2019-10-31 11:36:23
Dalsu Choi1¶, Hyun-Sig Kil1¶, and Sungho Lee1,2∗
1Carbon Composite Materials Research Center, Institute of Advanced Composite Materials,
Korea Institute of Science and Technology, 92 Chudong-ro, Bongdong-eup, Wanju-gun,
Jeollabuk-do 55324 Republic of Korea
2Department of Nano Material Engineering, Korea University of Science and Technology, 217
Gajeong-ro, Yuseong-gu, Daejeon 34113 Republic of Korea
Carbon fibers (CFs) were invented by Joseph Swan in 1860, who filed a patent for the use of a light bulb filament. Initially, biomaterials such as cotton and bamboo were used as precursors. According to historical records, researchers working in Edison’s institute searched for candidates and found adequate bamboo in a Japanese temple, Iwashimizu Hachimangu. Later, even though Rayon precursors were used for the mass production of CFs, process difficulty and limits of mechanical properties were obstacles to their continued usage. Instead, in 1971, Toray in Japan introduced the fist commercial polyacrylonitrile (PAN)-based CF, T300. By continuous research efforts and manufacturing development, one of those fibers, produced by Toray in Japan and called T1000 grade, shows a maximum tensile strength of 7 GPa. Toray has launched another type of PAN-based CF called M-series fibers, which have a higher tensile modulus of up to 500 GPa. Diverse CFs reinforce composites applied in various industries, such as defense, aerospace, automotive, and sports goods. It is noteworthy that tensile properties of those CFs originate from their microstructures, which differentiate electrical and thermal properties, as well. In 1963, Otani in Gunma University started to use coal tar pitch as a precursor for CFs . Because pitch is a different type of material compared to PAN with regards to chemical structures and physical properties, it is well known that pitch-based CFs exhibit different mechanical properties and microstructures. Currently, several companies provide carbon fibers in commercial market and ~ 10000 ton of CFs are produced yearly. More than 90% of CFs are PAN-based, and the remaining are pitch-based CFs.
There are five steps for CF manufacturing: precursor synthesis, fiber spinning, stabilization, carbonization, and surface treatment [11,12]. PAN and pitch are synthetic polymers, which can be wet and melt spun, respectively, for precursor fibers. Even though PAN is a semicrystalline polymer, the melting point of PAN is not observed in the usual heating conditions, which directs them to wet-spinning after being dissolved in a solvent such as DMSO, DMF, or salt. As the next step, stabilization is a complicated chemical reaction including cyclization, oxidation, and dehydrogenation in the case of PAN to convert from linear to ladder structure. It takes place in an oxygen atmosphere at 200-300 °C for more than 1 h. Pitch also requires oxygen uptake in a similar condition, indicating that oxygen diffuses into PAN and pitch fibers for thermal oxidation. The following step is carbonization, which provides continuous thermal treatment of stabilized fibers up to 1500 °C in an inert atmosphere. It is obvious that, in both thermal treatment steps, stabilization and carbonization, various gases are released depending on the precursors, leading to significant weight loss and dimensional changes, especially diameter reduction because constant winding hinders shrinkage in the longitudinal direction. In fact, surface treatment after fiber spinning and carbonization protects the fiber surface from damage during lots of winding, where friction occurs. In addition, surface treatment after carbonization, called sizing,provides interfacial compatibility to the matrix in composite applications.
Thus far, research efforts on CF production have mainly focused on enhancing mechanical performance . Because the production cost of CFs is extremely high, even though CFs boast exceptional mechanical properties over other structural materials, their use has been strictly limited to exotic applications, such as aircraft, aerospace, and military. However, given recently announced legislation in the USA and EU that restricts the CO2 emission of commercial vehicles, the CF research direction has experienced a huge turnaround. In 2012, the US government introduced a new corporate average fuel economy (CAFE) regulation stating that the average fuel efficiency of commercial vehicles should meet a 35.5-mpg standard by 2017 and then should be higher than 54.5 mpg by 2025, which doubles the 2011 standard of 27.3 mpg. The automotive industry concluded that reducing the weight of vehicles will be the best solution to meet the requirements and gave attention to carbon fiber-reinforced plastics (CFRPs), which are extremely lightweight structural materials with toughness. However, commercial vehicles have lower structural toughness requirements than aircraft and aerospace applications. Rather, commercial vehicles need structural materials with moderate strength but low cost. Therefore, huge demand for CFs with low cost and moderate or low mechanical properties was created by the automotive industry, which in turn translated into huge attention to the development of low-cost CFs. Considering the cost limitation of commercial vehicle production, it is speculated that the cost of CF production should be less than $ 11.00 - $ 15.40 per kg for successful application in the automotive industry. Simultaneously, low-cost CFs should satisfy a tensile strength of 1.72 GPa and a modulus of 172 GPa for use in commercial automobiles.
When the production cost of current PAN-based CFs is analyzed part-by-part, the precursor is responsible for 53% of the total cost [17]. Specialty PAN for CF fabrication is itself pricey, and the only available spinning option for processing PAN is also costly. As a result, the main strategy to reduce the CF production cost has been focused on adopting low-cost precursor other than PAN. In this review, we will cover low-cost precursors for CF fabrication andadvanced processing techniques that might be helpful for realizing low-cost CFs based on low-cost precursors.
2. Low-cost CFs based on low-cost precursors
In this chapter, we will first discuss about the previously reported low-cost precursors for fabricating low-cost CFs. Variety of researches reporting CF fabrication from acrylic precusors
including textile polyacrylonitrile (PAN) and meltable PAN, lignin, polyethylene, and pitch will be covered. A table below is the summary of the advantages and disadvantages of each CF precursors.
2.1. Acrylic fiber-based CFs
2.1.1. Introduction
Polyacrylonitrile (PAN)-based fibers are well known, and only a small fraction of these materials are used as a precursor to carbon fibers. The special-grade PAN fibers are currently the most important precursor material to produce lightweight and high-performance CFs for different industries, such as aerospace, military, turbine blades, construction, medical, automobile, sporting goods, etc. The manufacturing process of PAN-based CFs consists of a low-temperature thermal stabilization in air, followed by a low- and high-temperature carbonization in a nitrogen atmosphere. Among the CF production processes, the precursor is one of the major production cost factors (45~60% of production cost). Therefore, a lower-cost precursor is required to reduce the final-product CF cost. The high costs for typical commercialized special-grade PAN precursor processing initiated the search for low-cost alternatives. One method for achieving this is the use of cheap commercial acrylic fibers (textilegrade PAN fibers) and another is developing melt-spinnable acrylic precursor to bring down the precursor production cost by adopting cost-effective melt-spinning.
2.1.2. Textile-grade PAN fiber
Textile-grade PAN fibers are commonly used in the production of blankets, carpets, and clothes. The product unit cost of textile-grade PAN fibers is low because of the high volume of production. It is speculated that the CF production cost could be reduced by up to 26% by adopting low-cost textile PAN fibers as a CF precursor. The size of their tow (160,000-320,000 monofilaments) is larger than that of specialgrade PAN fibers (3,000-12,000 monofilaments). The textile-grade PAN fibers currently manufactured are composed of at least 85 wt.% acrylonitrile (AN) units. The remaining 15 wt.% consists of neutral and/or ionic comonomers, which are added to improve the properties of the fibers. Neutral comonomers, such as methyl acrylate (MA), vinyl acetate (VA) and methyl methacrylate (MMA), and ionic and acidic comonomers, such as sodium methallyl sulfonat(SMS), sodium 2-methyl-2-acrylamidopropane sulfonate (SAMPS), sodium p-styrene sulfonate (SSS), sodium p-sulfophenyl methallyl ether (SMPE), and itaconic acid (IA), are used to modify the solubility of the acrylic copolymers in spinning solvents, modify the morphology of acrylic fibers, and improve the rate of diffusion of dyes into the acrylic fiber [18-21]. PAN fibers used in the production of CFs are special-grade fibers that are different from textile-grade PAN fibers in terms of chemical composition, type and amount of comonomers, cross-section dimension, linear density and tensile strength. These fibers have larger cross-section areas and linear densities, lower tensile properties, and different types and amounts of comonomers compared to specialgrade PAN fibers. The chemical structure and properties of precursor fibers have critical effects on the properties of carbon fibers. Some of the critical characteristics of precursor fibers that cause enormous variations in the carbon fiber properties are linear density, types and amount of comonomers, molecular weight and its distribution, preferred orientation, surface coating, crimps, size of the tow, and the number of filaments in the tow. Special-grade PAN has higher purity and higher AN content (normally > 90%) for improved properties of precursor fibers and resultant CFs. However, the textile-grade PAN fibers are not readily processable and are not suitable for carbon fiber production due to uncontrollable oxidation or an extremely long time for stabilization, and the thermal behavior of PAN fibers depends on the type and amount of comonomer content. In recent years, some studies have attempted to use textile-grade PAN fibers by using some chemical and mechanical treatments before and after stabilization for production of CFs with suitable mechanical properties. In Shokuhfar and Sedghi et al. [22,23], three commercial textile-grade PAN fibers having different types of chemical composition (sodium 2-methyl-2-acrylamidopropane sulfonate (SAMPS), sodium methallyl sulfonate (SMS) and itaconic acid (IA)) were studied for the manufacture of CFs. They found that the presence of IA comonomer in the PAN fibers causes the production of carbon fibers with high tensile strength, while the presence of SAMPS and SMS comonomers has a negative effect on the properties of CFs because these comonomers cause a delay in the cyclization reaction of nitrile groups and increase the stabilization temperature. Thus, the stabilization process took a very long time, from 300 min to 380 min in the temperature range of 180 ~ 330 °C (Table 2.1.2).
Lee et al. [24] reported that textile-grade PAN fibers as a copolymer of acrylonitrile (AN) and methyl acrylate (MA) (AN/MA=90/10) were 200% to 400% drawn in a hot water bath at
90 °C or/and in a tubular furnace at 180 °C. The drawing process led to higher crystallinity and molecular orientation, contributing to a significant increase in tensile properties. In addition, the cyclization temperature shifted to a lower temperature in the DSC thermogram when fibers wereLee et al. [24] reported that textile-grade PAN fibers as a copolymer of acrylonitrile (AN) and methyl acrylate (MA) (AN/MA=90/10) were 200% to 400% drawn in a hot water bath at 90 °C or/and in a tubular furnace at 180 °C. The drawing process led to higher crystallinity and molecular orientation, contributing to a significant increase in tensile properties. In addition, the cyclization temperature shifted to a lower temperature in the DSC thermogram when fibers were drawn. However, the stabilization process is still long at 390 min at the relatively low temperature of 255 °C. It may be possible to modify some fabrication parameters, such as the stabilization duration and temperature, or use a different stabilization system to improve the properties and reduce the price of carbon fibers made from textile-grade PAN fibers.
In recent years, many attempts have been made to reduce the production cost by reducing the time and energy consumption during the most tedious thermal stabilization process. The
newly introduced stabilization systems, such as radiation and plasma treatment, were shown to be efficient and cost-effective to accelerate the thermal stabilization of textile-grade PAN fibers. Since the early 2000s, Oak Ridge National Laboratory (ORNL) researchers believed textilegrade PAN was a pathway to lower-cost CFs and developed a plasma oxidation process to replace thermal ovens to oxidize or stabilize the precursor fibers. A plasma oxidation system has been demonstrated to be much faster and use significantly less energy while making a fiber with
better qualities. The plasma oxidation process generates a highly reactive gas from the air that reacts with PAN much faster than molecular oxygen, reducing the processing time to less than 30
minutes and using 75% less energy [25-27]. The resultant CFs, with tensile strength and modulus of 1.72-2.96 GPa and 136-235 GPa, respectively [27].
Alternately, as discussed in Yoo et al. [28], an electron-beam was irradiated on textilegrade PAN fibers at various electron doses (200 ~ 1500 kGy) prior to the thermal stabilization to fabricate CFs. Textile-grade PAN fibers have experienced superficial fusion between filaments during the thermal stabilization due to the high contents of co-monomers. By only conducting a conventional thermal stabilization route, single-filament CFs were unable to be fabricated due to the superficial fusion between filaments. The superficial fusion of filaments was prevented and single-filament CFs were able to be fabricated by irradiating an electron-beam prior to thermal stabilization. Furthermore, the total stabilization time was shortened by 64% compared to the conventional stabilization for commercial PAN fibers. Electron-beam irradiation generated chain radicals of PAN molecules, and these radicals recombined with each other to form a crosslink.
As shown in Fig. 2.1.2(b), the ESR spectra of electron-beam irradiated textile-grade PAN fibers exhibited a superposition of several hyperfine structures, verifying the existence of various
radicals formed by irradiation. This crosslinking enhanced the thermal stability of the surface of textile-grade PAN fibers, which hindered the fusion between fibers during thermal stabilization.
As a result, the tensile strength, Young's modulus, and strain-to-failure of the resulting CFs were 1.83 ± 0.23 GPa, 147.44 ± 4.55 GPa, and 1.30 ± 0.15%, respectively. Based on these results,
textile-grade PAN precursor fibers could be an alternative low-cost precursor to reduce the production cost and fabricate CFs with comparable mechanical properties.
2.1.3. Meltable PAN fiber
Another alternative method for reducing the production cost is the use of melt spinning technology in the production of the precursor fibers. As is well known, precursor PAN fibers are
manufactured by wet-, dry- or dry-jet wet spinning because PAN-based polymers go through a thermally induced cyclization reaction below their melting points (~ 300 °C), which cannot be
processed in melt with conventional spinning techniques. The precursor PAN fibers are generally solution-spun (20-30 wt.% solution) in polar solvents, such as dimethylacetamide (DMAc),
dimethylformamide (DMF), dimethylsulfoxide (DMSO) and aqueous sodium thiocyanate solutions [29-31]. However, melt spinning of pure PAN is impossible unless large amounts of solvent and plasticizers are added. Furthermore, this production method is costly due to the relatively low spinning speed in comparison with melt spinning [32-34]. It is in this regard that melt-spinnable PAN precursors seems to be the better alternative for reducing the cost of the final-product CFs.
A melt spinning process has been developed to reduce the processing cost. Some of the literature reported the melt spinning of acrylic fibers by mixing the polymer with additives, such as water [35], ethylene carbonate [36], and propylene carbonate [37]. These additives are used to reduce the interaction between nitrile groups of the polymer chain, which plasticize the polymer and lower its melting point, thereby enabling the melt spinning without significant degradation. The reduction in the melting temperature of PAN attributed to the use of plasticizer is particularly well documented in the literature. Grove et al. [35] have obtained melt-spun PAN fibers using water as a plasticizer to lower the viscosity and the melting point of PAN. The melt-spun PAN fibers were stabilized from several hours at low temperatures to less than 5 min at higher temperatures. Carbonization was achieved by passing the stabilized fibers through two preheaters at 500 °C and 700 °C, followed by a Lindberg furnace at 1200 °C under an inert atmosphere. The morphology of meltspun fibers is similar to that of wet- and dry-spun PAN fibers. However, the obtained CFs exhibited the existence of surface defects and internal voids, and it was concluded that these defects (in Fig. 2.1.3) resulted in the relatively low mechanical properties of 2.5 GPa for the tensile strength and 173–214 GPa for the tensile modulus.
Min et al. [36] researched the melting behavior of plasticized PAN and physical properties of melt-spun acrylic fibers. Melt spinning was performed with various weight ratios of polymer/water/ethylene carbonate (P/W/EC). The combined use of water and ethylene carbonate exhibited a synergistic effect on the plasticization of PAN. Fig. 2.1.4 illustrates the effect of substituting water for EC on the melting and crystallization behaviors of PAN with a fixed plasticizer content of 23 wt.%. The melting and crystallization peaks shifted to a lower temperature, and both peaks were further broadened. The melt-spun fibers showed a density of 1.15 g/cm3, which is compared with the density of most acrylic fibers prepared by solution spinning, 1.18 g/cm3. The difference in densities is an indication of micropores within the meltspun fiber (Fig. 2.1.5).
The morphology of the melt-spun fibers was comparable to that of fibers obtained from wet- and dry-spinning. However, more surface defects and internal microvoids were observed in the plasticized melt-spun PAN fibers, which led to a decrease of the mechanical properties of the CFs. Furthermore, such as when using toxic and expensive plasticizers, it would involve a cumbersome solvent-recovery unit, the process for which would increase the cost of the manufacture of the melt-spun PAN precursor fibers and the CFs. For such a reason as mentioned above, the melt spinning of PAN fibers requires different melt-processable techniques and additionally requires special thermal stabilization because fibers can prematurely melt during stabilization. Some researchers have put effort into developing solvent-free melt-spinnable PAN precursors to reduce CF costs [38-41]. One possible approach is the use of suitable comonomers or termonomers in a particular amount to disturb the crystal structure and nitrile-nitrile
interactions so that a stable melt can be formed at reduced temperatures. Mukundan et al. [38] noticed that copolymers containing 85–90 mol% of AN melt below their degradation temperature and have melt viscosity in the processable range at 220 °C. However, the high content of comonomer may hinder the cyclization of the nitrile groups in the stabilization process, which lowers the properties of precursor fibers and the resultant CFs, with tensile strength and modulus of ~ 730 MPa and ~ 140 GPa, respectively.
Paiva and Naskar et al. [40,41] partially cross-linked and cyclized melt-spun PAN fibers by UV irradiation before subsequent thermal oxidative stabilization and carbonization. They prepared two types of PAN precursor polymers, terpolymer and copolymer, having an acrylonitrile/methyl acrylate/acryloyl benzophenone (AN/MA/ABP) mole ratio of 85/14/1 and an acrylonitrile/methyl acrylate (AN/MA) ratio of 85/15, respectively. The AN/MA/ABP terpolymer precursor (85:14:1 mole ratio) was found to be thermally stable and melt processable at 220 °C. An accelerated UV stabilization route was developed for the precursor, where ABP acts as the UV-sensitive component. However, the AN/MA copolymer precursor could not be stabilized to any significant extent due to the lack of UV sensitivity. The fibers that received thermal oxidation after UV irradiation could be successfully carbonized. Although the CFs exhibited low mechanical properties, there is the potential to substantially improve the properties by optimizing the composition of copolymer and terpolymer, molecular weight or distribution, and melt spinning conditions. The mechanism of UV-assisted stabilization will be discussed in the following chapter entitled ‘Advanced processing for the preparation of CFs.’ As a summary, the spinning methods, processing conditions for CFs, and mechanical properties for some acrylic fiber-based CFs are summarized in Table 2.1.3.
2.2. Lignin-based CFs
2.2.1 Introduction
Lignin is a biopolymer with branched polyphenolic network structures. It is one of three major components of lignocellulose, which is a feedstock for making paper [42-44]. During the papermaking process, lignin is usually separated from cellulose and hemi-cellulose, the other two major components of lignocelluose, because the presence of lignin induces yellowing of the paper and deteriorates the strength of the paper. As a consequence, a tremendous amount of lignin is extracted and wasted as a byproduct of the papermaking process [45]. The chemical structure of lignin includes p-hydroxylphenyl (H-unit), Syringyl (S-unit), and Guaiacyl (G-unit). Those three monomers are linked by a variety of linkages. Therefore, the definite chemical structure of lignin is varied depending on the sources and processing methods [47].
For a long time, lignin had only been considered as waste, and its only application was as a low-value thermal energy source. Recently, numerous research efforts have been focused on reviving lignin, a cellulosic waste, as a low-cost precursor for low-cost carbon fiber fabrication. The cost of melt-spun lignin precursor fiber is estimated as $3.4/kg, which is significantly lower than that of conventional PAN ($8.2/kg) [15,46]. Therefore, when compared to the PAN-based CF production process, it is expected that the CF production cost could be reduced by up to 21% by adopting lignin as a precursor.
Another merit of lignins is that lignins with certain structures have thermoplastic characteristics such that they can be extruded via melt-spinning [48]. In summary, due to their low price and aromatic and thermoplastic characteristics, lignin has been considered as a promising low-cost precursor for carbon fiber fabrication, and numerous research efforts have been conducted regarding the development of high-quality lignin-based carbon fiber [49]. However, the current mechanical properties of lignin-based carbon fibers are inferior when compared to carbon fibers from other precursors [49,50]. Even from the perspective of low-cost carbon fiber targeting low-end markets, the mechanical properties of lignin-based carbon fiber are not satisfactory. Nevertheless, strong advantages in cost and as a pathway of reviving wastes into a value-added product make lignin a robust candidate for low-cost carbon fiber precursor.
2.2.2 Classification of Lignins
Based on the sources, lignins can be classified into three major categories, including hardwood, softwood, and grass. Chemical structures and molecular weights differ based on the source and processing method of lignins. Hardwood and softwood are mostly used to fabricate carbon fiber, while lignins from other sources, such as swithgrass, bamboo, and corn straw, have also been reported. Hardwood lignins mainly consist of G and S unit, while softwood lignins are composed of G-unit. Additionally, softwood and hardwood lignins show a huge difference in how the monomers are connected. Softwood lignins have a larger number of carbon–carbon bonds and lower content of ether bonds than hardwood lignins. As a result, Tg of hardwood lignins and softwood lignins significantly varies, with softwood lignins exhibiting relatively higher Tg of 138 ~ 160 °C, while that of hardwood lignins is 110 ~ 130 °C. The high Tg of softwood lignins makes them unprocessable through melt-spinning, thereby requiring additional steps, such as chemical modification or blending. For hardwood lignins, even though they are generally melt-processable, the ratio between S and G-unit should be carefully tuned for efficient
melt-spinning. Higher S-unit content ensures easier melt-processability but simultaneously might cause the easier fusion of fibers during stabilization.
Extraction methods are the other criteria for categorizing classes of various lignins. Lignins are separated from lignocellulose through a variety of methods. First, the kraft method is the most widely used method to extract lignin from lignocellulose in the papermaking industry, and therefore kraft lignin constitutes the largest volume among all classes of lignins. The kraft process uses an aqueous solution of sodium hydroxide and sodium sulfide mixture. Lignocelluloses, wood biomasses, are treated in the alkaline solution at 140 ~ 180 °C for 2 ~ 4 hrs. Due to the harsh treatment environment, chemical linkages of native kraft lignins are seriously damaged, which in turn provides more phenolic –OH groups through bond breakage when processed for a longer time in alkaline conditions. Next, the extraction method to be introduced is the organosolv process. As the name indicates, the organosolv process is a pulping process that uses organic solvents, such as ethanol and acetic acid. The organosolv process is environmentally friendlier than the traditional kraft process and is also more efficient in separating lignocellulosic biomass into cellulose, hemicelluloses, and lignin. Lignins collected through the organosolv process possess lower molecular weights and Tg values and smaller polydispersities (PDIs). Moreover, when compared to kraft lignins, organosolv lignins have higher phenolic –OH content.
2.2.3 CFs from Lignins
S. Otani et al., for the first time, reported the fabrication of carbon fiber from lignin in 1965 [51]. Hardwood kraft lignin was melt-spun at 170 °C with CO flow to the surface of the extruded fibers. As a result, lignin fibers of 20 ~ 30 µm in diameter could be spun at rates of 15 m/min. Resultant lignin fibers were then thermal oxidatively stabilized at 150 °C for an hour. Subsequently, the stabilized fibers were carbonized under N2 at 700 °C, and carbon fibers with strength reaching 8000 kg-f/cm2 (0.785 GPa) were obtained.
Later, K. Sudo et al. reported a lignin preparation method using a high-pressure steam treatment followed by organic solvent or alkali extraction [52]. Extracted lignins were first hydrocracked to reduce their molecular weight and simplify their chemical structure. Then,hydrocracked lignins were again reassembled into lignins with a proper molecular weight suitable for melt-spinning. Through melt-spinning, lignin precursor fibers with diameters ranging between 10 ~ 40 µm were prepared, and carbonization of those lignin fibers yielded carbon fibers with strengths ranging between 30 ~ 80 kg-f/mm2 (0.29-0.78 GPa). The following study of K. Sudo et al. reported a phenolation method using phenol with p-toluene and sulfonic acid [53]. Such treatment provided melt-spinnable lignins, and the resultant lignin fibers could be carbonized into carbon fibers with a strength of 52.8 kg-f/mm2(0.518 GPa). The same authors also provided other processes, including hydrogenolysis and
phenolysis of steam-exploded lignins, which provided resultant carbon fibers with a strength of 0.45 GPa [54].
In 1995, Y. Uraki et al. introduced an aqueous acetic acid pulping method to extract melt-spinnable lignins [55]. Organosolv lignins went through additional treatments, including heat treatment or fractionation using acetic acid to tune their molecular weight for more efficient melt-spinning. Carbon fibers prepared from such lignin exhibited strengths up to 0.355 GPa and
a modulus of 39.1 GPa. More recently, a novel processing method that can be integrated into the conventional kraft pulping method was introduced in a publication by P. Tomani [56]. The Lignoboost technology allowed for production of lignins suitable for carbon fiber fabrication. Lignoboost technology recovered lignins from black liquor, a lignin-rich byproduct of the kraft process, through precipitation using a carbon dioxide treatment. Recovered lignins had relatively lower ash and sulfur content than traditional kraft lignins. Additionally, the Tg, molecular weight, and polydispersity, which are all important parameters judging processability, of Lignoboost lignins were lower than kraft lignins.Y. Nordstrom et. al. adopted ultrafiltration to tune the molecular weight and polydispersity and interrogated the role of those parameters [57]. Both hardwood and softwood lignins extracted through Lignoboost technology were studied. Changes in processability before and after ultrafiltration were assessed. Ultrafiltered lignins generally showed better processability.
Both filtered hardwood lignins (Tg of 114 °C) and softwood lignins (Tg of 146 °C), could be melt-spun, while melt-spinning of hardwood lignins and softwood lignins without filtration, whose Tg are 139 °C and 150 °C, was not possible. Interestingly, although the Tg of softwood lignins did not differ significantly before or after ultrafiltration, the molecular weight of filteredlignins was only half that of the nonfiltered lignins. Inversely, the Tg of hardwood lignin changed significantly after ultrafiltration, and the molecular weight of permeate also exhibited a drastic reduction. Thus, based on the premise that hardwood lignin has a more linear structure and lower Tg than softwood lignin, the authors expected membrane filtration to further accentuate the structural and thermal differences between softwood and hardwood lignins. As a result, the authors tried mixing ultrafiltered lignins having low Tg with parent lignins as a plasticizer, which could improve the processability.
I. Norberg et al. examined the various thermal oxidative stabilization profiles for ultrafiltered lignins [58]. There were some changes in lignin stabilization chemistry depending on the stabilization profile and sources. The most interesting part was that stabilization could be performed at a high speed with a ramping rate reaching 15 °C/min when ultrafiltered softwood kraft lignin was used. The authors also presented another impressive result that the ultrafiltered softwood kraft lignin could be stabilized without the presence of oxygen. DSC curves of samples showed that ultrafiltered softwood kraft lignin (SKLP) was well stabilized after 15 °C/min stabilization in air and 4 °C/min stabilization in nitrogen. The melting peak observed at ~ 160 °C completely disappeared (Fig. 2.2.3).
However, it was in 2013 when lignin carbon fibers with modest strength reaching 1 GPa were first introduced. Zhang et al. could fabricate high-quality carbon fiber out of softwood kraft lignins whose mechanical properties reached 1.04 GPa for strength and 52 GPa for the modulus [59]. The authors chemically modified softwood kraft lignin through acetylation to dry-spin lignin precursor using acetone. For acetylation, varied concentration of acetic anhydrous was used at an elevated temperature of ~ 85 °C. It could be observed from the FT-IR spectra that the eminent peak at ~ 3500 cm-1 representing the hydroxyl group was diminished as a higher concentration of acetic anhydrous was used (Fig. 2.2.4). Removal of the hydroxyl group enabled easier extrusion. However, excessive acetylation removed too many hydroxyl groups and inhibited proper stabilization. Thus, acetylation of hydroxyl groups had to be executed in a controlled manner, and lignin fibers could be successfully stabilized without deformation.
2.2.4 Carbon Fibers from a Lignin/Polymer mixture
An earlier example of a lignin/polymer mixture system for carbon fiber fabrication is a lignin/polyehtlyne oxide (PEO) system. Strong hydrogen bonds are generated when lignin and PEO are mixed together, which indirectly weakens the van der Waals force among lignin molecules. As a result, a lignin/PEO blend exhibits reduced Tg as the PEO fraction is increased. In 2001, J. Kadla et al., for the first time, reported a method using PEO as a plasticizer to reduce the Tg of hardwood kraft lignins [60]. Blending PEO allowed for spinning of a fine lignin fiber. From 5 wt.% to 25 wt.%, blends with varied mixing ratios were prepared. Upon blending, the temperature for spinning continuous fiber significantly dropped from 195 °C to 150 °C when 25 wt.% of PEO was mixed. However, a high PEO ratio inhibited the proper stabilization reaction, so the authors had to optimize the blending ratio. Through optimization, the authors found that less than 5 wt.% was adequate for spinning continuous fiber and for successful stabilization. Precursor fibers with 3 wt.% PEO exhibited optimal mechanical properties with a strength of 0.55 GP and modulus of 60 GPa. S. Kubo et al. also published studies about lignin blends with polypropylene (PP).
However, the authors could not overcome the miscibility issue of PP-lignin blends, and the resultant carbon fibers exhibited hollow or porous morphologies [61]. In the same study, the authors also blended polyethylene terephthalate (PET) with lignins and examined the properties of the resultant carbon fibers. PET/lignin blends exhibited better behavior than PP and showed sound mechanical properties, reaching 0.7 GPa for strength and 94 GPa for the modulus (Table 2.2.1).
More recent studies also tried blending lignins with PAN to obtain carbon fibers with better mechanical properties [62–65]. However, most of the studies used higher PAN content,exceeding the fraction occupied by lignins. Therefore, we categorize those studies as PAN-based carbon fiber studies with lignin additives for cost reduction. As a summary, we present a table displaying the mechanical properties of previously reported lignin-based CFs with brief processing remarks.
2.3. Polyethylene-based CFs
2.3.1 Introduction
As described in the previous chapter, because the precursor constitutes approximately 50% of the entire production cost, successful adoption of a low-cost precursor for carbon fiber fabrication might significantly reduce the total production cost. Among several candidates, details about PE (polyethylene) as a carbon fiber precursor will be discussed in this chapter.
PE is the most widely used polymer in the world. Based on the frequency and shape of the side branches attached to the main carbon single-bond backbone, PE is generally classified as LDPE (low-density polyethylene), LLDPE (linear low-density polyethylene), and HDPE (highdensity polyethylene). When the production of all the analogues, including LDPE, LLDPE, and HDPE, are combined, the production volume reaches 29.3% of the worldwide plastic production [66,67].
Given that it is a mass-production polymer, PE has definite advantages in terms of price and supply over conventional specialty polymeric precursors for carbon fiber fabrication. On top of that, PE can be shaped into a long fiber through a melt-spinning process, while most of the other precursors require pricey solution-spinning processes using large amounts of organic solvent. Cheap operation cost is not the only advantage of the melt-spinning process. Fibers can be produced in a much faster manner when compared to wet-spinning, which lowers the cost per unit mass of fiber production. As a result, the price of PE fibers per pound is only 1~2 dollars, while that of the carbon fiber-grade PAN fibers is 4~5 dollars per pound [16,17,68].
However, stabilizing PE polymer chains is relatively harder when compared to conventional precursors using simple heat treatment. Currently, only a limited number of studies regarding PE-based carbon fiber fabrication are available, and the only successful stabilization method reported so far is a sulfuric acid treatment [69-75]. A number of studies reported that treatment of PE fiber in sulfuric acid with free SO3, more specifically sulfuric acid with elevated temperature or fuming sulfuric acid, aromatizes the chemical structure, and carbon fibers with moderate mechanical properties could be fabricated by carbonizing sulfonated PE fibers.
Nevertheless, even considering the tedious sulfuric acid treatment, carbon fiber fabrication from PE has a great advantage in total production cost. The expected cost savings in other processes far exceed the cost added by the adoption of the sulfuric acid treatment, and a cost reduction is expected in total. Therefore, PE has great potential to be a great low-cost alternative to conventional carbon fiber. Here, we will first introduce a detailed mechanism for PE stabilization through sulfonation, which is a critical difference from other carbon fiber precursors. Then, exemplary works demonstrating successful carbon fiber fabrication from PE fiber will be reviewed.
2.3.2 Stabilization mechanism of PE through sulfonation
The first demonstration of PE-based carbon fiber fabrication was reported in US patent US4070446A by S. Horikiri et al., and the first academic study with more details was provided by A. Postema et al. in 1990 [69,76]. However, those pioneering studies did not provide details about the chemical structural transformation of PE fibers into a stabilized structure. Even though there have been a tremendous number of studies regarding sulfonation of PE, all of them have dealt with low-temperature sulfonation scenarios, which cannot be directly applied to interpret the high-temperature sulfonation process used for carbon fiber fabrication. It was in 2013 that J. Younker et al. published the first study discussing details of the chemical structural changes during high-temperature sulfonation of PE [77]. Later, in 2015, a more complete picture was presented by B. Barton et al. [75]. J. Younker et al. selected n-heptane-4-sulfonic acid (H4S) as a model material representing the sulfonated repeating unit of PE and studied two different probable mechanisms of reaction site formation during PE sulfonation at elevated temperature [77].
The first mechanism was five centered internal elimination (Ei5), which follows classic sulfoxide/sulfone chemistry, including hydrogen abstraction by internal attack by one of the
nucleophilic oxygens in the SO3H functional group attached to the main PE backbone. The second mechanism was radical chain reaction involving two interconnected reactions. Two possible pathways of radical initiation were proposed. One was through hydrogen abstraction via the SO3H radical homolytically detachment from the main chain. Another pathway the authors provided was decomposition of the SO3H radical, providing an OH radical that in turn carries radicals and abstract hydrogen. Through density functional theory (DFT) and transition state theory (TST), the authors determined the rate constants of two different mechanisms of reaction site formation through sulfonation. By comparing experimental TGA results and simulated TGA curves constructed through a kinetic Monte Carlo (kMC) method based on DFT/TST rate constants, the authors concluded that the radical chain reaction is the more probable mechanism for the creation of reaction sites through high-temperature sulfonation.
Unlike the study presented by J. Younker et al., which was more focused on defining the exact pathway of how the SO3H functional group detaches and provides reaction sites, B. Barton et al. provided a broader picture of the stabilization of PE through sulfonation [75]. The authors first checked the FT-IR spectra of sulfonated PE samples and specified the attachment of SO3H, ketone, and hydroxyl functional groups. Then, the gas evolution of sulfonated PE samples during the carbonization process was interrogated using evolved gas analysis-gas chromatography (EGA-GC). The evolution of a substantial amount of SO2 gas was detected at temperatures between 120 and 220 °C, giving a clue regarding the detachment of SO3H. FT-IR measurement of a sulfonated PE sample that was heat treated at 120 °C in sulfuric acid concomitantly indicated the loss of S=O vibrations in conjunction with decreased intensity of alkyl vibration. Such results represent the crosslinking and stabilization of PE chains via SO3H functional group detachment, as described in the previous study of J. Younker et al. Then, the gas evolution behavior in the higher temperature region was explored. Various gases, including H2, CO, CO2, and H2O, were evolved upon heat treatment in higher temperature, which resembled the dehydrogenation step in the stabilized PAN carbonization process, and the carbon structure was finally organized.
Fig. 2.3.3. A proposed mechanism describing the transformation of polyethylene through the sulfonation process. The process starts with extensive dehydrogenation of hydrocarbon, form of H2HO3 and further crosslink the chains. Above 600 °C, the oxidation of any remaining C–H bonds occurs thermally with the evolution of various gases, and transformation into the carbon structure is completed [75]. Adapted with permission from Elsevier. Copyright (2015) Elsevier.
2.3.3 Fabrication of low-cost carbon fiber from PE In this chapter, demonstration of carbon fiber production from PE precursor will be reviewed. The first work presenting PE-based carbon fibers was by A. Postema et al. in 1990 [69]. The authors used melt-spun LLDPE fiber. For stabilization, chlorosulfuric acid was used, and sulfonation time was varied all the way up to 40 hrs. One critical condition to note is that the sulfonation process was performed at room temperature. Upon sulfonation, the mass of LLDPE fiber increased up to 150% of its original mass, and the length of fiber shrunk approximately 60% as LLDPE fiber was crosslinked through adequate sulfonation. Then, the LLDPE fiber sulfonated for the designated time was pyrolyzed under N2 at 900 °C, and the maximum temperature was held for 5 min. When the sulfonation time was less than 8 hrs, stabilization was not sufficient, and carbon fiber was not formed. Interestingly, unsuccessful carbon fiber fabrication took place not only in the case of relatively short sulfonation time but also when the sulfonation time exceeded 32 hrs. The authors rationalized the result by referencing the overoxidationscenario in PAN-based carbon fiber production. After spotting the processing window for adequate sulfonation time, the authors optimized the stress applied to the fiber during the carbonization process. As well studied in PAN-based carbon fiber studies, stress application is a crucial step in guiding anisotropic carbon structure formation and is closely related to the mechanical properties of the carbon fiber. Optimal mechanical properties, with a tensile strength of 1.15 GPa and modulus of 60 GPa, were obtained when a stress of approximately 1.3 MPa was applied.
One interesting part was that the thickness of the final carbon fiber was 40 µm, which is relatively thick. In their later study, through improved spinning technology and a hot stretching method, they could produce a thinner precursor LLDPE fiber. Thinner precursor LLDPE fiber could be successfully organize carbon fiber with shorter sulfonation time (4 hrs), and the mechanical properties could also be significantly improved (tensile strength of 2.16 GPa and modulus of 130 GPa) [70].
A later study by D. Zhang et al. in 1996 introduced an improved method of producing PE-based carbon fiber [72]. They sulfonated PE precursor fiber at the elevated temperature and could significantly shorten the processing time for stabilization. In this case, concentrated sulfuric acid (95%) was first heated up to 130 °C, and the precursor fiber emerged. Then, the temperature of the sulfuric acid bath was raised to 180 °C for 75 min. The stabilized fiber was carbonized at 1100 °C. The resultant carbon fiber exhibited very good mechanical properties, reaching a tensile strength of 2.0 GPa and modulus of 200 GPa.
After the report of D. Zhang et al., it took almost 15 years for successive works to come out. In 2013, a group of researchers from Oak Ridge National Labs (ORNL) reported a study demonstrating fabrication of patterned carbon fibers from PE/Fugitive polymer bicomponent extrusion [73]. The study was mostly focused on fabricating carbon fiber with a patterned shape, but fabrication of plain LLDPE carbon fiber was briefly mentioned in the supporting information.
In this study, they used very reactive fuming sulfuric acid, which contains extra SO3. Fuming sulfuric acid was heated at 70 °C, and sulfonation took an hour to complete the stabilization.
After sulfonation, they carbonized the treated fiber at 1200 °C and could fabricate carbon fibers with a strength of 1.1 GPa and modulus of 100 GPa.
Then, in 2015, J. Kim et al. published a report regarding the fabrication of carbon fiber from gel-spun LLDPE fiber precursor [74]. In this work, they performed a comprehensive study
of how the stress during both sulfonation and carbonization influences the mechanical properties of PE-based carbon fiber. When 0.25 MPa and 0.26 MPa of stress were applied during the
sulfonation and carbonization processes, respectively, the authors could produce carbon fibers with tensile strength of 1.65 GPa and modulus of 110 GPa.
However, the most impressive works presented recently are the series of studies by B. Barton et al. [75,78]. They not only provided a detailed interrogation of the chemical structural evolution of PE during the sulfonation process, which was reviewed in the previous section, but also revealed the distinct nature of PE-derived carbon fiber by adopting a structure-property model relating tensile strength to angular orientation of microstructures, which was originally suggested by M. Northort et al. [79] Through careful interrogation of a single filament using synchrotron WAXD (wide angle X-ray diffraction), the authors could obtain information regarding the angular orientation of the graphitic microstructures constituting the carbon fiber [78]. With the obtained information, they could assess the modulus for the shear between neighboring graphitic microstructures and could find that the value was significantly lower when compared to PAN- or pitch-based carbon fibers. The result suggested that interlayer crosslinking via C‒C sp3 residues was weaker, and the authors concluded that such weak crosslinking might allow easier movement for better orientation during a high-temperature graphitization process.
Moreover, the same group even published a paper sharing their multiphase reactor design for PE-based carbon fiber fabrication, which can handle a large-tow (1000-6000 filaments)
precursor bundle [80]. The multiphase reactor was constructed with a total of four batches – three sulfonation batches and one deionized water batch for washing. The first three sulfonation batches contained fuming sulfuric acid (120% SO3) at 50 °C, sulfuric acid at 120 °C, and sulfuric acid at 140 °C, respectively. Additionally, a fiber guide roller coupled with winding motors allowed for subtle adjustment of the applied stress for each procedure. They used the exact setup for the fabrication of PE-based carbon fibers with tensile strength of 1.5 GPa and modulus of 90 GPa , which that was exhibited in their previous paper [81].
Finally, we are going to cover some recent studies that introduced novel PE stabilization methods other than sulfonation. Though the following methods could not be adopted to fabricate carbon fibers, noncarbonizable PE could be successfully transformed into carbonaceous materials. Thus, it is expected that those new PE stabilization strategies have a high potential as alternative routes for PE carbon fiber fabrication if further modifications and improvements are made in the future. The first study for review is about thermal oxidative stabilization of LLDPE [82]. D. Choi et al. explored the thermal oxidation behavior of LLDPE in the high temperature region reaching 350 °C where oxidative decomposition takes place. Some exothermic reactions other than melting were found in the upper region (200 ~ 330 °C) of the survey in the differential thermal analysis (DTA)/thermogravimetric analysis (TGA). The authors could confirm that those exothermic reactions were related to aromatization and oxygen functionalization of LLDPE using Raman, FT-IR, and XRD. For thermal stabilization through high-temperature oxidation, LLDPE samples were heated in a convection oven up to specific temperatures and removed as soon as the temperature reached the target value. As a result, LLDPE could be successfully transformed into a carbon with a high yield reaching ~ 45%. However, as the stabilization
temperature is far above the melting point of PE, the introduced technology could not be used for PE fiber stabilization.
Another interesting study by B. Barton et al. introduced an ammoxidation method for PE stabilization [83]. The authors focused on maximizing the crosslinking density of precursor fiber prior to the actual stabilization process to maintain fiber conformation. Thus, they prepared PE grafted with vinyltrimethoxysilane (PE-g-VTMS). PE-g-VTMS fibers were then crosslinked by submerging them in a solution of 5 wt.% NACURE B201 in isopropanol. After 15 hrs of immersion, the dried fiber was crosslinked in a moisture oven at 60 °C for 24 hrs. The gel content of the processed fiber was ~ 70%, confirming that PE chains were successfully crosslinked. Crosslinked PE fibers then went through ammoxidation. For ammoxidation, 200 ml of 5 M ammonium hydroxide (NH4OH) was prepared, and the air was flowed through a sparging bubbler at a rate of 60 ml/min. Treatment was performed at room temperature for 20 hrs. The carbon yield of crosslinked PE fibers after ammoxidation was ~54% at 1000 °C. Though they
have a high carbon yield, ammoxidized PE fibers could not avoid fiber fusion during the carbonization process.
In summary, we prepared a table including stabilization conditions and mechanical properties of the reviewed PE-based CFs.
As is well known, oxidation/stabilization is an important step in CF production because this process determines the mechanical properties after the high-temperature heat-treatment step referred to as carbonization. The as-spun pitch fiber is stabilized by heating it to 250 ~350 °C in air for a duration ranging from 30 min to several hours. During the stabilization process, the asspun pitch fibers crosslink among the precursor molecular to become an infusible thermosetting fiber. A thermosetting reaction that is too fast causes the melting of the fiber surface, and the fibers stick to each other. To avoid this phenomenon, the stabilization process must be optimized, and both overlong and insufficient stabilization will be detrimental.
The subsequent steps, carbonization and graphitization, are removal steps of the introduced oxygen containing groups and hydrogen bonding formed during the stabilization step. To avoid the defects
created by the excessive release of volatiles, it is preferred for the stabilized fibers to be precarbonized at 700 ~ 900 °C under a nitrogen atmosphere for a brief period to remove volatiles. The CFs are subsequently carbonized at 1500 ~ 1800 °C to obtain the required mechanical properties of CFs. To prepare mesophase pitch-based carbon fibers with high modulus, a graphitization step up to 3000 °C is even applied. To reduce the cost of CF manufacturing, changes will be considered in all steps, such as modification of the precursor pitch, improvement of spinning, and reduction of the stabilization temperature and duration.
2.4.2. Mesophase pitch-based carbon fibers
Pitch-based carbon fibers are commonly classified into mesophase pitch-based carbon fibers (MPCFs) and isotropic pitch-based carbon fibers (IPCFs) according to the type of precursor pitches based on their optical characteristics and into high-performance carbon fibers (HPCFs) and general-purpose carbon fibers (GPCFs) according to their mechanical properties [9,88]. As shown Fig. 2.4.2, HPCFs are produced from mesophase pitches, which have a high degree of molecular orientation [93]. As-spun mesophase pitch fibers, usually composed of large planar-shaped polycondensed aromatic compounds having high stacking units, can easily develop the graphitic structure during the carbonization and graphitization process.
The mesophase pitch can be formed either by thermal polymerization of petroleum- or coal tarderived pitches or catalytic polymerization of pure compounds, such as naphthalene. However, the spinnable 100% mesophase pitch with a relatively narrow molecular weight distribution could be prepared from special and complicated methods, such as nondehydrogenative polymerization of aromatic hydrocarbons with aluminum chloride as a catalyst[89], aromatic hydrocarbons using HF–BF3 as the condensation catalyst [90], supercritical fluid extraction of petroleum pitch [91], and a two-stage production process where it is under pressure at higher temperature in the first stage and in a vacuum in the second stage under vacuum [92]. In addition, for melt-spinning of mesophase pitch, both the molecular weight and the mesophase content must be optimized for melt-spinning conditions. Therefore, pitch purification can add significant costs to the production of mesophase pitch. These processes require a high manufacturing cost, which is the limitation for mass production and cost reduction of HPCFs. Due to the comparatively high production cost of MPCFs, their applications are mainly limited to
certain areas, such as the aerospace and sporting goods industries. Therefore, isotropic pitch-based CFs are attractive due to their reasonable manufacturing cost even though moderate properties
are expected.
2.4.3 Isotropic pitch-based carbon fibers
Isotropic pitch-based carbon fibers(IPCFs) for general-purpose carbon fibers (GPCFs) are normally produced from coal tar or petroleum pitch, which has relatively low elastic modulus and tensile strength because of the low-ordered stacking of aromatic molecular planes. IPCFs as carbon fiber-reinforced composites were applied to automobiles for parts of the automotive body, chassis, suspension systems, etc. and construction as components of reinforced concrete or heatresistive insulation. In the meantime, the research and development of IPCFs have been centered on preparation of spinnable pitches and optimization of the stabilization/carbonization process for cost reduction and the enhancement of mechanical properties. However, its tensile strength is still insufficient.
There are several prior research studies on modifying an isotropic pitch precursor to increase the softening point by thermal treatment, chemical additives, and catalytic reactions for good spinnability and high strength and to optimize the stabilization process for development of a cost-effective process. Kim et al.[94] have developed a novel isotropic pitch composed of linear chains of aromatic compounds from naphtha cracked oil (NCO) through a brominationdehydrobromination reaction (NB). The aromatic components of NB were primarily condensed cyclic compounds containing three and four aromatic
rings (Fig. 2.4.3(a)). The as-spun pitch fibers were stabilized at 270 °C for 1 hr with a heating rate of 0.5 °C/min and then carbonized at 800 °C for 5 min with a heating rate of 5°C/min.
Carbon fibers derived from this isotropic pitch exhibited extraordinarily high tensile strength of 1500 MPa and elongation of 3.2%. In contrast, a pitch prepared by simple distillation (ND) was composed of compounds containing three to six aromatic rings (Fig. 2.4.3(b)), which carried the tensile strength of carbon fiber by only 700 MPa and elongation of 2.5% with a similar fiber diameter. The molecular weight of the NB pitch (linear molecules) was greater than that of the ND pitch (nonlinear or kinked compounds). The NB pitch exhibited a higher degree of shear-thinning than did the ND pitch, which suggests a
greater degree of molecular orientation in the NB pitch during spinning (Fig. 2.4.4).
Later, Kim et al. [95] further studied the preparation of IPCFs from the co-carbonization of NCO/CT (naphtha-cracked oil/coal tar). IPCFs obtained from the co-carbonization of NCO/CT mixture via bromination-dehydrobromination exhibited unprecedentedly high tensile strengths and elongations at breaks of 1800 and 2050 MPa, and 3.2 and 2.8% following carbonization at 800 and 1100 °C for 5 min, respectively. The aromatic components derived from the co-carbonization process were primarily condensed cyclic compounds containing four or five aromatic rings (Fig. 2.4.3(c)). In addition, the molecular weight of NCB was much greater than that of NB. This indicates that the linear isotropic pitch molecules and the molecular orientation and molecular weight of precursor fibers play an important role in enhancing the mechanical properties of carbon fibers derived from isotropic pitches.
For the production process approach,
Jang and Kil et al. [96,97] reported that stabilized isotropic pitch fiber with a density range of 1.35-1.36 g/cm3 exhibited the highest tensile strength regardless of the stabilization conditions(Fig. 2.4.5(a) and (b)). Based on the significant correlation between density and tensile properties, they reported a systematic study on optimizing the stabilization process by varying heating conditions, such as the heating speed and the starting temperature. The densities of as spun fiber and stabilized pitch fibers that were heattreated up to a temperature of approximately 170 °C were 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 of holding (Fig. 2.4.6(a)). 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 the unnecessary region below 170 °C can shorten the stabilization process. As a result, the total stabilization time has been significantly reduced down to 56.5 min with comparable mechanical properties of the conventional
isotropic pitch-based CFs (Fig. 2.4.6(b)).
2.4.4 Liquefied wood- and biotar-derived pitch-based carbon fiber
The CFs with mechanical properties suitable for GPCFs can be produced from new precursor pitches, such as liquefied wood (LW) and biotar. In some of the research studies, cellulose and lignin have been used as precursors for carbon fibers, but utilization of wood is still very low due to the difficulty of wood component separation. In recent years, various techniques have been developed for the use of wood. Among them, wood liquefaction has been regarded as one of the effective techniques for converting wood into useful liquid materials. Through liquefaction, wood components are brought to low-molecular-weight compounds with high reactivity, and liquefied wood has been widely used as a raw material [98-100]. Ma et al.[99,100] have tried to prepare carbon fibers using liquefied wood (LWCFs) as a new precursor. The fibers from liquefied wood in phenol (WPFs) were prepared by the following procedure (Fig. 2.4.7(a)). Wood flour was mixed with phenol containing 8 wt.% H3PO4 as reaction catalyst. The hexamethylenetetramine (HMTA) as synthesis agent was added to liquefied wood to prepare the spinning solution. After melt-spinning using a homemade spinning apparatus (Fig. 2.4.7(b)), the
spun filaments were cured by soaking in a hydrochloric acid (HCl) and formaldehyde (HCHO) solution.
As seen from Fig. 2.4.8, the surfaces of WPFs and LWCFs were smooth without micropores, attachment or groove structures. WPFs with an average diameter of 27 ~ 42 µm, tensile strength of 230 ~ 356 MPa, and modulus of 15 ~ 31 GPa were obtained. WPFs were carbonized at 1100 °C for 2 hrs with a heating rate of 4 °C/min under a nitrogen atmosphere. The tensile strength, modulus
and yield of LWCFs are 1.22–1.7 GPs, 114–176 GPa and 55.8%, respectively. Those fibers not only improved the “value-added” of products from LW but also reduced the capital costs of carbon
fibers due to partial displacement of fossil resources during preparation. Biotar, which is a typical byproduct of the charcoal manufacturing industry, is usually produced during the pyrolysis
of biomass via a series of complex reactions. Unlike coal tar, biotar has been studied very little thus far as another type of pitch precursor for carbon fiber production. Development of applications for biopitch is important to increase the revenue of the charcoal manufacturing industry and to stimulate the use of biomass, thereby answering the appeals of environment preservation and the manufacture of low-cost carbon fibers. Therefore, to obtain the highly thermally stable and melt-spinnable biopitch, Prauchner et al. [101-103] presented biopitch as a potential CF precursor. The precursor biopitch was obtained by the pyrolysis and condensation of eucalyptus wood. The as-spun eucalyptus tar pitch fibers were stabilized by an oxidative thermal treatment. After stabilization, the fibers were carbonized at 1000 °C for 1 hr with a heating rate of 2 °C/min under a nitrogen atmosphere in a tubular furnace. The carbon fibers produced had an average diameter of 27 µm and an apparent density of 1.84 g/cm3. SEM micrographs (Fig. 2.4.9) show that the fibers present a smooth and homogeneous surface. Unfortunately, the mechanical properties of the resulting CFs were not satisfactory, with tensile strengths of 130 MPa and a tensile modulus of 14 GPa being found.
Qiao et al. [104,105] have reported two types of biomass tar (wood tar and bamboo tar)-based resin through a chemical modification. Such biomass resin is expected to be used as a new carbon precursor for carbon materials. Wood tar was collected during pyrolysis of mixed woods from beech, oak, and maple trees, and bamboo tar was collected during the pyrolysis of phyllostachys pubescens. Wood and bamboo tar was polymerized with a formaldehyde solution in a flask with stirring by using oxalic acid and HCl as catalysts (two-stage polymerization), respectively, to synthesize resin. Bamboo tar resin was obtained by removing water under a vacuum. Melt-spun biotar resin fibers were stabilized and carbonized. The mechanical properties of carbon fibers (maximum strength and modulus of 632 MPa and 44 GPa, respectively) are found to be comparable to and competitive with those of commercially available carbon fibers.
In the present studies for low-cost CFs made from biotar, Yang et al. [106] reported on the pitches obtained by low-temperature co-carbonization of biotar (BT) and ethylene bottom oil (EO) mixtures. In addition, they applied a developed method for preparation of spinnable pitch by a bromination-dehydrobromination (BD) method and compared with the conventional atmospheric distillation (AD) method. The CFs obtained from the BD-derived spinnable pitch had smoother and more homogenous surfaces than the AD-derived spinnable pitch, which showed some white spots on thesurface in Fig. 2.4.10(a and b). The BD method was more effective in improving the spinnability of BT-derived pitch and showed a higher tensile strength of BT-derived pitch-based CF compared with that prepared using the AD method.
The fabricated CFs using the obtained biotar-derived isotropic pitch precursor showed an enormous tensile strength of 1200 MPa by only carbonization at 800 °C for 5 min (Fig. 2.4.10(e)). As a summary,
we present a table with the mechanical properties of previously reported pitch precursor-based CFs and their stabilization and carbonization conditions.
1-Methyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyridin-2(1H)-oneCatalog No.:AA00017W CAS No.:1002309-52-5 MDL No.:MFCD11044683 MF:C12H18BNO3 MW:235.0872 |
5-(Trifluoromethyl)indoleCatalog No.:AA0002X0 CAS No.:100846-24-0 MDL No.:MFCD03095341 MF:C9H6F3N MW:185.1458 |
4-Amino-2,6-dichloropyrimidineCatalog No.:AA00045E CAS No.:10132-07-7 MDL No.:MFCD00038015 MF:C4H3Cl2N3 MW:163.9927 |
1-Cyclopropyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl);-1H-pyrazoleCatalog No.:AA000GB5 CAS No.:1151802-22-0 MDL No.:MFCD16659007 MF:C12H19BN2O2 MW:234.1025 |
4-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)pyridin-2(1H)-oneCatalog No.:AA000O3L CAS No.:1256358-90-3 MDL No.:MFCD11223287 MF:C11H16BNO3 MW:221.0606 |
Platinum(IV) oxideCatalog No.:AA000W0M CAS No.:1314-15-4 MDL No.:MFCD00066964 MF:O2Pt MW:227.0768 |
6-(Trifluoromethyl)picolinic acidCatalog No.:AA000ZKI CAS No.:131747-42-7 MDL No.:MFCD02180819 MF:C7H4F3NO2 MW:191.1074 |
3-QuinolinecarboxaldehydeCatalog No.:AA00125C CAS No.:13669-42-6 MDL No.:MFCD00006768 MF:C10H7NO MW:157.1687 |
1-Fluorocyclopropanecarboxylic AcidCatalog No.:AA0012QP CAS No.:137081-41-5 MDL No.:MFCD18205954 MF:C4H5FO2 MW:104.0797 |
5-Acetyl-2-bromopyridineCatalog No.:AA001A05 CAS No.:139042-59-4 MDL No.:MFCD04974527 MF:C7H6BrNO MW:200.0326 |
2-bromo-5-cyanopyridineCatalog No.:AA001B8J CAS No.:139585-70-9 MDL No.:MFCD00234141 MF:C6H3BrN2 MW:183.0054 |
4-Bromo-2,3-difluorophenolCatalog No.:AA001J5S CAS No.:144292-32-0 MDL No.:MFCD08061907 MF:C6H3BrF2O MW:208.9882 |
2-ChloroquinoxalineCatalog No.:AA001K4P CAS No.:1448-87-9 MDL No.:MFCD00043907 MF:C8H5ClN2 MW:164.5917 |
3-Methoxyazetidine hydrochlorideCatalog No.:AA001L0A CAS No.:148644-09-1 MDL No.:MFCD06804514 MF:C4H10ClNO MW:123.5813 |
4-MethoxyphenolCatalog No.:AA001M5C CAS No.:150-76-5 MDL No.:MFCD00002332 MF:C7H8O2 MW:124.1372 |
1-Methyl-1H-pyrazole-5-carboxylic acidCatalog No.:AA001RS3 CAS No.:16034-46-1 MDL No.:MFCD00464253 MF:C5H6N2O2 MW:126.1133 |
Tetrabutylammonium bromideCatalog No.:AA001UP1 CAS No.:1643-19-2 MDL No.:MFCD00011633 MF:C16H36BrN MW:322.3677 |
D-Galactosamine hydrochlorideCatalog No.:AA00253E CAS No.:1772-03-8 MDL No.:MFCD00135830 MF:C6H14ClNO5 MW:215.6321 |
2-Methyl-4,6-dichloropyrimidineCatalog No.:AA0025N8 CAS No.:1780-26-3 MDL No.:MFCD00090472 MF:C5H4Cl2N2 MW:163.0047 |
4-N-Boc-AminocyclohexanoneCatalog No.:AA0026UR CAS No.:179321-49-4 MDL No.:MFCD00798168 MF:C11H19NO3 MW:213.2735 |
3-(Trifluoromethyl)pyrazoleCatalog No.:AA002DEO CAS No.:20154-03-4 MDL No.:MFCD00115018 MF:C4H3F3N2 MW:136.0752 |
5-Bromo-2-chlorobenzaldehydeCatalog No.:AA002DQP CAS No.:189628-37-3 MDL No.:MFCD08445659 MF:C7H4BrClO MW:219.4631 |
1-Methyl-1H-indole-3-carbaldehydeCatalog No.:AA002E97 CAS No.:19012-03-4 MDL No.:MFCD00014570 MF:C10H9NO MW:159.1846 |
3-Cyanobenzeneacetic acidCatalog No.:AA002GF3 CAS No.:1878-71-3 MDL No.:MFCD06411079 MF:C9H7NO2 MW:161.1574 |
4-Chloro-2-pyridinecarboxylic acid methyl esterCatalog No.:AA002O7T CAS No.:24484-93-3 MDL No.:MFCD04116183 MF:C7H6ClNO2 MW:171.5810 |
2,5-DifluorobenzaldehydeCatalog No.:AA002T91 CAS No.:2646-90-4 MDL No.:MFCD00010327 MF:C7H4F2O MW:142.1029 |
4-Cyanophenylhydrazine hydrochlorideCatalog No.:AA002VTU CAS No.:2863-98-1 MDL No.:MFCD00673994 MF:C7H8ClN3 MW:169.6115 |
(S)-1-(Tert-Butoxycarbonyl)-3-HydroxypyrrolidineCatalog No.:AA0032DH CAS No.:101469-92-5 MDL No.:MFCD01317839 MF:C9H17NO3 MW:187.2362 |
2-(2-Pyridyl)ethylamineCatalog No.:AA0032RW CAS No.:2706-56-1 MDL No.:MFCD00006367 MF:C7H10N2 MW:122.1677 |
4-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)isoxazoleCatalog No.:AA0033Q5 CAS No.:928664-98-6 MDL No.:MFCD06657891 MF:C9H14BNO3 MW:195.0234 |
4,4-DifluorocyclohexanolCatalog No.:AA0033S1 CAS No.:22419-35-8 MDL No.:MFCD10000566 MF:C6H10F2O MW:136.1398 |
4-Bromo-2-methylbenzaldehydeCatalog No.:AA0033UU CAS No.:24078-12-4 MDL No.:MFCD07787171 MF:C8H7BrO MW:199.0446 |
4-Chloro-3-fluoropyridineCatalog No.:AA0033WP CAS No.:2546-56-7 MDL No.:MFCD03453233 MF:C5H3ClFN MW:131.5354 |
5-BromopicolinonitrileCatalog No.:AA00345P CAS No.:97483-77-7 MDL No.:MFCD00234144 MF:C6H3BrN2 MW:183.0054 |
Benzo[b]thiopheneCatalog No.:AA0034F2 CAS No.:95-15-8 MDL No.:MFCD00005864 MF:C8H6S MW:134.1982 |
Methyl 3-fluoro-4-nitrobenzoateCatalog No.:AA003543 CAS No.:185629-31-6 MDL No.:MFCD08444027 MF:C8H6FNO4 MW:199.1359 |
Tris(4-fluorophenyl)phosphineCatalog No.:AA0035P8 CAS No.:18437-78-0 MDL No.:MFCD00013553 MF:C18H12F3P MW:316.2569 |
1,5,2,4-Dioxadithiane, 2,2,4,4-tetraoxideCatalog No.:AA0036O2 CAS No.:99591-74-9 MDL No.:MFCD19440777 MF:C2H4O6S2 MW:188.1796 |
2-(Diphenylphosphino)benzoic acidCatalog No.:AA0037WS CAS No.:17261-28-8 MDL No.:MFCD00674024 MF:C19H15O2P MW:306.2950 |
1H-Indole-7-carbaldehydeCatalog No.:AA0037XP CAS No.:1074-88-0 MDL No.:MFCD01318152 MF:C9H7NO MW:145.1580 |
Oxazole-5-carboxylic acidCatalog No.:AA00388J CAS No.:118994-90-4 MDL No.:MFCD04114931 MF:C4H3NO3 MW:113.0715 |
(R)-1,2,3,4-Tetrahydronaphthalen-1-amineCatalog No.:AA003BZE CAS No.:23357-46-2 MDL No.:MFCD00671629 MF:C10H13N MW:147.2169 |
2-(Hydroxymethyl)benzonitrileCatalog No.:AA003EZV CAS No.:89942-45-0 MDL No.:MFCD11111887 MF:C8H7NO MW:133.1473 |
2-Amino-4-(trifluoromethyl)pyridineCatalog No.:AA003GBI CAS No.:106447-97-6 MDL No.:MFCD00077455 MF:C6H5F3N2 MW:162.1125 |
2-Bromo-5-methylphenolCatalog No.:AA003GNE CAS No.:14847-51-9 MDL No.:MFCD11100989 MF:C7H7BrO MW:187.0339 |
2-Chloro-5-methylpyrimidineCatalog No.:AA003GWH CAS No.:22536-61-4 MDL No.:MFCD09260903 MF:C5H5ClN2 MW:128.5596 |
2-MethoxybenzaldehydeCatalog No.:AA003HGX CAS No.:135-02-4 MDL No.:MFCD00003308 MF:C8H8O2 MW:136.1479 |
3-N-Boc-AminopiperidineCatalog No.:AA003JPK CAS No.:172603-05-3 MDL No.:MFCD03839941 MF:C10H20N2O2 MW:200.2780 |
Isoquinolin-5-amineCatalog No.:AA003MBI CAS No.:1125-60-6 MDL No.:MFCD00006907 MF:C9H8N2 MW:144.1732 |
5-Bromo-1H-pyrazolo[3,4-b]pyridineCatalog No.:AA003MCW CAS No.:875781-17-2 MDL No.:MFCD05663982 MF:C6H4BrN3 MW:198.0201 |
indoline-2,3-dioneCatalog No.:AA003QZB CAS No.:91-56-5 MDL No.:MFCD00005718 MF:C8H5NO2 MW:147.1308 |
Ethanimidic acid, 2,2,2-trichloro-, methyl esterCatalog No.:AA003RIG CAS No.:2533-69-9 MDL No.:MFCD00001759 MF:C3H4Cl3NO MW:176.4290 |
1,3-Benzenedicarboxylic acid, 1-methyl esterCatalog No.:AA003S2W CAS No.:1877-71-0 MDL No.:MFCD00029972 MF:C9H8O4 MW:180.1574 |
tert-Butyl 4-oxoazepane-1-carboxylateCatalog No.:AA003SR7 CAS No.:188975-88-4 MDL No.:MFCD03788435 MF:C11H19NO3 MW:213.2735 |
3-Methylpyrazole-4-boronic Acid Pinacol EsterCatalog No.:AA006EQU CAS No.:936250-20-3 MDL No.:MFCD08690235 MF:C10H17BN2O2 MW:208.0652 |
2-Aminopyridin-4-ylboronic acidCatalog No.:AA006RI4 CAS No.:903513-62-2 MDL No.:MFCD11520501 MF:C5H7BN2O2 MW:137.9323 |
4-(2-Aminoethyl)anilineCatalog No.:AA00703L CAS No.:13472-00-9 MDL No.:MFCD00008194 MF:C8H12N2 MW:136.1943 |
(R)-Pyrrolidin-3-olCatalog No.:AA007LRX CAS No.:2799-21-5 MDL No.:MFCD00145220 MF:C4H9NO MW:87.1204 |
4-Ethynyl-N,N-dimethylbenzenamineCatalog No.:AA00AOTU CAS No.:17573-94-3 MDL No.:MFCD00168816 MF:C10H11N MW:145.2010 |
1-Piperazinecarboxylic acid, 3-methyl-, 1,1-dimethylethyl esterCatalog No.:AA00HFN8 CAS No.:120737-59-9 MDL No.:MFCD03001706 MF:C10H20N2O2 MW:200.2780 |
4-Amino-5-hydroxynaphthalene-2,7-disulfonic acidCatalog No.:AA00IFV8 CAS No.:90-20-0 MDL No.:MFCD00035728 MF:C10H9NO7S2 MW:319.3110 |
4-(Bromomethyl)-1,2-dimethoxybenzeneCatalog No.:AA003835 CAS No.:21852-32-4 MDL No.:MFCD09833606 MF:C9H11BrO2 MW:231.0864 |