152
5 Optical Measurement Techniques
terms.
9 Features with a high oscillator strength typically exhibit a stronger optical
absorption due to a more pronounced light–matter interaction.
By measuring the reflectance R and the transmission T of a sample, the absorbance
A = 1 − T − R can be obtained. Spectrum acquisition can be either done with the
use of tunable monochromatic light at suitable controlled irradiances for serial frequency sampling or with the use of broadband white light. However, not all investigated samples have the desired transparency in the relevant spectral region, typically
due to their substrate material. This makes for instance transmission experiments in
the visible range on 2D material flakes isolated on SiO 2 /Si substrates impossible as
long as opaque silicon is attached on the reverse side of the sample. Similarly, gain
media deposited onto highly-reflective dielectric/Bragg mirrors cannot be probed in
transmission geometry.
The equipment used for measurements of this kind at low temperature and detection of signal is not different from that of steady-state PL spectroscopy (see for
instance [7]). Similarly, polarisation or time-dependent experiments can be performed. Yet, transmission experiments in a cryostat need optical access from opposite
sides. Furthermore, the employed broadband spectrum should not be blocked in the
desired spectral range by the used optical windows. Naturally, the detector’s responsivity over the whole spectral range and its dynamic range should be sufficiently
high.
Typically, 2D materials deposited onto substrates are characterised in the
reflection-contrast spectroscopy mode to determine excitonic resonances and to probe
their Rydberg-like energy series (see [64, 65], and for instance the monolayer and
bilayer study of [8] or the PL/absorption enhancement study of [6]). This requires a
comparison of the spectrum acquired on the flake to a position off the flake on the
same substrate in order to remove (i.e. cancel out) contributions from the supporting
substrate.
White-light spectroscopy at reasonably low incident powers generates such a low
amount of excited charge carriers that the spectra resemble a zero-density-regime
probe of a system’s resonance. This can be for instance useful to compare optical dispersion measurements obtained in PL and reflection contrast with respect
to density-dependent effects in the linear excitonic regime [66], or to characterise
cavity–polariton-mode anticrossing behaviour in basically-unexcited optical microcavities (e.g., shown in [27]). Note that high-enough spectral density at shorter wavelength can also lead to charge-carrier excitation during white-light exposure, particularly in combination with a strong focus achieved by microscope objectives.
At higher intensities (i.e. irradiances), saturable absorption leads to nonlinear
spectroscopy, where the active medium becomes transparent. This is needed to characterise for instance saturable-absorber mirrors used for mode-locked lasers by the
respective time-resolved method. On the other hand, an exciting light–matter cou9 The resonance(s) described by such (complex) dielectric function r (ω, k) in the ideal case feature a Lorentzian lineshape, which is representative for homogeneously-broadened resonances
(linewidth linked to lifetime by γ ≈ 1/τ 0 ). In contrast, inhomogeneously-broadened spectral lines
(e.g., statistically-broadened by size fluctuations) are well described by a Gaussian lineshape. A
convolution of both is given by the Voigt profile.
5 Optical Measurement Techniques
terms.
9 Features with a high oscillator strength typically exhibit a stronger optical
absorption due to a more pronounced light–matter interaction.
By measuring the reflectance R and the transmission T of a sample, the absorbance
A = 1 − T − R can be obtained. Spectrum acquisition can be either done with the
use of tunable monochromatic light at suitable controlled irradiances for serial frequency sampling or with the use of broadband white light. However, not all investigated samples have the desired transparency in the relevant spectral region, typically
due to their substrate material. This makes for instance transmission experiments in
the visible range on 2D material flakes isolated on SiO 2 /Si substrates impossible as
long as opaque silicon is attached on the reverse side of the sample. Similarly, gain
media deposited onto highly-reflective dielectric/Bragg mirrors cannot be probed in
transmission geometry.
The equipment used for measurements of this kind at low temperature and detection of signal is not different from that of steady-state PL spectroscopy (see for
instance [7]). Similarly, polarisation or time-dependent experiments can be performed. Yet, transmission experiments in a cryostat need optical access from opposite
sides. Furthermore, the employed broadband spectrum should not be blocked in the
desired spectral range by the used optical windows. Naturally, the detector’s responsivity over the whole spectral range and its dynamic range should be sufficiently
high.
Typically, 2D materials deposited onto substrates are characterised in the
reflection-contrast spectroscopy mode to determine excitonic resonances and to probe
their Rydberg-like energy series (see [64, 65], and for instance the monolayer and
bilayer study of [8] or the PL/absorption enhancement study of [6]). This requires a
comparison of the spectrum acquired on the flake to a position off the flake on the
same substrate in order to remove (i.e. cancel out) contributions from the supporting
substrate.
White-light spectroscopy at reasonably low incident powers generates such a low
amount of excited charge carriers that the spectra resemble a zero-density-regime
probe of a system’s resonance. This can be for instance useful to compare optical dispersion measurements obtained in PL and reflection contrast with respect
to density-dependent effects in the linear excitonic regime [66], or to characterise
cavity–polariton-mode anticrossing behaviour in basically-unexcited optical microcavities (e.g., shown in [27]). Note that high-enough spectral density at shorter wavelength can also lead to charge-carrier excitation during white-light exposure, particularly in combination with a strong focus achieved by microscope objectives.
At higher intensities (i.e. irradiances), saturable absorption leads to nonlinear
spectroscopy, where the active medium becomes transparent. This is needed to characterise for instance saturable-absorber mirrors used for mode-locked lasers by the
respective time-resolved method. On the other hand, an exciting light–matter cou9 The resonance(s) described by such (complex) dielectric function r (ω, k) in the ideal case feature a Lorentzian lineshape, which is representative for homogeneously-broadened resonances
(linewidth linked to lifetime by γ ≈ 1/τ 0 ). In contrast, inhomogeneously-broadened spectral lines
(e.g., statistically-broadened by size fluctuations) are well described by a Gaussian lineshape. A
convolution of both is given by the Voigt profile.