Dielectric barrier discharge assisted continuous plasma polypyrrole deposition for the surface modification of carbon nanotube-grafted carbon fibers

    Dielectric barrier discharge assisted continuous plasma polypyrrole deposition for the surface modification of carbon nanotube-grafted carbon fibers Hande Yavuz, Gr´egory Girard, Jinbo Bai PII: DOI: Reference:

S0040-6090(16)30450-3 doi: 10.1016/j.tsf.2016.08.028 TSF 35401

To appear in:

Thin Solid Films

Received date: Revised date: Accepted date:

29 December 2015 11 August 2016 11 August 2016

Please cite this article as: Hande Yavuz, Gr´egory Girard, Jinbo Bai, Dielectric barrier discharge assisted continuous plasma polypyrrole deposition for the surface modification of carbon nanotube-grafted carbon fibers, Thin Solid Films (2016), doi: 10.1016/j.tsf.2016.08.028

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 Dielectric Barrier Discharge Assisted Continuous Plasma Polypyrrole Deposition for the Surface Modification of Carbon Nanotube-grafted Carbon Fibers Hande Yavuz, Grégory Girard, Jinbo Bai

SC R

IP

T

Laboratoire de Mécanique des Sols, Structures et Matériaux, École Centrale Paris, CNRS UMR8579, Grande Voie des Vignes, 92290, Châtenay-Malabry, France

Abstract

The incorporation of carbon nanotube-grafted carbon fibers (CNT-CF) into polymer matrices

NU

provides highly-enhanced physical properties to the materials. They could be ideal candidates

MA

to be integrated into multi-functional composites applications in several industries. However, in order to take advantage of carbon nanotubes (CNTs), it is essential to keep all the CNTs on the surface of fibers during their implementation phase in the composite manufacturing

TE

D

process. Besides, some unresolved safety issues remain during the integration phase of such hybrid forms into polymer matrices. This research is making an attempt to develop an

CE P

effective plasma surface treatment method either to keep the maximum possible amount of CNTs on the fibers and to modify the surface properties of CNT-CF which is highly necessary

AC

prior to the composite fabrication. In order to provide better understanding of the modified structures in a domain ranging from microscopic to atomic scales, several characterization studies were realized by X-ray Photoelectron Spectroscopy (chemical structure) and Scanning Electron Microscopy (microstructure). This research is expected to provide valuable information for further studies to develop hybrid composites where multifunctionality is the main concern.

Keywords: Plasma assisted deposition; Dielectric barrier discharge; Carbon nanotube grafted fiber; Plasma polypyrrole.

1

ACCEPTED MANUSCRIPT 1. Introduction Carbon nanotubes (CNTs) vary in their morphology and their structural characteristics.

T

Resulting from these properties; chiral angle, tube diameter, C–C bond length of the

IP

hexagonal lattice, tube length, tube-end configuration (end-caps), as well as nanotube wall

SC R

structure are some of the parameters which can be used to characterize the nanotubes [1, 2]. Regarding their configuration (i.e., chiral angle and diameter), these nanotubes can possess different level of metallization (i.e., metallic, semi-metallic or semiconducting) [1]. They can

NU

produce conductive capabilities which can be characterized by several techniques [3-5]. CNTs

MA

can also ensure high elastic and strength properties [1, 6]. It has been found that when uniform forces are applied on the outer layer of multi-walled carbon nanotubes, the inner layer exhibits small axial displacements. This revealed that the force acting on the outer layer

TE

D

is transferred to the inner layer through long range intermolecular forces such as Van der Waals interactions [1]. Thanks to CNTs’ superior mechanical, electrical and thermal

CE P

properties, they have found great potential to be embedded in a wide variety of structural applications in order to obtain high performance, multi-functional, and lightweight

AC

components [7].

When CNTs are desired to be used in structural composites, one of the most important criteria is to increase their surface adhesion capability prior to manufacturing phases such as weaving and impregnation. In order to enhance the interfacial strength between the CNTs and polymer matrix, the amount of good bonding between CNTs and polymer matrix must be reached, and less graphitized more rough CNT surfaces should be provided. Therefore, the integration of functional groups such as hydroxyl, carboxyl, nitrogen or fluorine groups on the outer layers of CNTs by plasma modification [8-20] and the incorporation of nanowires [21] are the promising methods within the field of functionalization processes which may further give possibility to improve the micro/nanomechanical properties of polymer composites.

2

ACCEPTED MANUSCRIPT Either to quantify or qualify such properties of composites, nanoindentation tests can be combined with imaging techniques such as atomic force microscopy and scanning electron

T

microscopy [22].

IP

Plasma processing can be considered as a promising tool in materials surface treatment

SC R

because of the comparatively straightforward scale-up to larger dimensions, possibility of continuous surface treatment and compatibility of any type of substrate. Moreover, with well designed plasma conditions, functional structure deposition on materials is also possible in a

NU

very short time. This research demonstrates the possibility of plasma polypyrrole deposition at

MA

atmospheric pressure and characterization of the chemical and morphological properties. It is also stated that the encapsulation of CNTs diminishes the risk to inhale free CNTs that exists on the carbon fibers subsequent to chemical vapor deposition (CVD).

TE

D

2. Experimental 2.1. Materials

CE P

Polyacrylonitrile based Hexcel AS4-12K (AF) and Toray T700-12K (TF) fibers were used throughout the investigation. CNT-grafted fibers were chosen to be prepared from TF

AC

fibers (TF/CNT) by a CVD method [23]. Pyrrole (C4H5N, 98+ %, liquid at STP) and ptoluene sulfonic acid monohydrate (C7H8O3S.H2O, 98+ %, solid at STP, referred as dopant or pTSA.H2O throughout the paper) were purchased from Sigma-Aldrich. Both precursor was heated prior to plasma treatment and was fed easily into the plasma post-discharge region in the vapor phase. Nitrogen (Alphagaz B50, 9.4 m3, 20 MPa) which was used as a process gas for the initiation of plasma discharge and as a carrier gas for the feeding of precursors contained the following impurities: H2O < 3 ppm, O2 < 2 ppm, CnHn < 0.5 ppm. 2.2. Experimental set-up The plasma generation system used to treat AF and TF/CNT fibers was based on dielectric barrier discharge technique. Plasma generator and plasma source (Model ULD-120,

3

ACCEPTED MANUSCRIPT processed length is 120 mm) was developed by Acxys Technologies (Saint-Martin-le-Vinoux, France). Plasma generator consists of electric supply, gas control and touch screen; plasma

T

source consists of process gas distribution chamber and water cooled co-axial cylindrical

IP

electrodes. The atmospheric pressure glow discharge was produced between two co-axial

SC R

water-cooled cylindrical electrodes located inside the source. An insulator material covers the inner metal electrode and surrounded by a cylindrical slotted metal electrode. The inner electrode was connected to a power supply consisting of a low frequency generator (120

NU

kHz), a high voltage transformer (+/- 3.5 kV) and a power amplifier. The injected power can

MA

be adjusted from 1000 to 2000 W. The treatment duration per 120-mm-long sample was chosen to be between 45 s (treatment speed: 0.16 m/min) and 60 s (treatment speed: 0.12 m/min) for this study. N2 was used as process gas, pyrrole and pTSA.H2O were used for

TE

D

plasma deposition (Figure 1). Flow was adjusted by means of mass flowmeters and recorded using the LabView software. In the presence of dopant, injection points of precursor and

CE P

dopant were separated from each-other in order to avoid possible oxidation reactions between them. After each experiment, plasma treated fibers were washed with acetonitrile (Cas no. 75-

AC

05-8) and the solution was injected into a gas chromatograph in order to identify the solubility of the deposited layer. According to the results, no peaks were observed except acetonitrile which was an indication of insolubility of coating in that solvent. 2.3. Apparatus 2.3.1. X-ray photoelectron spectroscopy (XPS) The sample surfaces were analyzed with the Physical Electronics PHI 5000 Versaprobe X-ray photoelectron spectrometer. The spectra were collected using a monochromatic Al Kα X-ray source (hv =1486.6 eV) operated at 15 kV and 25 W. The pass energy for collecting survey and high resolution spectra were 187.85 eV (step size = 0.4 eV, time/step=20 ms) and 29.35 eV (step size=0.1 eV, time/step=50 ms), respectively. A combination of 1 eV electrons

4

ACCEPTED MANUSCRIPT and 6 eV Ar+ ions were used for the charge neutralization. For calibration purposes, the C 1s electron binding energy was referenced at 284.8 eV. Curve fitting was performed using

T

Multipak software (developed by Ulvac-Phi). The quantitative analysis of atomic surface

IP

composition of carbon fibers (C 1s, O 1s and N 1s photoelectron peaks) was calculated by

SC R

integration of the peak areas with Shirley method on the basis of Wagner and Scofield sensitivity factors. 2.3.2. Scanning electron microscopy (SEM)

NU

The topography of unmodified and plasma modified AF and TF/CNT surfaces was

MA

examined by LEO Gemini 1530 scanning electron microscope. Fiber samples were fixed on aluminum holders without any metallization prior to SEM examination. Magnification of the collected SEM images was ranging between 2 kX and 50 kX.

TE

D

3. Results and Discussion

3.1. X-ray Photoelectron Spectroscopy

CE P

The changes in the elemental composition of TF/CNT and AF fibers with plasma treatment were studied to examine the surface of these fibers treated at different plasma

AC

conditions by electron spectroscopy for chemical analysis (XPS). Untreated samples were also examined to compare chemical differences between unmodified and plasma modified fibers. The surface of untreated TF/CNT fibers were composed mainly of carbon in C-C/C-H and C-O/C-OH or C-N/C=N which were about 82 % and 9 %, respectively. In the case of plasma polypyrrole deposited fibers, the content of both oxygen and nitrogen based functional groups were increased compared to untreated fibers. The increase in the concentration of oxygen and nitrogen based active groups may alter the polarity and the structure of the fiber surface. Therefore, all results obtained showed that not only the plasma polypyrrole (PPPy) deposited on carbon nanotubes, but also it may improve the adhesion between matrix and fibers as a result of produced active groups on these fiber surfaces.

5

ACCEPTED MANUSCRIPT The elemental composition and the N/C and O/C of the studied fibers are represented in Table 1. In the case of untreated TF/CNT fiber, carbon, nitrogen and oxygen content was

T

found 96.0 %, 1.5 %, and 2.5 %, respectively. Survey spectra showed that the surface of

IP

untreated TF/CNT fiber contains low concentration of nitrogen (<1.5 %) and oxygen (<2.5 %)

SC R

which indicated that few nitrogen and oxygen based functional groups were presented on the surface of TF/CNT fibers compared to plasma treated fibers. The XPS measurements of plasma modified samples exhibited that the surface composition of each fiber was changed

NU

substantially during the plasma treatment (Figure 2). For the pyrrole plasma treated TF/CNT

MA

fibers, carbon to nitrogen ratio was ranging between 2 and 15 while it was equal to 64 for untreated TF/CNT fibers. The values of the binding energy (B.E.), the peak assignment (P.A.) and the relative area (R.A.) of each photopeak estimated from the curve fit of C 1s, N 1s, and

TE

D

O 1s for plasma modified TF/CNT samples are represented in Table 2. Increase of oxygen and nitrogen content in all plasma polymers is an indication of plasma polypyrrole deposition

CE P

(not ideal polypyrrole) which accounts for several advantages such as the improvement of adhesion between matrix and the fiber [24]. Neither the precursor pyrrole nor the carrier gas

AC

contains oxygen, but oxygen was detected at a considerable amount in all XPS measurements of plasma treated samples. Since all experiments took place in the open-air configuration, the most probable reason of the presence of oxygen peak in each spectrum would be related to the atmospheric oxygen and/or water vapor which might be trapped in the deposited layer during the plasma operation. Besides, it has been known that free radicals may remain on the plasma polymer surface and these radicals possess a long half-life which could be susceptible to react with oxygen in air and further form carbonyl, hydroxyl, and carboxyl groups [8, 9, 25]. When high resolution spectra of untreated TF/CNT were examined, some peaks appeared to be multiple peaks (Figure 3a-c). Carbon gave a major peak at 284.6 eV corresponds to aromatic, graphitic and/or aliphatic carbon (C=C, C-C, C-H) and the others

6

ACCEPTED MANUSCRIPT were detected around 286.3, 287.7, and 289.3 corresponds to C-O, C=O, and COO (i.e., carbonyl compounds) in ester and/or anhydrides, respectively [26]. Additionally, the

T

asymmetry of C 1s line (between the low and high energy levels) and the presence of π-

IP

plasmon at high energy levels (Figure 3a) may be related to electronic transitions in multi-

SC R

walled carbon nanotubes due to conductivity electrons that are located at the Fermi level, in the same fashion as in transition metals [27].

The TF/CNT fiber treated at 1400 W 45 s pyrrole plasma, carbon, nitrogen and oxygen

NU

content was found about 61 %, 21 %, and 18 %, respectively. Survey spectra showed that the

MA

surface of these fibers contains highly nitrogen (>21 %) and oxygen (>17 %) based functional groups on the surface of these fibers. When high resolution spectra were explored (Figure 3df), C 1s, N 1s, and O 1s peaks were most likely to be decomposed into five, three, and two

TE

D

lines, respectively. Carbon gave a major peak at 284.7 eV which may correspond to α-carbons in pyrrole ring and/or aromatic, graphitic and aliphatic carbons originated from TF/CNT fiber.

CE P

The lowest energy component of C 1s peak detected around 283.6 eV may correspond to βcarbons in pyrrole ring [28, 29]. The third line of C 1s located at 286.2 eV might be an

AC

overlap of C=N, C-N, C-OH or C-O groups [28] or it could be originated from the presence of polaron species =C-NH.+ [30]. The fourth line at 287.6 eV may be attributed to C=O, C=N or the excited state of C-N (i.e., C-N*) in pyrrole ring [31]. The highest energy component of C 1s photopeak gave a line around 288.8 eV which indicated the presence of carbonyl and/or ether groups in esters and anhydrides [26]. The major peak of N 1s (399.2 eV) was assigned to -NH- groups of pyrrole [32]. The lowest energy component of N 1s peak (~398 eV) may be attributed to N=C whereas the highest energy component (~400.4 eV) may correspond to the nitrogen atoms that are more positively charged than the N atoms of the main pyrrole peak [28, 31, 32]. Decomposition of O 1s peak indicates an overlap of O-N, O-C, and O=C groups

7

ACCEPTED MANUSCRIPT at 532.34 eV and the important signal of O 1s around 530.9 eV indicated the surface reaction with oxygen was likely to occur due to the open-air configuration of plasma system [32].

T

Survey spectrum of 1400 W 60 s pyrrole plasma treated TF/CNT fiber surface was

IP

composed of 71 % carbon, ~7 % nitrogen and ~22 % oxygen. The line shape analysis of C 1s

SC R

photopeak (Figure 3g-i) revealed that it can be decomposed into four lines and symmetric in both high and low binding energy sides. Carbon gave a major peak around 283.8 eV and another important peak at 285.0 eV may correspond to α-carbons and β-carbons in pyrrole

NU

ring, respectively [28, 31]. Similar to 1400 W 45 s treated sample, the third line of C 1s is

MA

located at 286.2 eV and it might be an overlap of C=N, C-N, C-OH or C-O groups [28]. The highest energy component of C 1s peak around 287.7 eV is probably due to the overlap of C=O, C=N or the excited state of C-N (i.e., C-N*) [31]. The main peak of N 1s (399.1 eV)

TE

D

could be attributable to -NH- groups of pyrrole [32]. The lowest energy component of N 1s peak (~398 eV) was attributed to N=C whereas the highest energy component (~400.4 eV)

CE P

may be corresponded to the nitrogen atoms that are more positively charged than the N atoms of the main pyrrole peak as it was observed in the other plasma treated samples [28, 31, 32].

AC

Decomposition of O 1s peak indicated an overlap of O-N, O-C, and O=C groups (532.3 eV) and similar to 1400 W 45 s treated TF/CNT fibers. The presence of O 1s signal around 530.9 eV is an indication of the surface reaction with oxygen due to the air exposure [32]. Survey spectrum of 1800 W 45 s pyrrole plasma treated TF/CNT fiber surface showed a composition of ~76 % carbon, ~5 % nitrogen and ~19 % oxygen. The C 1s high resolution spectrum was resolved in four peaks (Figure 3j-l). Besides, the fifth one (~291.6 eV, 0.5 %) could be attributed to the presence of the shake-up satellites (π-π*) [29]. When the high resolution C 1s photopeak was explored, the increase in binding energies showed that the C 1s orbital assume different conformations. In the case of 1800 W 45 s treated fiber surface, the lowest energy component (~285.2 eV) can be associated with carbon bonds in Cβ-COOH or

8

ACCEPTED MANUSCRIPT α- carbons in pyrrole ring [28, 31]. Because, it is noted that in all components of C 1s peaks the value of full width at half maximum (FWHM) was equal to 1.72 eV whereas the FWHM

T

values of the components of C 1s peaks in other samples were varied between 1.44 (for

IP

untreated TF/CNT) and 1.58 eV. Due to the high FWHM value of C 1s peak components of

SC R

1800 W 45 s treated sample, the components of C 1s peak may be shifted to higher energy levels compared to the other plasma treated samples. Therefore, the peak around 285.2 eV might also correspond to α-carbons in pyrrole ring. The clear shift can be seen in Figure 2

NU

which provided the comparison of C 1s curve fit spectrum of 1800 W 45 s plasma treated

MA

fiber with the other TF/CNT plasma treated fibers. The peaks recorded around 286.6 eV, 288.0 eV, 289.3 eV may be attributed to the similar functional groups which has already been discussed above. The main peak of N 1s (~400.5 eV) might be assigned to nitrogen atoms that

TE

D

are more positively charged than the atoms of the other components of N 1s peak [28, 31, 32]. The lowest energy component of N 1s peak (~399.3 eV) may correspond to -NH- groups of

CE P

pyrrole [32]. The highest binding region (~401.8 eV) of N 1s core level may be related to the extra charges that were presented on the selected N atoms in the deposited layer (Figure 3k).

AC

However, this charge does not occur at specific nitrogen sites, involves the whole pyrrole ring unit [28]. The main peak of O 1s (532.2 eV) was assigned to the O-N, O-C or O=C groups (Figure 3l). The second line, located at 533.6 eV might correspond to carbonyl compounds, alcohols or ethers [26]. In order to find out the effect of dopant, TF/CNT and AF fibers were treated at the same plasma conditions (1400 W 45 s) and examined by using XPS spectroscopy. At first, the absence of sulfur atom on their survey spectrum and decrease observed in their oxygen at- % (about 5 %) draws attention (Table 1, Figure 3o and 4f). Even though dopant (pTSA.H2O) contains sulphur, in none of their XPS spectrum sulphur atom was detected (Table 1 and Table 3). Therefore, it can be concluded that pTSA.H2O was not penetrated inside the plasma

9

ACCEPTED MANUSCRIPT polypyrrole as a tosylate ion (dopant ion); it might partly be dissociated into gaseous products (e.g., SO2) and lost during the plasma polymerization of pyrrole. During this process, it may

T

affect the structure by creating defectively α-β bonded carbon atoms (around 285.8 eV) [33]

IP

which has not been detected in the other plasma treated samples. As evidenced by the highly

SC R

similar decomposition of C 1s, N 1s and O 1s photopeaks, it can be said that the surface chemistry of those fibers have been changed in the same manner. As it was found by XPS curve fit, untreated AF has less aromatic and/or graphitic structure than TF/CNT fibers due to

NU

the lower C=C and C-C groups (Table 2 and 3). It was also found that the surface of as-

MA

received AF fiber includes oxygen based functionals at a considerable amount (~10 %) due to the surface oxidation carried out by the supplier [26]. Comparison of the asymmetric spectral lines indicates some important consequences.

TE

D

The C 1s spectrum was asymmetric on the high binding energy side which could be an indication of crosslinked chain terminating, non-α-α’-bonded carbons or carbons in partially

CE P

saturated rings in polypyrrole materials. These carbons are usually referred to measures of disorder of the material [28]. In the N 1s spectrum, the asymmetric shape implies the

AC

electrostatic interactions occurred in polypyrrole materials [28, 34]. The explored peak shapes in XPS spectra showed that the pyrrole plasma treated TF/CNT fibers have more symmetrical C 1s and N 1s peaks than that of untreated TF/CNT fibers due to the presence of relatively ordered polymer deposition. It was also shown that the most ordered material among the pyrrole plasma treated samples was belonged to 1400 W 60 s sample due to the obtained somewhat symmetrical peak shape compared to the others. It is also noted that the presence of the shake-up satellites (π-π*) around 291 eV might be due to electronic transitions in PPy ring [29] and/or electronic transitions (i.e., π-plasmon) in the carbon nanotubes [27].

3.2. Results of SEM

10

ACCEPTED MANUSCRIPT The SEM micrographs of plasma treated TF/CNT and AF fibers vary between 2 kX and 50 kX and are shown in Figures 5 and 6. Almost all pyrrole plasma treated TF/CNT samples

T

were exhibited tiny spherolites which is one of the common morphology observed in

IP

electrochemically polymerized polypyrrole [35] and plasma polymerized [36, 37]. Since the

SC R

deposited layer was not as thick as that was reported in literature, the tiny spherical formations of polypyrrole may not be as same as in those micrographs given in literature [36, 37]. On the other hand, the plasma polymerization of pyrrole on AF fibers clearly indicated

NU

that the polypyrrole spherolites (20-180 nm) formed on these fibers (Figure 6). In the plasma

MA

polymerization of pyrrole+dopant, these tiny spherolites have become smaller (20-60 nm) most likely due to the presence of dopant which affect the reactions in the post-discharge and resulted in tiny spherolites (Figure 6).

TE

D

The structure of deposited polymer on TF/CNT and AF fibers may not be ideal polypyrrole as typically observed in electrochemically polymerized polypyrrole, but it may

CE P

contain cross-linking as expected from the plasma polymers [38]. The possibility of trapped oligomers in the polymer, the presence of amorphous carbon deposits on carbon nanotubes

AC

due to CVD process and the agglomerated carbon nanotubes on TF/CNT fibers might also be contributed to the morphological differences observed on these micrographs. The plasma polypyrrole that was formed under the applied conditions can be explained by the diffusion controlled nucleation growth at the surface of the material. In the case of TF/CNT fibers, the carbon nanotubes on the fiber surface possibly serve as nucleation sites during the pyrrole polymerization.

4. Conclusions In the frame of this work, we adapted a dielectric barrier discharge technique for continuous plasma polypyrrole deposition on the surface CFs and CNT-CF. Then, we focused on the characterization of plasma polypyrrole deposited CFs and CNT-CF fibers by using

11

ACCEPTED MANUSCRIPT several techniques such as XPS and SEM. From low to high plasma powers (1400 W to 1800 W), deposited polymer tend to lose the memory of its monomeric structure probably due to

T

the loss of α and β-carbons in pyrrole ring. This kind of change might be due to more ring

IP

opening reactions and high loss of nitrogen-containing species at higher power levels.

SC R

Regardless of exposure time, this sharp change was not observed at low powers. We found that p-toluene sulfonic acid monohydrate (pTSA. H2O) was not acted as a dopant in the form of tosylate ion. We also diminished the risk to inhale free carbon nanotubes prior to composite

NU

manufacturing by encapsulating them into plasma polypyrrole structures.

MA

Acknowledgments

This research was supported under EADS FRANCE ANR PROCOM Project (contact grant no. ANR-08-MAPR-0025). The financial support from French Government and

TE

D

technical support from Acxys Technologies are gratefully acknowledged. We thank Mme. Françoise Garnier from ECP Lab. MSSMat, CNRS UMR 8579 and M. David Alamarguy

assistance.

AC

References

CE P

from CNRS UMR 8507 CentraleSupelec – Universites UPMC et UPSud for their technical

1. E.T. Thostenson, C. Li, T.-W. Chou, Nanocomposites in context (Review), Comp. Sci. & Tech. 65 (2005) 491-516.

2. H.S. Nalwa, Handbook of Nanostructured Materials and Nanotechnology, in: P.M. Ajayan (Ed.), Carbon Nanotubes, Academic Press, USA, 2000, pp. 375 – 403.

3. H. Dai, E.W. Wong, C.M. Lieber, Probing Electrical Transport in Nanomaterials: Conductivity of Individual Carbon Nanotubes, Science 272 (1996) 523–526.

4. J. Baker-Jarvis, M.D. Janezic, J.H. Lehman, Dielectric resonator method for measuring the electrical conductivity of carbon nanotubes from microwave to millimeter frequencies, J. Nanomaterials, vol. 2007, Article ID 24242, 2007.

12

ACCEPTED MANUSCRIPT 5. A. Allaoui, S.V. Hoa, P. Evesque, J. Bai, Electronic transport in carbon nanotube tangles under compression: The role of contact resistance, Scr. Mater. 61 (2009) 628–631.

IP

T

6. X.L. Xie, Y.W. Mai, X.P. Zhou, Dispersion and alignment of carbon nanotubes in polymer matrix: A review, Mat. Sci. and Eng. R 49 (2005) 89–112.

SC R

7. J.H. Du, J. Bai, H.M. Cheng, The present status and key problems of carbon nanotube based polymer composites, eXPRESS Pol. Lett. 1 (2007) 253–273.

NU

8. R. Ionescu, E.H. Espinosa, E. Sotter, E. Llobet, X. Vilanova, X. Correig, A. Felten, C. Bittencourt, G. Van Lier, J.C. Charlier, J.J. Pireaux, Oxygen functionalisation of MWNT and their use as gas sensitive thick-film layers, Sens. and Act. B 113 (2006) 36–46.

MA

9. T. Xu, J. Yang, J. Liu, Q. Fu, Surface modification of multi-walled carbon nanotubes by O2 plasma App. Surf. Sci. 253 (2007) 8945–8951.

TE

D

10. J.A. Kim, D.G. Seong, T.J. Kang, J.R. Youn, Effects of surface modification on rheological and mechanical properties of CNT/epoxy composites, Carbon 44 (2006) 1898–1905.

CE P

11. H. Bubert, S. Haiber, W. Brandl, G. Marginean, M. Heintze, V. Brüser, Characterization of the uppermost layer of plasma-treated carbon nanotubes, Dia. and Rel. Mat. 12 (2003) 811–815.

AC

12. G. Kalita, S. Adhikari, H.R. Aryal, R. Afre, T. Soga, M. Sharon, M. Umeno, Functionalization of multi-walled carbon nanotubes (MWCNTs) with nitrogen plasma for photovoltaic device application, Curr. App. Phy. 9 (2009) 346–351.

13. N.O.V. Plank, R. Cheung, Functionalisation of carbon nanotubes for molecular electronics, Micro. Eng. 73–74 (2004) 578–582.

14. Y.W. Zhu, F.C. Cheong, T. Yu, X. J. Xu, C.T. Lim, J.T.L. Thong, Z.X. Shen, C.K. Ong, Y.J. Liu, A.T.S. Wee, C.H. Sow, Effects of CF4 plasma on the field emission properties of aligned multi-wall carbon nanotube films, Carbon 43 (2005) 395–400.

15. Y.C. Hong, D.H. Shin, H.S. Uhm, Super-hydrophobicity of multi-walled carbon nanotubes treated by a glow discharge, Surf. & Coat. Tech. 201 (2007) 5025–5029.

13

ACCEPTED MANUSCRIPT 16. A. Felten, J. Ghijsen, J.J. Pireaux, R.L. Johnson, C.M. Whelan, D. Liang, G. Van Tendeloo, C. Bittencourt, Photoemission study of CF4 RF-Plasma treated multi-wall carbon nanotubes, Carbon 46 (2008) 1271–1275.

SC R

IP

T

17. L. Valentini, D. Puglia, F. Carniato, E. Boccaleri, L. Marchese, J. M. Kenny, Use of plasma fluorinated single-walled carbon nanotubes for the preparation of nanocomposites with epoxy matrix, Comp. Sci. and Tech. 68 (2008) 1008–1014.

NU

18. W.J. Chou, C.C. Wang, C.Y. Chen, Characteristics of polyimide-based nanocomposites containing plasma-modified multi-walled carbon nanotubes, Comp. Sci. and Tech. 68 (2008) 2208–2213.

MA

19. D. Shi, J. Lian, P. He, L.M. Wang, W.J. van Ooij, M. Schulz, Y. Liu, D.B. Mast, Plasma deposition of ultrathin polymer films on carbon nanotubes, App. Phy. Lett. 81 (2002) 1–3.

TE

D

20. K. Yua, Z. Zhu, Y. Zhang, Q. Li, W. Wang, L. Luo, X. Yu, H. Ma, Z. Li, T. Feng, Change of surface morphology and field emission property of carbon nanotube films treated using a hydrogen plasma App. Surf. Sci. 225 (2004) 380–388.

CE P

21. X. Tao, L. Dong, X. Wang, W. Zhang, B. J. Nelson, X. Li, B4C-Nanowires/Carbonmicrofiber hybrid structures and composites from cotton T-shirts, Adv. Mat. 22 (2010) 2055–2059.

AC

22. X. Li, P. Nardi, Micro/nanomechanical characterization of a natural nanocomposite material – the shell of Pectinidae, Nanotechnology 15 (2004) 211–217. 23. J. Bai, “Procede de synthese de nanotubes de carbone sur materiaux micrometriques longues et particuliares”, FR2939422-A1, issued date June 11, 2010.

24. X. Zhang, Y. Huang, T. Wang, Surface analysis of plasma grafted carbon fiber, App. Surf. Sci. 253 (2006) 2885–2892. 25. H. Yasuda, in Plasma chemistry of polymers, Ed., M. Shen, Marcell Dekker Inc., New York, 1976. 26. U. Zielke, K.J. Hüttinger, W.P. Hoffman, Surface oxidized carbon fibers: I. Surface structure and chemistry, Carbon 34 (1996) 983–998.

27. Y.M. Shulga, T.-C. Tien, C.-C. Huang, S.-C. Lo, V.E. Muradyan, N.V. Polyakova, Y.-C. Ling, R.O. Loutfy, A.P. Moravsky, XPS study of fluorinated carbon multi-walled nanotubes, J. Elect. Spect. and Rel. Phen. 160 (2007) 22–28. 14

ACCEPTED MANUSCRIPT

T

28. J. Zhou and E.R. Fisher, Synthesis and Properties of PPPy/Au Composite Nanofibers, J. Nano. Nanotechnol. 4 (2004) 539–547.

SC R

IP

29. P. Lisboa, G. Gilliland, G. Ceccone, A. Valsesia, F. Rossi, Surface functionalization of polypyrrole films using UV-light induced radical activation, App. Surf. Sci. 252 (2006) 4397–4401.

NU

30. L. Ruangchuay, J. Schwank, A. Sirivat, Surface degradation of a-naphthalene sulfonatedoped polypyrrole during XPS characterization App. Surf. Sci. 199 (2002) 128–137.

MA

31. M. Vasquez, G.J. Cruz, M.G. Olayo, T. Timoshina, J. Morales, R. Olayo, Chlorine dopants in plasma synthesized heteroaromatic polymers, Polym. 47 (2006) 7864–7870.

D

32. J. Wang, K.G. Neoh, E.T. Kang, Comparative study of chemically synthesized and plasma-polymerized pyrrole and thiophene thin films, Th. Sol. Fil. 446 (2004) 205–217.

CE P

TE

33. E. Pigois-Landureau, Y.F. Nicolau, M. Delamar, XPS study of layer-by-layer deposited polypyrrole thin films, Synth. Met. 72 (1995) 111–119.

AC

34. E. Benseddik, M. Makhlouki, J.C. Bernede, S. Lefrant, A. Pron, XPS studies of environmental stability of polypyrrole-poly(vinyl alcohol) composites Synth. Met. 72 (1995) 237–242.

35. B. Lin, R. Sureshkumar, J.L. Kardos, Electropolymerization of pyrrole on PAN-based carbon fibers: experimental observations and a multiscale modeling approach Chem. Eng. Sci. 56 (2001) 6563–6575.

36. K. Hosono, I. Matsubara, N. Murayama, W. Shin, N. Izu, S. Kanzaki, Structure and properties of plasma polymerized and 4-ethylbenzenesulfonicacid-doped polypyrrole films, Th. Sol. Fil. 441 (2003) 72–75.

37. K. Hosono, I. Matsubara, N. Murayama, W. Shin, N. Izu, Effects of discharge power on the structure and electrical properties of plasma polymerized polypyrrole films, Mater. Lett. 58 (2004) 1371–1374. 38. H. Biederman, D. lav ns , Plasma polymer films and their future prospects, Surf. and Coat. Tech. 125 (2000) 371–376.

15

ACCEPTED MANUSCRIPT List of Figure Captions

IP

Figure 2. High resolution C 1s comparison spectra of TF/CNT fibers

T

Figure 1. Plasma treatment system used for the modification of CFs and CNT-CFs

SC R

Figure 3. High resolution C 1s, N 1s, and O 1s spectra of TF/CNT fibers; untreated (a, b, c); 1400 W 45 s (d, e, f); 1400 W 60 s (g, h, i); 1800 W 45 s (j, k, l), 1400 W 45 s + dopant (m, n, o). Note that the row-wise graph numbering corresponds to C 1s, N 1s and O 1s photopeaks, respectively.

NU

Figure 4. High resolution C 1s, N 1s, O 1s spectra of AF fibers; untreated (a, b, c); 1400 W 45 s + dopant (d, e, f). Note that the row-wise graph numbering corresponds to C 1s, N 1s and O 1s photopeaks, respectively.

MA

Figure 5. SEM micrograph of TF/CNT fibers; (a) untreated, (b) 1400 W 45 s, (c) 1400W 45 s + dopant, (d) 1800 W 45 s

AC

CE P

TE

D

Figure 6. SEM micrograph of AF fibers; (a) untreated, (b) 1400 W 45 s, (c) 1400 W 45 s + dopant

16

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

17

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

18

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

19

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

20

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

21

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

22

ACCEPTED MANUSCRIPT Tables

C at %

N at %

O at %

96.0

1.5

2.5

64.0

38.4

21.1

17.8

2.8

3.4

71.0

7.2

21.8

9.8

3.2

75.5

5.2

19.3

14.5

3.9

70.8

16.4

12.8

4.3

5.5

85.4

4.9

9.7

17.4

8.8

71.7

16.2

12.2

4.4

5.8

Untreated 1400 W 45 s TF/CNT

Pyrrole

1400 W 60 s 1800 W 45 s 1400 W 45 s

NU

Pyrrole + pTSA.H2O Untreated AF

1400 W 45 s

AC

CE P

TE

D

MA

Pyrrole + pTSA.H2O

23

61.1

T

Plasma conditions

Precursor

IP

Fiber

SC R

Table 1. XPS results for surface composition of TF/CNT and AF fibers treated at different conditions C/N

C/O

ACCEPTED MANUSCRIPT Table 2. XPS C 1s, N 1s, and O 1s curve fit results of untreated and plasma treated TF/CNT fibers at different conditions

3

4

5

1

B.E. (eV)

284. 6

286. 2

287. 7

289. 3

290.9

398. 5

R.A (%)

81.6 6

9.0

3.71

2.39

3.24

C-C, P.A C-H

284. 7

286. 2

R.A (%)

7.47

44.7

16.9 7

CO, COH, CN, C=N

2

3

401.1

plasmo n

33.6 5

N=C

66.35

1

2

530. 6

532. 5

24.2 4

76.7 6

Graphitic N

O 1s

Or

OC, O=C

287. 6

288.8

397. 9

399.2

400. 4

530. 9

532. 3

19.7

11.16

17.8 2

58.19

23.9 9

72.1 2

27.8 8

CN*, C=N , C=O

ON, N=C CO-O

-NH-

N+

O 1s

=N-

OC, O=C

=CNH. +

B.E. (eV)

283. 8

285

286. 1

287. 7

398. 0

399.1

400. 4

530. 9

532. 3

R.A (%)

52.8 8

22.0 4

7.22

17.8 6

25.3 8

63.1

11.5 2

54.6 4

45.3 6

Cα

CO, COH, CN, C=N

P.A

Cβ

3

ON,

TE

CE P Cβ

AC

P.A

Cα and/o r C-C, C-H

1400 W 60 s pyrrole

COO

C=O

283. 6

1400 W 45 s

TF/CNT

C=N ,

B.E. (eV)

TF/CNT

pyrrole

C-O, COH, C-N, C=N

NU

untreated

T

2

O 1s photopeak

SC R

1

MA

TF/CNT

Propert y

D

Sample

N 1s photopeak

IP

C 1s photopeak

CN*,

ON, N=C

C=N ,

-NH=N-

C=O

C 1s photopeak

N+

O 1s

OC, O=C

N 1s photopeak

O 1s photopeak24

ACCEPTED MANUSCRIPT 2

3

4

5

1

2

3

1

2

B.E. (eV)

285.2

286. 6

288. 0

289. 3

291.6

399. 3

400. 5

401.8

532. 2

533. 6

R.A (%)

65.8

16.0 5

8.44

9.16

0.54

35.4 4

56.2 5

8.31

58.8 5

41.1 5

Cα

TF/CNT

IP plasm on

C= O

NH-

N+

O-N,

NH2/N H3

+

COH,

O=C

OH, ROR’

O-C,

3

B.E. (eV)

284.5

285. 8

287. 2

288. 5

289.9

398. 7

399. 9

400.9

531. 4

532. 1

533. 4

R.A (%)

52.94

20.5 7

11.2 6

12.9 9

2.25

22.9 4

58.7 7

18.29

42.2 0

40.8 9

16.9 1

ON,

COH,

Cαβ

CN*, C= N, C= O

OC,

OH, ROR’

1400 W 45 s

C-C, C-H, P.A

Cβ

COO

plasm on

N= C

NH-

N+

O=C =

=NO= C

CE P

pyrrole+ dopant

COO

SC R

P.A

CN*, C= N,

MA

pyrrole

CβCOOH,

CO, COH, CN, C= N

NU

1800 W 45 s

T

1

D

TF/CNT

Propert y

TE

Sample

Table 2 contd. XPS C 1s, N 1s, and O 1s curve fit results of untreated and plasma treated TF/CNT

AC

fibers at different conditions

Table 3. XPS C 1s, N1s, and O 1s curve fit results of untreated and plasma treated AF fiber C 1s photopeak Sample

Propert y

1

2

3

4

AF

B.E. (eV)

284. 5

285. 8

287. 3

R.A

73.3

16.4

6.81

untreated

N 1s photopeak 5

O 1s photopeak

1

2

3

1

2

3

288. 7

399. 2

400.6

402. 3

531. 2

532. 2

533. 5

3.41

42.7

48.74

8.52

64.9

29.8

5.20

25

ACCEPTED MANUSCRIPT

CC,

5

CO R

C-H

CO O

NH-

284. 7

286. 2

287. 8

288. 9

290.9

398. 4

R.A (%)

63.0 0

14.5 5

12.5 9

8.26

1.50

29.0 1

Cαβ

CN*, C= N, C= O

P.A

CH,

COO

NU

CC,

Cβ

-NH2

NOx

O=C =

plasm on

N= C

399.7 54.57

-NH-

1 ON,

COH,

OC,

-OH

O= C

ROR’

400. 7

531. 1

532. 3

533. 7

16.4 2

47.0 8

39.9 8

12.9 3

ON,

COH,

OC,

-OH

N+

O=C =

=NO= C

ROR’

AC

CE P

TE

D

*excited states [31]

9

SC R

B.E. (eV)

1400 W 45 s pyrrole + dopant

C= N, C= O

MA

AF

4

T

P.A

3

IP

(%)

26

ACCEPTED MANUSCRIPT Highlights

CE P

TE

D

MA

NU

SC R

IP

T

Hybrid materials were modified by a dielectric barrier discharge plasma generator. Carbon nanotubes (CNTs) were encapsulated by plasma polypyrrole deposition. Encapsulation dimisnishes the risk to inhale CNTs prior to composite manufacturing. Polypyrrole tend to lose the memory of its monomeric structure at high plasma powers.

AC

   

27