Electrospun Poly(lactic acid) - ACS Publications - American Chemical ...

Article Cite This: Biomacromolecules XXXX, XXX, XXX−XXX

pubs.acs.org/Biomac

Electrospun Poly(lactic acid)-Based Fibrous Nanocomposite Reinforced by Cellulose Nanocrystals: Impact of Fiber Uniaxial Alignment on Microstructure and Mechanical Properties Siqi Huan,† Guoxiang Liu,† Wanli Cheng, Guangping Han,* and Long Bai* Key Laboratory of Bio-based Material Science and Technology (Ministry of Education), Northeast Forestry University, Harbin 150040, P R China S Supporting Information *

ABSTRACT: Uniform poly(lactic acid)/cellulose nanocrystal (PLA/CNC) fibrous mats composed of either random or aligned fibers reinforced with up to 20 wt % CNCs were successfully produced by two different electrospinning processes. Various concentrations of CNCs could be stably dispersed in PLA solution prior to fiber manufacture. The microstructure of produced fibrous mats, regardless of random or aligned orientation, was transformed from smooth to nanoporous surface by changing CNC loading levels. Aligning process through secondary stretching during high-speed collection can also affect the porous structure of fibers. With the same CNC loading, fibrous mats produced with aligned fibers had higher degree of crystallinity than that of fibers with random structure. The thermal properties and mechanical performances of PLA/ CNC fibrous mats can be enhanced, showing better enhancement effect of aligned fibrous structure. This results from a synergistic effect of the increased crystallinity of fibers, the efficient stress transfer from PLA to CNCs, and the ordered arrangement of electrospun fibers in the mats. This research paves a way for developing an electrospinning system that can manufacture high-performance CNC-enhanced PLA fibrous nanocomposites. mechanically robust nanoscale fillers, for example, bioactive glass,10 into electrospun PLA matrices therefore is a promising approach to fabricate high-performance PLA fibrous nanocomposites. Because PLA is an environmentally friendly material, it is important from the perspective of sustainability to select suitable additives, resulting in the fact that bio- or green-nanocomposite of PLA-based electrospun fibers made from renewable or naturally derived nanofillers is in great necessity. Recently, one strong trend is to utilize nanocellulose acting as green nanofillers to improve the properties and versatility of polymer-based nanocomposites achieving a combination of desired properties and environmental benefits.11−14 The biocompatible and biodegradable cellulose nanocrystals (CNCs) that are readily isolated via controlled acid hydrolysis of cellulosic raw materials are short crystalline nanorods,15 showing a novel enhancement effect. The driving force to explore CNCs as reinforcing agents in electrospinning technique therefore is their large specific surface area, low density, high surface charge, and the ability to enhance properties at low loading levels as well as adaptability to processing conditions.16 In particular, CNCs are usually used to reinforce the mechanical properties of electrospun polymer materials due to their high Young’s modulus (∼138 GPa)17 and high mechanical strength (∼7 GPa)18 obtained from the

1. INTRODUCTION Poly(lactic acid), PLA, a biodegradable thermoplastic polyester derived from sustainable resources,1 has been considered to be one of the most attractive biopolymers due to its renewability, biodegradability, biocompatibility, and superior physical properties.2 Furthermore, hydrophobic PLA naturally displays higher hydrophilicity than common hydrophobic thermoplastic polymers due to a better access between water molecules and the polar oxygen linkages of its backbone,3 providing a promising opportunity to form biograde products. PLA-based materials can be fabricated by a number of commercially technologies including solvent casting, particulate leaching, membrane lamination, and melt molding as well as electrospinning and precise extrusion.4 Among these techniques, consecutive electrospinning to manufacture PLA-based microor nanofibers attracts considerable attention due to its simplicity, high efficiency, and desirable microstructure (e.g., porosity).5 However, electrospun pure PLA fibers normally have the drawbacks of showing weak mechanical properties and low thermal stability,6 narrowing their industrial applications. Enhancing the mechanical performances of electrospun PLA nanofibers, therefore, is highly desired for consumer applications, especially for PLA-based fibers that have to meet controllable mechanical requirements during transportation, reprocessing, and recovery. Fabrication of nanocomposites that occur on the nanoscale has been a facile method to develop or modify novel structural and functional heterogeneous materials.7−9 Incorporating © XXXX American Chemical Society

Received: January 5, 2018 Revised: February 11, 2018 Published: February 14, 2018 A

DOI: 10.1021/acs.biomac.8b00023 Biomacromolecules XXXX, XXX, XXX−XXX

Article

Biomacromolecules

Figure 1. Schematic drawing (not to scale) of (a) preparation of PLA/CNC as-spun solutions and electrospinning process of PLA/CNC fibrous mats with (b) random and (c) aligned fiber arrangement.

fiber alignment processes during electrospinning to controllably tune the surface microstructure (nanopores) of produced fibers. In this study, a series of PLA/CNC fibrous nanocomposite with random and aligned fibers were fabricated via an electrospinning process using PLA solutions with varying CNC loading levels (Figure 1). The main objective of this study was to investigate the influence of CNC loading levels and fiber alignment on the surface morphology and properties of electrospun PLA/CNC fibers. Surface microstructure of electrospun fibers was explored by atomic force microscopy (AFM) and scanning electron microscopy (SEM). Performance evaluations of PLA/CNC fibrous nanocomposite mats, especially mechanical and thermal properties, were undertaken to elucidate the structure−property relationship. The long-term perspective in this research is to develop a biograde electrospinning system to manufacture fully biocompatible, biodegradable, and environmentally friendly high-performance fibrous nanocomposites for potential biological applications.

densely and orderly crystallized structure after acid hydrolysis.19 Accordingly, CNCs are one of the most promising green nanofillers to modify the properties of electrospun PLA fibers. The effective enhancement of incorporating CNCs into hydrophobic polymer-based electrospun nanofibers, including polystyrene,20 poly(methyl methacrylate),21,22 and polymer mixture containing PLA,23 has been demonstrated; however, full discovery of the unique contribution of CNCs to controllably design and tailor-make microstructure and performances of nanocomposites under electrospinning process remains to be a challenge. In electrospun CNC-based nanocomposites, the local orientation of CNCs in fibers has been systematically reported,24,25 which can be considered as the main reason for significant modification of mechanical performance because efficient stress transfer within electrospun fibers and CNCs is expected only as CNCs are oriented along the fiber axis.26 Furthermore, it has also been verified that the alignment of electrospun fibers could further increase their performances and endow novel properties compared with those of randomly oriented ones.27 Chen et al. reported that the alignment of electrospun protein fibers reinforced with cellulose nanowhiskers greatly improved their mechanical properties,28 demonstrating the function of fiber alignment in final performance. Consequently, a combination of the enhancement effect of CNCs and the formation of uniaxially ordered fiber arrangement can offer a synergetic function to create highperformance electrospun PLA/CNC fibrous nanocomposite. However, although alignment of electrospun polymers functionalized with CNCs has been studied recently,29 few studies have investigated the influence of CNCs loading levels on the formation of aligned PLA/CNC fibers. Moreover, the structure−property relationship of uniaxially aligned electrospun PLA/CNC fibrous nanocomposites, particularly the effects of fiber alignment on their mechanical properties, have not been fully investigated and understood. More importantly, to our knowledge, this is the first report to combine CNCs and

2. EXPERIMENTAL SECTION 2.1. Materials. PLA particles (Mw = 270 000) were purchased from NatureWorks (USA) and dried in a vacuum oven at 60 °C for 24 h prior to use. Commercial microcrystalline cellulose (MCC, KY100S, 75% moisture, Daicel, Japan) was used as raw material for producing (CNCs. 98% sulfuric acid was purchased from Aladdin (Shanghai, China). N,N-Dimethylformamide (DMF) and chloroform (CHCl3) (Kermel, Tianjin, China) were of analytical grade and used as received without further purification. Deionized water was used throughout the study. 2.2. Preparation of Cellulose Nanocrystals. CNCs were prepared via sulfuric acid hydrolysis of MCC according to our previous study.8 In brief, 170 mL of 64 wt % sulfuric acid aqueous solution was heated to 45 °C in a water bath before adding 40 g of MCC under vigorous magnetic stirring. After 1 h, the hydrolysis process was stopped by adding additional water (10 times). The resulting suspension was then cooled to ambient temperature and washed with deionized water by successive centrifugations (10 000 rpm, 15 min) until neutral pH. Dialysis against deionized water was performed for 7 days to remove free acid molecules from the suspension. After dialysis, the yield was calculated by withdrawing a B


Article

Biomacromolecules

The neat PLA and PLA/CNC fibrous mats were analyzed using differential scanning calorimetry (DSC-204, Netzsch, Germany) to study the phase transitions and crystalline behavior. Samples of ∼5 mg were loaded into aluminum crucibles sealed using a crimping tool. An empty aluminum crucible served as a comparison sample. All DSC measurements were performed with respect to the signal baseline obtained for the two empty crucibles. The samples were heated from room temperature to 200 °C, held for 2 min to eliminate the thermal history, and then cooled to room temperature and finally heated to 200 °C. All of the processes were conducted at 10 °C/min heating rate in a 50 mL/min dynamic N2 atmosphere. The accuracy of the temperature measurement was 0.1 °C, and the accuracy of the enthalpy measurement for phase transitions was 0.2 J/g. All tests were carried out in duplicate. The crystallinity degree of neat PLA and PLA/ CNC fibers was calculated using the following equation31

known, small amount of the sample and obtaining its oven-dried weight. The pelleted CNCs obtained after dialysis were dried using a freeze-dryer (Scientz-10N, Xin Zhi, China). The shape and microstructure of produced CNCs were characterized and analyzed by AFM with scanning area of 3 μm × 3 μm. 2.3. Preparation of As-Spun Solutions. Spinning solutions were prepared from CNCs and a solution mixture of PLA, DMF, and CHCl3. After comparing the morphologies by electrospinning different concentrations of PLA solution without adding CNCs (shown in Figure S1), 10 wt % PLA was chosen to investigate the influence of fiber alignment during electrospinning on microstructure and properties of CNC/PLA fibrous nanocomposites. In brief, a certain amount of dry PLA particles was dissolved in the solvent mixture of CHCl3 and DMF (3:1), and the obtained solution was vigorously stirred in 65 °C water bath for 2 h to obtain 10 wt % PLA solution. Then, varied amounts of CNC powder produced as above were loaded into PLA solutions, and the mixtures were sonicated with high amplitude for 20 min to obtain final as-spun solutions. All of the asspun solutions were freshly made shortly before electrospinning experiments. The CNC loading concentrations in relation to the adding weight of PLA were 0, 5, 10, 15, and 20 wt %. The samples were designed as X-PLA/CNC-y, where X represents the random (R) or aligned (A) fiber arrangement and y corresponds to CNC loading level. The conductivity, viscosity, and surface tension of prepared PLA/ CNC as-spun solutions were characterized at ambient temperature using a conductivity meter (DDSJ-318, Lei Ci, Shanghai, China), digital rotational viscometer (SNB-1, Heng Ping, Shanghai, China), and a surface tension meter (JK99B, Zhong Chen, Shanghai, China), respectively. 2.4. Electrospinning Setup. The setup of electrospinning apparatus was purchased from Yong Kang Le Ye Company, China, as described in our previous research.20 The as-spun solutions were loaded into a 10 mL plastic, disposable syringe (Zhi Yu, Shanghai) with a stainless-steel needle (i.d.: 0.6 mm). The positive voltage generator applying a potential of 16 kV was fixed on the needle, while the negative electrode applying −3 kV potential was connected with the collector to collect the electrospun fibrous mats. For the collection of nonaligned fibers, a rectangular piece of aluminum foil (240 mm length and 150 mm width) was covered on the cylinder with rotating speed of 80 rpm. The aligned fibrous nanocomposites were collected on a smaller cylinder (480 mm length and 50 mm width) with ultrahigh rotating speed of ∼2800 rpm. The flow rate of the as-spun solutions was automatically controlled by a syringe pump. The temperature and humidity in the chamber during electrospinning were controlled at 25 °C and 22%, respectively. The obtained fibrous mats were carefully detached from the aluminum foil and stored at ambient for 24 h prior to characterization. 2.5. Characterization. The surface as well as the cross-section of PLA/CNC fibers were observed by SEM (QUANTA-200, FEI, Hillsboro, OR). For the surface morphology characterization, fibrous mats collected on the aluminum foil were cut into small pieces and then were mounted on metal stubs with double-sided adhesive tape. For the cross-section characterization, fibrous mats with aligned morphology were immersed in epoxy resin with the addition of initiator, dried in an oven under 100 °C, and then cut by a glass knife. Gold−palladium was coated on both the samples of surface and crosssection before observation at an accelerating voltage of 12.5 kV. The diameter of the electrospun fibers was analyzed from SEM images by ImageJ.30 The mean diameters, diameter distribution, and surface pore size of the fibers were calculated based on at least 100 fibers from the corresponding SEM micrographs. Surface topological structure in the bulk of PLA/CNC fibrous mats was determined via using AFM. Images were obtained from a commercial instrument (Dimension Icon, Bruker, Billerica, MA) under tapping mode at a scan rate of 0.999 Hz. To obtain typical images for the sample, scanning was made over a large area. After selecting the typical area, higher magnified images were obtained in the area of 5 μm × 5 μm. The digital resolution of all pictures was 256 × 256 points. Roughness data were analyzed by NanoScope software.

χc =

ΔHm − ΔHcc ωΔHm0

× 100%

where ω is the weight fraction of PLA in fibrous mats, ΔHm (J/g) is the heat of fusion from the second heating circle, ΔHcc (J/g) is the heat of cold crystallization, and ΔH0m is the heat of fusion of 100% crystalline PLA, which has a value of 93.6 J/g.32 Thermogravimetric analysis (TGA, TGA-209, Netzsch, Germany) was conducted to study thermal decomposition of PLA/CNC fibrous mats. Samples of 5−10 mg were heated from room temperature to 700 °C at a rate of 5 °C/min under an Ar atmosphere. The onset decomposition temperature and the maximum thermal decomposition temperature of samples were defined as Tonset and Tmax, respectively. The surface wettability of PLA/CNC fibrous mats was characterized by a contact angle meter (OCA20, Dataphysics, Bad Vilbel, Germany) at ambient temperature. Contact angle (CA) was measured using a sessile drop method with droplet volume being 5 μL. The CA values of the right side and the left side of the water droplet were both measured and averaged. All CA data were an average of at least five measurements at different locations on the surface. Mechanical properties of PLA/CNC fibrous mats were determined from stress−strain curves of tensile test. The tensile test was carried out on a model 3365 universal testing machine (Instron, Norwood, MA) with a tensile rate of 10 mm/min according to the ASTM D 88209 at ambient temperature and 30% humidity. The size of each sample was 15 mm length and 5 mm width. All tests were carried out at least in triplicate. Notably, the tensile for fibrous mats with aligned fibers was performed along the fiber direction.

3. RESULTS AND DISCUSSION 3.1. Characteristics of As-Spun Solutions. The microstructure of CNCs prepared by acid hydrolysis was observed by AFM (Figure 2). As shown in Figure 2, the pristine CNCs showed homogeneous morphology of rod-shaped, welldispersed nanoparticles, which can be attributed to the electrostatic repulsion from the negative sulfate half-ester groups introduced from acid hydrolysis process.33 Individual CNCs with negligibly lateral or longitudinal association were

Figure 2. AFM (a) height (19.4 nm light to dark) and (b) phase (15° from light to dark) topographic images of pristine CNCs. The scanning area was 3 μm × 3 μm. C


Article

Biomacromolecules also clearly identified, ensuring a prerequisite for fine enhancement effect. From AFM images, the average dimensions of CNCs were approximately 260 ± 32 nm in length and 42 ± 18 nm in width, indicating that the corresponding aspect ratio (L/W) was ∼7. The obtained CNCs were further freeze-dried and redispersed in 10 wt % PLA solution to prepare as-spun solutions. A photograph of asspun solutions with different CNC loading levels is shown in Figure S2. With increasing CNC amount, the transparency of prepared PLA/CNC solutions was gradually decreased, showing the highest turbidity at 20 wt % CNCs. On the contrary, the as-spun solution containing 20 wt % CNCs was still homogeneous without any sedimentation before and during electrospinning, which may be beneficial from the strong electrostatic repulsion of CNCs and the high-energy sonication process. Furthermore, no nozzle blockage for spinning solution with 20 wt % CNCs during electrospinning (Section 3.2) was observed, directly demonstrating the good dispersion of CNCs in PLA solution. Because physicochemical parameters of as-spun solution play crucial roles in determining the properties of electrospun fibers, viscosity, surface tension, and conductivity of as-spun solutions containing various CNC concentrations were measured (Table 1). The viscosity significantly increased with increasing CNC

density could increase the electrical force imposed on the ejected jets. Taken together, despite combining hydrophobic PLA and hydrophilic CNCs in single spinning solution system, the prepared PLA/CNC as-spun solutions display fine electrospinnability. 3.2. Fiber Morphology. The morphology of electrospun PLA/CNC fibrous mats with random or aligned fibers loading with various amounts of CNCs is shown in Figure 3. As shown in Figure 3R1−R5, all fibers produced showed similar diameter of ∼2 μm, and obvious beads or bead-on-string structure within single fibers were scarcely visible. In Figure 3R2, PLA/CNC fibers containing 5 wt % CNCs showed smooth surface but less uniform as fibers without adding CNCs. Because the solvent (DMF/CHCl3) used was identical for all formulations in current study, it can be concluded that the main change of fiber morphology was caused by varied CNC loading levels. Because the viscosity and conductivity of spinning solutions increased with loading more CNCs while surface tension was kept similar, the change of fiber morphology can be attributed to the fact that the higher conductivity resulting from incorporating CNCs could increase the electrical force applied on the ejected fluid, which can counteract the viscoelastic forces more efficiently, thereby stretching the PLA/CNC jets to a greater extent during electrospinning.34 Meanwhile, the higher viscosity of spinning solutions could provide sufficient resistance to reduce rapid shape transformation and recovery, leading to a steady electrospinning process. Increasing CNC concentrations, the surface of produced fibers showing random arrangement becomes porous, with pore size being ∼90 nm (Figure 3R3−R5 and Figure S3). The formation of porous structure on fiber surface may be caused by the poor compatibility between hydrophilic CNC nanorods and hydrophobic PLA chains at interfaces. CNCs that were well dispersed in as-spun solutions may weaken the interconnection of PLA chains, transforming this effect from interior to surface with increasing addition of CNCs. On the contrary, the differences of solvent evaporation rate of DMF and CHCl3 can also promote the formation of porous structure. Because the boiling point of DMF (153 °C) is much higher than that of CHCl3 (61 °C), PLA dissolved in solvent mixture can form a main fiber network after the rapid evaporation of CHCl3 as soon as spinning droplets were ejected. Meanwhile, DMF may be trapped inside fibers transitorily and afterward evaporated during electrospinning, thereby inducing a secondary structure formation on the fiber surface. Taken together, it can be reasonably assumed that the formation of nanopores on fiber

Table 1. Characteristics of As-Spun Solutions with Varying Amounts of CNCs code

viscosity (mPa·s)

surface tension (mN/m)

conductivity (F/m)

PLA PLA/CNC-5 PLA/CNC-10 PLA/CNC-15 PLA/CNC-20

385.3 399.5 444.3 484.8 511.6

32.804 33.625 33.912 34.212 34.993

1.458 1.794 2.377 2.734 3.176

loading levels, possibly attributed to the increased entanglement between CNCs and PLA chains that originated from their interconnection during sonication. The surface tension of asspun solutions was similar, showing slightly increase compared with that of pure PLA solution. Conductivity was also observed to increase with loading more CNCs into PLA solutions. This can be ascribed to the sulfate half-ester groups on CNC surface. The increased conductivity can provide a positive effect on the continuous spinning of smaller diameter and more homogeneous fibers in PLA/CNC systems because the high charge

Figure 3. SEM images of electrospun PLA/CNC fibrous mats with random (R1−R5) and aligned (A1−A5) fiber arrangement. The loading level of CNCs was 0, 5, 10, 15, and 20 wt % from 1 to 5. The scale bar was 2 μm. D


Article

Biomacromolecules

Figure 4. AFM images of electrospun PLA/CNC fibrous mats with (a) random fibers, R-PLA (R1) and R-PLA/CNC-20 (R5), and aligned fibers, APLA/CNC-20 (A5). (b,c) Mean roughness measurements and corresponding values of PLA/CNC fibers in panel a. The red line in height images indicates the measuring position of panel b. The scanning area was 5 μm × 5 μm.

deduced that the anomalistic increase in fiber diameter for higher CNC concentrations originated from the widened fiber lateral dimension by forming nanopores. In summary, the microstructure of electrospun PLA/CNC fibers, that is, random versus aligned morphology and nonporous versus hierarchical multiporous, can be controllably manufactured by changing CNC additions and fiber arrangement, demonstrating a finely tunable electrospinning system to produce fibrous biobased nanocomposites. Further confirmation of electrospun PLA/CNC fibers with multiple microstructures was revealed by AFM (Figure 4). From the phase image (Figure 4a), it can be clearly found that nanopores appeared regularly on the surface of produced fibers for both random (R5) and aligned (A5) structure, while electrospun fibers without adding CNCs (R1) were nonporous. These results again verified the generation of nanopores on PLA/CNC fibers during electrospinning. Notably, as observed from Figure 4a, no significant differences in surface roughness for electrospun fibers with or without aligned were observed, which might be due to the sensitivity of AFM at relatively large scanning area. However, the roughness measurements and corresponding mean values shown in Figure 4b,c clearly indicated that electrospun fibers in all formulations had rough surfaces but displaying different degree. The mean roughness value of R-PLA (without loading CNCs) collected on lowspeed roller was ∼250 nm, and the roughness of fibers with adding 20 wt % CNCs (R5) increased to 325 nm. After aligning the fibers during electrospinning, the mean roughness value of A5 decreased to ∼290 nm. We speculated that the variation of roughness for electrospun PLA/CNC fibers was caused by a combination effect of component compatibility, nanoporous structure, and secondary stretching process during collecting fibers on the high-speed roller. By comparing the roughness values of R1 and R5, it can be seen that incorporating CNCs into PLA matrix via electrospinning could increase the surface roughness of obtained fibers, attributed to the fact that a heterogeneous coarsening process could be induced during solvent evaporation and fiber formation process, which has been shown above. On the contrary, the formation of

surface is controlled by the synergistic effect of CNC addition and solvent evaporation rate. When depositing PLA/CNC fibers on a collector with high spin speed, all formed fibers can be aligned along the rotating collector, leading to a possibility to form fibrous mats with aligned fiber arrangement. The fiber alignment in PLA/CNC fibrous mats was microscopically visualized, and the obtained micrographs are shown in Figure 3A1−A5. As shown in Figure 3A1−A5, the PLA/CNC fibers produced displayed fine oriented arrangement with uniform fiber diameter distribution. Notably, compared with the electrospun PLA/CNC fibers showing random morphology, the diameter of aligned fibers was much smaller and more uniform at the same CNC loading levels, which may be attributed to the fact that the high-speed stretching during fiber orienting process could further extend fiber lengthways and decrease PLA deposit laterally. After aligning, nanopores were also observed on PLA/CNC fibers but only occurred at high CNC loading levels of 15 and 20 wt % (Figure 3A4,A5 and Figure S3). We speculated this discrepancy to be caused by the secondary stretching after fibers were deposited onto the high-speed collector. When CNC addition was 300 °C. Furthermore, the Tmax of R-PLA/ CNC mats displayed a similar variation trend, showing increased value as loading CNCs into PLA matrix. The electrospun PLA fibrous mats containing 5 wt % CNCs presented the highest Tonset and Tmax, indicating the most significant improvement of heat resistance, which can be ascribed to increased crystallinity of electrospun R-PLA/CNC nanocomposites (Table 2).36 From Figure 6d, compared with the TGA curves of R-PLA/CNC mats, those of A-PLA/CNC fibrous mats were fairly similar but showed higher Tonset and Tmax values at the same CNC loading levels, indicating that the fibrous mats with aligned fibers had better thermal stability, possibly attributed to higher crystallinity of A-PLA/CNC fibers than that of R-PLA/CNC. In summary, it can be concluded that electrospinning to produce aligned fibrous structure under identical components exhibits a better effect on modifying the thermal properties of PLA/CNC fibrous mats than that of random fibers; that is,

nanopores in R5 could also be responsible for the higher surface roughness. For electrospun PLA/CNC fiber containing aligned structure, as electrospun fibers were aligned by a highspeed collector, the produced fibers would undergo an additional stretching by the collector, leading to a flattening process to smooth the surface with decreasing the roughness (Figure 4c). As discussed above, at the highest CNC addition, the nanopores can be retained, although they may be affected accordingly, which was also demonstrated by AFM results. To investigate the interior structure of electrospun PLA/ CNC fibers with nanopores, cross-sections of the aligned fibers containing 10 and 20 wt % CNCs were observed by SEM (Figure 5). It should be noticed that because the fibrous mats

Figure 5. SEM images of the cross-section of electrospun aligned PLA/CNC fibers with (a,c) 10 wt % and (b,d) 20 wt % loading level of CNCs.

were immersed into epoxy resin and cured at high temperature, the fusion between fibers and resin could occur; however, individual fibers can still be identified. Furthermore, the wavelike structure in all images was artificially caused by cutting. As shown in Figure 5, the arrangement of produced fibers was well oriented along the same direction, again demonstrating the successful fabrication of aligned fiber by high-speed collecting method. Another interesting feature obtained from Figure 5 was that in the comparison of the cross-section of fibers with (20 wt %) or without (10 wt %) surface nanopores, both electrospun PLA/CNC fibers showed porous structure inside the fibers (Figure 5c,d and Figure S4), irrespective of surface structure of produced fibers. Because there is no difference between electrospun fibers with 15 and 20 wt % CNC loading levels, it can be concluded that the generation of inner pores was a generalized process during electrospinning for aligned PLA fibers with incorporating CNCs, which may be induced by the aforementioned compatible difference between PLA and CNCs and solvent evaporation effect. Therefore, it can be reasonably assumed that the different evaporation rate of DMF and CHCl3 and poor interfacial adhesion of CNC-PLA generate the pores first inside fibers during electrospinning process. As aligning process was performed, when the amount of CNCs was insufficient to support the nanoporous structure during secondary stretching, the nanopores would be merged, but interior pores could be retained. As enough CNCs were incorporated into PLA matrix, the pores distributed both on F


Article

Biomacromolecules

Figure 6. DSC and TGA curves of electrospun PLA and PLA/CNC fibrous mats composed of (a,c) random and (b,d) aligned fiber arrangement with different CNC loading levels. The inserts in panels c and d were corresponding differential thermogravimetry (DTG) curves of samples.

Table 2. Results of Thermal Properties of Electrospun PLA/CNC Fibrous Mats with or without Aligned Fiber Arrangement code

Tonset (°C)

Tmax (°C)

Tg (°C)

Tcc (°C)

Tm (°C)

ΔHm (J/g)

ΔHcc (J/g)

χc (%)

R-PLA R-PLA/CNC-5 R-PLA/CNC-10 R-PLA/CNC-15 R-PLA/CNC-20 A-PLA A-PLA/CNC-5 A-PLA/CNC-10 A-PLA/CNC-15 A-PLA/CNC-20

293.1 310.1 308.8 308.4 306.2 293.3 311.2 310.5 309.2 308.8

348.1 360.7 360.1 359.4 357.2 347.8 361.9 360.2 360.7 358.8

59.8 60.8 60.4 60.3 60.5 60.1 60.7 60.4 60.6 60.2

120.4 122.4 119.3 121.5 118.5 120.8 120.3 121.3 122.1 121.3

164.7 165.3 165.6 166.1 164.3 164.8 165.1 164.8 166.2 165.4

35.2 31.1 27.8 22.6 19.2 36.1 32.2 27.1 23.3 19.1

35.1 24.0 22.8 19.1 16.5 35.9 24.1 21.9 19.6 15.9

0 8.0 5.9 4.3 3.5 0 9.1 6.1 4.5 4.1

CNCs, the CA value of the mat made with random fibers was attained around 150°, while the CA for fibrous mat with aligned fibers was 140°. On the basis of these results, a mechanism combining surface microstructure of fibers and spatial arrangement of electrospun fibers was proposed to interpret the change of surface hydrophobic property of electrospun PLA/ CNC fibrous mats. Surface roughness is a decisive factor in determining the surface hydrophobicity of materials because rough surface can trap more air, whose CA value is considered to be 180°.37 With the addition of CNCs into PLA matrix, the surface roughness could be increased (shown in Figure 4c), thereby leading to an increase in CA values, regardless of random or aligned fibrous mats. Additionally, the structure of electrospun PLA/CNC fibrous mats was turned to be porous after reaching a certain amount of CNCs, further increasing uneven property of fiber surface, which also contributed to the change of CA. Interestingly, compared with R-PLA/CNC-20, A-PLA/CNC-20 showed smaller CA, although they both had identical CNC concentrations and similar surface porous

additional high-speed aligning process of fibers during electrospinning offers a secondary enhancement contribution for controllably tailoring the thermal performance of PLA/CNC nanocomposite. 3.4. Surface Wettability. CA is a convenient indicator to evaluate surface hydrophilic−hydrophobic nature. The surface hydrophobicity of electrospun PLA/CNC fibrous nanocomposite is an important factor for its application; however, the hydrophobicity difference of CNCs and PLA chains and the changeful microstructure of obtained fibers make it difficult to simply predict the surface properties of PLA/CNC mats. Therefore, the CA values of electrospun PLA/CNC fibrous mats with either random or aligned structures at varied loading amount of CNCs were explored (Figure 7). It can be observed that neat PLA fibrous mats containing random and aligned fibers had values of 127 and 123°, respectively. With increasing the addition of CNCs from 5 to 15 wt %, the CA values gradually increased to >130° for both fibrous mats, displaying similar variation trend. Significantly, after loading 20 wt % G


Article

Biomacromolecules

Table 3. Tensile Properties of Electrospun PLA/CNC Fibrous Mats with Different Fiber Arrangement code R-PLA R-PLA/CNC-5 R-PLA/CNC-10 R-PLA/CNC-15 R-PLA/CNC-20 A-PLA A-PLA/CNC-5 A-PLA/CNC-10 A-PLA/CNC-15 A-PLA/CNC-20

σmax (MPa)

εb (%)

Young’s modulus (MPa)

± ± ± ± ± ± ± ± ± ±

87.6 67.4 80.2 81.4 81.8 27.4 32.5 26.8 22.6 18.6

43.8 42.3 70.6 41.1 34.1 62.7 126.8 82.7 68.1 74.6

3.9 3.8 4.7 2.9 1.8 4.5 15.3 10.4 7.6 8.5

0.3 0.2 0.1 0.3 0.1 0.5 1.1 0.6 0.3 0.4

levels, the enhanced tensile properties of PLA/CNC fibrous mats should be attributed to the function of CNCs by increasing crystallinity for composite (Table 2) and the reduced loading force on PLA through efficient stress transfer from PLA to stiffer CNCs.38 Moreover, during the electrospinning process, part of CNCs can be aligned into ordered structure to disperse within PLA matrix by electric field force, which provides larger interfacial adhesion,39 leading to a possibility to achieve accelerated physical aging to transform PLA/CNC fibrous mats to a more stable and dense structure.8 At higher CNC concentrations, decreased tendency of mechanical properties of mats can be likely attributed to the aggregation of CNCs at high loading levels in electrospun fibers, especially on fiber surface, decreasing crystallinity of composites (Table 2), and weakened the cohesion between fibers. On the contrary, the obvious surface and interior pores generated during electrospinning may act as a “defect” that can negatively influence the mechanical performance of mats by forming cracks prematurely during tensile test. For electrospun A-PLA/CNC fibrous mats (Figure 8b and Table 3), the tensile properties have been dramatically improved, with highest σmax (15.3 MPa) and E (126.8 MPa) at 5 wt % CNCs. Increasing CNC contents, the tensile properties of fibrous mats gradually decreased, but all showed higher σmax and E than those of electrospun R-PLA mats. From Table 3, it can also be seen that the properties of A-PLA/CNC fibrous mats followed similar varying trend as R-PLA/CNC, although the values were significantly higher. These results suggested that adding CNCs into PLA through aligning

Figure 7. Contact angle (CA) of electrospun PLA/CNC fibrous mats with either random or aligned fiber arrangement. Corresponding images of water droplet on surface of A-PLA/CNC fibrous mats at certain CNC concentrations were inserted accordingly.

morphology. This is likely attributed to the fact that the porous structure of aligned fibers at 20 wt % CNC addition underwent a secondary stretching process during high-speed collecting, which may reduce the pore density on fiber surface, leading to a decreased surface roughness (Figure 4c). On the contrary, the ordered spatial arrangement of fibers may present a negative effect on the hydrophobicity of fibrous mats because randomly packed fibers could be considered as crisscross morphology, allowing water drops to be squeezed out,20 which leads to a relatively smaller CA for A-PLA/CNC fibrous mats at the same CNC concentrations, especially at 20 wt % CNC loading level. 3.5. Mechanical Property. Typical stress−strain curves of neat PLA and PLA/CNC nanocomposite mats with either random or aligned fibers are presented in Figure 8, and their maximum tensile stress (σmax), the elongation at break (εb), and Young’s modulus (E) are summarized in Table 3. As shown in Figure 8a and Table 3, with increased CNC contents, the σmax and E of electrospun PLA/CNC nanocomposite mats initially increased at low CNC loading levels, showing 4.7 ± 0.1 and 70.6 MPa at 10 wt % CNCs, respectively. Further increasing CNC concentrations, the σmax and E of obtained fibrous mats decreased accordingly. Furthermore, the εb of all R-PLA/CNC fibrous mats was approximately similar. At low CNC loading

Figure 8. Stress−strain curves of electrospun PLA/CNC fibrous mats with (a) random and (b) aligned fiber arrangement. H


Article

Biomacromolecules method during electrospinning could effectively enhance the mechanical performance of fibrous mats. On one hand, because the tensile test was along the fiber direction, the stretching force was also applied along fibers, leading to the fact that stronger stretching force could be exerted on the oriented fibers, thereby enabling much larger stress onto A-PLA/CNC fibrous mats. Furthermore, loading CNCs into PLA matrix also contributed to the improvement of mechanical properties, either by tuning the crystallinity of PLA or transferring the stress from PLA to CNCs. On the contrary, secondary stretching process during high-speed collecting may also promote fiber deformation to a more tight structure, thereby increasing the tensile performance.40,41 Similarly, lower crystallinity of PLA chains resulting from higher CNC additions could also weaken the tensile properties of mats, giving rise to smaller σmax and E. Interestingly, the εb of mats were decreased with increasing CNC addition. This is also likely attributed to the interior and surface porous structure of fibers. Because the surface pores were distributed on the fibers but not all along the fiber direction, it can be reasonably speculated that during stretching the surface pores might be transformed to large cracks, which promoted the rupture of bulk fibers, resulting in shorter elongation at break. In summary, with the more ordered fiber organization in fibrous mats, significant enhancement of tensile properties of A-PLA/CNC fibrous mats is achieved, directly confirming the microstructure of fibers originating from the addition of CNCs, and aligning method during electrospinning can be correlated with their mechanical properties.

friendly and biocompatible high-performance CNC-enhanced fibrous nanocomposites.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biomac.8b00023. Additional materials are provided related to SEM images of electrospun fibrous mats with different PLA concentrations without loading CNCs; visual appearance of the as-spun solutions with varying CNC loading levels; and SEM images of random and aligned fibers as well as cross-section of aligned fibers at higher magnification (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Tel: +86-451-8219-1871. *E-mail: [email protected]. Tel: +358-(0)50-3394495. ORCID

Siqi Huan: 0000-0003-1688-9484 Guangping Han: 0000-0001-5434-4878 Long Bai: 0000-0003-3356-9095 Author Contributions †

S.H. and G.L. authors contributed equally.

Notes

4. CONCLUSIONS A series of CNC-enhanced PLA fibrous nanocomposite mats with both random and aligned fiber arrangement were successfully electrospun by loading different concentrations of CNCs into PLA matrix. Uniform and fine PLA/CNC fibers with expected fiber stacking forms can be produced under experimental conditions due to the synergetic effect of electric conductivity, interfacial tension, and viscosity of spinning solutions. Morphological investigation of obtained fibrous mats indicated that the microstructure of fibers, regardless of random or aligned orientation, was transformed from smooth surface to surface containing nanopores by changing CNC loading levels. Nevertheless, porous surface of fibers can also be influenced by aligning process through secondary stretching during high-speed collection. The incorporation of CNCs into PLA matrix changed the crystallinity of PLA, showing higher crystallinity for aligned PLA/CNC fibers. It was also found that the thermal properties of PLA/CNC fibrous mats, especially for thermal stability of aligned mats, can be also enhanced. The mechanical properties of electrospun PLA/CNC fibrous mats were significantly improved by aligning fibers during electrospinning. This can be ascribed to a combined effect of the increased crystallinity of fibers, the efficient stress transfer from PLA to CNCs, and the ordered arrangement of electrospun fibers in the mats. Moreover, the surface porous structure also showed an impact on the mechanical properties of fibrous mats, particularly for the elongation at break, demonstrating the tunable properties of electrospun fibrous mats by controlling fiber microstructure. This study presented for the first time that combining CNCs and fiber aligning process during electrospinning can finely tune the surface microstructure of produced fibers. Finally, this research opens an avenue for developing an electrospinning system that can manufacture environmentally

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This work was financially supported by the National Natural Science Foundation of China (grant no. 31470580). REFERENCES

(1) Pirani, S.; Abushammala, H.; Hashaikeh, R. Preparation and Characterization of Electrospun PLA/nanocrystalline Cellulose-based Composites. J. Appl. Polym. Sci. 2013, 130, 3345−3354. (2) Scaffaro, R.; Botta, L.; Lopresti, F.; Maio, A.; Sutera, F. Polysaccharide Nanocrystals as Fillers for PLA Based Nanocomposites. Cellulose 2017, 24, 447−478. (3) Jalvo, B.; Mathew, A. P.; Rosal, R. Coaxial Poly(lactic acid) Electrospun Composite Membranes Incorporating Cellulose and Chitin Nanocrystals. J. Membr. Sci. 2017, 544, 261−271. (4) Gupta, B.; Revagade, N.; Hilborn, J. Poly(lactic acid) Fiber: An overview. Prog. Polym. Sci. 2007, 32, 455−482. (5) Zhou, C.; Shi, Q.; Guo, W.; Terrell, L.; Qureshi, A. T.; Hayes, D. J.; Wu, Q. Electrospun Bio-nanocomposite Scaffolds for Bone Tissue Engineering by Cellulose Nanocrystals Reinforcing Maleic Anhydride Grafted PLA. ACS Appl. Mater. Interfaces 2013, 5, 3847−3854. (6) Li, L.; Hashaikeh, R.; Arafat, H. A. Development of Eco-efficient Micro-porous Membranes via Electrospinning and Annealing of Poly(lactic acid). J. Membr. Sci. 2013, 436, 57−67. (7) Paul, D.; Robeson, L. M. Polymer Nanotechnology: Nanocomposites. Polymer 2008, 49, 3187−3204. (8) Huan, S.; Bai, L.; Cheng, W.; Han, G. Manufacture of Electrospun All-aqueous Poly(vinyl alcohol)/cellulose Nanocrystal Composite Nanofibrous Mats with Enhanced Properties Through Controlling Fibers Arrangement and Microstructure. Polymer 2016, 92, 25−35. (9) Bai, L.; Gu, J.; Huan, S.; Li, Z. Aqueous Poly(vinyl acetate)-based Core/shell Emulsion: Synthesis, Morphology, Properties and Application. RSC Adv. 2014, 4, 27363−27380.

I


Article

Biomacromolecules

Prolamin Protein Fibers. ACS Appl. Mater. Interfaces 2014, 6, 1709− 1718. (29) He, X.; Xiao, Q.; Lu, C.; Wang, Y.; Zhang, X.; Zhao, J.; Zhang, W.; Zhang, X.; Deng, Y. Uniaxially Aligned Electrospun All-cellulose Nanocomposite Nanofibers Reinforced with Cellulose Nanocrystals: Scaffold for Tissue Engineering. Biomacromolecules 2014, 15, 618−627. (30) Schneider, C. A.; Rasband, W. S.; Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 2012, 9, 671−675. (31) Frone, A. N.; Berlioz, S.; Chailan, J.-F.; Panaitescu, D. M. Morphology and Thermal Properties of PLA-cellulose Nanofibers Composites. Carbohydr. Polym. 2013, 91, 377−384. (32) Mihai, M.; Huneault, M. A.; Favis, B. D.; Li, H. Extrusion Foaming of Semi-Crystalline PLA and PLA/Thermoplastic Starch Blends. Macromol. Biosci. 2007, 7, 907−920. (33) Kalashnikova, I.; Bizot, H.; Cathala, B.; Capron, I. Modulation of Cellulose Nanocrystals Amphiphilic Properties to Stabilize Oil/water Interface. Biomacromolecules 2012, 13, 267−275. (34) Huan, S.; Liu, G.; Han, G.; Cheng, W.; Fu, Z.; Wu, Q.; Wang, Q. Effect of Experimental Parameters on Morphological, Mechanical and Hydrophobic Properties of Electrospun Polystyrene Fibers. Materials 2015, 8, 2718−2734. (35) Wei, L.; Agarwal, U. P.; Matuana, L.; Sabo, R. C.; Stark, N. M. Performance of High Lignin Content Cellulose Nanocrystals as Filler in Poly(lactic acid). Polymer 2018, 135, 305−313. (36) Kumar, A. P.; Depan, D.; Singh Tomer, N.; Singh, R. P. Nanoscale Particles for Polymer Degradation and Stabilization-trends and Future Perspectives. Prog. Polym. Sci. 2009, 34, 479−515. (37) Cassie, A.; Baxter, S. Wettability of Porous Surfaces. Trans. Faraday Soc. 1944, 40, 546−551. (38) Zhou, C.; Chu, R.; Wu, R.; Wu, Q. Electrospun Polyethylene oxide/cellulose nanocrystal Composite Nanofibrous Mats with Homogeneous and Heterogeneous Microstructures. Biomacromolecules 2011, 12, 2617−2625. (39) Ansari, F.; Salajková, M.; Zhou, Q.; Berglund, L. A. Strong Surface Treatment Effects on Reinforcement Efficiency in Biocomposites Based on Cellulose Nanocrystals in Poly(vinyl acetate) Matrix. Biomacromolecules 2015, 16, 3916−3924. (40) Lundahl, M. J.; Klar, V.; Wang, L.; Ago, M.; Rojas, O. J. Spinning of Cellulose Nanofibrils into Filaments: A Review. Ind. Eng. Chem. Res. 2017, 56, 8−19. (41) Lundahl, M. J.; Cunha, A. G.; Rojo, E.; Papageorgiou, A. C.; Rautkari, L.; Arboleda, J. C.; Rojas, O. J. Strength and Water Interactions of Cellulose I Filaments Wet-Spun from Cellulose Nanofibril Hydrogels. Sci. Rep. 2016, 6, 30695.

(10) Noh, K.-T.; Lee, H.-Y.; Shin, U.-S.; Kim, H.-W. Composite Nanofiber of Bioactive Glass Nanofiller Incorporated Poly(lactic acid) for Bone Regeneration. Mater. Lett. 2010, 64, 802−805. (11) De France, K. J.; Chan, K. J.; Cranston, E. D.; Hoare, T. Enhanced Mechanical Properties in Cellulose Nanocrystal-poly (oligoethylene glycol methacrylate) Injectable Nanocomposite Hydrogels Through Control of Physical and Chemical cross-linking. Biomacromolecules 2016, 17, 649−660. (12) Mittal, N.; Jansson, R.; Widhe, M.; Benselfelt, T.; Håkansson, K. M.; Lundell, F.; Hedhammar, M.; Söderberg, L. D. Ultrastrong and Bioactive Nanostructured Bio-Based Composites. ACS Nano 2017, 11, 5148−5159. (13) Chaabouni, O.; Boufi, S. Cellulose Nanofibrils/polyvinyl acetate Nanocomposite Adhesives with Improved Mechanical Properties. Carbohydr. Polym. 2017, 156, 64−70. (14) Huan, S.; Yokota, S.; Bai, L.; Ago, M.; Borghei, M.; Kondo, T.; Rojas, O. J. Formulation and composition effects in phase transitions of emulsions co-stabilized by cellulose nanofibrils and an ionic surfactant. Biomacromolecules 2017, 18, 4393−4404. (15) Han, G.; Huan, S.; Han, J.; Zhang, Z.; Wu, Q. Effect of Acid Hydrolysis Conditions on the Properties of Cellulose Nanoparticlereinforced Polymethylmethacrylate Composites. Materials 2014, 7, 16−29. (16) Xu, X.; Wang, H.; Jiang, L.; Wang, X.; Payne, S. A.; Zhu, J.; Li, R. Comparison Between Cellulose Nanocrystal and Cellulose Nanofibril Reinforced Poly(ethylene oxide) Nanofibers and Their Novel Shish-kebab-like Crystalline Structures. Macromolecules 2014, 47, 3409−3416. (17) Nishino, T.; Takano, K.; Nakamae, K. Elastic Modulus of the Crystalline Regions of Cellulose Polymorphs. J. Polym. Sci., Part B: Polym. Phys. 1995, 33, 1647−1651. (18) Shi, Q.; Zhou, C.; Yue, Y.; Guo, W.; Wu, Y.; Wu, Q. Mechanical Properties and In Vitro Degradation of Electrospun Bio-nanocomposite Mats from PLA and Cellulose Nanocrystals. Carbohydr. Polym. 2012, 90, 301−308. (19) Habibi, Y.; Lucia, L. A.; Rojas, O. J. Cellulose Nanocrystals: Chemistry, Self-assembly, and Applications. Chem. Rev. 2010, 110, 3479−3500. (20) Huan, S.; Bai, L.; Liu, G.; Cheng, W.; Han, G. Electrospun Nanofibrous Composites of Polystyrene and Cellulose Nanocrystals: Manufacture and Characterization. RSC Adv. 2015, 5, 50756−50766. (21) Dong, H.; Strawhecker, K. E.; Snyder, J. F.; Orlicki, J. A.; Reiner, R. S.; Rudie, A. W. Cellulose Nanocrystals as a Rreinforcing Material for Electrospun Poly(methyl methacrylate) Fibers: Formation, Properties and Nanomechanical Characterization. Carbohydr. Polym. 2012, 87, 2488−2495. (22) Li, C.; Li, Q.; Ni, X.; Liu, G.; Cheng, W.; Han, G. Coaxial Electrospinning and Characterization of Core-Shell Structured Cellulose Nanocrystal Reinforced PMMA/PAN Composite Fibers. Materials 2017, 10, 572−587. (23) Arrieta, M.; López, J.; López, D.; Kenny, J.; Peponi, L. Biodegradable Electrospun Bionanocomposite Fibers Based on Plasticized PLA-PHB Blends Reinforced with Cellulose Nanocrystals. Ind. Crops Prod. 2016, 93, 290−301. (24) Song, W.; Liu, D.; Prempeh, N.; Song, R. Fiber Alignment and Liquid Crystal Orientation of Cellulose Nanocrystals in the Electrospun Nanofibrous Mats. Biomacromolecules 2017, 18, 3273−3279. (25) Wanasekara, N.; Santos, R.; Douch, C.; Frollini, E.; Eichhorn, S. J. Orientation of Cellulose Nanocrystals in Electrospun Polymer Fibres. J. Mater. Sci. 2016, 51, 218−227. (26) Xue, J.; Xie, J.; Liu, W.; Xia, Y. Electrospun Nanofibers: New Concepts, Materials, and Applications. Acc. Chem. Res. 2017, 50, 1976−1987. (27) Liao, H.; Wu, Y.; Wu, M.; Zhan, X.; Liu, H. Aligned Electrospun Cellulose Fibers Reinforced Epoxy Resin Composite Films with High Visible Light Transmittance. Cellulose 2012, 19, 111−119. (28) Wang, Y.; Chen, L. Cellulose Nanowhiskers and Fiber Alignment Greatly Improve Mechanical Properties of Electrospun J
