Behind the Reactivity of Lactones: A Computational and Spectroscopic

Article pubs.acs.org/JPCA

Behind the Reactivity of Lactones: A Computational and Spectroscopic Study of Phenol·γ-Butyrolactone Iker León,*,† Jorge González,† Judith Millán,‡ Fernando Castaño,† and José A. Fernández*,† †

Departamento de Quı ́mica Fı ́sica, Facultad de Ciencia y Tecnologı ́a, Universidad del Paı ́s Vasco-UPV/EHU, B. Sarriena s/n, Leioa 48940, Spain ‡ Departamento de Quı ́mica, Facultad de Ciencias, Estudios Agroalimentarios e Informática, Universidad de La Rioja, Madre de Dios 51, Logroño 26006, Spain S Supporting Information *

ABSTRACT: In this work, the intermolecular interaction between phenol and γ-butyrolactone (GBL) has been studied by a combination of spectroscopic and computational techniques. The electronic and vibrational transitions of phenol·GBL were measured in a supersonic jet expansion by resonant two-photon ionization (R2PI) and ion dip IR (IDIR) spectroscopy. The results obtained were compared with calculations carried out with both M06-2X and MP2 molecular orbital methods in order to characterize the intermolecular interactions. The singly detected conformer is stabilized by a relatively strong hydrogen bond in which phenol acts as a proton donor to the carbonyl group of GBL. The phenol·GBL2 cluster has also been studied, finding up to three populated conformers. Nevertheless, in the three species, the main interaction between the phenolic hydroxyl group and the GBL’s carbonyl group remains similar to that of phenol·GBL. Furthermore, the CO···H interaction is reinforced.

■

INTRODUCTION γ-Butyrolactone (GBL) has received public attention because it is used as a prodrug for the synthesis of γ-hydroxybutyric acid (GHB), a popular recreational drug.1,2 While GHB is a controlled substance in some countries, GBL, the intramolecular ester (lactone) of GHB, is a freely available chemical and thus serves as starting point in the synthesis of GHB. GBL is converted to GHB via hydrolysis and re-esterification by a mechanism that interconverts the GBL closed-ring ester to the open-chain hydroxycarboxylic acid3 (see Scheme 1). Although there are some studies in alkaline media,4,5 slightly acidic media are more relevant in biologic terms6 and have not received much attention.7 Under acidic conditions, lactones are in reversible equilibrium with the corresponding open-chain hydroxycarboxylic acids.3 Scheme 1 shows the basic mechanism of this reaction; in a first step a protonation of GBL’s CO oxygen atom occurs, followed by a nucleophilic attack, leading to the ring-opening. In this work, we test whether this preprotonation interaction of GBL in the gas phase takes place at the carbonyl or ester oxygen atoms and whether there is a preference for a particular © 2014 American Chemical Society

binding site. An effective way to examine this interaction is the formation of noncovalent species between GBL and a biologically relevant acidic molecule by means of supersonic molecular beams as the cold (few K) isolated environment of the expansion allows exploration of their conformational preferences without perturbation of their (rovibronic) energy levels. Furthermore, while there exist some studies carried out in solution,4,5,8−10 only some studies based on its cyclation in the gas phase have been carried out for either GBL or GHB.11 The combination of molecular beams with laser-based spectroscopic methods, such as resonant two-photon ionization (R2PI) or ion dip IR (IDIR) double resonance spectroscopy, is a powerful approach for analyzing weakly bound clusters because of its mass and conformer selectivity. Here, phenol was chosen because of its chemical relevance, the presence of an aromatic chromophore, and its acidic character.12 Several studies have examined how electronic excitation increases Received: October 18, 2013 Revised: March 19, 2014 Published: March 19, 2014 2568

dx.doi.org/10.1021/jp4103417 | J. Phys. Chem. A 2014, 118, 2568−2575

The Journal of Physical Chemistry A

Article

chemistry to fully characterize the clusters. Using molecular mechanics, the conformational landscape of the clusters is thoroughly explored, and the structures found are further optimized at M06-2X/6-311++G(d,p) and MP2/6-311++G(d,p) methods, which have been reported to yield quantitative results for this kind of system.14,15

Scheme 1. (a) Chemical Equilibrium between GBL and GHB and (b) Acidic Hydrolysis Interconversion of GBL to GHB

■

METHODS

Experimental Section. A detailed description of the experimental system can be found elsewhere,14 and therefore, a very short summary is offered here. Phenol (99.5% Fluka) and GBL (>99% Aldrich) were seeded in He without further modifications, and the mixture was expanded inside of the chamber of a linear time-of-flight (Jordan Inc.). Heating the sample to 60 °C was enough to obtain sufficient vapor pressure to carry out the experiments. Typical backing pressures of 1−2 bar were employed in the expansion. After crossing a 1 mm skimmer, the supersonic beam was interrogated with a variety of UV and IR laser-based spectroscopic techniques in the ionization region. Two-color REMPI was obtained by setting the laser (second color) at 28169 cm−1, while UV/UV and IR/UV16−18 double resonance experiments were carried out with active subtraction in order to improve the S/N ratio (see ref 14 for a detailed description on such experiments). Calculations. An exploration of the intermolecular energy surface was done using a fast molecular mechanics method (MMFFs)19 to locate the most significant minima in a 25 kJ/ mol window. The conformational search was done using the Macromodel’s conformational search tool in the Schrodinger’s suite.20 No geometric restriction was imposed on the clusters. The structures were generated through an advanced search algorithm that uses a combination of the Monte Carlo

phenol’s acidity. The use of a supersonic expansion makes it possible to generate stable phenol·GBL complexes in the ground state, allowing the spectroscopic exploration of this system. In this paper, we rationalize the results obtained on the phenol·GBL cluster in terms of the intermolecular interactions between the monomers in order to shed light on the reactivity of the cyclic esters. The results are of current interest as many biologically important molecules contain cyclic phosphates and lactone structures.13 Introduction of a second GBL molecule to yield phenol·GBL2 has also been studied as an approach to simulate the evolution toward a more concentrated media. The experimental results are combined with computational

Figure 1. Two-color REMPI spectrum of phenol·GBL and phenol·GBL2 recorded in the 35400−37000 cm−1 region. The spectra of phenol, phenol· W1, and phenol2 are also shown for comparison. 2569



Article

procedures and ‘‘Large-scales Low-Mode’’ (which uses frequency modes to create new structures) and using the MMFFs force field. A minimum of 100 000 runs was performed for each stoichiometry. All of the structures found for phenol· GBL were subjected to full optimization using DFT and MP2 methods. More than 100 structures were found for phenol· GBL2, and thus, they were grouped into families. The most stable structures and at least one member of each family were considered during the optimization step. After careful examination of the obtained structures, around 50 of them were selected for optimization using DFT and MP2 methods. The complete set of optimized structures can be found in Figures S1−S4 of the Supporting Information. Tables S1 and S2 (Supporting Information) collect the energetics of the optimized structures. The final stable structures are named as 1, 2, ..., etc., where the number refers to their relative stability, starting with the most stable one. The structural parameters of the assigned species are collected in Table S3 (Supporting Information), and a detailed view is shown in Figures S7−S9 (Supporting Information). The calculations were performed using the Gaussian 09 (rev A.02) suite.21 The M06-2X/6-311++G(d,p) calculation level was used for optimization and normal-mode analysis as it was demonstrated to give accurate descriptions of both hydrogen bond and dispersion interactions.22,23 More precisely, it has been demonstrated in a recent paper that this calculation level accurately predicts the structure of the phenol dimer and trimer, as the comparison with results from MW spectroscopy demonstrates.24 All of the optimizations were accompanied by normal-mode analysis to confirm that the resulting structures are true minima. The relative energy values given in the present work are basis set superposition error (BSSE)- and zero-point energy (ZPE)-corrected. All of the structures were also recalculated at the MP2/6-311++G(d,p) level as a further test on the validity of the computational procedure.

Figure 2. Two-color REMPI spectrum of phenol·GBL recorded in the 35700−36700 cm−1 region together with the hole burning trace. The asterisk denotes the vibronic band used for the probe laser to record both the IDIR and hole burning spectra.

published data.16 Figure 3 shows that a single peak at 3409 cm−1, due to the stretching mode of the phenolic hydroxyl group, is found for phenol·GBL. Such a band is shifted 246 cm−1 from the phenol OH stretch, suggesting the formation of a hydrogen bond in which phenol is the proton donor. Such a red shift (246 cm−1) is even greater than those found for the phenol dimer25 and Ph(OH)···OH2,26 whose OH stretches are located at 3529 and 3522 cm−1, respectively. This fact suggests a stronger tendency to share the proton in phenol·GBL than in the other systems. Figure 3 also shows our theoretical predictions of the IR spectra for phenol, phenol·W1, and phenol2, in good agreement with the values in the literature.12,16,25 The calculations were done at the same level of theory than those for phenol·GBL for calibration purposes. It is important to have additional molecular references as election of an incorrect scaling factor may result in incorrect assignments for the single OH stretching mode in phenol·GBL. As observed in Figure 3, a correction factor of 0.934 satisfactorily reproduces the experimental shifts for all clusters; therefore, it was adopted in this work. Furthermore, this correction factor is in good agreement with those employed in previous experiments with the same calculation level.27−29 Figure 4 shows the experimental IDIR spectrum of phenol· GBL, together with some of the calculated structures and their predicted OH stretching vibrations (the rest of the structures can be found in Figures S1 and S2 of the Supporting Information). In the most stable structure 1, phenol is placed above GBL, acting as a proton donor to the GBL’s carbonyl group. Although the hydrogen bond somehow contributes to the cluster binding energy, the relative position of the two molecules is not optimum for such an interaction as the carbonyl’s electron lone pairs are located in the same plane as the CO group due to the sp2 hybridization, and therefore, it is not the primary interaction. This observation is reflected in the angle between the OH bond and the oxygen, which is ∼150°

■

RESULTS AND DISCUSSION Phenol·GBL. Figure 1 shows the two-color (2c) REMPI spectra of phenol·GBL and phenol·GBL2, together with those of phenol, phenol·W1, and phenol2 for comparison. Clearly, phenol·GBL presents a richer spectroscopy than phenol·W1. Phenol·GBL2 shows an even larger abundance of vibrational features than phenol·GBL or even the phenol dimer. It is also worthy to note that the red shift in the electronic origins of both phenol·GBL and phenol·GBL2 is larger than that for the other two clusters. Figure 2 shows the 2c-REMPI spectrum of phenol·GBL together with the “hole burning” spectrum recorded tuning the depopulation laser at 35876 cm−1. The red-most peak is located at 35876 cm−1, thus accounting for a 475 cm−1 red shift from the phenol 000 transition (36350 cm−1).16 This shift is considerably larger than that for phenol·W1,12 whose 000 transition is located at 35996 cm−1; that is, it is red-shifted by 355 cm−1. The hole burning spectrum shows that the most prominent transitions are due to the presence of a single isomer. Despite this fact, we cannot completely rule out the presence of additional weaker conformers. Figure 3 shows the IR spectrum of the detected phenol·GBL conformer, recorded using IR/UV, together with the IR spectra for phenol, phenol· W1, and phenol2, previously measured by other groups (refs 16, 12, and 23, respectively) and reanalyzed for this work. As can be observed, the OH stretching mode of a phenol bare molecule is located at 3655 cm−1, as expected from the 2570



Article

Figure 4. (Upper panel) Some of the calculated structures for phenol· GBL at the M06-2X/6-311++G(d,p) level. The relative stability of each structure is also shown in brackets. (Lower panel) IR spectra of the detected phenol·GBL conformer, recorded using IR/UV, together with the normal modes for some selected calculated conformers. The complete set of calculated IR spectra can be found in Figure S2 (Supporting Information).

Figure 3. IR spectra of the detected phenol·GBL conformer, recorded using IR/UV, together with the IR spectra of the phenol2, phenol·W1, and phenol. The OH stretches for the assigned structures in the literature, calculated at M06-2X/6-311++G(d,p), show that a correction factor of 0.934 may be applied in all cases to account for anharmonicity.

However, the absence of the most stable structure challenges both the assignment and the stability predicted by the computations. In order to rule out calculation pitfalls, all of the structures were also computed at the MP2/6-311++G(d,p) level. The results are shown in Table 1. Apart from an overall tendency of MP2 toward overestimating the interactions due to dispersive forces,32−35 no other significant variation was found. Furthermore, the comparison between the experimental IR spectra and those calculated at MP2 (Figure S5 of the Supporting Information) leads to the same conclusion as that with M06-2X, further reinforcing our assignment. In this case, a correction factor of 0.943 was employed after comparison with the results for phenol, phenol·W1, and phenol dimer (see Figure S6 of the Supporting Information). Finally, the calculations were also computed at the M06-2X/aug-cc-pVTZ level to test the effect of the basis set size. However, the results show no significant differences with previous results using M062X and MP2 as structure 1 remains as the global minimum. Therefore, the conformational energies predicted by the theory seem to be consistent, and a different reason must be behind the absence in the spectrum of the structure calculated as most stable. A likely possibility is a significant contribution of the entropy to the stability of the conformers at the preexpansion temperature. Thus, we calculated the Gibbs energies at the valve’s temperature (333.15 K) for the most stable conformers, and the results are shown in Table 1. It is clear from the table that the entropic contributions favor the hydrogen-bonded

for a distance of 2 Å. The main binding force is thus of dispersive nature and takes place between the two rings. We observed in Figure 4 that the predicted spectrum for such a structure does not reproduce the experimental results. Those structures in which the phenolic hydroxyl group has a weak interaction with one of the lone pairs of the GBL’s ester oxygen, such as structure 2, are 4 kJ/mol higher in energy. These types of structures can also be discarded as the predicted OH stretching mode lies ∼200 cm−1 to the blue of the experimental peak. Structure 9 is representative of those isomers where both molecules are stabilized by dispersive forces and where the phenol’s hydroxyl group is free from any interaction. Consequently, they present a peak at ∼3650 cm−1, which is again too far from the experimental determination. The spectra predicted for the other two structures 4 and 6 in Figure 4, in which the phenolic hydroxyl group presents a relatively strong hydrogen bond to the GBL’s carbonyl group, match very well the experimental results. The difference between both conformations is the relative orientation of the two molecules, perpendicular in structure 4 or almost parallel in structure 6 and, therefore, lacking the extra stabilization energy due to the ring−ring interaction. It is worth noting the similarity between structure 4 and that of the phenol dimer,24,25,30,31 which seems to reinforce our assignment. 2571



Article

electronic excitation, from 1.89 to 1.77 Å, although the structure of the neutral and excited states barely suffers from any change. On the other hand, structure 1 clearly experiences a large change in the geometry, potentially precluding its detection. In fact, the structure finally reached using different calculation methods is closer to that of the assigned conformer. Figure 6 shows the assignment proposed for phenol·GBL and phenol·GBL2, with the PhO−H···OC distance and angles highlighted. The same parameters for phenol·W112 and phenol·dimer,25 calculated at the same level of theory, are also shown for a direct comparison. Considering that the hydrogen bond angle and distance is significantly closer to ideality for phenol·W1 and phenol2 than those in phenol·GBL and phenol· GBL2, one would expect a stronger hydrogen bond in the former clusters, contrarily to what we observe assuming a direct relationship between the OH stretching shift and hydrogen bond strength. Consequently, the extra strength of the CO··· H interaction must be due to the ability of the oxygen atom to accept the proton as it also drags electron density from the carbonyl group, greatly increasing GBL’s affinity for accepting phenol’s proton. Furthermore, the spectral shifts also indicate whether phenol acts as a proton donor or acceptor. In molecules with an OH attached to an aromatic ring, such as phenol, the shift can be attributed to the reinforcement of the inductive effect produced by the phenol’s oxygen atom on the π* ← π transition, caused by the hydrogen bond. The π* ← π transition transfers electron density from the oxygen atom to the aromatic ring. Thus, structures with a larger electron density on the oxygen atom will be more stabilized in the excited state than in the ground state, resulting in a red shift on the electronic origin relative to the monomer. Therefore, if phenol acts as proton donor (acid), the formation of a hydrogen bond leads to a red shift. Such is the case of phenol·W1 or phenol2. In this line, our results clearly highlight that the shift in the excitation energy is an unpredictable property of noncovalent systems because it is the result of subtle changes in several weak interactions. Thus, in phenol·GBL, there are additional interactions that lead to the observation of a shift much greater than expected, as, for example, the contribution from the interaction between the phenolic and ester rings. Phenol·GBL2. The REMPI and hole burning spectra of phenol·GBL2 are shown in Figure 7. An increase in the number of experimental conformers, from one to three, is observed with the addition of a second GBL molecule. The first and third conformers present relatively close vibronic origins at 35675 and 35888 cm−1, respectively, while the second conformer’s 000 transition is not that evident as it is midway between both conformers. It is worthy to note that the 000 transition of conformer 3 almost overlaps with that of phenol·GBL, while the first and second conformers’ origins are shifted to the red. Figure 8 shows the IDIR spectra of the three detected conformers together with the predicted vibrational spectra for some of the calculated structures (the rest of the structures can be found in Figures S3 and S4 of the Supporting Information). The three conformers present a strong hydrogen bond. Also, the shift to the red in the OH stretch seems to correlate with the shift in the 000 transition, as can be inferred from the peaks located at 3356, 3364, and 3404 cm−1, respectively. Assignment of the detected isomers to the calculated structures by simple comparison is not easy as all the structures below 15 kJ/mol present IR spectra that reproduce the position of the observed OH stretch. Nevertheless, in all of them, the

Table 1. Comparison of the Relative Energies for the Phenol·GBL Structures, Calculated at M06-2X/6-311+ +G(d,p) and MP2/6-311++G(d,p) Levelsb M06-2X/6-311++G(d,p)

MP2/6-311++G(d,p)

structure

ΔE kJ/mol

ΔEZPE kJ/mol

ΔG333.15K kJ/mola

structure

ΔE kJ/mol

ΔEZPE kJ/mol

1 2 3 4c 5 6c 7 8c 9 10 11 12 13 14 15 16 17 18 19 20 22

0.00 4.88 6.27 7.51 11.00 14.50 14.65 18.58 21.37 21.47 21.60 23.74 22.46 25.01 26.67 27.26 30.94 31.31 31.23 31.66 34.36

0.00 4.14 4.25 5.47 8.93 11.41 12.11 16.32 18.31 18.35 19.36 20.13 20.28 21.89 23.02 23.58 26.65 26.66 27.02 27.45 28.93

1.73 5.91 2.69 0.00 8.57 2.86 5.85 4.58 14.81 14.86 16.90 13.97 18.05 15.84 16.93 17.39 20.41 17.07 20.63 20.27 21.27

1 2 3 4b 5 10 7b 9 11 13 14 12b 6c 15 16 8c 20 18 17 19 22

0.00 2.04 4.85 − 7.47 17.64 − 17.25 19.01 18.17 20.62 − 17.81 22.58 22.58 20.87 26.23 29.49 29.48 29.37 30.61

0.00 1.72 4.03 − 6.03 14.87 − 15.10 15.70 16.50 17.28 − 18.44 19.62 19.62 21.28 23.03 25.45 25.72 25.83 26.23

The ΔG column indicates the sum of electronic and thermal free energies at 333.15 K. bThe structures converge into slightly different structures for the MP2/6-311++G(d,p) calculation method. cThese structures present their predicted IR spectrum in agreement with the experimental results. a

species (such as structures 4, 6, and 8) over the stacked ones. In fact, the assigned species becomes the most stable at 333.15 K, while the structure previously predicted to be the global minimum at 0 K is now 1.7 kJ/mol above the energy origin. The complete absence of structure 1 can be explained due to large conformational changes in the excited state that hamper its detection. Calculations of the excited state using the timedependent (TD) M06-2X/6-311++G(d,p) method for the global minimum (structure 1) and the assigned structure (structure 4) are shown in Figure 5. Structure 4 presents a reduction in the O−H···O hydrogen bond distance upon

Figure 5. Comparison between the structures of phenol·GBL in the neutral and excited states. The left panel shows the most stable structure 1 at 0 K, while the right panel shows the structure for the assigned species (structure 4). The atoms of the excited states are depicted as darkened spheres for a better comparison. Both structures were calculated at the M06-2X/6-311++G(d,p) level. 2572



Article

Figure 6. (Upper panel) Phenol·W1 and phenol2 structures as reported in refs 12 and 23 and reoptimized at the M06-2X/6-311++G(d,p) level. (Lower panel) Phenol·GBL and phenol·GBL2 assigned structures. The hydrogen bond angles and distances (in Ångstroms) are also shown.

Figure 8. (Upper panel) Some of the calculated structures for phenol· GBL2 at the M06-2X/6-311++G(d,p) level. The relative stability of each structure is also shown in brackets. (Lower panel) IR spectra of the detected phenol·GBL2 conformers, recorded using IR/UV, together with the normal modes for some selected calculated conformers. The complete set of calculated IR spectra can be found in Figure S4 (Supporting Information).

Figure 7. Two-color REMPI spectrum of phenol·GBL2 recorded in the 35400−37000 cm−1 region together with the hole burning traces. The asterisks denote the vibronic bands probed to record both IDIRS and hole burning spectra for each of the conformers.

phenol molecule is strongly bound to the electronic pair of the CO moiety of one of the GBL molecules, adopting a structure similar to that of phenol·GBL, while the second GBL is placed above phenol, interacting with its partner by dispersive

forces. As an example, the conformers 1 and 3 shown in Figure 8 are two of the most stable structures among those whose predicted spectrum is in agreement with the experimental 2573



Article

(Consolider-Ingenio 2010/CSD2007-00013, CTQ2011-22923, and CTQ2009-14364). UPV/EHU (UFI 11/23).

results. The second GBL molecule introduces an extra stabilization due to a C−H···π interaction. In the cases where there are other significant structural changes, as in structures 34, 36, or 37 in Figure 8, the position of the OH stretch is significantly shifted with respect to the experimental determination. It must be noted that both IDIR and REMPI experiments indicate that phenol·GBL2 first and second conformers present a stronger hydrogen bond than that of phenol·GBL. Surprisingly, addition of more GBL molecules makes the C O proton affinity even greater, facilitating the acyl−oxygen cleavage observed in solution during the hydrolysis of this molecule. Reinforcement of an intermolecular hydrogen bond upon complexation with a third partner has been observed before,36,37 and here, we present another interesting case.

■

ABBREVIATIONS IR/UV, infrared/ultraviolet double resonance; REMPI, resonance enhanced multiphoton ionization; IDIR, and ion dip IR

■

■

CONCLUSIONS The phenol·GBL and phenol·GBL2 molecular clusters were characterized in a jet expansion using a number of massresolved laser spectroscopic techniques and computational chemistry methods. For phenol·GBL, a single conformer was found, while for phenol·GBL2, the spectroscopy gets more complicated due to the presence of three conformers. In both systems and for all of the detected conformers, both the electronic spectra and the IDIR experiments suggest that the phenol molecule acts as a proton donor to the carbonyl group of GBL by a relatively strong hydrogen bond. This supports the reaction mechanism by which GBL lactose ring-opening is produced by the protonation of the carbonyl oxygen followed by a nucleophilic attack. Using laser spectroscopy, we confirm that this noncovalent bond between the carbonyl oxygen and the phenol hydrogen is even stronger than that for phenol·W1. The calculated structural parameters further support our assignment.

■

ASSOCIATED CONTENT

S Supporting Information *

All calculated structures, predicted IR spectra, and the computed relative stability of the species studied in this work. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

(1) Ropero-Miller, J. D.; Goldberger, B. A. Clin. Lab. Med. 1998, 18, 727−746. (2) Elsohly, M. A.; Salamone, S. J. Anal. Toxicol. 1999, 23, 141−146. (3) March, J. Advanced Organic Chemistry; 3rd ed.; John Wiley & Sons: New York, 1985. (4) Ciolino, L.; Mesmer, M.; Satzger, D.; Machal, C.; McCauley, H.; Mohrhaus, A. J. Forensic Sci. 2001, 46, 1315−1323. (5) Hennesy, S.; Moane, S.; MacDermott, S. J. Forensic Sci. 2004, 49, 1220−1229. (6) Van Duuren, B. L.; Goldschmidt, B. M. J. Med. Chem. 1966, 9, 77. (7) Coffin, F. D.; Long, F. A. J. Am. Chem. Soc. 1952, 74, 5767. (8) Pérez-Prior, M. T.; Manso, J. A.; García-Santos, M. P.; Calle, E.; Casado, J. J. Org. Chem. 2005, 70, 420−426. (9) DeFrancesco, J. V.; Witkowski, M. R.; Ciolino, L. A. J. Forensic Sci. 2006, 59, 321. (10) Brewster, V. L.; Edwards, H. G. M.; Hargreaves, M. D.; Munshi, T. Drug Test. Anal. 2009, 1, 25. (11) Bordejé, M. C.; Mó, O.; Yáñez, M.; Herreros, M.; Abboud, J.-L. M. J. Am. Chem. Soc. 1993, 115, 7389−7396. (12) Watanabe, T.; Ebata, T.; Tanabe, S.; Mikami, N. J. Chem. Phys. 1996, 105, 408−419. (13) Kaiser, E. T.; Kézdy, F. J. Prog. Bioorg. Chem. 1976, 4, 239. (14) León, I.; Millán, J.; Castaño, F.; Fernández, J. A. ChemPhysChem 2012, 13, 3819. (15) León, I.; Cocinero, E. J.; Millán, J.; Rijs, A. M.; Usabiaga, I.; Lesarri, A.; Castaño, F.; Fernández, J. A. J. Chem. Phys. 2012, 137, 074303−1. (16) Pribble, R. N. Science 1994, 265, 75. (17) Tanabe, S.; Ebata, T.; Fujii, M-; Mikami, N. Chem. Phys. Lett. 1993, 215, 347−352. (18) Riehn, C.; Lahmann, Ch.; Wasserman, B.; Brutchy, B. Chem. Phys. Lett. 1992, 197, 443. (19) Halgren, T. A. J. Comput. Chem. 1999, 20, 730−748. (20) MacroModel, version 9.9 (Suite 2012); Schrödinger, LLC: New York, 2012. (21) Frisch, M.; et al. Gaussian 09, revision A02; Gaussian, Inc.: Wallingford, CT, 2009. (22) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157−167. (23) Hohenstein, E. G.; Chill, S. T.; Sherrill, C. D. J. Chem. Theory Comput. 2008, 4, 1996−2000. (24) Seifert, N. A.; Steber, A. L.; Neill, J. L.; Pérez, C.; Zaleski, D. P.; Pate, B. H.; Lesarri, A. Phys. Chem. Chem. Phys. 2013, 15, 11468. (25) Ebata, T.; Kayano, M.; Sato, S.; Mikami, N. J. Phys. Chem. 2001, 105, 8623−8628. (26) Ebata, T.; Mizuochi, N.; Watanabe, T.; Mikami, N. J. Phys. Chem. 1996, 100, 546−550. (27) León, I.; Cocinero, E. J.; Lesarri, A.; Castaño, F.; Fernández, J. A. J. Phys. Chem. A 2012, 116, 8934−8941. (28) León, I.; Cocinero, E. J.; Rijs, A. M.; Millán, J.; Alonso, E.; Lesarri, A.; Fernández, J. A. Phys. Chem. Chem. Phys. 2013, 15, 568− 575. (29) León, I.; Montero, R.; Castaño, F.; Longarte, A.; Fernández, J. A. J. Phys. Chem. A 2012, 116, 6798−6803. (30) Schmitt, M.; Böhm, M.; Ratzer, C.; Krügler, D.; Kleinermanns, K.; Kalkman, I.; Berden, G.; Meerts, W. L. ChemPhysChem 2006, 7, 1241−1249. (31) Steber, A.; Neill, J.; Zaleski, D.; Pate, B.; Lesarri, A.; Bird, R.; Vaquero-Vara, V.; Pratt, D. Faraday Discuss. 2011, 150, 227−242. (32) Hobza, P.; Riehn, C.; Weichert, A.; Brutchy, B. Chem. Phys. 2002, 283, 331−339.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Phone + 34 94 601 5387. Website: https://sites.google.com/site/gesemupv/ (I.L.). *E-mail: [email protected]. Fax: + 34 94 601 35 00. Phone + 34 94 601 5387. Website: https://sites.google.com/ site/gesemupv/ (J.A.F.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Technical and human support provided by the Laser Facility of the SGIKER (UPV/EHU, MICINN, GV/EJ, ESF) is gratefully acknowledged. I.L. thanks the GV for predoctoral and postdoctoral fellowships. J.G. thanks the UPV/EHU for a predoctoral fellowship. Computational resources from the SGI/ IZO-SGIker and from the i2BASQUE academic network were used for this work. The research leading to these results has received funding from the Spanish MICINN and MINECO 2574



Article

(33) Kolar, M.; Hobza, P. J. Phys. Chem. A 2007, 111, 5851−5854. (34) León, I.; Cocinero, E. J.; Millan, J.; Jaeqx, S.; Rijs, A. M.; Lesarri, A.; Castaño, F.; Fernández, J. A. Phys. Chem. Chem. Phys. 2012, 14, 4398−4409. (35) León, I.; Millán, J.; Cocinero, E. J.; Lesarri, A.; Castaño, F.; Fernández, J. A. Phys. Chem. Chem. Phys. 2012, 14, 8956. (36) Seurre, N.; Sepiol, J.; Lahmani, F.; Zehnacker-Rentien, A.; Barbu-Debus, K. L. Phys. Chem. Chem. Phys. 2004, 6, 4658−4664. (37) Brenner, V.; Millie, P.; Broquier, M.; Lahmani, F.; ZehnackerRentien, A.; Peremans, A. J. Phys. Chem. A 2001, 105, 6841−6850.

2575


Behind the Reactivity of Lactones: A Computational and Spectroscopic

Recommend Documents