Improving Processing and Performance of Pure Lignin Carbon ... - MDPI

International Journal of

Molecular Sciences Article

Improving Processing and Performance of Pure Lignin Carbon Fibers through Hardwood and Herbaceous Lignin Blends Omid Hosseinaei, David P. Harper *

, Joseph J. Bozell and Timothy G. Rials

Center for Renewable Carbon, University of Tennessee, 2506 Jacob Drive, Knoxville, TN 37996, USA; [email protected] (O.H.); [email protected] (J.J.B.); [email protected] (T.G.R.) * Correspondence: [email protected]; Tel.: +1-865-946-1121 Received: 18 April 2017; Accepted: 27 June 2017; Published: 1 July 2017

Abstract: Lignin/lignin blends were used to improve fiber spinning, stabilization rates, and properties of lignin-based carbon fibers. Organosolv lignin from Alamo switchgrass (Panicum virgatum) and yellow poplar (Liriodendron tulipifera) were used as blends for making lignin-based carbon fibers. Different ratios of yellow poplar:switchgrass lignin blends were prepared (50:50, 75:25, and 85:15 w/w). Chemical composition and thermal properties of lignin samples were determined. Thermal properties of lignins were analyzed using thermogravimetric analysis and differential scanning calorimetry. Thermal analysis confirmed switchgrass and yellow poplar lignin form miscible blends, as a single glass transition was observed. Lignin fibers were produced via melt-spinning by twin-screw extrusion. Lignin fibers were thermostabilized at different rates and subsequently carbonized. Spinnability of switchgrass lignin markedly improved by blending with yellow poplar lignin. On the other hand, switchgrass lignin significantly improved thermostabilization performance of yellow poplar fibers, preventing fusion of fibers during fast stabilization and improving mechanical properties of fibers. These results suggest a route towards a 100% renewable carbon fiber with significant decrease in production time and improved mechanical performance. Keywords: lignin; carbon fiber; hardwood; herbaceous; nuclear magnetic resonance (NMR) spectroscopy; thermal properties; thermostabilization; tensile properties

1. Introduction Lignin has been explored as a precursor for low-cost and bio-derived carbon fibers since the 1960s [1–3], but it has gained significant attention recently as many industries seek to find renewable replacements for petroleum derived carbon [4,5]. Different methods of spinning, including wet and dry spinning, melt-spinning, and electrospinning, have been investigated to manufacture lignin-based carbon fibers [1,2,4–11]. The most dominant and preferred method, especially for producing low cost fibers for composite and structural applications, is melt-spinning [2,4,12]. Melt-spinning requires lignin to be fusible (an ability to melt and flow), without undergoing extensive thermal-induced depolymerization and/or condensation reactions during extrusion. Depolymerization produces volatiles that lead to defects (mainly pores) on the surface of fibers, while condensation restricts lignin’s thermal mobility and melt flow characteristics, consequently affecting spinning performance. Manufacturing lignin carbon fibers involves multiple processing steps of melt-spinning, oxidative thermostabilization, and carbonization. Different lignins, depending on their botanical source and extraction process, have different characteristics and behave differently during melt-spinning and conversion to carbon fiber [2,3,13]. Hardwood lignin, especially organosolv, has demonstrated greater thermal mobility and better spinnability, while lignins from softwoods and grasses have better performance during thermostabilization and conversion processes [2,3,6,14,15]. Lignins with a Int. J. Mol. Sci. 2017, 18, 1410; doi:10.3390/ijms18071410

www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2017, 18, 1410

2 of 13

high number of guaiacyl (G) units (softwoods and grasses) have a more condensed and cross-linked structure, which limits thermal mobility and causes difficulty in melt-spinning [2,3,15]. In the case of organosolv hardwood lignin, ethoxylation of its side chain is responsible for its high thermal mobility and great spinnability [3,16]. On the other hand, the more condensed structure (C–C linkages) and unoccupied C5 sites in high G content lignins (grasses and softwood) promotes faster thermostabilization and reduces carbon fiber production time [3,15]. Increasing the thermal mobility of lignin to improve melt-spinnability has been the focus of much previous work, especially in the case of kraft lignin. This includes plasticizing lignin through blending with synthetic polymers such as polyethylene (PE) [17], polyethylene oxide (PEO) [2,17,18], polyethylene terephthalate (PET) [17–19], polypropylene (PP) [17–21], and polyvinyl alcohol (PVA) [18,22]. Thunga et al. [11] used a blend of chemically modified softwood kraft lignin (butyrated) and polylactic acid (PLA) for melt-spinning. In addition, chemically modified lignins, such as acetylated [23], hydrogenated [24], and phenolated [25] lignins also have been tried to improve melt-spinnability. Chemical modification is costly and in some cases, such as acetylation, produce porous fibers [23,26]. Purification and fractionation, through ultrafiltration [14], pH-fractionation [27], and solvent extraction [6], are other methods which have been used for obtaining lignin with high thermal mobility and improved melt-spinning performance. In principle, blending lignin with other polymers should be an easy and low-cost method to improve spinnability of lignin. However, in the case of polymer blends, miscibility is an important factor that is impacted by hydrogen bonding and acid-base interaction between components of the blend [18]. For example, it has been shown that most polymers such as PP, PVA, and PE are immiscible with lignin [17]. These immiscible blends can negatively affect melt-spinnability and degrade fiber processing in lignins such as Alcell or hardwood kraft, or result in the formation of hollow and porous carbon fiber [17,19,20]. Both PEO and PET can form a miscible blend with lignin, especially PEO, which forms a strong intermolecular hydrogen bonding network [16,18]. However, there are limitations, such as difficulty in thermostabilization of PEO/lignin blends and the high melting temperature of PET, which is above the thermal decomposition point for lignin [2,19]. Lignin/lignin blends, plasticizing an infusible lignin with a fusible lignin, could be an approach to improve melt processing while preventing problems related to the miscibility of lignin and synthetic polymers. Nordstrom et al. [14] were able to melt-spin softwood kraft lignin, which is originally infusible, by plasticizing it with membrane fractionated hardwood kraft lignin. In our previous work, we demonstrated how thermal properties and chemical structure of organosolv hardwood and herbaceous lignins results in differences in melt-spinning, thermostabilization performance, and properties of resulting carbon fibers [3]. Organosolv hardwood lignin, due to its lower number of aliphatic hydroxyl groups and phenolic acids, presented better spinnability and produced stronger fibers with fewer defects. On the other hand, switchgrass lignins were difficult to process into fibers and produced fibers with defects and lower tensile properties. However, the higher number of G units and more condensed structure in switchgrass lignin resulted in faster thermostabilization [3]. The goal of this paper is to use different ratios of organosolv switchgrass and yellow poplar lignin for producing melt-spun lignin-based carbon fibers. Organosolv hardwood lignin (yellow poplar) will plasticize and improve spinning of switchgrass lignin, while switchgrass lignin will help to increase the thermostabilization rate and decrease the processing time to convert lignin to carbon fiber. 2. Results and Discussion 2.1. Properties of Lignins We extensively evaluated physical and chemical properties of switchgrass and yellow poplar organosolv lignins from the same process in our previous study [3]. Here, we briefly discuss some important physical/chemical properties of lignin. Purity, acid insoluble and acid soluble lignin content, and elemental composition of lignin samples are summarized in Table 1. Both lignins have low

Int. J. Mol. Sci. 2017, 18, 1410 Int. J. Mol. Sci. 2017, 18, 1410

3 of 13 3 of 13

low amounts of impurities, which is desirable for making carbon fibers (Table 1). Similar to amounts of impurities, which is desirable for making carbon fibers (Table 1). Similar to observations observations in our previous paper, switchgrass lignin has a higher amount of impurities which in our previous paper, switchgrass lignin has a higher amount of impurities which include residual include residual hemicelluloses, ash, and protein-based components (as indicated by its higher hemicelluloses, ash, and protein-based components (as indicated by its higher nitrogen content). nitrogen content). Impurities can prevent fusion and flow during melt-spinning, cause defects on the Impurities can prevent fusion and flow during melt-spinning, cause defects on the surface of carbon surface of carbon fibers, and decrease fiber carbon yield [2,3]. fibers, and decrease fiber carbon yield [2,3]. Table 1. The composition of switchgrass and yellow poplar lignin samples. Table 1. The composition of switchgrass and yellow poplar lignin samples.

Parameter

Switchgrass Switchgrass 93.2 93.2 0.30 0.30 64.4 64.4 5.70 5.70 0.78 0.78 28.8 28.8

Parameter Purity (%) 1 1 PurityAsh (%) (%) Ash (%) C (%) C (%) H H (%) (%) N (%) N (%) O (%) O 2(%) 2 1

Yellow Poplar Yellow 96.2 Poplar 0.17 96.2 64.5 0.17 64.5 5.89 5.89 0.26 0.26 29.2 29.2

1 Purity 2 Subtracted of acid solubleand andacid acid insoluble insoluble lignin. from C, H, Purity was was sumsum of acid soluble lignin.2 Subtracted from C,N, H,and N, ash. and ash.

Hydroxyl groups are important functional groups in lignin that affect its physical properties, 31P NMR especially Tg, and consequently its functional melt and flow behavior Based on the properties, Hydroxyl groups are important groups in lignin[3,16,28]. that affect its physical 31 (phosphorus-31 nuclear magnetic resonance) results, poplar lignin Based has a higher of especially Tg , and consequently its melt and flow yellow behavior [3,16,28]. on the number P NMR phenolic hydroxyl groups while switchgrass lignin has a higher number of aliphatic hydroxyl groups (phosphorus-31 nuclear magnetic resonance) results, yellow poplar lignin has a higher number (Figure 1; Table 2). Aliphatic groups result the aformation of intermolecular of phenolic hydroxyl groups hydroxyl while switchgrass lignininhas higher number of aliphatic hydrogen hydroxyl bonds, motion, increase Tg, result and limit fusibility [16,28]. In addition, groups which (Figuredecrease 1; Table molecular 2). Aliphatic hydroxyl groups in the formation of intermolecular switchgrass lignin possesses a significant numberincrease of p-hydroxyphenyl hydroxyl groups, derived hydrogen bonds, which decrease molecular motion, Tg , and limit fusibility [16,28]. In addition, from p-hydroxyphenyl (H) units and p-coumarates 2), which also can negatively affect meltswitchgrass lignin possesses a significant number of (Table p-hydroxyphenyl hydroxyl groups, derived from spinning performance [3]. The combined presence of aliphatic hydroxyl groups and phenolic acids, p-hydroxyphenyl (H) units and p-coumarates (Table 2), which also can negatively affect melt-spinning such as p-coumarates switchgrass lignin,ofmakes continuous veryphenolic difficult acids, by increasing performance [3]. The in combined presence aliphatic hydroxylspinning groups and such as its melt viscosity and reducing its thermal [3]. spinning very difficult by increasing its melt p-coumarates in switchgrass lignin, makesstability continuous viscosity and reducing its thermal stability [3].

Internal standard

C5 substituted OH

Aliphatic OH

p-hydroxyphenyl OH

Guaiac yl OH S yringyl OH Carboxylic ac id OH

Switchgrass

Yellow poplar

ppm

Figure 1. Quantitative 31 P NMR spectra and signal assignment of lignin samples.

Figure 1. Quantitative 31P NMR spectra and signal assignment of lignin samples.

Int. J. Mol. Sci. 2017, 18, 1410

4 of 13

Table 2. Hydroxyl group contents of lignin samples obtained by quantitative 31 P NMR spectroscopy (mmol g−1 ). Sample ID Yellow poplar Switchgrass

Phenolic OH

Carboxylic Acid OH (COOH)

pHydroxyphenyl

Condensed Phenolic

Guaiacyl

0.05

0.00

0.35

0.11

0.52

0.40

Syringyl

Total Phenolic OH

Aliphatic OH

0.72

2.41

3.48

1.58

0.74

0.69

2.35

1.88

According to differential scanning calorimeter (DSC) results, Tg of switchgrass lignin was higher than yellow poplar lignin (Table 3). Higher Tg indicates less thermal mobility and restriction in molecular motion, which also will affect fusibility of lignin. This difference in the Tg of the lignins could be due to multiple factors such as the higher content of impurities, and a higher number of aliphatic hydroxyl groups in switchgrass lignin. On the other hand, lignin with a higher Tg can potentially thermostabilize faster and have shorter processing time compared to lignin with a lower Tg [4]. Tg also affects thermal softening and the melt flow temperature of lignin, as presented in Table 3; switchgrass lignin has a higher melt temperature. Table 3. Summary of key thermal properties of switchgrass and yellow poplar lignin samples. Parameter (◦ C)

Tg Delta Cp (J g−1 ◦ C−1 ) Tm (◦ C) Td (◦ C) DTG peak temperature (◦ C) DTG Peak value (% min−1 ) Mass at 300 ◦ C (%) Mass at 400 ◦ C (%) Mass at 500 ◦ C (%) Residual char (%)

Yellow Poplar

Switchgrass

119 0.43 150 259 382 −4.27 90.6 60.9 47.9 38.3

129 0.31 160 252 378 −3.27 88.0 62.1 48.8 38.1

Tg : Glass transition temperature; Td : thermal decomposition temperature (5% weight loss temperature); and Tm : zero shear melt point temperature; DTG: derivative thermogravimetric analysis.

The thermal decomposition profile of lignins shows a clear difference, as switchgrass lignin has lower thermal stability in the 200–300 ◦ C range (Figure 2; Table 3). The mass loss in this range is due to thermal decomposition of aliphatic groups, phenolic acids, and residual hemicellulose [3]. Carbon fibers made from switchgrass lignin have more defects and pores caused by the presence of less thermally stable and volatile materials [3]. The combination of higher processing temperatures and lower thermal stability in this processing range makes producing pure switchgrass lignin fibers of high quality nearly impossible without modification.

Int. Int. J. Mol. Sci.Sci. 2017, J. Mol. 2017,18, 18,1410 1410

5 of 13

100

Weight (%)

-2

60

-4

Derivative weight loss (% min-1)

0

80

40

5 of 13

Yellow poplar Switchgrass 200

400

600

800

Temperature (°C)

Figure 2. Thermal decomposition of switchgrass and yellow poplar lignins. Figure 2. Thermal decomposition of switchgrass and yellow poplar lignins.

2.2.2.2. Fiber Spinning toCarbon CarbonFiber Fiber Fiber Spinningand andConversion Conversion to The g valuesof ofdifferent different blend between Tg of switchgrass and yellow poplar poplar The TgTvalues blendratios ratiosstand stand between Tgpure of pure switchgrass and yellow lignin. expected,samples samples with lignin have the the highest Tg value (Table(Table 4). DSC4). DSC lignin. AsAs expected, with50% 50%switchgrass switchgrass lignin have highest Tg value curves of all blend samples have a single transition, which indicates a miscible blend (Figure 3) curves of all blend samples have a single transition, which indicates a miscible blend (Figure 3) [14,19]. [14,19]. Immiscible lignin/polymer blends, such as lignin/PP, present two separate Tg values for the Immiscible lignin/polymer blends, such as lignin/PP, present two separate Tg values for the individual individual polymers [19]. Immiscible blends also exhibit phase separation, fiber breakage, or a polymers [19]. Immiscible blends also exhibit phase separation, fiber breakage, or a twisted fiber twisted fiber geometry during extrusion or subsequent thermal processing. geometry during extrusion or subsequent thermal processing. Table 4. Tg of switchgrass (SG) and yellow polar (YP) lignin blends (w/w).

Table 4. Tg of switchgrass (SG) and yellow polar (YP) lignin blends (w/w). Parameter 50% YP:50% SG 75% YP:25% SG 85% YP:15% SG Tg (°C) 127 125 122 Parameter 50% YP:50% SG 75% YP:25% SG 85% YP:15% SG Delta Cp 0.35 0.37 125 0.41 Tg (◦ C)(J g−1 °C−1) 127 122 −1 1410 ◦ C−1 ) Int. J. Mol. Sci.Cp 2017, 6 of 13 0.35 0.37 0.41 Delta (J g18, All blends showed good spinnability and the ability to be continuously spun. Continuous melt spinning of pure switchgrass lignin was not possible, as it possessed a relatively high melt viscosity and presence of volatiles [3]. However, lignin remarkably Inc rease blending in Tg with switchgrass with yellow poplar YP improved spinning performance of switchgrass lignin, even for blends with high switchgrass ratio inc reased S G content (50%). 85% YP :15% SG 75% YP :25% SG 50% YP :50% SG SG

Figure 3. (DSC) thermograms of switchgrass, yellow poplar, and vary 3. Differential Differentialscanning scanningcalorimeter calorimeter (DSC) thermograms of switchgrass, yellow poplar, and ratios (YP:SG w/w) lignin composition. vary ratios (YP:SG w/w) lignin composition.

Although there was a slight difference in Tg and melt temperature of individual lignin samples, extrusion was performed at the same temperature (180 °C) for all ratios. This temperature is relatively low and far below the thermal decomposition temperature of lignins (252–259 °C). Keeping the extrusion temperature as low as possible prevents the formation of volatiles and helps produce fibers with fewer defects [14]. Yellow poplar lignin helps to decrease the melt viscosity of switchgrass lignin

Int. J. Mol. Sci. 2017, 18, 1410

6 of 13

All blends showed good spinnability and the ability to be continuously spun. Continuous melt spinning of pure switchgrass lignin was not possible, as it possessed a relatively high melt viscosity and presence of volatiles [3]. However, blending switchgrass with yellow poplar lignin remarkably improved spinning performance of switchgrass lignin, even for blends with high switchgrass ratio (50%). Although there was a slight difference in Tg and melt temperature of individual lignin samples, extrusion was performed at the same temperature (180 ◦ C) for all ratios. This temperature is relatively low and far below the thermal decomposition temperature of lignins (252–259 ◦ C). Keeping the extrusion temperature as low as possible prevents the formation of volatiles and helps produce fibers with fewer defects [14]. Yellow poplar lignin helps to decrease the melt viscosity of switchgrass lignin while making it possible to extrude at a lower temperature, decreasing the formation of volatiles. Previously, we reported non-continuous extrusion of neat switchgrass lignin at a temperature of 190 ◦ C [3]. In this work, continuous spinning of the blend at 180 ◦ C was possible. The winding speed also significantly improved. Fibers from blends were spun at about 60 m min−1 , while neat switchgrass lignin could only be spun at 18 m min−1 [3]. 2.3. Properties of Lignin-Based Carbon Fibers Prior to carbonization, lignin fibers were oxidatively thermostabilized at different rates to evaluate the effect of blending on thermostabilization performance (Table 5). The thermostabilization rate needs to be slow enough to prevent the fusion of fibers. Lower S/G ratios and a more condensed structure of switchgrass lignin (higher number of C–C linkages) facilitates faster stabilization [3]. The 50% blend had the best stabilization performance and did not fuse even at the fastest rate (Table 5; Figure 4). Increasing the ratio of yellow poplar lignin resulted in slightly decreasing stabilization rate performance and resulted in some fiber fusion, especially in the blend with the lowest level of switchgrass lignin (Table 5). Unlike previously reported results for melt spinning of neat switchgrass lignin [3], pores or surface defects were not observed on the fibers (Figure 4). The thermostabilization performance of blended fibers was even better than previously reported data for neat samples from both biomass sources [3]. In the case of yellow poplar lignin, the improvement is directly related to the effect of switchgrass lignin, which facilitates crosslinking and condensation reactions during thermostabilization. The higher Tg of switchgrass lignin in this paper compared to our previous report is the reason that fibers made from a blend of this lignin can be thermostabilized faster compared to fibers prepared neat switchgrass lignin [3]. Higher Tg facilitates thermostabilization and makes it Int. J.from Mol. Sci. 2017, 18, 1410 Int. J. Mol. Sci. 2017, 18, 1410 possible to decrease the processing time required for manufacturing carbon fiber [4].

7 of 13 7 of 13 7 of 13 7 of 13

Mol. Sci. 2017,Sci. 18,2017, 1410Int. Int. J. Mol. Sci. 2017,Int. 18,J.1410 Mol. Int. Sci. J. 2017, Int. 18, J.1410 Mol. 18,J.1410 Mol. Sci. 2017, 18, 1410 7 of 13 7 of 13 Int. J. Mol. Sci. 2017, 18, 1410 of 13 Table 5. Thermostabilization behavior of lignin fibers made from different ratios of7switchgrass and Int. J. Mol. Sci. 2017,Int. 18,J.1410 Mol. Sci. 2017,Int. 18, J. 1410 Mol. Sci. 2017, 18, 1410 7 of 13 7 of 13 and 7 of 13 Table 5. Thermostabilization behavior ofmade ligninfrom fibersdifferent made from different ratios ofand switchgrass Table 5. Thermostabilization behavior of lignin fibers ratios of switchgrass J. Mol. Sci. 2017, 18, 1410 7 of 13 yellow polar lignin blends. Table 5.Table Thermostabilization behavior ofdifferent lignin fibers made from different ratios ofratios switchgrass andratios Table Thermostabilization Table 5. Thermostabilization behavior of 5. lignin Thermostabilization behavior fibers Table made of 5. lignin Thermostabilization from behavior fibers made of lignin ratios from behavior of fibers different switchgrass made of lignin ratios from and of fibers different switchgrass made from and of different switchgrass and of switch Int. J. Mol. Sci.5.2017, Int. 18, J. 1410 Mol. Sci. 2017, 18, 1410 7 of 13 7 of 13 Table 5. Thermostabilization behavior of lignin fibers made from different ratios of switchgrass and

yellowblends. polar lignin blends. yellow polar lignin Table Thermostabilization Table 5.polar Thermostabilization behavior Table of 5.polar lignin Thermostabilization behavior fibers made ofpolar lignin from behavior fibers different made of lignin ratios fromof fibers different switchgrass made ratios from and ofdifferent ratios and of switchgrass and yellow polar lignin blends. yellow5.polar lignin yellow blends. yellow blends. lignin yellow blends. lignin −1switchgrass 7 ofblends. 13 Rate (°C min ) yellow polar ligninlignin blends. Table 5. Thermostabilization behavior of lignin fibers made from different ratios ofHeating switchgrass and Source Table Thermostabilization Table 5.polar Thermostabilization behavior ofpolar ligninbehavior fibersblends. made of lignin fromfibers different made ratios fromofdifferent switchgrass ratios and of−1switchgrass and Heating Rate (°C min ) yellow5.polar lignin yellow blends. lignin yellow blends. lignin ◦ − 1 −10.5 0.1) −1(°C 0.2 −1 Rate Heating Rate (0.05 C(°C min Source Rate min ) min−1)Rate (°C min−1) Heating Rate (°C Heating min Heating min )Rate (°C Heating yellow polar lignin blends. −1 Heating Rate)Heating (°C min ) Source yellowofpolar lignin yellow blends. polar lignin blends.ratiosSource 0.05 0.1 0.2 0.5 −1 Source Source Source Source zation behavior lignin fibers made from different of switchgrass and −1 −1 Source Heating Rate (°C Heating )Rate (°C Heating min )Rate (°C ) 0.05 0.1 0.5 0.05 0.1 0.2 0.05 0.50.1 0.05 0.50.2 0.1 0.2 0.05 0.50.1 0.2 0.5 0.05 0.1min 0.2 0.5min 50% YP:50% SG −1Source 0.05 0.1 0.2 0.2 0.5 SourceHeatingSource Rate (°C min ) min nds. −1)Rate (°C min−1) 50% YP:50% SG Heating Rate (°C Heating 0.05 0.1 0.2 0.05 0.5 0.1 0.2 0.05 0.5 0.1 0.2 0.5 Source 50% YP:50% SGSG SG50% YP:50% SG 50% YP:50% SG 0.05 50% YP:50% SG50% YP:50% SG 50% YP:50% Source Source 0.1 0.2 0.5 YP:25% 50% YP:50% SG75% −1) 0.05 0.1 0.2 0.05 0.50.1 0.5 Heating 75% Rate (°C min YP:25% SG 50% YP:50% SG50% YP:50% SG50% YP:50% SG 0.2 75% YP:25% SG Source YP:50% SG SG 75% YP:25% SGSG SG75% YP:25% SG 75% YP:25% SG YP:25% SG 75% YP:25% 0.0550%0.1 0.2 0.5 85% YP:15% 50% YP:50% SG75% 50% YP:50% SG YP:15% 75% YP:25% SG85%

75% YP:25% SG75% YP:25% SG75% YP:25% 85% YP:15% SG SG 75% YP:25% SGfusion 85% YP:15% SG Excellent (without and sticking together); Good (slightly stickYP:15% together);and 85% YP:15% SG 85% YP:15% SG 85% SG85% SG GoodModerately Excellent (without fusion andYP:15% sticking together); (slightlyfused. stick together); 75% YP:25% SG 75% YP:25% SG 85% YP:15% SG ExcellentSG85% (without fusion andYP:15% sticking together); Good (slightly stick together); 85% YP:15% YP:15% SG 85% SG and Moderately fused. Excellent (without fusion and (without sticking together); Good together); (slightly stick together); Excellent Excellent fusion (without andfusion Excellent sticking fusion (without together); and Excellent sticking fusion together); Good and sticking (slightly fusion Good stick and sticking together); (slightly stick together); (slightly Good stick together); (slightly stick 75% YP:25% SG (without 85% YP:15% SG Excellent (without andYP:15% sticking together); Goodtogether); (slightly stickGood together); andYP:15% Moderately fused. 85% SG 85% SG (without Excellent fusion (without and Excellent sticking fusion (without together); and sticking fusion together); Good and sticking (slightly together); Good stick together); (slightly Good stick together); (slightly stick together); and Moderately fused. and Excellent Moderately and fused. Moderately and fused. Moderately and fused. Moderately fused. Thesticking mechanical properties of carbon fibers stick are summarized in Table 6 and Figures 5 and 6. The and fusion Moderately fused. Excellent (without and together); Good (slightly together); in Table 6 and Figures 5 and 6. The 85% YP:15% SG The mechanical properties of carbon fibers are summarized (without fusion (without and sticking fusion together); and sticking together); Good (slightly Good stick together); (slightly stick test together); and Excellent Moderately and Excellent fused. Moderately and fused. Moderately fused. data were analyzed with analysis of variance (ANOVA) followed by Tukey for comparing Theproperties mechanical properties offibers carbon fibers are Table 6summarized and and 6.56means The The mechanical The properties mechanical of carbon The properties mechanical fibers ofare carbon The properties summarized mechanical of are carbon in properties summarized Table fibers 6 summarized and ofTable are Figures carbon insummarized Table fibers 5in 6and and are 6.Figures inThe Table 5Figures 66. and and 6.Figures in5The Table and and6.Figure The and Moderately fused. The mechanical of carbon fibers are summarized in 6 and Figures 5 and The data were analyzed with analysis of variance (ANOVA) followed by Tukey test for comparing means and Moderately and fused. Moderately fused. t fusion The andmechanical sticking together); Good (slightly stick together); using JMP software (Supplementary Materials, Table S1). In all blends, the diameter of the fibers The properties mechanical of carbon The properties mechanical fibers of are carbon properties summarized fibers of are carbon in summarized Table fibers 6 and are Figures in summarized Table 5 6 and and 6. Figures in The Table 5 6 and and 6. Figures The 5 and 6. data were analyzed with analysis of variance (ANOVA) followed by Tukey testfollowed formeans comparing means data were analyzed data were analysis analyzed data of with were variance analysis analyzed (ANOVA) data ofwith were variance analysis analyzed followed (ANOVA) ofby with variance Tukey analysis followed test (ANOVA) of for by variance comparing Tukey followed test (ANOVA) means for bycomparing Tukey test means for bycomparing Tukey testmeans forThe com dataproperties werewith analyzed with analysis of variance (ANOVA) followed by Tukey test for comparing using JMP software (Supplementary Materials, Table S1). In all blends, thethat diameter of the fibers The mechanical of carbon fibers are summarized in Table 6 and Figures 5 and 6. The used. increased slightly at increased stabilization rates (Table 6). It has been shown slower stabilization mechanical The properties mechanical ofofcarbon properties fibers ofare carbon summarized fibers are in summarized Table 6the and Figures in Table 56the and and 6. Figures The 5the and 6. The data were analyzed data with were analysis analyzed data with were variance analysis analyzed (ANOVA) of with variance analysis followed (ANOVA) of by variance Tukey followed test (ANOVA) for by comparing Tukey followed test means for by comparing Tukey test means for comparing means using JMP software (Supplementary Materials, Table S1). In all blends, the diameter of the fibers usingThe JMP software using (Supplementary JMP software using (Supplementary JMP Materials, software using Table (Supplementary JMP Materials, S1). software In all Table blends, (Supplementary Materials, S1). In all diameter Table blends, Materials, S1). of In the all diameter Table fibers blends, S1). of In the all diameter fibers blends, of the the diamete fibers using JMP software (Supplementary Materials, Table S1). Infor all blends, diameter of the increased slightly at increased stabilization rates (Table 6).decreased Itthe has been shown thatfibers slower stabilization a were analyzed with analysis of variance (ANOVA) followed by Tukey test comparing means rates induce oxidation, increase mass loss, and result in fiber diameters [3,29]. Therefore, data analyzed data with were analysis analyzed of with variance analysis (ANOVA) of variance followed (ANOVA) by Tukey followed test for by Tukey test means for comparing means using JMPslightly software using (Supplementary JMP software using (Supplementary JMP Materials, software Table (Supplementary Materials, S1). In all Table blends, Materials, S1). the In all diameter Table blends, S1). of the In the all diameter fibers blends, the the diameter fibers of the increased slightly at increased stabilization rates (Table 6). Itcomparing has been shown that stabilization increased increased atare increased slightly increased stabilization at increased slightly rates increased stabilization at (Table increased 6). slightly It rates has stabilization at (Table been increased shown 6). It rates has stabilization that (Table been slower shown 6). stabilization It rates has that (Table been slower shown 6).of stabilization Itslower has that been slower shown stabilization thatfibers slowe perties of were carbon fibers summarized in Table 6 and Figures 5 and 6. The increased slightly at increased stabilization rates (Table 6). It has been shown that slower stabilization rates Materials, induce oxidation, increase loss, and result fiber inofdecreased fiber with diameters [3,29].switchgrass Therefore, ng JMP software (Supplementary Table S1). In all mass blends, the diameter the fibers faster stabilization rates can increase the final carbon yield. Fibers the lowest using JMP software using (Supplementary JMP software (Supplementary Materials, Table Materials, S1). In all Table blends, S1). the In all diameter blends, of the the diameter fibers of the fibers increased slightly increased at increased slightly increased stabilization at increased slightly rates stabilization at (Table increased It rates has stabilization (Table been shown It rates has that (Table been slower shown stabilization It has that been slower shown stabilization that slower stabilization rates induce oxidation, increase mass loss, result in decreased fiber diameters [3,29]. Therefore, ratesofinduce oxidation, rates induce increase oxidation, rates mass induce loss, increase and oxidation, rates mass result induce loss, increase in6). decreased and oxidation, mass result fiber loss, increase in6). decreased diameters and mass result fiber [3,29]. loss, in6).decreased diameters and Therefore, result fiber [3,29]. inTherefore, decreased diameters Therefore, fiber [3,29]. diameters Therefore, [3,2 analysis variance (ANOVA) followed by Tukey test for comparing means rates induce oxidation, increase mass loss, and result in and decreased fiber diameters [3,29]. faster stabilization rates can increase the final carbon fiber yield. Fibers with the lowest switchgrass reased slightlyslightly at increased stabilization rates (Table 6). It has been shown that slower stabilization lignin content (15%) have the smallest diameter, which switchgrass lignin increases increased increased atrates increased slightly stabilization at increased rates stabilization (Table 6). It rates has (Table been shown 6). It has that been slower shown stabilization that slower stabilization rates induce oxidation, rates induce increase oxidation, rates mass induce loss, increase and oxidation, mass result loss, increase in decreased and mass result fiber loss, in decreased diameters and result fiber [3,29]. in indicates decreased diameters Therefore, fiber [3,29]. diameters Therefore, [3,29]. Therefore, faster stabilization rates can increase the final carbon fiber yield. Fibers with the lowest switchgrass faster stabilization faster stabilization can increase faster rates stabilization the can final increase faster carbon rates stabilization the fiber can final increase yield. carbon rates Fibers the fiber can final with increase yield. carbon the Fibers lowest the fiber final with switchgrass yield. carbon the Fibers lowest fiber with switchgrass yield. the Fibers lowest with switchgrass the lowe plementary Materials, Table S1). In all blends, the diameter of the fibers faster increase stabilization rates can increase the final fiber yield. Fibers with the lowest switchgrasslignin increases lignin content (15%)in have thecarbon smallest diameter, which indicates switchgrass es induce oxidation, mass loss, and result decreased fiber diameters [3,29]. Therefore, extensional viscosity and prevents stretching of the fibers during spinning. These observations are rates induce oxidation, rates induce increase oxidation, mass loss, increase and mass result loss, in decreased and result fiber inswitchgrass decreased diameters fiber [3,29]. diameters Therefore, [3,29]. Therefore, faster stabilization faster rates stabilization can increase faster rates stabilization the can final increase carbon rates the fiber can final increase yield. carbon Fibers the fiber final with yield. carbon the Fibers lowest fiber with switchgrass yield. the Fibers lowest with switchgrass the lowest switchgrass lignin content (15%) have the smallest diameter, which indicates switchgrass lignin increases content lignin (15%) content have the lignin (15%) smallest content have diameter, the lignin (15%) smallest content have which diameter, the indicates (15%) smallest have which diameter, the indicates smallest which lignin switchgrass diameter, indicates increases which lignin switchgrass indicates increases lignin switchgrass increases lig asedlignin stabilization rates (Table 6). It has been shown that slower stabilization lignin content (15%) the smallest diameter, which indicates switchgrass lignin increases extensional viscosity and prevents stretching of lowest the fibers during spinning. These observations are ter stabilization rates can increase thehave final carbon fiber yield. Fibers with the switchgrass consistent with results previously reported about the effect of botanical lignin source on stabilization faster stabilization faster rates stabilization can increase rates the can final increase carbon the fiber final yield. carbon Fibers fiber with yield. the Fibers lowest with switchgrass the lowest switchgrass lignin lignin (15%) content have the lignin (15%) smallest content have diameter, the (15%) smallest have which diameter, the indicates smallest which switchgrass diameter, indicates which lignin switchgrass indicates increases lignin switchgrass increases lignin are increases extensional viscosity and prevents stretching of the fibers during spinning. These observations extensional viscosity extensional and prevents viscosity extensional stretching and prevents viscosity extensional of the stretching and fibers prevents viscosity during of the stretching spinning. and fibers prevents during of These the stretching spinning. fibers observations during of These the are spinning. fibers observations during These are spinning. observations These are ob crease masscontent loss, and result in decreased fiber diameters [3,29]. Therefore, extensional viscosity anddiameter, prevents stretching of the fibers during spinning. These observations areon stabilization consistent with results previously reported about the effectincreases of botanical lignin source nin lignin contentcontent (15%) have the smallest which indicates switchgrass lignin rate and properties of lignin-based carbon fibers [3]. lignin (15%) content have the (15%) smallest have diameter, the smallest which diameter, indicates which switchgrass indicates lignin switchgrass increases lignin extensional viscosity extensional and prevents viscosity extensional stretching and prevents viscosity of the stretching and fibers prevents during of the stretching spinning. fibers during of These the spinning. fibers observations during These are spinning. observations These are observations areo consistent with results previously reported about the effect of botanical lignin source on stabilization consistent consistent results previously with consistent results reported previously with about consistent results reported the previously effect with about of results botanical reported the previously effect lignin about of botanical source reported thelignin effect on lignin stabilization about of botanical source the effect on lignin stabilization ofincreases botanical source onlignin stabilization source an increase thewith final carbon fiber yield. Fibers with the lowest switchgrass consistent with results previously reported about the effect of botanical source on stabilization ratestretching and properties lignin-based carbon fibers [3]. ensional viscosity and prevents ofprevents the of fibers during spinning. These observations areThese The tensile strength of fibers at all blend ratios decreased slightly by increasing the stabilization extensional viscosity extensional and prevents viscosity stretching and of the stretching fibers during of the spinning. fibers during These spinning. observations are observations are consistent with consistent results previously with consistent results reported previously with about results reported the previously effect about of botanical reported the effect lignin about of botanical source the effect on lignin stabilization of botanical source on lignin stabilization source on stabilization rate and properties of lignin-based carbon fibers [3]. ratesmallest and properties rate and of lignin-based properties rate and of carbon lignin-based properties fibers rate[3]. and of carbon lignin-based properties fibers [3]. of carbon lignin-based fibers [3]. carbon fibers [3]. ve the diameter, which indicates switchgrass lignin increases 50% YP:50% SG

Int. J. Mol. Sci. 2017, 18, 1410

7 of 13

The mechanical properties of carbon fibers are summarized in Table 6 and Figures 5 and 6. The data were analyzed with analysis of variance (ANOVA) followed by Tukey test for comparing means using JMP software (Supplementary Materials, Table S1). In all blends, the diameter of the fibers increased slightly at increased stabilization rates (Table 6). It has been shown that slower stabilization rates induce oxidation, increase mass loss, and result in decreased fiber diameters [3,29]. Therefore, faster stabilization rates can increase the final carbon fiber yield. Fibers with the lowest switchgrass lignin content (15%) have the smallest diameter, which indicates switchgrass lignin increases extensional viscosity and prevents stretching of the fibers during spinning. These observations are consistent with results previously reported about the effect of botanical lignin source on stabilization rate and properties of lignin-based carbon fibers [3]. The tensile strength of fibers at all blend ratios decreased slightly by increasing the stabilization rate (Table 6). The samples with the highest switchgrass content (50%) showed no statistically significant changes in tensile properties based on different stabilization rates (Table 6; Supplementary Materials, Table S1). This could be mainly based on stabilization performance of fibers, since the sample that had the best stabilization performance had no fusion observed, even at the highest stabilization rate. Fibers with the lowest content of switchgrass lignin (15%) and the lowest stabilization rate (0.05 ◦ C min−1 ) had the best tensile strength, which was significantly different from other samples (Table 6; Supplementary Materials, Table S1). Switchgrass lignin, compared to yellow poplar lignin, can produce more volatiles and form pores or defects on the surface of fibers. This decreases the mechanical properties of carbon fibers [3]. Therefore, fibers with high switchgrass lignin content (50%) had the lowest tensile strength at the same stabilization rate (0.05 ◦ C min−1 ). Based on these results, the sample with the lowest ratio of switchgrass lignin and which stabilized at the slowest rate (0.05 ◦ C min−1 ), to prevent fusion, had better tensile strength compared to the samples with high switchgrass lignin content. On the other hand, at the fastest stabilization rate (0. 5 ◦ C min−1 ), the samples with high switchgrass lignin content (50%), which did not fuse, had the highest tensile strength (Table 6, Figures 5 and 6). Tensile modulus of carbon fibers also followed the same trend as tensile strength, where the blend of 50% switchgrass lignin had the best performance in different stabilization rates, and the sample with 15% switchgrass lignin had the highest tensile modulus at the slowest stabilization rate (Table 6). However, based on statistical analysis, the differences between tensile modulus of all blend ratios stabilized at a rate of 0.05 ◦ C min−1 were not significant (Supplementary Materials, Table S1). For both tensile strength and tensile modulus, the lowest values were observed in samples with 15% and 25% switchgrass lignin when stabilized at the fastest stabilization rate (0.5 ◦ C min−1 ). These values were significantly different from other values (Table 6; Supplementary Materials, Table S1). The highest tensile strength was observed in the sample with 15% switchgrass lignin when stabilized at the slowest stabilization rate (0.05 ◦ C min−1 ). This value was significantly different from other values (Table 6; Supplementary Materials, Table S1). Based on the results, increasing switchgrass lignin content can accelerate stabilization and prevent fusion, but due to its effects on formation of defects, it can reduce tensile strength when used at high quantity.

Int. J. Mol. Sci. 2017, 18, 1410 Int. J. Mol. Sci. 2017, 18, 1410

8 of 13 8 of 13

a

b

c

d

e

f

Figure of of carbon fibers made from switchgrass and and yellow poplar ligninlignin blends: (a,b) Figure 4. 4. SEM SEMimages images carbon fibers made from switchgrass yellow poplar blends: −1− ◦ 1 , (c,d) 50% YP:50% 50% YP:50% SG thermostabilized at heating rate of 0.05 °C min (a,b) 50% YP:50% SG thermostabilized at heating rate of 0.05 C min , (c,d) 50% YP:50% SG SG −1 1 75% YP:25% SG thermostabilized at heating rate thermostabilized thermostabilized at atheating heatingrate rateofof0.5 0.5°C◦ Cmin min,−(e) , (e) 75% YP:25% SG thermostabilized at heating −1. ◦ C−1min −1(f) ◦C of 0.05 , and 85% SG thermostabilized at heating rate ofrate 0.05of°C min rate of °C 0.05min , and (f)YP:15% 85% YP:15% SG thermostabilized at heating 0.05 min−1 .

Tensile modulus of carbon fibers also followed the same trend as tensile strength, where the blend of 50% switchgrass lignin had the best performance in different stabilization rates, and the sample with 15% switchgrass lignin had the highest tensile modulus at the slowest stabilization rate (Table 6). However, based on statistical analysis, the differences between tensile modulus of all blend ratios stabilized at a rate of 0.05 °C min−1 were not significant (Supplementary Materials, Table S1). For both tensile strength and tensile modulus, the lowest values were observed in samples with 15% and 25% switchgrass lignin when stabilized at the fastest stabilization rate (0.5 °C min−1). These values were significantly different from other values (Table 6; Supplementary Materials, Table S1). The highest tensile strength was observed in the sample with 15% switchgrass lignin when stabilized at the slowest stabilization rate (0.05 °C min−1). This value was significantly different from other values (Table 6; Supplementary Materials, Table S1). Based on the results, increasing switchgrass lignin content can accelerate stabilization and prevent fusion, but due to its effects on formation of defects, it can reduce tensile strength when used at high quantity.

Int. J. Mol. Sci. 2017, 18, 1410

9 of 13

Table 6. Mechanical properties of carbon fibers made from different ratios of switchgrass and yellow poplar lignin blend a.

Stabilization Rate (°C min−1)

Tensile Tensile Strain at 9 of 13 Strength Modulus Break (MPa) (GPa) (%) Table 6. Mechanical properties of carbon fibers made from different ratios of switchgrass and yellow 24.1 571 39.4 1.07 a. 50% YP:50% 0.05 poplar lignin blendSG (2.1) (111) (2.8) (0.16) 26.8 556 39.6 1.23 Tensile Tensile 50% YP:50% SG 0.1 Diameter Strain(0.24) at Break Stabilization Rate (1.2) (131) (3.0) Modulus Strength Source (%) (◦ C min−1 ) (µm) (GPa)39.7 (MPa) 552 29.3 1.19 50% YP:50% SG 0.2 24.1 (1.4) 571 (130) 39.4(2.0) 1.07 (0.25) 50% YP:50% SG 0.05 (2.1) (111) (2.8) (0.16) 32.0 525 38.3 1.25 26.8 556 39.6 1.23 50% YP:50% SG 0.1 0.5 50% YP:50% SG (1.4) (87) (3.2) (0.18) (1.2) (131) (3.0) (0.24) 29.3 552 607 39.7 40.5 1.19 18.8 1.11 50% YP:50% SG 0.2 75% YP:25% SG 0.05 (1.4) (130) (2.0) (0.25) (1.3) (169) (3.1) (0.30) 32.0 525 38.3 1.25 50% YP:50% SG 0.5 21.4 1.32 (1.4) (87) 562 (3.2) 33.6 (0.18) 75% YP:25% SG 0.1 18.8 607 (137) 40.5(3.3) 1.11 (3.3) (0.13) 75% YP:25% SG 0.05 (1.3) (169) (3.1) (0.30) 22.9 1.41 21.4 562 558 33.6 33.2 1.32 75% YP:25% SG 0.1 0.2 75% YP:25% SG (3.3) (2.0) (137) (178) (3.3)(4.6) (0.13) (0.26) 22.9 558 33.2 1.41 31.7 332 30.5 1.36 75% YP:25% SG 0.2 (2.0) (178) (4.6) (0.26) 75% YP:25% SG 0.5 (3.4) (110) (5.1) (0.35) 31.7 332 30.5 1.36 75% YP:25% SG 0.5 (3.4) (110) 747 (5.1) 41.8 (0.35) 15.7 1.18 85% YP:15% SG 0.05 15.7 747 41.8 1.18 85% YP:15% SG 0.05 (1.1) (208) (3.9) (0.40) (1.1) (208) (3.9) (0.40) 17.0 1.09 17.0 569 569 40.1 40.1 1.09 85% YP:15% SG 85% YP:15% SG 0.1 0.1 (1.0) (1.0) (103) (103) (4.9)(4.9) (0.27) (0.27) 19.1 447 33.8 1.35 85% YP:15% SG 0.2 19.1 1.35 (1.6) (100) 447 (5.0) 33.8 (0.29) 85% YP:15% SG 0.2 23.4 229 (100) 30.4(5.0) 1.15 (1.6) (0.29) 85% YP:15% SG 0.5 (1.9) (82) 229 (5.9) 30.4 (0.35) 23.4 1.15 85% YP:15% SG 0.5 deviations are shown in parentheses. a Standard (1.9) (82) (5.9) (0.35)

Int. J. Mol. Sci. 2017, 18, 1410

Source

a

Diameter (µm)

Standard deviations are shown in parentheses.

1200 50% YP:50% S 75% YP:25% S 85% YP:15% S

Tensile strength (MPa)

1000

800

600

400

200

0 0.05

0.1

0.2

0.5

Stabilization rate (°C min-1)

Figure 5. Effect of thermostabilization rate on tensile strength of carbon fibers made from different Figure 5. Effect of thermostabilization rate on tensile strength of carbon fibers made from different ratios of switchgrass and yellow polar lignin blends. ratios of switchgrass and yellow polar lignin blends.

Int. J. Mol. Sci. 2017, 18, 1410 Int. J. Mol. Sci. 2017, 18, 1410

10 of 13 10 of 13

60 50% YP:50% S 75% YP:25% S 85% YP:15% S

Tensile modulus (GPa)

50

40

30

20

10

0 0.05

0.2

0.1

0.5

Stabilization rate (°C min-1)

Figure 6. Effect of thermostabilization rate on tensile modulus of carbon fibers made from different Figure of thermostabilization on tensile ratios6.ofEffect switchgrass and yellow polarrate lignin blends.modulus of carbon fibers made from different ratios of switchgrass and yellow polar lignin blends.

3. Materials and Methods 3. Materials and Methods 3.1. Materials 3.1. Materials Organosolv lignins were obtained from Alamo switchgrass (knife milled to 2–5 cm) and pulp grade 2 and thicknesses Organosolv lignins were obtained from Alamo switchgrass (knife milled todescribed 2–5 cm) and pulp yellow poplar chips (about 4 cm of 0.5–1 cm) using a previously organosolv 2 grade yellow poplar chips (about cm and thicknesses of 0.5–1 cm) using in a previously described fraction method [3,30,31]. In brief,4 organosolv fractionation was performed a flow-through reactor organosolv fraction method [3,30,31]. In brief, organosolv fractionation was performed in a flowwith a 16:34:50 w/w mixture of methyl isobutyl ketone (MIBK), ethanol, and water in the presence of ◦ C for isobutyl through withM) a 16:34:50 w/w mixture of methyl (MIBK), in sulfuricreactor acid (0.05 at a temperature of 160 120 min ketone (severity factorethanol, of 2.50).and Thewater collected the presence sulfuric to acid (0.05 M) atsolvent a temperature of a160 °C for 120 min (severity factor by of 2.50). black liquorof separated a lignin-rich phase and hemicellulose-rich aqueous phase adding The collected black liquor separated to a lignin-rich solvent phase and a hemicellulose-rich aqueous NaCl (10 g per 100 mL of deionized water in the initial solvent mixture) in a separatory funnel. Lignin phase adding NaCl g perby100 mL evaporation of deionizedofwater the until initialallsolvent in a from by both phases was (10 isolated rotary each in phase solventmixture) was removed. separatory funnel. Lignin from both phases was isolated rotary evaporation of each phasewashing until The isolation process continued by trituration of the solid by residue with diethyl ether and final allwith solvent was removed. The isolation process continued by trituration of the solid residue with deionized water to remove possible impurities. The final lignin was filtered through a paper filter diethyl etherinand final washing deionized and dried a vacuum oven atwith 80 ◦ C for 12 h. water to remove possible impurities. The final lignin was filtered through a paper filter and dried in a vacuum oven at 80 °C for 12 h. 3.2. Chemical and Elemental Composition of Lignin Samples 3.2. Chemical and Elemental Composition of Lignin Samples The ash content and lignin content (acid soluble plus acid insoluble lignin) of the lignin samples ash content andstandard lignin content (acid [32,33]. soluble Elemental plus acid insoluble lignin) of the lignin samples wasThe determined using procedures compositions of lignin samples (carbon, was determined using standard [32,33]. Elemental compositions of lignincombustion samples hydrogen, and nitrogen content) procedures were measured using a PerkinElmer 2400II CHNS/O (carbon, hydrogen, nitrogen content)were wereperformed measuredinusing a PerkinElmer 2400II CHNS/O elemental analyzer.and These measurements triplicate. combustion elemental analyzer. These measurements were performed in triplicate. 3.3. Thermal Properties of Lignin Samples 3.3. Thermal Properties Lignin Samples The lignin zero of shear melt flow temperature (T ) was determined optically using a Fisher-Johns m

melting point zero apparatus. Thermal lignin samples was studied using a PerkinElmer The lignin shear melt flow decomposition temperature (Tof m) was determined optically using a Fisher-Johns Pyris 1point thermogravimetric analyzer. Thermogravimetric were conducted in duplicate melting apparatus. Thermal decomposition of ligninanalyses samples (TGA) was studied using a PerkinElmer using approximately 5–7 mg specimens that were heatedanalyses from room temperature to 105 ◦in C at a heating Pyris 1 thermogravimetric analyzer. Thermogravimetric (TGA) were conducted duplicate ◦ C min−1 , held at this temperature for 10 min to remove moisture, and then heated to 950 ◦ C rate of 10 using approximately 5–7 mg specimens that were heated from room temperature to 105 °C at a −1 ). Glass transition temperatures at the same rate−1under a nitrogen atmosphere (10min mLto min heating rate ofheating 10 °C min , held at this temperature for 10 remove moisture, and then heated ) were determined in triplicate on approximately 2 mg samples of lignin a PerkinElmer to(T950 °C at the same heating rate under a nitrogen atmosphere (10 mL min−1using ). Glass transition g Diamond differential calorimeter (DSC) with indium standard 100 ◦using C mina−1 . temperatures (Tg) were scanning determined in triplicate on calibrated approximately 2 mg samples of at lignin PerkinElmer Diamond differential scanning calorimeter (DSC) calibrated with indium standard at 100 °C min−1. Each specimen was heated from 25 °C at a rate of 100 °C min−1 under nitrogen (UHP, 20

Int. J. Mol. Sci. 2017, 18, 1410

11 of 13

Each specimen was heated from 25 ◦ C at a rate of 100 ◦ C min−1 under nitrogen (UHP, 20 mL min−1 ) to 200 ◦ C and held at that temperature until the change in thermal energy from the sample was zero to expel any remaining moisture in the sample and erase the thermal history of the sample. The sample was then cooled to 25 ◦ C at a rate of 100 ◦ C min−1 and again heated to 220 ◦ C at the same heating rate. The second trace was used for the calculation of Tg and the corresponding heat capacity (∆Cp ). 3.4.

31 P

NMR Spectroscopy of Lignin Samples

Hydroxyl groups of lignin samples were measured using quantitative 31 P NMR spectroscopy after derivatization of lignin with 100 µL of 2-chloro-4,4,5,5-tetramethyl-1,3,2-dioxaphospholane (TMDP) [34,35]. The derivatized samples (30 mg) were dissolved in 0.75 mL of pyridine and deuterated chloroform (1.6:1 v/v) and mixed with 100 µL of a solution of N-hydroxy-5-norbornene-2,3-dicarboxylic acid imide (10 mg mL−1 ) and chromium(III) acetylacetonate (5 mg mL−1 ) as internal standard and relaxation agent, respectively. 31 P NMR spectra were acquired using an inverse-gated decoupling pulse sequence with a 90◦ pulse angle, 25 s relaxation delay, and 256 scans. 3.5. Spinning and Conversion to Carbon Fiber Yellow poplar and switchgrass lignins were mixed at ratios of 50:50, 75:25, and 85:15 (w/w) and the blend used for melt spinning. Melt spinning of the lignins was performed using a Haake MiniLab counter-rotating twin-screw extruder (Thermo Scientific), equipped with a 200 µm custom-built die. The temperature of the extruder and spinneret were 180 and 185 ◦ C, respectively. A rotating cylinder (76 mm diameter), rotating at 250 rpm, was used to collect fibers. Lignin fibers were oxidatively thermostabilized by heating the samples to 250 ◦ C under air at selected rates (0.05, 0.1, 0.2, and 0.5 ◦ C min−1 ) and then holding the samples for 30 min at 250 ◦ C in a Heratherm OGH60 oven (Thermo Scientific). The stabilized fibers were carbonized in a Lindberg/Blue M 25 mm tube furnace (Thermo Scientific) by heating from room temperature to 600 ◦ C at a rate of 3 ◦ C min−1 , holding for 5 min at 600 ◦ C, heating from 600 to 1000 ◦ C at a rate of 5 ◦ C min−1 and holding for 15 min at 1000 ◦ C, under a nitrogen flow of 0.2 L min−1 . The tensile properties of carbon fibers were measured according to the ASTM standard (ASTM C1557-03) using an Instron 5943 single column tabletop testing system [36]. Results were an average of 40 fibers per each sample. 4. Conclusions Yellow poplar and switchgrass lignins produce miscible blends that do not phase separate with thermal processing. Yellow poplar lignin improves the spinning performance of switchgrass blends. Higher switchgrass content enables thermostabilization of fibers at high rates without fusing. The fibers with the highest switchgrass lignin content (50%) did not show any fusion, even at the fastest stabilization rate (0.5 ◦ C min−1 ). Mechanical properties of carbon fibers were highest in the samples with the lowest switchgrass lignin content and stabilized at the lowest rate. However, samples with the highest switchgrass lignin content had consistent mechanical properties across all stabilization rates evaluated. This study provides a rational approach for further development of carbon fibers from lignin by combining the best performance properties of each lignin to improve overall fiber process and performance. Supplementary Materials: Supplementary materials can be found at www.mdpi.com/1422-0067/18/7/1410/s1. Acknowledgments: This research was supported by grants from the South Eastern Regional Sun Grant Center and Agriculture and Food Research Initiative Competitive Grant no. 2013-67021-21178 from the USDA National Institute of Food and Agriculture. Author Contributions: David P. Harper and Timothy G. Rials conceived and designed the experiments; O.H. performed the experiments; Omid Hosseinaei and David P. Harper analyzed the data; Joseph J. Bozell contributed reagents/materials/analysis tools and assisted with analysis; Omid Hosseinaei wrote the paper. All authors reviewed, contributed to the writing, and agreed to the article’s content. Conflicts of Interest: The authors declare no conflict of interest.

Int. J. Mol. Sci. 2017, 18, 1410

12 of 13

References 1. 2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

Otani, S.; Fukuoka, Y.; Igarashi, B.; Sasaki, K. Method For Producing Carbonized Lignin Fiber. U.S. Patent 3,461,082, 12 August 1969. Kadla, J.F.; Kubo, S.; Venditti, R.A.; Gilbert, R.D.; Compere, A.L.; Griffith, W. Lignin-based carbon fibers for composite fiber applications. Carbon 2002, 40, 2913–2920. [CrossRef] Hosseinaei, O.; Harper, D.P.; Bozell, J.J.; Rials, T.G. Role of physicochemical structure of organosolv hardwood and herbaceous lignins on carbon fiber performance. ACS Sustain. Chem. Eng. 2016, 4, 5785–5798. [CrossRef] Baker, D.A.; Rials, T.G. Recent advances in low-cost carbon fiber manufacture from lignin. J. Appl. Polym. Sci. 2013, 130, 713–728. [CrossRef] Frank, E.; Steudle, L.M.; Ingildeev, D.; Spörl, J.M.; Buchmeiser, M.R. Carbon fibers: Precursor systems, processing, structure, and properties. Angew. Chem. Int. Ed. 2014, 53, 5262–5298. [CrossRef] [PubMed] Baker, D.A.; Gallego, N.C.; Baker, F.S. On the characterization and spinning of an organic-purified lignin toward the manufacture of low-cost carbon fiber. J. Appl. Polym. Sci. 2012, 124, 227–234. [CrossRef] Dallmeyer, I.; Ko, F.; Kadla, J.F. Electrospinning of technical lignins for the production of fibrous networks. J. Wood Chem. Technol. 2010, 30, 315–329. [CrossRef] Ruiz-Rosas, R.; Bedia, J.; Lallave, M.; Loscertales, I.G.; Barrero, A.; Rodríguez-Mirasol, J.; Cordero, T. The production of submicron diameter carbon fibers by the electrospinning of lignin. Carbon 2010, 48, 696–705. [CrossRef] Baker, D.A.; Hosseinaei, O. High glass transition lignins and lignin derivatives for the manufacture of carbon and graphite fibers. U.S. Patent 20140271443 A1, 2014. Zhang, M.; Ogale, A.A. Carbon fibers from dry-spinning of acetylated softwood kraft lignin. Carbon 2014, 69, 626–629. [CrossRef] Thunga, M.; Chen, K.; Grewell, D.; Kessler, M.R. Bio-renewable precursor fibers from lignin/polylactide blends for conversion to carbon fibers. Carbon 2014, 68, 159–166. [CrossRef] Meek, N.; Penumadu, D.; Hosseinaei, O.; Harper, D.; Young, S.; Rials, T. Synthesis and characterization of lignin carbon fiber and composites. Compos. Sci. Technol. 2016, 137, 60–68. [CrossRef] Sammons, R.J.; Harper, D.P.; Labbé, N.; Bozell, J.J.; Elder, T.; Rials, T.G. Characterization of organosolv lignins using thermal and FT-IR spectroscopic analysis. BioResources 2013, 8, 2752–2767. [CrossRef] Nordström, Y.; Norberg, I.; Sjöholm, E.; Drougge, R. A new softening agent for melt spinning of softwood kraft lignin. J. Appl. Polym. Sci. 2013, 129, 1274–1279. [CrossRef] Norberg, I.; Nordström, Y.; Drougge, R.; Gellerstedt, G.; Sjöholm, E. A new method for stabilizing softwood kraft lignin fibers for carbon fiber production. J. Appl. Polym. Sci. 2012, 126, 3824–3830. [CrossRef] Kubo, S.; Kadla, J.F. Poly(ethylene oxide)/organosolv lignin blends: Relationship between thermal properties, chemical structure, and blend behavior. Macromolecules 2004, 37, 6904–6911. [CrossRef] Kadla, J.F.; Kubo, S.; Gilbert, R.D.; Venditti, R.A. Lignin-based carbon fibers. In Chemical Modification, Properties, and Usage of Lignin; Hu, T.Q., Ed.; Springer US: Boston, MA, USA, 2002; pp. 121–137. Kadla, J.F.; Kubo, S. Lignin-based polymer blends: Analysis of intermolecular interactions in lignin–synthetic polymer blends. Compos. Part A 2004, 35, 395–400. [CrossRef] Kubo, S.; Kadla, J.F. Lignin-based carbon fibers: Effect of synthetic polymer blending on fiber properties. J. Polym. Environ. 2005, 13, 97–105. [CrossRef] Kubo, S.; Yoshida, T.; Kadla, J.F. Surface porosity of lignin/pp blend carbon fibers. J. Wood Chem. Technol. 2007, 27, 257–271. [CrossRef] Kadla, J.F.; Kubo, S.; Venditti, R.A.; Gilbert, R.D. Novel hollow core fibers prepared from lignin polypropylene blends. J. Appl. Polym. Sci. 2002, 85, 1353–1355. [CrossRef] Kubo, S.; Kadla, J.F. The formation of strong intermolecular interactions in immiscible blends of poly(vinyl alcohol) (PVA) and lignin. Biomacromolecules 2003, 4, 561–567. [CrossRef] [PubMed] Uraki, Y.; Nakatani, A.; Kubo, S.; Sano, Y. Preparation of activated carbon fibers with large specific surface area from softwood acetic acid lignin. J. Wood Sci. 2001, 47, 465–469. [CrossRef] Sudo, K.; Shimizu, K. A new carbon fiber from lignin. J. Appl. Polym. Sci. 1992, 44, 127–134. [CrossRef] Sudo, K.; Shimizu, K.; Nakashima, N.; Yokoyama, A. A new modification method of exploded lignin for the preparation of a carbon fiber precursor. J. Appl. Polym. Sci. 1993, 48, 1485–1491. [CrossRef]

Int. J. Mol. Sci. 2017, 18, 1410

26. 27. 28. 29. 30.

31. 32. 33. 34.

35.

36.

13 of 13

Chatterjee, S.; Clingenpeel, A.; McKenna, A.; Rios, O.; Johs, A. Synthesis and characterization of lignin-based carbon materials with tunable microstructure. RSC Adv. 2014, 4, 4743–4753. [CrossRef] Kleinhans, H.; Salmén, L. Development of lignin carbon fibers: Evaluation of the carbonization process. J. Appl. Polym. Sci. 2016, 38. [CrossRef] Baumberger, S.; Dole, P.; Lapierre, C. Using transgenic poplars to elucidate the relationship between the structure and the thermal properties of lignins. J. Agric. Food Chem. 2002, 50, 2450–2453. [CrossRef] [PubMed] Brodin, I.; Ernstsson, M.; Gellerstedt, G.; Sjöholm, E. Oxidative stabilisation of kraft lignin for carbon fibre production. Holzforschung 2012, 66, 141–273. [CrossRef] Bozell, J.J.; Black, S.K.; Myers, M.; Cahill, D.; Miller, W.P.; Park, S. Solvent fractionation of renewable woody feedstocks: Organosolv generation of biorefinery process streams for the production of biobased chemicals. Biomass Bioenergy 2011, 35, 4197–4208. [CrossRef] Tao, J.; Hosseinaei, O.; Delbeck, L.; Kim, P.; Harper, D.P.; Bozell, J.J.; Rials, T.G.; Labbe, N. Effects of organosolv fractionation time on thermal and chemical properties of lignins. RSC Adv. 2016, 6, 79228–79235. [CrossRef] Sluiter, A.; Hames, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D. Determination of Ash in Biomass; National Renewable Energy Laboratory: Denver, CO, USA, January 2008. Sluiter, A.; Hames, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D.; Crocker, D. Determination of Structural Carbohydrates and Lignin in Biomass; National Renewable Energy Laboratory: Denver, CO, USA, April 2008. Granata, A.; Argyropoulos, D.S. 2-Chloro-4,4,5,5-tetramethyl-1,3,2-dioxaphospholane, a reagent for the accurate determination of the uncondensed and condensed phenolic moieties in lignins. J. Agric. Food Chem. 1995, 43, 1538–1544. [CrossRef] Crestini, C.; Argyropoulos, D.S. Structural analysis of wheat straw lignin by quantitative 31 P and 2D NMR spectroscopy. The occurrence of ester bonds and α-O-4 substructures. J. Agric. Food Chem. 1997, 45, 1212–1219. [CrossRef] ASTM C1557-14. Standard Test Method for Tensile Strength and Young’s Modulus of Fibers; ASTM International: West Conshohocken, PA, 2014; Available online: www.astm.org (accessed on 17 July 2015). © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).