High-power THz wave generation in plasma induced by polarization

adjusted two-color laser pulses

Yasuo Minami, Takayuki Kurihara, Keita Yamaguchi, Makoto Nakajima, and Tohru Suemoto

Citation: Appl. Phys. Lett. 102, 041105 (2013); doi: 10.1063/1.4789773

View online: http://dx.doi.org/10.1063/1.4789773
View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v102/i4
Published by the American Institute of Physics.

Additional information on Appl. Phys. Lett.

Journal Homepage: http://apl.aip.org/
Journal Information: http://apl.aip.org/about/about_the_journal
Top downloads: http://apl.aip.org/features/most_downloaded
Information for Authors: http://apl.aip.org/authors

Downloaded 11 Apr 2013 to 150.108.161.71. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
 APPLIED PHYSICS LETTERS 102, 041105 (2013)

High-power THz wave generation in plasma induced by polarization

adjusted two-color laser pulses
Yasuo Minami,1,a) Takayuki Kurihara,2 Keita Yamaguchi,2 Makoto Nakajima,3
and Tohru Suemoto2
1
Department of Physics, Faculty of Engineering, Yokohama National University, Yokohama 240-8501, Japan
2
Institute for Solid State Physics, University of Tokyo, Kashiwa 277-8581, Japan
3
Graduate School of Science, Chiba University, Chiba 263-8522, Japan
(Received 10 December 2012; accepted 15 January 2013; published online 28 January 2013)
We introduce a simple and efficient method of enhancing the terahertz field in an air plasma
produced by two-color laser pulses, by inserting a specially designed dual-wavelength wave plate
between the non-linear optical crystal and the plasma. Adjusting the polarization of the two laser
pulses yielded an electric field of 1.4 MV/cm, which was 1.7 times as intense as that obtained from
the unmodified system. Additionally, taking a dispersion of the group velocities of the two-color
laser pulses into account, we discussed the validity of the enhancement factor. V C 2013 American

Institute of Physics. [http://dx.doi.org/10.1063/1.4789773]

Terahertz waves are electromagnetic waves with frequen- the BBO crystal depends on the power of the incident pulse,
cies in the range of 0.1–100 THz, which defines the so-called and the THz wave polarization in turn depends on the SH
“THz gap.” A general lack of technology for the generation power. As a result, the THz wave polarization is a sensitive
and detection of radiation in this region of the electromagnetic function of the incident power.20–22 Therefore, the method
spectrum has hindered research into the applications of these requires delicate readjustment of the azimuthal alignment of
waves over the past decades. To fill the “THz gap,” many an electro-optic (EO) detecting crystal, and a probe pulse
THz generation methods have recently been developed, based polarization measurement is required every time the intensity
on approaches involving photoconductive antennas,1 optical of light changes. To avoid these inconveniences, schemes
rectification in nonlinear crystals,2 Cherenkov radiation from that use dual-wavelength wave plates (DWP) and phase-
LiNbO3 ,3 plasma currents,4,5 and free electron lasers,6 among compensating wedge plates have been proposed, and shown to
other.7–9 Especially, a high photon conversion efficiency be advantageous.23,24 However, proof-of-principle experi-
(40%) is attained.8 Recently, it was found that it is possible to ments along these lines have not been applied to the actual
generate an intense THz field by means of a Cherenkov enhancement of the generation of THz waves. In this study, we
phase-matching technique in a nonlinear crystal, such as report on the generating THz waves with electric field
LiNbO3 ,10–12 or from a plasma produced in air by two-color strengths of 1.4 MV/cm from an air plasma produced by two-
laser pulses.5,13,14 These new intense THz-wave sources are color laser pulses by adjusting the laser pulse polarizations
driving accelerated research in diverse areas of nonlinear using a specially designed DWP. Additionally, taking a disper-
physics, including anharmonic intermolecular oscillations,15 sion of the group velocities of the two-color laser pulses into
field ionization of excitons,16 avalanche effects in photogener- account, we discuss the validity of the enhancement factor. We
ated carriers,17 and more.18,19 THz wave generation using air also estimate that if the dispersion between the two-color laser
plasma produced by two-color laser pulses has some advan- pulses is eliminated, the THz field attain to 2 MV/cm.
tages over the Cherenkov phase-matching approach in gener- The THz generation and detection components used in
ating high-frequency THz wave, which enables us to explore this work are shown in Fig. 1. Figures 1(a) and 1(b) depict
sciences in THz region. the THz generating system without and with a DWP, respec-
In the air plasma approach, type I b-barium borate (BBO) tively. A Ti:sapphire amplifier system with a repetition rate
is placed between a convex lens and the plasma in order to of 1 kHz provided 20 fs laser pulses with a wavelength of
generate a second harmonic (SH) wave. The THz wave is gen- 800 nm (without a carrier-envelope phase control). Pulse
erated from an asymmetric ionization current in the plasma, energy was 4.5 mJ, and each pulse was focused through a
caused by an asymmetric electric field that is the result of 100 lm-thick type I b-BBO crystal by means of a 220 mm
superimposed two-color light waves. Usually, the ordinary focal length lens, which also included a DWP in the case of
axis of a BBO crystal is rotated to an angle mismatched for op- the wave plate experiment. The generated THz wave was
timum SH generation (SHG) in order to generate a SH electric forwarded to a 400 lm-thick GaP crystal using off-axis para-
field component (2x) parallel to that of the fundamental wave bolic mirrors with the diameters of 76 mm and focal lengths
(x). However, this rotation of the BBO crystal degrades SHG of 150 mm (for receiver) and 51 mm (for condenser), and
efficiency, resulting in a reduction in THz wave intensity. The observed using a balanced EO sampling scheme (Fig. 1(c)).
method also has a drawback of introducing instability into the Therefore, the focal depth for the THz wave is estimated to
polarization of the THz field, since the efficiency of SHG in 3.0 mm. The THz wave generated from that region is
observed in our experiment. The focal length used in the
probe pulse was 250 mm. To avoid saturation of the EO
a)
Email: [email protected]. detection, we attenuated the THz wave via Si plates. We

0003-6951/2013/102(4)/041105/4/$30.00 102, 041105-1 C 2013 American Institute of Physics

V

Downloaded 11 Apr 2013 to 150.108.161.71. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
 041105-2 Minami et al. Appl. Phys. Lett. 102, 041105 (2013)

FIG. 2. Time evolution of the THz wave field (a) without DWP, and
(b) with DWP. Figure insets: FFT power spectra. With a DWP included, a
maximum electric field of 1.4 MV/cm is obtained.

giving an attenuation factor of 0.24. To estimate the intensity of

the THz field propagating in air, we considered the Fresnel
transmission coefficient of the GaP crystal surface tGaP , given
FIG. 1. The THz generation schemes (a) without a DWP, and (b) with a by 2=ðn1 þ n3 Þ. n3 ¼ 3:18 is the refractive index of GaP, and
DWP. A Ti:sapphire amplifier system with a repetition rate of 1 kHz provides
800 nm laser pulses with pulsewidths of 20 fs. 4.5 mJ pulses are focused via a
we thus calculated a coefficient of 0.48. The time evolution of
220 mm focal length lens through a 100 lm-thick type I b-BBO crystal the calibrated THz field was recorded and is shown in
(as well as a DWP in (b)). (c) The THz-field detection system. Fig. 2(a).29 The maximum electric field was found to be
0.82 MV/cm. For this case, the angle between the ordinary axis
employed another Si plate for separating the THz wave from of the BBO crystal and the polarization of the fundamental light
any incident light, as shown in Fig. 1(c). of from the BBO was set to 50 , while the tilt angle of the BBO
The DWP used in this work was specially designed for crystal was set to 1 .20 The inset of Fig. 2(a) shows the FFT
the experiment. The DWP consists of a true zero-order power spectrum of the electric field. This type of THz wave
single-plate with a thickness of 43 lm, which minimizes the source has applications in spectroscopy over the broad fre-
temporal deviation between two color laser pulses. To sup- quency range of 0.5–7 THz, which is limited by the intrinsic
press reflections, we applied an anti-reflection (AR) coating response of GaP crystal.30,31 The upper frequency ceiling is the
for 400 nm and 800 nm on the surface of the DWP. The plate result of a dip structure at 8 THz. This dip can be seen in the
performs as a half-wave plate for laser pulses with frequency GaP crystal detection system, and arises from a phase mismatch
x, and as a full-wave plate for pulses with double the fre- between the THz wave and the probe pulse in the GaP crys-
quency (2x). tal.30,31 The mismatch manifests in the time evolution signal as
We also observed the THz wave electric field strength a ripple after the main peak (see Fig. 2(a)), which makes the
as a function of the BBO-to-focus distance, in an effort to peak asymmetric.
investigate the details of DWP enhancement. In a second round of experiments, we generated THz
In the first experiment, we generated THz waves with no waves with the addition of the DWP in the system, as shown
DWP in place, as shown in Fig. 1(a). The azimuthal angle and in Fig. 1(b). The DWP was placed between the BBO crystal
tilt angle (angle between the plane normal and the propagation and the plasma so as to rotate the polarization of the 800 nm
direction of the incident light) of the BBO crystal were adjusted laser pulse to match that of the 400 nm pulse. As a conse-
to maximize THz wave generation.22 The distance (d) between quence, the plasma is produced by polarization-parallelized
the BBO crystal and plasma was 50 mm.20,25–27 The electric two-color pulses at the focus, which generates THz waves
field of the THz waves was attenuated via Si plates, where the more effectively. In this approach, the azimuthal angle of the
attenuation factor tNSi may be expressed as ½4n1 n2 =ðn1 þ n2 Þ2 N , BBO crystal is fixed to maximize SHG efficiency (so the
with n1 and n2 denoting the respective refractive indices of air angle between the ordinary axis of the BBO crystal and the
and Si, respectively, and N being the number of Si attenuator polarization of the fundamental wave is 0 ). The distance d
plates. In our experiment, n1 ¼ 1:00; n2 ¼ 3:42,28 and N ¼ 4, between the BBO crystal and the focus is readjusted to be

Downloaded 11 Apr 2013 to 150.108.161.71. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
 041105-3 Minami et al. Appl. Phys. Lett. 102, 041105 (2013)

d ¼ 50 mm. The time evolution of the THz field in these DWP as follows: based on the thickness of the DWP and the differ-
experiments is shown in Fig. 2(b), where the maximum elec- ence in group velocity of the two laser pulses propagating
tric field can be seen to be 1.4 MV/cm. The inset in the figure within the DWP, the difference in DWP propagation times is
shows the associated power spectrum. estimated to be 7.9 fs. That is to say, the 800 nm laser pulse
To evaluate the enhancement factor arising from the use is 7.9 fs ahead of the 400 nm pulse post-DWP. This propaga-
of the DWP, we investigated the dependence of the THz tion time difference is non-negligible relative to the 20 fs
wave electric field intensity on the BBO-to-focus distance, temporal width of each laser pulse. The time deviation of the
as measured by EO detection. The results are shown in two pulses as they propagate through the air between the
Fig. 3, where the electric fields obtained with and without BBO crystal and the focus is an additional 5.6 fs. Therefore,
the DWP are compared. Local maxima (E > 0) and minima the total difference in arrival time for the two pulses leaving
(E < 0) appear at intervals of 25 mm, in good agreement the DWP-air path is 13.5 fs. We assume a Gaussian pulse en-
with values obtained in other experiments.20,25–27 The peri- velope, exp½t2 ln 2=ð2T02 Þ, for the electric fields of the x
odic modulation is due to phase velocity dispersion of the and 2x pulses, where 2T0 is the pulse width of a pulse
two-color laser pulses in the air between the BBO crystal (FWHM). For convenience, a pulse width of 2T0 ð¼ 20 fsÞ is
and the plasma. The observed behavior for the scenario with- assumed for both pulses. The generation mechanism for THz
out a DWP can be explained as follows: Wen et al. reported waves in the plasma-based method is best explained by a
that the polarization of a THz field rotates as a result of plasma ionizing model, rather than any third order nonlinear
changes in the BBO-to-focus distance.20 They found that effect. However, it is known that the maximum THz field
output extrema were separated by a distance of 25 mm, and ETHz that can be generated is approximately given by
that the relative phase difference between the vertically
polarized THz field and the horizontally polarized THz field ETHz ¼ AE2x E2x ; (1)
was p=2 radians. This is caused by the relative phase differ-
ence between x and 2x waves with non-parallel polariza- where A is the constant (and not equal to vð3Þ ).32 From
tion, because of the difference in the phase velocities of the Eq. (1), and defining the time difference between the arrival
two laser pulses. We observed that the vertical polarization of the two laser pulses as dt, the generated THz field may be
component with a phase shift, which was partly in agreement expressed as ETHz ðdtÞ ¼ AE2x E2x expð3dt2 ln 2=8T02 Þ:
with the results of Wen et al. The approximated envelopes For the setup without a DWP, an estimation of the effi-
for the electric field modulation are shown in Fig. 3 as ciency of THz wave generation must take into account the
dashed lines. From the figure, we can see that the THz field angular displacement (h) between the polarization of the wave
is enhanced by the addition of the DWP by a factor of 1:4 with frequency x and the azimuthal angle of the BBO crystal
at d ¼ 75 mm. as well as the angular displacement (90  h) between the
The measured enhancement of wave generation by the polarizations of the x and 2x waves. The efficiency is given
addition of the DWP merits some discussion. To increase the by R1 ðdt; hÞ ¼ cos2 h cosð90  hÞETHz ðdtÞ=ETHz ð0Þ. Substi-
generated THz field amplitude, it is important to maximize tuting dt ¼ 5:6 fs, and h ¼ 50 , we obtain R1 ð5:6 fs; 50 Þ
overlap of the time envelopes of the two laser pulses. How-  0:29. The polarizations of the x and 2x waves are parallel
ever, in our method, we cannot avoid a slight time deviation in experiments that include the DWP, thus, the efficiency with
between the two pulses due to group velocity dispersion in DWP inclusion depends only upon the difference in travelling
the DWP, despite the fact that the DWP is very thin. We esti- time of the two laser pulses, and the previous equation is sim-
mate the effect of this deviation on the THz field generation plified to R2 ðdtÞ ¼ ETHz ðdtÞ=ETHz ð0Þ. Upon substitution of
dt ¼ 13:5 fs, we obtain R2 ð13:5 fsÞ  0:62. Therefore, the
ideal enhancement in wave generation efficiency that might
be expected from the addition of a DWP becomes
R2 ð13:5 fsÞ=R1 ð5:6 fs; 50 Þ  2:1, which is 1.3 times larger
than what was found in our experiments. The source of this
discrepancy might be the time walk-off between 400 and
800 nm introduced by the BBO.
When using a DWP in the manner we have described,
the polarization of the two laser pulses is always parallel. In
practice, we found that the vertical polarization component
in our THz pulses was 15 times smaller than the horizontal
component. The polarization of the THz wave that is gener-
ated by this method is expected to be stable, which is of great
advantage when considering usability.
Since Thomas et al. reported that the intensity of THz
fields generated at 0.1–2 atm pressure is almost invariant,21 it
should be possible to reduce the ambient air pressure without
degrading the wave generation efficiency in the resulting
plasma. If one generates THz waves with the incorporation
FIG. 3. THz wave electric field intensity vs. the BBO-to-focus distance as
observed by EO detection. Closed circles: with DWP, open circles: without of a DWP at 0.1 atm of air, the time lag between the two
DWP. Dashed lines show the approximated envelopes. laser pulses due to air traversal will be reduced to 0.6 fs. In

Downloaded 11 Apr 2013 to 150.108.161.71. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
 041105-4 Minami et al. Appl. Phys. Lett. 102, 041105 (2013)

11
that case, the expected enhancement factor would become S. Watanabe, N. Minami, and R. Shimano, Opt. Express 19, 1528 (2011).
12
R2 ð8:5 fsÞ=R1 ð5:6 fs; 50 Þ  2:9, implying an achievable H. Hirori, A. Doi, F. Blanchard, and K. Tanaka, Appl. Phys. Lett. 98,
091106 (2011).
THz field of up to 1.9 MV/cm. 13
J. Chen, P. Han, and X.-C. Zhang, Appl. Phys. Lett. 95, 011118 (2009).
In conclusion, we have generated THz waves with field 14
Y. Minami, M. Nakajima, and T. Suemoto, Phys. Rev. A 83, 023828
strengths of 1.4 MV/cm in an air plasma produced by two- (2011).
15
M. Jewariya, M. Nagai, and K. Tanaka, Phys. Rev. Lett. 105, 203003
color laser pulses by including a DWP in the experimental
(2010).
setup. In the experiment, we adjusted the polarization of the 16
B. Ewers, N. S. K€ oster, R. Woscholski, M. Koch, S. Chatterjee, G. Khi-
laser pulses by means of said DWP. Although the system is trova, H. M. Gibbs, A. C. Klettke, M. Kira, and S. W. Koch, Phys. Rev. B
simple when compared to a similar approach that includes a 85, 075307 (2012).
17
H. Hirori, K. Shinokita, M. Shirai, S. Tani, Y. Kadoya, and K. Tanaka,
DWP in addition to phase-compensating wedge plates,22–24 Nat. Commun. 2, 594 (2011).
our method performed very well and achieved satisfactory 18
I. Katayama, H. Aoki, J. Takeda, H. Shimosato, M. Ashida, R. Kinjo, I.
THz wave amplitudes. The method provides THz waves Kawayama, M. Tonouchi, M. Nagai, and K. Tanaka, Phys. Rev. Lett. 108,
within the frequency band of 0.5–7 THz, which should allow 097401 (2012).
19
H. Wen, M. Wiczer, and A. M. Lindenberg, Phys. Rev. B 78, 125203
for a wide exploration of many new physical phenomena. (2008).
20
H. Wen and A. M. Lindenberg, Phys. Rev. Lett. 103, 023902 (2009).
21
This work has been supported in part by a Grant-in-Aid H. G. Roskos, M. D. Thomson, M. Kreß, and T. L€ offler, Laser Photon.
for Scientific Research (KAKENHI, Nos. 23244063, Rev. 1, 349 (2007).
22
K.-Y. Kim, A. J. Taylor, J. H. Glownia, and G. Rodriguez, Nat. Photon. 2,
23686012, 23760045, and 24656043) from JSPS, and by a 605 (2008).
research granted from The Murata Science Foundation. 23
J. Dai, N. Karpowicz, and X.-C. Zhang, Phys. Rev. Lett. 103, 023001
(2009).
24
1
C. Fattinger and D. Grischkowsky, Appl. Phys. Lett. 53, 1480 (1988). T.-J. Wang, C. Marceau, S. Yuan, Y. Chen, Q. Wang, F. Theberge,
2
A. Rice, Y. Jin, X. F. Ma, X.-C. Zhang, D. Bliss, J. Larkin, and M. M. Ch^ateauneuf, J. Dubois, and S. L. Chin, Laser Phys. Lett. 8, 57
Alexander, Appl. Phys. Lett. 64, 1324 (1994). (2011).
25
3
D. H. Auston, K. P. Cheung, J. A. Valdmanis, and D. A. Kleinman, Phys. M. Kress, T. L€ offler, S. Eden, M. Thomson, and H. G. Roskos, Opt. Lett.
Rev. Lett. 53, 1555 (1984). 29, 1120 (2004).
26
4
H. Hamster, A. Sullivan, S. Gordon, and R. W. Falcone, Phys. Rev. E 49, K. Y. Kim, J. H. Glownia, A. J. Taylor, and G. Rodriguez, Opt. Express
671 (1994). 15, 4577 (2007).
27
5
T. Bartel, P. Gaal, K. Reimann, M. Woerner, and T. Elsaesser, Opt. Lett. V. Vaičaitis, V. Smilgevičius, and V. Jarutis, Opt. Commun. 284, 2206
30, 2805 (2005). (2011).
28
6
D. A. Jaroszynski, B. Ersfeld, G. Giraud, S. Jamison, D. R. Jones, R. C. J. Dai, J. Zhang, W. Zhang, and D. Grischkowsky, J. Opt. Soc. Am. B 21,
Issac, B. M. W. McNeil, A. D. R. Phelps, G. R. M. Robb, H. Sandison, G. 1379 (2004).
29
Vieux, S. M. Wiggins, and K. Wynne, Nucl. Instrum. Methods Phys. Res. A The electric field of THz wave ET is calibrated as ET
445, 317 (2000). ¼ ðk0 n30 r41 LtGaP tNSi Þ1 sin1 ðDI=IÞ, where DI=I is the modulated ratio of
7
W. Shi, Y. J. Ding, N. Fernelius, and K. L. Vodopyanov, Opt. Lett. 27, the EO signal, k0 is the wave number of the probe pulse, n0 is the refractive
1454 (2002). index of GaP, r41 is the EO coefficient of GaP, and L is the thickness of
8
Y. Jiang, D. Li, Y. J. Ding, and I. B. Zotova, Opt. Lett. 36, 1608 (2011). the GaP, respectively. See also Ref. 12.
30
9
T. Kitada, F. Tanaka, T. Takahashi, K. Morita, and T. Isu, Appl. Phys. Q. Wu and X.-C. Zhang, Appl. Phys. Lett. 70, 1784 (1997).
31
Lett. 95, 111106 (2009). S. Casalbuoni, H. Schlarb, B. Schmidt, P. Schm€ user, B. Steffen, and A.
10
K.-L. Yeh, M. C. Hoffmann, J. Hebling, and K. A. Nelson, Appl. Phys. Winter, Phys. Rev. ST Accel. Beams 11, 072802 (2008).
32
Lett. 90, 171121 (2007). X. Xie, J. Dai, and X.-C. Zhang, Phys. Rev. Lett. 96, 075005 (2006).

Downloaded 11 Apr 2013 to 150.108.161.71. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions