Topics in Current Chemistry (2018) 376:28
1 3
Another central aspect in the state-of-the-art methods of time-resolved multidimensional spectroscopy is the scanning of the time delay between laser pulses.
Mechanically moving stages takes time and thus 2D spectroscopy methods involving three or more beams take intrinsically longer for acquiring the complete dataset
than techniques with a single delay to scan. It is important to note in this context that
in order to allow for averaging of signals, interferometric phase stability between
the beams is generally required in 2D spectroscopy methods, leading to challenging
experimental constraints when delays are actively scanned. However, interferometric phase stability is often required mainly between given pulse pairs, e.g., between
the first two pulses, or between the signal and the local oscillator. Phase stability to
achieve experimentally becomes the more challenging the shorter the wavelength of
the applied laser light is. Hence, the technically most demanding experiments are
spectrally located in the UV region, but advances in technical instrumentation have
made it possible to obtain phase stability as good as ca λ/187 over half an hour at ca.
310 nm [64]. Such drastic experimental requirements have led to the development of
several novel experimental approaches regarding the achievement of phase stability.
The desired phase stability can generally be achieved in several ways. For instance,
the use of diffractive optics [61, 65–67] helps to generate phase–coherent pairs of
laser beams at a given geometry. In general, the beams generated that way are then
delayed with optical wedges [68, 69] rather than delay stages to delay the associated
pulses, which allows an advantageous way to maintain phase stability. Alternatively,
one can use for example acousto-optic modulators (AOM), liquid crystal masks, and
“liquid–crystals on silicon” (LCOS) shapers to generate phase–coherent pulse pairs
[70–75]. Both variants allow for the generation of simple beam geometries and pulse
delaying with intrinsic phase stability and therefore reducing time for experiments.
In addition to these aspects, optical phase modulators allow for a facile and precise
generation of desired phases for independent pulses, thereby making it possible to
“switch” phases on demand, which is used for example to suppress or even eliminate
signal contributions upon averaging (phase cycling) [76].
The experimental time required to scan several delays can be further optimized
by exploiting on-the-fly scanning techniques, often called rapid-scan or fast scan
[77, 78]. Instead of scanning the delay “step-by-step” between two pulses, and for
each delay step averaging a certain number of laser pulses, rapid scan techniques
measure the signal only one time at a single time delay (i.e., a single pulse), after
which it moves to the next delay [79]. This method of data acquisition offers several
advantages. The first one being that a sequence of delays can be potentially measured much faster than by halting at each delay to acquire the signal. It requires, however, the ability to measure the delay instantaneously. This is offered off-the-shelf in
several commercial delay stages but has also been implemented by exploiting interferometric detection of continuous wave reference beams [80], or by using spatial
light modulators for generating pulse delay. Significantly reducing the time that it
takes to acquire a certain measurement, the main advantage of rapid-scan techniques
is the ability to filter out the low-frequency noise typical of ultrashort laser sources.
In a similar context of reducing measurement time, sophisticated sampling techniques have been developed (Fig. 5). These methods make use of the possibility to
drastically reduce for example the number of data points required to construct an
10
Reprinted from the journal
1 3
Another central aspect in the state-of-the-art methods of time-resolved multidimensional spectroscopy is the scanning of the time delay between laser pulses.
Mechanically moving stages takes time and thus 2D spectroscopy methods involving three or more beams take intrinsically longer for acquiring the complete dataset
than techniques with a single delay to scan. It is important to note in this context that
in order to allow for averaging of signals, interferometric phase stability between
the beams is generally required in 2D spectroscopy methods, leading to challenging
experimental constraints when delays are actively scanned. However, interferometric phase stability is often required mainly between given pulse pairs, e.g., between
the first two pulses, or between the signal and the local oscillator. Phase stability to
achieve experimentally becomes the more challenging the shorter the wavelength of
the applied laser light is. Hence, the technically most demanding experiments are
spectrally located in the UV region, but advances in technical instrumentation have
made it possible to obtain phase stability as good as ca λ/187 over half an hour at ca.
310 nm [64]. Such drastic experimental requirements have led to the development of
several novel experimental approaches regarding the achievement of phase stability.
The desired phase stability can generally be achieved in several ways. For instance,
the use of diffractive optics [61, 65–67] helps to generate phase–coherent pairs of
laser beams at a given geometry. In general, the beams generated that way are then
delayed with optical wedges [68, 69] rather than delay stages to delay the associated
pulses, which allows an advantageous way to maintain phase stability. Alternatively,
one can use for example acousto-optic modulators (AOM), liquid crystal masks, and
“liquid–crystals on silicon” (LCOS) shapers to generate phase–coherent pulse pairs
[70–75]. Both variants allow for the generation of simple beam geometries and pulse
delaying with intrinsic phase stability and therefore reducing time for experiments.
In addition to these aspects, optical phase modulators allow for a facile and precise
generation of desired phases for independent pulses, thereby making it possible to
“switch” phases on demand, which is used for example to suppress or even eliminate
signal contributions upon averaging (phase cycling) [76].
The experimental time required to scan several delays can be further optimized
by exploiting on-the-fly scanning techniques, often called rapid-scan or fast scan
[77, 78]. Instead of scanning the delay “step-by-step” between two pulses, and for
each delay step averaging a certain number of laser pulses, rapid scan techniques
measure the signal only one time at a single time delay (i.e., a single pulse), after
which it moves to the next delay [79]. This method of data acquisition offers several
advantages. The first one being that a sequence of delays can be potentially measured much faster than by halting at each delay to acquire the signal. It requires, however, the ability to measure the delay instantaneously. This is offered off-the-shelf in
several commercial delay stages but has also been implemented by exploiting interferometric detection of continuous wave reference beams [80], or by using spatial
light modulators for generating pulse delay. Significantly reducing the time that it
takes to acquire a certain measurement, the main advantage of rapid-scan techniques
is the ability to filter out the low-frequency noise typical of ultrashort laser sources.
In a similar context of reducing measurement time, sophisticated sampling techniques have been developed (Fig. 5). These methods make use of the possibility to
drastically reduce for example the number of data points required to construct an
10
Reprinted from the journal