Highly Biodegradable and Tough Polylactic Acid–Cellulose

Mar 21, 2017 - Atlantic Basin Conference Call for Abstracts: Deadline Extended to Oct 2nd. The American Chemical Society has partnered with seven othe...
6 downloads 15 Views 6MB Size
Download PDF
Research Article pubs.acs.org/journal/ascecg

Highly Biodegradable and Tough Polylactic Acid−Cellulose Nanocrystal Composite Joseph K. Muiruri,†,‡ Songlin Liu,*,‡ Wern Sze Teo,§ Junhua Kong,‡ and Chaobin He*,†,‡ †

Department of Materials Science and Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117576 Institute of Materials Research and Engineering, Agency for Science, Technology and Research (A*STAR), 2 Fusionopolis Way, Innovis #08-03, Singapore 138634 § Singapore Institute of Manufacturing Technology, Agency for Science, Technology and Research (A*STAR), 2 Fusionopolis Way, Innovis, #08-04, Singapore 138634 ‡

S Supporting Information *

ABSTRACT: Poly(L-lactide) cellulose nanocrystals-filled nanocomposites were fabricated by blending of cellulose nanocrystalsg-rubber-g-poly(D-lactide) (CNC-rD-PDLA) and commercial PLLA, in which CNC-g-rubber was synthesized by ring opening polymerization (ROP) of D-lactide and a ε-caprolactone mixture to obtain CNC-P(CL-DLA), followed by further polymerization of D-lactide to obtain CNC-rD-PDLA. X-ray diffraction (XRD), nuclear magnetic resonance (NMR), and solubility tests confirmed successful grafting of the rubber segment and the PDLA segment onto CNC. Stereocomplexation between CNC-rDPDLA nanofillers and PLLA matrix was confirmed by FT-IR, XRD, and differential scanning calorimetry (DSC) characterization. The PLLA/CNC-rD-PDLA nanocomposites exhibited greatly improved tensile toughness. With 2.5% CNC-rD-PDLA loading, strain at break of PLLA/CNC-rD-PDLA was increased 20-fold, and the composite shows potential to replace poly(ethylene terephthalate). SEM and small-angle X-ray scattering (SAXS) investigations revealed that fibrillation and crazing during deformation of PLLA/CNC-rD-PDLA nanocomposites were the major toughening mechanisms in this system. The highly biodegradable and tough cellulose nanocrystals-filled PLLA nanocomposites could tremendously widen the range of industrial applications of PLA. KEYWORDS: Stereocomplex, Sustainability, Poly(lactide), Small angle X-ray scattering

■

composites.7 However, PLA has weaknesses, for example, ingrained brittleness (elongation at break ∼4−7%), low heat deflection temperature (HDT) (∼55 °C), and poor impact resistance (impact strength ∼2.6 kJ/m2), which limits its largescale commercial applications.6,8 Many strategies have been devised to alter the bulk and surface properties of PLA, such as copolymerization, addition of nanofillers (organic or inorganic), and stereocomplexation.6,9,10 Stereocomplexation can occur mainly from stereoselective van der Waal forces interactions between PLLA and PDLA, resulting in formation of stereocomplex crystallites (SC) with better physical properties compared to PLLA or PDLA.11,12

INTRODUCTION Increased environmental awareness has revitalized a worldwide search for ecofriendly products in tandem with the United Nations (UN) program on environment and sustainable development.1 In the polymer composites industry, the sustainability agenda has stirred up interest in biopolymers because they are considered renewable, biodegradable, and biocompatible.2,3 Among the many biopolymers, poly(lactide) (PLA), which is mainly derived from corn and sugar cane, remains the most promising due to its unique properties such as biodegradability, compostability, renewability, and nontoxicity to humans and the environment.4 Moreover, some of the PLA’s physical properties such as stiffness and tensile strength are comparable to those of synthetic plastics such as polyethylene terepthalate (PET).5,6 In view of this, PLA-based composites would be a viable alternative to petroleum-based © 2017 American Chemical Society

Received: December 21, 2016 Revised: March 15, 2017 Published: March 21, 2017 3929

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering

Figure 1. Synthesis of CNC-rD-PDLA by ring opening polymerization (ROP).

is hypothesized that by incorporating CNC in the center, the modulus and tensile strength of the resulting composite will not be compromised while the toughness will increase. In addition, a copolymer of ε-caprolactone (ε-CL) and lactide (LA) will be grafted onto CNC, as a biodegradable rubber segment, and a PDLA chain will be subsequently grafted to the copolymer to facilitate stereocomplexation with the PLLA matrix, thus enhancing the filler−matrix interaction. It is hoped that the PLLA/CNC-rD-PDLA nanocomposites will exhibit improved toughness, similar to the PLLA/POSS-rubber-PDLA system. At the same time, CNC-rD-PDLA nanofillers as well as PLLA/ CNC-rD-PDLA nanocomposites are biodegradable.

The SC crystallites not only show high melting point (by 50 °C) but also exhibit higher mechanical performances, thermal stability, and hydrolytic resistance.13 Recently, advanced stereocomplex-based materials have been developed by combining stereocomplexation and nanocomposite approaches.14,15 So far, various fillers such as clay,16 lignin,17 carbon nanotubes (CNT),18,19 polyhedral oligomeric silsesquioxanes (POSS),20 graphene, graphene oxide (GO),21,22 and so on have been utilized in fabrication of stereocomplex PLLA nanocomposites. Although different kinds of nanofillers have been used to improve PLLA properties, most of them are nonbiodegradable and, therefore, not consistent with the concept of sustainability. On the other hand, cellulose, the most ubiquitous polydispersed linear polymer, which is widely investigated due to its unique characteristics, is renewable, biodegradable, and biocompatible with high strength (7.5 GPa), high modulus (145 GPa), high aspect ratio, high surface area, thermal stability, and dimensional stability.23−25 Nanocellulose has also been utilized as nucleating agents as well as multifunctional initiators in ring opening polymerization (ROP) of PLA.4,26,27 Purnama and Kim2 fabricated biostereocomplex−nanocomposite materials by stereocomplexation of equimolar PLLA-CNW and PDLA-CNW nanofillers through supercritical carbon dioxide technology. The nanofillers were prepared by grafting PLLA and PDLA onto acetylated cellulose nanowhiskers (CNWs), respectively. Another strategy of enhancing the mechanical properties of PLA, especially toughness and impact strength, is rubber toughening. Blending PLA with ductile polymers, such as natural rubber,28 poly(n-butyl acrylate) (PBA),29 poly(butylene succinate) (PBS),30 and star-shaped structures,31 has revealed enhanced overall toughness of PLA.27 In our earlier work,29 PBA-g-PDLA was synthesized and blended with PLLA to fabricate stereocomplex nanocomposites. The nanocomposites exhibited elongation at break of 29%. It is worth noting that most of the rubber phases used in PLLA toughening are nonbiodegradable. On the other hand, poly(ε-caprolactone) has a remarkable combination of ductility, toughness, and biocompatibility, a suitable biodegradable polymer widely used in biomedical applications.20,27 Random copolymers of ε-caprolactone (ε-CL) and lactide (LA) have been widely studied as biodegradable materials with rubbery characteristics for PLA toughening.32,33 However, CNC-g-rubber-g-PDLA nanofillers and their stereocomplex PLLA nanocomposites have not been reported so far. In this work, a tripartite approach has been designed to include cellulose nanocrystals (CNC) as rigid nanofillers due to their unique physical properties and environmental benefits. It

■

EXPERIMENTAL SECTION

Materials. Spray dried cellulose nanocrystals (CNCs) were purchased from the University of Maine. The D-lactide (D-LA) monomer was purchased from Corbion Purac. The ε-caprolactone (εCL) monomer, anhydrous toluene, and tin(II)-ethylhexanoate (Sn(Oct)2, 95%) were purchased from Sigma-Aldrich. Commercial PLLA (product number: 3051D; 96.5% of L-lactide; Mw: 160,000; PDI: 1.7) pellets were obtained from NatureWorks. All other chemicals were used as received. Synthesis of CNC-rD, CNC-rD-PDLA, and CNC-g-PDLA. Here, 2 g of dried and solvent-exchanged cellulose nanocrystals (CNC) and 9.19 g (80.6 mmol) of ε-caprolactone (ε-CL) were charged in a 100 mL round-bottomed flask to obtain a homogeneous mixture. Then, 0.2 g of Sn(Oct)2 (1% on weight of monomers), 5.81 g (40.3 mmol) of Dlactide, and 50 mL of anhydrous toluene were added to the mixture and homogenized at 50 °C. The homogeneous mixture was refluxed and stirred at 130 °C in nitrogen for 24 h. After that, 2 mL of the reaction mixture was obtained and precipitated in excess methanol and denoted as CNC-rD for characterization purposes. Finally, 5 g (34.7 mmol) of D-lactide was charged to the reaction mixture at 130 °C for another 24 h. The resultant product was separated from the mixture by precipitation in excess methanol, filtered by centrifuging in methanol several times, dried overnight in a vacuum oven at 60 °C, and denoted as CNC-rD-PDLA. Figure 1 shows the synthesis of CNC-rD-PDLA by ring opening polymerization. To clearly understand the contribution of the rubber segment in CNC-rD-PDLA to the properties of the fabricated nanocomposites, CNC-g-PDLA without the rubber segment was synthesized for comparison. Preparation of PLLA/CNC-rD-PDLA and PLLA/CNC-g-PDLA Nanocomposites. CNC-rD-PDLA and commercial PLLA were dissolved in chloroform separately, and their solutions were admixed and then cast in glass Petri dishes. The concentration of the solution was about 100 mg/mL. The film samples were obtained by slow evaporation of solvent for 12 h and then dried overnight at 60 °C under vacuum. The cast films were then ground and injection molded into dog-bone specimens according to ASTM D638, Type V,34 using a piston injection molding system (Haake MiniJet, Thermo Fisher Scientific) operated at an injection temperature of 205 °C, mold temperature of 65 °C, injection pressure/time of 950 bar/15 s, and 3930

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering

Figure 2. (a) WAXD profiles of CNC, CNC-rD, CNC-rD-PDLA, and CNC-g-PDLA showing different characteristic peaks. (b) Solubility tests of (1) CNC, (2) CNC-rD, (3) CNC-rD-PDLA, and (4) CNC-g-PDLA in chloroform after 15 min. post-pressure/time of 350 bar/45 s. Neat PLLA and PLLA/CNC-gPDLA were also prepared in the same way. Characterization. 1H NMR was used to evaluate the chemical structures of CNC-rD, CNC-rD-PDLA, and CNC-g-PDLA. The analyses were performed at room temperature with a Bruker 400 MHz spectrometer using CDCl3 as solvent and internal standard (δ = 7.26 ppm). The thermal properties of the synthesized nanofillers and the prepared nanocomposites were characterized using DSC (TA Instruments Q100). The samples were heated from 30 to 250 °C, cooled to 30 °C, and reheated to 250 °C using a ramping rate of 20 °C/min. The X-ray diffraction (XRD) diffratograms were obtained at ambient temperature on a Bruker GADDS X-ray diffractometer (D8 ADVANCE). The operating voltage was 40 kV for the area detector and a current of 40 mA, with Cu Kα radiation (λ = 1.5418 Å). The intensity in the spectra was measured as a function of 2θ in the range of 5−40°. The impact fracture surfaces of the nanocomposites were observed on a field emission scanning electron microscope (FESEM) (JEOL JSM 6700F) operated at 5.00 kV. The samples were sputter-coated with gold prior to SEM observations. Fourier transform infrared (FT-IR) spectra were obtained using a PerkinElmer 2000 spectrophotometer at a resolution of 4 cm−1 and 64 scans. Tensile tests were performed using an Instron 5569 tensile machine at a crosshead speed of 1 mm/min at room temperature. Five specimens of each nanocomposite were tested, and the data reported were the average values. Small angle X-ray scattering (SAXS) was performed on a Xenocs Xeuss 2.0 SAXS instrument at 50 kV and 0.6 mA. The scattering intensity of the SAXS results was normalized by sample thickness. Cu Kα radiation (1.5418 Å) was used.

Figure 3. DSC curves for CNC-rD, CNC-rD-PDLA, and CNC-gPDLA.

i.e., cellulose I (α at 2θ = 15°, 16.5°, 20.5°, and 34.5°) and II (β at 2θ = 12.5°, 20°). After grafting P(CL-LA) chains onto the CNC surface, CNC diffraction peaks designated to cellulose I (α) and cellulose II (β) become much weaker. This is probably due to the low cellulose content (ca. 10%) in CNC-rD. However, after grafting of the rubber segment onto CNC, new peaks appear at 2θ = 16.5°, 19°, 21°, and 24° which could be ascribed to PDLA and PCL blocks in the random copolymer.17,35 With subsequent grafting of the PDLA chain onto CNC-rD, the peaks at 2θ = 16.5° and 19° become stronger, which is evidence that they belong to the PDLA segment in CNC-rD-PDLA. In CNC-g-PDLA, strong peaks appearing at ∼2θ = 16.5°, 19°, 22.5°, and 34.5° are ascribed to PDLA and cellulose I (α), respectively. Solubility tests were performed on CNC, CNC-rD, CNCrD-PDLA, and CNC-g-PDLA in chloroform. As shown in Figure 2(b), CNC-rD, CNC-rD-PDLA, and CNC-g-PDLA form stable dispersions in chloroform, whereas CNC in chloroform is cloudy and will form precipitate upon being kept still. The improved solubility also confirms the effective grafting of CL-LA chains and PDLA chains to form CNC-rDPDLA.

■

RESULTS AND DISCUSSION Synthesis and Characterization of CNC-rD, CNC-rDPDLA, and CNC-g-PDLA. CNC-rD-PDLA was synthesized based on a protocol applied by our group.17 The fundamental idea was to graft a random copolymer of ε-CL and D-LA as a rubber segment (rD), followed by the polymerization of a PDLA block to form CNC-rD-PDLA. To understand the possible structure changes during grafting reactions, WAXD was used to characterize CNC, CNC-rD, CNC-rD-PDLA, and CNC-g-PDLA. As shown in Figure 2(a), unmodified CNC used in this study exists in two polymorphs, 3931

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering

Figure 4. 1HNMR spectra of CNC-rD-PDLA, CNC-rD, and CNC-g-PDLA.

Figure 6. WAXD profiles of neat PLLA and PLLA/CNC-rD-PDLA nanocomposites.

Figure 5. FTIR spectra of PLLA and PLLA/CNC-rD-PDLA nanocomposites.

CNC-rD-PDLA, the Tg is slightly increased to −14 °C. This could result from covalent bonding between PDLA and P(CLLA) chains, which reduces the mobility of P(CL-LA) chains. In contrast, CNC-g-PDLA exhibits a high melting peak at about 170 °C, which is from the PDLA chain and a Tg near 70 °C, which is slightly above the Tg of PLA. The chain structures of nanofillers were further investigated by using 1H NMR (Figure 4). The signals for proton assignment as reported in the literature are methine protons CHCH3 (δ ∼ 5.2 ppm) of the PLLA repeat unit and methyl

The thermal properties of the nanofillers were probed using DSC. From the DSC thermograms in Figure 3, CNC-rD exhibits a Tg of about −20 °C, which underscores the rubbery character of CNC-rD at room temperature. The thermogram also reveals a small melting peak near 130 °C, which is lower than the melting temperature of PLA. This demonstrates that the crystallites are small and quasi-random in nature.17 After subsequent grafting of the PDLA onto CNC-rD to obtain 3932

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering

Figure 9. SEM images of the fractured surfaces of (a) neat PLLA, (b) 2.5% CNC-rD-PDLA, (c) 5.0% CNC-rD-PDLA, and (d) 10.0% CNCrD-PDLA. Figure 7. DSC curves of PLLA and PLLA/CNC-rD-PDLA nanocomposites with various nanofiller loadings.

Figure 10. SEM microstructure of cryo-fractured surface of (A,B) PLLA/2.5% CNC-rD-PDLA showing homogeneous distribution of the fillers and (C,D) PLLA/10% CNC-rD-PDLA showing the “large” size fillers with good interface adhesion.

Figure 8. Typical nominal stress−strain curves for PLLA and PLLA/ CNC-rD-PDLA nanocomposites with different CNC-rD-PDLA contents.

Table 1. Tensile Properties of PLLA, PLLA/CNC-rD-PDLA, and PLLA/CNC-g-PDLA with Various Loadings sample neat PLLA 2.5% CNC-rD-PDLA 5.0% CNC-rD-PDLA 10.0% CNC-rD-PDLA 2.5% CNC-g-PDLA 5.0% CNC-g-PDLA

maximum tensile strength (MPa) 58.43 50.50 44.22 42.10 53.27 55.50

± ± ± ± ± ±

1.37 0.85 2.9 1.4 2.93 0.83

Young’s modulus (GPa) 3.88 3.24 2.70 2.64 3.39 3.30

± ± ± ± ± ±

0.26 0.38 0.29 0.4 0.48 0.42

strain at break (%) 8.07 187.58 208.89 246.78 11.84 8.54

± ± ± ± ± ±

3.57 16.7 16 30 0.89 1.6

Figure 11. SEM microstructure of fracture surface of PLLA/5% CNCg-PDLA: (a) low magnification and (b) high magnification.

protons CHCH3 (δ ∼ 1.4 ppm) of PLLA. The methylene protons (−CH2−O− and −CH2−COO−) of PCL repeat units are observed at δ ∼ 4.07−4.11 ppm and δ ∼ 2.23−2.32 ppm, respectively. The chemical shift for the methylene proton (−CH2−CH2−) of the PCL repeat unit is observed at δ ∼ 1.50−1.65 ppm.36 The presence of NMR peaks for CNC-rD

and CNC-rD-PDLA at chemical shifts 5.2, 4.16, 4.06, and 2.3− 2.4 ppm confirms the successful grafting. These peaks slightly increase in intensity after grafting the PDLA chain. The chemical shifts for PCL at 4.06 and 2.3−2.4 ppm are absent in CNC-g-PDLA, an indication that no rubber segment is present. 3933

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering

Figure 12. (a) Schematic representation of tensile direction vs incident X-ray and expected cross-like SAXS signature. (b) SAXS scattering pattern for undeformed neat PLLA. (c) Structural evolution of SAXS patterns at different strains for 2.5% CNC-rD-PDLA/PLLA blend. (d) Integrated intensity (I × q) versus scattering vector (q) curves of neat PLLA and 2.5% CNC-rD-PDLA/PLLA nanocomposites obtained by integrating the 35 sector areas in SAXS patterns, where q = 4π sin(θ)/λ (Å−1) and I × q (counts/s).

increased melting temperature by about 50 °C above that of homocrystallites counterparts.11 This increase in melting temperature is ascribed to strong hydrogen interactions (CH3···OC) in the stereocomplex (31 helix) crystalline structure.13,38 Having confirmed the stereocomplex formation, the thermal behavior of neat PLLA and PPLA/CNC-rD-PDLA nanocomposites was studied using DSC. From the DSC thermograms in Figure 7, with 10% CNC-rD-PDLA loading, an additional endothermic peak appears at 196 °C besides the normal melting peak at 150 °C, which is characteristic of a stereocomplex. The melting peak at 196 °C is below the expected 210 °C for stereocomplex crystals reported earlier,21 possibly because the PDLA chains are short which produces imperfect stereocomplex crystallites.17 The stereocomplex melting temperature is not prominent at 2.5% and 5% CNCrD-PDLA loadings, possibly due to either low filler loadings.39 Mechanical Properties. Improvements of properties in nanocomposites through SC formation are not limited to thermal properties but may also be extended to mechanical properties.17 From this point of view, it is necessary to investigate the mechanical properties of the nanocomposites. It is worth noting that improving the toughness of thermoplastic by adding a rubber phase plays an important role in the polymer industry. Figure 8 shows the typical stress−strain curves for neat PLLA and PLLA/CNC-rD-PDLA nanocomposites with different CNC-rD-PDLA contents. As shown, the addition of CNC-rD-PDLA leads to a dramatic increase in strain at break from 8% for neat PLLA to approximately 180%, 210%, and 250% for the nanocomposites containing 2.5%, 5.0%, and 10% CNC-rD-PDLA nanofillers,

This agrees with DSC results in Figure 3 where CNC-g-PDLA has a high Tg and melting temperature. Formation of Stereocomplexation in PLLA/CNC-rDPDLA Manocomposites. It is well known that PDLA and PLLA can form a stereocomplex from the melt or in solution as long as the D-lactyl unit sequences and L-lactyl unit sequences coexist in a system.13 With successful grafting of a rubber segment and a PDLA segment onto a CNC surface, it is likely that a stereocomplex between the PDLA segment in the CNCrD-PDLA and PLLA matrix will be formed as envisaged in the design of the experiment. Figure 5 shows the ATR-FTIR spectra of PLLA and the PLLA/CNC-rD-PDLA nanocomposites. Absorption at 920 cm−1 represents a 103 helix from PLLA homocrystals, while at 908 cm−1 represents a 31 helix characteristic of stereocomplex crystallites.3 As shown, the peak at 920 cm−1 decreases while the broad peak at 908 cm−1 for neat PLLA increases with the addition of the CNC-rDPDLA nanofillers. The stereocomplexation was further confirmed using WAXD. As shown in Figure 6, neat PLLA exhibits diffraction peaks at 2θ = 14.8°, 16.5°, 19°, and 22.3°. PLLA/CNC-rD-PDLA nanocomposites show additional diffraction peaks at 2θ = 11.9°, 20.8°, and 23.8°, which are the characteristic peaks from SC crystals.37 It is worth noting that both homocrystallites and SC crystals are present in the nanocomposites. The fraction of the SC crystals increases with increasing CNC-rD-PDLA loadings as the diffraction intensity becomes stronger. Thermal Properties. Stereocomplex formation can be considered as an easy and important approach in improving the physical properties of PLA, especially thermal properties with 3934

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering

in previous studies20,44 and further confirmed in this work by various characterizations. In order to study the interface bonding between the filler and the matrix, we purposely fractured the samples in liquid nitrogen to avoid the interference from the fibrillation. Figure 10 shows the SEM images of cryo-fractured surfaces of PLLA/2.5% CNC-rDPDLA and PLLA/10% CNC-rD-PDLA nanocomposites, respectively. Apparently, homogeneous distribution of the fillers is evidenced. Furthermore, no debonding of the filler particles from the matrix was observed in the samples fractured in cryogenic conditions, implying good matrix−filler interactions. It is, therefore, apparent that the slight weakening of the tensile strength and modulus in this work is caused by the rubber segment in the filler. In contrast, although stereocomplex formation is also evident in the control samples of PLLA/CNC-g-PDLA nanocomposites, the mechanical properties are distinct from that of PLLA/ CNC-rD-PDLA nanocomposites. For instance, PLLA/5% CNC-g-PDLA nanocomposites show high tensile strength and high Young’s modulus of 55.5 MPa and 3.3 GPa respectively, compared to 44.2 MPa and 2.7 GPa for PLLA/ 5% CNC-rD-PDLA nanocomposites. The strain at break of PLLA/CNC-g-PDLA is only comparable to that of neat PLLA. These observations justify the SEM study to help unravel the structure−property relationship for the control samples. Figure 11 shows SEM images of the PLLA/5% CNC-gPDLA nanocomposite. As shown, the fracture surface has a lot of cavities, showing the existence of agglomeration of the nanofillers, and is also smooth, without fibrillations observed. It is worth noting that both samples of PLLA/CNC-rD-PDLA and PLLA/CNC-g-PDLA can only be differentiated by the absence of the rubber segment in CNC-g-PDLA, which could be correlated to the differences in the resultant mechanical properties. In view of this, the most probable toughening mechanism for PLLA/CNC-rD-PDLA nanocomposites is singled out as rubber-induced crazing and fibrillation in which CNC-rDPDLA nanofillers act as craze initiators. This occurs when the stress required to initiate crazing is less than the yield stress; thus, multiple triaxial stress concentrations at the dispersed phase initiate crazes.45 To support this school of thought, the SAXS technique was employed to trace the formation of crazes in the PLLA/2.5% CNC-rD-PDLA nanocomposite. Figure 12(a) shows the cross-like SAXS signature expected from the scattering of craze fibrils in the equatorial direction and of the crack planes in the meridional direction, with the incident X-ray normal to the stretching direction. Figure 12(c) and (d) show the 2-D and 1-D SAXS structural evolution as a function of strain, respectively. The 2-D SAXS images are reduced to 1-D scattering curves by integration to obtain scattering intensity (I) as a function of scattering vector (q) (where q = 4π sin(θ)/ λ, 2θ is the scattering angle). This is followed by calculations of the relative invariants (Qr), making use of the 35° sector areas through invariant analysis as follows:

respectively. A tremendous improvement of strain at break is a strong indication of improvement of tensile toughness. A slight trade-off between the strain at break and the strength is observed. With 2.5% CNC-rD-PDLA nanofillers loading, strain at break is enhanced to ca. 180%. Considering that the strength is marginally reduced, these nanocomposites could open a wide range of applications for PLLA. Toughening Mechanism. To investigate the effect of the rubber segment in CNC-rD-PDLA nanofillers on the mechanical behavior of PLLA-based nanocomposites, CNC-gPDLA without the rubber segment was synthesized and used for PLLA modification. The mechanical properties of both PLLA/CNC-rD-PDLA and PLLA/CNC-g-PDLA are summarized in Table 1. PLLA/CNC-g-PDLA nanocomposites exhibit higher Young’s moduli and strengths compared with PLLA/ CNC-rD-PDLA nanocomposites but much lower strain at break, indicating much lower tensile toughness. The results could clearly unearth the role of a rubber segment on the enhancement of toughness of PLLA. From Table 1, it is apparent that the role of CNC-rD-PDLA nanofillers as toughening modifiers cannot be underestimated. The strain at break for PLLA/CNC-g-PDLA containing 2.5% and 5.0% of CNC-g-PDLA are only 11.8% and 8.5%, respectively, and PLLA/CNC/rD-PDLA containing 2.5% and 5.0% of CNC-rD-PDLA have strain at break of 187% and 208%, respectively. The toughening mechanism needs to be investigated in order to have a better understanding of the structure−property relationship. SEM studies were conducted to elucidate the morphology of the fracture surfaces. In Figure 9, the fracture surface of neat PLLA is smooth, indicating a brittle fracture character.33 This fracture character clearly correlates with the results presented in Table 1, low strain at break (8%) for neat PLLA and low energy absorption after yielding.20 When 2.5% CNC-rD-PDLA is incorporated into PLLA, a dramatic increase in strain at break is observed. Such a large strain at break is characteristic of a ductile material which undergoes yielding, followed by strain softening to an ultimate draw ratio and then strain hardening before fracture (Figure 8). With the addition of as low as 2.5% CNC-rD-PDLA nanofillers, the morphology of the nanocomposites changes to elongated fibril structures which are characteristic of a ductile fracture. The fibrillated structures are ascribed to resistance to deformation, which leads to higher absorption of energy and hence enhanced toughness.40 It is worth noting that in processing structure−property relationships, interplay of factors (e.g., matrix−filler interface adhesion, filler size, and filler loading) could result in enhancement or deterioration of mechanical properties. First, the toughness of the PLLA/CNC-rD-PDLA nanocomposites is enhanced. However, all nanocomposites have lower tensile strength and modulus compared to neat PLLA. In this study, the “as-received” spherical spray-dried nanocrystals are agglomerates (Figure S1), which is consistent with the conventional spray-drying process, and such morphology has been reported in previous studies.41−43 In our study, the presence of filler “particles” in Figure 10 is an indication of partial dispersion even after surface modification. Therefore, the reinforcement to improve strength and modulus from CNC could not be maximized. Second, compatibility between PLLA and the CNC-rDPDLA filler was controlled by the formation of a stereocomplex at the interface. This kind of interface control has been reported

q2

Qr =

∫q1

I(q)qdq

(1)

where Qr is a relative invariant, q is the scattering vector, and I(q) is the scattering intensity from crazes. As shown, distinct differences in structure evolution at different strains are evident. For the undeformed neat PLLA (Figure 12(b) and (d)), the scattering intensity is low due to lack of inhomogeneities. However, the undeformed PLLA/2.5% CNC-rD-PDLA nano3935

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering Author Contributions

composite has a higher scattering intensity; this could be ascribed to the presence of microdomains in the PLLA matrix Upon stretching of the PLLA/2.5% CNC-rD-PDLA nanocomposite, the 2-D SAXS pattern evolves from spherical to oblate in the equatorial direction. The evolution is quantified as craze fibrils using relative invariants (Qr) of neat PLLA at 0% strain and the PLLA/2.5% CNC-rD-PDLA nanocomposite at 0%, 10%, 50%, and 90% strains and was calculated to be 0.0006, 0.024, 0.027, 0.03, and 0.088, respectively. This phenomenon is attributed to the crowd of microcrazes, bridged by load-bearing fibrils, in the matrix, which often results in stress whitening at the macroscopic scale in tensile testing.28,46 From the calculated relative invariants, Qr, craze formation plays an important role in enhancing toughness of the nanocomposites. It is worth mentioning that the calculated relative invariants, which show the changes in total volume of scatterers, underestimate the real craze concentration due to two reasons: (1) The craze structure is subjected to remarkable shrinkage upon relaxation. (2) The craze fibrils are slowly reabsorbed into the bulk material upon removal of the load.47

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors sincerely thank Dr. Xiaoshan Fan, Dr. Du Yuan, and Mr. Jayven Chee Chuan Yeo for their valuable advice during synthesis and characterization of the nanocomposites.

■

■

CONCLUSIONS In this work, fully biodegradable cellulose nanocrystals-filled PLLA nanocomposites have been fabricated by solution blending of commercial PLLA and CNC-rD-PDLA nanofillers, followed by injection molding. The CNC-rD-PDLA nanofillers were synthesized by copolymerization of ε-caprolactone (CL) and lactide (D-LA) as a biodegradable rubber segment, followed by grafting of PDLA chains, via ring opening polymerization. The possible changes in structures during grafting reactions to produce CNC-rD-PDLA nanofillers and subsequent grafting of PDLA chains were confirmed by XRD, 1 H NMR, solubility tests, and DSC. Formation of a stereocomplex between PLLA and CNC-rD-PDLA nanofillers was revealed by FTIR, XRD, and DSC, providing strong interactions between the PLLA matrix and CNC-rD-PDLA nanofillers. Stereocomplex formation is known to enhance thermal properties of nanocomposites. The fabricated PLLA/ CNC-rD-PDLA exhibited enhanced melting temperature and an astonishing 20-fold increase in strain at break. SEM and SAXS studies revealed that crazing and fibrillation during deformation of the PLLA/CNC-rD-PDLA nanocomposites are the major mechanisms resulting in greatly improved toughness. The highly biodegradable and tough cellulose nanocrystalsfilled PLLA nanocomposites could tremendously widen the range of industrial applications of PLA.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acssuschemeng.6b03123. SEM image of the “as received” spray-dried cellulose nanocrystals (CNC). (PDF)

■

REFERENCES

(1) Haque, M. S. Environmental discourse and sustainable development: Linkages and limitations. Ethics and the Environment 2000, 5, 3−21. (2) Purnama, P.; Kim, S. H. Bio-based composite of stereocomplex polylactide and cellulose nanowhiskers. Polym. Degrad. Stab. 2014, 109, 430−435. (3) Fukushima, K.; Kimura, Y. Stereocomplexed polylactides (NeoPLA) as high-performance bio-based polymers: their formation, properties, and application. Polym. Int. 2006, 55 (6), 626−642. (4) Raquez, J. M.; Murena, Y.; Goffin, A. L.; Habibi, Y.; Ruelle, B.; DeBuyl, F.; Dubois, P. Surface-modification of cellulose nanowhiskers and their use as nanoreinforcers into polylactide: A sustainablyintegrated approach. Compos. Sci. Technol. 2012, 72 (5), 544−549. (5) Liu, H. Z.; Zhang, J. W. Research Progress in Toughening Modification of Poly(lactic acid). J. Polym. Sci., Part B: Polym. Phys. 2011, 49 (15), 1051−1083. (6) Anderson, K. S.; Schreck, K. M.; Hillmyer, M. A. Toughening polylactide. Polym. Rev. 2008, 48 (1), 85−108. (7) Raquez, J. M.; Habibi, Y.; Murariu, M.; Dubois, P. Polylactide (PLA)-based nanocomposites. Prog. Polym. Sci. 2013, 38 (10−11), 1504−1542. (8) Balakrishnan, H.; Hassan, A.; Imran, M.; Wahit, M. U. Toughening of Polylactic Acid Nanocomposites: A Short Review. Polym.-Plast. Technol. Eng. 2012, 51 (2), 175−192. (9) Li, Z.; Tan, B. H.; Lin, T.; He, C. Recent advances in stereocomplexation of enantiomeric PLA-based copolymers and applications. Prog. Polym. Sci. 2016, 62, 22. (10) Tan, B. H.; Muiruri, J. K.; Li, Z.; He, C. Recent Progress in Using Stereocomplexation for Enhancement of Thermal and Mechanical Property of Polylactide. ACS Sustainable Chem. Eng. 2016, 4 (10), 5370−5391. (11) Ikada, Y.; Jamshidi, K.; Tsuji, H.; Hyon, S. H. Stereocomplex formation between enantiomeric poly (lactides). Macromolecules 1987, 20 (4), 904−906. (12) Fan, X.; Tan, B. H.; Li, Z.; Loh, X. J. Control of PLA Stereoisomers-Based Polyurethane Elastomers as Highly Efficient Shape Memory Materials. ACS Sustainable Chem. Eng. 2016, 5, 1217. (13) Tsuji, H. Poly (lactide) stereocomplexes: formation, structure, properties, degradation, and applications. Macromol. Biosci. 2005, 5 (7), 569−597. (14) Jiang, L.; Shen, T.; Xu, P.; Zhao, X.; Li, X.; Dong, W.; Ma, P.; Chen, M. Crystallization modification of poly (lactide) by using nucleating agents and stereocomplexation. e-Polym. 2016, 16 (1), 1− 13. (15) Li, Z.; Yuan, D.; Jin, G.; Tan, B. H.; He, C. Facile Layer-byLayer Self-Assembly toward Enantiomeric Poly (lactide) Stereocomplex Coated Magnetite Nanocarrier for Highly Tunable Drug Deliveries. ACS Appl. Mater. Interfaces 2016, 8 (3), 1842−1853. (16) Furuhashi, Y.; Morioka, K.; Tamegai, H.; Yoshie, N. Preparation and some properties of stereocomplex-type poly (lactic acid)/layered silicate nanocomposites. J. Appl. Polym. Sci. 2013, 127 (3), 1615−1622. (17) Sun, Y.; Yang, L. P.; Lu, X. H.; He, C. B. Biodegradable and renewable poly(lactide)-lignin composites: synthesis, interface and toughening mechanism. J. Mater. Chem. A 2015, 3 (7), 3699−3709.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected](S. Liu) . *E-mail: [email protected] (C. He). ORCID

Chaobin He: 0000-0001-8200-8337 3936

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937

Research Article

ACS Sustainable Chemistry & Engineering (18) Sun, Y.; He, C. B. Synthesis, stereocomplex crystallization, morphology and mechanical property of poly(lactide)-carbon nanotube nanocomposites. RSC Adv. 2013, 3 (7), 2219−2226. (19) Dong, Q.; Li, Y.; Han, C.; Zhang, X.; Xu, K.; Zhang, H.; Dong, L. Poly (l-lactide)/poly (d-lactide)/multiwalled carbon nanotubes nanocomposites: Enhanced dispersion, crystallization, mechanical properties, and hydrolytic degradation. J. Appl. Polym. Sci. 2013, 130 (6), 3919−3929. (20) Sun, Y.; He, C. Biodegradable “core−shell” rubber nanoparticles and their toughening of poly (lactides). Macromolecules 2013, 46 (24), 9625−9633. (21) Sun, Y.; He, C. B. Synthesis and Stereocomplex Crystallization of Poly(lactide)-Graphene Oxide Nanocomposites. ACS Macro Lett. 2012, 1 (6), 709−713. (22) Xu, H.; Wu, D.; Yang, X.; Xie, L.; Hakkarainen, M. Thermostable and impermeable “nano-barrier walls” constructed by poly (lactic acid) stereocomplex crystal decorated graphene oxide nanosheets. Macromolecules 2015, 48 (7), 2127−2137. (23) Habibi, Y.; Lucia, L. A.; Rojas, O. J. Cellulose Nanocrystals: Chemistry, Self-Assembly, and Applications. Chem. Rev. 2010, 110 (6), 3479−3500. (24) Lin, N.; Huang, J.; Dufresne, A. Preparation, properties and applications of polysaccharide nanocrystals in advanced functional nanomaterials: a review. Nanoscale 2012, 4 (11), 3274−3294. (25) Ng, H. M.; Sin, L. T.; Tee, T. T.; Bee, S. T.; Hui, D.; Low, C. Y.; Rahmat, A. R. Extraction of cellulose nanocrystals from plant sources for application as reinforcing agent in polymers. Composites, Part B 2015, 75, 176−200. (26) Habibi, Y. Key advances in the chemical modification of nanocelluloses. Chem. Soc. Rev. 2014, 43 (5), 1519−1542. (27) Purnama, P.; Jung, Y.; Kim, S. H. Stereocomplexation of Poly(Llactide) and Random Copolymer Poly(D-lactide-co-epsilon-caprolactone) To Enhance Melt Stability. Macromolecules 2012, 45 (9), 4012− 4014. (28) Bitinis, N.; Fortunati, E.; Verdejo, R.; Bras, J.; Kenny, J. M.; Torre, L.; Lopez-Manchado, M. A. Poly(lactic acid)/natural rubber/ cellulose nanocrystal bionanocomposites. Part II: Properties evaluation. Carbohydr. Polym. 2013, 96 (2), 621−627. (29) Ye, S.; Ting Lin, T.; Weei Tjiu, W.; Kwan Wong, P.; He, C. Rubber toughening of poly (lactic acid): Effect of stereocomplex formation at the rubber-matrix interface. J. Appl. Polym. Sci. 2013, 128 (4), 2541−2547. (30) Stoyanova, N.; Mincheva, R.; Paneva, D.; Manolova, N.; Dubois, P.; Rashkov, I. Electrospun non-woven mats from stereocomplex between high molar mass poly(l-lactide) and poly(d-lactide)-blockpoly(butylene succinate) copoly(ester urethane)s. Eur. Polym. J. 2012, 48 (12), 1965−1975. (31) Sun, B.; Liu, Y.; Zhang, B.; Bian, X.; Li, G.; Chen, X. Enhanced toughness and strength of poly (d-lactide) by stereocomplexation with 5-arm poly (l-lactide). J. Appl. Polym. Sci. 2016, 133 (1), 42857. (32) Xie, J.; Ihara, M.; Jung, Y.; Kwon, I. K.; Kim, S. H.; Kim, Y. H.; Matsuda, T. Mechano-active scaffold design based on microporous poly (L-lactide-co-ε-caprolactone) for articular cartilage tissue engineering: Dependence of porosity on compression force-applied mechanical behaviors. Tissue Eng. 2006, 12 (3), 449−458. (33) Zhang, C.; Zhai, T.; Turng, L.-S.; Dan, Y. Morphological, mechanical, and crystallization behavior of polylactide/polycaprolactone blends compatibilized by l-lactide/caprolactone copolymer. Ind. Eng. Chem. Res. 2015, 54 (38), 9505−9511. (34) ASTM Standard D638: Standard Test Method for Tensile Properties of Plastics; ASTM International: West Conshohocken, PA, 2010. (35) Agarwal, S.; Speyerer, C. Degradable blends of semi-crystalline and amorphous branched poly (caprolactone): Effect of microstructure on blend properties. Polymer 2010, 51 (5), 1024−1032. (36) Sriputtirat, S.; Boonkong, W.; Pengprecha, S.; Petsom, A.; Thongchul, N. Low molecular weight poly(lactide-co-caprolactone) for tissue adhesion and tetracycline hydrochloride controlled release in wound management. Adv. Chem. Eng. Sci. 2012, 2 (1), 15−27.

(37) Tsuji, H. Poly(lactic acid) stereocomplexes: A decade of progress. Adv. Drug Delivery Rev. 2016, 107, 97. (38) de Arenaza, I. M.; Sarasua, J. R.; Amestoy, H.; Lopez-Rodriguez, N.; Zuza, E.; Meaurio, E.; Meyer, F.; Santos, J. I.; Raquez, J. M.; Dubois, P. Polylactide Stereocomplex Crystallization Prompted by Multiwall Carbon Nanotubes. J. Appl. Polym. Sci. 2013, 130 (6), 4327− 4337. (39) Geschwind, J.; Rathi, S.; Tonhauser, C.; Schomer, M.; Hsu, S. L.; Coughlin, E. B.; Frey, H. Stereocomplex Formation in Polylactide Multiarm Stars and Comb Copolymers with Linear and Hyperbranched Multifunctional PEG. Macromol. Chem. Phys. 2013, 214 (13), 1434−1444. (40) Ma, P. Tailoring the Properties of Bio-Based and Biocompostable Polymer Blends. Ph.D., Technische Universiteit Eindhoven, 2011. http://dx.doi.org/10.6100/IR712662 (accessed March 2017). (41) Khoshkava, V.; Kamal, M. R. Effect of cellulose nanocrystals (CNC) particle morphology on dispersion and rheological and mechanical properties of polypropylene/CNC nanocomposites. ACS Appl. Mater. Interfaces 2014, 6 (11), 8146−8157. (42) Kamal, M. R.; Khoshkava, V. Effect of cellulose nanocrystals (CNC) on rheological and mechanical properties and crystallization behavior of PLA/CNC nanocomposites. Carbohydr. Polym. 2015, 123, 105−114. (43) Lu, P.; Hsieh, Y.-L. Preparation and properties of cellulose nanocrystals: rods, spheres, and network. Carbohydr. Polym. 2010, 82 (2), 329−336. (44) Bai, H.; Bai, D.; Xiu, H.; Liu, H.; Zhang, Q.; Wang, K.; Deng, H.; Chen, F.; Fu, Q.; Chiu, F.-C. Towards high-performance poly (Llactide)/elastomer blends with tunable interfacial adhesion and matrix crystallization via constructing stereocomplex crystallites at the interface. RSC Adv. 2014, 4 (90), 49374−49385. (45) Nagarajan, V.; Mohanty, A. K.; Misra, M. Perspective on Polylactic Acid (PLA) based Sustainable Materials for Durable Applications: Focus on Toughness and Heat Resistance. ACS Sustainable Chem. Eng. 2016, 4, 2899. (46) He, C.; Liu, T.; Tjiu, W. C.; Sue, H.-J.; Yee, A. F. Microdeformation and fracture mechanisms in polyamide-6/organoclay nanocomposites. Macromolecules 2008, 41 (1), 193−202. (47) Salomons, G. J. Small-Angle X-ray Scattering Study of Crazing in Bulk Thermoplastic Polymers; Queen’s University Kingston, 1998.

3937

DOI: 10.1021/acssuschemeng.6b03123 ACS Sustainable Chem. Eng. 2017, 5, 3929−3937