Article pubs.acs.org/JPCA
Photodissociation Dynamics of Enolic 1,2-Cyclohexanedione at 266, 248, and 193 nm: Mechanism and Nascent State Product Distribution of OH Monali Kawade, Ankur Saha, Hari P. Upadhyaya,* Awadhesh Kumar, and Prakash D. Naik Radiation & Photochemistry Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India S Supporting Information *
ABSTRACT: The photodissociation dynamics of 1,2-cyclohexanedione (CHD), which exists in enolic form in gas phase, is studied using pulsed laser photolysis (LP)-laser induced fluorescence (LIF) “pump-and-probe” technique at room temperature. The nascent state distribution of the OH radical, formed after initial photoexcitation of the molecule to it is (π, π*) and Rydberg states, is determined. The initial (π, π*) and Rydberg states are prepared by excitation with the fourth harmonic output of Nd:YAG (266 nm)/KrF (248 nm) and ArF (193 nm) lasers, respectively. The ro-vibrational distribution of the nascent OH photofragment is measured in collision-free conditions using LIF. The OH fragments are formed in the vibrationally cold state at all the above wavelengths of excitation but differ in rotational state distributions. At 266 nm photolysis, the rotational population of OH shows a curvature in Boltzmann plot, which is fairly described by two types of Boltzmann-like distributions characterized by rotational temperatures of 3100 ± 100 and 900 ± 80 K. However, at 248 nm photolysis, the rotational distribution is described by a single rotational temperature of 950 ± 80 K. The spin−orbit and Λ-doublets ratios of OH fragments formed in the dissociation process are also measured. The average translational energy in the center-of-mass coordinate, partitioned into the photofragment pairs of the OH formation channels, is determined to be 12.5 ± 3.0, 12.7 ± 3.0, and 12.0 ± 3.0 kcal/mol at 266, 248, and 193 nm excitation, respectively. The energy partitioning into various degrees of freedom of products is interpreted with the help of different models, namely, statistical, impulsive, and hybrid models. To understand the nature of the dissociative potential energy surface involved in the OH formation channel, detailed ab initio calculations are performed using configuration interaction-singles (CIS) method. It is proposed that at 266 nm photolysis, the OH fragment is formed from two different excited state structures, one with a strong H bonding, similar to that in the ground state, and another without effective H bonding, whereas, at 248 nm photodissociation, it seems that the OH formation occurs mainly from the excited state, which lacks effective H-bonding. At 193 nm excitation, the initially prepared population in the Rydberg state crosses over to a nearby σ* repulsive state along the C−O bond, from where the dissociation takes place. The exit barrier for the OH dissociation channel is estimated to be 14 kcal/mol. The existence of dynamical constraint due to strong hydrogen bond in the ground state is effectively present in the dissociation process at 266 and somewhat deficient at 248 nm photolysis. Spectroscopic and electronic properties of α-diketones have been the subject of an important number of theoretical and experimental studies.1−6 In case of aromatic α-diketones, photorotamerism, related to the dihedral intercarbonyl bond has a strong influence on the ground state and excited state properties.7 The 1,2-diketo functional group in a six-membered ring system is an important structural unit found in a number of biologically important molecules and also important from synthetic viewpoints. The structural unit is considered to be the active site of the anticancer drug Quassinoid Bruceantin,8,9 and antitumor steroid 4-hydroxyandrost-4-ene-3,17-dione.10 A small member of this family, 3-methyl-1,2-cyclohexanedione, is considered to be responsible for coffee aroma.11 It has been
I. INTRODUCTION The α-diketones are the compounds having two keto groups on adjacent carbon atoms in an aliphatic chain or ring. These compounds show the tendency for interconversion between a keto and enol form, widely known as tautomerism. Keto−enol tautomerism arising due to this plays an important role in many organic reactions. This keto−enol interconversion is exploited by synthetic chemists as its enol form is responsible for a large number of important reactions, e.g., benzilic acid rearrangement, Michael reaction, Claisen rearrangement and several other photochemical reactions. Though keto−enol transformation is very important, there are very few studies that explore its fundamental nature and reactions associated with it. This is primarily due to the difficulty in isolating these molecules in an easily characterized form. As a result, experimental work involving tautomerism is mostly limited to the studies in the gas phase. © 2013 American Chemical Society
Received: November 14, 2012 Revised: February 27, 2013 Published: February 27, 2013 2415
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
found that α-diketones undergo facile Strating−Zwanenburg photodecarbonylation to produce the corresponding poly(acene).12 These diketones show a weak fluorescence and phosphorescence and have a small singlet−triplet energy gap that facilitates rapid intersystem crossing from the singlet to the triplet state. Both of these states of diketones are short-lived, and their decarbonylation is a rapid process, occurring within few nanoseconds. Photoaddition reaction of 1,2-diketones and silyl ketene acetals in benzene, acetone, or acetonitrile solutions is observed to promote formation of 1,4-dioxenes arising by Paterno−Buchi processes, and β-hydroxy-γ-ketoesters, generated by single electron transfer-promoted Claisen-type condensation. These competitive pathways from the excited states of the 1,2-diketones to these products are influenced by solvent polarity and the nature of the silyl ketene acetal and 1,2diketone.13 Monocarbonyl cyclohexanone does not display any enolic characteristics and exists mainly in keto form in solution (∼99%). However, introduction of second carbonyl group in the molecule can have profound effect on the enolizibility of the cyclohexanone. The effect depends on the position of the second carbonyl group on the ring. When the carbonyl group is introduced at the α-position, the diketone shows enolic character in both solution and gas phase. But when it is at the β-position, enolic character is seen in only solution phase. In contrast, on substitution at the γ- position, the diketone does not show enolic character, it exists only in keto form. Thus, 1,3and 1,4-cyclohexanedione are expected to exist mainly in keto form in the gas phase, implying a preference for a chair like structure. This was confirmed by the strong spectral similarities between these two compounds.14 Unlike 1,3- and 1,4cyclohexanedione, 1,2-cyclohexanedione (CHD) is expected to have dominant enolic character. Several studies have reported on the molecular structure and keto−enol tautomerism in 1,2-cyclohexanedione (CHD).14−19 All these studies suggest that in the gas phase, the enol form dominates. The theoretical calculations done for this compound show that the enol form with hydrogen bond is lower in energy than the keto form and the enol form without hydrogen bond is higher in energy than the keto form.16 Walzl et al.,20 in their detailed studies on various diketones using electron-impact spectroscopy have suggested that in CHD molecule, the two carbonyl groups can be twisted away from coplanarity due to constraint imposed by ring. This increases the overlap of highest occupied nonbonding (n) orbital and the lowest unoccupied π* orbital, resulting in a large value of the n,π* singlet−triplet splitting. CHD is synthetically an important compound and has various applications in complex chemistry. The reaction of CHD with specific ω-alkenyl organometallics proceeds readily at both carbonyls and leads to 1,2-diols bonded to terminal olefinic chains in aqueous tetrahydrofuran.21 Schwartz et al.22 have observed that the reactions of CHD with diamond (001) and Silicon (001) surfaces are quite different. It reacts with diamond (001) through the OH portion of the monoenol species via OH cleavage, and preferentially undergoes a 1,3-H shift with Si (001) and results in a product involving both O atoms. This difference is due to fact that the zwitterionic SiSi dimer induces an intramolecular 1,3-H shift, suggesting the charge transfer effect, which is an important element of uncatalyzed enol keto transformations. Another very interesting aspect of CHD is that it belongs to the α,β-enone class of compound. The detailed photochemistry of α,β-enone had been already studied in recent years.23−25
Previously, we have studied the photodissociation dynamics of a β-diketone, acetylacetone25 at 266, 248, and 193 nm and detected OH. CHD, a α-diketone, different from acetylacetone a β-diketone, its dissociation dynamics can be different. However, the enolic form of CHD is also an α,β-enone similar to enolic acetylacetone. The only difference between the two enone system is the position of the OH group. Thus, it will be interesting to investigate the dissociation dynamics of CHD and compare with that of acetylacetone. Here, we report the dynamics of generation of the transient OH radical, detected by laser induced fluorescence (LIF) method, on photoexcitation of CHD at 266, 248, and 193 nm in the gas phase. In present work, we have measured the partitioning of the available energy among translational, rotational, and vibrational degrees of freedom of the photoproducts. On the basis of experimental results, coupled with ab initio molecular orbital calculations, we have proposed a plausible mechanism of OH formation. We have also compared the results with our earlier work on the enolic acetylacetone photodissociation.
II. EXPERIMENTAL SECTION The present studies on photodissociation dynamics are carried out in a flow reactor, at millitorr level pressure, using the laser photolysis-laser induced fluorescence setup (LP-LIF). The experimental setup is discussed in detail in our earlier work.26,27 Briefly, an excimer laser (Lambda Physik model Compex-102, Fluorine version) is used for photolysis, and a seeded Nd:YAG laser (Quantel, model YG980 E20) pumped frequency-doubled dye laser (Quantel, TDL90) is used as the probe laser. The reaction chamber is made up of glass and has crossed right angle arms for photolysis and probe lasers. All the arms are equipped with baffles, and the windows are fixed at the Brewster angle to reduce scattering. The photolysis and the probe lasers intersect at the center of the reaction cell. The detection system views the intersection volume of photolysis and probe laser beams through the bottom arm window. The fluorescence is collected by a lens (focal length 50 mm, 38 mm diameter) and detected by a photomultiplier tube (Hamamatsu, model R 928P). A band-pass filter (λcenter = 310 nm, fwhm = 10 nm, %T310 nm = 10%) is placed between the collecting lens and the PMT to cut off the scattering from the photolysis laser light. The fluorescence signal is gate integrated by a boxcar (SRS 250), averaged for 30 laser shots. The dye laser is scanned via RS232 interface, and the data are collected through GPIB interface using control and data acquisition program on a personal computer. To correct for the laser intensity fluctuations, LIF intensities are normalized with respect to both the pump and the probe laser energies, monitored by suitable photodiodes. CHD (98%+, Alfa Aesar) vapor is flowed through the reaction chamber in a pressure range of mTorr level, monitored by a capacitance pressure gauge and photolyzed at 266, 248, and 193 nm as required. The OH fragments generated on photolysis are probed state selectively in the wavelength region 306−309 nm, by exciting the A2Σ ← X2Π (0−0) transition of OH, and monitoring the subsequent A → X fluorescence. A circular aperture is used to select the homogeneous part of the rectangular excimer laser beam profile. In the present studies, all the photolysis experiments are carried out at laser intensities less than 1.0 mJ/cm2. Both the laser beams are unfocused and attenuated to prevent any saturation effect and multiphotonic events that occur at high laser intensities. This has been checked by recording the variation of LIF intensity of the OH 2416
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
fragment with intensities of both lasers, which shows a linear dependence.
III. RESULTS AND ANALYSIS A typical LIF spectrum with appropriate assignments is shown in Figure 1. A detailed description of notation used can be seen
Figure 2. Boltzmann plots of rotational distributions of the nascent OH radical generated on (A) 266 nm and (B) 248 nm photolysis. The solid lines are the fit to the data points.
OH (1−1) transition is also scanned to determine the population of the OH fragment in v″ = 1, if any, but no LIF signal was observed. On the basis of the experimental detection limit and Franck−Condon factors relative to the OH (0−0) transition, it is estimated that less than 5% of the total OH yield is formed in the v″ = 1 state. B. Spin−Orbit State Distribution. The ground electronic state of OH being 2Π, there are two spin−orbit components, 2 Π3/2 and 2Π1/2. The LIF intensities of only P1(N) and P2(N) lines are used to obtain the spin−orbit population. Figure 3
Figure 1. Typical portion of LIF spectrum of OH after photodissociation of CHD (10 mTorr) at 248 nm photodissociation. (Pump−probe delay ≈ 50 ns).
in the literature.28 Briefly, the symbols P, Q, and R denote rotational transitions with ΔN = −1, 0, and +1, respectively. The subscripts 1 and 2 represent the transitions satisfying both ΔJ and ΔN, from Π3/2 and Π1/2 spin−orbit states, respectively, which form the main branches of the transition, whereas 21 and 12 represent the transition where only ΔN is satisfied and are satellite to the main branches represented by subscript 1 and 2. The Q and P/R branches originate respectively from Λ doublet states Π− (A″) and Π+ (A′) due to parity selection rule (+ ↔ −).29 The relative populations of the OH fragments are determined by normalizing the peak areas of the rotational lines with respect to pump and probe laser intensities, pressure change, if any, and the respective Einstein absorption coefficients.30 The spin−orbit and the Λ doublet ratios are calculated from the relative populations of different rotational states. The translational energy associated with the OH fragment is calculated from the Doppler profiles of the rotational lines. The detailed results are presented below. A. Rotational State Distribution. The nascent rotational state population of OH radicals, generated on photodissociation of CHD at 248 and 266 nm, is used to construct a Boltzmann plot for obtaining the rotational temperature of nascent OH fragments. The rotational state population at 193 nm could not be measured with a good signal-to-noise ratio due to low absorption coefficient of CHD at 193 nm. A typical Boltzmann plot for the photodissociation of CHD at 266 and 248 nm is shown in Figure 2. The rotational temperatures have been obtained by the best fit to all the rotational states. At 266 nm photolysis, the Boltzmann plot is nonlinear, showing a curvature. Hence, the rotational population is fairly described by two types of Boltzmann-like distributions, which are characterized by rotational temperatures of 3100 ± 100 and 900 ± 80 K. However, at 248 nm photolysis, the rotational distribution is described by a single rotational temperature of 950 ± 80 K. The
Figure 3. Statistically weighted spin−orbit ratios of nascent OH(v″=0) as a function of the rotational quantum number (N). The red filled squares and the blue filled circles denote the ratios at 266 and 248 nm photolysis, respectively.
shows the spin−orbit (Π3/2/Π1/2) ratios multiplied by appropriate statistical weights (2J + 1) plotted versus the OH rotational quantum number (N), for different wavelengths, namely, 266 and 248 nm. From the figure, it is evident that the average value is one at 248 nm photolysis, within experimental error, thereby indicating a statistical distribution, without any preference for either of the spin−orbit states in the dissociation process. However, at 266 nm photolysis the average seems to 2417
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
Gaussian function as depicted in Figure 5 for the P1(4) line, which implies that the translational energy follows the
slightly more than one. Here also, the nature of spin−orbit ratios for 266 and 248 nm photolysis shows a difference. The relative population of spin−orbit states gives a clue in terms of coupling between the initially prepared excited states with a nearby dissociating state.31 Whereas, a statistical distribution suggests that the initially excited state probably does not interact directly with a triplet dissociating state, a nonstatistical distribution indicates the role of triplet state in the photodissociation process. C. Population of the Λ-Doublets. Each spin−orbit state of OH has two Λ-doublet components, denoted as Π+ (or A′) and Π− (or A″), depending on the orientation of the π lobe of the unpaired electron on OH with respect to the plane of rotation. In the high-J limit, in Π+ (A′) state, the π lobe lies in the plane of rotation, whereas in the Π− (A″) state, π lobe is perpendicular to the plane of rotation. The relative populations of the Λ-doublets provide information about the exit channel dynamics during the breaking of chemical bond. In Figure 4,
Figure 5. Doppler profile of the P1(4) line in the spectrum for 248 nm photodissociation of CHD.
Maxwell−Boltzmann distribution. The width and the shape of the Doppler broadened LIF line include contributions from the fragment molecular velocity, the thermal motion of the parent molecule, and the finite probe laser line width. Thus, the actual Doppler width is calculated using deconvolution procedure using the width (fwhm) of the laser spectral profile of the probe laser beam, which is obtained from the OH Doppler profile measured in a thermalized condition. All the rotational lines exhibit the same width within an experimental error. More than 45 rotational line profiles are evaluated to estimate the average kinetic energy of OH fragment in the laboratory frame, ELAB t (OH). The product translational energy in the center of LAB mass frame ECM t (OH) is obtained from the Et (OH), and is found to be 12.5 ± 3.0, 12.7 ± 3.0, and 12.0 ± 3.0 kcal/mol at 266, 248, and 193 nm excitation, respectively. Taking the value of ∼90 kcal/mol as the bond dissociation energy for OH producing channel, similar to that in acetylacetone, the f T values are calculated as 0.73, 0.51, and 0.21 at 266, 248, and 193 nm excitation, respectively.
Figure 4. Λ-doublet ratio of nascent OH(v″=0) as a function of the rotational quantum number (N). The red filled squares and the blue filled circles denote the ratios at 266 and 248 nm photolysis, respectively.
the Λ-doublet ratio is plotted against the rotational number, N. Both at 248 and 266 nm photodissociation, the Λ-doublet ratio deviates from the statistical value of unity. The deviation is more prominent in the case of 266 nm photolysis. The result implies a role for exit channel interaction in formation of OH. D. Translational Energy in Products. From the OH Doppler profiles, the average kinetic energy of the OH radical in the laboratory frame, ELAB t (OH), can be determined. The Doppler profiles reflect the distribution, f(vz), of the velocity component vz of the absorbing species along the propagation direction of the probe laser beam via the linear Doppler shift Δν = ν − ν0 = vzν0/c. For an isotropic velocity distribution, f(vz) = f(vx) = f(vy), the average translational energy in the 2 laboratory frame is given by Elab T (OH) = 3/2mOH⟨vZ ⟩OH, 2 where ⟨vZ ⟩OH for a Gaussian Doppler profile, is represented by the equation: 2
⟨vZ ⟩OH
2 1 ⎛ FWHM ⎞ 2 = ⎜ ⎟c 2 ln 2 ⎝ 2ν0 ⎠
IV. COMPUTATIONAL DETAILS Ab initio molecular orbital (MO) calculations are performed to investigate the potential energy surface (PES) of the ground and excited electronic states of various conformers of CHD, using Gaussian suite of program. CHD exists in various conformers (see below), namely, the H-bonded enolic form (structure a), diketo form (structure b), and non-H-bonded enolic form (structure c). We optimized geometries of these conformers/isomers at HF level using 6-311++G** basis set. The relative energy is then calculated at the MP2 level using the 6-311++G(3df,2p) basis set. Ground state self-consistent field (SCF) wave functions are examined for their stabilities before doing excited state calculations. Excited electronic state calculations were performed at the configuration interaction with single electronic excitation (CIS) level, and various excited states of CHD accessible at 266, 248, and at 193 nm excitations are examined. Each of the excited state geometries (T1, T2, S1, and S2) was then fully optimized following CIS procedure for both types of enolic form (a and c, see below). For calculation of stationary point energies of various excited states, a hybrid approach was employed. Once the structure of different excited states, namely, T1, T2, S1, and
(1)
where c is the speed of light and fwhm is the full width at halfmaximum of the normalized Doppler profile of the OH radical with transition at ν0. In the present case, Doppler profile is a 2418
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
as S0-TS. This transition state structure lies 11.1 kcal/mol above the structure a. The diketo form of CHD, structure (b), is also optimized and the relative energy was calculated. The diketo structure (b) lies 4.5 kcal/mol above the hydrogenbonded structure (a). The non-hydrogen-bonded structure (c) is least stable among all the structure in the ground state. Although different reaction channels are possible from the ground state of CHD, in the present study our main aim is to characterize the structure responsible for the OH production. Generally, the dissociation process occurring on the ground state potential energy surface does not proceed through any transition state and also in the present case, the f T values indicates a clear involvement of excited states for OH formation channel. Hence, a detailed ab initio calculations are performed using CIS procedure, which will be presented in following section.
S2, were optimized using the CIS procedure, the vertical excitation energies (VEE) for these structures were then calculated using time-dependent density functional theory (TD-DFT) to get the total energy for better accuracy. Vibrational frequencies are calculated with the final optimized geometries, in both the ground and the excited states to confirm the nature of the stationary state geometries. Orbitals participating in the electronic transitions were assigned for each of the excited states and can be seen in Figure 6.
Figure 6. Computed MOs involved in the transition of both the conformers of enolic 1,2-cyclohexanedione (CHD). The green and the red color of the MO lobes represent two opposite phases of the MO wave function.
B. Nature of Excited States. The onset of optical absorption in enolic CHD occurs at ∼286 nm (4.34 eV). Its gas phase UV spectrum shows strong absorption extending up to 218 nm (5.7 eV), with λmax = 256 nm (4.84 eV). The nature of this band has been attributed to the π → π* transition. To understand the nature of the transitions involved in enolic CHD in the UV region, ab initio molecular orbital (MO) calculations were performed, in detail. We optimized the ground state geometries of enolic CHD for its various conformers, using the 6-311++G** set of basis sets. The augmented basis set, with diffuse and triple-ζ functions, namely, aug-cc-pVTZ, was then used for obtaining the vertical excitation energies for various transitions, using time-dependent density functional theory (TD-DFT). Although the calculated vertical transition energies slightly differ as compared to the
A. Profiles of Different Potential Energy Surfaces in the Ground State. Various conformers of CHD in the ground state are optimized using the 6-311++G** basis set, and the most stable geometry consists of a plane in which both CO and OH moieties lie on the plane making a 5-member ring structure with strong H-bonding. These structures can be seen in the graphic shown below. The hydrogen-bonded (a) and non-hydrogen-bonded structures (c) are optimized, and the energy difference between them is found to be 6.9 kcal/mol. The transition state correlating these two structures, namely, the hydrogen-bonded (a) and the non-hydrogen-bonded structures (c), is also optimized and can be seen in Figure 7 2419
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
Figure 7. Different optimized structures for various excited states of CHD (for details see the text).
experimental results, the nature of transitions and that of the orbitals involved are accurately predicted, using this method. The orbitals participating in the different electronic transitions were visualized, for better understanding of the process. The vertical excitation energies and the respective oscillator strengths of several low-lying triplet and singlet states for enolic CHD are shown in Table 1. A total of four excited states are considered, two each triplet and singlet states. These states are primarily due to excitations from the n (HOMO−1) and π (HOMO) orbitals, to π* orbitals, which is LUMO. Besides these states, Rydberg states involving excitations to a diffuse 3s
orbital are also present at higher energies. For H-bonded CHD, structure a, the T1 state is a π−π* transition, at 2.44 eV (508 nm). The second triplet state, T2, is an n−π* transition at 3.42 eV (363 nm). The S1 is mainly an n−π* transition with very low oscillator strengths (0.0006), at 3.98 eV (311 nm). The S2 is mainly a π−π* transition with considerable oscillator strength (0.1570), at 4.538 eV (274 nm). Similarly, for nonH-bonded CHD, structure c, the two lowest triplet states, namely, T1 and T2, are at 2.99 eV (414 nm) and 3.33 eV (372 nm), which are π−π* and n−π* transitions, respectively. The singlet states, namely, S1 and S2, are mainly n−π* and π−π* 2420
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
Table 1. Different Transition Types in CHD and Their Vertical Excitation Energies (eV) and Corresponding Values [kcal/mol, nm, and Respective Oscillator Strengths (in Parentheses)] of Various Excited States of Both the Conformers of Enolic 1,2Cyclohexanedione, Namely, H-Bonded Conformer (a) and Non-H-Bonded Conformer (c)a enolic 1,2-cyclohexanedione (H-bonded, structure a) excited sate 1 2 3 4 a
T1 T2 S1 S2
transition type (π → π*) (n → π*) 1 (n → π*) 1 (π → π*)
3 3
vertical excitation energy (eV) and (kcal/mol, nm, oscillator strength) 2.4418 3.4151 3.9817 4.5322
(56.30, 508, 0.0) (78.75, 363, 0.0) (91.82, 311, 0.0006) (104.51, 274, 0.1570)
relative energy (kcal/mol) and (nm) of the different stationary points 31.0 (922) 76.3 (375) 88.1 (325)
enolic 1,2-cyclohexanedione (non-H-bonded, structure c) vertical excitation energy (eV) and (kcal/mol, nm, oscillator strength) 2.9949 3.3304 3.9278 5.0298
(69.06, 414, 0.0) (76.80, 372, 0.0) (90.58, 316, 0.0003) (115.99, 247, 0.1558)
relative energy (kcal/mol) and (nm) of the different stationary points 34.0 76.3 87.3 110.0
(841) (375) (328) (260)
The table also reports the relative energy (kcal/mol and nm) of different stationary points in the excited states of both the conformers.
C2O2H1 backbone with C3 and H2) are nonplanar, making this structure a truly puckered one. Careful examinations of structural parameters and the natural orbital analysis confirmed this structure to be an nπ* state. It is quite reasonable to assume that the initial excitation is from the n orbital to the CO π* orbital weakening the CO bond to a large extent. The T2 minimum structure shows an increased CO bond length of 1.25 Å as compared to 1.190 Å in the ground state. The O1···H1 distance in the T2 minimum is 2.36 Å, indicating that the H bonding is less effective in the T2 state as compared to its ground state. The position of the T2 minimum is about 76.3 kcal/mol above the ground state. c. S1 State. The structure for the S1 minimum is a nonplanar, which is shown in the figure as S1-H. A comparison shows that the S1 minimum is quite similar to T2 3(nπ*) in structure. Natural orbital analysis and a close examination of the structure suggest that this structure results from a 1(nπ*) transition. The O1···H1 distance in the S1 minimum is 2.44 Å, indicating that the H bonding is even less effective in the S1 state as compared to either T2 or the ground state. The S1 state minimum is located about 88.1 kcal/mol above the ground state. d. S2 State. We failed to optimize the structure of the S2 minimum at the present level of theory for the H-bonded CHD. However, in a recent article by Chakraborty and coworkers18 the S2 state, along with the S1 state, for the H−bonded structure is reported. A careful examination of both their S1 and S2 structures and our own calculation reveal that they are equivalent in terms of structure and energy. We also repeated the theoretical calculations and confirmed that after a few iterations in CIS calculations, the S1 and S2 excited states cross over and finally produce the structure of S1 only. So the reported structure for the S2 state by Chakraborty and coworkers18 does not seem to be genuine. Moreover, we tried to optimize the structure of the S2 minima at various levels of theories but we failed to get the desired results. However, the absorption spectra of compound show a very intense peak at around 266 nm, indicating a bound state of ππ* character in nature, similar to a state that can be termed as S2. This discrepancy prompted us to search for the local minimum for the ππ* singlet state for non-hydrogen-bonded CHD, which will be discussed in detail in following section. ii. Non-Hydrogen-Bonded Enolic CHD (Structure c). a. T1 State. The T1 minimum at the CIS/6-311++G** level for nonhydrogen-bonded enolic CHD has a nonplanar structure similar to that for the hydrogen-bonded counterpart, except for the OH orientation shown in Figure 7 as T1-NH. The C2C3 double bond (1.32 Å) is converted to almost a single bond (1.48 Å) and the C1C2 single bond length (1.49 Å) is shortened to nearly a double bond (1.43 Å). Similarly, the
transitions with oscillator strengths of 0.0003 and 0.1558, at 3.93 eV (316 nm) and 5.03 eV (247 nm), respectively. The other singlet states, namely, S3 and S4 are Rydberg states in both types of structures. For H-bonded CHD, structure a, the S3 and S4 states are at 5.74 eV (216 nm) and 6.16 eV (202 nm) with oscillator strengths 0.0068 and 0.0008, respectively. Similarly, for non-H-bonded CHD, structure c, the S3 and S4 states are at 5.56 eV (223 nm) and 6.02 eV (206 nm) with oscillator strengths 0.0017 and 0.0084, respectively. Considering the corresponding wavelengths for vertical excitation energies and the respective oscillator strengths of the various transitions, it is evident that at 248 and 266 nm, CHD has a π−π* transition. The excited state, with π−π* transition, adiabatically correlates only with highly excited photoproducts (Supporting Information, Figure 1S). Because the formation of excited states of OH is not feasible in a single-photon excitation in the present case, it implies that OH is not produced from the initially excited π−π* state. Hence, it is assumed that, in the CHD, the initially prepared singlet π−π* state crosses over to the nearby states, mostly the n−π* state, which may be either triplet or singlet, or the triplet π−π* state, from where the OH formation may take place. At 193 nm excitation, the molecule is prepared in its Rydberg state from where various other processes may occur, including the formation of the OH radical. C. Profiles of Different Potential Energy Surfaces in Excited States. i. Hydrogen-Bonded Enolic CHD (Structure a). a. T1 state. The T1 minimum at the CIS/6-311++G** level has a nonplanar structure, shown in Figure 7 as T1-H. From the natural orbital analysis of the structure, it is quite reasonable to expect that the lowest triplet structure arises from the (ππ*) transition of the CC bond. One electron is initially excited from the π to π* orbital converting the C2C3 double bond (1.32 Å) to almost a single bond (1.49 Å) and the C1C2 single bond (1.49 Å) is shortened to 1.42 Å. Similarly, the C2O2 bond length is slightly shortened from 1.34 to 1.32 Å and C1O1 is lengthened from 1.19 to 1.21 Å. Apart from these changes, the O1···H1 distance is slightly elongated to ∼2.20 Å at the T1 minimum from the ground state value of ∼2.13 Å. It seems that the H-bonding is still effectively present in the T1 structure, because of the near planarity of the O1 C1C2O2H1 backbone. These atoms are also nearly planar with C3 as in the case of the ground state. However, the H2 atom that is on the plane in the ground state is now out of the plane in the T1 minima. These details can be seen in the Figure 7. The position of the T1 minimum is about 31.0 kcal/ mol above the S0 ground state. b. T2 State. The structure for the T2 minimum is shown in Figure 7 as T2-H. In this structure, all the atoms (O1C1 2421
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
forming a six-membered ring, the situation is different in CHD because of its ring structure. The ring structure does not allow the enolic H-bonded structure (a) to stabilize by balancing the charges effectively and simultaneously. This is the main reason we failed to optimize the S2 minimum structure for structure a. The only option to stabilize the structure by the H atom of the OH moiety is by rotation of COH bond with almost no barrier, thus giving rise to the S2 structure minimum. In this process, the H-bond is broken and the geometry changes to non-H-bonded structure (c). The above discussion clearly shows why we could get a S2 structure minimum for the non-Hbonded structure (c) and not for the H-bonded structure (a).
C2O2 bond length is slightly shortened, and the C1O1 bond is lengthened. The O1C1C2O2H1 backbone with a C3 atom is nearly planar, similar to the ground state. The only difference in structure is that H2 is in the plane in the ground state but out of the plane (at ∼61°) in the T1 minima. These details can be seen in the Figure 7. The position of T1 minimum is about 34.0 kcal/mol above the S0 ground state of structure a. b. T2 State. The structure for the T2 minimum is shown in Figure 7 as T2-NH. In this structure, all O1C1C2O2 H1 atoms are nonplanar, and the backbone connecting the H2C3C2O2H1 atoms are planar. In this structure, the initial excitation of the n orbital to the CO π* orbital weakens the CO bond to a large extent. The T2 minimum structure shows an increased CO bond length of 1.25 Å as compared to 1.19 Å in the ground state. The position of the T2 minimum is about 76.3 kcal/mol above the ground state. c. S1 State. As shown in Figure 7 (S1-NH) the structure for the S1 minimum is almost planar. A comparison shows that the S1 minimum is quite similar to T2 3(nπ*) in structure. Natural orbital analysis and a close examination of the structure suggest that this structure results from a 1(nπ*) transition. The position of S1 minimum is about 87.3 kcal/mol above the ground state. d. S2 State. We could optimize the structure of S2 minimum at the present level of theory, and it is shown in Figure 7 as S2NH. From the natural orbital analysis of the structure, it is quite reasonable to expect that this particular structure arises from the (ππ*) transition of the CC bond. One electron is initially excited from the π to π* orbital converting the C2C3 double bond (1.32 Å) to almost a single bond (1.43 Å) and shortening the C1C2 single bond length (1.49 Å) to 1.42 Å. Similarly, the C2O2 bond length is slightly shortened, and the C1 O1 bond is lengthened. Similar to the structure T1-NH, and that in the ground state, the O1C1C2O2H1 atoms with C3 atom are nearly planar. The only difference from the T1-NH structure and the ground state is that the H2 that lies in-plane in the ground state is now out of the plane at ∼30° as compared to ∼61° in T1 minima. These details can be seen in the Figure 7. The location of the S2 minimum is about 110.0 kcal/mol above the S0 ground state. Generally, in the case of α,β-enones, the S2 state originates from a ππ* transition centered at the double bond having a mixed character of π*(CO)/π*(CC). For acrylic acid, the S2 state structure calculated by Arendt et al.,32 shows transfer of a large amount of electron density from the CC double bond to the COOH group, which can be treated as intramolecular charge transfer. This structure is so different from the ground state that the double bond is actually shifted to the next carbon atom, and the initial double bond is elongated and behaves as a single bond. Similarly, in enolic acetylacetone25 a strong intramolecular H bonding counterbalances the charge at the O atom and the structure gets stabilized. This stable structure of S2 indeed lowers the transition energy in enolic acetylacetone with the 1(ππ*) transition showing a peak at 267 nm, compared to that in acrylic acid, which shows a peak at ∼198 nm. Similar to enolic acetylacetone, the 1(ππ*) transition in the CHD, shows an intense absorption peak in the 267 nm region. Like in acrylic acid, the 1(ππ*) state in CHD is expected to show transfer of a large amount of electron density from the CC double bond to the adjacent CC single bond and to the C O group, which can be treated as intramolecular charge transfer. Contrary to case in enolic acetylacetone, where the charges are effectively balanced by the strong H-bonding
V. DISCUSSION A. Average Energies of the Fragments. The partitioning of the available energy into various degrees of freedom of the fragments is mainly governed by the nature of transition state and its location on the dissociative potential energy surface. Dissociation of CHD at 266 and 248 nm occurs with a large fraction and similar amount of relative translational energy being imparted into the fragments, indicating the presence of an exit barrier. Because the statistical model predicts well only the energy partitioning of a dissociation process with no barrier, we have employed different models, including the statistical model, to understand the dissociation process at different wavelengths. i. Statistical Model. A statistical prior distribution is often the starting model to understand any dissociating event. In this model, the available energy, Eavail, is distributed among all the accessible states with equal probabilities under the constraint of conservation of energy. This model completely ignores angular momentum constraints in the photodissociation process. A statistical dissociation process occurs predominantly, if the photo-excited parent molecule is so long-lived that the excess energy is partitioned statistically among the available degrees of freedom of the products. This includes a process where a rapid internal conversion to the ground electronic state takes place, followed by the subsequent slow dissociation. Under these events, in a large molecule, with many low-frequency modes, a relatively small amount of the excess energy is partitioned into translational motion of the products. For this kind of dissociation process, a priori calculations33−35 were adopted, along with a simple analytical expression established by Klots,36 relating the mean translational energy release, ET, and the Eavail, for a statistical barrierless dissociation process (Supporting Information, Figure 2S). The density of states were estimated from the vibrational frequencies and the rotational constants of the fragments obtained at the HF/6-311++G(d,p) level of theory. The statistical model puts ∼7% of the available energy into the relative translational mode of the photofragments. Thus, the available energy appearing as the relative kinetic energy of the products is underestimated using statistical model. Hence, the statistical model fails to explain the observed partitioning of the available energy among photofragments at all the wavelengths studied. ii. Impulsive Model. In the impulsive model, the distribution of energy among product states is governed by the dissociative event, i.e., by the repulsive force acting during the breaking of the parent molecule into the products. This treatment conserves energy and both linear and angular momenta. The model assumes that all the available energy is stored within the breaking bond. We have employed the method for calculation as given by Crim and co-workers33 and Tuck37 to estimate the 2422
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
earlier. So, in the vicinity of the Franck−Condon region, the C−OH bond rotates very fast almost with no barrier and gives rise to the S2 structure minimum for the non-H-bonded form, depicted in Figure 7 as S2-NH. This shows that even though the non-H-bonded structure (c) is not excited initially from the ground state, the excited state dynamics is totally controlled by its structure. Due to fast nonradiative processes, the S2-NH state crosses over to various other states, either S1 or T1/T2 of both the structures, and the subsequent dissociation results in the formation of OH. However, we are unable to locate any transition state structure for the OH fragment formation, using the CIS method in any of these excited states. However, the observation that a similar amount of translational energy is released into the photofragments irrespective of the excitation energy or the available energy indicates the presence of an exit barrier for the dissociation process. To further investigate this issue, we have mapped the potential energy (PE) curves for various excited states. In Figure 8, we have presented calculated
energy partitioning using impulsive model. The impulsive model puts ∼48% of the available energy into the product translational mode. Like the statistical model, the impulsive model also is unable to explain the experimentally obtained translational energy partitioning between the fragments for 266 and 193 nm photodissociation processes, which are ∼73% and ∼21%, respectively. However, the prediction based on the impulsive model is closer to the experimental value at least at 248 nm, which is ∼51%. However, both statistical and impulsive models allocate low rotational and vibrational energies into the OH radical, as observed at 248 and 266 nm. The failure of both the statistical and impulsive models in explaining the partitioning of the available energy prompted us to apply the hybrid model, employed by North et al.38 for the case of reactions with barrier. In this model, the Eavail for the products is divided into the excess energy above the exit barrier (Estat) and the exit barrier energy (Eimp). The partitioning of Estat and Eimp is treated respectively by the statistical and modified impulsive models. The energy partitioned into each fragment is then obtained by adding contributions from each of these two models. The measured energy partitioning into the fragments is reproduced well with an assumed exit barrier of ∼14.0 kcal/mol. For both wavelengths, namely, 266 nm (∼108 kcal/mol) and 248 nm (∼115 kcal/mol) the Eimp value is taken as 14 kcal/mol whereas Estat values are taken as 4 and 11 kcal/ mol, respectively. Using these values, the expected translational energies released for both the wavelengths, namely, 266 nm and 248 nm, are calculated as 12 and 13 kcal/mol. This model predicts no vibrational excitation of OH. B. Mechanism of the Dissociation Process. We shall now present the complete picture of the dissociation process in the light of the above discussion in conjunction with the dynamics models for energy partitioning and the structure calculation with the help of the ab initio method. As discussed earlier, the major fraction of the available energy is channeled into the translational mode of the photofragments, which indicates either the impulsive dissociation of the C−OH bond, or the presence of an exit barrier. As shown earlier, the impulsive dissociation does not account for such a high fraction of the available energy being partitioned into the translational energy of the products, at least at 266 nm photolysis. Also, the amount of translational energy is almost constant irrespective of the excitation wavelength or the available energy. This kind of behavior in translational energy release in any photodissociation process is typically of the dissociation process with an exit barrier. Thus, there should be an exit barrier to the C−O bond cleavage. Such an exit barrier is observed for a simple bond cleavage leading to OH in photodissociation of acetic39 and acrylic acid27,40 and acetylacetone.25 The 266 and 248 nm excitation of CHD leads to the same electronic state through 1 (π → π*) transition. Dissociation mechanism of CHD at different wavelengths is discussed below. The absorptions of 266 and 248 nm photons by enolic CHD lead to the allowed (π → π*) transition near the S2 state origin. Subsequent to excitation, we are unable to get any fluorescence even though the excitation energy is near the origin of the transition. This suggests the importance of fast nonradiative processes in this region of excitation. As discussed earlier, the enolic CHD exists in two forms, namely, the H-bonded (a) and non-H-bonded form (c), with an energy difference of about ∼7 kcal/mol in favor of H-bonded structure. However, the S2 structure for the H-bonded structure (a) does not exist because of the nonstabilization of the charge separation as discussed
Figure 8. Potential energy curves for various excited electronic states of different CHD structures, namely, H-bonded (a) and non-Hbonded structures (c) calculated with the TD-DFT method as a function of the C2−O2 bond length.
PE curves as a function of the C2−O2 bond length for various excited electronic states, such as T2, S1, and S2 for both the Hbonded (a) and non-H-bonded structure (c) of CHD. The molecular geometries in the ground state for different discrete values of C2−O2 bond lengths were optimized at the HF/6311++G** level of theory, and the energies of such fully optimized configurations were used to generate the PE curve for the ground state. For excited states, the curves were generated by plotting the calculated vertical transition energies corresponding to different ground state optimized geometries using TD-DFT method, employing aug-cc-pvDZ basis sets. All the excited states presented in Figure 8 are bound states, and the relative energy is with respect to the H-bonded structure (a). However, it should be noted that the energies shown in Figure 8 are not those along the excited state reaction coordinate following the minimum energy path, but rather much higher energies. The first point in the PE curves belongs to the energy minimum structure for C2−O2 bond length of 1.32 Å. At equilibrium geometry, the excited states, namely, T2H and S1H, belonging to the H-bonded structure (a) have the minimum energy and are more stable than the corresponding structures for the non-H-bonded entity (c). However, as the C2−O2 bond starts elongating, excited states belonging to the non-H-bonded isomer become comparatively more stable and again this tendency is changed at longer C2− 2423
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
At 193 nm photolysis, we lack sufficient experimental data to discuss the mechanism of the OH formation unambiguously. Absorption of 193 nm light prepares CHD in a Rydberg state and due to fast internal conversion it can cross over to different lower potential energy surfaces. The translational energy release into photofragments estimated from the experiment is higher compared to the energy evaluated using the statistical method and somewhat lower compared to the impulsive model. Because we were not able to measure the rotational state distribution and other parameters like spin−orbit and Λ doublet ratios, it is difficult to suggest any explicit mechanism for 193 nm photolysis. At this point, we can only guess that the dissociation is taking place from a repulsive state, similar to acetylacetone.
O2 bond length. This clearly demonstrates that the crossingover of an excited state from the H-bonded to the non-Hbonded isomers and vice versa is very efficient. Possibly, a more accurate method like complete-active-space self-consistent field (CASSCF), can produce an optimized TS geometry for the OH formation. However, at present, we are not able to carry out this type of calculation for a medium large molecule, such as CHD. Although we could not optimize the TS structure, our present theoretical calculation in combination with experimental results, unambiguously suggest the presence of an exit barrier for the OH elimination channel. The large amount of translation energy release in the present case is very much similar to the photodissociation dynamics of similar enones, such as enolic acetylacetone. The transition state for the OH elimination channel in enolic acetylacetone was fully optimized at the T1 (ππ*) excited state.25 However, a combination of ultrafast electron diffraction (UED) studies and CASSCF calculation indicates the mixing of the S2 state with the S1 state in enolic acetylacetone is responsible for the OH elimination channel.41 Similar mixing or crossing of S2 and S1 states of the non-H-bonded structure of CHD is seen in the present case also (Figure 8). Moreover, the T1 state of both the H-bonded and non-H-bonded structure also has an exit barrier. Thus, the initially prepared S2 state crosses either to S1NH or to T2NH/T2H and finally dissociates to give OH radical as a photoproduct with an exit barrier. The different values of the rotational temperature and the nature of rotational distribution at 266 and 248 nm photolysis indicate different type of dissociation mechanisms taking place at these two wavelengths. At the 266 nm photolysis, the two types of rotational temperatures and hence the rotational energies were observed, which clearly indicate that there may be two types of molecular geometries responsible for the OH elimination channel. Similar observations were also reported for the rotational state distribution of nascent OH radical formed in the photolysis of nitro-benzoic acid42 and alkyl nitriles43 by Han and coworkers. They suggested the involvement of more than one structure in the OH formation channel giving rise to non Boltzmann distribution of OH rotational state. Likewise, we propose that the OH channel with high rotational temperature is originating from an excited state with the H bonding, e.g., T1H. The situation can be visualized as follows: the C−OH bond dissociation from an H-bonded structure is expected to generate a large torque generally giving rise to large rotation to the OH radical. Hence, we strongly believe that the dissociation at 266 nm involves the H-bonded structure, such as T1H, for the large rotational energy distribution and a non-H-bonded structure for the low rotational energy distribution. At 248 nm photolysis, the experimentally determined rotational temperature matches quite well with the lower value of rotational temperature at 266 nm photolysis. This indicates that a similar mechanism is responsible for these two cases and hence at 248 nm we have attributed the OH formation channel to a non-Hbonded structure. The value for the barrier height can be obtained from the hybrid model using the experimentally determined translational energy. The translational energy released into the photofragments is in good agreement with the modified impulsive model assuming an exit barrier of 14.0 kcal/mol. Hence, the transition state on the excited surface for the OH formation channel is located at ≈104 kcal/mol, from the ground state structure a, using an experimental value of 90.3 kcal/mol for the C−OH bond dissociation energy.
VI. CONCLUSION In summary, enolic CHD generates the OH radical upon excitation wavelengths of 266, 248, and 193 nm, which prepare the molecule in (π, π*) and Rydberg states. The nascent state of the photofragment OH is probed with the LIF technique. At 266 nm photolysis, the rotational population is fairly described by two types of Boltzmann-like distributions, which are characterized by rotational temperatures of 3100 ± 100, and 900 ± 80 K, whereas at 248 nm photolysis, the rotational distribution is described by a single rotational temperature of 950 ± 80 K. The average translational energy partitioned into the photofragment pairs in the center-of-mass coordinate is determined to be 12.5 ± 3.0, 12.7 ± 3.0, and 12.0 ± 3.0 kcal/ mol at 266, 248, and 193 nm excitation, respectively. The high percentage of translational energy partitioned into the products at 266 and 248 nm photolysis is consistent with the hybrid model showing the presence of an exit barrier for the OH formation channel. The spin−orbit and Λ doublet ratios are also measured to gain insights into dynamics of OH formation. Detailed ab initio quantum calculations suggest that at 266 nm photolysis, the OH fragment is formed from two different excited state structures, one with a strong H-bonding and another without effective H-bonding, whereas, at 248 nm photodissociation, we suggest that the OH formation occurs mainly from an excited state that is devoid of any effective Hbonding. At 266 nm photodissociation, the lowest planar T1H(ππ*) excited state, with strong H bond, can be responsible for the OH formation with higher rotational temperature, which is confirmed by preferential population of the Λ+(A′) state, whereas at 248 and 266 nm with lower rotational temperature, a mixture of S1/T2, where the H bonding is not effective, has been suggested. At 193 nm excitation, the initially prepared Rydberg state can cross over to a nearby σ* repulsive state along the C−O bond, from where the dissociation can take place. The exit barrier for the OH dissociation channel is estimated to be 14 kcal/mol. The existence of a dynamical constraint due to strong hydrogen bond in the ground state is effectively present in the dissociation process at 266, and somewhat minimized at 248 nm photolysis.
■
ASSOCIATED CONTENT
S Supporting Information *
Various important diabatic and adiabatic potential energy surfaces along the C2−O2 bond with the other geometrical parameters optimized for the ground state are shown in Figure 1S. Similarly, Figure 2S shows the ground state PES of Hbonded CHD structure as a function of the C2−O2 bond 2424
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
(17) Mukhopadhyay, A.; Ghosh, A. K.; Mukherjee, M.; Chakraborty, T. Electron Ionization Cross-Section and Fragmentation of αCyclohexanedione. Int. J. Mass Spectrom. 2012, 309, 192−199. (18) Mukhopadhyay, A.; Mukherjee, M.; Ghosh, A. K.; Chakraborty, T. UV Photolysis of α-Cyclohexanedione in the Gas Phase. J. Phys. Chem. A 2011, 115, 7494−7502. (19) Samanta, A. K.; Pandey, P.; Bandyopadhyay, B.; Chakraborty, T. Cooperative Strengthening of an Intramolecular OsH···O Hydrogen Bond by a Weak CsH···O Counterpart: Matrix-Isolation Infrared Spectroscopy and Quantum Chemical Studies on 3-Methyl-1,2Cyclohexanedione. J. Phys. Chem. A 2010, 114, 1650−1656. (20) Walzl, K. N.; Xavier, I. M., Jr.; Kuppermann, A. Electron Impact Spectroscopy of Various Diketone Compounds. J. Chem. Phys. 1987, 86, 6701−6706. (21) Balskus, E. P.; Mendez-Andino, J.; Arbit, R. M.; Paquette, L. A. Intercalation of Multiple Carbon Atoms Between the Carbonyls of αDiketones. J. Org. Chem. 2001, 66, 6695−6704. (22) Schwartz, M. P.; Barlow, D. E.; Russell, J. N., Jr.; Weidkamp, K. P.; Butler, J. E.; D’Evelyn, M. P.; Hamers, R. J. Semiconductor SurfaceInduced 1,3-Hydrogen Shift: The Role of Covalent vs Zwitterionic Character. J. Am. Chem. Soc. 2006, 128, 11054−11061. (23) Trivella, A.; Wassermann, T. N.; Mestdagh, J. M.; Manca, C. T.; Marinelli, F.; Roubin, P.; Coussan, S. New Insights into the Photodynamics of Acetylacetone: Isomerisation and Fragmentation in Low-Temp Matrixes. Phys. Chem. Chem. Phys. 2010, 12, 8300− 8310. (24) Horspool, W. M. Enone Cycloadditions and Rearrangements: Photoreactions of Dienones and Quinones. Photochemistry 2005, 35, 17−46. (25) Upadhyaya, H. P.; Kumar, A.; Naik, P. D. Photodissociation Dynamics of Enolic-Acetylacetone at 266, 248, and 193 nm: Mechanism and Nascent State Product Distribution of OH. J. Chem. Phys. 2003, 118, 2590−2598. (26) Naik, P. D.; Kumar, A.; Upadhyaya, H. P.; Bajaj, P.; Sarkar, S. K. Lasers in Chemistry; Wiley: New York, 2008; Vol. 1. (27) Naik, P. D.; Upadhyaya, H. P.; Kumar, A.; Sapre, A. V.; Mittal, J. P. Photodissociation of Carboxylic Acids: Dynamics of OH Formation. J. Photochem. Photobiol. C 2003, 3, 165−182. (28) Dieke, G. H.; Crosswhite, H. M. The Ultraviolet Bands of OH Fundamental Data. J. Quant. Spectrosc. Radiat. Transfer 1962, 2, 97− 199. (29) Alexander, M. H.; Andresen, P.; Bacis, R.; Bersohn, R.; Comes, F. J.; Dagdigian, P. J.; Dixon, R. N.; Field, R. W.; Flynn, G. W.; Gericke, K.-H.; et al. A Nomenclature for Λ-Doublet Levels in Rotating Linear Molecules. J. Chem. Phys. 1988, 89, 1749−1953. (30) Chidsey, I. L.; Crosley, D. R. Calculated Rotational Transition Probabilities for the A-X System of OH. J. Quant. Spectrosc. Radiat. Transfer 1980, 23, 187−199. (31) Vasudev, R.; Zare, R. N.; Dixon, R. N. State-Selected Photodissociation Dynamics: Complete Characterization of the OH Fragment Ejected by the HONO Ã State. J. Chem. Phys. 1984, 80, 4863−4878. (32) Arendt, M. F.; Browning, P. W.; Butler, L. J. Emission Spectroscopy of the Predissociative Excited State Dynamics of Acrolein, Acrylic Acid, and Acryloyl Chloride at 199 nm. J. Chem. Phys. 1995, 103, 5877−5885. (33) Galloway, D. B.; Glenewinkel-Meyer, T.; Bartz, J. A.; Huey, L. G.; Crim, F. F. The Kinetic and Internal Energy of NO from the Photodissociation of Nitrobenzene. J. Chem. Phys. 1994, 100, 1946− 1952. (34) Levine, R. D.; Kinsey, J. L. In Atom-Molecule Collision Theorys: A Guide for the Experimentalist; Bernstein, R. B., Ed.; Plenum Press: New York, 1979. (35) Muckermann, J. T. Information Theoretic Prior Functions for Large Molecular Systems. J. Phys. Chem. 1989, 93, 179−184. (36) Klots, C. E. Thermochemical and Kinetic Information from Metastable Decompositions of Ions. J. Chem. Phys. 1973, 58, 5364− 5367.
length. This information is available free of charge via the Internet at http://pubs.acs.org
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors thank Dr. S. K. Sarkar for their constant guidance and keen interest throughout this work.
■
REFERENCES
(1) Verheijdt, P. L.; Cerfontain, H. Dipole Moments, Spectroscopy, and Ground and Excited State Conformations of Cycloalkane-1,2Diones. J. Chem. Soc., Perkin Trans. 2 1982, 1541−1547. (2) Arnett, J. F.; McGlynn, S. P. Photorotamerism of Aromatic αDicarbonyls. J. Phys. Chem. 1975, 79, 626−629. (3) Nicodem, D. E.; Silva, R. S.; Togashi, D. M.; Fernanda, M.; da Cunha, V. Solvent Effects on the Quenching of the Equilibrating n,π* and π,π* Triplet States of 9,10-Phenanthrenequinone by 2-Propanol. J. Photochem. Photobiol. A 2005, 175, 154−158. (4) Becker, R. S.; Natarajan, L. V.; Lenoble, C.; Harvey, R. G. Photophysics, Photochemistry, and Theoretical Calculations of Some Benz[a]anthracene-3,4-Diones and Their Significance. J. Am. Chem. Soc. 1988, 110, 7163−7167. (5) Charney, E.; Tsai, L. Spectroscopic Examination of the Lower Excited States of α-Diketones. Camphorquinone. J. Am. Chem. Soc. 1971, 93, 7123−7132. (6) Rubin, M. B.; Gleiter, R. The Chemistry of Vicinal Polycarbonyl Compounds. Chem. Rev. 2000, 100, 1121−1164. (7) Malval, J.-P.; Dietlin, C.; Allonas, X.; Fouassier, J.-P. Sterically Tuned Photoreactivity of an Aromatic α-Diketone Family. J. Photochem. Photobiol. A 2007, 192, 66−73. (8) Kupchan, S. M.; Britton, R. W.; Lacadie, J. A.; Ziegler, M. F.; Sigel, C. W. The Isolation and Structural Elucidation of Bruceantin and Bruceantinol, New Potent Antileukemic Quassinoids from Brucea Antidysenterical. J. Org. Chem. 1975, 40, 648−654. (9) Fukamiya, N.; Lee, K.; Muhammad, I.; Murakami, C.; Okano, M.; Harvey, I.; Pelletier, J. Structure−Activity Relationships of Quassinoids for Eukaryotic Protein Synthesis. Cancer Lett. 2005, 220, 37−48. (10) Hoffken, K.; Jonat, W.; Possinger, K.; Kolbel, M.; Kunz, T.; Wagner, H.; Becher, R.; Callies, R.; Friederich, P.; Willmanns, W.; et al. Aromatase Inhibition with 4-Hydroxyandrostenedione in the Treatment of Postmenopausal Patients with Advanced Breast Cancer: A Phase II Study. J. Clin. Oncol. 1990, 8, 875−880. (11) Gianturco, M. A.; Giammarino, A. S.; Pitcher, R. G. The Structures of Five Cyclic Diketones Isolated from Coffee. Tetrahedron 1963, 19, 2051−2059. (12) Mondal, R.; Okhrimenko, A. N.; Shah, B. K.; Neckers, D. C. Photodecarbonylation of α-Diketones: A Mechanistic Study of Reactions Leading to Acenes. J. Phys. Chem. B 2008, 112, 11−15. (13) Cho, D. W.; Lee, H.-Y.; Oh, S. W.; Choi, J. H.; Park, H. J.; Mariano, P. S.; Yoon, U. C. Photoaddition Reactions of 1,2-Diketones with Silyl Ketene Acetals. Formation of β-Hydroxy-γ-Ketoesters. J. Org. Chem. 2008, 73, 4539−4547. (14) Francis, J. T.; Hitchcock, A. P. Distinguishing Keto and Enol Structures by Inner-Shell Spectroscopy. J. Phys. Chem. 1994, 98, 3650−3657. (15) Samanta, A. K.; Pandey, P.; Bandyopadhyay, B.; Chakraborty, T. Keto−Enol Tautomers of 1,2-Cyclohexanedione in Solid, Liquid, Vapour and a Cold Inert Gas Matrix: Infrared Spectroscopy and Quantum Chemistry Calculation. J. Mol. Struct. 2010, 963, 234−239. (16) Shen, Q.; Traetteberg, M.; Samdal, S. The Molecular Structure of Gaseous 1,2-Cyclohexanedione. J. Mol. Struct. 2009, 923, 94−97. 2425
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
The Journal of Physical Chemistry A
Article
(37) Tuck, A. F. Molecular Beam Studies of Ethyl Nitrite Photodissociation. J. Chem. Soc., Faraday Trans. 2 1977, 73, 689−708. (38) North, S. W.; Blank, D. A.; Gezelter, J. D.; Longfellow, C. A.; Lee, Y. T. Evidence for Stepwise Dissociation Dynamics in Acetone at 248 and 193 nm. J. Chem. Phys. 1995, 102, 4447−4460. (39) Naik, P. D.; Upadhyaya, H. P.; Kumar, A.; Sapre, A.; Mittal, J. P. Dynamics of Acetic Acid Dissociation at 193.3 nm: Selectivity in OH Reaction Channel. Chem. Phys. Lett. 2001, 340, 116−122. (40) Upadhyaya, H. P.; Kumar, A.; Naik, P. D.; Sapre, A. V.; Mittal, J. P. Dynamics of OH Formation in the Dissociation of Acrylic Acid in its (n,π*) and (π,π*) Transitions Excited at 248 and 193 nm. J. Chem. Phys. 2002, 117, 10097−10103. (41) Xu, S.; Park, S. T.; Feenstra, J. S.; Srinivasan, R.; Zewail, A. H. Ultrafast Electron Diffraction: Structural Dynamics of the Elimination Reaction of Acetylacetone. J. Phys. Chem. A 2004, 108, 6650−6655. (42) Zhou, C.-H.; Cheng, S.-B.; Sun, J.-L.; Yin, H.-M.; Han, K.-L.; He, G.-Z. Dynamics of OH Formation in the Photodissociation of oNitrobenzoic Acid at 295 and 355 nm. J. Phys. Chem. A 2009, 113, 4923−4929. (43) Yue, X.-F.; Sun, J.-L.; Yin, H.-M.; Wei, Q.; Han, K.-L. Photodissociation Dynamics of Alkyl Nitrites at 266 and 355 nm: The OH Product Channel. J. Phys. Chem. A 2009, 113, 3303−3310.
2426
dx.doi.org/10.1021/jp311251m | J. Phys. Chem. A 2013, 117, 2415−2426
