Composites: Part B 53 (2013) 342–346
Contents lists available at SciVerse ScienceDirect
Composites: Part B
journal homepage: www.elsevier.com/locate/compositesb
Improving mechanical and electrical properties of oriented polymer-free
multi-walled carbon nanotube paper by spraying while winding
Wei Liu a,b,c, Haibo Zhao d, Zhenzhong Yong e, Geng Xu e, Xin Wang d, Fujun Xu a,b, David Hui f,
Yiping Qiu a,b,⇑
a
Key Laboratory of Textile Science and Technology, Ministry of Education, China
College of Textiles, Donghua University, Shanghai 201620, China
College of Fashion, Shanghai University of Engineering Science, Shanghai 201620, China
d
North Carolina State University, Raleigh, NC 27695, USA
e
Suzhou Institute of Nano-Tech and Nano-Bionics, Suzhou 215123, China
f
Department of Mechanical Engineering, University of New Orleans, New Orleans, LA 70148, USA
b
c
a r t i c l e
i n f o
Article history:
Received 22 September 2012
Received in revised form 6 May 2013
Accepted 25 May 2013
Available online 6 June 2013
Keywords:
A. Nano-structures
A. Preform
B. Mechanical Properties
B. Electrical Properties
a b s t r a c t
In this study, a new method is introduced for fabricating carbon nanotube (CNT) paper, in which the solvent is sprayed on the CNT sheet while it is wound on a rotating mandrel. As the solvent evaporated, the
capillary force pulls CNT closer together, resulting in a CNT paper with a high degree of alignment and a
high packing density. Three batches of multi-walled CNTs with different wall thicknesses, tube diameters
and lengths are utilized for synthesizing highly oriented CNT papers. It is found that CNTs with smallest
diameter of 8 nm form strongest CNT paper with a tensile strength of 563 MPa and a tensile modulus of
15 GPa, while that made with CNTs of 10 nm diameter shows the highest electrical conductivity of
5.5 104 S/m.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
Due to the unique structure and extraordinary properties, carbon nanotubes (CNTs) have drawn much attention during the last
two decades. For individual CNTs, the tensile strengths are in the
range of 10–63 GPa, and Young’s moduli between 0.27 and
1.47 TPa [1–3]. The combination of their exceptional mechanical
properties and electrical conductivity [4–6] makes CNTs very
promising materials in various applications [7,8]. For the same reason, CNT paper formed by CNT network possesses great potential
for actuator, capacitors, battery electrode and reinforcement for
composites [9–16]. However, traditional methods for fabricating
CNT paper involve dispersion and filtration of suspended tubes,
usually single-walled carbon nanotubes, accompanied with crookedness and agglomeration of CNTs in the CNT paper, limiting their
strength and electrical conductivity [17,18]. To sufficiently utilize
the unique properties of CNTs, a desirable CNT paper structure
with compacted long and aligned nanotubes is critical.
Many new approaches have been developed to fabricate the
highly oriented CNT paper by utilizing the vertically aligned CNT
⇑ Corresponding author at: College of Textiles, Donghua University, Shanghai
201620, China. Tel.: +86 21 67792744; fax: +86 21 67792627.
E-mail address: ypqiu@dhu.edu.cn (Y. Qiu).
1359-8368/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.compositesb.2013.05.043
(VACNT) array. For example, Wang et al. have used a ‘‘domino
pushing’’ method by pushing down the VACNT array to the aligned
CNT paper [19]. The electrical conductivity along the alignment of
the CNT paper is 2.0 104 S/m. Similarly, by shear pressing the
VACNT array of about 1.4 mm height, Bradford et al. made the
aligned CNT preform with tensile strength of 16 MPa and electrical
conductivity of 1.2 104 S/m parallel to the alignment of the CNTs
[14]. However, in the above methods, the size of the CNT paper is
limited by the size of the substrate supporting the VACNT during
its synthesis. Recently, anisotropic A4 sheet CNT paper were obtained by Inoue et al. through firstly winding the draw-out CNT
web (10 m/s drawing speed) from the 2.0-mm-high drawable VACNT array and then ethanol spray was applied to densify the structure afterwards [20]. This approach allows the aligned CNT in the
draw-out web transformed into the CNT paper, forming a compact
structure without disrupting the alignment. The CNT paper has a
tensile strength of 75.6 MPa and electrical conductivity of
4.0 104 S/m [20]. However, the strength of the CNT paper produced using this method is still relatively low compared with that
of the component CNTs. Since the intermolecular bonds are the
only forces holding the CNTs together, if the intermolecular attraction between the CNTs is increased, the efficiency of load transfer
among the neighboring CNTs may also be improved, leading to
an elevated tensile strength.
W. Liu et al. / Composites: Part B 53 (2013) 342–346
In this paper, a new and facile method, namely spraying while
winding (SWW), was introduced for synthesizing CNT paper. It allows a decreasing of the inter-tube distances and thus increases
attractions among the neighboring CNTs, leading to an improved
strength of the CNT paper without sacrificing its electrical conductivity. Through this method, CNT paper with tunable orientation,
size and thickness can be obtained in only one-step. Three different
VACNT arrays were used to provide forest-drawn CNT sheet as the
nanotube sources.
2. Experimental
2.1. Preparation of CNT paper
343
2.4. Determination of mechanical and electrical properties
Tensile mechanical property of the three different CNT paper
was characterized using a Shimadzu EZ-S instrument with a
100 N load cell at a strain rate of 8.3% per min. Strain was measured from grip displacement. The specimens with a thickness of
around 10 mm were prepared by cutting the CNT paper into
1.5 cm 0.5 mm pieces (gauge length 6 mm). For each CNT paper
tensile testing sample, 7 specimens were tested. The electrical conductivity was calculated based on the resistance measured along
the aligned CNT direction by a 4-probe Agilent 34410A 6.5 digit
multi-meter. The specimens were prepared the same way as those
for the tensile test. Maximum value was chose from two tested
specimens for each sample.
Three different VACNT Arrays, namely CNT-1, CNT-2 and CNT-3,
were synthesized in a chemical vapor deposition system by different catalysts. CNT-1 and CNT-2 arrays were grown using CVD from
catalyst of deposited thin iron films [21], while CNT-3 arrays were
grown by one-step chloride-mediated CVD [22]. These vertically
aligned CNTs have uniform height and can be pulled out as a layer
of CNT sheet continuously. Fig. 1 demonstrates SWW approach for
fabricating aligned CNT paper, where a CNT sheet was drawn out of
a VACNT array and wound on a rotating mandrel covered by aluminum foil. During the processing, a solution with equal volumes of
deionized water and ethanol was sprayed on the CNT sheet. The
as-stacked sheets were densified due to the capillary force caused
by the evaporation of the solvents [23]. To allow the solvents fully
evaporated before covered by the subsequent wound CNT sheet,
the CNT sheet drawing speed was kept at 18 mm/s according to
our previous study [24]. By choosing an appropriate mandrel diameter, sheet width, and number of revolutions, highly oriented CNT
papers with the desired size and thickness were produced by pealing them off the aluminum foil. In this study, the diameter of the
mandrel was 4 cm and the CNT paper width was controlled around
5 mm to lower the CNT cost.
Fig. 2 shows the TEM images of the CNT-1, CNT-2 and CNT-3
nanotubes. These tubes differed from each other greatly in their
diameter and number of walls as shown in Table 1. The CNT-1 tubes
were mainly double- and triple-walled and about 8 nm in diameter.
The CNT-2 tubes had mostly 6 walls and diameters ranging from 8 to
10 nm. The CNT-3 tubes were many-walled (about 50) and the outer
diameter was also very large, around 45 nm. CNT-1, CNT-2 and CNT3 were 200, 300 and 700 lm in lengths respectively.
Fig. 3 shows the normalized Raman spectra of CNT-1, CNT-2 and
CNT-3. As indicated by the ratio ID/IG of the D and G band intensities
of the Raman scattering spectrum which was 0.54 and 0.63 for the
CNT-1 and CNT-2 arrays, more defects were found as the number
of walls increased. It suggested that the thinner CNT-1 tubes contained fewer defects and thus were stronger. Using a different synthesizing system, CNT-3 tubes with around 50 walls showed the
least defects with the ID/IG of 0.35 though it had large tube diameters.
2.2. Microscopy analysis
3.2. Analysis of the SWW process for fabricating aligned CNT paper
Transmission electron microscopy (TEM) images of individual
CNTs were collected in a TEM machine (Model JEOL 2010F) operation at 200 kV to measure the diameter and number of walls of
CNTs. Scanning electron microscopy analysis (Model JEOL 6400F)
was used to determine the arrangement of the CNTs in the CNT
paper.
The CNT-2 sheet (Fig. 4a), drawn from vertically aligned array, is
composed of aligned CNTs by entangling and van de Waals interactions. The veil-like CNT sheet is very easy to be torn due to the loose
structure and weak interaction between the adjacent nanotubes.
After continuous winding and spraying the ethanol–water solution,
CNT sheet can be tightly compacted. As shown in Fig. 4b, the as-produced CNT-2 paper has a denser structure than the CNT-2 sheet. It
has been verified by our previous study that the contact angle between the ethanol–water droplet and the CNT sheet is substantially
small than 90° [25]. Therefore, the liquid bridge [26] would pull the
neighboring nanotube together when the liquid volume diminished
due to evaporation (Fig. 5). The high surface energy of CNTs prevented the bundle from separating again. However, the neighboring
nanotubes not only gathered into bundles but also left vacancies between them, resulting in the porosity of the CNT paper. Furthermore,
the tubes were still highly aligned and only a few of them are
crooked, which indicated the spraying processing did not disrupt
the basic orientation of the nanotubes in the CNT sheet. The compacted structure and alignment of CNTs ensured greater intermolecular attraction forces, facilitating load transfer and sharing,
potentially resulting in a stronger CNT paper without degrading
electrical properties in the CNT oriented direction.
2.3. Determination of VACNT array quality
The quality of VACNT arrays was assessed using Renishaw
Ramascope. The laser of 514 nm wavelength was focused on the
sample through a Reinshaw microscope.
Fig. 1. Schematic view of the spraying while winding process for fabricating CNT
paper. A CNT sheet is drawn out of a drawable VACNT array and continuously
wound onto a rotating mandrel on which micrometer-sized droplets of DI water/
ethanol are deposited.
3. Results and discussion
3.1. Structure of the CNTs
3.3. Comparison of the tensile properties of CNT papers
Tensile strength of CNT paper made of few-walled (2–3), several-walled (6), and many-walled (50) nanotubes were compared
344
W. Liu et al. / Composites: Part B 53 (2013) 342–346
Fig. 2. TEM image of the individual carbon nanotubes (a) CNT-1. (b) CNT-2. (c) CNT-3.
Table 1
Tube diameter, number of walls, array height and ID/IG ratio of Raman intensities of
the CNT-1, CNT-2 and CNT-3.
Diameter (nm)
Walls
Height (lm)
ID/IG
CNT-1
CNT-2
CNT-3
8
2–3
200
0.54
8–10
6
300
0.63
45
50
700
0.35
Fig. 3. Normalized Raman spectra of CNT-1, CNT-2 and CNT-3.
in Table 1. Due to the long CNTs and high orientation in the condensed structure, CNT papers made by SWW showed much better
mechanical properties than other CNT papers fabricated using filtration process [17,27,28]. The strongest CNT paper made by
CNT-1 achieved an average strength of 562.5 MPa and a modulus
of 15.3 GPa. As only the outermost layer of a CNT carries the load
before fracture, the ratio of the effective cross-section area of the
outermost wall to the total cross-section area, Xeffective, can be utilized to estimate how much area contributed to the load transfer
(assuming the aligned MWCNTs are hexagonally closely packed)
[20]. Xeffective is expressed as:
2pdD
X effective ¼ pffiffiffi
2
3ðd þ DÞ
where d is the spacing between graphite (0 0 2) planes, D is the
diameter of the CNT. As shown in Table 2, CNT-1 paper has the highest caculated Xeffective of 14%, which indicated largest load transfer
among CNTs. Similar ratios of Xeffective from 11% to 14% were obtained for CNT-2 paper. However, the mean values of strength
and modulus were only 75% and 67% of those of the CNT-1 paper,
due to the more defects on the tube structure as well as much less
effective outermost wall per unit volume. Another reason could be
that thick CNTs might have less intermolecular attraction between
neighboring nanotubes due to the increased bending stiffness
which could hinder tight packing of the tubes. For CNT-3 paper,
only 3% of the total cross-section area for each tube was effective
concerning with the mechanical properties. Therefore, the tested
strength and modulus were 117.7 MPa and 5.2 GPa, respectively.
However, even this tensile strength was 42 MPa higher than that
of the oriented CNT paper obtained by winding the CNT sheet and
spraying ethanol afterwards, using CNTs with diameter in the same
range (30–50 nm). Instead of spraying afterwards, SWW achieved
abetter alignment and closer packing of the neighboring nanotubes
in the CNT paper.
Notice that there were different ways of CNT papers to respond
to the external strain based on the stress–strain curves in Fig. 6.
The tensile stress of CNT-1 and CNT-2 papers increased almost linearly upon stretching with much larger slope or modulus, while
the CNT-3 paper demonstrated a much lower slope or smaller
modulus. For the former ones, both of their stress–strain curves
showed relatively gentle slope at the beginning due to the realignment of the waviness tube after starting to bear the load. This
behavior is the same as the refinement of the CNT fiber during
stretching [29]. As the strain increased over 1%, all the aligned
tubes in the closely packed paper responded uniformly, resulting
in a large modulus and sufficient strain at break. However, for
CNT-3 paper, the situation changed for the diameter of nanotubes
was several times large. The smaller effective cross-section area
brought reduced strength and modulus. In addition, tube-tube
interaction weakened as the contact area per unit volume between
the CNTs decreased. Therefore, the nanotubes in CNT-3 paper slide
away from each other as strain increased and fractured gradually.
3.4. Comparison of the electrical conductivity of CNT paper
The three different CNT papers are also good electrical conductors with conductivity of 3.6 104, 5.5 104 and 2.9 104 S/m,
345
W. Liu et al. / Composites: Part B 53 (2013) 342–346
Fig. 4. (a) SEM image of the surface of the as-drawn CNT-2 sheet. (b) SEM image of the surface of the as-produced CNT-2 paper.
Fig. 5. Attractive force of liquid bridge pulled nanotube together upon evaporation when the contact angle of the liquid is 0° < h < 90° (after Rondeau et al. [26]).
Table 2
Mechanical properties and Xeffective of the CNT-1, CNT-2 and CNT-3 paper.
CNT-1 paper
CNT-2 paper
CNT-3 paper
Strength (MPa)
Standard deviation (MPa)
Modulus (GPa)
Standard deviation (GPa)
Strain (%)
Standard deviation (%)
Xeffective (%)
562.5
422.6
117.7
59.4
17.4
21.1
15.3
10.1
5.2
1.1
0.9
2.0
4.7
6.4
24.5
0.5
0.5
1.9
14
11–14
3
Fig. 6. Typical stress–strain curve of three types of CNT papers.
correspondently to the paper fabricated from CNT-1, CNT-2 and
CNT-3. The conductivity of CNT paper relies on the conductivity
of the tubes and the CNT organization, such as orientation and
compact density, in the paper. For example, the aligned and densely packed CNT paper made by ‘‘domino pushing’’ has an electrical
conductivity of 2 104 S/m, 25% higher than the randomly aligned
paper from same batch of CNTs [19]. Similarly, CNT performs made
by ‘‘shear pressing’’ showed better electrical conductivity of
1.2 104 S/m in the longitudinal direction than the electrical conductivity of 0.4 104 S/m in the transverse direction [14]. In a CNT
network, electric current was transferred through the main channel formed by the contact area, which can be improved magnificently by the alignment and density of CNT bundles. As a result,
CNT papers obtained by SWW showed a great improvement on
the conductivity over the CNT papers fabricated by conventional
filtration method.
In this study, since CNTs were highly oriented in the structure,
electron transported predominantly through the outermost wall.
However, the CNT-1 paper did not possess better electrical conductivity than other two CNT papers, which is counter intuitive that it
owned largest contact area between nanotubes and highest effective cross-section area due to the smallest dimension. For single
tube with length over 200 lm, the resistance along the tube is also
as high as the contact resistance [30,31], therefore the overall resistance of the CNT paper depends on not only the junction resistance
of CNTs but also the resistance of individual CNTs. On one hand,
CNT-1 has the shortest length, which contributed in more contact
resistance in the CNT paper. On the other hand, if the outermost
wall of MWNT can be considered as single wall nanotube, the electrical conductivity of the nanotube would be metallic or semi-conduct. The possibility of metallic nanotube would increase with the
increasing of the diameter. In other words, there could be a larger
number of semi-conductive nanotubes in CNT-1 paper than CNT-2
and CNT-3 paper, resulting in a lower electrical conductivity.
4. Conclusions
CNT papers were fabricated with desired structure of highly orientated CNTs tightly packed together by a simple SWW method. In
this method, continuous CNT sheet composed of aligned nanotubes
were wound on a rotation mandrel as DI water/ethanol solution
was sprayed. The capillary force during the solvent evaporation
drew neighboring tubes closer together while maintaining the
CNT alignment. As a result, high strength and good conductivity
were obtained along the oriented direction. Using CNTs with diameters around 8 nm, the CNT paper achieved mean strength of
346
W. Liu et al. / Composites: Part B 53 (2013) 342–346
562.5 MPa and mean modulus of 15.3 GPa. CNT paper fabricated
with CNTs of 8–10 nm diameters showed the highest electrical
conductivity of 5.5 104 S/m.
Acknowledgements
This work was supported by the National High Technology Research and Development Program of China (No. 2007AA03Z101),
the State Key Program of National Natural Science of China (No.
51035003), Natural Science Foundation for the Youth (Nos.
50803010 and 60904056), National Science Foundation for Postdoctoral Scientists of China (No. 20100470664), Shanghai Postdoctoral Research Funded Project (No. 09R21410100), the Program
of Introducing Talents of Discipline to Universities (No. B07024),
Shanghai University Young Teacher Training Program and the Fundamental Research Funds for the Central Universities.
References
[1] Yu MF, Files BS, Arepalli S, Ruoff RS. Tensile loading of ropes of single wall
carbon nanotubes and their mechanical properties. Phys Rev Lett
2000;84(24):5552–5.
[2] Yu MF, Lourie O, Dyer MJ, Moloni K, Kelly TF, Ruoff RS. Strength and breaking
mechanism of multiwalled carbon nanotubes under tensile load. Science
2000;287(5453):637–40.
[3] Wei XL, Chen Q, Peng LM, Cui RL, Li Y. Tensile loading of double-walled and
triple-walled carbon nanotubes and their mechanical properties. J Phys Chem
C 2009;113(39):17002–5.
[4] Ebbesen TW, Lezec HJ, Hiura H, Bennett JW, Ghaemi HF, Thio T. Electrical
conductivity of individual carbon nanotubes. Nature 1996;382(6586):54–6.
[5] Hongjie D. Carbon nanotubes: opportunities and challenges. Surf Sci
2002;500(1–3):218–41.
[6] Shimizu T, Abe H, Ando A, Nakayama Y, Tokumoto H. Electrical conductivity
measurements of a multi-walled carbon nanotube. Surf Interface Anal
2005;37(2):204–7.
[7] Baughman RH, Zakhidov AA, de Heer WA. Carbon nanotubes–the route toward
applications. Science 2002;297(5582):787–92.
[8] Dresselhaus MS. NT10: recent advances in carbon nanotube science and
applications. Acs Nano 2010;4(8):4344–9.
[9] Baughman RH, Cui C, Zakhidov AA, Iqbal Z, Barisci JN, Spinks GM, et al. Carbon
nanotube actuators. Science 1999;284(5418):1340–4.
[10] Bahr JL, Yang J, Kosynkin DV, Bronikowski MJ, Smalley RE, Tour JM.
Functionalization of carbon nanotubes by electrochemical reduction of aryl
diazonium salts: a bucky paper electrode. J Am Chem Soc
2001;123(27):6536–42.
[11] Ng SH, Wang J, Guo ZP, Chen J, Wang GX, Liu HK. Single wall carbon nanotube
paper as anode for lithium-ion battery. Electrochim Acta 2005;51(1):23–8.
[12] Futaba DN, Hata K, Yamada T, Hiraoka T, Hayamizu Y, Kakudate Y, et al. Shapeengineerable and highly densely packed single-walled carbon nanotubes and
their application as super-capacitor electrodes. Nat Mater 2006;5(12):987–94.
[13] Xu GH, Zheng C, Zhang Q, Huang JQ, Zhao MQ, Nie JQ, et al. Binder-free
activated carbon/carbon nanotube paper electrodes for use in supercapacitors.
Nano Res 2011;4(9):870–81.
[14] Bradford PD, Wang X, Zhao H, Maria J-P, Jia Q, Zhu YT. A novel approach to
fabricate high volume fraction nanocomposites with long aligned carbon
nanotubes. Compos Sci Technol 2010;70(13):1980–5.
[15] Wang SR, Liang ZY, Pham G, Park YB, Wang B, Zhang C, et al. Controlled
nanostructure and high loading of single-walled carbon nanotubes reinforced
polycarbonate composite. Nanotechnology 2007;18(9):095708.
[16] Gou JH. Single-walled nanotube bucky paper and nanocomposite. Polym Int
2006;55(11):1283–8.
[17] Berhan L, Yi YB, Sastry AM, Munoz E, Selvidge M, Baughman R. Mechanical
properties of nanotube sheets: alterations in joint morphology and achievable
moduli in manufacturable materials. J Appl Phys 2004;95(8):4335–45.
[18] Song PC, Liu CH, Fan SS. Improving the thermal conductivity of
nanocomposites by increasing the length efficiency of loading carbon
nanotubes. Appl Phys Lett 2006;88(15):153111.
[19] Wang D, Song P, Liu C, Wu W, Fan S. Highly oriented carbon nanotube papers
made of aligned carbon nanotubes. Nanotechnology 2008;19(7):075609.
[20] Inoue Y, Suzuki Y, Minami Y, Muramatsu J, Shimamura Y, Suzuki K, et al.
Anisotropic carbon nanotube papers fabricated from multiwalled carbon
nanotube webs. Carbon 2011;49(7):2437–43.
[21] Li QW, Zhang XF, DePaula RF, Zheng LX, Zhao YH, Stan L, et al. Sustained
growth of ultralong carbon nanotube arrays for fiber spinning. Adv Mater
2006;18(23):3160–3.
[22] Inoue Y, Kakihata K, Hirono Y, Horie T, Ishida A, Mimura H. One-step grown
aligned bulk carbon nanotubes by chloride mediated chemical vapor
deposition. Appl Phys Lett 2008;92(21).
[23] Whitten PG, Spinks GM, Wallace GG. Mechanical properties of carbon
nanotube paper in ionic liquid and aqueous electrolytes. Carbon
2005;43(9):1891–6.
[24] Liu W, Zhang X, Xu G, Bradford PD, Wang X, Zhao H, et al. Producing superior
composites by winding carbon nanotubes onto a mandrel under a poly(vinyl
alcohol) spray. Carbon 2011;49(14):4786–91.
[25] Liu W, Zhao H, Inoue Y, Wang X, Bradford PD, Kim H, et al. Poly(vinyl alcohol)
reinforced with large-diameter carbon nanotubes via spray winding.
Composites Part A 2012;43(4):587–92.
[26] Rondeau X, Affolter C, Komunjer L, Clausse D, Guigon P. Experimental
determination of capillary forces by crushing strength measurements.
Powder Technol 2003;130(1–3):124–31.
[27] Suppiger D, Busato S, Ermanni P. Characterization of single-walled carbon
nanotube mats and their performance as electromechanical actuators. Carbon
2008;46(7):1085–90.
[28] Xu G, Zhang Q, Zhou W, Huang J, Wei F. The feasibility of producing MWCNT
paper and strong MWCNT film from VACNT array. Appl Phys A
2008;92(3):531–9.
[29] Jia J, Zhao J, Xu G, Di J, Yong Z, Tao Y, et al. A comparison of the mechanical
properties of fibers spun from different carbon nanotubes. Carbon
2011;49(4):1333–9.
[30] Li S, Yu Z, Rutherglen C, Burke PJ. Electrical properties of 0.4 cm long singlewalled carbon nanotubes. Nano Lett 2004;4(10):2003–7.
[31] Hecht D, Hu L, Gruner G. Conductivity scaling with bundle length and diameter
in single walled carbon nanotube networks. Appl Phys Lett
2006;89(13):133112.