Effect of the Coagulation Bath on the Structure and ... - ACS Publications

Feb 12, 2017 - Department of Textile Engineering, Chemistry and Science, The Nonwovens Institute, North Carolina State University, 1020 Main. Campus ...
0 downloads 0 Views 2MB Size
Research Article pubs.acs.org/journal/ascecg

Effect of the Coagulation Bath on the Structure and Mechanical Properties of Gel-Spun Lignin/Poly(vinyl alcohol) Fibers Chunhong Lu, Charles Blackwell, Qingyuan Ren, and Ericka Ford* Department of Textile Engineering, Chemistry and Science, The Nonwovens Institute, North Carolina State University, 1020 Main Campus Drive, Raleigh, North Carolina 27695, United States

Downloaded via TUFTS UNIV on June 29, 2018 at 14:31:33 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Gel spinning was investigated as an alternative approach to the melt spinning of lignin-based fibers. Lignin/ poly(vinyl alcohol) (PVA) composites with various weight percentages of lignin were gel-spun into high-strength fibers. Although lignin is an amorphous biopolymer, incorporation of the rigid filler enhanced the mechanical properties of the PVA fibers and affected their structure. Lignin stabilized the gel structure of the thermoreversible PVA gel, as noted by higher gel melting temperatures. Methanol/acetone coagulation baths with high acetone content rendered the gel-spun fibers more drawable, helped to maintain lignin within the gel fibers, and increased the gel melting point. The best mechanical performance was observed for fibers containing 5% lignin, which had an average tensile strength of 1.1 GPa, a Young’s modulus of 37 GPa, and a toughness of 17 J/g. Structural analysis of the 5% lignin fibers showed them to possess the highest index of PVA crystallinity. The fibers were more drawable at higher weight percentages of lignin. This plasticizing behavior at elevated temperatures of drawing led to stronger lignin-based fibers. Evidence of hydrogen bonding between lignin and PVA within gel-drawn fibers was observed by infrared spectroscopy, and substituents of the lignin biopolymer were mildly aligned along the fiber axis. Lignin/PVA fibers resisted dissolution in boiling water, unlike neat PVA fibers. However, fiber swelling increased with lignin content. Structurally, gel-spun composites must have regions of PVA laced between lignin. In summary, biobased fibers containing 5−50% lignin were gel-spun into composites having mechanical properties that are suitable for industrial and high-performance fiber applications. KEYWORDS: Gel spinning, Lignin, Poly(vinyl alcohol), Coagulation bath, Raman anisotropy



separation.6,8 The low molecular weight of lignin and its lack of crystalline order9 are likely causes of poor mechanical performance among carbonized fibers. Nevertheless, lignin has the potential to blend with polymers. The polar nature of lignin and its tendency to hydrogen-bond with molecular groups aid its miscibility with polar polymers.10 Kubo and Kadla10 reported strong interactions between lignin and poly(vinyl alcohol) (PVA) in melt-spun fibers, but the mechanical properties of lignin/PVA fibers were not reported. Improved interactions between lignin and polymer were observed among solvent-cast blends. Xu et al.11 and Su et al.12 found increased mechanical performance among solutioncast lignin/PVA membranes in the range of 10−20% lignin. Insufficient studies in the area of solution-spun lignin-based fibers motivated the present work, which used PVA as the matrix polymer. PVA staple fibers are used as concrete reinforcement. These fibers prevent crack propagation, and their hydroxyl groups

INTRODUCTION High-performance fibers are versatile reinforcing materials for different industries and applications. With the high demand for cost-effective high-performance fibers, recent efforts have focused on the development of high-end materials from biopolymers, namely, lignin. Lignin is the second most abundant biopolymer next to plant-based cellulose and is a waste byproduct of the pulp and paper industry.1,2 Lignin has been investigated as a filler for biobased fibers. However, melt-extruded lignin/polypropylene (PP) and lignin/ polyethylene (PE)3 fibers showed significantly less tensile strength than the lignin-free fibers. Lignin is a candidate for a precursor carbon fiber because of its high carbon yield; however, lignin-derived fibers typically demonstrate mechanical properties that are inferior to those of conventional carbon fibers.4,5 The mechanical properties of both precursor6 and carbonized fibers7 from melt-extruded lignin/poly(lactic acid) (PLA) were poorer than those of fibers prepared from PLA and carbon fiber from wholly lignin, respectively. Lignin substituents branch out three dimensionally,2 which could hinder its molecular compatibility with other polymers, especially in the melt. Melt-spun fibers often show evidence of phase © 2017 American Chemical Society

Received: October 18, 2016 Revised: February 10, 2017 Published: February 12, 2017 2949

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering

Figure 1. Homogeneity of lignin/PVA dopes is visible within (a) photographs and (b) optical micrographs. Dopes of lignin/PVA contained (1) 0%, (2) 5%, or (3) 30% lignin. The gel fiber spinning process is illustrated in (c) as three steps: as-spun gel fiber formation (step 1), gel fiber aging (step 2), and multistage fiber drawing (step 3).

promote molecular adhesion between the fibers and concrete. Solution-spun PVA fibers have high tenacity values in the range of 14−18 g/den13 (1−2 GPa). In contrast to melt spinning, solvent is used to homogenize blends used for solution spinning. High-performance PVA fibers are produced by the gel-spinning technique, and the Young’s modulus values of gelspun PVA fibers range between 20 and 115 GPa.14 The mechanical performance of gel-spun fibers is influenced by several processing parameters: solvent mixture for dissolution, gelation temperature,15 drawing temperature, and ultimate draw ratio.16 Motivating this work is the prospect of fabricating sustainable biobased fibers whose mechanical performance and low cost are ideal for use in industrial and high-performance applications. To our knowledge, this is the first study to investigate the fabrication of lignin/PVA fibers by the gel spinning technique. The structure−property relationships of gel-spun lignin/PVA fibers were investigated with respect to lignin content, composition of the solvent coagulation bath for gelation, drawing temperature, and draw ratio.



homogeneity of lignin/PVA dopes was inspected visually and optically with a Nikon Eclipse 50i POL optical microscope. Gel Spinning. Lignin/PVA dopes were dark brown in color (see Figure 1a,b). Optical micrographs of dopes having 0%, 5%, and 30% lignin (Figure 1b) showed homogeneous solutions that were absent of aggregation. This suggests that lignin and PVA were completely dissolved in DMSO/wateran ideal case for any type of solution spinning. A schematic of the gel-spinning process is shown in Figure 1c. The lignin/PVA spinning dopes were dispensed from a steel highpressure syringe. The syringe was heated to 85 °C before extrusion of dopes through a 19-gauge syringe needle (0.69 mm inner diameter). Afterward, the dope gelled in a −25 °C coagulation bath. The syringe tip to coagulation bath distance was 3−5 mm. The resulting as-spun gel fibers were collected onto a rotating winder and later immersed in the 5 °C coagulation bath for 24 h. Fibers were drawn through one to four stages of silicone oil at elevated temperatures of 90−240 °C (step 3 in Figure 1). The draw ratio (DR) at each stage of fiber drawing was calculated as

DR =

V2 V1

(1)

where V1 is the velocity of the fiber feeding winder and V2 is the velocity of the fiber take-up winder. The coagulation baths contained mixtures of methanol and acetone. Lignin/PVA gelation (in particular the gel opacity) and lignin leaching were initially tested in 20 mL of solvent. Spinning dopes at 85 °C were added dropwise into −25 °C methanol/acetone mixtures. An ideal solvent system would yield a clear gel and negligible amounts of lignin leached into the coagulation bath. Gel opacity and lignin diffusion into the coagulant were observed after 10 min and 24 h of aging the spinning dopes in −25 and 5 °C baths. PVA gels are thermoreversible. At elevated temperatures, they return to liquids of dissolved polymer. Gel melting points of lignin/ PVA gels were measured according to the method described by Ryan and Fleischer.17 Their test method was slightly modified as described in the Supporting Information and shown in Figure S1. Lignin/PVA dopes were loaded into capillary tubes having one end-capped and then were gelled in −25 °C 100/0 and 15/85 methanol/acetone baths for 10 min. The capillary tube of the gelled polymer and a thermocouple were placed in a test tube that was positioned within the center of a Thiele tube. The Thiele tube filled with silicone oil was heated using a Bunsen burner. The gel melting point was identified as

EXPERIMENTAL SECTION

Materials. PVA, having a molecular weight of 146−186 kg/mol and 99% hydrolysis, was purchased from Sigma-Aldrich. Aqueous raw pine sawdust lignin paste (project no. L28) at pH 3 was provided by Pure Lignin Environmental Technology (PLET). Lignin was extracted from wood pulp using a weak-acid hydrolysis treatment. Solvents were used as-received: dimethyl sulfoxide (DMSO) from Sigma-Aldrich and acetone and methanol from BDH Chemicals. Spinning Dope Preparation. Aqueous PLET lignin paste was filtered to remove low-molecular-weight fractions. Lignin was extracted from the paste using acetone and vacuum filtration. Afterward, the lignin was dried at 85 °C for 24 h and finely ground into powder using a mortar and pestle. Spinning dopes of PVA and lignin/PVA were prepared. PVA powder (10 g) was dissolved in 100 mL of 80/20 (v/v) DMSO/ distilled water under constant stirring at 85 °C for 1 h. Lignin/PVA dopes, at weight ratios of up to 50% (w/w) lignin to polymer, were also dissolved in 80/20 (v/v) DMSO/distilled water at 85 °C. The final concentration of PVA in the spinning dopes was 10 g/dL. The 2950

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering

Figure 2. Coagulation of lignin/PVA dopes having (a) 5% and (b) 50% lignin in (A) 100/0, (B) 90/10, (C) 80/20, (D) 70/30, (E) 60/40, (F) 50/ 50, (G) 40/60, (H) 30/70, (I) 20/80, (J) 10/90, and (K) 0/100 (v/v) methanol/acetone mixtures for (1) 10 min at −25 °C and (2) 24 h at 5 °C.

Figure 3. Illustration of fiber microstructures at (a) low and (b) high lignin contents in (1) as-spun gel fibers and (2) gel-drawn fibers. the temperature at which the stationary gel transformed into a flowing liquid. Mechanical Testing. The mechanical properties of fibers were tested on the MTS-Q testing system according to ASTM D3379. Test parameters included a strain rate of 15 mm/min, a gauge length of 25 mm, and a sample size of 10. To normalize the data, the effective cross-sectional area A was calculated gravimetrically as A = d/ρ

later imaged using a LEXT OSL4000 3D measuring laser confocal microscope. Fiber fracture tips from mechanical testing were sputter-coated with gold and imaged by scanning electron microscopy (SEM) using an FEI Verios 460L scanning electron microscope at an accelerating voltage of 2 kV. Structural Characterization. Infrared spectra of as-received PVA powder, lignin powder, and lignin/PVA fibers were obtained using a Nicolet iS50 spectrophotometer. Spectra were collected at 128 scans and a spectral resolution of 4 cm−1. The absorbance at 1144 cm−1 represents symmetric C−C stretching21 along the polymer chain, wherein neighboring hydroxyl (−OH) groups engage in intramolecular hydrogen bonding.22 Absorbance spectra of PVA were normalized to the peak intensity of the reference band at 854 cm−1.22−24 The band at 854 cm−1 (C−C stretching) was chosen as the reference band since its absorbance is not affected by processing.22 The percent crystallinity (α) of the polymer was calculated using the following equation:22−24

(2)

where d is the linear density of the fiber and ρ is the density of the composite fiber. Linear density was measured by weighing a known length of fiber. Before weighing, fibers that were hot-drawn in oil were rinsed with isopropyl alcohol to remove residual silicone oil. The density of composite fibers was determined using the expression

ρ = ρPVA (1 − wf ) + ρlignin wf

(3)

where wf is the weight fraction of lignin. The densities of PVA (ρPVA)18 and lignin (ρlignin)19 are the same. Therefore, the composite fiber’s density was ρ = 1.3 g/cm3. Tensile toughness (Ut) was calculated by integrating stress−strain curves, which represent the energy absorbed before fiber breakage,20 according to eq 4: Ut =

∫ σεi i

⎛ A ⎞ α = ⎜B + C 1144 ⎟ × 100% A 854 ⎠ ⎝

(5)

where B and C are constants whose values were calculated from known values of α as determined by X-ray diffraction.23 The absorbances at 1144 cm−1 (A1144) and 854 cm−1 (A854) were obtained from infrared spectra.23 In eq 5, the absorbance of crystalline PVA at 1144 cm−1 is normalized by that of the reference band at 854 cm−1. The A1144/A854 ratio is an index of fiber crystallinity, and values of this ratio were used to compare the relative crystallinities of the fibers. Polarized Raman spectroscopy was used to quantify the molecular anisotropy of lignin embedded in lignin/PVA fibers. With a BaySpec

(4)

where σi and εi are the stress and strain, respectively, at each data point i. Imaging Analysis. Fibers were embedded in synthetic cork. Thin, perpendicular mounts revealed fiber cross-sectional areas that were 2951

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering Nomadic confocal Raman microscope, polarized spectra were collected parallel and perpendicular to the axis of fiber bundles (containing ∼50 fibers). Fiber measurements were taken with a 10× objective lens and a 785 nm laser at 104 mW (based on the filter setting). The sample exposure time was 1 s with an acquisition number of 10. Water Dissolution and Swelling. To investigate the water resistance of the composite fibers, 3 mg bundles of neat PVA, 5% lignin, and 50% lignin fibers were placed in 75 mL of water and gradually heated from room temperature to 100 °C on a hot plate. After immersion, fibers were imaged by confocal microscopy. Fiber swelling was tested among lignin/PVA fibers. Fiber bundles containing 0%, 5%, and 30% lignin were immersed in vials of distilled water for 24 h at 25 °C. Samples were taken from the vials, and excess water was removed with filter paper. The fiber swelling ratio (S) was calculated using eq 6: w − wd S= w × 100% wd (6)

Table 1. Lignin/PVA Gel Melting Points in Response to 100/0 and 15/85 Methanol/Acetone Coagulation lignin/PVA gel 0% lignin 5% lignin 30% lignin 50% lignin

gel melting temperature (°C) in 100/0 methanol/ acetone bath 101 108 111 119

± ± ± ±

2 6 4 2

gel melting temperature (°C) in 15/85 methanol/ acetone bath 103 114 118 122

± ± ± ±

3 4 5 2

at the same lignin concentration. Therefore, higher temperatures were expected for the drawing of gel fibers produced from coagulation in 15/85 methanol/acetone. Crystalline regions in the gel structures affects the gel melting temperature. As shown in Figure 3, lignin resides in semicrystalline polymer-rich domains. Its presence enhances the thermal resistance of PVA, thereby increasing the gel melting point of lignin/PVA gels. Coagulation in pure methanol reduces the lignin content in the gel fiber. As a result, gels formed in 100/0 methanol/acetone have lower gel melting temperatures than those formed in 15/85 methanol/ acetone. Effect of the Coagulation Bath on Fiber Drawing. On the basis of our study of solvent systems for coagulation (see Figure 2), the 15/85 methanol/acetone system was identified as an appropriate bath for lignin/PVA gelation. To understand the effect of the coagulating solvent on the processing of gelspun fibers, parameters for lignin/PVA fibers that were drawn in 100/0 and 15/85 methanol/acetone baths are summarized in Table 2. The following discussion will describe the effects of the coagulation bath, gel melting point, and lignin content on the drawing parameters. Changes in stage drawing temperature and draw ratio were observed. Overall, the total draw ratio of fibers increased with lignin content (Table 2). Fibers were drawn in multiple stages after gel formation in 100/0 methanol/acetone. During stage 1 drawing, the gel fibers solidify. In general, most of the imbibed solvent diffuses from the gel fibers into the high-temperature oil bath. The stage 1 drawing temperature decreased from 100 to 90 °C for gel fibers coagulated in pure methanol as the lignin concentration increased above 10% lignin to polymer. Further, the stage 1 draw ratio decreased with additional lignin. The draw ratio also increased with lignin content during stages 2 and 3 heat drawing after 100/0 methanol/acetone coagulation. On the basis of the results from the gel melting study (Table 1), the stage 1 drawing temperature was expected to increase with lignin content. Lower drawing temperatures were also expected for 100/0 methanol/acetone-coagulated gel fibers. Lignin leached into the 100/0 methanol/acetone bath and stage 1 drawing oil from structures of 15−20% lignin gel fibers (as represented by Figure 3b1). At 15−20% lignin, the drawing temperature decreased from 100 to 90 °C for fibers containing 0−10% lignin. Acetone in the 15/85 methanol/acetone coagulation bath precluded leaching of lignin into the gel-spinning bath and drawing oil. The as-spun draw ratio decreased with increasing lignin content among gels coagulated from 15/85 methanol/ acetone baths. In contrast to pure methanol coagulation baths, more lignin would reside in the as-spun gels from 15/85 methanol/acetone baths. Since lignin remained in the thermally drawn fibers, the temperatures at consecutive stages of drawing were significantly higher for gel fibers coagulated in 15/85 methanol/acetone and with higher lignin content. Further, the

where wd is the weight of the dry fiber and ww is the weight of the fiber after wetting.



RESULTS AND DISCUSSION Effect of the Methanol/Acetone Ratio on Lignin/PVA Gelation. In general, methanol facilitates PVA gelation.15 Wholly acetone coagulation baths were not optimum for PVA gelation. When exposed to acetone, neat PVA dopes turned opaque. PVA gels from acetone coagulation were brittle and less drawable than the transparent gels formed from PVA in chilled methanol. However, adding acetone to the methanolbased coagulation bath was beneficial to lignin/PVA gelation. Lignin/PVA dopes were added dropwise into methanol/ acetone mixtures with different volume ratios (see Figure 2). Yellowing of the solvent mixture was indicative of lignin diffusion into methanol from the gel polymer. Because lignin is soluble in methanol and insoluble in acetone, more lignin diffused into the coagulation bath at higher methanol/acetone ratios, greater percentages of lignin/PVA, and longer coagulation times. Negligible amounts of leaching were observed among baths having more than 80% acetone by volume. Further, these lignin/PVA gels remained soft and flexible. Gel structures at low to high lignin content, as interpreted from the results in Figure 2, are illustrated in Figure 3. At the onset of gelation, the polymer dope phase separates into polymer-rich and solvent-rich domains.25 Crystalline regions of PVA function as physical junctions for the gel network, and solvent domains are spaced throughout the network. On the basis of the observations in Figure 2, lignin more readily diffuses into methanol-rich coagulation baths at high weight percentages of lignin and high ratios of methanol/acetone in the bath. Therefore, lignin must preferentially reside in polymer-rich domains at low lignin content. Once lignin has saturated the polymer-rich domains, it will then reside in solvent-rich domains of the gel network. Acetone in the coagulation bath suppresses diffusion of lignin from both the polymer-rich and solvent-rich domains of the PVA gel into the coagulation bath. Even at high lignin content, 80−90% acetone in the bath maintained the higher amount of lignin in the gel structure. Effect of the Lignin Content on Gel Melting. On the basis of the data shown in Table 1, lignin effectively increased the melting temperature of gels coagulated in media of 100/0 and 15/85 methanol/acetone. Gels coagulated in 15/85 methanol/acetone exhibited even higher melting temperatures 2952

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering

Table 2. Drawing Parameters and Draw Ratios (DR) for Gel-Spun Fibers with Different Lignin Contents (0−50%) 100/0 methanol/acetone bath as-spun DR stage 1 drawing stage 2 drawing stage 3 drawing stage 4 drawing total heat DRb effective diameter (μm) linear density d (dtex) a

T (°C) DR T (°C) DR T (°C) DR T (°C) DR

15/85 Methanol/Acetone Bath

0%

5%

10%

15%

20%

5%

20%

30%

50%

2.5 100 4.2 130 1.4 180 1.3 −a − 7.6 52 27

2.2 100 4.0 150 1.5 180 1.3 − − 7.8 52 27

2.3 100 4.0 150 1.5 190 1.5 − − 9.0 52 27

2.5 90 4.0 160 1.5 190 1.4 − − 8.2 51 26

2.4 90 3.3 160 2.1 220 1.5 − − 10.1 35 13

2.6 100 5.8 180 1.7 210 1.3 230 1.2 14.7 32 10

1.9 100 6.8 180 1.5 210 1.3 230 1.2 16.0 41 17

1.4 100 7.0 190 1.5 210 1.4 230 1.2 17.5 48 23

1.4 120 8.5 190 1.4 225 1.5 240 1.2 20.6 40 17

Not applicable: methanol-coagulated fibers melted and relaxed at high temperatures. bTotal heat DR is the cumulative draw ratio from stages 1 to 4.

Figure 4. Tensile strength and Young’s modulus of lignin/PVA fibers with 0−50% lignin obtained using (a) 100/0 and (b) 15/85 methanol/acetone coagulation baths during gel fiber spinning.

yield further increases in fiber mechanical performance. Instead, the fiber properties diminished drastically to a tensile strength of 0.37 GPa and a Young’s modulus of 11 GPa with 20% lignin. Voids were observed throughout the fiber cross sections of 20% lignin fibers that were coagulated in 100/0 methanol/acetone (Figure 6). Fibers from 15/85 methanol/acetone coagulation had significantly better mechanical properties than those from 100/0 methanol/acetone coagulation (Figure 4). The average tensile strength and Young’s modulus for 5% lignin fibers were 1.1 and 37 GPa, respectively, which exceeded those of commercially available high-strength Kuralon PVA staple fiber (0.88 and 23 GPa, respectively). Above 5% lignin, the fibers exhibited poorer mechanical properties. The tensile strength values were 0.74 GPa for 20% lignin, 0.77 GPa for 30% lignin, and 0.76 GPa for 50% lignin. The Young’s modulus was in the range of 31−36 GPa, which is still higher than that of commercial Kuralon PVA staple fiber from Kuraray. The mechanical properties of fibers from 15/85 methanol/ acetone coagulation were superior to those from 100/0 methanol/acetone coagulation for the following reasons. The 15/85 methanol/acetone bath suppressed lignin leaching during gelation and fiber drawing. At 20% lignin, voids were not observed in the fiber cross sections of 15/85 methanol/ acetone-gelled fibers. Further, gel fibers from the 15/85 methanol/acetone bath were more drawable. The total draw ratios for 15/85 methanol/acetone fibers were higher than those for 100/0 methanol/acetone fibers (see Table 2). Higher

stage 1 drawing temperature increased from 100 to 120 °C by 50% lignin, and the stage 1 draw ratio increased from 5.8 for 5% lignin to 8.5 for 50% lignin. Higher values of the draw ratio at elevated temperatures were attributed to the plasticizing effect of lignin on PVA. Similarly, Kubo and Kadla10 and Baumberger et al.26 generalized the improved processing of PVA as due to lignin plasticization of PVA/lignin blends. Lignin has a lower molecular weight than PVA, and it can disrupt hydrogen bonding between PVA molecules. The glass transition temperature of lignin (Tg = 120−150 °C27) is expected to influence the fibers’ molecular mobility at drawing temperatures close to Tg. As a result, the extension of matrix PVA chains was more feasible at higher temperatures. Effect of the Coagulation Bath on the Fiber Mechanical Properties. The effect of the coagulation solvent on the tensile strength and Young’s modulus of lignin/PVA fibers (up to 50% lignin) is shown in Figure 4. Neat PVA fibers were spun using 100/0 methanol/acetone coagulation; however, it was not possible to spin neat PVA fibers using 15/85 methanol/acetone coagulation. Acetone is attributed with turning the neat PVA gel fibers opaque and making them too brittle for spinning. As the lignin content increased in the 100/0 methanol/ acetone-coagulated fibers, lignin did aid fiber strengthening among some PVA fibers (Figure 4a). The maximum tensile strength was 0.61 GPa at 5% lignin, and the maximum modulus was 24 GPa at 10% lignin. Higher lignin concentrations did not 2953

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering

lation from 15/85 methanol/acetone yielded tougher fibers than those obtained from wholly methanol baths. A maximum toughness value of 8.5 J/g was observed at 5% lignin upon 100/ 0 methanol/acetone coagulation. An even higher value of 17 J/ g was obtained for 5% lignin fibers upon 15/85 methanol/ acetone coagulation. As the lignin content increased, the toughness slightly decreased. Toughness values of 10−12 J/g were reported at 20−50% lignin with 15/85 methanol/acetone coagulationvalues that are greater than that of graphene oxide reinforced PVA composites (6 J/g).28 The toughness value of 12 J/g for 50% lignin fiber was greater than those at 20−30% lignin, although the tensile modulus at 50% lignin was less than the values at those lower weight fractions. Upon incorporation of lignin at 50% of the matrix polymer, its rigid structure increased the fiber toughness. A discussion of fiber morphology is used below to explain the apparent differences in mechanical performance. Effect of Lignin on Fiber Morphology. Neat PVA and 5% lignin fibers obtained from 100/0 methanol/acetone coagulation were smooth and had round cross sections. At higher lignin contents, the fiber cross sections were less circular (Figure 6) as a result of nonuniform solvent removal. Kidneybean-shaped cross sections were observed among fibers having 20% lignin obtained from coagulation in 100/0 methanol/ acetone. Similarly, 15/85 methanol/acetone coagulation also resulted in kidney-bean-shaped fibers. At 5% lignin, fibers from coagulation in 15/85 methanol/acetone were finer than fibers from the 100/0 methanol/acetone bath. Gel fibers from 15/85 methanol/acetone had higher total draw ratios and lower values of linear density (Table 2), which correspond to smallerdiameter fibers. The coagulation solvent and lignin content affected solvent removal from gel-spun fibers. Solvent removal occurred in the coagulation bath and high-temperature drawing

fiber draw ratios typically facilitate polymer chain alignment along the fiber axis for better mechanical performance. Furthermore, molecular interactions between lignin and the matrix polymer can disrupt hydrogen bonding between the PVA chains, so lignin plasticizes PVA’s mobility during hightemperature drawing. Evidence of PVA/lignin bonding will be discussed with infrared analysis. The effect of the coagulation bath on lignin/PVA fiber toughness is shown in Figure 5. Characteristics of tough fibers

Figure 5. Toughness of lignin/PVA fibers having increasing amounts of lignin that were coagulated in 100/0 (▲) or 15/85 (●) methanol/ acetone.

include high strength and ductility. The strain at break values for all of the lignin/PVA fibers ranged between 3.1% and 4.2% regardless of the coagulation bath. The lignin/PVA fibers were tougher than neat PVA fibers, as observed for fibers from pure methanol coagulation. Coagu-

Figure 6. Confocal micrographs of cross sections of (a) 0, (b) 5%, and (c) 20% lignin/PVA fibers obtained from (1) 100/0 and (2) 15/85 methanol/acetone coagulation baths. 2954

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering

Figure 7. (1) Low- and (2) high-resolution SEM images of fracture tips of (a) neat PVA fiber from 100/0 methanol/acetone coagulation and (b) 5%, (c) 20%, (d) 30%, and (e) 50% lignin fibers from 15/85 methanol/acetone coagulation after mechanical testing.

Figure 8. IR absorbance spectra of lignin powder, PVA powder, neat PVA fiber coagulated in 100/0 methanol/acetone, and lignin/PVA fibers coagulated in 15/85 methanol/acetone between (a) 4200−1000 cm−1 and (b) 1250−800 cm−1.

oil. Voids throughout the fiber cross section of 20% lignin fibers had resulted in significant losses in mechanical performance (Figure 4a). These voids were likely caused by the high content of lignin in the solvent-rich domains of the gel fiber (Figure 3b1), which was followed by the aggressive diffusion of lignin from the as-spun gel to methanol and from the gel fiber to the high-temperature drawing oil. Fibers having 20% lignin obtained from 15/85 methanol/ acetone coagulation were finer than the 20% lignin fibers obtained from 100/0 methanol/acetone. Also, their fiber cross sections were rougher and without voids. Methanol/acetone coagulation suppressed voiding and enabled better mechanical

performance (Figure 6). The reduction in mechanical performance among 15/85 methanol/acetone-coagulated fibers having more than 20% lignin was caused by other differences in fiber morphology, namely, lignin aggregation. After mechanical testing, fiber fracture tips were imaged by SEM, as shown in Figure 7. Neat PVA fibers showed a smooth fracture tip (Figure 7a). Its microstructure contrasts with the fibrillar, more ductile fracture tip of 5% lignin fibers (Figure 7b). PVA fibrils are associated with highly oriented and ordered chains of polymer. (The next section describes indices of crystallinity among gel-spun PVA fibers versus lignin content measured by infrared spectroscopy.) The fibrillar micro2955

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering

The Raman anisotropy (R), given by eq 7, has been used to quantify the molecular orientation of conjugated polymers:29 I R= I⊥ (7)

structure appeared as a result of lignin plasticization and is responsible for the good mechanical properties. PVA fibrils were noticeable among 30% lignin fibers (Figure 7d). At 50% lignin (Figure 7e), the fibers were less fibrillar, and lignin aggregation (in the form of small beads) was observed at high resolution (Figure 7e2). Lignin aggregation is attributed to its concentration within polymer-poor domains of the gel fiber at high lignin content (Figure 3b1). As a result, lignin was not homogeneously dispersed throughout the polymer. In summary, lignin promoted the formation of fibrillar PVA at high fiber draw ratios; however, lignin aggregates were structural defects that limited the mechanical performance of fiber at more than 20% lignin. Effect of Lignin Content on Fiber Microstructure. Since 15/85 methanol/acetone coagulation yielded the highest values of mechanical performance, the effect of the lignin content on PVA crystallization was investigated by IR spectroscopy (Figure 8). Amorphous PVA is associated with the C−O vibrational mode at 1094 cm−1, and PVA crystallinity affects the peak at 1144 cm−1. The value of the A1144/A854 ratio is an index of fiber crystallinity (Table 3). After gel fiber spinning, the crystallinity

where I∥ and I⊥ represent the peak Raman intensities at polarization angles that are parallel (∥) or perpendicular (⊥) to the fiber axis. This technique was used to measure the molecular orientation of lignin within fibers coagulated in 15/ 85 methanol/acetone. The Herman orientation factor30 f is often used to quantify the molecular orientation in the direction of fiber drawing. The parameters f and R are related by eq 8:29 R−1 f= (8) R+4 For perfectly aligned molecules, R = ∞ and f = 1. For randomly aligned molecules, R = 1 and f = 0. The peak at ∼1560 cm−1 was assigned to phenol in-plane stretching, and the peak at ∼1650 cm−1 was assigned to conjugated bonds in the lignin structure31 (Figure 9). The

Table 3. Lignin/PVA Fiber Crystallinity as Indicated by the A1144/A854 Infrared Absorbance Ratio materiala

A1144/A854

PVA powder neat PVA fiber 5% lignin/PVA 20% lignin/PVA 30% lignin/PVA 50% lignin/PVA

2.92 3.34 4.67 2.88 2.98 2.60

Neat PVA fiber was fabricated from 100/0 methanol/acetone coagulation, and 5−50% lignin/PVA fibers were obtained from 15/ 85 methanol/acetone coagulation. a

Figure 9. Polarized Raman spectra parallel (∥ or 0°) and perpendicular (⊥ or 90°) to the fiber axis for 5% lignin fibers from 15/85 methanol/ acetone coagulation.

of PVA fiber was greater than for the as-received powder. Fiber drawing caused dense packing of the polymer chains. The highest degree of crystallinity occurred at 5% lignin, but the PVA crystallinity decreased from this value at more than 20% lignin. Fibers having 30% lignin were slightly more crystalline than those containing 20 and 50% lignin. However, lignin aggregation and phase separation from PVA, as observed among fiber fracture tips, was detrimental to the properties and crystallinity of PVA fibers. These trends in fiber crystallinity agree with changes in the mechanical properties of lignin/PVA fibers from 15/85 methanol/acetone coagulation (Figure 4). Intermolecular bonding between lignin and PVA induces molecular adhesion. The IR absorbance spectrum from 3000 to 3700 cm−1 (Figure 8a) provides insight into hydrogen bonding for lignin powder (3384 cm−1) and PVA (3345 cm−1). Among 5% lignin fibers, the −OH band shifted toward lower frequencies of 3333 cm−1. This behavior was suggestive of hydrogen bonding between lignin and PVA.10 As a result, 5% lignin exhibited better mechanical performance and a fibrillar morphology. At high lignin contents, the peak positions were 3347 cm−1 for 20% lignin, 3342 cm−1 for 30% lignin, and 3360 cm−1 for 50% lignin. Those wavenumbers are associated with less intermolecular bonding between lignin and PVA. The lignin orientation within composite fibers was also studied to understand its structure in the fibers.

orientation parameters R and f were determined for both peaks for all of the lignin/PVA fibers (Table 4). R and f were the Table 4. Orientation Parameters for Lignin Embedded in 5− 50% Lignin/PVA Fibers from 15/85 Methanol/Acetone Coagulation Raman anisotropy (R) orientation factor ( f)

benzene ring conjugated CC benzene ring conjugated CC

5%

20%

30%

50%

1.65 1.62 0.12 0.11

1.10 1.16 0.02 0.03

1.45 1.50 0.08 0.09

1.27 1.33 0.05 0.06

greatest for 5% lignin fiber, although its anisotropy value was very low (R ∼ 2). The amorphous structure of lignin impedes its axial alignment; nevertheless, our results suggest that interactions between PVA and lignin cause lignin segments to align along the fiber axis. Lignin aggregation at higher weight fractions decreased the interactions between the biopolymer and PVA. Nevertheless, highly drawn fibers of fibrillar PVA resulted from the plasticizing behavior of lignin and the interactions between aligned PVA and lignin segments. The 2956

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering

Figure 10. Confocal micrographs of (a) PVA fibers from 100/0 methanol/acetone coagulation and (b) 5% and (c) 50% lignin-based fibers from 15/ 85 methanol/acetone coagulation after water immersion at 5, 85, and 100 °C.

values of R and f were the lowest for 20% lignin, but these increased slightly at 30−50% lignin in fiber. Moisture Resistance of Lignin/PVA Fibers. Carbon nanotube (CNT) fillers are known to enhance the solvent resistance of PVA composite fibers against DMSO.32 In this study, lignin was found to enhance the thermal resistance of composite PVA fibers to dissolution in water as well as its swelling behavior at room temperature. Lignin/PVA fiber structures were implied from these studies of their susceptibility to water dissolution at elevated temperatures (Figure 10). The dissolution of neat PVA, 5% lignin, and 50% lignin composite fibers was tested. All three fibers remained intact at room temperature (25 °C). In 85 °C water, the neat PVA fiber dissolved, whereas the lignin-based fibers exhibited diameter increases from swelling. At 5% lignin, the fiber partially dissolved in boiling water. Fibrils of lignin/PVA were suspended in water, leaving behind a fine fiber, as shown in Figure 10b3. Hydrogen bonding between lignin and PVA (as previously determined from IR spectra in Figure 8) stymied fiber dissolution. Fibrils of lignin/PVA remained after boiling in water. Their presence at elevated temperatures suggests that the PVA interfacing lignin is highly crystalline. Lignin located in the polymer-rich domains of the gel fiber (Figure 3a2) later causes the crystalline regions of the fibrillar polymer to resist dissolution at elevated temperatures, while the amorphous polymer dissolves in boiling water. At 100 °C, 50% lignin fiber behaved like a swollen gel (Figure 10c3). Its structure remained mostly intact. Intermolecular interactions between lignin and PVA cause the fibers to resist dissolution. Also, water-insoluble aggregates of lignin can hinder the dissolution of crystalline and amorphous polymer. The results of immersing lignin/PVA fibers in roomtemperature water are presented in Table 5. Increasing the

Table 5. Swelling (in %) of Lignin/PVA Fibers after 24 h of Water Immersiona neat PVA

5% lignin

30% lignin

14

19

82

Neat PVA fiber was fabricated from 100/0 methanol/acetone coagulation. Fibers having 5% and 30% lignin were obtained from 15/85 methanol/acetone coagulation. a

lignin content influenced the moisture swelling within gel-spun lignin/PVA fibers. Although lignin inhibits the thermal dissolution of PVA, it permits PVA swelling in confined regions. This behavior suggests that the structure of composite fibers has water-insoluble lignin that is dispersed throughout the PVA and in intimate contact with the matrix polymer.



CONCLUSION We have successfully gel-spun biobased lignin/PVA fibers from coagulation baths of 15/85 methanol/acetone that show promising results of good mechanical performance and water resistance. Even at 50% lignin, the mechanical strength and Young’s modulus of our biobased fibers were competitive with those of commercial PVA staple fiber. In summary, 5% lignin fiber possessed the highest value of mechanical strength. The structural enhancement at 5% lignin was evidenced by its high index of crystallinity, intermolecular bonding between lignin and PVA, and mild alignment of lignin upon fiber drawing. At 50% lignin, fibers exhibited improved toughness and resistance to water dissolution in comparison to neat fibers. The swelling results suggest that water is capable of penetrating the hydrophilic regions of PVA but that lignin kept the fibers’ structure loosely intact. Although lignin is an amorphous, three-dimensional biopolymer, this work has shown evidence of enhanced 2957

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering strength, which is generally observed with highly ordered fillers, such as one-dimensional CNTs32 and two-dimensional graphene oxide.33 For instance, 0.3 wt % single-walled CNT/ PVA34 and 0.1 wt % reduced graphene oxide-reinforced PVA35 fibers had similar values of mechanical performance: tensile strength of 2.2 GPa and Young’s modulus of 36 GPa. At 5% lignin, a similar Young’s modulus was obtained, although its tensile strength was half as much. For carbon-filler-reinforced PVA, the draw ratio of gel-spun fibers decreased from 13 for neat PVA to 11 for composites of PVA with 0.5 wt % reduced graphene oxide.36 Performance enhancements among gel-spun lignin/PVA fibers were thus attributed to high draw ratios (due to lignin plasticization) and molecular adhesion between PVA and lignin’s rigid, hydrophobic structure. At up to 50% lignin, the mechanical performance of gel-spun lignin/PVA composites indicates their potential for use in industrial and highperformance applications. Thus, gel spinning is an attractive alternative to melt spinning for low-cost lignin fiber production.



(7) Wang, S.; Li, Y.; Xiang, H.; Zhou, Z.; Chang, T.; Zhu, M. Low Cost Carbon Fibers from Bio-renewable Lignin/Poly(lactic acid) (PLA) Blends. Compos. Sci. Technol. 2015, 119, 20−25. (8) Kubo, S.; Kadla, J. Lignin-based Carbon Fibers: Effect of Synthetic Polymer Blending on Fiber Properties. J. Polym. Environ. 2005, 13, 97−105. (9) Oroumei, A.; Fox, B.; Naebe, M. Thermal and Rheological Characteristics of Biobased Carbon Fiber Precursor Derived from Low Molecular Weight Organosolv Lignin. ACS Sustainable Chem. Eng. 2015, 3, 758−769. (10) 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. (11) Xu, G.; Ren, S.; Wang, D.; Sun, L.; Fang, G. Fabrication and Properties of Alkaline Lignin/Poly(vinyl alcohol) Blend Membranes. BioResources 2013, 8, 2510−2520. (12) Su, L.; Xing, Z.; Wang, D.; Xu, G.; Ren, S.; Fang, G. Mechanical Properties Research and Structural Characterization of Alkali Lignin/ Poly(Vinyl Alcohol) Reaction Films. BioResources 2013, 8, 3532−3543. (13) Handbook of Fiber Chemistry; Lewin, M., Ed.; CRC Press: Boca Raton, FL, 2007. (14) Kunugi, T.; Kawasumi, T.; Ito, T. Preparation of Ultra-High Modulus Polyvinyl Alcohol Fibers by the Zone-Drawing Method. J. Appl. Polym. Sci. 1990, 40, 2101−2112. (15) Cha, W. I.; Hyon, S. H.; Ikada, Y. Gel spinning of Poly (vinyl alcohol) from Dimethyl Sulfoxide/Water Mixture. J. Polym. Sci., Part B: Polym. Phys. 1994, 32, 297−304. (16) Yamaura, K.; Tanigami, T.; Hayashi, N.; Kosuda, K. I.; Okuda, S.; Takemura, Y.; Itok, M.; Matsuzawa, S. Preparation of High Modulus Poly(Vinyl Alcohol) by Drawing. J. Appl. Polym. Sci. 1990, 40, 905−916. (17) Ryan, C. F.; Fleischer, P. C., Jr. The Gel Melting Point as a Measure of the Tacticity of Poly(Methyl Methacrylate). J. Phys. Chem. 1965, 69, 3384−3400. (18) Luo, J.; Li, Q.; Zhao, T.; Gao, S.; Sun, S. Bonding and Toughness Properties of PVA Fibre Reinforced Aqueous Epoxy Resin Cement Repair Mortar. Constr Build Mater. 2013, 49, 766−771. (19) Chemical Modification, Properties, and Usage of Lignin; Hu, T. Q., Ed.; Springer: New York, 2002. (20) Song, P. a.; Xu, Z.; Guo, Q. Bioinspired Strategy to Reinforce PVA with Improved Toughness and Thermal Properties via Hydrogen-Bond Self-Assembly. ACS Macro Lett. 2013, 2, 1100−1104. (21) Tadokoro, H.; Kǒzai, K.; Seki, S.; Nitta, I. On the Crystalline Band in the Infrared Absorption Spectrum of Polyvinyl Alcohol. J. Polym. Sci. 1957, 26, 379−382. (22) Mallapragada, S. K.; Peppas, N. A. Dissolution Mechanism of Semicrystalline Poly(Vinyl Alcohol) in Water. J. Polym. Sci., Part B: Polym. Phys. 1996, 34, 1339−1346. (23) Tretinnikov, O.; Zagorskaya, S. Determination of the Degree of Crystallinity of Poly(Vinyl Alcohol) by FTIR Spectroscopy. J. Appl. Spectrosc. 2012, 79, 521−526. (24) Peppas, N. A. Infrared Spectroscopy of Semicrystalline Poly(Vinyl Alcohol) Networks. Makromol. Chem. 1977, 178, 595−601. (25) Takahashi, N.; Kanaya, T.; Nishida, K.; Kaji, K. GelationInduced Phase Separation of Poly(Vinyl Alcohol) in Mixed Solvents of Dimethyl Sulfoxide and Water. Macromolecules 2007, 40, 8750−8755. (26) Baumberger, S.; Lapierre, C.; Monties, B.; Lourdin, D.; Colonna, P. Preparation and Properties of Thermally Moulded and Cast Lignosulfonates-Starch Blends. Ind. Crops Prod. 1997, 6, 253− 258. (27) Sadeghifar, H.; Wells, T.; Le, R. K.; Sadeghifar, F.; Yuan, J. S.; Jonas Ragauskas, A. Fractionation of Organosolv Lignin Using Acetone:Water and Properties of the Obtained Fractions. ACS Sustainable Chem. Eng. 2017, 5, 580−587. (28) Shin, M. K.; Lee, B.; Kim, S. H.; Lee, J. A.; Spinks, G. M.; Gambhir, S.; Wallace, G. G.; Kozlov, M. E.; Baughman, R. H.; Kim, S. J. Synergistic Toughening of Composite Fibres by Self-Alignment of Reduced Graphene Oxide and Carbon Nanotubes. Nat. Commun. 2012, 3, 650.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acssuschemeng.6b02423. Gel melting point measurement (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +1-919-515-8145. ORCID

Ericka Ford: 0000-0002-7172-3105 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The Nonwovens Institute (Project 14-176 NC) supported this research. This work was performed in part at the Analytical Instrumentation Facilities (AIF) at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation (Award ECCS-1542015). The AIF is a member of RTNN, a site in the National Nanotechnology Coordinated Infrastructure (NNCI). Pure Lignin Environmental Technology Ltd donated the lignin used in this study. LaTasha Nicholson is thanked for her experimental assistance with gel melting studies, and Judy Elson is thanked for her expertise with confocal microscopy.



REFERENCES

(1) Lignins: Occurrence, Formation, Structure and Reactions; Sarkanen, K. V., Ludwig, C. H., Eds.; Wiley-Interscience: New York, 1971. (2) Sjöström, E. Wood Chemistry: Fundamentals and Applications; Academic Press: San Diego, CA, 1993. (3) Alexy, P.; Košíková, B.; Podstránska, G. The Effect of Blending Lignin with Polyethylene and Polypropylene on Physical Properties. Polymer 2000, 41, 4901−4908. (4) Luo, J. Lignin-Based Carbon Fiber. M.S. Thesis, The University of Maine, Orono, ME, 2010. (5) Kadla, J.; Kubo, S.; Venditti, R.; Gilbert, R.; Compere, A.; Griffith, W. Lignin-based Carbon Fibers for Composite Fiber Applications. Carbon 2002, 40, 2913−2920. (6) 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. 2958

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959

Research Article

ACS Sustainable Chemistry & Engineering (29) Liem, H. M.; Etchegoin, P.; Whitehead, K. S.; Bradley, D. D. Raman Anisotropy Measurements: An Effective Probe of Molecular Orientation in Conjugated Polymer Thin Films. Adv. Funct. Mater. 2003, 13, 66−72. (30) White, J. L.; Spruiell, J. E. The Specification of Orientation and its Development in Polymer Processing. Polym. Eng. Sci. 1983, 23, 247−256. (31) Gierlinger, N.; Keplinger, T.; Harrington, M. Imaging of Plant Cell Walls by Confocal Raman Microscopy. Nat. Protoc. 2012, 7, 1694−1708. (32) Minus, M. L. The Study of Crystallization and Interfacial Morphology in Polymer/Carbon Nanotube Composites. Ph.D. Dissertation, Georgia Institute of Technology, Atlanta, GA, 2008. (33) Morimune, S.; Nishino, T.; Goto, T. Poly(Vinyl Alcohol)/ Graphene Oxide Nanocomposites Prepared by a Simple Eco-Process. Polym. J. 2012, 44, 1056−1063. (34) Xu, X.; Uddin, A. J.; Aoki, K.; Gotoh, Y.; Saito, T.; Yumura, M. Fabrication of High Strength PVA/SWCNT Composite Fibers by Gel Spinning. Carbon 2010, 48, 1977−1984. (35) Li, J.; Shao, L.; Zhou, X.; Wang, Y. Fabrication of High Strength PVA/rGO Composite Fibers by Gel Spinning. RSC Adv. 2014, 4, 43612−43618. (36) Abdou, J. P.; Reynolds, K. J.; Pfau, M. R.; van Staden, J.; Braggin, G. A.; Tajaddod, N.; Minus, M.; Reguero, V.; Vilatela, J. J.; Zhang, S. Interfacial Crystallization of Isotactic Polypropylene Surrounding Macroscopic Carbon Nanotube and Graphene Fibers. Polymer 2016, 91, 136−145.

2959

DOI: 10.1021/acssuschemeng.6b02423 ACS Sustainable Chem. Eng. 2017, 5, 2949−2959