Cellulose DOI 10.1007/s10570-017-1479-3

ORIGINAL PAPER

Enhanced dispersion and interface compatibilization of crystalline nanocellulose in polylactide by surfactant adsorption Kai Chi . Jeffrey M. Catchmark

Received: 1 June 2017 / Accepted: 29 August 2017 Ó Springer Science+Business Media B.V. 2017

Abstract In this study, the surface polarity of sulfated crystalline nanocellulose (CNC) was tailored using an ecologically compatible surface modification strategy. Lauric arginate, a novel biologically-derived cationic surfactant, was ionically bonded to the CNC surface sulfate groups forming a monolayer that significantly increased surface hydrophobicity. Both unmodified (P-CNC) and surfactant modified (FCNC) were incorporated into a non-polar PLA matrix to study their reinforcing effect. The P-CNC, ascribed to its inherent hydrophilic characteristic, had limited nucleating and reinforcing effect on the PLA matrix. Large nanoparticle aggregation and interface debonding were easily discernable in P-CNC/PLA nanocomposite films. The hydrophobic F-CNC, by contrast, had a much better dispersibility and interface compatibility within the PLA matrix. The cold crystallization rate, crystallinity, storage modulus (glassy and rubbery states), glass transition temperature, and tensile strength and modulus of F-CNC/PLA nanocomposite films were remarkably enhanced with appropriate loading level of F-CNC (\5 wt%). These results demonstrate an efficient route to increase the hydrophobicity of CNC for its enhanced nanoreinforcing effect in various non-polar matrices.

K. Chi  J. M. Catchmark (&) Department of Agricultural and Biological Engineering, The Pennsylvania State University, 308 Forest Resources Lab, University Park, PA 16802, USA e-mail: [email protected]

Keywords Polylactide  Crystalline nanocellulose  Surface modification  Cationic surfactant  Interface compatibilization  Ionic binding

Introduction Biologically derived plastics produced from renewable natural resources have attracted an extraordinary level of attention as a crucial component of creating a sustainable society that values environmental stewardship. These ecologically compatible materials can offer comparable performance to synthetic polymers in many industries including packaging, food handling, construction, textiles, automotive and electronics while providing improved capabilities such as biocompatibility in biomedical applications (Murariu and Dubois 2016). These bioplastics, such as polylactide (PLA), thermoplastic starch, poly(butylene succinate) (PBS), and poly(hydroxyalkanoates) (PHAs), are considered to be renewable, biodegradable and biocompatible, and have appreciable properties that can substitute or complement traditional fossil-fuel plastics (polyolefin, polystyrene, polyethylene terephthalate) (Farah et al. 2016; Zhang et al. 2012). The global market of the bioplastics is progressively growing and estimated to constitute 25–30% of the overall plastic market share by 2020 (Misra et al. 2015).

123

Cellulose

PLA, a linear aliphatic polyester derived completely from renewable sources (corn, sugar, beet, etc.), is one of the most promising and widely studied biopolymers. It is commercially available (first commercial production by Natureworks in 2002) and industrially obtained via either direct polymerization of lactic acid or ring-opening polymerization of lactide dimer (Pretula et al. 2016; Tan et al. 2016). PLA continues to maintain a key-position in the bioplastics market due to its desirable biocompatibility, high mechanical properties (tensile strength and elastic modulus), high transparency, good processibility and competitive cost (Reddy et al. 2013). These unique characteristics have made PLA a versatile material in a broad array of applications, especially in food packaging and biomedical devices (Saini et al. 2016; Siracusa et al. 2008). Nonetheless, many challenges impede full market acceptance of PLA, including some inherent drawbacks such as brittleness, low heat distortion temperature, slow crystallization rate, and unsatisfactory water vapor or oxygen barrier properties (Dhar et al. 2016; Goffin et al. 2011; Rasal et al. 2010). Therefore, extensive efforts have been made to address these drawbacks with the aim to broaden its scope of application. The most commonly employed strategies include copolymerization (Li et al. 2016), stereocomplexation (Fukushima and Kimura 2006; Tan et al. 2016), incorporation of nanofiller (Raquez et al. 2013), polymer blending (Jiang et al. 2006; Liu et al. 2011; Zhang et al. 2014), and plasticization (Choi et al. 2013). Incorporation of a nanofiller appears to be a powerful approach to enhance the thermo-mechanical properties of PLA and furthermore endow desired functionalities for its specific end-use (Raquez et al. 2013). Various types of nanoreforcing fillers, including layered silicates (Lai et al. 2014; San Roma´n et al. 2013; Zhang et al. 2015), carbon nanotubes (Park et al. 2013), nanocellulose (Habibi et al. 2013; Oksman et al. 2006; Suryanegara et al. 2010), silica (Zhu et al. 2010), and metal oxides (Murariu et al. 2011), have been added into PLA matrix and the resultant nanocomposites exhibit enhanced properties and specific end-use characteristics. Unfortunately, many of these nanofillers are not biocompatible or biodegradable, impacting the life cycle and sustainability aspects of these compositions. Crystalline nanocellulose (CNC), a novel class of nanoparticle isolated by acid hydrolysis from the world’s most abundant biopolymer, is regarded as an

123

emerging commercial bionanomaterial (Dufresne 2013; Klemm et al. 2011). It presents impressive characteristics such as renewability, biocompatibility, biodegradability, extraordinary mechanical properties, high surface area, light weight and many surface functionalization possibilities, and therefore is an ideal nanoreinforcing candidate for polymer nanocomposites (Dufresne 2017). The pioneering works conducted by Favier et al. (1995a, b) demonstrated significant improvement in storage modus of poly(styrene-cobutyl acrylate) latex with a low loading level of tunicate nanocrystals. Since then, numerous studies have been reported on the incorporation of nanocellulose into PLA (Fortunati et al. 2012; Iwatake et al. 2008; Johari et al. 2016; Lin et al. 2011; Oksman et al. 2006) and many other polymer matrices (Ansari et al. 2015; Grunert and Winter 2002; Habibi et al. 2008; Kaushik et al. 2010; Nagalakshmaiah et al. 2016a, b; Natterodt et al. 2017; Xu et al. 2013; Yu et al. 2014). However, the hydrophilic nature of nanocellulose originating from the abundant surface hydroxyl groups has limited its applications to polar polymer matrices (Khan et al. 2012; Paralikar et al. 2008; Wang et al. 2010). Indeed, the improvement of overall mechanical properties of nanocomposites is strongly dependent on the nanodispersion of filler and good filler–polymer interface adhesion. Consequently, to expand the utilization of nanocellulose in applications involving hydrophobic or non-polar media, various surface modification strategies have been explored and are typically classified into two categories: chemical functionalization and physical adsorption (Lin et al. 2012). Common chemical functionalization approaches include esterification (Lin et al. 2011), etherification (Hasani et al. 2008), silylation (Raquez et al. 2012), amidation (Follain et al. 2010), ‘‘grafting onto’’ (a technique involving the coupling of presynthesized polymer chains, such as polyethylene glycol and polyethylene oxide, to the surface hydroxyls of CNC) (Mangalam et al. 2009), and ‘‘grafting from’’ (a process of surface initiated polymerization from the immobilized monomer on the CNC surface) (Habibi et al. 2008). However, this methodology would not be acceptable from a cost, health and safety or environmental lifecycle standpoint due to several deficiencies, including tedious washing and purification process, negative environmental impact and possibility of destroying the native characteristics of nanocellulose (Hubbe et al. 2015). Comparatively

Cellulose

speaking, physical adsorption by macromolecules or surfactants is a green, industrially scalable and ecologically friendly method to engineer the surface polarity of nanocellulose. A variety of macromolecules (ionomers, copolymers, polyethylene oxide, etc.) (Ben Azouz et al. 2011; Huang et al. 2016; Sakakibara et al. 2016) or surfactants (ionic or nonionic) (Ansari et al. 2015; Kim et al. 2009) have been reported for the surface functionalization of nanocellulose. Coupling of CNC with surfactant to promote its dispersion and interface compatibilization in nonpolar polymer matrices has been explored in several studies (Ansari et al. 2015; Bondeson and Oksman 2007; Fortunati et al. 2012; Kim et al. 2009; Ljungberg et al. 2005; Nagalakshmaiah et al. 2016a). Huex et al. (2000) presented the first report on surface modification of CNC using an anionic surfactant (an acid phosphate ester of ethoxylated nonylphenol) and the modified CNC was found to largely enhance the mechanical performance of atactic propylene matrix (Ljungberg et al. 2005). Kim and coworkers (Kim et al. 2009) used a nonionic surfactant (sorbitan monostearate) to modify the dispersion of CNC in a polystyrene matrix. They concluded that the optimum addition of surfactant could improve the mechanical properties of the resulting nanocomposite ascribed to the enhanced compatibility between CNC and polymer matrix (Kim et al. 2009). Most recently, hydrophobization of negatively charged CNCs using quaternary ammonium surfactants were investigated and the cationic surfactant-coated CNCs were uniformly dispersed in polypropylene or poly(vinyl acetate) matrix, contributing to the enhanced thermomechanical performance of the ensuing nanocomposites (Ansari et al. 2015; Nagalakshmaiah et al. 2016a). However, the excess addition of surfactant could have a negative impact on the properties of the nanocomposites due to its plasticizing effect as observed in previous studies (Bondeson and Oksman 2007; Kim et al. 2009). In the current study, a novel food-grade cationic surfactant lauric arginate (LAE) was used to modulate the surface polarity of sulfated CNC. LAE is a family of amino acid-based surfactants with eco-friendly characteristics (Pinazo et al. 2016) and has never been utilized to modify CNC for its improved dispersion in hydrophobic polymer matrix. Our previous work indicated that the complete ionic binding between

sulfated CNC and LAE occurred at the molar ratio of 1:4 on the basis of the sulfate groups from CNC and cationic head groups from LAE (Chi and Catchmark 2017). The goal of this study is, therefore, to evaluate the reinforcing efficiency of such ionically-coated CNC in the PLA matrix.

Materials and methods Materials Lab-made CNC was prepared by subjecting the microcrystalline cellulose (Avicel PH 101, SigmaAldrich, Saint Louis, MO, USA) to a standard acid hydrolysis process (64 wt% sulfuric acid, 45 °C, 2 h, 10 mL/g acid-to-cellulose), according to our previous work (Chi and Catchmark 2017). The extracted nanocrystals had average length and diameter of 152 ± 48 and 4.0 ± 1.4 nm, respectively, and a sulfate content of 0.23 mmol/g on the basis of AFM and conductometric titration results (Chi and Catchmark 2017). An amino acid-based cationic surfactant (lauric arginate, LAE, C20H41N4O3Cl, C98% purity, MW = 421.0 g/mol, Lot # 2014-1201) was kindly provided by A&B Ingredients (Fairfield, NJ, USA). Commercial semi-crystalline PLA (grade: 4032 D; 98.5% of L-lactide; weight average (Mw) and number average (Mn) molecular weight of 200,000 and 150,000 Da, respectively) pellets were obtained from NatureWorks (Minnetonka, MN, USA) and ovendried at 60 °C for 24 h prior to use. Dichloromethane (DCM) was purchased from Sigma-Aldrich. All chemical reagents were used as received. Surface modification of CNC An environmentally benign approach was employed to tailor the surface polarity of CNC for its improved dispersion and compatiblization in a PLA matrix. Briefly, the LAE stock solution (45 mM) was added dropwise into 0.5 wt% CNC suspension at pH = 4.5 to yield a sulfate-to-LAE molar ratio of 1:4 (surfactant-to-CNC weight ratio of 0.39:1). As suggested in our previous study (Chi and Catchmark 2017), this molar ratio corresponded to the complete ionic binding of LAE to the CNC surface, meaning that all the sulfate groups on the CNC surface were electrostatically bonded to the cationic head groups of LAE

123

Cellulose

with hydrophobic alky chains facing outward, which increased the hydrophobicity of CNC. This complete ionic bonding of the surfactant results in a surfactant monolayer anchored on the CNC surface. The aqueous mixture was magnetically stirred at room temperature for 24 h to ensure complete coupling of surfactant to the CNC surface via ionic binding. Afterward, the mixture was vacuum filtered through a Buchner funnel using a 0.45 lm filter membrane, washed once and dialyzed against ultra-pure water to remove any unbound surfactant monomer and NaCl formed during surfactant adsorption. Finally, the surface functionalized CNC (denoted as F-CNC) was freeze-dried and redispersed in DCM at the concentration of 1% (w/v), and the resultant suspension was ultrasonicated by an ultrasonic processor (Sonics&Materials, Vibra-Cell, model VCX750) for 10 min at 50% amplitude. As a comparison, the pristine CNC (P-CNC) was freezedried, redispersed in DCM and ultrasonicated at the same conditions as the F-CNC. Preparation of nanocomposite films Nanocomposite films, including P-CNC reinforced PLA (P-CNC/PLA) and F-CNC reinforced PLA (FCNC/PLA), were fabricated by solution casting method. The PLA stock solution (10%, w/v) was prepared via dissolving the dried PLA pellets into DCM followed by stirring at 60 °C overnight to ensure complete dissolution. Then, desired amount of 1% (w/ v) P-CNC (or F-CNC) suspension was added into the PLA stock solution to obtain a 1 g mixture (dry weight) with the P-CNC (or F-CNC) loading level of 1, 3, 5, 10 or 15% (w/w). The mixture was thoroughly stirred at room temperature for at least 8 h, sonicated in a bath sonicator for 5 min, and casted into a glass petri dish (diameter = 10 cm). After two-day evaporation at ambient conditions in a chemical hood, the nanocomposite film with a thickness of *0.75 to 0.8 mm was peeled off from the glass petri dish and further dried in a vacuum oven (25 psi) at 60 °C for 48 h to remove the residual solvent. Finally, the solution-casted film was sandwiched between Teflon sheets and placed between two stainless steel plates, and compression molded using a Carver heated press at 170 °C for 3 min with a compression weight of 2 t. The hot pressed film was allowed to cool down inside the steel plates at ambient conditions until reaching the room temperature (from 170 °C to RT in 10 min). The

123

pure PLA film was also prepared following the same procedures as described above. All obtained films with approximate thickness of 0.1 mm were conditioned in a desiccator at ambient temperature for at least 3 days before any further analysis to minimize moisture adsorption. Characterization Atomic force microscopy (AFM) was used to evaluate the dispersion state of F-CNC in DCM. Before imaging, the previously prepared stock F-CNC/DCM mixture (1%, w/v) was diluted to 0.02 mg/mL by fresh DCM solvent and ultrasonicated for 5 min. A drop (10 lL) of diluted suspension was deposited onto a freshly cleaved mica substrate and dried in the air. A Dimension Icon AFM (Bruker, CA, USA) operated in ScanAsyst and PeakForce tapping mode was used for the imaging. Sample was scanned at ambient conditions (scan rate of 0.5 Hz; image resolution: 512 9 512 pixels; peak force: 500 pN) using SCANASYST-AIR ? probe. The images were processed by a Nanoscope Analysis software (Version 1.40). The pristine CNC dispersed in water was also imaged by AFM according to the aforementioned protocol. The optical property of pure PLA and nanocomposite films was assessed using a UV/Vis/NIR spectrophotometer (PerkinElmer, model LAMBDA 1050). The films with the thickness of *0.1 mm were placed in the test cell and the air was used as reference. The light transmittance of the samples was measured in the wavelength range between 300 and 800 nm at a scan rate of 200 nm/min. Three replicates for each sample were examined to ensure reproducibility. To study the distribution and interface compatiblization of P-CNC and F-CNC in PLA matrix, the cross section of film samples was imaged by a Nova NanoSEM 630 Field emission scanning electron microscopy (FE-SEM, FEI, Nova NanoSEM, USA) operating at an accelerating voltage of 5 kV, a beam current of 28 pA and a working distance of 3.5 mm. The samples were frozen in liquid nitrogen, fractured in bending and sputter coated with a thin layer (*3 to 5 nm) of iridium. The X-ray diffraction (XRD) measurements were performed on a PANalytical Empyrean diffractometer (PANalytical, USA) with Cu Ka radiation generated at 45 kV and 40 mA.

Cellulose

The diffractometer was operated in reflection mode with the automatic divergence slit. The XRD patterns were recorded at a rate of 2°/min over 2h angles from 5° to 40° with a step size of 0.026°. The non-isothermal crystallization behavior of film samples was studied using a differential scanning calorimeter (MDSC, Q2000, TA instruments, USA) under a nitrogen atmosphere (flow rate of 50 mL/min). Each sample (*5 to 8 mg) was scanned at 10 °C/min under the heat/cool/heat cycles. Specifically, each specimen was first heated to 190 °C and maintained at that temperature for 5 min to eliminate the previous thermal history. Subsequently, the sample was cooled to -20 °C, held for 3 min, and heated up to 190 °C. The main thermal parameters, such as cold crystallization temperature (Tcc ), melting temperature (Tm ), cold crystallization enthalpy (DHcc ) and melting enthalpy (DHm ), were determined from the second heating curves. The crystallinity (vc ) of each sample was calculated using the following equation: vc ¼

ðDHm  DHcc Þ  100% DHm0  w

where DHm is the enthalpy of melting, DHcc is the enthalpy of cold crystallization, DHm0 is the melting enthalpy of pure 100% crystalline PLA, as determined to be 93.7 J/g (Yasuniwa et al. 2008), and w is the weight fraction of PLA in the film sample. Thermomechanical properties of film samples were investigated by dynamic mechanical analysis (DMA) conducted on a dynamic mechanical analyzer (Q800, TA instruments, USA) in tension mode. Rectangular samples with the dimensions of 40 9 5 9 0.1 mm (L 9 W 9 T) were heated from 20 to 120 °C at a heating rate of 3 °C/min. Each test was performed at a constant frequency of 1 Hz and a strain amplitude of 0.05%. Thermomechanical data including the storage modulus (E0 ) at 20 and 90 °C, respectively, and the glass transition temperature (Tg) (determined from the peak of Tan d) were obtained. At least two measurements for each sample were performed to ensure repeatability. Tensile properties of nanocomposite films were evaluated at ambient conditions using the same DMA instrument. The dimensions of the specimen were the same as that for DMA tests. Each measurement was conducted at a strain ramp rate of 5%/min and a gauge length of 10 mm until break. Tensile strength,

Young’s modulus and strain at break were determined from the stress–strain curves. At least five replicates were characterized for each sample. The significant difference of the data was evaluated using the Tukey’s range tests with p \ 0.05 confident interval (Minitab Statistical Software; Release 17.3; Penn State University, University Park, PA).

Results and discussion Dispersion state of functionalized CNC in dichloromethane The dispersion test can be an effective and straightforward method to detect the surface polarity change of CNC induced by the adsorption of surfactant. Figure 1a shows the pristine CNC (P-CNC) and functionalized CNC (F-CNC) dispersed in water (left) and dichloromethane (DCM, right), respectively, and the corresponding AFM images are represented in Fig. 1 b and c. The P-CNC was a highly hydrophilic material and could be well-dispersed in a polar solvent such as water (polarity index of 10.2) due to the electrostatic repulsion from the negatively charged surface. These nanoparticles were approximately 100–250 and 2–5 nm in length and diameter, respectively, as estimated from AFM image (Fig. 1b). After surfactant adsorption, the F-CNC formed a homogeneous and stable suspension in a less polar solvent (DCM, polarity index of 3.1) and remained welldispersed after 24 h standing, indicating the surface polarity change (increased hydrophobicity) of CNC. The AFM image of F-CNC (Fig. 1c) showed that most modified nanoparticles were isolated and individualized with similar length (*150 nm) and diameter (*4 nm) as that of P-CNC. Nonetheless, small laterally associated aggregates/or bundles (bundle width *30 to 80 nm) were still observed and could scatter the light strongly, as evidenced by the turbid appearance of F-CNC suspension. Ansari et al. (2015) employed a cationic quaternary ammonium surfactant to surface modify CNC and the modified CNC showed well-individualized nanoparticles in toluene (polarity index of 2.4) without obvious agglomerates (from TEM image). In their study, the charge density of sulfate groups was 0.5 mmol/g, twice as high as that of CNC used in this study. It is speculated that high charge density of sulfate groups on the CNC surface

123

Cellulose

Fig. 1 (a) Photographs of suspensions of P-CNC and F-CNC in water and DCM, respectively (after 24 h); AFM amplitude images of P-CNC (b) and F-CNC (c) in selected solvents

may adsorb more surfactant monomer through ionic binding, exposing more hydrophobic tail groups to the outside of CNC. Therefore, the F-CNC is less hydrophobic as compared to that in Ansari’s work, which might be a possible explanation for the formation of small agglomerates in DCM. Additionally, it suggests that the surface polarity of CNC could be engineered by controlling the amount of adsorbed surfactant, assuming their interaction is electrostatic in nature. It is worth mentioning that the change in hydrophobicity of F-CNC in our study is achieved based on the appropriate mixing ratio of CNC and surfactant. In our previous work, we systematically investigated the binding interactions between sulfated CNC and LAE and their underlying binding mechanism (Chi and Catchmark 2017). It was found that the complete ionic binding occurred at the molar ratio of 1:4 (sulfate groups:LAE) from the isothermal titration calorimetry results (Chi and Catchmark 2017). The excess addition of surfactant could induce the surfactant aggregates on the CNC surface either in the form of admicelle or micelle (Brinatti et al. 2016; Chi and Catchmark 2017), undermining the hydrophobicity and thermal stability of the modified CNC as revealed from contact angle and TGA results (Chi and Catchmark 2017), respectively. Our result is consistent with several recent studies. Abitbol et al. (2014) studied the interactions between cationic cetyltrimethylammonium bromide (CTAB, another type of quaternary ammonium surfactant) and sulfated CNC, and found the 100% coupling efficiency of surfactant to CNC via ionic binding was achieved at the molar ratio of 1:4 (sulfate:CTAB) from the elementary analysis. Ansari

123

et al. (2015) compared the dispersion quality of CNCs modified with various CNC/cationic surfactant molar ratios in toluene in order to optimize the amount of surfactant. They concluded that a molar ratio of 1:4 (CNC sulfate groups: cationic surfactant) offered a stable suspension with high transmittance. When excessive surfactant was mixed with CNC (molar ratio of 1:8 or 1:12), the transmittance of the resultant suspensions decreased, possibly ascribed to the formation of surfactant micelles at high molar ratios. Therefore, it is critical to optimize the amount of surfactant required to ionically coat CNC as excessive surfactant could act as plasticizer and undermine the properties of the ensuing nanocomposites (Bondeson and Oksman 2007; Kim et al. 2009), as well as increase cost. Optical property The optical transparency of nanocomposite films is a good indicator of the dispersion quality of nanofiller in the polymer matrix. The visual appearance (Fig. 2a) of neat PLA and various nanocomposite films was examined to qualitatively assess the dispersibility of P-CNC and F-CNC in PLA matrix. The neat PLA film is highly transparent, making it widely used in the packaging industry. When incorporating P-CNC or F-CNC, the clarity of the PLA film was impaired, especially for P-CNC at loading levels larger than 3%. The PLA film filled with 15% P-CNC appeared to be extremely opaque, suggesting the formation of large CNC aggregates in the nanocomposite film. The F-CNC/PLA films, on the other hand, maintained the transparency of neat PLA film up to the loading level

Cellulose

Fig. 2 Photographs (a) and UV–Vis spectra of neat PLA and nanocomposite films reinforced with various loading levels of P-CNC (b) and F-CNC (c)

of 5%. With the further increase of F-CNC contents, the transparency of nanocomposite films was gradually lost, but still more transparent than the same content level of P-CNC. Further, the light transmittance of neat PLA and nanocomposite films was analyzed by UV–Vis spectroscopy (Fig. 2b, c). It is straightforward that the poor distribution of P-CNC in PLA matrix caused large CNC agglomerates, which scattered the light strongly and resulted in the decrease in light transmittance (Fig. 2b). At the wavelength of 550 nm, the light transmittance of neat PLA film (91%) decreased to 75 and 52%, respectively, with addition of 1 and 3% P-CNC. Further increase of the P-CNC content (5–15%) in PLA matrix caused the transmittance of nanocomposite films to be reduced to less than 20%. For the F-CNC/PLA nanocomposites, the transmittance of films were above 50% at a P-CNC loading level of 1–10%, indicating the better

dispersion of F-CNC in PLA matrix. When the F-CNC content reached 15%, the transmittance of nanocomposite film decreased to 45%, probably attributed to the nanofiller aggregation or increased crystallinity (as seen from the XRD results) (Herrera et al. 2015). Overall, the UV–Vis transparency results suggest that enhanced dispersion of CNC in the PLA matrix can be achieved by surface adsorption of surfactant, in agreement with previous studies (Ansari et al. 2015; Fortunati et al. 2012; Nagalakshmaiah et al. 2016a). Morphology The fractured morphologies (Fig. 3) of neat PLA and its nanocomposite films further confirmed the improved dispersion and interface compatibility of nanofiller within the matrix when surfactant was used.

123

Cellulose

Fig. 3 SEM images of the fracture cross-section of neat PLA (a, a1) and PLA nanocomposites reinforced with 5 wt% P-CNC (b, b1), 5 wt% F-CNC (c, c1) and 10 wt% F-CNC (d, d1)

The neat PLA film (Fig. 3a, a1) presented a clear and smooth fractured surface, indicating its brittle nature. PLA reinforced with 5 wt% P-CNC exhibited inhomogeneous distribution of P-CNC as indicated by large flake-shaped agglomerates consisting of multiple CNCs (Fig. 3b1). Moreover, large holes between P-CNC agglomerates and the PLA matrix were observed (Fig. 3b). These behaviors have been observed in many studies and were ascribed to the self-aggregation of CNC by inter-nanoparticle hydrogen bonding and interface incompatibility between hydrophilic CNC and hydrophobic matrices (Ansari et al. 2015; Bondeson and Oksman 2007; Kim et al. 2009; Nagalakshmaiah et al. 2016a, b). On the other hand, nanocomposite film reinforced with 5 wt% F-CNC showed much more uniform dispersion of nanofiller (Fig. 3c, white dots) and even individualized F-CNC within the matrix (Fig. 3c1). No obvious cavity between F-CNC and PLA matrix was observed. It is hypothesized that this improved dispersion and interface adhesion of F-CNC in nanocomposite films are a result of ionically adsorbed LAE monomers on the CNC surface, which prevent the nanoparticle selfaggregation and improve the interface compatibility through their hydrophobic tails. However, when the content of F-CNC was increased to 10 wt%, aggregation was still observed (Fig. 3d1), but to a lesser extent as displayed in unmodified CNC (Fig. 3b1). In addition, for 10 wt% F-CNC holes were observed again

123

(Fig. 3d1), which might be ascribed to the reduced interface interaction when F-CNC aggregates were present in the matrix. Such behavior has been observed in a previous study, which incorporated excessive surfactant modified CNC (10 wt%) into a polypropylene matrix (Nagalakshmaiah et al. 2016a). Crystalline structure The XRD patterns of neat PLA and its nanocomposite films are represented in Fig. 4. The crystallized PLA was also prepared and used as a reference. As seen from Fig. 4a, the sulfated CNC displayed three main cellulose Ib characteristic peaks at 2h angles around  (110) 14.7°, 16.6°, and 22.6°, representing the (110), and (200) crystallographic planes, respectively (Yu et al. 2014). The crystallized PLA exhibited typical diffraction peaks of a-form crystal. Diffraction peaks at 2h = 12.7°, 14.9°, 16.8°, 19.2° and 22.5° were corresponded to the crystallographic planes of (103), (010), (200)/(110), (203), (015), respectively (Yasuniwa et al. 2008). As for the neat PLA film, the diffraction pattern exhibited a broad bump without any discernable diffraction peaks, indicating its amorphous structure. It is not surprising that the neat PLA film is in the amorphous state as PLA typically has a slow crystallization rate (Rasal et al. 2010). When the PLA was incorporated with various contents of

Cellulose

Fig. 4 XRD patterns of (a) CNC, neat and crystallized PLA films; (b) P-CNC/PLA nanocomposite films; (c) F-CNC/PLA nanocomposite films. The crystallized PLA film was prepared by annealing treatment at 130 °C for 1 h

P-CNC, the diffraction patterns of nanocomposite films (Fig. 4b) showed increased peak intensity at 2h = 22.6°, which was the characteristic peak of incorporated P-CNC. The broad bumps were observed and the diffraction peaks at 2h = 16.8° and 19.2° (two prominent diffraction peaks of crystallized PLA) were absent in all P-CNC/PLA nanocomposite samples, suggesting that the PLA matrix was still amorphous in the presence of P-CNC. However, when F-CNC was added into the PLA matrix with different loading levels, the resultant nanocomposite films exhibited significant changes in diffraction patterns (Fig. 4c), as displayed by the gradual increase of the intensity of crystalline peaks at 2h = 16.8° and 19.2°, implying the increased crystallinity of nanocomposite samples. It is, therefore, concluded that the addition of F-CNC can promote the crystallization of PLA while the P-CNC has no such effect. Thermal properties The non-isothermal crystallization of neat PLA and nanocomposite films was studied by DSC. The second heating scan is usually used to access the thermal transitions in the material. DSC thermograms are shown in Fig. 5 and the characteristic thermal data are collected in Table 1. In Fig. 5, the glass transition, cold crystallization and melting of PLA matrix can be clearly seen, which are typical thermal transitions for a

semi-crystalline polymer. All DSC traces show a constant glass transition temperature (Tg) of *61 °C, independent of the P-CNC or F-CNC content, possibly attributed to the fact that DSC is less sensitive in revealing the molecular-level motion of polymer chains as compared to the DMA. The melting peaks of all P-CNC/PLA nanocomposite films were around 164.5 °C, independent of the content change of P-CNC. For all F-CNC/PLA nanocomposite films, the meting peaks showed a slight shift to higher temperature around 166.6 °C. It is also noted that there are two melting peaks on the DSC curves of P-CNC/ PLA (15 wt% P-CNC) and F-CNC/PLA (1 and 3 wt%) samples. This double melting behavior of PLA has been explained by different mechanisms, such as melt/ recrystallization model, presence of multiple crystal forms and crystalline lamellae populations with different thickness or perfection (Frone et al. 2016; Goffin et al. 2011). For the modified-CNC reinforced PLA composites, the presence of double melting peaks is often ascribed to the melt/recrystallization model (Habibi et al. 2013). The melting peak at lower temperature is probably attributed to less perfect or disordered a0 -form crystal, while the one at higher temperature might be assigned to the more perfected a-form crystal (Johari et al. 2016). During the DSC heating scan, the a0 -form crystals have enough time to melt and evolve into highly ordered and perfect aform crystals before re-melting at higher temperature.

123

Cellulose Fig. 5 DSC thermograms of neat PLA, (a) P-CNC/ PLA, and (b) F-CNC/PLA nanocomposite films

Table 1 Characteristic DSC and DMA parameters of PLA and its nanocomposite films Sample

Tcc (oC)

Tm (oC)

DHcc (J/g)

DHm (J/g)

vc (%)

E0 (20 °C) (MPa)

Neat PLA P-CNC 1%

129 128

164.3 164.8

28.1 27.4

30.9 30.8

3 3.7

2157 2278

P-CNC 3%

127

164.4

28.6

32.7

4.5

2286

9

73.1

P-CNC 5%

126

164.7

24.6

30.6

6.7

2519

17

71.5

P-CNC 10%

120

164.2

28.2

32.8

5.5

2816

22

72.7

E0 (90 °C) (MPa) 3 6

Tg (oC) 70.6 72.9

P-CNC 15%

116

164.3

27.5

30.7

4

3025

40

73.5

F-CNC 1%

111

167.1

28.7

36

7.9

2815

90

74.8

F-CNC 3%

108

166.8

24.8

34.9

11.1

3286

190

78.2

F-CNC 5%

105

166.8

20.8

33.4

14.2

4262

266

82.0

F-CNC 10%

102

166.3

19.5

34.5

17.8

3818

354

79.3

F-CNC 15%

98

166.1

15

37.8

28.6

3801

421

80.7

Interestingly, we also notice that the areas under the first and second melting peaks for P-CNC/PLA (15 wt%) and F-CNC/PLA (1–3 wt%) samples are different. It has been suggested that the areas under double melting peaks are directly associated with the crystal perfection, which is affected by the loading level and nature of CNC in the PLA matrix (Goffin et al. 2011). In our study, P-CNC/PLA (15 wt%) and F-CNC/PLA (1–3 wt%) samples are different in nanofiller characteristics (content, dispersion state and interfacial compatibility) and crystallinity. For P-CNC, due to its high aggregation state and limited nucleating role in the PLA matrix, the crystal perfection might be inhibited and only a small amount of a0 -form crystal could be reorganized into highly ordered a-form crystal (smaller area under the second melting peaks).

123

F-CNC provided much better dispersion and interfacial compatibility in the PLA matrix, which might largely contribute to the crystal perfection process (larger area under the second melting peaks). Besides, an increase in the content of F-CNC in PLA matrix induced an increase in the relative content of more perfect a-form crystal, in agreement with previous studies (Goffin et al. 2011; Habibi et al. 2013). A striking change was found for the cold crystallization temperature (Tcc) of nanocomposite films. Both P-CNC and F-CNC could promote the cold crystallization of PLA as represented by the shift of Tcc to lower temperatures. High loading levels of P-CNC or F-CNC caused larger Tcc decrease due to larger nucleus-promoting surface imparted by nanofiller. The addition of 15 wt% P-CNC decreased the

Cellulose

Tcc from 129 (neat PLA) to 116 °C, while addition of 15 wt% F-CNC had a more prominent effect (Tcc = 98 °C). Besides, the crystallinity of PLA increased by incorporating P-CNC or F-CNC due to the nucleating effect of nanofiller. Compared to P-CNC, F-CNC led to a notably greater crystallinity increase at the same concentration level. In practical applications, PLA is well known for its slow crystallization rate. As such, different strategies have been employed, such as adding nucleation agents, plasticizer, and annealing treatment (Murariu and Dubois 2016). The presence of nanofiller in PLA based nanocomposite not only significantly improves the mechanical performance but also acts as nucleation agent for the promotion of crystallization. The nucleating role of CNC and its inducted crystallization have been widely reported for various polymer matrices (Fortunati et al. 2012; Fujisawa et al. 2013; Habibi et al. 2008; Xu et al. 2017). For example, Fortunati et al. (Fortunati et al. 2012) found that the crystallization of PLA was significantly promoted in the presence of surfactant modified CNC as evidenced by the decrease of Tcc and increased crystallinity, similar to our results. It appears that the nucleating effect is strongly dependent on the uniform distribution of CNC, which results in larger interface area for nucleation. Therefore, in our study, the surfactant modified CNC has a stronger nucleating effect than pristine CNC due to its improved dispersibility and interface compatibility within PLA matrix. Thermomechanical properties The impact of P-CNC or F-CNC on the dynamic mechanical properties of nanocomposite films was investigated by DMA. Figure 6 shows the evolution of storage modulus (E0 , Fig. 6a, b) and Tand (Fig. 6a1, b1) as a function of temperature for various film samples. The corresponded E0 at glassy (20 °C) and rubbery state (90 °C) and a-relaxation temperatures (glass transition temperatures, Tg) of neat PLA and nanocomposite samples are listed in Table 1. Neat PLA film displayed a typical behavior of amorphous thermoplastic polymer with a high E0 (2157 MPa) at glassy region and a sharply decreased E0 above Tg at 70.6 °C. The storage modulus reached a plateau (*3 MPa) at the temperature range of 85–95 °C, and then increased progressively after 95 °C due to the cold crystallization (Zhang et al. 2012). With the

addition of P-CNC, the E0 of nanocomposite films at both glassy and rubbery states were increased. At a loading level of 15 wt% P-CNC, E0 of nanocomposite film was improved by 40% at glassy state and 12.3 times at rubbery state, respectively. This much more significant improvement of E0 at rubbery state can be ascribed to the higher reinforcing efficiency of CNC when the polymer matrix become softened at high temperature (Suryanegara et al. 2009). When PLA matrix was filled with F-CNC, extraordinary enhancement of E0 at glassy and rubbery regions was observed. The maximum E0 of nanocomposite films at glass state was 4262 MPa, enhanced by 98% as compared to the neat PLA film, when incorporating 5 wt% F-CNC. E0 of nanocomposite films at rubbery state steeply increased with the increase of F-CNC content and reached to 421 MPa, approximate 139 times higher than that of neat PLA, when 15% F-CNC was present in the matrix. In the presence of nanofiller, the increased E0 of polymer matrix at glassy and rubbery states might be correlated with the formation of rigid continuous nanofiller percolation network and nanofiller induced restriction of polymer chain mobility, respectively, according to previous work (Tingaut et al. 2009). Compared with P-CNC, the exceptional higher reinforcing effect induced by F-CNC was found, which could be explained by the fact that F-CNC had enhanced dispersion and interface compatibility within the matrix as indicated by SEM images. However, it should be noted that the dramatically enhanced E0 of F-CNC reinforced nanocomposite films is the consequence of both increase in crystallinity of PLA matrix as evidenced by XRD results and nanoreinforcement imparted by welldispersed rigid F-CNC network (Nagalakshmaiah et al. 2016a; Suryanegara et al. 2009). Glass transition temperature (Tg), determined from the peak of Tand curves (Fig. 6a1, b1), is a critical parameter to determine the properties and applications of a material. Tg is very sensitive to the interface bonding between polymer matrix and nanofiller. Generally, the uniform dispersion and good interface bonding of nanofiller within the matrix can result in a higher Tg due to the formation of an interphase layer surrounding the nanofiller, restricting the mobility of polymer chains (Kaleemullah et al. 2012). As seen from Table 1, the incorporation of P-CNC did not significantly alter the Tg of PLA matrix, mainly attributed to the poor distribution and interface

123

Cellulose

Fig. 6 Storage modulus (a, b) and Tan d (a1, b1) curves of neat PLA and its nanocomposite films

incompatibility of P-CNC as previously observed from SEM images (Fig. 3b, b1). In the case of nanocomposites reinforced with F-CNC, Tg shifted to higher temperature with the increasing F-CNC content from 70.6 °C (neat) to 74.8 °C (1 wt% F-CNC), 78.2 °C (3 wt% F-CNC) and 82 °C (5 wt% F-CNC), implying the improved nanofiller dispersion and interface compatibility effectively confine the motion of polymer chains. It is also worth noting that Tg of F-CNC/PLA nanocomposite films level off when the content of F-CNC exceeds 5 wt%, possibly attributed to the nanofiller aggregation as suggested in previous study (Qiao et al. 2011). Tensile properties Representative stress–strain curves of neat PLA and its nanocomposite films are presented in Fig. 7, and tensile strength, Young’s modulus and strain at break are summarized in Table 2. The tensile strength and Young’s modulus of neat PLA film were 41.95 MPa and 2.05 GPa, respectively. With the addition of P-CNC, the Young’s moduli of nanocomposite films slightly increased, and followed the trend observed in E0 at glassy state. The Young’s modulus of P-CNC/ PLA with 15 wt% P-CNC was enhanced by 34% compared to that of neat PLA film (statistically distinguishable with p \ 0.05), primarily due to the rigidity of nanocrystals. It is well known that the

123

increase in crystallinity can also strengthen the tensile strength and modulus of PLA (Suryanegara et al. 2009). Our XRD results showed that all the nanocomposite films reinforced with P-CNC were amorphous in nature. Thus, the influence of crystallinity is excluded and the improvement in Young’s modulus is mainly from the reinforcement effect of P-CNC. However, the tensile strength of P-CNC/PLA nanocomposite films were undermined, especially when the content of P-CNC was higher than 5 wt%. The tensile strength of P-CNC/PLA with 10 and 15 wt% P-CNC were statistically lowered (p \ 0.05) by 26 and 24%, respectively. It is likely that large P-CNC agglomerates act as stress concentration points, which is responsible for the impairment of tensile strength. Favier et al. (1995b) has proposed a widely accepted percolation model to explain the reinforcing mechanism for CNC filled nanocomposites. With uniform distribution of nanofiller and good interface bonding between nanofiller and matrix, CNC can form a continuous stiff percolation network above the percolation threshold, which effectively enables the stress transfer from soft matrix to rigid nanofiller (Favier et al. 1995b). In the case of P-CNC/PLA nanocomposites, the poor distribution of nanofiller and lack of intimate bonding between nanofiller and matrix resulted in large voids and nanofiller aggregates within the composite structure, negatively impacting the stress transfer from matrix to nanofiller. In

Cellulose Fig. 7 Representative stress-stain curves of PLA and its nanocomposite films: (a) P-CNC/PLA and (b) FCNC/PLA

Table 2 Tensile properties of PLA and its nanocomposite films

Sample

Young’s modulusa (GPa)

Neat PLA

2.1 ± 0.2G

P-CNC 1%

Mean values with different superscripts for each property are significantly different at the 5% significance level

b

Maximum strength from stress–strain curves

2.1 ± 0.1

7.9 ± 0.8A

DE

4.3 ± 0.6B

DE

40.7 ± 2.3

2.1 ± 0.2

39.2 ± 2.3

2.9 ± 0.5C

P-CNC 5%

2.3 ± 0.2FG

36.4 ± 2.1EF

2.3 ± 0.3CD

P-CNC 15% F-CNC 1%

G

42 ± 0.9D

Strain at break (%)

P-CNC 3% P-CNC 10% a

G

Tensile strengthb (MPa)

EFG

2.5 ± 0.2

DE

2.7 ± 0.2

EF

2.6 ± 0.3

1.5 ± 0.2CD

FG

1.4 ± 0.2D

C

6.8 ± 1.6A

B

31 ± 2.3 31.9 ± 1.3 47.8 ± 2.3

F-CNC 3% F-CNC 5%

3.1 ± 0.2 4.1 ± 0.2A

57.1 ± 3.2 69.6 ± 3.3A

4.6 ± 0.4B 2.6 ± 0.4CD

F-CNC 10%

3.7 ± 0.2B

42.9 ± 1.7D

1.5 ± 0.1D

F-CNC 15%

CD

G

BC

DE

3.5 ± 0.2

contrast, the F-CNC/PLA nanocomposite films showed much better tensile properties. With 5 wt% F-CNC, the tensile strength and Young’s modulus of nanocomposite films were enhanced by 66 and 100%, respectively, and statistically different from those of neat PLA film. This notable improvement in tensile strength and Young’s modulus is probably the result of increased crystallinity, and improved dispersion/interface compatibility of F-CNC within the PLA matrix, as indicated by the SEM and XRD results. Nonetheless, with the continuous increase of nanofiller contents, aggregation is inevitable, even for surfactant modified CNC (Ansari et al. 2015; Nagalakshmaiah et al. 2016a). Beyond 5 wt% F-CNC, nanofiller aggregation led to the decrease in both tensile strength and Young’s modulus. Besides, the increase in brittleness with the increase of nanofiller content was found for both P-CNC/PLA and F-CNC/ PLA nanocomposite films, which is a common behavior for many nanocellulose reinforced polymer

40 ± 2.2

1.5 ± 0.1D

nanocomposites (Dhar et al. 2016; Habibi et al. 2008; Ljungberg et al. 2005).

Conclusions Sulfuric acid-extracted CNC was successfully surface modified by a novel biologically-derived cationic surfactant (LAE) via complete ionic binding using a simple and environmental benign mixing procedure. The formation of stable and homogenous suspension of modified CNC (F-CNC) in non-polar dichloromethane (DCM) indicated the increase in surface hydrophobicity of CNC due to ionically bound LAE. AFM analysis revealed that most F-CNC were welldispersed and individualized in DCM, yet some aggregates were still observed. Thereafter, both unmodified (P-CNC) and F-CNC reinforced PLA nanocomposite films were fabricated by solution casting and compression molding. The P-CNC, due

123

Cellulose

to its interfacial incompatibility with the non-polar PLA matrix, showed limited improvement in the mechanical and thermal properties of the resulting nanocomposite films. By contrast, F-CNC exhibited notably improved dispersibility and interface compatibility, resulting in much more significant improvement in thermal and mechanical properties. For instance, with the addition of 5 wt% F-CNC, crystallinity, Tg, E0 (glassy and rubbery states) and tensile strength and modulus of resultant nanocomposite film were strongly increased. This strong enhancement is ascribed to a well-dispersed continuous F-CNC percolation network, which promotes the crystallization of PLA and enables the stress-transfer process upon loading. It is also noted that the high F-CNC content ([5 wt%) could still cause the nanoparticle selfaggregation, inducing the decrease in Tg and tensile properties. In conclusion, it is expected that the LAE modified CNC with appropriate loading level can serve as a promising nanofiller for enhancing the mechanical and thermal properties of PLA. Acknowledgments We gratefully acknowledge the financial support from USDA Forest Service (Agreement No. 11-JV11111129-121) and the Penn State College of Agricultural Sciences Graduate Student Competitive Grants Program. We also thank A&B Ingredients for providing the lauric arginate.

References Abitbol T, Marway H, Cranston ED (2014) Surface modification of cellulose nanocrystals with cetyltrimethylammonium bromide. Nordic Pulp Pap Res J 29:046–057 Ansari F, Salajkova´ M, Zhou Q, Berglund LA (2015) Strong surface treatment effects on reinforcement efficiency in biocomposites based on cellulose nanocrystals in poly(vinyl acetate) matrix. Biomacromolecules 16:3916–3924 Ben Azouz K, Ramires EC, Van den Fonteyne W, El Kissi N, Dufresne A (2011) Simple method for the melt extrusion of a cellulose nanocrystal reinforced hydrophobic polymer. ACS Macro Lett 1:236–240 Bondeson D, Oksman K (2007) Dispersion and characteristics of surfactant modified cellulose whiskers nanocomposites. Compos Interfaces 14:617–630 Brinatti C, Huang J, Berry RM, Tam KC, Loh W (2016) Structural and energetic studies on the interaction of cationic surfactants and cellulose nanocrystals. Langmuir 32:689–698 Chi K, Catchmark JM (2017) Crystalline nanocellulose/lauric arginate complexes. Carbohydr Polym 175:320–329 Choi K-M, Choi M-C, Han D-H, Park T-S, Ha C-S (2013) Plasticization of poly(lactic acid) (PLA) through chemical

123

grafting of poly(ethylene glycol) (PEG) via in situ reactive blending. Eur Polym J 49:2356–2364 Dhar P, Tarafder D, Kumar A, Katiyar V (2016) Thermally recyclable polylactic acid/cellulose nanocrystal films through reactive extrusion process. Polymer 87:268–282 Dufresne A (2013) Nanocellulose: a new ageless bionanomaterial. Mater Today 16:220–227 Dufresne A (2017) Cellulose nanomaterial reinforced polymer nanocomposites. Curr Opin Colloid Interface Sci 29:1–8 Farah S, Anderson DG, Langer R (2016) Physical and mechanical properties of PLA, and their functions in widespread applications—a comprehensive review. Adv Drug Deliv Rev 107:367–392 Favier V, Canova GR, Cavaille´ JY, Chanzy H, Dufresne A, Gauthier C (1995a) Nanocomposite materials from latex and cellulose whiskers. Polym Adv Technol 6:351–355 Favier V, Chanzy H, Cavaille JY (1995b) Polymer nanocomposites reinforced by cellulose whiskers. Macromolecules 28:6365–6367 Follain N, Marais M-F, Montanari S, Vignon MR (2010) Coupling onto surface carboxylated cellulose nanocrystals. Polymer 51:5332–5344 Fortunati E et al (2012) Multifunctional bionanocomposite films of poly(lactic acid), cellulose nanocrystals and silver nanoparticles. Carbohydr Polym 87:1596–1605 Frone AN, Panaitescu DM, Chiulan I, Nicolae CA, Vuluga Z, Vitelaru C, Damian CM (2016) The effect of cellulose nanofibers on the crystallinity and nanostructure of poly(lactic acid) composites. J Mater Sci 51:9771–9791 Fujisawa S, Saito T, Kimura S, Iwata T, Isogai A (2013) Surface engineering of ultrafine cellulose nanofibrils toward polymer nanocomposite materials. Biomacromolecules 14:1541–1546 Fukushima K, Kimura Y (2006) Stereocomplexed polylactides (Neo-PLA) as high-performance bio-based polymers: their formation, properties, and application. Polym Int 55:626–642 Goffin A-L, Raquez J-M, Duquesne E, Siqueira G, Habibi Y, Dufresne A, Dubois P (2011) From interfacial ring-opening polymerization to melt processing of cellulose nanowhisker-filled polylactide-based nanocomposites. Biomacromolecules 12:2456–2465 Grunert M, Winter WT (2002) Nanocomposites of cellulose acetate butyrate reinforced with cellulose nanocrystals. J Polym Environ 10:27–30 Habibi Y, Goffin A-L, Schiltz N, Duquesne E, Dubois P, Dufresne A (2008) Bionanocomposites based on poly(ecaprolactone)-grafted cellulose nanocrystals by ringopening polymerization. J Mater Chem 18:5002–5010 Habibi Y, Aouadi S, Raquez J-M, Dubois P (2013) Effects of interfacial stereocomplexation in cellulose nanocrystalfilled polylactide nanocomposites. Cellulose 20:2877–2885 Hasani M, Cranston ED, Westman G, Gray DG (2008) Cationic surface functionalization of cellulose nanocrystals. Soft Matter 4:2238–2244 Herrera N, Mathew AP, Oksman K (2015) Plasticized polylactic acid/cellulose nanocomposites prepared using melt-extrusion and liquid feeding: mechanical, thermal and optical properties. Compos Sci Technol 106:149–155

Cellulose Heux L, Chauve G, Bonini C (2000) Nonflocculating and chiralnematic self-ordering of cellulose microcrystals suspensions in nonpolar solvents. Langmuir 16:8210–8212 Huang L, Ye Z, Berry R (2016) Modification of cellulose nanocrystals with quaternary ammonium-containing hyperbranched polyethylene ionomers by ionic assembly. ACS Sustain Chem Eng 4:4937–4950 Hubbe MA, Rojas OJ, Lucia LA (2015) Green modification of surface characteristics of cellulosic materials at the molecular or nano scale: a review. BioResources 10:6095–6206 Iwatake A, Nogi M, Yano H (2008) Cellulose nanofiber-reinforced polylactic acid. Compos Sci Technol 68:2103–2106 Jiang L, Wolcott MP, Zhang J (2006) Study of biodegradable polylactide/poly(butylene adipate-co-terephthalate) blends. Biomacromolecules 7:199–207 Johari AP, Mohanty S, Kurmvanshi SK, Nayak SK (2016) Influence of different treated cellulose fibers on the mechanical and thermal properties of poly(lactic acid). ACS Sustain Chem Eng 4:1619–1629 Kaleemullah M, Khan SU, Kim J-K (2012) Effect of surfactant treatment on thermal stability and mechanical properties of CNT/polybenzoxazine nanocomposites. Compos Sci Technol 72:1968–1976 Kaushik A, Singh M, Verma G (2010) Green nanocomposites based on thermoplastic starch and steam exploded cellulose nanofibrils from wheat straw. Carbohydr Polym 82:337–345 Khan A et al (2012) Mechanical and barrier properties of nanocrystalline cellulose reinforced chitosan based nanocomposite films. Carbohydr Polym 90:1601–1608 Kim J, Montero G, Habibi Y, Hinestroza JP, Genzer J, Argyropoulos DS, Rojas OJ (2009) Dispersion of cellulose crystallites by nonionic surfactants in a hydrophobic polymer matrix. Polym Eng Sci 49:2054–2061 Klemm D, Kramer F, Moritz S, Lindstro¨m T, Ankerfors M, Gray D, Dorris A (2011) Nanocelluloses: a new family of naturebased materials. Angew Chem Int Ed 50:5438–5466 Lai S-M, Wu S-H, Lin G-G, Don T-M (2014) Unusual mechanical properties of melt-blended poly(lactic acid) (PLA)/clay nanocomposites. Eur Polym J 52:193–206 Li Z, Tan BH, Lin T, He C (2016) Recent advances in stereocomplexation of enantiomeric PLA-based copolymers and applications. Prog Polym Sci 62:22–72 Lin N, Huang J, Chang PR, Feng J, Yu J (2011) Surface acetylation of cellulose nanocrystal and its reinforcing function in poly(lactic acid). Carbohydr Polym 83:1834–1842 Lin N, Huang J, Dufresne A (2012) Preparation, properties and applications of polysaccharide nanocrystals in advanced functional nanomaterials: a review. Nanoscale 4:3274–3294 Liu H, Song W, Chen F, Guo L, Zhang J (2011) Interaction of microstructure and interfacial adhesion on impact performance of polylactide (PLA) ternary blends. Macromolecules 44:1513–1522 Ljungberg N, Bonini C, Bortolussi F, Boisson C, Heux L, Cavaille´ J-Y (2005) New nanocomposite materials reinforced with cellulose whiskers in atactic polypropylene: effect of surface and dispersion characteristics. Biomacromolecules 6:2732–2739

Mangalam AP, Simonsen J, Benight AS (2009) Cellulose/DNA hybrid nanomaterials. Biomacromolecules 10:497–504 Misra M, Pandey JK, Mohanty A (2015) Biocomposites: design and mechanical performance, 1st edn. Woodhead Publishing, Sawston Murariu M, Dubois P (2016) PLA composites: from production to properties. Adv Drug Del Rev 107:17–46 Murariu M et al (2011) High-performance polylactide/ZnO nanocomposites designed for films and fibers with special end-use properties. Biomacromolecules 12:1762–1771 Nagalakshmaiah M, El Kissi N, Dufresne A (2016a) Ionic compatibilization of cellulose nanocrystals with quaternary ammonium salt and their melt extrusion with polypropylene. ACS Appl Mater Interfaces 8:8755–8764 Nagalakshmaiah M, Pignon F, El Kissi N, Dufresne A (2016b) Surface adsorption of triblock copolymer (PEO–PPO– PEO) on cellulose nanocrystals and their melt extrusion with polyethylene. RSC Adv 6:66224–66232 Natterodt JC, Sapkota J, Foster EJ, Weder C (2017) Polymer nanocomposites with cellulose nanocrystals featuring adaptive surface groups. Biomacromolecules 18:517–525 Oksman K, Mathew AP, Bondeson D, Kvien I (2006) Manufacturing process of cellulose whiskers/polylactic acid nanocomposites. Compos Sci Technol 66:2776–2784 Paralikar SA, Simonsen J, Lombardi J (2008) Poly(vinyl alcohol)/cellulose nanocrystal barrier membranes. J Membr Sci 320:248–258 Park SH, Lee SG, Kim SH (2013) Isothermal crystallization behavior and mechanical properties of polylactide/carbon nanotube nanocomposites. Compos Part A Appl Sci Manuf 46:11–18 Pinazo A, Manresa MA, Marques AM, Bustelo M, Espuny MJ, Perez L (2016) Amino acid-based surfactants: new antimicrobial agents. Adv Colloid Interface Sci 228:17–39 Pretula J, Slomkowski S, Penczek S (2016) Polylactides— methods of synthesis and characterization. Adv Drug Del Rev 107:3–16 Qiao R, Deng H, Putz KW, Brinson LC (2011) Effect of particle agglomeration and interphase on the glass transition temperature of polymer nanocomposites. J Polym Sci Part B Polym Phys 49:740–748 Raquez JM, Murena Y, Goffin AL, Habibi Y, Ruelle B, DeBuyl F, Dubois P (2012) Surface-modification of cellulose nanowhiskers and their use as nanoreinforcers into polylactide: a sustainably-integrated approach. Compos Sci Technol 72:544–549 Raquez J-M, Habibi Y, Murariu M, Dubois P (2013) Polylactide (PLA)-based nanocomposites. Prog Polym Sci 38:1504–1542 Rasal RM, Janorkar AV, Hirt DE (2010) Poly(lactic acid) modifications. Prog Polym Sci 35:338–356 Reddy MM, Vivekanandhan S, Misra M, Bhatia SK, Mohanty AK (2013) Biobased plastics and bionanocomposites: current status and future opportunities. Prog Polym Sci 38:1653–1689 Saini P, Arora M, Kumar MNVR (2016) Poly(lactic acid) blends in biomedical applications. Adv Drug Del Rev 107:47–59 Sakakibara K, Yano H, Tsujii Y (2016) Surface engineering of cellulose nanofiber by adsorption of diblock copolymer dispersant for green nanocomposite materials. ACS Appl Mater Interfaces 8:24893–24900

123

Cellulose San Roma´n MS, Holgado MJ, Salinas B, Rives V (2013) Drug release from layered double hydroxides and from their polylactic acid (PLA) nanocomposites. Appl Clay Sci 71:1–7 Siracusa V, Rocculi P, Romani S, Dalla Rosa M (2008) Biodegradable polymers for food packaging: a review. Trends Food Sci Technol 19:634–643 Suryanegara L, Nakagaito AN, Yano H (2009) The effect of crystallization of PLA on the thermal and mechanical properties of microfibrillated cellulose-reinforced PLA composites. Compos Sci Technol 69:1187–1192 Suryanegara L, Nakagaito AN, Yano H (2010) Thermo-mechanical properties of microfibrillated cellulose-reinforced partially crystallized PLA composites. Cellulose 17:771–778 Tan BH, Muiruri JK, Li Z, He C (2016) Recent progress in using stereocomplexation for enhancement of thermal and mechanical property of polylactide. ACS Sustain Chem Eng 4:5370–5391 Tingaut P, Zimmermann T, Lopez-Suevos F (2009) Synthesis and characterization of bionanocomposites with tunable properties from poly(lactic acid) and acetylated microfibrillated cellulose. Biomacromolecules 11:454–464 Wang Y, Tian H, Zhang L (2010) Role of starch nanocrystals and cellulose whiskers in synergistic reinforcement of waterborne polyurethane. Carbohydr Polym 80:665–671 Xu X, Liu F, Jiang L, Zhu JY, Haagenson D, Wiesenborn DP (2013) Cellulose nanocrystals vs. cellulose nanofibrils: a comparative study on their microstructures and effects as

123

polymer reinforcing agents. ACS Appl Mater Interfaces 5:2999–3009 Xu C, Lv Q, Wu D, Wang Z (2017) Polylactide/cellulose nanocrystal composites: a comparative study on cold and melt crystallization. Cellulose 24:2163–2175 Yasuniwa M, Sakamo K, Ono Y, Kawahara W (2008) Melting behavior of poly(L-lactic acid): X-ray and DSC analyses of the melting process. Polymer 49:1943–1951 Yu H, Yan C, Yao J (2014) Fully biodegradable food packaging materials based on functionalized cellulose nanocrystals/ poly(3-hydroxybutyrate-co-3-hydroxyvalerate) nanocomposites. RSC Adv 4:59792–59802 Zhang K, Mohanty AK, Misra M (2012) Fully biodegradable and biorenewable ternary blends from polylactide, poly(3hydroxybutyrate-co-hydroxyvalerate) and poly(butylene succinate) with balanced properties. ACS Appl Mater Interfaces 4:3091–3101 Zhang K, Nagarajan V, Misra M, Mohanty AK (2014) Supertoughened renewable PLA reactive multiphase blends system: phase morphology and performance. ACS Appl Mater Interfaces 6:12436–12448 Zhang L, Li Y, Wang H, Qiao Y, Chen J, Cao S (2015) Strong and ductile poly(lactic acid) nanocomposite films reinforced with alkylated graphene nanosheets. Chem Eng J 264:538–546 Zhu A, Diao H, Rong Q, Cai A (2010) Preparation and properties of polylactide–silica nanocomposites. J Appl Polym Sci 116:2866–2873