ones, respectively

the alcoholic OH and benzene ring, and thus the lower frequency band at 3652 cm−1 could be assigned to the alcoholic OH and the higher frequency one at 3665 cm−1 is assigned to phenolic OH of these conformers. Figure 2.18 (right) (k) and (l) shows the IR spectra of 18C6·TS complex measured by monitoring bands I and II, respectively. Though it is clear that in both species, I and II, phenolic OH is H-bonded. However, many bands appear in both IR spectra. The complication of the IR spectra is interpreted that each H-bonded OH stretch is accompanied by low-frequency modes, as are observed in the IR spectra of 3nC6·phenol (Fig. 2.3) and 18C6·RE (Fig. 2.7f) complexes. Such the combination modes sometimes appear when the H-bonded OH stretching vibration strongly couples with low-frequency modes. In that case, instead of infrared selection rule (υ = ±1), vibrationally adiabatic model [60] or vibrational Born–Oppenheimer approximation model is appropriate [61]. Then, the infrared spectrum from υ O H = 0 to distorted υ O H = 1 exhibits Franck–Condon-like structure showing combination bands with strongly coupled low-frequency vibrations. Thus, in the IR spectra obtained by monitoring band I, the OH stretch at 3387 cm−1 is accompanied by 54 cm−1 vibration, and the OH stretch at 3551 cm−1 is accompanied by 18 cm−1 vibration. The appearance of such the combination bands is more prominent for band II, and both the OH stretching vibrations at

56

T. Ebata

3289 and 3487 cm−1 are accompanied many low-frequency vibrations. These results indicate that in both 18C6·TA and 18C6·TS, the OH group(s) are strongly coupled with low-frequency intermolecular modes via H-bonding. In both complexes, it is clear that the number of isomers is considerably reduced upon the complex formation. So, we will take a look at the structures of the complexes with an aid of DFT calculation. Figure 2.19 shows the most stable six isomers of 18C6·TA. We see that the binding structure of phenolic part with 18C6 is almost the same in these complexes. They are stabilized via the OH··O and CH··π H-bonding. The main difference is the conformation of “–CH2 –CH2 –NH2 ” chain. Table 2.2 lists the relative total energies of the complexes (E), interaction energies (E int ), and relative energies of conformers of the 18C6 part and TA part with respect to the most stable ones, respectively. 18C6·TA-I is the lowest and 18C6·TA-II is the second lowest energy isomer. The conformations of TA portion in the 18C6·TA-I and TAIV complexes can be classified to TA-Gat, and in analogous way that of TA in the 18C6·TA-II and TA-V complex can be classified to TA-Ggt. In these complexes, NH is bound to benzene ring via NH··π H-bond. That is, even in the complex, TA prefers the conformation in which NH forms the NH··π H-bonding. The difference between 18C6·TA-I and TA-IV, and between 18C6·TA-II and TA-V is the difference of the conformation of 18C6 part. In 18C6·TA-I and 18C6·TA-II, the benzene CH group is forming a weak CH··O H-bonding with the oxygen(s) of 18C6. This interaction makes an additional stabilization of the complexes. Figure 2.18m shows the calculated IR

Fig. 2.19 Lowest energy six isomers of 18C6·TA obtained by DFT calculation at the ωB97X-D/631++G** level. The dashed lines represent the OH··O (r H ···O < 2.7 Å) and CH··π (r H ···π < 3.0 Å) interactions between TA and 18C6

2 Microscopic Study on Molecular Recognition of Host–Guest …

57

Table 2.2 Relative total energies (E) of the complexes, interaction energies (E int ), relative energies of the conformers of 18C6 (E CE ) and TA (E TA ) of the isomers of 18C6·TA complex. The energies are in unit of kJ/mol. Calculation is performed at the ωB97X-D/6-31++G** level E int

E 18C6

E TA

18C6·TA-I

E 0.00

103

9.38

3.30

18C6·TA-II

3.28

98.1

8.91

2.58

18C6·TA-III

7.99

96.5

8.89

6.90

18C6·TA-IV

8.93

94.4

9.58

2.87

18C6·TA-V

10.8

90.8

8.55

2.78

18C6·TA-VI

10.9

92.6

8.98

6.18

spectra of 18C6·TA-I. The position of the H-bonded OH stretching vibration well reproduces the observed one of band A. Figure 2.20 shows the most stable six isomers of 18C6·TS, and Table 2.3 lists the relative total energy of the complexes (E), interaction energy (E int ), and relative energies of conformers of the 18C6 part and TS part with respect to the most stable

Fig. 2.20 Lowest energy six isomers of 18C6·TS obtained by DFT calculation at the ωB97X-D/631++G** level. The dashed lines represent the OH··O (r H ···O < 2.7 Å) and CH··π (r H ···π < 3.0 Å) interactions between TS and 18C6

58

T. Ebata

Table 2.3 Relative total energies (E) of the complexes, interaction energies (E int ), relative energies of the conformers of 18C6 (E CE ) and TS (E TA ) of the isomers of 18C6·TS complexes. The energies are in unit of kJ/mol. Calculation is performed at the ωB97X-D/6-31++G** level E

E int

E 18C6

E TS

18C6·TS-I

0.00

116

10.3

11.4

18C6·TS-II

2.42

118

10.1

17.5

18C6·TS-III

2.56

112

11.1

18C6·TS-IV

2.94

102

18C6·TS-V

5.38

112

18C6·TS-VI

7.40

94.7

9.35 11.4 9.10

9.11 3.42 10.9 1.72

ones, respectively. As is expected, all the complexes have the same binding structure between phenolic part and 18C6, and are stabilized via the OH··O(18C6) and CH··π H-bonding. A characteristic feature of these isomers is that the alcoholic OH of TS is also H-bonded to the oxygen of 18C6. In bare form, TS has the stable TS-Gg’c and TS-Gg’t structures, in which the alcoholic OH is H-bonded to benzene ring, but in the complex the alcoholic OH prefers the oxygen of 18C6 as an H-bonding accepter instead of benzene. This is in good contrast with the case of 18C6·TA, in which NH of TA prefers free from H-bond in the complex with 18C6. Due to this intermolecular H-bonding, the interaction energy of 18C6·TS is roughly ~ 15 kJ/mol larger than that of 18C6·TA. Figure 2.18n shows the calculated IR spectra of the six isomers of 18C6·TS. Here, we focus on the four isomers within 5 kJ/mol of the relative total energy. Among them, the IR spectrum of 18C6·T-IV does not reproduce the observed one of either band I or II, since the alcoholic OH is free from H-bond in this conformer. For other three isomers, it is difficult to give a definitive assignment by the comparison of the OH stretching vibration. Finally, we summarize the similarity and difference between 18C6·TA and 18C6·TS. As guest species, both TA and TS are the para-substituted phenol. This leads to a similarity of the complex structures of 18C6·TA and 18C6·TS, in which phenolic part is wedged by bending 18C6, and are stabilized via the OH··O and CH··π H-bonding. The difference of the structure comes from that whether the end of the methylene chain is either NH2 or OH group. In the monomer, both of them prefer the intramolecular H-bonded conformation in bare form. However, in the complex, TA prefers its NH2 to be free from H-bond, while TS prefers its OH to be H-bonded to the oxygen of 18C6 instead of intramolecular H-bond. The difference between them comes from the fact that NH2 has two NH-bonds. If one NH forms NH··O H-bond with the oxygen of 18C6, the other HN has to point to opposite direction close to the CH of 18C6. This causes a steric hindrance and instability. Thus, TA is forced to choose the conformation to avoid the steric hindrance between CH and NH group rather than form an NH–O H-bond. The molecular recognition is the concerted conformer modification involving not only of the guest but also the host, and even a tiny difference is important in the “induced-fit” complexation process. The laser spectro-

2 Microscopic Study on Molecular Recognition of Host–Guest …

59

scopic study for the gas-phase cold complex combined with the quantum chemical calculations can reveal such the mechanism in detail at molecular level. Acknowledgements The author greatly appreciates Dr. R. Kusaka and Dr. F. Morishima for their help to complete this chapter.

References 1. Pedersen, C.J.: Cyclic polyethers and their complexes with metal salts. J. Am. Chem. Soc. 89, 7017–7036 (1988) 2. Pedersen, C.J.: The discovery of crown ethers. Science 241, 536–540 (1988) 3. Stuart, A.M., Vidal, J.A.: Perfluoroalkylated 4,13-diaza-18-crown-6 ethers: synthesis, phasetransfer catalysis, and recycling studies. J. Org. Chem. 72, 3735–3740 (2007) 4. Jose, N., Sengupta, S., Basu, J.K.: Selective production of benzaldehyde by permanganate oxidation of benzyl alcohol using 18-crown-6 as phase transfer catalyst. J. Mol. Catal. A: Chem. 309, 153–158 (2009) 5. Yang, J. –S., Hwang, C. –Y., Hsieh, C.-C., Chiou, S.-Y.: Spectroscopic correlations between supermolecules and molecules. Anatomy of the ion-modulated electronic properties of the nitrogen donor in monoazacrown-derived intrinsic fluoroionophores. J. Org. Chem. 124, 719–726 (2004) 6. Kimura, K., Mizutani, R., Yokoyama, M., Arakawa, R., Sakurai, Y.: Metal-ion complexation and photochromism of triphenylmethane dye derivatives incorporating monoaza-15-crown-5 moieties. J. Am. Chem. Soc. 51, 111–117 (2005) 7. Uchegbu, I.F., Vyas, S.P.: Non-ionic surfactant based vesicles (niosomes) in drug delivery. Int. J. Pharm. 172, 33–70 (1998) 8. Izatt, R.M., Rytting, J.H., Nelson, D.P., Haymore, B.L., Christensen, J.J.: Binding of alkali metal ions by cyclic polyethers: significance in ion transport processes. Science 164, 443–444 (1969) 9. Izatt, R.M., Nelson, D.P., Rytting, J.H., Haymore, B.L., Christensen, J.J.: A calorimetric study of the interaction in aqueous solution of several uni- and bivalent metal ions with cyclic popyether dicyclohexyl-18-crown-6 at 10, 25 and 40°. J. Am. Chem. Soc. 93, 1619–1623 (1971) 10. Pedersen, C.J., Frensdorff, H.K.: Macrocyclic polyethers and their complexes. Angew. Chem. Int. Ed. Engl. 11, 16–25 (1972) 11. Izatt, R.M., Terry, R.E., Haymore, B.L., Hansen, L.D., Dalley, N.K., Avondet, A.G., Christensen, J.J.: Calorimetric titration study of the interaction of several uni- and bivalent cations with 15-crown-5, 18-crown-6, and two isomers of dicyclohexo-18-crown-6 in aqueous solution at 25 °C and μ = 0.1. J. Am. Chem. Soc. 98, 7620–7626 (1976) 12. Lamb, J.D., Izatt, R.M., Swain, C.S., Christensen, J.J.: A systematic study of the effect of macrocycle ring size and donor atom type on the log K, H, and TS of reactions at 25 °C in methanol of mono- and divalent cations with crown ethers. J. Am. Chem. Soc. 102, 475–479 (1980) 13. Kumbhar, R.R., Dagade, D.H., Terdale, S.S., Patil, K.J.: Thermodynamic equilibrium constant studies on aqueous electrolytic (Alkaline Earth Chlorides) solutions containing 18-crown-6 at 298.15 K. J. Sol. Chem. 36, 259–273 (2007) 14. Zhang, H., Chu, J.H., Leming, S., Dearden, D.V.: Gas-phase molecular recognition: gas-phase crown ether-alkali metal ion complexes and their reactions with neutral crowns. J. Am. Chem. Soc. 113, 7415–7417 (1991) 15. Chen, Q., Cannell, K., Nicoll, J., Dearden, D.V.: The macrobicyclic cryptate effect in the gas phase. J. Am. Chem. Soc. 118, 6335–6344 (1996)

60

T. Ebata

16. Maleknia, S., Brodbelt, J.: Gas-phase selectivities of crown ethers for alkali metal ion complexation. J. Am. Chem. Soc. 114, 4295–4298 (1992) 17. Chu, I.H., Dearden, D.V., Bradshaw, J.S., Huszthy, P., Izatt, R.M.: Chiral host-guest recognition in an ion-molecule ration. J. Am. Chem. Soc. 115, 4318–4320 (1993) 18. Dearden, D.V., Zhang, H., Chu, I.H., Wong, P., Chen, Q.Z.: Macrocyclic chemistry without solvents: gas phase reaction rates. Pure Appl. Chem. 65, 423–428 (1993) 19. Ray, D., Feller, D., More, M.B., Glendening, E.D., Armentrout, P.B.: Cation–ether complexes in the gas phase: bond dissociation energies and equilibrium structures of Li+(1,2dimethoxyethane)x, x = 1 and 2, and Li+(12-crown-4). J. Phys. Chem. 100, 16116–16125 (1996) 20. Sobott, F., Kleinekofort, W., Brutschy, B.: Cation selectivity of natural and synthetic ionophores probed with laser-induced liquid beam mass spectrometry. Anal. Chem. 69, 3587–3594 (1997) 21. Anderson, J.D., Paulsen, E.S., Dearden, D.V.: Alkali metal binding energies of dibenzo-18crown-6: experimental and computational results. Int. J. Mass Spectrom. 227, 63–76 (2003) 22. Armentrout, P.B.: Cation-ether complexes in the gas phase: thermodynamic insight into molecular recognition. Int. J. Mass Spectrom. 193, 227–240 (1999) 23. Armentrout, P.B.: Mass spectrometry—not just a structural tool: The use of guided ion beam tandem mass spectrometry to determine thermochemistry. Am. Chem. Soc. Mass Spectrom. 13, 419–434 (2002) 24. Lee, S., Wyttenbach, T., Vonhelden, G., Bowers, M.T.: Gas phase conformations of Li+ , Na+ , K+ , and Cs+ complexed with 18-crown-6. J. Am. Chem. Soc. 117, 10159–10160 (1995) 25. Wyttenbach, T., von Helden, G., Bowers, M.T.: Conformations of alkali ion cationized polyethers in the gas phase: polyethylene glycol and bis[(benzo-15-crown-5)-15-ylmethyl] pimelate. Int. J. Mass Spectrom. 165, 377–390 (1997) 26. Hilderbrand, A.E., Myung, S., Clemmer, D.E.: Exploring crown ethers as shift reagents for ion mobility spectrometry. Anal. Chem. 78, 6792–6800 (2006) 27. Göth, M., Leremyte, F., Schmitt, X.J., Warnke, S., von Helden, G., Sobott, F., Pagel, K.: Gasphase microsolvation of ubiquitin: investigation of crown ether complexation sites using ion mobility-mass spectrometry. Analyst, 141, 5502–5510 (2016) 28. Kusaka, R., Inokuchi, Y., T. Ebata, T.: Laser spectroscopic study on the conformations and hydrated structures of benzo-18-crown-6-ether and dibenzo-18-crown-6-ether in supersonic jets. Phys. Chem. Chem. Phys. 9, 4452–4459 (2007) 29. Kusaka, R., Inokuchi, Y., T. Ebata, T.: Structure of hydrated clusters of dibenzo-18-crown-6ether in a supersonic jet—encapsulation of water molecules in the crown cavity. Phys. Chem. Chem. Phys. 10, 6238–6244 (2008) 30. Kusaka, R., Inokuchi, Y., Ebata, T.: Water-mediated conformer optimization in benzo-18crown-6-ether/water system. Phys. Chem. Chem. Phys. 11, 9132–9140 (2009) 31. Kokubu, S., Kusaka, R., Inokuchi, Y., Haino, T., Ebata, T.: Laser spectroscopic study on (dibenzo-24-crown-8-ether) -water and -methanol complexes in supersonic jets. Phys. Chem. Chem. Phys. 12, 3559–3565 (2010) 32. Kusaka, R., Inokuchi, Y., Xantheas, S.S., Ebata, T.: Structures and encapsulation motifs of functional molecules probed by laser spectroscopic and theoretical methods. Sensors 10, 3519–3548 (2010) 33. Kusaka, R., Kokubu, S., Inokuchi, Y., Haino, T., Ebata, T.: The structure of host-guest complexes between dibenzo-18-crown-6 and water, ammonia, methanol, and acetylene—evidence of molecular recognition on the complexation. Phys. Chem. Chem. Phys. 13, 6827–6836 (2011) 34. Kusaka, R., Inokuchi, Y., Haino, T., Ebata, T.: Structures of (3n-crown-n)-phenol (n = 4, 5, 6, 8) host-guest complexes: formation of a uniquely stable complex for n = 6 via collective intermolecular interaction. J. Phys. Chem. Lett. 3, 1414–1420 (2012) 35. Inokuchi, Y., Kusaka, R., Ebata, T., Boyarkin, O.V., Rizzo, T.R.: Laser spectroscopic study of cold host-guest complexes of crown ethers in the gas phase. ChemPhysChem 14, 649–660 (2013) 36. Morishima, F., Kusaka, R., Inokuchi, Y., Haino, T., Ebata, T.: Anomalous cage effect of the excited state dynamics of catechol in the 18C6-catecol host-guest complex. J. Phys. Chem. B 119, 2557–2565 (2015)

2 Microscopic Study on Molecular Recognition of Host–Guest …

61

37. Morishima, F., Kusaka, R., Inokuchi, Y., Haino, T., Ebata, T.: Cage effect on conformational preference and photophysics in the host-guest complex of benzenediols with 18-crown-6. Phys. Chem. Chem. Phys. 18, 8027–8038 (2016) 38. Ebata, T., Inokuchi, Y.: Laser spectroscopic study of cold gas phase host-guest complexes of crown ethers. Chem. Records 16, 1034–1053 (2016) 39. Shubert, V.A., James, W.H., Zwier, T.S.: Jet-cooled electronic and vibrational spectroscopy of crown ethers: benzo-15-crown-5 ether and 4 -amino-benzo-15-crown-5 Ether. J. Phys. Chem. A 113, 8055–8066 (2009) 40. Ebata, T., Kusaka, R, Inokuchi, Y.: Vibrational spectroscopy of gas phase functional molecules and their complexes cooled in supersonic beams. In: de Caro, D. (Ed.) Vibrational Spectroscopy, pp. 79–114. InTech (2012) 41. Kolossváry, I., Guida, W.C.: Low mode search. an efficient, automated computational method for conformational analysis: application to cyclic and acyclic alkanes and cyclic peptides, J. Am. Chem. Soc. 118, 5011–5019 (1996) 42. MacroModel, version 9.1, Schrödinger, LLC, New York (2005) 43. Halgren, T.A.: MMFF VI. MMFF94s option for energy minimization studies, J. Comput. Chem. 20, 730–748 (1999) 44. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Scalmani, G., Barone, V., Mennucci, B., Petersson, G.A., et al.: Gaussian 09, Revision D.01. Gaussian, Inc., Wallingford CT (2009) 45. Watanabe, T., Ebata, T., Tanabe, S., Mikami, N.: Size-selected vibrational spectra of phenol(H2 O)n (n = 1-4) clusters observed by IR-UV double resonance and stimulated Raman-UV double resonance spectroscopies. J. Chem. Phys. 105, 408–419 (1996) 46. Humphrey, S.J., Pratt, D.W.: High resolution S1 –S0 fluorescence excitation spectra of hydroquinone. Distinguishing the cis and trans rotamers by their nuclear spin statistical weights. J. Chem. Phys. 99, 5078 (1993) 47. Melandri, S., Maccaferri, G., Caminati, W., Favero, P.G.: Conformational equilibrium in resorcinol by means of the free-jet absorption millimeter wave spectrum. Chem. Phys. Lett. 256, 513–517 (1996) 48. Bürgi, T.S., Leutwyler, S.: O–H torsional vibrations in the S0 and S1 states of catechol. J. Chem. Phys. 101, 8418–8429 (1994) 49. Gerhards, M., Perl, W., Schumm, S., Henrichs, U., Jacoby, C., Kleinermanns, K.: Structure and vibrations of catechol and catechol-H2 O(D2 O) in the S0 and S1 state. J. Chem. Phys. 104, 9362–9375 (1996) 50. Koshkand Jr., D.E.: Application of a theory of enzyme specificity to protein synthesis. Proc. Natl. Acad. Sci. USA 44, 98–104 (1958) 51. Livingstone, R.A., Thompson, J.O.F., Iljina, M., Donaldson, R.J., Sussman, B.J., Paterson, M.J., Townsend, D.: Time-resolved photoelectron imaging of excited state relaxation dynamics in phenol, catechol, resorcinol, and hydroquinone. J. Chem. Phys. 137,184304 (17 pp.) (2012) 52. Smetana, A.J., Popov, A.I.: Resonance and calorimetric study of Lithium crown complexes in various solvents. J. Solut. Chem. 9, 183–196 (1980) 53. Lin, J.D., Popov, A.I.: Nuclear magnetic resonance studies of some sodium ion complexes with crown ethers and [2]-cryptands in various solvents. J. Am. Chem. Soc. 103, 3773–3777 (1981) 54. Izatt, R.M., Lamb, J.D., Christensen, J.J., Haymore, B.L.: Anomalous stability sequence of Lanthanide(III) chloride complexes with 18-crown-6 in methanol. Abrupt decrease to zero from Gd3+ ion to Tb3+ ion. J. Am. Chem. Soc. 99, 8344–8346 (1977) 55. Izatt, R.M., Lamb, J.D., Rossiter, B.E., Izatt, N.E., Christensen, J.J., Haymore, B.L.: Thermodynamics of formation of 18-crown-6 complexes with Arenediazonium and Anilinium salts in methanol at 25 °C. J. Chem. Soc. Chem. Commun. 9, 386–387 (1978) 56. Teh, C.K., Sulkes, M.J.: Spectroscopy of jet-cooled tyrosine derivatives. J. Chem. Phys. 94, 5826–5832 (1991) 57. Yoon, I., Seo, K., Lee, S., Lee, Y., Kim, B.: Conformational study of tyramine and its water clusters by laser spectroscopy. J. Phys. Chem. A 111, 1800–1807 (2007) 58. Makara, K., Misawa, K., Miyazaki, M., Mitsuda, H., Ishiuchi, S., Fujii, M.: J. Phys. Chem. A 112, 13463–13469 (2008)

62

T. Ebata

59. Hockridge, M.R., Knight, S.M., Robertson, E.G., Simons, J.P., McCombie, J., Walker, M.: Conformational landscapes in flexible organic molecules: 4-Hydroxy phenyl ethanol (p-tyrosol) and its singly hydrated complex. Phys. Chem. Chem. Phys. 1, 407–413 (1999) 60. Craig, S.M., Menges, F.S., Duong, C.H., Denton, J.K., Madison, L.R., McCoy, A.B., Johnson, M.A.: Hidden role of intermolecular proton transfer in the anomalously diffuse vibrational spectrum of a trapped hydronium ion. Proc. Natl. Acad. Sci. USA 114, 4706–4713 (2017) 61. Ebata, T., Nagao, K., Mikami, N.: Mode-dependent anharmonic coupling between OH stretching and intermolecular vibrations of the hydrogen-bonded clusters of phenol. Chem. Phys. 231, 199–204 (1998)

Chapter 3

Chirality Effects in Jet-Cooled Cyclic Dipeptides Ariel Pérez-Mellor and Anne Zehnacker

OH Substitution of the aromatic residues stabilizes the stacked conformation of the diketopiperazine dipeptide for identical absolute configuration of the residues only. Opposite absolute configuration of the residues results in folded-extended conformations similar to those observed for the nonsubstituted dipeptide

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_3

63

64

A. Pérez-Mellor and A. Zehnacker

Abstract Jet-cooled bichromophoric cyclic dipeptides built on a diketopiperazine (DKP) ring are studied by combining conformer-specific vibrational spectroscopy with quantum chemical calculations. The dependence of the c-LL and c-LD dipeptides structure upon relative absolute configuration, L or D, of the residues is investigated for two residues: phenylalanine (Phe) and tyrosine (Tyr). A folded-extended structure is systematically observed for all systems, like in solution or in the solid. This structure is stabilized by NH…π and CH…π interactions and shows limited stereoselectivity; the only difference between c-LL and c-LD is the nature of the CH…π interaction—Cα …π in c-LD and Cβ …π in c-LL—and a stronger NH…π interaction in c-LD. For all the species studied, the electronic excitation and the charge in the radical cation are localized on the extended ring. The c-LL diastereomer of cyclo di-tyrosyl stands out by the existence of a stacked structure, in which formation of an OH…O hydrogen bond stabilizes parallel aromatic rings orientation. In this structure, the electronic excitation and the major part of the charge in the cation are localized on the H-bond donor. The OH…O H-bond is possible in c-LL and not c-LD, which explains the high stereoselectivity. Keywords Chirality · Diketopiperazine · DKP · Laser spectroscopy · Supersonic expansion

3.1 Introduction Dipeptides with aromatic residues such as phenylalanine (Phe) or tyrosine (Tyr) spontaneously undergo intramolecular peptide bond formation in the solid phase, with concomitant loss of water [1, 2]. The cyclic dipeptide formed thereby is called diketopiperazine (DKP). DKP formation sometimes is an unwanted reaction, as in overheated sweetener aspartame [3, 4]. However, many DKP peptides with aromatic residues have beneficial medicinal applications as antivirals, antiparasitics, anticancer therapy [5–7], or as catalysts [8, 9]. The condensed-phase structures of aromatic DKP dipeptides are very diverse. Supramolecular assemblies can form, thanks to the conjunction of hydrophobic interactions between non-polar residues and NH…O=C hydrogen bonds between two DKP rings. Long ladder-like structures built on the repetition of a double NH…O=C hydrogen bond motif are observed in cyclo diphenylalanine of natural chirality L [10]. Tyr containing dipeptides spontaneously cyclize to DKP nanotubes, while linear diphenylalanine cyclizes to DKP nanofibrils and nanowires [11]. Determining the DKP monomer structure helps to understand the interactions that shape these nanostructures. Unless rare exceptions like cyclo diglycine [12, 13], the peptide ring is non-planar. Its conformation results from a compromise between the steric hinA. Pérez-Mellor · A. Zehnacker (B) Institut des Sciences Moléculaires d’Orsay (ISMO), CNRS, Univ. Paris-Sud, Université Paris-Saclay, 91405 Orsay, France e-mail: [email protected]

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

65

drance due to the substituents and the planarity of the amide bond. It ranges from pseudo boat or chair to intermediate twisted shapes [14, 15]. Most of the structural studies conducted so far for determining the substituents geometry rest either on NMR in solution or X-ray experiments in the solid. Bulky residues such as Phe or Tyr adopt a flagpole position of the substituent [16, 17]. Steric hindrance is increased by the presence of two bulky substituents. In the cyclo diphenylalanine crystal, one of them is folded over the DKP ring in a flagpole position, while the second one is extended [18, 19]. In what follows, we will focus on cyclo diphenylalanine and cyclo dityrosine, noted in short c-Phe-Phe and c-Tyr-Tyr. The absolute configuration of the residues will be denoted by L and D. The notation c-LD or c-LL will be used for discussing the common aspects of Phe and Tyr containing dipeptides. Both c-LPhe-LPhe and c-LTyr-LTyr have been studied by NMR in polar or protic solvents [16]. The results seem to point at rather symmetrical structures. However, electronic circular dichroism has suggested that c-LTyr-LTyr structures with almost parallel rings coexist with fully extended structures in solution [20]. We have undertaken the study of DKP dipeptides in different environments, either in the solid state by vibrational circular dichroism (VCD), protonated in a room temperature ion trap, or under supersonic expansion conditions [21–25]. This study aims to understand the factors that determine the structural differences between the molecule with residues of natural chirality L and that containing one D residue. This work is part of our recent studies on chirality effects in cyclic systems at low temperature [26–29]. We will describe here the spectroscopic properties of jet-cooled c-Phe-Phe and c-Tyr-Tyr, which are both built from two aromatic residues. c-Phe-Phe will serve as a model system in which the interaction between the two aromatic residues is mainly due to dispersion. In cTyr-Tyr, the presence of the hydroxyl on the aromatic rings may cause additional interactions such as OH…O or OH…π hydrogen bond. The presence of the DKP ring limits the conformational freedom to the motions of the aromatic substituents. These two molecules are therefore good model systems for studying the localization of the electronic excitation in a bichromophoric system in a constrained geometry [30–35]. The question of the localization of the excitation can also be raised for vibrations, and DKP dipeptides allow probing the coupling between vibrational modes in a system of well-defined geometry. The studied cyclic dipeptides are shown in Fig. 3.1.

Fig. 3.1 Scheme of the DKP dipeptides, with atom numbering. The chiral centers are indicated by *. c-Phe-Phe: R = H and c-Tyr-Tyr: R = OH

66

A. Pérez-Mellor and A. Zehnacker

3.2 Methodology 3.2.1 Experimental Methods The experimental set-up rests on a supersonic expansion equipped with a homemade laser desorption source and a time-of-flight mass spectrometer [36, 37]. Argon was used as a carrier gas for sufficient cooling of the studied dipeptide. The adiabatic expansion obtained thereby results in fast efficient cooling of the internal degrees of freedom and kinetic trapping of the most stable conformers present at the source temperature. The potential energy surfaces of the studied dipeptides are expected to show several minima separated by energy barriers and the conformational temperature describing the isomer population distribution is closer to room temperature rather than to the low rotational or vibrational temperature achieved in the jet [38]. Mass-resolved S0 –S1 spectra were obtained by one-color resonance-enhanced two-photon ionization (RE2PI) spectroscopy. Vibrational spectra were obtained using the IR-UV double-resonance technique in the ν(NH) and ν(OH) stretch region [39, 40]. Fixing the UV probe on the main vibronic bands of the electronic spectrum and scanning the IR pump in the 3 μm region allowed for the measurement of mass-resolved conformer-selective vibrational absorption spectra, as dip spectra in the UV probe-induced ion current. The IR pulse was triggered 80 ns before the UV pulse for recording the IR spectrum of the neutral molecules in their electronic ground state, and 50 ns after the UV pulse for the ion. A homemade active baseline subtraction scheme was used for monitoring the IR absorption as the difference in ion signal produced by successive UV laser pulses (one without and one with the IR laser present). Additional double-resonance experiments were performed by setting the IR to the vibrational transitions observed in the vibrational spectrum and scanning the UV probe. They did not reveal new UV absorptions, which indicates that all the conformers were detected. The dipeptides described here bear two identical chromophores and one can raise the question of the localization of the electronic transition. To answer this question, we followed the approach successfully used by Leutwyler’s and Zwier’s groups. It rests on the symmetry breaking arising from the presence of a single 13 C atom in natural isotopic abundance (~20% for molecules of this size). The RE2PI spectrum of the singly 13 C substituted isotopologue reflects the convolution of the exciton splitting and the site splitting due to the dissymmetry [41–43]. It corresponds to the average of the differences in the zero-point energies (ZPE) between all 12 C molecules and those containing one 13 C. Statistically, the probability for the 13 C to be located on one of the aromatic rings (12 carbon atoms) is much larger than that of being located on the peptide ring (4 carbon atoms). We can therefore assume that the two bands of the doublet correspond to the excitation of the all 12 C ring and of that containing 13 C. The peptides under study were purchased from Novopep Limited (Shanghai—China) and used without further purification.

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

67

3.2.2 Theoretical Methods DKP cyclic dipeptides display a complex conformational landscape due to the possible orientation of the aromatic substituents. The potential energy surface of c-LPheLPhe and c-LPhe-DPhe was explored using the OPLS-2005 force field combined with the advanced conformational search implemented in the MacroModel program of the Schrödinger package [44]. The potential energy surface of c-LTyr-LTyr and cLTyr-DTyr was then manually explored, starting from the six local minima obtained for c-Phe-Phe. All local minima found thereby were fully optimized within the frame of the DFT theory using the dispersion-corrected functional B3LYP-D3 [45, 46] associated with the Pople 6-311++g(d, p) split-valence basis set [47]. This level of theory satisfactorily accounts for the vibrational spectroscopy of similar systems at an acceptable calculation cost [21, 24, 48]. In particular, inclusion of dispersion is important for aromatic bichromophoric species [45, 46]. The charge distribution was obtained from the Natural Bond Orbital (NBO) analysis [49]. The electronic excited state energies were calculated by the time-dependent DFT (TD-DFT) method for the first ten singlet excited states, at the WB97XD/aug-cc-pVDZ level of theory. This level of theory satisfactory describes the electronic excited states of aromatic molecules [50, 51]. The vertical ionization energy was calculated at the unrestrictedDFT/B3LYP level. Optimization of the radical cation was performed at the same level of theory as the neutrals by removing an electron from the calculated neutral forms, which reflects the vertical ionization process. The vibrational frequencies were first calculated within the frame of the harmonic approximation at the same level of theory. The absence of imaginary frequency was checked for all local minima found. The structural differences between diastereoisomers are often subtle [36, 52] and the definite assignment of the observed structures rests on a reliable comparison between observed and simulated spectra in the region of 3 μm. To this end, one can resort to several strategies. The most commonly used is to use tabulated or empirical scaling factors to account for anharmonicity and basis set incompleteness [53]. Better agreement between experimental and simulated spectra is obtained when using mode-dependent scaling factors based on an extensive library of similar systems [54, 55]. An additional difficulty comes from the fact that combination bands or overtones of the amide I and II modes, that is ν(CO) and δ(NH), appear in this region [56]. Description of the vibrational modes beyond the harmonic approximation is therefore desirable. An efficient but computationally expensive approach is full anharmonic calculations using the variational perturbation theory [57–59]. To circumvent the computational cost, one can also limit the calculations to relevant selected modes. It gives similar results if the modes are carefully selected. Another compromise between accuracy and computational cost is to define the scaling factor as the slope of the linear regression between harmonic and anharmonic frequencies, for all the computed structures of a given molecule. This results in different scaling factors for different spectral ranges. For example, full harmonic calculations were performed for the model system cyclo tyrosyl proline

68

A. Pérez-Mellor and A. Zehnacker

[23]. The scaling factors for the fingerprint, the ν(CH), and the ν(NH)/ν(OH) region are 0.977, 0.957, and 0.953, respectively. These scaling factors are usually similar for molecules of similar structure: in the case of cyclo Phe-Phe, the scaling factor defined by the ratio between mode-selected anharmonic frequencies and harmonic ones is 0.952 in the ν(OH) and ν(NH) stretch region. In what follows, we will use the full anharmonic frequencies calculations for the most stable structures of c-Phe-Phe. In parallel, anharmonic frequencies including only ν(OH) and ν(NH) stretches were computed for all the structures of c-Phe-Phe and c-Tyr-Tyr. In the case of c-Phe-Phe, the agreement between full or selected modes calculation and the experiment is excellent. Less satisfactory agreement is obtained for c-Tyr-Tyr, probably because the modes included in the calculations are not sufficient. We will therefore use the harmonic frequencies scaled by 0.952 for the latter. The harmonic frequencies given in the text and tables include the scaling factor. All calculations were performed with the Gaussian 09 package [60].

3.3 Results and Discussion 3.3.1 Nomenclature of the Studied Systems The DKP ring has limited conformational mobility; the most important parameters defining the dipeptides geometry are related to the orientation of the aromatic substituent relative to the amide bond (Fig. 3.2). Three geometries are minima of the potential energy surface: two gauche geometries, g+ and g− , correspond to dihedral angles τ1 (N C1 C5 C6 ) and τ2 (N C3 C12 C13 ) of ~60° and −60° for the L residue, respectively, with opposite sign of the angles for the D residue. The trans t geometry

Fig. 3.2 Nomenclature for the aromatic substituent conformation in a D residue (left) or an L residue (right). In all the schemes, the substituent is in g+ conformation. The two other possible orientations are indicated by the corresponding letters t and g− . Nomenclature for the hydroxyl substituent orientation: anticlockwise (type I) orientation of the hydroxyl group in a Newman projection or clockwise orientation (type II)

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

69

corresponds to τ1 and τ2 angles of ~180°. For c-Tyr-Tyr, an additional parameter describes the orientation of the tyrosine hydroxyls relative to the DKP ring, I and II (see Fig. 3.2). For each benzyl substituent orientation, the number of isomers is thus multiplied by four in c-Tyr-Tyr relative to c-Phe-Phe. The nomenclature used, in what follows, starts with “c-” for the cyclic nature of the peptide, followed by the orientation of the substituent g+ , g− , or t. L or D in subscript denotes the configuration of each residue. In the case of c-Tyr-Tyr, it is followed by I or II for the tyrosyl OH positions.

3.3.2 Symmetry Properties Particular symmetry properties arise from the fact that the DKP dipeptides studied here are made of two identical residues, as it is the case for the simple model cyclo diglycine. The latter is not chiral and possesses Ci symmetry because the DKP ring is planar [12]. In the systems studied here, the presence of substituents is the cause of dissymmetry. First, the DKP ring becomes out of plane to accommodate the bulky aromatic rings, which has consequences on the symmetry of the system. In contrast with planar cyclo diglycine for which the two H atoms are in equivalent positions, the substituents of the Cα carbons now occupy distinct positions, axial and equatorial [61]. From the stereochemical point of view, chirality is introduced by substituting the Cα atoms (Cα1 and Cα3 ) by benzyl (c-Phe-Phe) or hydroxybenzyl (c-Tyr-Tyr). The molecules therefore exist as four stereoisomers. In c-LL, the two substituents are both in equatorial or axial positions. The DKP ring itself may belong to the C2 or C1 (no symmetry) point group. All the C1 or C2 conformations are chiral and their enantiomer is their mirror image, c-DD. In c-LD, one of the substituents is in equatorial, and the other is in axial position. As the two residues have opposite chirality, c-LD can be seen at first sight as a meso compound, which is not chiral. The DKP ring may have Ci symmetry, or no symmetry (C1 ). Indeed, Ci conformers are not chiral, for example cyclo D-alanylL-alanine has planar DKP ring and Ci symmetry [13]. In contrast, C1 geometries are chiral because the two aromatic substituents have non-equivalent orientations. This happens when dissymmetry is brought about by the interaction between the two residues. This type of chirality is transient at room temperature but is frozen under supersonic expansion conditions. The dissymmetric structures therefore exist as pairs of non-superimposable mirror images under supersonic jet conditions, c-LD and c-DL. We will limit the discussion to the former in what follows. As will be seen later, all local minima of c-LD are non-symmetrical. The non-chiral Ci structures are transition states. This contrasts with c-LL for which some of the local minima belong to the C2 symmetry point group, as described in what follows.

70

A. Pérez-Mellor and A. Zehnacker

3.3.3 Theoretical Results 3.3.3.1

Calculated Structure in the Ground Electronic State

The calculated structures are separated into six families resulting from the combination of the g+ , g− , and t orientations of the aromatic substituents, defined by the angles τ1 and τ2 introduced earlier. We will limit the discussion to the most stable families, shown in Fig. 3.3. We will also discuss the pseudo equatorial or pseudo axial position of the benzyl substituents. The orientation of the tyrosine OH increases the

Fig. 3.3 Structures of c-Phe-Phe and c-Tyr-Tyr. The relative energy is given in parentheses in kcal/mol

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

71

number of conformers in c-Tyr-Tyr, without modifying neither the energetics nor the spectroscopy, except in the stacked conformation described below. This is reminiscent of the amino acid tyrosine whose conformers due to hydroxyl rotation cannot be discriminated by their IR signature [62, 63]. This is why we will discuss the calculated structures in terms of families that include the four OH orientations. c-g + g − (“Folded-Extended”) structures: This geometry is asymmetric, with the substituents in dissymmetric orientations: the g + aromatic ring is folded over the amide bond and g − is extended out of the DKP ring, which results in stabilizing combination of CH…π and NH…π interactions. The DKP geometry is in a twisted boat conformation. The c-g L+ g L− structure of c-LL is stabilized by a C12 H…π interaction and a weak NH…π interaction. Both substituents, in particular the folded one, are in pseudo axial position. For the c-g L+ g − D conformer of c-LD, the pseudo axial folded L residue acts as an acceptor in the C3 H…π interaction, while the extended D residue is equatorial and acts as the donor. Whatever the chirality, the pseudo axial orientation of the folded substituent is of prime importance as it allows formation of the secondary CH…π and NH…π interactions. The “folded-extended” geometry c-g+ g− is the most stable structure in all systems. It is more stable by 1.3 kcal/mol and contributes by more than 90% to the total population of c-LPhe-LPhe. The energetic advantage is slightly less (0.95 kcal/mol) for c-LPhe-DPhe but is still large enough for c-g+ g− to be the only one present in − the jet. The energy of the four c-g+ L gL conformers is the same within 0.3 kcal/mol in c-Tyr-Tyr and they are probably present in the supersonic expansion. Altogether, − c-g+ L gD conformers account for 80% of the population of c-LTyr-DTyr. However, − c-LTyr-LTyr stands out as c-g+ L gL conformers only amount to ~60% of the total population, a point to which we shall return later. The c-g+ g− (“Folded-Extended”) structures show little dependence upon absolute configuration of the residues. However, the fact that both residues have identical or opposite absolute configuration changes the nature of the CH…π interaction. In c-LD, due to the equatorial nature of the extended g− substituent, there is a − Cα H…π interaction in c-g+ L gD while two axial substituents in c-LL result in Cβ H…π + − interaction in c-gL gL . In addition, the NH…π interaction is slightly stronger in − + − + − c-g+ L gD as indicated by the N2 H…π distance, shorter in c-gL gD than in c-gL gL by ~0.06 Å. c-g − g − (“Fully-Extended”) structures: They have C2 symmetry in c-LL, thus, − the two benzyl substituents are equivalent. In c-g− L gL , both aromatic rings, in axial − position, are fully extended. The equivalent structure in c-LD, c-g− L gD , only shows minor differences in the position of the phenyl rings related to the fact that one residue is in axial and the other in equatorial position. Fully extended structures are higher in energy by at least 1.8 kcal/mol in all the systems and will not be discussed further. Their relative energy illustrates the importance of correctly including dispersion, as expected in a system with two strongly polarizable aromatic rings. Non-inclusion of dispersion stabilizes the extended structures that become the most stable, which is in contradiction to the experimental results. c-g S+ g S+ (“Stacked”) structures: They also belong to the C2 symmetry point group for c-LL. The equatorial position is favored over axial as it releases the repulsion

72

A. Pérez-Mellor and A. Zehnacker

arising from parallel benzene rings. Still, c-g L+ g L+ lies 3 kcal/mol higher in energy than the most stable form in c-LPhe-LPhe. Indeed, dispersion is the only stabilizing interaction there and cannot counterbalance the expected repulsion between the benzene rings. In c-LPhe-DPhe, c-g L+ g L+ is not a stacked geometry because both the substituents cannot be in equatorial position. The geometry with maximum interaction between the aromatic rings is c-g L+ t D . Its relative energy exceeds 2 kcal/mol for c-Phe-Phe and c-Tyr-Tyr and we will not discuss it further. The energetics of the stacked structures are completely different in c-LTyr-LTyr. The stacked c-g L+ g L+ geometry facilitates the interaction between the two hydroxyls, which stabilizes this structure. For example, the Gibbs energy of c-g L+I g L+I is as low as 0.2 kcal/mol. In contrast with the other structures, the orientation of the OHs slightly influences the energetics. In c-g L+I g L+I , the two OH groups show antiparallel orientation of their electric dipoles and mostly interact through dipole–dipole interaction, as described for amide stacking in γ-peptides [64]. c-g L+I g L+I I differs from c-g L+I g L+I by the parallel orientation of the two OH that favors OH…O hydrogen bond formation. It is interesting to note that the dipole–dipole interaction is competitive with the hydrogen bond formation and that these two structures are almost isoenergetic. However, c-g L+I g L+I I reproduces the experimental findings better than c-g L+I g L+I , as described later. Interestingly, the most stable conformer of c-LL is always more stable than the most stable conformer of c-LD, by 0.4 kcal/mol for c-Phe-Phe and 2.0 kcal/mol for c-Tyr-Tyr.

3.3.3.2

Calculated Structure in the Cation

Optimization of the cation of c-Phe-Phe indicates charge localization on the nitrogen atoms. For this reason, the NH stretches are shifted down in energy and not observed experimentally. We have therefore focused on c-LTyr-LTyr and c-LTyr-DTyr. The calculated structures of their cations are shown in Fig. 3.4. Starting points for the − optimization are the most stable neutral “folded-extended” structures c-g+ LII gLI and + − + + c-gLI gDI , and the “stacked” geometry of c-LTyr-LTyr, c-gLI gLII , reflects the vertical ionization process. For the “folded-extended” structure, the main difference relative to the neutral is that the g− conformation of the extended benzyl is not stable in the ion. Instead, the g− aromatic ring undergoes rotation toward a t conformation. This leads to the cion -g+ t structure. Indeed, the NH…π interaction in the neutral becomes repulsive in the ion and is replaced by an interaction between the amide CO and the positively charged aromatic ring. + The geometry of the stacked conformer c-g+ LI gLII is not much modified upon ionization, except the OH…O distance, which is considerably shorter in the ion (1.94 vs. 2.23 Å). The stronger H-bond associated with this shorter distance stabilizes the + + hydrogen-bonded form cion -g+ LI gLII by 2.7 kcal/mol relative to c-gLII tLI . The NBO charges distribution, given in Fig. 3.4, is very dissymmetrical. In the folded-extended structure, cion -g+ t, most of the charge is on the extended g+ LI benzyl. Then, the ν(OH) stretch frequency of the folded tyrosyl is similar to that in the neutral.

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

73

Fig. 3.4 Calculated structure of the c-Tyr-Tyr cations, together with the charges on the aromatic rings and the Cα H groups. Only charges larger than 0.1 are indicated

The rest of the charge is mainly borne by Cα H groups. Interestingly, the amide bonds are neutral. + The charge of the hydrogen-bonded cation cion -g+ LI gLII is distributed on both aromatic rings. Still, it is dissymmetric, being 0.654 on the H-bond donor versus 0.232 on the acceptor. Thus, the ν(OH) stretch frequency of the H-bond acceptor is less shifted relative to the neutral than that of the donor. We will keep these remarks in mind when discussing the vibrational spectroscopy of the cation.

3.3.4 Experimental Results 3.3.4.1

Electronic Spectroscopy

The electronic spectra are shown in Figs. 3.5 and 3.6. That of c-Phe-Phe (Fig. 3.5) shows a simple vibronic pattern, almost identical for c-LPhe-Lhe and c-LPhe-DPhe. It shows an intense origin located at very similar energies, 37,603 and 37,600 cm−1 for c-LPhe-LPhe and c-LPhe-DPhe, respectively, in the region of the phenylalanine monomer [65–67]. It is followed by a strong Hertzberg-Teller allowed transition at 532 and 529 cm−1 for c-LPhe-LPhe and c-LPhe-DPhe, respectively. Almost no other vibronic activity is observed. These spectra indicate similar rigid structures for the two diastereomers. As observed for c-Phe-Phe, the spectrum of c-Tyr-Tyr appears in the region of the amino acid monomer. However, it contrasts strongly with that of c-Phe-Phe. c-LTyrDTyr only shows a featureless absorption with two maxima at 35,500 and 35,800 cm−1 , which is assigned to a species denoted by Bc-LD (Fig. 3.6). The spectrum of

74

A. Pérez-Mellor and A. Zehnacker

Fig. 3.5 RE2PI spectrum of a c-LPhe-LPhe, b c-LPhe-DPhe. The inset shows the comparison between the origin recorded at the mass of all 12 C c-Phe-Phe and at the mass of a single 13 C c-PhePhe. The 13 C spectrum is multiplied by 5. The calculated frequencies are labeled by νn . ν26 is the Herzberg-Teller allowed mode. The bands probed by the UV laser for measuring the IR spectra are indicated by *. Adapted from Ref. [24] (license 4551230911971) Copyright 2018, with permission from Elsevier

c-LTyr-LTyr shows similar broad maxima, at ~35,600 and 35,800 cm−1 , assigned to a species denoted by Bc-LL , and superimposed with narrow lines, with an origin at 35,274 cm−1 , assigned to a species denoted by Ac-LL . We exclude that the broad absorption arises from non-radiative processes in the electronic excited state because the spectra of cold tyrosine or protonated tyrosine are well resolved [63, 68, 69]. A more likely hypothesis is spectral congestion and/or insufficient cooling due to OH isomerism.

3.3.4.2

Vibrational Spectroscopy and Assignment

c-g + g − Structures: Identical IR-UV spectra are obtained whatever the band probed in the RE2PI spectrum of c-LPhe-LPhe or c-LPhe-DPhe, which indicates the presence of a single conformer under supersonic jet conditions. c-LPhe-LPhe and c-LPheDPhe show similar spectroscopic signatures with a triplet at 3393–3407–3416 cm−1 for c-LPhe-LPhe and at 3386–3417–3424 cm−1 for c-LPhe-DPhe. The different frequencies for the ν(NH) stretch point at dissymmetric structures. We will therefore

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

75

Fig. 3.6 RE2PI electronic spectrum of a c-LTyr-LTyr, b c-LTyr-DTyr. The bands probed by the UV laser for measuring the IR dip spectra are indicated by their frequency. Adapted and reproduced from Ref. [22] with permission from the Royal Society of Chemistry

discard the C2 structures for the assignment and focus on the most stable “foldedextended” c-gL+ gL− and c-gL+ gD− structures. Figure 3.7 shows the comparison between the experimental spectra and those resulting from the anharmonic frequencies calculations for c-g+ g− . The band at 3416 (c-LPhe-LPhe) or 3417 cm−1 (c-LPhe-DPhe) is assigned to the free ν(NH) stretch of the g+ extended benzyl, perfectly reproduced by the anharmonic calculations at 3411 cm−1 (c-gL+ gL− ) and 3412 cm−1 (c-gL+ gD− ). − − −1 (c-g+ The lower-energy band at 3393 cm−1 (c-g+ L gL ) or 3386 cm L gD ) is assigned − to the ν(NH) stretch of the g folded benzyl, involved in the NH…π interaction. The downshift of the frequency is very well reproduced by the anharmonic calculations − − −1 for (c-g+ that yield 3392 cm−1 for (c-g+ L gL ) and 3387 cm L gD ). For all systems studied here, a third band appears in the ν(NH) stretch region, the intensity of which decreases more rapidly than that of the others when reducing the laser power. It can be explained by taking into account overtones or combination − bands involving the ν(CO) stretch and the δ(NH) bend. In c-g+ L gL the two ν(CO) stretches are uncoupled and calculated at different frequencies, the lower-energy band at 1705 cm−1 corresponds to the CO of the amide interacting with the folded benzyl and the high-energy ν(CO) at 1713 cm−1 to the other one. Anharmonic calculations predict active ν(CO) overtones at 3395 and 3410 cm−1 , in the vicinity of the ν(NH) stretch. They could be responsible for the band at 3407 cm−1 .

76

A. Pérez-Mellor and A. Zehnacker

Fig. 3.7 Comparison between the IR-UV spectra and those of the structures to which they are assigned for c-Phe-Phe (left) and c-Tyr-Tyr (right). a c-LL experimental spectra. b Calculated spectra for c-LL. c c-LD experimental spectra. d Calculated spectra for c-LD. Adapted and reproduced from Ref. [22] with permission from the Royal Society of Chemistry. Adapted and reprinted Ref. [24] (license 4551230911971) Copyright 2018, with permission from Elsevier − In contrast, the two ν(CO) modes are strongly coupled in c-g+ L gD , and result in a −1 forbidden symmetric mode at 1714 cm and a strongly allowed asymmetric mode at 1716 cm−1 . Anharmonic calculations predict an intense combination band of the two ν(CO) stretches at 3413 cm−1 , in resonance with the free ν(NH). It could be responsible for the band at 3424 cm−1 . Such intense combination bands or overtones have been observed already in dipeptides, including those built on the diketopiperazine ring [70] or β sheets models [56]. Similar vibrational spectra are observed in the ν(NH) stretch region when setting the probe on the broad absorption observed in the RE2PI spectra of c-LTyr-LTyr and cLTyr-DTyr. Again, the IR spectrum of each diastereomer does not depend on the probe position; again, c-LTyr-LTyr and c-LTyr-DTyr show similar spectroscopic signatures in the region of ν(NH), with three congested bands at 3400, 3412, and 3424 cm−1 for c-LTyr-LTyr and a doublet at 3394–3432 cm−1 accompanied by a shoulder at 3417 cm−1 for c-LTyr-DTyr. The spectra of c-Tyr-Tyr only differ from that of c-PhePhe by the presence of a narrow band at 3656 ± 1 cm−1 , characteristic of a free ν(OH) stretch. The spectra of Bc-LL and Bc-LD can be interpreted in terms of “folded-extended structures” identical to those calculated for c-Phe-Phe. They compare well to that + − gLI simulated for the c-g + g − structures; in particular, the most stable of them c-gLII + − − 1 and c-gLI gDI shown in Fig. 3.3. The sharp band at 3656 cm is the superposition of the two free ν(OH). The assignment of the triplet in the 3400–3412 cm−1 range parallels that described for cyclo Phe-Phe. We therefore assign the broad absorption − structures differentiating only by the tyrosyl OH to the superposition of c-gL+ gSL orientation.

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

77

The major difference between the two diastereoisomers, for both c-Tyr-Tyr and cPhe-Phe, is the energy difference between the bound and free ν(NH) stretches, which is always larger in c-LL than c-LD (experimental value of 23 cm−1 vs. 31 cm−1 for c-Phe-Phe and 12 vs. 38 cm−1 for c-Tyr-Tyr. This difference reflects the stronger NH…π interaction in c-LD. It should be noted that scaled harmonic frequencies satisfactorily account for the frequency gap between c-LL and c-LD, for both studied molecules. c-g + g + Structures: The IR-UV spectrum of c-LTyr-LTyr recorded with the probe at the origin or any of the narrow transitions is shown in Fig. 3.7. Compared to the spectra described above, the free ν(OH) stretch is slightly shifted down in energy (3648 cm−1 ). The doublet at 3409 and 3428 cm−1 is similar to those described for ν(NH) in the other systems and points at a dissymmetric structure with nonequivalent benzyls. Last, the intense peak at 3554 cm−1 appears in the range of bound ν(OH) stretches. This spectral pattern is very well reproduced by that calculated + −1 for c-g + LI g LII , which is the only structure that reproduces the feature at 3554 cm − 1 assigned to the bound OH. The band at 3648 cm is assigned to the free ν(OH) calculated at 3649 cm−1 , which is slightly shifted down in energy due to its role as a hydrogen bond donor. The band at 3409 cm−1 is assigned to the strong asymmetric ν(NH) stretch overlapped with the weak symmetric combination, calculated at 3413/3414 cm−1 . Finally, the band at 3428 cm−1 is assigned to an overtone or combination band.

3.3.5 Localization of the Electronic Transition 3.3.5.1

Experimental Spectra

c-LPhe-LPhe: The 000 transition of singly 13 C-substituted c-Phe-Phe, shown in the inset of Fig. 3.5, is split by ~4 cm−1 . This splitting is the convolution of the exciton splitting due to the coupling between the locally excited states and the site splitting due to the dissymmetry of the molecule and the non-equivalence of the benzyl rings. Because of the relatively large distance between the chromophores, the exciton splitting is smaller than the experimental resolution in similar systems such as 1,3diphenoxymethane [43] or bis-phenoxymethane [30] and is negligible. The different intensities within the doublet confirm that c-LPhe-DPhe is not a symmetrical conformation. The observed spectrum is therefore characteristic of a bichromophoric system with non-equivalent subunits and well-localized transitions. Indeed, the change in electron density between S0 and S1 electronic states (vertical transition) reflects the localized character of the excitation, both in c-Phe-Phe and c-Tyr-Tyr (see Fig. 3.8). The changes in electron density are mainly located on a single benzene ring, pointing at the localization of the transition.

78

A. Pérez-Mellor and A. Zehnacker

Fig. 3.8 Difference in electron density between S0 and S1 in the structures corresponding to the − + − experiment for the folded-extended forms of a c-LPhe-LPhe c-g + L g L , b c-LTyr-LTyr c-g L I I g L I , + − + + c c-LTyr-DTyr c-g L I g D I and the stacked structure of d c-LTyr-LTyr c-g L I g L I . The electron density isovalue is 0.004 a.u. The electron density difference is coded in blue for an increase upon electronic excitation and red for a decrease. The S0 → S1 energy is given in parentheses for the different conformers of c-Tyr-Tyr

3.3.5.2

Simulated Spectra

Optimization of the first electronic excited state and simulation of the S0 → S1 − transition of the c-g+ L gL conformer of c-LPhe-LPhe, introducing both Franck–Condon and Hertzberg terms, confirm the experimental findings. After shifting down the calculated origin by ~2230 cm−1 to scale it on the experimental value, excellent agreement between simulated and experimental vibronic patterns is obtained, as shown in Fig. 3.5. The intense band at +532 cm−1 is assigned to the Hertzberg-Teller allowed transition, akin to ν6 of benzene. The geometry of the S1 state is very close to that of the ground state, apart from a minor rotation of the g− S extended residue. This minor change manifests itself by the presence of a weak vibrational progression built on a 25 cm−1 mode, assigned to the ν1 (36 cm−1 ) mode involving this rotation. c-LTyr-LTyr: The same 13 C experiments were not possible for c-Tyr-Tyr because of weak intensity and spectral congestion. Instead, we have compared the calculated vertical S0 –S1 energies for the conformers to which the experimental spectrum + is assigned (see Fig. 3.8). The order of the calculated S0 –S1 energies, c-g+ LI gLII < + − + − c-gL gD < c-gL gL , is in qualitative agreement with the experiment. In particular, the + 35274 cm−1 band assigned to the origin of c- g+ LI gLII is shifted down in energy relative

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

79

to the origin of the tyrosine conformers (35491–35650 cm−1 ) [63]. This is reminiscent of the S0 –S1 transition of the dipeptide tyrosyl-glycine, which is red-shifted by ~400 cm−1 relative to tyrosine when the hydroxyl group is not free [71]. In all the c-g+ g− structures, for c-Phe-Phe and c-Tyr-Tyr alike, the S0 –S1 transition is of ππ* nature and is localized on the extended g− aromatic ring. IR spectroscopy of the electronic excited state [72, 73] has been proposed recently as a tool for determining the localization of the energy [74]. The systems studied here should show two families of NH stretches, localized on the locally excited aromatic ring or not, with frequencies characteristic of the ground and electronic excited states. However, the lifetimes are not long enough for this experiment to be possible with nanosecond lasers. It would be interesting to perform this experiment on DKP peptides with chromophores with longer S1 lifetimes.

3.3.6 Localization of the Charge in the Cation The double-resonance spectrum of the c-LPhe-LPhe radical cation does not show any transition, probably because the ν(NH) stretches are shifted down out of the accessible frequency range. The double-resonance spectrum of the c-LTyr-LTyr cation is recorded with the UV probe set at the same positions as for the neutral ground state. The obtained spectra do not depend on the position of the probe. Monitoring the depletion of the parent at m/z 326 or the intensity of the fragment resulting from Cα –Cβ cleavage at m/z 220 also leads to the same spectra. This observation indicates that the measured spectra are due to structures populated after intramolecular vibrational energy redistribution (IVR). The same independence of the spectrum upon the probe wavelength is observed for c-LTyr-DTyr. The experimental spectra shown in Fig. 3.9 are analyzed in what follows at the light of the charge density calculations described in Sect. 3.2.2. c-LTyr-DTyr: The experimental spectrum (Fig. 3.9) shows excellent agreement with that simulated for cion -g+ LI tDI and reflects the dissymmetry of the phenol rings, which was apparent from the charge distribution (Sect. 3.2.2). The band at 3642 cm−1 −1 corresponds to the free ν(OH) of the neutral g+ LI ring calculated at 3638 cm . The −1 intense band at 3572 cm is assigned to the free ν(OH) of the charged tDI ring, calculated at 3586 cm−1 . This value is close to that observed for the cyclo LtyrosylLproline radical cation (3563 cm−1 ), where the charge is necessarily localized on Tyr [23]. It is also close to that of the phenol:argon cation (3535 cm−1 ) [75]. Lastly, the weak band at 3410 cm−1 is assigned to the two ν(NH) stretches calculated at identical values (3400 cm−1 ). The NH…π interaction observed in the neutral is absent in the cation, which makes the two NH equivalent with identical ν(NH) stretch frequencies, close to that of the ν(NH) of the neutral ground state, due to the very small charges on the two NH. c-LTyr-LTyr: The spectrum of c-LTyr-LTyr is more complex than that of c-LTyrDTyr and is accounted for by the contribution of two structures. The first one, cion -g+ LII tLII , is similar to that described for c-LTyr-DTyr. The only difference rel-

80

A. Pérez-Mellor and A. Zehnacker

Fig. 3.9 Experimental and simulated spectrum of the ionic state of a c-LTyr-LTyr and b c-LTyr-DTyr together with corresponding calculated structures. The relative Gibbs energy is given in parentheses in kcal/mol. Adapted and reproduced from Ref. [22] with permission from the Royal Society of Chemistry

ative to c-LTyr-DTyr is a small redshift (7 cm−1 ) of the free ν(OH) of the neutral g+ LII ring, observed at 3635 cm−1 and calculated at 3630 cm−1 . The free ν(OH) of the “charged” tLII ring appears at a similar value (3569 cm−1 ) as in c-LTyr-DTyr within the experimental error and is calculated at 3590 cm−1 . The ν(NH) stretches are also slightly shifted down in frequency (3394 cm−1 ) relative to c-LTyr-DTyr. The second + structure, cion -g+ LI gLI , is the hydrogen-bonded structure. As the hydrogen bond also influences the frequency of the donor, the free ν(OH) is slightly shifted down in −1 energy (3619 cm−1 ) relative to that of cion -g+ LII tLII and is calculated at 3613 cm . −1 The large intensity of the band at 3394 cm is explained by the superposition of + the intense bound ν(OH) of cion -g+ LI gLI , superimposed with the ν(NH) stretches of + + + cion -gLII tLII and cion -gLI gLI , all calculated at 3399 ± 3 cm−1 . Lastly, the free ν(OH) stretch shows a shoulder at its low-energy side that may be assigned to a hot band, as often observed in photo-cations of cyclic molecules [28]. The potential energy surface of the cation reflects that of the neutral: IVR populates one conformer only for c-LTyr-DTyr, while two different structures are observed for c-LTyr-LTyr, despite their energy difference. Interestingly, the stereo-selectivity is preserved in the cation. This contrast with other cyclic systems bearing two chiral

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

81

centers. In molecules where chirality is due to an asymmetric nitrogen atom, the nitrogen becomes planar upon ionization because it bears the positive charge [28]. As a result, the effects of stereochemistry are lost. Opening of the cycle upon ionization also results in a loss of stereochemical effects [27].

3.4 Conclusion and Perspectives Compared to the linear peptides, DKP peptides are rigid structures, with only one conformation of the DKP ring and the substituent orientation for the two diastereomers of c-Phe-Phe or c-LTyr-DTyr, and two for c-LTyr-LTyr. This contrasts with the parent amino acids, with six conformers for jet-cooled phenylalanine and 12 for tyrosine [63, 67]. None of the structures found is symmetrical. The asymmetry of the c-g+ g− conformers is due to the different orientation of the two aromatic rings that are folded and extended. As a result, the molecule cannot be seen like a meso − + − compound and c-g+ L gD is a mirror image to c-gD gL . Going from one enantiomer to the other only requires large amplitude motions inverting the folded and extended positions. At room temperature, interconversion is easy and the two transient enantiomers cannot be distinguished, unless embedded in special environments like a chiral liquid crystal, in which NMR experiments should discriminate between them [76, 77]. It should be noticed too that folded-extended structures are systematically found in many neutral DKP peptides, in condensed or in the gas phase, unless one introduces an interaction stronger than the NH…π or CH……π that are responsible for the foldedextended structures [19, 21, 23, 25]. In the case of c-LTyr-LTyr reported here, the interaction is an OH…O hydrogen bond. Future perspectives would be to modulate this interaction by changing the distance between the donor and the acceptor, for example by replacing one tyrosine by homo tyrosine. Adding a solvent molecule for bridging the two hydroxyls would be interesting too [78, 79]. We have reached the same conclusion concerning the predominance of folded-extended structures for protonated DKP peptides isolated in a room temperature ion trap [21]. It would be interesting to test this hypothesis further by resorting to cryogenic ion traps for isolating the most stable structures of DKP dipeptides and studying the relationship between their structure and their photoreactivity. Indeed, cryogenic ion traps coupled with IR or UV laser spectroscopy have proven to be a powerful tool for studying chiral recognition in clusters of chiral molecules [80, 81]. Both experimental and theoretical results point out at the localization of the electronic excitation, and most of the charge for the cation, to one aromatic ring. The coupling between the two moieties is therefore limited. However, more coupling effects between the two chromophores should be observed in electronic circular dichroism [82, 83]. Finally, calculations indicate that the c-LL structure is more stable than the c-LD, for both studied molecules. Although one cannot draw general conclusions from two systems only, this observation agrees well with the so-called homochirality of life.

82

A. Pérez-Mellor and A. Zehnacker

References 1. Borthwick, A.D.: 2,5-diketopiperazines: synthesis, reactions, medicinal chemistry, and bioactive natural products. Chem. Rev. 112(7), 3641–3716 (2012). https://doi.org/10.1021/ cr200398y 2. Basiuk, V.A., Gromovoy, T.Y.: The gas solid-phase 2,5-dioxopiperazine synthesis—cyclization of vaporous dipeptides on silica surface. Collect. Czech. Chem. Commun. 59(2), 461–466 (1994). https://doi.org/10.1135/cccc19940461 3. Berset, J.-D., Ochsenbein, N.: Stability considerations of aspartame in the direct analysis of artificial sweeteners in water samples using high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). Chemosphere 88(5), 563–569 (2012). https://doi.org/10. 1016/j.chemosphere.2012.03.030 4. Lin, S.Y., Wang, S.L.: Advances in simultaneous DSC-FTIR microspectroscopy for rapid solidstate chemical stability studies: some dipeptide drugs as examples. Adv. Drug Deliv. Rev. 64(5), 461–478 (2012). https://doi.org/10.1016/j.addr.2012.01.009 5. Sinha, S., Srivastava, R., De Clereq, E., Singh, R.K.: Synthesis and antiviral properties of arabino and ribonucleosides of 1,3-dideazaadenine, 4-nitro-1,3-dideazapurine and diketopiperazine. Nucleosides Nucleotides Nucl. Acids 23(12), 1815–1824 (2004). https://doi.org/10. 1081/ncn-200040614 6. Grande, F., Garofalo, A., Neamati, N.: Small molecules anti-HIV therapeutics targeting CXCR6. Curr. Pharm. Des. 14(4), 385–404 (2008) 7. Walchshofer, N., Sarciron, M.E., Garnier, F., Delatour, P., Petavy, A.F., Paris, J.: Anthelmintic activity of 3,6-dibenzyl-2,5-dioxopiperazine, cyclo(L-Phe-L-Phe). Amino Acids 12(1), 41–47 (1997). https://doi.org/10.1007/bf01373425 8. Danda, H.: Essential factors in asymmetric hydrocyanation catalyzed by cyclo (-(R)-Phe-(R)His-). Synlett 1991(04), 263–264 (1991) 9. Tanaka, K., Mori, A., Inoue, S.: The cyclic dipeptide cyclo [(S)-phenylalanyl-(S)-histidyl] as a catalyst for asymmetric addition of hydrogen cyanide to aldehydes. J. Org. Chem. 55(1), 181–185 (1990) 10. Jeon, J., Shell, M.S.: Self-assembly of cyclo-diphenylalanine peptides in vacuum. J. Phys. Chem. B 118(24), 6644–6652 (2014). https://doi.org/10.1021/jp501503x 11. Jaworska, M., Jeziorna, A., Drabik, E., Potrzebowski, M.J.: Solid state NMR study of thermal processes in nanoassemblies formed by dipeptides. J. Phys. Chem. C 116(22), 12330–12338 (2012). https://doi.org/10.1021/jp302616n 12. Gross, E.: The Peptides Analysis. Modern Techniques of Conformational Structural, and Configurational Analysis. Elsevier, Synthesis, Biology (2012) 13. Sletten, E.: Conformation of cyclic dipeptides. Crystal and molecular structures of cyclo-dalanyl-l-alanyl and cyclo-l-alanyl-l-alanyl (3,6-dimethylpiperazine-2,5-dione). J. Am. Chem. Soc. 92(1), 172–177 (1970). https://doi.org/10.1021/ja00704a028 14. Carlson, K.L., Lowe, S.L., Hoffmann, M.R., Thomasson, K.A.: Theoretical UV circular dichroism of aliphatic cyclic dipeptides. J. Phys. Chem. A 109(24), 5463–5470 (2005). https://doi. org/10.1021/jp046580u 15. Milne, P.J., Oliver, D.W., Roos, H.M.: Cyclodipeptides—structure and conformation of cyclo(Tyrosyl Prolyl). J Cryst. Spectrosc. 22(6), 643–649 (1992). https://doi.org/10.1007/ bf01160980 16. Kopple, K.D., Marr, D.H. (1967) Conformation of cyclic peptides. The folding of cyclic dipeptides containing an aromatic side chain. J. Am. Chem. Soc. 89(24), 6193–6200 (1967). https:// doi.org/10.1021/ja01000a035 17. Lin, C.-F., Webb, L.E.: Crystal structures and conformations of the cyclic dipeptides cyclo(Glycyl-L-tyrosyl) and cyclo-(L-seryl-L-tyrosyl) Monohydrate. J. Am. Chem. Soc. 95(20), 6803–6811 (1973). https://doi.org/10.1021/ja00801a046 18. Benedetti, E., Marsh, R.E., Goodman, M.: Conformational studies on peptides. X-ray structure determinations of six N-methylated cyclic dipeptides derived from alanine, valine, and phenylalanine. J. Am. Chem. Soc. 98(21), 6676–6684 (1976). https://doi.org/10.1021/ja00437a041

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

83

19. Gdaniec, M., Liberek, B.: Structure of cyclo(-L-Phenylalanyl-L-Phenylalanyl-). Acta Crystallogr. Section C-Crystal Struct. Commun. 42, 1343–1345 (1986). https://doi.org/10.1107/ s0108270186092338 20. Fleischhauer, J., Grotzinger, J., Kramer, B., Kruger, P., Wollmer, A., Woody, R.W., Zobel, E.: Calculation of the circular-dichroism spectrum of cyclo(L-Tyr-L-Tyr) based on a moleculardynamics simulation. Biophys. Chem. 49(2), 141–152 (1994). https://doi.org/10.1016/03014622(93)e0065-d 21. Alata, I., Perez-Mellor, A., Ben Nasr, F., Scuderi, D., Steinmetz, V., Gobert, F., Jaidane, N.E., Zehnacker-Rentien, A.: Does the residues chirality modify the conformation of a cyclodipeptide? Vibrational spectroscopy of protonated cyclo-diphenylalanine in the gas phase. J. Phys. Chem. A 121(38), 7130–7138 (2017). https://doi.org/10.1021/acs.jpca.7b06159 22. BenNasr, F., Perez-Mellor, A., Alata, I., Lepere, V., Jaidane, N.-E., Zehnacker, A.: Stereochemistry-dependent hydrogen bonds stabilise stacked conformations in jet-cooled cyclic dipeptides: (LD) vs. (LL) cyclo tyrosine-tyrosine. Faraday Discuss. (2018). https://doi. org/10.1039/c8fd00079d 23. Pérez Mellor, A., Alata, I., Zehnacker, A.: Conformational landscape of diastereomer cyclic dipeptide: IR spectroscopy of jet-cooled cyclo LTyr-LPro vs. cyclo LTyr-DPro. https://doi.org/ 10.1021/acs.jpcb.9b04529 24. Perez-Mellor, A., Alata, I., Lepère, V., Zehnacker, A.: Chirality effects in the structures of jet-cooled bichromophoric dipeptides. J. Mol. Spectrosc. 349, 71–84 (2018) 25. Perez-Mellor, A., Zehnacker, A.: Vibrational circular dichroism of a 2,5-diketopiperazine (DKP) peptide: evidence for dimer formation in cyclo LL or LD diphenylalanine in the solid state. Chirality 29(2), 89–96 (2017). https://doi.org/10.1002/chir.22674 26. Bouchet, A., Klyne, J., Ishiuchi, S.I., Dopfer, O., Fujii, M., Zehnacker, A.: Stereochemistrydependent structure of hydrogen-bonded protonated dimers: the case of 1-amino-2-indanol. Phys. Chem. Chem. Phys. 20(18), 12430–12443 (2018). https://doi.org/10.1039/c8cp00787j 27. Bouchet, A., Klyne, J., Piani, G., Dopfer, O., Zehnacker, A.: Diastereo-specific conformational properties of neutral, protonated and radical cation forms of (1R,2S)-cis and (1R,2R)-trans amino-indanol by gas phase spectroscopy. Phys. Chem. Chem. Phys. 17, 25809–25821 (2015). https://doi.org/10.1039/C5CP00576K 28. Mahjoub, A., Chakraborty, A., Lepere, V., Le Barbu-Debus, K., Guchhait, N., Zehnacker, A.: Chirality-dependent hydrogen bond direction in jet-cooled (S)-1,2,3,4-tetrahydro-3isoquinoline methanol (THIQM): IR-ion dip vibrational spectroscopy of the neutral and the ion. Phys. Chem. Chem. Phys. 11(25), 5160–5169 (2009) 29. Alauddin, M., Gloaguen, E., Brenner, V., Tardivel, B., Mons, M., Zehnacker-Rentien, A., Declerck, V., Aitken, D.J.: Intrinsic folding proclivities in cyclic-peptide building blocks: configuration and heteroatom effects analyzed by conformer-selective spectroscopy and quantum chemistry. Chem. Eur. J. 21(46), 16479–16493 (2015). https://doi.org/10.1002/chem. 201501794 30. Zehnacker, A., Lahmani, F., Breheret, E., Desvergne, J.P., BouasLaurent, H., Germain, A., Brenner, V., Millie, P.: Laser induced fluorescence of jet-cooled non-conjugated bichromophores: bis-phenoxymethane and bis-2,6-dimethylphenoxymethane. Chem. Phys. 208(2), 243–257 (1996) 31. Lahmani, F., Zehnacker, A., Desvergne, J.P., Bouas-Laurent, H., Colomes, M., Kruger, A.: Fluorescence emission in fluid solution and supersonic jet of symmetrical bisbenzenes incorporating a three membered flexible chain. J. Photochem. Photobiol. A Chem. 113(3), 203–212 (1998) 32. Baquero, E.E., James III, W.H., Choi, T.H., Jordan, K.D., Zwier, T.S.: Single conformation spectroscopy of a flexible bichromophore: 3-(4-Hydroxyphenyl)-N-benzylpropionamide. J. Phys. Chem. A 112(44), 11115–11123 (2008). https://doi.org/10.1021/jp806787p 33. Buchanan, E.G., Gord, J.R., Zwier, T.S.: Solvent effects on vibronic coupling in a flexible bichromophore: electronic localization and energy transfer induced by a single water molecule. J. Phys. Chem. Lett. 4(10), 1644–1648 (2013). https://doi.org/10.1021/jz400641p

84

A. Pérez-Mellor and A. Zehnacker

34. Pillsbury, N.R., Kidwell, N.M., Nebgen, B., Slipchenko, L.V., Douglass, K.O., Cable, J.R., Plusquellic, D.F., Zwier, T.S.: Vibronic coupling in asymmetric bichromophores: experimental investigation of diphenylmethane-d(5). J. Chem. Phys. 141(6). https://doi.org/10.1063/1. 4892344 35. Walsh, P.S., Buchanan, E.G., Gord, J.R., Zwier, T.S.: Binding water to a PEG-linked flexible bichromophore: IR spectra of diphenoxyethane-(H2O)(n) clusters, n = 2–4. J. Chem. Phys. 142(15). https://doi.org/10.1063/1.4917305 36. Sen, A., Bouchet, A., Lepere, V., Le Barbu-Debus, K., Scuderi, D., Piuzzi, F., ZehnackerRentien, A.: Conformational analysis of quinine and its pseudo enantiomer quinidine: a combined jet-cooled spectroscopy and vibrational circular dichroism study. J. Phys. Chem. A 116(32), 8334–8344 (2012) 37. Mengesha, E.T., Zehnacker-Rentien, A., Sepiol, J., Kijak, M., Waluk, J.: Spectroscopic study of jet-cooled deuterated porphycenes: unusual isotopic effects on proton tunneling. J. Phys. Chem. B 119(6), 2193–2203 (2015). https://doi.org/10.1021/jp505553z 38. Gloaguen, E., de Courcy, B., Piquemal, J.P., Pilmé, J., Parisel, O., Pollet, R., Biswal, H.S., Piuzzi, F., Tardivel, B., Broquier, M., Mons, M.: Gas-phase folding of a two-residue model peptide chain: on the importance of an interplay between experiment and theory. J. Am. Chem. Soc. 132(34), 11860–11863. https://doi.org/10.1021/ja103996q 39. Tanabe, S., Ebata, T., Fujii, M., Mikami, N.: OH stretching vibrations of phenol-(H2O)n(n = 1–3) complexes observed by IR-UV double-resonance spectroscopy. Chem. Phys. Lett. 215(4), 347–352 (1993) 40. Pribble, R.N., Zwier, T.S.: Size-specific infrared-spectra of benzene-(H2 O)(n) clusters (n = 1 through 7)—evidence for noncyclic (H2 O)(n) structures. Science 265(5168), 75–79 (1994) 41. Ottiger, P., Leutwyler, S., Koppel, H.: Vibrational quenching of excitonic splittings in H-bonded molecular dimers: the electronic Davydov splittings cannot match experiment. J. Chem. Phys. 136(17). https://doi.org/10.1063/1.4705119 42. Balmer, F.A., Kopec, S., Koppel, H., Leutwyler, S.: Excitonic splitting and vibronic coupling analysis of the m-Cyanophenol dimer. J. Phys. Chem. A 121(1), 73–87 (2017). https://doi.org/ 10.1021/acs.jpca.6b10416 43. Buchanan, E.G., Walsh, P.S., Plusquellic, D.F., Zwier, T.S.: Excitonic splitting and vibronic coupling in 1,2-diphenoxyethane: conformation-specific effects in the weak coupling limit. J. Chem. Phys. 138(20). https://doi.org/10.1063/1.4807300 44. MacroModel version 9.8. ed. Schrödinger, LLC, New York, NY (2010) 45. Grimme, S., Antony, J., Ehrlich, S., Krieg, H.: A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132(15). https://doi.org/10.1063/1.3382344 46. Grimme, S., Ehrlich, S., Goerigk, L.: Effect of the damping function in dispersion corrected density functional theory. J. Comput. Chem. 32(7), 1456–1465 (2011). https://doi.org/10.1002/ jcc.21759 47. Frisch, M.J., Pople, J.A., Binkley, J.S.: Self-consistent molecular-orbital methods. 25. Supplementary functions for Gaussian-basis sets. J. Chem. Phys. 80(7), 3265–3269. https://doi.org/ 10.1063/1.447079 48. Lepere, V., Le Barbu-Debus, K., Clavaguera, C., Scuderi, D., Piani, G., Simon, A.-L., Chirot, F., MacAleese, L., Dugourd, P., Zehnacker, A.: Chirality-dependent structuration of protonated or sodiated polyphenylalanines: IRMPD and ion mobility studies. Phys. Chem. Chem. Phys. 18(3), 1807–1817 (2016). https://doi.org/10.1039/c5cp06768e 49. Reed, A.E., Curtiss, L.A., Weinhold, F.: Intermolecular interactions for a natural bond orbital, donor-acceptor viewpoint. Chem. Rev. 88(6), 899–926 (1988) 50. Chai, J.D., Head-Gordon, M.: Long-range corrected hybrid density functionals with damped atom-atom dispersion corrections. Phys. Chem. Chem. Phys. 10(44), 6615–6620 (2008). https:// doi.org/10.1039/b810189b 51. Jacquemin, D., Perpete, E.A., Ciofini, I., Adamo, C.: Assessment of the omega B97 family for excited-state calculations. Theoret. Chem. Acc. 128(1), 127–136 (2011). https://doi.org/ 10.1007/s00214-010-0783-x

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

85

52. Sen, A., Lepere, V., Le Barbu-Debus, K., Zehnacker, A.: How do pseudoenantiomers structurally differ in the gas phase? An IR/UV spectroscopy study of jet-cooled hydroquinine and hydroquinidine. Chemphyschem. Eur. J. Chem. Phys. Phys. Chem. 14(15), 3559–3568 (2013). https://doi.org/10.1002/cphc.201300643 53. Johnson, R.D.: NIST computational chemistry comparison and benchmark database NIST standard reference database number 101. http://cccbdb.nist.gov/ (2018). Accessed 15 Jan 2019 54. Gloaguen, E., Mons, M.: Isolated neutral peptides. In: Rijs, A.M., Oomens, J. (eds) Gas-phase IR spectroscopy and structure of biological molecules. Topics in Current Chemistry-Series, vol. 364, pp. 225–270 (2015). https://doi.org/10.1007/128_2014_580 55. Plowright, R.J., Gloaguen, E., Mons, M.: Compact folding of isolated four-residue neutral peptide chains: H-bonding patterns and entropy effects. ChemPhysChem 12(10), 1889–1899 (2011). https://doi.org/10.1002/cphc.201001023 56. Gerhards, M., Unterberg, C.: Structures of the protected amino acid Ac-Phe-OMe and its dimer: a beta-sheet model system in the gas phase. Phys. Chem. Chem. Phys. 4(10), 1760–1765 (2002). https://doi.org/10.1039/b110029g 57. Bloino, J., Barone, V.: A second-order perturbation theory route to vibrational averages and transition properties of molecules: general formulation and application to infrared and vibrational circular dichroism spectroscopies. J. Chem. Phys. 136(12), 124108 (2012). https://doi. org/10.1063/1.3695210 58. Bloino, J., Baiardi, A., Biczysko, M.: Aiming at an accurate prediction of vibrational and electronic spectra for medium-to-large molecules: an overview. Int. J. Quantum Chem. 116(21), 1543–1574 (2016). https://doi.org/10.1002/qua.25188 59. Bloino, J., Biczysko, M., Barone, V.: Anharmonic effects on vibrational spectra intensities: infrared, Raman, vibrational circular dichroism, and Raman optical activity. J. Phys. Chem. A 119(49), 11862–11874 (2015). https://doi.org/10.1021/acs.jpca.5b10067 60. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Scalmani, G., Barone, V., Mennucci, B., Petersson, G.A., Nakatsuji, H., Caricato, M., Li, X.J., Hratchian, H.P., Izmaylov, A.F., Bloino, J., Zheng, G., Sonnenberg, J.L., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery, J.J.A., Peralta, J.E., Ogliaro, F., Bearpark, M., Heyd, J.J., Brothers, E., Kudin, K.N., Staroverov, V.N., Kobayashi, R., Normand, J., Raghavachari, K., Rendell, A., Burant, J.C., Iyengar, S.S., Tomasi, J., Cossi, M., Rega, N., Millam, J.M., Klene, M., Knox, J.E., Cross, J.B., Bakken, V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Martin, R.L., Morokuma, K., Zakrzewski, V.G., Voth, G.A., Salvador, P., Dannenberg, J.J., Dapprich, S., Daniels, A.D., Farkas, O., Foresman, J.B., Ortiz, J.V., Cioslowski, J., Fox, D.J.S.: Gaussian 09, Revision D.01. Gaussian Inc., Wallingford, CT (2009) 61. Urago, H., Suga, T., Hirata, T., Kodama, H., Unno, M.: Raman optical activity of a cyclic dipeptide analyzed by quantum chemical calculations combined with molecular dynamics simulations. J. Phys. Chem. B 118(24), 6767–6774 (2014). https://doi.org/10.1021/jp503874z 62. Inokuchi, Y., Kobayashi, Y., Ito, T., Ebata, T.: Conformation of L-tyrosine studied by fluorescence-detected UV-UV and IR-UV double-resonance spectroscopy. J. Phys. Chem. A 111(17), 3209–3215 (2007). https://doi.org/10.1021/jp070163a 63. Shimozono, Y., Yamada, K., Ishiuchi, S-i, Tsukiyama, K., Fujii, M.: Revised conformational assignments and conformational evolution of tyrosine by laser desorption supersonic jet laser spectroscopy. Phys. Chem. Chem. Phys. 15(14), 5163–5175 (2013). https://doi.org/10.1039/ c3cp43573c 64. James III, W.H., Buchanan, E.G., Guo, L., Geman, S.H., Zwier, T.S.: Competition between amide stacking and intramolecular H bonds in gamma-peptide derivatives: controlling nearestneighbor preferences. J. Phys. Chem. A 115(43), 11960–11970 (2011). https://doi.org/10.1021/ jp2081319 65. Martinez, S.J., Alfano, J.C., Levy, D.H.: The electronic spectroscopy of the amino-acids tyrosine and phenylalanine in a supersonic jet. J. Mol. Spectrosc. 156(2), 421–430 (1992)

86

A. Pérez-Mellor and A. Zehnacker

66. Lee, K.T., Sung, J., Lee, K.J., Kim, S.K., Park, Y.D.: Resonant two-photon ionization study of jet-cooled amino acid: L-phenylalanine and its monohydrated complex. J. Chem. Phys. 116(19), 8251–8254 (2002). https://doi.org/10.1063/1.1477452 67. Hashimoto, T., Takasu, Y., Yamada, Y., Ebata, T.: Anomalous conformer dependent S-1 lifetime of L-phenylalanine. Chem. Phys. Lett. 421(1–3), 227–231 (2006). https://doi.org/10.1016/j. cplett.2006.01.074 68. Boyarkin, O.V., Mercier, S.R., Kamariotis, A., Rizzo, T.R.: Electronic spectroscopy of cold, protonated tryptophan and tyrosine. J. Am. Chem. Soc. 128(9), 2816–2817 (2006). https://doi. org/10.1021/ja058383u 69. Ishiuchi, S., Wako, H., Kato, D., Fujii, M.: High-cooling-efficiency cryogenic quadrupole ion trap and UV-UV hole burning spectroscopy of protonated tyrosine. J. Mol. Spectrosc. 332, 45–51 (2017). https://doi.org/10.1016/j.jms.2016.10.011 70. Abo-Riziq, A.G., Crews, B., Bushnell, J.E., Callahan, M.P., De Vries, M.S.: Conformational analysis of cyclo(Phe-Ser) by UV-UV and IR-UV double resonance spectroscopy and ab initio calculations. Mol. Phys. 103(11–12), 1491–1495 (2005). https://doi.org/10.1080/ 00268970500095923 71. Abo-Riziq, A., Grace, L., Crews, B., Callahan, M.P., van Mourik, T., de Vries, M.S.: Conformational structure of Tyrosine, Tyrosyl-glycine, and Tyrosyl-glycyl-glycine by double resonance spectroscopy. J. Phys. Chem. A 115(23), 6077–6087 (2011). https://doi.org/10.1021/ jp110601w 72. Seurre, N., Sepiol, J., Lahmani, F., Zehnacker-Rentien, A., Le Barbu-Debus, K.: Vibrational study of the S-0 and S-1 states of 2-naphthyl-1-ethanol/(water)(2) and 2-naphthyl1-ethanol/(methanol)(2) complexes by IR/UV double resonance spectroscopy. Phys. Chem. Chem. Phys. 6(19), 4658–4664 (2004) 73. Loquais, Y., Gloaguen, E., Habka, S., Vaquero-Vara, V., Brenner, V., Tardivel, B., Mons, M.: Secondary structures in Phe-containing isolated dipeptide chains: laser spectroscopy vs quantum chemistry. J. Phys. Chem. A 119(23), 5932–5941 (2015). https://doi.org/10.1021/ jp509494c 74. Scutelnic, V., Prlj, A., Zabuga, A., Corminboeuf, C., Rizzo, T.R.: Infrared spectroscopy as a probe of electronic energy transfer. J. Phys. Chem. Lett. 9(12), 3217–3223 (2018). https://doi. org/10.1021/acs.jpclett.8b01216 75. Fujii, A., Sawamura, T., Tanabe, S., Ebata, T., Mikami, N.: Infrared dissociation spectroscopy of the OH stretching vibration of phenol-rare gas van der Waals cluster ions. Chem. Phys. Lett. 225(1–3), 104–107 (1994) 76. Aroulanda, C., Merlet, D., Courtieu, J., Lesot, P.: NMR experimental evidence of the differentiation of enantiotopic directions in C-s and C-2v molecules using partially oriented, chiral media. J. Am. Chem. Soc. 123(48), 12059–12066 (2001). https://doi.org/10.1021/ja011685l 77. Emsley, J.W., Lesot, P., Merlet, D.: The orientational order and conformational distributions of the two enantiomers in a racemic mixture of a chiral, flexible molecule dissolved in a chiral nematic liquid crystalline solvent. Phys. Chem. Chem. Phys. 6(3), 522–530 (2004). https://doi. org/10.1039/b312512b 78. Lahmani, F., Douhal, A., Breheret, E., Zehnacker-Rentien, A.: Solvation effects in jet-cooled 7-hydroxyquinoline. Chem. Phys. Lett. 220(3–5), 235–242 (1994) 79. Bach, A., Coussan, S., Muller, A., Leutwyler, S.: Water-chain clusters: vibronic spectra of 7-hydroxyquinoline center dot(H2 O)2 . J. Chem. Phys. 112(3), 1192–1203 (2000) 80. Klyne, J., Bouchet, A., Ishiuchi, S., Fujii, M., Schneider, M., Baldauf, C., Dopfer, O.: Probing chirality recognition of protonated glutamic acid dimers by gas-phase vibrational spectroscopy and first-principles simulations. Phys. Chem. Chem. Phys. 20(45), 28452–28464 (2018). https:// doi.org/10.1039/c8cp05855e 81. Fujihara, A., Sato, T., Hayakawa, S.: Enantiomer-selective ultraviolet photolysis of temperature-controlled protonated tryptophan on a chiral crown ether in the gas phase. Chem. Phys. Lett. 610, 228–233 (2014). https://doi.org/10.1016/j.cplett.2014.07.045

3 Chirality Effects in Jet-Cooled Cyclic Dipeptides

87

82. Matile, S., Berova, N., Nakanishi, K., Novkova, S., Philipova, I., Blagoev, B.: Porphyrinspowerful chromophores for structural studies by exciton-coupled circular-dichroism. J. Am. Chem. Soc. 117(26), 7021–7022 (1995). https://doi.org/10.1021/ja00131a033 83. Hong, A., Choi, C.M., Eun, H.J., Jeong, C., Heo, J., Kim, N.J.: Conformation-specific circular dichroism spectroscopy of cold. Isolated Chiral Molecules. Angewandte Chemie-International Edition 53(30), 7805–7808 (2014). https://doi.org/10.1002/anie.201403916

Chapter 4

Hydrogen Bond Networks Formed by Several Dozens to Hundreds of Molecules in the Gas Phase Asuka Fujii

Infrared spectra of neutral and protonated clusters of water, methanol, and ammonia are observed in the size range of several dozens to hundreds. Convergence of the spectra with increasing cluster size is discussed

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_4

89

90

A. Fujii

Abstract Infrared spectroscopy was applied to neutral and protonated clusters of fundamental protic molecules, water, methanol (MeOH), and ammonia, in the size range of dozens to hundreds. For neutral (H2 O)n and (MeOH)n clusters, a phenol molecule was introduced as a chromophore for the resonant multiphoton ionization detection, and the infrared-ultraviolet double-resonance scheme with mass spectrometry was used to perform moderately size-selective infrared spectroscopy in the OH stretch region. Infrared spectra of neutral (NH3 )n clusters in the NH stretch region were also measured with the same scheme, but without addition of a chromophore. For protonated clusters, H+ (H2 O)n , H+ (MeOH)n , and H+ (NH3 )n , infrared dissociation spectroscopy was applied to observe definitely size-selective spectra of these clusters in the large size region. The size dependence of the hydrogen bond networks of the neutral and protonated clusters is discussed on the basis of the observed spectra. Convergence of the spectra of the protonated clusters to those of the neutral clusters is seen with increasing size, and the convergence size tells us the maximum size range of the influence of the excess proton to the surrounding hydrogen bond network. Convergence of the spectra of large-sized clusters to those of the bulk is also seen, and the dependence of the convergence size on the hydrogen bond network is discussed. Keywords Hydrogen-bonded clusters · Infrared spectroscopy · Water · Methanol · Ammonia

4.1 Introduction Hydrogen bond is a strongly directional intermolecular interaction. It is basically formed between a pair of proton donor (OH or NH) and acceptor (lone pair or π electrons). Therefore, structures of hydrogen bond networks depend heavily on the hydrogen bond coordination property of their constituent molecule. A variety of different types of hydrogen bond networks, from simple one-dimensional chains to complicated three-dimensional cages, have been observed in hydrogen-bonded crystals [1, 2]. Total number of molecules (size, n) in the system can be a crucial factor of restriction on possible hydrogen bond networks. When the number of molecules in the system is finite, that is, in molecular clusters in the gas phase, many molecules are located on the surface, in which the molecules cannot fully satisfy their hydrogen bond coordination ability. Especially in small-sized clusters, all the molecules are necessarily on the surface, and their hydrogen bond networks tend to be largely different from those seen in bulk [3, 4]. With increase of size, however, the “interior” should be formed in the cluster, and the hydrogen bond network of this interior moiety is expected to finally converge to that of the bulk (or bulk-like structures). Such a convergence process from the gas phase cluster to the bulk may also depend strongly A. Fujii (B) Department of Chemistry, Graduate School of Science, Tohoku University, Sendai 980-8578, Japan e-mail: [email protected]

4 Hydrogen Bond Networks Formed by Several …

91

on the hydrogen coordination property of the constituent molecule; it is reasonable to expect that convergence to a very simple network, like one-dimensional chain, should be faster than that to a complicated three-dimensional cage structure. Therefore, hydrogen bond structures of gas phase clusters in the size range of several dozens to hundreds are of great interest, since the interior formation and the convergence of the interior moiety to the bulk are expected in this size region [5–44]. To explore hydrogen bond network structures of clusters, infrared (IR) spectroscopy would be the most powerful technique. This is because hydrogen bond structures are reflected well in frequency shifts of their OH and NH stretching vibrations. In the last two decades, rapid development of quantum chemical computations and computational resources enables us to precisely predict stable cluster structures. Direct comparison between observed IR spectrum and simulated ones based on energy-optimized structures is now a standard technique to determine structures of hydrogen-bonded clusters. Number of possible stable isomer structures, however, rapidly increases with increasing cluster size, and survey of the potential landscape is practically difficult in the size range of several dozens to hundreds. Moreover, preparation of a single isomer structure in experimental observation is also unrealistic because of existence of densely degenerated low-energy isomers. Molecular dynamics simulations can handle large-sized clusters and will be a promising approach to analyze large-sized clusters. Their precision is, however, still limited in comparison with high-level quantum chemical computations. Therefore, analyses of IR spectra of large-sized clusters are forced to be qualitative at the present stage, and our focus is to catch the structural trend in the size development of the cluster. To this end, comparison between spectra of large-sized clusters and those of the bulk is very informative. In IR spectroscopic measurements of clusters in the size range of several dozens to hundreds, the most serious problem is selection of cluster size. For neutral clusters, there has been no universal technique to perform size-selective IR spectroscopy. Definitely size-selective measurements by using crossed beam scattering or infraredultraviolet double-resonance techniques are applicable only for clusters of less than ~10 molecules [10]. Therefore, average cluster size (n) control by tuning of the cluster source condition has been performed in IR spectroscopy of large-sized hydrogenbonded neutral clusters, such as water and ammonia [9, 12, 13, 36–39]. On the other hand, size of charged clusters (actually protonated clusters for many protic molecules) can be freely selected by mass spectrometric techniques, and combination with IR photodissociation spectroscopy enables us to measure definitely size-selected IR spectra at any size, in principle [17–29, 31, 33–35, 40]. It has been well known that an excess proton largely changes hydrogen bond network structures of surrounding molecules because of the different hydrogen bond coordination property of the protonated site and significant enhancement of hydrogen bond strength [22, 34]. With increasing cluster size, however, the influence of the excess proton is diluted, and the gross feature of the hydrogen bond network should finally converge to that of the neutral bulk system. When the influence of the excess proton is enough diluted, the IR spectrum of the protonated cluster becomes essentially identical with that of the neutral cluster of the same cluster size. Systematic comparison between IR

92

A. Fujii

spectra of neutral and protonated clusters, therefore, would tell us the maximum size range of the influence of the excess proton on surrounding molecules [34, 35, 40]. Once the size of the protonated cluster goes over this convergence size to the neutral, IR spectra of protonated large clusters can be used to probe the convergence of the hydrogen bond network of the clusters to that of the bulk. In this chapter, we review our studies on the IR spectral convergence of hydrogenbonded clusters with increasing size. Neutral and protonated clusters of the simple protic molecules, water, methanol (MeOH), and ammonia, are focused. In Sect. 4.2, the experimental techniques to measure size-selective IR spectra of large-sized neutral and protonated clusters are introduced. In Sects. 4.3–4.5, IR spectra of neutral and protonated clusters in the range of several dozens to hundreds are reviewed. Two spectral convergence processes, protonated clusters to neutral clusters and clusters to the bulk, are discussed based on the size dependence of the observed IR spectra and the comparison with the spectra of the bulk.

4.2 Experimental Techniques for Size-Selective IR Spectroscopy of Large-Sized Clusters 4.2.1 Neutral Clusters The most successful IR spectroscopic technique for neutral clusters in the gas phase is infrared-ultraviolet (IR-UV) double-resonance technique described in Sect. 1.3. 2 of this book [45]. Its scheme is shown in Fig. 4.1. A pulsed UV laser, which is resonant on the vibronic transition of the cluster, is introduced, and the resulting resonant two-photon ionization signal is monitored as a measure of the population at the neutral ground vibrational level. Prior to the UV laser pulse (typically ~50 ns), a pulsed IR laser is introduced and its frequency is scanned. When the IR frequency is resonant on a vibrational transition of the cluster, the decrease of the ground-level population is detected as a depletion of the ionization signal. Size-selection by IR-UV double-resonance spectroscopy is essentially achieved by tuning the UV frequency to be resonant with the electronic transition of the cluster. Therefore, a sharp vibronic band structure is requested for the cluster to apply this technique. Neat clusters of water and methanol lack a proper electronic transition in the UV region. Therefore, introduction of a chromophore molecule such as benzene and phenol in the cluster enables us to apply IR-UV double-resonance spectroscopy to these neutral clusters [46, 47]. The chromophore molecule should be almost inert (benzene) or compatible (phenol) to the hydrogen bond network of the cluster. However, in large-sized clusters, in which the number of water/methanol molecules is larger than ~10, the electronic transition of the aromatic chromophore in the cluster is totally broadened, and the resonant frequency tuning becomes helpless to select the cluster size. Therefore, we used the scheme of IR-UV double-resonance spectroscopy with the phenol chromophore but we relied only on mass spectrometry to select the cluster size

4 Hydrogen Bond Networks Formed by Several …

93

Fig. 4.1 Scheme of infrared-ultraviolet (IR-UV) double-resonance spectroscopy. The produced ions are analyzed by a time-of-flight mass spectrometer

[30, 32, 34, 35]. The broadened S1 –S0 electronic transition of the phenol-(X)n clusters (X = H2 O or MeOH) was excited to ionize all large-sized clusters simultaneously. The produced [phenol-(X)n ]+ ions were mass-selected with a time-of-flight mass spectrometer and were detected as a measure of the population of phenol-(X)n+n . Here, n is the number of molecules evaporated upon ionization, that is, the size uncertainty of the spectral carrier. It is estimated to be 0 ≤ n ≤ ~6 and ~10 for X = H2 O and MeOH, respectively, by the energetics calculations of the ionization [30, 35]. This size uncertainty is acceptable enough to observe size dependence of hydrogen bond networks in the large-sized region. When an IR transition occurs in phenol-(X)n+n , the monitored [phenol-(X)n ]+ signal intensity decreases because of the vibrational predissociation of the clusters. As for neutral ammonia clusters (NH3 )n , they show the broadened S1 –S0 electronic absorption in the UV region, and we applied the same IR-UV spectroscopic scheme without an additional chromophore [40]. The size uncertainty was also estimated to be 0 ≤ n ≤ ~10.

4.2.2 Protonated Clusters Protonated clusters have an excess charge, and their sizes can be definitely selected by using mass spectrometric techniques. IR spectroscopy of size-selected protonated clusters was performed by photodissociation spectroscopy combined with multi-

94

A. Fujii

stage mass spectrometry [34, 48]. In the present study, protonated clusters, H+ (H2 O)n , H+ (MeOH)n , and H+ (NH3 )n , were generated by a supersonic jet expansion crossed by an electron beam at the collisional region. Protonated clusters of interest were precisely size-selected by the first quadrupole mass spectrometer and were introduced into an octopole ion guide. In the ion guide, the protonated clusters were irradiated by the IR light. IR spectra were measured by monitoring the fragment or parent ion intensity selectively detected by the second quadrupole mass spectrometer while scanning the IR frequency. Vibrational predissociation causes enhancement of the fragment ion intensity and depletion of the parent ion intensity when the IR frequency is resonant on a vibrational transition. For H+ (H2 O)n and H+ (MeOH)n , depletion of the parent cluster ion was monitored. This enables us to avoid the interference by the secondary IR dissociation of fragment ions, which is prominent in large-sized clusters. For H+ (NH3 )n , detection of depletion of the parent ion was utilized only in the relatively large-sized clusters (n ≥ 50) and the fragment ions were monitored in the smaller-sized clusters (n ≤ 40) since the IR absorption intensity of the NH stretching vibrations in H+ (NH3 )n is much weaker than that of the OH stretching vibrations in H+ (H2 O)n and H+ (MeOH)n , and depletion of the parent ion intensity could not be detected in n ≤ 40.

4.3 IR Spectroscopy of Large-Sized Water and Protonated Water Clusters [31, 32] Figure 4.2a shows the observed IR spectra of size-selected phenol-(H2 O)n (n = 19–49) in the OH stretching region [32]. In these clusters, the phenol molecule is introduced as a chromophore for the ionization detection, and the hydroxyl group of phenol is expected to be compatible to a water molecule at the cluster surface, as schematically shown in the inset of Fig. 4.1. Therefore, the spectra of phenol(H2 O)n can be approximately regarded as those of (H2 O)n+1 . The Na-doped neutral water cluster study has supported this approximation at n = 20 [14]. Each observed spectrum includes the contribution of larger-sized clusters of phenol-(H2 O)n+n (0 ≤ n ≤ ~6). The broad absorption below 3600 cm−1 is attributed to hydrogen-bonded OH stretching vibrations, and the sharp band near 3700 cm−1 is due to free (dangling) OH stretching vibrations. In the spectra, the gap around 3500 cm−1 was caused by the depletion of the IR laser power due to the self-absorption of the nonlinear optical crystal used to generate the IR light. With these spectra, we can infer size-dependence of hydrogen bond structures of the water clusters in terms of coordination numbers of water molecules. Though the hydrogen bond network development process in this size region has already been partly demonstrated in the average-size-controlled cluster studies [8, 9, 11–13], the much better size selectivity in the present measurements enables for more rigorous discussion on the size dependence.

4 Hydrogen Bond Networks Formed by Several …

95

Fig. 4.2 IR spectra of a phenol-(H2 O)n and b H+ (H2 O)n in the OH stretching region. The spectra of phenol-(H2 O)n are approximately regarded as those of neat (H2 O)n+1 . The gap in the 3500 cm−1 region is due to the reduction of the IR light. The red trace overlapped with the spectrum of H+ (H2 O)50 is the reproduction of the spectrum of phenol-(H2 O)49

The IR spectra of phenol-(H2 O)n in Fig. 4.2a show that the intensity of the free OH band relative to that of the hydrogen-bonded OH band gradually decreases with increasing n. Because four-coordinated water molecules have no free OH, the decrease of the free OH band intensity is a signature of the increase of fourcoordinated water molecules. According to this observation, spectral evidence of the four-coordinated water molecules is also expected to appear in the hydrogen-bonded OH stretch region. In the spectra, the relative intensity of the ~3350 cm−1 region increases with increasing n. This is accompanied by the suppression of the relative band intensities in the ~3100 and ~3500 cm−1 regions. While the ~3350 cm−1 region is almost flat in the spectrum of n = 19, a clear hump is seen in the spectra of n = 29. When a cluster has “interior” molecules, these molecules should be fully solvated, that is, four-coordinated in water clusters. Therefore, the growth of the hump at the ~3350 cm−1 region is regarded as a signature of the interior growth in the water clusters, and it begins at least in the size region of n = 20–30. Similar trends have also been reported for the average-size-controlled water clusters [8, 9, 11–13]. Figure 4.2b shows IR photodissociation spectra of the precisely size-selected H+ (H2 O)n (n = 20–50) [31]. The red trace overlapped in the spectrum of n = 50 is the spectrum of phenol-(H2 O)49 for comparison. As in the case of the neutral water clusters, the relative intensity of the free OH band decreases with increasing cluster size, meaning that four-coordinated water molecules become dominant in the larger clusters. On the other hand, the spectral evolution of the hydrogen-bonded OH bands in n < ~50 is different from that in the neutral clusters, and the emergent process of

96

A. Fujii

the four-coordinated water band is not clear. This is due to the excess proton, which largely changes the surrounding network structure. However, the spectra of phenol(H2 O)49 and H+ (H2 O)50 are nearly identical, as demonstrated by the comparison in the figure. This means that the influence of the excess proton is diluted enough in n ≥ ~50. In other words, the influence of the excess proton on the hydrogen bond network is limited to surrounding ~50 water molecules at the maximum. The IR spectra of larger-sized H+ (H2 O)+n (50 ≤ n ≤ 221) are shown in Fig. 4.3a [31]. In this size region, hydrogen bond networks of the clusters are approximately the same as those of neutral (H2 O)n . The spectra show a gradual low-frequency shift of the maximum of the hydrogen-bonded OH band with increasing cluster size. This shift cannot simply be attributed to increase of four-coordinated water molecules. It implies that further structural evolution occurs, shifting the four-coordinated water band to lower frequency. To interpret this result, we compared the IR spectra of H+ (H2 O)n with those of liquid water, supercooled water, and hexagonal ice (Fig. 4.3b) Fig. 4.3 a IR spectra of H+ (H2 O)n (n = 50–221) in the OH stretching region. b IR spectra of liquid water (298 K), supercooled water (240 K), and hexagonal ice (235 K). These condensed phase spectra are reconstructed by the reported numerical data (Refs. [49, 50]). The dashed line is a guide to the eye

4 Hydrogen Bond Networks Formed by Several …

97

[49, 50]. The hydrogen-bonded OH band of the n ≤ ~60 clusters is similar to that of liquid water with respect to the absorption maximum. On the other hand, in the larger clusters, the maximum of the band is shifted to ~3300 cm−1 , approaching those of supercooled water and ice. Hexagonal ice has a hydrogen bond network consisting only of four-coordinated water molecules, and it has been supposed that supercooled water has partially ordered (crystalline) networks [51]. These results, then, suggest that H+ (H2 O)n forms liquid-like and disordered structures in the size range of n ≤ ~100, while they form at least partially ordered, crystalline-like structures in the interior of the larger sizes (n ≥ ~100). Here we note that even at n = ~200, the number of surface molecules in the cluster is larger than that in the interior. The surface is still influential to the interior and hinders the complete crystallization. However, this crystalline core formation in large H+ (H2 O)n clusters would finally lead to the bulky ice nanoparticle motifs. This has been suggested for large neutral clusters by electron diffraction [5, 6] and IR spectroscopic studies of the averagesize-controlled neutral clusters [12, 13]. After our results were published, Pradzynski et al. have reported IR spectra of size-selected Na-doped water clusters Na(H2 O)n (n = 85–475), and they have concluded that the onset of crystallization is about n = 275 and the characteristic band of ice around 3200 cm−1 is seen at n = 475 [14, 15]. While this conclusion essentially agrees with our findings, their observed size dependence shows an evident difference from our observation [14, 31]. The absorption maximum frequency of Na(H2 O)n (e.g., ~3400 cm−1 at n = 225) actually tends to be higher than that of H+ (H2 O)n (~3300 cm−1 at n = 221). If this result is simply interpreted, crystallization of H+ (H2 O)n occurs from a smaller size than Na(H2 O)n . The IR spectra of M (H2 O)n (M = La3+ , Ca2+ , Na+ , I− , and SO4 2− , n = ~250) have also been reported by Williams et al., and these spectra show that the hydrogen-bonded OH stretch band features remarkable changes by the ions [26]. Moreover, they also demonstrated that the onset of the crystallization of water clusters clearly delayed (n ~ 100) with the addition of a La3+ ion [27]. Dopant (ion) effects on the water network ordering should be different from each other. Furthermore, cluster conditions (e.g., temperature) should be considered in discussions of the crystallization process. Further studies will be expected to clarify the nature of the complicated water networks in the large water clusters. The free OH band of the H+ (H2 O)n also provides insight on the surface structure of the cluster. The free OH band shows a shift to high frequency in the size region of n < ~60. This is due to the dilution of the charge effect of the excess proton. In the n > ~60 region, the free OH frequency is almost constant (3699–3700 cm−1 ). The converged value can be compared with the free OH frequency of the surfaces of liquid water and ice. The free OH frequency of liquid water surface is precisely determined to be 3697 cm−1 by the recent sum frequency generation spectroscopic study [52]. According to the FTIR nanoparticle study, the free OH frequency of small ice particles (particle diameter is

Fig. 4.4 IR spectra of a phenol-(MeOH)n and b H+ (MeOH)n in the CH and OH stretching region. The spectra of phenol-(MeOH)n are approximately regarded as those of neat (MeOH)n+1 . The gap in the 3500 cm−1 region is due to the reduction of the IR light. Reproduced from Ref. [35] with permission from the PCCP Owner Societies

4 Hydrogen Bond Networks Formed by Several …

99

~3600 cm−1 ). Remarkable changes of these bands are not found with increasing n. Such size-independence of the spectra of phenol-(MeOH)n quite contrasts with the clear size dependence of the spectra of phenol-(H2 O)n in the same size range, shown in the previous subsection. The hydrogen-bonded OH stretch band position of phenol-(MeOH)n agrees with that of the smaller-sized neat (MeOH)n clusters (n = 5–9) measured by the crossed molecular beam or vacuum ultraviolet ionization techniques [10, 55–60]. This means that the basic hydrogen bond network motif of the clusters does not change in the size region of n = ~5–40. For neat (MeOH)n clusters, cyclic (single-ring) structures have been confirmed in n = 3–9. The hydrogen-bonded OH band shows a shift to low frequency with the ring size expansion, but the band position soon converges to ~3300 cm−1 at n = 5–6. The present spectra show that this single-ring motif is held in the large-sized methanol clusters at least up to n = ~40. In addition, a study on the decomposition of the OH stretch band has been performed, and it has been demonstrated that side-chain formation of the ring moiety is very scarce in this size range [35]. In the large cyclic structures, their hydrogen bond networks are essentially one-dimensional. Number of neighboring methanol molecules of a site (number of molecules within the first to nth shells of a site) is constant even though the ring size is largely expanded. The cooperative enhancement of a hydrogen bond is caused only by its neighboring hydrogen bonds. Therefore, it is reasonable that no further shift of the hydrogen-bonded OH band occurs with increasing cluster size, once the ring size goes over a threshold size (an effective size range of the cooperative enhancement), n = ~5. In addition, closed cyclic structures have no “surface” (dangling OH) in the hydrogen bond network, and therefore they are free from the influence of the surface. This contrasts with the case of water. Water prefers to form three-dimensional hydrogen bond networks, and saturation of number of neighboring molecules occurs at much larger sizes. Moreover, their three-dimensional networks are largely influenced by the surface moiety, which are predominant in small sizes. Here, we note on the suggestion by the recent theoretical studies [61, 62]. Dispersion-corrected density functional theory (DFT) computations have predicted that stacked double-ring structures, in which two independent hydrogen-bonded rings are bound by dispersion, can compete with single-ring structures in n ≥ ~15. Since the hydrogen-bonded OH stretch band does not show clear size dependence in n > ~5, it is difficult to distinguish these two structure types by IR spectroscopy. Coexistence of the double- and single-ring type structures is quite plausible in the observed size region. The IR spectra of size-selected H+ (MeOH)n (n = 10–50) are shown in Fig. 4.4b [35]. The hydrogen-bonded OH stretch band centered at 3300 cm−1 appears and the CH stretch bands partially overlap with the low-frequency tail of the OH band. No apparent free OH stretch band is seen. In contrast to the neutral clusters, remarkable size-dependence of the hydrogen-bonded OH stretch band shape is seen in the protonated clusters. The width of the OH band becomes obviously narrower with increasing n. At the origin of this band narrowing, there are two possibilities. One is evaporation cooling in larger-sized clusters (i.e., reduction of hot bands) and the other is the dilution of the effect of the excess proton (contribution of hydrogen-bonded

100

A. Fujii

OHs around the excess proton). The narrowing of the hydrogen-bonded OH band converges at n = 30. After this convergence, the spectral features of H+ (MeOH)n (n = 30–50) are almost identical with those of the neutral phenol-(MeOH)n clusters, as seen in Fig. 4.4. This fact indicates that the hydrogen bond network motif of H+ (MeOH)n (n = 30–50) can be approximately regarded as that of the neutral clusters. Hydrogen bond networks of methanol can be more complicated by the existence of the excess proton because the protonated methanol site prefers to be a double donor (DD) site, which is forbidden in neutral methanol. It has been demonstrated that the hydrogen bond networks of H+ (MeOH)n develop from linear chains to “bicyclic” structures which are composed of double rings, with increasing size [63–65]. A bicyclic structure contains one DD site (protonated site) and two threecoordinated sites (double-acceptor single-donor, AAD), but all of the other sites are two-coordinated sites (acceptor donor, AD), which are the components of ring structures. Because the relative weight of these three exceptional sites (DD and two AAD) can be negligible in large-sized clusters, it is reasonably expected that the IR spectra of H+ (MeOH)n finally converges to those of neutral (MeOH)n , as we observed in n ≥ 30. The agreement of the IR spectra of the protonated and neutral clusters means that AD sites (one-dimensional hydrogen bond network) are dominant in both the clusters, and the influence of the excess proton can be negligible. Convergence of the cluster IR spectra into those of the bulk is examined also for methanol. Here, we compare the observed IR spectrum of H+ (MeOH)50 , which is the largest size we observed, with the previously reported bulk spectra of methanol in the CH and OH stretching vibrational regions [66]. Bulk spectra of (a) liquid (300 K), (b) vitreous solid (93 K), and (c) crystal (93 K, α phase) methanol are reproduced in Fig. 4.5 with (d) spectrum of H+ (MeOH)50 [35]. The OH stretch band peak position of the cluster is quite similar to those of the vitreous solid and crystal though the size of the network in the cluster is still much smaller than those in the solid phase. This quite contrasts with the case of water, in which the convergence of the band position of the cluster (into that of the crystal spectrum) occurs in the much larger size region (n = ~475) [14, 26, 27, 31]. As discussed earlier, the essentially one-dimensional hydrogen bond network of methanol restricts the effective area of the cooperative enhancement among hydrogen bonds, and this causes the fast convergence of the OH stretch frequency with increase of cluster size. On the other hand, there is a remarkable difference between the spectra of liquid methanol and the cluster. The OH frequency of liquid is clearly higher than that of the cluster. This difference suggests that the major hydrogen bond network size (chain length) in liquid is much shorter than n = ~10. This notion agrees with the prediction of the molecular dynamics simulations [67–70].

4 Hydrogen Bond Networks Formed by Several …

101

Fig. 4.5 Comparison of the IR spectra of methanol in the condensed phases with that of H+ (MeOH)50 . a Liquid at 300 K, b vitreous solid at 93 K, c crystal at 93 K, and d H+ (MeOH)50 (a)–(c). Reprinted with permission from Ref. [66]. Copyright (1961), American Institute of Physics

4.5 IR Spectroscopy of Large-Sized Ammonia and Protonated Ammonia Clusters [40] Ammonia has three donating protons but has only one proton-accepting site (nonbonding orbital). Because of this imbalance between the donor and acceptor sites, the one-to-one relation between donor and acceptor in a hydrogen bond is not kept in crystalline ammonia [71], and a molecule is six-coordinated as schematically shown in Fig. 4.6a. An ammonia monomer has two normal modes of NH stretching vibrations; symmetric stretch (ν1 ) and doubly degenerated asymmetric stretch (ν3 ). The ν1 fundamental is strongly coupled with the overtone of the NH bending mode (2ν4 ) [36, 38, 72, 73]. The infrared (IR) transition intensity ratio between the ν1 (with coupled 2ν4 ) and ν3 bands is ~2.5:1 in the monomer [74]. In the crystalline structure, however, the IR intensity of the ν1 band is almost cancelled among molecules while the ν3 band intensity is enhanced by hydrogen bond formation. Therefore, the ν3 band is dominant over the ν1 (+2ν4 ) band in crystalline ammonia [37, 75–80]. This intensity alternation is the spectral signature of the compact multi-ring hydrogen bond network structure in crystalline ammonia. In other words, the decrease of the ν1 band intensity relative to the ν3 band indicates the progress of the crystallization process of ammonia [38]. Figure 4.7 shows moderately size-selected IR spectra of (NH3 )n+n (n = ~5 to ~80) in the NH stretch region [40]. The effective size uncertainty of the spectra is estimated to be 0 ≤ n ≤ ~10. In the observed NH stretch region, the 2ν4 , ν1 , and ν3

102

A. Fujii

Fig. 4.6 a Multiple-ring motif with a six-coordinated ammonia molecule. b Radial chain motif seen in small-sized protonated ammonia clusters. Reproduced from Ref. [40] with permission from the PCCP Owner Societies Fig. 4.7 IR spectra of (NH3 )n in the NH stretching region. Reproduced from Ref. [40] with permission from the PCCP Owner Societies. Note that the band assignments are partly revised from the original figure in Ref. [40] (see text)

4 Hydrogen Bond Networks Formed by Several …

103

bands are seen. The assignments of the ν1 and 2ν4 bands have been controversial [36, 38, 40, 42, 72–74]. In the monomer, the 2ν4 frequency is lower than ν1 , but the reverse of the energy order is expected to occur in large-sized clusters since the enhancement of the hydrogen bond strength lowers the ν1 frequency while it raises the ν4 frequency. The reverse has been predicted to occur in the size range n = 10–100 on the basis of the empirical potential computations, but no experimental evidence has been found [36]. After we published the spectra in Fig. 4.7 [40], we analyzed the anharmonic coupling of small-sized (NH3 )n (n ≤ 5), and we found that the energy order reverse actually occurs at much smaller size, n = 3–4, without obvious frequency shifts [72]. Therefore, in this time, we re-assign the lower frequency bands at 3200–3250 cm−1 to ν1 , and the band at ~3300 cm−1 to 2ν4 . The highest frequency band is uniquely attributed to the ν3 band. The suppression of the intensities of the ν1 (and coupling 2ν4 bands) is a spectral sign of the multiple-ring motif formation leading to the crystalline structure. The present measurements indicate that the multiple-ring motif is rapidly formed up to n = ~20, and the progress (slope of the ν3 band relative intensity increases) is rather moderate in the larger region up to n = 80. This size dependence agrees well with that observed in the previous studies on the average-size-controlled clusters in a molecular beam and in He droplets [36, 38]. The initial size of the crystallization (n = ~20) is much smaller than the case of water, which requests at least ~100 molecules to show the sign of the partially crystalline structure formation. However, as shown in the comparison with the spectra of the condensed phase ammonia ((a) amorphous film and (b) crystalline film) and (c) much larger-sized clusters (n = 104 ) in He droplets in Fig. 4.8 [38, 76, 77], the spectrum of n = 80 (spectrum (d)) is still clearly different from that of crystalline ammonia but is rather similar to amorphous solid Fig. 4.8 Comparison of the IR spectra of ammonia in the condensed phases with those of clusters. a Amorphous film deposited at 20K (Ref. [76]). b Crystalline film deposited at 90K (Ref. [79]). c (NH3 )4 in He droplets at 0.38K (Ref. [38]). d (NH3 )80 . Spectra (a)–(c) are taken from Ref. [38]. Copyright (2008), American Institute of Physics

104

A. Fujii

(film). The almost complete convergence to the crystal spectrum has been reported to occur in the average-size-controlled cluster of n = 104 in He droplets and in an aggregation cell [37, 38]. Figure 4.9 shows IR spectra of H+ (NH3 )n (n = 8–100) in the NH stretch region [40]. In the case of these spectra of the protonated clusters, the size selection is rigorous (n = 0). The most significant feature in the spectra is the intense band of the NH stretching vibrations of the NH4 + ion core moiety, which appears at Fig. 4.9 IR spectra of H+ (NH3 )n the NH stretching region. The intense band at around 2800 cm−1 is the NH stretch of the NH4 + ion core. Reproduced from Ref. [40] with permission from the PCCP Owner Societies

4 Hydrogen Bond Networks Formed by Several …

105

~2800 cm−1 [81, 82]. The ν1 , 2ν4 , and ν3 bands are also seen in the 3200–3400 cm−1 region, being similar to those of the neutral clusters. Addition of an excess proton to neutral ammonia clusters largely changes their hydrogen bond networks of ammonia. The protonated site becomes ammonium ion, NH4 + , and strongly prefers being four-coordinated. Lee et al. have reported the IR spectra of H+ (NH3 )n (n = 2–11), and they have demonstrated that the solvation of the ammonium ion core is completed at n = 5 to form the first shell (i.e., NH4 + (NH3 )4 ) and extension of the four hydrogen-bonded chains from the first shell occurs to form the second shell in the size region of n = 6–9 [81, 82]. A schematic representation of such a radial chain structure is shown in Fig. 4.6b. The mid-IR spectra (the NH bending vibrations and the excess proton vibration) of n = 2–8 and the theoretical studies have also supported this picture [82–87]. No ring motifs are included in these radial chain-type hydrogen bond networks of the protonated clusters, and this contrasts with the networks of the neutral ammonia clusters, which prefer having ring motifs even in the very small sizes. The spectra of n = 8 and 10 are very different from those of the neutral clusters at the corresponding sizes, and this reflects the different H-bond network motif. It is, however, expected that structures of protonated clusters finally converge to those of neutral clusters with increasing cluster size since the influence of the excess proton is diluted enough in large clusters. In the radial chain structures (Fig. 4.6b), the lower ν3 intensity relative to ν1 + 2ν4 is their spectral signature. This is because no ring motif is involved and the intensity cancellation does not occur in the ν1 (and 2ν4 ) mode. In the size region of n = 12–15, the ν1 + 2ν4 intensity rapidly decreases and the peak height of the ν3 band is rather higher in n ≥ 18. This is an indication of the ring motif formation, which finally leads to the crystallization. In other words, additional NH3 molecules begin to connect the radial chains in the hydrogen bond network development at this size region. In n ≥ 30, as seen in Fig. 4.9, the ν1 –ν3 region of the spectra is quite similar to that of the neutral clusters at the corresponding sizes. The NH stretch band of the NH4 + ion core moiety also becomes weaker with increasing n, and it almost disappears at n = 40. The spectra are almost identical to those of the neutral clusters in n ≥ 40, and this means that the influence of the excess proton is diluted enough and the hydrogen bond networks can be approximately regarded as those of neutral clusters.

4.6 Conclusions In this chapter, we have discussed two spectral convergences of hydrogen-bonded clusters with increase of cluster size. One is the dilution process of the influence of the excess proton on the hydrogen bond networks of water, methanol, and ammonia. The excess proton changes the structure and strength of the surrounding hydrogen bond networks. To probe the size range of such influence of the excess proton, we have performed the comparison between the IR spectra of the large-sized neutral and protonated clusters of these molecules in the OH/NH stretch regions, which are sensitive to hydrogen bonds. The IR spectra of H+ (H2 O)n become almost identical to

106

A. Fujii

those of (H2 O)n in n ≥ ~50 and H+ (MeOH)n also shows such spectral convergence to the corresponding neutral at n = ~30. The convergence of H+ (NH3 )n to (NH3 )n occurs at n = ~40 and this value is just in between the convergence sizes of methanol and water. Since the spectral convergence of the protonated cluster to the neutral occurs at the similar size regions in these typical protic molecules, it would be a general trend that the influence of an excess proton on hydrogen bond networks is effectively limited to surrounding 30–50 molecules. In hydration of higher charged ions, the range of the influence should be wider because of much stronger interactions with surrounding molecules. Structural influence over 100 water molecules has been recently reported for La3+ and Fe(CN)46− [27, 28]. The other convergence is hydrogen bond networks of large-sized clusters to the crystalline structure. The spectra of (H2 O)n (and their equivalents) show the beginning of the shift to low frequency of the hydrogen-bonded OH stretch band of fourcoordinated water sites at n = ~100. This is interpreted as the initial sign of the interior crystallization. However, the completion of the interior crystallization (agreement of the peak position of the hydrogen-bonded OH band) occurs at much larger size (n = ~475) [14]. On the other hand, the hydrogen-bonded OH stretch band position of (MeOH)n converges to that of amorphous and crystalline solids at n = ~5, and no obvious change of the hydrogen-bonded OH band is seen up to n = 50. In (NH3 )n , the initial sign of the crystallization (multi-ring structure formation) is seen in relatively small size (n = ~20). Its progress, however, seems to be rather moderate up to n = 80, and the completion of the crystallization may occur at the quite large size (n = ~104 ) [38]. These three examples demonstrate that the convergence process of each species is quite different. The hydrogen bond strength and coordination property of each species should govern the convergence process to the bulk. Water can be four-coordinated, and therefore, its hydrogen bond networks cannot be free from the influence of the surface (presence of dangling OHs at the surface). To be equivalent to the bulk, the interior of the cluster should be away from the surface. One hundred molecules would be the minimum number to separate the interior enough from the surface. Methanol prefers to be two-coordinated (acceptor–donor), and its hydrogen bond networks become essentially one-dimensional (infinite linear chains in the crystal and large closed rings in clusters). Such networks have no surface (dangling OHs). In addition, the number of neighboring molecules in the network is limited, and this restricts the range of the cooperative enhancement of the hydrogen bond strength. Therefore, the spectral convergence of the hydrogen bond networks to the bulk-like condition occurs at much smaller size in methanol. In ammonia, the strength of hydrogen bonds is much weaker than that in water and methanol. Moreover, the one-to-one relation between donor and acceptor in a hydrogen bond is not kept because of the imbalance between the numbers of donor and acceptor sites. Therefore, though three-dimensional cage structures are formed, the influence of the surface would be much less in (NH3 )n . This is an advantage to have the initial step to the crystallization at the small size (n = ~20). However, weak hydrogen bonds (and missing of the one-to-one relation) are easy to disorder, and this might be one of the reasons of the moderate progress of the crystallization in the clusters of ammonia.

4 Hydrogen Bond Networks Formed by Several …

107

Acknowledgments The authors acknowledge the essential contribution of the following collaborators: Prof. Naohiko Mikami, Dr. Kenta Mizuse, Dr. Marusu Katada, Dr. Ryunosuke Shishido, Mr. Toru Hamashima, and Mr. Tomohiro Kobayashi.

References 1. Jeffrey, G.A.: An Introduction to Hydrogen Bonding. Oxford University Press, New York (1997) 2. Maréchal, Y.: The Hydrogen Bond and the Water Molecule. Elsevier, Amsterdam (2007) 3. Hobza, P., Müller-Dethlefs, K.: Non-covalent Interactions: Theory and Experiment. Royal Society of Chemistry, Cambridge (2010) 4. Ludwig, R.: Water: from clusters to the bulk. Angew. Chem. Int. Ed. 40, 1809–1827 (2001) 5. Torchet, G., Schwartz, P., Farges, J., de Feraudy, M.F., Raoult, B.: Structure of solid water clusters formed in a free jet expansion. J. Chem. Phys. 79, 6196–6202 (1983) 6. Gartmann, T.E., Hartweg, S., Ban, L., Chasovskikh, E., Yoder, B.L., Signorell, R.: Electron scattering in large water clusters from photoelectron imaging with high harmonic radiation. Phys. Chem. Chem. Phys. 20, 16364–16371 (2018) 7. Coker, D.F., Miller, R.E., Watts, R.O.: The infrared predissociation spectra of water clusters. J. Chem. Phys. 82, 3554–3562 (1985) 8. Page, R.H., Vernon, M.F., Shen, Y.R., Lee, Y.T.: Infrared vibrational predissociation spectra of large water clusters. Chem. Phys. Lett. 141, 1–6 (1987) 9. Devlin, J.P., Joyce, C., Buch, V.: Structural relaxation rates near the ice surface: basis for separation of the surface and subsurface spectra. J. Phys. Chem. A 104, 1974–1977 (2000) 10. Buck, U., Huisken, F.: Infrared spectroscopy of size-selected water and methanol clusters. Chem. Rev. 100, 3863–3890 (2000) 11. Andersson, P., Steinbach, C., Buck, U.: Vibrational spectroscopy of large water clusters of known size. Euro. Phys. J. D 24, 53–56 (2003) 12. Buch, V., Bauerecker, S., Devlin, J.P., Buck, U., Kazimirski, J.K.: Solid water clusters in the size range of tens-thousands of H2 O: a combined computational/spectroscopic outlook. Int. Rev. Phys. Chem. 23, 375–433 (2004) 13. Steinbach, C., Andersson, P., Kazimirski, J.K., Buck, U., Buch, V., Beu, T.A.: Infrared predissociation spectroscopy of large water clusters: a unique probe of cluster surfaces. J. Phys. Chem. A 108, 6165–6174 (2004) 14. Pradzynski, C.C., Forck, R.M., Zeuch, T., Slavíˇcek, P., Buck, U.: A fully size-resolved perspective on the crystallization of water clusters. Science 337, 1529–1532 (2012) 15. Buck, U., Pradzynski, C.C., Zeuch, T., Dieterich, J.M., Hartke, B.: A size resolved investigation of large water clusters. Phys. Chem. Chem. Phys. 16, 6859–6871 (2014) 16. Pradzynski, C.C., Dierking, C.W., Zurheide, F., Forck, R.M., Buck, U., Zeuch, T., Xantheas, S.S.: Infrared detection of (H2 O)20 isomers of exceptional stability: a drop-like and a facesharing pentagonal prism cluster. Phys. Chem. Chem. Phys. 16, 26691–26696 (2014) 17. Miyazaki, M., Fujii, A., Ebata, T., Mikami, N.: Infrared spectroscopic evidence for protonated water clusters forming nanoscale cages. Science 304, 1134–1137 (2004) 18. Shin, J.-W., Hammer, N.I., Diken, E.G., Johnson, M.A., Walters, R.S., Jaeger, T.D., Duncan, M.A., Christie, R.A., Jordoan, K.D.: Infrared signature of structures associated with the H+ (H2 O)n (n = 6–27) clusters. Science 304, 1137–1140 (2004) 19. Wu, C.-C., Lin, C.-K., Chang, H.-C., Jiang, J.-C., Kuo, J.-L., Klein, M.L.: Protonated clathrate cages enclosing neutral water molecules: H+ (H2 O)21 and H+ (H2 O)28 . J. Chem. Phys. 122, 074315 (2005) 20. Douberly, G.E., Ricks, A.M., Duncan, M.A.: Infrared spectroscopy of perdeuterated protonated water clusters in the vicinity of the clathrate cage. J. Phys. Chem. A 113, 8449–8453 (2009)

108

A. Fujii

21. Fournier, J.A., Johson, C.J., Wolke, C.T., Weddle, G.H., Wolk, A.B., Johnson, M.A.: Vibrational spectral signature of the proton defect in the three-dimensional H+ (H2 O)21 cluster. Science 344, 1009–1012 (2014) 22. Fournier, J.A., Wolke, C.T., Johnson, M.A., Odbadrakh, T.T., Jordan, K.D., Kathmann, S.M., Xantheas, S.S.: Snapshots of proton accommodation at microscopic water surface: understanding the vibrational spectral signature of the charge defect in cryogenically cooled H+ (H2 O)n = 2–18 clusters. J. Phys. Chem. A 119, 9425–9440 (2015) 23. Bush, M.F., Saykally, R.J., Williams, E.R.: Infrared action spectra of Ca2+ (H2 O)11–69 exhibit spectral signatures for condensed-phase structures with increasing cluster size. J. Am. Chem. Soc. 130, 15482–15489 (2008) 24. O’Brien, J.T., Prell, J.S., Bush, M.F., Williams, E.R.: Sulfate ion patterns water at long distance. J. Am. Chem. Soc. 132, 8248–8249 (2010) 25. Prell, J.S., O’Brien, J.T., Williams, E.R.: Structural and electric field effects of ions in aqueous nanodrops. J. Am. Chem. Soc. 133, 4810–4818 (2011) 26. O’Brien, J.T., Williams, E.R.: Effects of ions on hydrogen-bonding water networks in large aqueous nanodrops. J. Am. Chem. Soc. 134, 10228–10236 (2012) 27. Cooper, R.J., DiTucci, M.J., Chang, T.M., Williams, E.R.: Delayed onset of crystallinity in ion-containing aqueous nanodrops. J. Am. Chem. Soc. 138, 96–99 (2016) 28. DiTucci, M.J., Williams, E.R.: Nanometer patterning of water by tetra anionic ferrocyanide stabilized in aqueous nanodrops. Chem. Sci. 8, 1391–1399 (2017) 29. Mizuse, K., Fujii, A., Mikami, N.: Long rage influence of an excess proton on the architecture of hydrogen bond network in large-sized water clusters. J. Chem. Phys. 126, 231101 (2007) 30. Mizuse, K., Hamashima, T., Fujii, A.: Infrared spectroscopy of phenol-(H2 O)n>10 : structural strains in hydrogen bond networks of neutral water clusters. J. Phys. Chem. A 113, 12134–12141 (2009) 31. Mizuse, K., Mikami, N., Fujii, A.: Infrared spectra and hydrogen-bonded network structures of large protonated water clusters H+ (H2 O)n (n = 20–200). Angew. Chem. Int. Ed. 49, 10119–10122 (2010) 32. Hamashima, T., Mizuse, K., Fujii, A.: Spectral signatures of four coordinated sites in water clusters: Infrared spectroscopy of phenol-(H2O)n (~20 ≤ n ≤ ~50). J. Phys. Chem. A 115, 620–625 (2011) 33. Mizuse, K., Fujii, A.: Infrared spectroscopy of large protonated water clusters H+ (H2 O)20–50 cooled by inert gas tagging. Chem. Phys. 419, 2–7 (2013) 34. Fujii, A., Mizuse, K.: Infrared spectroscopic studies on hydrogen-bonded water networks in the gas phase clusters. Int. Rev. Phys. Chem. 32, 266–307 (2013) 35. Kobayashi, T., Shishido, R., Mizuse, K., Fujii, A., Kuo, J.-L.: Structures of hydrogen bond networks formed by a few tens of methanol molecules in the gas phase: size-selective infrared spectroscopy of neutral and protonated methanol clusters. Phys. Chem. Chem. Phys. 15, 9523–9530 (2013) 36. Steinbach, C., Buck, U., Beu, T.A.: Infrared spectroscopy of large ammonia clusters as a function of size. J. Chem. Phys. 125, 133403 (2006) 37. Jetzki, M., Bonnamy, A., Signorell, R.: Vibrational delocalization in ammonia aerosol particles. J. Chem. Phys. 120, 11775–11784 (2004) 38. Slipchenko, M.N., Sartakov, B.G., Vilesov, A.F.: Evolution of the vibrational spectrum of ammonia from single molecule to bulk. J. Chem. Phys. 128, 134509 (2008) 39. Matsumoto, Y., Honma, K.: Crystalline structures in large ammonia clusters studied by IR cavity ringdown spectroscopy and a density functional theory calculation. Chem. Phys. Lett. 490, 9–13 (2010) 40. Katada, M., Shishido, R., Fujii, A.: Infrared spectroscopy of large-sized neutral and protonated ammonia clusters. Phys. Chem. Chem. Phys. 16, 7595–7601 (2014) 41. Beu, T.A., Buck, U.: Structure of ammonia clusters from n = 3 to 18. J. Chem. Phys. 114, 7848–7852 (2001) 42. Beu, T.A., Buck, U.: Vibrational spectra of ammonia clusters from n = 3 to 18. J. Chem. Phys. 114, 7853–7858 (2001)

4 Hydrogen Bond Networks Formed by Several …

109

43. Janeiro-Barral, P.E., Mella, M., Curotto, E.: Structure and energetics of ammonia clusters (NH3 )n (n = 3–20) investigated using a rigid-polarizable model derived from ab initio calculations. J. Phys. Chem. A 112, 2888–2898 (2008) 44. Malloum, A., Fifen, J.J., Conradie, J.: Structures and spectroscopy of the ammonia eicosamer, (NH3 )n = 20 . J. Chem. Phys. 149, 024304 (2018) 45. Page, R.H., Shen, Y.R., Lee, Y.T.: Infrared-ultraviolet double resonance studies of benzene molecules in a supersonic beam. J. Chem. Phys. 88, 5362–5376 (1988) 46. Pribble, R.N., Zwier, T.S.: Size-specific infrared spectra of benzene-(H2 O)N clusters (N = 1–7): Evidence for noncyclic (H2 O)N structures. Science 265, 75–79 (1994) 47. Watanabe, T., Ebata, T., Tanabe, S., Mikami, N.: Size-selective vibrational spectra of phenol(H2 O)n (n = 1–4) clusters observed by IR-UV double resonance and stimulated Raman-UV double resonance spectroscopies. J. Chem. Phys. 105, 408–419 (1996) 48. Mizuse, K., Fujii, A.: Infrared photodissociation spectroscopy of H+ (H2 O)6 ·Mm (M = Ne, Ar, Kr, Xe, H2 , N2 , and CH4 ): messenger-dependent balance between H3 O+ and H2 O5 + core isomers. Phys. Chem. Chem. Phys. 13, 7129–7135 (2011) 49. Zasetsky, A.Y., Khalizov, A.F., Earle, M.E., Sloan, J.J.: Frequency dependent complex refractive indices of supercooled liquid water and ice determined from aerosol extinction spectra. J. Phys. Chem. A 109, 2760–2764 (2005) 50. Bertie, J.E., Lan, Z.: Infrared intensities of liquids XX: the intensity of the OH stretching band of liquid water revisited, and the best current values of the optical constants of (H2 O)(l) at 25 °C between 15,000 and 1 cm−1 . Appl. Spectrosc. 50, 1047–1057 (1996) 51. Zasetsky, A.Y., Khalizov, A.F., Earle, M.E., Sloan, J.J.: Local order and dynamics in supercooled water: a study by IR spectroscopy and molecular dynamics simulations. J. Chem. Phys. 121, 6941–6947 (2004) 52. Yamaguchi, S.: Development of single-channel heterodyne-detected sum frequency generation spectroscopy and its application to the water/vapor interface. J. Chem. Phys. 143, 034202 (2015) 53. Rowland, B., Fisher, M., Devlin, J.P.: Probing icy surfaces with the dangling-OH-mode absorption: large ice clusters and microporous amorphous ice. J. Chem. Phys. 95, 1378–1384 (1991) 54. Cooper, R.J., O’Brien, J.T., Chang, T.M., Williams, E.R.: Structural and electrostatic effects at the surfaces of size- and charge-selected aqueous nanodrops. Chem. Sci. 8, 5201–5213 (2017) 55. Provencal, R.A., Paul, J.B., Roth, K., Chapo, C., Casaes, R.N., Saykally, R.J., Tschumper, G.S., Schaefer III, H.F.: Infrared cavity ring down spectroscopy of methanol clusters: single donor hydrogen bonding. J. Chem. Phys. 110, 4258–4267 (1999) 56. Buck, U., Siebers, J.-G., Wheatley, R.J.: Structure and vibrational spectra of methanol clusters from a new potential model. J. Chem. Phys. 108, 20–32 (1998) 57. Fu, H.B., Hu, Y.J., Bernstein, E.R.: IR + vacuum ultraviolet (118 nm) nonresonant ionization spectroscopy of methanol monomers and clusters: neutral cluster distribution and size-specific detection of the OH stretch vibrations. J. Chem. Phys. 124, 024302 (2006) 58. Larsen, R.W., Zielke, P., Suhm, M.A.: Hydrogen-bonded OH stretching modes of methanol clusters: a combined IR and Raman isotopomer study. J. Chem. Phys. 126, 194307 (2007) 59. Han, H.-L., Camacho, C., Witek, H.A., Lee, Y.-P.: Infrared absorption of methanol clusters (CH3 OH)n with n = 2–6 recorded with a time-of-flight mass spectrometer using infrared depletion and vacuum-ultraviolet ionization. J. Chem. Phys. 134, 144309 (2011) 60. Pribble, R.N., Hagemeister, F.C., Zwier, T.S.: Resonant ion-dip infrared spectroscopy of benzene-(methanol)m clusters with m = 1–6. J. Chem. Phys. 106, 2145–2157 (1997) 61. Kazachenko, S., Bulusu, S., Thakkar, A.J.: Methanol clusters (CH3 OH)n : putative global minimum-energy structures from model potentials and dispersion-corrected density functional theory. J. Chem. Phys. 138, 224303 (2013) 62. Hsu, P.J., Ho, K.-L., Lin, S.-H., Kuo, J.-L.: Exploration of hydrogen bond networks and potential energy surfaces of methanol clusters using a two-stage clustering algorithm. Phys. Chem. Chem. Phys. 19, 544–556 (2017) 63. Li, Y.-C., Hamashima, T., Yamazaki, R., Kobayashi, T., Suzuki, Y., Mizuse, K., Fujii, A., Kuo, J.-L.: Hydrogen-bonded ring closing and opening of protonated methanol clusters H+ (CH3 OH)n (n = 4–8) with the inert gas tagging. Phys. Chem. Chem. Phys. 17, 22042–22053 (2015)

110

A. Fujii

64. Fujii, A., Sugawara, N., Hsu, P-.J., Shimamori, T., Li, Y.-C., Hamashima, T., Kuo, J.-L.: Hydrogen bond network structures of protonated short-chain alcohol clusters. Phys. Chem. Chem. Phys. 20, 14971–14991 (2018) 65. Fifen, J.J., Nsangou, M., Dhaouadi, Z., Motapon, O., Jaidane, N.-E.: Structures of protonated methanol clusters and temperature effects. J. Chem. Phys. 138, 184301 (2013) 66. Falk, M., Whalley, E.: Infrared spectra of methanol and deuterated methanols in gas, liquid and solid phases. J. Chem. Phys. 34, 1554–1568 (1961) 67. Matsumoto, M., Gubbins, K.E.: Hydrogen bonding in liquid methanol. J. Chem. Phys. 93, 1981–1994 (1990) 68. Morrone, J.A., Tuckerman, M.E.: Ab initio molecular dynamics study of proton mobility in liquid methanol. J. Chem. Phys. 117, 4403–4413 (2002) 69. Pagliai, M., Cardini, G., Righini, R., Schettino, V.: Hydrogen bond dynamics in liquid methanol. J. Chem. Phys. 119, 6655–6662 (2003) 70. Handgraaf, J.-W., Meijer, E.J., Gaigeot, M.-P.: Density-functional theory-based molecular simulation study of liquid methanol. J. Chem. Phys. 121, 10111–10119 (2004) 71. Olovsson, I., Templeton, D.H.: X-ray study of solid ammonia. Acta Crystallogr. 12, 832–836 (1959) 72. Ho, K.-L., Lee, L.-Y., Katada, M., Fujii, A., Kuo, J.-L.: An ab initio anharmonic approach to study vibrational spectra of small ammonia clusters. Phys. Chem. Chem. Phys. 18, 30498–30506 (2016) 73. Slipchenko, M.N., Sartakov, B.G., Vilesov, A.F., Xantheas, S.S.: Study of NH stretching vibrations in small ammonia clusters by infrared spectroscopy in He droplets and ab initio calculations. J. Phys. Chem. A 111, 7460–7471 (2007) 74. Kleiner, I., Brown, L.R., Tarrago, G., Kou, Q.-L., Pcqué, N., Guelachvili, G., Dana, V., Madin, J.-Y.: Positions and intensities in the 2ν4 /ν1 /ν3 vibrational system of 14 NH3 near 3 μm. J. Mol. Spectrosc. 193, 46–71(1999) 75. Reding, F.P., Hornig, D.F.: The vibrational spectra of molecules and complex ions in crystals. V. Ammonia and deutero-ammonia. J. Chem. Phys. 19, 594–601(1951) 76. Wolff, H., Rollar, H.G., Wolff, E.: Infrared spectra and vapor pressure isotope effect of crystallized ammonia and its deuterium derivatives. J. Chem. Phys. 55, 1373–1378 (1971) 77. Binbrek, O.S., Anderson, A.: Raman spectra of molecular crystals. Ammonia and 3-deuteroammonia. Chem. Phys. Lett. 15, 421–427(1972) 78. Bromberg, A., Kimel, S., Ron, A.: Infrared spectrum of liquid and crystalline ammonia. Chem. Phys. Lett. 46, 262–266 (1977) 79. Sill, G., Fink, U., Ferraro, J.R.: Absorption coefficients of solid NH3 from 50 to 7000 cm−1 . J. Opt. Soc. Am. 70, 724–739 (1980) 80. Holt, J.S., Sadoskas, D., Pursell, C.J.: Infrared spectroscopy of the solid phases of ammonia. J. Chem. Phys. 120, 7153–7157 (2004) 81. Price, J.M., Crofton, M.W., Lee, Y.T.J.: Observation of internal rotation in the NH4 + (NH3 )4 ionic cluster. J. Chem. Phys. 91, 2749–2751 (1989) 82. Price, J.M., Crofton, M.W., Lee, Y.T.J.: Vibrational spectroscopy of the ammoniated ammonium ions NH4 + (NH3 )n (n = 1–0). J. Phys. Chem. 95, 2182–2195 (1991) 83. Ichihashi, M., Yamabe, J., Murai, K., Nonose, S., Hirao, K., Kondow, T.: Infrared spectroscopy of NH4 + (NH3 )n–1 (n = 6–9): Shell structures and collective ν 2 Vibrations. J. Phys. Chem. 100, 10050–10054 (1996) 84. Tono, K., Bito, K., Kondoh, H., Ohta, T., Tsukiyama, K.: Infrared photodissociation spectroscopy of protonated ammonia cluster ions, NH4 + (NH3 )n (n = 5–8), by using infrared free electron laser. J. Chem. Phys. 125, 224305 (2006) 85. Yang, Y., Kühn, O., Santambrogio, G., Goebbert, D.J., Asmis, K.R.: Vibrational signatures of hydrogen bonding in the protonated ammonia clusters NH 4 + (NH3 )1–4 . J. Chem. Phys. 129, 224302 (2008)

4 Hydrogen Bond Networks Formed by Several …

111

86. Malloum, A., Fifen, J.J., Dhaouadi, Z., Emgo, S.G.N., Jaidane, N.-E.: Structures and spectroscopy of protonated ammonia clusters at different temperatures. Phys. Chem. Chem. Phys. 18, 26827–26843 (2016) 87. Malloum, A., Fifen, J.J., Dhaouadi, Z., Emgo, S.G.N., Jaidane, N.-E.: Structures and spectroscopy of medium size protonated ammonia clusters at different temperatures, H+ (NH3 )10–16 . J. Chem. Phys. 146, 044305 (2017)

Chapter 5

Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes Yoshiya Inokuchi

UV and IR spectroscopy of benzo-crown ether complexes with metal ions are performed under cold, gas-phase conditions, providing detailed information on the geometric and electronic structures of the complexes

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_5

113

114

Y. Inokuchi

Abstract UV and IR spectra of benzo-crown ether complexes with metal ions are obtained under cold, gas-phase conditions with an electrospray ion source and cryogenically cooled ion traps. UV photodissociation spectroscopy is performed to obtain the UV spectra of the complexes. Conformer-specific UV and IR spectra are measured by UV-UV hole-burning and IR-UV double-resonance spectroscopy. Geometric and electronic structures of the crown ether complexes in addition to the number of the conformers are determined on the basis of the UV and IR spectra with the aid of quantum chemical calculations. We examine solvation effects on the encapsulation of metal ions by crown ethers by using microsolvated complexes, which contain several numbers of solvent molecules such as water and methanol. The relation between the function characteristic of crown ethers such as ion selectivity and the structures is discussed. Keywords Crown ethers · Metal ions · Host–guest chemistry

5.1 UV and IR Spectroscopy of Cold Metal Ion–Benzo-Crown Ether Complexes in the Gas Phase 5.1.1 Introduction Crown ethers (CEs) have played important roles in host–guest chemistry for the selective capture of guest species, especially metal cations [1–4]. In the initial stage of the CE studies by Pedersen, the complexation of CEs with metal salts [1, 4] was investigated using UV spectroscopy [1, 3]. However, broad features of the UV spectra due to solvent and temperature effects prevented providing further information about the conformation of the complexes. In the same paper, it was suggested that among the alkali metal cations K+ was the most strongly bonded to dibenzo18-crown-6 (DB18C6) by solubility measurements [1]. In the 1990s, metal ion–CE complexes were extensively investigated in the gas phase with mass spectrometric techniques [5–17], ion mobility methods [18, 19], IR spectroscopy [20–26], and UV spectroscopy [27–29]. In this section, we report spectroscopic studies on metal cation–CE complexes by gas-phase UV spectroscopy combined with cooling in an ion trap [30, 31]. Scheme 5.1 shows the CEs examined in this study: benzo-12-crown-4 (B12C4), benzo-15-crown5 (B15C5), dibenzo-15-crown-5 (DB15C5), benzo-18-crown-6 (B18C6), DB18C6, dibenzo-21-crown-7 (DB21C7), and dibenzo-24-crown-8 (DB24C8). We report UV photodissociation (UVPD) spectra of benzo-CE complexes with alkali metal ions. The CE complexes are produced by electrospray ion source and cooled down to ~10 K in a cold ion trap [30, 31]. Isolating and cooling the complexes in the gas Y. Inokuchi (B) Department of Chemistry, Graduate School of Science, Hiroshima University, Kagamiyama 1-3-1, Higashi-hiroshima, Hiroshima 739-8526, Japan e-mail: [email protected]

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

115

Scheme 5.1 Benzo-crown ethers examined in this study

phase greatly simplify the UV spectra and provide well-resolved vibronic bands. Furthermore, we measure conformer-specific IR and UV spectra via IR-UV doubleresonance and UV-UV hole-burning (HB) spectroscopy. We can use these spectra to distinguish peaks in the UV spectra that belong to different conformers. The use of density functional theory (DFT) allows us to determine the conformation of the complexes. The number, geometric structure, and electronic structure of the metal ion–CE complexes in the gas phase are discussed on the basis of the UV and IR spectroscopic results.

116

Y. Inokuchi

5.1.2 DB18C61 Figure 5.1 displays the UVPD spectra of (a) room temperature K+ ·DB18C6 and (b) K+ ·DB18C6 that is cooled in a cold ion trap [32]. The uncooled complex has a broad absorption around 36,300 cm−1 . The UV spectrum of the cooled K+ ·DB18C6 complex consists of many sharp bands, with the band origin clearly observed at 36,415 cm−1 . From the intensity and the vibrational frequency of hot bands, we estimate the temperature of the cooled complex to be ~10 K. This result indicates that UVPD spectroscopy in a cold ion trap provides rich spectroscopic information even for large CE complexes. Figure 5.2 shows the UVPD spectra of the cooled M+ ·DB18C6 (M = Li, Na, K, Rb, and Cs) complexes with the LIF spectrum of bare, jet-cooled DB18C6 [33]. All the M+ ·DB18C6 complexes show a blueshift of the absorption relative to uncomplexed DB18C6; the shift in the band origin in the UVPD spectra is an indication of the interaction strength [32]. The Li+ complex shows the most blueshifted band origin, indicating the strongest interaction. For the Rb+ and Cs+ complexes, the band origin gradually shifts to the red with respect to the position of the K+ complex. This indicates that the interaction between DB18C6 and the metal ion becomes progressively weaker from K+ to Cs+ . In order to distinguish vibronic bands due to different conformers, IR-UV doubleresonance spectra are measured using strong vibronic bands. Figure 5.3 shows the Fig. 5.1 The UVPD spectra of a uncooled K+ ·DB18C6 and b K+ ·DB18C6 that is cooled in the cold ion trap. The temperature of the cooled complex is estimated to be ~10 K. Reprinted with permission from [32] J. Am. Chem. Soc., 133, 12256–12263 (2011). Copyright 2011 American Chemical Society

1 This

section is based on the following article: [32] Inokuchi Y., Boyarkin O. V., Kusaka R., Haino T., Ebata T., Rizzo T. R.: UV and IR spectroscopic studies of cold alkali metal ion-crown ether complexes in the gas phase. J. Am. Chem. Soc. 133, 12256–12263 (2011). Copyright (2011) American Chemical Society.

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

117

Fig. 5.2 The UVPD spectra of the M+ ·DB18C6 (M = Li, Na, K, Rb, and Cs) complexes with the LIF spectrum of jet-cooled DB18C6 reported by Kusaka et al. Reprinted with permission from [32] J. Am. Chem. Soc., 133, 12256–12263 (2011). Copyright 2011 American Chemical Society Fig. 5.3 The IR-UV double-resonance spectra of the M+ ·DB18C6 (M = Li, Na, K, Rb, and Cs) complexes in the CH stretching region. Reprinted with permission from [32] J. Am. Chem. Soc., 133, 12256–12263 (2011). Copyright 2011 American Chemical Society

118

Y. Inokuchi

IR-UV spectra of the M+ ·DB18C6 complexes in the CH stretching region. For the Li+ and Na+ complexes, two kinds of IR spectra are found, indicating that there exist at least two conformers for the Li+ and Na+ complexes. The K+ , Rb+ , and Cs+ complexes show only one IR spectrum; there is only one stable isomer for each of them. The similarity of the IR spectra of the K+ , Rb+ , and Cs+ complexes implies that the conformation of DB18C6 in each of them will be similar. We determine the structure of the M+ ·DB18C6 complexes with the aid of DFT and time-dependent DFT (TD-DFT). Figure 5.4 displays the lowest-energy structures of the M+ ·DB18C6 complexes calculated at the M05-2X/6-31+G(d) level. For the Li+ and Na+ complexes, the two conformers identified in the UVPD and IR-UV spectra can be attributed to the most and the second most stable forms calculated at the M052X level, because these are substantially more stable than others. The K+ complex has a boat-type C2v conformer (K-a in Fig. 5.4e). This is the most stable conformer at calculation levels using the M05-2X functionals [32]. From the similarities among the UVPD spectra and the IR-UV spectra, we expect the Rb+ and Cs+ complexes to have a C2v structure similar to K-a. The assignment of the structure can be confirmed with TD-DFT calculations. The transition energy calculated for Rb-a (C2v ) and Csb (C2v ) is coincident with the position of the origin bands for the Rb+ and Cs+

Fig. 5.4 The structure of a, b Li+ ·DB18C6, c, d Na+ ·DB18C6, e K+ ·DB18C6, f Rb+ ·DB18C6, and g Cs+ ·DB18C6 complexes. Reprinted with permission from [32] J. Am. Chem. Soc., 133, 12256–12263 (2011). Copyright 2011 American Chemical Society

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

119

complexes [32]. These results reinforce the assignment of the structure of the Rb+ and Cs+ complexes to Rb-a and Cs-b [32]. In both Li-a and Li-b, DB18C6 distorts its ether ring to hold the Li+ ion, because the cavity size of DB18C6 is much larger than Li+ . This distortion is also seen also for Na-a and Na-b. However, the distance between the oxygen atoms and the Na+ ion is longer (2.39 Å) than that of Li-a and Li-b (2.09 Å), and the ether ring is not so highly distorted as in the Li+ complex. In K-a, Rb-a, and Cs-b, the ether ring of DB18C6 opens the most, adopting a boat (C2v ) conformation. The K+ ion is in the center of the cavity, while the Rb+ and Cs+ ions sit on top of the open cavity because their ion diameters are larger than the cavity size. Complexes that are highly distorted to hold metal ions open the ether ring largely upon S1 –S0 excitation. As a result, the encapsulation of metal ions in the cavity of DB18C6 could perhaps be controlled by the UV irradiation [32].

5.1.3 B15C5 and B18C62 UVPD spectra of the cooled M+ ·B15C5 (M = Li, Na, K, Rb, and Cs) complexes also show well-resolved vibronic bands [34]. Similar to the case of DB18C6, all the M+ ·B15C5 complexes show a blueshift of the UV absorption relative to uncomplexed B15C5. We determine the number of conformers and examine which peaks in the UV spectra belong to different conformers by IR-UV double-resonance spectroscopy. For the Li+ ·B15C5 complex, we observe two different IR-UV spectra, indicating that it can adopt two different conformations in the ion trap, whereas the Na+ ·B15C5 complex shows only one IR spectrum and hence exists as a single conformer. The K+ ·B15C5, Rb+ ·B15C5, and Cs+ ·B15C5 complexes each show three different IR-UV spectra, which confirms that they arise from three different conformers. In addition, these three IR-UV spectra show similar spectral features in the CH stretching region for the K+ , Rb+ , and Cs+ complexes, indicating a similar conformation. We determine the structure of these conformers with the aid of quantum chemical calculations, under the assumption that conformers showing similar IR-UV spectra in the CH stretching region have similar conformations of the crown ring [34]. All the M+ ·B18C6 complexes show sharp vibronic bands with a blueshift relative to uncomplexed B18C6 [34]. Conformer-specific IR spectra in the CH stretching region are measured also for the M+ ·B18C6 complexes via IR-UV double resonance. On this basis, we determine the number of conformers as 2, 3, 2, 1, and 1 for the Li+ , Na+ , K+ , Rb+ , and Cs+ complexes in the cold ion trap, respectively. The structure of the conformers is determined with quantum chemical calculations [34]. Since the structures of the M+ ·B15C5, M+ ·B18C6, and M+ ·DB18C6 complexes in the gas 2 This section is based on the following

article: [34] Inokuchi Y., Boyarkin O. V., Kusaka R., Haino T., Ebata T., Rizzo T. R.: Ion selectivity of crown ethers investigated by UV and IR spectroscopy in a cold ion trap. J. Phys. Chem. A 116, 4057–4068 (2012). Copyright (2012) American Chemical Society.

120

Y. Inokuchi

phase were determined, the effect of phenyl group substitution and ether ring size to the encapsulation of alkali metal ions was discussed on the basis of these conformers [32, 34, 35]. For smaller ions, the CEs deform the ether ring to decrease the distance and increase the interaction between the metal ions and oxygen atoms; the metal ions are completely surrounded by the ether ring. In the case of larger ions, the metal ions are too large to enter the crown cavity and are positioned on it, leaving one of its sides open for further solvation. Thermochemistry data calculated on the basis of the stable conformers of the complexes suggest that the ion selectivity of CEs is controlled primarily by the enthalpy change for the complex formation in solution, which depends strongly on the complex structure [34].

5.1.4 B12C43 UVPD spectroscopy is applied also to metal ion complexes with a smaller benzoCE, B12C4 [36]. Figure 5.5 shows the UVPD spectra of the M+ ·B12C4 (M = Li,

Fig. 5.5 The UVPD spectra of the M+ ·B12C4 (M = Li, Na, K, Rb, and Cs) complexes. Reprinted with permission from [36] J. Phys. Chem. A, 120, 6394–6401 (2016). Copyright 2016 American Chemical Society 3 This

section is based on the following article: [36] Inokuchi Y., Nakatsuma M., Kida M., Ebata T.: Conformation of alkali metal ion-benzo-12-crown-4 complexes investigated by UV photodissociation and UV-UV hole-burning spectroscopy. J. Phys. Chem. A 120, 6394–6401 (2016). Copyright (2016) American Chemical Society.

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

121

Na, K, Rb, and Cs) complexes in the 36,300–37,600 cm−1 region. Thanks to the cooling of the ions to ~10 K [37], all the complexes show sharp (~3 cm−1 FWHM) and well-resolved vibronic bands. The UV spectra of M+ ·B12C4 show a strong origin band at 36,673, 36,617, 36,543, 36,510, and 36,472 cm−1 for M = Li, Na, K, Rb, and Cs, respectively. The existence of strong origin bands in the UV spectra suggests a small geometry change upon the excitation from the S0 to S1 state for the M+ ·B12C4 complexes, which is in contrast to the results that several conformers of the M+ ·B15C5 and M+ ·B18C6 complexes show a very weak origin band and substantially extended low-frequency progressions [34]. UV-UV HB spectra enable us to discriminate vibronic bands due to a single conformer. Figure 5.6 shows the UV-UV HB spectra of the M+ ·B12C4 (M = Na, K, Rb, and Cs) complexes observed by fixing the probe laser frequency at each Fig. 5.6 The UV-UV HB and UVPD spectra of the M+ ·B12C4 complexes. Reprinted with permission from [36] J. Phys. Chem. A, 120, 6394–6401 (2016). Copyright 2016 American Chemical Society

122 Table 5.1 The number of conformers for the M+ ·B12C4, M+ ·B15C5, and M+ ·B18C6 (M = Li, Na, K, Rb, and Cs) complexes

Y. Inokuchi M

M+ ·B12C4

M+ ·B15C5a

M+ ·B18C6a

Li

2

2

2

Na

1

1

3

K

1

3

2

Rb

1

3

1

Cs

1

3

1

a Reference

[34]

origin band. All the strong bands in the UVPD spectra emerge in the UV-UV HB spectra for all the ion complexes. Hence, the M+ ·B12C4 (M = Na, K, Rb, and Cs) complexes each have only one conformation under the current cold conditions. For the Li+ ·B12C4 complex, it is not possible to measure a UV-UV HB spectrum because of much weaker intensity of the fragment Li+ ion. The vibronic structure in the UVPD spectra of the M+ ·B12C4 complexes is similar to each other with a small, gradual shift, suggesting that the conformation of the complexes is similar to each other. The most stable conformers of the M+ ·B12C4 (M = Li, Na, K, Rb, and Cs) complexes obtained by quantum chemical calculations have a quite similar conformation in the B12C4 part, which is well coincident to the experimental results described above [36]. Here, we discuss the relation between the conformation of M+ ·B12C4 and the ion selectivity of 12C4. In solutions, 12C4 has the maximum value of the equilibrium constant for the complex formation with Na+ among alkali metal ions [8, 38]. Table 5.1 shows the number of conformers found under cold conditions in the gas phase [34]. B15C5 and B18C6 have a flexibility to take different conformations for some metal ions. For the M+ ·B12C4 complexes, there is one conformer for M = Na to Cs and two conformers for M = Li. Since the size of the crown ring of B12C4 is too small to have the conformation flexibility, the entropic advantage by taking different conformations cannot be obtained for any alkali metal ion in the M+ ·B12C4 complexes [36].

5.1.5 DB15C54 The UVPD spectra of the M+ ·DB15C5 (M = Li, Na, K, Rb, and Cs) complexes in the 36,000–37,500 cm−1 region are shown in Fig. 5.7 [39]. The K+ , Rb+ , and Cs+ complexes show sharp and well-resolved vibronic bands (Fig. 5.7c–e). The UVPD spectrum of the K+ ·DB15C5 complex (Fig. 5.7c) has an extensive and intense progression in the 36,600–36,800 cm−1 region with an interval of 22 cm−1 , but with no strong ori4 This section is based on the following article: [39] Inokuchi Y., Kida M., Ebata T.: Geometric and electronic structures of dibenzo-15-crown-5 complexes with alkali metal ions studied by UV photodissociation and UV-UV hole-burning spectroscopy. J. Phys. Chem. A 121, 954–962 (2017). Copyright (2017) American Chemical Society.

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

123

Fig. 5.7 The UVPD spectra of the M+ ·DB15C5 (M = Li, Na, K, Rb, and Cs) complexes. Reprinted with permission from [39] J. Phys. Chem. A, 121, 954–962 (2017). Copyright 2017 American Chemical Society

gin band. In addition to this progression, a strong band also appears at 36,839 cm−1 , followed by a progression of several vibronic bands with an interval of ~26 cm−1 . These two progressions can be ascribed to different isomers or different electronic states, because the vibronic patterns are different between them. The Rb+ ·DB15C5 and Cs+ ·DB15C5 complexes show similar spectral features (Fig. 5.7d, e) to those of K+ ·DB15C5; a strong band is observed at 36,810 and 36,768 cm−1 , accompanied by several vibronic bands on the higher frequency side, and an extensive progression on the lower frequency side. The M+ ·DB15C5 (M = K, Rb, and Cs) complexes also have very weak, well-resolved bands in the 36,200–36,500 cm−1 region. The spectral pattern of the Na+ ·DB15C5 complex (Fig. 5.7b) seems to be similar to that of the K+ complex. The absorption above 36,600 cm−1 shows broader features with a few resolved bands, and the low-frequency component in the 36,300–36,600 cm−1 region becomes stronger than the case of the K+ complex. The UVPD spectrum of the Li+ complex (Fig. 5.7a) has congested features around 36,750 cm−1 with a smaller signal-to-noise ratio than that of the other complexes. For discriminating bands of different conformers in the UVPD spectra, we perform UV-UV HB spectroscopy to the K+ , Rb+ , and Cs+ complexes. Figure 5.8 displays the UV-UV HB spectra with the UVPD spectra for the M+ ·DB15C5 (M = K, Rb, and Cs) complexes. The positions of the probe laser are shown with arrows in the UVPD spectra in Fig. 5.8. The UV-UV HB spectra seem to be saturated and the rel-

124

Y. Inokuchi

Fig. 5.8 The UV-UV HB and UVPD spectra of the M+ ·DB15C5 (M = K, Rb, and Cs) complexes. The positions of the probe laser are shown with arrows in the UVPD spectra. Reprinted with permission from [39] J. Phys. Chem. A, 121, 954–962 (2017). Copyright 2017 American Chemical Society

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

125

ative intensity of the vibronic bands is different between the UVPD and the UV-UV HB spectra, but all the vibronic bands, which include the extensive progression on the lower frequency side, appear in the UV-UV HB spectra. For the K+ and Rb+ complexes, UV-UV HB spectra are also measured by probing one of the extensive progressions at 36,765 and 36,685 cm−1 , respectively (the UV-UV HB2 curves in Fig. 5.8a, b). The UV-UV HB2 spectra in Fig. 5.8a, b also show all the vibronic bands in the UVPD spectra. These HB results indicate that all the vibronic bands appearing in the 36,500–37,000 cm−1 region are due to a single isomer. Hence, the two progressions in the UVPD spectra can be assigned to the S1 –S0 and S2 –S0 transitions of the single isomer. For the K+ ·DB15C5 complex, we extend the measurement region of UV-UV HB spectroscopy up to 36,200 cm−1 (Fig. 5.8d) with the probe position of 36,839 cm−1 to see if the weak absorption in the 36,200–36,500 cm−1 region is due to the same conformer. Since the UV-UV HB spectrum in Fig. 5.8d does not show any noticeable depletion in the 36,200–36,500 cm−1 region, the weak components of the M+ ·DB15C5 (M = K, Rb, and Cs) complexes in the 36,200–36,500 region are assigned to a less stable, another isomer. The geometric structure of the M+ ·DB15C5 complexes is determined on the basis of the UVPD and UV-UV HB results with the aid of quantum chemical calculations [39]. Here, we discuss the electronic interaction in the M+ ·DB15C5 and M+ ·DB18C6 complexes [32, 39]. It was reported that the K+ ·DB18C6 complex has only one conformer under cold gas-phase conditions [32], the same as the case of K+ ·DB15C5, but the electronic spectrum is substantially different between K+ ·DB18C6 and K+ ·DB15C5. These complexes have two benzene chromophores, and they interact with each other. In the UVPD spectrum of the K+ ·DB18C6 complex, the interaction appears as an exciton splitting for vibronic bands with an interval of 2.7 cm−1 [32]. This complex belongs to C2v point group, and two benzene rings are symmetrically equivalent. In contrast, in the spectrum of the K+ ·DB15C5 complex the S1 –S0 and S2 –S0 transitions are separated by more than 100 cm−1 with quite different vibronic structures; the S1 –S0 transition shows an intense and extensive progression, while the S2 –S0 transition has a strong origin band followed by several vibronic bands. The large separation of the electronic transitions and different vibronic structures will be due to the fact that the two benzene rings are symmetrically inequivalent in K+ ·DB15C5 [39]. Figure 5.9 presents molecular orbitals (MOs) that contribute the most to the S1 –S0 and S2 –S0 transitions of the K+ ·DB15C5 (K-A) and K+ ·DB18C6 (K-a) complexes. Interestingly, the MOs and resulting electronic transitions are almost localized in one of the two benzene rings for the K+ ·DB15C5 complex (Fig. 5.9a); the electronic interaction between the two benzene rings is quite small. In contrast, the K+ ·DB18C6 complex (K-a) belongs to C2v point group, and as a result the MOs are equally distributed to the two benzene rings. For the K+ ·DB18C6 complex (K-a), the distance between the centers of the benzene rings is 9.1 Å. In the case of the K+ ·DB15C5 complex (K-A), the distance is 8.0 Å, slightly shorter than that of K+ ·DB18C6. In spite of the shorter distance, the electronic interaction between the benzene rings seems to be very weak for K+ ·DB15C5. Hence, it is probable that the main reason for very weak electronic interaction in the K+ ·DB15C5 complex is a substantially large

126

Y. Inokuchi

Fig. 5.9 MOs that contribute the most to the S1 –S0 and S2 –S0 transitions of a K+ ·DB15C5 and b K+ ·DB18C6. Reprinted with permission from [39] J. Phys. Chem. A, 121, 954–962 (2017). Copyright 2017 American Chemical Society

difference in the electronic transition energy of the two benzene rings due to their inequivalent natures. In addition, it is also plausible that the relative configuration of the benzene rings is not suitable very much for the interaction in the K+ ·DB15C5 complex [39]. The electronic excitonic interaction energy can be estimated with the transition dipole moments of the two chromophores. In the weak interaction model, the S1 –S2 splitting energy of vibronic transitions (E) for systems having two equivalent chromophores is given by the following equations: E = 2 · F · Vab 3 −1 Vab = |µa | · |µb | · (4πε0 Rab ) · (2cosθa · cosθb −sinθa · sinθb · cosφ)

(5.1) (5.2)

where µa and µb are the transition dipole moments of the two chromophores, Rab is the distance between the two chromophores, θ a and θ b are the angles of the transition

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

127

Table 5.2 Transition dipole moments, distance of benzene rings, and estimated exciton splitting energies for the K+ ·DB15C5 and K+ ·DB18C6 complexes |μa | (10−30 cm)

|μb | (10−30 cm)

Rab (Å)

|μa ||μb |/4πε0 R3ab V aab (cm−1 ) (cm−1 )

V bab (cm−1 )

K+ ·DB18C6

5.6

5.6

9.1

18

33

27

K+ ·DB15C5

4.4

4.5

8.0

17

17

10

a Calculated b Estimated

with Eq. (5.2) with the results of the TD-DFT calculations shown in Fig. 5.10

dipole moments to the line connecting the two chromophores, and φ is the dihedral angle between µa and µb [40–44]. V ab represents the electronic part, corresponding to the electronic interaction energy estimated above, and F is the vibrational part, which takes the Franck–Condon factor into account. We estimate the V ab value of the K+ ·DB15C5 and K+ ·DB18C6 complexes using Eq. (5.2), the geometry of the most stable conformers, and the calculated transition dipole moments of the two chromophores described above. Table 5.2 collects |µa |, |µb |, Rab , the energy part of Eq. (5.2), and the estimated V ab values. The transition dipole moments of the two chromophores are almost parallel to the benzene plane and perpendicular to benzene C–C bonds forming the ether ring. For the K+ ·DB18C6 complex, V ab = 33 cm−1 is obtained with Eq. (5.2). This value agrees well with the interaction energy (27 cm−1 ) obtained with the calculated energies of the electronic states (Fig. 5.10). For the K+ ·DB15C5 complex, the energy part of Eq. (5.2) (17 cm−1 , see the third column from the right in Table 5.2) is almost the same as that of the K+ ·DB18C6 complex (18 cm−1 ). However, since the angle part of the DB15C5 complex is fairly smaller than that of the DB18C6 one, the V ab value of the K+ ·DB15C5 complex (17 cm−1 ) is substantially smaller than that of K+ ·DB18C6 (33 cm−1 ), which agrees with the trend of the TD-DFT results described above or in the rightmost column of Table 5.2. Hence, the main reason for very weak electronic interaction in the K+ ·DB15C5 complex is its less suitable arrangement (not distance but relative angles) of the two benzene rings than that of the K+ ·DB18C6 complex. In order to examine the degree of localization for the electronic transitions of the K+ ·DB15C5 complex quantitatively, we estimate the contribution (%) of the electronic transition on each benzene ring to the S1 –S0 and S2 –S0 transitions of the K+ ·DB15C5 complex (Table 5.3). The excited states can be expressed by linear combinations of φ a * ·φ b and φ a ·φ b * : ψ+ = α · φa ∗ · φb + β · φa · φb ∗

(5.3a)

ψ− = β · φa ∗ · φb − α · φa · φb ∗

(5.3b)

where φ α(β ) * and φ α(β ) are wavefunctions of two chromophores (a and b) in the S1 and S0 states, respectively. These values are estimated in two ways. One is based on

128

Y. Inokuchi

Fig. 5.10 Energy levels of the electronic excited states of the two chromophores and the complexes for a K+ ·DB18C6 and b K+ ·DB15C5. The vertical axis is the energy with respect to the electronic ground state of each species. The structures in the figure are the K+ complexes used for the calculation of the UV transition energy of the two chromophores and the complexes. The scaling factor of 0.8340 is employed also for the calculated energy in this figure. The numbers in the figure show the energy difference in cm−1 . The arrows in part a show the direction of the transition dipole moment of the two chromophores. Reprinted with permission from [39] J. Phys. Chem. A, 121, 954–962 (2017). Copyright 2017 American Chemical Society

the calculated S1 –S0 and S2 –S0 transition energies of the complexes and the calculated S1 –S0 transition energy of the two chromophores; the contribution corresponds to the square of coefficients α and β in Eq. (5.3). In the other, the contribution is estimated from the results of the TD-DFT calculations. The calculations provide coefficients for electron promotion between two MOs for the electronic transitions of the complexes. Hence, we can estimate the contribution of each chromophore to the electronic transitions of the complexes from these coefficients. It is quite obvious in Table 5.3 that the S1 –S0 and S2 –S0 transitions of the K+ ·DB15C5 complex are localized in one of the two benzene rings. In addition, the UVPD spectrum of the K+ ·DB15C5 complex in the present study shows separated S1 and S2 excited states. The vibronic structures of the S1 –S0 and S2 –S0 transitions of the K+ ·DB15C5 complex are quite different from each other; the S1 –S0 transition shows an intense and extensive progression, whereas the S2 –S0 transition has a strong origin band fol-

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

129

Table 5.3 Contribution (%) of the electronic transition of two chromophores (benzene a and b) to the S1 –S0 and S2 –S0 transition of the K+ ·DB15C5 complex Conformer K+ ·DB15C5

K-A

Transition

Contribution (%)a

Contribution (%)b

Benzene a

Benzene a

Benzene b

Benzene b

S1 –S0

97

3

72

0

S2 –S0

3

97

3

67

a Calculated on the basis of the S

1 –S0 and S2 –S0 transition energies of the complexes and the S1 –S0 transition energy of two chromophores b Estimated from the results of the TD-DFT calculations (coefficients for electron promotion between two MOs)

lowed by several vibronic bands. These theoretical and experimental results strongly suggest that the K+ ·DB15C5 complex has no or very weak excitonic interaction between the benzene rings, different from the case of the K+ ·DB18C6 complex [32]. This conclusion indicates that the K+ ·DB15C5 complex belongs to the “very weak” category of excitonic interactions [43].

5.1.6 DB21C7 and DB24C85 Figure 5.11 shows UVPD spectra of the K+ ·DB21C7 and K+ ·DB24C8 complexes with that of K+ ·DB15C5 and K+ ·DB18C6 [32, 39, 45]. The spectra of the K+ ·DB15C5 and K+ ·DB18C6 complexes exhibit well-resolved sharp vibronic bands [32, 39]. Figure 5.12a, b shows the structure with the highest occupied MO (HOMO) and second highest occupied MO (HOMO-1) of the K+ ·DB15C5 (K-A) and K+ ·DB18C6 (K-a) complexes determined in the previous studies [32, 39]. Numbers in Fig. 5.12 represent the contribution (%) of each benzene part to the MOs. The crown cavity of DB15C5 is too small to encapsulate K+ ion, and the K+ ·DB15C5 complex has a C1 structure [39]. In the K+ ·DB18C6 complex, the crown opens the cavity the most, and the K+ ion is located at the center of the cavity, providing a boat-type, C2v form [32]. The UVPD spectra of the K+ ·DB21C7 and K+ ·DB24C8 complexes (Fig. 5.11c, d) show different features from those of K+ ·DB15C5 and K+ ·DB18C6. The K+ ·DB21C7 complex exhibits several sharp bands on the high-frequency side of ~36,777 cm−1 . In sharp contrast, the K+ ·DB24C8 complex shows very broad features with maxima at ~36,400 and ~36,670 cm−1 , accompanied by very weak, sharp bands around 36,143 cm−1 . We determine the most probable structure of the K+ ·DB21C7 and K+ ·DB24C8 complexes on the basis of quantum chemical calculations [45]. Figure 5.12c, d depicts the most stable form of the K+ ·DB21C7 and K+ ·DB24C8 5 This section is based on the following article: [45] Kida M., Kubo M., Ujihira T., Ebata T., Abe M.,

Inokuchi Y.: Selective probing of potassium ion in solution by intramolecular excimer fluorescence of dibenzo-crown ethers. ChemPhysChem 19, 1331–1335 (2018). Copyright (2018) Wiley-VCH Verlag GmbH & Co. KGaA.

130

Y. Inokuchi

Fig. 5.11 The UVPD spectra (curves) and the calculated oscillator strengths (bars) of the K+ ·DB15C5, K+ ·DB18C6, K+ ·DB21C7, and K+ ·DB24C8 complexes. Reprinted with permission from [45] ChemPhysChem., 19, 1331–1335 (2018). Copyright 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

complexes with the HOMO and HOMO-1. Vertical electronic transition energies obtained by the TD-DFT calculations are also displayed in Fig. 5.11 with bars. For the K+ ·DB21C7 complex, the most stable structure (K-A in Fig. 5.12c) has an open conformation with benzene rings distant from each other. The electronic transitions of K-A well reproduce the position of the sharp UVPD bands (Fig. 5.11c). Under the present cold (~10 K) condition, the UVPD spectrum of the K+ ·DB21C7 complex is attributed mainly to K-A in Fig. 5.12c [45]. In the case of the K+ ·DB24C8 complex, the most stable structure (K-a in Fig. 5.12d) shows two electronic transitions at positions similar to the broad component of the UVPD spectrum, as shown in Fig. 5.11d. The maxima at ~36,400 and ~36,670 cm−1 can be ascribed to the two electronic transitions of K-a. As shown in Fig. 5.12d, isomer K-a of K+ ·DB24C8 has a highly folded conformation for the encapsulation of K+ ion, different from the case of K+ ·DB15C5, K+ ·DB18C6, and K+ ·DB21C7. As a result, the distance between the benzene rings is very short ( Na+ > K+ > Rb+ > Cs+ . This conclusion indicates that solvent plays a crucial role in the selectivity of K+ ion in solution [15, 16, 18, 54]. A number of theoretical and experimental studies have been devoted to CE complexes in the gas phase under microsolvated conditions, in solutions, and in crystals to understand the ion selectivity of CEs in solutions [20, 22, 66–80]. These experimental and theoretical studies both suggest that the net effects due to solvation of not only the M+ ions but also of the M+ ·18C6 complexes are of great importance for determining the ion-binding selectivity in solution. It was also pointed out that entropic effects, which were missing in the above studies, should be included to fully understand the ion selectivity, because the encapsulation process remains endothermic simply by considering only the enthalpy contribution [15, 16, 70]. In order to examine the entropic contribution, one has to

136

Y. Inokuchi

determine the structure of the CE complexes. In addition, the number of conformers will also affect the ion selectivity; the larger the number of conformations a specific complex adopts, the more favorable is its formation. In this section, we will present our recent studies on the microsolvation of cold complexes of CEs with monovalent and divalent metal ions; the CEs are DB18C6, B18C6, and B15C5, and the metal ions are K+ , Rb+ , Cs+ , Ca2+ , Sr2+ , Ba2+ , and Mn2+ ; molecules used for microsolvation are H2 O and CH3 OH [81–83]. The experimental details were described in the previous papers [81–83]. First, UVPD spectra are observed for the microsolvated complexes under cold, gas-phase conditions. Then IR-UV double-resonance spectroscopy is employed to measure conformer-selective IR spectra of the cold species in the OH stretching (3200–3800 cm−1 ) region. Based on the IR-UV results, the number and the structure of the conformers are determined with the aid of quantum chemical calculations. Finally, the relation between the results of the microsolvation in the gas phase and the ion selectivity in solutions is discussed.

5.2.2 K+ ·DB18C6·(H2 O)n 6 Figure 5.16 shows the UVPD spectra of the cooled K+ ·DB18C6·(H2 O)n (n = 0–5) complexes [81]. The spectra for n = 1–5 are measured by monitoring the yield of the bare K+ ·DB18C6 photofragment ion. All the UVPD spectra in Fig. 5.16 show a number of vibronically resolved bands and a redshift relative to non-hydrated K+ ·DB18C6 [32]. We measure the IR-UV spectra of the K+ ·DB18C6·(H2 O)n complexes by fixing the UV wavenumber at the positions pointed by the arrows in Fig. 5.16 and scanning the wavenumber of the IR OPO. The IR-UV spectra recorded by fixing the UV frequency where there is no vibronic band provide IR gain spectra that are a sum of different conformers, if they exist [34]. We first measure IR gain spectra at nonresonant UV frequencies to see the IR absorption of all the conformers and then measure IR dip spectra with the UV fixed on specific vibronic bands to attribute IR bands to each conformer. Figure 5.17 displays the IR-UV and calculated IR spectra of the K+ ·DB18C6·(H2 O)n (n = 1−3) in the OH stretching region [81]. The top curve in each panel of Fig. 5.12b, c is the IR-UV gain spectrum, which is well reproduced by the summation of the IR-UV dip spectra shown below. For the n = 1 complex (Fig. 5.17a), only one kind of IR-UV spectra is observed, indicating that there is a single conformer. In the case of the n = 2 and 3 complexes (Fig. 5.17b, c), two different IR-UV spectra are observed, showing that the n = 2 and 3 complexes each has at least two conformers. Figure 5.18 displays the IR-UV and calculated IR spectra of the K+ ·DB18C6·(H2 O)n (n = 4, 5) in the OH stretching region. The n = 4 and 5 6 This section is based on the following article: [81] Inokuchi Y., Ebata T., Rizzo T. R., Boyarkin O. V.: Microhydration effects on the encapsulation of potassium ion by dibenzo-18-crown-6. J. Am. Chem. Soc. 136, 1815–1824 (2014). Copyright (2014) American Chemical Society.

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

137

Fig. 5.16 The UVPD spectra of the K+ ·DB18C6·(H2 O)n (n = 1–5) complexes with that of bare K+ ·DB18C6 complex. The arrows show the UV positions at which the IR-UV spectra are measured. Reprinted with permission from [81] J. Am. Chem. Soc., 136, 1815–1824 (2014). Copyright 2014 American Chemical Society

complexes show only one IR-UV spectrum each. Hence, one dominant conformer exists for the n = 4 and 5 complexes. In geometry optimization with quantum chemical calculations, many stable isomers are found for the K+ ·DB18C6·(H2 O)n (n = 1–5) complexes. In all of these isomers, the K+ ·DB18C6 part takes a boat form, similar to the structure of the nonhydrated K+ ·DB18C6, with different hydration manners [81]. Since these isomers are predicted to have substantially different IR spectra in the OH stretching region, we can determine the structure of the K+ ·DB18C6·(H2 O)n complexes by comparison of the IR-UV spectra with the calculated ones [81]. The structure of the complexes determined on the basis of IR spectroscopy is shown in Fig. 5.19; the calculated IR spectra of the complexes in Fig. 5.19 are shown in Figs. 5.17 and 5.18. The IR spectral patterns in the IR-UV spectra are well reproduced by those of the calculated IR spectra. The K+ ·DB18C6 complex has a boat-type C2v conformer in the gas phase [32]. This bent structure of K+ ·DB18C6, which is due to structural constraints by the two benzene rings, affects the hydration manner. As shown in Fig. 5.17c, the n = 3 complex has two stable conformers in the ion trap. This is characteristic of the K+ complex compared to the Rb+ and Cs+ complexes. The measured and calculated IR spectra of the Rb+ ·DB18C6·(H2 O)3 and Cs+ ·DB18C6·(H2 O)3 complexes are shown in Fig. 5.20. The UVPD spectra of these complexes are displayed in the

138

Y. Inokuchi

Fig. 5.17 The IR-UV and calculated IR spectra of the K+ ·DB18C6·(H2 O)1–3 complexes in the OH stretching region. Numbers in parentheses are the wavenumber of the UV probe laser used for the IR-UV measurements. Characters in parentheses show the conformation in Fig. 5.19. Reprinted with permission from [81] J. Am. Chem. Soc., 136, 1815–1824 (2014). Copyright 2014 American Chemical Society

next section. The IR-UV gain spectra (the top curve in each panel of Fig. 5.20a, b), which are measured at nonresonant UV positions, are similar to the IR-UV depletion spectra monitored at strong vibronic bands (the middle curve in Fig. 5.20a, b). This suggests that there is only one conformer for each of the Rb+ ·DB18C6·(H2 O)3 and Cs+ ·DB18C6·(H2 O)3 complexes. The structures of the Rb+ ·DB18C6·(H2 O)3 and Cs+ ·DB18C6·(H2 O)3 complexes are attributed to the most stable isomers, Rb3a and Cs3a, respectively, on the basis of the similarity between the IR-UV spectra and the calculated IR spectra (Fig. 5.20). Both conformers Rb3a and Cs3a strongly resemble conformer K3a of the K+ ·DB18C6·(H2 O)3 complex (Fig. 5.19d). The high stability of K3a, Rb3a, and Cs3a can be ascribed to the displaced position of the metal ions in the M+ ·DB18C6 complexes. The metal ions in the non-hydrated K+ ·DB18C6, Rb+ ·DB18C6, and Cs+ ·DB18C6 complexes deviate from the oxygen mean plane of

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes Fig. 5.18 The IR-UV and calculated IR spectra of the K+ ·DB18C6·(H2 O)4, 5 complexes in the OH stretching region. Numbers in parentheses are the wavenumber of the UV probe laser used for the IR-UV measurements. Characters in parentheses show the conformation in Fig. 5.19. Reprinted with permission from [81] J. Am. Chem. Soc., 136, 1815–1824 (2014). Copyright 2014 American Chemical Society

Fig. 5.19 Structure of the K+ ·DB18C6·(H2 O)1–5 complexes determined by the comparison of the IR spectra observed and calculated. Reprinted with permission from [81] J. Am. Chem. Soc., 136, 1815–1824 (2014). Copyright 2014 American Chemical Society

139

140

Y. Inokuchi

Fig. 5.20 The IR-UV and calculated IR spectra of the M+ ·DB18C6·(H2 O)3 (M+ = Rb+ and Cs+ ) complexes in the OH stretching region. Numbers in parentheses are the wavenumber of the UV probe laser used for the IR-UV measurements. Characters in parentheses show the conformation. Reprinted with permission from [81] J. Am. Chem. Soc., 136, 1815–1824 (2014). Copyright 2014 American Chemical Society

DB18C6 by 0.51, 1.00, and 1.36 Å, respectively [32, 34]. Hence, the H2 O molecules in the M+ ·DB18C6·(H2 O)3 complexes tend to stay above the M+ ion to have direct intermolecular bonds with the metal ions. On the other hand, another conformer (K3g) coexists for the K+ ·DB18C6·(H2 O)3 complex. This is likely due to the optimum matching between the K+ ion and the crown cavity. Compared to Rb+ and Cs+ ions, the K+ ion is effectively encapsulated by DB18C6, making the interaction between K+ and H2 O molecules weaker. As a result, structures with different modes of hydration, such as K3a and K3g, are less different in energy for the K+ ·DB18C6 complex compared to Rb+ and Cs+ and can coexist. Multiple conformations for the K+ ·DB18C6·(H2 O)3 complex are therefore evidence of the effective capture of the K+ ion by DB18C6 over the Rb+ and Cs+ ions in water. In contrast, one conformer is predominantly stable for the larger K+ ·DB18C6·(H2 O)n (n = 4 and 5) complexes. In these complexes, the H2 O molecules are bound to the K+ ion cooperatively by forming an H2 O ring, and the K+ ion is pulled out from the DB18C6 cavity more than the smaller complexes. In particular, the n = 4 complex has a quite symmetric structure (K4a in Fig. 5.19f); the ring with four H2 O molecules looks the most suitable for the solvation to the K+ ·DB18C6 complex. The appearance of one stable conformation for the n = 4 and 5 complexes is mainly due to the high stability of the rings formed

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

141

with four or five water molecules and also due to a good matching in size between K+ ion and these water rings.

5.2.3 Rb+ ·DB18C6·(H2 O)n and Rb+ ·DB18C6·(H2 O)n 7 Figure 5.21a shows the UVPD spectra of the Rb+ ·DB18C6·(H2 O)n (n = 0–8) complexes, which exhibit a similar trend as the corresponding K+ complexes [81, 83]. The 0-0 band in the spectrum of Rb+ ·DB18C6 appears at 36,319 cm−1 [32], with those of the hydrated complexes progressively shifting to lower energy for n = 1 to 4

Fig. 5.21 The UVPD spectra of the M+ ·DB18C6·(H2 O)n (M = Rb and Cs, n = 1–8) complexes with that of bare (n = 0) complex. The arrows show the UV positions at which the IR-UV spectra are measured. Reprinted with permission from [83] J. Phys. Chem. A, 122, 3754–3763 (2018). Copyright 2018 American Chemical Society

7 This

section is based on the following article: [83] Inokuchi Y., Ebata T., Rizzo T. R.: Microhydration of dibenzo-18-crown-6 complexes with K+ , Rb+ , and Cs+ investigated by cold UV and IR spectroscopy in the gas phase. J. Phys. Chem. A 122, 3754–3763 (2018). Copyright (2018) American Chemical Society.

142

Y. Inokuchi

before shifting to higher frequency again at n = 5. The UVPD spectrum of the n = 4 complex shows congested features, similar to that of the K+ ·DB18C6·(H2 O)4 complex [81, 83]. Unlike the case of the doubly hydrated potassiated species (Fig. 5.16c), the Rb+ ·DB18C6·(H2 O)2 complex shows congested spectral features with no strong origin band. The UVPD spectra of the Cs+ ·DB18C6·(H2 O)n (n = 0 − 8) complexes are shown in Fig. 5.21b. The origin band of Cs+ ·DB18C6 appears at 36,234 cm−1 [32]. Similar to K+ and Rb+ , the UV bands of the Cs+ complexes shift to lower frequency with increasing hydration. However, one difference for the Cs+ spectra from the K+ and Rb+ spectra is that the UV bands continuously shift to lower frequency from n = 1 to 5 before exhibiting a blueshift at n = 6. Moreover, in the case of the Cs+ complexes, it is the n = 5 species that exhibits particularly congested vibronic features. In the spectra of the n = 2 and 4 complexes, a number of sharp bands appear much more closely than the case of Rb+ . These spectral features will represent the complex structure, which will be described in more detail later. The structure of the hydrated complexes was determined on the basis of UVPD and IR-UV spectra and quantum chemical calculations; the details of the determination were explained in the previous study [83]. Figure 5.22 shows the structure of the M+ ·DB18C6·(H2 O)n (M = K, Rb, Cs; n = 1 − 5) complexes in the gas phase. The bare complexes, M+ ·DB18C6 (M = K, Rb, and Cs), all adopt the boat conformation (K-a, Rb-a, and Cs-b) in which the cavity of the DB18C6 is most open, and the metal ions are located almost at the center of the cavity. The difference in the structure among the bare complexes is the distance between the crown cavity and the metal ions [32]. This small difference leads to a remarkable difference in the structure of the hydrated species. For the n = 1 complexes, the structure is similar for K+ , Rb+ , and Cs+ (K1a, Rb1a, and Cs1b). For n = 2, Rb+ and Cs+ complexes have a similar hydration structure (Rb2a and Cs2a); the two H2 O molecules are bound to Rb+ or Cs+ independently and form an O–H···π hydrogen bond. In contrast, the K+ ·DB18C6·(H2 O)2 complex has two types of hydration structure, both different from that of the Rb+ and Cs+ complexes; the two H2 O molecules are bound to each other through an O–H···O hydrogen bond, and are located either on top (K2d) or at the bottom (K2f) of the boat-type K+ ·DB18C6 conformer. The K+ ·DB18C6·(H2 O)3 , Rb+ ·DB18C6·(H2 O)3 , and Cs+ ·DB18C6·(H2 O)3 complexes all have a similar hydration structure. One of the two isomers of the K+ complex (K3a) and the isomers of the Rb+ and Cs+ complexes (Rb3a and Cs3a) have all the three H2 O molecules on top of the M+ ·DB18C6 part with a similar hydration structure. The K+ ·DB18C6·(H2 O)3 complex has an additional isomer (K3g) with two H2 O molecules on top and one at the bottom of the boat K+ ·DB18C6 conformer. For the n = 4 and 5 complexes, the K+ and Rb+ complexes have a similar structure to each other (K4a and Rb4a, and K5a and Rb5a), while the Cs+ complexes have different hydration patterns (Cs4a and Cs5a). In the case of the n = 5 complexes, one of the five H2 O molecules is bound to an oxygen atom of the DB18C6 component. For the K+ and Rb+ complexes, the third and fifth H2 O molecules are close to the benzene rings, forming the O–H···π hydrogen bond. In contrast, the distance between the H2 O ring and the benzene rings is substantially longer for the Cs+ complex than that for the K+ and Rb+ complexes because the Cs+ ion is displaced largely from the DB18C6 part.

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

143

Fig. 5.22 Optimized structures of the M+ ·DB18C6·(H2 O)n (M = K, Rb, Cs; n = 0–5) complexes calculated at the M05-2X/6-311++G(d, p) level. It was confirmed that the conformers of the K+ ·DB18C6·(H2 O)1–5 complexes in this figure exist in the experiment under cold gas-phase conditions. For the hydrated Rb+ ·DB18C6 and Cs+ ·DB18C6 complexes, the most stable conformers (or the second most stable one for Cs+ ·DB18C6·(H2 O)1 ) are displayed in this figure. Reprinted with permission from [83] J. Phys. Chem. A, 122, 3754–3763 (2018). Copyright 2018 American Chemical Society

Because the K+ ion is encapsulated deeply in the DB18C6 cavity, H2 O molecules can be bound to the K+ ion on both sides of the K+ ·DB18C6 complex, which results in multiple isomers in both experiment and theory. For the Rb+ and Cs+ ions, the distance between the metal ions and the DB18C6 cavity is slightly larger than in the case of the K+ ion [32], which allows H2 O molecules to interact with the Rb+ or Cs+ ion only on top of the M+ ·DB18C6 part, providing a single stable conformer. The existence of multiple isomers for the hydrated K+ ·DB18C6 complexes can contribute to the effective formation of the K+ ·DB18C6 complex and preferential capture of

144

Y. Inokuchi

K+ ion by DB18C6 in solution because of “conformational” entropic effects. These are different from usual entropic effects, which are related to the Gibbs free energy of a single conformation, but the more the number of complex conformers, the more preferred the complex formation. In this sense, the results of the Rb+ and Cs+ complexes reinforce the uniqueness of K+ ion in the encapsulation by DB18C6 [81, 83].

5.2.4 M2+ ·B15C5·L and M2+ ·B18C6·L (M = Ca, Sr, Ba, Mn; L = H2 O and CH3 OH)8 A similar trend for the conformation number of microsolvated complexes is found for divalent metal ions [82]. Figure 5.23 shows the UVPD spectra of the M2+ ·CE·L (M = Ca, Sr, Ba, Mn; CE = B15C5 and B18C6; L = H2 O and CH3 OH) complexes [82]. The band origins of neutral B15C5 is found at 35,645, 35,653, and 36,217 cm−1 for three conformers [84]. Neutral B18C6 shows the band origins at 35,167 cm−1 and around 35,650 cm−1 for four conformers [85]. Thus, all the UVPD spectra in Fig. 5.23 show blueshifted transitions compared to neutral B15C5 or B18C6; this is the same trend as that of B15C5 and B18C6 complexes with alkali metal ions [34]. All the UVPD spectra in Fig. 5.23 show a number of vibronically resolved bands. In order to distinguish vibronic bands due to different conformers and determine their structure, we measure the IR-UV spectra of the complexes by fixing the UV wavenumber at one of the vibronic bands and scanning the wavenumber of the IR OPO in the OH stretching region. In addition, we measure conformer-specific UVPD spectra by fixing the IR frequency to one of the vibrational bands of a particular conformer and scanning the UV frequency. By examining the depletion yield of each vibronic band by the IR excitation, we can distinguish those bands in the UVPD spectra that belong to the same conformer [82]. The IR-UV spectra of the M2+ ·CE·L (M = Ca and Sr; CE = B15C5 and B18C6; L = H2 O and CH3 OH) complexes are displayed in Fig. 5.24. In this IR region, the solvated complexes show the OH stretching vibration of the H2 O and CH3 OH components. Quantum chemical calculations of the M2+ ·CE·L complexes suggest that the H2 O and CH3 OH molecules are directly attached to the M2+ ions through an M2+ ···O intermolecular bond in all the calculated conformers; the OH groups are almost free from an intermolecular bond [82]. As a result, the OH stretching bands of the H2 O and CH3 OH components show sharp spectral features. Hence, the IRUV spectra can be used to distinguish vibronic bands in the UVPD spectra due to different conformers, though the difference in the band position of the OH stretching vibrations is quite small for different conformers. Figure 5.24a shows the IR-UV spectrum of the Ca2+ ·B15C5·H2 O complex. Vibrational bands are observed at 3610 8 This

section is based on the following article: [82] Inokuchi Y., Ebata T., Rizzo T. R.: Solvent effects on the encapsulation of divalent ions by benzo-18-crown-6 and benzo-15-crown-5. J. Phys. Chem. A 119, 8097–8105 (2015). Copyright (2015) American Chemical Society.

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

145

Fig. 5.23 The UVPD spectra of the M2+ ·CE·L (M = Ca, Sr, Ba, Mn; CE = B15C5 and B18C6; L = H2 O and CH3 OH) complexes. The arrows show the UV positions at which the IR-UV spectra are measured. Reprinted with permission from [82] J. Phys. Chem. A, 119, 8097–8105 (2015). Copyright 2015 American Chemical Society

146

Y. Inokuchi

Fig. 5.24 IR-UV double-resonance spectra of the M2+ ·CE·L (M = Ca and Sr; CE = B15C5 and B18C6; L = H2 O and CH3 OH) complexes. The numbers in parentheses show the UV position for measuring each IR-UV spectrum (see Fig. 5.23). Reprinted with permission from [82] J. Phys. Chem. A, 119, 8097–8105 (2015). Copyright 2015 American Chemical Society

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

147

and 3676 cm−1 ; these can be assigned to the symmetric and asymmetric OH stretching vibrations at the H2 O part. Setting the UV laser on most of the strong vibronic bands in the UVPD spectrum of Ca2+ ·B15C5·H2 O results in IR bands at the same positions as those in Fig. 5.24a with a precision of ~1 cm−1 in the band position. This result suggests that there is only one conformer for the Ca2+ ·B15C5·H2 O complex in our experiment. Also for the Ca2+ ·B15C5·CH3 OH complex, the IR-UV measurement provides only one IR spectrum (Fig. 5.24c); there is also only one conformer for Ca2+ ·B15C5·CH3 OH in our experiment. In sharp contrast to the case of the Ca2+ ·B15C5·H2 O and Ca2+ ·B15C5·CH3 OH complexes, the B18C6 complexes of Ca2+ ion show multiple conformations. Figure 5.24b displays the IR-UV spectra of the Ca2+ ·B18C6·H2 O complex. All the IR-UV spectra show the bands at similar positions (~3615 and ~3695 cm−1 ), but noticeable differences in the frequencies can be seen among the three spectra. The IR-UV spectrum measured at 36,951 cm−1 (the upper spectrum in Fig. 5.24b) has both gain and depletion signals around 3615 and 3695 cm−1 . The UV band at 36,951 cm−1 is relatively weaker than the other bands. In the IR-UV measurement at 36,951 cm−1 , electronic transitions from vibrationally excited states of main conformers also occur and provide the same photofragment Ca2+ ·B18C6 ion. This excitation process gives gain signals in the IR-UV spectrum of the minor conformer measured at 36,951 cm−1 [32, 34]. From the IR-UV results in Fig. 5.24b, it was concluded that the Ca2+ ·B18C6·H2 O complex has at least three conformers in our experiment. In the case of the Ca2+ ·B18C6·CH3 OH complex, three types of IR-UV spectra are also observed in the OH stretching region (Fig. 5.24d), indicating that there are at least three isomers. Figures 5.24e–h, 5.25a–d, and 5.25e–h show the IR-UV spectra of the Sr2+ , Ba2+ , and Mn2+ complexes, respectively. All the complexes show OH stretching vibrations in the 3550–3700 cm−1 region. Based on their stretching frequencies, the OH groups are free from a strong intermolecular bond, and the H2 O or CH3 OH is bound to the M2+ ion through an M2+ ···O bond. The IR-UV spectrum of the Sr2+ ·B15C5·CH3 OH complex is expanded around 3640 cm−1 (Fig. 5.24g). Though the difference in the OH frequency is quite small (~1.1 cm−1 ), they are reproducibly different; there are at least two conformers for Sr2+ ·B15C5·CH3 OH. For most of the B15C5 conformers, the OH bands appear at ~3610 and ~3677 cm−1 for the H2 O complexes and at ~3640 cm−1 for the CH3 OH complexes. One conformer of the Ba2+ ·B15C5·H2 O complex shows OH bands at 3587 and 3670 cm−1 (Fig. 5.25a), which are smaller than those of the other B15C5·H2 O complexes; it is probable that the OH groups are bonded to the benzene ring of B15C5 through the O–H···π hydrogen bonds in this conformer. In contrast to the B15C5 complexes, those containing B18C6 show a wider range of OH frequencies for Ca2+ , Sr2+ , Ba2+ , and Mn2+ . The symmetric and asymmetric OH stretching frequencies of the H2 O component are in the range of 3583–3616 and 3674–3699 cm−1 , respectively. The OH bands of the CH3 OH complexes are located in the 3598–3646 cm−1 range. The difference in the OH frequencies reflects the orientation of the H2 O or CH3 OH component against the metal ion–CE complexes. The B18C6 complexes can have more different types of solvation geometries than those with B15C5.

148

Y. Inokuchi

Fig. 5.25 IR-UV double-resonance spectra of the M2+ ·CE·L (M = Ba and Mn; CE = B15C5 and B18C6; L = H2 O and CH3 OH) complexes. The numbers in parentheses show the UV position for measuring each IR-UV spectrum (see Fig. 5.23). Reprinted with permission from [82] J. Phys. Chem. A, 119, 8097–8105 (2015). Copyright 2015 American Chemical Society

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

149

Table 5.4 The number of isomers of M2+ ·B15C5·L and M2+ ·B18C6·L complexes with M2+ = Ca2+ , Sr2+ , Ba2+ , and Mn2+ , and L = H2 O and CH3 OH M2+

L = H2 O

L = CH3 OH

B15C5

B18C6

B15C5

B18C6

Ca2+

1

3

1

3

Sr2+

1

3

2

5

Ba2+

2

1

1

2

Mn2+

1

2

1

3

We collect the number of conformers determined by IR-UV double-resonance spectroscopy in Table 5.4. The number of conformers of the B18C6 complexes is obviously larger than that of the B15C5 complexes for Ca2+ , Sr2+ , and Mn2+ ions. In contrast, the Ba2+ complexes do not show such a clear trend. These results can be attributed to the matching in size between the crown cavity and the metal ions, as seen in the case of the K+ ·DB18C6·(H2 O)3 , Rb+ ·DB18C6·(H2 O)3 , and Cs+ ·DB18C6·(H2 O)3 complexes [81]. The calculated structure of the M2+ ·CE·L (M = Ca, Sr, Ba, Mn; CE = B15C5 and B18C6; L = H2 O and CH3 OH) complexes was shown in the previous study [82]. In the bare complexes of B15C5 with Ca2+ , Sr2+ , and Ba2+ , the B15C5 cavity is too small to encapsulate these ions completely; one side of the metal ions is open for direct solvation. The H2 O or CH3 OH molecule is bound on the open side, opposite to the B15C5. This situation is quite similar to the Rb+ ·DB18C6 and Cs+ ·DB18C6 complexes [81]. In contrast, the B18C6 component in the bare complexes holds the Ca2+ , Sr2+ , and Ba2+ ions more deeply than B15C5. In the solvated complexes, the H2 O or CH3 OH molecule tends to extract the metal ions from the crown cavity by being attached directly to the metal ions through the M2+ ···O bond. From a thermochemical point of view, a larger number of conformations can provide larger stability in solution. In the encapsulation systems where an optimal matching in size occurs between host and guest species, there can be some flexibility in the position of the guest against the host. As a result, a larger number of solvation manners become possible for such systems when it is dissolved, which enhances the capability of encapsulation in solution.

References 1. Pedersen, C.J.: Cyclic polyethers and their complexes with metal salts. J. Am. Chem. Soc. 89, 7017–7036 (1967) 2. Pedersen C. J.: Crystalline salt complexes of macrocyclic polyethers. J. Am. Chem. Soc. 92, 386 (1970) 3. Pedersen, C.J.: New macrocyclic polyethers. J. Am. Chem. Soc. 92, 391–394 (1970) 4. Pedersen, C.J.: The discovery of crown ethers. Science 241, 536–540 (1988) 5. Zhang, H., Chu, J.H., Leming, S., Dearden, D.V.: Gas-phase molecular recognition—gas-phase crown-ether alkali-metal ion complexes and their reactions with neutral crowns. J. Am. Chem. Soc. 113, 7415–7417 (1991)

150

Y. Inokuchi

6. Zhang, H., Dearden, D.V.: The gas-phase macrocyclic effect—reaction-rates for crown ethers and the corresponding glymes with alkali-metal cations. J. Am. Chem. Soc. 114, 2754–2755 (1992) 7. Maleknia, S., Brodbelt, J.: Gas-phase selectivities of crown ethers for alkali-metal ion complexation. J. Am. Chem. Soc. 114, 4295–4298 (1992) 8. Chu, I.H., Zhang, H., Dearden, D.V.: Macrocyclic chemistry in the gas-phase—intrinsic cation affinities and complexation rates for alkali-metal cation complexes of crown-ethers and glymes. J. Am. Chem. Soc. 115, 5736–5744 (1993) 9. Brodbelt, J.S., Liou, C.C.: New frontiers in host-guest chemistry—the gas-phase. Pure Appl. Chem. 65, 409–414 (1993) 10. Dearden, D.V., Zhang, H., Chu, I.H., Wong, P., Chen, Q.Z.: Macrocyclic chemistry without solvents—gas-phase reaction-rates. Pure Appl. Chem. 65, 423–428 (1993) 11. Ray, D., Feller, D., More, M.B., Glendening, E.D., Armentrout, P.B.: Cation-ether complexes in the gas phase: bond dissociation energies and equilibrium structures of Li+ (1,2dimethoxyethane)x , x = 1 and 2, and Li+ (12-crown-4). J. Phys. Chem. 100, 16116–16125 (1996) 12. Sobott, F., Kleinekofort, W., Brutschy, B.: Cation selectivity of natural and synthetic ionophores probed with laser-induced liquid beam mass spectrometry. Anal. Chem. 69, 3587–3594 (1997) 13. More, M.B., Ray, D., Armentrout, P.B.: Cation-ether complexes in the gas phase: bond dissociation energies of K+ (dimethyl ether)x , x = 1 − 4; K+ (1,2-dimethoxyethane)x , x = 1 and 2; and K+ (12-crown-4). J. Phys. Chem. A 101, 4254–4262 (1997) 14. More, M.B., Ray, D., Armentrout, P.B.: Cation-ether complexes in the gas phase: bond dissociation energies of M+ (dimethyl ether)x , x = 1-3, M+ (1,2-dimethoxyethane)x , x = 1 and 2, and M+ (12-crown-4) where M = Rb and Cs. J. Phys. Chem. A 101, 7007–7017 (1997) 15. More, M.B., Ray, D., Armentrout, P.B.: Intrinsic affinities of alkali cations for 15-crown-5 and 18-crown-6: bond dissociation energies of gas-phase M+ –crown ether complexes. J. Am. Chem. Soc. 121, 417–423 (1998) 16. Armentrout, P.B.: Cation-ether complexes in the gas phase: thermodynamic insight into molecular recognition. Int. J. Mass Spectrom. 193, 227–240 (1999) 17. Anderson, J.D., Paulsen, E.S., Dearden, D.V.: Alkali metal binding energies of dibenzo-18crown-6: experimental and computational results. Int. J. Mass Spectrom. 227, 63–76 (2003) 18. Lee, S., Wyttenbach, T., Vonhelden, G., Bowers, M.T.: Gas-phase conformations of Li+ , Na+ , K+ , and Cs+ complexed with 18-crown-6. J. Am. Chem. Soc. 117, 10159–10160 (1995) 19. Wyttenbach, T., Von Helden, G., Bowers, M.T.: Conformations of alkali ion cationized polyethers in the gas phase: Polyethylene glycol and bis (benzo-15-crown-5)-15-ylmethyl pimelate. Int. J. Mass Spectrom. 165, 377–390 (1997) 20. Rodriguez, J.D., Vaden, T.D., Lisy, J.M.: Infrared spectroscopy of ionophore-model systems: hydrated alkali metal ion 18-crown-6 ether complexes. J. Am. Chem. Soc. 131, 17277–17285 (2009) 21. Rodriguez, J.D., Lisy, J.M.: Infrared spectroscopy of multiply charged metal ions: methanolsolvated divalent manganese 18-crown-6 ether systems. J. Phys. Chem. A 113, 6462–6467 (2009) 22. Rodriguez, J.D., Lisy, J.M.: Infrared spectroscopy of gas-phase hydrated K+ : 18-crown-6 complexes: evidence for high energy conformer trapping using the argon tagging method. Int. J. Mass Spectrom. 283, 135–139 (2009) 23. Rodriguez, J.D., Kim, D., Tarakeshwar, P., Lisy, J.M.: Exploring gas-phase ion-ionophore interactions: infrared spectroscopy of argon-tagged alkali ion-crown ether complexes. J. Phys. Chem. A 114, 1514–1520 (2010) 24. Martinez-Haya, B., Hurtado, P., Hortal, A.R., Steill, J.D., Oomens, J., Merkling, P.J.: Spectroscopic investigation of the gas-phase conformations of 15-crown-5 ether complexes with K+ . J. Phys. Chem. A 113, 7748–7752 (2009) 25. Martinez-Haya, B., Hurtado, P., Hortal, A.R., Hamad, S., Steill, J.D., Oomens, J.: Emergence of symmetry and chirality in crown ether complexes with alkali metal cations. J. Phys. Chem. A 114, 7048–7054 (2010)

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

151

26. Hurtado, P., Hortal, A.R., Gámez, F., Hamad, S., Martínez-Haya, B.: Gas–phase complexes of cyclic and linear polyethers with alkali cations. Phys. Chem. Chem. Phys. 12, 13752 (2010) 27. Kim, H.J., Shin, W.J., Choi, C.M., Lee, J.H., Kim, N.J.: Electronic photodepletion spectroscopy of dibenzo-18-crown-6 with a potassium ion. Bull. Korean Chem. Soc. 29, 1973–1976 (2008) 28. Choi, C.M., Kim, H.J., Lee, J.H., Shin, W.J., Yoon, T.O., Kim, N.J., Heo, J.: Ultraviolet photodepletion spectroscopy of dibenzo-18-crown-6-ether complexes with alkali metal cations. J. Phys. Chem. A 113, 8343–8350 (2009) 29. Choi, C.M., Lee, J.H., Choi, Y.H., Kim, H.J., Kim, N.J., Heo, J.: Ultraviolet photodepletion spectroscopy of dibenzo-18-crown-6-ether complexes with alkaline earth metal divalent cations. J. Phys. Chem. A 114, 11167–11174 (2010) 30. Boyarkin, O.V., Mercier, S.R., Kamariotis, A., Rizzo, T.R.: Electronic spectroscopy of cold, protonated tryptophan and tyrosine. J. Am. Chem. Soc. 128, 2816–2817 (2006) 31. Rizzo, T.R., Stearns, J.A., Boyarkin, O.V.: Spectroscopic studies of cold, gas-phase biomolecular ions. Int. Rev. Phys. Chem. 28, 481–515 (2009) 32. Inokuchi, Y., Boyarkin, O.V., Kusaka, R., Haino, T., Ebata, T., Rizzo, T.R.: UV and IR spectroscopic studies of cold alkali metal ion-crown ether complexes in the gas phase. J. Am. Chem. Soc. 133, 12256–12263 (2011) 33. Kusaka, R., Inokuchi, Y., Ebata, T.: Laser spectroscopic study on the conformations and the hydrated structures of benzo-18-crown-6-ether and dibenzo-18-crown-6-ether in supersonic jets. Phys. Chem. Chem. Phys. 9, 4452–4459 (2007) 34. Inokuchi, Y., Boyarkin, O.V., Kusaka, R., Haino, T., Ebata, T., Rizzo, T.R.: Ion selectivity of crown ethers investigated by UV and IR spectroscopy in a cold ion trap. J. Phys. Chem. A 116, 4057–4068 (2012) 35. Inokuchi, Y., Kusaka, R., Ebata, T., Boyarkin, O.V., Rizzo, T.R.: Laser spectroscopic study of cold host-guest complexes of crown ethers in the gas phase. ChemPhysChem 14, 649–660 (2013) 36. Inokuchi, Y., Nakatsuma, M., Kida, M., Ebata, T.: Conformation of alkali metal ion-benzo12-crown-4 complexes investigated by UV photodissociation and UV-UV hole-burning spectroscopy. J. Phys. Chem. A 120, 6394–6401 (2016) 37. Inokuchi, Y., Soga, K., Hirai, K., Kida, M., Morishima, F., Ebata, T.: Ultraviolet photodissociation spectroscopy of the cold K+ ·Calix[4]arene complex in the gas phase. J. Phys. Chem. A 119, 8512–8518 (2015) 38. Izatt, R.M., Bradshaw, J.S., Nielsen, S.A., Lamb, J.D., Christensen, J.J., Sen, D.: Thermodynamic and kinetic data for cation macrocycle interaction. Chem. Rev. 85, 271–339 (1985) 39. Inokuchi, Y., Kida, M., Ebata, T.: Geometric and electronic structures of dibenzo-15-crown-5 complexes with alkali metal ions studied by UV photodissociation and UV-UV hole-burning spectroscopy. J. Phys. Chem. A 121, 954–962 (2017) 40. Müller, A., Talbot, F., Leutwyler, S.: S1 /S2 exciton splitting in the (2-pyridone)2 dimer. J. Chem. Phys. 116, 2836–2847 (2002) 41. Wessel, J.E., Syage, J.A.: Excitonic interactions in naphthalene clusters. J. Phys. Chem. 94, 737–747 (1990) 42. Kusaka, R., Inokuchi, Y., Ebata, T.: Structure of hydrated clusters of dibenzo-18-crown-6-ether in a supersonic jet-encapsulation of water molecules in the crown cavity. Phys. Chem. Chem. Phys. 10, 6238–6244 (2008) 43. Ottiger, P., Köppel, H., Leutwyler, S.: Excitonic splittings in molecular dimers: why static ab initio calculations cannot match them. Chem. Sci. 6, 6059–6068 (2015) 44. Balmer, F.A., Kopec, S., Köppel, H., Leutwyler, S.: Excitonic splitting and vibronic coupling analysis of the m-cyanophenol dimer. J. Phys. Chem. A (2016) 45. Kida, M., Kub, M., Ujihira, T., Ebata, T., Abe, M., Inokuchi, Y.: Selective probing of potassium ion in solution by intramolecular excimer fluorescence of dibenzo-crown ethers. ChemPhysChem 19, 1331–1335 (2018) 46. Förster, T.: Excimers. Angew. Chem. Int. Ed. 8, 333–343 (1969) 47. Hirayama, F., Lipsky, S.: Excimer fluorescence of benzene and its alkyl derivatives—concentration and temperature dependence. J. Chem. Phys. 51, 1939–1951 (1969)

152

Y. Inokuchi

48. Yamauchi, A., Hayashita, T., Kato, A., Nishizawa, S., Watanabe, M., Teramae, N.: Selective potassium ion recognition by benzo-15-crown-5 fluoroionophore/γ-cyclodextrin complex sensors in water. Anal. Chem. 72, 5841–5846 (2000) 49. Patra, S., Gunupuru, R., Lo, R., Suresh, E., Ganguly, B., Paul, P.: Cation-induced fluorescent excimer emission in calix[4]arene-chemosensors bearing quinoline as a fluorogenic unit: experimental, molecular modeling and crystallographic studies. New J. Chem. 36, 988–1002 (2012) 50. Jarolimova, Z., Vishe, M., Lacour, J., Bakker, E.: Potassium ion-selective fluorescent and pH independent nanosensors based on functionalized polyether macrocycles. Chem. Sci. 7, 525–533 (2016) 51. Izatt, R.M., Terry, R.E., Haymore, B.L., Hansen, L.D., Dalley, N.K., Avondet, A.G., Christensen, J.J.: Calorimetric titration study of interaction of several univalent and bivalent-cations with 15-crown-5, 18-crown-6, and 2 isomers of dicyclohexo-18-crown-6 in aqueous-solution at 25 °C and mu = 0.1. J. Am. Chem. Soc. 98, 7620–7626 (1976) 52. Allen, F.H.: The cambridge structural database: a quarter of a million crystal structures and rising. Acta Crystallogr. Sect. B: Struct. Sci. 58, 380–388 (2002) 53. More, M.B., Ray, D., Armentrout, P.B.: Cation-ether complexes in the gas phase: bond dissociation energies of Na+ (dimethyl ether)x , x = 1 − 4; Na+ (1,2-dimethoxyethane)x , x = 1 and 2; and Na+ (12-crown-4). J. Phys. Chem. A 101, 831–839 (1997) 54. Armentrout, P.B., Austin, C.A., Rodgers, M.T.: Alkali metal cation interactions with 12-crown4 in the gas phase: revisited. Int. J. Mass Spectrom. 330, 16–26 (2012) 55. Dearden, D.V., Liang, Y.J., Nicoll, J.B., Kellersberger, K.A.: Study of gas-phase molecular recognition using fourier transform ion cyclotron resonance mass spectrometry (FTICR/MS). J. Mass Spectrom. 36, 989–997 (2001) 56. Brodbelt, J.S.: Probing molecular recognition by mass spectrometry. Int. J. Mass Spectrom. 200, 57–69 (2000) 57. Kempen, E.C., Brodbelt, J.S.: A method for the determination of binding constants by electrospray ionization mass spectrometry. Anal. Chem. 72, 5411–5416 (2000) 58. Franski, R.: Complexes of large crown ethers with the lithium cation studied by electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 23, 3488–3491 (2009) 59. Franski, R.: Cation-pi interactions in gas-phase complexes formed by benzo-crown ethers and alkali metal cations. Rapid Commun. Mass Spectrom. 25, 672–674 (2011) 60. Katritzky, A.R., Malhotra, N., Ramanathan, R., Kemerait, R.C., Zimmerman, J.A., Eyler, J.R.: Measurement of gas-phase binding-energies of crown ethers with metal-ions by fouriertransform ion-cyclotron resonance mass-spectrometry. Rapid Commun. Mass Spectrom. 6, 25–27 (1992) 61. Leize, E., Jaffrezic, A., Vandorsselaer, A.: Correlation between solvation energies and electrospray mass spectrometric response factors. Study by electrospray mass spectrometry of supramolecular complexes in thermodynamic equilibrium in solution. J. Mass Spectrom. 31, 537–544 (1996) 62. Malhotra, N., Roepstorff, P., Hansen, T.K., Becher, J.: Alkali-metal ion complexation of crown ethers and related ligands studied by Cf-252 plasma desorption mass-spectrometry. J. Am. Chem. Soc. 112, 3709–3710 (1990) 63. Peiris, D.M., Yang, Y.J., Ramanathan, R., Williams, K.R., Watson, C.H., Eyler, J.R.: Infrared multiphoton dissociation of electrosprayed crown ether complexes. Int. J. Mass Spectrom. Ion Processes 157, 365–378 (1996) 64. Schalley, C.A.: Supramolecular chemistry goes gas phase: the mass spectrometric examination of noncovalent interactions in host-guest chemistry and molecular recognition. Int. J. Mass Spectrom. 194, 11–39 (2000) 65. Schalley, C.A.: Molecular recognition and supramolecular chemistry in the gas phase. Mass Spectrom. Rev. 20, 253–309 (2001) 66. Wipff, G., Weiner, P., Kollman, P.: A molecular-mechanics study of 18-crown-6 and its alkali complexes: an analysis of structural flexibility, ligand specificity, and the macrocyclic effect. J. Am. Chem. Soc. 104, 3249–3258 (1982)

5 Gas-Phase Spectroscopy of Metal Ion–Benzo-Crown Ether Complexes

153

67. Van Eerden, J., Harkema, S., Feil, D.: Molecular dynamics of 18-crown-6 complexes with alkali-metal cations: calculation of relative free energies of complexation. J. Phys. Chem. 92, 5076–5079 (1988) 68. Dang, L.X.: Mechanism and thermodynamics of ion selectivity in aqueous-solutions of 18crown-6 ether—a molecular-dynamics study. J. Am. Chem. Soc. 117, 6954–6960 (1995) 69. Glendening, E.D., Feller, D., Thompson, M.A.: An ab initio investigation of the structure and alkali metal cation selectivity of 18-crown-6. J. Am. Chem. Soc. 116, 10657–10669 (1994) 70. Feller, D.: Ab initio study of M+ : 18-crown-6 microsolvation. J. Phys. Chem. A 101, 2723–2731 (1997) 71. Dang, L.X., Kollman, P.A.: Free energy of association of the 18-crown-6: K+ complex in water: a molecular dynamics simulation. J. Am. Chem. Soc. 112, 5716–5720 (1990) 72. Dang, L.X.: Free-energies for association of Cs+ to 18-crown-6 in water—a moleculardynamics study including counter ions. Chem. Phys. Lett. 227, 211–214 (1994) 73. Dang, L.X., Kollman, P.A.: Free-energy of association of the K+ 18-crown-6 complex in water— a new molecular-dynamics study. J. Phys. Chem. 99, 55–58 (1995) 74. Thompson, M.A., Glendening, E.D., Feller, D.: The nature of K+ crown-ether interactions—a hybrid quantum mechanical-molecular mechanical study. J. Phys. Chem. 98, 10465–10476 (1994) 75. Zwier, M.C., Kaus, J.W., Chong, L.T.: Efficient explicit-solvent molecular dynamics simulations of molecular association kinetics: Methane/methane, Na+ /Cl- , methane/benzene, and K+ /18-crown-6 ether. J. Chem. Theory Comput. 7, 1189–1197 (2011) 76. Hay, B.P., Rustad, J.R., Hostetler, C.J.: Quantitative structure stability relationship for potassium-ion complexation by crown-ethers—a molecular mechanics and ab-initio study. J. Am. Chem. Soc. 115, 11158–11164 (1993) 77. Poonia, N.S.: Coordination chemistry of sodium and potassium complexation with macrocyclic polyethers. J. Am. Chem. Soc. 96, 1012–1019 (1974) 78. Live, D., Chan, S.I.: Nuclear magnetic resonance study of the solution structures of some crown ethers and their cation complexes. J. Am. Chem. Soc. 98, 3769–3778 (1976) 79. Bajaj, A.V., Poonia, N.S.: Comprehensive coordination chemistry of alkali and alkaline-earth cations with macrocyclic multidentates—latest position. Coord. Chem. Rev. 87, 55–213 (1988) 80. Rodriguez, J.D., Lisy, J.M.: Probing ionophore selectivity in argon-tagged hydrated alkali metal ion-crown ether systems. J. Am. Chem. Soc. 133, 11136–11146 (2011) 81. Inokuchi, Y., Ebata, T., Rizzo, T.R., Boyarkin, O.V.: Microhydration effects on the encapsulation of potassium ion by dibenzo-18-crown-6. J. Am. Chem. Soc. 136, 1815–1824 (2014) 82. Inokuchi, Y., Ebata, T., Rizzo, T.R.: Solvent effects on the encapsulation of divalent ions by benzo-18-crown-6 and benzo-15-crown-5. J. Phys. Chem. A 119, 8097–8105 (2015) 83. Inokuchi, Y., Ebata, T., Rizzo, T.R.: Microhydration of dibenzo-18-crown-6 complexes with K+ , Rb+ , and Cs+ investigated by cold UV and IR spectroscopy in the gas phase. J. Phys. Chem. A 122, 3754–3763 (2018) 84. Shubert, V.A., James, W.H., Zwier, T.S.: Jet-cooled electronic and vibrational spectroscopy of crown ethers: Benzo-15-crown-5 ether and 4 ‘-amino-benzo-15-crown-5 ether. J. Phys. Chem. A 113, 8055–8066 (2009) 85. Kusaka, R., Inokuchi, Y., Ebata, T.: Water-mediated conformer optimization in benzo-18crown-6-ether/water system. Phys. Chem. Chem. Phys. 11, 9132–9140 (2009)

Part III

Spectroscopy and Characterization of Metal Clusters

Chapter 6

Metal Cation Coordination and Solvation Studied with Infrared Spectroscopy in the Gas Phase Michael A. Duncan

Metal ion complex infrared spectroscopy and theory to determine structures and coordination numbers

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_6

157

158

M. A. Duncan

Abstract Transition metal cation–molecular complexes are produced in the gas phase environment of a molecular beam using laser ablation in a supersonic expansion. Complexes with carbon monoxide, carbon dioxide, water, acetylene, or benzene are produced by entraining small partial pressures of these molecules into an expansion of either argon or helium. A specially designed time-of-flight mass spectrometer is used to analyze the ions produced and to mass-select them for spectroscopy. Mass-selected ions are excited in the infrared region of the spectrum with a tunable IR optical parametric oscillator laser system to measure photodissociation spectroscopy in the 2000–4500 cm−1 region. Infrared band patterns, combined with structures and spectra predicted by density functional theory, reveal the coordination and solvation interactions in these systems, and how binding to metal distorts the structures of small molecules. Keywords Ion–molecule complexes · Mass spectrometry · Ion spectroscopy · Photodissociation

6.1 Introduction Metal–molecular interactions lie at the heart of heterogeneous [1–5] and homogeneous [6, 7] catalysis, metal–ligand bonding [7–11], metal ion solvation [12–14], metal chelation and sequestration [15], and the function of many biological systems [16]. Additionally, new composite materials, such as metal-decorated nanotubes, metal-intercalated graphene, or metal-organic frameworks (MOFs), involve many of the same metal–molecular interactions [17–26]. These areas are critically important in petroleum processing, solar energy generation, hydrogen storage, battery materials, and related areas such as water splitting, CO2 reduction, or heavy metal waste disposal. However, the molecular-level understanding of such systems is limited because of the complexity of metal electronic structure and bonding. Conventional chemistry has documented the properties of stable metal complexes and compounds [7–11]. Likewise, heterogeneous catalysis has been studied extensively on well-characterized metal surfaces [1–3]. However, emerging catalytic systems often involve oxide-supported clusters in the ultra-small size range, with a distribution of particle sizes [4, 5, 27–32]. Homogeneous catalysis is mediated by metal complexes in solution with a delicate relationship between coordination and solvation [33–36]. Metal-organic and metal-carbon materials involve cation–π interactions [17–19], and metal ion solvation involves many subtleties of covalent versus electrostatic interactions [37–43]. Unfortunately, detailed insights into this rich and varied chemistry are often limited because theory and experiments cannot study the same systems in the same environment. Isolated metal complexes provide model systems, more tractable for theory, that can elucidate key interactions. Careful investigations of elecM. A. Duncan (B) Department of Chemistry, University of Georgia, Athens, GA 30602, USA e-mail: [email protected]

6 Metal Cation Coordination and Solvation Studied …

159

tronic structure, geometries, bonding energies, and reactivity are therefore possible. As discussed in this chapter, our research focuses on these model metal–molecular complexes and their clusters using molecular beam sources, mass spectrometers, and infrared laser spectroscopy, in combination with computational chemistry. To investigate metal systems containing a specific composition, we study ionized clusters and complexes which can be mass-selected. Isolated metal centers or those with specific numbers of attached ligand or solvent molecules can be produced and studied. Transition metal ion–molecule complexes have been studied with mass spectrometry for many years, providing reaction products and rates, as well as bond energies [44–57]. While these data are valuable, spectroscopy is needed on these systems, evaluated with corresponding computational studies, to make real progress in the understanding of metal electronic structure and bonding. Vibrational spectroscopy provides the best probe of structure and bonding for metal complexes, and both IR and Raman spectroscopy have been used for many years in this area [58, 59]. However, although these methods are straightforward for conventional inorganic complexes [58], and can be adapted for adsorbates on surfaces [59], they are not easily applied to low-density samples in the gas phase. Vibrational information can be obtained via electronic [60–81] or photoelectron spectroscopy [82–97], but IR spectra can be compared more directly to the predictions of theory. Small metal ions have been studied with infrared absorption spectroscopy in rare gas matrices [98], but the identification of the spectral carrier in these experiments can be ambiguous. Ionized complexes in the gas phase can be size-selected with mass spectrometers, but the resulting density is too low for absorption spectroscopy. Ion spectroscopy is often further complicated by the conditions in ion sources, generally involving discharges, hot plasmas, or other forms of energetic excitation. Until recently, these issues severely limited the IR spectroscopy of ions. However, much recent progress has been made in this area [99–118]. Improved ion sources using laser ablation or electrospray ionization (ESI) now produce a wide variety of metal–ligand and metal–solvent complexes. Ion cooling, needed for sharp spectra, has been implemented with supersonic expansions or cryogenic ion traps. Sensitivity limitations have been addressed by using laser photodissociation spectroscopy rather than absorption. Finally, new IR lasers provide intense sources with broad tuning ranges to access the full-vibrational spectrum [119, 120]. Infrared spectroscopy of gas phase ions is now a rapidly expanding area of research, in which our group has been actively engaged. Our experiments use laser vaporization in pulsed-nozzle/supersonic molecular beam sources to produce cold metal-containing ions [121]. Although other groups use electrospray ionization (ESI) sources, we find the laser source to be better suited to the ions we study. The ions produced are analyzed and mass-selected with a specially designed reflectron time-of-flight spectrometer (RTOF) [122]. This instrument provides high throughput for maximized ion density, while maintaining the cold temperatures produced by the source. The selected ions are spatially bunched at the turning point in the reflectron field to optimize overlap with the laser. Excitation here allows the full mass spectrum to be detected for each laser shot so that different fragment channels can be recorded simultaneously. The experiment uses the high-intensity,

160

M. A. Duncan

broadly tunable, optical parametric oscillator/amplifier (OPO/OPA) laser sold by LaserVision [119]. Its main configuration covers the range of 2000–4500 cm−1 , corresponding to the higher frequency vibrations of small ligand or solvent molecules. A second OPA configuration uses silver-gallium-selenide crystals, extending the tuning range to the 600–2300 cm−1 region [120]. Here, we study lower frequency ligand or solvent vibrations, as well as the M–O stretches of oxide clusters. The two laser configurations allow almost full coverage of the infrared spectrum. Combined with our ion source and mass spectrometer, these lasers have produced spectroscopy for many transition metal–molecular complexes [106, 107, 123–169]. This work is complemented by that of other groups using similar infrared OPO laser systems [108–118, 170–200] or free-electron lasers (FELs) [201–212]. This kind of mass spectrometry combined with IR spectroscopy provides the coordination numbers, geometries and electronic structures of metal–molecular complexes. Mass spectra and photodissociation patterns reveal the number of ligand or solvent molecules attached to a specific metal center. The vibrations in these complexes typically occur near those of the corresponding free ligand or solvent molecules, indicating that binding usually takes place without dissociation or insertion chemistry. The number of IR bands, their shifts compared to the vibrations of the isolated ligand or solvent molecules, and the relative band intensities provide distinctive patterns that can be compared with the predictions of theory to determine structures. Density functional theory (DFT) is used in these studies, although we use due caution in its applications. For example, a well-known issue in transition metal complexes is determination of the correct spin configuration giving rise to the ground state [213–219]. DFT has trouble with the relative energies of spin states, but usually predicts a valid infrared pattern for each spin state. The measured vibrational patterns are then compared with the (scaled harmonic) predictions of theory for different electronic states. These vibrational patterns, rather than the computed relative energies, are generally good enough to determine the spin of the ground state or to reveal the presence of more than one electronic state. Infrared patterns can also reveal the occurrence of intracluster reactions through the appearance of new spectra corresponding to reaction products. A key aspect of this work is the ligand or solvent molecule binding energy and our ability to cause fragmentation with IR photons. Dissociation energies for many metal ion complexes are known via methods such as collision-induced dissociation [44–55]. Bond energies range from 5000 cm−1 (12–15 kcal/mol) for electrostatically bonded metal–CO2 ions, up to as high as 30,000 cm−1 (70–80 kcal/mol) for metal–benzene complexes with strong covalent bonds. Across this range, one photon infrared excitation on vibrational fundamentals is not energetic enough to cause photodissociation. Bond energies usually decrease in complexes with more ligands, but these systems have the same problem until the metal coordination is completed. However, when ligands are present beyond the inner-sphere coordination, they are bound by weaker electrostatic forces and their elimination is efficient, providing good spectra. The onset of greater dissociation yields therefore identifies the coordination number. To study smaller complexes with partial coordination and stronger bonding, we attach weakly bound rare gas atoms, using the “tagging” method first

6 Metal Cation Coordination and Solvation Studied …

161

used by Lee et al. [99–101]. This general method is now used throughout ion spectroscopy [102–118]. To document the effects of tagging, we use theory to investigate the spectra of complexes with and without the tag atom. Another essential requirement for this work has been the extension of IR laser coverage to the fingerprint region. In the past, this could only be done using FEL’s, such as the FELIX system in the Netherlands. Previous studies there by our group and others used infrared resonance-enhanced multiphoton dissociation (IR-REMPD) of cations to obtain vibrational spectra [201–212]. However, the quality of these spectra is often poor because of the laser line width and power broadening from the multiphoton processes. The new OPO’s have broader wavelength coverage (600–4500 cm−1 ) and higher resolution. Using tagging, or elimination of external ligands, spectra can be measured via single-photon dissociation, and line widths are much improved (1–5 cm−1 , limited by predissociation). As shown in Fig. 6.1, the quality of the spectra obtained is excellent. In this chapter, we describe the work from our lab investigating metal ion complexes with carbon monoxide [143–155], carbon dioxide [123–128], water [129–142], acetylene [156–162], and benzene [165–168]. These experiments show that IR photodissociation spectroscopy can be applied to ions containing virtually any metal or ligand. It provides the dissociation products, the number of IR-active vibrations, the frequency shifts that occur when ligands bind to metal, and the rela-

Fig. 6.1 The infrared spectra of V+ (CO2 )6 complexes, illustrating the broad tunability of our IROPO lasers. The CO2 bend and asymmetric stretch have two bands corresponding respectively to molecules coordinated to the metal ion and those attached only to other CO2 molecules. The symmetric stretch is only IR-active for molecules attached to the metal and a single band is observed here

162

M. A. Duncan

tive intensities of different bands. Computational chemistry complements the experiments, predicting the structures of complexes, their electronic configurations, and their spectra.

6.2 Metal-Carbonyl Complexes Transition metal carbonyls provide classic examples of inorganic complexes [7–11], and CO is the classic probe molecule for surface science and catalysis [1–3, 58]. In both contexts, the carbonyl stretch reveals the nature of the bonding. Metal carbonyls are characterized by the positions of the C–O stretch relative to the vibration of the isolated CO diatomic (2143 cm−1 ) [220]. Unsaturated carbonyls, including ions, have been investigated by mass spectrometry [45–55, 221–226], matrix isolation spectroscopy [98], and photoelectron spectroscopy [82, 83, 85, 87]. Computational studies have explored the mechanism of the shifts that occur for the C–O stretches in different systems, including the familiar effects of σ donation and π backbonding [219, 227–235]. Our group has investigated transition metal carbonyl cations [143–155] to compare these to well-known neutrals. Other research groups have used similar methods to investigate other atomic metal cation–carbonyls [190–197] or the carbonyls of metal atom clusters [205–209]. We first studied cation complexes that could provide isoelectronic analogs to known neutral metal carbonyl complexes. The Co+ (CO)5 complex provided an analog to Fe(CO)5 [145], Mn+ (CO)6 provided an analog to Cr(CO)6 [147], and Cu+ (CO)4 provided an analog to Ni(CO)4 [150]. In each case, the cations were found to have the same coordination and structures as the corresponding neutrals (trigonal pyramid, octahedral, and tetrahedral, respectively) and the same closed-shell singlet ground states. However, significant differences were apparent in the spectroscopy between the neutrals and the corresponding cations. In the neutrals, the C–O stretch vibrations are strongly red-shifted compared to the stretch of molecular CO by 100–150 cm−1 . However, as shown in Fig. 6.2, the C–O stretch vibrations were hardly shifted at all for the Co and Mn cations and they were blue-shifted for the Cu cations. The frequencies for these neutrals and ions are summarized in Table 6.1. It is well known in inorganic chemistry that the shifts of the C–O stretches arise from the competing effects of σ donation and π back-bonding [219, 227–235]. For most neutral metals, π back-bonding is the more significant factor, and the carbonyl stretches occur at much lower frequencies than that of CO itself. The smaller red shifts seen for the cations here are attributed to their reduced π back-bonding [149]. Blue shifts to higher frequencies are known to occur for certain metals with filled d shells that are inefficient at back donation. We find this behavior not only for the copper carbonyl cations shown here [150], but also for Au+ , Pt+ and Rh+ carbonyl complexes [143, 144, 155]. The fully coordinated ions and their corresponding neutrals shown in Table 6.1 all have the 18-electron configuration, which is recognized as a guiding principle in transition metal chemistry. We were also interested to see how robust this rule

6 Metal Cation Coordination and Solvation Studied …

163

Fig. 6.2 The infrared spectra of the saturated carbonyl complexes for selected metal cations. The red dashed line shows the frequency of the isolated CO molecule

Table 6.1 A comparison of the carbonyl stretch frequencies for isoelectronic neutral and cationic complexes

Complex

Experimental C–O stretch (cm−1 )

Cr(CO)6

2003 [236]

Mn+ (CO)6

2115

Fe(CO)5

2013, 2034 [237]

Co+ (CO)5

2140, 2150

Ni(CO)4

2056 [238]

Cu+ (CO)4

2193

is, and what its limitations are, if any. Early transition metals have fewer valence d electrons, and therefore would require more carbonyl ligands to achieve the 18electron configuration. We investigated the cation carbonyls of the group V metals (V, Nb, Ta) [146, 151], which would need seven carbonyls to reach this limit, and those of Sc and Y [154], which would need eight carbonyls, to see if such higher coordination numbers were possible. According to theory, these higher coordination complexes are stable for each of these metals. However, we found experimentally that vanadium did not form the seven-coordinate (7C) carbonyl, but instead formed the six-coordinate (6C) complex (see spectrum in Fig. 6.2). Niobium and tantalum, on the other hand, did form the 7C complexes. The spectrum of Ta+ (CO)7 , which forms a capped octahedral structure rather than the pentagonal bipyramid, is presented in Fig. 6.3. Likewise, scandium did not form the 8C complex, but yttrium did. In both groups, only the heavier metals formed the expected high coordination. We explained

164

M. A. Duncan

Fig. 6.3 The infrared spectrum of Ta+ (CO)7 compared to the spectra predicted by density functional theory for two different isomers. The spectrum agrees with that predicted for the capped octahedron structure

this trend in terms of the kinetics of carbonyl addition to these metals. In both systems, the filled coordination produces a singlet ground state, while the n − 1 complex is a triplet. Adding the last CO therefore requires a spin change, which may inhibit the rate of this process. Our clusters grow in a 1–3 ms time frame (defined by the pulsed jet expansion) by sequential addition of ligands to the ablated metal cations, and therefore slower growth rates may inhibit the formation of complexes, even if they are stable. The heavier metals with stronger spin-orbit coupling should change spin more readily, possibly explaining how these species could achieve the higher coordination. This reasoning was used previously by Weitz et al. to explain similar results for CO addition to unsaturated neutral carbonyls [239, 240]. Harvey et al. used computational studies to model these spin-controlled kinetics [241–243]. Since our work on these systems, Zhou and coworkers have found eight-coordinate complexes for other early transition metal cations [196], and they have found unexpected 8C complexes for the alkaline earth metal cations [197]. The group IV metals (Ti, Zr, Hf) all have an odd number of electrons as cations, and it is therefore unclear what coordination would be expected for these systems. We found that they all formed 6C complexes rather than the 7C (17-electron) or 8C (19-electron) species [152, 195]. Rhodium carbonyls provided another interesting

6 Metal Cation Coordination and Solvation Studied …

165

case [155]. Rh+ is a d8 species, which is generally expected to form 4C square planar complexes [11], even though the 18-electron species would be 5C. We found a primary coordination of four carbonyls but a secondary coordination of five. The fifth ligand had an intermediate binding energy, weaker than the first four, but not as weak as the external ligands. The spectrum of the n = 4 complex indicated a square-planar structure, while that of the n = 5 species was a square pyramid. We have also studied metal oxide carbonyls. Oxidation of the metal center removes the d electron density available for back-bonding, which reduces or eliminates the red shifts in these systems. The V+ (CO)6 complex spectrum has a slightly red-shifted C–O stretch, with one main band because of the octahedral structure [146, 151]. VO+ (CO)5 , VO2 (CO)+4 , and VO3 (CO)+3 (each measured by elimination of one excess CO from the next larger complex) have C–O stretches shifted progressively further to the blue because of the reduced back-bonding (Fig. 6.4) [153]. Similar blue-shifted carbonyl stretches are observed for CO binding on metal oxide surfaces [2, 59]. The metal–oxygen stretches in these clusters can be compared with those of the corresponding VO+ , VO2 + , and VO3 + ions recently measured by Asmis et al. [244]. The oxide stretches in the carbonyls shift to the red compared to those in the isolated oxides, another result of partial charge transfer in these systems.

Fig. 6.4 The IR spectra of vanadium oxide carbonyls and the structures predicted to be most stable for these complexes. Each complex has one external ligand, which is eliminated in the photodissociation process to obtain the spectrum. The C–O stretch vibrations are shifted to the same or higher frequencies than the free-CO vibration, indicated as the dashed red line. The bands at 2169–2174 cm−1 come from the external CO ligand

166

M. A. Duncan

6.3 Metal–CO2 Complexes The binding of CO2 to metals is of widespread importance for CO2 capture and catalytic conversion to small organics [20, 21, 25, 34, 35]. However, the binding of CO2 to metal ions in the gas phase involves primarily electrostatic interactions [245–247], and these systems have been studied less than the corresponding carbonyl complexes. Our lab has studied the electronic spectroscopy of Mg+ (CO2 ) [65] and Ca+ (CO2 ) [67], and the infrared photodissociation spectroscopy of several M+ (CO2 )n cation complexes [123–127]. Figure 6.1 shows the infrared spectrum of V+ (CO2 )6 , which illustrates the behavior seen for many of these systems. Of the normal modes for CO2 , the degenerate bending mode (ν2 , 667 cm−1 ) and the asymmetric stretch (ν3 , 2349 cm−1 ) are IR active for the isolated molecule, whereas the symmetric stretch (ν1 , 1333 cm−1 ) is inactive [248]. However, theory and experiments agree that metal cation binding is most favorable in the M+ –O=C=O linear configuration. In this structure, all three vibrations are IR active. As shown in the figure, CO2 molecules attached to metal have these three vibrations, with shifts to higher frequencies than the vibrations of the isolated molecule. In a cluster like V+ (CO2 )6 , there are coordinated molecules, which give rise to shifted vibrations, and second-sphere molecules not attached to the metal whose vibrations are mostly unshifted. This gives rise to doublet features (shifted plus unshifted bands) for the asymmetric stretch and bending vibrations, and a single band (shifted only) for the symmetric stretch, which is only IR-active when it is attached to the metal. This pattern of bands has been seen for almost all the metal ion–CO2 complexes that we have studied. An exception to this general behavior occurs in larger V+ (CO2 )n clusters, in which additional bands were seen beyond the coordination and solvation features. As shown in Fig. 6.5, new bands at 1140, 1800, 2402, and 3008 cm−1 were seen for clusters with seven or more CO2 ligands, and these bands became more intense in the larger clusters. These new bands suggest that there was an intracluster reaction producing a new kind of structure. Because the clusters were mass-selected, the reaction product must have the same mass as one or more CO2 units, which could be true for an oxide-carbonyl species, VO+ (CO)(CO2 )n−1 , a metal carbonyl-carboxylate species, V+ (CO)(CO2 )n−1 (CO3 ), or a metal oxylate species, V+ (CO2 )n−2 (C2 O4 ). To explore these possibilities, we made the oxide-carbonyl species direct, and found that the VO+ and carbonyl stretches in these systems (Fig. 6.4) do not match the new bands. Instead, we found that the reaction product is an oxalate species (C2 O4 – ), with covalently linked CO2 molecules. The lower frequency vibrations (1140 and 1800 cm−1 ) are those of the oxalate moiety, and the higher frequency vibrations (2402 and 3008 cm−1 ) are those of solvating CO2 molecules interacting with the new kind of charge center in the clusters. Although we cannot determine the exact charge states in this system, oxalate is most stable when it carries a negative charge. This reaction therefore apparently occurs by electron transfer from the V+ ion to CO2 , producing a V2+ , C2 O4 − ion pair. This suggestion would explain the onset at larger cluster sizes. Solvation from the surrounding excess CO2 molecules could stabilize the higher charge state of V2+ and that of the oxalate. Theory on this system is plagued

6 Metal Cation Coordination and Solvation Studied …

167

Fig. 6.5 The infrared spectrum of V+ (CO2 )7 compared to that for V+ (CO2 )6 showing the sudden appearance of several new vibrations associated with an intracluster reaction

by multireference issues, and we were not able to determine whether the ion pair is in contact or solvent-separated. However, the same kind of chemistry has also been seen by Weber and coworkers [174] for negative ion M− (CO2 )n clusters. Apparently, the negative charge on CO2 activates it to enable a rich variety of chemistry.

6.4 Metal–Water Complexes The interaction of water with metal ions is fundamental to the chemistry of solvation [12–14, 37–43]. Unfortunately, the details of cation–water interactions are difficult to obtain from solution measurements, which involve ensemble averaging over many structures. Gas phase measurements have investigated the thermochemistry of cation–water bonding [47, 48, 50, 54, 55, 249–264], and computational studies have studied structures and energetics of these systems [57, 265–277]. However, infrared spectroscopy probes the structures of these systems more directly. Our work has examined several M+ (H2 O)n and M2+ (H2 O)n systems [129–142], focusing on both the mono-hydrated complexes and the coordination behavior when multiple water molecules condense around the metal ion. Other groups have also explored the same kinds of systems using similar methods [170–173, 178–181, 183–187, 192]. We have studied nearly all of the singly charged first-row transition metals in complexes with a single water molecule [129–142]. The spectra in the O–H stretching

168

M. A. Duncan

region are shown in Fig. 6.6. The binding energies of argon are very different for the early versus late transition metals. Consequently, the late transition metals require the attachment of two or more argon atoms before photodissociation can be measured in this region of the IR. As shown in the left frame of the figure, those complexes tagged with a single argon have more complex vibrational patterns than those tagged with two argons. The additional structure at higher frequency arises from partially resolved rotational structure (K-type bands) on the asymmetric stretch band. This structure is discussed in more detail later. We found that water bound to metal ions generally has O–H stretching frequencies that are shifted to the red compared to those of the free molecule (3657 and 3756 cm−1 for the symmetric and asymmetric stretches, respectively) [248]. In a charge-transfer process not unlike that for metal carbonyls, the cation polarizes water, removing electron density from its highest occupied molecular orbital. This orbital involves not only the non-bonding lone pairs on oxygen but also has bonding character along the O–H bonds; weakening these bonds lowers the vibrational frequency. Figure 6.7 shows the electron density map of the Ti+ (H2 O) complex in its doublet ground state compared to that in the separated

Fig. 6.6 The infrared spectra in the O–H stretching region for different transition metal cation–water complexes compared to the symmetric and asymmetric O–H stretch frequencies for the isolated water molecule (dashed red lines). The early transition metal complexes are tagged with a single argon, producing partially resolved rotational structure in some cases, whereas the late-transition metal complexes are usually tagged with two or three argons. The spectra for the early transition metal complexes are generally shifted further to the red than those of the late transition metals. The lowest frequency band for Fe+ (H2 O) is from an isomer with argon attached to an OH of water, inducing an even greater red shift [130]

6 Metal Cation Coordination and Solvation Studied …

169

Fig. 6.7 The charge density map of the Ti+ (H2 O) complex in its doublet ground state. Red color shows an increase in charge density compared to the separated cation–water system, whereas blue shows a reduction in charge density

cation–molecule system. This illustrates the effects of the charge transfer between the water and the metal. As shown in the two views, electron density increases on the metal ion center, and decreases in the vicinity of the O–H bonds. This charge transfer drives the shift in the vibrational frequencies. In addition to the red shift in the frequencies—which varies considerably with different metals—the relative intensities of the two O–H stretches change, with the symmetric stretch gaining relative to the asymmetric. In the free water molecule, the asymmetric/symmetric stretch intensity ratio is about 18:1, whereas in the cation water complexes this ratio is closer to 1:1. The symmetric stretch in these metal complexes oscillates charge more effectively along the molecular axis, enhancing the dynamic dipole and the IR intensity. The shifts seen for these singly charged metal complexes have been compared with selected examples of doubly charged complexes [135, 137–139]. In those systems, the shifts of the vibrational frequencies and the enhancement of the symmetric stretch intensity are both greater than that for the corresponding singly charged complexes. Interestingly, the shifts of the O–H stretching frequencies measured are generally greater for the early transition metals than they are for the late transition metals. Figure 6.8 shows a plot of the O–H stretch frequencies across the periodic table groups and a comparison to the corresponding M+ –(H2 O) bond energies determined in other labs. Surprisingly, the magnitudes of the red shifts for the two O–H stretches

170

M. A. Duncan

Fig. 6.8 Plots of the shifts in the O–H stretches for different transition metal cation–water complexes compared to the cation–water binding energies

of water are greater for the early transition metal, and less for the late transition metals, with a local maximum for the manganese cation. The binding energies are greater for the late transition metals. It therefore seems that there is no clear correlation between binding energies and vibrational band shifts, even though the charge transfer that causes the vibrational shifts should have at least some relevance for the electrostatic bonding in these systems. However, the bonding in these transition metal–water complexes is a complex mixture of both electrostatic and covalent interactions, and so it may be oversimplified to assume a correlation between these two properties. It is worth noting that density functional theory accurately predicts both the trends in binding energies and vibrational frequency shifts. In Ar–M+ (H2 O) complexes when the tag atom binds opposite water, the complex has C2v symmetry and is nearly a symmetric top, with only the light hydrogen atoms located off the C2 symmetry axis. This causes the A rotational constant to be relatively large (>10 cm−1 ) and K-type rotational sub-band structure can be resolved, even with our modest 1 cm−1 laser line width. This is apparent in the spectra for the early transition metals, in the left frame of Fig. 6.6, with the exception of the Mn+ (H2 O) spectrum (it binds argon in a bent position, which produces a much smaller A constant, and the structure is not resolved [138]). In these systems, a multiplet structure arises

6 Metal Cation Coordination and Solvation Studied …

171

for the asymmetric stretch, which is a perpendicular-type band. The symmetric stretch is a parallel-type band, with more closely spaced rotational structure that cannot be resolved under these conditions. Because the hydrogen molecules of water are equivalent by symmetry, ortho–para symmetry rules must be applied, resulting in a 3:1 statistical weight for transitions originating in the K = 1 versus K = 0 levels. At low temperature, only the K = 0 and 1 levels are populated significantly, and K = 1 cannot relax to K = 0 because of the nuclear spin symmetry. The only transitions seen are those originating from these two levels. The K = 0 → K = 1 transition (labeled 0,1 in the figure) is then lower in relative intensity than the K = 1 → K = 0 or K = 1 → K = 2 transitions (labeled 0,1 and 2,1 in the figure). The rotational structure can be simulated using the PGopher software [278], and the parameters are adjusted to get the best match with the experiment, as shown in Fig. 6.9. The best fit produces the A rotational constant and the temperature of the ions. As shown in the figure, the non-equilibrium conditions of the supersonic molecular beam produces slightly different temperatures for the J and K quantum states, an effect that is not uncommon in such molecular beam experiments. Assuming that the O–H bond distances remain nearly constant (suggested by theory), then the A rotational constant reveals the H–O–H angle, which is often expanded by the cation–water polarization interaction. In the scandium example shown here, this angle is estimated to be 107.13°, which is significantly larger than the angle in an isolated water molecule (104.7°). Our rotationally resolved studies on the Sc+ , Ti+ , V+ , Nb+ , and Cr+ systems all found H–O–H angles expanded with respect to that of water [135, 137, 141, 142]. In the case of the vanadium and niobium complexes, the analysis of the rotational structure was complicated by an unexpected quenching of the ortho–para separation catalyzed by the metal ions, changing the selection rules and the appearance of the spectra [142]. IR spectroscopy of metal cations solvated by multiple water molecules can reveal their coordination numbers. In small clusters, water coordinated directly to metal has free O–H stretching vibrations near those of the isolated water molecule. However, when water adds to the second sphere, hydrogen bonding causes a strong red shift of 200–400 cm−1 in the O–H stretches, and the IR intensity increases. The first appearance of vibrations in the hydrogen bonding region therefore identifies the coordination number for the metal cation. We found in the past that this is four water molecules for Ni+ [132], and three for Zn+ [140]. Figure 6.10 shows spectra for different sizes of V+ (H2 O)n , in which the first evidence for a band in the hydrogen bonding region occurs for the n = 4 cluster, indicating that the coordination is complete with three molecules [278]. In related work, Nishi et al. studied V+ (H2 O)n complexes without tagging, finding a coordination of four molecules [172]. Our result here can be rationalized to agree with their result, if we assume that argon acts as a coordinating ligand in at least some of the n = 4 complexes. V+ ions exhibited a coordination of six for carbonyl ligands [146, 151] and four for CO2 ligands [127], contrasting with the behavior seen here for water. Coordination numbers for the single positive ions we have studied are generally lower than those expected for

172

M. A. Duncan

Fig. 6.9 An expanded view of the IR spectrum in the O–H stretching region for scandium–water cations compared to a simulated spectrum including partially resolved rotational structure. The rotational structure is consistent with expectations for a C2v structure, with a triplet for the asymmetric stretch and a 3:1 intensity alternation from the nuclear spin statistics. A, B, and C are the rotational constants in the ground ( ) and excited ( ) vibrational states, and B.O. indicates the band origins. B and C values come from the theoretical structure, whereas the A values are adjusted to fit the spacings in the spectrum. The temperature is adjusted to fit the relative band intensities (TK ) and line widths from unresolved structure (TJ ) in the spectrum. Figure used from Ref. [137] with permission from the American Institute of Physics, Copyright 2011

the more highly charged metal ions found in normal solutions. The highly charged metal ions in solution have fewer electrons occupying the valence orbitals than the singly charged species. It is likely that ligand–electron repulsion from the occupied orbits causes the lower coordination numbers for the singly charged species.

6 Metal Cation Coordination and Solvation Studied …

173

Fig. 6.10 Infrared spectra for V+ (H2 O)n clusters in the n = 1–4 size range. The first evidence for a band in the hydrogen bonding region occurs for the n = 4 cluster, with the band at 3504 cm−1 marked with the red arrow. This suggests that the solvation sphere is filled with the next-smallest n = 3 cluster

6.5 Metal–Acetylene Complexes Metal–acetylene and metal–ethylene complexes form the simplest examples of cation-π interactions relevant in many areas of catalysis and biological chemistry [6–11, 16, 279–282]. These systems have been studied often in ion chemistry and investigated with computational chemistry [283–288]. In some of the first spectroscopic work, our group measured electronic spectra for Ca+ (C2 H2 ) and Mg+ (C2 H2 ) complexes [69, 70]. In the infrared, we investigated the C–H stretches in several transition metal ion complexes with a single acetylene [157], comparing the vibrations to the known symmetric and asymmetric stretches of acetylene (3374 and 3289 cm−1 , respectively) [248]. Figure 6.11 shows a comparison of several M+ (C2 H2 ) complexes, including new examples from more recent work. As shown in the figure, all the C–H stretches for these metal ion complexes occur at frequencies lower than those of acetylene itself. The cation–π interaction transfers charge from the molecule to the metal in much the same way seen already for metal-carbonyls and metal–water complexes. In acetylene complexes, polarization removes electron density from the C–C and C–H bonds, lowering their frequencies. The C–C and symmetric C–H stretches

174

M. A. Duncan

Fig. 6.11 Infrared spectra of cation–acetylene complexes in the C–H stretching region

of acetylene are not IR active in the free molecule, whereas the asymmetric C–H stretch is IR-active. However, in cation–acetylene complexes, the C–C and symmetric C–H stretches can become weakly IR active from the distortion of the molecule (e.g., CH groups bending away from linear) or the changing dipole produced by concerted metal and molecular motion. Consequently, the spectra shown in Fig. 6.11 have stronger asymmetric stretch vibrations at lower frequency and weaker symmetric stretch bands at higher frequency. The intensity of the weaker symmetric stretch band varies for different metals depending on the degree of “activation” induced by the metal. The exception to this trend is the V+ (C2 H2 ) complex, which has two bands with nearly equal intensities. This suggests that the bonding in this complex is somehow different from that in the other species considered here. Computational studies were insightful for these systems. We found that most metals form cation–π complexes, with the cation in a two-fold position above the π cloud and some slight bending of the CH groups away from the metal. However, the V+ complex formed a very different structure—that of a VC2 metallacycle with the

6 Metal Cation Coordination and Solvation Studied …

175

CH groups bent strongly away from the metal (computed CCH angle = 37.7°). In this bent configuration, both C–H stretches are IR active with comparable intensity and there are much greater red shifts in the two frequencies, all consistent with the experimental spectrum. There are covalent bonds between the metal and the carbons, and the C–C bond has lengthened (computed from 1.199 in acetylene to 1.301 Å in V+ -acetylene), consistent with its reduced bond order. The interaction between V+ and acetylene is clearly very different from that of the other metal ions studied so far. Complexes with multiple acetylene molecules coordinated to a single metal ion make it possible to investigate the coordination sphere and possible reactions between ligands mediated by the metal. In the case of multiple acetylene complexes of Ni+ , a coordination of four acetylenes was determined in a near-tetrahedral structure [158]. In larger clusters, a new band appeared which indicated an intracluster reaction forming cyclobutadiene [156]. In recent work, we examined the multiple acetylene complexes of Cu+ , finding an inner coordination of three acetylenes and a secondary solvation of three additional acetylenes in the highly symmetric Cu+ (C2 H2 )6 complex [159]. In this structure, whose spectrum is shown in Fig. 6.12, each acetylene molecule in the second coordination sphere is bonded to two inner-sphere molecules via bifurcated CH–π hydrogen bonds. Because of the highly symmetric structure, the IR spectrum has only two bands corresponding to the in-phase and out-of-phase asymmetric stretches of the core (3172 cm−1 ) and outer (3258 cm−1 ) ligands. Gold

Fig. 6.12 The infrared spectrum of Cu+ (C2 H2 )6 and the spectrum predicted by theory for the structure with three inner-sphere and three outer-sphere molecules. Figure used from Ref. [159] with permission from the American Chemical Society, Copyright 2015

176

M. A. Duncan

cation also forms a three-fold inner-sphere coordination, but with less symmetric second-sphere structures [160]. As noted earlier, the interaction between vanadium ions and acetylene molecules is quite different from that of the other transition metals, prompting us to examine its behavior as multiple acetylenes are added around the metal. Figure 6.13 shows some of the spectra and structures obtained. The di-acetylene complex forms a bow–tie structure, with each of the two acetylenes bound in a three-membered ring metallacycle like that seen for the mono-acetylene complex. When three acetylenes are added, the spectrum becomes more complex, with several more vibrational bands spread over a wider frequency region. Additionally, the spectrum varies with the concentration of acetylene added to the experiment. The third trace down in the figure shows the spectrum measured at lower acetylene concentration (2.5% in argon), while the lower trace shows the spectrum measured with higher concentration (15%). The multiband spectrum at lower concentration can be assigned to two isomers, primarily the one shown with both three- and five-membered metallacycle rings, and a Fig. 6.13 The infrared spectra and structures formed from the addition of multiple acetylene ligands around vanadium ions. The third trace down shows the spectrum for V+ (C2 H2 ) at low concentration, while the bottom trace shows the same mass ion when acetylene is added at higher concentration. Cyclization chemistry occurs, which eventually forms benzene

6 Metal Cation Coordination and Solvation Studied …

177

secondary one with three, three-membered metallacycle rings. The single band in the lower spectrum is assigned to the V+ (benzene) complex! This is predicted by theory, which shows that this isomer is the most stable for this composition, and it can also be confirmed by producing the same mass ion directly from benzene and measuring its spectrum, which is identical to that shown here. The structures mentioned for the spectrum of V+ (C2 H2 )3 at low concentration have been implicated in previous theoretical work on other metal–acetylene systems as intermediates along the reaction path to form benzene via the cyclization of acetylene. Apparently, we have observed the same kind of cyclization chemistry here for the vanadium cation system too. Although the cyclization of acetylene to form benzene is a known chemistry on a number of different catalysts, the mechanism for the reaction has always been uncertain. Our infrared spectra at low concentration reveal for the first time the specific intermediate structures involved. Additional work will be necessary to understand the concentration dependence in more detail and to determine whether other metal ions might catalyze similar cyclization chemistry.

6.6 Metal–Benzene Complexes Metal–benzene complexes are known for the formation of sandwiches, and cation–π interactions are well studied in organometallic chemistry [6–11, 16, 279–282]. These systems have been studied in gas phase ion chemistry and in computational chemistry [289–306]. As shown in previous studies in our lab, the interaction of metal cations with the aromatic π system has distinctive effects on vibrational spectra. Charge transfer from the ring system toward the metal induces a red shift on the in-plane carbon ring distortion, ν19 (1486 cm−1 in isolated benzene), while this also causes a blue shift in the out-of-plane hydrogen bend, ν11 (673 cm−1 in isolated benzene) [167, 302]. We documented these patterns for several cation–benzene systems in work done at the FELIX free electron laser using infrared multiphoton photodissociation (IR-MPD) spectroscopy on the ions without tagging [201–203], as the cluster source available at that time did not allow sufficient cooling for the formation of rare gas adducts. Unfortunately, the conditions used for the IR-MPD process can cause significant power broadening in spectral lines and shifts to lower frequencies. Rare gas tagging has not yet been applied to transition metal–benzene complexes, except for spectra in the C–H stretching region [165, 166]. The vibrations most sensitive to the metal–benzene charge transfer are in the fingerprint region, and therefore the details of this chemistry are yet to be revealed. The most well-studied metal ion–benzene complex is Al+ (benzene)n , for which we have measured spectra for the n = 1–4 complexes using argon tagging [168]. The spectrum for the n = 1 complex is shown in Fig. 6.14, where it is compared to the spectrum reported previously for this complex using IR-MPD with the FELIX free electron laser [202]. As shown, the quality of these tagged spectra is now far superior to the previous work in signal levels and resolution. Bands which were broad in the IR-MPD spectrum are much sharper, and the shifts from the IR-MPD process

178

M. A. Duncan

Fig. 6.14 The IR spectrum of Al+ (bz) in the C–H and fingerprint regions, measured with argon tagging (middle) compared to that measured with IR multiphoton dissociation spectroscopy with the FELIX free electron laser (top). The lower trace (blue) shows the spectrum predicted by theory. The C–H stretch region has a triplet structure from a Fermi resonance, as seen in isolated benzene. Figure used from Ref. [168] with permission from the American Chemical Society, Copyright 2014

apparently occur in an unpredictable way at both higher and lower frequencies for different bands. The light red dashed lines in the figure show the positions of the freebenzene IR-active vibrations, including the well-known Fermi resonance that splits the single C–H stretch expected into a triplet [307]. The purple dashed lines show the positions of Raman-active (IR-inactive) vibrations, which appear in the IR spectrum of the metal ion complex because of its reduced symmetry. The red shift in the ν19 band associated with charge transfer is only 10 cm−1 , whereas the blue shift of the ν11 out-of-plane bending mode is 75 cm−1 . The former is much smaller than the shifts seen for transition metal complexes, consistent with their expected greater charge transfer, but the latter is comparable to the shifts seen before because it arises from

6 Metal Cation Coordination and Solvation Studied …

179

the mechanical action of the bending hydrogen molecules bumping into the metal. The data on the larger complexes reveals that the coordination around Al+ contains three benzene molecules, that is, it does not form the same kind of sandwich seen for transition metals. Clearly, the quality of spectra for tagged ions is highly desirable, and our lab is working to get similar data for transition metal ion complexes. Ongoing work has obtained partial spectra for V+ (bz) and for Co+ (bz)2 [308]. Both of these systems exhibit multiplet structure in the ν19 vibration, indicating that the benzene ring is distorted from its D6h symmetric structure by the strong metal binding. Although metal ion–benzene systems have been studied for many years, their electronic structure remains a significant challenge. In the case of V+ (benzene) and V+ (benzene)2 , ordinary DFT (B3LYP or BP86 functionals with large basis sets) misses the ground state spin configuration (triplet predicted; quintet agrees with experiment and higher level theory) [165, 201]. Higher levels of theory get the correct quintet spin state for this system [304, 305]. For Ni+ (benzene)2 , DFT apparently gets the wrong ground state structure (η4 sandwich predicted; η6 observed) [166]. As in the case of other systems, the 18-electron rule is a useful guiding principle for metal ion–benzene complexes. Mn+ (benzene)2 is isoelectronic to the known neutral dibenzene chromium species, but the infrared spectrum of this ion has not yet been measured in the gas phase. Its expected η6 coordination on the six-fold axis of benzene is common for many metal ions. However, later transition metal ions have more valence electrons, and do not need to interact with all six π electrons to achieve the 18-electron configuration. Some of these systems are known to adopt η4 or lower coordination in the condensed phase, and then their sandwich structures should have the two rings offset from each other. In extreme cases, some transition metals are predicted to bind strongly enough to distort the planarity of the ring (e.g., Fe+ , V+ , Co+ , Ni+ ). All of these structures will lead to recognizable patterns in the fingerprint region. Future studies of these systems with tagging are therefore highly desirable. A final aspect of these metal-benzene complexes worth mentioning is that the early transition metal systems, particularly vanadium, form multiple decker sandwich structures with unusual electronic structure and bonding [296–298]. IR spectra have been obtained for neutrals following cation deposition on surfaces, but not for ions. These systems will be even more challenging for future experiments and theory.

6.7 Conclusions The studies described in this chapter illustrate how infrared photodissociation spectroscopy can be applied to a variety of metal–molecular complexes in the gas phase. These gas phase studies eliminate the effects of solvent or counterions and make it possible to investigate isolated molecules with different numbers of ligand or solvent molecules. Vibrational band patterns, in coordination with computational predictions, make it possible to determine the structures of these complexes, and the effects that metal binding has on the geometry and charge distribution of the molecular adducts. Additionally, the spectral patterns reveal the number of ligand or

180

M. A. Duncan

solvent molecules making up the first (and sometimes higher) coordination sphere(s). In the case of carbonyl ligands, this provides an opportunity to make comparisons to several well-known neutral complexes that are isoelectronic analogs to the cations studied here. Because vibrational band patterns vary with the electronic state and spin multiplicity of the system, these spectra also make it possible to investigate the electronic structure of these complexes, and to identify strengths and weaknesses of density functional theory computations. We find examples in which DFT fails to describe the system adequately, such as the transition metal–benzene spin states, but also find many examples where it performs quite well to describe vibrational band patterns. The computations presented here usually use the B3LYP functional. We have tried other functionals, especially including dispersion-corrected versions which are believed to describe the energetics of bonding more accurately. However, our experiments do not probe bonding energetics; they measure IR spectra. For this application, we find that harmonic DFT/B3LYP calculations with proper scaling to account for anharmonicity provide the best description of vibrational patterns. Although we have presented a variety of metal–molecular complexes here, there are clearly many more which could be investigated. Complexes with larger ligand or solvent molecules become more chemically interesting, and these studies can also be extended to metals other than the main group and transition metal species described here (e.g., lanthanides, actinides). We anticipate that this general area of activity will continue to provide fundamental insights into metal–molecular interactions for the foreseeable future. Acknowledgments This research was supported by the U. S. Department of Energy (grant no. DE-SC0018835), the National Science Foundation (grant no. CHE-1764111), and the Air Force Office of Scientific Research (grant no. FA-9550-15-1-0088).

References 1. Somorjai, G.A.: Introduction to Surface Chemistry and Catalysis. Wiley, New York (1994) 2. Henrich, V.E., Cox, P.A.: The Surface Science of Metal Oxides. Cambridge University Press, Cambridge (1994) 3. Ertl, G.: Reactions at Solid Surfaces. Wiley, Hoboken, NJ (2009) 4. Ozin, G.A., Arsenault, A.C.: Nanochemistry. Royal Society of Chemistry Publishing, Cambridge (2005) 5. Heiz, U., Landman, U. (eds.): Nanocatalysis. Springer, Berlin (2007) 6. van Leeuwen, P.W.N.M.: Homogeneous Catalysis. Kluwer, Dordrecht (2004) 7. Hartwig, J.F.: Organotransition Metal Chemistry: From Bonding to Catalysis. University Science Books, Sausalito, CA (2010) 8. Cotton, F.A., Wilkinson, G., Murillo, C.A., Bochmann, M.: Advanced Inorganic Chemistry, 6th edn. Wiley Interscience, New York (1999) 9. Huhey, J.E., Keiter, E.A., Keiter, R.L.: Inorganic Chemistry, 4th edn. Harper Collins, New York (1993) 10. Long, N.J.: Metallocenes. Blackwell Science, Oxford, U.K. (1998) 11. Crabtree, R.H.: The Organometallic Chemistry of the Transition Metals, 5th edn. Wiley, Hoboken, N.J. (2009)

6 Metal Cation Coordination and Solvation Studied … 12. 13. 14. 15. 16. 17. 18. 19.

20. 21. 22. 23. 24. 25.

26. 27. 28. 29.

30. 31. 32.

33. 34.

35.

36. 37.

181

Burgess, J.: Ions in Solution. Horwood Publishing, Chichester, U.K. (1999) Marcus, Y.: Ion Solvation. Wiley, Chichester, U.K. (1985) Richens, D.T.: The Chemistry of Aqua Ions. Wiley, Chichester, U. K. (1997) Steed, J.W., Atwood, J.L.: Supramolecular Chemistry. Wiley, Chichester, U.K. (2009) Bertini, I., Gray, H.B., Stiefel, E.I., Valentine, J.S.: Biological Inorganic Chemistry: Structure and Reactivity. University Science Books, Sausalito, CA (2007) Stoyanov, S.R., Titov, A.V., Král, P.: Transition metal and nitrogen doped carbon nanostructures. Coord. Chem. Rev. 253, 2852–2871 (2009) Kamat, P.V.: Graphene-based nanoarchitechtures. Anchoring semiconductor and metal nanoparticles on a two-dimensional carbon support. J. Phys. Chem. Lett. 1, 520–527 (2010) Lebedeva, M.A., Chamberlain, T.W., Khlobystov, A.N.: Harnessing the synergistic and complementary properties of fullerene and transition-metal compounds for nanomaterial applications. Chem. Rev. 115, 11301–11351 (2015) Low, J., Yu, J., Ho, W.: Graphene-based photocatalysts for CO2 reduction to solar fuel. J. Phys. Chem. Lett. 6, 4244–4251 (2015) D’Alessandro, D.M., Smit, B., Long, J.R.: Carbon dioxide capture: prospects for new materials. Angew. Chem. Int. Ed. 49, 6058–6082 (2010) Jena, P.: Materials for hydrogen storage: past, present and future. J. Phys. Chem. Lett. 2, 206–211 (2011) Suh, M.P., Park, H.J., Prasad, T.K., Lim, D.-W.: Hydrogen storage in metal-organic frameworks. Chem. Rev. 112, 782–835 (2012) Yoon, M., Srirambalaji, R., Kim, K.: Homochiral metal-organic frameworks for asymmetric heterogeneous catalysis. Chem. Rev. 112, 1196–1231 (2012) Poloni, R., Lee, K., Berger, R.F., Smit, B., Neaton, J.B.: Understanding trends in CO2 adsorption in metal-organic frameworks with open metal sites. J. Phys. Chem. Lett. 5, 861–865 (2014) Pera-Titus, M.: Porous inorganic membranes for CO2 capture: present and prospects. Chem. Rev. 114, 1413–1492 (2014) Gao, F., Goodman, D.W.: Model catalysts: simulating the complexities of heterogeneous catalysts. Annu. Rev. Phys. Chem. 63, 265–286 (2012) Häkkinen, H.: Atomic and electronic structure of gold clusters: understanding flakes, cages and superatoms from simple concepts. Chem. Soc. Rev. 37, 1847–1859 (2008) Lee, I., Albiter, M.A., Zhang, Q., Ge, J., Yin, Y., Zaera, F.: New nanostructured heterogeneous catalysts with increased selectivity and stability. Phys. Chem. Chem. Phys. 13, 2449–2456 (2011) Schauermann, S., Nilius, N., Shaikhutdinov, S., Freund, H.-J.: Nanoparticles for heterogeneous catalysis: new mechanistic insights. Acc. Chem. Res. 46, 1673–1681 (2013) McFarland, E.W., Metiu, H.: Catalysis by doped oxides. Chem. Rev. 113, 4391–4427 (2013) Reichert, R., Jusys, Z., Behm, R.J.: Au/TiO2 photo(electro)catalysis: the role of Au cocatalyst in photoelectrochemical water splitting and photocatalytic H2 evolution. J. Phys. Chem. C 119, 24750–24759 (2015) Esswein, A.J., Nocera, D.G.: Hydrogen production by molecular photocatalysis. Chem. Rev. 107, 4022–4047 (2007) Kortlever, R., Shen, J., Schouten, K.J.P., Calle-Vallejo, F., Koper, M.T.M.: Catalysts and reaction pathways for the electrochemical reduction of carbon dioxide. J. Phys. Chem. Lett. 6, 4073–4082 (2015) White, J.L., Baruch, M.F., Pander III, J.E., Hu, Y., Fortmeyer, I.C., Park, J.E., Zhang, T., Liao, K., Gu, J., Yan, Y., Shaw, T.W., Abelev, E., Bocarsly, A.B.: Light-driven heterogeneous reduction of carbon dioxide: photocatalysis and photoelectrodes. Chem. Rev. 115, 12888–12935 (2015) Blakemore, J.D., Crabtree, R.H., Brudvig, G.W.: Molecular catalysts for water oxidation. Chem. Rev. 115, 12974–13005 (2015) Jungwirth, P., Tobias, D.J.: Specific ion effects at the air/water interface. Chem. Rev. 106, 1259–1281 (2006)

182

M. A. Duncan

38. Jungwirth, P., Winter, B.: Ions at aqueous interfaces: from water surface to hydrated proteins. Annu. Rev. Phys. Chem. 59, 343–366 (2008) 39. Kohagen, M., Pluhaˇrová, E., Mason, P.E., Jungwirth, P.: Exploring ion-ion interactions in aqueous solutions by a combination of molecular dynamics and neutron scattering. J. Phys. Chem. Lett. 6, 1563–1567 (2015) 40. Tielrooij, K.J., Garcia-Araez, N., Bonn, M., Bakker, H.J.: Cooperativity in ion hydration. Science 328, 1006–1009 (2010) 41. Tobias, D.J., Stern, A.C., Baer, M.D., Levin, Y., Mundy, C.J.: Simulation and theory of ions at atmospherically relevant aqueous liquid-air interfaces. Annu. Rev. Phys. Chem. 64, 339–359 (2013) 42. Chen, H., Ruckenstein, E.: Hydrated ions: from individual ions to ion pairs to ion clusters. J. Phys. Chem. B 119, 12671–12676 (2015) 43. Kanjana, K., Courtin, B., MacConnel, A., Bartels, D.M.: Reactions of hexa-aquo transition metal ions with the hydrated electron up to 300 °C. J. Phys. Chem. A 119, 11094–11104 (2015) 44. Eller, K., Schwarz, H.: Organometallic chemistry in the gas phase. Chem. Rev. 91, 1121–1177 (1991) 45. Freiser, B.S. (ed.): Organometallic Ion Chemistry. Kluwer, Dordrecht (1996) 46. Armentrout, P.B., Baer, T.: Gas phase ion dynamics and chemistry. J. Phys. Chem. 100, 12866–12877 (1996) 47. Rodgers, M.T., Armentrout, P.B.: Noncovalent metal-ligand bond energies as studied by threshold collision-induced dissociation. Mass Spectrom. Rev. 19, 215–247 (2000) 48. Armentrout, P.B.: Reactions and thermochemistry of small transition metal cluster ions. Ann. Rev. Phys. Chem. 52, 423–461 (2001) 49. Ervin, K.M.: Metal-ligand interactions: gas phase transition metal carbonyls. Int. Rev. Phys. Chem. 20, 127–164 (2001) 50. Armentrout, P.B.: Guided ion beam studies of transition metal-ligand thermochemistry. Int. J. Mass Spectrom. 227, 289–302 (2003) 51. Bohme, D.K., Schwarz, H.: Gas-phase catalysis by atomic and cluster metal ions: the ultimate single-site catalysts. Angew. Chem. Int. Ed. 44, 2336–2354 (2005) 52. Operti, L., Rabezzana, R.: Gas phase ion chemistry in organometallic systems. Mass Spectrom. Rev. 25, 483–513 (2006) 53. Roithova, J., Schröder, D.: Selective activation of alkanes by gas phase metal ions. Chem. Rev. 110, 1170–1211 (2010) 54. Armentrout, P.B.: Fifty years of ion and neutral thermochemistry by mass spectrometry. Int. J. Mass Spectrometry 377, 54–63 (2015) 55. Rodgers, M.T., Armentrout, P.B.: Cationic noncovalent interactions: energetics and periodic trends. Chem. Rev. 116, 5642–5678 (2016) 56. Ricca, A., Bauschlicher Jr., C.W., Rosi, M.: Second-order Moller-Plesset perturbation theory for systems involving first transition row metals. J. Phys. Chem. 98, 9498–9502 (1994) 57. Bauschlicher Jr., C.W., Partridge, H., Langhoff, S.R.: The calculation of accurate metal-ligand bond energies. Adv. Met. Semicond. Clusters 2, 165–202 (1994) 58. Nakamoto, K.: Infrared and Raman Spectra of Inorganic and Coordination Compounds, Parts A and B, 5th edn. Wiley, New York (1997) 59. Yates, J.T., Madey, T.E.: Vibrational Spectroscopy of Molecules on Surfaces. Plenum, New York (1987) 60. Lessen, D.E., Asher, R.L., Brucat, P.J.: Cluster ion photodissociation spectroscopy. Adv. Metal Semiconductor Clusters 1, 267–302 (1993) 61. Asher, R.L., Bellert, D., Buthelezi, T., Brucat, P.J.: Optical excitation of Co+ N2 . J. Phys. Chem. 99, 1068–1072 (1995) 62. Hayes, T., Bellert, D., Buthelezi, T., Brucat, P.J.: The photodissociation of V+ –CH4 . Chem. Phys. Lett. 264, 220–224 (1997) 63. Bellert, D., Buthelezi, T., Brucat, P.J.: The structure of Co+ –CO2 . Chem. Phys. Lett. 290, 316–322 (1998)

6 Metal Cation Coordination and Solvation Studied …

183

64. Willey, K.F., Yeh, C.S., Robbins, D.L., Pilgrim, J.S., Duncan, M.A.: Photodissociation spectroscopy of Mg+ –H2 O and Mg+ –D2 O. J. Chem. Phys. 97, 8886–8895 (1992) 65. Yeh, C.S., Willey, K.F., Robbins, D.L., Pilgrim, J.S., Duncan, M.A.: Photodissociation spectroscopy of the Mg+ –CO2 complex and its isotopic analogs. J. Chem. Phys. 98, 1867–1875 (1993) 66. Scurlock, C.T., Pullins, S.H., Reddic, J.E., Duncan, M.A.: Photodissociation spectroscopy of Ca+ –H2 O and Ca+ –D2 O. J. Chem. Phys. 104, 4591–4599 (1996) 67. Scurlock, C.T., Pullins, S.H., Duncan, M.A.: Photodissociation spectroscopy of Ca+ photodissociation spectroscopy of Ca+ –H2 O and Ca+ –D2 OCO2 . J. Chem. Phys. 105, 3579–3585 (1996) 68. Pullins, S.H., Reddic, J.E., France, M.R., Duncan, M.A.: Photodissociation spectroscopy of Ca+ –N2 . J. Chem. Phys. 108, 2725–2732 (1998) 69. France, M.R., Pullins, S.H., Duncan, M.A.: Photodissociation spectroscopy of Ca+ –C2 H2 and Ca+ –C2 D2 π-complexes. J. Chem. Phys. 109, 8842–8850 (1998) 70. Duncan, M.A.: Spectroscopy of metal ion complexes: gas phase models for solvation. Ann. Rev. Phys. Chem. 48, 63–93 (1997) 71. Farrar, J.M.: Size-dependent reactivity in open shell metal-ion polar solvent clusters: spectroscopic probes of electronic-vibration coupling, oxidation and ionization. Int. Rev. Phys. Chem. 22, 593–640 (2003) 72. Sanekata, M., Misaizu, F., Fuke, K.: Photodissociation study on Ca+ (H2 O)n , n = 1–6: electron structure and photoinduced dehydrogenation reaction. J. Chem. Phys. 104, 9768–9778 (1996) 73. Misaizu, F., Sanekata, M., Tsukamoto, K., Fuke, K., Iwata, S.: Photodissociation of sizeselected aquamagnesium (Mg+ (H2 O)n ) ions for n = 1 and 2. J. Phys. Chem. 96, 8259–8264 (1992) 74. Chen, J., Wong, T.H., Cheng, Y.C., Montgomery, K., Kleiber, P.D.: Photodissociation spectroscopy and dynamics of MgC2 H4 + . J. Chem. Phys. 108, 2285–2296 (1998) 75. Chen, J., Wong, T.H., Kleiber, P.D., Wang, K.H.: Photofragmentation spectroscopy of Al+ (C2 H4 ). J. Chem. Phys. 110, 11798–11805 (1999) 76. Abate, Y., Kleiber, P.D.: Photodissociation spectroscopy of Zn+ (H2 O) and Zn+ (D2 O). J. Chem. Phys. 122, 084305 (2005) 77. Metz, R.B.: Photofragment spectroscopy of covalently bound transition metal complexes: a window into C–H and C–C bond activation by transition metal ions. Int. Rev. Phys. Chem. 23, 79–108 (2004) 78. Metz, R.B.: Optical spectroscopy and photodissociation dynamics of multiply charged ions. Int. J. Mass Spectrom. 235, 131–143 (2004) 79. Stringer, K.L., Citir, M., Metz, R.B.: Photofragment spectroscopy of π complexes: Au+ (C2 H4 ) and Pt+ (C2 H4 ). J. Phys. Chem. A 108, 6996–7002 (2004) 80. Thompson, C.J., Faherty, K.P., Stringer, K.L., Metz, R.B.: Electronic spectroscopy and photodissociation dynamics of Co2+ -methanol clusters: Co2+ (CH3 OH)n (n = 4–7). Phys. Chem. Chem. Phys. 7, 814–818 (2005) 81. Cox, H., Stace, A.J.: Recent advances in the visible and UV spectroscopy of metal dication complexes. Int. Rev. Phys. Chem. 29, 555–588 (2010) 82. Engelking, P.C., Lineberger, W.C.: Laser photoelectron spectrometry of the negative ions of iron and iron carbonyls. Electron affinity determination for the series Fe(CO)n , n = 0, 1, 2, 3, 4. J. Am. Chem. Soc. 101, 5569–5573 (1979) 83. Stevens, A.E., Feigerle, C.S., Lineberger, W.C.: Laser photoelectron spectrometry of Ni(CO)− n , n = 1–3. J. Am. Chem. Soc. 104, 5026–5031 (1982) 84. Rathbone, G.J., Sanford, T., Andrews, D., Lineberger, W.C.: Photoelectron imaging spectroscopy of Cu− (H2 O)1,2 anion complexes. Chem. Phys. Lett. 401, 570–574 (2005) 85. Bengali, A.A., Casey, S.M., Cheng, C.L., Dick, J.P., Fenn, P.T., Villalta, P.W., Leopold, D.G.: Negative ion photoelectron spectroscopy of coordinatively unsaturated group VI metal carbonyls: Cr(CO)3 , Mo(CO)3 and W(CO)3 . J. Am. Chem. Soc. 114, 5257–5268 (1992) 86. Miller, S.R., Marcy, T.P., Millam, E.L., Leopold, D.G.: Photoelectron spectroscopic characterization of the niobium-benzene anion produced by reaction of niobium with ethylene. J. Am. Chem. Soc. 129, 3482–3483 (2007)

184

M. A. Duncan

87. Klopcic, S.A., Moravec, V.D., Jarrold, C.C.: Anion photoelectron spectroscopy of PdCO– and PdCN– : reactivity of Pd– . J. Chem. Phys. 110, 8986–8991 (1999) 88. Moravec, V.D., Jarrold, C.C.: The electronic structure of PdC2 H and PdC2 HN determined by anion photoelectron spectroscopy. J. Chem. Phys. 112, 792–798 (2000) 89. Judai, K., Hirano, M., Kawamata, H., Yabushita, S., Nakajima, A., Kaya, K.: Formation of vanadium-arene complex anions and their photoelectron spectroscopy. Chem. Phys. Lett. 270, 23–30 (1997) 90. Masabuchi, T., Ohi, K., Iwasa, T., Nakajima, A.: Experimental and theoretical studies on the electronic properties of vanadium-benzene sandwich cluster anions, Vn Bz–n+1 (n = 1–5). J. Chem. Phys. 137, 224305 (2012) 91. Gerhards, M., Thomas, O.C., Nilles, J.M., Zheng, W.-J., Bowen Jr., K.H.: Cobalt-benzene cluster anions: mass spectrometry and negative ion photoelectron spectroscopy. J. Chem. Phys. 116, 10247–10252 (2002) 92. Zheng, W., Nilles, J.M., Thomas, O.C., Bowen Jr., K.H.: Photoelectron spectroscopy of nickelbenzene cluster anions. J. Chem. Phys. 122, 044306 (2005) 93. Zheng, W., Nilles, J.M., Thomas, O.C., Bowen Jr., K.H.: Photoelectron spectroscopy of titanium-benzene cluster anions. Chem. Phys. Lett. 401, 266–270 (2005) 94. Zheng, W., Eustis, S.N., Li, X., Nilles, J.M., Thomas, O.C., Bowen Jr., K.H., Kandalam, A.K.: Photoelectron spectroscopic study of iron-benzene cluster anions. Chem. Phys. Lett. 462, 35–39 (2008) 95. Zhang, X., Lim, E., Kim, S.K., Bowen Jr., K.H.: Photoelectron spectroscopic and computational study of (M–CO2 )− anions, M = Cu, Ag. Au. J. Chem. Phys. 143, 174305 (2015) 96. Yang, D.-S.: Zero electron kinetic energy photoelectron spectra of metal clusters and complexes. Adv. Metal Semiconductor Clusters 5, 187–225 (2001) 97. Yang, D.-S.: High resolution electron spectroscopy of gas phase metal-aromatic complexes. J. Phys. Chem. Lett. 2, 25–33 (2011) 98. Zhou, M., Andrews, L., Bauschlicher Jr., C.W.: Spectroscopic and theoretical investigations of vibrational frequencies in binary unsaturated transition metal carbonyl cations, neutrals and anions. Chem. Rev. 101, 1931–1962 (2001) 99. Okumura, M., Yeh, L.I., Lee, Y.T.: The vibrational predissociation spectroscopy of hydrogen cluster ions. J. Chem. Phys. 83, 3705–3706 (1985) 100. Okumura, M., Yeh, L.I., Meyers, J.D., Lee, Y.T.: Infrared spectra of the cluster ions triaquahydrogen(1+ )-molecular hydrogen (H7 O3 + •H2 ) and tetraaquahydrogen(1+ )-molecular hydrogen (H9 O4 + •H2 ). J. Chem. Phys. 85, 2328–2329 (1986) 101. Okumura, M., Yeh, L.I., Meyers, J.D., Lee, Y.T.: Infrared spectra of the solvated hydronium ion: vibrational predissociation spectroscopy of mass-selected H3 O+ •(H2 O)n •(H2 )m . J. Phys. Chem. 94, 3416–3427 (1990) 102. Lisy, J.M.: Spectroscopy and structure of solvated alkali-metal ions. Int. Rev. Phys. Chem. 16, 267–289 (1997) 103. Ebata, T., Fujii, A., Mikami, N.: Vibrational spectroscopy of small-sized hydrogen-bonded clusters and their ions. Int. Rev. Phys. Chem. 17, 331–361 (1998) 104. Wang, L.-S., Ding, C.F., Wang, X.B., Barlow, S.E.: Photodetachment photoelectron spectroscopy of multiply charged anions using electrospray ionization. Rev. Sci. Instrum. 70, 1957–1966 (1999) 105. Bieske, E.J., Dopfer, O.: High resolution spectroscopy of cluster ions. Chem. Rev. 100, 3963–3998 (2000) 106. Duncan, M.A.: Infrared spectroscopy to probe structure and dynamics in metal ion-molecule complexes. Int. Rev. Phys. Chem. 22, 407–435 (2003) 107. Duncan, M.A.: Frontiers in the spectroscopy of mass-selected molecular ions. Int. J. Mass Spectrom. 200, 545–569 (2000) 108. Robertson, W.H., Johnson, M.A.: Molecular aspects of halide ion hydration: the cluster approach. Annu. Rev. Phys. Chem. 54, 173–213 (2003) 109. Rizzo, T.R., Stearns, J.A., Boyarkin, O.V.: Spectroscopic studies of cold, gas-phase biomolecular ions. Int. Rev. Phys. Chem. 28, 481–515 (2009)

6 Metal Cation Coordination and Solvation Studied …

185

110. Baer, T., Dunbar, R.C.: Ion spectroscopy: where did it come from; where is it now; and where is it going? J. Am. Soc. Mass Spectrom. 21, 681–693 (2010) 111. Duncan, M.A.: Infrared laser spectroscopy of mass-selected carbocations. J. Phys. Chem. A 116, 11477–11491 (2012) 112. Wolk, A.B., Leavitt, C.M., Garand, E., Johnson, M.A.: Cryogenic ion chemistry and spectroscopy. Acc. Chem. Res. 47, 202–210 (2014) 113. Hock, C., Kim, J.B., Weichman, M.L., Yacovitch, T.I., Neumark, D.M.: Slow photoelectron velocity-map imaging spectroscopy of cold negative ions. J. Chem. Phys. 137, 224201 (2012) 114. Redwine, J.G., Davis, Z.A., Burke, N.L., Oglesbee, R.A., McLuckey, S.A., Zwier, T.S.: A novel ion trap based tandem mass spectrometer for the spectroscopic study of cold gas phase polyatomic ions. Int. J. Mass Spectrom. 348, 9–14 (2013) 115. Chakrabarty, S., Holtz, M., Campbell, E.K., Banerjee, A., Gerlich, D., Maier, J.P.: A novel method to measure electronic spectra of cold molecular ions. J. Phys. Chem. Lett. 4, 4051–4054 (2013) 116. Duffy, E.M., Marsh, B.M., Garand, E.: Probing the hydrogen-bonded water network at the active site of a water oxidation catalyst: Ru(bpy)(tpy)(H2 O)+2 •(H2 O)0−4 . J. Phys. Chem. A 119, 6326–6332 (2015) 117. Heine, N., Asmis, K.R.: Cryogenic ion trap vibrational spectroscopy of hydrogen-bonded clusters relevant to atmospheric chemistry. Int. Rev. Phys. Chem. 34, 1–34 (2015) 118. Roithová, J., Gray, A., Andris, E., Jašík, J., Gerlich, D.: Helium tagging infrared photodissociation spectroscopy of reactive ions. Acc. Chem. Res. 49, 223–230 (2016) 119. Bosenberg, W.R., Guyer, D.R.: Broadly tunable, single-frequency optical parametric frequency-conversion system. J. Opt. Soc. Am. B 10, 1716–1722 (1993) 120. Gerhards, M.: High energy and narrow bandwidth mid IR nanosecond laser system. Opt. Commun. 241, 493–497 (2004) 121. Duncan, M.A.: Laser vaporization cluster sources. Rev. Sci. Instrum. 83, 041101 (2012) 122. Cornett, D.S., Peschke, M., LaiHing, K., Cheng, P.Y., Willey, K.F., Duncan, M.A.: A reflectron time-of-flight mass spectrometer for laser photodissociation. Rev. Sci. Instrum. 63, 2177–2186 (1992) 123. Gregoire, G., Duncan, M.A.: Infrared spectroscopy to probe structure and growth dynamics in Fe+ (CO2 )n complexes. J. Chem. Phys. 117, 2120–2130 (2002) 124. Gregoire, G., Brinkman, N., Schaefer, H.F., Duncan, M.A.: Infrared photodissociation spectroscopy of Mg+ (CO2 )n and Mg+ (CO2 )n Arm complexes. J. Phys. Chem. A 107, 218–227 (2003) 125. Walters, R.S., Jaeger, T.D., Brinkman, N., Schaefer, H.F., Duncan, M.A.: Infrared photodissociation spectroscopy of Al+ (CO2 )n and Al+ (CO2 )n Arm complexes. J. Phys. Chem. A 107, 7396–7405 (2003) 126. Walker, N.R., Grieves, G.A., Walters, R.S., Duncan, M.A.: Growth dynamics and intracluster reactions in Ni+ (CO2 )n complexes via infrared spectroscopy. J. Chem. Phys. 121, 10498–10507 (2004) 127. Walker, N.R., Walters, R.S., Duncan, M.A.: Infrared photodissociation spectroscopy of V+ (CO2 )n and V+ (CO2 )n- Ar complexes. J. Chem. Phys. 120, 10037–10045 (2004) 128. Ricks, A.M., Brathwaite, A.D., Duncan, M.A.: IR spectroscopy of V+ (CO2 )n clusters: Solvation-induced electron transfer and activation of CO2 . J. Phys. Chem. A 117, 11490–11498 (2013) 129. Walters, R.S., Walker, N.R., Pillai, E.D., Duncan, M.A.: Infrared spectroscopy of V+ (H2 O) and V+ (D2 O) complexes: ligand deformation and an incipient reaction. J. Chem. Phys. 119, 10471–10474 (2003) 130. Walters, R.S., Duncan, M.A.: Infrared spectroscopy of solvation and isomers in Fe+ (H2 O)1,2 Arm complexes. Austr. J. Chem. 57, 1145–1148 (2004) 131. Walker, N.R., Walters, R.S., Tsai, C.-S., Jordan, K.D., Duncan, M.A.: Infrared photodissociation spectroscopy of Mg+ (H2 O)Arn complexes: isomers in progressive microsolvation. J. Phys. Chem. A 109, 7057–7067 (2005)

186

M. A. Duncan

132. Walters, R.S., Pillai, E.D., Duncan, M.A.: Solvation processes in Ni+ (H2 O)n complexes revealed by infrared photodissociation spectroscopy. J. Am. Chem. Soc. 127, 16599–16610 (2005) 133. Vaden, T.D., Lisy, J.M., Carnegie, P.D., Pillai, E.D., Duncan, M.A.: Infrared spectroscopy of the Li+ (H2 O)Ar complex: the role of internal energy and its dependence on ion preparation. Phys. Chem. Chem. Phys. 8, 3078–3082 (2006) 134. Kasalova, V., Allen, W.D., Schaefer, H.F., Pillai, E.D., Duncan, M.A.: Model systems for probing metal cation hydration: the V+ (H2 O) and V+ (H2 O)Ar complexes. J. Phys. Chem. A 111, 7599–7610 (2007) 135. Carnegie, P.D., Bandyopadhyay, B., Duncan, M.A.: Infrared spectroscopy of Cr+ (H2 O) and Cr2+ (H2 O): the role of charge in cation hydration. J. Phys. Chem. A 112, 6237–6243 (2008) 136. Carnegie, P.D., McCoy, A.B., Duncan, M.A.: Infrared spectroscopy and theory of Cu+ (H2 O)Ar2 and Cu+ (D2 O)Ar2 : fundamentals and combination bands. J. Phys. Chem. A 113, 4849–4854 (2009) 137. Carnegie, P.D., Bandyopadhyay, B., Duncan, M.A.: Infrared spectroscopy of Sc+ (H2 O) and Sc2+ (H2 O) via argon complex predissociation: the charge dependence of cation hydration. J. Chem. Phys. 134, 014302 (2011) 138. Bandyopadhyay, B., Carnegie, P.D., Duncan, M.A.: Infrared spectroscopy of Mn+ (H2 O)n and Mn2+ (H2 O) complexes via argon complex predissociation. J. Phys. Chem. A 115, 7602–7609 (2011) 139. Bandyopadhyay, B., Duncan, M.A.: Infrared spectroscopy of V2+ (H2 O) complexes. Chem. Phys. Lett. 530, 10–15 (2012) 140. Bandyopadhyay, B., Reishus, K.N., Duncan, M.A.: Infrared spectroscopy of solvation in small Zn+ (H2 O)n complexes. J. Phys. Chem. A 117, 7794–7803 (2013) 141. Ward, T.B., Carnegie, P.D., Duncan, M.A.: Infrared spectroscopy of the Ti(H2 O)Ar+ ionmolecule complex: electronic state switching induced by argon. Chem. Phys. Lett. 654, 1–5 (2016) 142. Ward, T.B., Miliordos, E., Carnegie, P.D., Xantheas, S.S., Duncan, M.A.: Ortho-para interconversion in cation-water complexes: the case of V+ (H2 O) and Nb+ (H2 O) clusters. J. Chem. Phys. 146, 224305 (2017) 143. Velasquez III, J., Njegic, B., Gordon, M.S., Duncan, M.A.: IR photodissociation spectroscopy and theory of Au+ (CO)n complexes: nonclassical carbonyls in the gas phase. J. Phys. Chem. A 112, 1907–1913 (2008) 144. Velasquez III, J., Duncan, M.A.: IR photodissociation spectroscopy of Pt+ (CO)n complexes. Chem. Phys. Lett. 461, 28–32 (2008) 145. Ricks, A.M., Bakker, J.M., Douberly, G.E., Duncan, M.A.: IR spectroscopy of Co+ (CO)n complexes in the gas phase. J. Phys. Chem. A 113, 4701–4708 (2009) 146. Ricks, A.M., Reed, Z.D., Duncan, M.A.: Seven-coordinate homoleptic metal carbonyls in the gas phase. J. Am. Chem. Soc. 131, 9176–9177 (2009) 147. Reed, Z.D., Duncan, M.A.: Infrared spectroscopy and structures of manganese carbonyl cations, Mn(CO)+n (n = 1–9). J. Am. Soc. Mass Spectrom. 21, 739–749 (2010) 148. Ricks, A.M., Gagliardi, L., Duncan, M.A.: Infrared spectroscopy of extreme coordination: the carbonyls of U+ and UO2 + . J. Am. Chem. Soc. 132, 15905–15907 (2010) 149. Ricks, A.M., Reed, Z.D., Duncan, M.A.: IR spectroscopy of gas phase metal carbonyl cations. J. Mol. Spec. 266, 63–74 (2011) 150. Brathwaite, A.D., Reed, Z.D., Duncan, M.A.: Infrared spectroscopy of copper carbonyl cations. J. Phys. Chem. A 115, 10461–10469 (2011) 151. Ricks, A.M., Brathwaite, A.D., Duncan, M.A.: Coordination and spin states of V+ (CO)n clusters revealed by IR spectroscopy. J. Phys. Chem. A 117, 1001–1010 (2013) 152. Brathwaite, A.D., Duncan, M.A.: Infrared photodissociation spectroscopy of saturated group IV (Ti, Zr, Hf) metal carbonyl cations. J. Phys. Chem. A 117, 11695–11703 (2013) 153. Brathwaite, A.D., Ricks, A.M., Duncan, M.A.: Infrared spectroscopy of vanadium oxide carbonyl cations. J. Phys. Chem. A 117, 13435–13442 (2013)

6 Metal Cation Coordination and Solvation Studied …

187

154. Brathwaite, A.D., Maner, J.A., Duncan, M.A.: Testing the limits of the 18-electron rule: the gas phase carbonyls of Sc+ and Y+ . Inorg. Chem. 53, 1166–1169 (2014) 155. Brathwaite, A.D., Abbott-Lyon, H.L., Duncan, M.A.: Distinctive coordination of CO vs N2 to rhodium cations: an infrared and computational study. J. Phys. Chem. A 120, 7659–7670 (2016) 156. Walters, R.S., Jaeger, T.D., Duncan, M.A.: Infrared spectroscopy of Ni+ (C2 H2 )n complexes: evidence for intracluster cyclization reactions. J. Phys. Chem. A 106, 10482–10487 (2002) 157. Walters, R.S., Schleyer, P.v.R., Corminboeuf, C., Duncan, M.A.: Structural trends in transition metal cation-acetylene complexes revealed through the C–H stretch fundamentals. J. Am. Chem. Soc. 127, 1100–1101 (2005) 158. Walters, R.S., Pillai, E.D., Schleyer, P.v.R., Duncan, M.A.: Vibrational spectroscopy of Ni+ (C2 H2 )n (n = 1–4) complexes. J. Am. Chem. Soc. 127, 17030–17042 (2005) 159. Brathwaite, A.D., Ward, T.B., Walters, R.S., Duncan, M.A.: Cation-π and CH-π interactions in the coordination and solvation of Cu+ (acetylene)n complexes. J. Phys. Chem. A 119, 5658–5667 (2015) 160. Ward, T.B., Brathwaite, A.D., Duncan, M.A.: Infrared spectroscopy of Au(Acetylene)+n complexes in the gas phase. Top. Catal. 61, 49–61 (2018) 161. Marks, J.H., Ward, T.B., Duncan, M.A.: Infrared spectroscopy of the coordination and solvation in Cu+ (ethylene)n (n = 1–9) complexes. Int. J. Mass Spectrom. 435, 107–113 (2019) 162. Ward, T.B., Marks, J.H., Brathwaite, A.D., Duncan, M.A.: Cyclotrimerization of acetylene in gas phase V+ (C2 H2 )n complexes detected with infrared spectroscopy. To be submitted 163. Pillai, E.D., Jaeger, T.D., Duncan, M.A.: Infrared spectroscopy and density functional theory of small V+ (N2 )n clusters. J. Phys. Chem. A 109, 3521–3526 (2005) 164. Pillai, E.D., Jaeger, T.D., Duncan, M.A.: Infrared spectroscopy of Nb+ (N2 )n complexes: coordination, structures and spin states. J. Am. Chem. Soc. 129, 2297–2307 (2007) 165. Jaeger, T.D., Duncan, M.A.: Structure, coordination and solvation in V+ (benzene)n complexes via gas phase infrared spectroscopy. J. Phys. Chem. A 108, 6605–6610 (2004) 166. Jaeger, T.D., Duncan, M.A.: Infrared photodissociation spectroscopy of Ni+ (benzene)x complexes. J. Phys. Chem. A 109, 3311–3317 (2005) 167. Duncan, M.A.: Structures, energetics and spectroscopy of gas phase transition metal ionbenzene complexes. Int. J. Mass Spectrom. 272, 99–118 (2008) 168. Reishus, K.N., Brathwaite, A.D., Mosley, J.D., Duncan, M.A.: Infrared spectroscopy of coordination versus solvation in Al+ (benzene)1−4 complexes. J. Phys. Chem. A 118, 7516–7525 (2014) 169. Inokuchi, Y., Ohshimo, K., Misaizu, F., Nishi, N.: Structures of [Mg(H2 O)1,2 ]+ and [Al(H2 O)1,2 ]+ ions studied by infrared photodissociation spectroscopy: evidence of [HO–Al–H]+ ion core structure in [Al(H2 O)2 ]+ . Chem. Phys. Lett. 390, 140–144 (2004) 170. Inokuchi, Y., Ohshimo, K., Misaizu, F., Nishi, N.: Infrared photodissociation spectroscopy of [Mg(H2 O)1−4 ]+ and [Mg(H2 O)1−4 Ar]+ . J. Phys. Chem. A 108, 5034–5040 (2004) 171. Iino, T., Ohashi, K., Inoue, K., Judai, K., Nishi, N., Sekiya, H.: Infrared spectroscopy of Cu+ (H2 O)n and Ag+ (H2 O)n : coordination and solvation of noble-metal ions. J. Chem. Phys. 126, 194302 (2007) 172. Sasaki, J., Ohashi, K., Inoue, K., Imamura, T., Judai, K., Nishi, N., Sekiya, H.: Infrared photodissociation spectroscopy of V+ (H2 O)n (n = 2–8): coordinative saturation of V+ with four H2 O molecules. Chem. Phys. Lett. 474, 36–40 (2009) 173. Furukawa, K., Ohashi, K., Koga, N., Imamura, T., Judai, K., Nishi, N., Sekiya, H.: Coordinatively unsaturated cobalt ion in Co+ (H2 O)n (n = 4–6) probed with infrared photodissociation spectroscopy. Chem. Phys. Lett. 508, 202–206 (2011) 174. Weber, J.M.: The interaction of negative charge with carbon dioxide – insight into solvation, speciation and reductive activation from cluster studies. Int. Rev. Phys. Chem. 33, 489–519 (2014) 175. Dodson, L.G., Thompson, M.C., Weber, J.M.: Characterization of intermediate oxidation states in CO2 activation. Annu. Rev. Phys. Chem. 69, 231–252 (2018)

188

M. A. Duncan

176. Citir, M., Altinay, G., Metz, R.B.: Electronically and vibrational spectroscopy and vibrationally mediated photodissociation of V+ (OCO). J. Phys. Chem. A 110, 5051–5057 (2006) 177. Citir, M., Altinay, G., Austein-Miller, G., Metz, R.B.: Vibrational spectroscopy and theory of Fe+ (CH4 )n (n = 1–4). J. Phys. Chem. A 114, 11322–11329 (2010) 178. Daluz, J.S., Kocak, A., Metz, R.B.: Photodissociation studies of the electronic and vibrational spectroscopy of Ni+ (H2 O). J. Phys. Chem. A 116, 1344–1352 (2012) 179. Kocak, A., Austein-Miller, G., Pearson III, W.L., Altinay, G., Metz, R.B.: Dissociation energy and electronic and vibrational spectroscopy of Co+ (H2 O) and its isotopomers. J. Phys. Chem. A 117, 1254–1264 (2013) 180. Kocak, A., Sallese, Z., Johnston, M.D., Metz, R.B.: Vibrational spectroscopy of Co+ (CH4 )n and Ni+ (CH4 )n (n = 1–4). J. Phys. Chem. A 118, 3253–3265 (2014) 181. Pearson III, W.L., Copeland, C., Kocak, A., Sallese, Z., Metz, R.B.: Near ultraviolet photodissociation spectroscopy of Mn+ (H2 O) and Mn+ (D2 O). J. Chem. Phys. 141, 204305 (2014) 182. Kocak, A., Ashraf, M.A., Metz, R.B.: Vibrational spectroscopy reveals varying structural motifs in Cu+ (CH4 )n and Ag+ (CH4 )n (n = 1 − 6). J. Phys. Chem. A 119, 9653–9665 (2015) 183. Bush, M.F., Saykally, R.J., Williams, E.R.: Hydration of the calcium dication: direct evidence for second shell formation from infrared spectroscopy. ChemPhysChem 8, 2245–2253 (2007) 184. O’Brien, J.T., Williams, E.R.: Hydration of gaseous copper dications probed by IR action spectroscopy. J. Phys. Chem. A 112, 5893–5901 (2008) 185. Bush, M.F., Saykally, R.J., Williams, E.R.: Reactivity and infrared spectroscopy of gaseous hydrated trivalent metal ions. J. Am. Chem. Soc. 130, 9122–9128 (2008) 186. Bush, M.F., O’Brien, J.T., Prell, J.S., Wu, C.-C., Saykally, R.J., Williams, E.R.: Hydration of alkaline earth metal dications: effects of metal ion size determined using infrared action spectroscopy. J. Am. Chem. Soc. 131, 13270–13277 (2009) 187. O’Brien, J.T., Williams, E.R.: Coordination numbers of hydrated divalent transition metal ions investigated with IRPD spectroscopy. J. Phys. Chem. A 115, 14612–14619 (2011) 188. Dryza, V., Poad, B.L.J., Bieske, E.J.: Attaching molecular hydrogen to metal cations: perspectives from gas-phase infrared spectroscopy. Phys. Chem. Chem. Phys. 14, 14954–14965 (2012) 189. Dryza, V., Bieske, E.J.: Non-covalent interactions between metal cations and molecular hydrogen: spectroscopic studies of M+ –H2 complexes. Int. Rev. Phys. Chem. 32, 559–587 (2013) 190. Wang, G., Chi, C., Cui, J., Xing, X., Zhou, M.: Infrared photodissociation spectroscopy of mononuclear iron carbonyl anions. J. Phys. Chem. A 116, 2484–2489 (2012) 191. Chi, C., Cui, J., Xing, X., Wang, G., Zhou, M.: Infrared photodissociation spectroscopy of trigonal bipyramidal 19-electron Ni(CO)+5 cation. Chem. Phys. Lett. 542, 33–36 (2012) 192. Li, Y., Wang, G., Wang, C., Zhou, M.: Coordination and solvation of the Au+ cation: infrared photodissociation spectroscopy of mass-selected Au(H2 O)+n (n = 1–8) complexes. J. Phys. Chem. A 116, 10793–10801 (2012) 193. Zhou, X., Cui, J., Li, Z.H., Wang, G., Zhou, M.: Infrared photodissociation spectroscopy and theoretical study of homoleptic dinuclear chromium carbonyl cluster cations with a linear bridging carbonyl group. J. Phys. Chem. A 116, 12349–12356 (2012) 194. Wang, G., Cui, J., Cui, C., Zhou, X., Li, Z.H., Xing, X., Zhou, M.: Bonding in homoleptic iron carbonyl cluster cations: a combined infrared photodissociation and theoretical study. Chem. Sci. 3, 3272–3279 (2012) 195. Zhou, X., Cui, J., Li, Z.H., Wang, G., Liu, Z., Zhou, M.: Carbonyl bonding on oxophilic metal centers: infrared photodissociation spectra of mononuclear and dinuclear titanium carbonyl cation complexes. J. Phys. Chem. A 117, 1514–1521 (2013) 196. Jin, J., Yang, T., Xin, K., Wang, G., Jin, X., Zhou, M., Frenking, G.: Octacarbonyl anion complexes of group three transition metals [TM(CO)8 ]− (TM = Sc, Y, La) and the 18-electron rule. Angew. Chem. Int. Ed. 57, 6236–6241 (2018) 197. Wu, X., Zhao, L., Jin, J., Pan, S., Li, W., Jin, X., Wang, G., Zhou, M., Frenking, G.: Observation of alkaline earth complexes M(CO)8 (M = Ca, Sr, or Ba) that mimic transition metals. Science 361, 912–916 (2018)

6 Metal Cation Coordination and Solvation Studied …

189

198. Garand, E., Fournier, J.A., Kamrath, M.Z., Schley, N.D., Crabtree, R.H., Johnson, M.A.: Characterization of an activated iridium water splitting catalyst using infrared photodissociation of H2 tagged ions. Phys. Chem. Chem. Phys. 14, 10109–10113 (2012) 199. Johnson, C.J., Dzugan, L.C., Wolk, A.B., Leavitt, C.M., Fournier, J.A., McCoy, A.B., Johnson, M.A.: Microhydration of contact ion pairs in M2+ OH− (H2 O)n = 1–5 (M = Mg, Ca) clusters: spectral manifestations of a mobile proton defect in the first hydration sphere. J. Phys. Chem. A 118, 7590–7597 (2014) 200. DePalma, J.W., Kelleher, P.J., Johnson, C.J., Fournier, J.A., Johnson, M.A.: Vibrational signatures of solvent-mediated deformation of the ternary core ion in size-selected [MgSO4 Mg(H2 O)n = 4–11 ]2+ clusters. J. Phys. Chem. A 119, 8294–8302 (2015) 201. van Heijnsbergen, D., von Helden, G., Meijer, G., Maître, P., Duncan, M.A.: Infrared spectroscopy of V+ (benzene) and V+ (benzene)2 complexes in the gas phase. J. Am. Chem. Soc. 124, 1562–1563 (2002) 202. van Heijnsbergen, D., Jaeger, T.D., von Helden, G., Meijer, G., Duncan, M.A.: Infrared spectroscopy of Al+ (benzene) in the gas phase. Chem. Phys. Lett. 364, 345–351 (2002) 203. Jaeger, T.D., van Heijnsbergen, D., Klippenstein, S., von Helden, G., Meijer, G., Duncan, M.A.: Infrared spectroscopy and density functional theory of transition metal ion-benzene and dibenzene complexes. J. Am. Chem. Soc. 126, 10981–10991 (2004) 204. Oomens, J., Moore, D.T., von Helden, G., Meijer, G., Dunbar, R.C.: The site of Cr+ attachment to gas-phase aniline from infrared spectroscopy. J. Am. Chem. Soc. 126, 724–725 (2004) 205. Moore, D.T., Oomens, J., Eyler, J.R., Meijer, G., von Helden, G., Ridge, D.P.: Gas-phase IR spectroscopy of anionic iron carbonyl clusters. J. Am. Chem. Soc. 126, 14726–14727 (2004) 206. Fielicke, A., von Helden, G., Meijer, G., Petersen, D.B., Simard, B., Rayner, D.M.: Size and charge effects on the binding of CO to small isolated rhodium clusters. J. Phys. Chem. B 108, 14591–14598 (2004) 207. Moore, D.T., Oomens, J., Eyler, J.R., von Helden, G., Meijer, G., Dunbar, R.C.: Infrared spectroscopy of gas phase Cr+ coordination complexes: determination of binding sites and electronic states. J. Am. Chem. Soc. 127, 7243–7254 (2005) 208. Fielicke, A., von Helden, G., Meijer, G., Petersen, D.B., Simard, B., Rayner, D.M.: Size and charge effects on the binding of CO to late transition metal clusters. J. Chem. Phys. 124, 194305 (2006) 209. Fielicke, A., von Helden, G., Meijer, G., Simard, B., Rayner, D.M.: Gold cluster carbonyls: vibrational spectroscopy of the anions and the effects of cluster size, charge, and coverage on the CO stretching frequency. J. Phys. Chem. B 109, 23935–23940 (2005) 210. Simon, A., Jones, W., Ortega, J.-M., Boissel, P., Lemaire, J., Maître, P.: Infrared multiphoton dissociation spectroscopy of gas-phase mass-selected hydrocarbon-Fe+ complexes. J. Am. Chem. Soc. 126, 11666–11674 (2004) 211. MacAleese, L., Maître, P.: Infrared spectroscopy of organometallic ions in the gas phase: from model to real world complexes. Mass Spectrom. Rev. 26, 583–605 (2007) 212. Reinhard, B.M., Lagutschenkov, A., Lemaire, J., Maître, P., Boissel, P., Niedner-Schatteburg, G.: Reductive nitrile coupling in niobium-acetonitrile complexes probed by free electron laser multiphoton dissociation spectroscopy. J. Phys. Chem. A 108, 3350–3355 (2004) 213. Harvey, J.N.: DFT computation of relative spin-state energetics of transition metal compounds. Struct. Bond. 112, 151–183 (2004) 214. Cramer, C.J., Truhlar, D.J.: Density functional theory for transition metals and transition metal chemistry. Phys. Chem. Chem. Phys. 11, 10757–10816 (2009) 215. Li, S., Hennigan, J.M., Dixon, D.A., Peterson, K.A.: Accurate thermochemistry for transition metal oxide clusters. J. Phys. Chem. A 113, 7861–7877 (2009) 216. Cohen, A.J., Mori-Sanchez, P., Yang, W.: Challenges for density functional theory. Chem. Rev. 112, 289–320 (2012) 217. Ramabhadran, R.O., Becher III, E.L., Chowdhury, A., Raghavachari, K.: Fluxionality in the chemical reactions of transition metal oxide clusters: the role of metal, spin state, and the reactant molecule. J. Phys. Chem. A 116, 7189–7195 (2012)

190

M. A. Duncan

218. Shil, S., Bhattacharya, D., Sarkar, S., Misra, A.: Performance of the widely used Minnesota density functionals for the prediction of heat of formations, ionization potentials of some benchmarked first row transition metal complexes. J. Phys. Chem. A 117, 4945–4955 (2013) 219. Frenking, G., Fröhlich, N.: The nature of the bonding in transition-metal compounds. Chem. Rev. 100, 717–774 (2000) 220. Huber, K.P., Herzberg, G.: Constants of Diatomic Molecules. Van Nostrand Reinhold, New York (1979) 221. Khan, F.A., Clemmer, D.E., Schultz, R.H., Armentrout, P.B.: Sequential bond energies of chromium carbonyls (Cr(CO)+x , x = 1–6). J. Phys. Chem. 97, 7978–7987 (1993) 222. Sievers, M.R., Armentrout, P.B.: Collision-induced dissociation studies of V(CO)+x , x = 1–7: sequential bond energies and the heat of formation of V(CO)6 . J. Phys. Chem. 99, 8135–8141 (1995) 223. Meyer, F., Chen, Y.M., Armentrout, P.B.: Sequential bond energies of Cu(CO)+x and Ag(CO)+x (x = 1–4). J. Am. Chem. Soc. 117, 4071–4081 (1995) 224. Goebel, S., Haynes, C.L., Khan, F.A., Armentrout, P.B.: Collision-induced dissociation studies of Co(CO)+x , x = 1–5: sequential bond energies and heats of formation of Co(CO)4 . J. Am. Chem. Soc. 117, 6994–7002 (1995) 225. Meyer, F., Armentrout, P.B.: Sequential bond energies of Ti(CO)+x , x = 1–7. Mol. Phys. 88, 187–197 (1996) 226. Zhang, X.G., Armentrout, P.B.: Sequential bond energies of Pt(CO)+x (x = 1–4) determined by collision-induced dissociation. Organometallics 20, 4266–4273 (2001) 227. Bauschlicher Jr., C.W.: Transition metal-ligand bonding, II. J. Chem. Phys. 84, 260–267 (1986) 228. Bauschlicher Jr., C.W., Bagus, P.S., Nelin, C.J., Roos, B.J.: The nature of bonding in XCO for X = Fe, Ni and Cu. J. Chem. Phys. 85, 354–364 (1986) 229. Bauschlicher Jr., C.W., Barnes, L.A.: On the dissociation energies and bonding in NiCO+ and TiCO+ . Chem. Phys. 124, 383–394 (1988) 230. Barnes, L.A., Rosi, M., Bauschlicher Jr., C.W.: Theoretical studies of the first- and second-row mono- and di-carbonyl positive ions. J. Chem. Phys. 93, 609–624 (1990) 231. Sodupe, M., Bauschlicher Jr., C.W., Lee, T.J.: The calculation of the vibrational frequencies of CuCO+ , NiCO and CuCH3 . Chem. Phys. Lett. 189, 266–272 (1992) 232. Goldman, A.S., Krogh-Jespersen, K.: Why do cationic carbon monoxide complexes have high CO stretching force constants and short CO bonds? Electrostatic effects, not σ bonding. J. Am. Chem. Soc. 118, 12159–12166 (1996) 233. Lupinetti, A.J., Fau, S., Frenking, G., Strauss, S.H.: Theoretical analysis of the bonding between CO and positively charged atoms. J. Phys. Chem. 101, 9551–9559 (1997) 234. Lupinetti, A.J., Frenking, G., Strauss, S.H.: Nonclassical metal carbonyls. Angew. Chem. Int. Ed. 37, 2113–2116 (1998) 235. Lupinetti, A.J., Strauss, S.H., Frenking, G.: Non-classical metal carbonyls. Prog. Inorg. Chem. 49, 1–112 (2001) 236. O’Dwyer, M.F.: Infrared spectra and normal coordinate analysis of iron pentacarbonyl from vibrational spectra of isotopic species. J. Mol. Spec. 2, 144 (1958) 237. Shufler, S.L., Sternberg, H.W., Friedel, R.A.: Infrared spectrum and structure of chromium hexacarbonyl, Cr(CO)6 . J. Am. Chem. Soc. 78, 2687 (1956) 238. Boquet, G., Birgone, M.: Infrared spectra of Ni(CO)4 in the gas phase. Spectrochim. Acta 27, 139 (1971) 239. Ryther, R.J., Weitz, E.: Reaction kinetics of coordinatively unsaturated iron carbonyls formed on gas-phase excimer laser photolysis of iron pentacarbonyl. J. Phys. Chem. 95, 9841–9852 (1991) 240. Weitz, E.: Transient infrared spectroscopy as a probe of coordinatively unsaturated metal carbonyls in the gas phase. J. Phys. Chem. 98, 11256–11264 (1994) 241. Harvey, J.N., Aschi, M.: Modeling spin-forbidden reactions: recombination of carbon monoxide with iron tetracarbonyl. Faraday Discuss. 124, 129–143 (2003)

6 Metal Cation Coordination and Solvation Studied …

191

242. Besora, M., Carreón-Macedo, J.-L., Cimas, Á., Harvey, J.N.: Spin-state changes and reactivity in transition metal chemistry: reactivity of iron tetracarbonyl. Adv. Inorg. Chem. 61, 573–623 (2009) 243. Besora, M., Carreón-Macedo, J.-L., Cowan, A.J., George, M.W., Harvey, J.N., Portius, P., Ronayne, K.L., Sun, X.-Z., Towrie, M.A.: Combined theoretical and experimental study on the role of spin states in the chemistry of Fe(CO)5 photoproducts. J. Am. Chem. Soc. 131, 3583–3592 (2009) 244. Asmis, K.R., Sauer, J.: Mass-selective vibrational spectroscopy of vanadium oxide cluster ions. Mass Spectrom. Rev. 26, 542–562 (2007) 245. Sodupe, M., Branchadell, V., Bauschlicher Jr., C.W.: Theoretical study of M+ –CO2 and OM+ CO systems for the first transition row metal atoms. J. Phys. Chem. A 101, 7854–7859 (1997) 246. Sievers, M.R., Armentrout, P.B.: Activation of carbon dioxide: gas-phase reactions of Y+ , YO+ and YO2 + with CO and CO2 . Inorg. Chem. 38, 397–402 (1999) 247. Armentrout, P.B., Koizumi, H., MacKenna, M.: Sequential bond energies of Fe+ (CO2 )n , n = 1–5, determined by threshold collision-induced dissociation and ab initio theory. J. Phys. Chem. A 109, 11365–11375 (2005) 248. Shimanouchi, T.: Molecular vibrational frequencies. In: Linstrom, P.J., Mallard, W.H. (eds.) NIST Chemistry WebBook, NIST Standard Reference Database Number 69, National Institute of Standards and Technology, Gaithersburg MD, 20899 (http://webbook.nist.gov) 249. Marinelli, P.J., Squires, R.R.: Sequential solvation of atomic transition-metal ions. The second solvent molecule can bind more strongly than the first. J. Am. Chem. Soc. 111, 4101–4103 (1989) 250. Dalleska, N.F., Honma, K., Sunderlin, L.S., Armentrout, P.B.: Solvation of transition metal ions by water. Sequential binding energies of M+ (H2 O)x (x = 1–4) for Ti to Cu determined by collision induced dissociation. J. Am. Chem. Soc. 116, 3519–3528 (1994) 251. Rosi, M., Bauschlicher Jr., C.W.: The binding energies of one and two water molecules to the first transition-row metal positive ions. J. Chem. Phys. 90, 7264–7272 (1989) 252. Rosi, M., Bauschlicher Jr., C.W.: The binding energies of one and two water molecules to the first transition-row metal positive ions II. J. Chem. Phys. 92, 1876–1878 (1990) 253. Poisson, L., Pradel, P., Lepetit, F., Reau, F., Mestdagh, J.-M., Visticot, J.-P.: Binding energies of first and second shell water molecules in the Fe(H2 O)2+ , Co(H2 O)2+ and Au(H2 O)2+ cluster ions. Eur. Phys. J. D 14, 89–95 (2001) 254. Jayaweera, P., Blades, A.T., Ikonomou, M.G., Kebarle, P.: Production and study in the gas phase of multiply charged solvated or coordinated metal ions. J. Am. Chem. Soc. 112, 2452–2454 (1990) 255. Stace, A.J.: Metal ion solvation in the gas phase: the quest for higher oxidation states. J. Phys. Chem. A 106, 7993–8005 (2002) 256. Schröder, D., Schwarz, H.: Generation, stability, and reactivity of small multiply charged ions in the gas phase. J. Phys. Chem. A 103, 7385–7394 (1999) 257. Beyer, M., Williams, E.R., Bondybey, V.E.: Unimolecular reactions of dihydrated alkaline earth metal dications M2+ (H2 O)2 , M = Be, Mg, Ca, Sr, and Ba: salt-bridge mechanism in the proton-transfer reaction M2+ (H2 O)2 → MOH+ + H3 O+ . J. Am. Chem. Soc. 121, 1565–1573 (1999) 258. Rodriquez-Cruz, S.E., Jockusch, R.A., Williams, E.R.: Binding energies of hexahydrated alkaline earth metal ions, M2+ (H2 O)6 , M = Mg, Ca, Sr, Ba: evidence of isomeric structures for magnesium. J. Am. Chem. Soc. 121, 1986–1987 (1999) 259. Rodriquez-Cruz, S.E., Klassen, J.S., Williams, E.R.: Hydration of gas-phase ions formed by electrospray ionization. J. Am. Soc. Mass Spectrom. 10, 958–968 (1999) 260. Rodriquez-Cruz, S.E., Jockusch, R.A., Williams, E.R.: Hydration energies and structures of alkaline earth metal ions, M2+ (H2 O)n , n = 5–7, M = Mg, Ca, Sr and Ba. J. Am. Chem. Soc. 121, 8898–8906 (1999) 261. Bush, M.F., Saykally, R.J., Williams, E.R.: Formation of hydrated triply charged metal ions from aqueous solutions using nanodrop mass spectrometry. Int. J. Mass Spectrom. 253, 256–262 (2006)

192

M. A. Duncan

262. Shvartsburg, A.A., Siu, K.W.M.: Is there a minimum size for aqueous doubly charged metal cations? J. Am. Chem. Soc. 123, 10071–10075 (2001) 263. Walker, N.R., Grieves, G.A., Jaeger, J.B., Walters, R.S., Duncan, M.A.: Generation of unstable doubly charged metal ion complexes in a laser vaporization cluster source. Int. J. Mass Spec. 228, 285–295 (2003) 264. Beyer, M.K.: Hydrated metal ions in the gas phase. Mass Spectrom. Rev. 26, 517–541 (2007) 265. Trachtman, M., Markham, G.D., Glusker, J.P., George, P., Bock, C.W.: Interactions of metal ions with water: Ab initio molecular orbital studies of structure, bonding enthalpies, vibrational frequencies and charge distributions. 1. Monohydrates. Inorg. Chem. 37, 4421–4431 (1998) 266. Markham, G.D., Glusker, J.P., George, P., Bock, C.W.: The arrangement of first- and secondsphere water molecules in divalent magnesium complexes: results from molecular orbital and density functional theory and from structural crystallography. J. Phys. Chem. B 106, 5118–5134 (2002) 267. Irigoras, A., Fowler, J.E., Ugalde, J.M.: Reaction of Sc+ and V+ with water. J. Am. Chem. Soc. 121, 574–580 (1999) 268. Irigoras, A., Fowler, J.E., Ugalde, J.M.: Reaction of Cr+ , Mn+ and Fe+ with water. J. Am. Chem. Soc. 121, 8549–8558 (1999) 269. Irigoras, A., Elizalde, O., Silanes, I., Fowler, J.E., Ugalde, J.M.: Reactivity of Co+ (3 F,5 F), Ni+ (2 D,4 F), and Cu+ (1 S,3 D): reaction of Co+ , Ni+ and Cu+ with water. J. Am. Chem. Soc. 122, 114–122 (2000) 270. Feller, D., Glendening, E.D., de Jong, W.A.: Structures and binding enthalpies of M+ (H2 O)n clusters, M = Cu, Ag. Au. J. Chem. Phys. 110, 1475–1491 (1999) 271. Lee, E.C., Lee, H.M., Tarakeshwar, P., Kim, K.S.: Structures, energies and spectra of aquasilver (I) complexes. J. Chem. Phys. 119, 7725–7736 (2003) 272. Lee, H.M., Min, S.K., Lee, E.C., Min, J.-H., Odde, S., Kim, K.S.: Hydrated copper and gold monovalent cations: Ab initio study. J. Chem. Phys. 122, 064314 (2005) 273. Taylor, M.S., Muntean, F., Lineberger, W.C., McCoy, A.B.: A theoretical and computational study of the anion, neutral and cation Cu(H2 O) complexes. J. Chem. Phys. 121, 5688–5699 (2004) 274. Reveles, J.U., Calaminich, P., Beltran, M.R., Köster, A.M., Khanna, S.N.: H2 O nucleation about Au+ . J. Am. Chem. Soc. 129, 15565–15571 (2007) 275. Miliordos, E., Xantheas, S.S.: Elucidating the mechanism behind the stabilization of multicharged metal cations in water: a case study of the electronic states of microhydrated Mg2+ , Ca2+ and Al3+ . Phys. Chem. Chem. Phys. 16, 6886–6892 (2014) 276. Miliordos, E., Xantheas, S.S.: Unimolecular and hydrolysis channels for the detachment of water from microsolvated alkaline earth dication (Mg2+ , Ca2+ , Sr2+ , Ba2+ ) clusters. Theor. Chem. Acc. 133, 1450 (2014) 277. Garza-Galindo, R., Castro, M., Duncan, M.A.: Theoretical study of nascent hydration in the Fe+ (H2 O)n system. J. Phys. Chem. A 116, 1906–1913 (2012) 278. Western, C.M.: PGOPHER: a program for simulating rotational, vibrational, and electronic spectra. J. Quant. Spectros. Rad. Trans. 186, 221–242 (2017) 279. Muertterties, E.L., Bleeke, J.R., Wucherer, E.J., Albright, T.A.: Structural, stereochemical and electronic features of arene-metal complexes. Chem. Rev. 82, 499–525 (1982) 280. Ma, J.C., Dougherty, D.A.: The cation-π interaction. Chem. Rev. 97, 1303–1324 (1997) 281. Dougherty, D.A.: The cation-π interaction. Acc. Chem. Res. 46, 853–885 (2013) 282. Mahadevi, A.S., Sastry, G.N.: Cation-π interaction: its role and relevance in chemistry, biology, and materials science. Chem. Rev. 113, 2100–2138 (2013) 283. Sodupe, M., Bauschlicher Jr., C.W.: Theoretical study of the bonding of the first- and secondrow transition metal positive ions to acetylene. J. Phys. Chem. 95, 8640–8645 (1991) 284. Stockigt, D., Schwarz, J., Schwarz, H.: Theoretical and experimental studies on the bond dissociation energies of Al(methane)+ , Al(acetylene)+ , Al(ethene)+ , and Al(ethane)+ . J. Phys. Chem. 100, 8786–8790 (1996)

6 Metal Cation Coordination and Solvation Studied …

193

285. Gidden, J., Van Koppen, P.A.M., Bowers, M.T.: Dehydrogenation of ethene by Ti+ and V+ : excited state effects on the mechanism for C−H bond activation from kinetic energy release distributions. J. Am. Chem. Soc. 119, 3935–3941 (1997) 286. Sievers, M.R., Jarvis, L.M., Armentrout, P.B.: Transition metal ethene bonds: thermochemistry of M+ (C2 H4 )n (M = Ti – Cu, n = 1 and 2) complexes. J. Am. Chem. Soc. 120, 1891–1899 (1998) 287. Manard, M.J., Kemper, P.R., Carpenter, C.J., Bowers, M.T.: Dissociation reactions of diatomic silver cations with small alkenes: experiment and theory. Int. J. Mass Spectrom. 241, 99–108 (2005) 288. Sharma, P., Attah, I., Momoh, P., El-Shall, M.S.: Metal acetylene cluster ions M+ (C2 H2 )n as a model for studying reactivity of laser-generated transition metal cations. Int. J. Mass Spectrom. 300, 81–90 (2011) 289. Meyer, F., Khan, F.A., Armentrout, P.B.: Thermochemistry of transition metal benzene complexes: binding energies of M(C6 H6 )+x (x = 1, 2) for Ti to Cu. J. Am. Chem. Soc. 117, 9740–9748 (1995) 290. Lin, C.-Y., Dunbar, R.C.: Radiative association kinetics and binding energies of chromium ions with benzene and benzene derivatives. Organometallics 16, 2691–2697 (1997) 291. Li, Y., Baer, T.: Dissociation kinetics of energy-selected (C6 H6 )Cr+ ions: benzene-chromium neutral and ionic bond energies. J. Phys. Chem. A 106, 9820–9826 (2002) 292. Willey, K.F., Cheng, P.Y., Pearce, K.D., Duncan, M.A.: Photoinitiated charge transfer and dissociation in mass-selected metalloorganic complexes. J. Phys. Chem. A 94, 4769–4772 (1990) 293. Willey, K.F., Cheng, P.Y., Bishop, M.B., Duncan, M.A.: Charge-transfer photochemistry in ion-molecule cluster complexes of silver. J. Am. Chem. Soc. 113, 4721–4728 (1991) 294. Willey, K.F., Yeh, C.S., Robbins, D.L., Duncan, M.A.: Charge-transfer in the photodissociation of metal ion-benzene complexes. J. Phys. Chem. 96, 9106–9111 (1992) 295. Jaeger, T.D., Duncan, M.A.: Photodissociation of M+ (benzene)x complexes (M = Ti, V, Ni) at 355 nm. Int. J. Mass Spectrom. 241, 165–171 (2005) 296. Hoshino, K., Kurikawa, T., Takeda, H., Nakajima, A., Kaya, K.: Structures and ionization energies of sandwich clusters (Vn (benzene)m ). J. Phys. Chem. 99, 3053–3055 (1995) 297. Yasuike, T., Nakajima, A., Yabushita, S., Kaya, K.: Why do vanadium atoms form multipledecker sandwich clusters with benzene molecules efficiently? J. Phys. Chem. A 101, 5360–5367 (1997) 298. Nakajima, A., Kaya, K.: A novel network structure of organometallic clusters in the gas phase. J. Phys. Chem. A 104, 176–191 (2000) 299. Bauschlicher Jr., C.W., Partridge, H., Langhoff, S.R.: Theoretical study of transition-metal ions bound to benzene. J. Phys. Chem. 96, 3273–3278 (1992) 300. Dargel, T.K., Hertwig, R.H., Koch, W.: How do coinage metal ions bind to benzene? Mol. Phys. 96, 583–591 (1999) 301. Yang, C.N., Klippenstein, S.J.: Theory and modeling of the binding in cationic transition metal-benzene complexes. J. Phys. Chem. 103, 1094–1103 (1999) 302. Chaquin, P., Costa, D., Lepetit, C., Che, M.: Structure and bonding in a series of neutral and cationic transition metal-benzene η6 complexes [M(η6 –C6 H6 )]n+ (M = Ti, V, Cr, Fe, Co, Ni, and Cu). Correlation of charge transfer with the bathochromic shift of the e1 ring vibration. J. Phys. Chem. A 105, 4541–4545 (2001) 303. Kim, D., Hu, S., Tarakeshwar, P., Kim, K.S., Lisy, J.M.: Cation-π interactions: a theoretical investigation of the interaction of metallic and organic cations with alkenes, arenes, and heteroarenes. J. Phys. Chem. A 107, 1228–1238 (2003) 304. Horváthová, L., Dubecký, M., Mitas, L., Štich, I.: Spin multiplicity and symmetry breaking in vanadium-benzene complexes. Phys. Rev. Lett. 109, 053001 (2012) 305. Horváthová, L., Dubecký, M., Mitas, L., Štich, I.: Quantum Monte Carlo study of π-bonded transition metal organometallics: Neutral and cation vanadium-benzene and cobalt-benzene half sandwiches. J. Chem. Theory Comput. 9, 390–400 (2013)

194

M. A. Duncan

306. Maner, J.A., Mauney, D.T., Duncan, M.A.: Imaging charge transfer in a cation-π system: velocity-map imaging of Ag+ (benzene) photodissociation. J. Phys. Chem. Lett. 6, 4493–4498 (2015) 307. Snavely, D.L., Walters, V.A., Colson, S.D., Wiberg, K.B.: FTIR spectrum of benzene in a supersonic expansion. Chem. Phys. Lett. 103, 423–429 (1984) 308. Reishus, N.R., Duncan, M.A.: IR spectroscopy in the fingerprint region for vanadium- and cobalt-benzene complexes. Work in progress 309. Carnegie, P. D., Duncan, M. A.: Water solvation shells around vanadium cations, unpublished work

Chapter 7

Superatomic Nanoclusters Comprising Silicon or Aluminum Cages Atsushi Nakajima

X@Al12 39

40

41

B@Al12 Si@Al12 P@Al12

Ih

2P shell

Binary Cage Superatom M@Si16 Td

67

68

69

Sc@Si16 Ti@Si16 V@Si16

2D shell Periodic table for binary cage superatoms of X@Al12 and M@Si16

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_7

195

196

A. Nakajima

Abstract This chapter describes two superatoms, each comprising a central atom and a silicon or aluminum cage. Binary nanoclusters (NCs) at optimized mixing ratios are key components in designing the functionalities relevant to their electronic properties. To form chemically robust functional NCs, it is important to design the cooperatively synergistic effects between the electronic and geometric structures because these stabilize the individual NCs not only against charge transfer into the corresponding cations or anions but also against structural perturbations in their assemblies. Among binary NCs, synergistic effects are particularly expected when one central atom encapsulating cage structure completes a specific electron shell because electronic and geometric factors can operate simultaneously. Although the term “superatom” is widely used when the valence electrons in NCs complete an electron shell, more synergistic effects appear when the superatom adopts a closepacked structure, such as a highly symmetric cage as a binary cage superatom. Representative examples are given for one central atom encapsulated by silicon and aluminum cages, M@Si16 and X@Al12 , their formation and characterization are described, and a large-scale synthetic approach is established for M@Si16 . The perspectives for binary cage superatom assembly are discussed in terms of theoretical calculations. Keywords Nanocluster · Superatom · Binary cage superatom · Silicon cage · Aluminum cage, superatom salt · Dimeric superatom · Superatom assembly

7.1 Introduction Aluminum (Al) and silicon (Si) are elements adjacent to each other in the periodic table. However, their electrical characteristics are distinctly different; Al is highly conductive, while Si is semiconductive. Both are found abundant on the earth, and they are indispensable elements for technological civilizations in modern society. Si element has greatly contributed to modern society as an excellent semiconductor electronic material from the mid-twentieth century, in terms of innovative progress in informationalization. Particularly, photolithographic cycles for patterning circuits on Si substrates have allowed the development of highly integrated electronic devices by large-scale integration (LSI). Moreover, LSI has been continuously downsized with shorter wavelength lithography until the twenty-first century according to Moore’s law [1]. Even though the miniaturization is finally approaching a physical limit of a scale of several nanometers, Si element still plays a central role as the main electronic material in modern society. In contrast to the top-down approach such as A. Nakajima (B) Department of Chemistry, Faculty of Science and Technology, Keio University, 3-14-1 Hiyoshi, Kohoku-Ku, 223-8522 Yokohama, Japan e-mail: [email protected] Keio Institute of Pure and Applied Sciences (KiPAS), Keio University, 3-14-1 Hiyoshi, Kohoku-Ku, 223-8522 Yokohama, Japan

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

197

fine photolithography toward bulk Si, it is also crucial to explore electronic materials by a bottom-up approach using Si atoms. Beyond the physical limit of several nanometers, creating nanoscale Si compounds can provide novel functionalities, including luminescence and thermoelectric properties [2–5]. Above all, nanoclusters (NCs) of several to hundreds of atomic aggregates allow us to develop the potential for promising Si compounds for exploring the bottom-up Si nanotechnology. When fullerene C60 was discovered in a NC beam from carbon vapor [6], much attention was also focused on creating nanoscale Si species having the same cage structure as elemental Si [7, 8], because they belong to the same group in the periodic table. Although further development was required to establish caged Si NCs compared to the progress with C60 , recently these showed rich chemistry between the gas phase and the condensed phase [9–13]. A study on the cage structure of Si atoms was reported in 1987 [7]. Based on the mass spectrum, Dr. Beck reported that when one molybdenum (Mo) atom was mixed with Si vapor, MoSi15 + and MoSi16 + cations were generated. He deduced that a Mo atom was contained in a Si cage, which was inferred from analogous metallofullerenes in which metal atoms are contained inside the hollow cage of a carbon fullerene [14]. Although experimental and theoretical studies were extensively performed [15–25], it was not easy to distinguish new nanoscale Si compounds, including a Si cage. On the other hand, Al element is a low-density metal that is resistant to corrosion, so it is widely used as a building material in the aerospace industry. Al element forms alloys with various elements, widening the diversity of physical properties in various materials. When miniaturizing Al to form NCs, an emerging feature is an electron shell structure [26–28], which was found in NCs consisting of alkali or coinage metals. Particularly, Al13 − affords a 40 electron-completing 2P shell together with an icosahedral close-packed structure, and this is known as a “superatom”, mimicking the atomic configuration in the periodic table [29–33]. The novel electronic properties of Al superatoms are intriguing with respect to fabricating nanostructured functional materials. This chapter describes two binary cage superatoms (BCSs) of metal-atomencapsulating Si (M@Si16 ) [9–13] and heteroatom-encapsulating Al (X@Al12 ) [34–40] (Fig. 7.1); M = group 3–5 transition metals and X = boron (B), Si, phosphorus (P), scandium (Sc), and titanium (Ti). To form chemically robust functional NCs, it is important to design the cooperatively synergistic effects between electronic and geometric structures because these stabilize the NCs individually not only against charge transfer into the corresponding cations or anions but also against structural perturbations in their assemblies. The synergistic effect is particularly expected for BCSs in which one central atom encapsulating cage structure completes a specific electron counting because electronic and geometric factors can then work simultaneously, retaining their structural symmetry. Although the term “superatom” is widely used when the valence electrons in NCs complete an electron shell [29–33], more synergistic effects appear when the superatom takes a close-packed structure such as a highly symmetric cage as a BCS [13, 38, 39]. For representative examples of M@Si16 and X@Al12 , which have T d and I h symmetries, respectively, the formation

198

A. Nakajima

Fig. 7.1 Binary cage superatoms (BCSs) of M@Si16 (M = Ti and Ta) and Si@Al12 . Reprinted with permission from Ref. [13]. Copyright 2017 American Chemical Society

and characterization are described and a large-scale synthetic approach is established for M@Si16 [12, 13]. The perspectives for the binary cage superatom assembly are discussed in terms of theoretical calculations [39, 41–43].

7.2 Experimental Methods for Gas Phase Nanoclusters 7.2.1 Dual-Laser Vaporization Nanocluster Source Figure 7.2 shows a schematic of the NC source of face-to-face laser plasma mixed with pulsed helium (He) carrier gas [44, 45]. An Even-Lavie pulsed valve with a high stagnation pressure of 60–100 atm. [46] was used at a repetition rate of 10 Hz, which and XAl+/0/– . Intense supersonic was suitable for the effective formation of MSi+/0/– n n He gas pulses were operated to recombine and cool the laser-ablated metal vapor. Fig. 7.2 Schematic of a dual-laser vaporization NC source with two sample rods, two pulsed valves, and a reaction room. Reproduced from Ref. [45] with permission from The Chemical Society of Japan

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

199

The optimum laser fluence required to maximize the NC intensities depends on the properties of the target under investigation; specifically, the dual-laser vaporization method allows control of the elemental mixing ratio. The NC source was modified to be “face-to-face” so that the NCs could be more effectively generated [44]. In the primary NC source, two target rods were laterally separated by 4–5 mm, and two lasers independently vaporized their front surfaces. The two lasers then irradiated the two rods with a delay of about 5 μs in synchronization with the speed of the He carrier gas. Hence, in addition to binary mixed NCs, individual NCs consisting of each single element were generated, because the time lag between the vaporization and cooling processes significantly reduces the formation efficiency of the binary NCs. The focusing positions of the two vaporization lasers were then shifted toward each other by 1.5–2 mm and moved off the front surfaces of the target rods. The shift in the focusing positions facilitated vaporization of the curved surface of the sample rods in a face-to-face manner [47]. The effective mixing of the hot sample plasmas allowed efficient formation of the binary NCs when the two rods were almost simultaneously vaporized, within a few hundred nanoseconds (ns). The simultaneous vaporization enabled us to form binary NCs with one pulsed laser to vaporize the two independent rods, although optimization of the laser fluences became more complex. The mixed vapor formed binary NCs, and their reactivity was examined by exposing reactant gas in the downstream reaction room [48], as shown in Fig. 7.2. The binary NCs passed through a source exit and expanded into a differentially pumped chamber through a skimmer.

7.2.2 Spectroscopic Methods for Nanoclusters The electronic and geometric structures of the binary NCs were investigated by mass spectrometry, anion photoelectron spectroscopy (PES), and photoionization spectroscopy (PIS). Mass analysis of the neutral binary NCs ionized with an ArF (193 nm; 6.43 eV) or F2 (157 nm; 7.90 eV) laser and of the charged binary NCs was performed using time-of-flight (TOF) mass spectrometry. For cationic and anionic NCs, the beam was directly accelerated with a pulsed voltage approximately 2 keV, while neutral photoionization with the ArF/F2 laser was applied in a static electric field of 2 keV. To achieve the appropriate conditions for one-photon ionization of the NCs with the F2 laser, the laser power dependence was measured by changing the flow rate of He gas toward the laser path tubing between the laser exit and the CaF2 window of the chamber. A laser fluence below 1 mJ/cm2 was used, where the ion intensity was linearly dependent on the laser power. For the PIS measurements of the binary NC neutrals, photoionization efficiency curves were measured with a tunable photoionization laser (5.2–6.4 eV) of an optical parametric oscillator (OPO), and the ionization energy was determined from the threshold energy. The laser fluence (typically around 300 μJ/cm2 ) was monitored during the measurements to normalize the ion intensities [49].

200

A. Nakajima

For the PES measurements of the binary NC anions, a magnetic-bottle-type TOF electron spectrometer was used [50–52]. After mass selection in their TOF with pulse acceleration at 900 eV, their kinetic energy was considerably reduced with a pulsed electric decelerator before they entered the photodetachment region. The fifth harmonic (213 nm, 5.83 eV) of a pulsed Nd3+ :YAG laser was then used to irradiate the mass-selected NC anions to detach photoelectrons. The electrons were guided by a strong, inhomogeneous magnetic field with an Nd–Fe–B-based permanent magnet [53], and subsequently by a weak guiding magnetic field produced with an electric current (2–3 A), and detected with a microchannel plate (MCP). Their kinetic energy was analyzed from their TOF and calibrated using the Au– (2 S1/2 ← 1 S0 ) transition [54, 55]. The photoelectron signal was typically accumulated over 20,000–40,000 laser shots.

7.3 Silicon (Si)-based Binary Cage Superatoms (BCSs) 7.3.1 Mass Spectrometry for Si Cage Compounds In contrast to C60 , stable nanoscale compounds comprising only Si atoms have not yet been discovered, but we found that a Si NC containing one metal atom was strongly distributed in the mass spectrum as a magic number by systematically changing the metal atom (M) [9, 44, 45, 56]. To explore the binary M–Si formation, we used the NC source with dual-laser vaporization (Fig. 7.2) to enhance the mixing of hot atomic Si and M vapor, and the NCs produced were mass-analyzed using the TOF spectrometer. The magic number behavior featured with a specific composition coupled with the charge state [9, 45]. Figure 7.3 shows mass spectra for an M–Si NC beam obtained by mixing metal atom M vapor of Sc, Ti, and vanadium (V) of groups 3–5 with Si atoms in the dual-laser vaporization source in cationic, neutral, and anionic charge states. In some TOF mass spectra, the magic number (black arrow) appeared when one M atom is mixed with 16 Si atoms. In nine spectra in Fig. 7.3a–c, the mass spectra in which the magic number appears move from the lower left to the upper right due to the combination of the metal element and charge state [9, 45]. The origin of the magic number is that the NC affords a total valence electron number of 68, and in the Jellium model for NCs with uniform charge distribution [26–28], the 68 electrons correspond to the number of closed shell electrons, up to the 2D shell. Furthermore, the most distinct magic number behavior appears with tantalum (Ta) atoms and in their cations (Fig. 7.3e); TaSi16 + ions are predominant compared to their neighbors [44, 45, 56]. This appears attributable to the interplay between electronic and geometric structures; the TaSi16 + cations afford the total valence electron number of the closed shell and are also stabilized by the size of the Ta atom being most suitable for the inner diameter of the Si16 cage. Similarly, TiSi16 neutrals are exclusively produced among neutral forms. Therefore, we structurally

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

titanium (Ti)

vanadium (V)

anion

neutral

cation

scandium (Sc)

201

200

400

200

600

400

600

cation

Si16Nb+

400

400

600

tantalum (Ta)

niobium (Nb)

200

200

Si16Ta+

600

200

400

Mass number (m/z)

600

Fig. 7.3 Mass spectra of cations, neutrals, and anions for M–Si NC beam (M = scandium (Sc), titanium (Ti), and vanadium (V)), together with those of niobium (Nb) and tantalum (Ta) cations mixed with Si. Black arrows show magic number peaks, while white arrows show non-magic number peaks. Reproduced from Ref. [45] with permission from The Chemical Society of Japan

evaluated metal-encapsulating Si-based BCSs by further spectroscopic characterization coupled with large-scale synthesis of Ta@Si16 and Ti@Si16 . To quantitatively evaluate the electronic properties of M@Si16 , anion PES was used [9]. Figure 7.4 shows the photoelectron spectra for Sc@Si16 – , Ti@Si16 – , and V@Si16 – using 266 nm (4.66 eV) and 213 nm (5.82 eV) detachment lasers. The binding energies of Sc@Si16 – and V@Si16 – reach 3 eV or more, whereas that of Ti@Si16 – is small at around 2 eV. This shows that the electronic stabilization is large due to the pairing energy when the total number of valence electrons in the anion is an even number, while an odd number makes the anion unstable. The threshold binding energy corresponds to the EA of the corresponding neutral, and the EA is shown in Fig. 7.4. Moreover, since Sc@Si16 – possesses 68 electrons to complete the 2D shell, Ti@Si16 – has one excess electron, and thus the peak labeled X in the Ti@Si16 – spectrum appears to be an additional peak to that of Sc@Si16 – . To confirm that peak X corresponds to a singly occupied molecular orbital (SOMO),

202

A. Nakajima

X

Relative Electron Intensity

(a) 3.41

2.03

0.12

(f)

0.09

A

A (d)

X

3.08

X

X

(b)

A

(c)

B

0.13

(g)

A

1.90

X

X 0

1

2

3

4

5

0

1

2

3

0

1

2

3

4

5

0.32

3.23

1

5

X’ A’

(e)

0

4

B

2

3

4

5

Electron Binding Energy / eV Fig. 7.4 Photoelectron spectra of BCS anions of Sc@Si16 – a, b, Ti@Si16 – c, d, and V@Si16 – f, g at 266 nm (4.66 eV; top three spectra) and at 213 nm (5.82 eV; bottom four spectra). Comparing the photoelectron spectrum of Ti@Si16 – with that of Ti@Si16 F– e enables us to assign the HOMO–LUMO gap. Reprinted with permission from Ref. [9]. Copyright 2005 American Chemical Society

Ti@Si16 F– was generated by adding an F atom to Ti@Si16 – and the photoelectron spectrum was obtained [57–59]. As shown in Fig. 7.4e, peak X disappears for Ti@Si16 F– while the spectral features in the higher binding energy region are retained. This is because the F atom (with one electron deficient) effectively accommodates one electron into a deeper MO. The spectral change enables us to assign peak X as the SOMO and to evaluate the gap between the highest occupied MO (HOMO) and the lowest unoccupied MO (LUMO) to be 1.90 eV, as shown in Fig. 7.4d. The HOMO–LUMO gap is considerably larger than that of C60 (1.57 eV) [60], demonstrating the high stability of the Ti@Si16 BCS. As suggested by theoretical calculations [61], cage “aromaticity” might be an important determinant of the electronic stability of the BCSs.

7.3.2 Development of Intense Nanocluster Source with Magnetron Sputtering The dual-laser vaporization source is very powerful to allow investigation of binary NC formation by changing the combinations of two different elements. However, structural analysis methods, such as nuclear magnetic resonance (NMR) and Raman spectroscopy, often require enhancement of the total number of NCs. Furthermore,

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

203

for well-controlled soft-landing, it is necessary to reduce the kinetic energy distribution of the beam, because a wide energy distribution might cause dissociation of the NCs in collisions with the substrate to which a bias voltage is applied [62]. Based on the production amount and narrow kinetic energy distribution, compared to laser evaporation, it is advantageous to use a magnetron sputtering method in which larger targets are available and NCs are formed in a steady flow of inert cooling gas, as shown in Fig. 1.14 [63, 64]. To form binary NCs, two choices are available: dual magnetron sputtering with two targets or single magnetron sputtering with a mixed target. Since the targeted BCSs of M@Si16 have a specific composition, it is convenient to use a mixed target to produce a stable beam from merging nascent atomic vapor. Considering the sputtering rates of each element, the mixing ratios of the target are optimized for generating M@Si16 BCSs. Particularly, size-selective soft-landing is performed only for charged NCs excluding neutrals, and then the NC ion density should be enhanced to efficiently accumulate the selected NCs on a substrate. For magnetron sputtering, high-power impulse magnetron sputtering (HiPIMS) [65–67] with a pulsed power supply increases the ion density while maintaining the average output. The HiPIMS NC source can provide 2–10 times higher ion density compared to conventional magnetron sputtering using a static DC power supply.

7.3.3 Immobilization and Characterization of Si-based BCSs on Solid Surface Using a quadrupole mass filter, only Ta@Si16 + BCS ions formed in the developed NC source were selected, and these were deposited and immobilized on a solid substrate [10, 68]. During the selective deposition, the Ta@Si16 produced is deposited under soft-landing conditions (kinetic energy below 1 eV per atom) to avoid destruction when colliding with the substrate. When the deposited TaSi16 was characterized by X-ray photoelectron spectroscopy (XPS) (Fig. 7.5) [11, 69, 70], both Si 2p and Ta 4f exhibit sharp peaks, showing (1) the preserved 1:16 composition of Ta:Si based on their intensity ratio and (2) a single chemical environment for each element based on their peak envelopes. These results can be explained by the Ta atom being encapsulated by the Si16 cage, because other structures, such as linear, two-dimensional planar ones would have provided broader XPS peaks due to different chemical environments for Si and Ta atoms. When the deposited substrate is heated to 350°C, although TaSi16 reacts slightly with residual oxygen, its oxidation initially occurs at Si atoms (Fig. 7.5, bottom), while Ta is not oxidized [11]. Generally, since naked Ta is more easily oxidized than Si [71, 72], the Ta atom must be encapsulated inside the Si16 cage (hereafter referred to as Ta@Si16 ). In this soft-landing method, Ta@Si16 can be deposited on the substrate to form several layers, but it is highly desirable to construct a more efficient synthesis methodology for further nanomaterial science as well as developing detailed structural analyses.

A. Nakajima

XPS intensity (count

104)

204

(a) Si 2p 10

(b) Ta 4f

4f7/2

4 4f5/2

2p3/2 2p1/2 2

5

0

0

(c) Si 2p Heated 6

Asdeposited 2 104 count

100

(d) Ta 4f

1 104 count

6

95 25 Binding energy (eV)

20

Fig. 7.5 XPS spectra of Ta@Si16 BCS deposited on HOPG around a Si 2p and b Ta 4f core levels. The fitted results (red line) and spin-orbit contributions (orange dotted line and green dash-dotted line) are superimposed in (a) and (b). Background-subtracted XPS spectra of the Ta@Si16 BCS film for (c) Si 2p and (d) Ta 4f before and after heating (720 K, 16 h) are shown by blue and red lines, respectively. Reprinted with permission from Ref. [11]. Copyright 2015 American Chemical Society

7.3.4 Dispersion Trapping of Si-Based BCSs in Liquid To identify the structure of the metal-atom-encapsulating Si16 cage, a new apparatus was designed and constructed, as shown in Fig. 7.6, in which species generated in the beam with HiPIMS are dispersed in liquids by a direct liquid embedded trapping (DiLET) [12, 13]. The DiLET is based on the idea that chemical isolation toward liquids trapping all beam species is much more efficient than the size-selective deposition, because (1) neutral NCs can also be captured by liquids and (2) the ion transmittance in the mass selection is as low as 20%. Since the NC source operates under vacuum, a liquid with a low vapor pressure must be used to achieve a vacuum of around 10−2 Pa, and thus NCs are injected into polyethylene glycol dimethyl ether (PEG-DME) liquid. During injection, a fresh liquid surface must be prepared to prevent aggregation of various NCs, and then the liquid is stirred vigorously during beam injection. Liquid trapping was applied to Ti@Si16 as well as Ta@Si16 using a Ti or Ta mixed Si disk target. As the beam was injected into the PEG-DME liquid, the transparent liquid became brown after about 1 h, and after about 3 h, the liquid was removed from the chamber and treated in a glove box under reduced oxygen and moisture. While purifying and isolating the liquid with appropriate solvents, fractionation was repeated three times by changing the mixing ratio of nonpolar hexane and large polar tetrahydrofuran (THF), and finally a fraction soluble in pure THF was obtained.

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

205

Fig. 7.6 Schematic of NC synthesis apparatus based on HiPIMS and DiLET. Reprinted with permission from Ref. [12]. Copyright 2017 American Chemical Society

7.3.5 Structural Analysis of Si-based BCSs The resulting fraction was analyzed by XPS, 29 Si-NMR, Raman spectroscopy, and quantum chemistry calculations [12]. First, XPS was performed for the obtained Ta–Si fraction, and the Si 2p and Ta 4f spectra were consistent with those in Fig. 7.5. This clearly demonstrates that Ta@Si16 BCSs in the beam can be successfully dispersed in PEG-DME liquid and that Ta@Si16 BCSs can be obtained as stable species even after both solvent extraction and ligation with PEG-DME. In fact, the isolation of Ta@Si16 BCSs was also confirmed by mass spectrometry [12]. Furthermore, in the Raman spectra, broad peaks were observed at around 120, 300, and 450 cm−1 for both Ti–Si and Ta–Si (upper traces in Fig. 7.7a, b). Importantly, the features of these Raman spectra are consistent with those of naked BCSs in the surfaceenhanced Raman spectra (SERS) (lower traces in Fig. 7.7a, b), which are obtained by depositing exclusively Ti@Si16 or Ta@Si16 BCS ions on the substrate. Furthermore, the successful synthesis of M@Si16 BCSs on a 100 mg scale enables us to apply NMR measurements; 29 Si–NMR spectra for Ti@Si16 and Ta@Si16 exhibit one and two peaks, respectively, between 100 and −100 ppm (Fig. 7.8). Although these peak positions are inconsistent with the predictions obtained from quantum chemistry calculations, the result suggests (1) the difficulty of calculation-based prediction, (2) the weak coordination effect of PEG-DME, and (3) structural fluctuations of the Si16 cage. Based on these spectroscopic structural evaluations, it was concluded that both Ti@Si16 and Ta@Si16 BCSs are identified as tetrahedral structures of Si16

206 Fig. 7.7 Raman spectra excited at 532 nm for isolated M@Si16 :PEG-DME BCS and size-selected naked M@Si16 BCS on the SERS substrate (M@Si16 /Ag/SrTiO3 ); M = (a) Ti and (b) Ta. Stick bars represent the selected Raman active modes calculated by DFT for FK, dist-FK, and f-D4d isomers. Reprinted with permission from Ref. [12]. Copyright 2017 American Chemical Society

Fig. 7.8 29 Si NMR spectra of M@Si16 :PEG-DME BCS dispersed in THF for M = Ti (300 K) and for M = Ta (318 K). Stick bars represent chemical shifts (CSs) calculated by ZORA-DFT for FK, dist-FK, and f-D4d isomers at the PBE0/TZ2P level. The CSs averaged over the sites are shown with faint colors. Reprinted with permission from Ref. [12]. Copyright 2017 American Chemical Society

A. Nakajima

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

207

derived from the Frank-Kasper (FK) structure: a metal-encapsulating tetrahedral Si cage (METS) (Fig. 7.1) [12, 13].

7.4 Aluminum (Al)-based Binary Cage Superatoms (BCSs) 7.4.1 Mass Spectrometry of Al-based BCSs 7.4.1.1

Aluminum–Silicon

Figure 7.9 shows the mass spectra of the Al–Si NC anions, neutrals, and cations produced [38, 39]. In the photoionization of the Aln Sim neutral, the laser power dependence indicated that one-photon ionization occurred with the F2 laser (7.90 eV). Since the mass of Si (28 u) is very close to that of Al (27 u), the mixed Al–Si NCs form bundles of mass peaks for each n + m. Although the most intense mass peaks were observed at n + m = 13 in the mass spectra, high-resolution TOF mass spectrometer [73] revealed that the most abundant peak is systematically changed. The right-hand figures in Fig. 7.9 show intensity distributions at n + m = 13, the most abundant 13-mers being Al13 − for anions, Al12 Si for neutrals, and Al11 Si2 + for cations. The

(a)

n+m=13

Anions: AlnSim

Al13

(a’)

Al12 Si

Intensity (arb. unit)

Fig. 7.9 Mass spectra of the Al–Si NC anions, neutrals, and cations. The Aln Sim neutrals are photoionized with the F2 laser (7.90 eV). The most intense mass peaks are observed at n + m = 13 in the mass spectra; intensity distributions at n + m = 13 are shown in the right-hand figures, The most abundant 13-mers are the Al13 – anion, Al12 Si neutral, and Al11 Si2 + cation. Reprinted with permission from Ref. [38]. Copyright 2006 American Chemical Society

350 352 354 356

(b)

n+m=13

Al12Si

(b’)

Neutrals: AlnSim Al13 Al11Si

350 352 354 356

(c)

n+m=13

Cations: AlnSim+

Al11Si2+ (c’) Al 12Si + Al13+

350 352 354 356

200

300 400 500 Mass number (m/z)

Mass number (m/z)

208

A. Nakajima

charge state dependence for the most abundant Al13 – , Al12 Si, and Al11 Si2 + shows that the doped Si atoms function as a tetravalent atom to satisfy 2P shell closure(40 e) in the Al to Si substitution, because both Al12 Si and Al11 Si2 + NCs possess 40 valence electrons in common in addition to Al13 − . Furthermore, a charge state dependence is observed in the size-dependent abundance in the mass spectra. Figure 7.10 shows − + the intensity distributions of the Al− n and Aln Si anions, Aln Si neutrals, and Aln Si + and Aln Si2 cations. Some intensity distributions exhibit magic number behavior at n + m = 13, representing the above-mentioned Al13 – , Al12 Si, and Al11 Si2 + . These results clearly confirm that the substitution of Si atoms corresponds to one-electron addition toward the total number of valence electrons in the Al–Si binary NCs. As reported earlier [15, 38], the chemical stability of the Aln X NCs can be examined using a chemical probe method. When the adsorption reactivity of the Al–Si binary NCs is measured toward O2 reactant gas, Al12 Si neutrals show prominent chemical inertness compared to the others [39], as shown in Fig. 7.11. In fact, Al12 Si neutrals are favored electronically as well as geometrically. A plausible explanation is that Al12 Si adopts a closed electron configuration when a tetravalent Si atom dopant completes the 2P electron shell combined with the valence electrons of sp-hybridized Al atoms [36, 37, 74, 75]. Furthermore, the diameter of a Si atom (1.18 Å) is slightly

Intensity (arb. units)

13-0

(a) AlnSi+

5 5

10 10

15 15

11-2+

20 20

(d)

25 25

AlnSi2+

12-1

(e) Aln

(b) AlnSi

(c) AlnSi

10 15 20 25 55 10 15 20 25 Number of Al atoms (n)

5 10 15 20 25 5 10 15 20 25 Number of Al atoms (n)

5 10 15 20 25 Number of Al atoms (n)

Fig. 7.10 Intensity distributions of the Aln Si+ and Aln Si2 + cations, Aln Si neutrals, and Al–n and Aln Si– anions. A prominent peak appears at n + m = 13 for Aln Si2 + , Aln Si, and Aln Si– . Solid arrows show the positions for Al12 Si+/– . Reprinted with permission from Ref. [38]. Copyright 2006 American Chemical Society

Fig. 7.11 Plots of the relative reactivity of the Aln Si1 NCs against exposure to O2 at n = 7–17 with experimental uncertainties. Reproduced from Ref. [39] with permission from the PCCP Owner Societies

Relative Reactivity (arb. unit)

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

209

1.0 0.8 0.6 0.4 0.2 0.0 8

10

12

14

16

Number of Al Atoms

smaller than that of an Al atom (1.43 Å) [76], leading to geometric stabilization owing to less distorted icosahedral structures. In order to make the icosahedral Al13 more geometrically stable, the Al12 cage favors a smaller central atom, because the distance between adjacent surface atoms is distortedly extended by 5% compared to the distance between the central and surface atoms [34, 77].

7.4.1.2

Aluminum–Boron

21-2 22-1

500

600

29-0

23-0

19-0

10-0

16-0

11-2

AlnBm

7-0

Fig. 7.12 Mass spectrum of the Al–B nanocluster anions (Aln B–m ). The most intense mass peaks are observed at n + m = 13–0 and 12–1 along with some enhancements around n + m = 23. Reprinted from Ref. [34] with permission from Elsevier

12-1 13-0

As described in the previous section, when one Al atom is substituted with a Si atom in the Al13 – superatoms, the Si atom is located at a central atom in the icosahedron, resulting in: (1) relaxation of structural distortions intrinsically relevant to the icosahedral structure and (2) neutralization of the negative Al13 – ion due to one-electron addition with a tetravalent Si atom, forming Si@Al12 BCS. Similar structural relaxation is expected for boron (B), because B is a smaller atom than Al, with both belonging to group 13. Figure 7.12 shows a mass spectrum of Al NC anions mixed with B atoms [34]. As seen in the distributions for the number of boron atoms (Fig. 7.13; m = 0, 1, 2,

200

300

400

Mass number (m/z)

700

210

A. Nakajima 150

(a) 13-0

(c)

Aln

125

AlnB2

11-2

40

21-2

Relative intensities (Arb. units)

100 75

23-0

50

20

25 0

0

5

10

15

20

(b) 12-1

100

25

30

AlnB

75

22-1

50

5

10

40

(d)

30

10-3

15

20

25

30

AlnB3 20-3

20 10

25 0

0

5

10

15

20

25

Number of Al atoms

30

5

10

15

20

25

30

Number of Al atoms

Fig. 7.13 Intensity distributions of the Aln B–m anions a m = 0, b m = 1, c m = 2, and d m = 3). Magic number behavior appears at n + m = 13 and 23, while it disappears at n − m = 11–2 and 10–3. Reprinted from Ref. [34] with permission from Elsevier

and 3), a distribution maximum is observed at n − m = 13–0 and similarly at 12–1, where an Al atom is substituted with a B atom. However, when two B atoms are substituted, no local maximum is observed at 11–2, showing that the second B atom destabilizes 12–1. The results suggest that at 12–1 the B atom is encapsulated in an Al12 cage, and that the second B atom destabilizes the icosahedral cage from Al12 to Al11 B due to a vertex-replaced icosahedron (exohedral B@Al11 B) with a smaller surface B atom. In the distributions shown in Fig. 7.13a, larger Al NC anions also exhibit magic number behavior, particularly at n = 23, although this is less prominent compared to n = 13. Al23 – electronically satisfies the 3S shell closure by 70 electrons [34]. The behavior of the Al–B distributions in Fig. 7.13b–d suggests that the local maximum at n + m = 23 is retained with substitution of up to two B atoms but the third B atom (m = 3) makes the Al23 derivatives less stable. This suggests that the two B atoms are encapsulated in the Al cage, while the third B atom becomes a surface atom. The structure of Al21 B2 is discussed below together with that of Al21 Si2 .

7.4.2 Anion Photoelectron Spectroscopy for Al-based BCSs To quantitatively evaluate the electronic properties, anion PES was used for these two binary NCs. Figure 7.14 shows photoelectron spectra of Al12 Si– , Al12 SiF– , Al12 B– , and Al13 Cs– [37, 38, 75]. In the Al12 Si– spectrum, a small peak (label X) is observed

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

A (a)

Al12Si

(c)

Al12B

B 1.53

Electron intensity

Fig. 7.14 Photoelectron spectra of a Al12 Si– , b Al12 SiF– , c Al12 B– , and d Al13 Cs– at 213 nm (5.82 eV). Reprinted with permission from Ref. [38]. Copyright 2006 American Chemical Society

211

x (b)

Al12SiF

A

Al13Cs

(d)

1.50

x 0

1

2

3

4

5

0

1

2

3

4

5

Electron binding energy / eV

at 1.5 eV following a large peak at around 3.5 eV, and the EA is as small as about 1.5 eV. The spectral feature is attributed to one-electron addition to the 40 electron closed shell of neutral Al12 Si, resulting in the excess electron occupying the orbital above the HOMO of Al12 Si. In fact, the F-atom adduct of Al12 SiF– anion can be selectively produced when Al–Si NC anions are reacted with fluorine (F2 ) gas [38, 39], which is seemingly promoted by the excess electron in Al12 Si– . When the photoelectron spectrum of the Al12 SiF– product is obtained, the peak X observed in Fig. 7.14a disappears while maintaining the other features, as shown in Fig. 7.14b. Furthermore, the spectrum of Al12 SiF– is almost the same as those of Al13 – and Al12 B– (Fig. 7.14c). Namely, by adding the F atom [57–59], the excess 41st electron is scavenged into a deeper level by the F atom, forming a combination of Al12 Si neutral and F– , as discussed for Ti@Si16 – and Ti@Si16 F– in Sec. 7.3.1. The results show that the peak X of Al12 Si– is attributable to a SOMO, with a gap of 1.53 eV between the HOMO and the LUMO for Al12 Si from the two peak intervals. Furthermore, the NC anions of both Al13 – and Al12 B– satisfy 40 electron shell closure, and they become 41st electron systems when a cesium alkali metal atom (Cs) is added [35, 36, 38, 78]. Interestingly, as shown in Fig. 7.14d, the peak X appears in a low binding energy region, and the photoelectron spectrum is almost the same as that of Al12 Si– . Specifically, the Al13 Cs neutral is a 40 electron species, forming a superatomic salt of (Al13 – )(Cs+ ) between the halogen-like Al13 superatom and the alkali metal atom [35, 38].

212

A. Nakajima

7.4.3 Photoionization Spectroscopy for Al-based BCSs The E i of the neutral NCs can be evaluated by measuring the photoionization efficiency curve with a tunable photoionization laser [38, 39, 49]. Figure 7.15 shows the size-dependent E i values of Aln , Aln Si, and Aln Si2 [39], and these asymptotically approach 4.08 eV of the Al metal work function [76] as the size increases. Along with the asymptotic E i decrease, a discontinuous change in E i is found when crossing the closed shell of a specific electron shell, the 40 electron boundary. The E i values of the Al13 and Al12 Si neutrals reach 6.42 eV or more. When one Al atom is added to these to form Al14 and Al13 Si, their E i values drop drastically to around 5.6 eV. A similar E i drop is observed when one Al atom in Al12 Si is substituted with a Si atom to form Al11 Si2 . These discontinuous drops in E i values can be explained in terms of electron occupation above the 40 electron shell closure. A local maximum/minimum in E i attributable to a similar electron shell is also observed around 70 electrons (3S shell), the local E i maximum for Al22 Si and the local E i minimum for Al21 Si2 , which are treated as a closed shell of 70 electrons and a one-electron excess of 71 electrons, respectively.

6.6 6.4 6.2 6.0 5.8 5.6 5.4

Ionization Energy / eV

Fig. 7.15 Ionization energies of neutral Aln Sim NCs in eV; a m = 0, b m = 1, and c m = 2. The open squares around 6.4 eV show that E i value is between 6.42 and 7.90 eV, where they can be ionized not by ArF laser (193 nm; 6.42 eV) but by F2 laser (157 nm; 7.90 eV). Calculated ionization energies for some NCs are shown along with their values, where open and solid circles show adiabatic and vertical ionization energies, respectively. The compositions of local maxima and minima are shown with the total valence electrons in parentheses. Reproduced from Ref. [39] with permission from the PCCP Owner Societies

6.6 6.4 6.2 6.0 5.8 5.6 5.4 6.6 6.4 6.2 6.0 5.8 5.6 5.4

(a) Aln

13(39)

24 (72) 14 (42)

10-1(34) 6.75

6.55 12-1

6.56

(b) AlnSi1

(40)

6.28

18-1 22-1 (58) (70)

13-1 17-1 (43) (55)

23-1 (73) 27-1 (85)

8-2 (32)

(c) AlnSi2 14-2 (50)

9-2 (35)

5.68 5.34

11-2 15-2 (41) (53)

10

20-2 (68)

28-2 (92)

21-2 (71)

20

30

Number of Al atoms (n)

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

213

For another Al-based binary NC doped with phosphorus (P) atoms, the substitution of an Al atom with P adds two valence electrons because of the pentavalent P atom [38]. The E i of P@Al12 is as low as 5.37 eV, showing a one-electron excess against the 40 electron shell closure, and the E i is the lowest value compared with those of similarly sized NCs. This feature is consistent with the total valence electron number of P@Al12 of 41, causing P@Al12 to act as an alkali metal-like BCS.

7.4.4 Theoretical Calculations for Al-Based BCSs These quantitative experimental evaluations allow us to clarify the electronic and geometric properties of Al12 X and Al21 X2 NCs in detail using quantum chemistry calculations [39, 40, 42, 43]. Figure 7.16 shows optimized structures of cations, neutrals, and anions of Al12 Si, where two isomeric forms are shown; endohedral Si in an icosahedral Al12 cage and a vertex-replaced icosahedron (exohedral Si atom) [39]. In any charge state, Si@Al12 with a central Si atom is calculated to be more stable than the vertex-replaced one. When one electron is depleted or in excess to form Al12 Si+ or Al12 Si– , the encapsulated Al12 cage structure is unchanged, but the structural symmetry is lowered from Ih symmetry. The lowering of symmetry is caused by the Jahn–Teller effect, in which the NCs are electronically stabilized by structural distortion associated with the undegeneration of electronic states [76]. Retaining the cage structure against charge exchange among Al12 Si+/0/– implies that the Si@Al12 superatoms are stable against charge transfer, indicating that these are promising BCSs for fabricating BSC assemblies for an electronic device. Additionally, molecular orbital (MO) diagrams are obtained from the optimized Al13 – , Al12 Si– , and Al12 SiF– , as shown in Fig. 7.17 [38, 39]. In Al13 – and Al12 SiF– , the HOMO–LUMO gap is large, which is consistent with the experimental results shown in Fig. 7.14. Furthermore, in Al12 Si– , the SOMO level appears below the LUMO level, represented by a small bump in the photoelectron spectrum in Fig. 7.14. X@Al12 (X = B, Si, and P) BCSs behave as halogen-like, rare gas-like, and alkali metal-like superatoms, respectively [38, 39, 79–81], where smaller atoms, from trivalent to pentavalent atoms of main group elements, are preferred as the central atom. To verify whether electron shell closure occurs by doping transition metals, the electronic properties were examined for Al12 M doped with trivalent Sc and tetravalent Ti [40]. It was found that neither Sc nor Ti atoms were encapsulated in the Al12 cage due to their large atomic radius; instead, they formed a vertex-replaced structure, exohedral Al12 M. In addition, their electronic states are described not by 40 electron shell closure, but according to the Wade-Mingos rule, in which Al@Al11 is bonded to a transition metal atom, as shown in Fig. 7.18.

214 Fig. 7.16 Calculated equilibrium structures and energy differences between isomers with central Si and surface Si; a Al12 Si1 + , b Al12 Si1 , and c Al12 Si1 – . Al and Si atoms are shown as red and blue, respectively. Reproduced from Ref. [39] with permission from the PCCP Owner Societies

A. Nakajima

(a) Al12Si1+

E= 0.00 eV Cs, 1A’

E= 0.62 eV C1, 1A

(b) Al12Si1

E= 0.00 eV Ih, 1Ag

E= 0.62 eV C1 , 1A

(c) Al12Si1−

E= 0.00 eV C2h, 2Ag

E= 0.28 eV C5v, 2A1

7.4.5 Size Evolution of Al-Based Nanoclusters for Assembled Materials In the size dependence of E i shown in Fig. 7.15, features based on electron shell closure appear not only in the vicinity of Al13 but also around Al23 ; for Al21 Si2 , the local E i minimum is due to 3S shell closure (70 e). It appears reasonable that both B and Si atoms tend to be encapsulated in Al NCs as demonstrated for B@Al12 and Si@Al12 . Based on the icosahedral Al12 cage, a face-sharing bi-icosahedral structure is conceivable for Al21 B2 and Al21 Si2 [39, 43], and then the lowest-energy isomer can be calculated.

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

Al13−(Al12Si1) (Ih)

Al12Si1−(C2h)

215

Al12Si1F−(C5v)

Kohn-Sham Orbital Energy (eV)

0.00 -1.00 -2.00

SOMO 1.27 eV

-3.00 -4.00 -5.00

2p (F)

-6.00 -7.00 -8.00 -9.00 10.00

Unoccupied MO Occupied MO

Fig. 7.17 Molecular orbital diagrams of Al13 – (Si@Al12 ), Si@Al12 – , and Si@Al12 F– . Both Al13 – (Si@Al12 ) and Si@Al12 F– have large HOMO–LUMO gap, whereas Si@Al12 – has SOMO. Reproduced from Ref. [39] with permission from the PCCP Owner Societies

Fig. 7.18 Equilibrium structure of Al12 Sc– NCs calculated at the PBE0/def-SV(P) level (Al; red and Sc; blue). Reproduced from Ref. [40] with permission from IOP Publishing

Singlet, C5v For Al21 B2 , optimization from the initial structure of the face-sharing biicosahedral structure affords a triangular rice-ball structure containing two B atoms [43]. For Al21 Si2 , however, it was calculated that a face-sharing bi-icosahedron is more stable than a triangular structure for cations and neutrals [39], as shown in Fig. 7.19. The structure shares an Al trimer on the cage surface, with a partial overlap of the two icosahedral Si@Al12 . More interestingly, it was found that the superatomic orbital (SAO) of Al21 Si2 can be represented by superposing two Si@Al12 superatoms. Figure 7.20 shows the energy diagrams of the Al21 Si2 bi-icosahedron, which shares an Al trimer between two Si@Al12 . Using a linear combination of SAOs (LCSAO) [39], a MO picture for dimeric superatoms (di-SAs) is obtained similar to linear combinations of atomic orbitals (LCAOs); the 2Pσ* LCSAO-MOs for Al21 Si2 + cations and Al21 Si2 neutrals are vacant and SOMO, respectively, while up to the 1Fδ* LCSAO-MO the MOs are occupied. For di-SA, the orbital shapes of the SAOs allow wavefunction overlap at closer distances between the two superatoms compared to that between the atomic

216

A. Nakajima

(a) Bi-icosahedron

(b) Triangular

cation ΔE = 0.0 eV

ΔE = +0.02 eV

neutral

ΔE = 0.0 eV

ΔE = +0.17 eV

Fig. 7.19 Calculated equilibrium structures and energy differences between two isomers of a biicosahedron and b triangular form for cationic and neutral Al21 Si2 with PBE1PBE/6-311+G*. The experimental ionization energy is reproduced by the bi-icosahedrons. Reproduced from Ref. [39] with permission from the PCCP Owner Societies

orbitals of homonuclear diatomics. This feature gives an orbital stability of di-SA in the order of σ < π < δ < φ, because the nature of the LCSAO-MOs is closer to the united (super)atom limit than that of the homonuclear diatomics (LCAO-MOs). Furthermore, the electronic properties of the superatom dimers of X@Al12 –Y@Al12 (X–Y = Si–Si, B–P, Al–P), hetero-assemblies of endohedral Al–based BCSs, are theoretically predicted, where the optimized dimers are obtained by facing the sides of the monomers in a staggered fashion [42]. When the electronic absorption spectra are calculated for the B@Al12 –P@Al12 and Al13 –P@Al12 heterodimers (a combination between halogen-like and alkali metal-like superatoms), a CT band from B/Al@Al12 to P@Al12 is found in the visible region. The charge distributions in the heterodimer of B@Al12 –P@Al12 are unchanged by inserting Si@Al12 between the two superatoms, and the dipole moment of the heterotrimer (3.89 D) is larger than that of the heterodimer (2.38 D). Similarly, a heterodimer and trimer comprising Si-based BCSs, M@Si16 (M = Sc, Ti, and V), are predicted to exhibit electronic excitation involving CT states; this is characterized as electron transfer from V@Si16 to Sc@Si16 in the heterodimer of V@Si16 –Sc@Si16 with a dipole moment of 7.63 D [41]. When the Ti@Si16 BCS is inserted between the V@Si16 –Sc@Si16 dimer, the linear heterotrimer of Sc@Si16 –Ti@Si16 –V@Si16 has a larger dipole moment of 15.6 D and one or more localized frontier orbitals compared to the dimer. These dimers and trimers are the smallest assembled BCSs, and the theoretical insight allows combination of

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

217

1F σ∗ 1F σ 1F π∗ 2P σ∗ (SOMO) -6

1F 2P

2P π∗ 2P σ 2P π

Energy (eV)

-8

-10

1D σ∗

1D

2S σ∗

1D π∗ 1D σ 1D δ∗

2S

2S σ

1D π 1D δ

1F δ* 1F π 1F φ* 1F φ2 1F δ 1F φ1

1P σ∗

-12 1P -14

-16 1S

(a) 2 × Si@Al12 (Ih)

1P π∗ 1P σ 1P π

1S σ∗

-4

-6

-8

-10

-12

-14

-16

1S σ

(b) Bi-icosahedral Al21Si2 (di-SAs)

Fig. 7.20 Calculated energy diagram of a icosahedral Si@Al12 and b bi-icosahedral Al21 Si2 neutral using the linear combination of SAOs (LCSAO) of the dimeric Si@Al12 superatom. With face sharing of Al3 , nine electrons are subtracted in the Al21 Si2 neutral compared to two Si@Al12 . Solid lines show filled or half-filled states, whereas dotted lines show unoccupied states. The LCSAO-MO of 2P σ* is a SOMO for the Al21 Si2 neutral, while the LCSAO-MO of 1F δ* is a HOMO for the Al21 Si2 cation. Reproduced from Ref. [39] with permission from the PCCP Owner Societies

different BCSs having various central atoms to create new nanoscale materials [82, 83], which will establish a new area in the field of NC science.

7.5 Conclusions The metal-atom-encapsulating Si16 cage (METS) is a novel nanostructure that was synthesized in the gas phase, and the structural properties were successfully characterized in 2017 after a long research period of over 30 years [12]. The BCS formation features synergistic generation from Si and M atomic vapor. In the highly symmetrical Td structure, in which 16 Si atoms form a spherical outer shell, the total number of valence electrons is controlled by replacing the central metal atom. M@Si16 is regarded as a representative “superatoms” in which 17 atoms in total behave as a new “atom”. For example, Ta@Si16 is a superatom with one excess electron, which is like an alkali-metal atom such as Li and Na. By replacing the central metal atom with various metal atoms, the electronic properties are designed while retaining the structural motif, and then their aggregates and hetero-interface exhibit distinct chemical and

218

A. Nakajima

physical properties. Furthermore, chemical ligation would modify their properties to form designer nanomaterials. The BCS nanomaterials will show novel optical and conduction properties, and M@Si16 BCS itself exhibits new bonding features. The BCS might enable a paradigm shift in Si-based nanoscale material to overcome the integration limit of current top-down Si electronics together with M@Ge16 BCSs [61, 84]. For Al-based BCS, X@Al12 (X = B, Si, and P) BCSs behave as halogen-like, rare gas-like, and alkali metal-like superatoms, where small atoms are preferred as central atoms from trivalent to pentavalent atoms of main group elements. Since BCSs retain the cage structure by charge exchange from the Si@Al12 Si@Al+/0/– 12 neutral in Ih symmetry, the Si@Al12 superatoms are stable against charge transfer, affording promising BCS assemblies for an electronic device. The theoretical analysis showed that the SAO of bi-icosahedral Al21 Si2 can be represented by superposing two Si@Al12 BCSs. Using the LCSAO, a MO picture for di-SAs is obtained similar to the LCAO. Finally, the BCS assembly can apparently create new nanoscale materials; heteroassembly of an M@Si16 or X@Al12 BCS comprising different central atom BCSs is theoretically predicted to show an electronic transition relevant to charge transfer. Beyond M@Si16 and X@Al12 BCSs themselves, these BCS assemblies will further widen a rich diversity to fabricate nanoscale functional materials, and along with that superatom periodic table would evolve from these BCS family members. Acknowledgments This work is partly supported by the program of Exploratory Research for Advanced Technology (ERATO) in Japan Science and Technology Agency (JST) entitled with “Nakajima Designer Nanocluster Assembly Project”, by JSPS KAKENHI of Grant-in-Aids for Scientific Research (A) no. 15H02002, and by JSPS KAKENHI of Challenging Research (Pioneering) no. 17H06226. This research is in collaboration with co-authors of Refs. [9–13], including Dr. Hironori Tsunoyama, Dr. Masahiro Shibuta, Dr. Masato Nakaya, Dr. Toyoaki Eguchi, Dr. Takeshi Iwasa, Dr. Kiichirou Koyasu, Professor Norihiro Tokitoh (Kyoto Univ.), and Associate Professor Yoshiyuki Mizuhata (Kyoto Univ.).

References 1. Moore, G.E.: Cramming more components onto integrated circuits. Electronics 114–117 (1965) 2. Canham, L.T.: Silicon quantum wire array fabrication by electrochemical and chemical dissolution of wafers. Appl. Phys. Lett. 57, 1046–1048 (1990) 3. Walters, R.J., Bourianoff, G.I., Atwater, H.A.: Field-effect electroluminescence in silicon nanocrystals. Nat. Mater. 4, 143–146 (2005) 4. Hochbaum, A.I., Chen, R., Delgado, R.D., Liang, W., Garnett, E.C., Najarian, M., Majumdar, A., Yang, P.: Enhanced thermoelectric performance of rough silicon nanowires. Nature 451, 163–167 (2008) 5. Boukai, A.I., Bunimovich, Y., Tahir-Kheli, J., Yu, J.-K., Goddard III, W.A., Heath, J.R.: Silicon nanowires as efficient thermoelectric materials. Nature 451, 168–171 (2008) 6. Kroto, H.W., Heath, J.R., O’Brien, S.C., Curl, R.F., Smalley, R.E.: C60 : Buckminsterfullerene. Nature 318, 162–163 (1985)

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

219

7. Beck, S.M.: Studies of silicon cluster-metal atom compound formation in a supersonic molecular beam. J. Chem. Phys. 87, 4233–4234 (1987) 8. Beck, S.M.: Mixed metal-silicon clusters formed by chemical reaction in a supersonic molecular beam: implications for reactions at the metal/silicon interface. J. Chem. Phys. 90, 6306–6312 (1989) 9. Koyasu, K., Akutsu, M., Mitsui, M., Nakajima, A.: Selective formation of MSi16 (M = Sc, Ti, and V). J. Am. Chem. Soc. 127, 4998–4999 (2005) 10. Nakaya, M., Iwasa, T., Tsunoyama, H., Eguchi, T., Nakajima, A.: Formation of a superatom monolayer using gas-phase-synthesized Ta@Si16 nanocluster ions. Nanoscale 6, 14702–14707 (2014) 11. Shibuta, M., Ohta, T., Nakaya, M., Tsunoyama, H., Eguchi, T., Nakajima, A.: Chemical characterization of an alkali-like superatom consisting of a Ta-encapsulating Si16 Cage. J. Am. Chem. Soc. 137, 14015–14018 (2015) 12. Tsunoyama, H., Akatsuka, H., Shibuta, M., Iwasa, T., Mizuhata, Y., Tokitoh, N., Nakajima, A.: Development of integrated dry-wet synthesis method for metal encapsulating silicon cage superatoms of M@Si16 (M = Ti and Ta). J. Phys. Chem. C 121, 20507–20516 (2017) 13. Tsunoyama, H., Shibuta, M., Nakaya, M., Eguchi, T., Nakajima, A.: Synthesis and characterization of metal-encapsulating Si16 cage superatoms. Acc. Chem. Res. 51, 1735–1745 (2018) 14. Heath, J.R., O’Brien, S.C., Zhang, Q., Liu, Y., Curl, R.F., Kroto, H.W., Tittel, F.K., Smalley, R.E.: Lanthanum complexes of spheroidal carbon shells. J. Am. Chem. Soc. 107, 7779–7780 (1985) 15. Sanekata, M., Koya, T., Nagano, S., Negishi, Y., Nakajima, A., Kaya, K.: Electronic and geometric structures of metal-silicide clusters. Trans. Mater. Res. Soc. Jpn 25, 1003–1006 (2000) 16. Kumar, V., Kawazoe, Y.: Metal-encapsulated fullerenelike and cubic caged clusters of silicon. Phys. Rev. Lett. 87, 045503 (2001) 17. Hiura, H., Miyazaki, T., Kanayama, T.: Formation of metal-encapsulating Si cage clusters. Phys. Rev. Lett. 86, 1733–1736 (2001) 18. Lu, J., Nagase, S.: Structural and electronic properties of metal-encapsulated silicon clusters in a large size range. Phys. Rev. Lett. 90, 115506 (2003) 19. Ohara, M., Koyasu, K., Nakajima, A., Kaya, K.: Geometric and electronic structures of metal (M)-doped silicon clusters (M = Ti, Hf, Mo and W). Chem. Phys. Lett. 371, 490–497 (2003) 20. Zheng, W., Nilles, J.M., Radisic, D., Bowen, K.H.: Photoelectron spectroscopy of chromiumdoped silicon cluster anions. J. Chem. Phys. 122, 071101 (2005) 21. Jaeger, J.B., Jaeger, T.D., Duncan, M.A.: Photodissociation of metal-silicon clusters: encapsulated versus surface-bound metal. J. Phys. Chem. A 110, 9310–9314 (2006) 22. Reveles, J.U., Khanna, S.N.: Electronic counting rules for the stability of metal-silicon clusters. Phys. Rev. B 74, 035435 (2006) 23. Torres, M.B., Fernández, E.M., Balbás, L.C.: Theoretical study of isoelectronic Sin M clusters (M = Sc− , Ti, V+ ; n = 14–18). Phys. Rev. B 75, 205425 (2007) 24. Lau, J.T., Hirsch, K., Klar, P., Langenberg, A., Lofink, F., Richter, R., Rittmann, J., Vogel, M., Zamudio-Bayer, V., Möller, T., von Issendorff, B.: X-ray spectroscopy reveals high symmetry and electronic shell structure of transition-metal-doped silicon clusters. Phys. Rev. A 79, 053201 (2009) 25. Claes, P., Janssens, E., Ngan, V.T., Gruene, P., Lyon, J.T., Harding, D.J., Fielicke, A., Nguyen, M.T., Lievens, P.: Structural identification of caged vanadium doped silicon clusters. Phys. Rev. Lett. 107, 173401 (2011) 26. Knight, W.D., de Heer, W.A., Clemenger, K., Saunders, W.A., Chou, M.Y., Cohen, M.L.: Electronic shell structure and abundances of sodium clusters. Phys. Rev. Lett. 52, 2141–2143 (1984) 27. de Heer, W.A.: The physics of simple metal clusters: experimental aspects and simple models. Rev. Mod. Phys. 65, 611–676 (1993) 28. Brack, M.: The physics of simple metal clusters: self-consistent jellium model and semiclassical approaches. Rev. Mod. Phys. 65, 677–732 (1993)

220

A. Nakajima

29. Jena, P.: Beyond the periodic table of elements: the role of superatoms. J. Phys. Chem. Lett. 4, 1432–1442 (2013) 30. Luo, Z., Castleman Jr., A.W., Khanna, S.N.: Reactivity of metal clusters. Chem. Rev. 116, 14456–14492 (2016) 31. Tomalia, D.A., Khanna, S.N.: A systematic framework and nanoperiodic concept for unifying nanoscience: hard/soft nanoelements, superatoms, meta-atoms, new emerging properties, periodic property patterns, and predictive mendeleev-like nanoperiodic tables. Chem. Rev. 116, 2705–2774 (2016) 32. Reber, A.C., Khanna, S.N.: Superatoms: electronic and geometric effects on reactivity. Acc. Chem. Res. 50, 255–263 (2017) 33. Jena, P., Sun, Q.: Super atomic clusters: design rules and potential for building blocks of materials. Chem. Rev. 118, 5755–5870 (2018) 34. Nakajima, A., Kishi, T., Sugioka, T., Kaya, K.: Electronic and geometric structures of aluminum-boron negative cluster ions (Aln B–m ). Chem. Phys. Lett. 187, 239–244 (1991) 35. Nakajima, A., Hoshino, K., Naganuma, T., Sone, Y., Kaya, K.: Ionization potentials of aluminum-sodium bimetallic clusters (Aln Nam ). J. Chem. Phys. 95, 7061–7066 (1991) 36. Hoshino, K., Watanabe, K., Konishi, Y., Taguwa, T., Nakajima, A., Kaya, K.: Ionization potentials of aluminum-cesium bimetallic clusters (Aln Csm ). Chem. Phys. Lett. 231, 499–503 (1994) 37. Kawamata, H., Negishi, Y., Nakajima, A., Kaya, K.: Electronic properties of substituted aluminum clusters by boron and carbon atoms (Aln B–m /Aln C–m ); new insights into s-p hybridization and perturbed shell structures. Chem. Phys. Lett. 337, 255–262 (2001) 38. Akutsu, M., Koyasu, K., Atobe, J., Hosoya, N., Miyajima, K., Mitsui, M., Nakajima, A.: Experimental and theoretical characterization of aluminum-based binary superatoms of Al12 X and their cluster salts. J. Phys. Chem. A 110, 12073–12076 (2006) 39. Akutsu, M., Koyasu, K., Atobe, J., Miyajima, K., Mitsui, M., Tsunoyama, H., Nakajima, A.: Geometric and electronic properties of Si-atom doped Al clusters: robustness of binary superatoms against charging. Phys. Chem. Chem. Phys. 19, 20401–20411 (2017) 40. Tsunoyama, H., Akutsu, M., Koyasu, K., Nakajima, A.: Stability for binary Al12 X nanoclusters (X = Sc and Ti): superatom or Wade’s polyhedron. J. Phys. Condensed Matter 30, 494004 (2018) 41. Iwasa, T., Nakajima, A.: Geometric, electronic, and optical properties of a superatomic heterodimer and trimer: Sc@Si16 –V@Si16 and Sc@Si16 –Ti@Si16 –V@Si16 . J. Phys. Chem. C 116, 14071–14077 (2012) 42. Iwasa, T., Nakajima, A.: Geometric, electronic, and optical properties of monomer and assembly of endohedral aluminum superatomic clusters. J. Phys. Chem. C 117, 21551–21557 (2013) 43. Iwasa, T., Nakajima, A.: Geometric, electronic, and optical properties of a boron-doped aluminum cluster of B2 Al21 − . Chem. Phys. Lett. 582, 100–104 (2013) 44. Koyasu, K., Atobe, J., Akutsu, M., Mitsui, M., Nakajima, A.: Electronic and geometric stabilities of clusters with transition metal encapsulated by silicon. J. Phys. Chem. A 111, 42–49 (2007) 45. Nakajima, A.: Study on electronic properties of composite clusters toward nanoscale functional advanced materials. Bull. Chem. Soc. Jpn 86, 414–437 (2013) 46. Even, U., Jortner, J., Noy, D., Lavie, N., Cossart-Magos, C.: Cooling of large molecules below 1 K and He clusters formation. J. Chem. Phys. 112, 8068–8071 (2000) 47. Geohegan, D.B.: Imaging and blackbody emission spectra of particulates generated in the KrF-laser ablation of BN and YBa2 Cu3 O7−x . Appl. Phys. Lett. 62, 1463–1465 (1993) 48. Geusic, M.E., Morse, M.D., O’Brien, S.C., Smalley, R.E.: Surface reactions of metal clusters I: the fast flow cluster reactor. Rev. Sci. Instrum. 56, 2123–2130 (1985) 49. Miyajima, K., Muraoka, K., Hashimoto, M., Yasuike, T., Yabushita, S., Nakajima, A., Kaya, K.: The quasi-band electronic structure of Vn (Benzene)n+1 clusters. J. Phys. Chem. A 106, 10777–10781 (2002) 50. Cheshnovsky, O., Yang, S.H., Pettiette, C.L., Craycraft, M.J., Smalley, R.E.: Magnetic timeof-flight photoelectron spectrometer for mass-selected negative cluster ions. Rev. Sci. Instrum. 58, 2131–2137 (1987)

7 Superatomic Nanoclusters Comprising Silicon or Aluminum Cages

221

51. Ganteför, G., Meiwes-Broer, K.H., Lutz, H.O.: Photodetachment spectroscopy of cold aluminum cluster anions. Phys. Rev. A 37, 2716–2718 (1988) 52. Ganteför, G., Gausa, M., Meiwes-Broer, K.H., Lutz, H.O.: Ultraviolet photodetachment spectroscopy on jet-cooled metal-cluster anions. Faraday Discuss. Chem. Soc. 86, 197–208 (1988) 53. Nakajima, A., Taguwa, T., Hoshino, K., Sugioka, T., Naganuma, T., Ono, F., Watanabe, K., Nakao, K., Konishi, Y., Kishi, R., Kaya, K.: Photoelectron spectroscopy of (C6 F6 )–n and (AuC6 F6 )– clusters. Chem. Phys. Lett. 214, 22–26 (1993) 54. Ho, J., Ervin, K.M., Lineberger, W.C.: Photoelectron spectroscopy of metal cluster anions: Cu–n , Ag–n , and Au–n . J. Chem. Phys. 93, 6987–7002 (1990) 55. Moore, C.E.: Atomic energy levels as derived from the analyses of optical spectra. NSRDSNBS, 35, vol. 3. U.S. National Bureau of Standards: Washington, DC (1971) 56. Koyasu, K., Atobe, J., Furuse, S., Nakajima, A.: Anion photoelectron spectroscopy of transition metal- and lanthanide metal-silicon clusters: MSi− n (n = 6–20). J. Chem. Phys. 129, 214301 (2008) 57. Kawamata, H., Negishi, Y., Kishi, R., Iwata, S., Nakajima, A., Kaya, K.: Photoelectron spectroscopy of silicon-fluorine binary cluster anions (SinFm− ). J. Chem. Phys. 105, 5369–5376 (1996) 58. Negishi, Y., Kawamata, H., Hayase, T., Gomei, M., Kishi, R., Hayakawa, F., Nakajima, A., Kaya, K.: Photoelectron spectroscopy of germanium-fluorine binary cluster anions: the HOMO-LUMO gap estimation of gen clusters. Chem. Phys. Lett. 269, 199–207 (1997) 59. Kishi, R., Negishi, Y., Kawamata, H., Iwata, S., Nakajima, A., Kaya, K.: Geometric and electronic structures of fluorine bound silicon clusters. J. Chem. Phys. 108, 8039–8058 (1998) 60. Wang, X.B., Ding, C.F., Wang, L.S.: High resolution photoelectron spectroscopy of C60 − . J. Chem. Phys. 110, 8217–8220 (1999) 61. Furuse, S., Koyasu, K., Atobe, J., Nakajima, A.: Experimental and theoretical characterization of MSi16 – , MGe–16 , MSn–16 , and MPb16 – (M = Ti, Zr, and Hf): the role of cage aromaticity. J. Chem. Phys. 129, 064311 (2008) 62. Bromann, K., Felix, C., Brune, H., Harbich, W., Monot, R., Buttet, J., Kern, K.: Controlled deposition of size-selected silver nanoclusters. Science 274, 956–958 (1996) 63. Haberland, H., Karrais, M., Mall, M.: A new type of cluster and cluster ion source. Z. Phys. D 20, 413–415 (1991) 64. Haberland, H., Karrais, M., Mall, M., Thurner, Y.: Thin films from energetic cluster impact: a feasibility study. J. Vac. Sci. Technol., A 10, 3266–3271 (1992) 65. Tsunoyama, H., Zhang, C., Akatsuka, H., Sekiya, H., Nagase, T., Nakajima, A.: Development of high-flux ion source for size-selected nanocluster ions based on high-power impulse magnetron sputtering. Chem. Lett. 42, 857–859 (2013) 66. Zhang, C., Tsunoyama, H., Akatsuka, H., Sekiya, H., Nagase, T., Nakajima, A.: Advanced nanocluster ion source based on high-power impulse magnetron sputtering and time-resolved measurements of nanocluster formation. J. Phys. Chem. A 117, 10211–10217 (2013) 67. Zhang, C., Tsunoyama, H., Feng, Y., Nakajima, A.: Extended Smoluchowski model for the formation of size-selected silver nanoclusters generated via modulated pulsed power magnetron sputtering. J. Phys. Chem. C 120, 5667–5672 (2016) 68. Nakaya, M., Iwasa, T., Tsunoyama, H., Eguchi, T., Nakajima, A.: Heterodimerization via the covalent bonding of Ta@Si16 nanoclusters and C60 molecules. J. Phys. Chem. C 119, 10962–10968 (2015) 69. Ohta, T., Shibuta, M., Tsunoyama, H., Eguchi, T., Nakajima, A.: Charge transfer complexation of Ta-encapsulating Ta@Si16 superatom with C60 . J. Phys. Chem. C 120, 15265–15271 (2016) 70. Shibuta, M., Kamoshida, T., Ohta, T., Tsunoyama, H., Nakajima, A.: Alkali-like superatom chemistry of group-5 metal encapsulating Si16 cage; M@Si16 (M = V, Nb, Ta). Comm. Chem. 1, 50 (2018) 71. Gomoyunova, M.V., Pronin, I.I., Malygin, D.E., Gall, N.R., Vyalikh, D.V., Molodtsov, S.L.: Photoemission study of cobalt interaction with the oxidized Si(100) 2 × 1 surface. Surf. Sci. 600, 2449–2456 (2006)

222

A. Nakajima

72. van der Veen, J.F., Himpsel, F.J., Eastman, D.E.: Chemisorption-induced 4f-core-electron binding-energy shifts for surface atoms of W(111), W(100), and Ta(111). Phys. Rev. B 25, 7388–7397 (1982) 73. Even, U., Dick, B.: Computer optimization for high-resolution time-of-flight mass spectrometer. Rev. Sci. Instrum. 71, 4415–4420 (2000) 74. Li, X., Wu, H., Wang, X.B., Wang, L.S.: s-p hybridization and electron shell structures in aluminum clusters: a photoelectron spectroscopy study. Phys. Rev. Lett. 81, 1909–1912 (1998) 75. Koyasu, K., Akutsu, M., Atobe, J., Mitsui, M., Nakajima, A.: Electronic properties of Cs-atom doped aluminum and silicon clusters: Aln Csm and Sin Csm . Chem. Phys. Lett. 421, 534–539 (2006) 76. Shriver, D. F.; Atkins, P. W., Inorganic Chemistry. 1999, Oxford Univ. Press, 3rd edn 77. Mackay, A.L.: A dense non-crystallographic packing of equal spheres. Acta Crystallogr. 15, 916–918 (1962) 78. Kishi, R., Iwata, S., Nakajima, A., Kaya, K.: Geometric and electronic structures of silicon— sodium binary clusters. I. Ionization energy of Sin Nam . J. Chem. Phys. 107, 3056–3070 (1997) 79. Gong, X.G., Kumar, V.: Enhanced stability of magic clusters: a case study of icosahedral Al12 X, X = B, Al, Ga, C, Si, Ge, Ti, As. Phys. Rev. Lett. 70, 2078–2081 (1993) 80. Li, X., Wang, L.-S.: Experimental search and characterization of icosahedral clusters: Al12 X– (X = C, Ge, Sn, Pb). Phys. Rev. B 65, 153404 (2002) 81. Lu, Q.L., Jalbout, A.F., Luo, Q.Q., Wan, J.G., Wang, G.H.: Theoretical study of hydrogenated Mg, Ca@Al12 Clusters. J. Chem. Phys. 128, 224707 (2008) 82. Khanna, S.N., Jena, P.: Assembling Crystals from Clusters. Phys. Rev. Lett. 69, 1664–1667 (1992) 83. Claridge, S.A., Castleman Jr., A.W., Khanna, S.N., Murray, C.B., Sen, A., Weiss, P.S.: Clusterassembled materials. ACS Nano 3, 244–255 (2009) 84. Atobe, J., Koyasu, K., Furuse, S., Nakajima, A.: Anion photoelectron spectroscopy of germanium and tin clusters containing a transition- or lanthanide-metal atom; MGen− (n = 8–20) and MSnn− (n = 15–17) (M = Sc–V, Y–Nb, and Lu–Ta). Phys. Chem. Chem. Phys. 14, 9403–9410 (2012)

Chapter 8

Characterization of Chemically Modified Gold/Silver Superatoms in the Gas Phase Kiichirou Koyasu, Keisuke Hirata and Tatsuya Tsukuda

ESI & MALDI mass spectrometry

Ion mobility mass spectrometry Collision induced dissociation mass spectrometry

Photoelectron spectroscopy Photodissociation spectroscopy

Characterization methods applied to ligand protected Au/Ag clusters in gas phase

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_8

223

224

K. Koyasu et al.

Abstract Atomically precise Au and Ag clusters protected by organic ligands can be viewed as chemically modified superatoms. These chemically modified Au/Ag superatoms have gained interests as promising building units of functional materials as well as ideal platforms to study the size-dependent evolution of structures and physicochemical properties. Mass spectrometry not only allows us to determine the chemical compositions of the synthesized superatoms but also gives us molecularlevel insights into the mechanism of complex processes in solution. A variety of the gas-phase methods including ion-mobility–mass spectrometry, collision- or surfaceinduced dissociation mass spectrometry, photoelectron spectroscopy, and photodissociation mass spectrometry have been applied to the chemically modified Au/Ag superatom ions isolated in the gas phase. These studies have provided novel and complementary information on their intrinsic geometric and electronic structures that cannot be obtained by conventional characterization methods. This chapter surveys the recent progress in the gas-phase studies on chemically synthesized Au/Ag superatoms. Keywords Ligand-protected gold/silver clusters · Superatoms · Mass spectrometry

8.1 Introduction 8.1.1 Metal Clusters as Novel Functional Units Metal clusters composed of less than a few hundred atoms are located between the nanoparticles and atoms of the corresponding metal (Scheme 8.1) and have attracted the attention of scientists over the last four decades [1]. The central interest at the early stage of the research was observing the finite-size effects on the physical properties of metal clusters and understanding their origins from a microscopic viewpoint. To address these fundamental questions, versatile and efficient production methods for naked clusters (laser vaporization [2] and magnetron sputtering [3] methods) and ultrasensitive characterization methods coupled with mass spectrometry have been developed [4]. Experimental approaches for the study of size-specific properties of naked metal cluster ions are summarized in Fig. 8.1. Magic numbers were searched by mass spectrometry [4]. Electronic structures have been studied by photoelectron spectroscopy (PES) [5–10] and photodissociation spectroscopy (PDS) [11], whereas geometric structures (overall motif and atomic packing) have been studied by ionmobility–mass spectrometry (IM MS) [12, 13], electron diffraction (ED) [14], and K. Koyasu · K. Hirata · T. Tsukuda (B) Department of Chemistry, School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-Ku, Tokyo 113-0033, Japan e-mail: [email protected] K. Koyasu · T. Tsukuda Elements Strategy Initiative for Catalysts and Batteries (ESICB), Kyoto University, Katsura, Kyoto 615-8520, Japan

8 Characterization of Chemically Modified Gold/Silver Superatoms …

225

Scheme 8.1 Comparison between metal nanoparticles and metal clusters

Fig. 8.1 Summary of experimental approaches for the investigation of size-selected naked metal cluster ions and composition-defined protected metal cluster ions

226

K. Koyasu et al.

vibrational spectroscopy [15]. Chemical properties and binding strength were studied by collision-induced dissociation/reaction mass spectrometry (CID/CIR MS) [16, 17]. These experimental results with the help of theoretical calculations have led to discoveries of a variety of remarkable size-specific phenomena and physicochemical properties. For example, observations of magic numbers in the size distributions have led to the concept of electron shell closing based on the Jellium model [18] and superatoms [19]. It is widely recognized that various physicochemical properties of metal clusters deviate significantly from their bulk counterparts due to the unique geometric and electronic structures (Scheme 8.1) and evolve dramatically as a function of size, as exemplified by the metal–insulator transition [20]. Rapid progress in the research under the catchphrases, “small is different” [21, 22] and “every atom counts” [23, 24], has convinced the community that metal clusters are promising functional units of novel materials.

8.1.2 Naked Gold Clusters as Prototypical Superatoms Since the discovery of oxidation catalysis of nanosized gold (Au) by Haruta [25], Au clusters have been the most extensively studied systems among the naked metal clusters [4–17]. Here, the knowledge on the structures and properties of Au clusters accumulated so far is summarized as a reference to that for chemically synthesized Au clusters. The stability of Au clusters is governed by the electronic structure: Au clusters gain a special stability when superatomic electronic shells (1S, 1P, 1D, 2S, 1F, 2P, 1G, 2D,…) are closed with valence electrons totaling 2, 8, 18, 20, 34, 40, 58, and so on [4, 26]. Neutral Aun clusters with n = 20, 34, 40, 58, … correspond to magic clusters since each Au atom provides a single 6s electron as a valence electron. Electronic shell closure at these magic clusters manifests itself in the remarkably smaller electron affinity (EA) than those of the neighboring sizes (Fig. 8.2a) [4–6, 9, 26]. The size-dependent and size-specific chemical reactivity toward molecular oxygen [17], relevant to aerobic oxidation catalysis [27], is also correlated with the electronic structure (Fig. 8.2b): the molecular adsorption proceeds via electron transfer from Aun − into the adsorbed π* of O2 . Small Aun clusters exhibit diverse, non-FCC structures and unexpected evolution in geometric structures [26]. Figure 8.2c plots the collision cross section (CCSs) of Aun − as a function of size n. The decrease in the CCSs at n = 13 is explained in terms of the structural transition from planar to three-dimensional motifs [13, 28]. Au13 takes neither cuboctahedral nor icosahedral structures [7]. The magic Au20 has an FCC structure with a pyramidal motif [6], while Au34 and Au58 have Au4 @Au30 [9, 26] and Au10 @Au48 [29] core–shell structures, respectively (Fig. 8.2c). We can see that these magic Au clusters provide nearly spherical potential wells for the confinement of the valence electrons. Although the FCC structure appears at n = 20 after hollow cage structures in the size range of 16–18 [8], hollow structures appear again at n = 21 and 24 [14].

8 Characterization of Chemically Modified Gold/Silver Superatoms …

227

Fig. 8.2 Size dependence of a AEA of naked Aun and b reactivity and c CCS of naked Aun − [5, 13, 17]. Inset shows theoretically predicted structures of Au13 , Au20 , Au34 , and Au58

8.1.3 Ligand-Protected Gold Clusters as Chemically Modified Superatoms Chemical synthesis has been a challenge for developing the materials science of metal clusters, as evidenced by the explosive growth in the materials science of nanocarbons after the large-scale production of C60 by Smalley and Krätschmer [30]. Au clusters have been studied most extensively as representative systems of the chemical synthesis because of their robustness against oxidation under ambient conditions. The first requirement for the chemical synthesis is to stabilize individual Au clusters against aggregation so that they can be treated as conventional chemical compounds. To this end, the surface of the Au cluster has been passivated by ligands L, such as thiolates (RS) [31], alkynes (RC≡C) [32], phosphines (R3 P) [33], carbenes [34], and halides to yield [Aux Ly ]z . The second requirement is to define the chemical composition of [Aux Ly ]z with atomic and molecular precision. In recent decades, a large variety of [Aux Ly ]z with atomically defined sizes in the range of x < ~500 have been synthesized [35–38]. The nonbulk-like atomic packing of the Au cores has been elucidated by single-crystal X-ray diffraction (SCXRD) analysis and transmission electron microscopy [39]. The discrete nature of the electronic structures has been demonstrated by UV–Vis optical spectroscopy, photoluminescence, and voltammetry. Figure 8.3 lists single-crystal structures and the optical absorption spectra of representative small Au- or Ag-based clusters protected by ligands

228

K. Koyasu et al.

Fig. 8.3 Geometric structures determined by single-crystal X-ray diffraction, core structures, and optical spectra of [Au9 (PPh3 )8 ]3+ (1), [PdAu8 (PPh3 )8 ]2+ (2), [Au11 (PPh3 )8 Cl2 ]+ (3), [Au11 (PPh3 )7 (NHCiPr )Cl2 ]+ (4), [Au25 (SC2 H4 Ph)18 ]− (5), [Ag25 (SC6 H3 Me2 )18 ]− (6), [PtAg24 (SC6 H3 Me2 )18 ]2– (7), and [PdAg24 (SC6 H3 Me2 )18 ]2– (8). The red wireframe in 4 represents an N-heterocyclic carbene ligand

8 Characterization of Chemically Modified Gold/Silver Superatoms …

229

[(Au/Ag)x Ly ]z . The structures of the metallic cores are highlighted in the figure. [Au9 (PPh3 )8 ]3+ (1) [40] and [PdAu8 (PPh3 )8 ]2+ (2) [41] have a crown-shaped M@Au8 (M = Au, Pd) core with the unligated M atom at the center. [Au11 (PPh3 )8 Cl2 ]+ (3) [42] has a quasi-icosahedral Au11 core protected by mixed ligands of Cl and PPh3 , whereas one of the PPh3 ligands of 3 is replaced with N-heterocyclic carbene (NHCiPr ) to form [Au11 (PPh3 )7 (NHCiPr )Cl2 ]+ (4) [34]. [Au25 (SC2 H4 Ph)18 ]− (5) [43] and [Ag25 (SC6 H3 Me2 )18 ]− (6) [44] have icosahedral cores of Au13 and Ag13 , respectively, protected by six M2 (SR)3 (M = Au, Ag) bidentate units with staple motifs. The central Ag atom of the Ag13 core of 6 is replaced with a Pt and Pd atom to form [PtAg24 (SC6 H3 Me2 )18 ]2– (7) and [PdAg24 (SC6 H3 Me2 )18 ]2– (8), respectively [45]. The ligands not only sterically protect the Au/Ag-based core from aggregation but also adjust the number of valence electrons in the Au/Ag cores and the atomic packing of the Au/Ag cores. A simple electron counting scheme has been proposed for [Aux Ly ]z [46]. The formal number of valence electrons (n*) in the Au core is calculated by the following equation: n∗ = x − y × N − z

(8.1)

where x represents the total number of valence electrons supplied by Au constituent atoms and is equal to the number of the Au atoms in the cluster. The N value is the number of electrons taken from the Au core by a single ligand L, which is dependent on the nature of the Au–L interaction. A thiolate, alkynyl, or halogen ligand takes one electron (N = 1), whereas a phosphine or carbene ligand does not take any electrons (N = 0). The n* values of [Aux Ly ]z adopt the values of 8, 18, 20, 34, 40, 58, and so on as long as the Au core can be viewed as a sphere. This simple counting scheme can be applied to bimetallic [(Au/Ag)x Ly ]z clusters and the analogues doped with Pd or Pt [47, 48]: each Ag atom provides one valence electron, whereas Pd and Pt atoms do not. The n* values thus calculated are 6 for 1 and 2 and 8 for 3–8 (Fig. 8.3). Thus, the crown-shaped (M@Au8 )2+ (M = Au+ , Pd) cores in 1 and 2 can be viewed as oblate-shaped superatoms, where three 1P superatomic orbitals are split into two subgroups. Two of the 1P superatomic orbitals (1Px , 1Py ) correspond to HOMO and the 1Pz superatomic orbital to LUMO. In contrast, the quasi-icosahedral (Au11 )3+ cores of 3 and 4, icosahedral (M13 )5+ cores of 5 (M = Au) and 6 (M = Ag), and (M@Ag12 )4+ (M = Pt, Pd) cores of 7 and 8 can be viewed as nearly spherical superatoms. The 1P and 1D superatoms of 3–8 correspond to the HOMO and LUMO, respectively. Figure 8.3 also presents optical absorption spectra, showing clear absorption onsets and well-resolved structures due to the quantized electronic structures. The onsets for 3–8 are assigned to HOMO (1P)–LUMO (1D) transition, whereas those for 1 and 2 are not due to the HOMO (1Px , 1Py )–LUMO (1Pz ) transitions because it is optically forbidden.

230

K. Koyasu et al.

A variety of the gas-phase methods have been applied to the chemically modified Au/Ag superatoms including 1–8 by introducing them in the gas phase using electrospray ionization (ESI) or matrix-assisted laser desorption/ionization (MALDI) method (Fig. 8.1). These studies have provided novel and complementary information on their intrinsic geometric and electronic structures that cannot be obtained by conventional characterization methods. This chapter examines the recent progress in the gas-phase studies on chemically synthesized Au/Ag-based clusters [(Au/Ag)x Ly ]z .

8.2 Mass Spectrometry (MS) Mass spectrometry plays several important roles in the gas-phase studies on [(Au/Ag)x Ly ]z : (1) determination of chemical compositions and net charge, which are the key descriptors of [(Au/Ag)x Ly ]z ; (2) detection and identification of intermediate species formed in solution during the formation and reactions of the clusters; and (3) as an interface for the gas-phase measurements listed in Fig. 8.1. The key requirement is the ionization of the chemically synthesized clusters in the intact form. To this end, the conventional methods such as ESI [49] and MALDI [50] have been used. This section showcases typical examples of mass spectrometric characterization of the synthesized products and mass spectrometric detection of reaction intermediates.

8.2.1 Experimental Methods Schematic setups for the ESI and MALDI mass spectrometry (ESI/MALDI MS) are shown in Fig. 8.1. [(Au/Ag)x Ly ]z intrinsically charged (z = 0) in solution like 1–8 can be directly introduced into the mass spectrometer as a continuous (cw) beam by desolvation in the ESI source. The ESI method can be applied even to neutral clusters (z = 0) if they are charged either by the protonation or deprotonation of the ligands before ESI or attachment of cationic species such as Cs+ during ESI. The ESI method also allows us to sample intermediate and transient species nascently formed in solution. In contrast, a pulsed beam of [(Au/Ag)x Ly ]z can be generated by the MALDI method: solidified cluster samples with a matrix (most typically DCTB) are irradiated with a pulsed laser light. The MALDI method is also used to desorb Au clusters stabilized by polymer [51]. Portions of the cw ion beam from the ESI source or the pulsed ion beams from the MALDI source are injected into the TOF MS by applying a pulsed electric field.

8 Characterization of Chemically Modified Gold/Silver Superatoms …

231

8.2.2 Mass and Charge Determination of Synthesized Clusters Determination of the chemical compositions and charge states of the isolated clusters is the primary purpose of mass spectrometry. Since the first successful application to a series of Au clusters protected by glutathione (GSH) [49], mass spectrometry has been widely and routinely used to check the purity of the synthesized cluster samples. Figure 8.4a, b shows the first successful applications: the ESI mass spectrum for [Au25 (SG)18–x (SG– )x ]− , where x represents the number of the deprotonated GS ligands (SG– ), and the MALDI mass spectrum for [Au25 (SC2 H4 Ph)18 ]− (5), respectively [50]. In Fig. 8.4a, the [Au25 (SG)18–x (SG− )x ]− clusters are multiply charged due to the multiple deprotonations of the SG ligands. Figure 8.4c shows the ESI and MALDI mass spectra of Au940±20 (SC2 H4 Ph)160±4 , which is among the largest clusters (core diameter of 2.9 nm) studied by mass spectrometry so far [52]. MALDI MS has also been successfully applied to characterization of the Au clusters stabilized by polyvinylpyrrolidone (PVP). Figure 8.4d demonstrates the formation of Au34 magic clusters and doping of a single Rh atom [53, 54].

8.2.3 Characterization of Transient Clusters in Solution Another important application of MS is the detection of transient species produced in solution, which will provide fundamental information on the reaction mechanism at the molecular level [55–63]. During the formation of [Au25 (SG)18 ]− , 29 stable intermediate species were detected by retarding the reduction rate of Au(I) precursors using CO as the reducing agent [55]. All the intermediates featured even-numbered valence electrons and their sequential appearance indicated a 2 e− reduction growth mechanism. In situ ESI MS has been applied to a variety of solution processes: the seed-mediated growth of [Au25 (SR)18 ]− to [Au44 (SR)26 ]− [56]; alloying process between [Au2 (SR)2 Cl]− complexes and [Ag44 (SR)30 ]4− [57]; ligand-exchange-induced size transformation of Au38 (SC2 H4 Ph)24 to Au36 (SC6 H4 (t-Bu))24 [58]; intercluster reaction between [Au25 (SC2 H4 Ph)18 ]– (5) and [Ag44 (SC6 H4 F)30 ]4– [59]; and spontaneous alloying between [Au25 (SC2 H4 Ph)18 ]– (5) and [Ag25 (SC6 H3 Me2 )18 ]– (6) [60]. In the last reaction, the formation of dianionic species [Ag25 Au25 (SC6 H3 Me2 )18 (SC2 H4 Ph)18 ]2− was elucidated at the initial stage (Fig. 8.5). The formation of hydride (H– ) adduct [HPdAu8 (PPh3 )8 ]+ was observed in the ESI mass spectrum of the reaction mixture of [PdAu8 (PPh3 )8 ]2+ (2) with NaBH4 (Fig. 8.6a) [61]: 

PdAu8 (PPh3 )8

2+

 + (2) + H− → HPdAu8 (PPh3 )8

(8.2)

232

K. Koyasu et al.

Fig. 8.4 Negative-mode a ESI mass spectrum of [Au25 (SG)18–x (SG− )x ]− (Adapted with permission from Ref. [49]. Copyright (2005) American Chemical Society.) and b MALDI mass spectrum of [Au25 (SC2 H4 Ph)18 ]− (5). (Adapted with permission from Ref. [50]. Copyright (2008) American Chemical Society.) c Negative-mode ESI and positive-mode MALDI mass spectra of Au940±20 (SC2 H4 Ph)160±4 . (Adapted with permission from Ref. [52]. Copyright (2014) American Chemical Society.) d Negative-mode MALDI mass spectra of Rh-doped Au:PVP clusters (Ref. [54]—Reproduced by permission of The Royal Society of Chemistry)

8 Characterization of Chemically Modified Gold/Silver Superatoms …

233

Fig. 8.5 a ESI mass spectrum of the mixture of [Au25 (SC2 H4 Ph)18 ]– (5) and [Ag25 (SC6 H3 Me2 )18 ]– (6) and b DFT-optimized structure of [Ag25 Au25 (SC6 H3 Me2 )18 (SC2 H4 Ph)18 ]2− (Reprinted from Ref. [60] by The Author(s) licensed under CC BY 4.0)

Theoretical studies suggested that the hydride is adsorbed onto the unligated central Pd atom of [PdAu8 (PPh3 )8 ]2+ (2) (Fig. 8.3). Time-resolved ESI mass spectra during the reactions of [HPdAu8 (PPh3 )8 ]+ with Au(I)Cl(PPh3 ) revealed the formation of [HPdAu10 (PPh3 )8 Cl2 ]+ via [HPdAu9 (PPh3 )8 Cl]+ (Fig. 8.6b) while retaining the n* value of eight: +  +  HPdAu8 (PPh3 )8 + AuCl → HPdAu9 (PPh3 )8 Cl

(8.3)

 + +  HPdAu9 (PPh3 )8 Cl + AuCl → HPdAu10 (PPh3 )8 Cl2

(8.4)

234

K. Koyasu et al.

Fig. 8.6 a ESI mass spectra of the mixture of [PdAu8 (PPh3 )8 ]2+ (2) and NaBH4 or NaBD4 . b Time-resolved ESI mass spectra recorded after mixing [HPdAu8 (PPh3 )8 ]+ and AuClPPh3 . c Core structures of the DFT-optimized structures of [PdAu8 (PMe3 )8 ]2+ , [HPdAu8 (PMe3 )8 ]+ , [HPdAu9 (PMe3 )8 Cl]+ , and [HPdAu10 (PMe3 )8 Cl2 ]+ . The ligands are omitted for simplicity (Adapted with permission from Ref. [61]. Copyright 2018 American Chemical Society.)

The H atom remained bonded throughout the growth processes. The core structures during the hydride-mediated growth reactions (8.2)–(8.4) are summarized in Fig. 8.6c. This hydride-mediated growth process is different from that of [Au9 (PPh3 )8 ]3+ (1) to [Au11 (PPh3 )8 Cl2 ]+ (3), where the hydrogen is lost in the form of a proton at the first stage of the growth [62]: 3+ 2+   Au9 (PPh3 )8 (1) + H− → HAu9 (PPh3 )8

(8.5)

 2+  + HAu9 (PPh3 )8 + AuCl → Au10 (PPh3 )8 Cl + H+

(8.6)

 + +  Au10 (PPh3 )8 Cl + AuCl → Au11 (PPh3 )8 Cl2 (3)

(8.7)

These hydride-mediated growth processes represent a promising method of sizecontrolled synthesis based on the bottom-up approach [63].

8 Characterization of Chemically Modified Gold/Silver Superatoms …

235

8.3 Collision-Induced Dissociation Mass Spectrometry (CID MS) In-source CID MS allows us to investigate the fragmentation patterns of [(Au/Ag)x Ly ]z in the electronically ground state upon collisional activation with the atmospheric molecules. The fragmentation patterns provide information on the thermal stability of [(Au/Ag)x Ly ]z , the relative binding affinities of different ligands with respect to the clusters, and the effect of the ligand structure on their binding affinities.

8.3.1 Experimental Methods A typical setup is schematically shown in Fig. 8.1. The cw beam of chemically purified [(Au/Ag)x Ly ]z ions from the ESI source is introduced into a differentially pumped area in which the typical pressure is approximately several hundred Pa. Isolated ions [(Au/Ag)x Ly ]z undergo dissociation upon collision with the buffer gas: z  (Au/Ag)x L y → [(Au/Ag)m Ln ]z + neutral fragments

(8.8)

The voltage applied to the electrodes in the differentially pumped area (V CID in Fig. 8.1) is adjusted to control the nominal collision energy of the [(Au/Ag)x Ly ]z ions with the background gas: the collision energy increases with the VCID .

8.3.2 Fragmentations from Synthesized Clusters In the CID of the representative phosphine-protected Au cluster [Au11 (PPh3 )8 Cl2 ]+ (3), the loss of the AuCl(PPh3 ) units proceeded competitively with that of the PPh3 ligands (Fig. 8.7a, Eqs. 8.9 and 8.10) [64].  +  + Au (PPh3 )8−n Cl2 + n PPh3 + Au11 (PPh3 )8 Cl2 (3) →  11 Au11−m (PPh3 )8−m Cl2−m + m AuCl(PPh3 )

(8.9) (8.10)

The numbers of released PPh3 and AuCl(PPh3 ) units increased with the increase in the collision energy (Fig. 8.7a). According to Eq. 8.1, the n* values in the daughter ions formed in Eqs. 8.9 and 8.10 are calculated to be eight, which agrees with that predicted by the electronic shell model. The direct loss of anionic ligand Cl– while

236

K. Koyasu et al.

Fig. 8.7 a A typical CID mass spectrum of [Au11 (PPh3 )8 Cl2 ]+ (3) as a function of the CID voltage (V CID ) (Adapted with permission from Ref. [64]. Copyright 2018 American Chemical Society). The notation (n, m) represents the numbers of released PPh3 and AuCl(PPh3 ) units. Collision energy-resolved fragmentation curves of b [Ag25 (SC6 H3 Me2 )18 ]− (6) and c [Ag29 (S2 C6 H4 )12 ]3− (Adapted with permission from Ref. [69]. Copyright 2017 American Chemical Society)

leaving the daughter ions with nine electrons did not proceed, indicating that the heterolytic dissociation of the Au–Cl bond is not energetically favored. These results revealed that the CID pathways are governed by the electronic stability of the daughter ions. In the low-energy CID of the representative thiolate-protected Au cluster [Au25 (SC2 H4 Ph)18 ]− (5), [Au21 (SC2 H4 Ph)14 ]− was formed as a major daughter ion by losing an Au4 (SC2 H4 Ph)4 unit (Eq. 8.11) [65, 66]. The loss of an Au4 (SR)4 unit is commonly observed not only in the CID of other [Aux (SR)y ]+/− clusters [67] but also in the MALDI of [Aux (SR)y ] recorded under high laser fluence [50]. Theoretical calculations predicted that the dissociation of [Au25 (SMe)18 ]− into [Au21 (SMe)14 ]− and Au4 (SMe)4 is exothermic only by 0.82 eV and proceeds via complex rearrangement of the intracluster chemical bonds [68]. The preferential loss of Au4 (SC2 H4 Ph)4 was ascribed to the electronic stability of the daughter anion [Au21 (SC2 H4 Ph)14 ]− (n* = 8) and the high stability of neutral fragment Au4 (SC2 H4 Ph)4 with a cyclic structure. When the collision energy was further increased, loss of the second Au4 (SC2 H4 Ph)4 as well as a (C2 H4 Ph)2 unit was observed (Eqs. 8.12 and 8.13):

8 Characterization of Chemically Modified Gold/Silver Superatoms …

  − − Au25 (SC2 H4 Ph)18 (5) → Au21 (SC2 H4 Ph)14 + Au4 (SC2 H4 Ph)4  −  − Au21 (SC2 H4 Ph)12 S2 + (C2 H4 Ph)2 − Au21 (SC2 H4 Ph)14 →  Au17 (SC2 H4 Ph)10 + Au4 (SC2 H4 Ph)4

237

(8.11) (8.12) (8.13)

The n* values in the daughter ions in Eqs. 8.12 and 8.13 are eight if we consider that two electrons are removed by an S2 unit. This estimation supports the conclusion that the CID channels are governed by the electronic stability of the daughter ions. Thiolate-protected Ag clusters such as [Ag25 (SC6 H3 Me2 )18 ]− (6), [Ag29 (S2 C6 H4 )12 ]3− , and [Ag44 (SC6 H4 F)30 ]4− exhibit different CID patterns from the Au analogues. [Ag25 (SC6 H3 Me2 )18 ]− (6) undergoes sequential loss of an Ag3 (SR)3 unit (Fig. 8.7b, Eqs. 8.14 and 8.15) [69]:   − − Ag25 (SC6 H3 Me2 )18 (6) → Ag22 (SC6 H3 Me2 )15 + Ag3 (SC6 H3 Me2 )3 

Ag22 (SC6 H3 Me2 )15

−

−  → Ag19 (SC6 H3 Me2 )12 + Ag3 (SC6 H3 Me2 )3

(8.14) (8.15)

The above fragmentation pathways are determined by the electronic stability of the daughter ions (n* = 8) and the geometric stability of Ag3 (SC6 H3 Me2 )3 units. [Ag29 (S2 C6 H4 )12 ]3− dissociates into two anionic fragments (Fig. 8.7c, Eqs. 8.16 and 8.17): 3− 2−  −   Ag29 (S2 C6 H4 )12 → Ag24 (S2 C6 H4 )9 + Ag5 (S2 C6 H4 )3

(8.16)

2− −  −   Ag24 (S2 C6 H4 )9 → Ag19 (S2 C6 H4 )4 + Ag5 (S2 C6 H4 )3

(8.17)

Both of the fragments [Ag24 (S2 C6 H4 )9 ]2− and [Ag19 (S2 C6 H4 )4 ]− of Eqs. 8.16 and 8.17 are eight electron systems. [Ag44 (SC6 H4 F)30 ]4− (n* = 18) dissociates into [Ag43 (SC6 H4 F)28 ]3− (n* = 18) and [Ag1 (SC6 H4 F)2 ]− :  3−  −  4− Ag43 (SC6 H4 F)28 + Ag1 (SC6 H4 F)2 3−  − Ag44 (SC6 H4 F)30 →  Ag42 (SC6 H4 F)27 + Ag2 (SC6 H4 F)3

(8.18) (8.19)

The CID pattern also provides the information on the relative binding affinity of the different ligands [71]. In the low-energy CID of [Au11 (PPh3 )7 (NHCiPr )Cl2 ]+ (4), synthesized by replacing one of the PPh3 ligands of [Au11 (PPh3 )8 Cl2 ]+ (3) with an N-heterocyclic carbene (NHCiPr ), the loss of NHCiPr was suppressed and that of AuCl(NHCiPr ) or AuCl(PPh3 ) was dominant (Eqs. 8.20 and 8.21) [34]:

238

K. Koyasu et al.

   + Au11 (PPh3 )7 NHCiPr Cl2 (4) →

  +  Au (PPh3 )7 Cl + AuCl NHCiPr (8.20)     10 + Au10 (PPh3 )6 NHCiPr Cl + AuCl(PPh3 ) (8.21)

This result demonstrates that NHCiPr has significantly higher binding affinity to Au than PPh3 through the strong Au–C bond [71]. The CID MS illustrates how the structures of the ligands affect by the interaction with the Au clusters. For example, the stability against CID processes was dependent on the position of carboxyl group on mercaptobenzoic acid used as protecting ligands of the Au25 clusters: the stability decreased in the order of [Au25 (pSC6 H4 CO2 H)18 ]− > [Au25 (m-SC6 H4 CO2 H)18 ]− > [Au25 (o-SC6 H4 CO2 H)18 ]− [72]. This trend was explained in terms of weakening of the Au–S bonds by the steric effect of the carboxyl group at the ortho position.

8.3.3 Fragmentations from Transient Clusters in Solution In-source CID coupled with the ESI source also allows us to monitor the fragmentation pathways of the metal clusters produced in situ in solution. For example, all phosphine-protected Au clusters such as [Au20 (PPh3 )8 ]2+ [70] and [Au11 (dppp)5 ]3+ (dppp = 1,3-bis(diphenylphosphino)propane) [73] underwent dissociation of phosphine ligands (Eq. 8.22), indicating that the Au–P bonds are weakest within the systems (Fig. 8.8a, b): 2+ 2+   → Au20 (PPh3 )8−n + n PPh3 Au20 (PPh3 )8

(8.22)

The n* values of the fragments are 18, illustrating the importance of electronic stability of the fragments. The PPh3 ligands were sequentially lost up to n = 4 according to Eq. 8.22, but the loss of the fifth PPh3 was significantly retarded. This suggests that eight PPh3 ligands are divided into two groups in terms of the binding affinity to the Au20 core. This inference is consistent with the theoretical prediction that four PPh3 ligands are weakly adsorbed on four (111) facets of the pyramidal Au20 core, while the remaining four PPh3 ligands are more strongly coordinated to the apex sites of the Au20 core (Fig. 8.8c, d).

8.4 Ion-Mobility–Mass Spectrometry (IM MS) IM MS determines the collision cross section (CCS) of [(Au/Ag)x Ly ]z , which directly reflects the geometrical motif including the ligand layer (Fig. 8.1). Thus, IM MS also allows us to identify structural isomers if present and to monitor an isomerization process induced by collisional excitation.

8 Characterization of Chemically Modified Gold/Silver Superatoms …

239

Fig. 8.8 a Mass spectrum and b CID mass spectrum of [Au20 (PPh3 )8 ]2+ . The collision gas was benzene and the average collision energy was ~0.64 eV. DFT-optimized structures of c [Au20 (PH3 )8 ]2+ and d [Au20 (PH3 )4 ]2+ (Adapted with permission from Ref. [70]. Copyright 2004 American Chemical Society.)

8.4.1 Experimental Methods In a commercially available apparatus (Synapt HDMS, Waters UK Ltd.), the massselected beam of [(Au/Ag)x Ly ]z prepared using the ESI source and a quadrupole mass filter is injected into the traveling wave ion mobility (TWIM) cell [74] by applying a pulsed voltage to a gate electrode. The [(Au/Ag)x Ly ]z ions in the TWIM cell are propelled by the continuous sequence of traveling waves with a triangular shape, but are unable to keep up with the wavefront in the presence of the buffer gas: larger ions spend longer drift time in the cell (T cell ). The experimental CCS exp is estimated by the following equation: √ Tcell =

μ B Aexp z

(8.23)

where the terms z and μ are the charge and the reduced mass of the clusters and the buffer gas, respectively. A and B are constants determined by calibration.

240

K. Koyasu et al.

8.4.2 Morphology The arrival time distribution (ATD) exhibited single peaks for [Au25 (SC2 H4 Ph)18 ]− (5) [65] and binary clusters [X@Ag14 (C ≡ CBu)12 ]+ and [X@Ag8 Cu6 (C ≡ CBu)12 ]+ (X = Cl, Br), showing the absence of distinct isomers (Fig. 8.9a) [75]. Their experimental CCS values were, qualitatively, reproduced by those calculated based on their single-crystal structures. The agreement suggests that the clusters retain their geometrical motifs in the gas phase [65, 75]. The presence of two isomers in [Au8 (PPh3 )x (PPh2 Me)7–x ]2+ was identified from the doublet ATD profiles [76]. IM MS revealed the formation of the dimer of [Au25 (SC2 H4 Ph)18 ]– (5) in solution, which cannot be identified by mass spectrometry based on the mass-to-charge ratio [77]. The following dimerization of [Ag29 (S2 C6 H4 )12 ]3– mediated by alkali metal cations was observed by IM MS measurement (Fig. 8.9b) [78]: 3−  4−  + 2Na+ → Ag29 (S2 C6 H4 )12 Na 2 2 Ag29 (S2 C6 H4 )12

(8.24)

8.4.3 Collision-Induced Isomerization The CCSs of [Au9 (PPh3 )8 ]3+ (1) and [PdAu8 (PPh3 )8 ]2+ (2) having a crown motif (Fig. 8.10) in methanol electrosprayed into the gas phase were determined to be 442 and 422 Å2 , respectively [79]. The CCS values calculated for their single-crystal structures using the exact hard-sphere scattering method [80] were 413 and 420 Å2 , respectively. In contrast, those calculated using the projection approximation [81] and diffuse-hard-sphere scattering methods [82] deviate significantly from the experimental values, suggesting that it is not trivial to predict the packing structures of the clusters from the CCS values alone. The arrival time distributions of [Au9 (PPh3 )8 ]3+ (1) and [PdAu8 (PPh3 )8 ]2+ (2) were monitored by increasing the collision energy by reducing the He pressure in the cell to test the possibility of detecting structural isomers of [Au9 (PPh3 )8 ]3+ having the Au9 cores with a butterfly motif [83]. [Au9 (PPh3 )8 ]3+ (1) and [PdAu8 (PPh3 )8 ]2+ (2) underwent isomerization to smaller species with the CCS values of 404 and 402 Å2 , respectively (Fig. 8.10). We interpret the results in terms of the collision-induced transformation of the ligand layer structures from the disordered phase into the densely packed phase due to CH–π and π–π interactions found in the single crystal. This observation suggests that the ESI allows the isolation of the clusters while retaining the structures including the ligand layers in dispersing media, which are determined by the subtle balance between ligand–solvent and ligand–ligand interactions.

8 Characterization of Chemically Modified Gold/Silver Superatoms …

241

Fig. 8.9 a Arrival time distribution and DFT-optimized structures of [Cl@Ag14 (C ≡ CBu)12 ]+ and [Cl@Ag8 Cu6 (C ≡ CBu)12 ]+ (Reprinted with permission from Ref. [75]. Copyright 2017 American Chemical Society). b Arrival time distribution of [Ag29 (S2 C6 H4 )12 Na]2− and its dimer 4− [Ag29 (S2 C6 H4 )12 Na]4− 2 . Inset shows the DFT-optimized structure of [Ag29 (S2 C6 H4 )12 Na]2 (Reproduced with permission of RSC Pub in the format Book via Copyright Clearance Center. From Ref. [78])

242

K. Koyasu et al.

Fig. 8.10 Arrival time distributions of a [Au9 (PPh3 )8 ]3+ (1) and b [PdAu8 (PPh3 )8 ]2+ (2). The numbers indicate the He flow rates supplied to the cell in mL min−1 (Adapted with permission from Ref. [79]. Copyright 2018 American Chemical Society)

8.5 Photoelectron Spectroscopy (PES) Photoelectron spectroscopy (PES) on negatively charged [(Au/Ag)x Ly ]z (z < 0) allows us to directly probe the electronic structures of occupied states: energy levels with respect to the vacuum level and density of states.

8.5.1 Experimental Methods The negatively charged [(Au/Ag)x Ly ]z (z < 0) clusters from the ESI source are irradiated by pulsed laser light, after the mass selection (Fig. 8.1):  z z+1  + e− (Au/Ag)x L y + h ν → (Au/Ag)x L y

(8.25)

The kinetic energy of the photoelectron (E kin ) is determined by measuring the TOF of the electrons using a magnetic-bottle-type photoelectron spectrometer (MB PES), by which the photoelectrons detached toward all the solid angles are collected by an inhomogeneous magnetic field. The electron binding energy (E B ) is obtained by the energy conservation law expressed as E B = E h[ν] −E kin

(8.26)

8 Characterization of Chemically Modified Gold/Silver Superatoms …

243

Fig. 8.11 a Schematic presentation of electronic structures of (M13 )5+ superatoms (M = Au, Ag). (Ref. [84, 85]—Reproduced by permission of The Royal Society of Chemistry). b PE spectra of [Au25 (SC12 H25 )18 ]− (Ref. [84]), [Au25 (SC2 H4 Ph)18 ]− (5) (Ref. [85]), and [Ag25 (SC6 H3 Me2 )18 ]− (6) (Ref. [85]) recorded at 355 nm

where Eh[ν ] represents the photon energy (Fig. 8.11a).

8.5.2 Electron Affinities of (M13 )5+ (M = Au, Ag) Superatoms Figure 8.11b shows the PE spectra of [Au25 (SC12 H25 )18 ]− [84], [Au25 (SC2 H4 Ph)18 ]− (5) [85], and [Ag25 (SC6 H3 Me2 )18 ]− (6) [85] recorded at 355 nm. The spectra of [Au25 (SC12 H25 )18 ]− and [Au25 (SC2 H4 Ph)18 ]− (5) exhibit small humps and intense bands at the similar energies which are assigned to the photodetachment from the 1P superatomic orbital and Au 5d bands, respectively. Since SCXRD results showed that the structural difference between [Au25 (SC2 H4 Ph)18 ]− and [Au25 (SC2 H4 Ph)18 ]0 is small [86], the spectral onset corresponds to the adiabatic electron affinity (AEA) of [Au25 (SC2 H4 Ph)18 ]0 . The AEA values of [Au25 (SC12 H25 )18 ]0 and [Au25 (SC2 H4 Ph)18 ]0 thus determined were 2.2 and 2.36 ± 0.01 eV, suggesting that the electronic structure of the superatomic (Au13 )5+ core is not affected seriously by the structures of the thiolates. These experimental AEA values are significantly smaller than those of a model system [Au25 (SCH3 )18 ]0 predicted theoretically (3.0 eV

244

K. Koyasu et al.

[87] and 3.17 eV [88]). The PE spectrum of Ag analogues [Ag25 (SC6 H3 Me2 )18 ]− (6) also exhibits two bands [85], which were assigned to the photodetachment from superatomic 1P orbitals and 4d orbitals localized on the Ag atoms, respectively. The AEA of [Ag25 (SC6 H3 Me2 )18 ]0 was determined to be 2.02 ± 0.01 eV, which is comparable to that of [Au25 (SR)18 ]0 . Doping of heteroatom(s) is a promising approach to enhancing the stability and further improving the properties of [(Au/Ag)x Ly ]z . State-of-the-art synthesis based on coreduction and galvanic replacement allowed us to precisely define the number, element, and location of heteroatom(s) introduced into [(Au/Ag)x Ly ]z [89, 90]. For examples, [Ag25 (SC6 H3 Me2 )18 ]− (6), [PtAg24 (SC6 H3 Me2 )18 ]2– (7), and [PdAg24 (SC6 H3 Me2 )18 ]2– (8) in Fig. 8.1 have M@(Ag+ )12 (M = Ag+ , Pd, Pt) superatomic cores and provide an ideal platform to study the effect of single-atom doping on the electronic structures. PES measurements revealed that the binding energies of valence electrons in M@(Ag+ )12 (M = Pd, Pt) were smaller than that in (Ag+ )13 due to the reduction in formal charge of the core potential [91].

8.5.3 Thermionic Emission from (M13 )5+ Superatoms The PE spectra of [Au25 (SC2 H4 Ph)18 ]− (5) and [Ag25 (SC6 H3 Me2 )18 ]− (6) recorded at 266 nm (Fig. 8.12a) exhibit completely different profiles from those at 355 nm (Fig. 8.11b). The PE spectra at 266 nm do not show the band structures observed at 355 nm (Fig. 8.11b) but are dominated by new bands at E B > 4.0 eV. These findings indicate that slow electron emission dominates over direct photodetachment upon photoirradiation of [Au25 (SC2 H4 Ph)18 ]− (5) and [Ag25 (SC6 H3 Me2 )18 ]− (6) at 266 nm [85]. The emission of slow photoelectrons from the naked cluster anions of W, Nb, and Pt [92] has been observed and ascribed to thermionic emission (TE) from vibrationally excited anionic states. The absence of photoinduced TE from the naked cluster anions of Au and Ag was ascribed to their large VDEs and small vaporization energies in the corresponding bulk metal [92]. If the photon energy adsorbed (4.66 eV) is equally distributed to the vibrational degrees of freedom of the Au13 and Ag13 cores, the temperature of the core reaches 1.6 × 103 K. Kinetic energy distributions simulated at 1.6 × 103 K (black dotted lines in Fig. 8.12a) reproduce the experimentally observed profiles. This agreement supports the assignment to TE from the Au13 and Ag13 cores of [Au25 (SC2 H4 Ph)18 ]− (5) and [Ag25 (SC6 H3 Me2 )18 ]− (6), respectively. The above discussion suggests that the key to TE is the suppression of fragmentation of cluster anions electronically excited above the electron detachment threshold. The PD MS of [Au25 (SC2 H4 Ph)18 ]− (5) and [Ag25 (SC6 H3 Me2 )18 ]− (6) recorded at 266 nm were dominated by the depletion of parent ions: no photofragment ions from [Au25 (SC2 H4 Ph)18 ]− (5) were observed, whereas [Ag22 (SC6 H3 Me2 )15 ]− was observed as a minor fragment from [Ag25 (SC6 H3 Me2 )18 ]− (6). These results indicate that the TE is a major decay pathway of [M25 (SR)18 ]− (M = Au, Ag) photoexcited at 266 nm.

8 Characterization of Chemically Modified Gold/Silver Superatoms …

245

Fig. 8.12 a PE spectra of [Au25 (SC2 H4 Ph)18 ]− (5) and [Ag25 (SC6 H3 Me2 )18 ]− (6) recorded at 266 nm. Red solid and black dotted lines correspond to experimental data and simulated curves for thermionic emission of the M13 core, respectively [85]. b Jablonski diagram illustrating the TE pathway of [M25 (SR)18 ]− following the photoexcitation at 266 nm. Horizontal lines are vibrational levels. Gray area indicates the electron detachment continuum. PA = photoabsorption; PD = photodissociation; IC = internal conversion; TE = thermionic emission [85]. (c) Electron photodetachment yield measured as a function of the laser wavelength for [Au25 (SG)12 (SG– )6 ]– (Adapted with permission from Ref. [85, 95]. Copyright 2010 American Chemical Society.)

246

K. Koyasu et al.

The relaxation pathways of [M25 (SR)18 ]− upon photoexcitation at 266 nm are explained by a Jablonski diagram (Fig. 8.12b). The initial step is selective excitation of [M25 (SR)18 ]− into [M25 (SR)18 ]− *, which is embedded in the photodetachment continuum (shaded area in Fig. 8.12b). Electronic transitions within the M13 core as well as metal-to-ligand transitions [93] would be involved in the electronic transitions at 266 nm. Electronically excited [M25 (SR)18 ]− * quickly undergoes internal conversion (IC) to form vibrationally excited [M25 (SR)18 ]− followed by TE leaving internal energy in the remaining neutral [M25 (SR)18 ]0 . Photodissociation, a competing process of IC, is almost completely suppressed even though the photon energy exceeds the energy required for dissociation into [M21 (SR)14 ]− and M4 (SR)4 (2.9 eV for M = Au) [68]. Protection of the M13 core by stiff M2 (SR)3 units [94] may contribute to the promotion of the IC process of [M25 (SR)18 ]− * by retarding the nuclear motion toward the dissociation. TE is also involved in the much more efficient photoelectron detachment from [Au25 (SG)12 (SG– )6 ]– at 200–300 nm than at 430–670 nm (Fig. 8.12c) [95].

8.6 Other Methods This section briefly touches on other methods used for the characterization of [(Au/Ag)x Ly ]z : photodissociation mass spectrometry (PD MS) and surface-induced dissociation mass spectrometry (SID MS).

8.6.1 Photodissociation Mass Spectrometry (PD MS) Isolated ions [(Au/Ag)x Ly ]z undergo dissociation upon photoexcitation when the energy of absorbed laser light exceeds that required for dissociation: 

(Au/Ag)x L y

z

z †  + hν → (Au/Ag)x L y → [(Au/Ag)m Ln ]z + neutral fragments

(8.27)

Clusters in an electronically excited state [(Au/Ag)x Ly ]z† may dissociate directly or undergo IC (Fig. 8.12b) followed by intracluster energy redistribution. Instead of detecting the depletion of the incident light as in the case of conventional spectroscopy in solution, the depletion of parent ions or the yield of daughter ions is monitored by mass spectrometry. The action spectra recorded as a function of the laser wavelength in principle correspond to the optical absorption spectra of [(Au/Ag)x Ly ]z isolated in gas phase. PD mass spectra can be measured by irradiating the chemically purified [(Au/Ag)x Ly ]z just before introduction into the mass spectrometer (Fig. 8.1) [85]. A unique feature of PD MS is that the energy deposited to the clusters is more precise

8 Characterization of Chemically Modified Gold/Silver Superatoms …

247

and can be larger than that by CID. As a result, the PD at ultraviolet (UV) light results in richer fragmentation patterns compared to those by the CID method (Sect. 9.3). Novel dissociation processes were found in the PD of [Au25 (SC6 H4 CO2 H)18 ]– (NH4 + )3 at 193 nm [96]. The initial stage of the PD was the sequential loss of ammonium salts leading to [Au25 (SC6 H4 CO2 H)15 ]2+ : −    Au25 (SC6 H4 CO2 H)18 NH+ 4 3 + hν(193 nm)  −1+x  +    → Au25 (SC6 H4 CO2 H)18−x NH4 3−x + x(SC6 H4 CO2 H− ) NH+ 4 , x = 1−3 (8.28) This step is followed by two competing fragmentation pathways: 2+  + hν(193 nm) Au25 (SC6 H4 CO2 H)15  2+ + y S(C6 H4 CO2 H)2 , y = 1−7 Au25 S y (SC6 H4 CO2 H)15−2y 2+ →  Au25 S2y (SC6 H4 CO2 H)15−2y + y (C6 H4 CO2 H)2 , y = 1−7

(8.29) (8.30)

These PD processes (8.28)–(8.30) lead to nearly complete removal of the thiolates while retaining the total number of Au atoms. This phenomenon is similar to the observation of [Au25 S~12 ]– in the laser desorption process of [Au25 (SC2 H4 Ph)18 ]− (5) [97]. Also observed in the PD MS was [Au4 (SC6 H4 CO2 H)4 + Na]+ , which may be produced via vibrationally excited states in the electronically ground state as in the case of the CID processes.

8.6.2 Surface-Induced Dissociation (SID) Surface-induced dissociation (SID) is another method used to study the dissociation processes of [(Au/Ag)x Ly ]z . It has been demonstrated that ~10% of the collisional energy can be deposited into a projectile by collision with a solid surface on a timescale of femtoseconds [98]. Since the collision energy of the ionic species can be easily tuned over a wide range, collision with the surface can impart much larger energy than that by PD and CID. Efficient and ultrafast energy conversion may promote further fragmentation of [(Au/Ag)N (L)M ]Z . It is reported that multiply charged phosphine-protected Au clusters, such as [Au7 (PPh3 )6 ]2+ , [Au8 (PPh3 )6 ]2+ , [Au8 (PPh3 )7 ]2+ , and [Au9 (PPh3 )7 ]2+ , underwent fissions into ionic fragments in addition to loss of a neutral PPh3 [99]: 2+ +  +   → Au6 (PPh3 )4 + Au1 (PPh3 )2 Au7 (PPh3 )6

(8.31)

The results showed strong contrast to the results of CID measurement on the similar system, in which the loss of a neutral PPh3 was mainly observed [99]. However, the interpretation of the SID mass spectra is not straightforward due to the concur-

248

K. Koyasu et al.

rent charge transfer processes between the projectiles and surfaces. For example, charge stripping from [Ag11 (SG)7 ]3− was observed upon the collision with a selfassembled monolayer of a fluorocarbon on the gold surface [100]. It was proposed that this charge stripping is associated with the release of anionic fragments, such as Ag(SG)− and Ag4 (SG)− 3.

8.7 Summary This chapter summarizes the recent progress in the gas-phase studies on ligandprotected Au/Ag clusters [(Au/Ag)x Ly ]z synthesized in solution with atomic precision. Mass spectrometry (MS) coupled with soft ionization such as electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI) methods has been routinely used to determine the chemical compositions of [(Au/Ag)x Ly ]z . Another important application of MS is the in situ detection of transient species produced in solution, which will help the molecular-level understanding of the mechanism of complex processes in solutions: seed- or hydride-mediated growth processes, alloying processes, ligand exchange reactions, ligand-exchange-induced size transformation, and metal exchange reactions between the clusters. Novel intermediates such as hydride-doped clusters and dimers of the clusters were observed. Collision-induced dissociation (CID) MS allows us to investigate the fragmentation patterns of [(Au/Ag)x Ly ]z upon collisional activation with the gaseous molecules. The fragmentation patterns provide information on the thermal stability of [(Au/Ag)x Ly ]z , the relative binding affinities of different ligands with respect to the clusters, and the effect of the ligand structure on their binding affinities. Ion-mobility (IM) MS determines the collision cross section (CCS) of [(Au/Ag)x Ly ]z , which directly reflects the geometrical motif including the ligand layer. IM MS also allows us to identify structural isomers if present and to monitor an isomerization process induced by collisional excitation. Photoelectron spectrometry (PES) on negatively charged [(Au/Ag)x Ly ]z (z < 0) allows us to directly probe the electronic structures of occupied states: energy levels with respect to the vacuum level and density of states. Thiolate-protected Au/Ag clusters exhibit thermionic emission upon UV laser irradiation, whereas naked Au/Ag clusters do not. The characterization of [(Au/Ag)x Ly ]z in the gas phase will elucidate their intrinsic geometric and electronic structures in the absence of the perturbation from the environments. Future studies with a combination with the gas-phase methods and the conventional methods such as single-crystal X-ray crystallography and X-ray absorption spectroscopy [101] will deepen our understanding on the structures–properties correlations of [(Au/Ag)x Ly ]z and contribute to the development of the materials science of chemically modified superatoms.

8 Characterization of Chemically Modified Gold/Silver Superatoms …

249

References 1. Haberland, H. (ed.): Clusters of Atoms and Molecules. Springer, Berlin (1994) 2. Dietz, T.G., Duncan, M.A., Powers, D.E., Smalley, R.E.: Laser production of supersonic metal cluster beams. J. Chem. Phys. 74, 6511 (1981); Bondybey, V.E., English, J.H.: Laser induced fluorescence of metal clusters produced by laser vaporization: gas phase spectrum of Pb2 . J. Chem. Phys. 74, 6978 (1982) 3. Haberland, H., Karrais, M., Mall, M.: A new type of cluster and cluster ion source. Z. Phys. D 20, 413 (1991) 4. de Heer, W.A.: The physics of simple metal clusters: experimental aspects and simple models. Rev. Mod. Phys. 65, 611 (1993) 5. Taylor, K.J., Pettiette-Hall, C.L., Cheshnovsky, O., Smalley, R.E.: Ultraviolet photoelectron spectra of coinage metal clusters. J. Chem. Phys. 96, 3319 (1992) 6. Li, J., Li, X., Zhai, H.-J., Wang, L.-S.: Au20 : a tetrahedral cluster. Science 299, 864 (2003) 7. Häkkinen, H., Yoon, B., Landman, U., Li, X., Zhai, H.-J., Wang, L.-S.: On the electronic and atomic structures of small Au− n (N = 4–14) clusters: a photoelectron spectroscopy and density-functional study. J. Phys. Chem. A 107, 6168 (2003) 8. Bulusu, S., Li, X., Wang, L.-S., Zeng, X.C.: Evidence of hollow golden cages. Proc. Natl. Acad. Sci. 103, 8326 (2006) 9. Lechtken, A., Schooss, D., Stairs, J.R., Blom, M.N., Furche, F., Morgner, N., Kostko, O., von Issendorff, B., Kappes, M.M.: Au34 − : a chiral gold cluster? Angew. Chem., Int. Ed. 46, 2944 (2007) 10. Huang, W., Ji, M., Dong, C.-D., Gu, X., Wang, L.-M., Gong, X.G., Wang, L.-S.: Relativistic effects and the unique low-symmetry structures of gold nanoclusters. ACS Nano 2, 897 (2008) 11. Gilb, S., Jacobsen, K., Schooss, D., Furche, F., Ahrichs, R., Kappes, M.M.: Electronic photodissociation spectroscopy of Au− n ·Xe (n = 7–11) versus time-dependent density functional theory prediction. J. Chem. Phys. 121, 4619 (2004) 12. Glib, S., Weis, P., Furche, F., Ahlrichs, R., Kappes, M.M.: Structures of small gold cluster cations (Au+n , n < 14): ion mobility measurements versus density functional calculations. J. Chem. Phys. 116, 4094 (2002) 13. Furche, F., Ahlrichs, R., Weis, P., Jacob, C., Gilb, S., Bierweiler, T., Kappes, M.M.: The structures of small gold cluster anions as determined by a combination of ion mobility measurements and density functional calculations. J. Chem. Phys. 117, 6982 (2002) 14. Xing, X., Yoon, B., Landman, U., Parks, J.H.: Structural evolution of Au nanoclusters: from planar to cage to tubular motifs. Phys. Rev. B 74, 165423 (2006) 15. Gruene, P., Rayner, D.M., Redlich, B., van der Meer, A.F.G., Lyon, J.T., Meijer, G., Fielicke, A.: Structures of neutral Au7 , Au19 , and Au20 clusters in the gas phase. Science 321, 674 (2008) 16. St. Becker, Dietrich, G., Hase, H.-U., Kluge, H.-J., Kreisle, D., St. Krücheberg, Lindinger, M., Lützenkirchen, K., Weidele, H., Ziegler, J.: Collision induced dissociation of stored gold cluster ions. Z. Phys. D 30, 341 (1994) 17. Wallace, W.T., Whetten, R.L.: Coadsorption of CO and O2 on selected gold clusters: evidence for efficient room-temperature CO2 generation. J. Am. Chem. Soc. 124, 7499 (2002) 18. Cohen, M.L., Chou, M.Y., Knight, W.D., de Heer, W.A.: Physics of metal clusters. J. Phys. Chem. 91, 3141 (1987) 19. Castleman, A.W., Khanna, S.N.: Clusters, superatoms, and building blocks of new materials. J. Phys. Chem. C 113, 2664 (2009) 20. Issendorff, B., Cheshnovsky, O.: Metal to insulator transitions in clusters. Annu. Rev. Phys. Chem. 56, 549 (2005) 21. El-Sayed, M.A.: Small is different: shape-, size-, and composition-dependent properties of some colloidal semiconductor nanocrystals. Acc. Chem. Res. 37, 326 (2004) 22. Landman, U., Luedtke, W.D.: Small is different: energetic, structural, thermal, and mechanical properties of passivated nanocluster assemblies. Faraday Discuss. 125, 1 (2004)

250

K. Koyasu et al.

23. Heiz, U., Sanchez, A., Abbet, S., Schneider, W.D.: Catalytic oxidation of carbon monoxide on monodispersed platinum clusters: each atom counts. J. Am. Chem. Soc. 121, 3214 (1999) 24. Landman, U.: Materials by numbers: computations as tools of discovery. Proc. Natl. Acad. Sci. 102, 6671 (2005) 25. Haruta, M., Kobayashi, T., Sano, H., Yamada, N.: Novel gold catalysts for the oxidation of carbon monoxide at a temperature far below 0 °C. Chem. Lett. 16, 405 (1987) 26. Wang, L.-M., Wang, L.-S.: Probing the electronic properties and structural evolution of anionic gold clusters in the gas phase. Nanoscale 4, 4038 (2012) 27. Yamazoe, S., Koyasu, K., Tsukuda, T.: Nonscalable oxidation catalysis of gold clusters. Acc. Chem. Res. 47, 816 (2014) 28. Weis, P.: Structure determination of gaseous metal and semi-metal cluster ions by ion mobility spectrometry. Int. J. Mass Spectrom. 245, 1 (2005) 29. Ouyang, R., Xie, Y., Jiang, D.-E.: Global minimization of gold clusters by combining neural network potentials and the basin-hopping method. Nanoscale 7, 14817 (2015) 30. Parker, D.H., Wurz, P., Chatterjee, K., Lykke, K.R., Hunt, J.E., Pellin, M.J., Hemminger, J.C., Gruen, D.M., Stock, L.M.: High-yield synthesis, separation, and mass-spectrometric characterization of fullerenes C60 to C266 . J. Am. Chem. Soc. 113, 7499 (1991) 31. Whetten, R.L., Khoury, J.T., Alvarez, M.M., Murthy, S., Vezmar, I., Wang, Z.L., Stephens, P.W., Cleveland, C.L., Luedtke, W.D., Landman, U.: Nanocrystal gold molecules. Adv. Mater. 8, 428 (1996) 32. Maity, P., Tsunoyama, H., Yamauchi, M., Xie, S., Tsukuda, T.: Organogold clusters protected by phenylacetylene. J. Am. Chem. Soc. 133, 20123 (2011) 33. Schmid, G. (ed.), Nanoparticles: From Theory to Application, 2nd, Completely Revised and Updated Ed. (Wiley, New Jersey, 2010) 34. Narouz, M.R., Osten, K.M., Unsworth, P.J., Man, R.W.Y., Salorinne, K., Takano, S., Tomihara, R., Kaappa, S., Malola, S., Dinh, C.-T., Padmos, J.D., Ayoo, K., Garrett, P.J., Nambo, M., Horton, J.H., Sargent, E.H., Häkkinen, H., Tsukuda, T., Crudden, C.M.: N-Heterocyclic carbene-functionalized magic number gold nanoclusters. Nature Chem. 11, 419 (2019) 35. Tsukuda, T.: Toward an atomic-level understanding of size-specific properties of protected and stabilized gold clusters. Bull. Chem. Soc. Jpn. 85, 151 (2012) 36. Tsukuda, T., Häkkinen, H. (eds.): Protected metal clusters: from fundamentals to applications. Elsevier, Amsterdam (2015) 37. Jin, R., Zeng, C., Zhou, M., Chen, Y.: Atomically precise colloidal metal nanoclusters and nanoparticles: fundamentals and opportunities. Chem. Rev. 116, 10346 (2016) 38. Chakraborty, I., Pradeep, T.: Atomically precise clusters of noble metals: emerging link between atoms and nanoparticles. Chem. Rev. 117, 8208 (2017) 39. Azubel, M., Koivisto, J., Malola, S., Bushnell, D., Hura, G.L., Koh, A.L., Tsunoyama, H., Tsukuda, T., Pettersson, M., Häkkinen, H., Kornberg, R.D.: Electron microscopy of gold nanoparticles at atomic resolution. Science 345, 909 (2014) 40. Bellon, P.L., Cariati, F., Manassero, M., Naldini, L., Sansoni, M.: Novel gold clusters. Preparation, properties, and x-ray structure determination of salts of octakis(triarylphosphine)enneagold, [Au9 L8 ]X3 . J. Chem. Soc. D 1423 (1971) 41. Matsuo, S., Takano, S., Yamazoe, S., Koyasu, K., Tsukuda, T.: Selective and high-yield synthesis of oblate superatom [PdAu8 (PPh3 )8 ]2+ . ChemElectroChem 3, 1206 (2016) 42. McKenzie, L.C., Zaikova, T.O., Hutchison, J.E.: Structurally similar triphenylphosphinestabilized undecagolds, Au11 (PPh3 )7 Cl3 and [Au11 (PPh3 )8 Cl2 ]Cl, exhibit distinct ligand exchange pathways with glutathione. J. Am. Chem. Soc. 136, 13426 (2014) 43. Heaven, M.W., Dass, A., White, P.S., Holt, K.M., Murray, R.M.: Crystal structure of the gold nanoparticle [N(C8 H17 )4 ][Au25 (SCH2 CH2 Ph)18 ]. J. Am. Chem. Soc. 130, 3754 (2008) 44. Joshi, C.P., Bootharaju, M.S., Alhilaly, M.J., Bakr, O.M.: [Ag25 (SR)18 ]− : the “golden” silver nanoparticle. J. Am. Chem. Soc. 137, 11578 (2015) 45. Yan, J., Su, H., Yang, H., Malola, S., Lin, S., Häkkinen, H., Zheng, N.: Total structure and electronic structure analysis of doped thiolated silver [MAg24 (SR)18 ]2− (M = Pd, Pt) clusters. J. Am. Chem. Soc. 137, 11880 (2015)

8 Characterization of Chemically Modified Gold/Silver Superatoms …

251

46. Walter, M., Akola, J., Lopez-Acevedo, O., Jadzinsky, P.D., Calero, G., Ackerson, C.J., Whetten, R.L., Grönbeck, H., Häkkinen, H.: A unified view of ligand-protected gold clusters as superatom complexes. Proc. Natl. Acad. Sci. 105, 9157 (2008) 47. Kwak, K., Tang, Q., Kim, M., Jiang, D.-E., Lee, D.: Interconversion between superatomic 6-electron and 8-electron configurations of M@Au24 (SR)18 clusters (M = Pd, Pt). J. Am. Chem. Soc. 137, 10833 (2015) 48. Liu, Y., Chai, X., Cai, X., Chen, M., Jin, R., Ding, W., Zhu, Y.: Central doping of a foreign atom into the silver cluster for catalytic conversion of CO2 toward C–C bond formation. Angew. Chem., Int. Ed. 57, 9775 (2018) 49. Negishi, Y., Nobusada, K., Tsukuda, T.: Glutathione-protected gold clusters revisited: bridging the gap between Gold(I)−Thiolate complexes and Thiolate-protected gold nanocrystals. J. Am. Chem. Soc. 127, 5261 (2005) 50. Dass, A., Stevenson, A., Dubay, G.R., Tracy, J.B., Murray, R.W.: Nanoparticle MALDITOF mass spectrometry without fragmentation: Au25 (SCH2 CH2 Ph)18 and mixed monolayer Au25 (SCH2 CH2 Ph)18−x (L)x . J. Am. Chem. Soc. 130, 5940 (2008) 51. Tsunoyama, H., Tsukuda, T.: Magic numbers of gold clusters stabilized by PVP. J. Am. Chem. Soc. 131, 18216 (2009) 52. Kumara, C., Zuo, X., Cullen, D.A., Dass, A.: Faradaurate-940: synthesis, mass spectrometry, electron microscopy, high-energy x-ray diffraction, and x-ray scattering study of Au∼940±20 (SR)∼160±4 nanocrystals. ACS Nano 8, 6431 (2014) 53. Hayashi, S., Ishida, R., Hasegawa, S., Yamazoe, S., Tsukuda, T.: Doping a single palladium atom into gold superatoms stabilized by PVP: emergence of hydrogenation catalysis. Top. Catal. 61, 136 (2018) 54. Hasegawa, S., Takano, S., Yamazoe, S., Tsukuda, T.: Prominent hydrogenation catalysis of a PVP-stabilized Au34 superatom provided by doping a single Rh atom. Chem. Commun. 54, 5915 (2018) 55. Luo, Z.T., Nachammai, V., Zhang, B., Yan, N., Leong, D.T., Jiang, D.E., Xie, J.P.: Toward understanding the growth mechanism: tracing all stable intermediate species from reduction of Au(I)–Thiolate complexes to evolution of Au25 nanoclusters. J. Am. Chem. Soc. 136, 10577 (2014) 56. Yao, Q.F., Yuan, X., Fung, V., Yu, Y., Leong, D.T., Jiang, D.E., Xie, J.P.: Understanding seed-mediated growth of gold nanoclusters at molecular level. Nat. Commun. 8, 927 (2017) 57. Yao, Q.F., Feng, Y., Fung, V., Yu, Y., Jiang, D.E., Yang, J., Xie, J.P.: Precise control of alloying sites of bimetallic nanoclusters via surface motif exchange reaction. Nat. Commun. 8, 1555 (2017) 58. Zeng, C., Liu, C., Pei, Y., Jin, R.: Thiol ligand-induced transformation of Au38 (SC2 H4 Ph)24 to Au36 (SPh-t-Bu)24 . ACS Nano. 7, 6138 (2013) 59. Krishnadas, K.R., Ghosh, A., Baksi, A., Chakraborty, I., Natarajan, G., Pradeep, T.: Intercluster reactions between Au25 (SR)18 and Ag44 (SR)30 . J. Am. Chem. Soc. 138, 140 (2016) 60. Krishnadas, K.R., Baksi, A., Ghosh, A., Natarajan, G., Pradeep, T.: Structure-conserving spontaneous transformations between nanoparticles. Nat. Commun. 7, 13447 (2016) 61. Takano, S., Hirai, H., Muramatsu, S., Tsukuda, T.: Hydride-mediated controlled growth of a bimetallic (Pd@Au8 )2+ superatom to a hydride-doped (HPd@Au10 )3+ superatom. J. Am. Chem. Soc. 140, 12314 (2018) 62. Takano, S., Hirai, H., Muramatsu, S., Tsukuda, T.: Hydrogen-doped gold superatoms (Au9 H)2+ : synthesis, structure and transformation. J. Am. Chem. Soc. 140, 8380 (2018) 63. Takano, S., Hasegawa, S., Suyama, M., Tsukuda, T.: Hydride doping to chemically-modified gold-based superatoms. Acc. Chem. Res. 51, 3074 (2018) 64. Tomihara, R., Hirata, K., Yamamoto, H., Takano, S., Koyasu, K., Tsukuda, T.: Collisioninduced dissociation of undecagold clusters protected by mixed ligands [Au11 (PPh3 )8 X2 ]+ (X = Cl, C ≡ CPh). ACS Omega 3, 6237 (2018) 65. Angel, L.A., Majors, L.T., Dharmaratne, A.C., Dass, A.: Ion mobility mass spectrometry of Au25 (SCH2 CH2 Ph)18 nanoclusters. ACS Nano 4, 4691 (2010)

252

K. Koyasu et al.

66. Fields-Zinna, C.A., Sampson, J.S., Crowe, M.C., Tracy, J.B., Parker, J.F., deNey, A.M., Muddiman, D.C., Murray, R.W.: Tandem mass spectrometry of thiolate-protected Au nanoparticles Nax Au25 (SC2 H4 Ph)18−y (S(C2 H4 O)5 CH3 )y . J. Am. Chem. Soc. 131, 13844 (2009) 67. Yao, Q., Fung, V., Sun, C., Huang, S., Chen, T., Jiang, D., Lee, J.Y., Xie, J.: Revealing isoelectronic size conversion dynamics of metal nanoclusters by a noncrystallization approach. Nat. Commun. 9, 1979 (2018) 68. Liu, C., Lin, S., Pei, Y., Zeng, X.C.: Semiring chemistry of Au25 (SR)18 : fragmentation pathway and catalytic active site. J. Am. Chem. Soc. 135, 18067 (2013) 69. Chakraborty, P., Baksi, A., Khatun, E., Nag, A., Ghosh, A., Pradeep, T.: Dissociation of gas phase ions of atomically precise silver clusters reflects their solution phase stability. J. Phys. Chem. C 121, 10971 (2017) 70. Zhang, H.-F., Stender, M., Zhang, R., Wang, C., Li, J., Wang, L.-S.: Toward the solution synthesis of the tetrahedral Au20 cluster. J. Phys. Chem. B 108, 12259 (2004) 71. Tang, Q., Hu, G., Fung, V., Jiang, D.-E.: Insights into interfaces, stability, electronic properties, and catalytic activities of atomically precise metal nanoclusters from first principles. Acc. Chem. Res. 51, 2793 (2018) 72. Bertorelle, F., Russier-Antoine, I., Comby-Zerbino, C., Chirot, F., Dugourd, P., Brevet, P.-F., Antione, R.: Isomeric effect of mercaptobenzoic acids on the synthesis, stability, and optical properties of Au25 (MBA)18 nanoclusters. ACS Omega 3, 15635 (2018) 73. Bergeron, D.E., Hudgens, J.W.: Ligand dissociation and core fission from diphosphineprotected gold clusters. J. Phys. Chem. C 111, 8195 (2007) 74. Shvartsburg, A.A., Smith, R.D.: Fundamentals of traveling wave ion mobility spectrometry. Anal. Chem. 80, 9689 (2008) 75. Daly, S., Choi, C.M., Zavras, A., Krsti´c, M., Chirot, F., Connell, T.U., Williams, S.J., Donnelly, P.S., Antoine, R., Giuliani, A., Bonaˇci´c-Koutecký, V., Dugourd, P., O’Hair, R.A.J.: Gasphase structural and optical properties of homo- and heterobimetallic rhombic dodecahedral nanoclusters [Ag14–n Cun (C ≡ CtBu)12 X]+ (X = Cl and Br): ion mobility, VUV and UV spectroscopy, and DFT calculations. J. Phys. Chem. C 121, 10719 (2017) 76. Ligare, M.R., Baker, E.S., Laskin, J., Johnson, G.E.: Ligand induced structural isomerism in phosphine coordinated gold clusters revealed by ion mobility mass spectrometry. Chem. Commun. 53, 7389 (2017) 77. Baksi, A., Chakraborty, P., Bhat, S., Natarajan, G., Pradeep, T.: [Au25 (SR)18 ]2− 2 : a noble metal cluster dimer in the gas phase. Chem. Commun. 52, 8397 (2016) 78. Chakraborty, P., Baksi, A., Mudedla, S.K., Nag, A., Paramasivam, G., Subramanian, V., Pradeep, T.: Understanding proton capture and cation-induced dimerization of [Ag29 (BDT)12 ]3− clusters by ion mobility mass spectrometry. Phys. Chem. Chem. Phys. 20, 7593 (2018) 79. Hirata, K., Chakraborty, P., Nag, A., Takano, S., Koyasu, K., Pradeep, T., Tsukuda, T.: Interconversions of structural isomers of [PdAu8 (PPh3 )8 ]2+ and [Au9 (PPh3 )8 ]3+ revealed by ion mobility mass spectrometry. J. Phys. Chem. C 122, 23123 (2018) 80. Shvartsburg, A.A., Jarrold, M.F.: An exact hard-spheres scattering model for the mobilities of polyatomic ions. Chem. Phys. Lett. 261, 86 (1996) 81. von Helden, G., Hsu, M.-T., Gotts, N., Bowers, M.T.: Carbon cluster cations with up to 84 atoms: structures, formation mechanism, and reactivity. J. Phys. Chem. 97, 8182 (1993) 82. Larriba, C., Hogan, C.J.: Ion mobilities in diatomic gases: measurement versus prediction with non-specular scattering models. J. Phys. Chem. A 117, 3887 (2013) 83. Schulz-Dobrick, M., Jansen, M.: Supramolecular intercluster compounds consisting of gold clusters and Keggin anions. Eur. J. Inorg. Chem. 2006, 4498 (2006) 84. Hirata, K., Yamashita, K., Muramatsu, S., Takano, S., Ohshimo, K., Azuma, T., Nakanishi, R., Nagata, T., Yamazoe, S., Koyasu, K., Tsukuda, T.: Anion photoelectron spectroscopy of free [Au25 (SC12 H25 )18 ]− . Nanoscale 9, 13409 (2017) 85. Hirata, K., Kim, K., Nakamura, K., Kitazawa, H., Hayashi, S., Koyasu, K., Tsukuda, T.: Photoinduced thermionic emission from [M25 (SR)18 ]− (M = Au, Ag) revealed by anion photoelectron spectroscopy. J. Phys. Chem. C. 123, 13174 (2019)

8 Characterization of Chemically Modified Gold/Silver Superatoms …

253

86. Tofanelli, M.A., Salorinne, K., Ni, T.W., Malola, S., Newell, B., Phillips, B., Häkkinen, H., Ackerson, C.J.: Jahn-Teller effects in Au25 (SR)18 . Chem. Sci. 7, 1882 (2016) 87. Akola, J., Walter, M., Whetten, R.L., Häkkinen, H., Grönbeck, H.: On the structure of thiolateprotected Au25 . J. Am. Chem. Soc. 130, 3756 (2008) 88. Kacprzak, K.A., Lehtovaara, L., Akola, J., Lopez-Acevedo, O., Häkkinen, H.: A density functional investigation of thiolate-protected bimetal PdAu24 (SR)z18 clusters: doping the superatom complex. Phys. Chem. Chem. Phys. 11, 7123 (2009) 89. Wang, S., Li, Q., Kang, X., Zhu, M.: Customizing the structure, composition, and properties of alloy nanoclusters by metal exchange. Acc. Chem. Res. 51, 2784 (2018) 90. Hossain, S., Niihori, Y., Nair, L.V., Kumar, B., Kurashige, W., Negishi, Y.: Alloy clusters: precise synthesis and mixing effects. Acc. Chem. Res. 51, 3114 (2018) 91. Kim, K., Hirata, K., Nakamura, K., Kitazawa, H., Hayashi, S., Koyasu, K., Tsukuda, T.: Elucidating the doping effect on the electronic structure of thiolate-protected silver superatoms by photoelectron spectroscopy. Angew. Chem. Int. Ed. (2019) (in press) 92. Ganteför, G., Eberhardt, W., Weidele, H., Kreisle, D., Recknagel, E.: Energy dissipation in small clusters: direct photoemission, dissociation, and thermionic emission. Phys. Rev. Lett. 77, 4524 (1996) 93. Schacht, J., Gaston, N.: From the superatom model to a diverse array of super-elements: a systematic study of dopant influence on the electronic structure of thiolate-protected gold clusters. ChemPhysChem 17, 3237 (2016) 94. Yamazoe, S., Takano, S., Kurashige, W., Yokoyama, T., Nitta, K., Negishi, Y., Tsukuda, T.: Hierarchy of bond stiffnesses within icosahedral-based gold clusters protected by thiolates. Nat. Commun. 7, 10414 (2016) 95. Hamouda, R., Bellina, B., Bertorelle, F., Compagnon, I., Antoine, R., Broyer, M., Rayane, D., Dugourd, P.: Electron emission of gas-phase [Au25 (SG)18 -6H]7− gold cluster and its action spectroscopy. J. Phys. Chem. Lett. 1, 3189 (2010) 96. Black, D.M., Crittenden, C.M., Brodbelt, J.S., Whetten, R.L.: Ultraviolet photodissociation of selected gold clusters: ultraefficient unstapling and ligand stripping of Au25 (pMBA)18 and Au36 (pMBA)24 . J. Phys. Chem. Lett. 8, 1283 (2017) 97. Wu, Z., Gayathri, C., Gil, R.R., Jin, R.: Probing the structure and charge state of glutathionecapped Au25 (SG)18 clusters by NMR and mass spectrometry. J. Am. Chem. Soc. 131, 6535 (2009) 98. Beck, R.D., St. John, P., Homer, M.L., Whetten, R.L.: Impact-induced cleaving and melting of alkali-halide nanocrystals. Science 253, 879 (1991) 99. Johnson, G.E., Laskin, J.: Understanding ligand effects in gold clusters using mass spectrometry. Analyst 141, 3573 (2016) 100. Baksi, A., Harvey, S.R., Natarajan, G., Wysocki, V.H., Pradeep, T.: Possible isomers in ligand protected Ag11 cluster ions identified by ion mobility mass spectrometry and fragmented by surface induced dissociation. Chem. Commun. 52, 3805 (2016) 101. Yamazoe, S., Tsukuda, T.: X-ray absorption spectroscopy on atomically precise metal clusters. Bull. Chem. Soc. Jpn. 92, 193 (2019)

Part IV

Dynamics of Vibrationally and Electronically Excited State Molecules, Ions, and Clusters

Chapter 9

Time-Resolved Study on Vibrational Energy Relaxation of Aromatic Molecules and Their Clusters in the Gas Phase Takayuki Ebata

Time resolved IR-UV pump-probe study reveals the mechanism of IVR and vibrational predissociation, and the time scale of molecules and clusters

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_9

257

258

T. Ebata

Abstract In this chapter, time-resolved study on the vibrational energy relaxation (VER) of gas-phase aromatic molecules and their clusters cooled in supersonic jets are described. The experiment is performed by picosecond IR-UV pump-probe spectroscopy, where the molecules are excited to a specific vibrational level of the electronic ground state (S0 ) by a picosecond IR pulse laser, and the time evolution of the excited level as well as the energy transferred levels are monitored by a picosecond UV pule laser. In the molecule in isolated condition, VER involves intramolecular vibrational energy redistribution (IVR), and in the cluster IVR is followed by vibrational predissociation (VP), if the initial vibrational energy is larger than the binding energy of the cluster. We focus VER of high frequency vibrations; OH, NH and CH stretching vibrations. Among these vibrations, OH and NH starching vibrations are localized motion, while the CH stretching vibration is the collective motion of several CH groups. The molecules studied in this chapter is phenol, aniline, 2-aminopyridine and benzene. We will see how the difference of the vibrational motion affects the IVR process and its speed in these molecules. We also investigate the IVR and VP of the H-bonded clusters and van der Waals clusters. We discuss the route of the energy flow starting from the IR excited vibration to other vibrational levels within the molecules and clusters and the time scales. Keywords Vibrational energy relaxation (VER) · Anharmonic coupling · Intramolecular vibrational energy redistribution (IVR) · Vibrational predissociaiton (VP) · Hydrogen-bond · Pump-probe spectroscopy

9.1 Introduction Vibrational energy relaxation (VER) is a common phenomenon ocurring in liquid [1–20], surface [21, 22], crystal [23] as well as in the gas phase [24–30]. In general, normal mode picture is approximate expression and vibrational modes are not completely orthogonal to each other. So, even in the isolated molecule in the gas phase, the energy put into a vibrational level is redistributed into nearby vibrational levels via anharmoic coupling, which is called intramolecular vibrational energy redistribution (IVR) [24–42]. In solution, the energy put into specific vibrational level of the solute molecule is transferred either directly to the solvent molecules (vibrational energy transfer, VET) or after IVR within the solute molecule [10, 16, 19]. In this sense, investigation of IVR of the isolated molecule and VET in the molecular cluster with the structure well-defined, is ideal to understand mechanism of VER in the T. Ebata (B) Department of Chemistry, Graduate School of Science, Hiroshima University, Higashi-hiroshima 739-8526, Japan e-mail: [email protected] Present Address: Department of Applied Chemistry, National Chaio Tung University, Hsinchu, Taiwan

9 Time-Resolved Study on Vibrational Energy Relaxation …

259

gas phase as well as condensed phase. In this chapter, we discuss VER of aromatic molecules in the gas phase, under supersonically jet-cooled condition. We focus on the IVR of high frequency vibration, namely OH, NH and CH stretching vibrations. First, we will show IVR of the OH stretch of bare phenol [31, 32, 34], and NH2 stretch of bare aniline [35–37] and 2-aminopyridine [39]. Second we discuss the VER of hydrogen(H)-bonded clusters of phenol [33, 36, 38, 42], phenol dimer [31], and benzene clusters [40, 41]. All the experiments are carried out by picosecond infrared (IR)-ultraviolet (UV) pump-probe spectroscopy. We will discuss the route of the energy flow beginning from the IR excited vibration to other vibrational levels within the molecules and clusters and the time scale of each step.

9.2 VER from the OH, CH and NH Stretching Vibrations of Bare Molecules The OH and NH stretching vibrations in aromatic molecules are localized modes different from other modes, because these vibrational motions involve mostly the oscillation of H atom with the fixed heavy atom. Another characteristic of these modes is that they exhibit large red-shift and broadening upon hydrogen(H)-bonding [43, 44]. Here we show our study of VER of the OH stretching vibration of phenol and the stretching vibration of the NH2 group of aniline and their isotopomers. The OH stretching frequency of phenol-h6 is 3657 cm−1 [34, 45, 46] and that of the OD stretch of phenol-d 1 is 2701 cm−1 [34, 45]. Aniline has two NH2 stretching modes, symmetric (νs ) and anti-symmetric (νa ) stretches with the frequencies of 3423 and 3509 cm−1 , respectively [47]. We show the time-resolved study of IVR of the OH stretch of phenol, NH2 stretches of aniline and 2-aminopyridine under cold gas phase condition in supersonic beam. The experiments are carried by the picosecond IRUV pump-probe spectroscopic technique; the picosecond (ps) IR laser pulse (12 ps FWHM) excites the OH or NH2 stretching vibration, and after certain delay time a tunable picosecond UV laser pulse monitors the decay of the IR pumped level as well as the growth of the redistributed levels by resonant two-photon (R2PI) ionization via the S1 state as shown in Fig. 9.1. VER of OH (OD) stretching vibration of phenol: Fig. 9.2 shows the IR spectrum of bare phenol, cooled in a supersonic jet (See Part I), in the OH and CH stretching vibrational region. Upper panel shows the IR-UV double resonance spectrum obtained by nanosecond laser with 0.2 cm−1 resolution. The detail of this spectroscopy is described in Part I. Lower panel shows the ionization-gain IR spectrum obtained with the picosecond laser system with 5 cm−1 resolution, where the UV probe laser pulse is introduced at a delay time of 40 ps from the IR laser pulse and its frequency is fixed to the transition from the IVR redistributed levels.

260

T. Ebata

Fig. 9.1 Energy level diagram and IR-UV pump-probe scheme to investigate VER from the OH stretching vibration of phenol. V1 and V2 are the anharmonic coupling strengths. Reprinted with permission from J. Chem. Phys. 121, 11530–11534 (2004). Copyright 2004 American Institute of Physics

Fig. 9.2 (Upper) IR-UV double resonance spectrum of bare phenol in the energy region of the OH stretching and CH stretching vibration obtained by nanosecond laser. (Lower) IR spectra in the same region obtained by picosecond IR-UV pump-probe scheme. Reprinted with permission from J. Chem. Phys. 120, 7400–7409 (2004). Copyright 2004 American Institute of Physics

Figure 9.3a shows the transient R2PI spectra of phenol-h6 (C6 H5 OH) measured at several delay times (t) after exciting the OH stretching vibration at 3657 cm-1 . The transitions from the IR pumped OH stretching vibrational level appear at 32,965 cm−1 (OH01 ) and several sharp vibronic bands at higher frequency region. These sharp bands disappear with the delay time (t), while the transitions from doorway states and

9 Time-Resolved Study on Vibrational Energy Relaxation …

261

Fig. 9.3 a Transient R2PI spectra of phenol-h6 (C6 H5 OH) measured at several delay times (t) after exciting the OH stretching vibration. The electronic transition from the IR pumped OH stretching vibrational level appear as sharp vibronic bands, while the transitions from the doorway states and dense bath states appear very broad. Reprinted with permission from J. Chem. Phys. 120, 7400–7409 (2004). Copyright 2004 American Institute of Physics. b Time profile of the OH stretching vibration (upper) and that bath states (lower) for phenol-h6 (phenol-d 0 ) and phenol-d 5 . Solid line the fitted time profiles, giving τIVR = 15 ps and 80 ps phenol-h6 and phenol-d 5 , respectively. Reprinted with permission from J. Chem. Phys. 121, 11530–11534 (2004). Copyright 2004 American Institute of Physics

262

T. Ebata

dense bath levels appear very broad in the region higher than 34,000 cm−1 . The broadness is due to the overlap of many transitions from the redistributed levels. The intensity of the broad band increases with t. Figure 9.3b shows the time-profiles of the OH stretching level and the redistributed levels of phenol-h6 (phenol-d 0 ), which are compared with those of phenol-d 5 (C6 D5 OH). In both species, the transition from the OH stretch level decays exponentially and that of the redistributed levels shows single exponential rise with the same time constant of the decay. By deconvoluting the time profiles with the pump and probe laser pulse shapes, the IVR lifetime is determined to be 14 ps for phenol-h6 and 80 ps for phenol-d 5 . These results are, at first glance, very strange. Based on Fermi’s-Golden rule, the IVR rate constant (k IVR ) can be expressed as [48], kIVR =

2π |Vanh |2 ρ. 

(9.1)

Here, |Vanh |2 is the anharmonic coupling strength and ρ is the density of states of the bath modes. Since the vibrations involving the movement of H atom have higher frequencies than those of D atom, the vibrational density of states of the bath mode of phenol-h6 in the OH stretching energy region should be less than that of phenol-d 5 . Actually, the estimated ρ of phenol-d 5 (350 states/cm−1 ) having a symmetry under C s point group is three times larger than phenol-h6 (110 states/cm−1 ) at 3657 cm−1 . Thus a simple density of states description predicts an opposite result for the IVR rate constant between phenol-h6 and phenol-d 5 , if we assume a similar magnitude of V anh . The result indicates that the IVR dose not proceed with a simple “OH → bath modes”, coupling model but will proceed via some doorway state(s), that is “OH → doorway state → bath modes” as shown in Fig. 9.1. The observed difference of k IVR between phenol-h6 and phenol-d 5 , is ascribed to the difference of the coupling strength (V anh ) in the “OH → doorway state” energy redistribution route. To make it more clear about the validity of the “OH → doorway state → bath modes” for the IVR model, we investigate IVR of OD stretching vibration instead of OH. The OD stretching frequency of phenol-d 1 (C6 H5 OD) and phenol-d 6 (C6 D5 OD) is 2701 cm−1 , and ρ is estimated to be 15 and 43 states/cm−1 , respectively. Thus, the number of the coupled states is reduced. Figure 9.4 shows the transient R2PI spectra measured at several delay times after the excitation of the OD stretching vibration of phenol-d 1 by picosecond IR laser pulse. The spectra consists of sharp vibronic bands, and broad bands due to bath modes similar to phenol-h6 . However, when we take a look more detail the transient spectra, we realize that the OD01 band decays with oscillating manner, and new sharp vibronic bands (bands A and B) exhit a time-profile different from the OD band. Figure 9.5 (left) shows the time-profiles of OD01 band and band A. It is clear that both bands show quantum beats but with different phases. To reproduce the timeprofiles of these bands, we postulate the model in which two doorway states are anharmonically coupled to the OD stretch vibrational level and the OD stretch level and the doorway states further relax to the dense bath states as shown in Fig. 9.5(b). Then, the 0-th order OD stretch, |OD, and two doorway states, |lm , are coupled by

9 Time-Resolved Study on Vibrational Energy Relaxation …

263

Fig. 9.4 Transient R2PI spectra of phenol-d 1 (C6 H5 OD) measured at several delay times (t) after exciting the OD stretching vibration. The transitions from the IR pumped OD stretching vibrational level appear as sharp vibronic bands. In addition, new sharp bands, A and B, from the doorway states and broad bands due to dense bath states appeared. Reprinted with permission from J. Chem. Phys. 121, 11530–11534 (2004). Copyright 2004 American Institute of Physics

anharmonic coupling, producing the following nonstationary states, |n = αn |OD +

2 

βnlm |lm .

(9.2)

m=1

Here, we set the total number of coupled states (n) to be 3. The picosecond IR laser coherently excited the three levels, and the time-profiles of the coherently excited quasistationary states are expressed as, |ϕ(t) =

3 

αn |ne−i(En /)t .

(9.3)

n=1

In this equation, En = En − iγn /2, where En is the energy of the stationary state and γn is the width of the state, corresponding to the IVR decay rate constant to the dense bath mode. The time profile of the OD stretching state, IOD , the doorway state, Il1 , and the bath state, Ibath , are given by,

264

T. Ebata

Fig. 9.5 a Time profiles of the OD stretching vibration, doorway state (band A) and bath states for phenol-d 1 . Solid lines are the fitted time profiles, based on the coupling scheme of (b). See text. Reprinted with permission from J. Chem. Phys. 121, 11530–11534 (2004). Copyright 2004 American Institute of Physics 3 

IOD = |OD | ϕ(t)|2 = +2



|αn |4 e−(  )t γn

n=1

|αn | |αm | cos((Enm /)t)e− [(γn +γm )/2]t 2

2

(9.4)

n=m

Il1 = |l1 | ϕ(t)|2 = +2



3 

 2 γn |αn |4 βnl1  e−(  )t

n=1

αn αm βnl1 βml1

cos((Enm /)t)e− [(γn +γm )/2]t ,

(9.5)

n=m

and Ibath = 1 −

3 

|αn |2 e−(  )t , γn

(9.6)

n=1

  respectively. Here, γn = 1 − |αn |2 γ . IOD , Il1 , and Ibath are the time-profiles of the OD01 band, band A, and broad continuum, respectively. Actually, the time profile of band A has contribution of Il1 , and Ibath . The coefficients (αn ), energy spacing

9 Time-Resolved Study on Vibrational Energy Relaxation …

265

Table 9.1 Coefficients and energy spacing of the three stationary states, and bandwidths of the doorway states in the OD stretching vibration of phenol-d 1 . The values are taken from Ref. [34] βnl1

βnl2

E (cm−1 )b

n

αn

1

0.152

0.784

–

a

0.439

2

0.663

−0.465

–

a

0

3

0.424

−0.411

–

a

a These

−0.152

γ −1 (ps) 60 90 80

values could not be determined energy with respect to the n = 2 state

b Relative

(Enm ), and width (γ ) are determined by fitting thetime  profile of Eq. 9.4 to the observed one of the OD01 band, and The coefficients βnl1 are obtained by fitting the time profile of Eq. 9.5 to the observed file of band A. Obtained coefficients, energy spacing and bandwidth corresponding to IVR decay lifetime of the doorway states to dense bath states, are listed in Table 9.1. Though a similar treatment can be applied to the IVR of the OH stretch, the decay profile of the OH stretch and that of the rise of the bath mode showed a single exponential decay and rise curves, respectively, with the same time constant (lifetime) with each other. This means that a larger number of the doorway   states are coupled to the OH stretch and the IVR lifetime of the doorway state γ −1 to the dense bath state is shorter than that of OD stretch to the doorway state. In that sense, it is very interesting that the value of γ −1 of phenol-d 1 is very similar to that of the IVR lifetime of OH stretch of phenol-d 5 in spite of large difference of the density of states (ρ) between them; 350 state/cm−1 for phenol-d 5 vs 15 states/cm−1 for phenol-d 5 . This means that a simple statistical model description based on the density of bath states cannot be applied to the IVR of the OH stretching vibration of bare molecules. VER from the CH stretching vibration of phenol: Different from the OH stretching vibration, the CH stretching vibrations are not localized in one CH group but are the mixed motions of several CH groups. In addition, CH stretching vibrations frequently exhibits Fermi resonance with combination or overtone bands involving the CH bending and other vibrational modes. Actually, as seen in Fig. 9.2, roughly 20 bands appeared in the CH stretching region although phenol has only five CH groups. Such an appearance of large number of vibrational bands indicates an intense Fermi resonance in this region. We investigate how such the Fermi resonance affects IVR of the CH stretch level. The time profile of the R2PI signals after the picosecond IR pulse excitations of the CH and OH stretch levels are compared in Fig. 9.6. In case of the CH stretch excitation, IR wavenumber is fixed at 3056 cm−1 . Since the spectral bandwidth of the picosecond laser is ~5 cm−1 , the excitation causes simultaneous and partly coherent excitation of several vibrational levels. As seen in Fig. 9.6a, the decay of the CH stretch band is much faster than that of the OH stretch band. The solid curve of the CH stretch band is the convoluted curve assuming a single exponential decay with a lifetime of 5 ps. The true lifetime is thought to be shorter than 5 ps. Figure 9.6b shows the rise curves of the bath state and the convoluted

266

T. Ebata

Fig. 9.6 Comparison of the time profile of R2PI signals after the picosecond IR excitation of the OH and CH stretching levels. a Time profiles of the R2PI signals from the OH and CH stretch levels. b Time profiles of the bath states. The dashed and solid curves are the convoluted curve by assuming the IVR lifetime of 14 ps and 5 ps, respectively. Reprinted with permission from J. Chem. Phys. 120, 7400–7409 (2004). Copyright 2004 American Institute of Physics

curves by assuming the IVR lifetime of 5 ps for the CH stretch (solid curve) and 14 ps for the OH stretch (dashed curve). Thus, the IVR of the CH stretch vibration occurs much faster than the OH stretch vibration in spite of that the energy of the CH stretching vibration is ~600 cm−1 lower than that of the OH stretch. The reason for the faster IVR decay of the CH stretch than that of the OH stretch may as follows, (1) Coherent exaction of several vibrational levels by the picosecond IR pulse. (2) IVR decay of those levels to the dense bath state is very fast. To examine the above possibilities, we apply the IVR model used for the OD stretching vibration of phenol-d 1 to the CH stretching vibration. The picosecond IR laser pulse coherently excites several levels and they decay to bath mode with a lifetime of γ−1 . The result is shown in Fig. 9.7. In the figure, three cases are shown; (1) single level excitation (N = 1), (2) coherent excitation of three levels (N = 3) and (3) coherent excitation of five levels (N = 5). We assume the energy spacing  = 0.5 cm−1 . The simulation of the decay time profile is carried out for three γ−1 values (5, 20, and 100 ps). We find that the model of the coherent excitation of five levels with their lifetime γ−1 = 20 ps reasonably reproduces the observed decay profile. Thus, we conclude that in the CH stretch region, several vibrations appear via Fermi resonance and their IVR lifetime (γ−1 ) to the dense bath states is shorter than 20 ps. Coherent excitation of these levels results in very fast IVR decay with a lifetime less than 5 ps. VER of NH (ND) stretching vibration of aniline: The study is extended to the VER of the symmetric (sym) and anti-symmetric (anti-sym) stretching NH2 vibrations of

9 Time-Resolved Study on Vibrational Energy Relaxation …

267

Fig. 9.7 a–c Calculated time profiles of the CH levels after coherent excitation of single-level, three-levels and five-levels. In the inset is shown the coupling scheme,  is the energy spacing of each level and γ−1 is the IVR lifetime to the dense bath modes. d Comparison between the observed time profile of the CH01 band and simulated one using a laser pulse width of 14 ps. Reprinted with permission from J. Chem. Phys. 120, 7400–7409 (2004). Copyright 2004 American Institute of Physics

aniline [35]. Left panel of Fig. 9.8 shows (upper) the IR-UV DR and (lower) IRenhance spectra of aniline in the CH and NH2 stretch region. As seen in the figure, the frequency of sym (νs ) and anti-sym (νa ) of the HN2 stretching mode of aniline is 3423 and 3509 cm−1 , respectively. In aniline, we examine two questions; first is which vibrational mode shows faster IVR, and second is whether the vibrations involving the CH group works as the doorway state. The right panel of Fig. 9.8 shows 0 band shows a single exponential the time profiles obtained by exciting νs . The νs1 decay with a lifetime τIVR = 18 ps. On the other hand, the time constant of the rise of the intensity of broad band due to the transition from dense bath state cannot be 0 reproduced by a single exponential rise by using the same decay lifetime of the νs1 band (18 ps).   Ibath (t) = I0 1 − e−t/τI V R

(9.7)

Actually, we obtain τIVR = 5 ps when we fit the rise profile monitored at 32,714 cm−1 by using the rise curve given by Eq. 9.7. This value dose not agree with the decay lifetime (18 ps) of the NH2 stretch band. To fit the rise curves of the continuum, we need to apply following two-step relaxation model by postulating

268

T. Ebata

Fig. 9.8 (left) IR spectrum of aniline observed by (top) IR-zUV DR spectroscopy by nanosecond laser and (bottom) IR-gain spectroscopy by picosecond laser. (right) IR-UV pump-probe time profiles of a NH stretching level and b–e bath mode at several probe frequencies. Solid curves are simulated ones; doorway state, bath states, and sum of them. Reprinted with permission from J. Chem. Phys. 123, 124316–124324 (2005). Copyright 2005 American Institute of Physics

that the electronic transition of the doorway state is overlapped in the broad transient spectrum. INH2 (t) = I0 e−k1 t Idoorway (t) = I0  Ibath (t) = I0 1 +

k1 {e−k2 t − e−k1 t } k1 − k2

k2 k1 e−k1 t − e−k2 t k1 − k2 k1 − k2

(9.8) (9.9) (9.10)

Here, k 1 (= 1/τ1 ) and k 2 (= 1/τ2 ) are the rate constants for the first and second relaxation process, respectively. Different from the case of phenol-d 6 , any sharp transitions from the doorway states are not observed, and the broad continuum includes the transitions of both the doorway and the bath states. Thus, we reproduce the rise profiles shown in the right panel of Fig. 9.8 by the sum of the components of the doorway and bath states. The time constants obtained from the fittings are listed in Table 9.2. Also shown are the time constants for the deuterated species, aniline-d 5 (C6 D5 NH2 ). First, the IVR of sym-NH2 stretch is slightly faster than that of antiNH2 stretch but it is not so obvious. Thus, the symmetric property of the vibration is not so important. Second, different from phenol we do not see remarkable effect of deuterium substitution to the CH group. Thus, the vibrational modes involving the

9 Time-Resolved Study on Vibrational Energy Relaxation …

269

Table 9.2 Obtained IVR lifetimes of the NH2 stretching vibrations of aniline-d 0 and -d 5 , 2aminopyridine, and the OH stretching vibration of phenol-d 0 and -d 5 . The values are in the unit of ps. Also listed are the density of state in unit of/cm−1 Aniline-d a0 (cm−1 )

Aniline-d b5

τ1

τ2

ρ

τ1

τ2

νs (3423)

18 ± 2

20–200

185

27 ± 3

30–300

νa (3509)

34 ± 5

35–300

257

30 ± 6

30–300

Mode

2-Aminopyridinec νs (3441)

6.5

20–33

204

–

–

νa (3548)

–

d

20–33

244

–

–

Phenol-d d0 νOH (3657) a Ref.

14 ± 1

Phenol-d e5 –

110

80 ± 5

–

[35], b Ref. [37], c Ref. [39], d Ref. [34], e Ref. [32]

CH group is not the common effective doorway state in aromatic molecules. Other noticeable point is that the lifetime of the first step (τ1 = 1/k 1 ) in the IVR of the NH stretching vibration is not so different from that of OH stretching vibration of phenol. However, the second step is quite different. In case of phenol, the decay of the OH stretch level and the rise of the bath states have the same time constant, indicating the second step is faster than the second step, τ1 > τ2 , in the IVR of the OH stretch of phenol. While, IVR of the NH2 stretching vibration is opposite, τ2 > τ1 . We do not have clear explanation of the different behavior between them at present. VER of 2-aminopyridine: 2-aminopyridine has the structure in which the CH at ortho-position of aniline is replaced by a nitrogen atom. We found that IVR of the NH2 stretching vibration of 2-aminopyridine is much faster than that of aniline [39]. Upper panel of Fig. 9.9 shows the transient R2PI spectra after the picosecond IR excitation of (a) sym-NH2 stretch and (b) anti-sym-NH2 stretch of 2-aminopyridine. Due to the nature of symmetry of the NH2 stretch mode, the direct transition from the pumped level is observed only for (a) sym-NH2 stretch as seen in the figure. From the fitting of the decay profile of the pumped level, τ1 (= 1/k 1 ) is determined to be 6–7 ps. This lifetime is the shortest one determined by our picosecond laser. The lifetime is three times shorter than that of aniline sym-NH2 stretch. As seen in the lower panel of Fig. 9.9a, the time constant of the rise of the broad continuum (corresponding to bath mode) at higher than 32,500 cm−1 is at different monitoring UV frequencies, and all of them are slower than the decay of the NH2 stretch. Thus, again we have to apply the two-step IVR model similar to aniline. The time constants obtained by using this model are listed in Table 9.2. The faster lifetime of step 1 than that of aniline may be attributed to the lower symmetry of 2-aminopyridone compared to aniline. On the contrary, the lifetime of the second step is not so different between them, although τ2 has large uncertainty in aniline.

270

T. Ebata

Fig. 9.9 a (Upper) Transient R2PI spectra measured after the IR excitation of the sym-NH2 stretch vibration of 2-aminopyridine. (Lower) Time profiles of the transient R2PI signals measured at several UV frequencies. b (Upper) Transient R2PI spectra after the anti-sym-NH stretch vibration. (Lower) Time profiles of the transient R2PI signals measured at several UV frequencies. The vibrational motion of each normal mode is shown. The solid curves are ones convoluted by assuming single exponential decay. Reprinted with permission from Proc. Nat. Acad. Sci. U.S.A. 105, 12690–12695 (2008). Copyright (2008) The National Academy of Science of the USA

9 Time-Resolved Study on Vibrational Energy Relaxation …

271

9.3 IVR of the OH Stretching Vibration of Hydrogen-Bonded Clusters of Phenol It is well known that the OH stretching band shows remarkable red-shift, broadening and an increase of IR intensity upon the hydrogen (H)-bonding. The red-shift indicates the weakening of the OH bond strength as the H-bond donor, and the broadening indicates the fast VER. IVR and dissociation of phenol-H2 O cluster: We first show the study of VER of phenol-water 1:1 cluster (phenol-H2 O). Phenol-H2 O cluster has the structure in which phenolic OH forms an H-bond with the oxygen atom of H2 O as a donor, and the H2 O plane is perpendicular to the phenol plane. Figure 9.10 shows the potential energy diagram along the “phenol + H2 O” dissociation coordinate. The binding energy of the phenol-H2 O cluster is reported to be 1958 ± 38 cm−1 [49]. The OH stretching vibration of phenol-h1 -H2 O is 3525 cm−1 and that of the OD stretching vibration of phenol-d 1 -D2 O is 2600 cm−1 . Thus, the cluster will finally dissociate via vibrational predissociation (VP). In the cluster, we investigate the following problems. (1) How fast will the IVR of the phenolic OH (OD) stretching vibration be accelerated upon the H-bond formation? (2) How will the energy the OH stretching vibration be distributed to the bath mode of the cluster? (3) Can a statistical model such as RRKM be applied to the dissociation of the cluster?

Fig. 9.10 Schematic potential energy diagram of the phenol-h1 -H2 O and phenol-d 1 -D2 O clusters along the hydrogen-bond coordinate. Reprinted with permission from Chem. Phys. 419, 205–211 (2013). Copyright 2013 Elsevier

272

T. Ebata

Left pannel of Fig. 9.11 shows the transient R2PI spectra measured after the IR excitation of the H-bonded OD stretching vibration of phenol-d 1 -D2 O. Here, [phenold 1 -D2 O]+ mass is observed. The OD01 band at 33,410 cm−1 immediately disappears after the IR excitation. The broad transitions from the bath states appeared in the region higher than 34,500 cm−1 reaches to maximum intensity at a delay time of 20 ps, and diminishes at longer delay time. Right pannel shows the time profiles of R2PI signals observed at the marked uv positions. The OD01 band decays with a lifetime of 12 ps. This lifetime is more than 6 times shorter than the IVR lifetime (γ−1 ) of bare phenol-d 1 . The broad transition from the bath states shows rise with the same time constant of the decay of the OD01 band (12 ps) but the decay profiles are different at different probe uv frequencies, that is the decay lifetime becomes longer at higher uv frequencies. This is due to that in the early stage of IVR the OH stretching energy is mostly redistributed within the phenolic site (bath 1 of Fig. 9.12), so that the broad transition consists of hot bands of phenolic site vibrations having higher frequencies. With the time, the energy is redistributed to all the vibrational modes of the cluster, mainly the low frequency intramolecular vibrational modes (bath 2 of Fig. 9.12), and the electronic transitions of such the low frequency vibrations appear near the (0,0) band. Finally, the cluster dissociates via VP. Thus, the detailed energy flow from the OH stretch to the dissociation will be as follows, “OH stretch → randomization in the bath state of phenolic site → complete randomization bath state of whole the cluster → VP for fragmentation”. Such the VER route was also observed

Fig. 9.11 (left) Transient R2PI spectra of phenol-d 1 -D2 O after IR excitation of OD stretching vibration. (right) Time profiles of R2PI at different UV frequencies. Reprinted with permission from Chem. Phys. 419, 205–211 (2013). Copyright 2013 Elsevier

9 Time-Resolved Study on Vibrational Energy Relaxation …

273

Fig. 9.12 Energy level diagram and proposed energy flow scheme starting from the IR excitation of the H-bonded OD stretch to the dissociation of the H-bond of phenol-d 1 -D2 O. Bath 1 consists of the vibrations of phenolic site and bath 2 consists of the vibrations of whole the cluster. Reprinted with permission from Chem. Phys. 419, 205–211 (2013). Copyright 2013 Elsevier

in various H-bonded clusters of phenol [33, 36]. The fitting of the time profiles of Fig. 9.11 is carried out by assuming the three step model shown in Fig. 9.12. The red curve is the time profile of the bath 1, and the blue one is that of bath 2. Black curve is the sum of the two components. The obtained rate constant of each step is listed in Table 9.3. As was expected, the IVR of the first step is 4–5 times faster than the monomer, which reflects the stronger anharmonic coupling strength. We then examine whether the vibrational predissociation (VP) occurs with statistical manner, and can be estimated by RRKM theory. In the RRKM theory, the dissociation rate constant of the cluster is obtained by using following equation [50], Table 9.3 Obtained IVR(τ1 , τ2 ) and VP (τ3 ) lifetimes, and calculated dissociation lifetimes (τRRKM ) of phenol-h1 -H2 O and phenol-d 1 -D2 O, from the H-bonded OH and OD stretching vibration, respectively. The values are in the unit of ps. Data taken form ref. [42]. See Fig. 9.8 τ1

τ2

Phenol-h1 -H2 O

4

6

40

100

Phenol-d 1 -D2 O

12

24

100

330

Cluster

τ3

τRRKM

274

T. Ebata

k(E) =

1 W (E − E0 ) h ρ(E)

(9.11)

Here ρ(E) is the vibrational state densities at the phenolic OH/OD stretching energy E of the complex, E 0 is the dissociation energy of the hydrogen-bond, and W (E – E 0 ) is the total number of states of the complex at the transition state with an excess energy of E – E 0 . It is difficult to predict the transition state since the dissociation is thought to occur without a barrier. Instead, we assume that the intermolecular vibrational modes at the transition state can be replaced by the rotational motion of the fragments with a restriction of the conservation of angular momentum between the complex and the fragments. The structure and the vibrational frequencies are obtained by optimizing the structure at M05-2X/6-31++G(d, p) level, and ρ(E) is obtained by direct counting method of harmonic frequencies. The obtained rate constants are also listed in Table 9.3. The RRKM calculated lifetimes are ~3 times longer than the observed ones. This is probably insufficient estimation of ρ(E) and W (E – E 0 ) because we did not take into account the anharmonicity of the vibrations. Even if it is so, the calculated values show relatively good agreement with the observed ones, and we can conclude that the clusters dissociate after complete energy randomization. VER of phenol dimer: We next show the study of phenol dimer to investigate the difference of VER between the H-bond donor OH and acceptor OH. Left upper panel of Fig. 9.13a shows the IR spectrum of phenol dimer in the OH stretching region by IR-UV DR spectroscopy. Here, band D at 3530 cm−1 is the H-bond donor OH and band A at 3654 cm−1 is the acceptor OH stretch. Band D is 127 cm−1 red-shifted compared to free OH and its intensity is stronger and width is broader than band A, indicating IVR becomes faster by H-bonding. Left lower pannel shows the timeresolved R2PI spectra showing the ν10 (D) band at 35,200 cm−1 , and broad transitions from the bath mode, at the delay times of 10 and 200 ps after the IR excitation of band D. IVR occurs very fast in the dimer so that broad band due to the transition of the bath mode appears even at the delay time of 10 ps and the intensity becomes very weak at 200 ps. Right pannel of Fig. 9.13 shows the time-profiles of the R2PI intensities of (a) νOH (D)01 transition. In (a), the donor stretch decays with a lifetime less than 5 ps. The transition from the dense bath states rises with the same decay time constant of the IR pumped level, and decays with a lifetime of 90 ps due to vibrational predissociation (VP). (b) and (d) are the time-profiles of te dense bath states observed after exciting band A. The profile of the rise (15 ps) is slower than the case of band A excitation, but that of the decay (55 ps) due to vibrational predissociation is faster. The former indicates that the anharmonic coupling strength and the number of the coupled doorway or the dense bath states are not so different from that of bare phenol. On the other hand, the faster VP decay than that of the donor OH exciation is explained by the higher density of states along the dissociation coordinate, because the energy of the acceptor OH stretch is higher\than the donor OH stretch. Thus, similar to the phenol-water cluster, it is concluded that “OH stretch → (IVR) → dense bath state

9 Time-Resolved Study on Vibrational Energy Relaxation …

275

Fig. 9.13 (Upper left) IR-UV DR spectrum of phenol dimer, and the energy level diagram and picosecond IR-UV pump-probe excitation scheme. (Lower left) Time-resolved R2PI spectra of the νOH (D)01 transition, and transitions from the bath mode after IR exciation to the band D (donor OH). (Right) a, b and d: Time profiles of the R2PI signals after band D exciation. c and e: Time profiles of the R2PI signals after band A exciation. Reprinted with permission from J. Phys. Chem. A 105, 8623–8628 (2001). Copyright 2001 American Chemical Society

→ (VP) → fragments” is common VER process in all the H-bonded clusters of aromatic molelcules.

9.4 VER of the CH Vibration of Benzene Dimer and Trimer One of the advantages of the investigation of the molecular clusters in the gas phase is that the molecule at different site in the cluster exhibits different vibronic transitions so that we can monitor them seperately by tuning the probe UV frequency. Typical example is a benzene dimer, (benzene)2 . There have been many experimental and theorteical investigations on the structure of the benzene dimer, and it is now commonly understood that the dimer has a flexible T-shape structure [51, 52]. Figure 9.14 shows the structure of benzene dimer calculated at ωB97X-D/6-311++G(3df, 2p) level of theory [41]. The figure also shows the vector motion of the IR active CH stretching vibration (ν20 ). As seen the figure, the vibration is almost localized in each site of benzene. In bare benzene, ν20 is the only IR active mode having e1u symmetry among the six CH stretching vibrational modes.

276

T. Ebata

Fig. 9.14 Structure of benzene dimer calculated at ωB97X-D/6-311++G(3df, 2p) level of theory. The arrows indicate the normal mode of the CH stretching vibration of benzene at the stem site (left) and the top site (right), respectively. Reprinted with permission from J. Chem. Phys. 136, 044304 (2012). Copyright 2012 American Institute of Physics

Figure 9.15a shows the IR-UV DR spectrum of bare benzene (benzene-h6 ) in the region of ν20 obtained by nanosecond laser. The spectrum is shown as an inverted manner. The IR spectrum of benzene in the CH stretching region was analyzed in detail by Lee and coworkers [52], and the appearance of three bands, a, b, and c, is due to Fermi-resonance between ν20 and ν8 + ν19 /ν1 + ν6 + ν19 combination

Fig. 9.15 a IR-UV DR spectrum of benzne-h6 monomer in the CH(ν20 ) stretching region. The appearance of three bands is due to Fermi-resonance with ν8 + ν19 /ν1 + ν6 + ν19 combination levels. b, c IR-gain spectrum of hd- and hh-benzene dimer. The arrows indicate the positions of IR excitation by picosecond laser. Reprinted with permission from J. Chem. Phys. 136, 044304 (2012). Copyright 2012 American Institute of Physics

9 Time-Resolved Study on Vibrational Energy Relaxation …

277

levels, where each component is the in-plane vibration. Figure 9.15c shows the ionization-gain IR spectra of (benzene-h6 )2 , obtained by picosecond IR-UV pumpprobe spectroscopy. All the three bands of (benzene-h6 )2 , we call “hh-dimer”, are 2–4 cm−1 red-shifted from each of the three bands of bare benzene-h6 . In addition, the vibrational spectrum of (benzene-h6 ) (benzene-d 6 ), “hd-dimer”, is observed as shown in Fig. 9.15b. The spectrum of hd-dimer is almost the same with that of hhdimer. In hh-dimer, the difference of the vibrational frequency of ν20 between the stem and top site benzenes is reported to be ~1 cm−1 [53]. This means that in hh-dimer the picosecond IR laser pulse will coherently excite the CH stretching vibrations of the stem- and top-sites of benzene-h6 . In the hd-dimer, on the other hand, the IR pulse will simultaneously excite the benzene-h6 vibration of the two isomers, that is (stem)h(top)d and (top)h(stem)d. The R2PI spectrum of the S1 –S0 transition of benzene dimer was extensively studied by Schlag and coworkers [54–57]. The electronic spectrum of the dimer in the 610 region consists of the transitions of the benzenes at the top- and stem sites. The stem site benzene exhibits a sharp vibronic band, while the top site benzene exhibits a progression of ~15 cm−1 interval in the higher-wavenumber side of the sharp band of the stem-site benzene. Thus, the transient R2PI spectrum after exciting the CH stretching vibration will be overlapped by the transitions of benzenes in both sites. Figure 9.16 shows the transient R2PI spectra of hd- and hh-dimers observed at three different delay times (30 (black), 100 (red) and 300 (blue) ps) after the picosecond IR laser excitation of bands (a), (b) and (c). Also shown are the R2PI spectra in the (0,0) band region (green). Though the (0, 0) band of the S1 –S0 transition is symmetrically forbidden in the monomer, symmetry lowering in the dimer made this transition allowed. In these measurements, we monitored the [hd]+ and [hh]+ mass channels, respectively. In the transient spectra, sharp transitions are assigned to the vibronic transitions from the IR pumped CH stretch levels to 191 , 61 191 , and 62 191 levels in S1 . The appearances of these modes are due to that all the bands, (a), (b) and (c), have modes 19 and 6 characters via Fermi-resonance. These vibronic bands contain the overlapped transitions of the stem- and top-site benzenes, that is a sharp vibronic band of the stem-site benzene, and ∼15 cm−1 progression of stem-site benzene. These sharp vibronic bands disappear, while broad bands become stronger with an increase of delay time. For the measurement of the time profile of the pumped level, we chose the CH01 1910 band. As to this band, we can identify a sharp band of stem-site benzene, while 15 cm−1 progression of the top-site benzene is not resolved but exhibits a broad band feature with its peak at 50 cm−1 higher-wavenumber side of the stem-site benzene band due to low spectral resolution (5 cm−1 ) of the picosecond laser. They are marked by arrows, and the time profiles are obtained by monitoring those marked positions. Figure 9.17 show the decay time profiles of the transient R2PI signals measured after the IR pulse excitations of bands (b) and (c). All the decay curves can be fitted with single exponential decay function by deconvolution with 14 ps Gaussian pulses for IR and UV lasers. We see that the stem-site CH stretch relaxes with the lifetimes of 120–160 ps, and the top-site with the lifetimes of 370–420 ps in both hd- and hh-dimers. That is, IVR of the CH stretching vibration of benzene at the stem-site

278

T. Ebata

Fig. 9.16 Transient R2PI spectra observed at t = 30 (black), 100 (red), and 300 ps (blue) after the IR excitation of bands a, b, and c of a hd-dimer, and b hh-dimer, respectively. Green curves are R2PI spectra without IR excitation. In panel c, UV power is plotted. Arrows show the positions chosen for the measurement of the time profiles. Reprinted with permission from J. Chem. Phys. 136, 044304 (2012). Copyright 2012 American Institute of Physics

occurs 2.5–3 times faster than that at the top-site. These lifetimes are more or less the same for the bands (b) and (c) excitation. These results provides us with following conclusion on the anharmonic coupling of the CH stretching vibration of benzene dimer leading IVR. (1) The anharmonic coupling between Fermi-polyads and bath modes at the stemsite benzene is 2.5–3 times stronger than the benzene at the top-site. (2) The anharmonic coupling strength between the Fermi-polyad level and bath modes is almost the same among the three Fermi-polyads, since the decay lifetimes are almost the same with each other. (3) The excitation-exchange coupling between the two stationary states of the stemand top-site benzenes is very weak even in the hh-dimer. As to the third issue, the hh[(stem)h(top)h] dimer has two nearly degenerated vibrational levels with the energy difference of ∼1 cm−1 . A coherent excitation of these levels by a picosecond IR pulse will lead a quantum beat with roughly

9 Time-Resolved Study on Vibrational Energy Relaxation …

279

Fig. 9.17 Decay time profiles of hd- and hh-benzene dimer after the IR excitations of bands b and c, where [hd-dimer]+ and [hh-dimer]+ mass channels monitored, respectively. Two decay profiles in each panel correspond to stem-site and top-site decay. Red curves correspond to the fitted single-exponential decay function convoluted with 12 ps FWHM IR and UV Gaussian pulses. The obtained decay lifetimes are shown in each panel. Reprinted with permission from J. Chem. Phys. 136, 044304 (2012). Copyright 2012 American Institute of Physics

30 ps period in the decay profiles of the hh-dimer. The lack of the quantum beat in spite of the coherent vibrational excitation of hh-homo dimer indicates a negligible excitation-exchange coupling between the vibrations of the stem- and top-sites. This is probably due to its T-shaped structure. As seen in Fig. 9.14, the vector motion of ν20 . The vibrational motions clearly show that the stem- and top-site benzene independently vibrates. This situation is thought to be same in other in-plane vibrations. Thus, the calculation supports the negligible coupling between the vibrations of the stem and top site benzene molecules. We then discuss the time evolution of the bath states. Figure 9.18 shows the time-profile of the broad transitions from the bath states. The signals show maximum intensity at the delay time of 200–1000 ps, and gradually decay with time. Similar to the case of the decay of the H-bonded clusters of phenol, the time profiles are different at different monitoring uv frequencies. So, we analyze the decay profiles by using two bath-modes model shown in Fig. 9.19. Here, bath 1 consists of the intramolecular vibrations of monomer and the bath 2 consists the low frequency intermolecular vibrations of the dimer. Obtained lifetime of each step is listed in Table 9.4. Though, an experimental evidence is not shown here, we confirmed that the time constant of the rise of the benzene fragment is ~4000 ps, which agrees well with the value of τVP in the table.

280

T. Ebata

Fig. 9.18 Time evolutions of bath mode observed at three different UV frequencies after the IR excitation of bands a and b. These profiles were observed by monitoring a hd + and b hh+ mass channels. Red curves are fitted by using a similar model of Fig. 9.19, with τ2 (= 1/k 2 ) = 400 and τvp (= 1/k vp ) = 4000 ps. The τ1 (= 1/k 1 ) value used is 140 ps for band b. For τ1 of band a, we adopted 150 and 130 ps for the hd and hh dimers, respectively. Blue and green curves represent bath 1 and bath 2, respectively. Reprinted with permission from J. Chem. Phys. 136, 044304 (2012). Copyright 2012 American Institute of Physics

It is very interesting that IVR is very slow in benzene dimer. It should be remind that in bare phenol the IVR of the CH stretch occurs less than 5 ps, while that in benzene dimer is ~140 and ~360 ps, depending on the site. Thus, even in the dimer, there is not effective doorway states or bath mode to promote IVR of the CH stretch. We extended the study to benzene trimer, (benzene)3 , to examine the size effect. An increase of the size may reduce the symmetry of the cluster, and increase the density of states of the bath mode. The structure of benzene trimer has been investiagted by several groups [58–60]. Engkvist et al. proposed a cyclic structure is the most stable one by using nonempirical model potential calibrated from the CCSD(T) benzene dimer energies [58]. However, more recent calculation by Tauer et al. suggested the energies of other isomers are within a few kJ/mol [60]. Experimentally, Iimori et al. re-examined the electronic spectrum of benzene trimer and its deuterated species by R2PI and UV-UV HB spectroscopy. They concluded that the experimental results supports the cyclic structure [61]. In the experiment of benzene trimer, we have a detection problem. That is, it is reported that the trimer dissociates after the ionization into dimer cation and monomer [61]. So, we have to measure the dimer cation fragment to study VER of the trimer. However, the dimer cation signal is also generated by the ionization of dimer coexisting in the jet. The ion signal will

9 Time-Resolved Study on Vibrational Energy Relaxation …

281

Fig. 9.19 Schematic energy level diagram in VER of benzene dimer and a pump-probe excitation scheme. Energies of S1 , and IP for benzene and benzene dimer as well as the dissociation energy in neutral and ionic states are shown in cm−1 unit. These values are taken from Reprinted with permission from J. Chem. Phys. 136, 044304 (2012). Copyright 2012 American Institute of Physics Table 9.4 Time constants (ps) on vibrational energy relaxation of benzene dimer and trimer after the IR excitation of Fermi-polyad (band a, b, and c) of the CH stretching vibration [41] Dimer

Trimer

Band a hd τ1

Stem

–

a

Top

–

a

Band b

Band c

hd

hh

hd

hh

hdd b

–

a

140 ± 10

120 ± 10

160 ± 10

140 ± 10

50 ± 10

–

a

370 ± 20

360 ± 20

400 ± 20

420 ± 30

hh

τ2

200–700

300

τVP

2000–6000

1000

a Decay

profile could not be obtained from time profiles observed by monitoring dd + signal

b Determined

282

T. Ebata

interfere with the detection of fragment dimer ion coming from the timer ionization. So, instead of investigating (benzene-h6 )3 , we chose [(benzene-h6 ) (benzene-d 6 )2 ] hetero-trimer. Here we abbreviate (benzene-h6 )3 , [(benzene-h6 ) (benzene-d 6 )2 ], and [(benzene-h6 )2 (benzene-d 6 )] to (hhh), (hhd), and (hdd), respectively. Among them, only (hdd) gives [dd]+ mass signal in the picosecond IR-UV pump-probe study, so we chose this cluster for the investigation. Figure 9.20 shows the time profile of hdd-trimer after the picosecond IR pulse exaction of the Fermi-polyad bands. Obtained time constants for the second and third steps of the relaxation are listed also in Table 9.4. The IVR decay lifetime of the pumped revel is 50 ps, and that of VP is 1000 ps. Both lifetimes are shorter than the corresponding values of the dimer. The faster IVR decay of the trimer than the dimer is attributed to the larger anharmonic coupling constant of the trimer, while the faster VP of benzene trimer can be described by statistical (RRKM) theory as is the case of phenol-H2 O complex described above. The VP rate constant k(E) given by Eq. 9.11 is proportional to the ratio between the total number of possible internal states of the fragment, W(E – E 0 ), at the available energy and the density of states of the parent cluster, ρ(E). First ρ(E) of the T-shaped dimer is larger than that of cyclictrimer, because Felker and coworkers showed that the dimer has the intermolecular vibrations with the frequencies lower than the trimer [62]. On the other hand, W(E − E 0 ) of the [(benzene)2 + benzene] fragments from the trimer will be larger than

Fig. 9.20 Time evolution of the IR pumped level and the broad continuum attributed to bath mode of hdd-benzene trimer. a Decay profile of the CH stretch level. b, c Time evolutions of bath mode observed at two different UV frequencies after the IR excitation of bands a and b. Blue and green curves represent bath 1 and bath 2, respectively. Reprinted with permission from J. Chem. Phys. 136, 044304 (2012). Copyright 2012 American Institute of Physics

9 Time-Resolved Study on Vibrational Energy Relaxation …

283

that of the (benzene + benzene) fragments from the dimer, when one assumes that the dissociation energy is not so different between them. Thus, The lower ρ(E) and larger W(E – E 0 ) of trimer than dimer will result in faster VP rate constant of trimer, which is in good agreement with the observed tendency shown in Table 9.4.

9.5 Summary In this chapter, we investigated Vibrational Energy Relaxation (VER) from high frequency vibrations (3000–3700 cm−1 ), such as OH, NH and CH stretching vibrations, of aromatic molecules and their clusters in the gas-phase. In both the bare molecules and clusters, VER occurs by stepwise or hierarchical manner. Initial step is intramolecular vibrational energy redistribution (IVR) occurring within some limited vibrational levels consisted of combination or overtone vibrational levels, which we call doorway states. The doorway states are not the special modes and are different in different molecules. In case of phenol, the vibrational modes involving the CH groups are the doorway state of the first IVR step of the OH stretch. On the other hand, such the doorway state is not clearly identified for the NH2 stretching vibration of aniline. The time scale of the initial step is 6–35 ps for OH and NH stretching vibrations. VER of the CH stretching vibration is a little bit different from those vibrations. The CH stretching vibration is not localized in a specific CH group, different from OH and NH stretch modes, but is a coupled motions of several CH groups. In addition, CH stretch is anharmonically coupled with overtone or combination bands involving the CH bending vibration via Fermi-resonance. These factors lead very fast IVR decay of the CH stretch mode as shown in phenol, where IVR lifetime is shorter than 5 ps. On the other hand, in benzene the IVR of the CH stretch is slow even in the dimer, 100–400 ps depending the site of the dimer. Such a very slow IVR decay of benzene dimer is ascribed by that high symmetry of benzene restricts only limited number of vibrational levels which can anharmonically couple to the CH stretch. The energy is further dissipated to many low frequency bath modes. This stepwise process is common in bare molecules and clusters. In the cluster, the energy is further redistributed to the intermolecular modes and the clusters finally dissociates by vibrational predissociation (VP). The time scale of VP is in the order of 100–1000 ps, and such the long time scale leads the dissociation of the clsuter to occur by statistical manner.

References 1. Laubereau, A., Kaiser, W.: Vibrational dynamics of liquids and solids investigated by picosecond light pulses. Rev. Mod. Phys. 50, 607–666 (1978) 2. Faltermeier, B., Protz, R., Maier, M.: Concentration and temperature dependence of electronic and vibrational energy relaxation of O2 in liquid mixtures. Chem. Phys. 62, 377–385 (1981)

284

T. Ebata

3. Heilweil, E.J., Doany, F.E., Moore, R., Hochstrasser, R.M.: Vibrational energy relaxation of the cyanide ion in aqueous solution. J. Chem. Phys. 76, 5632–5634 (1982) 4. Harris, A.L., Brown, J.K., Harris, C.B.: The nature of simple photodissociation reactions in liquids on ultrafast time scales. Annu. Rev. Phys. Chem. 39, 341 (1988) 5. Kliner, D.A.V., Alfano, J.C., Barbara, P.F.: Photodissociation and vibrational relaxation of I2 − in ethanol. J. Chem. Phys. 98, 5375–5389 (1993) 6. Kühne, T.: Vibrational relaxation and geminate recombination in the femtosecondphotodissociation of triiodide in solution. J. Chem. Phys. 105, 10788–10802 (1996) 7. Egorov, S.A., Skinner, J.L.: A theory of vibrational energy relaxation in liquids. J. Chem. Phys. 105, 7058–7074 (1996) 8. Tokmakoff, A., Sauter, B., Fayer, M.D.: Temperature-dependent vibrational relaxation in polyatomic liquids: Picosecond infrared pump–probe experiments. J. Chem. Phys. 100, 9035–9043 (1994) 9. Hamm, P., Lim, M., Hochstrasser, R.M.: Vibrational energy relaxation of the cyanide ion in water. J. Chem. Phys. 107, 10523–10531 (1997) 10. Schultz, S.L., Qian, J., Jean, J.M.: Separability of intra- and intermolecular vibrational relaxation processes in the S1 state of trans-stilbene. J. Phys. Chem. A 101, 1000–1006 (1997) 11. Bingemann, D., King, A.M., Crim, F.F.: Transient electronic absorption of vibrationally excited CH2 I2 : watching energy flow in solution. J. Chem. Phys. 113, 5018–5025 (2000) 12. Levinger, N.E., Davis, P.H., Fayer, M.D.: Vibrational relaxation of the free terminal hydroxyl stretch in methanol oligomers: indirect pathway to hydrogen bond breaking. J. Chem. Phys. 115, 9352–9360 (2001) 13. Charvat, A., Jens Aβmann, J., Bernd Abel, B.: Real-time probing of intramolecular vibrational energy redistribution and intermolecular vibrational energy transfer of selectively excited CH2 I2 molecules in solution. J. Phys. Chem. A 105, 5071–5080 (2001) 14. Gaffney, K.J., Piletic, I.R., Fayer, M.D.: Hydrogen bond breaking and reformation in alcohol oligomers following vibrational relaxation of a non-hydrogen-bond donating hydroxyl stretch. J. Phys. Chem. A 106, 9428–9435 (2002) 15. Schwarzer, D., Hanisch, C., Kutne, P., Troe, J.: Vibrational energy transfer in highly excited bridged azulene-rryl compounds: direct observation of energy flow through aliphatic chains and into the solvent. J. Phys. Chem. A 106, 8019–8028 (2002) 16. Sekiguchi, K., Shimojima, A., Kajimoto, O.: Intramolecular and intermolecular vibrational energy relaxation of CH2 I2 dissolved in supercritical fluid. Chem. Phys. Lett. 356, 84–90 (2002) 17. Lock, A.J., Cilljamse, J.J., Woutersen, S., Bakker, H.J.: Vibrational relaxation and coupling of two OH-stretch oscillators with an intramolecular hydrogen bond. J. Chem. Phys. 120, 2351–2358 (2004) 18. Yamaguchi, S., Hirai, S., Ikeshita, K., Banno, M., Ohta, K., Tominaga, K.: Vibrational dynamics of hydrogen-bonded complexes studied by infrared pump-probe spectroscopy. Rev. Laser Eng. Suppl. 36, 1024–1027 (2008) 19. Tesar, S.L., Kasyanenko, V.M., Rubtsov, I.V., Rubtsov, G.I., Burin, A.L.: Theoretical study of internal vibrational relaxation and energy transport in polyatomic molecules. J. Phys. Chem. A 117, 315–323 (2013) 20. Grubb, M.P., Coulter, M.P., Marroux, H.J.B., Orr-Ewing, A.J., Ashfold, M.N.R.: Unravelling the mechanisms of vibrational relaxation in solution. Chem. Sci. 8, 3062–3069 (2017) 21. Heilweil, E.J., Casassa, M.P., Cavanagh, R.R., Stephenson, J.C.: Vibrational energy decay of surface adsorbates. In: Laubereau, A., et al. (eds.) Time-Resolved Vibrational Spectroscopy. Springer Proceedings in Physics, pp 71–75. Springer, Berlin (1985) 22. Hirose, C., Goto., Y., Akamatsu, N., Kondo, J., Domen, K.: Measurement of vibrational energy relaxation of a surface hydroxyl group by ultrashort tunable infrared pulses. Surf. Sci. 283, 244–247 (1993) 23. Heilweil, E.J.: Vibrational population lifetimes of OH(v = 1) in natural crystalline micas. Chem. Phys. Lett. 129, 48–54 (1986)

9 Time-Resolved Study on Vibrational Energy Relaxation …

285

24. Stewart, G., Ruoff, R., Kulp, T., NcDonlad, J.D.: Intramolecular vibrational relaxation in dimethyl ether. J. Chem. Phys. 80, 5353–5358 (1984) 25. Parmenter, C.S., Stone, B.M.: The methyl rotor as an accelerating functional group for IVR. J. Chem. Phys. 84, 4710–4711 (1986) 26. Riedle, E., Neusser, H.J., Schlag, E.W., Lin, S.H.: Intramolecular vibrational relaxation of benzene. J. Phys. Chem. 88, 198–202 (1984) 27. Bondybey, V.E.: Relaxation and vibrational energy redistribution processes in polyatomic molecules. Ann. Rev. Phys. Chem. 35, 591–612 (1984) 28. Gascooke, J.R., Virgo, E.A., Lawrance, W.D.: Torsion-vibration coupling in S1 toluene: implications for IVR, the torsional barrier height, and rotational constants. J. Chem. Phys. 143, 044313 (13 pp.) (2015) 29. Borst, D.R., Pratt, D.W.: Toluene: structure, dynamics, and barrier to methyl group rotation in its electronically excited state. A route to IVR. J. Chem. Phys. 113, 3658–3669 (2000) 30. Kushnarenko, A., Miloglyadov, E., Quack, M., Seyfang, G.: Intramolecular vibrational energy redistribution in HCCCH2 X (X = Cl, Br, I) measured by femtosecond pump–probe experiments in a hollow waveguide. Phys. Chem. Chem. Phys. 20, 10949–10959 (2018) 31. Ebata, T., Kayano, M., Sato, S., Mikami, N.: Picosecond IR-UV pump-probe spectroscopy. IVR of OH stretching vibration of phenol and phenol dimer. J. Phys. Chem. A 105, 8623–8628 (2001) 32. Yamada, Y., Ebata, T., Kayano, M., Mikami, N.: Picosecond IR-UV pump-probe spectroscopic study on the dynamics of vibrational relaxation of jet-cooled phenol I. IVR of the OH and CH stretching vibrations of bare phenol. J. Chem. Phys. 120, 7400–7409 (2004) 33. Kayano, M., Ebata, T., Yamada, Y., Mikami, N.: Picosecond IR-UV pump-probe spectroscopic study on the dynamics of vibrational relaxation of jet-cooled phenol II. IVR of the OH stretching vibration of hydrogen-bonded clusters. J. Chem. Phys. 120, 7410–7417 (2004) 34. Yamada, Y., Mikami, N., Ebata, T.: Real time detection of doorway states in the intramolecular vibrational energy redistribution of the OH/OD stretching vibration of phenol. J. Chem. Phys. 121, 11530–11534 (2004) 35. Yamada, Y., Okanao, J., Mikami, N., Ebata, T.: Picosecond IR-UV pump-probe spectroscopic study on the intramolecular vibrational energy redistribution of NH2 and CH stretching vibrations of jet-cooled aniline. J. Chem. Phys. 123, 124316–124324 (2005) 36. Yamada, Y., Kayano, M., Mikami, N., Ebata, T.: Picosecond IR-UV pump-probe study on the vibrational relaxation of phenol-ethylene hydrogen-bonded cluster: difference of relaxation route/rate between the donor and the acceptor site excitations. J. Phys. Chem. A 110, 6250–6255 (2006) 37. Yamada, Y., Okano, J., Mikami, N., Ebata, T.: Picosecond time-resolved study on the intramolecular vibrational energy redistribution of NH stretching vibration of jet-cooled aniline and its isotopomer. Chem. Phys. Lett. 432, 421–425 (2006) 38. Yamada, Y., Katsumoto, Y., Ebata, T.: Picosecond IR-UV pump-probe spectroscopic study on the vibrational energy flow in isolated molecules and clusters. Phys. Chem. Chem. Phys. 9, 1170–1185 (2007) 39. Yamada, Y., Mikami, N., Ebata, T.: Relaxation dynamics of NH stretching vibrations of 2-aminopyridine and its dimer in a supersonic beam. Proc. Nat. Acad. Sci. U.S.A. 105, 12690–12695 (2008) 40. Kusaka, R., Ebata, T.: Remarkable site difference of vibrational energy relaxation in benzene dimer: picosecond time-resolved IR-UV pump-probe spectroscopic study. Angew. Chemie Int. Ed. 122, 7143–7146 (2010) 41. Kusaka, R., Inokuchi, Y., Ebata, T.: Vibrational energy relaxation of benzene dimer and trimer in the CH stretching region studied by picosecond time-resolved IR-UV pump-probe spectroscopy. J. Chem. Phys. 136, 044304 (8 pp.) (2012) 42. Miyazaki, Y., Inokuchi, Y., Ebata, T., Petkovi´c, M.: Study on vibrational relaxation dynamics of phenol-water complex by picosecond time-resolved IR-UV pump-probe spectroscopy in a supersonic molecular beam. Chem. Phys. 419, 205–211 (2013)

286

T. Ebata

43. Pimentel, G.C., McClellan, A.L.: The Hydrogen Bond. W. H. Freeman and Company, San Francisco and London (1960) 44. Colthup, N.B., Daly, L.H., Wiberley, S.E.: Introduction to infrared and Raman spectroscopy, 3rd ed. Academic Press. Inc. Hartcourt Brace & Company, Publishers, Boston, San Diego, New York, London, Sydney, Tokyo, Toronto (1990) 45. Bist, H.D., Brand, J.C.D., Williams, D.R.: The vibrational spectrum and torsion of phenol. J. Mol. Spectrosc. 24, 402–412 (1967) 46. Watanabe, T., Ebata, T., Tanabe, S., Mikami, N.: Size-selected vibrational spectra of phenol(H2 O)n (n = 1–4) clusters observed by IR-UV double resonance and stimulated Raman-UV double resonance spectroscopies. J. Chem. Phys. 105, 408–419 (1996) 47. Honda, M., Fujii, A., Fujimaki, E., Ebata, T., Mikami, N.: NH stretching vibrations of jetcooled aniline and its derivatives in the neutral and cationic ground states. J. Phys. Chem. A 107, 3678–3686 (2003) 48. Dirac, P.A.M.: The quantum theory of emission and absorption of radiation. Proc. Roy. Soc. (London) A 114, 243–265 (1927) 49. Courty, A., Mons, M., Dimicoli, I., Piuzzi, F., Brenner, V., Millie, P.: Ionization, energetics, and geometry of the phenol-S complexes (S = H2 O, CH3 OH, and CH3 OCH3 ). J. Phys. Chem. A 102, 4890–4898 (1998) 50. Holbrook, K.A., Pilling, M.J., Robertson, S.H.: Unimolecular Reaction, 2nd ed. Wiley (1996) 51. van der Avoird, A., Podeszwa, R., Szalewicz, K., Leforestier, C., van Harrevelt, R., Bunker, P.R., Schnell, M., von Helden, G., Meijer, G.: Vibration–rotation-tunneling states of the benzene dimer: an ab initio study. Phys. Chem. Chem. Phys. 12, 8219–8240 (2010) 52. Page, R.H., Shen, Y.R., Lee, Y.Y.: Local modes of benzene and benzene dimer, studied by infrared–ultraviolet double resonance in a supersonic beam. J. Chem. Phys. 88, 4621–4636 (1988) 53. Erlekam, U., Frankowski, M., Meijer, G., von Helden, G.: An experimental value for the B1u C–H stretch mode in benzene. J. Chem. Phys. 124, 171101 (5 pp.) (2006) 54. Börnsen, K.O., Selzle, H.L., Schlag, E.W.: The interactions in the benzene dimer in a supersonic jet: study of the S1 level with isotopic labeling. Z. Naturforsch 39a, 1255–1258 (1984) 55. Börnsen, K.O., Selzle, H.L., Schlag, E.W.: Spectra of isotopically mixed benzene dimers: details on the interaction in the vdW bond. J. Chem. Phys. 85, 1726–1732 (1986) 56. Kiermeier, A., Ernstberger, B., Neusser, H.J., Schlag, E.W.: Benzene clusters in a supersonic beam. Multiphoton ionization, mass analysis and dissociation kinetics. Z. Phys. D At. Mol. Clusters 10, 311–317 (1988) 57. Scherzer, W., Krätzschmar, O., Selzle, H.L., Schlag, E.W.: Structural isomers of the benzene dimer from mass selective hole-burning spectroscopy. Z. Naturforsch 47a, 1248–1252 (1992) 58. Engkvist, O., Hobza, P., Selzle, H.L., Schlag, E.W.: Benzene trimer and benzene tetramer: structures and properties determined by the nonempirical model (NEMO) potential calibrated from the CCSD (T) benzene dimer energies. J. Chem. Phys. 110, 5758–5762 (1999) 59. Ye, X.Y., Li, Z.H., Wang, W.N., Fan, K.N., Xu, W., Hua, Z.Y.: The parallel π–π stacking: a model study with MP2 and DFT methods. Chem. Phys. Lett. 397, 56–61 (2004) 60. Tauer, T.P., Sherrill, C.D.: Beyond the benzene dimer: an investigation of the additivity of interactions. J. Phys. Chem. A 109, 10475–10478 (2005) 61. Iimori, T., Aoki, Y., Ohshima, Y.: S1 -S0 vibronic spectra of benzene clusters revisited. II. The trimer. J. Chem. Phys. 117, 3675–3686 (2002) 62. Schaeffer, M.W., Maxton, P.M., Felker, P.M.: The size dependence of ground-state collective vibrational modes in molecular clusters. Benzene dimer through pentamer. Chem. Phys. Lett. 224, 544–550 (1994)

Chapter 10

Non-adiabatic Dynamics of Molecules Studied Using Vacuum-Ultraviolet Ultrafast Photoelectron Spectroscopy Shunsuke Adachi and Toshinori Suzuki

Helvetica

The ultrafast photodynamics are followed entirely till the formation of the end products using timeresolved photoelectron spectroscopy with probe photon energies greater than the electron binding energies (eBEs) of transient states and the products

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_10

287

288

S. Adachi and T. Suzuki

Abstract Quantitative prediction of branching ratios and photoproduct quantum yields when an excited electronic state has multiple conical intersections (CIs) with other electronic states remains a challenging problem for theoretical chemists. Experimental benchmarking of these observables is thus highly desirable for evaluating the appropriateness of various approximations made in theoretical calculations. In this chapter, we present time-resolved photoelectron spectroscopy (TRPES) using a vacuum-ultraviolet (VUV) laser as a means to observe ultrafast dynamics through CIs in real time. We describe the details of our apparatus, and discuss some representative examples from our recent studies. In the UV photochemistry of furan following ππ* photoexcitation, the ring-puckering CI dominates the dynamics, safely returning more than 90% of the excited molecules to the original ground state. The remaining 10% undergo irreversible isomerization after passing through the puckering CI. In the ultrafast photodissociation of nitromethane, we find that ππ* electronic excitation leads to ultrafast cascading internal conversion (S3 → S2 → S1 + S0 ) prior to dissociation into CH3 + NO2 fragments, and that the dissociation predominantly proceeds on the S1 surface, leading to comparable production of NO2 (A) and NO2 (X). Keywords Time-resolved photoelectron spectroscopy · Vacuum-ultraviolet · Reaction dynamics · Conical intersection

10.1 Introduction Ultrafast photo-induced dynamics of polyatomic molecules frequently involve non-adiabatic transitions through conical intersections (CIs) [1]. Although recent advances in computational chemistry have enabled the identification of CI geometries and the prediction of possible non-radiative deactivation pathways [2–5], the excited electronic states of polyatomic molecules often have multiple CIs at different geometries, which facilitate non-adiabatic transitions to different electronic states and ultimately lead to different products [3]. In such situations, it is difficult to quantitatively predict these branching ratios and photoproduct quantum yields (QYs). For example, ring-puckering and ring-opening CIs play significant roles in photophysics and photochemistry of heterocycles. The former mediates ultrafast internal conversion (IC) to the original ground state, and is often associated with the intrinsic photostability of the DNA bases (nitrogen heterocycles) [2, 6–8], while the latter leads to permanent photodamage with high probability [9]. Theoretical calculations S. Adachi · T. Suzuki (B) Department of Chemistry, Graduate School of Science, Kyoto University, Kyoto, Japan e-mail: [email protected] S. Adachi e-mail: [email protected]

10 Non-adiabatic Dynamics of Molecules Studied …

289

on the UV photochemistry of furan (a five-membered heterocycle) are controversial with regard to the main relaxation pathway as well as the photoproducts and their QYs [2, 8]. Experimentally benchmarking the dynamics is thus highly desirable for quantifying the appropriateness of various approximations made in the theoretical calculations. Time-resolved photoelectron spectroscopy (TRPES) can serve as a powerful tool for investigating non-radiative dynamics and photochemistry. A schematic energy diagram illustrating TRPES is shown in Fig. 10.1. A molecule is first excited by a pump pulse, and the subsequent dynamics are interrogated through photoionization by a time-delayed probe pulse. The photoelectron kinetic energy (PKE) distribution is measured as a function of the delay between the pump and probe pulses, and the electron binding energy (eBE) is given by the difference between the probe photon energy and the measured PKE. Upon photoexcitation, a molecular wavepacket is created in the Franck-Condon (FC) region of the excited state, and it starts moving in the direction of descending potential energy (arrow A in Fig. 10.1) on the excitedstate potential energy surface (PES), which generally increases the eBE [2, 8, 10]. [Rigorously speaking, eBE depends on both the lower (neutral) and upper (cationic) state potential energies, and it varies with the transient molecular structure in a complex manner.] After the wavepacket passes through the CI region, it transits to the product ground state (arrow B) or returns to the reactant ground state (arrow B ), and eBE becomes even higher in either case. It is impossible to follow the ultrafast dynamics to these end products when the probe photon energy does not exceed their eBEs. Since most organic molecules have an eBE of about 10 eV, vacuum-UV (VUV) pulses are required to detect IC to the ground electronic state. High-harmonic generation (HHG) using intense Ti:sapphire lasers is a wellestablished approach to access the VUV and soft X-ray regions [12], and HHG sources have been used for photoionization in ultrafast spectroscopy [13]. Since an HHG spectrum is composed of many odd-order components, a time-compensating or time-preserving monochromator [14] is widely used to select a single harmonic

Fig. 10.1 Schematic diagram of TRPES. It is impossible to follow ultrafast dynamics to the final end products when the probe photon energy does not exceed their eBEs. Modified from [11] Copyright 2018, The Authors licensed under CC BY 4.0

290

S. Adachi and T. Suzuki

out of the HHG spectrum. As an alternative, we developed a laser system to generate single-order VUV radiation at 90 nm (hv = 14 eV) without the need for a monochromator [15], and integrated it with a TRPES apparatus [16–22]. The remainder of this chapter is structured as follows. In Sect. 10.2, we describe the experimental details of the VUV TRPES apparatus. In Sects. 10.3 and 10.4, we show representative results from our recent VUV TRPES studies. Finally, in Sect. 10.5 we present our conclusions.

10.2 Experimental Figure 10.2 schematically illustrates VUV TRPES using 205-nm pump and 90-nm probe pulses. Output pulses from a Ti:sapphire amplifier (810 nm, 35 fs, 12 mJ at 1 kHz) are split by an uncoated quartz plate. The reflected beam (~3%) is used for fourth-harmonic generation (FHG, 205 nm) and the transmitted beam (~97%) is sent to a third-harmonic generation (THG, 270 nm) stage. The single-pass THG stage is composed of a second-harmonic generation (SHG) crystal [β-BaB2 O4 (BBO)], a delay-compensation plate (α-BBO), a quartz waveplate for polarization control, and a THG crystal (β-BBO) [15]. The 270-nm pulses are gently focused onto a gas (Ar, Kr, and Xe) flow cell in a vacuum chamber to drive HHG. While the ninth harmonic (= 3rd × 3rd, 90 nm) is the lowest order obtained from the 270-nm driving laser and thus is generated most efficiently, higher-order harmonics [15th (= 3rd × 5th, 54 nm), 21st (= 3rd × 7th, 39 nm), etc.] are also generated. Since the ninth harmonic is well separated in energy (>9 eV) from other higher-order harmonics, it can be isolated quite easily using aluminum mirrors, as described below. Among the three target gas species, Kr has provided the highest conversion efficiency for ninth harmonic generation [15]. The ninth harmonic was sent to another vacuum chamber through an aperture for differential pumping. To separate the ninth harmonic from the 270-nm driving laser, the beam was reflected twice by silicon carbide (SiC) plates [23]. The incident angle

Fig. 10.2 VUV TRPES apparatus. THG: third-harmonic generation; FHG: fourth-harmonic generation; AP: apertures for differential pumping; SiC: silicon carbide plates; Al: aluminum mirrors; TOF: time-of-flight electron energy analyzer. Modified from [11] Copyright 2018, The Authors licensed under CC BY 4.0

10 Non-adiabatic Dynamics of Molecules Studied …

291

to the SiC plates was set at 71.8° (Brewster’s angle), at which the reflectivity at 270 nm is minimized. The calculated reflectivity values at 270 and 90 nm per bounce are 90% of the excited molecules safely returns to the original ground state geometry, without undergoing irreversible conversion into photoproducts. So far we have shown that the puckering CI facilitates efficient ultrafast deactivation to the original ground state. The obtained reaction timescales and branching ratios are summarized in Fig. 10.4. Now we move on to a discussion of the products. Figure 10.8a (solid curve) shows a TRPES spectrum averaged for longer delays (700–1500 fs). The averaged spectrum contains positive bands attributed to photoproducts and negative bands attributed to ground-state bleach of furan. Thus, the product spectrum (Fig. 10.8b) is obtained by subtracting the one-color spectrum of furan (dashed curve in Fig. 10.8a normalized at eBE < 9 eV) from the averaged TRPES spectrum. For the sake of comparison, calculated vertical ionization energies for possible photoproducts via the puckering CI (oxabicyclopentene and cyclopropenecarbaldehyde, see Fig. 10.9a for their molecular structures) and the ring-opening CI [1,3-, 1,2-, and 2,3-butadienals, see Fig. 10.9b] are presented in Fig. 10.8c, and their expected band positions are shown as colored rectangles. The experimental product spectrum (Fig. 10.8b) is reasonably consistent with the possible photoproducts from the puckering pathway (oxabicyclopentene and cyclopropene-carbaldehyde). Oesterling et al. stressed the importance of ring-opening trajectories [8], arguing that a good 50% of the trajectories that pass near the ring-opening CI remain on the excited-state surface while the ring opens completely. However, we did not observe any signature of ring-opening trajectories in the excited-state TRPES spectra (Fig. 10.6a) or in the product spectrum (Fig. 10.8b). Presumably, as the authors

10 Non-adiabatic Dynamics of Molecules Studied …

297

Fig. 10.8 a TRPES spectrum averaged for long delays (700–1500 fs, solid line) and one-color spectrum of furan (dashed line). b Product spectrum. c Calculated vertical ionization energies for possible photoproducts via the puckering and ring-opening pathways. Expected band positions of photoproducts via the two pathways are shown as colored rectangles. Adapted from [22] with permission from the PCCP Owner Societies Fig. 10.9 Molecular structures of furan isomers potentially formed after passing through the a puckering and b ring-opening CIs. Adapted from [22] with permission from the PCCP Owner Societies

298

S. Adachi and T. Suzuki

described [8], their complete active space self-consistent field calculations overestimated the QY for ring-opening due to a vanishing energy barrier along the pathway. In our dynamical simulations, oxabicyclopentene and cyclopropene-carbaldehyde can be formed in a few picoseconds via ground-state reactions that occur after passing through the puckering CI [22]. Although there exists fairly large uncertainty in the theoretically predicted photoproduct QY (η = 0.03–0.3), it appears qualitatively consistent with the experimental QY (η = 0.09).

10.4 Nitromethane The UV photoabsorption spectrum of nitromethane (NM, [CH3 NO2 ]) has two maxima: a weak band at ~270 nm and a strong band at ~200 nm, corresponding to the nπ* and ππ* transitions, respectively. The primary photochemical reaction following the ππ* excitation has been demonstrated to be C−N bond cleavage (CH3 NO2 + hυ → CH3 + NO2 ) [27–31]. The high available energy of this reaction and the low-lying excited states (A2 B2 , B2 B1 , and C2 A2 ) of NO2 enable formation of the NO2 product in both the ground and excited states [32–34]. The electronic structure of NM, as obtained from quantum chemistry calculations [30, 32–34], is illustrated in Fig. 10.10. The ππ* absorption corresponds to the S3 ← S0 transition. Isegawa et al. [34] carried out a theoretical reaction path analysis and argued that NM undergoes the following cascaded IC: S3 → S2 → S1 → S0 (Pathway A). Interestingly, the S3 , S2 , S1 , and S0 states of NM are predicted to dissociate into different electronic states of NO2 [30, 33, 34] (Pathways B–E) based on the state correlations between NM and its photoproducts. Thus, identification of the electronic state of the NO2 prodFig. 10.10 Ultrafast cascading photochemical reactions of NM following ππ* electronic excitation. The timescales and branching ratios were obtained in the present study. Adapted with permission from [21]. Copyright 2018 American Chemical Society

10 Non-adiabatic Dynamics of Molecules Studied …

299

uct is important in studies on NM photodissociation dynamics. Although product kinetic energy distributions have indicated that NO2 is formed with a high amount of internal energy [27, 30, 31, 35, 36], it remains unclear whether the internal energy is predominantly electronic or vibrational. Moreover, the reaction timescale has never been measured. TRPES spectra of NM in (a) low and (b) high eBE regions are shown in Fig. 10.11. The photoelectron signals in Fig. 10.11a arose from NM in S3 . The eBE of S3 is estimated to be roughly 5.3 eV based on the eBE of S0 (11.3 eV) and the pump photon energy (hvpump = 6.0 eV), and this value is in close accord with the leading edge of the TRPES spectrum (see Fig. 10.11a inset). In contrast to the excited-state TRPES spectra of furan (Fig. 10.6a), no gradual peak shift toward higher eBE— which would have been an indication of excited-state wavepacket motion—is evident in Fig. 10.11a (see dashed vertical line for eye guide). This indicates that the cascaded IC (S3 → S2 → S1 → S0 ) occurs in considerably shorter time than the experimental time resolution, and is presumably associated with only a slight geometrical change.

Fig. 10.11 TRPES spectra in a low and b high eBE regions. The inset in a shows the photoelectron spectrum averaged over delay times of –200 to +200 fs. c Static (one-color) photoelectron spectra. Adapted with permission from [21]. Copyright 2018 American Chemical Society

300

S. Adachi and T. Suzuki

The time profiles of the photoelectron intensity observed at eBE = 5.4, 6.5, and 8.1 eV appeared almost identical [21], and the timescale of the cascaded IC was evaluated to be τ ex = 24 ± 10 fs by fitting the three time profiles to single-exponential curves. NM molecules that are unexcited by the pump pulses generate a one-color background signal in the high eBE region. As with furan, we evaluated the one-color signal by averaging the photoelectron spectra over negative delays (−1400 to −400 fs), and subtracted it to obtain Fig. 10.11b. The positive signal near the time origin is attributed to NM ionization from the excited states, and the spectral features in the region of τ ≥ 500 fs correspond to products of NM photodissociation. Figure 10.12a shows a TRPES spectrum measured at τ = 1000 fs. The positive and negative peaks in this spectrum can be assigned from static (one-color) photoelectron spectra of the parent and possible products (Fig. 10.12c). The one-color spectrum of NM was measured in our laboratory using 14-eV pulses, and those for CH3 and NO2 (X) were acquired from the literature [37, 38]. For NO2 (A), a simulated TRPES spectrum at τ = 0 fs [26] was used. Each spectrum has been convolved with a Gaussian distribution representing the experimental energy resolution (~0.2 eV). In Fig. 10.12a, the positive bands peaked at 9.8 and 10.8 eV are assigned to the CH3 and NO2 fragments, respectively, while the negative band at 11–12 eV is assigned to the parent (NM). Fig. 10.12 a Experimental TRPES spectrum at τ = 1000 fs. b Spectrum reproduced using a linear combination of four one-color spectra [NM, CH3 , NO2 (X), and NO2 (A)]. c Individual contributions used in the linear combination. Adapted with permission from [21]. Copyright 2018 American Chemical Society

10 Non-adiabatic Dynamics of Molecules Studied …

301

The observed spectrum is well reproduced (Fig. 10.12b) by a linear combination of the four one-color spectra [NM, CH3 , NO2 (X), and NO2 (A)]. This suggests that the NO2 fragments are produced both in the X and A states. We note that our TRPES spectra exhibit negligible contributions from NO2 in the higher excited states (the B and C states), indicating that Pathways B and C (see Fig. 10.10) are less important in the ultrafast dissociation. This is consistent with the theoretical prediction [34] that NM excited to S3 will immediately relax via S2 to S1 , and that dissociation takes place from the S1 state. Square symbols in Fig. 10.13a show a cut along the TRPES spectra of Fig. 10.11b at the CH3 peak (eBE = 9.9 eV). Least-squares fitting (solid curve) of the experimental data to a linear combination of exponential decay and rise components was performed, with the former expressing the excited-state population of NM. The exponential-decay time-constant of τ decay = 26 fs obtained from the fitting agrees with the timescale of the cascaded IC (τ ex = 24 ± 10 fs) discussed earlier. The exponent growth in Fig. 10.13a is attributed to population of the CH3 fragment. The fitting error increased monotonically with τ diss when it was assumed to be >50 fs, and thus τ diss was considered to be ≤50 fs [21]. This implies that the timescale of fragment formation is similar to the excited-state lifetime (τ ex = 24 fs). Since all dissociation

Fig. 10.13 a Cut along the TRPES spectra of Fig. 10.11b at eBE = 9.9 eV (squares) and leastsquares fit (solid curve) using a linear combination of exponential decay and rise components. The dashed curves show individual components. b Ratio between the NO2 (X) and NO2 (A) populations. The gray envelope represents the error bounds. Adapted with permission from [21]. Copyright 2018 American Chemical Society

302

S. Adachi and T. Suzuki

pathways lead to CH3 formation, the dissociation is assumed to be complete within 50 fs. Fig. 10.12c shows individual contributions from the four species [NM, CH3 , NO2 (X), and NO2 (A)], and the population of NO2 (X) and NO2 (A) appear similar (For a rigorous comparison, the difference in their photoionization cross-sections must be considered.) Isegawa et al. argued that the S1 population mostly undergo IC to S0 via S1 /S0 CIs with a nearly equilibrium (1.45 Å) and/or an almost dissociated (~3 Å) C–N bond lengths [34]. The former pathway yields vibrationally hot NM in the electronic ground state, while the latter pathway may facilitate bifurcation into CH3 + NO2 (A) and CH3 + NO2 (X) products (Pathways D and D , respectively). In addition, NO2 (X) can also be produced through Pathway F (Fig. 10.10), in which NO2 (A) undergoes IC to NO2 (X) during the first 300 fs after photoexcitation, as reported in earlier studies [39, 40]. However, this pathway seems to play a less significant role in the cascading reactions. In Fig. 10.13b, the ratio between the NO2 (X) and NO2 (A) population is evaluated for τ > 200 fs (the population at τ = 1000 fs was shown in Fig. 10.12c). The ratio remains relatively unchanged, indicating that population transfer from A to X is minor. We can estimate QY for the CH3 + NO2 dissociation from the excitation and dissociation probabilities. The former is proportional to the absorption cross-section of NM (~2 × 10−17 cm2 at 205 nm [41]) and the pump laser fluence (17 J/m2 ), and 3.5% of the molecules in the laser volume were expected to undergo excitation in our experiment. The latter was evaluated to be 1.0% from averaged TRPES spectra for negative and positive time delays [21]. These two provide a dissociation QY of 0.29 (= 1.0/3.5), which is a comparable value to the nπ* excitation (0.24) evaluated in non-adiabatic molecular dynamics calculations [42]. This might suggest that the dynamics on the S1 surface play a critical role in both the ππ* and nπ* excitations. It is worth mentioning the reaction pathways not considered in our analysis. In addition to the ultrafast reactions shown in Fig. 10.10, various thermal reaction pathways are accessible via the S0 state (Pathway E), because of the large internal energy of NM (hvpump = 6.0 eV) available for overcoming barriers on the ground state. Examples include isomerization to methyl nitrite (CH3 ONO) and three-body dissociation (CH3 NO2 * → CH3 + NO + O) [43, 44]. However, there is no peak assignable to CH3 ONO (I e ~ 10.9 eV [45]) or NO (I e ~ 9.3 eV [46]) in the experimental TRPES spectrum (Fig. 10.12a) because these thermal reactions take much longer than the observation time window employed in this study (~2 ps). Note that our reported dissociation QY does not include thermal reactions that might affect the “final” photoproduct QYs on longer timescales. In fact, vibrationally hot NM will eventually dissociate, as evidenced by the nearly unit QY of all the dissociation channels in 193-nm laser photolysis [36].

10 Non-adiabatic Dynamics of Molecules Studied …

303

10.5 Conclusions and Future Prospects In this chapter, we presented TRPES using a VUV laser, which enables the observation of ultrafast dynamics involving non-adiabatic transitions through CIs. VUV pulses are useful because they enable one to follow ultrafast dynamics to the final products. In VUV TRPES of furan, ultrafast non-radiative deactivation to the ground state, ground-state bleach recovery, and photoproduct formation were observed. Following photoexcitation to the ππ* state, the ring-puckering CI plays a prominent role in the dynamics, funneling more than 90% of the excited molecules safely back to the original ground state. The remaining 10% undergoes isomerization after passing through this CI. In photodissociation dynamics studies of NM, identification of the electronic state of the NO2 product was a central subject. We found that the ππ* electronic excitation leads to S3 → S2 → S1 + S0 ultrafast cascading IC prior to dissociation into CH3 and NO2 fragments, and that the dissociation predominantly proceeds on the S1 PES, leading to comparable production of NO2 (A) and NO2 (X). Ultrafast non-adiabatic transitions among PESs are important in studying biologically relevant reactions such as light harvesting, UV photodamage, and cis-trans isomerization in vision. It is becoming possible to perform quantum computations on substantially large systems like proteins. We believe results obtained by VUV TRPES will help improve the accuracy of quantitative molecular dynamics simulations and further stimulate them. Acknowledgments The authors thank Prof. Roland Mitric and Dr. Alexander Humeniuk for theoretical simulations of photodynamics of furan and Tom Schatteburg for his experimental assistance. The work on nitromethane was a joint study with Prof. Hiroshi Kohguchi.

References 1. Domcke, W., Yarkony, D., Köppel, H.: Conical Intersections: Theory, Computation and Experiment. World Scientific, Singapore (2011) 2. Fuji, T., Suzuki, Y.-I., Horio, T., et al.: Ultrafast photodynamics of furan. J. Chem. Phys. 133, 234303 (2010). https://doi.org/10.1063/1.3518441 3. Thompson, A.L., Martínez, T.J.: Time-resolved photoelectron spectroscopy from first principles: excited state dynamics of benzene. Faraday Discuss. 150, 293 (2011). https://doi.org/10. 1039/c1fd00003a 4. Neville, S.P., Wang, Y., Boguslavskiy, A.E., et al.: Substituent effects on dynamics at conical intersections: allene and methyl allenes. J. Chem. Phys. 144, 014305 (2016). https://doi.org/ 10.1063/1.4938561 5. Humeniuk, A., Wohlgemuth, M., Suzuki, T., Mitri´c, R.: Time-resolved photoelectron imaging spectra from non-adiabatic molecular dynamics simulations. J. Chem. Phys. 139, 134104 (2013). https://doi.org/10.1063/1.4820238 6. Perun, S., Sobolewski, A.L., Domcke, W.: Ab initio studies on the radiationless decay mechanisms of the lowest excited singlet states of 9H-adenine. J. Am. Chem. Soc. 127, 6257–6265 (2005). https://doi.org/10.1021/ja044321c

304

S. Adachi and T. Suzuki

7. Barbatti, M., Aquino, A.J.A., Szymczak, J.J., et al.: Relaxation mechanisms of UV-photoexcited DNA and RNA nucleobases. Proc. Natl. Acad. Sci. 107, 21453–21458 (2010). https://doi.org/ 10.1073/pnas.1014982107 8. Oesterling, S., Schalk, O., Geng, T., et al.: Substituent effects on the relaxation dynamics of furan, furfural and β-furfural: a combined theoretical and experimental approach. Phys. Chem. Chem. Phys. 19, 2025–2035 (2017). https://doi.org/10.1039/C6CP06240G 9. Ashfold, M.N.R., Bain, M., Hansen, C.S., et al.: Exploring the dynamics of the photoinduced ring-opening of heterocyclic molecules. J. Phys. Chem. Lett. 8, 3440–3451 (2017). https://doi. org/10.1021/acs.jpclett.7b01219 10. Boguslavskiy, A.E., Schalk, O., Gador, N., et al.: Excited state non-adiabatic dynamics of the smallest polyene, trans 1,3-butadiene. I. Time-resolved photoelectron-photoion coincidence spectroscopy. J. Chem. Phys. 148, 164302 (2018). https://doi.org/10.1063/1.5016452 11. Adachi, S., Suzuki, T.: UV-driven harmonic generation for time-resolved photoelectron spectroscopy of polyatomic molecules. Appl. Sci. 8, 1784 (2018). https://doi.org/10.3390/ app8101784 12. Corkum, P.B., Krausz, F.: Attosecond science. Nat. Phys. 3, 381–387 (2007). https://doi.org/ 10.1038/nphys620 13. Nugent-Glandorf, L., Scheer, M., Samuels, D.A., et al.: Ultrafast time-resolved soft X-ray photoelectron spectroscopy of dissociating Br 2. Phys. Rev. Lett. 87, 17–20 (2001). https://doi. org/10.1103/PhysRevLett.87.193002 14. Frassetto, F., Cacho, C., Froud, C.A., et al.: Single-grating monochromator for extremeultraviolet ultrashort pulses. Opt. Express 19, 19169 (2011). https://doi.org/10.1364/OE.19. 019169 15. Adachi, S., Horio, T., Suzuki, T.: Generation of intense single-order harmonic pulse in the vacuum ultraviolet region using a deep ultraviolet driving laser. Opt. Lett. 37, 2118–2120 (2012). https://doi.org/10.1364/OL.37.002118 16. Adachi, S., Sato, M., Suzuki, T.: Direct observation of ground-state product formation in a 1,3-cyclohexadiene ring-opening reaction. J. Phys. Chem. Lett. 6, 343–346 (2015). https://doi. org/10.1021/jz502487r 17. Sato, M., Adachi, S., Suzuki, T.: Photoisomerization of vibrationally hot tetramethylethylene produced by ultrafast internal conversion from the excited state. J. Phys. Chem. A 120, 5099–5102 (2016). https://doi.org/10.1021/acs.jpca.6b00410 18. Yamamoto, Y.I., Karashima, S., Adachi, S., Suzuki, T.: Wavelength dependence of UV photoemission from solvated electrons in bulk water, methanol, and ethanol. J. Phys. Chem. A 120, 1153–1159 (2016). https://doi.org/10.1021/acs.jpca.5b09601 19. Sato, M., Suzuki, Y., Suzuki, T., Adachi, S.: Pump–probe photoelectron spectroscopy by a high-power 90 nm vacuum-ultraviolet laser. Appl. Phys. Express 9, 022401 (2016). https://doi. org/10.7567/APEX.9.022401 20. Adachi, S., Sato, M., Suzuki, T., Grebenshchikov, S.Y.: Unexpectedly broad photoelectron spectrum as a signature of ultrafast electronic relaxation of Rydberg states of carbon dioxide. Phys. Rev. A 95, 1–9 (2017). https://doi.org/10.1103/PhysRevA.95.033422 21. Adachi, S., Kohguchi, H., Suzuki, T.: Unravelling the electronic state of NO2 product in ultrafast photodissociation of nitromethane. J. Phys. Chem. Lett. 9, 270–273 (2018). https://doi.org/10. 1021/acs.jpclett.7b03032 22. Adachi, S., Schatteburg, T., Humeniuk, A., et al.: Probing ultrafast dynamics during and after passing through conical intersections. Phys. Chem. Chem. Phys. (2019). https://doi.org/10. 1039/C8CP04426K 23. Takahashi, E.J., Nabekawa, Y., Mashiko, H., et al.: Generation of strong optical field in soft Xray region by using high-order harmonics. IEEE J. Sel. Top. Quantum Electron. 10, 1315–1328 (2004). https://doi.org/10.1109/JSTQE.2004.838077 24. Hellmann, S., Rossnagel, K., Marczynski-Bühlow, M., Kipp, L.: Vacuum space-charge effects in solid-state photoemission. Phys. Rev. B Condens. Matter Mater. Phys. 79, 1–12 (2009). https://doi.org/10.1103/PhysRevB.79.035402

10 Non-adiabatic Dynamics of Molecules Studied …

305

25. Feldstein, M.J., Vöhringer, P., Scherer, N.F.: Rapid-scan pump–probe spectroscopy with high time and wave-number resolution: optical-Kerr-effect measurements of neat liquids. J. Opt. Soc. Am. B 12, 1500 (1995). https://doi.org/10.1364/JOSAB.12.001500 26. Berera, R., van Grondelle, R., Kennis, J.T.M.: Ultrafast transient absorption spectroscopy: principles and application to photosynthetic systems. Photosynth. Res. 101, 105–118 (2009). https://doi.org/10.1007/s11120-009-9454-y 27. Butler, L.J., Krajnovich, D., Lee, Y.T., et al.: The photodissociation of nitromethane at 193 nm. J. Chem. Phys. 79, 1708–1722 (1983). https://doi.org/10.1063/1.446015 28. Guo, Y.Q., Bhattacharya, A., Bernstein, E.R.: Photodissociation dynamics of nitromethane at 226 and 271 nm at both nanosecond and femtosecond time scales. J. Phys. Chem. A 113, 85–96 (2009). https://doi.org/10.1021/jp806230p 29. Kohge, Y., Hanada, T., Sumida, M., et al.: Photodissociation dynamics of nitromethane at 213 nm studied by ion-imaging. Chem. Phys. Lett. 556, 49–54 (2013). https://doi.org/10.1016/ j.cplett.2012.11.076 30. Rodríguez, J.D., González, M.G., Rubio-Lago, L., et al.: Stereodynamics of the photodissociation of nitromethane at 193 nm: unravelling the dissociation mechanism. J. Phys. Chem. A 117, 8175–8183 (2013). https://doi.org/10.1021/jp403272x 31. Sumida, M., Kohge, Y., Yamasaki, K., Kohguchi, H.: Multiple product pathways in photodissociation of nitromethane at 213 nm. J. Chem. Phys. 144, 064304 (2016). https://doi.org/10. 1063/1.4941090 32. Arenas, J.F., Otero, J.C., Peláez, D., Soto, J.: CASPT2 study of the decomposition of nitrosomethane and its tautomerization reactions in the ground and low-lying excited states. J. Org. Chem. 71, 983–991 (2006). https://doi.org/10.1021/jo051897r 33. Arenas, J.F., Otero, J.C., Peláez, D., Soto, J.: Role of surface crossings in the photochemistry of nitromethane. J. Chem. Phys. 122, 084324 (2005). https://doi.org/10.1063/1.1851977 34. Isegawa, M., Liu, F., Maeda, S., Morokuma, K.: Ab initio reaction pathways for photodissociation and isomerization of nitromethane on four singlet potential energy surfaces with three roaming paths. J. Chem. Phys. 140, 244310 (2014). https://doi.org/10.1063/1.4883916 35. Moss, D.B., Trentelman, K.A., Houston, P.L.: 193 Nm photodissociation dynamics of nitromethane. J. Chem. Phys. 96, 237 (1992). https://doi.org/10.1063/1.462510 36. Blais, N.C.: Photofragmentation of nitromethane in a molecular beam at 193 nm. J. Chem. Phys. 79, 1723 (1983). https://doi.org/10.1063/1.446016 37. Houle, F.A., Beauchamp, J.L.: Photoelectron spectroscopy of methyl, ethyl, isopropyl, and tert-butyl radicals. Implications for the thermochemistry and structures of the radicals and their corresponding carbonium ions. J. Am. Chem. Soc. 101, 4067–4074 (1979). https://doi. org/10.1021/ja00509a010 38. Katsumata, S., Shiromaru, H., Mitani, K., et al.: Photoelectron angular distribution and assignments of photoelectron spectra of nitrogen dioxide, nitromethane and nitrobenzene. Chem. Phys. 69, 423–431 (1982). https://doi.org/10.1016/0301-0104(82)88080-4 39. Arasaki, Y., Takatsuka, K.: Quantum wavepacket dynamics for time-resolved photoelectron spectroscopy of the NO2 conical intersection. Chem. Phys. 338, 175–185 (2007). https://doi. org/10.1016/j.chemphys.2007.04.025 40. Worner, H.J., Bertrand, J.B., Fabre, B., et al.: Conical intersection dynamics in NO2 probed by homodyne high-harmonic spectroscopy. Science 80(334), 208–212 (2011). https://doi.org/10. 1126/science.1208664 41. Napier, I.M., Norrish, R.G.W.: The photolysis and pyrolysis of nitromethane and methyl nitrite. Proc. R. Soc. A Math. Phys. Eng. Sci. 299, 317–336 (1967). https://doi.org/10.1098/rspa.1967. 0139 42. Nelson, T., Bjorgaard, J., Greenfield, M., et al.: Ultrafast photodissociation dynamics of nitromethane. J. Phys. Chem. A 120, 519–526 (2016). https://doi.org/10.1021/acs.jpca. 5b09776 43. Wodtke, A.M., Hintsa, E.J., Lee, Y.T.: The observation of CH3 O in the collision free multiphoton dissociation of CH3 NO2 . J. Chem. Phys. 84, 1044–1045 (1986). https://doi.org/10.1063/ 1.450590

306

S. Adachi and T. Suzuki

44. Dey, A., Fernando, R., Abeysekera, C., et al.: Photodissociation dynamics of nitromethane and methyl nitrite by infrared multiphoton dissociation imaging with quasiclassical trajectory calculations: signatures of the roaming pathway. J. Chem. Phys. 140, 054305 (2014). https:// doi.org/10.1063/1.4862691 45. Ogden, I.K., Shaw, N., Danby, C.J., Powis, I.: Completing dissociation channels of nitromethane and methyl nitrite ions and the role of electronic and internal modes of excitation. Int. J. Mass Spectrom. Ion Process. 54, 41–53 (1983). https://doi.org/10.1016/0168-1176(83)85004-6 46. Pratt, S.T.: Electric field effects in the near-threshold photoionization spectrum of nitric oxide. J. Chem. Phys. 98, 9241–9250 (1993). https://doi.org/10.1063/1.464404

Chapter 11

Femtosecond Time-Resolved Photoelectron Spectroscopy of Molecular Anions Alice Kunin and Daniel M. Neumark

We explore the application of this technique to probe electron attachment and photodissociation dynamics in iodide–nucleobase clusters. The pump pulse initiates intracluster charge transfer, creating transient nucleobase anions that model DNA damage pathways induced by low-energy electron attachment. Image is reproduced from Ref. [57] with permission from the PCCP Owner Societies

© Springer Nature Singapore Pte Ltd. 2019 T. Ebata and M. Fujii (eds.), Physical Chemistry of Cold Gas-Phase Functional Molecules and Clusters, https://doi.org/10.1007/978-981-13-9371-6_11

307

308

A. Kunin and D. M. Neumark

Abstract Femtosecond time-resolved photoelectron spectroscopy (TRPES) is a powerful technique to probe the ultrafast excited state dynamics of molecules. TRPES applied to gas-phase molecular anions and clusters is capable of probing not only excited state formation and relaxation but also electron accommodation dynamics upon injection of an excess electron into a solvent or molecule. We review the basics of TRPES as it applies to molecular anions and several applications including the study of electron solvation dynamics in clusters and excited state relaxation in several biomolecules. We then explore in detail the dynamics of electron attachment and photodissociation in iodide–nucleobase clusters studied by TRPES as a model system for examining radiative damage of DNA induced by low-energy electrons. By initiating charge transfer from iodide to the nucleobase and following the dynamics of the resulting transient negative ions with femtosecond time resolution, TRPES provides a novel window into the chemistry triggered by the attachment of low-energy electrons to nucleobases. Keywords Femtochemistry · Photodissociation · Photoexcitation · Nucleobases · Anion photoelectron spectroscopy

11.1 Introduction Anions are ubiquitous in nature and are important in many biological processes and chemical phenomena. Anionic clusters, which are gas-phase size-selected aggregates of atoms or molecules with one or more excess electrons, can readily be mass-selected and hence are particularly useful model systems to study the evolution of electronic and vibrational structure as a function of size for many systems, including carbon clusters [1, 2], metal and semiconductor clusters [3–6], and solute–solvent clusters [7–9]. One can also investigate electron accommodation or solvation dynamics in an isolated environment [10, 11], thus gaining new insights into the energetics and mechanism of electron solvation in water and other solvents [12]. Anionic clusters can also model charge transfer processes such as charge-transfer-to-solvent (CTTS) transitions with the use of an anionic dopant that, upon photoexcitation, injects the excess electron into the solvent [12, 13]. Modeling electron transfer and attachment dynamics is especially relevant for the study of a number of biological processes such as single- and double-strand DNA damage induced by low-energy electron attachment [14], or dynamics in electron transport chains found in photosynthesis and cellular respiration processes [15]. A. Kunin · D. M. Neumark (B) Department of Chemistry, University of California, Berkeley, CA 94720, USA e-mail: [email protected] A. Kunin e-mail: [email protected] D. M. Neumark Lawrence Berkeley National Laboratory, Chemical Sciences Division, Berkeley, CA 94720, USA

11 Femtosecond Time-Resolved Photoelectron Spectroscopy…

309

Femtosecond (fs) time-resolved photoelectron spectroscopy (TRPES) is a powerful technique to probe the excited states and ultrafast relaxation or dissociation dynamics in molecules and clusters [16–18]. In anions, this pump–probe technique creates an excited state upon pump excitation of a ground state anion; the probe pulse photodetaches the excess electron and the resulting time-evolving photoelectron energy and angular distribution follow the relaxation or decay dynamics. Negative ions are particularly well-suited to study with TRPES as the excess electron binding energy is typically within the range of energies that can be easily generated by traditional Ti:Sapphire ultrafast lasers. The basic principles and several applications of TRPES have been thoroughly reviewed [11, 15–28], so we focus here on only the key concepts as they relate to the study of anions. Single-photon anion photoelectron spectroscopy (PES) [29–35], shown schematically in Fig. 11.1a, employs a laser beam of photon energy hν to photodetach the excess electron of a prepared, stable anion. Only if the photon energy exceeds the electron binding energy (eBE) of the electron to the anion can the excess electron be detached. The kinetic energy (eKE) distribution of the outgoing photodetached electrons is measured, and the principle of energy conservation, as shown in Eq. 11.1, may then be used to determine accurate eBEs: eBE = hν − eKE

(11.1)

For a one-electron transition, photodetachment can occur to any neutral vibrational (and electronic) states within the photon energy range, provided there is sufficient Franck–Condon overlap between the anion and neutral vibrational wavefunctions. Identification of the transition between the anion and neutral vibrational ground states

Fig. 11.1 Example scheme for a anion PES and b TRPES. Anion* indicates the photoexcited state. The blue lines indicate the resultant kinetic energies of the photodetached electrons. Reproduced from Ref. [57] with permission from the PCCP Owner Societies

310

A. Kunin and D. M. Neumark

yields the electron affinity (EA). The vertical detachment energy (VDE) corresponds to the difference in energy between the anion and the neutral at the equilibrium geometry of the anion. In spectra that do not show any vibrational structure, the VDE is identifiable as the peak or maximum intensity (maximum Franck–Condon overlap) of the photoelectron spectrum, and the width of the spectrum is an indication of the geometry change that occurs upon photodetachment. Figure 11.1b shows a schematic diagram for fs anion TRPES. A fs pump pulse is used to generate an electronically excited anion, which can decay by multiple mechanisms including dissociation, internal conversion, or autodetachment. A timedelayed fs probe pulse photodetaches this transient negative ion (TNI) to monitor its temporal evolution, enabling characterization of these various pathways. A sufficiently energetic probe pulse can interrogate not only TNIs but also any anionic dissociation products that may form due to fragmentation, as well as radiationless transitions to other excited states or the anion ground state. The ability to follow ground state dynamics is a notable advantage of anion TRPES as compared to neutral TRPES, in which ionization from the ground state is often not energetically feasible. Thus, anion TRPES can offer a complete picture of the relaxation and dissociation dynamics subsequent to electronic excitation. Anion TRPES has been previously applied to study size-dependent electron solvation dynamics or CTTS dynamics with an iodide dopant atom in a cluster of solvent molecules or atoms [12, 36]. Size-dependent relaxation dynamics have also been probed for carbon clusters [37, 38] and transition metal clusters [39–45] with anion TRPES. TRPES has also been applied to probe fundamental dynamics of longrange interactions involved in excess electron binding, such as non-valence-bound anionic states [46–48] and multiply charged anions (MCAs) [49–52]. In recent years, electronic resonances and electron accommodation dynamics in many biologically relevant species have been of great interest [53–57]. In this chapter, Sect. 11.2 provides a brief description of selected past studies to illustrate the nature of the information that is uniquely gained by the versatile application of TRPES to anionic clusters. Section 11.3 delves into the specific work that has been done in the Neumark group on the dynamics of TNI formation and decay in iodide–nucleobase (I− ·N) clusters as a model system for the reductive damage of DNA. Section 11.4 covers the experimental and computational methodologies specific to the study of I− ·N complexes, and Sect. 11.5 examines the results of these TRPES studies of iodide–uracil (I− ·U) and iodide–thymine (I− ·T) clusters, as well as the simpler, model system of photoexcited iodide–nitromethane (I− ·CH3 NO2 ) clusters that provide an illustrative framework for understanding the more complex dynamics of the larger nucleobase species. We conclude with a summary and outlook for future applications of anion TRPES of iodide-containing clusters to advance our understanding of reductive damage pathways in DNA.

11 Femtosecond Time-Resolved Photoelectron Spectroscopy…

311

11.2 A Brief Overview of Past Anion TRPES Studies Several TRPES studies by the Neumark and Zewail groups have been directed at probing relaxation dynamics in size-selected anionic clusters A− n and iodide-associated I− ·An clusters, including carbon clusters [37, 38, 58, 59], oxygen and solvated oxygen species [60–63], mercury [45, 64–67], and I2 − and I2 − ·An [68–82], among others [83, 84]. Others have worked to theoretically simulate the TRPE spectra of I2 − and I2 − ·Y complexes to aid and improve the dynamical analysis of these studies [85–88], and the I2 − studies have also been extended by Sanov [89, 90] to probe the photodissociation dynamics of I2 Br− and IBr− anions. TRPES has also been used by the Eberhardt and Ganteför groups to explore dynamics of metal thermalization in transition metal A− n clusters [39–44, 91, 92] and desorption in metals with anionic adsorbates [93–98]. These transition metal cluster studies are able to probe the metal band structure and relaxation processes on a molecular level. Much of this work has been previously reviewed [11]. TRPES has been employed by Johnson [99], Neumark [100–104], and Zewail [36] − to study size-selected (H2 O)− n and (D2 O)n water clusters to probe the time-resolved dynamics of solvated electrons. These studies probed the excited state lifetimes of electronically excited water clusters as a model for the relaxation of the bulk solvated electron. The relaxation mechanism is expected to be initial relaxation along the excited state surface, internal conversion (IC), and finally ground state relaxation, but the timescales for each step are a matter of debate [12]. Transient absorption studies of hydrated electrons have measured a rapid, initial ~50 fs lifetime that has been ascribed either to relaxation along the excited state surface (“adiabatic” model) [105, 106] or IC (“nonadiabatic” model) [107] as the fastest observable step. TRPES of (H2 O)− n clusters by Neumark and co-workers [101, 103, 108] measured abrupt appearance and decay of an excited state feature with concomitant dynamics for ground state depletion and recovery, which is indicative of decay by IC. A dramatic decrease in excited state lifetime was observed with increasing cluster size, with sub100 fs IC of the excited state in the largest clusters studied (n = 70–200) [103]. Thus, this rapid timescale for IC strongly supports the nonadiabatic model for hydrated electron relaxation. This nonadiabatic mechanism is also more recently supported by TRPES studies of hydrated electrons in liquid microjets by Neumark [109, 110] and Suzuki [111]. TRPES of I− ·(H2 O)n and I− ·(D2 O)n clusters [112–114] allows one to study the injection of an excess electron into a solvent network as in a CTTS transition since a UV pump pulse can be used to promote charge transfer from the iodide to the solvent moiety (Sn ), as in Eq. 11.2: hνpump

I − ·Sn −−−→ I · · ·Sn∗−

(11.2)

TRPES of I− ·(H2 O)n most notably exhibited a strong shift to higher VDEs of the excited state feature after 1–2 ps, and greater magnitude shifting was observed in larger clusters. When compared to the VDEs measured for different isomers of

312

A. Kunin and D. M. Neumark

(H2 O)− n clusters, this shift suggests that the water cluster relaxation proceeds in a few ps via isomerization from an initially surface bound excess electron to one that may be in a more tightly bound configuration [112–116]. − Other S− n and I ·Sn solvent clusters have also been investigated in time-resolved experiments by Neumark and Zewail, including those of methanol and various iodide–alcohol complexes [117–120], ammonia [121, 122], acetonitrile (CH3 CN)− n [123, 124], and tetrahydrofuran (C4 H8 O) [125, 126]. Many of these solvents are suggested to share some similarities with the proposed relaxation schemes for water, although differences in hydrogen bonding and solvent molecule packing result in different sites or cavities that the excess electron can occupy. Additionally, the presence of different vibrational modes in these solvents compared to those of water affects the IC and solvent motion-driven relaxation pathways. These studies have been previously reviewed in considerable detail [11, 12]. Anion TRPES has also been employed to analyze the properties and dynamics of several nontraditional valence anions. While conventional anions have the excess electron occupying a valence orbital, non-valence-bound anions, in which there is a long-range attraction between the molecular core and the excess electron, are also important species [127]. These anions can be dipole-bound or multipole-bound if the neutral species possesses a sufficiently large dipole or multipole moment, as discussed in considerable detail in the later portions of this chapter. More recently, Verlet and co-workers have reported observation of correlation-bound states (CBSs) of the para-toluquinone trimer cluster anion (pTQ)− 3 [47], and the iodide–hexafluorobenzene cluster (I− ·C6 F6 ) using TRPES [48]. Such CBSs have been described by calculations to be non-valence-bound states that arise from correlation forces between the excess electron and the molecular valence electrons [128–133]. With TRPES, the pump pulse is used to either excite the species to a π* excited state that appears to internally convert to a transient CBS in