Optimum stabilization processing parameters for polyacrylonitrile-based carbon nanofibers and their difference with carbon (micro) fibers

Accepted Manuscript Optimum stabilization processing parameters for polyacrylonitrile-based carbon nanofibers and their difference with carbon (micro) fibers Shahram Arbab, Arash Teimoury, Hamideh Mirbaha, Dominique C. Adolphe, Babak Noroozi, Parviz Nourpanah PII:

S0141-3910(17)30190-8

DOI:

10.1016/j.polymdegradstab.2017.06.026

Reference:

PDST 8279

To appear in:

Polymer Degradation and Stability

Received Date: 6 March 2017 Revised Date:

29 June 2017

Accepted Date: 30 June 2017

Please cite this article as: Arbab S, Teimoury A, Mirbaha H, Adolphe DC, Noroozi B, Nourpanah P, Optimum stabilization processing parameters for polyacrylonitrile-based carbon nanofibers and their difference with carbon (micro) fibers, Polymer Degradation and Stability (2017), doi: 10.1016/ j.polymdegradstab.2017.06.026. 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

Optimum stabilization processing parameters for polyacrylonitrile-based

RI PT

carbon nanofibers and their difference with carbon (micro) fibers

Shahram Arbab1*, Arash Teimoury2, Hamideh Mirbaha3, Dominique C. Adolphe4, Babak

1

SC

Noroozi2, Parviz Nourpanah3

ATMT Research Institute, Department of Textile Engineering, Amirkabir University of Technology, P.O. Box 15875-4413, Tehran, Iran 3

Department of Textile Engineering, Amirkabir University of Technology, Tehran, Iran

Université de Haute-Alsace - ENSISA- Laboratoire de Physique et Mécanique Textiles, Mulhouse CEDEX, France

AC C

EP

TE D

4

Department of Textile Engineering, University of Guilan, Rasht, Iran

M AN U

2

* Corresponding author. E-mail: [email protected] (Shahram Arbab) Tel. +989123721728

1

ACCEPTED MANUSCRIPT

Abstract Carbon nanofiber webs have high electrical and thermal conductivity, porosity and surface area, etc. making them favorable in many applications. In this paper, three types of polyacrylonitrile

RI PT

(PAN) copolymers are electrospun (average diameter of 150 to 500 nm) and carbon nanofibers are produced (average diameter of 110 to 300 nm). The effects of chemical composition and processing parameters on the formation of graphitic structure, morphology and electrical conductivity of the carbon nanofibers are studied. Unlike carbon fibers in micro scale, using PAN without acidic comonomers is suitable for production of carbon nanofibers. Common

SC

processing parameters for stabilization of PAN microfibers are not applicable to PAN nanofibers. Nanofibers stabilized using common microfibers procedure cannot tolerate high

M AN U

carbonization temperatures. While, thermal stabilization at higher temperature (300°C) results in proper stabilized structure with ability to tolerate carbonization conditions. Progress of stabilization reactions higher than 98% is inappropriate for obtaining electrically conductive nanofiber webs, whereas 85-90% progress is considered adequate for development of proper structure during carbonization to obtain optimum electrical conductivity. Formation of nanofiber mats in shape of an interconnected sponge-like structure is believed to be necessary for obtaining

TE D

much higher electrical conductivity (17-26 S/cm compared to 1-8 S/cm). Reducing fiber diameter from micro to nanoscale, the effect of processing parameters as well as the rate of thermochemical reactions can be different.

EP

Keywords: Carbon Nanofibers; PAN Nanofibers; Stabilization; Morphology; Graphitic Structure; Electrical Conductivity.

AC C

Introduction

Carbon microfibers possess high mechanical strength and modulus, low density, excellent electrical and thermal conductivities; therefore, they have been widely used for numerous applications particularly for the development of large load-bearing composites [1-5]. PAN copolymers are the best precursors for the production of carbon micro and nanofibers [6-9]. Production of carbon microfibers from PAN precursor fibers includes stabilization in air (200350 °C), followed by high temperature carbonization (800-2000 °C) in high purity nitrogen atmosphere [10-13]. Beside other important structural modifications to improve mechanical

2

ACCEPTED MANUSCRIPT

properties of carbon fibers, reduction of fiber diameter is also one of the effective methods to minimize structural defects of carbon fibers and improve mechanical properties. Therefore, obtaining carbon fibers with smaller diameter is one of the important objectives in carbon fiber industry. Moreover, reduction of fiber diameter results in increased surface area per unit mass,

RI PT

which facilitates the stabilization of precursor fiber and prevents from formation of core – shell structures [2, 4, 14, 15].

Carbon nanofibers are receiving increasing attention because of their large length to diameter ratio, high strength, elastic modulus, and relatively low density. Therefore, they are used in

SC

composite reinforcement and nanocomposites [1, 16-19]. Furthermore, carbon nanofiber webs have high thermal and electrical conductivity, porosity and surface area as well as uniform pore

M AN U

size distribution, high chemical stability and excellent corrosion resistance, which makes them proper templates for nanotubes, filters, gas storage systems (mainly hydrogen), fuel cells, rechargeable batteries, catalyst supports, supercapacitors, etc. [1, 6, 7, 16-18, 20-27]. The electrospun nanofibers produced from PAN copolymers are uniform, with high degree of molecular orientation. Their structural defects are also less than PAN homopolymer nanofibers. These nanofibers are approximately 30 times finer than PAN microfibers. Additionally, the

TE D

electrospun PAN nanofiber webs have relatively uniform pore size distribution, high interconnectivity of pores and high porosity [3-5, 27-29]. Highly polar nitrile groups in PAN homopolymer make spinning process difficult [30]. Stabilization process of PAN homopolymer fibers is performed at relatively high temperatures,

EP

where the reactions are difficult to control due to sudden heat release [3, 9]. Incorporation of comonomers into PAN chains reduces the nitrile – nitrile interactions and increases the solubility

AC C

of the polymer. Hence, polymer chains can be more readily oriented beside each other and fiber orientation would improve. In addition, presence of comonomers facilitates the thermochemical reactions during oxidative stabilization by reducing the activation energy and initiation temperature of the reactions. Moreover, the temperature range of exothermic reactions is broadened, which improves the uniformity and mechanical properties of stabilized fibers and resulting carbon fibers [2, 9, 31-33]. Stabilization is the most complicated and time-consuming process in the production of carbon fibers. It is a thermo-mechanical and multi-step process usually carried out in air atmosphere in the temperature range of 200-350 °C, especially in production of carbon microfibers [9-11, 34-36]. During this process, chemical and physical 3

ACCEPTED MANUSCRIPT

changes such as cyclization, dehydrogenation, oxidation, crosslinking and chain fragmentation take place in fibers. The linear structure of PAN is converted into an infusible ladder-like structure that can tolerate higher temperatures in carbonization process [1, 10, 11,31, 34-42]. If PAN fibers are not properly stabilized, either obtaining carbon fibers will be impossible or low

RI PT

quality carbon fibers will be produced [43]. It is accepted that a maximum stabilization progress of about 60% in case of PAN microfibers is preferable for production of carbon microfibers, whereas higher progress of reactions leads to fiber damage in carbonization process [9, 35]. High performance carbon microfibers are usually made from copolymers of PAN. Acidic comonomers

SC

such as itaconic acid are used to decrease initiation temperature of the cyclization reactions in the stabilization process, whereas esteric comonomers such as methyl acrylate are used to improve

M AN U

solubility, spinnability and stretchability of the fibers [2]. Coleman and Sivy [43] studied degradation rate of PAN copolymers containing vinyl acetate, acrylamide and methacrylic acid by FT-IR spectroscopy. They observed that the degradation rate strongly depends on chemical composition of the comonomers. The effect of different comonomers on stabilization reactions of PAN nanofibers has also been investigated. But the research on this subject is limited compared to PAN microfibers. Cetiner et al. [30] showed that presence of vinyl acetate

TE D

comonomer in PAN copolymers, decreases PAN nanofiber diameter and provides excellent mechanical stability for the resulting carbon nanofibers. Liu et al. [2] compared nanofibers of PAN dipolymer (including itaconic acid comonomers) with PAN terpolymer (including itaconic acid, and methyl acrylate comonomers). They observed that PAN dipolymer nanofibers have

EP

higher macromolecular regularity, better orientation of macromolecular chains and crystals, and more dimensional stability. Therefore, the structural defects during thermal treatments decreases

AC C

and nanofibers have desirable structural and thermochemical properties. The difference in the content and composition of comonomers affects the sequence and progress of stabilization reactions [10, 35, 44]. Duan et al. [45] showed that PAN copolymer nanofibers containing 8 wt.% monobutyl itaconate (MBI) have higher cyclization degree than PAN copolymer nanofibers containing 5 wt.% MBI in similar stabilization conditions. After stabilization, carbonization process is carried out in an inert atmosphere at temperatures as high as 800-2000 °C. During this process, structural changes including crosslinking, and integration of cyclized segments convert the ladder-like structure of the stabilized fibers to the graphitic structure of carbon fibers [12, 13]. Raman spectroscopy is used to study the 4

ACCEPTED MANUSCRIPT

microstructure of carbon nanofibers and formation of ordered graphitic structures [4, 5, 46, 47]. A number of researchers showed that at higher carbonization temperatures, the conversion of disordered carbonaceous (turbostratic) structures to ordered graphitic structures are increased and the graphite crystal size is enhanced [4, 5, 19, 20, 46-50]. Electrical conductivity as one of the

RI PT

important properties of carbon nanofibers, is usually lower than 20 S/cm [20]. At low temperatures soon after stabilization, the basic structural units of carbon nanofibers are isolated due to existence of heteroatoms and voids. At higher temperatures, as the number of the heteroatoms decreases during carbonization process and the void sizes decrease, these basic

SC

structural units are bonded to each other. Once the microdomains become continuous, and the layers of turbostratic carbon appear across the nanofibers, constituting a conducting channel, the

M AN U

nanofibers start to be conductive. This usually takes place above 800 °C [51, 52]. Several factors can increase the electrical conductivity of carbon nanofiber webs, such as higher proportion of graphitic structure, bigger size of graphitic crystallites, higher chain orientation and less structural defects (such as holes and fractures) within the individual nanofibers as well as more connection points and higher compression between nanofibers in the web [5, 13, 46, 51, 53, 54]. Some researchers showed that electrical conductivity of carbon nanofibers increases with

TE D

temperature and time of the carbonization process. They both result in more graphitic and ordered structure within the individual nanofibers and increased connections between carbon nanofibers in the web. Increasing the temperature of carbonization process leads to a reduction in the voids between basic structural units of individual carbon nanofibers by joining graphite-like

EP

sheets. The distorted sheets of structural units of individual carbon nanofibers are bonded to one another, wherever the boundaries of adjacent sheets meet [5, 48, 51, 52, 54-56]. However, Kim

AC C

et al. [46] observed that despite the increase in graphitic microstructures from carbonization temperature of 2000 to 2800 °C, the electrical conductivity of carbon nanofiber webs is decreased from 55.41 to 20.15 S/cm. This is attributed to the reduced number of connection points within carbon nanofiber webs. Compared to carbon microfibers, not many research works have been done on the production of carbon nanofibers from different PAN precursors with different chemical compositions. Due to substantial difference in fiber dimensions between micro and nanofibers, the results obtained for carbon microfibers are not expected to be completely applicable to nanofibers. Accordingly, in this work three different types of PAN copolymers, which are used to produce commercial PAN 5

ACCEPTED MANUSCRIPT

fibers and two of them (P2 and P3) are used to produce industrial carbon microfibers [34, 44], are employed to make PAN nanofibers. Two different procedures -one of them used to produce carbon microfibers in industry- are employed for the stabilization of nanofibers, followed by carbonization. In order to investigate the effect of processing parameters in production of carbon

RI PT

nanofiber webs, different analysis techniques including, Fourier Transform Infrared Spectrometry (FT-IR), Differential Scanning Calorimetry (DSC), Field-Emission Scanning Electron Microscopy (FE-SEM), Raman Spectroscopy and Electrical Conductivity measurement

SC

are used. 2. Experimental

M AN U

2.1. Materials

Three types of PAN copolymers with different chemical compositions are used for production of PAN nanofibers (Table 1). N, N-dimethylformamide solvent (99.5%, Merck) is used to prepare electrospinning solutions with various concentrations. P2 and P3, which are copolymers with acidic comonomers are used to produce carbon microfibers in industry and their full characterization, stabilization and thermal behavior have been studied and reported elsewhere [9,

TE D

34-36, 44].

Table 1. Characteristics of PAN copolymers.

Manufacturer company

Chemical composition

P1

Polyacryl Co.

AN-MA-SMS

Monomer feed in polymerization reactor (%) 95-4.5-0.5

Courtaulds Co. Ltd

AN-MA-IA AN-MAA-AM

P2 P3

EP

Polymer code

Jilin Chemical Fiber Group, Co. Ltd

[η] (dL/g)

MV (g/mol)

1.441

112452

95-4-1

2.216

202000

95-1.5-3.5

2.174

197000

AC C

P: PAN Copolymer, AN: Acrylonitrile, MA: Methyl Acrylate, IA: Itaconic Acid, MAA: Methacrylic Acid, AM: Acrylamide, SMS: Sodium Methallyl Solfunate.

2.2. Nanofiber production

Electrospinning is carried out at room temperature with a positive high voltage of 11 kV and solution flow rate of 0.25 ml/h to produce electrospun PAN nanofiber webs with random orientation on a fixed metal collector. The distance between the spinneret and fixed metal collector is 15 cm. The electrospun PAN nanofiber webs are stabilized using two different

6

ACCEPTED MANUSCRIPT

procedures in air atmosphere. The stabilized nanofiber webs are then carbonized in pure nitrogen atmosphere. Nomenclature of PAN nanofibers (PF), stabilized PAN nanofibers from procedures 1 and 2 (S1F and S2F) and carbon nanofibers (CNF) at different spinning concentrations are explained in table 2. Stabilization and carbonization conditions are presented in Table 3. The

RI PT

stabilization conditions (temperature and time) in the procedure No. 1 are chosen based on thermal behavior of initial fibers [34, 44], as it is used in the commercial production lines of carbon microfibers (P2 and P3 copolymers are used to produce commercial carbon fibers, whereas P1 copolymers are only used to produce commercial textile-grade PAN fibers) [34-36,

SC

44]. The stabilized nanofiber webs using procedure No. 1 could not tolerate higher temperatures of carbonization process and were damaged. That means they were turned into ashes during

M AN U

carbonization process and lost their fiber form and finally nothing was remained to be characterized. Therefore, carbon nanofibers could only be obtained from the nanofibers stabilized in procedure No. 2.

Table 2. Nomenclature of initial PAN nanofibers, stabilized PAN nanofibers and carbon nanofibers. Electrospinning solution concentration (wt.%)

PAN nanofibers

Stabilized PAN nanofibers by procedure No. 1

Stabilized PAN nanofibers by procedure No. 2

Carbon nanofibers

PF1-9

S1F1-9

S2F1-9

CNF1-9

10

PF1-10

S1F1-10

S2F1-10

CNF1-10

12

PF1-12

S1F1-12

S2F1-12

CNF1-12

8

PF2-8

S1F2-8

S2F2-8

CNF2-8

10

PF2-10

S1F2-10

S2F2-10

CNF2-10

PF2-12

S1F2-12

S2F2-12

CNF212

PF3-7

S1F3-7

S2F3-7

CNF3-7

8

EP

TE D

9

PF3-8

S1F3-8

S2F3-8

CNF3-8

10

PF3-10

S1F3-10

S2F3-10

CNF3-10

12

PF3-12

S1F3-12

S2F3-12

CNF3-12

12

AC C

7

PFX: PAN fiber of polymer Px, S1F: Stabilized fiber by procedure No. 1, S2F: Stabilized fiber by procedure No. 2, CNF: Carbon nanofibers

Table 3. Stabilization and carbonization conditions of PAN nanofibers. Stabilization procedure No. 1

Stabilization procedure No. 2

7

Carbonization

ACCEPTED MANUSCRIPT

PAN nanofibers

Heating rate (°C/min)

Time in Final Temp. (min)

Final Temp. (°C)

Heating rate (°C/min)

Time in Final Temp. (min)

Final Temp. (°C)

Heating rate (°C/min)

Time in Final Temp. (min)

PF1

270

5

90

300

5

90

1250

40

5

PF2

240

5

90

300

5

90

1250

40

5

PF3

250

5

90

300

5

90

RI PT

Final Temp. (°C)

40

5

SC

1250

2.3. Characterization

M AN U

The chemical structure of nanofibers is investigated using a Nicolet Magna IR560F FT-IR spectrometer in the range of 600-4000 cm-1 with a resolution of ±4 cm-1 in transmission mode. Thermal behavior of PAN nanofibers (temperature range of 50- 450 °C) and stabilized PAN nanofibers (temperature range of 50- 600 °C) is studied using a Mettler Toledo DSC 802 in air and at a heating rate of 5 °C/min. In order to estimate the progress of the stabilization reactions,

CI =

∆H PF − ∆H SF × 100 ∆H PF

TE D

cyclization index (CI) is calculated using Equation 1 [34, 57-60]:

(1)

respectively.

EP

Where ∆HPF and ∆HSF are enthalpy of reactions of initial PAN and stabilized PAN nanofibers,

A Hitachi U-4080 FE-SEM is employed in order to examine the surface morphology of PAN and

AC C

carbon nanofibers. Prior to SEM observations, the specimens were sputter-coated with a thin layer of gold to avoid charge accumulations. The average diameter of nanofibers is calculated using Image J software. Diameter measurements are done in 100 different points randomly selected in the micrographs of each sample. The average diameter and standard deviation of each sample is presented [2, 18]. Microstructure of the carbon nanofibers is investigated using a Bruker Senterra Raman spectrometer with 9-15 cm-1 resolution in spectral range of 144-3800 cm-1. RI is calculated using

8

ACCEPTED MANUSCRIPT

Equation 2, which is an index for determination of the amount of ordered graphite crystallites in the carbon nanofibers [5, 46, 61]:

ID IG

(2)

RI PT

RI =

Where ID is the intensity of D-band (about 1340 cm-1) and IG is the intensity of G-band (about 1580 cm-1). Graphite crystallite size is determined from Equation 3 using RI values [48, 49, 61]:

4.4 RI

SC

La =

(3)

M AN U

Electrical conductivity of carbon nanofiber webs with random orientation is measured by standard four-point probe method using the Keithley 196 System DMM_2 [46, 55, 62]. Resistivity, resistance and conductivity of the webs are measured using Equations 4, 5 and 6, respectively, the equation (4) was adopted to calculate the resistivity because the thickness of all samples (0.0007-0.003 cm) was much less than the distance between the probes in four-probe cell [5, 46, 55, 62-64]: πt

R=ρ

σ =

TE D

V ×( ) Ln 2 I L A

L RA

EP

ρ=

(4) (5) (6)

AC C

Where t is the thickness of carbon nanofibers web (cm), V is voltage in (v), I is electric current (A), L is distance between two electrodes (cm) and A is the area of carbon nanofibers web (cm2). 3. Results and discussion

3.1. FT-IR study of PAN nanofibers and stabilized nanofibers webs FT-IR spectra of the initial and stabilized PAN nanofibers in different concentrations of electrospinning dope are shown in Figure 1. Main FT-IR characteristic bands of PAN nanofibers are the stretching vibration of C≡N at 2243 cm-1, the stretching vibration of C=O at 1732 cm-1

9

ACCEPTED MANUSCRIPT

(due to the presence of acidic and ester comonomers), vibrations of aliphatic CH groups at 2935 cm-1 and bending vibration of CH2 at 1451 cm-1. Upon stabilization of PAN nanofibers, there is a decrease in the intensity of the characteristic bands at 2243 cm-1, 1732 cm-1 and different vibrations of aliphatic CH groups at 1238, 1360, 1451, and 2935 cm-1 [1, 10,22, 35, 40, 42, 65].

RI PT

At the same time, new bands appear at 1589 cm-1 (C=N, C=C, and N-H groups in cyclized structure), 1373 cm-1 (bending vibration of methylene in cyclized structure) and 806 cm-1 (vibration motion of C-H bond in the aromatic structure of PAN nanofibers), as well as a shoulder-like peak at 1715 cm-1 attributed to C=O groups in the structure of cyclized PAN [1, 22,

SC

32, 40, 42, 66]. According to theoretical prediction, C=O groups in acidic comonomers initiate the cyclization reactions and crosslink the structure with ionic mechanism (in PF2 and PF3 nanofibers). Therefore, with progress in cyclization reactions, the number of C=O groups in

M AN U

linear structure (peak at 1732 cm-1) decreases and the number of these groups in cyclized structure (shoulder-like peak at 1715 cm-1) increases due to oxidation reactions [22, 31, 34, 35,

AC C

EP

TE D

44].

10

ACCEPTED MANUSCRIPT

Figure 1. FT-IR spectra of nanofibers in different concentrations of electrospinning dope: (a) PF1-10, S1F1, S2F1 (b) PF2-10, S1F2, S2F2 (c) PF3-10, S1F3, S2F3

The FT-IR study of PAN nanofibers stabilized by procedure No. 1 shows that the intensity of nitrile, carbonyl, and aliphatic CH bands has decreased to a great extent. However, these groups

RI PT

still exist in the structure of the stabilized PAN nanofibers, indicating that formation of cyclized structures in these fibers is not complete. Stabilization of the nanofibers using procedure No. 2 causes following changes in FT-IR spectra: Complete elimination of aliphatic CH2 groups (2935 cm-1), incorporation of bands at 1451 and 1373 cm-1 together, significant reduction of the

SC

intensity of carbonyl band (1732 cm-1) and its overlap with C=N and C=C bands (1589 cm-1), broadening of nitrile band and significant decrease in its intensity. These changes are more obvious in FT-IR spectra of S2F1 nanofibers compared to S2F2 and S2F3 nanofibers due to the

M AN U

differences in chemical compositions of precursor nanofibers. As there is no acidic comonomers in PF1 copolymers, the cyclization reactions initiate at higher temperatures and progress in faster rate compared to copolymers containing acidic comonomers [31]. The nitrile band (at 2243 cm-1) in S2F1 nanofibers shows more pronounced reduction. The bands at 1373 and 1589 cm-1 become broader, and they become incorporated together gradually by increasing the solution

TE D

concentration. These results qualitatively indicate that the stabilization procedure No. 2 results in higher progress of the stabilization reactions, as it is expected from higher processing temperature compared to the procedure No. 1. Moreover, it is observed that despite the absence of acidic comonomers in PF1 nanofibers, stabilization reactions in both procedures have

EP

significant progress compared to PF2 and PF3.

3.2. DSC study of PAN and stabilized nanofiber webs

AC C

DSC curves of initial and stabilized PAN nanofibers using two different stabilization procedures are shown in Figure 2. Cyclization indices of the stabilized PAN nanofibers are presented in Table 4. Acidic comonomers in PF2 and PF3 nanofibers initiates cyclization reactions through ionic mechanism [31, 35, 67]. As a result, the exothermic peaks are broader and their initiation temperatures are lower compared to PF1 nanofibers, which is at 254 °C (Figure 2). The exothermic peak in PF2 nanofiber appears at lower temperature (~210 °C) compared to PF3 nanofiber (~235 °C) due to the presence of itaconic acid comonomers in PF2. Itaconic acid contains two carboxylic acid groups in its structure, therefore it is beneficial for initiation of 11

ACCEPTED MANUSCRIPT

stabilization reactions [9, 31, 44]. The presence of itaconic acid also makes the exothermic peak of PF2 nanofiber broader compared to PF3. According to Figure 2 and Table 4, the progress of stabilization reactions in procedure No. 1 is very limited. About 10 to 40% of stabilization progress is observed depending on the chemical composition of nanofibers. Therefore, many

RI PT

reactions are still remained to take place in nanofibers. The remaining reactions take place during heating process of DSC tests, resulting in sharp exothermic peaks (Dashed red lines). However, due to application of higher temperature and longer heating time in stabilization procedure No. 2, the intensity of exothermic peaks and the enthalpy of reactions are decreased significantly in

SC

PAN nanofibers stabilized using the procedure No. 2 (Dashed blue lines) and cyclization indices (CI) in procedure No. 2 are much higher than procedure No. 1. It should be noted that in DSC

M AN U

thermograms of fibers stabilized using procedure No.2, the first shoulder below 450 °C is related to stabilization reactions. The bigger peak, which has appeared at higher temperatures with the maximum at about 500-550 °C is attributed to carbonization reactions. Unlike PF2 and PF3, PF1 nanofibers do not have acidic comonomers. Despite similar stabilization conditions, progress of stabilization reactions in PF1 nanofibers is approximately 85-90%, whereas the progress of stabilization reactions in PF2 and PF3 nanofibers containing acidic

TE D

comonomers is more than 98%. This is also the reason for bigger DSC peaks of stabilized PF1 fibers in procedure No. 2 (S2F1) compared to PF2 and PF3 (S2F2 and S2F3 peaks). Since progress of stabilization reactions for PF1 fibers has been less than PF2 and PF3, more reactions are left to take place during DSC heating process. It should be noted that quantitative results of DSC are in

AC C

EP

good agreement with qualitative results of FT-IR (Section 3.1).

12

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 2. DSC curves of initial and stabilized PAN nanofibers.

Compared to initial PF1 nanofibers (Tpk of 314 °C), exothermic DSC peak temperatures of stabilized S2F1 nanofibers (shoulder-like peak) in all concentrations of electrospinning dope shifts to higher temperatures of about 417 °C due to the progress of stabilization reactions during

TE D

stabilization process, which indicates the formation of cyclized structure in stabilized PAN nanofibers. In addition, because of the presence of acidic comonomers, DSC exothermic peaks of stabilized S2F2 and S2F3 nanofibers have appeared at lower temperatures (386 and 406 °C, respectively) compared to S2F1 nanofibers at 417 °C. The intensity of exothermic peaks of S2F2

EP

and S2F3 are also much lower than S2F1, because of higher progress of stabilization reactions. However, cyclization indices of S2F2 and S2F3 nanofibers are higher than S2F1. Considering the fact that fibers stabilized with procedure No. 1 were damaged in the

AC C

carbonization process at 1250 °C (as described in section 2.2), it can be concluded that nonoptimal stabilization conditions, i.e. temperature and time of the heating process, is the reason that we could not produce carbon nanofibers from fibers stabilized with procedure No. 1. However, the chosen time and temperature of stabilization process in procedure No. 1 is common and suitable for industrial production of carbon microfibers from PAN copolymers having acidic comonomers (Table 1). Table 4. Characteristics of exothermic peaks of initial PAN and stabilized PAN nanofibersand cyclization index (CI) of stabilized PAN nanofibers.

13

ACCEPTED MANUSCRIPT

Samples

∆H of PAN nanofibers (J/g)

∆H of stabilized nanofibers in procedure No. 1 (J/g)

CI of stabilized nanofibers in procedure No. 1 (%)

∆H of stabilized nanofibers in procedure No. 2 (J/g)

CI of stabilized nanofibers in procedure No. 2 (%)

3398.0

3116.3

8.29

322.4

90.51

SF1-10

5720.0

5241.6

9.13

729.4

87.25

SF1-12

3127.6

2894.7

7.45

413.0

86.79

SF2-8

10654.3

8502.0

20.20

90.7

99.15

SF2-10

11148.7

9847.2

11.67

157.1

98.59

SF2-12

10724.4

9580.8

10.66

292.9

97.27

SF3-7

11458.7

9094.8

20.63

136.9

98.81

SF3-8

11458.7

7335.6

35.98

193.9

98.31

SF3-10

11320.0

8350.8

26.23

94.2

99.17

SF3-12

10698.4

8096.4

24.32

39.0

99.64

M AN U

SC

RI PT

SF1-9

Many researchers believe that presence of acidic comonomers in PAN structure is necessary to have successful stabilization process and to obtain desirable carbon microfibers [31, 35,36,67]. Unlike PF2 and PF3 nanofibers, PF1 does not contain acidic comonomers. However, the progress of stabilization reactions in PF1 is significant (CI of about 85-90%) and its carbonization process

TE D

is successful meaning that the nanofiber web can tolerate higher temperature of carbonization and it is not destroyed during it. As it was mentioned, a maximum progress of about 60% in stabilization reactions of PAN microfibers seems favorable for production of carbon microfibers

EP

[9, 35], whereas higher progress of reactions leads to formation of core-shell structure in stabilized microfibers and fiber damage in carbonization process [9, 35]. However, stabilization of PAN nanofibers using common procedure for microfiber stabilization (Procedure No. 1) is not

AC C

suitable for production of carbon nanofibers. Higher progress of stabilization reactions (CI of 8590%) is necessary for successful production of carbon nanofibers (As it was obtained in Procedure No. 2). Moreover, uniform distribution of heat within the cross section of fibers and formation of a uniform intermediate structure is essential. In stabilization process of microfibers, uniform distribution of heat within the cross section of fibers is more difficult, leading to formation of core-shell structure [9, 68]. However, obtaining uniform structure in stabilization process of nanofibers is easier due to the significant reducion in fiber diameter. It seems that formation of defects in smaller diameter fibers during stabilization process is less than thicker

14

ACCEPTED MANUSCRIPT

fibers. At its extreme, a single polymer chain (filament of minimum diameter) cannot exist with any defects present, and contains no fraction of misoriented chains [4, 18, 69, 70]. Therefore, smaller dimension of nanofibers compared to microfibers leads to a significant impact on the role

3.3. Morphology study of PAN and carbon nanofiber webs

RI PT

of processing parameters on structure formation.

FE-SEM images of PAN and carbon nanofiber webs are shown in Figure 3 and their statistical data are reported in Table 5. Initial PAN nanofibers have a uniform and bead-free surface.The

SC

porous structure of nanofiber webs can be observed in the images. The diameter of different PAN nanofibers at different spinning dope concentrations is in the range of 150-559 nm, with a

M AN U

noticeable increasing trend with the increase in PAN concentration in electrospinning solution (Table 5). In general, when PAN concentration and viscosity of the electrospinning solutions increases, the number of macromolecular chains in the solution as well as chain entanglements increase, whereas the size of Taylor cone remains relatively constant. Hence charged nanofibers

AC C

EP

TE D

jet becomes longer and thicker, resulting in the increase of nanofibers diameter [71-75].

15

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

TE D

Figure 3. FE-SEM images of PAN nanofibers (magnification: 7500X): (a) PF1-9, (b) PF1-10, (c) PF2-8, and (d) PF3-8, and FE-SEM images of carbon nanofibers (magnification: 10000X): (e) CNF1-9, (f) CNF110, (g) CNF2-8, and (h) CNF3-8. (Inset images are taken with magnification of 30000X)

Stabilization and carbonization processes lead to emission of different molecules (such as H2O, N2, HCN, and …) from the PAN structure. As a result, the diameter of nanofibers decreases to about 110-312 nm during carbonization process (Table 5). The morphology of CNF1 (Figures 3

EP

(e) and (f)) has been changed significantly compared to PF1 precursor nanofibers (Figures 3 (a) and (b)). During carbonization process, straight PF1 nanofibers are merged together in form of an

AC C

interconnected network, creating a sponge-like structure with many lateral connections. However, the morphology of CNF2 and CNF3 nanofibers has not changed considerably during thermal treatment and no merging of nanofibers can be observed (Figure 3 (g) and (h)). Only a number of fractures have been introduced in the structure as a result of thermal treatment. We suppose that absence of acidic monomers in PF1 copolymers, which is considered unfavorable in carbon microfiber industry is one of the reasons for change in morphology of CNF1 fibers during carbonization compared to CNF2 and CNF3. The thermal reactions start at higher temperature in absence of acidic comonomers and they progress faster, which makes their control more difficult. Thus, merging of fibers would take place or the formed intermediate 16

ACCEPTED MANUSCRIPT

structure would allow the merging of fibers at higher temperatures. However, the certain reason for this phenomenon is still unclear and should be investigated more in our future research. Table 5. Statistical quantification results of PAN and carbon nanofibers FE-SEM images. PAN nanofiber diameter ± standard deviation (nm)

Carbon nanofiber

Carbon nanofiber diameter ± standard deviation (nm)

Diameter reduction of carbon nanofiber (%)

PF1-9

159 ± 32

CNF1-9

126 ± 35

20.75

PF1-10

223 ± 49

CNF1-10

110 ± 30

50.67

PF1-12

244 ± 43

CNF1-12

168 ± 29

31.15

PF2-8

159 ± 20

CNF2-8

PF2-10

240 ± 50

CNF2-10

PF2-12

313 ± 31

CNF2-12

PF3-7

221 ± 40

PF3-8

150 ± 23

PF3-10

529 ± 61

PF3-12

559 ± 89

SC

RI PT

Initial nanofiber

12.58

178 ± 36

25.83

186 ± 35

40.54

M AN U

139 ± 31

134 ± 30

39.37

CNF3-8

132 ± 32

12.00

CNF3-10

312 ± 55

41.02

CNF3-12

307 ± 58

45.08

TE D

CNF3-7

In similar spinning conditions, the diameter of PF1 nanofibers is lower than those of PF2 and PF3 nanofibers. According to Table 1, this can be attributed to the lower molecular weight and as a result, lower viscosity of PF1 polymer compared to PF2 and PF3 (the molecular weight of PF1 is

EP

almost half the molecular weight of PF2 and PF3). Lower diameter of PF1 nanofibers may lead to better diffusion of oxygen into nanofiber structure during stabilization process. Oxygen has an

AC C

important role in initiation and progress of stabilization reactions. Despite absence of acidic comonomers in PF1 nanofibers, better diffusion of oxygen results in proper oxidation of these nanofibers during stabilization process.

3.4. Microstructure and electrical conductivity of carbon nanofibers

Raman is used to study microstructure of carbon nanofibers (Figure 4 and Table 6). Raman characteristic peaks of carbonaceous materials include a D-band related to disordered turbostratic structures and a G-band related to ordered graphitic structures [5, 46, 48, 49, 61, 76].

17

ACCEPTED MANUSCRIPT

The relative intensity ratio (RI) of D-band to G-band can be used to determine the carbonization degree of carbon nanofibers [46-47, 60]. With an increase in graphitic structure and graphite crystallite size in carbon nanofibers, RI decreases [5, 46, 48, 49, 61, 76]. D-band (at about 1310 cm-1) and G-band (at about 1580 cm-1) overlap with each other in Raman spectra of CNF1, CNF2

RI PT

and CNF3 (Figure 4). RI values of CNF1 are lower than those of CNF2 and CNF3 at different concentrations of electrospinning dope (Table 6), demonstrating the formation of more graphitic

AC C

EP

TE D

M AN U

SC

ordered structures and bigger graphite crystallites in CNF1 compared to CNF2 and CNF3.

Figure 4. Raman spectra of carbon nanofibers in various concentrations of electrospinning dope: (a) CNF1, (b) CNF2 and (c) CNF3

As discussed in section 3.2, the cyclization index of PF1 nanofibers in stabilization process is about 90%, while it is over 98% in CNF2 and CNF3. Hence, it is expected to obtain better carbonaceous structure in CNF2 and CNF3. The results, however, indicate formation of more graphitic ordered structure in CNF1, which leads to carbon nanofibers with lower RI and larger 18

ACCEPTED MANUSCRIPT

graphitic crystallite sizes (Table 6). Due to the presence of acidic comonomers in PF2 and PF3 nanofibers, the progress of stabilization reactions is higher than PF1 nanofibers. However, the carbon nanofibers obtained from PF1 show better graphitic structure. Therefore, it can be concluded that the maximum progress of stabilization reactions is not favorable for production of

RI PT

carbon nanofibers. An optimum progress of stabilization reactions (about 85-90%) seems to be sufficient for production of carbon nanofibers.

Comparing two stabilization procedures (section 3.2), it is observed that the progress of stabilization reactions in procedure No. 1 (based on commercial stabilization process of

SC

microfibers) is only 10-40% in each type of copolymers. Carbon nanofibers cannot be produced from nanofibers stabilized using procedure No. 1 and the fibers are damaged during

M AN U

carbonization process at 1250 °C. The procedure No. 2 results in about 85-99% of progress in stabilization reactions and high-quality carbon nanofibers can be produced. As a result, it can be concluded that conditions for obtaining favorable carbon structure at nanoscale is different from microscale.

Table 6. Microstructure and electrical characterization of carbon nanofibers produced from different PAN precursors. IG

RI = ID / IG

La (nm)

CNF1-9

1766.5

1292.1

1.367

3.219

17.34 ± 0.17

CNF1-10

2263.7

1621.9

1.396

3.152

16.70 ± 0.38

CNF1-12

2945.2

2117.5

1.391

3.163

25.91 ± 1.04

CNF2-8

1494.5

1011.0

1.478

2.977

2.00 ± 0.02

CNF2-10

1188.3

839.1

1.416

3.107

2.33 ± 0.02

CNF2-12

2008.6

1443.7

1.391

3.163

7.81 ± 0.07

CNF3-7

3375.3

2444.9

1.381

3.186

8.20 ± 0.07

CNF3-8

886.7

603.4

1.469

2.993

3.09 ± 0.04

CNF3-10

982.8

672.1

1.462

3.010

1.10 ± 0.01

CNF3-12

780.9

546.3

1.429

3.079

6.21 ± 0.08

TE D

ID

AC C

EP

Sample

Electrical conductivity (S/cm)

Electrical conductivity of CNF1 webs is in the range of 17 to 26 S/cm, significantly higher than CNF2 and CNF3 (about 1 to 8 S/cm) (Table 6). The graphitic structure of CNF1 is also more complete (lower RI and higher La) compared to CNF2 and CNF3. The significantly higher electrical conductivity of CNF1 compared to CNF2 and CNF3 can also be attributed to formation 19

ACCEPTED MANUSCRIPT

of connection points between nanofibers during carbonization process, resulting in a sponge-like compact network structure of carbon nanofibers webs (Figure 3). The electrical conductivity of carbon nanofibers does not show a significant dependance on the nanofiber diameter. It is mainly influenced from formation of graphitic structures, compactness

RI PT

of structure and formation of connection points due to thermal treatment. The more compact nanofiber webs are, the more probable is the presence of connection points and enhancement of electrical conductivity of carbon nanofibers.

PF2 and PF3 nanofiber webs show higher progress in stabilization reactions due to presence of

SC

acidic comonomers. However, the properties of their carbon nanofiber webs (such as electrical conductivity, graphitic crystallite sizes, etc) are worse than PF1 nanofiber webs. Therefore, it can

M AN U

be concluded that obtaining proper carbon structure with higher electrical conductivity, requires optimum stabilization process with optimum progress in stabilization reactions and formation of a sponge-like connected network of carbon nanofibers. This morphology seems beneficial for having higher electrical conductivity in carbon nanofiber webs. 4. Conclusion

TE D

Carbon nanofibers were produced from three different types of PAN precursor. The results showed that despite the common viewpoint in production technology of carbon fibers in microscale, presence of acidic comonomers in the copolymer stucture is not necessary for production of carbon fibers at nanoscale. Two stabilization procedures were used in this research

EP

and the result showed that common procedure for stabilization of microfibers is not suitable for the stabilization of nanofibers. The nanofibers stabilized using common stabilization procedure

AC C

of microfibers could not tolerate harsh carbonization conditions and were damaged at high temperatures. While, thermal stabilization at higher temperature of 300 °C –uncommon in carbon microfiber industry- results in proper stucture of stabilized nanofibers that tolerate high temperaure of carbonization process. These observations can be attributed to differences in the size of micro and nanofibers. With this significant change in diameter of fibers, the effects of processing parameters as well as the rate of thermochemical reactions can be different. The results revealed that progress of stabilization reactions higher than 98% is not favorable for electrical conductivity of carbon nanofibers, whereas the stabilization progress in the range of 85-90% leads to development of proper graphitic structure during carbonization process and 20

ACCEPTED MANUSCRIPT

formation of nanofiber mats in shape of a network interconnected sponge-like structure with much higher electrical conductivity. References

RI PT

[1] Arshad SN, Naraghi M, Chasiotis I. Strong carbon nanofibers from electrospun polyacrylonitrile. Carbon 2011; 49: 1710-1719.doi: 10.1016/j.carbon.2010.12.056.

[2] Liu J, He L, Ma S, Liang J, Zhao Y, Fong H. Effects of chemical composition and postspinning stretching process on the morphological, structural, and thermo-chemical properties of

SC

electrospun polyacrylonitrile copolymer precursor nanofibers. Polymer 2015; 61: 20-28.doi: 10.1016/j.polymer.2015.01.063.

[3] Liu J, Zhou P, Zhang L, Ma Z, Liang J, Fong H. Thermo-chemical reactions occurring during

M AN U

the oxidative stabilization of electrospun polyacrylonitrile precursor nanofibers and the resulting structural conversions. Carbon 2009; 47: 1087-1095.doi: 10.1016/j.carbon.2008.12.033. [4] Zhou Z, Liu K, Lai C, Zhang L, Li J, Hou H, Reneker DH, Fong H. Graphitic carbon nanofibers developed from bundles of aligned electrospun polyacrylonitrile nanofibers containing phosphoric acid. Polymer 2010; 51: 2360-2367.doi: 10.1016/j.polymer.2010.03.044. [5] Zhou Z, Lai C, Zhang L, Qian Y, Hou H, Reneker DH, Fong H. Development of carbon

TE D

nanofibers from aligned electrospun polyacrylonitrile nanofiber bundles and characterization of their microstructural, electrical, and mechanical properties. Polymer 2009; 50: 2999-3006. doi: 10.1016/j.polymer.2009.04.058.

[6] Zhang B, Kang F, Tarascon JM, Kim JK. Recent advances in electrospun carbon nanofibers

EP

and their application in electrochemical energy storage. Prog Mater Sci 2016; 76: 319-380.doi: 10.1016/j.pmatsci.2015.08.002.

AC C

[7] Kim JH, Ganapathy HS, Hong SS, Gal YS, Lim KT. Preparation of polyacrylonitrile nanofibers as a precursor of carbon nanofibers by supercritical fluid process. J Supercrit Fluids 2008; 47: 103-107.doi: 10.1016/j.supflu.2008.05.011. [8] Lee HM, Kim HG, Kang SJ, Park SJ, An KH, Kim BJ. Effects of pore structures on electrochemical behaviors of polyacrylonitrile (PAN)-based activated carbon nanofibers. J Ind Eng Chem 2015; 21: 736-740. doi: 10.1016/j.jiec.2014.04.004. [9] Morgan P. Carbon fibers and their composites. Taylor & Francis Group, CRC Press, Boca Raton, FL, 2005. doi: 10.1201/9781420028744.

21

ACCEPTED MANUSCRIPT

[10] Xue Y, Liu J, Liang J. Correlative study of critical reactions in polyacrylonitrile based carbon fiber precursors during thermal-oxidative stabilization. Polym Degrad Stab 2013; 98: 219-229.doi: 10.1016/j.polymdegradstab.2012.10.018. [11] Rafiei S, Noroozi B, Arbab S, Haghi A. Characteristic assessment of stabilized

RI PT

polyacrylonitrile nanowebs for the production of activated carbon nano-sorbents. Chin J Polym Sci 2013; 32(4): 449-457. doi: 10.1007/s10118-014-1410-4.

[12] Rahaman MSA, Ismail AF, Mustafa A. A review of heat treatment on polyacrylonitrile fiber. Polym Degrad Stab 2007; 92: 1421-1432.doi: 10.1016/j.polymdegradstab.2007.03.023.

SC

[13] Liu CK,Lai K, Liu W, Yao M, Sun RJ. Preparation of carbon nanofibres through electrospinning and thermal treatment. Polym Int 2009; 58: 1341-1349.doi: 10.1002/pi.2669.

M AN U

[14] Xie Z, Niu H, Lin T. Continuous polyacrylonitrile nanofiber yarns: preparation and drydrawing treatment for carbon nanofiber production. RSC Adv 2015; 5: 15147-15153.doi: 10.1039/C4RA16247A.

[15] Ali AA, El-Hamid MA. Electro-spinning optimization for precursor carbon nanofibers. Composites Part A 2006; 37: 1681-1687.doi: 10.1016/j.compositesa.2005.10.008. [16] Gu SY, Wu QL, Ren J. Preparation and surface structures of carbon nanofibers produced

5805(08)60021-9.

TE D

from electrospun PAN precursors. New Carbon Mater 2008; 23: 171-176.doi:10.1016/S1872-

[17]Gu SY, Ren J, Wu QL, Preparation and structures of electrospun PAN nanofibers as a precursor

of

carbon

nanofibers.

Synth

Met

2005;

155:

157-161.doi:

EP

10.1016/j.synthmet.2005.07.340.

[18] Moon SC, Farris RJ. Strong electrospun nanometer-diameter polyacrylonitrile carbon fiber

AC C

yarns. Carbon 2009; 47: 2829-2839.10.1016/j.carbon.2009.06.027. [19] Zussman E, Chen X, Ding W, Calabri L, Dikin DA, Quintana JP, Ruoff RS. Mechanical and structural characterization of electrospun PAN-derived carbon nanofibers. Carbon 2005; 43: 2175-2185.doi: 10.1016/j.carbon.2005.03.031. [20] Yang Y, Simeon F, Hatton TA, Rutledge GC. Polyacrylonitrile-based electrospun carbon paper

for

electrode

applications.

J

Appl

Polym

Sci

2011;

124:

3861-

3870.doi: 10.1002/app.35485. [21] Karthikeyan KK, Biji P. A Novel Biphasic Approach for Direct Fabrication of Highly Porous, Flexible Conducting Carbon Nanofiber Mats from Polyacrylonitrile (PAN)/NaHCO3 22

ACCEPTED MANUSCRIPT

Nanocomposite.

Microporous

and

Mesoporous

Mater

2016;

224:

372-383.

doi:

10.1016/j.micromeso.2015.12.055. [22] Wu SH, Qin XH. Effects of the stabilization temperature on the structure and properties of polyacrylonitrile-based stabilized electrospun nanofiber microyarns. J Therm Anal Calorim

RI PT

2014; 116: 303-308.doi:10.1007/s10973-013-3530-4.

[23] Kim M, Kim Y, Lee KM, Jeong SY, Lee E, Baeck SH, Shim SE. Electrochemical improvement due to alignment of carbon nanofibers fabricated by electrospinning as an electrode for supercapacitor. Carbon 2016; 99: 607-618.doi: 10.1016/j.carbon.2015.12.068.

SC

[24] Aran H, Benito SP, Luiten-Olieman M, Er S, Wessling M, Lefferts L, Benes N, Lammertink R. Carbon nanofibers in catalytic membrane microreactors. J MembrSci 2011; 381: 244-250.doi:

M AN U

10.1016/j.memsci.2011.07.037.

[25] Benito SP, Lefferts L. The production of a homogeneous and well-attached layer of carbon nanofibers on metal foils. Carbon 2010; 48: 2862-2872.doi: 10.1016/j.carbon.2010.04.018. [26] Chinthaginjala J, Thakur D, Seshan K, Lefferts L. How carbon-nano-fibers attach to Ni foam. Carbon 2008; 46: 1638-1647.doi: 10.1016/j.carbon.2008.07.002. [27] Aslan T, Arslan S, Eyvaz M, Güçlü S, Yüksel E, Koyuncu I. A novel nanofiber

TE D

microfiltration membrane: Fabrication and characterization of tubular electrospun nanofiber (TuEN) membrane. J Membr Sci 2016; 520: 616-629.doi: 10.1016/j.memsci.2016.08.014. [28] Ahmed FE, Lalia BS, Hashaikeh R. A review on electrospinning for membrane fabrication: challenges and applications. Desalin 2015; 356: 15-30.doi: 10.1016/j.desal.2014.09.033.

scaffolds

for

EP

[29] Zhu X, Cui W, Li X, Jin Y. Electrospun fibrous mats with high porosity as potential skin

tissue

engineering.

Biomacromolecules

2008;

9:

1795-1801.doi:

AC C

10.1021/bm800476u.

[30] Cetiner S, Sen S, Arman B, Sarac AS. Acrylonitrile/vinyl acetate copolymer nanofibers with different vinylacetate content. J Appl Polym Sci 2013; 127: 3830-3838.doi: 10.1002/app.37690. [31] Ouyang Q, Cheng L, Wang H, Li K. Mechanism and kinetics of the stabilization reactions of itaconic acid-modified polyacrylonitrile. Polym Degrad Stab 2008; 93: 1415-1421.doi: 10.1016/j.polymdegradstab.2008.05.021. [32] Wangxi Z, Jie L, Gang W. Evolution of structure and properties of PAN precursors during their conversion to carbon fibers. Carbon 2003; 41: 2805-2812.doi: 10.1016/S00086223(03)00391-9. 23

ACCEPTED MANUSCRIPT

[33] Zhang L, Aboagye A, Kelkar A, Lai C, Fong H, A review: carbon nanofibers from electrospun polyacrylonitrile and their

applications. J Mater

Sci 2014; 49: 463-

480.doi:10.1007/s10853-013-7705-y. [34] Arbab S, Mirbaha H, Zeinolebadi A, Nourpanah P. Indicators for evaluation of progress in

RI PT

thermal stabilization reactions of polyacrylonitrile fibers. J Appl Polym Sci 2014; 131 (11).doi: 10.1002/app.40343.

[35] Arbab S, Zeinolebadi A. A procedure for precise determination of thermal stabilization reactions in carbon fiber precursors. Polym Degrad Stab 2013; 98: 2537-2545.doi:

SC

10.1016/j.polymdegradstab.2013.09.014.

[36]Mirbaha H, Arbab S, Zeinolebadi A, Nourpanah P. An investigation on actuation behavior of

M AN U

polyacrylonitrile gel fibers as a function of microstructure and stabilization temperature. Smart Mater Struct 2013; 22 (4): 045019. doi:10.1088/0964-1726/22/4/045019. [37] Dhakate SR, Gupta A, Chaudhari A, Tawale J, Mathur RB. Morphology and thermal properties of PAN copolymer based electrospun nanofibers. Synth Met 2011; 161: 411-419.doi: 10.1016/j.synthmet.2010.12.019.

[38] Esrafilzadeh D, Morshed M, Tavanai H. An investigation on the stabilization of special

TE D

polyacrylonitrile nanofibers as carbon or activated carbon nanofiber precursor. Synth Met 2009; 159: 267-272.doi: 10.1016/j.synthmet.2008.09.014. [39] Cipriani E, Zanetti M, Bracco P, Brunella V, Luda M, Costa L. Crosslinking and carbonization processes in PAN films and nanofibers. Polym Degrad Stab 2016; 123: 178-

EP

188.doi: 10.1016/j.polymdegradstab.2015.11.008. [40] Xiao S, Cao W, Wang B, Xu L, Chen B. Mechanism and kinetics of oxidation during the

AC C

thermal stabilization of polyacrylonitrile fibers. J Appl Polym Sci 2013; 127: 3198-3203.doi: 10.1002/app.37733.

[41] Fu Z, Gui Y, Cao C, Liu B, Zhou C, Zhang H. Structure evolution and mechanism of polyacrylonitrile and related copolymers during the stabilization. J Mater Sci 2014; 49: 28642874.doi:10.1007/s10853-013-7992-3. [42] Simitzis JC, Georgiou PC. Functional group changes of polyacrylonitrile fibres during their oxidative, carbonization and electrochemical treatment. J Mater Sci 2015; 50: 45474564.doi:10.1007/s10853-015-9004-2.

24

ACCEPTED MANUSCRIPT

[43] Coleman M, Sivy G. Fourier transform IR studies of the degradation of polyacrylonitrile copolymers—I: introduction and comparative rates of the degradation of three copolymers below 200 C and under reduced pressure. Carbon 1981; 19: 123-126.doi:10.1016/0008-6223(81)901184.

RI PT

[44] Arbab S, Zeinolebadi A. Quantitative analysis of the effects of comonomers and heating conditions on the stabilization reactions of polyacrylonitrile fibers as carbon fiber precursors. Polym Degrad Stab 2017; 139: 107-116. doi: 10.1016/j.polymdegradstab.2017.04.003.

[45] Duan G, Zhang H, Jiang S, Xie M, Peng X, Chen S, Hanif M, Hou H. Modification of

SC

precursor polymer using co-polymerization: A good way to high performance electrospun carbon nanofiber bundles. Mater Lett 2014; 122: 178-181.doi: 10.1016/j.matlet.2014.02.023.

M AN U

[46] Kim C,Yang KS, Kojima M, Yoshida K, Kim YJ, Kim YA, Endo M. Fabrication of Electrospinning-Derived Carbon Nanofiber Webs for the Anode Material of Lithium-Ion Secondary Batteries. Adv Funct Mater 2006; 16: 2393-2397.doi: 10.1002/adfm.200500911. [47] Ghasemi M, Shahgaldi S, Ismail M, Kim BH, Yaakob Z, Daud WRW. Activated carbon nanofibers as an alternative cathode catalyst to platinum in a two-chamber microbial fuel cell.Int J Hydrogen Energy 2011; 36: 13746-13752. doi: 10.1016/j.ijhydene.2011.07.118.

TE D

[48] Kim C, Park SH, Cho JI, Lee DY, Park TJ, Lee WJ, Yang KS. Raman spectroscopic evaluation of polyacrylonitrile-based carbon nanofibers prepared by electrospinning. Raman Spectrosc 2004; 35: 928-933.doi: 10.1002/jrs.1233. [49] Wang Y, Serrano S, Santiago-Aviles JJ. Raman characterization of carbon nanofibers

6779(02)00472-1.

EP

prepared using electrospinning. Synth Met 2003; 138: 423-427.doi: 10.1016/S0379-

AC C

[50] Li W, Li M, Wang M, Zeng L, Yu Y. Electrospinning with partially carbonization in air: Highly porous carbon nanofibers optimized for high-performance flexible lithium-ion batteries. Nano Energy 2015; 13: 693-701.doi: 10.1016/j.nanoen.2015.03.027. [51] Wang Y, Santiago-Aviles JJ, Furlan R, Ramos I. Pyrolysis Temperature and Time Dependence of Electrical Conductivity Evolution for Electrostatically Generated Carbon Nanofibers.

IEEE

Trans.

On

Nanotechnology

10.1109/TNANO.2003.808510.

25

2003;

2:

39-43.doi:

ACCEPTED MANUSCRIPT

[52] Agend F, Naderi N, Fareghi-Alamdari R. Fabrication and Electrical Characterization of Electrospun Polyacrylonitrile-Derived Carbon Nanofibers. J Appl Polym Sci 2007; 106: 255259.doi: 10.1002/app.26476. [53] Sharma CS, Katepalli H, Sharma A, Madou M. Fabrication and electrical conductivity of carbon

nanofiber

arrays.

Carbon

10.1016/j.carbon.2010.12.058.

2011;

49:

1727-1732.doi:

RI PT

suspended

[54] Li M, Han G, Yang B. Fabrication of the catalytic electrodes for methanol oxidation on electrospinning-derived carbon fibrous mats. Electrochem Commun 2008; 10: 880-883.doi:

SC

10.1016/j.elecom.2008.04.002.

[55] Hedin N, Sobolev V, Zhang L, Zhu Z, Fong H. Electrical properties of electrospun carbon

M AN U

nanofibers. Mater Sci 2011; 46: 6453-6456.doi:10.1007/s10853-011-5725-z. [56] Prilutsky S, Schechner P, Bubis E, Makarov V, Zussman E, Cohen Y. Anodes for glucose fuel cells based on carbonized nanofibers with embedded carbon nanotubes. Electrochim Acta 2010; 55: 3694-3702.doi: 10.1016/j.electacta.2010.02.005.

[57] Qin XH. Structure and property of electrospinning PAN nanofibers by different preoxidation temperatures. J Therm Anal Calorim 2010; 99: 571-575. doi: 10.1007/s10973-009-

TE D

0126-0.

[58] Tsai JS,Hsu HN. Determination of the aromatization index for oxidized polyacrylonitrile fibre by the differential scanning calorimetry method. J Mater Sci Lett 1992; 11: 14031405.doi:10.1007/BF00729641.

EP

[59] Tsai JS, Lin CH. The effect of molecular weight on the cross section and properties of polyacrylonitrile precursor and resulting carbon fiber. J Appl Polym Sci 1992; 42: 3045-

AC C

3050.doi: 10.1002/app.1991.070421124. [60] Bang Y, Lee S, Cho H. Effect of methyl acrylate composition on the microstructure changes of high molecular weight polyacrylonitrile for heat treatment. J Appl Polym Sci 1998; 68: 22052213.doi: 10.1002/(SICI)1097-4628. [61] Liu CK, Feng Y, He HJ, Zhang J, Sun RJ,Chen MY. Effect of carbonization temperature on properties of aligned electrospun polyacrylonitrile carbon nanofibers. Mater Des 2015; 85: 483486.doi: 10.1016/j.matdes.2015.07.021. [62] Kim C,Zhang S. Conductivity of carbon nanofiber/polypyrrole conducting nanocomposites, Mech Sci Technol 2009; 23: 75-80.doi:10.1007/s12206-008-0911-x. 26

ACCEPTED MANUSCRIPT

[63]Schroder DK.Semiconductor material and device characterization. John Wiley & Sons, 3rd ed. 2006. doi: 10.1002/0471749095. [64] Halliday D, Resnick R, Walker J. Fundamentals of Physics. John Wiley & Sons, 8th ed. 2008.

RI PT

[65] Ma S, Liu J, Liu Q, Liang J, Zhao Y, Fong H. Investigation of structural conversion and size effect from stretched bundle of electrospun polyacrylonitrile copolymer nanofibers during oxidative stabilization. Mater Des 2016; 95: 387-397.doi: 10.1016/j.matdes.2016.01.134.

[66] Devasia R, Reghunadhan Nair CP, Sadhana R, Babu NS, Ninan KN. Fourier Transform

SC

Infrared and Wide-Angle X-Ray Diffraction Studies of the Thermal Cyclization Reactions of High-Molar-Mass Poly(acrylonitrile-co-itaconic acid). J Appl Polym Sci 2006; 100: 3055-

M AN U

3062.doi: 10.1002/app.23705.

[67] Ouyang Q, Cheng L, Wang HJ, Li KX. DSC Study of Stabilization Reactions In Poly(Acrylonitrile– co–Itaconic Acid) With Peak-Resolving Method. J Therm Anal Calorim 2008; 94: 85-88.doi:10.1007/s10973-007-8773-5.

[68] Lai C, Zhong G, Yue Z, Chen G, Zhang L, Vakili A, Wang Y, Zhu L, Liu, J, and Fong H. Investigation of post-spinning stretching process on morphological, structural, and mechanical

TE D

properties of electrospun polyacrylonitrile copolymer nanofibers. Polymer 2011; 52: 519-528. doi: 10.1016/j.polymer.2010.11.044.

[69] Morris EA, Weisenberger MC, Bradley SB, Abdallah MG, Mecham SJ, Pisipati P, McGrath JE. Synthesis, spinning, and properties of very high molecular weight poly(acrylonitrile-co-

EP

methyl acrylate) for high performance precursors for carbon fiber. Polymer 2014; 55: 64716482. doi: 10.1016/j.polymer.2014.10.029

AC C

[70] Griffith AA. The Phenomena of Rupture and Flow in Solids. Philos Trans R Soc Lond A 1921; 221: 163-198. doi: 10.1098/rsta.1921.0006 [71] Gu SY, Ren J, Vansco GJ. Process optimization and empirical modeling for electrospun polyacrylonitrile (PAN) nanofiber precursor of carbon nanofibers. Eur Polym J 2005; 41: 25592668.doi: 10.1016/j.eurpolymj.2005.05.008. [72] Wang C, Chien HS, Hsu CH, Wang YC, Wang CT, Lu HA. Electrospinning of Polyacrylonitrile Solutions at Elevated Temperatures. Macromol 2007; 40: 7973-7983.doi: 10.1021/ma070508n.

27

ACCEPTED MANUSCRIPT

[73]Yordem OS,Papila M, Menceloglu YZ. Effects of electrospinning parameters on polyacrylonitrile nanofiber diameter: An investigation by response surface methodology. Mater Des 2008; 29: 34-44.doi: 10.1016/j.matdes.2006.12.013. [74] He JH, Wan YQ, Yu JY. Effect of Concentration on Electrospun Polyacrylonitrile (PAN)

RI PT

Nanofibers. Fibers Polym 2008; 9: 140-142.doi:10.1007/s12221-008-0023-3.

[75] Todd D, Mérida W. Morphologically controlled fuel cell transport layers enabled via electrospun

carbon

nonwovens.

J

Power

10.1016/j.jpowsour.2014.09.095.

Sources

2015;

273:

312-316.doi:

SC

[76] Zhu J, Chen C, Lu Y, Ge Y, Jiang H, Fu K, Zhang X. Nitrogen-doped carbon nanofibers derived from polyacrylonitrile for use as anode material in sodium-ion batteries. Carbon 2015;

AC C

EP

TE D

M AN U

94: 189-195.doi: 10.1016/j.carbon.2015.06.076.

28