Chemical Physics 330 (2006) 138–145 www.elsevier.com/locate/chemphys

The –water and aniline– complexes in the S1 excited state

G. Piani a, M. Pasquini a,b,I.Lo´pez-Toco´n a,c, G. Pietraperzia a,b, M. Becucci a,b,*, E. Castellucci a,b

a LENS, via N. Carrara 1, Polo Scientifico Universita` di Firenze, 50019 Sesto Fiorentino (FI), Italy b Dipartimento di Chimica, Universita` di Firenze, Firenze, Italy c Departamento de Quı´mica-Fı´sica, Universidad de Ma´laga, Spain

Received 27 March 2006; accepted 2 August 2006 Available online 5 August 2006

Abstract

We report an experimental and theoretical study on the properties of the aniline–water and aniline–methanol jet cooled complexes. In both complexes the ligand (H2OorCH3OH) is bonded to aniline, the interaction taking place at the of the , in the amino group. The S1 S0 electronic excitation spectrum in both cases exhibits a very broad and weak band, blue shifted with respect to the origin band of aniline by 700 cm1. Ab initio calculations, at different levels of theory with the cc-pvdz electronic basis set, were performed on aniline–water and predict a strong binding energy decrease in the excited state and a large change in geometry, in agreement with experimental results. Ó 2006 Elsevier B.V. All rights reserved.

1. Introduction [3,4]. Recently, different authors have reported on the properties of the anisole–H2O complex [5–7]. The structure Recent studies have reported on the properties of com- of anisole closely resembles that of : the only differ- plex formed by water with organic molecules in gas ence is the change of the OH group with the OCH3 group. supersonic expansion [1,2]. In most of the cases, water acts The electronic transition has practically the same character as a and the complex is stable, both in the ground and for both molecules [8]. Instead, in the case of water com- in the first electronic excited state. In this respect a repre- plex formation the two systems behave quite differently sentative system is the phenol–water complex. In the phe- as water is binding as an acid to anisole. Therefore, due nol–H2O complex, water is bound as a base to the to the decrease of with electronic excitation phenol OH group [1,3,4]. Due to the changes in the elec- in the oxygen atom lone pairs, the origin of the electronic tronic density with the electronic transition, the hydrogen transition for the complex is blue shifted with respect to bond is stronger in the excited state with respect to the the anisole monomer and the hydrogen bond distance ground state. A measure of this change in the interaction increases. A discussion is still open on the nature of the energy is given by the red shift of the electronic transition structural rearrangement of the anisole–water cluster with in the complex with respect to the isolated phenol [4]. Also electronic excitation and on the possible presence of a sec- the hydrogen-bond distance decreases in the excited state ond relevant interaction point between the oxygen atom of water and the aromatic hydrogen atom in ortho position of anisole. Different results were reported for adenine–H O * Corresponding author. Address: Dipartimento di Chimica, Universita` 2 di Firenze, Firenze, Italy. Fax: +39 055 4572 451. [9,10]. The adenine–H2O complex was clearly observed E-mail address: maurizio.becucci@unifi.it (M. Becucci). with high energy excitation (i.e., photons below 200 nm

0301-0104/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chemphys.2006.08.004 G. Piani et al. / Chemical Physics 330 (2006) 138–145 139 wavelength or direct ionization by electronic impact). If the interval between 0 °C and 25 °C. Mass spectra were adenine–H2O complex is excited to the S1 excited state, obtained via multi-photon ionization with different photon then the corresponding cation is observed only if a probe energy, in- and out-of-resonance for the aniline– pulse (used for ionization) is provided with a delay not water complex. The same procedures were applied for the longer than 220 fs with respect to the pump pulse. The study of the aniline–methanol complex. structure of the adenine–H2O 1:1 complex has not been determined by experiments but different models for this 2.2. Ab initio methods system have been reported [11,12]. In some of these models, water is connected to a hydrogen atom of the amino group In order to evaluate the properties of the aniline–H2O and one of the aromatic N atoms [11]. complex in the excited state, we have followed the guide- In this respect, it seems quite interesting to investigate lines provided by Fang for a similar study on phenol– the properties of the simplest aromatic amino molecule, H2O [20]. The equilibrium structure and the interaction i.e., aniline and the complex it forms with water. Our aim energy of the aniline–H2O complex have been calculated is to contribute, with experimental data and quantum cal- by second-order Møller–Plesset (MP2) perturbation theory culation, to the description of the changes in the properties and complete active space-self consistent field (CASSCF) of clusters with electronic excitation. methods in the ground electronic state, S0, and by single It has been shown by microwave spectroscopy experi- configuration interaction (CIS) and CASSCF methods in ments [17] that, in the ground state, aniline acts as a base the first excited electronic state, S1. In order to account with water placed in its symmetry plane above the aromatic for the correlation energy, the CASSCF calculations were ring, connected to the amino group via hydrogen bonding, followed by single point CASPT2 energy calculations. the of the amino group being on the opposite The cc-pvdz basis set has been employed in all calculations side of water. The properties of aniline in the gas phase as suggested by previous reports [21,22]. We also used an have been studied in detail both in the ground and the first augmented (aug-cc-pvdz) basis set in order to know its excited electronic state by rotationally resolved experi- effect on the results. The CASSCF active space includes ments [13–15]. Very recently, photon-induced dissociation all p electrons and p orbitals of the aromatic ring (three of anionic clusters containing aniline has been reported. p bonds, three corresponding antibonding orbitals) and There, the electron photo-detachment process have been the nitrogen n lone pair of aniline. This results in an active studied for negative ionic clusters formed by iodide with space of eight electrons in seven orbitals and is denoted by aniline [16]. By the measure in coincidence of the detached CAS(8,7). The oxygen lone pair orbitals of water were not electron and the neutral particles, it has been demonstrated included in the active space due to the difference in the that the cluster is dissociating with quantum yield almost orbital energies from those of p orbitals, which are near- unitary after the electron removal. degenerate. In the case of phenol–H2O complex [20],it We report here on the measure, by resonance-enhanced has been shown that the CASSCF convergence is seriously multi-photon ionization spectroscopy (REMPI), of the degraded when the n orbitals of water were included. The S1 S0 electronic transition of aniline–H2O and the clo- interaction energy derived from this type of calculation is sely related aniline–methanol complex, and on quantum subject to the basis set superposition error (BSSE) that calculations aiming at the description of the system in both can be corrected using the counterpoise method of Boys the ground and the excited. and Bernardi [23]. Therefore, the corrected interaction energy is calculated as: 2. Experimental and computational methods AB AB AB Eint ¼ EABðABÞ EABðAÞ EABðBÞ Z 2.1. Experimental methods where we define EY(X) as the energy of a molecular sys- tem X at the equilibrium geometry of the molecular system We have made a search for the S1 S0 electronic tran- Y with the set of basis functions related to the molecular sition of the aniline–H2O complex by REMPI spectroscopy system Z. A better evaluation of the binding energy in experiments in both 2 photon-1 colour and 2 photon-2 col- the complex, called stabilization energy, can be obtained our excitation schemes. The laser spectrometer coupled to a introducing the fragment relaxation in the counterpoise time-of-flight mass spectrometer was already described in scheme according to the expression: detail elsewhere [5]. The molecular beam was prepared by E ¼ E þ E ðAÞA þ E ðBÞB E ðAÞA E ðBÞB: flowing an aniline–helium gas mixture through a pulsed stab int AB AB A B valve (500 lm diameter). The stagnation pressure was The first term is the corrected interaction energy and the fol- 300 KPa. Aniline was placed in a temperature controlled lowing four terms represent the relaxation contribution, compartment and its partial pressure regulated controlling which compensates for the geometry distortion of the sub- A B its temperature (in a range between 10 °C and 30 °C). In system in the supermolecule, EAB(A) and EAB(B) , with re- A B order to produce aniline–water complexes, the helium was gard to the isolated optimum geometry, EA(A) and EB(B) flowing also through another external compartment con- [24]. Therefore, seven energy calculation had to be carried taining liquid water held at a temperature ranging in an out, instead of the three needed for the determination of 140 G. Piani et al. / Chemical Physics 330 (2006) 138–145 the interaction energy. All ab initio calculations have been any new sharp resonance in the REMPI spectrum either carried out using the GAUSSIAN 98 [25] programs pack- integrating the ion signal at the time of flight of mass 93 age on an IBM PWR3 computer. or 111 amu (corresponding to aniline or aniline–H2O molecular ions, respectively). The search for the origin 3. Results and discussion band of the aniline–water complex was extended to a rather large spectral region, ±1000 cm1 around the origin band 3.1. and REMPI spectroscopy of the S1 S0 transition of aniline. A weak and broad sig- nal in the REMPI spectrum was observed from the ions at The presence of the aniline–water complex in the molec- mass 111 amu (the 1:1 complex) – see Fig. 2 blue shifted ular beam was revealed by the measure of the mass with respect to the aniline origin band of about spectrum of the expanding gas mixture ionized by laser 700 cm1. The shift of the band to higher transition ener- multi-photon processes. The measurements were repeated gies can be predicted with some simple considerations for different stagnation pressures and different relative con- about the nature of the electronic transition involved. It centrations of helium, water and aniline. The energy of the is known that the nature of the aniline S1 S0 electronic exciting photons was varied between 34,000 and transition is essentially n ! p*. Then, a displacement of 35,000 cm1. The relative amounts of the aniline+ and (ani- electron density from the nitrogen lone pair to the aromatic + line–H2O) ions were very different and the quantitative ring is expected when aniline is excited to S1 state. As water measure of the ion signal for the complex was very difficult is hydrogen bonded to the nitrogen lone pair, it is then very due to saturation effects on the detector. Therefore, a real- reasonable to take these electronic factors as the most rel- time control method for the gain of the ion detector was evant for the understanding of the REMPI experimental devised and successfully applied [26]. Then we were able data: the complex is less stable in the S1 state as the electron to observe the presence of many aniline–water aggregates density involved in the hydrogen bonding is removed and with a predominant signal corresponding to the 1:1 ani- the electronic transition has to be blue shifted with respect line–water complex, as shown in Fig. 1. The mass spectrum to the one of the aniline monomer. The two potential sur- of the aniline–(CH3OH)n system of complexes was rather faces involved in the electronic transitions are possibly similar and it was obtained with the same methods (also shifted, the transition takes place in a region with a very in Fig. 1). high density of states and the complex is prepared in the The REMPI spectrum of pure aniline was readily S1 state vibrationally excited: this accounts for the observed with a good signal to noise ratio using a one observed spectral features. color-two photon excitation scheme. When water was However, other phenomena could possibly lead to the added in the expanding gas mixture we did not observe broad band observed in the REMPI spectrum, as, for

Fig. 1. The TOF mass spectrum obtained with multi-photon ionization (at 34,980 cm1) of a jet cooled gas mixture containing aniline and water (upper trace) or methanol (lower trace) in helium. The three peaks structure present in both TOF spectra around 35.9 ls (marked*) is due to the presence of a contaminant (C7H6O2, 122 amu) in the gas line: the intensity of both mass spectra has been normalized on these signals. G. Piani et al. / Chemical Physics 330 (2006) 138–145 141

Fig. 2. The aniline–H2O complex REMPI spectrum. instance, an insufficient cooling of the sample. At first, the observed for clusters. We have set-up another experimental possibility of a contribution from hot bands to the conges- scheme using a weak, tunable, pump pulse around 290 nm tion of the spectrum was checked by repeating the experi- and a ionization pulse at 355 nm: under the previous ment at different expansion conditions. The overall shape hypothesis, the ionization process takes place just above of the spectrum was always the same and, in the REMPI threshold and the ions produced have a maximum internal 1 spectrum of aniline, the I1 hot band was weak. We have also energy of about 0.05 eV. The 2 color-2 photon REMPI measured the yield in the production of different ions vary- spectrum of the aniline–water cluster is identical to the ing the excitation scheme. The ionization energy of aniline is one measured using two photons of the same energy. 7.7200(2) eV [19], assuming the same ionization energy for If the above described mechanism is responsible for the the aniline–watern clusters, using in the ionization process observed band shape, then the same spectroscopic behavior two photons at 4.275 eV (290 nm) the ions produced are is expected also for other clusters where the same kind of possibly left with an internal vibrational excitation as large intermolecular interaction is dominant. Then, we decided as 0.83 eV, at least 4 times larger than the calculated ground to extend the study to the aniline–methanol complex as state binding energy. Then, possibly fragmentation can the structure of this complex is very similar to that of ani- occur and clusters of different size can contribute to the ani- line–water and the interaction is the same [18]. As can be line–water 1:1 ion signal or no signal at all could be seen in Fig. 3, no relevant differences appears in the

Fig. 3. The aniline–CH3OH REMPI spectrum. 142 G. Piani et al. / Chemical Physics 330 (2006) 138–145 aniline–CH3OH REMPI spectrum with respect to the one lations both in the Cs symmetry and without symmetry con- observed for aniline–H2O. We reveal a similar weak and strains starting from different geometries, including some broad ionization signal, blue shifted with respect to the ori- with water interacting as a base with the amino group. gin of the aniline electronic transition by about 700 cm1. The geometrical parameters of the optimized structure of aniline and the two aniline–H2O conformers in the S0 and S1 3.2. Ab initio calculations: aniline–water complex states in Cs symmetry are shown in Table 1. It is to be noted that the aniline geometry in the complex is practically equal The theoretical description of the aniline–H2O complex to that of the free aniline. The amino inversion angle /, in the ground state provided by Spoerel and Stahl is in very defined as the angle between the ring and the amino group good agreement with their microwave data [17]. In the planes, is about 40° in the S0 state and the nitrogen atom is equilibrium conformation, water is hydrogen bonded to slightly displaced from the aromatic ring with a c angle the nitrogen lone pair of the amino group. The complex between the ring plane and the CN bond of about 3°. Fur- maintains the same Cs symmetry as aniline. The two sides thermore, for the S0 state the optimized structure of aniline of the aromatic ring are non-equivalent in the complex, agrees very well with the experimental bond distances and due to the pyramidal structure of the amino group in the angles [13,27]. The analysis of the available rotational S0 state. It is possible to distinguish two aniline–H2O con- spectra for the S1 state yields a large decrease in the A formers; anti and syn. The anti conformer has the nitrogen and an increase in the B rotational constants [15]. This atom of the amino group and the water molecule on the behavior suggests that the ring expands on the excitation, same side of the aromatic ring while the two amino hydro- and the molecule contracts along CN axis. In fact, the com- gen atoms are on the opposite side. In the case of the syn parison of the ab initio calculations in the S1 and S0 states conformer the hydrogen atoms of the amino group and shows that the CC bond lengths increase, the CH bond the water molecule are on the same side of the ring and lengths decrease and the CN bonds shortens (Table 1). the nitrogen atom is on the opposite side. These changes in the S1 are to be attributed to an increase We have optimized the structure of the two conformers in conjugation between the aromatic ring and the amino in the S0 and S1 electronic states. We have performed calcu- group [28,29].

Table 1 Experimental and calculated geometry of aniline in both ground and first electronic state for the isolated molecule and the two conformers of the aniline–

H2O complex Experimental Calculated a b c S0 S0 S1

An An An–H2O syn An–H2O anti An An–H2O syn An–H2O anti Distance (A˚ ) N–H 1.001(10) 1.019 1.018 1.020 0.997 0.997 1.001

C1–N 1.402(2) 1.411 1.406 1.418 1.346 1.341 1.359 C1–C2 1.397(2) 1.411 1.413 1.409 1.424 1.426 1.422 C2–C3 1.394(4) 1.403 1.403 1.403 1.413 1.414 1.412 C3–C4 1.396(3) 1.405 1.406 1.405 1.410 1.410 1.411 C2–H2 1.082(4) 1.096 1.096 1.096 1.081 1.081 1.080 C3–H3 1.083(2) 1.095 1.095 1.095 1.079 1.079 1.079 C4–H4 1.080(2) 1.094 1.094 1.094 1.082 1.082 1.082 Angles (°) HNH 113.1(100) 108.5 108.9 108.2 116.7 103.3 114.1

C6C1C2 119.4(9) 118.6 118.4 119.0 121.9 121.9 121.9 C3C4C5 118.9(17) 119.1 119.1 119.2 122.2 122.1 122.1 C1C2C3 120.1(9) 120.6 120.7 120.4 118.6 118.6 118.7 C2C3C4 120.7(17) 120.5 120.5 120.5 119.3 119.4 119.3 H2C2C3 120.1(17) 120.0 119.9 120.1 121.8 121.8 121.7 H2C3C2 119.4(17) 119.3 119.2 120.1 120.5 120.1 120.5 / 43d 47.0 39.1 46.4 18.6 9.7 30.0 c 0.6 3.0 3.4 3.2 1.1 0.7 2.1

H2O O–H 0.957 0.965 0.965 0.965 0.946 0.946 HOH 104.5 101.9 100.1 100.7 103.3 103.9 a Ref. [13]. b MP2/c-pvdz values. c CIS/cc-pvdz values. d Ref. [27]. G. Piani et al. / Chemical Physics 330 (2006) 138–145 143

Fig. 4 shows the optimized structure for the anti and syn Starting from the S0 anti structure, the calculation in the conformers for the aniline–H2O complex in the S0 and S1 S1 state yields another anti conformation but major differ- ˚ electronic states in the Cs symmetry. For the S0 state, it ences exist. The NH distance increases from 2.578 A in ˚ can be seen that in the anti conformer there is a hydrogen S0 to 3.876 A in S1 state. Therefore, there is significant bond between the amino nitrogen and the hydrogen of the elongation of the hydrogen bond upon electronic excita- water, while in the syn conformation the water molecule is tion. During the optimization of the S0 syn structure in positioned above the aromatic ring and far away from the S1 state, the complex acquires a new anti conformation. amino group. For the anti conformer, the hydrogen bond This conformation, labeled as anti II, (see Table 1 and NH distance and the NH–O (H2O) angle are about Fig. 4), is quite different to the previous one. The water 2.578 A˚ and 22.5°, respectively. This means that there is a molecule is above the aromatic ring and far away from non-linear NH–O hydrogen bond. The NO distance the amino nitrogen. All this means that the water molecule evaluated along the two bond is about 3.543 A˚ . For the is going to be only slightly bonded to the aniline molecule syn conformer, the closest hydrogen to the aromatic ring in S1. Moreover, the amino group is almost planar in S1. is located about 2.861 A˚ above the center of the ring and is displaced away by 0.36 A˚ toward the nitrogen. In this case, 3.2.1. Binding energy there is no hydrogen bond between the aniline nitrogen The MP2/cc-pvdz energy difference between the opti- atom and a hydrogen atom of water. This evidence could mized structures of the two conformers is about 870 cm1 be simply understood if we have in mind that, in the syn in the S0 state, with the anti structure being the most stable. configuration, the nitrogen lone pair is on the opposite side Because of the BSSE, the position of the minimum of the of the aromatic ring with respect to the water molecule. potential curve and the optimum interaction energy had Other local minima in the ground state are found, with to change upon counterpoise correction. We have calcu- the water interacting directly with amino group and placed lated the corrected potential energy surface as a function out of the aromatic ring. Those local minima will not be of the water position along the z-axis (perpendicular to discussed here as their calculated stabilization energy is the aromatic ring). Fig. 5 shows the potential energy curve smaller than that obtained for the Cs structure and the Cs of the aniline–H2O complex as a function of the water posi- structure is compatible with the experimental result pro- tion along the z-axis for the two conformers in S0 state. vided by microwave spectroscopy. The MP2 stabilization energy for both conformers which 1 1 For the S1 state we have optimized the geometry of the amounts to 1294 cm (3.7 Kcal/mol) and 862 cm complex starting from the two equilibrium structures (2.5 Kcal/mol) for the anti and syn conformers, respectively. obtained in the S0 state. We have directly evaluated also We have also repeated the same calculations (MP2) with other geometries without the symmetry plane and with the augmented aug-cc-pvdz basis set and the same behavior the water placed close to the aromatic ring or the amino has been found. The stabilization energies for the anti and group. We will discuss briefly here only those related to syn conformers are 1670 and 1064 cm1 (4.7 and 3.0 KCal/ the geometries calculated in the ground state in Cs mol), respectively, with a difference energy as large as symmetry. 600 cm1. These results agree with the experimental finding (5.17 Kcal/mol) [17]. In view of these results we conclude that the use of augmented basis set do not change the qual- itative behavior predicted for the complex in the S0 state

Fig. 4. The aniline–water complex calculated structures for ground and Fig. 5. Potential energy surface explored along the z-axis (perpendicular

first excited states in Cs symmetry. to aromatic ring plane). 144 G. Piani et al. / Chemical Physics 330 (2006) 138–145 although the energies are slightly affected by the basis set described for the anti II conformation, the water molecule employed. being far away from the nitrogen lone pair. The CAS energy For the S1 state in Cs symmetry, the CIS/cc-pvdz energy for the optimized anti II complex (361.686230 Hartree) is difference between the optimized structure of the two con- equal to that of the supermolecule (361.686230 Hartree) formers is about 132 cm1, with the anti conformation the showing that there is not an interaction between the water most stable one. The correct evaluation of the binding energy molecule and the aniline in the S1 state for a conformation is a rather difficult task due to the lack in size consistency of similar to the one observed in the ground state. The the CIS method. The way to perform this calculation sug- CASPT2 calculation in the excited state gives a small, gested by Fang is to evaluate the binding energy considering non-zero, stabilization energy only for the anti II conforma- the energy of the excited state of aniline and the energy of the tion (285 cm1). In comparison to the other computational ground state of water because of the localization of the tran- methods and to the experimental results, in the ground state sition [20]. Therefore, for a better description of the binding the CASPT2 calculation provides a rather large value for energy in S1, the structure and energy of separated aniline the stabilization energy in this system. However, the com- and water were determined by a supermolecule calculation. parison of the results obtained at CASPT2 level for the The supermolecule, including both monomers, is optimized ground and the excited state clearly indicate a large decrease with the same basis set as for the optimization of the com- in the intermolecular interactions with the electronic plex. In the resulting structure of the supermolecule, aniline excitation. and water are the same as the corresponding monomers, with Other geometries of the complex have been evaluated ˚ aniline–H2O separation of about 20 A. also in the excited state. Local minima are found with the In this way, the CIS/cc-pvdz binding energy is about oxygen atom of water bounded as a base to the hydrogen 362 cm1 (1.0 Kcal/mol) for the anti II conformation being atoms of the amino group but the stabilization energy is slightly more stable than the anti one (300 cm1 or always negligible. 0.9 Kcal/mol). We have also employed in a CIS calculation a larger basis set (aug-cc-pvdz) in order to reduce the BSSE 4. Conclusions error as pointed out by some authors [30,31]. In this case, the energy difference between the two optimized conform- The calculated property of the aniline–H O complex ers become larger, about 600 cm1, than in the S state, 2 0 (weaker interaction in the excited state and a rather large with the anti II structure being the most stable. By making difference in geometry between the ground and the first use of the supermolecule calculation we have corrected the electronically excited state) are consistent with the experi- energy values of the two conformers. In the case of the mental finding that shows a very weak, broad and blue anti II structure the energy of the supermolecule shifted signal for the REMPI spectrum with nanosecond (361.622601 Hartree) is equal to that of the optimized excitation. Similar spectral features can be related to small complex (361.622601 Hartree) indicating, therefore, that Frank–Condon coefficients for the electronic transition and the aniline and water in S state are not bounded. The 1 then to the contemporary excitation of many different other conformer has a slightly higher energy, about intermolecular vibrations with the electronic transition 12 cm1, showing then that the interaction between water (due to a large change of the equilibrium geometry of the and aniline is completely lost in the excited state. In order complex in the two states) and/or to a fast dissociation pro- to provide a common level of theoretical treatment for cess that takes place in the S potential energy surface of both the ground and the first electronic excited state we 1 the complex. The quantum calculations, we performed have also performed CAS(8,7) optimization of the complex point to the occurrence of a fast dissociation process with in both states. In the ground state, the CAS(8,7) stabiliza- the electronic excitation of the complex as a negligible tion energies agree satisfactorily with those of the MP2 cal- interaction energy is predicted in the excited state and the culations, indicating an even stronger preference for the changes in geometry are quite large. The occurrence of a anti conformation. The CAS(8,7) stabilization energy for fast dissociation process could be shown with fs time- the syn conformer is calculated at 748 cm1, very close to resolved pump and probe experiments and we are currently the value obtained with MP2 calculation (887 cm1). In working in this direction. These properties seem to be quite the case of the anti conformer, the CAS method gives a general for molecules containing an aromatic amino group much larger value for the interaction energy (1883 cm1) hydrogen bounded to acidic groups, as the same behavior than the MP2 method (1320 cm1). This trend is ever more is shown by both aniline–water and aniline–methanol clus- emphasized in the CASPT2 calculation that gives a stabil- ters: two systems with very similar intermolecular interac- ization energy of 3307 and 1259 cm1 for the anti and syn tions and equilibrium geometry. conformer, respectively. In summary, for the ground state all methods indicate a stronger relative stabilization for the anti conformer with respect to the syn conformer. Acknowledgements In the first electronic excited state, the CAS(8,7) calcula- tion yields a similar result to that obtained by CIS calcula- Financial support from Italian MIUR and EU (under tions with a structure much similar to that previously Contract No. RII3-CT-2003-506350) is kindly acknowl- G. Piani et al. / Chemical Physics 330 (2006) 138–145 145 edged. ILT gratefully acknowledges to Access to Research [13] D.G. Lister, J.K. Tyler, J.H. Hog, N.W.J. Larson, Mol. Struct. 23 Infrastructures EU program for financial support during (1974) 253. her stay in Florence. [14] B. Kleibo¨mer, D.H.Z. Sutter, Naturforsch 43a (1988) 561. [15] E.R.Th. Kerstel, M. Becucci, G. Pietraperzia, E. Castellucci, Chem. Phys. 199 (1995) 263. References [16] M.S. Bowen, M. Becucci, R.E.J. Continetti, Phys. Chem. A 109 (2005) 11781. [1] K. Mu¨ller-Dethlefs, O. Dopfer, T.G. Wright, Chem. Rev. 94 (1994) [17] U. Spoerel, W.J. Stahl, Mol. Spectrosc. 190 (1998) 278. 1845. [18] M. Haeckel, W.J. Stahl, Mol. Spectrosc. 198 (1999) 263. [2] B. Brutschy, Chem. Rev. 100 (2000) 3891. [19] J. Hager, M. Smith, S.J. Wallace, Chem. Phys. 83 (1985) 4820. [3] M. Gerhards, M. Schmitt, K.J. Kleinermanns, Chem. Phys. 104 [20] W.H.J. Fang, Chem. Phys. 112 (2000) 1204. (1996) 967. [21] P. Hobza, H.L. Selzle, E.W. Schlag, Chem. Rev. 94 (1994) 1767. [4] G. Berden, W.L. Meerts, M. Schmitt, K.J. Kleinermanns, Chem. [22] I. Lo´pez-To´con, J.C. Otero, M. Becucci, G. Pietraperzia, E. Castel- Phys. 104 (1996) 972. lucci, Chem. Phys. 249 (1999) 113. [5] M. Becucci, G. Pietraperzia, M. Pasquini, G. Piani, A. Zoppi, R. [23] S.F. Boys, F. Bernardi, Mol. Phys. 19 (1970) 553. Chelli, E. Castellucci, W.J. Demtroeder, Chem. Phys. 120 (2004) [24] P. Salvador, M. Duran, J.J. Dannenberg, J. Phys. Chem. A 106 (2002) 5601. 6883. [6] B.M. Giuliano, W. Caminati, Angew. Chem. Int. Ed. 44 (2005) 603. [25] M.J. Frisch et al., GAUSSIAN 98, Revision A.7, Gaussian Inc., [7] J.W. Ribblet, W.E. Sinclair, D.R. Borst, J.T. Yi, D.W. Pratt, J. Phys. Pittsburgh, Pennsylvania, 1998. Chem. A 110 (2006) 1478. [26] M. Pasquini, G. Piani, G. Pietraperzia, W. Demtroeder, M. Giuntini, [8] C. Eisenhardt, M. Becucci, G. Pietraperzia, Phys. Chem. Chem. Phys. M. Becucci, Rev. Sci. Instrum. 76 (2005) 113105. 3 (2001) 1407. [27] J. Christoffersen, J.M. Hollas, G.H. Kirby, Mol. Phys. 16 (1969) 441. [9] N.J. Kim, H. Kang, G. Yeong, Y.S. Kim, K.T. Lee, S.K.J. Kim, [28] A.D. Gorse, M.J. Pesquer, Mol. Struct. (Theochem) 281 (1993) 21. Phys. Chem. A 104 (2000) 6552. [29] J.C. Jiang, C.E.J. Lin, Mol. Struct. (Theochem) 392 (1997) 181. [10] H. Kang, K.T. Lee, S.K. Kim, Chem. Phys. Lett. 359 (2002) 213. [30] D.W. Schwenke, D.G.J. Truhlar, Chem. Phys. 82 (1985) 2418. [11] A.F. Jalbout, L.J. Adamowicz, Phys. Chem. A 105 (2001) 1033. [31] M.J. Frish, J.E. del Bene, J.S. Binkley, H.F.J. Schaeffer III, Chem. [12] H.T.J. Kim, Mol. Struct. (Theochem) 673 (2004) 121. Phys. 84 (1986) 2279.