Time-resolved photoelectron spectroscopy: from wavepackets to observables

Time-resolved photoelectron spectroscopy (TRPES) is a powerful tool for the study of intramolecular dynamics, particularly excited state non-adiabatic dynamics in polyatomic molecules. Depending on the problem at hand, different levels of TRPES measurements can be performed: time-resolved photoelectron yield; time- and energy-resolved photoelectron yield; time-, energy-, and angle-resolved photoelectron yield. In this pedagogical overview, a conceptual framework for time-resolved photoionization measurements is presented, together with discussion of relevant theory for the different aspects of TRPES. Simple models are used to illustrate the theory, and key concepts are further amplified by experimental examples. These examples are chosen to show the application of TRPES to the investigation of a range of problems in the excited state dynamics of molecules: from the simplest vibrational wavepacket on a single potential energy surface; to disentangling intrinsically coupled electronic and nuclear motions; to identifying the electronic character of the intermediate states involved in non-adiabatic dynamics by angle-resolved measurements in the molecular frame, the most complete measurement.     

Transient vibronic structure in ultrafast fluorescence spectra of photoactive yellow protein

The ultrafast photo-induced dynamics of wild-type photoactive yellow protein and its site-directed mutant of E46Q in aqueous solution was studied at room temperature by femtosecond fluorescence spectroscopy using the optical Kerr-gate method. The vibronic structure appears, depending on the excitation photon energy, in the time-resolved fluorescence spectra just after photoexcitation, which winds with time and disappears on a time scale of sub-picoseconds. This result indicates that the wavepacket is localized in the electronic excited state followed by dumped oscillations and broadening, and also that the initial condition of the wavepacket prepared depending on the excitation photon energy affects much the following ultrafast dynamics in the electronic excited state.     

Real-time observation of intramolecular proton transfer in the electronic ground state of chloromalonaldehyde: an ab initio study of time-resolved photoelectron spectra

The authors report on studies of time-resolved photoelectron spectra of intramolecular proton transfer in the ground state of chloromalonaldehyde, employing ab initio photoionization matrix elements and effective potential surfaces of reduced dimensionality, wherein the couplings of proton motion to the other molecular vibrational modes are embedded by averaging over classical trajectories. In the simulations, population is transferred from the vibrational ground state to vibrationally hot wave packets by pumping to an excited electronic state and dumping with a time-delayed pulse. These pump-dump-probe simulations demonstrate that the time-resolved photoelectron spectra track proton transfer in the electronic ground state well and, furthermore, that the geometry dependence of the matrix elements enhances the tracking compared with signals obtained with the Condon approximation. Photoelectron kinetic energy distributions arising from wave packets localized in different basins are also distinguishable and could be understood, as expected, on the basis of the strength of the optical couplings in different regions of the ground state potential surface and the Franck-Condon overlaps of the ground state wave packets with the vibrational eigenstates of the ion potential surface.     

Observation of a simple vibrational wavepacket in a polyatomic molecule via time-resolved photoelectron velocity-map imaging: a prototype for time-resolved IVR studies

We have prepared a coherent superposition of the two components of a Fermi resonance in the S1 state of toluene at approximately 460 cm(-1) with a approximately 1 ps laser pulse and monitored time-resolved photoelectron velocity-map images. The photoelectron intensities oscillate with time in a manner that depends on their kinetic energy, even though full vibrational resolution in the cation is not achieved. Analysis of the time-dependent photoelectron spectra enables information on the composition of the S1 wavepacket to be deduced. Such an experiment, in which a whole set of partially dispersed cation vibrational states are detected simultaneously, suggests an efficient method of studying intramolecular vibrational energy redistribution processes in excited states.     

Towards disentangling coupled electronic-vibrational dynamics in ultrafast non-adiabatic processes

Femtosecond time-resolved photoelectron spectroscopy is emerging as a new technique for investigating polyatomic excited state dynamics. Due to the sensitivity of photoelectron spectroscopy to both electronic configurations and vibrational dynamics, it is well suited to the study of non-adiabatic processes such as internal conversion, which often occur on sub-picosecond time scales. We discuss the technical requirements for such experiments, including lasers systems, energy- and angle-resolved photoelectron spectrometers and new detectors for coincidence experiments. We present a few examples of these methods applied to problems in diatomic wavepacket dynamics and ultrafast non-adiabatic processes in polyatomic molecules.     

Excited-state decay of hydrocarbon radicals, investigated by femtosecond time-resolved photoionization: ethyl, propargyl, and benzyl

The excited state decay of the hydrocarbon radicals ethyl, C(2)H(5); propargyl, C(3)H(3); and benzyl, C(7)H(7) was investigated by femtosecond time-resolved photoionization. Radicals were generated by flash pyrolysis of n-propyl nitrite, propargyl bromide, and toluene, respectively. It is shown that the 2 (2)A(') (3s) Rydberg state of ethyl excited at 250 nm decays with a time constant of 20 fs. No residual signal was observed at longer delay times. For the 3 (2)B(1) state of propargyl excited at 255 nm a slower decay with a time constant 50+/-10 fs was determined. The 4 (2)B(2) state of benzyl excited at 255 nm decays within 150+/-30 fs.     

Time-dependent photoionization of azulene: competition between ionization and relaxation in highly excited states

Pump-probe photoionization has been used to map the relaxation processes taking place from highly vibrationally excited levels of the S(2) state of azulene, populated directly or via internal conversion from the S(4) state. Photoelectron spectra obtained by 1+2(') two-color time-resolved photoelectron imaging are invariant (apart from in intensity) to the pump-probe time delay and to the pump wavelength. This reveals a photoionization process which is driven by an unstable electronic state (e.g., doubly excited state) lying below the ionization potential. This state is postulated to be populated by a probe transition from S(2) and to rapidly relax via an Auger-like process onto highly vibrationally excited Rydberg states. This accounts for the time invariance of the photoelectron spectrum. The intensity of the photoelectron spectrum is proportional to the population in S(2). An exponential energy gap law is used to describe the internal conversion rate from S(2) to S(0). The vibronic coupling strength is found to be larger than 60+/-5 microeV.     

Ultrafast dynamics of isolated fluorenone

The ultrafast dynamics of isolated 9-fluorenone was studied by femtosecond time-resolved photoionization and photoelectron spectroscopy. The molecule was excited around 264-266 nm into the S(6) state. The experimental results indicate that the excitation is followed by a multistep deactivation. A time constant of 50 fs or less corresponds to a fast redistribution of energy within the initially excited manifold of states, i.e., a motion away from the Franck-Condon region. Internal conversion to the S(1) state then proceeds within 0.4 ps. The S(1) state is long-lived, and only a lower bound of 20 ps can be derived. In addition, we computed excited state energies and oscillator strengths by TD-DFT theory, supporting the interpretation of the experimental data.     

Monitoring the effect of a control pulse on a conical intersection by time-resolved photoelectron spectroscopy

We have previously shown how femtosecond angle- and energy-resolved photoelectron spectroscopy can be used to monitor quantum wavepacket bifurcation at an avoided crossing or conical intersection and also how a symmetry-allowed conical intersection can be effectively morphed into an avoided crossing by photo-induced symmetry breaking. The latter result suggests that varying the parameters of a laser to modify a conical intersection might control the rate of passage of wavepackets through such regions, providing a gating process for different chemical products. In this paper, we show with full quantum mechanical calculations that such optical control of conical intersections can actually be monitored in real time with femtosecond angle- and energy-resolved photoelectron spectroscopy. In turn, this suggests that one can optimally control the gating process at a conical intersection by monitoring the photoelectron velocity map images, which should provide far more efficient and rapid optimal control than measuring the ratio of products. To demonstrate the sensitivity of time-resolved photoelectron spectra for detecting the consequences of such optical control, as well as for monitoring how the wavepacket bifurcation is affected by the control, we report results for quantum wavepackets going through the region of the symmetry-allowed conical intersection between the first two (2)A' states of NO(2) that is transformed to an avoided crossing. Geometry- and energy-dependent photoionization matrix elements are explicitly incorporated in these studies. Time-resolved photoelectron angular distributions and photoelectron images are seen to systematically reflect the effects of the control pulse.     

Femtosecond dynamics of isolated phenylcarbenes

Understanding the primary photophysical processes in molecules is essential for interpreting their photochemistry, because molecules rarely react from the initially excited electronic state. In this study the ultrafast excited-state dynamics of chlorophenylcarbene (CPC) and trifluoromethylphenylcarbene (TFPC), two species that are considered as models for carbene dynamics, were investigated by femtosecond time-resolved pump probe spectroscopy in the gas phase. Their dynamics was followed in real time by time-resolved photoionization and photoelectron imaging. CPC was excited at 265 nm into the 3 1A' state, corresponding to excitation from a pi-orbital of the aromatic ring into the LUMO. The LUMO contains a contribution of the p-orbital at the carbene center. Three time constants are apparent in the photoelectron images: A fast decay process with tau1 approximately 40 fs, a second time constant of tau2 approximatley 350 fs, and an additional time constant of tau3 approximately 1 ps. The third time constant is only visible in the time-dependence of low kinetic energy electrons. Due to the dense manifold of excited states between 3.9 and 5 eV, known from ab initio calculations, the recorded time-resolved electron images show broad and unstructured bands. A clear population transfer between the states thus can not directly be observed. The fast deactivation process is linked to either a population transfer between the strongly coupled excited states between 3.9 and 5.0 eV or the movement of the produced wave packet out of the Franck-Condon region. Since the third long time constant is only visible for photoelectrons at low kinetic energy, evidence is given that this time constant corresponds to the lifetime of the lowest excited A 1A' state. The remaining time constant reflects a deactivation of the manifold of states in the range 3.9-5.0 eV down to the A 1A' state.     

Aqueous photochemistry of bisphenol <Emphasis Type="Italic">E</Emphasis> in the presence of β-cyclodextrin

Nanosecond laser flash photolysis and time-resolved fluorescence were used to study photochemistry of bis(4-hydroxyphenyl)ethane (Bisphenol E, BPE) and complex of BPE with β-cyclodextrin (BPE-CD) in aqueous solutions. For both systems the primary photochemical process was found to be photoionization with the formation of a hydrated electron—phenoxyl radical pair. Inclusion of BPE in cyclodextrin cavity leads to the increase of photoionization and fluorescence quantum yield (from 0.009 to 0.16) as well as fluorescence lifetime (from 0.07 to 2.5 ns) due to decreasing of the quenching rate of the singlet excited state of complexed BPE by solvent molecules.     

EXCITED STATE DYNAMICS OF RETINAL PROTEINS AS STUDIED BY FOURIER TRANSFORM OF OPTICAL ABSORPTION SPECTRUM—I. DEVELOPMENT OF ANALYTICAL METHOD

Abstract— We significantly improved the analytical method for the study of excited state dynamics of pigments, by means of the time correlation function (tcf) of the vibrational wavepacket which is produced by the Fourier transform of experimentally obtained optical absorption spectra (FTOA). Applying the tcf method to the spectra of rhodopsin at 0°C and -180°C, we observed specific peaks which are slightly different between 0°C and -180°C in the early time region (1–130 fs) of the absolute value of tcf, representing a characteristic propagation of the wavepacket along a reaction coordinate pertinent to the cis-trans photoisomerization of the chromophore accompanying the motion of protein moiety. From the analysis of phase angle propagation, we obtained a rather small relaxation energy, 6–7 kcal/mol. Based on these results, we can say that FTOA analysis is useful as one of the most powerful techniques for the study of very early procedures in the excited state dynamics of pigments.     

Pronounced non-condon effect as the origin of the quantum beat observed in the time-resolved absorption signal from excited-state cis-stilbene

We carried out wavelength-dispersed time-resolved absorption measurements of cis-stilbene to investigate the mechanism of the appearance of the approximately 220 cm-1 oscillation, which has been assigned to the nuclear wavepacket motion of the S1 state. The observed oscillatory pattern showed almost same amplitude and phase across the absorption peak at 645 nm, which indicates that the modulation of the transition intensity gives rise to the quantum beat. We also carried out a semiquantitative numerical simulation of the time-resolved absorption spectra based on the effective linear response theory, in which we newly incorporated the Herzberg-Teller coupling model by introducing a coordinate-dependence of the transition moment. The results of these experiments and simulation clearly showed that the intensity of the quantum beat arises from a significant coordinate dependence of the Sn<--S1 transition moment, i.e., non-Condon effect. It was concluded that the vibronic coupling of the Sn state with other electronic states is so large that the Herzberg-Teller coupling predominantly contributes to the intensity of the quantum beat of the totally symmetric approximately 220 cm-1 vibration. The present work suggests a general importance of the non-Condon effect in spectroscopy involving highly excited electronic states.     

Quantum control spectroscopy of vibrational modes: Comparison of control scenarios for ground and excited states in β-carotene

Quantum control spectroscopy (QCS) is used as a tool to study, address selectively and enhance vibrational wavepacket motion in large solvated molecules. By contrasting the application of Fourier-limited and phase-modulated excitation on different electronic states, the interplay between the controllability of vibrational coherence and electronic resonance is revealed. We contrast control on electronic ground and excited state by introducing an additional pump beam prior to a DFWM-sequence (Pump-DFWM). Via phase modulation of this initial pump pulse, coherent control is extended to structural evolution on the vibrationally hot ground state (hot-S₀) and lowest lying excited state (S₁) of β-carotene. In an open loop setup, the control scenarios for these different electronic states are compared in their effectiveness and mechanism.     

Relevance of pi sigma* states in the photoinduced processes of adenine, adenine dimer, and adenine-water complexes

Ab initio calculations and time-resolved photoionization spectroscopy were carried out to characterize the role of the lowest two pi sigma* excited states for the photoinduced processes in the adenine monomer, adenine dimer, and adenine-water clusters. The calculations show--with respect to the monomer--a stabilization of 0.11-0.14 eV for the pi sigma* states in different isomers of adenine dimer and an even bigger stabilization of 0.14-0.36 eV for isomers of adenine-(H2O)1 and adenine-(H2O)3. Hence, the stabilized pi sigma* states should play an important role in the excited-state relaxation of partially or fully solvated adenine. This conclusion is supported by experimental results: In the adenine monomer, strong n pi* state signals are observed. Those signals are reduced in adenine dimer and vanish in water clusters due to the competing relaxation via the pi sigma* states.     

Near threshold photoionization of the ground and first excited states of C(2)

Calculations using the multichannel Schwinger configuration-interaction method are presented for the photoionization from the ground and the first excited states of the C(2) molecule. Both single channel and multichannel calculations are presented in a photon energy range from the threshold to about 50 eV of photon energy. For the ground state, inclusion of both intrinsic and dynamical correlation effects is seen to strongly alter the picture of the photoionization process inferred from single-channel frozen-core Hartree-Fock calculations. Furthermore, the photoionization study of the first excited state of molecular carbon has revealed the presence of strong interchannel coupling between the 3sigma(g)-->ksigma(u) channel and the photoionization channels leading to the A (4)Pi(g) and f (2)Pi(g) ionic states in the near threshold region.     

Dual fluorescence of ellipticine: excited state proton transfer from solvent versus solvent mediated intramolecular proton transfer

Photophysical properties of a natural plant alkaloid, ellipticine (5,11-dimethyl-6H-pyrido[4,3-b]carbazole), which comprises both proton donating and accepting sites, have been studied in different solvents using steady state and time-resolved fluorescence techniques primarily to understand the origin of dual fluorescence that this molecule exhibits in some specific alcoholic solvents. Ground and excited state calculations based on density functional theory have also been carried out to help interpretation of the experimental data. It is shown that the long-wavelength emission of the molecule is dependent on the hydrogen bond donating ability of the solvent, and in methanol, this emission band arises solely from an excited state reaction. However, in ethylene glycol, both ground and excited state reactions contribute to the long wavelength emission. The time-resolved fluorescence data of the system in methanol and ethylene glycol indicates the presence of two different hydrogen bonded species of ellipticine of which only one participates in the excited state reaction. The rate constant of the excited state reaction in these solvents is estimated to be around 4.2-8.0 × 10(8) s(-1). It appears that the present results are better understood in terms of solvent-mediated excited state intramolecular proton transfer reaction from the pyrrole nitrogen to the pyridine nitrogen leading to the formation of the tautomeric form of the molecule rather than excited state proton transfer from the solvents leading to the formation of the protonated form of ellipticine.     

Stark-assisted population control of coherent CS(2) 4f and 5p Rydberg wave packets studied by femtosecond time-resolved photoelectron spectroscopy

A two-color (3+1(')) pump-probe scheme is employed to investigate Rydberg wave packet dynamics in carbon disulfide (CS(2) (*)). The state superpositions are created within the 4f and 5p Rydberg manifolds by three photons of the 400 nm pump pulse, and their temporal evolution is monitored with femtosecond time-resolved photoelectron spectroscopy using an 800 nm ionizing probe pulse. The coherent behavior of the non-stationary superpositions are observed through wavepacket revivals upon ionization to either the upper (12) or lower (32) spin-orbit components of CS(2) (+). The results show clearly that the composition of the wavepacket can be efficiently controlled by the power density of the excitation pulse over a range from 500 GWcm(2) to 10 TWcm(2). The results are consistent with the anticipated ac-Stark shift for 400 nm light and demonstrate an effective method for population control in molecular systems. Moreover, it is shown that Rydberg wavepackets can be formed in CS(2) with excitation power densities up to 10 TWcm(2) without significant fragmentation. The exponential 1e population decay (T(1)) of specific excited Rydberg states are recovered by analysis of the coherent part of the signal. The dissociation lifetimes of these states are typically 1.5 ps. However, a region exhibiting a more rapid decay ( approximately 800 fs) is observed for states residing in the energy range of 74 450-74 550 cm(-1), suggestive of an enhanced surface crossing in this region.     

Excited-state dynamics of carotenoids in light-harvesting complexes. 1. Exploring the relationship between the S1 and S* states

Dispersed transient absorption spectra collected at variable excitation intensities in combination with time-resolved signals were used to explore the underlying connectivity of the electronic excited-state manifold of the carotenoid rhodopin glucoside in the light-harvesting 2 complex isolated from Rhodopseudomonas acidophila. We find that the S state, which was recently identified as an excited state in carotenoids bound in bacterial light-harvesting complexes, exhibits a different response to the increase of excitation intensity than the S(1) state, which suggests that the models used so far to describe the excited states of carotenoids are incomplete. We propose two new models that can describe both the time-resolved and the intensity-dependent data; the first postulates that S(1) and S* are not populated in parallel after the decay of the initially excited S(2) state but instead result from the excitation of distinct ground-state subpopulations. The second model introduces a resonantly enhanced light-induced transition during excitation, which promotes population to higher-lying excited states that favors the formation of S* over S(1). Multiwavelength target analysis of the time-resolved and excitation-intensity dependence measurements were used to characterize the involved states and their responses. We show that both proposed models adequately fit the measured data, although it is not possible to determine which model is most apt. The physical origins and implications of both models are explored.