7 Label-Free Pump–Probe Nanoscopy
175
highly absorbing chromophores by detecting probe absorption changes induced by
nonlinear absorption processes (presented in Chap. 1). Absorption-based measurements are advantageous because they permit the investigation of non-emissive and
dark states, broadening the range of available targets. Moreover, compared to scattering techniques, they are less dependent on the particle size and they can be used
to study smaller structures [12, 17]. The implementation of these absorption-based
methods in a nonlinear microscopy platform with a proper detection scheme [3–6]
completes the temporal and spectral information of the ultrafast phenomena with the
spatial localization at the microscopic scale.
7.2.1 Pump–Probe Microscope Design
The typical pump–probe microscope is based on a nonlinear laser-scanning microscope where two ultrashort pulsed laser beams with a proper spatial and temporal
overlap are used in order to achieve highly sensitive nonlinear imaging and, moreover,
to follow ultrafast processes with sub-picosecond temporal resolution. An example
of this setup is shown in Fig. 7.2a.
The two pump and probe beams may be generated by two different pulsed laser
sources synchronized with each other in order to have precise control over the timing
of the two pulse trains. More typically, the two beams come from the same laser
source, usually from a mode-locked Ti:sapphire, tuneable in the red and NIR part
of the spectrum, with a pulse repetition frequency around 70–90 MHz and a pulse
width between 50 and 10 ps. The single laser output can be split into two beams, and
one of them frequency doubled in a barium borate (BBO) nonlinear crystal in order
to obtain a different wavelength in the visible range, at exactly half the initial value.
Another approach consists of using the Ti:sapphire output to synchronously pump an
optical parametric oscillator (OPO), which is a coherent light source based on optical
gain from parametric amplification in a nonlinear crystal. OPOs are more versatile
because their synchronized outputs can be tuned over a wide wavelength range,
reaching wavelengths that cannot be provided by conventional lasers. Compared to
the direct frequency doubling which always provides two wavelengths, one half of
the other, the OPO allows for different pump–probe wavelength combinations, with
the possibility of working only in the NIR region. The intensity of the beams may
be controlled through variable neutral density filters or through a combination of
half-wave plates and linear polarizers.
The two beams are spatially combined with a dichroic mirror (DM), sent to a
scanning unit (SU) made by a pair of galvanometric mirrors, and collinearly focused
onto the sample (S) by a high numerical aperture objective (O). Three-dimensional
imaging can be achieved through an axial piezo-stage with nanometer resolution.
Owing to chromatic aberrations, pump and probe foci at the sample plane may differ
(see Fig. 7.2b). Adding two relay lenses with regulating distance in one of the two
optical paths may help in adjusting the input divergence of one beam in the objective
to match the other focus. Alternatively, the focusing of the probe may be optimized,
175
highly absorbing chromophores by detecting probe absorption changes induced by
nonlinear absorption processes (presented in Chap. 1). Absorption-based measurements are advantageous because they permit the investigation of non-emissive and
dark states, broadening the range of available targets. Moreover, compared to scattering techniques, they are less dependent on the particle size and they can be used
to study smaller structures [12, 17]. The implementation of these absorption-based
methods in a nonlinear microscopy platform with a proper detection scheme [3–6]
completes the temporal and spectral information of the ultrafast phenomena with the
spatial localization at the microscopic scale.
7.2.1 Pump–Probe Microscope Design
The typical pump–probe microscope is based on a nonlinear laser-scanning microscope where two ultrashort pulsed laser beams with a proper spatial and temporal
overlap are used in order to achieve highly sensitive nonlinear imaging and, moreover,
to follow ultrafast processes with sub-picosecond temporal resolution. An example
of this setup is shown in Fig. 7.2a.
The two pump and probe beams may be generated by two different pulsed laser
sources synchronized with each other in order to have precise control over the timing
of the two pulse trains. More typically, the two beams come from the same laser
source, usually from a mode-locked Ti:sapphire, tuneable in the red and NIR part
of the spectrum, with a pulse repetition frequency around 70–90 MHz and a pulse
width between 50 and 10 ps. The single laser output can be split into two beams, and
one of them frequency doubled in a barium borate (BBO) nonlinear crystal in order
to obtain a different wavelength in the visible range, at exactly half the initial value.
Another approach consists of using the Ti:sapphire output to synchronously pump an
optical parametric oscillator (OPO), which is a coherent light source based on optical
gain from parametric amplification in a nonlinear crystal. OPOs are more versatile
because their synchronized outputs can be tuned over a wide wavelength range,
reaching wavelengths that cannot be provided by conventional lasers. Compared to
the direct frequency doubling which always provides two wavelengths, one half of
the other, the OPO allows for different pump–probe wavelength combinations, with
the possibility of working only in the NIR region. The intensity of the beams may
be controlled through variable neutral density filters or through a combination of
half-wave plates and linear polarizers.
The two beams are spatially combined with a dichroic mirror (DM), sent to a
scanning unit (SU) made by a pair of galvanometric mirrors, and collinearly focused
onto the sample (S) by a high numerical aperture objective (O). Three-dimensional
imaging can be achieved through an axial piezo-stage with nanometer resolution.
Owing to chromatic aberrations, pump and probe foci at the sample plane may differ
(see Fig. 7.2b). Adding two relay lenses with regulating distance in one of the two
optical paths may help in adjusting the input divergence of one beam in the objective
to match the other focus. Alternatively, the focusing of the probe may be optimized,