- Email: [email protected]
Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers
G Model JIEC 3651 No. of Pages 8
Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers Hyun-Sig Kila,1, So Young Janga,b,1, Seunghyun Koc,d , Young Pyo Jeonc,d , Hwan-Chul Kimb , Han-Ik Johe, Sungho Leea,d,* a Carbon Composite Materials Research Center, Korea Institute of Science and Technology, 92 Chudong-ro, Bongdong-eup, Wanju-gun, Jeonbuk 55324, Republic of Korea b Department of Organic Materials and Fiber Engineering, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju, Jeonbuk 54896, Republic of Korea c Center for C-Industry Incubation, Korea Research Institute of Chemical Technology, 141 Gajeong-ro, Yuseong-gu, Daejeon 34114, Republic of Korea d Department of Nano Material Engineering, Korea University of Science and Technology, 217 Gajeong-ro, Yuseong-gu, Daejeon 34113, Republic of Korea e Department of Energy Engineering, Konkuk University, 120 Neungdong-ro, Gwangjin-gu, Seoul 05029, Republic of Korea
A R T I C L E I N F O
Article history: Received 31 August 2017 Received in revised form 28 September 2017 Accepted 28 September 2017 Available online xxx Keywords: Isotropic pitch Stabilization Tensile properties Carbon fiber
A B S T R A C T
Optimization of the stabilization process for manufacturing low-cost carbon fibers was performed. By simply varying both heating rate (0.5–10 C/min) and starting temperature (25–230 C), the resulting fibers possessed different properties such as densities, oxygen contents, and weight gains due to the oxygen up-takes. Subsequent oxidative reactions led to changes in chemical compositions of the fibers. Therefore, this study suggests that more concerned design with complementary parameters such as starting temperature and heating rate, significantly reducing stabilization time down to 56.5 min with the comparable mechanical properties of the conventional isotropic pitch based CFs. © 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction Carbon fibers (CFs) are used in various applications such as automobile, aerospace, sports, and many other components because of their excellent mechanical, thermal, and electrical properties [1–4]. However, the aforementioned applications using CFs have been limited so far since a large scale CFs production cost is prohibitively expensive for the commercialization. Therefore, significant research effort in the last several decades has been devoted to reducing CFs production cost by developing cheaper precursor materials, minimizing stabilization time, simplifying carbonization, or etc. A typical CFs manufacturing process flow consists of fiber spinning, oxidative stabilization, carbonization, graphitization, and surface treatment process [5]. Among these processes, the stabilization is the most important step because the oxidative behavior of precursor fiber greatly affects the final mechanical properties of the CFs. Also,
* Corresponding author at: Carbon Composite Materials Research Center, Korea Institute of Science and Technology, 92 Chudong-ro, Bongdong-eup, Wanju-gun, Jeonbuk 55324, Republic of Korea. Fax: +82 63 710 7589. E-mail address: [email protected] (S. Lee). 1 Authors contributed equally to this work.
stabilization process is considered as the most time consuming and expensive step. Long time stabilization at relatively low temperature used in the conventional process flow was considered to allowing oxidant to uniformly diffuse through precursor fiber [6]. Without this uniform oxidant diffusion through the fiber precursor, uneven stabilization or rapid oxidation can yield the final carbon fiber possessing poor properties [7,8]. In this regard, it is challenging to achieving the reduced stabilization time while maintaining a suitable degree of stabilization. Various process modifications such as stabilization temperature, time, heating rate, atmosphere, pressure, and multi-step stabilization [9–13] have been proposed. Many companies and researchers have been striving for reducing stabilization process, but it still remained difficult to optimize and deploy reliably. In this study, we report a systematic study on optimizing the stabilization process by varying heating conditions such as a heating speed and the starting temperature. Based on our previous study [14], the stabilized fibers with a density range of 1.35–1.36 g/ cm3 yielded CF with the highest tensile strength regardless of the stabilization conditions. Here, petroleum-based isotropic pitch fibers were first stabilized at various temperatures in ambient air atmosphere at heating rate of 0.5–10 C/min and then densities of the corresponding fibers were characterized. After the oxidative
https://doi.org/10.1016/j.jiec.2017.09.048 1226-086X/© 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
2
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
stabilization at various conditions, the stabilized fibers were carbonized at 1200 C under nitrogen atmosphere. Then, chemical and physical properties of stabilized fibers and tensile properties of the CFs were measured and then correlated. The study revealed a possibility to shorten the stabilization process time by eliminating unnecessary temperature ranges. In addition, a combination of variables such as starting temperature and heating rate was investigated for a further reducing stabilization time. We, therefore, demonstrated a procedure to optimize stabilization for developing a cost effective process. Experimental Preparation of pitch-based carbon fibers A commercial isotropic pitch (SN-280, Anshan, China) was melt-spun into pitch fibers using a 12-hole spinneret at 335 C with a diameter of 150 mm and an aspect ratio of 3 under nitrogen atmosphere. The prepared pitch fibers possessed the diameter of 12.2 0.8 mm. The as-spun pitch fibers were stabilized at 290 C for 30 min using a convection oven in ambient air atmosphere with various starting temperatures of 25, 150, 170, 190 and 230 C and various heating rates of 0.5, 1, 5 and 10 C/min. The stabilized fibers were then carbonized at 1200 C under nitrogen flow in a tube furnace at a heating rate of 5 C/min. Characterization The thermal properties of as-spun pitch fibers were analyzed using a thermogravimetric apparatus (SETSYS Evolution TGA, SETARAM Instrumentation, Caluire, France). Specimens were heated from 40 C to 400 C at a heating rate of 0.5, 1, 5 and 10 C with an air flow rate of 50 mL/min. The stabilization starting temperatures and the amounts of oxygen uptake were monitored using a thermogravimetric analyzer. The weight of pitch fibers used for the each analysis was approximately 10 mg. The densities of as-spun pitch fibers and stabilized fibers were determined by a Sartorius YDK03 Density Determination Kit (Sartorius AG, Goettingen, Germany). The Archimedean principle is applied for measuring the density of pitch fibers. Another method of density measurement, a density gradient column of liquid (POLYTEST, RayRan, UK) was prepared to determine the densities carbonized fibers. Two liquids, 1,1,2,2-tetrabromoethane (98%, Daejung Chemical Co., Korea) and benzene (99.5%, Daejung Chemical Co., Korea), were used to build the density gradient, and glass beads with accurately known densities floated in the column. The each fiber samples were inserted and left for 8 h in the column. At least three specimens were tested to calculate the densities of the fibers in each condition.
FT-IR spectroscopy (Nicolet IS10, USA) was performed to investigate the change in the oxygen functional groups during the stabilization using the KBr pellet technique. The mixed pellet was prepared with 400 mg of spectrometric grade KBr and 2 mg of sample. The mixture of the sample and the salt was crushed using a pestle and a mortar. The powder mixture was placed in a KBr die kit and pressed with a hydraulic laboratory press. Each sample was scanned 16 times at a resolution of 16 cm1 with a range of 4000– 400 cm1. All spectra were collected in the absorbance mode with an automatic baseline correction and then transformed to transmittance. Elemental analysis (Flash 2000, Thermo Scientific, USA) was carried out to measure the amounts of oxygen in the bulk of the CF samples. Oxygen was analyzed using helium gas at 1060 C for 500 s. The mechanical properties of the CFs were examined by a mechanical tester for a single fiber (FAVIMAT+, Textechno, Germany) with a test speed of 5 mm/min. The gauge length was 25 mm, and 20 specimens were measured for each experimental point. X-ray photoelectron spectroscopy (XPS, K-alpha, Thermo Scientific, USA) analysis was performed using monochromated Al Ka (1486.6 eV) X-rays to examine the type of chemical bond of nitrogen and the elemental composition on the surface of as-spun and stabilized pitch fiber samples. The survey spectrum was collected from 0 eV to 1350 eV, and the binding energies were referenced to the C1s line at 284.8 eV. Results and discussion Thermal analysis of the stabilized pitch fibers The thermal characteristics of as-spun pitch fibers were analyzed using TGA. Fig. 1a shows a TGA thermogram obtained from as-spun pitch fiber to mimic oxidative stabilization at a heating rate of 1 C/min under ambient air atmosphere. There were three regions showing different weight changes as a function of temperature. A slight increase in weight was observed from 40 to 175 C (region I) with increasing the temperature. Then, the weight significantly increased up to 310 C (region II) with a weight gain of 10 wt.%. Finally, at temperature above 310 C, there was a significant weight loss (region III). In the region I, a small amount of weight gain was observed even though pitch fibers were exposed under oxygen, indicating that temperature was not high enough to overcome energy barrier for inducing oxygen diffusion into the fibers. It is interesting to observe that the weight gain dramatically increased from 175 C due to oxygen diffusion in the region II. From our previous result by in-situ mass spectroscopy, oxygen uptake and gas evolution
Fig. 1. (a) TGA thermogram of as-spun pitch fiber from 40 to 400 C at a heating rate of 1 C/min under an air atmosphere and (b) density changes of stabilized pitch fibers as a function of stabilization temperature.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
occurred simultaneously with isotropic pitch fibers [16]. It is well known that oxygen uptake took place under ambient air atmosphere, leading to a weight gain in thermal treatment of pitch fibers [15]. On the other hand, the stabilization can also induce a mass loss due to the evolution of various gases such as H2O, CO, and CO2 upon oxidative heating. Therefore, it is evident that oxygen uptake is dominant over weight loss by gas removal in the region II. However, there is a volume limitation for pitch fibers to uptake oxygen, so that in the region III, overall weight loss was observed in the TGA thermogram because gas removal by oxidative reaction on the fiber surface was dominant than the oxidative cross-linking. Densities of the fibers as a function of stabilization temperature are shown in Fig. 1b. Densities of as-spun pitch fibers and stabilized fibers were determined using the Archimedean technique. Pitch fibers were thermally treated in a convection oven under an air with a heating rate of 1 C/min from 25 to a given temperature up to 290 C, and the last sample was held for 30 min. The density of the as-spun fiber was 1.20 g/cm3. The densities of stabilized fibers heat treated up to temperature about 170 C did not significantly change. However, the densities increased with further increasing temperature, and the stabilized fibers at 290 C with 30 min holding reached the highest density of 1.39 g/cm3. Even though heat treatment atmosphere used in TGA was not exactly identical as that in convection oven, it is interesting to note that weight gain and density increase started at a similar temperature. Density is a physical value to reflect chemical composition of pitch fibers, which indicates that oxygen uptake results in increasing density when dimension of fibers remained. Therefore, it is suggested that density can be a good representative property for chemical reactions and weight changes of the fibers. Effects of stabilization on tensile properties of CFs FT-IR spectroscopy was used to investigate the chemical structural change of as-spun pitch fiber and stabilized pitch fibers heat treated from 25 to 150, 190, 230, and 290 C for 30 min with a heating rate of 1 C/min. As shown in Fig. 2a, the FT-IR spectra of the as-spun fiber and stabilized fiber at 150 C were similar, possessing bands at around 3045, 2915, 1470, and 900–700 cm1 corresponded to the aromatic C H stretching, aliphatic C H stretching, aromatic CH2 bending and aromatic C H bending (outof-plane), respectively. In addition, bands at around 2960, 2850, and 1375 cm1 are assigned to the CH3, CH2/CH3, and CH3 groups. When the stabilization temperature was higher, these bands were weakened and/or disappeared. The peaks at around 1700 cm1 contributing to a carbonyl C¼O stretching appeared with fibers heat treated up to 230 C. The appearances of new bands at around 1840, 1770 and 1300–1110 cm1 from fibers heat treated up to 290 C for 30 min were ascribed to the formation of anhydride
3
group, ester and various C O C, and OC O stretches, respectively. These appeared and intensified bands suggested that a condensation of oxygenated groups occurred, producing the cross-linked structures. As expected, these chemical structural changes were accompanied with oxygen uptake. The oxygen contents and weight gains of as-spun fiber and stabilized fibers as a function of the stabilization temperature are shown as Fig. 2b. The oxygen contents and weight gains were analyzed by elemental and TGA analysis, respectively. As Fig. 2b shows both weight gain and oxygen contents remained constant up to temperature 170 C, and then increased with further increasing temperature. However, the difference between oxygen contents and weight gain were bigger when heat treatment temperature increased. As discussed above, a higher temperature rendered a more gas evolution, which resulted in that an amount of oxygen uptake was not completely added to weight of pith fibers. From these results, the starting temperature of stabilization can be determined by weight gain, oxygen content, or density in a given heating condition. It is likely that the fibers heat treated below the temperature, in which significant changes of weight gain, oxygen content, or density were not observed, did not lead to suitable degree of stabilization. Considering a continuous stabilization process flow, the first oven temperature should be decided. In addition, residence time and temperature in consecutive oven will be critical to design cost effective stabilization, which is also related to maximize performance of resulting CFs after carbonization. In a commercial process, stabilization for pitch fibers starts around 150 C, while most pitch fibers were stabilized from 25 C to around softening point for a long time without reasonable explanation in the open literature for reporting pitch based CFs experienced batch type stabilization [17,18]. However, it is clarified that over stabilization could remove gases such as CO2 and CO, which reduce carbon yield of CFs (Fig. 1a). Therefore, we demonstrated a systematic study on stabilization of pitch fibers to control starting temperature and residence time by a heating rate. For a further study to optimize temperature conditions and correlate these to performance of resulting CFs, stabilized fibers were prepared with varied conditions. Fig. 3a–d shows TGA thermograms with heating rates of 0.5, 1, 5, 10 C/min, respectively. When heating rate was 0.5 C/min, time and temperature for reaching starting point of significant increase in weight were found to be 260 min and 170 C. As expected, increasing heating rate reduced the time to reach the maximum weight from 260 to 16 min and increased the temperature from 170 to 200 C (Table 1). In order to investigate the effect of starting temperature and heating speed, the pitch fibers were stabilized with starting temperatures of 25, 150, 170, 190 and 230 C and heating rates of 0.5 to 10 C/min (Table 2). All fibers held for 30 min at a final temperature of 290 C, and it took for 36 to 560 min depending on the condition. While densities were similar with a given heating
Fig. 2. (a) FT-IR spectra and (b) oxygen contents and weight gain of stabilized pitch fibers at various temperatures.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
4
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 3. Oxygen take-ups of pitch fibers at various heating rates of 0.5, 1, 5, and 10 C/min.
Table 1 Effect of heating rate on performance of thermal stabilization. Heating rate ( C/min)
Tonset ( C)
Tmax ( C)
Max. weight gain (%)
Oxygen content (wt.%)
0.5 1 5 10
170 175 200 200
298 312 348 354
110.20 109.88 108.03 106.03
15.91 14.46 12.68 11.79
Table 2 The run-times of thermal stabilization process with various heating rate-starting temperature profiles. Heating rate ( C/min)
Starting temperature ( C)/run-time (min)
Target temperature ( C)/holding time (min)
25
150
170
190
230
290
0.5 1 5 10
530 265 53 26.5
280 140 28 14
240 120 24 12
200 100 20 10
120 60 12 6
30 30 30 30
Fig. 4. (a) Densities of stabilized pitch fibers as a function of starting temperature of stabilization with various heating rates of 0.5, 1, 5, and 10 C/min and (b) their oxygen contents and weight gain from TGA in Fig. 3.
rate regardless of starting temperature, a slower heating rendered higher densities (Fig. 4a). Even though a lower heating rate led to higher oxygen content, a starting temperature change did not render significant difference in oxygen content (Fig. 4b). Using these TGA thermograms, the activation energy of stabilization
were calculated. Kissinger suggested a mathematical equation to obtain activation energy of decomposition with minerals based on a first-order law and the Arrhenius equation by differential thermal analysis with various heating rates [19]. The following equation has been widely used to calculate activation energies for polymer
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
crystallization [20], nucleation [21], decomposition [22], and even carbonization [23], h i d ln ’=T 2 DE ¼ R dð1=T Þ where ’, T, R, and DE are heating rate, peak temperature, the gas constant, and activation energy, respectively. With details such as peak temperatures and a straight line in the plot of lnð’=T 2 Þ vs. 1/T, DE was found to be 141 kJ/mol (Table S1 and Fig. S1). Because there is no open literature to reveal the activation energy of stabilization, it is difficult to compare the value. However, it is interesting to note that activation energy of decomposition of polymers such as high density polyethylene (HDPE), low density polyethylene (LDPE), linear low density polyethylene (LLDPE), cross-linked polyethylene (XLPE), and polypropylene (PP) ranged 333–343 kJ/mol, 188– 199 kJ/mol, 219–230 kJ/mol, 170 kJ/mol and 179–188 kJ/mol, respectively [24–26], indicating that the calculated value for the activation energy of stabilization is quite acceptable. All stabilized fibers were carbonized up to 1200 C with a heating rate of 5 C/min. Tensile properties of resulting CFs are summarized in Fig. 5. CFs stabilized with the starting temperatures from 25 to 170 C showed similar tensile strength and modulus of 1.0 and 50 GPa, respectively, without a regard of heating rate. The maximum elongation of CFs was 2% with starting temperature of 25 C. When starting temperature was 190 C, both tensile strength and modulus slightly decreased down to 0.9 and 47 GPa, respectively. However, a significant decrease of tensile properties was observed with starting temperature of 230 C. In case of heating rate of 10 C/min, tensile strengths were 0.9 and 0.8 GPa with starting temperatures of 25 and 190 C, which were slightly lower compared to values from CFs experienced lower heating rates. Despite of the significant changes in the densities and chemical compositions during stabilization, no significant change in diameter of 11 mm was observed. In case of starting temperature of 230 C, samples were fused even with a slow heating rate, which made hard to separate a single fiber from a
5
bundle for tensile test as shown in Fig. 6b and d. It indicates that the temperature was too high as a starting point to stabilize pitch fibers. It is noteworthy that even though total process time for sample with a heating rate and starting temperature of 0.5 C/min and 170 C (270 min), respectively, was more than four times longer than that for sample with a heating rate and starting temperature of 10 C/min and 25 C (56.5 min), respectively, they showed comparable properties with tensile strength of 0.97 0.07 and 0.94 0.09 and modulus of 52.85 5.48 and 49.72 1.47 GPa, respectively. When a heating rate was very low (0.5 C/min), stabilization at a low temperature range (25–150 C) was not necessary, so that starting temperature can be adjusted by TGA and the total stabilization time had decreased by half. However, it still took a long residence time for 270 min. On the contrary, a high heating rate (10 C/min) needed a low starting temperature to expose fibers in an air atmosphere at a low temperature range (25– 150 C) even though an amount of oxygen uptake was not significant, which takes an advantage of saving much times down to 56.5 min. It is recalled in Table 3 that pitch fibers in most literatures were stabilized for a long time from 180 to 1900 min because process started at a room temperature with a very slow heating rate from 0.5 to 2 C/min [27–38]. It was believed that the significantly long treatment was required for complete oxidation and stabilization due to a slow diffusion of oxygen into pitch fibers. Therefore, it took 1900 min when tape-shape fibers with a thickness of 20 mm were stabilized. Obviously, those profiles are not affordable and feasible so that TGA thermograms have been utilized to decide starting temperature in a commercial manufacturing. The target temperature could depend on softening points and types of pitches. We mimicked those concerns and realized temperature profiles, which were determined by TGA thermograms and tensile properties of resulting CFs. It seemed to be successful because stabilization time reduced from 560 to 270 min maintaining tensile strength. However, one of combinations with two variables such as a starting temperature and heating rate of 25 C and 10 C/min,
Fig. 5. (a) Tensile strength, (b) tensile modulus, and (c) elongation of carbon fibers as a function of starting temperature of stabilization with various heating rates of 0.5, 1, 5, and 10 C/min.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
6
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 6. SEM images of stabilized fibers at 290 C for 30 min at different heating rates and starting temperatures. (a) 0.5 C/min—190 C, (b) 0.5 C/min—230 C, (c) 10 C/min— 190 C, (d) 10 C/min—230 C.
Table 3 Examples of stabilization profiles of pitch-based fiber [27–38]. Raw material
Softening point ( C)
Starting temperature ( C)
Heating rate ( C/min)
Target Holing time (min) temperature ( C)
Total time (min)
Elucidatory notes
References
MPa (CTc)
275
Room temperature
0.5
320
15
605
–
Matsumoto et al. [27]
1 2 0.5
340 360 300
15 15 –
330 182 550
–
Miura et al. [28]
6 (to 180 C)
280
30 (at 180 C)
176–456
Robinson et al. [29]
1 (to 280 C) –
180–280
20-300 (at 280 C) A few hours
Ribbon-shaped fiber –
Ogale et al. [30]
5 (to 130 C)
300
60 (at 220 C)
A few hours 371
HNO3 treatment
Vilaplana-Ortego et al. [31]
IPb, MP (petroleum) MP (naphthalene)
IP: 249 MP: 270–300 –
MP
–
IP (petroleum)
Low SP: 162– 168 High SP: 267
Petroleum pitch IP (2 MNd)
295
Room temperature Room temperature Room temperature Room temperature
Room temperature Room temperature
Low SP: 90– 183 High SP: 219– 252 e IP (NCO ) 260 Room temperature 250 Room IP (PFOf) temperature 240–260 Room IP (HPCg) temperature Room 260 IP (PFO, slurry oil, CT) temperature Room MP 265 temperature (naphthalene) 290 170 IP (petroleum) Room temperature a b MP: mesophase IP: isotropic pitch pitch f PFO: pyrolyzed gHPC: hyperfuel oil coal
1 (to 220 C) 1 (to 300 C) 2
120 (at 300 C) 320
120
267.5
–
Wazir et al. [32]
1
300
60
335
–
Kim et al. [33]
0.5
270
60
550
–
Kim et al. [34]
1
250
120
345
E-beam treatment
Park et al. [35]
0.5
270
60
550
–
Yang et al. [36]
1
280
60
315
–
Kim et al. [37]
2
260
60-1800
Tape-shaped fiber
Yuan et al. [38]
0.5 10
290 290
30 30
177.51917.5 270 56.5
–
Our research
c
CT: coal tar
respectively, led to a total residence time of 56.5 min with a similar tensile strength. This indicates that when pitch fibers were heated from a low temperature, a relatively high heating rate could be
d
2 MN: 2methylnaphthalene
e NCO: naphthacracked oil
suitable for stabilization. On the contrary, when starting temperature was 150 C with a heating rate of 10 C/min, a decrease of tensile strength was observed.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
7
stabilization. It is supposed in region I that oxygen could be absorbed on the surface of pitch fibers, and at most only very thin skin could be oxidized, resulting in infusible out-layer by crosslinking, because thermal energy was too low for oxygen to diffuse into the fibers inside. All of sudden, heating above 150 C catalyzed the diffusion of a large amount of oxygen significantly, which were readily on the surface. Therefore, such a high heating rate of 10 C/ min rendered an efficient temperature profile for preparing CFs. Considering stabilization from 150 C without a pre-heating, adsorption and diffusion occurred simultaneously by a sudden heat shock, which required a low heating rate for homogeneous oxygen distribution. As a result, the combination of 150 C and 10 C/min reduced tensile strength down to 0.85 0.08 GPa. Therefore, starting stabilization at such as a high temperature was not prerequisite for a cost effective process. These suggest that more concerned design is required to choose complementary variables such as starting temperature and heating rate, which significantly reduce residence time, maintaining mechanical properties of isotropic pitch based CFs. Conclusion
Fig. 7. C1s X-ray photoelectron spectra of (a) as-spun pitch fiber and stabilized pitch fibers with a heating rate of 10 C/min and different stabilization temperature profiles: (b) 25–150 C, (c) 150–290 C and (d) 25–290 C.
Fig. 7 presents the curve-fitted C1s region of the XPS for as-spun pitch fiber and stabilized pitch fibers with a heating rate of 10 C/ min and different stabilization temperature profiles. The C1s peaks are deconvoluted into six individual peaks that aromatic and aliphatic carbon (peak 1, 284.6–285.1 eV), single C O bond (peak 2, 286.1–286.6 eV), double bonded carbon oxygen groups (peak 3, 287.3 eV), carboxyl groups (peak 4, 288.6–289.1 eV), carbonate or COx groups (peak 5, 290.6 eV), and plasmon or shake-up satellite peaks due to p–p* transitions in aromatic rings (peak 6, 291.6 eV) [39–43]. The deconvolution details are summarized in Table S2. The XPS spectrum of as-spun pitch fiber mainly exhibited the peak 1, which corresponded to aromatic and aliphatic carbon with 70.4% of the total C1s. There is a small amount of oxygen related bonds, which could originate from an air atmosphere. Compared to as-spun pitch fiber, stabilized fiber up to 150 C (Fig. 7b) showed a decrease of aromatic and aliphatic carbon down to 33.7% of relative area and an increase in all oxygen-containing functional groups on the surface. It is recalled that there was no significant difference in the elemental analysis and weight gain (Fig. 2b) between as-spun and stabilized fibers up to 150 C, indicating that oxygen related bonds exist dominantly on the surface of the fibers. On the other hand, stabilized fibers from 150 to 290 C and 25 to 290 C (Fig. 7c, d) showed significant decreases of oxygencontaining functionalities. It is believed that oxygen, which was existed in a state of adsorbed on the surface of fiber, diffused into the fiber when temperature increased above 150 C. It is evident that even though there was no significant weight gain from 25 C to 150 C (region I, Fig. 1) in TGA thermograms, thermal history at this range play an important role in accelerating
Petroleum-based isotropic pitch fibers were stabilized at various starting temperatures and heating rates at 290 C for 30 min. Thermogravimetric analysis of spun fibers showed three regions depending on weight changes with increasing temperature under all heating rates. In the case of heating rate of 1 C/min, the weight slightly increased less than 1% up to 175 C, then gradually increased up to 310 C and showed a maximum weight gain of 9.88%. Finally, an apparent weight loss occurred over 310 C. Interestingly, the densities of as-spun fiber and stabilized pitch fibers heat treated up to temperature about 170 C was 1.20 g/cm3. The densities increased with increasing stabilization temperature, and the maximum value of 1.39 g/cm3 was found to be at 290 C with 30 min holding. From these results, the temperature of 175 C was considered as a starting temperature to yield a suitable degree of stabilization for fibers. Thus, skipping unnecessary region (temperature below 170 C) can shorten the stabilization process. To further optimize stabilization conditions, pitch fibers were stabilized with starting temperatures of 25, 150, 170, 190 and 230 C and heating rates of 0.5, 1, 5, and 10 C/min. As a result, total stabilization time has decreased by half down to 270 min (starting temperature of 170 C with heating rate of 0.5 C/min) compared to the conventional process, which stabilization started at 25 C. In addition, a heating rate and starting temperature of 10 C/min and 25 C, respectively, led to the total stabilization time of 56.5 min. The fibers showed comparable mechanical properties with tensile strength of 0.97 0.07 and 0.94 0.09 GPa and modulus of 52.85 5.48 and 49.72 1.47 GPa, respectively. Therefore, we believe that choosing starting temperature by TGA thermograms was not prerequisite and combination of starting temperature and heating rate play an important role to optimize stabilization condition. Acknowledgements This work was supported by a grant from the Korea Institute of Science and Technology (KIST) Institutional program and the Industrial Core Technology Development Program funded by the Ministry of Trade, Industry and Energy, Republic of Korea (No. 10052760). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.jiec.2017.09.048.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
W. Ruland, Carbon Fibers Adv. Mater. 2 (1990) 528. X. Huang, Materials 2 (2009) 2369. Y. Liu, S. Kumar, Polym. Rev. 52 (2012) 234. R. Eslami Farsani, R. Fazaeli, Int. J. Chem. Mol. Nucl. Mater. Metall. Eng. 6 (2012) 836. J.D. Buckley, D.D. Edie, Carbon–Carbon Materials and Composites, William Andrew, New Jersey, 1993. S.-H. Yoon, Y. Korai, I. Mochida, Carbon 32 (1994) 281. M.S.A. Rahaman, A.F. Ismail, A. Mustafa, Polym. Degrad. Stabil. 92 (2007) 1421. T. Matsumoto, Carbon 31 (1993) 143. R. Didchenko, C.D. Amata, US Patent 3954947 (1976). E. Fitzer, D.J. Muller, Carbon 13 (1975) 63. S.-S. Tzeng, J.-H. Pan, Mater. Chem. Phys. 74 (2002) 214. J.-I. Hayashi, M. Nakashima, K. Kusakabe, S. Morooka, S. Mitsuda, Carbon 33 (1995) 1567. E. Fitzer, W. Frohs, M. Heine, Carbon 24 (1986) 387. S.Y. Jang, S. Ko, Y.P. Jeon, J. Choi, N. Kang, H.C. Kim, H.-I. Joh, S. Lee, J. Ind. Eng. Chem. 45 (2017) 316. J. Drbohlav, W.T.K. Stevenson, Carbon 33 (1995) 693. J. Zhu, S.W. Park, H.-I. Joh, H.C. Kim, S. Lee, Macromol. Res. 23 (2015) 79. V. Ahmad, Y. Zhongren, F. Youqing, FTA Report (2012). P. Morgan, Carbon Fibers and Their Composites, CRC Press, Taylor & Francis Group, Florida, 2005. H.E. Kissinger, J. Res. Natl. Bur. Stand. 57 (1956) 217. S. Vyazovkin, N. Sbirrazzuoli, J. Therm. Anal. Calorim. 72 (2003) 681. K.M. Seven, J.M. Cogen, J.F. Gilchrist, Polym. Eng. Sci. 56 (2016) 541. P. Rantuch, D. Ka9 cíková, B. Nagypál, Eur. J. Environ. Saf. Sci. 2 (2014) 12. J. Ma, X. Meng, X. Guo, Y. Lan, S. Zang, PLoS One 12 (2017) 1.
[24] Y.-S. Cho, M.-J. Shim, S.-W. Kim, Mater. Chem. Phys. 52 (1998) 94. [25] J.W. Park, S.C. Oh, H.P. Lee, H.T. Kim, K.O. Yoo, Korean J. Chem. Eng. 17 (2000) 489. [26] A. Aboulkas, K. El Harfi, A. El Bouadili, Energy Convers. Manage. 51 (2010) 1363. [27] T. Matsumoto, I. Mochida, Carbon 30 (1992) 1041. [28] K. Miura, H. Nakagawa, K. Hashimoto, Carbon 33 (1995) 275. [29] K.E. Robinson, D.D. Edie, Carbon 34 (1996) 13. [30] A.A. Ogale, C. Lin, D.P. Anderson, K.M. Kearns, Carbon 40 (2002) 1309. [31] E. Vilaplana-Ortego, J. Alcañiz-Monge, D. Cazorla-Amorós, A. Linares-Solano, Carbon 41 (2003) 1001. [32] A.H. Wazir, L. Kakakhel, New Carbon Mater. 24 (2009) 83. [33] B.-J. Kim, H. Kil, N. Watanabe, M.-H. Seo, B.-H. Kim, K.S. Yang, O. Kato, J. Miyawaki, I. Mochida, S.-H. Yoon, Curr. Org. Chem. 17 (2013) 1463. [34] B.-J. Kim, Y. Eom, O. Kato, J. Miyawaki, B.C. Kim, I. Mochida, S.-H. Yoon, Carbon 77 (2014) 747. [35] M.-S. Park, Y. Ko, M.-J. Jung, Y.-S. Lee, Carbon Lett. 16 (2015) 121. [36] J. Yang, K. Nakabayashi, J. Miyawaki, S.-H. Yoon, Carbon 106 (2016) 28. [37] J. Kim, U.-S. Im, B. Lee, D.-H. Peck, S.-H. Yoon, D.-H. Jung, Carbon Lett. 19 (2016) 72. [38] G. Yuan, X. Li, X. Xiong, Z. Dong, A. Westwood, B. Li, C. Ye, G. Ma, Z. Cui, Y. Cong, J. Zhang, Y. Li, Carbon 115 (2017) 59. [39] E. Desimoni, G.I. Casella, A. Morone, A.M. Salvi, Surf. Interface Anal. 15 (1990) 627. [40] H. Shono, M. Matsumoto, X. Wang, M. Takeuchi, Y. Sendoda, A. Ishihara, T. Kabe, Sekiyu Gakkaishi 34 (1991) 510. [41] J.P. Boudou, J.I. Paredes, A. Cuesta, A. Martinez-Alonso, J.M.D. Tascon, Carbon 41 (2003) 41. [42] C. Moreno-Castilla, M.V. López-Ramón, F. Carrasco-Marín, Carbon 38 (2000) 1995. [43] T.H. Lim, S.Y. Yeo, Sci. Rep. 7 (2017) 4733.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers Hyun-Sig Kila,1, So Young Janga,b,1, Seunghyun Koc,d , Young Pyo Jeonc,d , Hwan-Chul Kimb , Han-Ik Johe, Sungho Leea,d,* a Carbon Composite Materials Research Center, Korea Institute of Science and Technology, 92 Chudong-ro, Bongdong-eup, Wanju-gun, Jeonbuk 55324, Republic of Korea b Department of Organic Materials and Fiber Engineering, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju, Jeonbuk 54896, Republic of Korea c Center for C-Industry Incubation, Korea Research Institute of Chemical Technology, 141 Gajeong-ro, Yuseong-gu, Daejeon 34114, Republic of Korea d Department of Nano Material Engineering, Korea University of Science and Technology, 217 Gajeong-ro, Yuseong-gu, Daejeon 34113, Republic of Korea e Department of Energy Engineering, Konkuk University, 120 Neungdong-ro, Gwangjin-gu, Seoul 05029, Republic of Korea
A R T I C L E I N F O
Article history: Received 31 August 2017 Received in revised form 28 September 2017 Accepted 28 September 2017 Available online xxx Keywords: Isotropic pitch Stabilization Tensile properties Carbon fiber
A B S T R A C T
Optimization of the stabilization process for manufacturing low-cost carbon fibers was performed. By simply varying both heating rate (0.5–10 C/min) and starting temperature (25–230 C), the resulting fibers possessed different properties such as densities, oxygen contents, and weight gains due to the oxygen up-takes. Subsequent oxidative reactions led to changes in chemical compositions of the fibers. Therefore, this study suggests that more concerned design with complementary parameters such as starting temperature and heating rate, significantly reducing stabilization time down to 56.5 min with the comparable mechanical properties of the conventional isotropic pitch based CFs. © 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction Carbon fibers (CFs) are used in various applications such as automobile, aerospace, sports, and many other components because of their excellent mechanical, thermal, and electrical properties [1–4]. However, the aforementioned applications using CFs have been limited so far since a large scale CFs production cost is prohibitively expensive for the commercialization. Therefore, significant research effort in the last several decades has been devoted to reducing CFs production cost by developing cheaper precursor materials, minimizing stabilization time, simplifying carbonization, or etc. A typical CFs manufacturing process flow consists of fiber spinning, oxidative stabilization, carbonization, graphitization, and surface treatment process [5]. Among these processes, the stabilization is the most important step because the oxidative behavior of precursor fiber greatly affects the final mechanical properties of the CFs. Also,
* Corresponding author at: Carbon Composite Materials Research Center, Korea Institute of Science and Technology, 92 Chudong-ro, Bongdong-eup, Wanju-gun, Jeonbuk 55324, Republic of Korea. Fax: +82 63 710 7589. E-mail address: [email protected] (S. Lee). 1 Authors contributed equally to this work.
stabilization process is considered as the most time consuming and expensive step. Long time stabilization at relatively low temperature used in the conventional process flow was considered to allowing oxidant to uniformly diffuse through precursor fiber [6]. Without this uniform oxidant diffusion through the fiber precursor, uneven stabilization or rapid oxidation can yield the final carbon fiber possessing poor properties [7,8]. In this regard, it is challenging to achieving the reduced stabilization time while maintaining a suitable degree of stabilization. Various process modifications such as stabilization temperature, time, heating rate, atmosphere, pressure, and multi-step stabilization [9–13] have been proposed. Many companies and researchers have been striving for reducing stabilization process, but it still remained difficult to optimize and deploy reliably. In this study, we report a systematic study on optimizing the stabilization process by varying heating conditions such as a heating speed and the starting temperature. Based on our previous study [14], the stabilized fibers with a density range of 1.35–1.36 g/ cm3 yielded CF with the highest tensile strength regardless of the stabilization conditions. Here, petroleum-based isotropic pitch fibers were first stabilized at various temperatures in ambient air atmosphere at heating rate of 0.5–10 C/min and then densities of the corresponding fibers were characterized. After the oxidative
https://doi.org/10.1016/j.jiec.2017.09.048 1226-086X/© 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
2
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
stabilization at various conditions, the stabilized fibers were carbonized at 1200 C under nitrogen atmosphere. Then, chemical and physical properties of stabilized fibers and tensile properties of the CFs were measured and then correlated. The study revealed a possibility to shorten the stabilization process time by eliminating unnecessary temperature ranges. In addition, a combination of variables such as starting temperature and heating rate was investigated for a further reducing stabilization time. We, therefore, demonstrated a procedure to optimize stabilization for developing a cost effective process. Experimental Preparation of pitch-based carbon fibers A commercial isotropic pitch (SN-280, Anshan, China) was melt-spun into pitch fibers using a 12-hole spinneret at 335 C with a diameter of 150 mm and an aspect ratio of 3 under nitrogen atmosphere. The prepared pitch fibers possessed the diameter of 12.2 0.8 mm. The as-spun pitch fibers were stabilized at 290 C for 30 min using a convection oven in ambient air atmosphere with various starting temperatures of 25, 150, 170, 190 and 230 C and various heating rates of 0.5, 1, 5 and 10 C/min. The stabilized fibers were then carbonized at 1200 C under nitrogen flow in a tube furnace at a heating rate of 5 C/min. Characterization The thermal properties of as-spun pitch fibers were analyzed using a thermogravimetric apparatus (SETSYS Evolution TGA, SETARAM Instrumentation, Caluire, France). Specimens were heated from 40 C to 400 C at a heating rate of 0.5, 1, 5 and 10 C with an air flow rate of 50 mL/min. The stabilization starting temperatures and the amounts of oxygen uptake were monitored using a thermogravimetric analyzer. The weight of pitch fibers used for the each analysis was approximately 10 mg. The densities of as-spun pitch fibers and stabilized fibers were determined by a Sartorius YDK03 Density Determination Kit (Sartorius AG, Goettingen, Germany). The Archimedean principle is applied for measuring the density of pitch fibers. Another method of density measurement, a density gradient column of liquid (POLYTEST, RayRan, UK) was prepared to determine the densities carbonized fibers. Two liquids, 1,1,2,2-tetrabromoethane (98%, Daejung Chemical Co., Korea) and benzene (99.5%, Daejung Chemical Co., Korea), were used to build the density gradient, and glass beads with accurately known densities floated in the column. The each fiber samples were inserted and left for 8 h in the column. At least three specimens were tested to calculate the densities of the fibers in each condition.
FT-IR spectroscopy (Nicolet IS10, USA) was performed to investigate the change in the oxygen functional groups during the stabilization using the KBr pellet technique. The mixed pellet was prepared with 400 mg of spectrometric grade KBr and 2 mg of sample. The mixture of the sample and the salt was crushed using a pestle and a mortar. The powder mixture was placed in a KBr die kit and pressed with a hydraulic laboratory press. Each sample was scanned 16 times at a resolution of 16 cm1 with a range of 4000– 400 cm1. All spectra were collected in the absorbance mode with an automatic baseline correction and then transformed to transmittance. Elemental analysis (Flash 2000, Thermo Scientific, USA) was carried out to measure the amounts of oxygen in the bulk of the CF samples. Oxygen was analyzed using helium gas at 1060 C for 500 s. The mechanical properties of the CFs were examined by a mechanical tester for a single fiber (FAVIMAT+, Textechno, Germany) with a test speed of 5 mm/min. The gauge length was 25 mm, and 20 specimens were measured for each experimental point. X-ray photoelectron spectroscopy (XPS, K-alpha, Thermo Scientific, USA) analysis was performed using monochromated Al Ka (1486.6 eV) X-rays to examine the type of chemical bond of nitrogen and the elemental composition on the surface of as-spun and stabilized pitch fiber samples. The survey spectrum was collected from 0 eV to 1350 eV, and the binding energies were referenced to the C1s line at 284.8 eV. Results and discussion Thermal analysis of the stabilized pitch fibers The thermal characteristics of as-spun pitch fibers were analyzed using TGA. Fig. 1a shows a TGA thermogram obtained from as-spun pitch fiber to mimic oxidative stabilization at a heating rate of 1 C/min under ambient air atmosphere. There were three regions showing different weight changes as a function of temperature. A slight increase in weight was observed from 40 to 175 C (region I) with increasing the temperature. Then, the weight significantly increased up to 310 C (region II) with a weight gain of 10 wt.%. Finally, at temperature above 310 C, there was a significant weight loss (region III). In the region I, a small amount of weight gain was observed even though pitch fibers were exposed under oxygen, indicating that temperature was not high enough to overcome energy barrier for inducing oxygen diffusion into the fibers. It is interesting to observe that the weight gain dramatically increased from 175 C due to oxygen diffusion in the region II. From our previous result by in-situ mass spectroscopy, oxygen uptake and gas evolution
Fig. 1. (a) TGA thermogram of as-spun pitch fiber from 40 to 400 C at a heating rate of 1 C/min under an air atmosphere and (b) density changes of stabilized pitch fibers as a function of stabilization temperature.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
occurred simultaneously with isotropic pitch fibers [16]. It is well known that oxygen uptake took place under ambient air atmosphere, leading to a weight gain in thermal treatment of pitch fibers [15]. On the other hand, the stabilization can also induce a mass loss due to the evolution of various gases such as H2O, CO, and CO2 upon oxidative heating. Therefore, it is evident that oxygen uptake is dominant over weight loss by gas removal in the region II. However, there is a volume limitation for pitch fibers to uptake oxygen, so that in the region III, overall weight loss was observed in the TGA thermogram because gas removal by oxidative reaction on the fiber surface was dominant than the oxidative cross-linking. Densities of the fibers as a function of stabilization temperature are shown in Fig. 1b. Densities of as-spun pitch fibers and stabilized fibers were determined using the Archimedean technique. Pitch fibers were thermally treated in a convection oven under an air with a heating rate of 1 C/min from 25 to a given temperature up to 290 C, and the last sample was held for 30 min. The density of the as-spun fiber was 1.20 g/cm3. The densities of stabilized fibers heat treated up to temperature about 170 C did not significantly change. However, the densities increased with further increasing temperature, and the stabilized fibers at 290 C with 30 min holding reached the highest density of 1.39 g/cm3. Even though heat treatment atmosphere used in TGA was not exactly identical as that in convection oven, it is interesting to note that weight gain and density increase started at a similar temperature. Density is a physical value to reflect chemical composition of pitch fibers, which indicates that oxygen uptake results in increasing density when dimension of fibers remained. Therefore, it is suggested that density can be a good representative property for chemical reactions and weight changes of the fibers. Effects of stabilization on tensile properties of CFs FT-IR spectroscopy was used to investigate the chemical structural change of as-spun pitch fiber and stabilized pitch fibers heat treated from 25 to 150, 190, 230, and 290 C for 30 min with a heating rate of 1 C/min. As shown in Fig. 2a, the FT-IR spectra of the as-spun fiber and stabilized fiber at 150 C were similar, possessing bands at around 3045, 2915, 1470, and 900–700 cm1 corresponded to the aromatic C H stretching, aliphatic C H stretching, aromatic CH2 bending and aromatic C H bending (outof-plane), respectively. In addition, bands at around 2960, 2850, and 1375 cm1 are assigned to the CH3, CH2/CH3, and CH3 groups. When the stabilization temperature was higher, these bands were weakened and/or disappeared. The peaks at around 1700 cm1 contributing to a carbonyl C¼O stretching appeared with fibers heat treated up to 230 C. The appearances of new bands at around 1840, 1770 and 1300–1110 cm1 from fibers heat treated up to 290 C for 30 min were ascribed to the formation of anhydride
3
group, ester and various C O C, and OC O stretches, respectively. These appeared and intensified bands suggested that a condensation of oxygenated groups occurred, producing the cross-linked structures. As expected, these chemical structural changes were accompanied with oxygen uptake. The oxygen contents and weight gains of as-spun fiber and stabilized fibers as a function of the stabilization temperature are shown as Fig. 2b. The oxygen contents and weight gains were analyzed by elemental and TGA analysis, respectively. As Fig. 2b shows both weight gain and oxygen contents remained constant up to temperature 170 C, and then increased with further increasing temperature. However, the difference between oxygen contents and weight gain were bigger when heat treatment temperature increased. As discussed above, a higher temperature rendered a more gas evolution, which resulted in that an amount of oxygen uptake was not completely added to weight of pith fibers. From these results, the starting temperature of stabilization can be determined by weight gain, oxygen content, or density in a given heating condition. It is likely that the fibers heat treated below the temperature, in which significant changes of weight gain, oxygen content, or density were not observed, did not lead to suitable degree of stabilization. Considering a continuous stabilization process flow, the first oven temperature should be decided. In addition, residence time and temperature in consecutive oven will be critical to design cost effective stabilization, which is also related to maximize performance of resulting CFs after carbonization. In a commercial process, stabilization for pitch fibers starts around 150 C, while most pitch fibers were stabilized from 25 C to around softening point for a long time without reasonable explanation in the open literature for reporting pitch based CFs experienced batch type stabilization [17,18]. However, it is clarified that over stabilization could remove gases such as CO2 and CO, which reduce carbon yield of CFs (Fig. 1a). Therefore, we demonstrated a systematic study on stabilization of pitch fibers to control starting temperature and residence time by a heating rate. For a further study to optimize temperature conditions and correlate these to performance of resulting CFs, stabilized fibers were prepared with varied conditions. Fig. 3a–d shows TGA thermograms with heating rates of 0.5, 1, 5, 10 C/min, respectively. When heating rate was 0.5 C/min, time and temperature for reaching starting point of significant increase in weight were found to be 260 min and 170 C. As expected, increasing heating rate reduced the time to reach the maximum weight from 260 to 16 min and increased the temperature from 170 to 200 C (Table 1). In order to investigate the effect of starting temperature and heating speed, the pitch fibers were stabilized with starting temperatures of 25, 150, 170, 190 and 230 C and heating rates of 0.5 to 10 C/min (Table 2). All fibers held for 30 min at a final temperature of 290 C, and it took for 36 to 560 min depending on the condition. While densities were similar with a given heating
Fig. 2. (a) FT-IR spectra and (b) oxygen contents and weight gain of stabilized pitch fibers at various temperatures.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
4
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 3. Oxygen take-ups of pitch fibers at various heating rates of 0.5, 1, 5, and 10 C/min.
Table 1 Effect of heating rate on performance of thermal stabilization. Heating rate ( C/min)
Tonset ( C)
Tmax ( C)
Max. weight gain (%)
Oxygen content (wt.%)
0.5 1 5 10
170 175 200 200
298 312 348 354
110.20 109.88 108.03 106.03
15.91 14.46 12.68 11.79
Table 2 The run-times of thermal stabilization process with various heating rate-starting temperature profiles. Heating rate ( C/min)
Starting temperature ( C)/run-time (min)
Target temperature ( C)/holding time (min)
25
150
170
190
230
290
0.5 1 5 10
530 265 53 26.5
280 140 28 14
240 120 24 12
200 100 20 10
120 60 12 6
30 30 30 30
Fig. 4. (a) Densities of stabilized pitch fibers as a function of starting temperature of stabilization with various heating rates of 0.5, 1, 5, and 10 C/min and (b) their oxygen contents and weight gain from TGA in Fig. 3.
rate regardless of starting temperature, a slower heating rendered higher densities (Fig. 4a). Even though a lower heating rate led to higher oxygen content, a starting temperature change did not render significant difference in oxygen content (Fig. 4b). Using these TGA thermograms, the activation energy of stabilization
were calculated. Kissinger suggested a mathematical equation to obtain activation energy of decomposition with minerals based on a first-order law and the Arrhenius equation by differential thermal analysis with various heating rates [19]. The following equation has been widely used to calculate activation energies for polymer
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
crystallization [20], nucleation [21], decomposition [22], and even carbonization [23], h i d ln ’=T 2 DE ¼ R dð1=T Þ where ’, T, R, and DE are heating rate, peak temperature, the gas constant, and activation energy, respectively. With details such as peak temperatures and a straight line in the plot of lnð’=T 2 Þ vs. 1/T, DE was found to be 141 kJ/mol (Table S1 and Fig. S1). Because there is no open literature to reveal the activation energy of stabilization, it is difficult to compare the value. However, it is interesting to note that activation energy of decomposition of polymers such as high density polyethylene (HDPE), low density polyethylene (LDPE), linear low density polyethylene (LLDPE), cross-linked polyethylene (XLPE), and polypropylene (PP) ranged 333–343 kJ/mol, 188– 199 kJ/mol, 219–230 kJ/mol, 170 kJ/mol and 179–188 kJ/mol, respectively [24–26], indicating that the calculated value for the activation energy of stabilization is quite acceptable. All stabilized fibers were carbonized up to 1200 C with a heating rate of 5 C/min. Tensile properties of resulting CFs are summarized in Fig. 5. CFs stabilized with the starting temperatures from 25 to 170 C showed similar tensile strength and modulus of 1.0 and 50 GPa, respectively, without a regard of heating rate. The maximum elongation of CFs was 2% with starting temperature of 25 C. When starting temperature was 190 C, both tensile strength and modulus slightly decreased down to 0.9 and 47 GPa, respectively. However, a significant decrease of tensile properties was observed with starting temperature of 230 C. In case of heating rate of 10 C/min, tensile strengths were 0.9 and 0.8 GPa with starting temperatures of 25 and 190 C, which were slightly lower compared to values from CFs experienced lower heating rates. Despite of the significant changes in the densities and chemical compositions during stabilization, no significant change in diameter of 11 mm was observed. In case of starting temperature of 230 C, samples were fused even with a slow heating rate, which made hard to separate a single fiber from a
5
bundle for tensile test as shown in Fig. 6b and d. It indicates that the temperature was too high as a starting point to stabilize pitch fibers. It is noteworthy that even though total process time for sample with a heating rate and starting temperature of 0.5 C/min and 170 C (270 min), respectively, was more than four times longer than that for sample with a heating rate and starting temperature of 10 C/min and 25 C (56.5 min), respectively, they showed comparable properties with tensile strength of 0.97 0.07 and 0.94 0.09 and modulus of 52.85 5.48 and 49.72 1.47 GPa, respectively. When a heating rate was very low (0.5 C/min), stabilization at a low temperature range (25–150 C) was not necessary, so that starting temperature can be adjusted by TGA and the total stabilization time had decreased by half. However, it still took a long residence time for 270 min. On the contrary, a high heating rate (10 C/min) needed a low starting temperature to expose fibers in an air atmosphere at a low temperature range (25– 150 C) even though an amount of oxygen uptake was not significant, which takes an advantage of saving much times down to 56.5 min. It is recalled in Table 3 that pitch fibers in most literatures were stabilized for a long time from 180 to 1900 min because process started at a room temperature with a very slow heating rate from 0.5 to 2 C/min [27–38]. It was believed that the significantly long treatment was required for complete oxidation and stabilization due to a slow diffusion of oxygen into pitch fibers. Therefore, it took 1900 min when tape-shape fibers with a thickness of 20 mm were stabilized. Obviously, those profiles are not affordable and feasible so that TGA thermograms have been utilized to decide starting temperature in a commercial manufacturing. The target temperature could depend on softening points and types of pitches. We mimicked those concerns and realized temperature profiles, which were determined by TGA thermograms and tensile properties of resulting CFs. It seemed to be successful because stabilization time reduced from 560 to 270 min maintaining tensile strength. However, one of combinations with two variables such as a starting temperature and heating rate of 25 C and 10 C/min,
Fig. 5. (a) Tensile strength, (b) tensile modulus, and (c) elongation of carbon fibers as a function of starting temperature of stabilization with various heating rates of 0.5, 1, 5, and 10 C/min.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
6
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 6. SEM images of stabilized fibers at 290 C for 30 min at different heating rates and starting temperatures. (a) 0.5 C/min—190 C, (b) 0.5 C/min—230 C, (c) 10 C/min— 190 C, (d) 10 C/min—230 C.
Table 3 Examples of stabilization profiles of pitch-based fiber [27–38]. Raw material
Softening point ( C)
Starting temperature ( C)
Heating rate ( C/min)
Target Holing time (min) temperature ( C)
Total time (min)
Elucidatory notes
References
MPa (CTc)
275
Room temperature
0.5
320
15
605
–
Matsumoto et al. [27]
1 2 0.5
340 360 300
15 15 –
330 182 550
–
Miura et al. [28]
6 (to 180 C)
280
30 (at 180 C)
176–456
Robinson et al. [29]
1 (to 280 C) –
180–280
20-300 (at 280 C) A few hours
Ribbon-shaped fiber –
Ogale et al. [30]
5 (to 130 C)
300
60 (at 220 C)
A few hours 371
HNO3 treatment
Vilaplana-Ortego et al. [31]
IPb, MP (petroleum) MP (naphthalene)
IP: 249 MP: 270–300 –
MP
–
IP (petroleum)
Low SP: 162– 168 High SP: 267
Petroleum pitch IP (2 MNd)
295
Room temperature Room temperature Room temperature Room temperature
Room temperature Room temperature
Low SP: 90– 183 High SP: 219– 252 e IP (NCO ) 260 Room temperature 250 Room IP (PFOf) temperature 240–260 Room IP (HPCg) temperature Room 260 IP (PFO, slurry oil, CT) temperature Room MP 265 temperature (naphthalene) 290 170 IP (petroleum) Room temperature a b MP: mesophase IP: isotropic pitch pitch f PFO: pyrolyzed gHPC: hyperfuel oil coal
1 (to 220 C) 1 (to 300 C) 2
120 (at 300 C) 320
120
267.5
–
Wazir et al. [32]
1
300
60
335
–
Kim et al. [33]
0.5
270
60
550
–
Kim et al. [34]
1
250
120
345
E-beam treatment
Park et al. [35]
0.5
270
60
550
–
Yang et al. [36]
1
280
60
315
–
Kim et al. [37]
2
260
60-1800
Tape-shaped fiber
Yuan et al. [38]
0.5 10
290 290
30 30
177.51917.5 270 56.5
–
Our research
c
CT: coal tar
respectively, led to a total residence time of 56.5 min with a similar tensile strength. This indicates that when pitch fibers were heated from a low temperature, a relatively high heating rate could be
d
2 MN: 2methylnaphthalene
e NCO: naphthacracked oil
suitable for stabilization. On the contrary, when starting temperature was 150 C with a heating rate of 10 C/min, a decrease of tensile strength was observed.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
7
stabilization. It is supposed in region I that oxygen could be absorbed on the surface of pitch fibers, and at most only very thin skin could be oxidized, resulting in infusible out-layer by crosslinking, because thermal energy was too low for oxygen to diffuse into the fibers inside. All of sudden, heating above 150 C catalyzed the diffusion of a large amount of oxygen significantly, which were readily on the surface. Therefore, such a high heating rate of 10 C/ min rendered an efficient temperature profile for preparing CFs. Considering stabilization from 150 C without a pre-heating, adsorption and diffusion occurred simultaneously by a sudden heat shock, which required a low heating rate for homogeneous oxygen distribution. As a result, the combination of 150 C and 10 C/min reduced tensile strength down to 0.85 0.08 GPa. Therefore, starting stabilization at such as a high temperature was not prerequisite for a cost effective process. These suggest that more concerned design is required to choose complementary variables such as starting temperature and heating rate, which significantly reduce residence time, maintaining mechanical properties of isotropic pitch based CFs. Conclusion
Fig. 7. C1s X-ray photoelectron spectra of (a) as-spun pitch fiber and stabilized pitch fibers with a heating rate of 10 C/min and different stabilization temperature profiles: (b) 25–150 C, (c) 150–290 C and (d) 25–290 C.
Fig. 7 presents the curve-fitted C1s region of the XPS for as-spun pitch fiber and stabilized pitch fibers with a heating rate of 10 C/ min and different stabilization temperature profiles. The C1s peaks are deconvoluted into six individual peaks that aromatic and aliphatic carbon (peak 1, 284.6–285.1 eV), single C O bond (peak 2, 286.1–286.6 eV), double bonded carbon oxygen groups (peak 3, 287.3 eV), carboxyl groups (peak 4, 288.6–289.1 eV), carbonate or COx groups (peak 5, 290.6 eV), and plasmon or shake-up satellite peaks due to p–p* transitions in aromatic rings (peak 6, 291.6 eV) [39–43]. The deconvolution details are summarized in Table S2. The XPS spectrum of as-spun pitch fiber mainly exhibited the peak 1, which corresponded to aromatic and aliphatic carbon with 70.4% of the total C1s. There is a small amount of oxygen related bonds, which could originate from an air atmosphere. Compared to as-spun pitch fiber, stabilized fiber up to 150 C (Fig. 7b) showed a decrease of aromatic and aliphatic carbon down to 33.7% of relative area and an increase in all oxygen-containing functional groups on the surface. It is recalled that there was no significant difference in the elemental analysis and weight gain (Fig. 2b) between as-spun and stabilized fibers up to 150 C, indicating that oxygen related bonds exist dominantly on the surface of the fibers. On the other hand, stabilized fibers from 150 to 290 C and 25 to 290 C (Fig. 7c, d) showed significant decreases of oxygencontaining functionalities. It is believed that oxygen, which was existed in a state of adsorbed on the surface of fiber, diffused into the fiber when temperature increased above 150 C. It is evident that even though there was no significant weight gain from 25 C to 150 C (region I, Fig. 1) in TGA thermograms, thermal history at this range play an important role in accelerating
Petroleum-based isotropic pitch fibers were stabilized at various starting temperatures and heating rates at 290 C for 30 min. Thermogravimetric analysis of spun fibers showed three regions depending on weight changes with increasing temperature under all heating rates. In the case of heating rate of 1 C/min, the weight slightly increased less than 1% up to 175 C, then gradually increased up to 310 C and showed a maximum weight gain of 9.88%. Finally, an apparent weight loss occurred over 310 C. Interestingly, the densities of as-spun fiber and stabilized pitch fibers heat treated up to temperature about 170 C was 1.20 g/cm3. The densities increased with increasing stabilization temperature, and the maximum value of 1.39 g/cm3 was found to be at 290 C with 30 min holding. From these results, the temperature of 175 C was considered as a starting temperature to yield a suitable degree of stabilization for fibers. Thus, skipping unnecessary region (temperature below 170 C) can shorten the stabilization process. To further optimize stabilization conditions, pitch fibers were stabilized with starting temperatures of 25, 150, 170, 190 and 230 C and heating rates of 0.5, 1, 5, and 10 C/min. As a result, total stabilization time has decreased by half down to 270 min (starting temperature of 170 C with heating rate of 0.5 C/min) compared to the conventional process, which stabilization started at 25 C. In addition, a heating rate and starting temperature of 10 C/min and 25 C, respectively, led to the total stabilization time of 56.5 min. The fibers showed comparable mechanical properties with tensile strength of 0.97 0.07 and 0.94 0.09 GPa and modulus of 52.85 5.48 and 49.72 1.47 GPa, respectively. Therefore, we believe that choosing starting temperature by TGA thermograms was not prerequisite and combination of starting temperature and heating rate play an important role to optimize stabilization condition. Acknowledgements This work was supported by a grant from the Korea Institute of Science and Technology (KIST) Institutional program and the Industrial Core Technology Development Program funded by the Ministry of Trade, Industry and Energy, Republic of Korea (No. 10052760). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.jiec.2017.09.048.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048
G Model JIEC 3651 No. of Pages 8
8
H.-S. Kil et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
W. Ruland, Carbon Fibers Adv. Mater. 2 (1990) 528. X. Huang, Materials 2 (2009) 2369. Y. Liu, S. Kumar, Polym. Rev. 52 (2012) 234. R. Eslami Farsani, R. Fazaeli, Int. J. Chem. Mol. Nucl. Mater. Metall. Eng. 6 (2012) 836. J.D. Buckley, D.D. Edie, Carbon–Carbon Materials and Composites, William Andrew, New Jersey, 1993. S.-H. Yoon, Y. Korai, I. Mochida, Carbon 32 (1994) 281. M.S.A. Rahaman, A.F. Ismail, A. Mustafa, Polym. Degrad. Stabil. 92 (2007) 1421. T. Matsumoto, Carbon 31 (1993) 143. R. Didchenko, C.D. Amata, US Patent 3954947 (1976). E. Fitzer, D.J. Muller, Carbon 13 (1975) 63. S.-S. Tzeng, J.-H. Pan, Mater. Chem. Phys. 74 (2002) 214. J.-I. Hayashi, M. Nakashima, K. Kusakabe, S. Morooka, S. Mitsuda, Carbon 33 (1995) 1567. E. Fitzer, W. Frohs, M. Heine, Carbon 24 (1986) 387. S.Y. Jang, S. Ko, Y.P. Jeon, J. Choi, N. Kang, H.C. Kim, H.-I. Joh, S. Lee, J. Ind. Eng. Chem. 45 (2017) 316. J. Drbohlav, W.T.K. Stevenson, Carbon 33 (1995) 693. J. Zhu, S.W. Park, H.-I. Joh, H.C. Kim, S. Lee, Macromol. Res. 23 (2015) 79. V. Ahmad, Y. Zhongren, F. Youqing, FTA Report (2012). P. Morgan, Carbon Fibers and Their Composites, CRC Press, Taylor & Francis Group, Florida, 2005. H.E. Kissinger, J. Res. Natl. Bur. Stand. 57 (1956) 217. S. Vyazovkin, N. Sbirrazzuoli, J. Therm. Anal. Calorim. 72 (2003) 681. K.M. Seven, J.M. Cogen, J.F. Gilchrist, Polym. Eng. Sci. 56 (2016) 541. P. Rantuch, D. Ka9 cíková, B. Nagypál, Eur. J. Environ. Saf. Sci. 2 (2014) 12. J. Ma, X. Meng, X. Guo, Y. Lan, S. Zang, PLoS One 12 (2017) 1.
[24] Y.-S. Cho, M.-J. Shim, S.-W. Kim, Mater. Chem. Phys. 52 (1998) 94. [25] J.W. Park, S.C. Oh, H.P. Lee, H.T. Kim, K.O. Yoo, Korean J. Chem. Eng. 17 (2000) 489. [26] A. Aboulkas, K. El Harfi, A. El Bouadili, Energy Convers. Manage. 51 (2010) 1363. [27] T. Matsumoto, I. Mochida, Carbon 30 (1992) 1041. [28] K. Miura, H. Nakagawa, K. Hashimoto, Carbon 33 (1995) 275. [29] K.E. Robinson, D.D. Edie, Carbon 34 (1996) 13. [30] A.A. Ogale, C. Lin, D.P. Anderson, K.M. Kearns, Carbon 40 (2002) 1309. [31] E. Vilaplana-Ortego, J. Alcañiz-Monge, D. Cazorla-Amorós, A. Linares-Solano, Carbon 41 (2003) 1001. [32] A.H. Wazir, L. Kakakhel, New Carbon Mater. 24 (2009) 83. [33] B.-J. Kim, H. Kil, N. Watanabe, M.-H. Seo, B.-H. Kim, K.S. Yang, O. Kato, J. Miyawaki, I. Mochida, S.-H. Yoon, Curr. Org. Chem. 17 (2013) 1463. [34] B.-J. Kim, Y. Eom, O. Kato, J. Miyawaki, B.C. Kim, I. Mochida, S.-H. Yoon, Carbon 77 (2014) 747. [35] M.-S. Park, Y. Ko, M.-J. Jung, Y.-S. Lee, Carbon Lett. 16 (2015) 121. [36] J. Yang, K. Nakabayashi, J. Miyawaki, S.-H. Yoon, Carbon 106 (2016) 28. [37] J. Kim, U.-S. Im, B. Lee, D.-H. Peck, S.-H. Yoon, D.-H. Jung, Carbon Lett. 19 (2016) 72. [38] G. Yuan, X. Li, X. Xiong, Z. Dong, A. Westwood, B. Li, C. Ye, G. Ma, Z. Cui, Y. Cong, J. Zhang, Y. Li, Carbon 115 (2017) 59. [39] E. Desimoni, G.I. Casella, A. Morone, A.M. Salvi, Surf. Interface Anal. 15 (1990) 627. [40] H. Shono, M. Matsumoto, X. Wang, M. Takeuchi, Y. Sendoda, A. Ishihara, T. Kabe, Sekiyu Gakkaishi 34 (1991) 510. [41] J.P. Boudou, J.I. Paredes, A. Cuesta, A. Martinez-Alonso, J.M.D. Tascon, Carbon 41 (2003) 41. [42] C. Moreno-Castilla, M.V. López-Ramón, F. Carrasco-Marín, Carbon 38 (2000) 1995. [43] T.H. Lim, S.Y. Yeo, Sci. Rep. 7 (2017) 4733.
Please cite this article in press as: H.-S. Kil, et al., Effects of stabilization variables on mechanical properties of isotropic pitch based carbon fibers, J. Ind. Eng. Chem. (2017), https://doi.org/10.1016/j.jiec.2017.09.048