216
T. Buckup et al.
the energetic positions of the high frequency modes, the lack of changes in the vibrational frequencies and their lifetimes (not shown) [29, 30] when the different
excitation spectra were used, helps to assign high-frequency modes to ground-state
wave packet motion.
The experimentally observed dependence of the relative contributions of the
high-frequency modes on the detection wavelength for a specific excitation spectrum exactly matches theoretical expectations for spectrally-resolved DFWM [29].
The highest relative contributions are expected on both blue- as well as red-detuned
detection wavelengths while these contributions are at a minimal level near the centre of the excitation spectrum. The decreasing intensity of the high-frequency modes
for spectrum 1 of RPSB (Fig. 9.6(b)) and spectrum 1 and blue-detuned detection
wavelengths for BR (Fig. 9.6(b)) has, however, different origins [29]. The first origin is based on the fact that red-detuned excitation spectra, for which a large part
of the applied spectral intensity is off-resonant (e.g. spectrum 2 for RPSB and spectrum 2 for BR), favour the excitation of ground state vibrational coherences over
the induction of excited state population. The second effect is based on interference
between ground and excited state response pathways which is due to the homodyne
signal detection in DFWM: In spectral regions of intense ESA, the relative intensity
of ground state dynamics is controlled by an interference term between ground state
and excited state response pathways. The intensity of ground state dynamics is determined by the different ratios of transition dipole moments for ground state bleach
and ESA.
In contrast to the high-frequency dynamics, the observed out-of-plane modes
(only for BR, 800–1000 cm −1 ), which are observed only for excitation spectrum 1,
exhibit negligible Franck-Condon activity [38]. For out-of-plane modes from the
ground state, one would expect such dynamics to contribute to both excitation spectra for BR, which is, however, not observed. As the modes are observed with high
relative intensity only in spectral regions of ESA 1 (500–570 nm) this indicates
that the out-of-plane modes are detected through a probing pathway exploiting ESA
transitions. In total, the above mentioned points show that the out-of-plane modes
originate from excited state dynamics in agreement with earlier speculations on the
basis of pump-probe experiments [15].
In this context, it is interesting to note that out-of-plane modes are clearly observable for excited state dynamics of BR while similar vibrational dynamics seem to
be negligible for RPSB even for signal detection in spectral regions throughout the
intense ESA band centred at 500 nm (Fig. 9.6). The importance of this observation
concerns information about the excitation mechanism of these modes: For a direct
excitation of the modes by the laser pulse in the excited state, these modes need
to be normal modes of the all-trans retinal chromophore. However, retinal normal
modes cannot simply be absent in case of RPSB in comparison to the BR dynamics.
This, together with the clear dependence of the out-of-plane modes on the excitation
wavelength, indicates that the activation of such modes in BR play an important role
in the relaxation pathway of the excited state leading to double bond isomerization.
This point will be discussed in more detail below in the context of the low-frequency
modes.
T. Buckup et al.
the energetic positions of the high frequency modes, the lack of changes in the vibrational frequencies and their lifetimes (not shown) [29, 30] when the different
excitation spectra were used, helps to assign high-frequency modes to ground-state
wave packet motion.
The experimentally observed dependence of the relative contributions of the
high-frequency modes on the detection wavelength for a specific excitation spectrum exactly matches theoretical expectations for spectrally-resolved DFWM [29].
The highest relative contributions are expected on both blue- as well as red-detuned
detection wavelengths while these contributions are at a minimal level near the centre of the excitation spectrum. The decreasing intensity of the high-frequency modes
for spectrum 1 of RPSB (Fig. 9.6(b)) and spectrum 1 and blue-detuned detection
wavelengths for BR (Fig. 9.6(b)) has, however, different origins [29]. The first origin is based on the fact that red-detuned excitation spectra, for which a large part
of the applied spectral intensity is off-resonant (e.g. spectrum 2 for RPSB and spectrum 2 for BR), favour the excitation of ground state vibrational coherences over
the induction of excited state population. The second effect is based on interference
between ground and excited state response pathways which is due to the homodyne
signal detection in DFWM: In spectral regions of intense ESA, the relative intensity
of ground state dynamics is controlled by an interference term between ground state
and excited state response pathways. The intensity of ground state dynamics is determined by the different ratios of transition dipole moments for ground state bleach
and ESA.
In contrast to the high-frequency dynamics, the observed out-of-plane modes
(only for BR, 800–1000 cm −1 ), which are observed only for excitation spectrum 1,
exhibit negligible Franck-Condon activity [38]. For out-of-plane modes from the
ground state, one would expect such dynamics to contribute to both excitation spectra for BR, which is, however, not observed. As the modes are observed with high
relative intensity only in spectral regions of ESA 1 (500–570 nm) this indicates
that the out-of-plane modes are detected through a probing pathway exploiting ESA
transitions. In total, the above mentioned points show that the out-of-plane modes
originate from excited state dynamics in agreement with earlier speculations on the
basis of pump-probe experiments [15].
In this context, it is interesting to note that out-of-plane modes are clearly observable for excited state dynamics of BR while similar vibrational dynamics seem to
be negligible for RPSB even for signal detection in spectral regions throughout the
intense ESA band centred at 500 nm (Fig. 9.6). The importance of this observation
concerns information about the excitation mechanism of these modes: For a direct
excitation of the modes by the laser pulse in the excited state, these modes need
to be normal modes of the all-trans retinal chromophore. However, retinal normal
modes cannot simply be absent in case of RPSB in comparison to the BR dynamics.
This, together with the clear dependence of the out-of-plane modes on the excitation
wavelength, indicates that the activation of such modes in BR play an important role
in the relaxation pathway of the excited state leading to double bond isomerization.
This point will be discussed in more detail below in the context of the low-frequency
modes.