polymer composite films with excellent mechanical strength and electromagnetic interference shielding

polymer composite films with excellent mechanical strength and electromagnetic interference shielding

Journal Pre-proof Ultrathin and flexible carbon nanotube/polymer composite films with excellent mechanical strength and electromagnetic interference s...

5MB Sizes 0 Downloads 2 Views

Journal Pre-proof Ultrathin and flexible carbon nanotube/polymer composite films with excellent mechanical strength and electromagnetic interference shielding Guang Wu, Yun Chen, Hang Zhan, Hai Tao Chen, Jia Hao Lin, Jian Nong Wang, Li Qiang Wan, Fa Rong Huang PII:

S0008-6223(19)31138-8

DOI:

https://doi.org/10.1016/j.carbon.2019.11.014

Reference:

CARBON 14771

To appear in:

Carbon

Received Date: 28 August 2019 Revised Date:

28 October 2019

Accepted Date: 5 November 2019

Please cite this article as: G. Wu, Y. Chen, H. Zhan, H.T. Chen, J.H. Lin, J.N. Wang, L.Q. Wan, F.R. Huang, Ultrathin and flexible carbon nanotube/polymer composite films with excellent mechanical strength and electromagnetic interference shielding, Carbon (2019), doi: https://doi.org/10.1016/ j.carbon.2019.11.014. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Published by Elsevier Ltd.

Ultrathin

and

flexible

carbon

nanotube/polymer

composite films with excellent mechanical strength and electromagnetic interference shielding Guang Wua, Yun Chena, Hang Zhana, Hai Tao Chena, Jia Hao Lina, Jian Nong Wanga,*, Li Qiang Wanb, Fa Rong Huangb a

School of Mechanical and Power Engineering, East China University of Science and

Technology, 130 Meilong Road, Shanghai 200237, China b

School of Materials Science and Engineering, East China University of Science and

Technology, 130 Meilong Road, Shanghai 200237, China

*

Corresponding author. Tel: 86-21-64252360. E-mail address: [email protected] (Jian Nong Wang)

Ultrathin

and

flexible

carbon

nanotube/polymer

composite films with excellent mechanical strength and electromagnetic interference shielding Guang Wua, Yun Chena, Hang Zhana, Hai Tao Chena, Jia Hao Lina, Jian Nong Wanga,*, Li Qiang Wanb, Fa Rong Huangb a

School of Mechanical and Power Engineering, East China University of Science and

Technology, 130 Meilong Road, Shanghai 200237, China b

School of Materials Science and Engineering, East China University of Science and

Technology, 130 Meilong Road, Shanghai 200237, China

*

Corresponding author. Tel: 86-21-64252360. E-mail address: [email protected] (Jian Nong Wang)

Abstract

Up till now, metals, conductive polymers and carbon materials have been widely applied in electromagnetic interference (EMI) shielding. However, EMI shielding materials that are light-weight, flexible, ultra-thin, and mechanically robust are strongly needed for many civilian and military applications. In this study, a carbon nanotube (CNT)/polymer composite film of only one-micrometer thick is prepared by continuous winding and deposition of a cylinder-like CNT assembly impregnated with a polymer solution. The in-situ impregnation leads to homogeneous mixing and strong tube-tube interfacial bonding. Further optimization of the CNT content and alignment to high levels endows the thin composite film with a high tensile strength of 1250 MPa as well as a high EMI shielding effectiveness of 30 dB in the frequency range from 1 GHz to 18 GHz. Such a combination of mechanical and shielding properties surpasses all previous observations, and thus provides a new strategy for developing novel shielding materials for wide applications.

1 Introduction Electromagnetic interference (EMI) is created by all electronic facilities that transmit, distribute, or use electrical energy. The detrimental effects of EMI have been of big concern since the proliferation of cellular towers, wireless networks, and electronic devices. For example, EMI may lead to malfunctioning and degradation of electronic devices [1-2], induce health risks to human beings [3-4], and even cause the loss of digital privacy as a result of data theft and hardware security breaches. To achieve effective EMI shielding in day-to-day applications, a minimum EMI shielding effectiveness (EMI-SE) of 20 dB is necessary [5]. Other requirements include a thin thickness to reduce cost and space occupation and good stability in physical, chemical, and mechanical properties for long-term use [6]. In addition, with the advent of flexible electronics, wearable devices, and implantable biomedical systems, effective and practical EMI shielding solutions have to be ultra-thin, lightweight, and

importantly, mechanically robust and flexible [7-8]. Great efforts have been made on developing novel materials for shielding EMI. Traditional shielding systems are metal-based [9] and suffer from deficiencies of high density, low corrosion-resistance, and poor mechanical properties. Conductive polymers such as polyaniline [10-11], polypyrrole [12-13], or polythiophene [13-14] are considered as alternative shielding materials because of their lightweight, good processability and tunable electromagnetic properties. Besides, porous composite foams based on carbon [15] and graphene [1,16] have also received wide interests as a means to reduce the weight while maintaining a high shielding effectiveness. The critical shortcomings of these conductive polymers and porous composite foams are that, to achieve an adequate EMI-SE, they have to have large thicknesses of a few millimeters due to their low electrical conductivity, and it is hard to improve their mechanical strength in the presence of pores. Conductive composites have been well studied for EMI shielding. In such materials, conductive fillers are dispersed in a ceramic or polymer matrix to form conductive networks for the purpose of attenuating electromagnetic waves [17-18]. In the case of a ceramic matrix (mullite, BaTiO3, Al2O3, SiO2), although good EMI SE (20~35 dB) and even high strength (380~506 MPa) can be achieved, the samples are thick (1.4~2.4 mm) and inflexible [19]. Many studies have focused on conductive polymer composites (CPCs) because of their potential flexibility. Examples have been shown for a transparent film composed of calcium alginate, silver nanowires, and polyurethane with a high EMI SE of 20.7 dB [20], a CNT/natural rubber composite film with a low EMI SE of 20 dB at a small thickness of 50 µm but a high EMI SE of 44.7 dB at a large thickness of 250 µm [21], and a robust graphene/calcium alginate film with an EMI SE of 25.7 dB at a thickness of 12 µm and a high tensile strength of 118 MPa [22]. In the case of CPCs, the conductive fillers are mainly carbon based, including carbon particles, carbon fibers, carbon nanotubes (CNTs), and graphene sheets [9]. At low contents of conductive fillers, a high electrical conductivity and good EMI SE have been achieved by introducing double-percolation and segregated structures with

the conductive fillers being selectively located in one continuous phase or interfaces [23-25], but the mechanical properties of the CPCs with such structures are low due to insufficient strengthening. Simply increasing the content of conductive fillers can increase the conductivity and EMI SE of CPCs. However, the high content of conductive fillers usually leads to the deterioration of mechanical and processing properties because of the poor dispersion and easy aggregation of the nanofillers [26-27]. To tackle this problem, great effects have been devoted on controlling the structure, aspect ratio, and surface functionalization of nanofillers [28-29]. Nevertheless, achieving a high EMI SE and mechanical strength at the same time has not been realized. CNT as an extraordinary one-dimensional material possesses high electrical and mechanical properties and thus is considered as an ideal filler for CPCs. Here, a cylinder-like macroscopic CNT assembly is fabricated by spray pyrolysis. When this CNT assembly just comes out from a high temperature reactor, it is infiltrated with a polymer solution of polytriazole (PTA) for making a composite film. At a high CNT content of 59.8 wt.% and a small thickness of 1 µm, the composite film has an electrical conductivity of 2040 S cm-1 and EMI SE exceeding 26 dB over the range from 1 GHz (L band) to 18 GHz (Ku band). By further controlling the alignment of CNTs, the composite film is shown to have good flexibility and a tensile strength as high as 1250 MPa. 2. Experimental methods 2.1 Synthesis of hollow cylindrical CNT assembly The experimental setup is schematically illustrated in Fig. 1. CNTs were synthesized by floating-catalyst spray pyrolysis in a horizontal furnace with a tubular reactor at a temperature of 1100−1150 °C . Ethanol dissolved with a catalyst precursor (1 wt.% ferrocene) and growth promoter (1.2 wt.% thiophene) was used as a feedstock and injected into the high-temperature zone. The carrier gas of nitrogen was also supplied to drive out the CNT assembly from the reactor.

Hollow CNT Assembly

Winding Drum Resin Container

Furnace

Carrier Gas

Precursors

Fig. 1. Schematic of the experimental process. A CNT assembly is produced from a reactor and deposited, together with a resin, on a winding drum to form a composite film.

2.2 Fabrication of CNT/PTA film PTA was used to make CPC films with CNTs. This resin [30-31] was synthesized from alkynyl compounds and azido compounds by 1,3-dipolar cycloaddition. It is a new kind of polymers with excellent thermal stability and mechanical property and can be cured at temperatures below 100 °C in the absence of any curing agent. Benefited from its good solubility in polar solvents such as acetone, PTA resin can be diluted into any concentrations as needed. However, it should be noted that PTA was used in this study only as an example for making CPCs with CNTs. Other high-performance thermosetting resins such as cyanate ester [32] and thermoplastic resins could also be used as long as they can be dissolved in a common solvent. For making the composite films, the CNT assembly was introduced and winded on a cylindrical drum. This winding drum was covered with an aluminum foil as the substrate for the deposition of the CNT assembly at the exit of the tubular reactor. Under the drum, there was a container containing the solution made from acetone and dissolved PTA. The height of the surface of the solution was adjusted to be just in contact with the drum. With the drum wetted with the solution, the CNT assembly shrank into a narrow film of ~10 nm thick and deposited on the Al foil due to the strong contraction force of acetone. As this deposition process proceeded, PTA was

simultaneously introduced into the film. The rotating rate of the drum was controlled to match the running rate of the CNT assembly. For comparison, two different rates of 20 and 30 resolutions per minute (rpm) were used. The whole deposition process lasted for 30 minutes for the preparation of each film. After 4 hours in vacuum for the removal of acetone, the resin-containing film was cured in a plate vulcanizing machine. The film was kept at 80 °C for 6 hours at the pressure of 2 MPa. After the curing, the film was cooled down to room temperature, and the original substrate of Al foil was removed. 2.3 Characterization CNTs were characterized by high-resolution transmission electron microscopy (HRTEM, JEOL JEM-2100, accelerating voltage of 200 kV) and the thickness and fracture surface of the prepared film were examined by scanning electron microscopy (SEM, Hitachi S3400N and S4800, accelerating voltage of 15 kV). The constituents of the composite film were analyzed by thermogravimetric analysis (TGA, Nietzsche STA 449 F3 Jupiter) at the heating rate of 10 °C min-1 in air. The degree of CNT graphitization was characterized by Raman spectroscopy (RAM, Senterra R200-L, excitation wavelength of 514 nm), using the ratio of the intensities of G and D bands. The CNT alignment in the composite material was evaluated by polarized Raman spectroscopy. A polarized laser beam was incident on the sample surface with the polarization direction parallel to the winding direction of the CNT film. The returned signals from the sample with the polarized direction parallel (0º) and perpendicular (90º) to that of the incident laser were collected separately. The ratio of the intensities of the G bands derived from these two spectra, IG(0º)/IG(90º), was introduced to describe the degree of the CNT alignment in the sample. 2.4 Testing of mechanical and physical properties The as-prepared composite film was cut into several small strips of 40 mm long and 2 mm wide with the length parallel to the winding direction or the orientation of

CNT alignment. Such strip samples were tested in a tensile tester (XS(08)X-15, Shanghai Xusai Co., China) equipped with a load measuring system with a resolution of 0.01 cN. The tensile tests were performed at a gauge length of 10 mm and a displacement rate 20 mm min-1 based on the standard of ASTM D3822. At least 5 specimens were tested to examine the repeatability of the result. A four-point probe was used to measure the electrical conductivity of the specimen. The measurement was also repeated over 20 times to get an average value. The electrical conductivity κ is calculated from κ = L/RA, where L, R, and A are the length, square resistance, and cross-sectional area of the sample, respectively. The EMI shielding performance of the composite material was measured by a vector network analyzer (VNA, Ceyear AV3672B, CETC41) with a circular flange coaxial device. The measurement was carried out in the frequency range from 1 GHz to 18 GHz at room temperature. The sample was cut into a wafer with a diameter of 60 mm and inserted in a copper sample holder. The scattering parameter S12 or S21 which is related to the transmissivity coefficient (T) was obtained, and the EMI SE (expressed in decibels) was calculated from the following equations: 2

T = S12 = S21

2

(1)

(

EMI SE = 10log (Pi Pt ) = 10log(1 T ) = 20log 1 S12

)

,

(2)

where Pi and Pt are the power of incident and transmitted electromagnetic wave, respectively.

3. Results and Discussion Under the present experimental conditions, a hollow cylinder-like CNT assembly was generated and continuously driven out into the open-air environment, as shown in Fig. 2a. The CNT assembly consisted of CNT individuals and bundles, and these CNTs constituted an interconnected network-like structure with a large portion of them aligned at one orientation (the extending direction of the cylindrical assembly), as showed in Fig. 2b. This network structure was conducive to enable rapid electron conduction. TEM examination revealed that the CNT assembly contained not only

CNTs but Fe nanoparticles as well (Fig. 2c). The presence of Fe nanoparticles is confirmed with high resolution TEM imaging (Supplementary Fig. S1). These Fe particles originated from the catalyst used for the growth of CNTs. CNTs appeared to be few-walled and well graphitized with diameters of 4−12 nm (Fig. 2d). The degree of graphitization was also studied by Raman spectroscopy by considering the graphite-associated G band appearing at ~1580 cm-1 and the disorder-induced D band appearing at ~1350 cm-1 [33]. A high ratio of the G and D peak intensities (IG/ID=5.15) confirmed a high degree of graphitization (Fig. 2e). The TGA result of the CNT assembly is shown in Fig. 2f. At ~500 ºC, the material began to lose weight significantly as carbonaceous substance was oxidized, and at ~800 ºC, the material ended up with a residue which is presumably Fe2O3 resulting from the complete oxidation of Fe particles. From the residual weight of Fe2O3 and according to the law of indestructibility of matter, the initial CNT assembly was estimated to contain 20.8 wt.% Fe and 79.2 wt.% CNTs. The winding of the CNT assembly with the addition of a resin solution resulted in a composite film (Fig. 3a). This as-prepared film was soft and flexible, and could even be rolled to a cylinder (Fig. 3b). Examination in SEM illustrated that the composite had a uniform surface and thickness of 1 µm in the sense of the mixing of CNTs and PTA resin (Fig. 3c and 3d). Previous CNT aggregation and poor dispersion were scarcely observed. This is a result of the on-line mixing of CNTs with PTA resin. That is, as the CNT assembly was introduced onto the winding drum, it was shrunk immediately with an acetone solution containing the PTA resin. Under such a circumstance, individual CNTs were wetted with the resin solution before they got aggregated. In addition to CNTs and PTA, the composite film also contained Fe particles which was 15.7 wt.% from TGA (Fig. 2f). Based on the contents of Fe in the original CNT film and the composite film, the contents of CNT and PTA in the composite film were calculated to be 59.8 wt.% and 24.5 wt.%, respectively. Since the Fe particles were originally attached on CNTs, they were also distributed in the composite film uniformly.

(a)

(b)

Hollow CNT Assembly

Tubular Reactor 1 µm

20 mm

(c)

(d)

200 nm

(f) 100

G peak

IG / ID = 5.15

80 Neat CNT film 29.70% residue (20.8% Fe, 79.2% CNT)

60

D peak

40 CNT/PTA film, 22.39% residue (15.7% Fe, 24.5% PTA, 59.8% CNT)

1000

1200

1400

1600

1800

-1

Raman Shift (cm )

2000

20

100 200 300 400 500 600 700 800 900 o

Temperature ( C)

Fig. 2. Experimental results of the produced CNT assembly. (a) Optical image of a hollow CNT assembly blown out from the reactor, corresponding to the marked rectangular part in Fig. 1; (b) Typical SEM image of the CNT assembly, showing the alignment of most CNTs along one orientation; (c) Typical TEM image of the assembly, showing the presence of CNTs as well as fine particles of Fe; (d) HRTEM image of the CNTs, showing their few-walled structures; (e) Raman spectrum showing a high intensity ratio of G and D bands of 5.15, indicating a highly graphitized structure; (f) TGA curves of neat CNT film and CNT/PTA composite film. The constituents of each film are shown in wt.%.

Mass (%)

Intensity (a.u.)

(e)

5 nm

(b)

(a) (c)

(d)

~ 1 µm 5 µm

2 µm 1.2

50

(f)

(e) 1.1

R/Ro

EMI-SE (dB)

40

30

Bending

1.0

Recovering 0.9

20

10

0.8 2

4

6

8

10

12

14

Frequency (GHz)

16

18

0

100

200

300

400

500

600

Cycles

Fig. 3. Experimental results of the composite film prepared with a solution containing 0.05 vol.% PTA resin at the winding rate of 30 rpm. (a) Optical image of the film after curing; (b) Optical image of the film rolled to a cylinder, showing its flexibility; (c) SEM image of the surface of the film; (d) SEM image of the cross section of this film, showing a thickness of about 1 µm; (e) EMI SE of this film at 1~18 GHz; (f) Variation of the resistance of this composite film after repeated bending and recovering (R) relative to the original resistance (R0).

EMI shielding is the concern of this study. When encountering a shielding material, an incident electromagnetic wave is divided into four parts: reflected, absorbed, internally multi-reflected, and transmitted. The shielding efficiency of a dense and conductive material as an electromagnetic attenuator is mainly determined by direct reflection and internal absorption as the incident wave can rarely penetrate into the material and thus the internal multi-reflection is so weak that it is negligible [34]. Measurement over the range from 1 GHz (L band) to 18 GHz (Ku band) showed that the EMI SE of the composite film winded at 30 rpm (Fig. 3a) exceeded 26 dB (Fig. 3e). In particular, in the ranges between 6.27 GHz and 7.61 GHz and between 8.18 GHz and 18 GHz (frequency bandwidth of 11.16 GHz), the shielding efficiency exceeded 30 dB, meeting most civil and military shielding requirements. However, the composite film prepared at the winding rate of 20 rpm had an EMI SE lower than 24 dB over the range from 1 to 18 GHz (Supplementary Fig. 2S). Four-point probing measurements revealed that the composite film prepared at 30 rpm had an electrical conductivity of 2040 ± 122 S cm-1, and that prepared at 20 rpm 1353 ± 104 S cm-1. These high conductivities are apparently related to the large content of CNTs in the film (~60 wt.%). For flexible electronic applications, it is necessary to examine the stability of the conductivity under repeated mechanical deformation. Here, a sample was bent by 180º and then recovered to its orginal shape. The electrical resistance and EMI SE were almost the same after 600 bending-recovering cycles as shown in Fig. 3f. According to the parallel resistor-capacitor model [35], the electrical conductivity may be taken as a good prediction of the shielding performance. The p-electrons of carbon atoms with a hexagonal structure in highly graphitized CNTs form a wide range of delocalized π-bond. No matter what the CNT is, metallic or semiconducting, the transition of electrons from valence band to conduction band is relatively easier than other materials. For this reason, CNT itself has a high electrical conductivity. Second, CNTs in the cylindrical assembly are mutually connected (Fig. 2c), constituting a network-like pathway for electron conduction in the composite matrix. Third, the content of CNTs, in other words, the concentration of conductive

nanofillers, is as high as 59.8 wt.%. All these contribute the observed high electrical conductivity and EMI shielding performance. Comparison between the films prepared at 30 rpm and 20 rpm having different conductivities (2040 vs 1353 S cm-1) and EMI SEs (>26 vs <24 dB) further suggests the importance of electrical conductivity in controlling EMI shielding performance. Because of the uniqueness of the present preparation process, the iron content in the composite film could be controlled by adjusting the amount of ferrocene added in the original precursor solution. We focused on the composite film having 15.7 wt.% Fe as randomly distributed nanoparticles (Fig. 2f, Fig. 3). With the presence of such Fe nanoparticles, the impedance mismatch of the film could be mitigated [36], and the coercivity slightly decreased with the specific saturation magnetic moment increased [37]. In this case, iron particles, as a kind of ferromagnetic substance, contribute to the magnetic loss of electromagnetic waves. That is, if the electromagnetic wave goes into the film, it would be consumed and converted into heat not only by dielectric loss but magnetic loss as well. Thus, the origin of the high shielding performance observed for the present composite film could be due to the synergetic effect of the conductive CNTs and magnetic nanoparticles as observed before [38]. As an ultrathin film, its EMI shielding performance has met the basic requirement of day-day applications. If its mechanical property, especially the tensile strength, is promoted to the order of GPa, the range of its potential applications would be greatly expanded. The tensile strength of the composite film is mainly decided by two factors: the proportion of PTA resin and the degree of CNT alignment in the composite material. These two factors could be controlled by changing the PTA concentration in the resin solution and the winding rate of the CNT assembly or the rotating rate of the winding drum, respectively. First, the rotating rate was set at 20 rpm, and the deposition time at 30 minutes, but the PTA concentration was varied. The overall tendency of the variation of the tensile strength of the composite film deposited with shrinking solutions of different PTA concentrations is shown in Fig. 4a. When the film was shrunk with pure acetone, CNTs intertwined with each other loosely under the effect of van der Waals force. A

lot of voids existed between the individuals or bundles of CNTs, which gave rise to a low tensile strength of 119 MPa. When 0.025 vol.% PTA resin was added in the shrinking solution, the average tensile strength of the CNT/PTA composite film was increased to 419 MPa. Compared to the neat CNT film, this increment reflects the effect of the PTA resin infiltrated between CNTs, binding individuals or bundles of CNTs together and providing an additional approach for load transfer. When the PTA concentration was doubled to 0.05 vol.%, the average tensile strength was enhanced further up to 824 MPa with an elongation at break of 5−6% (Fig. 4b). Such a large enhancement results from the further tight packing between CNTs and PTA resin and the further densification of the overall material and limitation of the slippage between CNTs (Supplementary Fig. S3a). However, when 0.1 vol.% PTA resin was added, the tensile strength decreased to 575 MPa. This experimental phenomenon suggests that excessive PTA resin could induce a detrimental effect on tensile strength. The reason may be that, when all the interfaces between CNTs have just been bound with resin, the addition of excessive resin actually increases the intertube spacing and thus weakens the direct load transfer between CNTs. The addition of excessive resin in the film can be clearly seen from the irregular resin blocks found at the fracture surface (Supplementary Fig. S3b). Therefore, the selection of a proper resin concentration to control the resin content in the composite material is important to the optimization of tensile strength. CNT orientation is another important factor deciding the mechanical property of the composite material. In the present experiment, as the CNT assembly was pulled out from the reactor and winded on the winding drum, it was stretched and thus CNT alignment was induced along the winding direction in the film. Tensile testing at the winding direction or the main CNT orientation generated a tensile strength of ~800 MPa, but that at the direction perpendicular to the CNT orientation gave a strength of only ~200 MPa (Fig. 4c). This observation manifests the anisotropy of the composite film and higher strength at the main CNT orientation. The presence of CNT alignment was further confirmed with polarized Raman spectroscopy. The ratio of the intensities of the G band at the parallel and perpendicular directions, IG(0º)/IG(90º), is used as a

measure of the degree of alignment. For the case of randomly oriented CNTs, the theoretical and experimentally measured ratio is 1 [39]. For the present composite film, the measured ratio is 1.69 (Fig. 4d), indicating the presence of CNT alignment.

1000

1000

(b) 800

800

Stress (MPa)

Tensile Strength (MPa)

(a)

600 400

600 400 200

200 0

0 0

0.025

0.05

0.1

0

2

4

6

8

10

Strain (%)

PTA resin concentration (%) 1000 800

(d)

σb(0o) / σb(90o)

90° 0°

Intensity (a.u.)

Stress (MPa)

(c)

= 3.8 600 400

90° 0°

ΙG(0o) / ΙG(90o) = 1.69

200 0 0

2

4

6

Strain (%)

8

10

1000

1200

1400

1600

1800

-1

Raman Shift (cm )

Fig. 4. Mechanical property and orientation characterization of the composite film prepared at the rotating rate of 20 rpm. (a) Variation of tensile strength of the composite films prepared at different PTA resin concentrations; (b) Stress-strain curve of the composite film prepared at the concentration of 0.05 vol.% PTA resin; (c) Representative stress-strain curves of the composite film, showing the ratio of tensile strengths between the parallel and perpendicular directions; (d) Polarized Raman spectra of the composite film, showing the ratio of G peak intensities measured at the parallel and perpendicular directions.

The degree of CNT alignment is dependent upon the winding rate of the CNT assembly or the rotating rate of the winding drum because the faster the rate, the large the stretching induced to the CNT assembly. With an attempt to induce a higher degree of CNT alignment, we increased the rotating rate from 20 rpm to 30 rpm while

2000

keeping the PTA concentration at the optimum level of 0.05 vol.% with all other conditions unchanged. The typical stress-strain curves of such a CNT/PTA composite are shown in Fig. 5a. Compared to the preparation at a lower rate, the tensile strength at the faster rate was raised up to 1250 ± 119 MPa, indicating better CNT alignment. A typical tensile fracture surface is illustrated in Fig. 5b. The whole fracture is relatively neat and regularly arranged, suggesting that the material underwent tensile deformation as a whole. CNT filaments drawn out of the sample were at least tens of micrometer long.

1500

(a) Stress (MPa)

1200 900 600 300

(b)

0 0

2

4

6

8

10 µm

10

Strain (%) 1500

σb(0o) / σ b( 90o)

= 6.1

Intensity (a.u.)

Stress (MPa)

1200

90 ° 0°

(d)

90 ° 0°

900 600

Ι G(0o) / ΙG( 90o) = 2.05

300

(c) 0 0

2

4

6

8

10

1000

Strain (%)

1200

1400

1600

1800 -1

Raman Shift (cm )

Fig. 5. Mechanical property and orientation characterization of the composite film with 24.5 wt.% PTA resin prepared at the rotating rate of 30 rpm. (a) Stress-strain curve of the film; (b) Fracture surface of the film; (c) Typical stress-strain curves of the film, showing the ratio of tensile strengths between the parallel and perpendicular directions; (d) Polarized Raman spectra of the film, showing the ratio of G peak intensities measured at the parallel and perpendicular directions.

2000

In order to better understand the effect of CNT alignment, we continued to do tensile tests perpendicular to the main CNT orientation or the winding direction. As shown in Fig. 5c, the average tensile strength at the CNT orientation is 6.1 times of that at the perpendicular direction. Polarized Raman spectroscopy gave rise to a ratio of IG(0º)/IG(90º)) of 2.05 (Fig. 5d). Both the tensile strength ratio and the G band intensity ratio provide evidence that the composite film prepared at the higher winding rate had a higher degree of CNT alignment and further that it is this higher degree of CNT alignment that contributed the higher tensile strength (1250 MPa vs 824 MPa). We attempted to wind the CNT assembly at a rate higher than 30 rpm, but the CNT assembly was easy to be broken and the winding process could not proceed without intermittent stops. This high strength composite film was chosen to be studied for EMI shielding. The results shown in Fig. 3 are for this composite film. It is now known that this composite film has a high content of aligned CNTs (59.8 wt.%), a high tensile strength (1.25 GPa), a high electrical conductivity (2040 S cm-1), and a high EMI SE (30 dB) even at a thickness as low as 1 µm. For a given shielding material, in addition to its electrical and magnetic properties, the shielding performance is also determined by its thickness. That is, a higher EMI SE can be measured if the material with a larger thickness is used. Almost all EMI SE data are measured from samples of different thicknesses in the literature. Therefore, direct comparison cannot be made among different materials. Here, we introduce specific EMI SE (electromagnetic interference shielding effectiveness per unit thickness) for comparison. This is to minimize the effect of sample thickness. Comparison diagrams of tensile strength versus EMI SE per millimeter thickness and tensile strength versus electrical conductivity are illustrated in Fig. 6a and Fig. 6b, respectively. The details of the results reported for different shielding materials are listed in Supplementary Table S1. Comparing with traditional carbon-based shielding materials with disconnected CNT structures [40-41], the performance of the present network-like structure with aligned CNTs is outstanding. This unique interconnected structure is beneficial to both load transfer and current conduction and endowed our composite film with high

tensile strength and electrical conductivity. With the addition of PTA resin, the voids between CNTs were filled and the interface between CNTs was bound. Unlike the cases with graphene used [22, 42], internal stratification, which prejudices both mechanical and shielding performance, did not appear in our sample. This is because of the on-line process of solution impregnation in which CNTs and resin molecules were mutually impregnated and assembled together almost nondestructively.

Tensile Strength (MPa)

1400

(a)

1200

CNT CF PEDOT:PSS PANI Graphene Metal Foil

1000

This work

800 600 400 200 0 10

1

10

2

10

3

10

4

EMI SE/t (dB/mm)

Tensile Strength (MPa)

1400 1200

CNT CF PEDOT:PSS PANI Graphene Metal Foil

(b)

1000

This work

800 600 400 200 0 10

-2

10

0

10

2

-1 Conductivity (S cm )

10

4

10

6

Fig. 6. Property comparisons between the present shielding material and previous ones involving the uses of carbon nanotubes (CNT), carbon fibers (CF), poly(3,4-ethylenedio xythiophene): poly(styrenesulfonate) (PEDOT:PSS), polyaniline (PANI), graphene, and metal foils. The details of data sources are listed in the Supplemental Data (Table S1). (a) Tensile strength versus EMI SE per millimeter thickness (t); (b) Tensile strength versus electrical conductivity.

A material having a high EMI SE only at a large thickness is hard to be used for many practical applications. This is true particularly for electronic devices which are now getting smarter, being made smaller, and operating at faster speeds, and thus desperately demand shielding materials with an ultrathin thickness for space and weight saving. Comparison with previous studies reveals that this target can be reached by developing composite films with a high content of CNTs as well as some magnetic nanoparticles. For the applications in flexible electronics, wearable devices, and even aerospace structural components, the shielding material is also required to possess a high mechanical strength to resist the exertion of external forces. A poor mechanical strength apparently restricts the long-term application under harsh mechanical conditions. Comparison with previous studies suggests that this goal can be realized by controlling the CNT content and alignment in the CPC material without sacrificing the shielding performance.

4. Conclusion In summary, a CNT/PTA composite film with a thickness of 1 µm was prepared by winding a cylinder-like CNT assembly together with a PTA resin solution and subsequent hot pressing. By optimizing the CNT content and CNT alignment, the film had a high strength of 1.25 GPa, electrical conductivity of 2040 S cm-1 and EMI SE of 30 dB from 1 GHz to 18 GHz. This excellent combination of mechanical, electrical, and electromagnetic shielding properties outperforms all existing shielding materials reported so far. Such an outstanding combination originates from the on-line impregnation of CNTs with PTA resin at the molecular level, strong intertube binding, high content of aligned CNTs, and interconnected network-like CNT structure. The present ultrathin, flexible, and robust film may find wide applications in both civilian and military shielding fields.

Acknowledgments This research was supported by National Key R&D Program of China (2018YFA0208404), National Natural Science Foundation of China (U1362104), and Innovation Program of Shanghai Municipal Education Commission.

References [1] Chen ZP, Xu C, Ma CQ, Ren WC, Cheng HM. Lightweight and flexible graphene foam composites for high-performance electromagnetic interference shielding. Adv Mater 2013; 25(9): 1296-1300. https://doi.org/10.1002/adma.201204196. [2] Yousefi N, Sun XY, Lin XY, Shen X, Jia JJ, Kim JK, et al. Highly aligned graphene/polymer nanocomposites with excellent dielectric properties for high‐performance electromagnetic interference shielding. Adv Mater 2014; 26(31): 5480-5487. https://doi.org/10.1002/adma.201305293. [3] Wu Y, Wang ZY, Liu X, Shen X, Zheng QB, Kim JK, et al. Ultralight graphene foam/conductive

polymer

composites

for

exceptional

electromagnetic

interference shielding. ACS Appl Mater Interfaces 2017; 9(10): 9059-9069. https://doi.org/10.1021/acsami.7b01017. [4] Frey AH. Headaches from cellular telephones: Are they real and what are the implications? Environ Health Persp 1998; 106(3): 101-103. https://doi.org/10.1289/ehp.98106101. [5] Shen B, Li Y, Zhai WT, Zheng WG. Compressible graphene-coated polymer foams with ultralow density for adjustable electromagnetic interference (EMI) shielding. ACS Appl Mater Interfaces 2016; 8(12): 8050-8057. https://doi.org/10.1021/acsami.5b11715. [6] Shen B, Zhai WT, Zheng WG. Ultrathin flexible graphene film: An excellent thermal conducting material with efficient EMI shielding. Adv Funct Mater 2014; 24(28): 4542-4548. https://doi.org/10.1002/adfm.201400079. [7] Lee SH, Kang D, Oh IK. Multilayered graphene-carbon nanotube-iron oxide three-dimensional heterostructure for flexible electromagnetic interference

shielding film. Carbon 2017; 111: 248-257. https://doi.org/10.1016/j.carbon.2016.10.003. [8] Ling JQ, Zhai WT, Feng WW, Shen B, Zhang JF, Zheng WG. Facile preparation of lightweight microcellular polyetherimide/graphene composite foams for electromagnetic interference shielding. ACS Appl Mater Interfaces 2013; 5(7): 2677-2684. https://doi.org/10.1021/am303289m. [9] Abbasi H, Antunes M, Velasco JI. Recent advances in carbon-based polymer nanocomposites for electromagnetic interference shielding. Prog Mater Sci 2019; 103: 319-373. https://doi.org/10.1016/j.pmatsci.2019.02.003. [10] Bhadra S, Khastgir D, Singha NK, Lee JH. Progress in preparation, processing and applications of polyaniline. Prog Polym Sci 2009; 34(8): 783-810. https://doi.org/10.1016/j.progpolymsci.2009.04.003. [11] Hussain ST, Abbas F, Kausar A, Khan MR. New polyaniline/polypyrrole /polythiophene and functionalized multiwalled carbon nanotube-based nanocomposites: Layer-by-layer in situ polymerization. High Perform Polym 2013; 25(1): 70-78.https://doi.org/10.1177/0954008312456048. [12] Kim MS, Kim HK, Byun SW, Jeong SH, Joo JS, Lee JY, et al. PET fabric/polypyrrole composite with high electrical conductivity for EMI shielding. Synthetic Met 2002; 126(2-3): 233-239. https://doi.org/10.1016/S0379-6779(01)00562-8. [13] Saini P, Choudhary V. Conducting polymer coated textile based multilayered shields for suppression of microwave radiations in 8.2-12.4 GHz range. J Appl Polym Sci 2013; 129(5): 2832-2839. https://doi.org/10.1002/app.38994. [14] Iqbal S, Shah J, Kotnala RK, Ahmad S. Highly efficient low cost EMI shielding by barium ferrite encapsulated polythiophene nanocomposite. J Alloy Compd 2019; 779: 487-496. https://doi.org/10.1016/j.jallcom.2018.11.307. [15] Zhang LY, Liu M, Roy S, Chu EK, See KY, Hu X. Phthalonitrile-based carbon foam with high specific mechanical strength and superior electromagnetic interference shielding performance. ACS Appl Mater Interfaces 2016; 8(11): 7422-7430. https://doi.org/10.1021/acsami.5b12072.

[16] Shen B, Zhai WT, Tao MM, Ling JQ, Zheng WG, Lightweight, multifunctional polyetherimide/[email protected]

composite

foams

for

shielding

of

electromagnetic pollution. ACS Appl Mater Interfaces 2013; 5: 11383-11391. https://doi.org/10.1021/am4036527. [17] Arjmand M, Chizari K, Krause B, Sundararaj U. Effect of synthesis catalyst on structure of nitrogen-doped carbon nanotubes and electrical conductivity and electromagnetic interference shielding of their polymeric nanocomposites. Carbon 2016; 98: 358-372. https://doi.org/10.1016/j.carbon.2015.11.024. [18] Lee SH, Yu S, Shahzad F, Hong JP, Kim WN, Koo CM, et al. Highly anisotropic Cu oblate ellipsoids incorporated polymer composites with excellent performance for broadband electromagnetic interference shielding. Compos Sci Technol 2017; 144: 57-62. https://doi.org/10.1016/j.compscitech.2017.03.016. [19] Ru JH, Fan YC, Zhou WW, Zhou ZX, Wang T, Liu RH, et al. Electrically conductive and mechanically strong graphene/mullite ceramic composites for high-performance electromagnetic interference shielding. ACS Appl Mater Interfaces 2018; 10: 39245-39256. https://doi.org/10.1021/acsami.8b12933. [20] Jia LC, Yan DX, Liu XF, Ma RJ, Wu HY, Li ZM. Highly efficient and reliable transparent electromagnetic interference shielding film. ACS Appl Mater Interfaces 2018; 10 (14): 11941-11949. https://doi.org/10.1021/acsami.8b00492. [21] Jia LC, Li MZ, Yan DX, Cui CH, Wu HY, Li ZM. A strong and tough polymer–carbon nanotube film for flexible and efficient electromagnetic interference shielding. J Mater Chem C 2017; 5: 8944-8951. https://doi.org/10.1039/c7tc02259j. [22] Jia LC, Sun WJ, Zhou CG, Li ZM. Integrated strength and toughness in graphene/calcium alginate films for highly efficient electromagnetic interference shielding. J Mater Chem C 2018; 6(34): 9166-9174. https://doi.org/10.1039/c8tc03151g. [23] Shi YD, Lei M, Chen YF, Zhang K, Zeng JB, Wang M. Ultralow percolation threshold in poly(L-lactide)/poly(ε-caprolactone)/multiwall carbon nanotubes composites with a segregated electrically conductive network. J Phys Chem C

2017; 121(5): 3087-3098. https://doi.org/10.1021/acs.jpcc.6b11351. [24] Jia LC, Yan DX, Cui CH, Ji X, Li ZM. A unique double percolated polymer composite for highly efficient electromagnetic interference shielding. Macromol Mater Eng 2016; 301(10): 1232-1241. https://doi.org/10.1002/mame.201600145. [25] Yan DX, Pang H, Xu L, Bao Y, Ren PG, Li ZM, et al. Electromagnetic interference shielding of segregated polymer composite with an ultralow loading of in situ thermally reduced graphene oxide. Nanotechnology 2014; 25(14): 145705. https://doi.org/10.1088/0957-4484/25/14/145705. [26] Zhan CX, Yu GQ, Lu Y, Wang LY, Wujcik E, Wei SY. Conductive polymer nanocomposites: A critical review of modern advanced devices. J Mater Chem C 2017; 5(7): 1569-1585. https://doi.org/10.1039/c6tc04269d. [27] Pang H, Xu L, Yan DX, Li ZM. Conductive polymer composites with segregated structures. Prog Polym Sci 2014; 39(11): 1908-1933. https://doi.org/10.1016/j.progpolymsci.2014.07.007. [28] Thomassin JM, Jérôme C, Pardoen T, Bailly C, Huynen I, Detrembleur C. Polymer/carbon based composites as electromagnetic interference (EMI) shielding materials. Mat Sci Eng R 2013; 74(7): 211-232. https://doi.org/10.1016/j.mser.2013.06.001. [29] Wang C, Murugadoss V, Kong J, He ZF, Mai XM, Shao Q, et al. Overview of carbon nanostructures and nanocomposites for electromagnetic wave shielding. Carbon 2018; 140: 696-733. https://doi.org/10.1016/j.carbon.2018.09.006. [30] Wang YY, Wan LQ, Han DJ, Ye LY, Huang FR. Investigation of novel polytriazole resins. Des Monomers Polym 2016; 19(7): 688-695. https://doi.org/10.1080/15685551.2016.1209628. [31] Zhang CC, Wan LQ, Xu JW, Guo B, Huang FR. Preparation and characterization of polytriazole/carbon nanotube composites. Chin J Mater Res 2014; 28(7): 555-560. https://doi.org/10.11901/1005.3093.2014.118. [32] Ren F, Song DP, Li Z, Jia LC, Zhao YZ, Yan DX, et al. Synergistic effect of graphene nanosheets and carbonyl iron–nickel alloy hybrid filler on electromagnetic interference shielding and thermal conductivity of cyanate ester

composites. J Mater Chem C 2018; 6(6): 1476-1486. https://doi.org/10.1039/c7tc05213h. [33] Dresselhaus MS, Dresselhaus G, Saito R, Jorio A. Raman spectroscopy of carbon nanotubes. Phys Rep 2005; 409(2): 47-99 (2005). https://doi.org/10.1016/j.physrep.2004.10.006. [34] Duan YP, Liu SH, Guan HT. Investigation of electrical conductivity and electromagnetic shielding effectiveness of polyaniline composite. Sci Technol Adv Mat 2005; 6(5): 513-518. https://doi.org/10.1016/j.stam.2005.01.002. [35] Park SH, Thielemann P, Asbeck P, Bandaru PR. Enhanced dielectric constants and shielding effectiveness of, uniformly dispersed, functionalized carbon nanotube composites. Appl Phys Lett 2009; 94(24): 243111. https://doi.org/10.1063/1.3156032. [36] Mei H, Zhao X, Gui XC, Lu DW, Han DY, Xiao SS, et al. SiC encapsulated [email protected] ultra-high absorptive shielding material for high temperature resistant EMI shielding. Ceram Int 2019; 45(14): 17144-17151. https://doi.org/10.1016/j.ceramint.2019.05.268. [37] Xu Z, Huang YA, Yang Y, Shen JY, Tang T, Huang RS. Dispersion of iron nano-particles on expanded graphite for the shielding of electromagnetic radiation. J Magn Magn Mater 2010; 322(20): 3084-3087. https://doi.org/10.1016/j.jmmm.2010.05.034. [38] Ren F, Guo ZZ, Shi YF, Jia LC, Qing YC, Ren PG, et al. Lightweight and highly efficient electromagnetic wave-absorbing of 3D CNTs/[email protected] ternary composite aerogels. J Alloy Compd 2018; 768: 6-14. https://doi.org/10.1016/j.jallcom.2018.07.209. [39] Xu W, Chen Y, Zhan H, Wang JN. High-strength carbon nanotube film from improving alignment and densification. Nano Lett 2016; 16(2): 946-952. https://doi.org/10.1021/acs.nanolett.5b03863. [40] Shajari S, Arjmand M, Sundararaj U, Sudak LJ. Synergistic effect of hybrid stainless steel fiber and carbon nanotube on mechanical properties and electromagnetic interference shielding of polypropylene nanocomposites.

Compos Part B-Eng 2019; 165: 662-670. https://doi.org/10.1016/j.compositesb.2019.02.044. [41] Farukh M, Singh AP, Dhawan SK. Enhanced electromagnetic shielding behavior of multi-walled carbon nanotube entrenched poly (3,4-ethylenedioxythiophene) nanocomposites. Compos Sci Technol 2015; 114: 94-102. https://doi.org/10.1016/j.compscitech.2015.04.004. [42] Mohan RR, Varma SJ, Faisal M, Jayalekshmi S. Polyaniline/graphene hybrid film as an effective broadband electromagnetic shield. RSC Adv 2015; 5(8): 5917-5923. https://doi.org/10.1039/c4ra13704c.

Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: