52
3 Rotatable-Polarisation Terahertz Time-Domain Spectroscopy of Anisotropic Media
be seen in Fig. 3.5a–e, at the corresponding angles to those of the resistance maps
in Fig. 3.3a–e. The ellipticity and orientation angle of the pulses was then extracted
using the method described in Sect. 2.4.2.
The ellipticity of the pulses at 1.0 THz as a function of ψ em is reported in Fig. 3.5f.
The average χ of THz pulses from the emitter was 0.925
◦ , which varied by 0.75
◦ over
a 180
◦ rotation, demonstrating that the polarisation remained close to linear for all
ψ em . The small measured ellipticity may be an artifact of a slight misalignment of the
gate and THz beams while propagating through the detection crystal [27], or it may
arise from some slight misligment of optics in the THz beam path. However, this small
ellipticity, reported here for the first time for an interdigitated THz emitter, is much
smaller than the ellipticity χ ∼ 10
◦ of wide-area emitters [13] and dipole antennas
[28]. While a small quadrupole moment may contribute to the finite ellipticity in
those two cases [28], the small ellipticity in this case may result from the gold
fingers, which in essence act as a wire-grid polarizer within the near-field of the
generated THz radiation.
The precise orientation of ψ = 0 relative to a sample under investigation will be an
important factor in applications of this rotatable polarisation technique, for example
in optical component design and material characterisation. In order to determine
precisely which ψ em corresponded to purely horizontally polarised pulses, where
ψ = 0, the orientation angle of the incident pulse at 1.0 THz was measured as a
function of ψ em , shown in Fig. 3.5g. A linear fit to the data gives an emitter angle
of 83.85
◦ for ψ = 0. Thus the incident orientation angle ψ in of the THz pulses that
will be incident on the sample in the RP-THz-TDS system was calibrated precisely
for each ψ em . Zero ψ in corresponds to a horizontally polarized incident pulse and
ψ in = ±90
◦ corresponds to vertically polarized incident pulses of opposite polarity.
3.3 Comparison of Rotatable Polarisation to Projection via
Wire-Grid Polarisers
To illustrate the advantages of the RP-THz-TDS method over other methods of
anisotropic spectroscopy, a relevant comparison is to that of polarisation projection
via WGPs. The polarization state of THz pulses projected to a certain angle using a
WGP was compared to that achieved when the polarisation state is directly rotated
using the rotatable emitter method described in this chapter. To explore this, a WGP
was fabricated by Michele Failla, using a similar UV photolithography process used
to create the THz emitter described in Sect. 3.2.1, which consisted of 300 nm thick
gold wires with 8 μm width and 20 μm period deposited onto a SI-GaAs substrate.
A schematic diagram of the experiment is shown in Fig. 3.6a, b. The WGP was
held in a manual rotation mount at the sample position of the spectrometer, and was
rotated to an arbitrary angle. This angle at which the THz polarisation is perpendicular to the wires was precisely identified as ψ in = −35.0
◦ , by scanning ψ in over a
range of 20
◦ in 0.625
◦ steps around the approximate direction perpendicular to the
3 Rotatable-Polarisation Terahertz Time-Domain Spectroscopy of Anisotropic Media
be seen in Fig. 3.5a–e, at the corresponding angles to those of the resistance maps
in Fig. 3.3a–e. The ellipticity and orientation angle of the pulses was then extracted
using the method described in Sect. 2.4.2.
The ellipticity of the pulses at 1.0 THz as a function of ψ em is reported in Fig. 3.5f.
The average χ of THz pulses from the emitter was 0.925
◦ , which varied by 0.75
◦ over
a 180
◦ rotation, demonstrating that the polarisation remained close to linear for all
ψ em . The small measured ellipticity may be an artifact of a slight misalignment of the
gate and THz beams while propagating through the detection crystal [27], or it may
arise from some slight misligment of optics in the THz beam path. However, this small
ellipticity, reported here for the first time for an interdigitated THz emitter, is much
smaller than the ellipticity χ ∼ 10
◦ of wide-area emitters [13] and dipole antennas
[28]. While a small quadrupole moment may contribute to the finite ellipticity in
those two cases [28], the small ellipticity in this case may result from the gold
fingers, which in essence act as a wire-grid polarizer within the near-field of the
generated THz radiation.
The precise orientation of ψ = 0 relative to a sample under investigation will be an
important factor in applications of this rotatable polarisation technique, for example
in optical component design and material characterisation. In order to determine
precisely which ψ em corresponded to purely horizontally polarised pulses, where
ψ = 0, the orientation angle of the incident pulse at 1.0 THz was measured as a
function of ψ em , shown in Fig. 3.5g. A linear fit to the data gives an emitter angle
of 83.85
◦ for ψ = 0. Thus the incident orientation angle ψ in of the THz pulses that
will be incident on the sample in the RP-THz-TDS system was calibrated precisely
for each ψ em . Zero ψ in corresponds to a horizontally polarized incident pulse and
ψ in = ±90
◦ corresponds to vertically polarized incident pulses of opposite polarity.
3.3 Comparison of Rotatable Polarisation to Projection via
Wire-Grid Polarisers
To illustrate the advantages of the RP-THz-TDS method over other methods of
anisotropic spectroscopy, a relevant comparison is to that of polarisation projection
via WGPs. The polarization state of THz pulses projected to a certain angle using a
WGP was compared to that achieved when the polarisation state is directly rotated
using the rotatable emitter method described in this chapter. To explore this, a WGP
was fabricated by Michele Failla, using a similar UV photolithography process used
to create the THz emitter described in Sect. 3.2.1, which consisted of 300 nm thick
gold wires with 8 μm width and 20 μm period deposited onto a SI-GaAs substrate.
A schematic diagram of the experiment is shown in Fig. 3.6a, b. The WGP was
held in a manual rotation mount at the sample position of the spectrometer, and was
rotated to an arbitrary angle. This angle at which the THz polarisation is perpendicular to the wires was precisely identified as ψ in = −35.0
◦ , by scanning ψ in over a
range of 20
◦ in 0.625
◦ steps around the approximate direction perpendicular to the