signal. Contrary to 2D IR spectroscopy, no stimulated emission signals occur in the
response due to the large frequency difference in the excitation and probing pulses
for 2D VE spectroscopy. Note that only exemplary non-rephasing signals are shown
in and complementary rephasing signals contribute as well. A more extensive
theoretical treatment of the signals involving a full set of response pathways has
been given elsewhere [273]. The signal strengths depend on transition dipole
moments of both IR, as well as electronic transitions. It is, therefore, important to
note that vibrational modes dominate the signal that are coupled to the electronic
transitions. However, also non-resonant electronic interactions can contribute to due
to the modulation of the polarizability of the sample by IR pre-excitation. That has
important consequences since it implies signals from the sample as well as from
solvent molecules. The latter response can be separately analyzed or even
subtracted. What ultimately counts in the 2D VE signal is the frequency difference
between vibrational modes in ground and excited electronic states (Fig. 25b and c).
Because the frequency in the excited electronic state can be higher or lower than in
the ground state, the magnitude and the sign of the frequency shift both affect the
position of the excited state absorption signal with respect to the ground state bleach
signal. That is an important difference to 2D IR spectroscopy, where anharmonic
shifts in electronic potentials generally shift the excited state absorption bands to
lower frequencies, although exceptions exist in special cases [274].
To highlight the applicability of 2D VE spectroscopy, Fig. 26c shows exemplary
2D VE spectra at two indicated population delays of a model compound (Fig. 26d)
dissolved in bulk formamide, i.e. [(CN) 5 Fe II CNRu
III (NH 3 ) 5 ]
- (FeRu, upper two
panels), along with complementary 2D VE measurements performed on the neat
solvent (lower two panels) [272]. The sample exhibits four vibrational modes in the
considered spectral range, which are attributable to CN stretching modes (colored
arrows in Fig. 26d and top panels for FT IR spectra in c). The 2D VE signals of the
FeRu sample exhibit a comparatively narrow width in the x 1 domain but
considerably larger widths in the x 3 domain due to the involvement of the
spectrally broad electronic transition, which is a metal-to-metal charge transfer
transition in this case. The signals are dominated by excited state absorption
contributions (blue) that are persistent at both delays and change shape as well as
intensity with increasing population time. Interestingly, not all modes contribute to
the 2D VE signals with an intensity that would reflect their IR absorption spectrum.
That observation reflects the different couplings of the modes to the electronic
transition. The lineshapes in 2D VE spectra appear strongly different from the ones
that are obtained from 2D IR signals. This is a result from the sensitivity of the 2D
VE signals on the fluctuations of the coupled vibrational and electronic frequencies.
A detailed analysis similar to the CLS method for 2D IR spectra [100] revealed that
the CLS exhibit (i) nonlinear components and (ii) positive or negative CLS values
for different transitions. Such behavior has been assigned to the different interaction
of the modes with their environment, i.e. bridge modes modulates the Fe-Ru
separation, whereas the trans modes couple to the solvent [273]. Overall, the loss of
correlation was determined as very fast (\ 1 ps), which reflects similar timescales
as determined for diagonal peaks from 2D IR measurements [273].
Top Curr Chem (Z) (2017) 375:86
123
174
Reprinted from the journal
response due to the large frequency difference in the excitation and probing pulses
for 2D VE spectroscopy. Note that only exemplary non-rephasing signals are shown
in and complementary rephasing signals contribute as well. A more extensive
theoretical treatment of the signals involving a full set of response pathways has
been given elsewhere [273]. The signal strengths depend on transition dipole
moments of both IR, as well as electronic transitions. It is, therefore, important to
note that vibrational modes dominate the signal that are coupled to the electronic
transitions. However, also non-resonant electronic interactions can contribute to due
to the modulation of the polarizability of the sample by IR pre-excitation. That has
important consequences since it implies signals from the sample as well as from
solvent molecules. The latter response can be separately analyzed or even
subtracted. What ultimately counts in the 2D VE signal is the frequency difference
between vibrational modes in ground and excited electronic states (Fig. 25b and c).
Because the frequency in the excited electronic state can be higher or lower than in
the ground state, the magnitude and the sign of the frequency shift both affect the
position of the excited state absorption signal with respect to the ground state bleach
signal. That is an important difference to 2D IR spectroscopy, where anharmonic
shifts in electronic potentials generally shift the excited state absorption bands to
lower frequencies, although exceptions exist in special cases [274].
To highlight the applicability of 2D VE spectroscopy, Fig. 26c shows exemplary
2D VE spectra at two indicated population delays of a model compound (Fig. 26d)
dissolved in bulk formamide, i.e. [(CN) 5 Fe II CNRu
III (NH 3 ) 5 ]
- (FeRu, upper two
panels), along with complementary 2D VE measurements performed on the neat
solvent (lower two panels) [272]. The sample exhibits four vibrational modes in the
considered spectral range, which are attributable to CN stretching modes (colored
arrows in Fig. 26d and top panels for FT IR spectra in c). The 2D VE signals of the
FeRu sample exhibit a comparatively narrow width in the x 1 domain but
considerably larger widths in the x 3 domain due to the involvement of the
spectrally broad electronic transition, which is a metal-to-metal charge transfer
transition in this case. The signals are dominated by excited state absorption
contributions (blue) that are persistent at both delays and change shape as well as
intensity with increasing population time. Interestingly, not all modes contribute to
the 2D VE signals with an intensity that would reflect their IR absorption spectrum.
That observation reflects the different couplings of the modes to the electronic
transition. The lineshapes in 2D VE spectra appear strongly different from the ones
that are obtained from 2D IR signals. This is a result from the sensitivity of the 2D
VE signals on the fluctuations of the coupled vibrational and electronic frequencies.
A detailed analysis similar to the CLS method for 2D IR spectra [100] revealed that
the CLS exhibit (i) nonlinear components and (ii) positive or negative CLS values
for different transitions. Such behavior has been assigned to the different interaction
of the modes with their environment, i.e. bridge modes modulates the Fe-Ru
separation, whereas the trans modes couple to the solvent [273]. Overall, the loss of
correlation was determined as very fast (\ 1 ps), which reflects similar timescales
as determined for diagonal peaks from 2D IR measurements [273].
Top Curr Chem (Z) (2017) 375:86
123
174
Reprinted from the journal