Transparent multi-walled carbon nanotube-silica composite prepared by hot-pressed sintering and its nonlinear optical properties

Article abstract of DOI:10.1016/j.jallcom.2009.02.153

Transparent multi-walled carbon nanotube (MWNT)-fused silica bulk composites were prepared by the hot-pressing technique. The results of transmission electron microscopy (TEM) proved that MWNTs were relatively well dispersed in the fused silica matrix. Th

Full text of DOI:10.1016/j.jallcom.2009.02.153

Journal of Alloys and Compounds 481 (2009) L4–L7
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Letter
Transparent multi-walled carbon nanotube–silica composite prepared by
hot-pressed sintering and its nonlinear optical properties
Heng Xua,b, Yubai Pana,∗, Yong Zhua,b, Huamin Koua, Jingkun Guoa
a State Key Lab of High Performance Ceramics and Superfine Microstructures, Shanghai Institute of Ceramics, Chinese Academy of Sciences,
1295 Ding Xi Road, Shanghai 200050, PR China
b Graduate School of the Chinese Academy of Sciences, 19A Yuquan Road, Beijing 100039, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Transparent multi-walled carbon nanotube (MWNT)-fused silica bulk composites were prepared by the
hot-pressing technique. The results of transmission electron microscopy (TEM) proved that MWNTs were
relatively well dispersed in the fused silica matrix. The linear transmittance of the composites reached
70% in the visible light region and about 76% near 1100 nm. The z-scan method was used to investigate
the nonlinear optical properties of the composites, which exhibited optical limiting properties at 800 nm
at the incident power of 30 mW and 50 mW. The main mechanism of the optical limiting performances
of the composites was deduced to be nonlinear absorption. The composites may be a good candidate for
optical limiting application.
Received 12 January 2009
Received in revised form 24 February 2009
Accepted 25 February 2009
Available online 17 March 2009
Keywords:
Amorphous materials
Sol–gel synthesis
Nonlinear optics
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
have observed strong and broad-band optical limiting in single-
walled and multi-walled nanotube suspensions [10,11] or solutions
Nowadays, laser sources are widely used in the fields of indus-
try, medicine, biology and military. The development of laser has
brought us many benefits; however, it has also brought potential
hazards for human eyes and optical sensors. Therefore, increasing
interests have been inspired in preparing materials with strong
optical limiting properties. This kind of material can effectively
attenuate intense laser beam and possesses high transmittance of
low-intensity ambient light. So, it can not only protect human eyes
and optical sensors from the high energy laser beam, but also make
sure the reception of normal signals. Different kinds of materials
such as fullerenes [1,2], organometallics [3,4], carbon black sus-
pensions (CBS) [5,6], semiconductors and liquid crystals, have been
investigated and selected as candidates for good optical limiters in
the visible domain for short laser pulse.
Carbon nanotubes (CNTs) have been the focus of extensive physi-
cal studies concerning their fascinating electronic, mechanical and
other properties [7,8]. Moreover, theoretical calculations [9] have
shown that CNTs with large second hyperpolarizabilities ꢀ (also
called third-order optical nonlinear coefficient) correlate with the
high optical limiting properties. Furthermore, many researchers
[12,13], as well as in organic films [11]. For example, Wang et al.
[14] have investigated the optical limiting performances of the
aqueous dispersions of multi-walled carbon nanotubols and [C60]
fullerols using the z-scan measurements at 532. All the carbon
nanotube materials, especially multi-walled carbon nanotubols,
showed much better optical limiting performances than [C60]
fullerols and C60. But most of these investigations were carried
out in suspensions or solutions, which brought many difficulties
in practical use. Zhan et al. [15,16] prepared MWNT–silica xerogel
composite and found its optical limiting properties were better than
those of MWNT suspension. However, this xerogel without sinter-
ing lacked strength and stability for practical use. Furthermore, it
required a long aging procedure of 2 months. Considering the above
reasons, it is urgent and also interesting to prepare CNT-ceramic
bulk material with improved strength and stability for optical lim-
iter.
An ideal optical limiter should have broad-band optical limit-
ing efficiency over the whole visible spectrum. On the other hand,
it should also possess high linear transmittance at low incident
energy and neutral colorimetry, to ensure the acceptance of sig-
nals for eyes and optical sensors. Silica (SiO2) is a widely used
optical material, which has a high transmittance (more than 90%)
and environmental stability (such as resistance to creep, excessive
hardness and anti-abrasion properties, high resistance to oxidation
and chemical attack). So it is suitable for the matrix of optical lim-
iter. In addition, SiO2 composites with varied contents of CNTs have
been prepared successfully and the composites possess improved
∗
Corresponding author at: State Key Lab of High Performance Ceramics and
Superfine Microstructures, Shanghai Institute of Ceramics, Chinese Academy of Sci-
ences, 1295 Ding Xi Road, Shanghai 200050, PR China. Tel.: +86 21 52412820;
fax: +86 21 52413903.
E-mail address: ybpan@mail.sic.ac.cn (Y.B. Pan).
0925-8388/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2009.02.153

H. Xu et al. / Journal of Alloys and Compounds 481 (2009) L4–L7
L5
Fig. 1. TEM images of (a) pristine MWNTs and (b) MWNTs coated with SiO2.
mechanical properties [17,18] and excellent microwave attenuation
properties [19]. But to the best of our knowledge, there is still no
report about the optical limiting properties of the CNT/SiO2 com-
posite prepared by hot-pressed sintering.
In this paper, we described the preparation of MWNT/SiO2
powder and the hot-pressing process of the bulk MWNT/SiO2 com-
posite, and then discussed the microstructures and optical limiting
properties.
(Fig. 1(b)), which is confirmed as amorphous in the XRD analysis
(not shown).
In an ideal optical limiter, the linear transmittance at low inci-
dent energy must be reasonable, at least of 70% [23]. Fused SiO2
has high linear transmittance, but many researches show that,
crystallization in fused SiO2 will ruin the transparency and other
properties. SiO2 powders obtained from melted quartz may retain
some crystalline structures as nucleation centers, which is bene-
ficial for crystallization. So it is unsuitable for the preparation of
the transparent material. The number of the O–H groups in sol–gel
SiO2 powders is greater than that of in SiO2 powders obtained
by other methods. The O–H groups could effectively depress the
crystallization of SiO2 during the hot-pressed sintering [24], and
then benefit of the high linear transmittance of the material. Fur-
thermore, the diameters of the sol–gel SiO2 powders are around
nanometer, which would help to reduce the sintering temperature.
Another challenge comes from the contamination of the graphite
die. Under higher temperature, more carbon atoms in the graphite
die are intent to diffuse into the composite during the sintering
procedure. For the above reasons, we chose the sol–gel SiO2 pow-
der for sintering. The composite was sintered at 1250 ◦C, which
was 50 ◦C lower than the sintering temperature we usually used
[17,19]. Finally, we got the transparent bulk material of MWNT/SiO2
2. Experimental procedures
MWNTs with diameters from 20 nm to 40 nm and lengths from 5 ␮m to 15 ␮m
were provided by Shenzhen NANO Tech. Port. Co. Ltd. of China. In order to achieve
better performance of the composite, the received MWNTs were coated with SiO2
by a sol–gel method as described in Ref. [20]. The modified MWNTs were marked as
SiO2@MWNT. Pure SiO2 powder was prepared by a rapid sol–gel method. The molar
ratio of tetraethyl orthosilicate (TEOS):ethanol:distilled water in the precursor was
1:4:8. The hot-pressed sintering technique was used to obtain bulk material.
The experiments were conducted as follows. The SiO2@MWNT and the pure
SiO2 powder prepared previously were mixed according to the designed ratio. Then
the mixture was ball-milled with absolute alcohol for 12 h, and using agate balls as
the grinding media. After been dried and sieved through a screen, the MWNT/SiO2
powder with a MWNT content of 0.02 wt% was obtained. Finally, the MWNT/SiO2
powder was added into a graphite die and hot-pressed at 1250 ◦C under an applied
stress of 30 MPa in N2 atmosphere for 0.5 h. The sintered composites were cut into
30 mm × 20 mm × 0.1 mm, and both sides of the composites were polished to fit for
the transmittance and optical limiting properties tests.
Transmission electron microscopy (TEM) (JEOL JEM 2100F FETEM) was used to
observe the microstructures of SiO2@ MWNT powder and the bulk composites. The
transmittance test was carried out on a Model U-2800 Spectrophotometer, Hitachi,
Japan. The z-scan experimental setup was based on a Spectra-Physics mode-locked
Ti:sapphire laser which produces laser pulses of about 100 fs duration with 82 MHz
repetition rate at 800 nm wavelength. The measurements were performed using TZ-
scan [21,22] techniques with a 10 cm focal-length objective lens and the beam waist
radius at the focus was about 10 ␮m.
3. Results and discussion
MWNT/SiO2 composite, which combines MWNT and transpar-
ent silica matrix, is expected to have excellent broad-band optical
limiting properties. The critical challenge, however, is how to
enhance the dispersion and alignment of MWNTs in the matrix.
Therefore, it is necessary to coat MWNT with silica, which could
eliminate the undesirable attractive interactions between the nan-
otubes and improve the interfacial bonding with matrix [20]. Fig. 1
shows the TEM images of the MWNTs without modifying and
the MWNTs coated with SiO2. The pristine MWNTs are curved
and twisted with each other (Fig. 1(a)). After coating process,
MWNTs have been coated with a uniform SiO2 layer about 5–10 nm
Fig. 2. Optical transmission spectrum of the MWNTs/SiO2 composite.

L6
H. Xu et al. / Journal of Alloys and Compounds 481 (2009) L4–L7
Fig. 3. The z-scan experimental apparatus in which the ratio D2/D1 is recorded as a
function of the sample position z.
by the hot-pressing method. Fig. 2 shows the optical transmission
spectrum of the composite from 200 nm to 1100 nm. The transmit-
tance reaches 70% in the visible light region and about 76% near
1100 nm. The linear optical transmittance of the composite meets
the requirement of an ideal optical limiter, which ensures the qual-
ity of observation/detection. And its transmittance is much better
than the previous reported MWNT-silica xerogel composite [15],
which is 55% at 532 nm and about 60% at 800 nm.
We performed a z-scan experiment to determine the optical lim-
iting properties of the sample. The z-scan technique is a sensitive
and simple characterization method of the intensity-dependent
optical properties of materials. When the measurement begins, a
sample is moved along the axis of propagation (z) of a focused
Gaussian beam through its focal plane. The input and output ener-
gies are simultaneously measured by two high sensitive detector
(D1 and D2) for each z position. When all the transmitted light
is detected, open-aperture z-scan provides information about the
nonlinear absorption of the sample. The detailed experiment set-up
is shown in Fig. 3.
Fig. 4 displays the open-aperture z-scan results, using the 100 fs,
800 nm laser pulses. The laser beam waist radius at the focus is
about 10 ␮m. The hollow circle and filled circle are the data mea-
sured at incident power of 30 mW and 50 mW, respectively. The
corresponding incident laser intensities at the focus are about
0.88 GW/cm2 and 1.47 GW/cm2. As the sample moves toward the
focus, the normalized transmittance decreases from 100% to 83%
and 73%, respectively. The pronounced valleys of transmission
curves around Z = 0 are the signatures of either nonlinear absorption
or nonlinear scattering. Nonlinear refraction is excluded, because
there is no pronounced signal in the close-aperture z-scan experi-
Fig. 4. Open-aperture z-scan measurement on the MWNTs/SiO2 composite, by using
the 100 fs, 800 nm laser pulses, with a power: (a) 30 mW and (b) 50 mW.
ments. Furthermore, it is regarded that the dominant contribution
to the observed transmission valley in Fig. 4 is from nonlinear
absorption because there is no light emitted from the sample dur-
ing the measured process. Moreover, the 50 mW curve has deeper
valley than 30 mW, which indicates that MWNTs have stronger non-
linear effect under higher incident intensity. In fact, it has been
previously reported that optical limiting behavior in the MWNT
suspension was attributed to the nonlinear scattering arising from
expanding microplasmas as in the carbon black suspension [25].
And contributions from other mechanisms such as self-defocusing
and thermal lensing effect have also been suggested [10]. Accord-
ing to our work, it seems that the optical limiting mechanism of
MWNT in the solid is different from that of MWNT in the suspen-
sion, which is consistent with Ref. [15]. Further research will be
necessary to identify all the physical mechanisms involved in the
optical limiting properties of the composite.
After the z-scan measurement, the composite was characterized
by TEM to confirm whether MWNTs were damaged or not. Fig. 5 dis-
plays the TEM image of MWNTs in the fused SiO2 matrix, (a) and
(b) are low-magnification image and high-resolution image, respec-
tively. It is found that MWNTs are maintained after sintering and
z-scan experiment. We can clearly see the layers of MWNTs from
Fig. 5(b), which proves that the laser beam irradiation in the z-scan
Fig. 5. TEM images of MWNTs in the silica matrix after laser beam irradiation: (a) low-magnification image and (b) high-resolution image.

H. Xu et al. / Journal of Alloys and Compounds 481 (2009) L4–L7
L7
experiment did not destroy the structure of MWNTs. The result con-
Acknowledgments
firms the previously deduction of nonlinear absorption mechanism,
because nonlinear scattering in the suspension will lead to sublima-
tion of carbon nanotubes. The nondestructive nonlinear absorption
mechanism of the MWNTs/SiO2 bulk material makes it recyclable,
and this turns out to be an advantage for practical use.
The authors are grateful for the financial supports from the
National Natural Science Foundation of China (No. 50702067) and
the Major Basic Research Programs of Shanghai (No. 07DJ14002).
As an optical limiter, the material should have not only optical
limiting properties and high linear transmittance, but also envi-
ronmental stability, especially under severe conditions. When the
laser beam is induced, energy absorbed by CNT will transform into
heat energy, leading to a high temperature in the matrix. The CNT
suspension is not stable, and polymer matrix cannot withstand the
high temperature of the incident laser. Silica is a ceramic mate-
rial, with good room and high temperature mechanical strength
and resistance to creep, excessive hardness and anti-abrasion prop-
erties, high resistance to oxidation and chemical attack, etc. All
of these factors make it suitable as the matrix for optical limiter.
Therefore, MWNT/SiO2 bulk materials prepared by hot-pressed sin-
tering might be a competitive candidate for practical use of optical
limiting.
References
[1] M.P. Joshi, S.R. Mishra, H.S. Rawat, S.C. Mehendale, K.C. Rustagi, Appl. Phys. Lett.
62 (15) (1993) 1763–1765.
[2] M. Meneghetti, R. Signorini, S. Sartori, R. Bozio, M. Maggini, G. Scorrano, M.
Prato, G. Brusatin, M. Guglielmi, Synth. Met. 103 (1999) 2474–2475.
[3] J. Lang, Z. Sun, Q. Xu, H. Yu, K. Tatsumi, Mater. Chem. Phys. 82 (2003) 493–498.
[4] M. Garcia, P. Florindo, M. Piedade, M. Duarte, M. Robalo, J. Heck, C. Wittenburg,
J. Holtmann, E. Licandro, J. Organomet. Chem. 693 (2008) 2987–2999.
[5] K.M. Nashold, D.P. Walter, J. Opt. Soc. Am. B 12 (1995) 1228–1237.
[6] R. Goedert, R. Becker, A. Clements, T. Whittaker III, J. Opt. Soc. Am. B 15 (1998)
1442–1462.
[7] K. Mohan Kumar, V. Kripesh, Andrew A.O. Tay, J. Alloy Compd. 450 (2008)
229–237.
[8] J.P. Cheng, X.B. Zhang, G.F. Yi, Y. Ye, M.S. Xia, J. Alloy Compd. 455 (2008) 5–9.
[9] R.H. Xie, J. Jiang, Appl. Phys. Lett. 71 (1997) 1029–1031.
[10] L. Vivien, E. Anglaret, D. Riehl, F. Bacou, C. Jurnet, C. Goze, M. Andrieux, M.
Brunet, F. Lafonta, P. Bernier, F. Hache, Chem. Phys. Lett. 307 (1999) 317–319.
[11] X. Sun, R.Q. Yu, G.Q. Xu, T.S.A. Hor, W. Ji, Appl. Phys. Lett. 73 (1998)
3632–3634.
4. Conclusions
[12] L. Liu, S. Zhang, T. Hu, Z. Guo, C. Ye, L. Dai, D. Zhu, Chem. Phys. Lett. 359 (2002)
191–195.
MWNT/SiO2 bulk materials were prepared by hot-pressing
method and its optical limiting properties were investigated. The
transmittance of the composite reached 70% in the visible light
region and about 76% near 1100 nm, which met the requirements
of an ideal optical limiter (>70%). In the z-scan experiments, the
composite showed optical limiting properties in different incident
intensity, and the main mechanism was deduced to be nonlinear
absorption. TEM proved that the hot-pressed sintering and the laser
beam irradiation did not destroy the MWNTs in the matrix. The non-
destructive nonlinear absorption mechanism of the MWNTs/SiO2
bulk material makes it recyclable, and this turns out to be an
advantage for practical use. Further work will be focused on the
improvement of MWNT dispersion and the resultant optical limit-
ing properties of the composites.
[13] H. Wu, R. Tong, X. Qiu, H. Yang, Y. Lin, R. Cai, S. Qian, Carbon 45 (2007) 152–159.
[14] Q. Wang, Y. Qin, Y. Zhu, X. Huang, Y. Tian, P. Zhang, Z. Guo, Y. Wang, Chem. Phys.
Lett. 457 (2008) 159–162.
[15] H. Zhan, W. Chen, M. Wang, C. Zheng, C. Zou, Chem. Phys. Lett. 382 (2003)
313–317.
[16] H. Zhan, C. Zheng, W. Chen, M. Wang, Chem. Phys. Lett. 411 (2005) 373–377.
[17] J. Ning, J. Zhang, Y. Pan, J. Guo, J. Mater. Sci. Lett. 22 (2003) 1019–1021.
[18] J. Ning, J. Zhang, Y. Pan, J. Guo, Ceram. Int. 30 (2004) 63–67.
[19] C. Xiang, Y. Pan, X. Liu, X. Sun, X. Shi, J. Guo, Appl. Phys. Lett. 87 (2005) 123103.
[20] Y. Zhu, Y. Pan, H. Xu, C. Xiang, Chem. Lett. 36 (2007) 1098–1099.
[21] M. Sheik-Bahae, A.A. Said, T.H. Wei, D.J. Hagan, E.W. Van Stryland, IEEE J. Quant.
Elect. 26 (1990) 760–769.
[22] M. Sheik-Bahae, A. Said, E. Van Stryland, Opt. Lett. 14 (17) (1989) 955–957.
[23] L. Vivien, P. Lancon, D. Riehl, F. Hache, E. Anglaret, Carbon 40 (2002)
1789–1797.
[24] L. Wang, B. Tsai, Mater. Lett. 43 (2000) 309–314.
[25] P. Chen, X. Wu, X. Sun, J. Lin, W. Ji, K.L. Tan, Phys. Rev. Lett. 82 (1999) 2548–2551.