Picosecond IR-UV pump-probe spectroscopic study on the intramolecular vibrational energy redistribution of NH2 and CH stretching vibrations of jet-cooled aniline

Intramolecular vibrational energy redistribution (IVR) of the NH2 symmetric and asymmetric stretching vibrations of jet-cooled aniline has been investigated by picosecond time-resolved IR-UV pump-probe spectroscopy. A picosecond IR laser pulse excited the NH2 symmetric or asymmetric stretching vibration of aniline in the electronic ground state and the subsequent time evolutions of the excited level as well as redistributed levels were observed by a picosecond UV pulse. The IVR lifetimes for symmetric and asymmetric stretches were obtained to be 18 and 34 ps, respectively. In addition, we obtained the direct evidence that IVR proceeds via two-step bath states; that is, the NH2 stretch energy first flows into the doorway state and the energy is further dissipated into dense bath states. The rate constants of the second step were estimated to be comparable to or slower than those of the first step IVR. The relaxation behavior was compared with that of IVR of the OH stretching vibration of phenol [Y. Yamada, T. Ebata, M. Kayano, and M. Mikami J. Chem. Phys. 120, 7400 (2004)]. We found that the second step IVR process of aniline is much slower than that of phenol, suggesting a large difference of the "doorway state increasing the dense bath states" anharmonic coupling strength between the two molecules. We also observed IVR of the CH stretching vibrations, which showed much faster IVR behavior than that of the NH2 stretches. The fast relaxation is described by the interference effect, which is caused by the coherent excitation of the quasistationary states.     

Theoretical investigation of highly excited vibrational states in DFCO: calculation of the out-of-plane bending states and simulation of the intramolecular vibrational energy redistribution

A previously developed modified Davidson scheme [C. Iung and F. Ribeiro, J. Chem. Phys. 121, 174105 (2005)] is applied to compute and analyze highly excited (nu2,nu6) eigenstates in DFCO. The present paper is also devoted to the simulations of the intramolecular vibrational energy redistribution (IVR) initiated by an excitation of the out-of-plane bending vibration (nnu6, n=2,4,6, . . . ,18, and 20). The multiconfiguration time-dependent Hartree method is exploited to propagate the corresponding six-dimensional wave packets. A comprehensive comparison with experimental data as well as with previous simulations of IVR in HFCO [G. Pasin et al. J. Chem. Phys. 124, 194304 (2006)] is presented.     

Intramolecular vibrational energy redistribution in bridged azulene-anthracene compounds: ballistic energy transport through molecular chains

Intramolecular vibrational energy flow in excited bridged azulene-anthracene compounds is investigated by time-resolved pump-probe laser spectroscopy. The bridges consist of molecular chains and are of the type (CH(2))(m) with m up to 6 as well as (CH(2)OCH(2))(n) (n=1,2) and CH(2)SCH(2). After light absorption into the azulene S(1) band and subsequent fast internal conversion, excited molecules are formed where the vibrational energy is localized at the azulene side. The vibrational energy transfer through the molecular bridge to the anthracene side and, finally, to the surrounding medium is followed by probing the red edge of the azulene S(3) absorption band at 300 nm and/or the anthracene S(1) absorption band at 400 nm. In order to separate the time scales for intramolecular and intermolecular energy transfer, most of the experiments were performed in supercritical xenon where vibrational energy transfer to the bath is comparably slow. The intramolecular equilibration proceeds in two steps. About 15%-20% of the excitation energy leaves the azulene side within a short period of 300 fs. This component accompanies the intramolecular vibrational energy redistribution (IVR) within the azulene chromophore and it is caused by dephasing of normal modes contributing to the initial local excitation of the azulene side and extending over large parts of the molecule. Later, IVR in the whole molecule takes place transferring vibrational energy from the azulene through the bridge to the anthracene side and thereby leading to microcanonical equilibrium. The corresponding time constants tau(IVR) for short bridges increase with the chain length. For longer bridges consisting of more than three elements, however, tau(IVR) is constant at around 4-5 ps. Comparison with molecular dynamics simulations suggests that the coupling of these chains to the two chromophores limits the rate of intramolecular vibrational energy transfer. Inside the bridges the energy transport is essentially ballistic and, therefore, tau(IVR) is independent on the length.     

Quasiclassical trajectory simulations of OH(v) + NO2 --> HONO2* --> OH(v') + NO2: capture and vibrational deactivation rate constants

Quasiclassical trajectory calculations are used to investigate the dynamics of the OH(v) + NO(2) --> HONO(2) --> OH(v') + NO(2) recombination/dissociation reaction on an analytic potential energy surface (PES) that gives good agreement with the known structure and vibrational frequencies of nitric acid. The calculated recombination rate constants depend only weakly on temperature and on the initial vibrational energy level of OH(v). The magnitude of the recombination rate constant is sensitive to the potential function describing the newly formed bond and to the switching functions in the PES that attenuate inter-mode interactions at long range. The lifetime of the nascent excited HONO(2) depends strongly not only on its internal energy but also on the identity of the initial state, in disagreement with statistical theory. This disagreement is probably due to the effects of slow intramolecular vibrational energy redistribution (IVR) from the initially excited OH stretching mode. The vibrational energy distribution of product OH(v') radicals is different from statistical distributions, a result consistent with the effects of slow IVR. Nonetheless, the trajectory results predict that vibrational deactivation of OH(v) via the HONO(2) transient complex is approximately 90% efficient, almost independent of initial OH(v) vibrational level, in qualitative agreement with recent experiments. Tests are also carried out using the HONO(2) PES, but assuming the weaker O-O bond strength found in HOONO (peroxynitrous acid). In this case, the predicted vibrational deactivation efficiencies are significantly lower and depend strongly on the initial vibrational state of OH(v), in disagreement with experiments. This disagreement suggests that the actual HOONO PES may contain more inter-mode coupling than found in the present model PES, which is based on HONO(2). For nitric acid, the measured vibrational deactivation rate constant is a useful proxy for the recombination rate, but IVR randomization of energy is not complete, suggesting that the efficacy of the proxy method must be evaluated on a case-by-case basis.     

Femtosecond Time-Resolved Stimulated Raman Spectroscopy: Application to the Ultrafast Internal Conversion in beta-Carotene

We have developed the technique of femtosecond stimulated Raman spectroscopy (FSRS), which allows the rapid collection of high-resolution vibrational spectra on the femtosecond time scale. FSRS combines a sub-50 fs actinic pump pulse with a two-pulse stimulated Raman probe to obtain vibrational spectra whose frequency resolution limits are uncoupled from the time resolution. This allows the acquisition of spectra with <100 fs time resolution and <30 cm(-1) frequency resolution. Additionally, FSRS is unaffected by background fluorescence, provides rapid (100 ms) acquisition times, and exhibits traditional spontaneous Raman line shapes. FSRS is used here to study the relaxation dynamics of beta-carotene. Following optical excitation to S(2) (1B(u) (+)) the molecule relaxes in 160 fs to S(1) (2A(g) (-)) and then undergoes two distinct stages of intramolecular vibrational energy redistribution (IVR) with 200 and 450 fs time constants. These processes are attributed to rapid (200 fs) distribution of the internal conversion energy from the S(1) C=C modes into a restricted bath of anharmonically coupled modes followed by complete IVR in 450 fs. FSRS is a valuable new technique for studying the vibrational structure of chemical reaction intermediates and transition states.     

Intramolecular vibrational redistribution in Ne-Br2: the signature of intermediate resonances in the excitation spectrum

Quantum-mechanical simulations of the Ne-Br(2)(B,v') excitation spectra produced after vibrational predissociation in the v'=20-35 range are reported. The aim is to investigate the signature in the excitation spectra of intermediate resonances lying in the lower v相似文献   

Vibrational relaxation of CH3I in the gas phase and in solution

Transient electronic absorption measurements reveal the vibrational relaxation dynamics of CH(3)I following excitation of the C-H stretch overtone in the gas phase and in liquid solutions. The isolated molecule relaxes through two stages of intramolecular vibrational relaxation (IVR), a fast component that occurs in a few picoseconds and a slow component that takes place in about 400 ps. In contrast, a single 5-7 ps component of IVR precedes intermolecular energy transfer (IET) to the solvent, which dissipates energy from the molecule in 50 ps, 44 ps, and 16 ps for 1 M solutions of CH(3)I in CCl(4), CDCl(3), and (CD(3))(2)CO, respectively. The vibrational state structure suggests a model for the relaxation dynamics in which a fast component of IVR populates the states that are most strongly coupled to the initially excited C-H stretch overtone, regardless of the environment, and the remaining, weakly coupled states result in a secondary relaxation only in the absence of IET.     

On the excess-energy dependence of radiationless decay rate constants

The results of calculations of the dependence of the radiationless rate constant on the excess of excitation energy within the two-electronic states model under the weak coupling and statistical limits are presented. It is assumed that the exact molecular states for a given electronic configuration are global in character containing equal contributions from all degenerated vibrational levels at a given excitation energy due to intramolecular vibrational relaxation (IVR). The results of calculations indicate an important role of the low-frequency vibrational modes, the potential energy surfaces of which cross between the two electronic states involved into the radiationless process. The sharp increase of the rate constant is predicted for the excitation energy below the diabatic crossing point, followed by saturation at higher energies. The calculated rate constants for the T₁→S₀ intersystem crossing in pyrazine and benzene are in good agreement with experimental observations. Some comments concerning the “channel-three” phenomenon in benzene are presented.     

Impact of methyl rotor in the excited state level mixing of doubly hydrogen-bonded complexes of 2-pyridone

We have presented in this paper the laser-induced fluorescence excitation and resolved fluorescence spectra of five 1:1 hydrogen-bonded complexes of 2-pyridone (2PY) with formic acid (FA), acetic acid (AA), propanoic acid (PA), formamide (FM), and acetamide (AM). The resolved fluorescence spectra, measured following excitation to different single vibronic levels of the dimers indicate that the intermolecular hydrogen bond vibrations undergo mixing with a number of intramolecular modes of the 2PY moiety in the excited state. A comparison of the emission spectral features of these dimers clearly indicates that the methyl groups belonging to the AA and AM moieties spectacularly accelerate the vibrational energy redistribution (IVR) in the 2PY moiety. On the other hand, although the molecular size of PA is bigger than AA, the spectral features of the 2PY-PA dimer bear signatures of a slower IVR rate compared to those of 2PY-AA. We propose that hyperconjugation of the methyl group with the cyclic hydrogen-bonded network involving AA and AM is responsible for the observed phenomenon.     

The intramolecular vibrational energy redistribution threshold in S(1) deuterated p-difluorobenzene

The intramolecular vibrational energy redistribution (IVR) in S(1) deuterated p-difluorobenzene (pDFB-d(4) or -d(4)) has been studied to determine the IVR threshold. For this, the S(1) <-- S(0) fluorescence excitation (FE) spectrum of jet-cooled d(4) was investigated in the 2000-3250 cm(-1) vibronic energy range of the S(1) electronic state, and single vibronic level fluorescence (SVLF) spectra have been acquired by exciting selected levels lying between 750 and 2850 cm(-1) in vibrational energy in the S(1) excited state. Congestion of the dispersed fluorescence in this molecule first appears as the vibrational level energy climbs above 2000 cm(-1). By comparing the SVLF spectra of pDFB-d(4) with those of p-difluorobenzene (pDFB or -h(4)), it is obvious that IVR threshold in -d(4) is localized with a few hundreds cm(-1) lower than that in pDFB. This decrease is entirely due to the increase in vibrational state density due to deuteration.     

Molecular dynamics simulations and instantaneous normal-mode analysis of the vibrational relaxation of the C-H stretching modes of N-methylacetamide-d in liquid deuterated water

Nonequilibrium molecular dynamics (MD) simulations and instantaneous normal mode (INMs) analyses are used to study the vibrational relaxation of the C-H stretching modes (ν(s)(CH?)) of deuterated N-methylacetamide (NMAD) in aqueous (D2O) solution. The INMs are identified unequivocally in terms of the equilibrium normal modes (ENMs), or groups of them, using a restricted version of the recently proposed Min-Cost assignment method. After excitation of the parent ν(s)(CH?) modes with one vibrational quantum, the vibrational energy is shown to dissipate through both intramolecular vibrational redistribution (IVR) and intermolecular vibrational energy transfer (VET). The decay of the vibrational energy of the ν(s)(CH?) modes is well fitted to a triple exponential function, with each characterizing a well-defined stage of the entire relaxation process. The first, and major, relaxation stage corresponds to a coherent ultrashort (τ(rel) = 0.07 ps) energy transfer from the parent ν(s)(CH?) modes to the methyl bending modes δ(CH?), so that the initially excited state rapidly evolves into a mixed stretch-bend state. In the second stage, characterized by a time of 0.92 ps, the vibrational energy flows through IVR to a number of mid-range-energy vibrations of the solute. In the third stage, the vibrational energy accumulated in the excited modes dissipates into the bath through an indirect VET process mediated by lower-energy modes, on a time scale of 10.6 ps. All the specific relaxation channels participating in the whole relaxation process are properly identified. The results from the simulations are finally compared with the recent experimental measurements of the ν(s)(CH?) vibrational energy relaxation in NMAD/D?O(l) reported by Dlott et al. (J. Phys. Chem. A 2009, 113, 75.) using ultrafast infrared-Raman spectroscopy.     

Intramolecular vibrational dynamics in S1 p-fluorotoluene. I. Direct observation of doorway states

Picosecond time-resolved photoelectron spectroscopy is used to investigate intramolecular vibrational redistribution (IVR) following excitation of S(1) 18a(1) in p-fluorotoluene (pFT) at an internal energy of 845 cm(-1), where ν(18a) is a ring bending vibrational mode. Characteristic oscillations with periods of 8 ps and 5 ps are observed in the photoelectron signal and attributed to coupling between the initially excited zero-order bright state and two doorway states. Values for the coupling coefficients connecting these three vibrational states have been determined. In addition, an exponential change in photoelectron signal with a lifetime of 17 ps is attributed to weaker couplings with a bath of dark states that play a more significant role during the latter stages of IVR. A tier model has been used to assign the most strongly coupled doorway state to S(1) 17a(1) 6a(2)('), where ν(17a) is a CH out-of-plane vibrational mode and 6a(2)(') is a methyl torsional level. This assignment signifies that a torsion-vibration coupling mechanism mediates the observed dynamics, thus demonstrating the important role played by the methyl torsional mode in accelerating IVR.     

Vibrational overtone initiated unimolecular dissociation of HOCH2OOH and HOCD2OOH: evidence for mode selective behavior

The vibrational overtone induced unimolecular dissociation of HMHP (HOCH(2)OOH) and HMHP-d(2) (HOCD(2)OOH) into OH and HOCH(2)O (HOCD(2)O) fragments is investigated in the region of the 4nu(OH) and 5nu(OH) bands. The unimolecular dissociation rates in the threshold region, corresponding to the 4nu(OH) band, exhibit measurable differences associated with excitation of the OH stretch of the alcohol versus the peroxide functional group, with the higher energy alcohol OH stretching state exhibiting a slower dissociation rate compared to the lower energy peroxide OH stretch in both HMHP and HMHP-d(2). Predictions using the Rice-Ramsperger-Kassel-Marcus theory give rates that are in reasonably good agreement with the measured dissociation rate for the alcohol OH stretch but considerably differ from the measured rates for the peroxide OH stretch in both isotopomers. The present results are interpreted as suggesting that the extent of intramolecular vibrational energy redistribution (IVR) is different for the two OH stretching states associated with the two functional groups in HMHP, with IVR being substantially less complete for the peroxide OH stretch. Analysis of the OH fragment product state distributions in conjunction with phase-space theory simulation gives a D(0) value of 38+/-0.7 kcal/mole for breaking the peroxide bond in HMHP.     

Evaluation of data for atmospheric models: master equation/RRKM calculations on the combination reaction, BrO + NO2 --> BrONO2, a conundrum

Experimental data for the title reaction were modeled using master equation (ME)/RRKM methods based on the Multiwell suite of programs. The starting point for the exercise was the empirical fitting provided by the NASA (Sander, S. P.; Finlayson-Pitts, B. J.; Friedl, R. R.; Golden, D. M.; Huie, R. E.; Kolb, C. E.; Kurylo, M. J.; Molina, M. J.; Moortgat, G. K.; Orkin, V. L.; Ravishankara, A. R. Chemical Kinetics and Photochemical Data for Use in Atmospheric Studies, Evaluation Number 15; Jet Propulsion Laboratory: Pasadena, California, 2006)1 and IUPAC (Atkinson, R.; Baulch, D. L.; Cox, R. A.; R. F. Hampson, J.; Kerr, J. A.; Rossi, M. J.; Troe, J. J. Phys. Chem. Ref. Data 2000, 29, 167)2 data evaluation panels, which represents the data in the experimental pressure ranges rather well. Despite the availability of quite reliable parameters for these calculations (molecular vibrational frequencies (Parthiban, S.; Lee, T. J. J. Chem. Phys. 2000, 113, 145)3 and a value (Orlando, J. J.; Tyndall, G. S. J. Phys. Chem. 1996, 100, 19398)4 of the bond dissociation energy, D298(BrO-NO2) = 118 kJ mol-1, corresponding to DeltaH0o = 114.3 kJ mol-1 at 0 K) and the use of RRKM/ME methods, fitting calculations to the reported data or the empirical equations was anything but straightforward. Using these molecular parameters resulted in a discrepancy between the calculations and the database of rate constants of a factor of ca. 4 at, or close to, the low-pressure limit. Agreement between calculation and experiment could be achieved in two ways, either by increasing DeltaH0o to an unrealistically high value (149.3 kJ mol-1) or by increasing DeltaEd, the average energy transferred in a downward collision, to an unusually large value (>5000 cm-1). The discrepancy could also be reduced by making all overall rotations fully active. The system was relatively insensitive to changing the moments of inertia in the transition state to increase the centrifugal effect. The possibility of involvement of BrOONO was tested and cannot account for the difficulties of fitting the data.     

Theoretical study on the vibrational structure of S0 thiophosgene from the origin to dissociation

In this work, using our vibrational variational calculation method and a recently derived refined quartic potential energy surface for S₀ thiophosgene, we have carried out large scale vibrational calculations to analyze the vibrational structure of this electronic state in the whole range of vibrational excitation energies down from the origin and up to the dissociation limit (at ↼20,000 cm^↙1). In the lower excited vibrational range we have achieved satisfactory coincidence of calculated to experimentally measured frequencies, while at the higher vibrational excitations our main objective has been to estimate what part of the available vibrational level density is effectively involved into the vibrational mixing and IVR. The results from our calculations have been compared to the available experimentally obtained dataset (obtained from synchrotron IR, SEP and LIF spectra) as well as to conclusions from the analyses by other authors using local coupling models.     

Test of a chemical timing method for measuring absolute vibrational relaxation rate constants for S1 p-difluorobenzene

A chemical timing (CT) method for measuring absolute rate constants for collisional vibrational relaxation has been tested for the 5(1) state of S(1) p-difluorobenzene (pDFB) where an alternative method exists to provide benchmark values. The CT method was originally developed to treat vibrational energy transfer (VET) in large molecules excited to high vibrational levels where the intramolecular vibrational redistribution (IVR) resulting from large vibrational state densities completely eliminates vibrational structure in the emission spectrum. Here we apply the same method to a low-lying state (5(1) with epsilon(vib) = 818 cm(-1)) located in the low-density region of the vibrational manifold where IVR plays no role. For high vibrational levels, the chemical timing method involves addition of high O(2) pressures (kTorr) to a low-pressure pDFB sample, introducing vibrational structure in the fluorescence spectrum. Response of this spectrum to vibrational relaxation by Ar is then examined. For levels such as 5(1), the fully structured fluorescence spectrum allows the rate constant for single-collision VET into the surrounding vibrational field to be measured directly without the presence of O(2). The measurements of 5(1) VET have been repeated with various O(2) pressures (kTorr) for comparison with the O(2)-free benchmark. In the presence of O(2), the rate constant for VET by Ar is (4.0 +/- 0.5) x 10(6) Torr(-1) s(-1) and independent of high O(2) pressure variations. The rate constant as found by the standard O(2)-free method is (3.6 +/- 0.4) x 10(6) Torr(-1) s(-1). This comparison suggests that the chemical timing method is capable of providing a reasonably accurate measure of the VET rate constant for high vibrational levels provided that details of the kinetics are known.     

Accurate potential energy curve for B2. Ab initio elucidation of the experimentally elusive ground state rotation-vibration spectrum

The electron-deficient diatomic boron molecule has long puzzled scientists. As yet, the complete set of bound vibrational energy levels is far from being known, experimentally as well as theoretically. In the present ab initio study, all rotational-vibrational levels of the X (3)Σ(g)(-) ground state are determined up to the dissociation limit with near-spectroscopic accuracy (<10 cm(-1)). Two complete sets of bound vibrational levels for the (11)B(2) and (11)B-(10)B isotopomers, containing 38 and 37 levels, respectively, are reported. The results are based on a highly accurate potential energy curve, which also includes relativistic effects. The calculated set of all vibrational levels of the (11)B(2) isotopomer is compared with the few results derived from experiment [Bredohl, H.; Dubois, I.; and Nzohabonayo, P. J. Mol. Spectrosc. 1982, 93, 281; Bredohl, H.; Dubois, I.; and Melen, F. J. Mol. Spectrosc. 1987, 121, 128]. Theory agrees with experiment within 4.5 cm(-1) on average for the four vibrational level spacings that are so far known empirically. In addition, the present theoretical analysis suggests, however, that the transitions from higher electronic states to the ground state vibrational levels v = 12-15 deserve to be reanalyzed. Whereas previous experimental investigators considered them to originate from the v' = 0 vibrational level of the upper state (2)(3)Σ(u)(-), the present results make it likely that these transitions originate from a different upper state, namely the v' = 16 or the v' = 17 vibrational level of the (1)(3)Σ(u)(-) state. The ground state dissociation energy D(0) is predicted to be 23164 cm(-1).     

Effects of Vibrational and Rotational Excitations on Dissociative Chemisorption Dynamics of N2 on Fe(111)

The dissociative chemisorption of N\begin{document}$ _2 $\end{document} is the rate-limiting step for ammonia synthesis in industry. Here, we investigated the role of initially vibrational excitation and rotational excitation of N\begin{document}$ _2 $\end{document} for its reactivity on the Fe(111) surface, based on a recently developed six-dimensional potential energy surface. Six-dimensional quantum dynamics study was carried out to investigate the effect of vibrational excitation for incidence energy below 1.6 eV, due to significant quantum effects for this reaction. The effects of vibrational and rotational excitations at high incidence energies were revealed by quasiclassical trajectory calculations. We found that raising the translational energy can enhance the dissociation probability to some extent, however, the vibrational excitation or rotational excitation can promote dissociation more efficiently than the same amount of translational energy. This study provides valuable insight into the mode-specific dynamics of this heavy diatom-surface reaction.     

Ultrafast decay of electronically excited singlet cytosine via a pi,pi* to n(O),pi* state switch

Singlet fluorescence lifetimes of adenosine, cytidine, guanosine, and thymidine, determined by femtosecond pump-probe spectroscopy (Pecourt, J.-M. L.; Peon, J.; Kohler, B. J. Am. Chem. Soc. 2000, 122, 9348. Pecourt, J.-M. L.; Peon, J.; Kohler, B. J. Am. Chem. Soc. 2001, 123, 10370), show that the excited states produced by 263 nm light in these nucleosides decay in the subpicosecond range (290-720 fs). Ultrafast radiationless decay to the ground state greatly reduces the probability of photochemical damage. In this work we present a theoretical study of isolated cytosine, the chromophore of cytidine. The experimental lifetime of 720 fs indicates that there must be an ultrafast decay channel for this species. We have documented the possible decay channels and approximate energetics, using a valence-bond derived analysis to rationalize the structural details of the paths. The mechanism favored by our calculations and the experimental data involves (1) a two-mode decay coordinate composed of initial bond length inversion followed by internal vibrational energy redistribution (IVR) to populate a carbon pyramidalization mode, (2) a state switch between the pi,pi* and nO,pi* (excitation from oxygen lone pair) excited states, and (3) decay to the ground state through a conical intersention. A second decay path through the nN,pi* state (excitation from the nitrogen lone pair), with a higher barrier, involves out-of-plane bending of the amino substituent.     

Ab initio molecular dynamics studies of tetrasulfur. Dynamics coupling the C2v open and D2h closed forms of S4

Ab initio molecular dynamics (AIMD) simulations were performed on the closed D(2h) and open C(2v) isomers of tetrasulfur. After a careful calibration of the electronic structure method, the calculations were done using the BPW91/aug-cc-pVTZ method. This combination of method/basis set adequately reproduces the relative benchmark CCSD(T) energy difference [Matus, M.; Dixon, D.; Peterson, K. A.; Harkless, J. A. W.; Francisco, J. S. J. Chem. Phys. 2007, 127, 174305] between these two isomers and, crucially, the fact that the D(2h) structure is a transition state linking two equivalent (mirror images) C(2v) isomers. The trajectories show that the symmetric open C(2v) isomers interconvert when passing through the D(2h) closed transition state structure and that, unlike tetraoxygen, no three-dimensional structures arise. The dynamic vibrational analysis yields peaks in good agreement with the static CCSD(T) harmonic frequencies and explains higher peaks as overtones, thus showing that unlike previous AIMD DFT-based approaches, carefully calibrated exchange-correlation functionals can produce reliable molecular dynamics results for complex PESs as the one corresponding to the lowest singlet of S(4).