- Email: [email protected]
Terahertz emission from vertically aligned multi-wall carbon nanotubes and their composites by optical excitation
Accepted Manuscript Terahertz emission from vertically aligned multi-wall carbon nanotubes and their composites by optical excitation Shan Huang, Weilong Li, Lipeng Zhu, Mi He, Zehan Yao, Yixuan Zhou, Xinlong Xu, Zhaoyu Ren, Jinbo Bai PII:
S0008-6223(18)30191-X
DOI:
10.1016/j.carbon.2018.02.067
Reference:
CARBON 12903
To appear in:
Carbon
Received Date: 29 October 2017 Revised Date:
11 February 2018
Accepted Date: 16 February 2018
Please cite this article as: S. Huang, W. Li, L. Zhu, M. He, Z. Yao, Y. Zhou, X. Xu, Z. Ren, J. Bai, Terahertz emission from vertically aligned multi-wall carbon nanotubes and their composites by optical excitation, Carbon (2018), doi: 10.1016/j.carbon.2018.02.067. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Graphical Abstract
ACCEPTED MANUSCRIPT
Terahertz Emission from Vertically Aligned Multi-wall Carbon Nanotubes and their Composites by Optical Excitation
RI PT
Shan Huanga, Weilong Lia,*, Lipeng Zhua, Mi Hea, Zehan Yaoa, Yixuan Zhoub, Xinlong Xua, Zhaoyu Rena,*, Jinbo Baic a
State Key Lab Incubation Base of Photoelectric Technology and Functional
SC
Materials, and Institute of Photonics & Photon-Technology, Northwest University,
M AN U
Xi’an 710069, China b
Physics Department, Northwest University, Xi’an 710069, China
c
Lab. MSSMAT, CNRS UMR8579, Ecole Centrale Paris, Grande Voie des Vignes,
AC C
EP
TE D
92290 Châtenay-Malabry, France
∗
Corresponding authors.
Weilong Li, Tel: +86 29 88303697, e-mail address: [email protected] (Weilong Li); Zhaoyu Ren, Tel: +86 29 88303336, e-mail address: [email protected] (Zhaoyu Ren)
ACCEPTED MANUSCRIPT Abstract Terahertz (THz) emission has been successfully observed from vertically aligned multi-wall carbon nanotube (VAMCNT) materials excited with polarized laser pulse
RI PT
in a reflection configuration. The effects of experimental parameters on THz generation are investigated systematically. The results indicate that the linear dependence of THz electric field on the pump power shows a typical second-order
SC
nonlinear optical effect and the transient photocurrent for generating THz waves is
M AN U
mainly originated from photon drag effect. By comparing with THz emission from polymer-coated VAMCNT composites, the detailed generation process of THz waves from VAMCNT materials is clarified: the free carriers in VAMCNT films could transport both along single nanotube and between the adjacent nanotubes, while free
TE D
carriers in VAMCNT composite films transport mainly along the tube axis due to the isolated carrier transport between the adjacent nanotubes by the coated polymer. Thus, transient photocurrent for generating THz emission from VAMCNT composite films
EP
is only along the tube axis and its intensity depends on the component of the pump
AC C
light electric field along the tube axis. In addition, THz emission from VAMCNT composites shows more regular response than that from pure VAMCNT films, which paves a way for stable performance of VAMCNT-based THz emitter.
ACCEPTED MANUSCRIPT 1. Introduction Recently, carbon nanomaterials such as carbon nanotubes (CNTs) and graphene are emerging as the attractive candidate materials for next-generation photonic and
RI PT
optoelectronic devices working in terahertz (THz) region due to their unique optoelectronic properties[1-8]. In particular, THz emitters based on carbon nanomaterials have been reported by several pioneering works[9-12]. P.A. Obraztsov
SC
et al.[11] have reported the investigation of a strong interband photon drag effect in
M AN U
multilayer graphene leading to efficient emission of THz radiation. Plasmon enhanced THz emission from single-layer graphene on a gold substrate has also been observed by Young-Mi Bahk et al.[12]. Our group recently reported that the enhanced THz radiation had been observed from vertical grown graphene due to the strong
TE D
light-matter interaction compared with single-layer graphene[13]. On the other hand, THz modulator and detector based on aligned or unaligned CNTs have been reported experimentally and theoretically by many groups [14-21], while THz emitters based
EP
on CNTs were mainly investigated theoretically[22-26]. Kibis et al. [22] demonstrated
AC C
theoretically that quasi-metallic CNTs could emit THz radiation due to heating of the electron gas and resulting population inversion under an applied voltage. N.R. Sadykov et al.[26] have also reported the theoretical study on THz radiation generation process in the medium based on noninteracting parallel aligned CNTs. As one of rarely reported experimental researches on CNTs-based THz emitters[27, 28], L.V. Titova et al.[28] successfully generated broadband THz pulses from horizontally aligned single-wall CNTs excited by femtosecond optical pulses without externally
ACCEPTED MANUSCRIPT applied bias, and provided that the built-in electric field resulting in the transient photocurrent surge along the nanotube axis was responsible for the THz generation. However, THz generation in CNTs has until now only been considered in single-wall
RI PT
CNTs, while THz generation in multi-wall CNTs has been rarely investigated. In addition, the experimental researches on CNTs-based THz emitter are so few that the detailed generation mechanism of THz waves from CNTs is still ambiguous.
SC
Therefore, more results about THz generation in all types of CNTs by kinds of
M AN U
experimental configurations are eagerly anticipated to enrich the understanding of THz generation process from CNTs.
In this paper, we have successfully generated THz pulses from vertically aligned multi-wall carbon nanotube (VAMCNT) materials excited with polarized laser pulse
TE D
in a reflection configuration. VAMCNT materials are easily fabricated and low-cost compared with horizontally aligned single-wall CNTs, which is helpful to further practical application of CNTs-based THz emitter. Meanwhile, compared with THz
EP
generation in transmission configuration which may result in strong reabsorption of
AC C
the emitted THz pulses by the aligned CNTs acting as a THz polarizer[18, 28], THz generation in a reflection configuration is more effective and the thickness of THz generation materials is not strictly required. The effects of experimental parameters (sample angle, pump power and polarization angle of pump light, etc.) on the waveform and intensity of THz pulses are evaluated in detail. In addition, by comparing with THz pulses from polymer-coated VAMCNT composites, the generation process of THz waves from VAMCNT materials is explored. The
ACCEPTED MANUSCRIPT comparative study on THz generation from VAMCNTs and their composites enriches the understanding of THz generation mechanism from CNTs and paves a way for CNTs-based THz emitter with stable performance.
2.1 Preparation of the VAMCNT sample
RI PT
2. Experimental
The VAMCNTs grown on quartz substrate (1×1 cm2, the thickness is 500±20 µm)
SC
are prepared by chemical vapor deposition (CVD) method with C2H2 as carbon source.
M AN U
The detailed procedure is similar as our previous work[29]. Briefly, the quartz substrate is placed into the center of a quartz tube (110 cm long, inner diameter 5 cm) and heated to the set temperature (750 oC) by a horizontal tube furnace under the carrier gases (argon and hydrogen) with certain flow rate ratio. Then, ferrocene
TE D
(Fe(C5H5)2) dissolved in xylene (C8H10) at a concentration of 0.06 g·ml-1 as the catalyst precursor is sprayed into the tube furnace by a syringe with the help of carrier gases (Ar and H2). At the same time, C2H2 is introduced into the tube furnace at the
EP
flow rate of 42 ml·min-1. After the growth for 6 min, the VAMCNT sample is cooled
AC C
to room temperature in Ar atmosphere. 2.2 Preparation of the VAMCNT-based composites The epoxy resin is selected as the polymer matrix. The polymer solution is
obtained by mixing the Epon812, twelve carbon succinic anhydride (DDSA), four hydrogen methyl benzene acid methyl (MNA) and phenol (DMP-30), and continuously heated at 40 oC for 2 hours in order to remove air bubbles. Then, the VAMCNT materials are placed into the prepared solution and heated at 40 oC for 48
ACCEPTED MANUSCRIPT hours. After drying at 60 oC for 48 hours, polymer-coated VAMCNT composites are acquired.
M AN U
SC
RI PT
2.3 THz experimental setup
Figure1. (a) The schematic diagram of THz generation experimental setup. (b) Schematic of VAMCNTs excited in reflection configuration. θpump is the angle between pump beam electric field vector E and the direction of X axis. The incidence angle of the pump light is shown as α.
TE D
Typical THz generation experimental setup in reflection configuration is shown in Figure 1(a). The femtosecond laser source is Ti: Sapphire oscillator, centered at a wavelength of 800 nm with a repetition rate of 1 KHz. The output of laser beam is
EP
divided into two parts by a beam splitter M1: a strong energy light as a pump beam
AC C
and a weak energy light as a reference beam, the pump beam is focused on the sample and the generated THz pulses in reflection direction are collected onto the electro-optic detection crystal (ZnTe(110)) by two the off-axis parabolic mirrors (P1 and P2). At the same time, the reference beam (or named time delay light) by reflection mirrors (M2, M3) is also gathered to the electro-optic detection crystal (ZnTe(110)) and overlapped with the emitted THz pulses. In THz generation setup, the VAMCNT samples are installed in the sample frame, and the different sample
ACCEPTED MANUSCRIPT angles θsample (0°，30°，45°，60°，90° and 180°) are attained by rotating the frame around the CNT axis, as shown as Figure 1(b). The incident angle ɑ of the pump light is fixed on 45°, and the pump polarization angle θpump, which is defined as the angle
RI PT
between the electric field vector E of the pump beam and the direction of X axis, could be changed with a half-wave plate between the sample and beam splitter M1.
field E with the θpump are shown in Table 1.
SC
The corresponding relations of the angle between CNT axis and pump beam electric
M AN U
Table 1. Corresponding relations of the angle between CNT axis and pump beam electric field with θpump Corresponding relations of the angle between CNT axis and pump beam electric field with θpump Pump 0o 30o 45o 60o 90o 180o polarization angle (θpump) 128o
120o
111o
90o
45o
TE D
135o
EP
The angle between CNT axis and pump beam electric field
3. Results and discussion
AC C
The photograph of VAMCNT with an area of about 1×1 cm2 grown on a quartz
substrate has been shown in Figure 2(a). In order to affirm the micro-structure of the CNT array, fractured edge of the sample has been characterized by a scanning electron microscopy (SEM, FEI Quanta 400 ESEM-FEG) as shown in Figure 2(b-d). From SEM picture of a large ratio we can clearly see that the length of VAMCNT is about 54 µm, and the structural asymmetry exists between the top and bottom of VAMCNT. At the bottom of VAMCNT, the diameter is larger, about 90 nm.
ACCEPTED MANUSCRIPT However, the CNT diameter gradually becomes smaller at the top of VAMCNTs (Figure 2(c)). The diameter change is mainly determined by the bottom growth pattern[30] of VAMCNTs in this experiment. That is to say, the catalyst particles are
RI PT
always attached on the surface of the substrate during the growth process of CNTs, and CNTs are grown on the top side of catalyst particles. Therefore, the CNT diameter is equivalent to the corresponding catalyst particle diameter. At the
SC
beginning of the CNT growth, the diameter of catalyst particles is smaller, which
M AN U
results in the small diameter CNTs at the top of final VAMCNTs. However, the diameter of catalyst particles increases with the growth of CNTs due to the continuing feedstock of catalyst source and the aggregation of neighboring catalysts, which is responsible for the large diameter CNTs at the bottom of final VAMCNTs. Moreover,
AC C
EP
TE D
the array orientation degree of VAMCNTs is different between the bottom section and
Figure 2. (a) Photograph of VAMCNT on a quartz substrate. (b) SEM images of the VAMCNT at the fractured edge of the sample under the lower magnification. SEM images of the top (c) and bottom (d) ends of VAMCNT under the higher magnification.
ACCEPTED MANUSCRIPT the top section. The orientation degree at the VAMCNT bottom is better than that at the top. The change of orientation degree of VAMCNTs is mainly attributed to the evolution of CNT diameter during the growth and the influence of the dual role of
M AN U
SC
RI PT
hydrogen [31].
Figure 3. (a) Typical THz pulse waveform emitted by the VAMCNT when exited with polarized pump light. Inset: corresponding Fourier amplitude spectrum. (b) The emitted THz domain waveform from the VAMCNT and the pure quartz substrate (not covered by VAMCNT), respectively.
A typically excited THz waveform in the time domain is shown in Figure 3(a) and
TE D
the corresponding amplitude spectrum by the Fourier transform is shown in the inset of Figure 3(a). THz waveform is obtained in the reflection direction of the 800 nm
EP
pump pulse when θpump and θsample are 0° and 90°, respectively. Thus we indicate that VAMCNTs could effectively generate THz pulse, and the maximum amplitude is
AC C
acquired at 0.5 THz. Meanwhile, we can observe that the bandwidth of generated THz pulse is broad in range of 0.1-2.3 THz. We also demonstrate that the quartz substrate does not contribute to THz emission of VAMCNTs, because THz radiation is not detected under the same excitation condition from the pure quartz substrate (not covered by VAMCNTs), as shown in Figure 3(b). In addition, THz generation from VAMCNT samples with different thickness is also investigated as shown in Figure S1 in the Supporting Information. From the Figures, we can see that THz waves could be
ACCEPTED MANUSCRIPT successfully generated from VAMCNT samples with different thickness, and the effects of experimental parameters on the emitted THz pulses from VAMCNT samples with different thickness are similar. Therefore, only surface (with certain skin
RI PT
depth) of the VAMCNT films contributes into THz generation while volume contribution is negligible.
Figure 4(a) demonstrates the dependence of the excited THz radiation waveform on
SC
the CNT orientation with fixed pump polarization at 0°. With the increase of the
M AN U
sample angle, the intensity and waveform of emitted THz pulses have a little change. We can imagine that, if the VAMCNT was ideally and perfectly aligned along the substrate, the emitted THz pulses should be completely same when the sample angle changes according to the configurations of our experimental setup. Therefore, the
TE D
slight difference of the emitted THz pulses should be resulted from the nonuniformity of the VAMCNT sample (as shown in Figure 2). In addition, the corresponding THz amplitude spectra of the emitted THz waves exhibit the same situation as their time
AC C
EP
domain signals in Figure 4(b).
Figure 4. (a) The time domain waveforms of the emitted THz signals from the VAMCNTs on a quartz substrate with different sample orientations (θpump = 0o). (b) Corresponding THz amplitude spectra by the Fourier transform.
ACCEPTED MANUSCRIPT The influence of the incident laser power on the generated THz pulse waveform for θpump=0°and θsample=0°is shown in Figure 5(a). With the increasing of pump power, the electric field intensity of THz pulse is gradually strengthened. The pump
RI PT
power (P) can be determined as P=ε/τhω, where ε and τhω are the energy of pump laser and the duration of pump laser pulse [32], respectively. Therefore, the energy of the excitation light is increased with its pump power, and it will result in the increase of
SC
THz pulse intensity. The corresponding THz amplitude spectra by the Fourier
M AN U
transform are shown in Figure 5(b). The frequency bandwidth is unchanged when the pump power increases, which is in range of 0.1-2.3 THz for 800 nm pump light and the frequency is approximately 0.5 THz for the maximum amplitude of the emitted
AC C
EP
TE D
THz signals.
Figure 5. (a) The time domain waveforms of the emitted THz signals from the VAMCNT on a quartz substrate with different pump powers. (b) Corresponding THz amplitude spectra. (c) Emitted THz electric field from the VAMCNT sample on a quartz substrate vs pump power.
ACCEPTED MANUSCRIPT Figure 5(c) shows that THz electric field is linearly changed with the increasing pump power. The black spots are the experimental data and the red solid line emphasizes the linear dependence. This is a typical second-order nonlinear optical
power and thus quadratic with the pump electric field.
RI PT
effect for THz generation process, because THz emission is proportional to the pump
Figure 6(a) shows THz radiations with different pump polarization angles ranged
SC
from 0o to 180o (the corresponding angle between the electric field of pump light and
M AN U
the CNT axis varied from 135o to 45o, as shown in Table 1). The maximum THz pulse intensity is obtained when the θpump is 90°, while the emitted THz pulse intensity is lower for θpump =0o or 180o. The corresponding THz amplitude spectra by the Fourier transform are shown in Figure 6(b), and the change trend of the amplitude spectra is
TE D
the same as that of the time domain signals. In Figure 6(c), we exhibit THz electric field intensity as a function of the pump polarization angle θpump. The θpump is obtained by rotating the half wave-plate, and the rotation angle of the half wave-plate is
EP
represented as φ. The relation of θpump and φ is: θpump =2φ. The results indicate that the
AC C
intensity of THz electric field in VAMCNT sample shows sin2φ dependence behavior. Optical rectification effect mainly occurs on the condition that pump photon energy (ħω) is smaller than the band gap (Eg) of some materials[33], while photon drag effect (PDE) and photogalvanic effect can occur in the process of intra-band or inter-band[34, 35]. In our materials (ħω » Eg), the optical rectification effect will be eliminated. P. A. Obraztsov et al.[35] demonstrated that the PDE and photogalvanic effect could be occurred simultaneously in micrometer thick nanographite with light
ACCEPTED MANUSCRIPT excitation. However, as reported by many references[36, 37], photo-galvanic effects are only present in noncentrosymmetric systems. In ideal VAMCNTs (with centrosymmetry) all photo-galvanic effects are strictly forbidden by symmetry. The
RI PT
VAMCNT samples in our study are not perfectly centrosymmetric, thus the photo-galvanic effects resulted from the imperfect symmetry may be existing (but not predominant). Therefore, it seems that PDE is more suitable for being employed to
AC C
EP
TE D
M AN U
SC
explain the mechanism of THz generation from the VAMCNT film.
Figure 6. (a) The time domain waveform of the emitted THz signal from the VAMCNTs on a quartz substrate with different pump polarization angles. (b)Fourier-transformed THz amplitude as a function of frequency. (c)Emitted THz electric field intensity from the VAMCNT sample on a quartz substrate vs pump polarization angle
With regard to PDE, the momentum of electrons (free carriers) in the VAMCNT film is increased by the incident photons with the large momentum when the sample is excited by the laser pulse. The process could give rise to the direct movement of the
ACCEPTED MANUSCRIPT electrons (free carriers) along the specific direction. Consequently, the electrons (free carriers) are dragged by photons, which will lead to the transient photocurrent and further generate THz radiation in the VAMCNT films. In this process of the photon
RI PT
momentum transformation, the participation of the phonon is needed and it could be obtained by the conservation of energy and momentum[38]. Meanwhile, the conservation reveals that the light-induced current exists only during the momentum
SC
relaxation time, which is determined by the electron-photon coupling[39-42]. With
M AN U
the enhancing of the coupling interaction the conversion efficiency of the momentum is higher. In our experiment, as analyzed and mentioned above (Fig. 6(a)), the strongest amplitude of emitted THz electric field from VAMCNT film is obtained in the reflection direction of pump light when the electric field of pump light is
TE D
perpendicular to the CNT axis, and the transient photocurrent induced by PDE reaches the maximum value. L.V. Titova et al.[28] reported the similar dependence behavior of emitted THz amplitude from horizontally aligned CNT film on the
EP
electric field of pump light in the transmission direction of pump light. They
AC C
considered that this behavior was caused by the anisotropy of optical absorption and reflectance of aligned CNT films. However, the inner carrier transport mechanism of CNT array samples was not involved. It is worth notifying that the free carriers not only could transport along single CNT but also may transport between the adjacent CNTs (due to the interaction between CNTs), and both aspects could form the transient photocurrent induced by PDE along the corresponding directions, which further lead to the generation of polarized THz radiation. Therefore, the amplitude of
ACCEPTED MANUSCRIPT emitted THz electric field from aligned CNT films should be determined by the
TE D
M AN U
SC
RI PT
carrier transport direction and the electric field of pump light.
Figure 7. SEM images of the VAMCNT composite films at the fractured edge of the sample under the different magnifications.
EP
In order to further elucidate the mechanism of THz generation from VAMCNT films, we prepared VAMCNT composite films by in situ coating epoxy resin polymer.
AC C
As illustrated by Figure 7, each tube of the VAMCNT is completely coated and the array orientation of the VAMCNT is well kept. The tube diameter is in the range of 100-120 nm, which is larger than that of pure VAMCNTs due to the polymer coating. The polymer-coated method can efficiently isolate the interaction between the adjacent CNTs (that is to say, isolate the carrier transport across neighboring CNTs), and hence THz generation from VAMCNT composite films should be only related to the carrier transport along the tube axis. For comparing to the pure VAMCNT films,
ACCEPTED MANUSCRIPT the VAMCNT composite films is also excited by laser pulse to detect THz generation with different sample angles, pump polarization angles and pump powers (the varied parameters are consistent with those in pure VAMCNT films), respectively. The
RI PT
evolutions of the emitted THz pulses from VAMCNT composite films with the sample angle and pump power are similar to the results of the pure VAMCNT films, except that the amplitude of the emitted THz electric field decreased slightly
SC
compared with pure VAMCNT films (Figure S2 and Figure S3 in Supporting
M AN U
Information). However, when changing the pump polarization angle, the evolution of the emitted THz pulse from VAMCNT composite films is different from that of the
AC C
EP
TE D
pure VAMCNT films, as shown in Figure 8.
Figure 8. (a) The time domain waveform of the emitted THz signal from VAMCNT composite films with different pump polarization angles. (b)Emitted THz electric field intensity from VAMCNT composite films vs pump polarization angle. The comparison of the emitted THz waveform from pure VAMCNT films and VAMCNT composite films: (c) θpump = 90o; (d) θpump =180o.
ACCEPTED MANUSCRIPT Figure 8(a) shows THz radiations from VAMCNT composite films with different pump polarization angles ranged from 0o to 180o. The higher THz pulse intensity is obtained when the θpump is 0o or 180o, while the emitted THz pulse intensity is
RI PT
minimum for θpump = 90°. In addition, the intensity of THz electric field emitted from VAMCNT composite films as a function of the pump polarization angle θpump (=2φ) shows cos2φ dependence behavior (Figure 8(b)).The results above are totally different
SC
from those of pure VAMCNT films. As illustrated in Figure 8(c), for θpump= 90o, the
M AN U
emitted THz signal from pure VAMCNT films is very strong, while the emitted THz signal from polymer-coated VAMCNT films nearly disappears. For θpump= 180o, the amplitude of the emitted THz signal from pure VAMCNT films is almost same as that from polymer-coated VAMCNT films (Figure 8(d)). The results further indicate that
TE D
the free carriers in VAMCNT films could transport both along single CNT and between the adjacent CNTs (Figure 9(a)), while free carriers in VAMCNT composite films transport mainly along the tube axis due to the isolated carrier transport between
EP
the adjacent CNTs by the coated polymer (Figure 9(b)). Therefore, for θpump= 90o (the
AC C
angle between the electric field of pump light and the CNT axis is also 90o, as shown in Table 1), the PDE-induced transient photocurrent in pure VAMCNT films transports only perpendicular to the tube axis, which leads to THz wave generation in the corresponding direction. While pure VAMCNTs are coated by epoxy resin polymer to form VAMCNT composite films, the only path of carrier transport perpendicular to the tube axis is interdicted for the same pump polarization angle (90o), thus the transient photocurrent for generating THz waves will not appear. For
ACCEPTED MANUSCRIPT the VAMCNT composite films, regardless of pump polarization angle, the carrier transport is along the tube axis due to the isolated carrier transport between the adjacent CNTs. When the pump polarization angle changes from 0o to 60o (the
RI PT
corresponding angle between the electric field of pump light and the CNT axis varies from 135o to 111o, see Table 1), the component of the pump light electric field along the tube axis decreases gradually (although the component of the pump light electric
SC
field perpendicular to the tube axis increases gradually, this component does not
M AN U
contribute to THz generation from VAMCNT composite films), which result in the weakened intensity of the emitted THz electric field from VAMCNT composite films,
AC C
EP
TE D
as shown in Figure S4 in Supporting Information.
Figure 9. Schematic diagram of the carrier transport paths in (a) pure VAMCNT films and (b) VAMCNT composite films.
In addition, we find that THz response of VAMCNT composite films is more
coherent than that of pure VAMCNT films. The intensity of the emitted THz electric field from VAMCNT composite films with different sample angles is more uniform than that from pure VAMCNT films (Figure S2). The linear dependence of the
ACCEPTED MANUSCRIPT emitted THz electric field on the pump power for VAMCNT composite films is better than that for pure VAMCNT films (Figure S3(c)). The intensity of the emitted THz electric field from VAMCNT composite films decreases with the pump polarization
RI PT
angle changed from 0o to 90o, while the evolution of the emitted THz electric field from VAMCNT films with the pump polarization angle is irregular. The results above indicate that coating non-conductive polymer on CNT array could effectively
SC
suppress the interaction of the adjacent nanotubes and make THz response of CNT
M AN U
array composites more foreseeable (decided only by the structure parameters of CNT itself), because the interaction of the adjacent nanotubes is complex and determined by many factors (besides the structure parameters of CNT itself, relative position of the adjacent CNTs, the irregular bridge-linkage between CNTs, etc.). Therefore, THz
TE D
devices based on VAMCNT composites will be more stable and enable more regular response properties than those based on pure VAMCNTs. 4. Conclusion
EP
The generation of broadband (0.1-2.3 THz) THz waves has been observed in
AC C
VAMCNT materials excited with 800 nm laser pulse in reflection configuration. The intensity and waveform of emitted THz pulses from pure VAMCNT films with different sample angles have a slight difference due to the nonuniformity of the VAMCNT sample. THz electric field is linearly changed with the pump power, which is a typical second-order nonlinear optical effect for THz generation process. With pump polarization angles changed from 0o to 180o, the maximum THz pulse intensity from pure VAMCNT films is obtained for θpump=90°, and the intensity of emitted THz
ACCEPTED MANUSCRIPT electric field shows sin2φ dependence behavior. The results indicate that PDE is suitable for being employed to explain THz generation from the VAMCNT film. The polymer-coated VAMCNT composites are prepared and excited by laser pulse on the
RI PT
same conditions as the pure VAMCNT films to comparatively study the THz generation mechanism. The higher THz pulse intensity from polymer-coated VAMCNT films is obtained when the θpump is 0o or 180o, while the emitted THz
SC
signal nearly disappears for θpump= 90o. The totally different results from pure
M AN U
VAMCNT films indicate that the free carriers in VAMCNT films could transport both along single CNT and between the adjacent CNTs, while free carriers in VAMCNT composite films transport mainly along the tube axis due to the isolated carrier transport between the adjacent CNTs by the coated polymer. Thus, the transient
TE D
photocurrent for generating THz pulses from VAMCNT composite films is only along the tube axis and its intensity depends on the component of the pump light electric field along the tube axis. In addition, we find that THz response of VAMCNT
EP
composite films is more coherent than that of pure VAMCNT films. Such
AC C
polymer-coated VAMCNT composites have great potentials in improving the stability and response regularity of THz devices. Acknowledgements
This work was financially supported by the National Natural Science Foundation of
China (No. 11304249, 61605160, 11774288), Key Science and Technology Innovation Team Project of Natural Science Foundation of Shaanxi Province (2017KCT-01), and the International Cooperative Program (NO. 201410780).
ACCEPTED MANUSCRIPT References [1] Xinlong Xu, Ken Chuang, Robin J. Nicholas, Michael B. Johnston, Laura M. Herz. Terahertz Excitonic Response of Isolated Single-Walled Carbon Nanotubes. Journal of Physical Chemistry C 2009, 113(42): 18106-9.
RI PT
[2] Riccardo Degl Innocenti, David S. Jessop, Christian W. O. Sol, Long Xiao, Stephen J. Kindness, Hungyen Lin, et al. Fast Modulation of Terahertz Quantum Cascade Lasers Using Graphene Loaded Plasmonic Antennas. Acs Photonics 2016,
SC
3(3): 467-70.
[3] Matthew C. Beard, Jeffrey L. Blackburn, Michael J. Heben. Photogenerated free
Nano Letters 2008, 8(12): 4238-42.
M AN U
carrier dynamics in metal and semiconductor single-walled carbon nanotube films.
[4] Rana, F. Graphene Terahertz Plasmon Oscillators. IEEE Transactions on Nanotechnology 2008, 7(1): 91-9.
TE D
[5] Q. Zhang, E. H. H´aroz, Z. Jin, L. Ren, X. Wang, R. S. Arvidson, et al. Plasmonic Nature of the Terahertz Conductivity Peak in Single-Wall Carbon Nanotubes. Nano Letters 2013, 13(12): 5991-6.
EP
[6] Yixuan Zhou, Yiwen E, Lipeng Zhu, Mei Qi, Xinlong Xu, Jintao Bai, et al.
AC C
Terahertz wave reflection impedance matching properties of graphene layers at oblique incidence. Carbon 2015, 96(10): 1129-37. [7] M. J. Paul, N. A. Kuhta, J. L. Tomaino, A. D. Jameson, L. P. Maizy, T. Sharf, et al. Terahertz transmission ellipsometry of vertically aligned multi-walled carbon nanotubes. Applied Physics Letters 2012, 101(11): 111107-10. [8] X. L. Xu, P. Parkinson, K.-C. Chuang, M. B. Johnston, R. J. Nicholas, L. M. Herz. Dynamic terahertz polarization in single-walled carbon nanotubes. Physical Review B Condensed Matter 2010, 82(8): 085441-4.
ACCEPTED MANUSCRIPT [9] Leonhard Prechtel, Li Song, Dieter Schuh, Pulickel Ajayan, Werner Wegscheider, Alexander W. Holleitner. Time-resolved ultrafast photocurrents and terahertz generation in freely suspended graphene. Nature Communications 2012, 3(48): 19596-600.
RI PT
[10] J. Maysonnave, S. Huppert, F. Wang, S. Maero, C. Berger, W. de Heer, et al. Terahertz Generation by Dynamical Photon Drag Effect in Graphene Excited by Femtosecond Optical Pulses. NANO LETTERS 2015, 14:5797-802.
SC
[11] Obraztsov Petr A, Natsuki Kanda, Kuniaki Konishi, Makoto Kuwata-Gonokami, Sergey V. Garnov, Alexander N, et al. Photon-drag-induced terahertz emission from
M AN U
graphene. Physical Review B 2014, 90(24): 241416-20.
[12] Young-Mi Bahk, Gopakumar Ramakrishnan, Jongho Choi, Hyelynn Song, Geunchang Choi, Yong Hyup Kim, et al. Plasmon Enhanced Terahertz Emission from Single Layer Graphene. Acs Nano 2014, 8(9): 9089-96.
TE D
[13] L. Zhu, Y. Huang, Z. Yao, B. Quan, L. Zhang, J. Li, et al, Enhanced polarization-sensitive terahertz emission from vertically grown graphene by a dynamical photon drag effect. Nanoscale 2017, 9(29): 10301-11.
EP
[14] Xiaowei He, Naoki Fujimura, J. Meagan Lloyd, Kristopher J. Erickson, A. Alec
AC C
Talin, Qi Zhang, et al. Carbon nanotube terahertz detector. Nano Letters 2014, 14(7): 3953-8.
[15] Callum J. Docherty, Samuel D. Stranks, Severin N. Habisreutinger, Hannah J. Joyce, Laura M. Herz, Robin J. Nicholas, et al. An ultrafast carbon nanotube terahertz polarisation modulator. Journal of Applied Physics 2014, 115(20): 203108-10. [16] Jisoo Kyoung, Eui Yun Jang, Marcio D. Lima, Hyeong-Ryeol Park, Raquel Ovalle Robles, Xavier Lepr, et al. A Reel-Wound Carbon Nanotube Polarizer for Terahertz Frequencies. Nano Letters 2011, 11(10): 4227-31.
ACCEPTED MANUSCRIPT [17] Lei Ren, Cary L. Pint, Layla G. Booshehri, William D. Rice, Xiangfeng Wang, David J. Hilton, et al. Carbon Nanotube Terahertz Polarizer. Nano Letters 2009, 9(7): 2610-2. [18] Lei Ren, Cary L. Pint, Takashi Arikawa, Kei Takeya, Iwao Kawayama,
RI PT
Masayoshi Tonouchi, et al. Broadband Terahertz Polarizers with Ideal Performance Based on Aligned Carbon Nanotube Stacks. Nano Letters 2012, 12(2): 787-90.
[19] Ahmed Zubair, Dmitri E. Tsentalovich, Colin C. Young, Martin S. Heimbeck,
Applied Physics Letters 2016, 108(14): 141107-10.
SC
Henry O. Everitt, Matteo Pasquali, et al. Carbon nanotube fiber terahertz polarizer.
M AN U
[20] Yixuan Zhou, Yiwen E, Xinlong Xu, Weilong Li, Huan Wang, Lipeng Zhu, et al. Angular dependent anisotropic terahertz response of vertically aligned multi-walled carbon nanotube arrays with spatial dispersion. Scientific Reports 2016, 6: 38515-23. [21] Ruili Wu, Weilong Li, Yun Wan, Zhaoyu Ren, Xinlong Xu, Yixuan Zhou.
TE D
Anisotropic terahertz response of stretch-aligned composite films based on carbon nanotube-SiC hybrid structures. Rsc Advances 2015, 5(34): 26985-90. [22] Kibis O.V., Rosenau da Costa M., Portnoi M.E. Generation of Terahertz
EP
Radiation by Hot Electrons in Carbon Nanotubes. Nano Letters 2007, 7(11): 3414-20.
AC C
[23] Portnoi M.E, Kibis O.V, Rosenau da Costa M. Terahertz Applications of Carbon Nanotubes. Superlattices & Microstructures 2008, 43(5): 399-407. [24] Parashar J, H. Sharma. Optical rectification in a carbon nanotube array and terahertz
radiation
generation.
Physica
E
Low-dimensional
Systems
and
Nanostructures 2012, 44(10): 2069-71. [25] Jain S, J. Parashar, R. Kurchania. Effect of magnetic field on terahertz generation via laser interaction with a carbon nanotube array. International Nano Letters 2013, 3(1): 1-4.
ACCEPTED MANUSCRIPT [26] N.R Sadykov, A.V. Aporoski, D.A. Peshkov. Terahertz radiation generation process in the medium based on the array of the noninteracting nanotubes. Optical and Quantum Electronics 2016, 48(7): 1-10. [27] Muthee M, E. Carrion, J. Nicholson, S.K. Yngvesson. Antenna-coupled terahertz
RI PT
radiation from joule-heated single-wall carbon nanotubes. Aip Advances 2011, 1(4): 042131-6.
[28] Lyubov V. Titova, Cary L. Pint, Qi Zhang, Robert H. Hauge, Junichiro Kono,
SC
Frank A. Hegmann. Generation of Terahertz Radiation by Optical Excitation of Aligned Carbon Nanotubes. Nano Letters 2015, 15(5): 3267-72.
M AN U
[29] Weilong Li, Jinkai Yuan, YouqinLin, ShenghongYao, ZhaoyuRen, HuiWang, et al. The controlled formation of hybrid structures of multi-walled carbon nanotubes on SiC plate-like particles and their synergetic effect as a filler in poly(vinylidene fluoride) based composites. Carbon, 2013, 51(1): 355-64.
graphene
carbon
TE D
[30] Lin J , Zhang C , Yan Z , Zhu Y , Peng Z , Hauge RH , et al. 3-Dimensional nanotube
carpet-based
microsupercapacitors
with
high
electrochemical performance. Nano Letters 2012, 13(1): 72-8.
EP
[31] Johannes Vanpaemel, Marleen H. van der Veen, Daire J. Cott, Masahito
AC C
Sugiura, Inge Asselberghs, Stefan De Gendt, et al. Dual Role of Hydrogen in Low Temperature Plasma Enhanced Carbon Nanotube Growth. Journal of Physical Chemistry C 2015, 119(32): 18293-302 [32] Gennady M. Mikheev, Albert G. Nasibulin, Ruslan G. Zonov, Antti Kaskela, Esko I. Kauppinen. Photon-drag effect in single-walled carbon nanotube films. Nano Letters 2012, 12(1): 77-83. [33] K Liu, J Xu, T Yuan, XC Zhang. Terahertz radiation from InAs induced by carrier diffusion and drift. Physical Review B 2006, 73(15): 155330-5.
ACCEPTED MANUSCRIPT [34] J. Maysonnave, S. Huppert, F. Wang, S. Maero, C. Berger, W. de Heer, et al. Room temperature broadband coherent terahertz emission induced by dynamical photon drag in graphene. Nano Letters 2014, 14: 5797-813. [35] Obraztsov P.A, Gennady M. Mikheev, Sergei V. Garnov, Alexander N.
RI PT
Obraztsov,Yuri P. Svirko. Polarization-sensitive photoresponse of nanographite. Applied Physics Letters 2011, 98(9): 091903-5.
Physica status solidi (b) 2012. 249(12): 2538-43.
SC
[36] E. L. Ivchenko. Photoinduced currents in graphene and carbon nanotubes.
M AN U
[37] M.M. Glazov, S.D. Ganichev. High frequency electric field induced nonlinear effects in graphene. Physics Reports 2014. 535(3): 101-38.
[38] Ryan W. Newson, Jean-Michel Ménard, Christian Sames, Markus Betz, Henry M. van Driel. Coherently Controlled Ballistic Charge Currents Injected in Single-Walled
TE D
Carbon Nanotubes and Graphite. Nano Letters 2008, 8(6): 1586-9. [39] Obraztsov P.A, Tommi Kaplas, Sergey V. Garnov, Makoto Kuwata-Gonokami, Alexander N. Obraztsov, Yuri P. Svirko. All-optical control of ultrafast photocurrents
EP
in unbiased graphene. Scientific Reports 2014, 4(2): 4007-12. [40] Marchetti S, Simili R, Bernardini M, M Giorgi. Detection of FIR radiation by
AC C
photon drag of free carriers in semimetal thin film. Physica Scripta 1988, 37(5): 820-2.
[41] Alexander N. Obraztsov, Dmitry A. Lyashenko, Shaoli Fang, Ray H. Baughman, Petr A. Obraztsov, Sergei V. Garnov, et al. Photon drag effect in carbon nanotube yarns. Applied Physics Letters 2009, 94(23): 231112-4. [42] Pfeiffer R, Kuzmany H, Simon F, Bokova S. N, Obraztsova E. Resonance Raman scattering from phonon overtones in double-wall carbon nanotubes. Physical Review B Condensed Matter 2005, 71(15): 155409-16.
S0008-6223(18)30191-X
DOI:
10.1016/j.carbon.2018.02.067
Reference:
CARBON 12903
To appear in:
Carbon
Received Date: 29 October 2017 Revised Date:
11 February 2018
Accepted Date: 16 February 2018
Please cite this article as: S. Huang, W. Li, L. Zhu, M. He, Z. Yao, Y. Zhou, X. Xu, Z. Ren, J. Bai, Terahertz emission from vertically aligned multi-wall carbon nanotubes and their composites by optical excitation, Carbon (2018), doi: 10.1016/j.carbon.2018.02.067. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Graphical Abstract
ACCEPTED MANUSCRIPT
Terahertz Emission from Vertically Aligned Multi-wall Carbon Nanotubes and their Composites by Optical Excitation
RI PT
Shan Huanga, Weilong Lia,*, Lipeng Zhua, Mi Hea, Zehan Yaoa, Yixuan Zhoub, Xinlong Xua, Zhaoyu Rena,*, Jinbo Baic a
State Key Lab Incubation Base of Photoelectric Technology and Functional
SC
Materials, and Institute of Photonics & Photon-Technology, Northwest University,
M AN U
Xi’an 710069, China b
Physics Department, Northwest University, Xi’an 710069, China
c
Lab. MSSMAT, CNRS UMR8579, Ecole Centrale Paris, Grande Voie des Vignes,
AC C
EP
TE D
92290 Châtenay-Malabry, France
∗
Corresponding authors.
Weilong Li, Tel: +86 29 88303697, e-mail address: [email protected] (Weilong Li); Zhaoyu Ren, Tel: +86 29 88303336, e-mail address: [email protected] (Zhaoyu Ren)
ACCEPTED MANUSCRIPT Abstract Terahertz (THz) emission has been successfully observed from vertically aligned multi-wall carbon nanotube (VAMCNT) materials excited with polarized laser pulse
RI PT
in a reflection configuration. The effects of experimental parameters on THz generation are investigated systematically. The results indicate that the linear dependence of THz electric field on the pump power shows a typical second-order
SC
nonlinear optical effect and the transient photocurrent for generating THz waves is
M AN U
mainly originated from photon drag effect. By comparing with THz emission from polymer-coated VAMCNT composites, the detailed generation process of THz waves from VAMCNT materials is clarified: the free carriers in VAMCNT films could transport both along single nanotube and between the adjacent nanotubes, while free
TE D
carriers in VAMCNT composite films transport mainly along the tube axis due to the isolated carrier transport between the adjacent nanotubes by the coated polymer. Thus, transient photocurrent for generating THz emission from VAMCNT composite films
EP
is only along the tube axis and its intensity depends on the component of the pump
AC C
light electric field along the tube axis. In addition, THz emission from VAMCNT composites shows more regular response than that from pure VAMCNT films, which paves a way for stable performance of VAMCNT-based THz emitter.
ACCEPTED MANUSCRIPT 1. Introduction Recently, carbon nanomaterials such as carbon nanotubes (CNTs) and graphene are emerging as the attractive candidate materials for next-generation photonic and
RI PT
optoelectronic devices working in terahertz (THz) region due to their unique optoelectronic properties[1-8]. In particular, THz emitters based on carbon nanomaterials have been reported by several pioneering works[9-12]. P.A. Obraztsov
SC
et al.[11] have reported the investigation of a strong interband photon drag effect in
M AN U
multilayer graphene leading to efficient emission of THz radiation. Plasmon enhanced THz emission from single-layer graphene on a gold substrate has also been observed by Young-Mi Bahk et al.[12]. Our group recently reported that the enhanced THz radiation had been observed from vertical grown graphene due to the strong
TE D
light-matter interaction compared with single-layer graphene[13]. On the other hand, THz modulator and detector based on aligned or unaligned CNTs have been reported experimentally and theoretically by many groups [14-21], while THz emitters based
EP
on CNTs were mainly investigated theoretically[22-26]. Kibis et al. [22] demonstrated
AC C
theoretically that quasi-metallic CNTs could emit THz radiation due to heating of the electron gas and resulting population inversion under an applied voltage. N.R. Sadykov et al.[26] have also reported the theoretical study on THz radiation generation process in the medium based on noninteracting parallel aligned CNTs. As one of rarely reported experimental researches on CNTs-based THz emitters[27, 28], L.V. Titova et al.[28] successfully generated broadband THz pulses from horizontally aligned single-wall CNTs excited by femtosecond optical pulses without externally
ACCEPTED MANUSCRIPT applied bias, and provided that the built-in electric field resulting in the transient photocurrent surge along the nanotube axis was responsible for the THz generation. However, THz generation in CNTs has until now only been considered in single-wall
RI PT
CNTs, while THz generation in multi-wall CNTs has been rarely investigated. In addition, the experimental researches on CNTs-based THz emitter are so few that the detailed generation mechanism of THz waves from CNTs is still ambiguous.
SC
Therefore, more results about THz generation in all types of CNTs by kinds of
M AN U
experimental configurations are eagerly anticipated to enrich the understanding of THz generation process from CNTs.
In this paper, we have successfully generated THz pulses from vertically aligned multi-wall carbon nanotube (VAMCNT) materials excited with polarized laser pulse
TE D
in a reflection configuration. VAMCNT materials are easily fabricated and low-cost compared with horizontally aligned single-wall CNTs, which is helpful to further practical application of CNTs-based THz emitter. Meanwhile, compared with THz
EP
generation in transmission configuration which may result in strong reabsorption of
AC C
the emitted THz pulses by the aligned CNTs acting as a THz polarizer[18, 28], THz generation in a reflection configuration is more effective and the thickness of THz generation materials is not strictly required. The effects of experimental parameters (sample angle, pump power and polarization angle of pump light, etc.) on the waveform and intensity of THz pulses are evaluated in detail. In addition, by comparing with THz pulses from polymer-coated VAMCNT composites, the generation process of THz waves from VAMCNT materials is explored. The
ACCEPTED MANUSCRIPT comparative study on THz generation from VAMCNTs and their composites enriches the understanding of THz generation mechanism from CNTs and paves a way for CNTs-based THz emitter with stable performance.
2.1 Preparation of the VAMCNT sample
RI PT
2. Experimental
The VAMCNTs grown on quartz substrate (1×1 cm2, the thickness is 500±20 µm)
SC
are prepared by chemical vapor deposition (CVD) method with C2H2 as carbon source.
M AN U
The detailed procedure is similar as our previous work[29]. Briefly, the quartz substrate is placed into the center of a quartz tube (110 cm long, inner diameter 5 cm) and heated to the set temperature (750 oC) by a horizontal tube furnace under the carrier gases (argon and hydrogen) with certain flow rate ratio. Then, ferrocene
TE D
(Fe(C5H5)2) dissolved in xylene (C8H10) at a concentration of 0.06 g·ml-1 as the catalyst precursor is sprayed into the tube furnace by a syringe with the help of carrier gases (Ar and H2). At the same time, C2H2 is introduced into the tube furnace at the
EP
flow rate of 42 ml·min-1. After the growth for 6 min, the VAMCNT sample is cooled
AC C
to room temperature in Ar atmosphere. 2.2 Preparation of the VAMCNT-based composites The epoxy resin is selected as the polymer matrix. The polymer solution is
obtained by mixing the Epon812, twelve carbon succinic anhydride (DDSA), four hydrogen methyl benzene acid methyl (MNA) and phenol (DMP-30), and continuously heated at 40 oC for 2 hours in order to remove air bubbles. Then, the VAMCNT materials are placed into the prepared solution and heated at 40 oC for 48
ACCEPTED MANUSCRIPT hours. After drying at 60 oC for 48 hours, polymer-coated VAMCNT composites are acquired.
M AN U
SC
RI PT
2.3 THz experimental setup
Figure1. (a) The schematic diagram of THz generation experimental setup. (b) Schematic of VAMCNTs excited in reflection configuration. θpump is the angle between pump beam electric field vector E and the direction of X axis. The incidence angle of the pump light is shown as α.
TE D
Typical THz generation experimental setup in reflection configuration is shown in Figure 1(a). The femtosecond laser source is Ti: Sapphire oscillator, centered at a wavelength of 800 nm with a repetition rate of 1 KHz. The output of laser beam is
EP
divided into two parts by a beam splitter M1: a strong energy light as a pump beam
AC C
and a weak energy light as a reference beam, the pump beam is focused on the sample and the generated THz pulses in reflection direction are collected onto the electro-optic detection crystal (ZnTe(110)) by two the off-axis parabolic mirrors (P1 and P2). At the same time, the reference beam (or named time delay light) by reflection mirrors (M2, M3) is also gathered to the electro-optic detection crystal (ZnTe(110)) and overlapped with the emitted THz pulses. In THz generation setup, the VAMCNT samples are installed in the sample frame, and the different sample
ACCEPTED MANUSCRIPT angles θsample (0°，30°，45°，60°，90° and 180°) are attained by rotating the frame around the CNT axis, as shown as Figure 1(b). The incident angle ɑ of the pump light is fixed on 45°, and the pump polarization angle θpump, which is defined as the angle
RI PT
between the electric field vector E of the pump beam and the direction of X axis, could be changed with a half-wave plate between the sample and beam splitter M1.
field E with the θpump are shown in Table 1.
SC
The corresponding relations of the angle between CNT axis and pump beam electric
M AN U
Table 1. Corresponding relations of the angle between CNT axis and pump beam electric field with θpump Corresponding relations of the angle between CNT axis and pump beam electric field with θpump Pump 0o 30o 45o 60o 90o 180o polarization angle (θpump) 128o
120o
111o
90o
45o
TE D
135o
EP
The angle between CNT axis and pump beam electric field
3. Results and discussion
AC C
The photograph of VAMCNT with an area of about 1×1 cm2 grown on a quartz
substrate has been shown in Figure 2(a). In order to affirm the micro-structure of the CNT array, fractured edge of the sample has been characterized by a scanning electron microscopy (SEM, FEI Quanta 400 ESEM-FEG) as shown in Figure 2(b-d). From SEM picture of a large ratio we can clearly see that the length of VAMCNT is about 54 µm, and the structural asymmetry exists between the top and bottom of VAMCNT. At the bottom of VAMCNT, the diameter is larger, about 90 nm.
ACCEPTED MANUSCRIPT However, the CNT diameter gradually becomes smaller at the top of VAMCNTs (Figure 2(c)). The diameter change is mainly determined by the bottom growth pattern[30] of VAMCNTs in this experiment. That is to say, the catalyst particles are
RI PT
always attached on the surface of the substrate during the growth process of CNTs, and CNTs are grown on the top side of catalyst particles. Therefore, the CNT diameter is equivalent to the corresponding catalyst particle diameter. At the
SC
beginning of the CNT growth, the diameter of catalyst particles is smaller, which
M AN U
results in the small diameter CNTs at the top of final VAMCNTs. However, the diameter of catalyst particles increases with the growth of CNTs due to the continuing feedstock of catalyst source and the aggregation of neighboring catalysts, which is responsible for the large diameter CNTs at the bottom of final VAMCNTs. Moreover,
AC C
EP
TE D
the array orientation degree of VAMCNTs is different between the bottom section and
Figure 2. (a) Photograph of VAMCNT on a quartz substrate. (b) SEM images of the VAMCNT at the fractured edge of the sample under the lower magnification. SEM images of the top (c) and bottom (d) ends of VAMCNT under the higher magnification.
ACCEPTED MANUSCRIPT the top section. The orientation degree at the VAMCNT bottom is better than that at the top. The change of orientation degree of VAMCNTs is mainly attributed to the evolution of CNT diameter during the growth and the influence of the dual role of
M AN U
SC
RI PT
hydrogen [31].
Figure 3. (a) Typical THz pulse waveform emitted by the VAMCNT when exited with polarized pump light. Inset: corresponding Fourier amplitude spectrum. (b) The emitted THz domain waveform from the VAMCNT and the pure quartz substrate (not covered by VAMCNT), respectively.
A typically excited THz waveform in the time domain is shown in Figure 3(a) and
TE D
the corresponding amplitude spectrum by the Fourier transform is shown in the inset of Figure 3(a). THz waveform is obtained in the reflection direction of the 800 nm
EP
pump pulse when θpump and θsample are 0° and 90°, respectively. Thus we indicate that VAMCNTs could effectively generate THz pulse, and the maximum amplitude is
AC C
acquired at 0.5 THz. Meanwhile, we can observe that the bandwidth of generated THz pulse is broad in range of 0.1-2.3 THz. We also demonstrate that the quartz substrate does not contribute to THz emission of VAMCNTs, because THz radiation is not detected under the same excitation condition from the pure quartz substrate (not covered by VAMCNTs), as shown in Figure 3(b). In addition, THz generation from VAMCNT samples with different thickness is also investigated as shown in Figure S1 in the Supporting Information. From the Figures, we can see that THz waves could be
ACCEPTED MANUSCRIPT successfully generated from VAMCNT samples with different thickness, and the effects of experimental parameters on the emitted THz pulses from VAMCNT samples with different thickness are similar. Therefore, only surface (with certain skin
RI PT
depth) of the VAMCNT films contributes into THz generation while volume contribution is negligible.
Figure 4(a) demonstrates the dependence of the excited THz radiation waveform on
SC
the CNT orientation with fixed pump polarization at 0°. With the increase of the
M AN U
sample angle, the intensity and waveform of emitted THz pulses have a little change. We can imagine that, if the VAMCNT was ideally and perfectly aligned along the substrate, the emitted THz pulses should be completely same when the sample angle changes according to the configurations of our experimental setup. Therefore, the
TE D
slight difference of the emitted THz pulses should be resulted from the nonuniformity of the VAMCNT sample (as shown in Figure 2). In addition, the corresponding THz amplitude spectra of the emitted THz waves exhibit the same situation as their time
AC C
EP
domain signals in Figure 4(b).
Figure 4. (a) The time domain waveforms of the emitted THz signals from the VAMCNTs on a quartz substrate with different sample orientations (θpump = 0o). (b) Corresponding THz amplitude spectra by the Fourier transform.
ACCEPTED MANUSCRIPT The influence of the incident laser power on the generated THz pulse waveform for θpump=0°and θsample=0°is shown in Figure 5(a). With the increasing of pump power, the electric field intensity of THz pulse is gradually strengthened. The pump
RI PT
power (P) can be determined as P=ε/τhω, where ε and τhω are the energy of pump laser and the duration of pump laser pulse [32], respectively. Therefore, the energy of the excitation light is increased with its pump power, and it will result in the increase of
SC
THz pulse intensity. The corresponding THz amplitude spectra by the Fourier
M AN U
transform are shown in Figure 5(b). The frequency bandwidth is unchanged when the pump power increases, which is in range of 0.1-2.3 THz for 800 nm pump light and the frequency is approximately 0.5 THz for the maximum amplitude of the emitted
AC C
EP
TE D
THz signals.
Figure 5. (a) The time domain waveforms of the emitted THz signals from the VAMCNT on a quartz substrate with different pump powers. (b) Corresponding THz amplitude spectra. (c) Emitted THz electric field from the VAMCNT sample on a quartz substrate vs pump power.
ACCEPTED MANUSCRIPT Figure 5(c) shows that THz electric field is linearly changed with the increasing pump power. The black spots are the experimental data and the red solid line emphasizes the linear dependence. This is a typical second-order nonlinear optical
power and thus quadratic with the pump electric field.
RI PT
effect for THz generation process, because THz emission is proportional to the pump
Figure 6(a) shows THz radiations with different pump polarization angles ranged
SC
from 0o to 180o (the corresponding angle between the electric field of pump light and
M AN U
the CNT axis varied from 135o to 45o, as shown in Table 1). The maximum THz pulse intensity is obtained when the θpump is 90°, while the emitted THz pulse intensity is lower for θpump =0o or 180o. The corresponding THz amplitude spectra by the Fourier transform are shown in Figure 6(b), and the change trend of the amplitude spectra is
TE D
the same as that of the time domain signals. In Figure 6(c), we exhibit THz electric field intensity as a function of the pump polarization angle θpump. The θpump is obtained by rotating the half wave-plate, and the rotation angle of the half wave-plate is
EP
represented as φ. The relation of θpump and φ is: θpump =2φ. The results indicate that the
AC C
intensity of THz electric field in VAMCNT sample shows sin2φ dependence behavior. Optical rectification effect mainly occurs on the condition that pump photon energy (ħω) is smaller than the band gap (Eg) of some materials[33], while photon drag effect (PDE) and photogalvanic effect can occur in the process of intra-band or inter-band[34, 35]. In our materials (ħω » Eg), the optical rectification effect will be eliminated. P. A. Obraztsov et al.[35] demonstrated that the PDE and photogalvanic effect could be occurred simultaneously in micrometer thick nanographite with light
ACCEPTED MANUSCRIPT excitation. However, as reported by many references[36, 37], photo-galvanic effects are only present in noncentrosymmetric systems. In ideal VAMCNTs (with centrosymmetry) all photo-galvanic effects are strictly forbidden by symmetry. The
RI PT
VAMCNT samples in our study are not perfectly centrosymmetric, thus the photo-galvanic effects resulted from the imperfect symmetry may be existing (but not predominant). Therefore, it seems that PDE is more suitable for being employed to
AC C
EP
TE D
M AN U
SC
explain the mechanism of THz generation from the VAMCNT film.
Figure 6. (a) The time domain waveform of the emitted THz signal from the VAMCNTs on a quartz substrate with different pump polarization angles. (b)Fourier-transformed THz amplitude as a function of frequency. (c)Emitted THz electric field intensity from the VAMCNT sample on a quartz substrate vs pump polarization angle
With regard to PDE, the momentum of electrons (free carriers) in the VAMCNT film is increased by the incident photons with the large momentum when the sample is excited by the laser pulse. The process could give rise to the direct movement of the
ACCEPTED MANUSCRIPT electrons (free carriers) along the specific direction. Consequently, the electrons (free carriers) are dragged by photons, which will lead to the transient photocurrent and further generate THz radiation in the VAMCNT films. In this process of the photon
RI PT
momentum transformation, the participation of the phonon is needed and it could be obtained by the conservation of energy and momentum[38]. Meanwhile, the conservation reveals that the light-induced current exists only during the momentum
SC
relaxation time, which is determined by the electron-photon coupling[39-42]. With
M AN U
the enhancing of the coupling interaction the conversion efficiency of the momentum is higher. In our experiment, as analyzed and mentioned above (Fig. 6(a)), the strongest amplitude of emitted THz electric field from VAMCNT film is obtained in the reflection direction of pump light when the electric field of pump light is
TE D
perpendicular to the CNT axis, and the transient photocurrent induced by PDE reaches the maximum value. L.V. Titova et al.[28] reported the similar dependence behavior of emitted THz amplitude from horizontally aligned CNT film on the
EP
electric field of pump light in the transmission direction of pump light. They
AC C
considered that this behavior was caused by the anisotropy of optical absorption and reflectance of aligned CNT films. However, the inner carrier transport mechanism of CNT array samples was not involved. It is worth notifying that the free carriers not only could transport along single CNT but also may transport between the adjacent CNTs (due to the interaction between CNTs), and both aspects could form the transient photocurrent induced by PDE along the corresponding directions, which further lead to the generation of polarized THz radiation. Therefore, the amplitude of
ACCEPTED MANUSCRIPT emitted THz electric field from aligned CNT films should be determined by the
TE D
M AN U
SC
RI PT
carrier transport direction and the electric field of pump light.
Figure 7. SEM images of the VAMCNT composite films at the fractured edge of the sample under the different magnifications.
EP
In order to further elucidate the mechanism of THz generation from VAMCNT films, we prepared VAMCNT composite films by in situ coating epoxy resin polymer.
AC C
As illustrated by Figure 7, each tube of the VAMCNT is completely coated and the array orientation of the VAMCNT is well kept. The tube diameter is in the range of 100-120 nm, which is larger than that of pure VAMCNTs due to the polymer coating. The polymer-coated method can efficiently isolate the interaction between the adjacent CNTs (that is to say, isolate the carrier transport across neighboring CNTs), and hence THz generation from VAMCNT composite films should be only related to the carrier transport along the tube axis. For comparing to the pure VAMCNT films,
ACCEPTED MANUSCRIPT the VAMCNT composite films is also excited by laser pulse to detect THz generation with different sample angles, pump polarization angles and pump powers (the varied parameters are consistent with those in pure VAMCNT films), respectively. The
RI PT
evolutions of the emitted THz pulses from VAMCNT composite films with the sample angle and pump power are similar to the results of the pure VAMCNT films, except that the amplitude of the emitted THz electric field decreased slightly
SC
compared with pure VAMCNT films (Figure S2 and Figure S3 in Supporting
M AN U
Information). However, when changing the pump polarization angle, the evolution of the emitted THz pulse from VAMCNT composite films is different from that of the
AC C
EP
TE D
pure VAMCNT films, as shown in Figure 8.
Figure 8. (a) The time domain waveform of the emitted THz signal from VAMCNT composite films with different pump polarization angles. (b)Emitted THz electric field intensity from VAMCNT composite films vs pump polarization angle. The comparison of the emitted THz waveform from pure VAMCNT films and VAMCNT composite films: (c) θpump = 90o; (d) θpump =180o.
ACCEPTED MANUSCRIPT Figure 8(a) shows THz radiations from VAMCNT composite films with different pump polarization angles ranged from 0o to 180o. The higher THz pulse intensity is obtained when the θpump is 0o or 180o, while the emitted THz pulse intensity is
RI PT
minimum for θpump = 90°. In addition, the intensity of THz electric field emitted from VAMCNT composite films as a function of the pump polarization angle θpump (=2φ) shows cos2φ dependence behavior (Figure 8(b)).The results above are totally different
SC
from those of pure VAMCNT films. As illustrated in Figure 8(c), for θpump= 90o, the
M AN U
emitted THz signal from pure VAMCNT films is very strong, while the emitted THz signal from polymer-coated VAMCNT films nearly disappears. For θpump= 180o, the amplitude of the emitted THz signal from pure VAMCNT films is almost same as that from polymer-coated VAMCNT films (Figure 8(d)). The results further indicate that
TE D
the free carriers in VAMCNT films could transport both along single CNT and between the adjacent CNTs (Figure 9(a)), while free carriers in VAMCNT composite films transport mainly along the tube axis due to the isolated carrier transport between
EP
the adjacent CNTs by the coated polymer (Figure 9(b)). Therefore, for θpump= 90o (the
AC C
angle between the electric field of pump light and the CNT axis is also 90o, as shown in Table 1), the PDE-induced transient photocurrent in pure VAMCNT films transports only perpendicular to the tube axis, which leads to THz wave generation in the corresponding direction. While pure VAMCNTs are coated by epoxy resin polymer to form VAMCNT composite films, the only path of carrier transport perpendicular to the tube axis is interdicted for the same pump polarization angle (90o), thus the transient photocurrent for generating THz waves will not appear. For
ACCEPTED MANUSCRIPT the VAMCNT composite films, regardless of pump polarization angle, the carrier transport is along the tube axis due to the isolated carrier transport between the adjacent CNTs. When the pump polarization angle changes from 0o to 60o (the
RI PT
corresponding angle between the electric field of pump light and the CNT axis varies from 135o to 111o, see Table 1), the component of the pump light electric field along the tube axis decreases gradually (although the component of the pump light electric
SC
field perpendicular to the tube axis increases gradually, this component does not
M AN U
contribute to THz generation from VAMCNT composite films), which result in the weakened intensity of the emitted THz electric field from VAMCNT composite films,
AC C
EP
TE D
as shown in Figure S4 in Supporting Information.
Figure 9. Schematic diagram of the carrier transport paths in (a) pure VAMCNT films and (b) VAMCNT composite films.
In addition, we find that THz response of VAMCNT composite films is more
coherent than that of pure VAMCNT films. The intensity of the emitted THz electric field from VAMCNT composite films with different sample angles is more uniform than that from pure VAMCNT films (Figure S2). The linear dependence of the
ACCEPTED MANUSCRIPT emitted THz electric field on the pump power for VAMCNT composite films is better than that for pure VAMCNT films (Figure S3(c)). The intensity of the emitted THz electric field from VAMCNT composite films decreases with the pump polarization
RI PT
angle changed from 0o to 90o, while the evolution of the emitted THz electric field from VAMCNT films with the pump polarization angle is irregular. The results above indicate that coating non-conductive polymer on CNT array could effectively
SC
suppress the interaction of the adjacent nanotubes and make THz response of CNT
M AN U
array composites more foreseeable (decided only by the structure parameters of CNT itself), because the interaction of the adjacent nanotubes is complex and determined by many factors (besides the structure parameters of CNT itself, relative position of the adjacent CNTs, the irregular bridge-linkage between CNTs, etc.). Therefore, THz
TE D
devices based on VAMCNT composites will be more stable and enable more regular response properties than those based on pure VAMCNTs. 4. Conclusion
EP
The generation of broadband (0.1-2.3 THz) THz waves has been observed in
AC C
VAMCNT materials excited with 800 nm laser pulse in reflection configuration. The intensity and waveform of emitted THz pulses from pure VAMCNT films with different sample angles have a slight difference due to the nonuniformity of the VAMCNT sample. THz electric field is linearly changed with the pump power, which is a typical second-order nonlinear optical effect for THz generation process. With pump polarization angles changed from 0o to 180o, the maximum THz pulse intensity from pure VAMCNT films is obtained for θpump=90°, and the intensity of emitted THz
ACCEPTED MANUSCRIPT electric field shows sin2φ dependence behavior. The results indicate that PDE is suitable for being employed to explain THz generation from the VAMCNT film. The polymer-coated VAMCNT composites are prepared and excited by laser pulse on the
RI PT
same conditions as the pure VAMCNT films to comparatively study the THz generation mechanism. The higher THz pulse intensity from polymer-coated VAMCNT films is obtained when the θpump is 0o or 180o, while the emitted THz
SC
signal nearly disappears for θpump= 90o. The totally different results from pure
M AN U
VAMCNT films indicate that the free carriers in VAMCNT films could transport both along single CNT and between the adjacent CNTs, while free carriers in VAMCNT composite films transport mainly along the tube axis due to the isolated carrier transport between the adjacent CNTs by the coated polymer. Thus, the transient
TE D
photocurrent for generating THz pulses from VAMCNT composite films is only along the tube axis and its intensity depends on the component of the pump light electric field along the tube axis. In addition, we find that THz response of VAMCNT
EP
composite films is more coherent than that of pure VAMCNT films. Such
AC C
polymer-coated VAMCNT composites have great potentials in improving the stability and response regularity of THz devices. Acknowledgements
This work was financially supported by the National Natural Science Foundation of
China (No. 11304249, 61605160, 11774288), Key Science and Technology Innovation Team Project of Natural Science Foundation of Shaanxi Province (2017KCT-01), and the International Cooperative Program (NO. 201410780).
ACCEPTED MANUSCRIPT References [1] Xinlong Xu, Ken Chuang, Robin J. Nicholas, Michael B. Johnston, Laura M. Herz. Terahertz Excitonic Response of Isolated Single-Walled Carbon Nanotubes. Journal of Physical Chemistry C 2009, 113(42): 18106-9.
RI PT
[2] Riccardo Degl Innocenti, David S. Jessop, Christian W. O. Sol, Long Xiao, Stephen J. Kindness, Hungyen Lin, et al. Fast Modulation of Terahertz Quantum Cascade Lasers Using Graphene Loaded Plasmonic Antennas. Acs Photonics 2016,
SC
3(3): 467-70.
[3] Matthew C. Beard, Jeffrey L. Blackburn, Michael J. Heben. Photogenerated free
Nano Letters 2008, 8(12): 4238-42.
M AN U
carrier dynamics in metal and semiconductor single-walled carbon nanotube films.
[4] Rana, F. Graphene Terahertz Plasmon Oscillators. IEEE Transactions on Nanotechnology 2008, 7(1): 91-9.
TE D
[5] Q. Zhang, E. H. H´aroz, Z. Jin, L. Ren, X. Wang, R. S. Arvidson, et al. Plasmonic Nature of the Terahertz Conductivity Peak in Single-Wall Carbon Nanotubes. Nano Letters 2013, 13(12): 5991-6.
EP
[6] Yixuan Zhou, Yiwen E, Lipeng Zhu, Mei Qi, Xinlong Xu, Jintao Bai, et al.
AC C
Terahertz wave reflection impedance matching properties of graphene layers at oblique incidence. Carbon 2015, 96(10): 1129-37. [7] M. J. Paul, N. A. Kuhta, J. L. Tomaino, A. D. Jameson, L. P. Maizy, T. Sharf, et al. Terahertz transmission ellipsometry of vertically aligned multi-walled carbon nanotubes. Applied Physics Letters 2012, 101(11): 111107-10. [8] X. L. Xu, P. Parkinson, K.-C. Chuang, M. B. Johnston, R. J. Nicholas, L. M. Herz. Dynamic terahertz polarization in single-walled carbon nanotubes. Physical Review B Condensed Matter 2010, 82(8): 085441-4.
ACCEPTED MANUSCRIPT [9] Leonhard Prechtel, Li Song, Dieter Schuh, Pulickel Ajayan, Werner Wegscheider, Alexander W. Holleitner. Time-resolved ultrafast photocurrents and terahertz generation in freely suspended graphene. Nature Communications 2012, 3(48): 19596-600.
RI PT
[10] J. Maysonnave, S. Huppert, F. Wang, S. Maero, C. Berger, W. de Heer, et al. Terahertz Generation by Dynamical Photon Drag Effect in Graphene Excited by Femtosecond Optical Pulses. NANO LETTERS 2015, 14:5797-802.
SC
[11] Obraztsov Petr A, Natsuki Kanda, Kuniaki Konishi, Makoto Kuwata-Gonokami, Sergey V. Garnov, Alexander N, et al. Photon-drag-induced terahertz emission from
M AN U
graphene. Physical Review B 2014, 90(24): 241416-20.
[12] Young-Mi Bahk, Gopakumar Ramakrishnan, Jongho Choi, Hyelynn Song, Geunchang Choi, Yong Hyup Kim, et al. Plasmon Enhanced Terahertz Emission from Single Layer Graphene. Acs Nano 2014, 8(9): 9089-96.
TE D
[13] L. Zhu, Y. Huang, Z. Yao, B. Quan, L. Zhang, J. Li, et al, Enhanced polarization-sensitive terahertz emission from vertically grown graphene by a dynamical photon drag effect. Nanoscale 2017, 9(29): 10301-11.
EP
[14] Xiaowei He, Naoki Fujimura, J. Meagan Lloyd, Kristopher J. Erickson, A. Alec
AC C
Talin, Qi Zhang, et al. Carbon nanotube terahertz detector. Nano Letters 2014, 14(7): 3953-8.
[15] Callum J. Docherty, Samuel D. Stranks, Severin N. Habisreutinger, Hannah J. Joyce, Laura M. Herz, Robin J. Nicholas, et al. An ultrafast carbon nanotube terahertz polarisation modulator. Journal of Applied Physics 2014, 115(20): 203108-10. [16] Jisoo Kyoung, Eui Yun Jang, Marcio D. Lima, Hyeong-Ryeol Park, Raquel Ovalle Robles, Xavier Lepr, et al. A Reel-Wound Carbon Nanotube Polarizer for Terahertz Frequencies. Nano Letters 2011, 11(10): 4227-31.
ACCEPTED MANUSCRIPT [17] Lei Ren, Cary L. Pint, Layla G. Booshehri, William D. Rice, Xiangfeng Wang, David J. Hilton, et al. Carbon Nanotube Terahertz Polarizer. Nano Letters 2009, 9(7): 2610-2. [18] Lei Ren, Cary L. Pint, Takashi Arikawa, Kei Takeya, Iwao Kawayama,
RI PT
Masayoshi Tonouchi, et al. Broadband Terahertz Polarizers with Ideal Performance Based on Aligned Carbon Nanotube Stacks. Nano Letters 2012, 12(2): 787-90.
[19] Ahmed Zubair, Dmitri E. Tsentalovich, Colin C. Young, Martin S. Heimbeck,
Applied Physics Letters 2016, 108(14): 141107-10.
SC
Henry O. Everitt, Matteo Pasquali, et al. Carbon nanotube fiber terahertz polarizer.
M AN U
[20] Yixuan Zhou, Yiwen E, Xinlong Xu, Weilong Li, Huan Wang, Lipeng Zhu, et al. Angular dependent anisotropic terahertz response of vertically aligned multi-walled carbon nanotube arrays with spatial dispersion. Scientific Reports 2016, 6: 38515-23. [21] Ruili Wu, Weilong Li, Yun Wan, Zhaoyu Ren, Xinlong Xu, Yixuan Zhou.
TE D
Anisotropic terahertz response of stretch-aligned composite films based on carbon nanotube-SiC hybrid structures. Rsc Advances 2015, 5(34): 26985-90. [22] Kibis O.V., Rosenau da Costa M., Portnoi M.E. Generation of Terahertz
EP
Radiation by Hot Electrons in Carbon Nanotubes. Nano Letters 2007, 7(11): 3414-20.
AC C
[23] Portnoi M.E, Kibis O.V, Rosenau da Costa M. Terahertz Applications of Carbon Nanotubes. Superlattices & Microstructures 2008, 43(5): 399-407. [24] Parashar J, H. Sharma. Optical rectification in a carbon nanotube array and terahertz
radiation
generation.
Physica
E
Low-dimensional
Systems
and
Nanostructures 2012, 44(10): 2069-71. [25] Jain S, J. Parashar, R. Kurchania. Effect of magnetic field on terahertz generation via laser interaction with a carbon nanotube array. International Nano Letters 2013, 3(1): 1-4.
ACCEPTED MANUSCRIPT [26] N.R Sadykov, A.V. Aporoski, D.A. Peshkov. Terahertz radiation generation process in the medium based on the array of the noninteracting nanotubes. Optical and Quantum Electronics 2016, 48(7): 1-10. [27] Muthee M, E. Carrion, J. Nicholson, S.K. Yngvesson. Antenna-coupled terahertz
RI PT
radiation from joule-heated single-wall carbon nanotubes. Aip Advances 2011, 1(4): 042131-6.
[28] Lyubov V. Titova, Cary L. Pint, Qi Zhang, Robert H. Hauge, Junichiro Kono,
SC
Frank A. Hegmann. Generation of Terahertz Radiation by Optical Excitation of Aligned Carbon Nanotubes. Nano Letters 2015, 15(5): 3267-72.
M AN U
[29] Weilong Li, Jinkai Yuan, YouqinLin, ShenghongYao, ZhaoyuRen, HuiWang, et al. The controlled formation of hybrid structures of multi-walled carbon nanotubes on SiC plate-like particles and their synergetic effect as a filler in poly(vinylidene fluoride) based composites. Carbon, 2013, 51(1): 355-64.
graphene
carbon
TE D
[30] Lin J , Zhang C , Yan Z , Zhu Y , Peng Z , Hauge RH , et al. 3-Dimensional nanotube
carpet-based
microsupercapacitors
with
high
electrochemical performance. Nano Letters 2012, 13(1): 72-8.
EP
[31] Johannes Vanpaemel, Marleen H. van der Veen, Daire J. Cott, Masahito
AC C
Sugiura, Inge Asselberghs, Stefan De Gendt, et al. Dual Role of Hydrogen in Low Temperature Plasma Enhanced Carbon Nanotube Growth. Journal of Physical Chemistry C 2015, 119(32): 18293-302 [32] Gennady M. Mikheev, Albert G. Nasibulin, Ruslan G. Zonov, Antti Kaskela, Esko I. Kauppinen. Photon-drag effect in single-walled carbon nanotube films. Nano Letters 2012, 12(1): 77-83. [33] K Liu, J Xu, T Yuan, XC Zhang. Terahertz radiation from InAs induced by carrier diffusion and drift. Physical Review B 2006, 73(15): 155330-5.
ACCEPTED MANUSCRIPT [34] J. Maysonnave, S. Huppert, F. Wang, S. Maero, C. Berger, W. de Heer, et al. Room temperature broadband coherent terahertz emission induced by dynamical photon drag in graphene. Nano Letters 2014, 14: 5797-813. [35] Obraztsov P.A, Gennady M. Mikheev, Sergei V. Garnov, Alexander N.
RI PT
Obraztsov,Yuri P. Svirko. Polarization-sensitive photoresponse of nanographite. Applied Physics Letters 2011, 98(9): 091903-5.
Physica status solidi (b) 2012. 249(12): 2538-43.
SC
[36] E. L. Ivchenko. Photoinduced currents in graphene and carbon nanotubes.
M AN U
[37] M.M. Glazov, S.D. Ganichev. High frequency electric field induced nonlinear effects in graphene. Physics Reports 2014. 535(3): 101-38.
[38] Ryan W. Newson, Jean-Michel Ménard, Christian Sames, Markus Betz, Henry M. van Driel. Coherently Controlled Ballistic Charge Currents Injected in Single-Walled
TE D
Carbon Nanotubes and Graphite. Nano Letters 2008, 8(6): 1586-9. [39] Obraztsov P.A, Tommi Kaplas, Sergey V. Garnov, Makoto Kuwata-Gonokami, Alexander N. Obraztsov, Yuri P. Svirko. All-optical control of ultrafast photocurrents
EP
in unbiased graphene. Scientific Reports 2014, 4(2): 4007-12. [40] Marchetti S, Simili R, Bernardini M, M Giorgi. Detection of FIR radiation by
AC C
photon drag of free carriers in semimetal thin film. Physica Scripta 1988, 37(5): 820-2.
[41] Alexander N. Obraztsov, Dmitry A. Lyashenko, Shaoli Fang, Ray H. Baughman, Petr A. Obraztsov, Sergei V. Garnov, et al. Photon drag effect in carbon nanotube yarns. Applied Physics Letters 2009, 94(23): 231112-4. [42] Pfeiffer R, Kuzmany H, Simon F, Bokova S. N, Obraztsova E. Resonance Raman scattering from phonon overtones in double-wall carbon nanotubes. Physical Review B Condensed Matter 2005, 71(15): 155409-16.