Ionization-induced π → H site-switching in phenol–CH₄ complexes studied using IR dip spectroscopy

Abstract

IR spectra of phenol–CH₄ complexes generated in a supersonic expansion were measured before and after photoionization. The IR spectrum before ionization shows the free OH stretching vibration (ν_OH) and the structure of neutral phenol–CH₄ in the electronic ground state (S₀) is assigned to a π-bound geometry, in which the CH₄ ligand is located above the phenol ring. The IR spectrum after ionization to the cationic ground state (D₀) exhibits a red shifted ν_OH band assigned to a hydrogen-bonded cationic structure, in which the CH₄ ligand binds to the phenolic OH group. In contrast to phenol–Ar/Kr, the observed ionization-induced π → H migration has unity yield for CH₄. This difference is attributed to intracluster vibrational energy redistribution processes.

---

1. Introduction

Noncovalent interactions, such as hydrogen bonds (H-bonds) and van der Waals interactions (π-stacking), are responsible for a large number of phenomena in chemistry, biology, and their related research fields.^1–3 Advanced quantum chemical calculations have proven that van der Waals interactions, particularly those generated by aromatic rings, are of comparable strength to H-bonds.³ Thus, their competition controls the structure, dynamics, and function of supramolecular systems such as DNA. Phenol (PhOH) and its van der Waals complexes with rare gas (Rg) or simple molecular ligands are benchmark systems to understand the competition between π-stacking and H-bonding, because they are the smallest molecular systems featuring both π-stacking to the aromatic ring and H-bonding to the OH group.^4–38

The structures of PhOH–Rg_n complexes have been studied extensively. In the neutral ground state (S₀), analyses of intermolecular vibrations in S₁–S₀ excitation spectra, high resolution UV spectra, and infrared (IR) dip spectra demonstrated a π-bound structure for PhOH–Rg (Rg = Ar and Kr), in which the Rg ligand binds above the benzene ring by π-stacking interactions, denoted as (10).^4,5,7,18,22 The notation (pq) indicates PhOH–Rg_n structures in which p and q π-bonded Rg ligands are located above and below the aromatic ring, respectively. For larger PhOH–Ar_n clusters (n = 2 to 4), IR dip spectra revealed that all of them have π-bound structural motifs in S₀, but various isomers can coexist, such as (11) and (20) for n = 2, (21) and (30) for n = 3, and (31) for n = 4.^27,30 Recently, high-level quantum chemical calculations indicated that the π-bound PhOH–Ar structure is a global minimum in the neutral ground state, while the H-bound structure was found to be a transition state or at most a shallow local minimum.^16,20,21,30 All the isomers can readily be separated by their different S₁–S₀ electronic transitions, enabling isomer selective excitation and ionization.^27,30

Structural studies of PhOH⁺–Rg cations in their cationic ground state (D₀) were performed using photoionization efficiency, mass-analyzed threshold ionization, and zero-kinetic-energy photoelectron spectroscopy.^{4,8–10,14,23,24,26} These spectroscopic studies suggested a π-bound structure for PhOH⁺–Rg dimers (Rg = Ar, Kr, and Xe) in the D₀ states. Early IR dip spectra also supported a π-bound structure, because almost no shift of the OH stretching vibration (ν_OH) was observed upon complex formation.⁷ However, IR photodissociation spectra of PhOH⁺–Ar/Kr generated in an electron impact (EI) ion source showed that the most stable isomer in the D₀ state is in fact the H-bound form.^11–13,19 Using photoionization, only the π-bound structure can directly be populated according to the Franck–Condon principle. On the other hand, EI does not have such a restriction and produces the most stable H-bonded structure.³² In this ion source, the first step of the reaction is EI ionization of the phenol monomer, which is followed by three-body aggregation forming the PhOH⁺–Rg cluster predominantly in its most stable H-bonded structure.¹³

The structural restriction in photoionization given by the Franck–Condon principle has opened a new research direction in the study of dynamical processes of noncovalent interactions.³¹ The π-bound structures in PhOH⁺–Rg_n generated by photoionization are a metastable state in D₀, and isomerization toward the H-bound global minima is possible. This photoionization induced π → H site switching reaction was observed for PhOH–Kr, the (11) isomer of PhOH–Ar₂, and the (30) isomer of PhOH–Ar₃ using nanosecond (“static”) and picosecond (time-resolved) UV-UV-IR dip spectroscopy.^{15,17–19,27–29,31} In nanosecond spectroscopy, the IR dip spectra of PhOH⁺–Rg_n show the H-bonded ν_OH vibration for the photoionized complexes (in D₀), although all of them are assigned to π-bound structures in S₀.^17,19,27,29 The picosecond time-resolved UV-UV-IR spectra demonstrate the formation and subsequent decay of the free ν_OH band of the π-bound complexes as a function of the delay time from the ionization event, with timescales of the order of 1–100 picoseconds.^{15,17,27,28,31} Simultaneously, a gradual growth of the bound ν_OH band of the H-bonded complex is observed. Thus, picosecond time-resolved IR spectroscopy clearly elucidates the formation of the metastable π-bound cation and its π → H switching reaction in real time.³¹ This π → H switching reaction is induced by the sudden change of the phenol–ligand interaction upon photoionization. In this respect, PhOH⁺–Rg_n clusters can also serve as a molecular model system to probe solvation dynamics. Similar molecular migration reactions in clusters have recently been observed for water moving around a peptide linkage in the acetanilide–water dimer and for related molecular complexes with polar ligands.^32,39–49

In these solvent migration events, the reaction yield is one of the important parameters for understanding their mechanism. For example, the IR spectrum of PhOH⁺–Kr recorded after photoionization shows both the free and the H-bonded ν_OH vibrations even for nanosecond delays after ionization.^19,28,29 Similar results were observed for PhOH⁺–Ar.^17,27 Thus, it has been concluded that the initially generated π-bound PhOH⁺–Rg isomer isomerizes towards the H-bound isomer, but the H → π back reaction also takes place, and finally both species coexist in equilibrium. As a result, the π → H switching reaction yield in PhOH⁺–Rg is less than unity. On the other hand, nanosecond IR spectra of PhOH–Ar₂ and PhOH–Ar₃ detect only H-bound species long after photoionization, indicating that the π → H reaction yield is unity.^15,17,27 This significant difference in the yield can be understood by the presence of a second (and further) Rg ligand(s) in the larger PhOH⁺–Rg_n complexes with n ≥ 2. The second ligand can remain at the original π-binding site after the π → H site switching of the first ligand. Its three intermolecular vibrational modes act as a bath for the switching reaction. As a result, the H-bound reaction product can immediately be stabilized by intracluster vibrational energy redistribution (IVR). IVR prevents the H → π back reaction to the π-bound structure leading to a 100% reaction yield for the π → H forward reaction. As the PhOH–Rg dimers do not have a second ligand, which can prevent the back reaction, the isomerization yield is non-unity.³¹

However, recent results on water migration from the CO to the NH site in the acetanilide⁺–water and related dimers show 100% reaction yields although the complex does not have a second ligand,^39–41 indicating that a refined explanation for the yield is required. Actually, the H₂O ligand has internal structure giving rise to more intermolecular degrees of freedom compared to the Rg complexes. To this end, we investigate here the complex of PhOH, the molecular benchmark for π → H site switching, with a CH₄ ligand. Similar to Rg ligands, CH₄ interacts with neutral and ionic phenol mainly via its polarizability. Thus, the topologies of the potential energy surfaces of PhOH–CH₄ and PhOH–Rg are expected to be similar in both charge states.¹³ However, CH₄ has internal structure and is not spherical like Rg atoms, which can influence the reaction dynamics and the reaction yield. With this motivation, we have measured IR spectra of PhOH–CH₄ in S₀ and D₀. From the IR spectra before and after ionization, we will explore the photoionization-induced π → H site switching reaction and discuss its yield in terms of IVR. The only spectral information available for PhOH–CH₄, namely the complexation-induced red shifts in the S₁ and ionization energies³⁴ and Raman spectra of PhOH–CH₄ in the fingerprint range³³ are consistent with a π-bound geometry in S₀. In contrast, IR spectra of PhOH⁺–(CH₄)_n generated in the EI source prove that the H-bound structure is the global minimum in the cationic D₀ state.¹³ Thus, it was anticipated that π → H isomerization of PhOH⁺–CH₄ could be detected in the current study by photoionization of PhOH–CH₄ complexes generated in a molecular beam.

2. Experimental and computational techniques

The setup for IR dip spectroscopy has been described previously.⁵⁰ PhOH purchased from SIGMA-ALDRICH was used after purification by vacuum sublimation. The sample vapor was diluted at room temperature in a gas mixture of ∼10% CH₄ in He at 3 bar. The resulting mixture was expanded into a vacuum chamber (typically 2 × 10⁻⁵ Torr) using a pulsed nozzle to produce PhOH–CH₄ clusters in a supersonic jet. The clusters were ionized by two-colour two-photon resonance-enhanced multiphoton ionization (REMPI) using two tuneable UV lasers generating ν₁ and ν₂. Here, PhOH–CH₄ was pumped to the S₁ origin by ν₁ and subsequently ionized by ν₂. Two-colour soft ionization keeps the excess energy after ionization small and prevents dissociation. The produced cluster ions were guided into a quadrupole mass filter and detected by a channel multiplier. The ion signal was digitized by a digital boxcar integrator after amplification and recorded in a personal computer as a function of the laser frequency. The UV photons ν₁ and ν₂ were generated by frequency-doubled outputs of dye lasers pumped by nanosecond Nd:YAG lasers. Tuneable IR light (ν_IR) was generated by difference frequency generation in a LiNbO₃ crystal between the second harmonic of a Nd:YAG laser and the output of a dye laser. The UV and ν_IR laser beams were focused on the molecular beam in a coaxial and counter-propagating geometry using lenses with f = 400 mm. The experiment was operated at 10 Hz. IR spectra in S₀ were obtained using IR dip spectroscopy. To this end, clusters were first irradiated with ν_IR, and after a 50 ns delay with ν₁ and ν₂. While monitoring the ion signal of PhOH⁺–CH₄, ν_IR was scanned through the OH stretching region. When ν_IR is resonant with a vibrational level, the population of the vibrational ground state decreases, and the ion signal is reduced (IR dip). Thus, the IR spectrum in S₀ was measured as a depletion of the ion signal. The IR spectra in D₀ were measured for two different ionization processes, namely photoionization (REMPI) and electron-impact ionization (EI). For the IR spectrum after REMPI (briefly REMPI-IR), the same setup as described above was used. In this case, ν_IR was introduced 50 ns after ν₁ and ν₂. Vibrational excitation of PhOH⁺–CH₄ causes vibrational predissociation of the cluster, and the IR spectrum was obtained by monitoring the depletion of the parent mass signal. The IR spectrum of PhOH⁺–CH₄ generated by EI was obtained in a tandem quadrupole mass spectrometer coupled with an electron impact ionization (EI) source and an octopole ion trap, as described elsewhere.^13,31,32,51 The EI-IR spectrum exhibits the vibrational transitions of the most stable structure of the cold cation complexes and provides a reference OH stretching frequency (ν_OH) for the structural assignments in the D₀ state.

Ab initio quantum chemical calculations were carried out using the Gaussian09 program⁵² to obtain stable structures, theoretical IR spectra, and binding energies of PhOH⁺–CH₄ (Table 1). Geometry optimization was performed at the MP2/aug-cc-pVDZ level, and normal mode analysis of the stationary points confirmed their nature as minima and was used to predict their IR spectra. Harmonic vibrational frequencies were scaled by factors of 0.9605 and 0.9509 for the neutral and the cationic species, respectively, to reproduce the ν_OH frequencies of PhOH⁺ in the respective charge states (3657 and 3534 cm⁻¹).^7,37,53 The binding energy of the cluster was estimated from single-point calculations on the minima at the MP2/aug-cc-pVTZ level, including basis set super-position error (BSSE) corrections and harmonic zero-point vibrational energy (ZPE) corrections obtained using the aug-cc-pVDZ basis. As spin contamination at the MP2 level is severe for the PhOH⁺ cation,³⁰ additional calculations have been performed at the M06-2X/aug-cc-pVTZ level (Table 1). As will be shown below, the intermolecular binding energies obtained at the M06-2X level are systematically larger than those at the MP2 level and match better with the available experimental data.

Table 1 Observed (ν_OH obs), calculated (ν_OH calc) OH stretching frequencies, IR intensities of ν_OH, and binding energies of the neutral and ionic PhOH–CH₄ cluster

	ν OH obs/cm^−1	ν OH calc ^e/cm^−1	I IR/km mol^−1	D0 (De)^f/cm^−1	D0 (De)^g/cm^−1
a Ref. 60. b Ref. 53 and 63. c From REMPI-IR spectrum, see text. d From EI-IR spectrum (ref. 13). e Calculated at the MP2/aug-cc-pVDZ level and scaled by 0.9605 and 0.9509 for neutral and cationic species, respectively. f Estimated from BSSE-corrected single-point calculations at MP2/aug-cc-pVTZ level on structures optimized at MP2/aug-cc-pVDZ level, at which the harmonic ZPE corrections are evaluated. Binding energies without ZPE corrections are given in parentheses. g Evaluated at the M06-2X/aug-cc-pVTZ level. Binding energies without ZPE corrections are given in parentheses.
PhOH	3657^a	3657	64	—	—
PhOH–CH4(π)	3656	3655	64	354 (650)	388 (654)
PhOH–CH4(OH)	Not observed	3652	190	259 (556)	282 (603)
PhOH^+	3534^b	3534	332	—	—
PhOH^+–CH4(π)	Not observed	3534	325	494 (748)	647 (944)
PhOH^+–CH4(OH)	∼3390^c, 3365^d	3423	1097	1129 (1495)	1620 (1809)

---

3. Results and discussion

To the best of our knowledge, the S₁–S₀ electronic spectrum of PhOH–CH₄ has not been reported except for the frequency of the S₁ origin.^33,34Fig. 1 shows the S₁–S₀ REMPI spectrum of PhOH–CH₄ obtained in a supersonic jet in the present work. The strongest transition at 36 290 cm⁻¹ is red shifted by ΔS₁ = −59 cm⁻¹ from the S₁ origin of bare PhOH. The band position is slightly lower in energy than those reported previously (36 302³³ and 36 298 cm⁻¹ (ref. 34)), while the shift matches closely (ΔS₁ = −58 cm⁻¹). Several low-frequency transitions appearing above the S₁ origin are attributed to intermolecular vibrations by comparison to the similar spectra of benzene–CH₄,⁵⁴ toluene–CH₄,⁵⁵ and aniline–CH₄.^56–59 To measure IR dip spectra in S₀, ν₁ was fixed to the S₁ origin, while ν₂ was tuned to 32 140 cm⁻¹. Fig. 2 shows the IR dip spectra of (a) PhOH and (b) PhOH–CH₄ in S₀, respectively. The free ν_OH band of PhOH at 3657 cm⁻¹ is consistent with the previous report.⁶⁰ The IR dip spectrum of the PhOH–CH₄ complex shows a sharp vibrational transition at 3656 cm⁻¹, which is readily assigned to ν_OH of PhOH in the complex. Since ν_OH is essentially unchanged by complex formation with CH₄, the complex is concluded to have a π-bound structure, in which CH₄ is located above the benzene ring. Similar π-bound structures were found for PhOH–Ar_n (n = 1 and 2),²² PhOH–Kr,^5,7 aniline–Ar,^58,61 and 4-aminobenzonitrile–Ar⁶² in S₀. The MP2 and M06-2X calculations support this assignment as they predict the π-bound isomer to be significantly more stable than the H-bound one, D₀(π) = 388 cm⁻¹ and D₀(OH) = 262 cm⁻¹ at the M06-2X level (Table 1). Moreover, the predicted red shift of Δν_OH = −4 cm⁻¹ for H-bound PhOH–CH₄ is not compatible with the measured IR spectrum. The reported Raman spectrum of PhOH–CH₄ is also inconsistent with a H-bonded structure in terms of complexation-induced frequency shifts and line widths.³³ Finally, the intermolecular structure of the S₁–S₀ spectrum of PhOH–CH₄ is quite similar to that of benzene–CH₄ and toluene–CH₄,^54,55 both of which cannot have H-bound structures. Thus, all spectral and quantum chemical evidence is consistent with a π-bound equilibrium structure of PhOH–CH₄ in S₀.

Fig. 1 Two-color two-photon REMPI spectrum of the S₁–S₀ transition of PhOH–CH₄. The S₁ band origin of PhOH is indicated by a dashed line for comparison.

Fig. 2 IR dip spectra of PhOH (a) and PhOH–CH₄ (b) in the S₀ state compared to theoretical spectra of PhOH–CH₄(π) (c) and PhOH–CH₄(OH) (d) calculated at the MP2/aug-cc-pVDZ level, respectively. The optimized structures are also shown.

Fig. 3(a) shows the REMPI-IR spectrum of PhOH⁺–CH₄ in the ν_OH region. A broad and intense vibrational transition was observed at ∼3390 cm⁻¹. Significantly, no transition was found in the vicinity of ν_OH of bare PhOH⁺ at 3534 cm⁻¹ (ref. 53 and 63) indicated by a dashed line. Thus, we assign the broad band to ν_OH of the cationic complex. The red shift and broadening of ν_OH is a clear signature of the formation of a H-bond. In fact, such a large red shift of Δν_OH ∼ −144 cm⁻¹ is only reproducible by the H-bound isomer as shown by the theoretical IR spectra in Fig. 3(c) and (d). Therefore, it is concluded that PhOH⁺–CH₄ generated by photoionization has an H-bonded structure in which CH₄ binds to the OH group. The EI-IR spectrum of cold PhOH⁺–CH₄ shown in Fig. 3(b) arises from the global minimum of the cation cluster and confirms this assignment.¹³ The H-bonded ν_OH transition is found at 3365 cm⁻¹, which is consistent with the predicted value of 3423 cm⁻¹ (MP2 level). The red shift predicted by the MP2 calculations is somewhat smaller than the experimental one (Δν_OH = −111 and −169 cm⁻¹), as they underestimate the interaction energy. For comparison, the red shift calculated at the M06-2X level (Δν_OH = −136 cm⁻¹) is closer to the experimental value, indicating that this level provides a closer approximation to the true interaction energy. The ν_OH frequency of cold H-bonded PhOH⁺–CH₄ clusters observed in the EI-IR spectrum corresponds to the red edge of the broad ν_OH band observed in the REMPI-IR spectrum. This difference is rationalized by the different temperatures of PhOH⁺–CH₄ generated by EI and REMPI. For the EI-IR spectrum, PhOH⁺–CH₄ complexes were generated and cooled directly in the supersonic expansion. In the REMPI process, on the other hand, the complexes were ionized after the cooling process. The complexes are then under collision-free conditions, with the ionization excess energy remaining in the complexes. It is also known that H-bonded ν_OH bands of hot clusters are blue shifted because H-bonds are weakened by excitation of intramolecular proton donor vibrations.^13,31,32 Therefore, the observed spectral feature in the REMPI-IR spectrum of PhOH⁺–CH₄ arises from ν_OH of “hot” H-bound dimers (vide infra).

Fig. 3 IR spectra of the PhOH⁺–CH₄ cluster cation in the D₀ state obtained by REMPI-IR (a) and EI-IR (b) spectroscopy compared to theoretical spectra of PhOH⁺–CH₄(OH) (c) and PhOH⁺–CH₄(π) (d) calculated at MP2/aug-cc-pVDZ level. The optimized structures are also shown. The ν_OH position of PhOH⁺ is indicated by a dashed line.

The detection of the H-bound structure after photoionization indicates that CH₄ migrates from the aromatic ring to the OH site in the cationic ground state, because the structure in S₀ is π-bound and photoionization keeps the same structure due to the Franck–Condon principle. The migration may occur in S₁; however, this scenario does not match the spectral features observed in the S₁–S₀ REMPI spectrum (Fig. 1). In the REMPI spectrum, the S₁ origin is the strongest band and low frequency vibrations assigned to intermolecular modes are weak and without prominent progression. In addition, the S₁–S₀ excitation spectrum is very similar in appearance to that of benzene–CH₄,⁵⁴ which cannot exhibit any type of π → H isomerization. Thus, it is concluded that the structure remains π-bound also in S₁, and the CH₄ ligand switches the binding motif from π-bonded to H-bonded after photoionization in the D₀ state.

Similar to PhOH–Ar₂,^15,17 the yield of the π → H isomerization reaction is 100% in the PhOH–CH₄ dimer. In the (11) isomer of PhOH–Ar₂, the two Ar atoms are bound to opposite sides of the benzene ring in S₀.²² One of the Ar atoms migrates to the OH group after photoionization yielding the H-bonded structure denoted as (H10).^15,17 No vibrational signature of the doubly π-bound (11) structure was found long after ionization, i.e., the reaction yield for the ionization-induced π → H site switching is unity. In contrast, the similar π-bound PhOH–Kr dimer shows both H-bonded and free ν_OH bands long after photoionization, with a finite equilibrium population between π-bound and H-bound structures.^19,28,29 The reaction yield was evaluated as only ∼40%. Interestingly, for the PhOH–CH₄ dimer, no free ν_OH band was found in the REMPI-IR spectrum, and the π → H site switching yield is unity, similar to that in the PhOH–Ar₂ trimer. Thus, although both PhOH–Kr and PhOH–CH₄ have only a single ligand and both are probed under isolated collision-free conditions in a supersonic jet, the reaction yields are qualitatively different.

For the PhOH–Ar₂ trimer, the 100% reaction yield was rationalized by the stabilization of the H-bound (H10) isomer via fast IVR.¹⁷ After the π → H site switching reaction, the H-bound complex has significant vibrational excess energy originating from the energy difference between the π-bound and the H-bound structures. In the H-bound (H10) trimer, the spectator Ar atom remains located on the π-ring, and thus its three intermolecular vibrations can act as bath modes and accept vibrational excess energy of the H-bound Ar atom. In this way, the vibrational excess energy in the reaction coordinate decreases significantly and the H → π back reaction is efficiently quenched.

On the other hand, the π ⇌ H equilibrium observed after photoionization of PhOH–Rg (Rg = Ar, Kr) dimers can be explained by the lack of bath modes. All three intermolecular modes in PhOH⁺–Rg originating from the three translational degrees of freedom of the single Rg atom are required to describe the reaction coordinate for Rg isomerization. Just after the π → H site switching, the isomerization excess energy is localized in the reaction coordinate. However, no low frequency bath modes in the complex are available to accept the excess energy because all three intermolecular vibrational modes are already involved in the reaction coordinate and cannot act as bath modes for IVR. Therefore, the H → π back reaction can take place in accordance with the classical description, and finally the π ⇌ H equilibrium is achieved.

Similar to PhOH–Rg, the PhOH–CH₄ dimer has a single spherical top ligand. The CH₄ ligand, however, is not a sphere like a Rg atom but has internal structure and three rotational degrees of freedom. The PhOH–CH₄ dimer has six intermolecular modes, three originating from translations and three from rotations. Analogous to the Rg case, the reaction coordinate for the CH₄ site switching requires all three translational motions. As the coupling of the rotation-derived modes with the isomerization coordinate is weak, at least one of them can act as the bath mode for IVR. Those modes can accept the excess energy after the isomerization, stabilize the “hot” nascent H-bound complex by IVR and thus prevent the H → π back reaction. Therefore, the 100% reaction yield can be explained by fast IVR, mainly by coupling of the isomerization coordinates to the intermolecular vibrations related to the internal rotation of CH₄.

Fig. 4 summarizes the reaction mechanism of the photoionization-induced π → H site switching reaction in PhOH⁺–CH₄ in a potential energy diagram. In the following, we combine the available experimental information with the data from the M06-2X calculations (Table 1) to derive a consistent quantitative picture of the energetics of this reaction potential because of the importance of energetics in the reaction.⁶⁴ The binding energy of H-bound PhOH⁺–CH₄ predicted by the M06-X level (D₀ = 1620 cm⁻¹) is quite similar to that of the related H-bound PhOH⁺–N₂ complex precisely measured as 1640 ± 10 cm⁻¹ by mass-analyzed threshold ionization spectroscopy.^35,38 This similarity is expected, because both complexes exhibit the same complexation-induced red shifts Δν_OH = −169 cm⁻¹.¹³ Moreover, similar H-bond energies for CH₄ and N₂ ligands are due to their similar proton affinities.^13,32,36 Clearly, the binding energy derived for H-bound PhOH⁺–CH₄ at the MP2 level (D₀ = 1129 cm⁻¹) is too low and not consistent with the magnitude of the measured Δν_OH shift. Similarly, the binding energy predicted for the π-bonded PhOH⁺–CH₄ isomer at the M06-2X level, (D₀ = 647 cm⁻¹) is more realistic than that at the MP2 level (D₀ = 494 cm⁻¹) by comparison to experimental dissociation energies estimated from photodissociation data of CH₄-containing complexes of aromatic ions.^65,66 On the basis of the M06-2X calculations, the gain in internal energy upon π → H isomerization is of the order of 1000 cm⁻¹. Additionally, the initially prepared π-bound complex may already contain some amount of internal energy, which depends on the Franck–Condon factors for photoionization from the S₁ origin and the ionization excess energy. The difference between the two-photon energy used for two-color soft ionization (ν₁ + ν₂ = 36 290 + 32 140 = 68 430 cm⁻¹) and the adiabatic ionization energy of π-bound PhOH⁺–CH₄ (IE = 68 334 cm⁻¹)^34,67 indicates that the maximum internal energy available upon photoionization is around 100 cm⁻¹. As a result, the barrier for π → H isomerization in the D₀ state of PhOH⁺–CH₄ must be relatively small. Moreover, the total internal energy of the H-bound PhOH⁺–CH₄ product after ionization is estimated as ∼1050 ± 100 cm⁻¹. Finally, we note that both the adiabatic IE and its complexation-induced red shift measured for π-bound PhOH⁺–CH₄ (IE = 68 334 and ΔIE = −294 cm⁻¹),^34,67 are in good agreement with the predictions of the M06-2X calculations (IE = 68 064 and ΔIE = −365 cm⁻¹), confirming that this theoretical level indeed makes reliable predictions for the interaction between PhOH⁺ and CH₄ in both charge states. In fact, ΔIE corresponds to the difference of the binding energies of π-bound PhOH⁺–CH₄ in the S₀ and D₀ states.

Fig. 4 Schematic potential energy curve of PhOH⁺–CH₄ along the reaction coordinate. The initial π-bound structure reacts to a “Hot” H-bound cluster, in which the excess energy remains as vibrational excitation. The “Hot” H-bound cluster may go back to the π-bound structure using this energy. Eventually, the excess energy flows to bath modes due to IVR. The rotation-derived intermolecular modes of the CH₄ ligand can act as bath modes to suppress the back reaction.

4. Conclusions

In summary, IR spectra of the PhOH⁺–CH₄ dimer in the S₀ and D₀ states were recorded in the ν_OH range using IR dip spectroscopy. While the neutral dimer has a π-bound structure in S₀, the cation dimer has a H-bound structure in D₀. From the Franck–Condon principle for photoionization, one can conclude that the CH₄ ligand switches its binding motif from π-bonding above the aromatic ring to H-bonding at the OH site. The 100% reaction yield observed for this prototypical π → H site switching triggered by ionization is explained by rapid IVR of the H-bound structure mainly due to the intermolecular energy transfer from the reaction coordinate to intermolecular modes originating from internal rotational degrees of freedom of CH₄. Following aromatic complexes with atomic and nonpolar Rg ligands, and those with polar H₂O and CH₃OH ligands, PhOH⁺–CH₄ is the first CH₄-containing complex for which the ionization-induced site-switching of the preferred binding motif is observed. Typical time scales for such site switching reactions are of the order of 1–100 ps.^15,17,40 Future efforts will be aimed at the time-resolved investigation of this prototypical intermolecular isomerization reaction.^{15,17,27,31,40}

Acknowledgements

This work was supported in part by the Core-to-Core Program 22003 from the Japan Society for the Promotion of Science (JSPS), KAKENHI on Innovative Area (2503) “Studying the Function of Soft Molecular Systems by the Concerted Use of Theory and Experiment”, the Cooperative Research Program of “Network Joint Research Center for Materials and Devices”, from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, and the Deutsche Forschungsgemeinschaft (DFG, DO 729/4). F.M. was supported by the Fonds National de la Recherche, Luxembourg (AFR/1357660).

References

1. P. Hobza and K. Müller-Dethlefs, Non-covalent interactions, The Royal Society of Chemistry, Cambridge, 2010.
2. E. A. Meyer, R. K. Castellano and F. Diederich, Angew. Chem., Int. Ed., 2003, 42, 1210–1250.
3. P. Jurečka and P. Hobza, J. Am. Chem. Soc., 2003, 125, 15608–15613.
4. N. Gonohe, H. Abe, N. Mikami and M. Ito, J. Phys. Chem., 1985, 89, 3642–3648.
5. M. Mons, J. L. Calvé, F. Piuzzi and I. Dimicoli, J. Chem. Phys., 1990, 92, 2155–2165.
6. O. Dopfer, G. Reiser, K. Müller-Dethlefs, E. W. Schlag and S. D. Colson, J. Chem. Phys., 1994, 101, 974–989.
7. A. Fujii, T. Sawamura, S. Tanabe, T. Ebata and N. Mikami, Chem. Phys. Lett., 1994, 225, 104–107.
8. O. Dopfer, M. Melf and K. Müller-Dethlefs, Chem. Phys., 1996, 207, 437–449.
9. C. E. H. Dessent, S. R. Haines and K. Müller-Dethlefs, Chem. Phys. Lett., 1999, 315, 103–108.
10. S. R. Haines, C. E. H. Desent and K. Müller-Dethlefs, J. Electron Spectrosc. Relat. Phenom., 2000, 108, 1–11.
11. N. Solcà and O. Dopfer, Chem. Phys. Lett., 2000, 325, 354–359.
12. N. Solcà and O. Dopfer, J. Mol. Struct., 2001, 563/564, 241–244.
13. N. Solcà and O. Dopfer, J. Phys. Chem. A, 2001, 105, 5637–5645.
14. S. Ullrich, G. Tarczay and K. Müller-Dethlefs, J. Phys. Chem. A, 2002, 106, 1496–1503.
15. S. Ishiuchi, M. Sakai, Y. Tsuchida, A. Takeda, Y. Kawashima, M. Fujii, O. Dopfer and K. Müller-Dethlefs, Angew. Chem., Int. Ed., 2005, 44, 6149–6151.
16. J. Makarewicz, J. Chem. Phys., 2006, 124, 084310.
17. S. Ishiuchi, M. Sakai, Y. Tsuchida, A. Takeda, Y. Kawashima, O. Dopfer, K. Müller-Dethlefs and M. Fujii, J. Chem. Phys., 2007, 127, 114307.
18. S. Ishiuchi, Y. Tsuchida, O. Dopfer, K. Müller-Dethlefs and M. Fujii, J. Phys. Chem. A, 2007, 111, 7569–7575.
19. A. Takeda, H.-S. Andrei, M. Miyazaki, S. Ishiuchi, M. Sakai, M. Fujii and O. Dopfer, Chem. Phys. Lett., 2007, 443, 227–231.
20. J. Černý, X. Tong, P. Hobza and K. Müller-Dethlefs, Phys. Chem. Chem. Phys., 2008, 10, 2780–2784.
21. J. Černý, X. Tong, P. Hobza and K. Müller-Dethlefs, J. Chem. Phys., 2008, 128, 114319.
22. I. Kalkman, C. Brand, T.-B. C. Vu, W. L. Meerts, Y. N. Svartsov, O. Dopfer, K. Müller-Dethlefs, S. Grimme and M. Schmitt, J. Chem. Phys., 2009, 130, 224303.
23. A. Armentano, M. Riese, M. Taherkahani, M. B. Yezzar, K. Müller-Dethlefs, M. Fujii and O. Dopfer, J. Phys. Chem. A, 2010, 114, 11139–11143.
24. X. Tong, A. Armentano, M. Riese, M. B. Yezzar, S. M. Pimblott, K. Müller-Dethlefs, S. Ishiuchi, M. Sakai, A. Takeda, M. Fujii and O. Dopfer, J. Chem. Phys., 2010, 133, 154308.
25. C. Walter, R. Kritzer, A. Schubert, C. Meier, O. Dopfer and V. Engel, J. Phys. Chem. A, 2010, 114, 9743–9748.
26. A. Armentano, X. Tong, M. Riese, S. M. Pimblott, K. Müller-Dethlefs, M. Fujii and O. Dopfer, Phys. Chem. Chem. Phys., 2011, 13, 6071–6076.
27. S. Ishiuchi, M. Miyazaki, M. Sakai, M. Fujii, M. Schmies and O. Dopfer, Phys. Chem. Chem. Phys., 2011, 13, 2409–2416.
28. M. Miyazaki, A. Takeda, S. Ishiuchi, M. Sakai, O. Dopfer and M. Fujii, Phys. Chem. Chem. Phys., 2011, 13, 2744–2747.
29. M. Miyazaki, S. Tanaka, S. Ishiuchi, O. Dopfer and M. Fujii, Chem. Phys. Lett., 2011, 513, 208–211.
30. M. Schmies, A. Patzer, M. Fujii and O. Dopfer, Phys. Chem. Chem. Phys., 2011, 13, 13926–13941.
31. M. Fujii and O. Dopfer, Int. Rev. Phys. Chem., 2012, 31, 131–173.
32. O. Dopfer, Z. Phys. Chem., 2005, 219, 125–168.
33. G. V. Hartland, B. F. Henson, V. A. Venturo and P. M. Felker, J. Phys. Chem., 1992, 96, 1164–1173.
34. X. Zhang and J. L. Knee, Faraday Discuss., 1994, 97, 299–313.
35. D. M. Chapman, K. Müller-Dethlefs and J. B. Peel, J. Chem. Phys., 1999, 111, 1955–1963.
36. A. Patzer, H. Knorke, J. Langer and O. Dopfer, Chem. Phys. Lett., 2008, 457, 298–302.
37. A. Fujii, M. Miyazaki, T. Ebata and N. Mikami, J. Chem. Phys., 1999, 110, 11125–11128.
38. S. R. Haines, W. D. Geppert, D. M. Chapman, M. J. Watkins, C. E. H. Dessent, M. C. R. Cockett and K. Müller-Dethlefs, J. Chem. Phys., 1998, 109, 9244–9251.
39. M. Weiler, T. Nakamura, H. Sekiya, O. Dopfer, M. Miyazaki and M. Fujii, ChemPhysChem, 2012, 13, 3875–3881.
40. K. Tanabe, M. Miyazaki, M. Schmies, A. Patzer, M. Schütz, H. Sekiya, M. Sakai, O. Dopfer and M. Fujii, Angew. Chem., Int. Ed., 2012, 51, 6604–6607.
41. K. Sakota, S. Harada, Y. Shimazaki and H. Sekiya, J. Phys. Chem. A, 2011, 115, 626–630.
42. K. Sakota, Y. Shimazaki and H. Sekiya, Phys. Chem. Chem. Phys., 2011, 13, 6411–6415.
43. T. Ikeda, K. Sakota, Y. Kawashima, Y. Shimazaki and H. Sekiya, J. Phys. Chem. A, 2012, 116, 3816–3823.
44. K. Sakota, Y. Kouno, S. Harada, M. Miyazaki, M. Fujii and H. Sekiya, J. Chem. Phys., 2012, 137, 224311.
45. M. Gerhards, A. Jansen, C. Unterberg and A. Gerlach, J. Chem. Phys., 2005, 123, 074321.
46. H. M. Kim, K. Y. Han, J. Park, G. S. Kim and S. K. Kim, J. Chem. Phys., 2008, 128, 041104.
47. N. Solcà and O. Dopfer, Chem. Phys. Lett., 2001, 347, 59–64.
48. M. Miyazaki, A. Fujii, T. Ebata and N. Mikami, Chem. Phys. Lett., 2001, 349, 431–436.
49. T. Nakamura, M. Schmies, A. Patzer, M. Miyazaki, S. Ishiuchi, M. Weiler, O. Dopfer and M. Fujii, Chem. Eur. J. DOI:10.1002/chem.201303321.
50. T. Omi, H. Shitomi, N. Sekiya, K. Takazawa and M. Fujii, Chem. Phys. Lett., 1996, 252, 287–293.
51. O. Dopfer, Int. Rev. Phys. Chem., 2003, 22, 437–495.
52. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. J. A. Montgomery, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford, CT, 2009.
53. E. Fujimaki, A. Fujii, T. Ebata and N. Mikami, J. Chem. Phys., 1999, 110, 4238–4247.
54. M. Schauer and E. R. Bernstein, J. Chem. Phys., 1985, 82, 726–735.
55. M. Schauer, K. S. Law and E. R. Bernstein, J. Chem. Phys., 1985, 82, 736–746.
56. E. R. Bernstein, K. Law and M. Schauer, J. Chem. Phys., 1984, 80, 634–644.
57. K. Yamanouchi, S. Isogai and S. Tsuchiya, J. Mol. Struct., 1986, 146, 349–359.
58. K. Yamanouchi, S. Isogai, S. Tsuchiya and K. Kuchitsu, Chem. Phys., 1987, 116, 123–132.
59. J. M. Smith, X. Zhang and J. L. Knee, J. Chem. Phys., 1993, 99, 2550–2559.
60. T. Watanabe, T. Ebata, S. Tanabe and N. Mikami, J. Chem. Phys., 1996, 105, 408–419.
61. P. Hermine, P. Parneix, B. Coutant, F. G. Amar and P. Bréchignac, Z. Phys. D: At., Mol. Clusters, 1992, 22, 529–539.
62. T. Nakamura, M. Miyazaki, S. Ishiuchi, M. Weiler, M. Schmies, O. Dopfer and M. Fujii, ChemPhysChem, 2013, 14, 741–745.
63. A. Fujii, A. Iwasaki, T. Ebata and N. Mikami, J. Phys. Chem. A, 1997, 101, 5963–5965.
64. A. Mahjoub, A. Chakraborty, V. Lepere, K. Le Barbu-Debus, N. Guchhait and A. Zehnacker, Phys. Chem. Chem. Phys., 2009, 11, 5160–5169.
65. O. Dopfer, R. V. Olkhov and J. P. Maier, J. Chem. Phys., 1999, 111, 10754–10757.
66. N. Solcà and O. Dopfer, Chem.–Eur. J., 2003, 9, 3154–3163.
67. O. Dopfer, G. Reiser, K. Müller-Dethlefs, E. W. Schlag and S. D. Colson, J. Chem. Phys., 1994, 101, 974–989.

---