J Mater Sci (2012) 47:2675–2686 DOI 10.1007/s10853-011-6093-4
Physico-chemical and mechanical properties of nanocomposites prepared using cellulose nanowhiskers and poly(lactic acid) Kazi M. Zakir Hossain • Ifty Ahmed • Andrew J. Parsons • Colin A. Scotchford • Gavin S. Walker • Wim Thielemans • Chris D. Rudd
Received: 27 July 2011 / Accepted: 28 October 2011 / Published online: 5 November 2011 Ó Springer Science+Business Media, LLC 2011
Abstract A range of nanocomposites were prepared using cellulose nanowhiskers (CNWs) and poly(lactic acid) (PLA) via a solvent casting process. Acid hydrolysis process was used to produce CNWs from bleached cotton. Structural morphology and surface topography of the CNWs and nanocomposites were examined using transmission (TEM) and scanning electron microscopy. TEM images revealed rod-like whiskers in the nano-scale region which were dispersed within the PLA matrix. The presence of the functional groups of CNWs and PLA were confirmed via FTIR analysis. Tensile tests were conducted on thin films and the nanocomposites containing 1 wt% CNWs showed a 34 and 31% increase in tensile strength and modulus, respectively, compared to pure PLA. The dynamic mechanical analysis showed that the tensile storage modulus also increased in the visco-elastic temperature region with increasing CNWs content in the nanocomposites. Thermogravimetric analysis showed that all the materials investigated were thermally stable from room temperature to 210 °C. A positive effect of CNWs on the crystal nucleation of PLA polymer in the nanocomposites was observed using differential scanning calorimetry and X-ray diffraction analysis. The degradation profiles of the
K. M. Z. Hossain (&) I. Ahmed A. J. Parsons C. A. Scotchford G. S. Walker C. D. Rudd Division of Materials, Mechanics and Structures, Faculty of Engineering, University of Nottingham, University Park, Nottingham NG7 2RD, UK e-mail:
[email protected] W. Thielemans School of Chemistry and Process and Environmental Research Division, Faculty of Engineering, University of Nottingham, University Park, Nottingham NG7 2RD, UK
nanocomposites in deionised water over 1 week revealed a mass loss of 1.5–5.6% at alternate temperatures (25, 37 and 50 °C) and at the same conditions the swelling ratio and water uptake were seen to increase with CNWs content in the nanocomposites, which was strongly influenced by the presence of crystalline CNWs.
Introduction Bio-based nanocomposites have attracted significant attention due to their biodegradability, biocompatibility and better processability compared to the use of pure polymer. Materials, derived from biological origin, having at least one dimension in the nano-scale can be incorporated into a natural polymer matrix [1, 2] to produce biobased nanocomposites. These types of composites have shown potential applications in the field of food packaging, biomedical and tissue engineering with improved thermal and mechanical properties [3–5]. High strength cellulose nanowhiskers (CNWs) embedded within a biopolymer matrix has huge scope as bone implants, in soft tissue repair and as resorbable sutures. Cellulosic fibre, a polysaccharide composed of several hundred to tens of thousands of b-glycoside units in a linear orientation [6], is a natural polymer obtained from vegetable origin and is both biodegradable [7–9] and biocompatible [10, 11]. Cellulose nanocrystals are generally produced via acid hydrolysis using sulphuric acid [12, 13], hydrochloric acid [14, 15], or nitric acid [16]. Recently, CNWs have been used to reinforce biopolymeric matrices and several studies have investigated the mode of their reinforcement [17, 18]. CNWs have high mechanical properties; for instance they have a tensile modulus of *130 GPa [19] and a tensile strength of around 10 GPa [19].
123
2676
Polymeric biomaterials are compatible with biological systems and have been used to treat, augment, or replace tissues, organs and/or functions of the biological systems. Poly(lactic acid) (PLA) has been studied extensively and has many potential applications in the medical, textile, food and packaging industries [20–23]. Within the medical field PLA has been investigated for use in drug delivery systems, drug encapsulation, surgical implants, sutures, soft tissue and bone fixative materials [24–28]. For composite preparation, PLA has been blended with reinforcing fibres [24, 26, 28–31], polymers including starch [30, 32], chitosan [2, 33, 34] and inorganic fillers [35]. Peterson et al. investigated the structural and thermomechanical properties of composites made using solution cast PLA/cellulose whiskers materials using tert-butanol with reasonably good dispersion of the whiskers (5 wt%) within the polymer matrix. Enhanced thermal stability in the region between 25 and 220 °C and improved storage modulus at higher temperature (64% increase at 60 °C) [36] was also reported. Bondeson and Oksman showed that nanocomposites prepared via extrusion moulding of PLA/ CNWs modified using polyvinyl alcohol followed by compression moulding showed improved mechanical properties, for example, tensile modulus increased from 3.31 to 3.71 GPa for the composite containing 65% PLA, 30% PVA and 5% CNWs compared with pure PLA [37]. Cellulose nanofibre (produced from kenaf pulp) reinforced PLA composites produced via melt compounding and extrusion also showed a tensile modulus and strength increase from 2.9 to 3.2 GPa and from 58 to 71 MPa, respectively, for nanocomposites containing 5% cellulose nanofibre. The dynamic mechanical analysis (DMA) results also reported that the storage modulus increased for all nanocomposites that were investigated, compared to PLA in the higher temperature region (70 °C) [38]. Liu et al. also investigated composite films prepared using PLA and flax cellulose obtained via acid hydrolysis of flax yarns and reported that composites containing 5% flax cellulose showed the tensile strength and modulus increased by 59 and 47%, respectively, [39]. In contrast with the studies above [36–38], it has also been reported that the tensile modulus of nanocomposites containing 1, 2, 3 and 5% freeze-dried CNWs isolated from highly purified alpha microcrystalline cellulose reduced by 37, 47, 43 and 35%, respectively, compared to pure PLA. However, in this study the dispersion of CNWs in chloroform was performed using a homogeniser and then stirred with PLA solution at 40 °C for 30 min. A significant decrease in tensile strength (up to 54%) was also reported in these nanocomposites [40]. In this study, we wanted to ascertain the effect of cotton-based nanowhiskers within a PLA matrix. This study focused on varying concentrations of nanowhiskers
123
J Mater Sci (2012) 47:2675–2686
(CNWs) isolated from cotton incorporated within PLA to improve mechanical properties. In addition, morphological, void content, thermomechanical, crystallisation, degradation and water uptake properties of these biodegradable nanocomposites were also investigated. Nanocomposites (reinforced with 1, 3 and 5% cotton-based CNWs prepared via a solvent casting process) were investigated in this study utilising Natureworks PLA as matrix.
Experimental Materials Sulphuric acid (purity 95%, specific gravity 1.83), cotton wool (Bleached), chloroform and amberlite MB 6113 resin were obtained from Fisher Scientific (UK). Uranyl acetate was supplied by Sigma-Aldrich (UK). PLA beads purTM chased from NatureWorks LLC (Ingeo Grade 3251D average Mw * 90,000–120,000 g mol-1, Density = -3 1.24 g cm , PDI = 1.636) were used as the matrix for the nanocomposites. CNWs preparation CNWs were prepared via hydrolysis of cotton wool (Bleached) for 45 min at 45 °C in 64 wt% aqueous H2SO4 with constant stirring [41]. The resulting suspension was then treated as described in the literature to obtain the cellulose nanocrystals [42]. In brief, the acid hydrolysed suspension was washed with deionised water utilising three successive centrifugations at 10,000 rpm and 10 °C for 15 min. For the removal of residual-free acid from the suspension a dialysis under running tap water was conducted for 48 h after which the pH of the eluent was measured (as it was required to be neutral). The dispersion of CNWs in water was homogenised using a Branson sonicator for 4 min and then filtered over a no. 2 fritted glass filter. The filtrate was stirred overnight with amberlite resin to remove non H3O? cations from the CNWs suspension and then filtered again to obtain the final CNWs in deionised water. Finally, crystalline CNWs were obtained by freezing the suspension in liquid nitrogen and freeze drying [42]. Before producing the nanocomposites, a homogeneous suspension of CNWs in chloroform was prepared by stirring and sonication. Nanocomposites preparation The solvent cast nanocomposites were prepared by dissolving PLA pellets in chloroform (10% w/v), which was stirred constantly using a magnetic stirrer at 40 °C until the pellets were fully dissolved before adding the CNWs
J Mater Sci (2012) 47:2675–2686
2677
suspension in chloroform and mixing thoroughly for 2 h. The formulations (Table 1) were poured into glass petri dishes placed on a horizontal surface and previously greased with silicon. The mixture was allowed to evaporate at room temperature (25 °C) and after 24 h the resulting films (approximate thickness 0.25 mm) were dried in a vacuum oven at 50 °C for 48 h to remove residual solvent.
qTheoretical ¼ h
1 WCNWs qCNWs
þ Wq PLA
i;
ð2Þ
PLA
where WCNWs and WPLA are the weight fractions of CNWs and PLA, respectively, and qCNWs and qPLA are the density of CNWs and PLA, respectively. Fourier transform infrared (FTIR) spectroscopy analysis
Characterisation Electron microscopic analysis The freeze-dried CNWs and their dispersion in the PLA matrix were examined using transmission electron microscopy (TEM). This was performed on a JEOL (JEM2000FXII, UK), using an accelerating voltage of 120 kV and a copper electron microscopic grid supported by a porous carbon (mesh size 300) film. For the characterisation of nanocomposites, a small sample was embedded in epoxy resin and cured for 1 day at room temperature (25 °C). The epoxy embedded films were cut into thin slices (*100 nm) using an ultra microtome cutter equipped with a sharp glass edge. To obtain better contrast in TEM the Cu-grid containing the samples was stained negatively using uranyl acetate (2 wt%) for 5 min. The surface topography of the pure PLA and the nanocomposites was characterised using a scanning electron microscopy (Philips XL30, FEI, USA) at an accelerating voltage of 12 kV and a working distance of 10 mm. A sputtered coating of platinum was used to avoid image distortion due to charging. Void content of nanocomposites The void content for the solvent cast pure PLA and the nanocomposites was determined according to ASTM D 2734-94 method [43] using the theoretical (qTheoretical ) and experimental (qExperimental ) density of the nanocomposites (see Eqs. 1, 2): qTheoretical qExperimental Void content ¼ ð1Þ qTheoretical
Table 1 Formulations and sample codes for the nanocomposites investigated in this study Sample codes used in this study
PLA (wt%)
CNWs (wt%)
PLA
100
–
PLA-1
99
1
PLA-3
97
3
PLA-5
95
5
Identification of functional groups of the CNWs produced was confirmed after acid hydrolysis of cotton using FTIR (Tensor-27, Bruker, Germany). All spectra were analysed TM with Opus software version 5.5. The nanocrystals and the thin films were scanned in transmittance mode in the region of 4000 and 550 cm-1 (wave numbers) using standard pike attenuated total reflectance (ATR) cell (Pike Technology, UK). Tensile tests The nanocomposite films produced were cut into dog bone shapes using a 10 9 64 mm dog bone cutter and the tensile properties were characterised using a Hounsfield Series-S tensile test machine (Software-QMAT) with a cross head speed of 1 mm min-1, gauge length 25 mm and a 1 kN load cell. The tensile strength, modulus and elongation at break were calculated from experimental data according to the European standard (ISO/DIS 527-1:2010). Dynamic mechanical analysis Thermomechanical properties of the PLA films and the nanocomposites were measured using a DMA (Q-800 from TA Instruments, USA) in mutifrequency strain mode using the tension film clamp. The characterisation was conducted using Q series (Q-800) software and parameters maintained during the analysis were: 0.05% strain, 0.01 N preload force, 125% force track and 1 Hz constant frequency. The temperature was ramped from room temperature (25 °C) to 80 °C with a heating rate of 5 °C min-1 and a gap distance of around 20 mm was maintained. Three repeat samples were used to characterise each material. The samples were prepared by cutting strips from the films with a width of 5 mm and length of 30 mm. The tensile storage modulus (E0 ) and tan delta of the nanocomposites were recorded with respect to increasing temperature. Thermogravimetric analysis (TGA) The TG analysis was conducted from room temperature (25 °C) to 480 °C using a SDT Q600 thermogravimetric analyser from TA instruments (USA) with a heating rate of
123
2678
J Mater Sci (2012) 47:2675–2686
10 °C min-1 under 100 mL min-1 nitrogen gas flow. For background correction a blank analysis was conducted. The weight loss (%) with temperature of freeze-dried CNWs and the nanocomposite films (6–10 mg) were determined. Data acquisition and processing was performed using TA Universal analysis 2000 software. Differential scanning calorimetry (DSC) analysis The percentage crystallinity of PLA in the nanocomposite films was investigated using a DSC (Q2000, TA instruments, UK). A sample size of *6 mg was used and heated from 20 to 200 °C at a heating rate of 10 °C min-1 under nitrogen gas flow (50 mL min-1) and then isothermally held at 200 °C for 5 min. After that the samples were subsequently cooled down to 20 °C at 5 °C min-1[39] and for the second time held isothermally at 20 °C for 5 min before heating once again to 200 °C at the same heating rate. Data acquisition and processing was performed using TA Universal analysis 2000 software. A blank pan measurement was conducted for background and at least three tests were done for each material to ensure repeatability. All the data acquired was from the second heating cycle of the DSC scan to eliminate any thermal history effects. The percentage crystallinity (Xc) of PLA in the nanocomposites was calculated according to Eq. 3 [44]: Xc ½% ¼
ðDHm Þ=UPLA 100% DHm0
ð3Þ
part. I(am) is the peak intensity around 18° and represents the amorphous part of the cellulose whiskers. Degradation and swelling profile analysis The degradation of the nanocomposite films was evaluated by means of mass loss measurements. The initial weight of four repeat specimens of each material (30 9 5 9 0.2 mm) was measured before being immersed in glass vials containing 20-mL deionised water at different temperatures (25, 37 and 50 °C). After 1 week, specimens were recovered from the media and dried in a vacuum oven at 50 °C until constant weight. After that, the weight of the dried specimens and the pH of the media containing the degraded materials were recorded. The percentage mass loss was determined using Eq. 5: Mass loss ð% Þ ¼
ðM0 Mds Þ 100 M0
where M0 is the mass of the dry sample and Mds is the mass of the dry sample after degradation. The swelling ratio of the nanocomposites were calculated gravimetrically by measuring the mass of the sample before and after swelling in deionised water at varying temperatures (25, 37 and 50 °C) for 1 week using Eq. 6. The swelled films were placed onto a tissue paper to blot dry and weighed immediately. Swelling ratio ð% Þ ¼
-1
where DHm is the enthalpy of fusion (J g ) of the polymer nanocomposites, DHm0 is the enthalpy of fusion for a PLA crystal of infinite size (taken to be 93.6 J g-1) [45] and UPLA is the fraction of PLA in the nanocomposites. X-ray diffraction (XRD) analysis The XRD patterns of freeze-dried CNWs and the nanocomposites films were obtained using a D500 diffractometer (SIEMENS) operated at 30 kV and 15 mA, utilising a Cu-Ka radiation source (k = 0.154). The diffraction patterns were recorded for 2h values between 2° and 50°, using a step size of 0.04°, providing 1200 steps, and a scan step time of 2 s. The scans were controlled by the Diffrac-AC software programme. The degree of crystallinity of CNWs was calculated according to Eq. 4 [46, 47]: Ic ½% ¼
IðCrysþamÞ IðamÞ 100 IðCrysþamÞ
ð4Þ
where IðCrysþamÞ represents the peak intensity (count per second) around 22.8° for the crystalline and amorphous
123
ð5Þ
ðMW M0 Þ 100 M0
ð6Þ
where Mw is the mass of wet sample after swelling and M0 is the mass of dry sample before swelling. After 1 week, the swelled films were dried in a vacuum oven at 50 °C until constant weight was obtained and the percentage water uptake by the films was calculated using Eq. 7: Water uptake ð% Þ ¼
ðMs Mds Þ 100 Mds
ð7Þ
where Ms is the mass of the swelled sample and Mds is the mass of the dry sample after swelling. Statistical analysis Unpaired t tests were conducted to investigate any statistically significant differences between the means of the data sets that were obtained. To interpret the t test results, a two tailed P value (P) was considered with a 95% confidence interval and a significance level of 0.05. The difference was accepted as significant when the value of P obtained was less than 0.05.
J Mater Sci (2012) 47:2675–2686
Results and discussion
2679
film processing stage, which has also been reported previously [52–54].
Morphological properties FTIR analysis The TEM image of the CNWs presented in Fig. 1a revealed a rod-like structure for the nanowhiskers and addition of 2 wt% uranyl acetate solution provided some contrast for these images. Due to acid hydrolysis, the amorphous portions of the long chain cellulose were hydrolysed, and the crystalline portion remained unaffected to produce rod-like whiskers in the nano-scale region which was confirmed via TEM. The average length and width of the CNWs were *300 ± 100 and 10 ± 2 nm, respectively, obtained from several TEM images, which was found to be very similar to the analysis conducted by Elazzouzi-Hafraoui et al. [48], who also obtained cotton nanocrystals of dimensions ranging from 25 to 320 nm in length and 6–70 nm in width. The dispersion of CNWs in the polymer matrix was also confirmed, as shown by the TEM image (Fig. 1b) and was seen to be aggregated into small bundles in the PLA matrix. This was suggested to be due to strong self-association tendency of CNWs via hydrogen bonding of their hydroxyl groups and their high surface area [49–51]. The surface topography of the pure PLA and nanocomposites was also analysed via SEM, as presented in Fig. 1c, d, where the presence of voids were identified in both PLA and the nanocomposite films, which was suggested to be due to the highly volatile chloroform escaping during the film drying process. A representative image of the solvent cast nanocomposites (approximate dimensions 15 9 7 9 0.2 mm) prepared using different percentages of CNWs are shown in Fig. 2. Images obtained showed that the PLA film was fairly transparent, whilst the PLA-1, PLA-3 and PLA-5 became increasingly opaque with increasing additions of CNWs in the nanocomposites [36], which were expected due to the aggregated dispersion of CNWs in the polymer matrix and also the presence of voids in the nanocomposites unintentionally obtained during the processing into films. Void content The presence of voids in the nanocomposites can have a significant influence on the mechanical properties of the nanocomposite films. The void content (%) of nanocomposite films was seen to increase with an increase in CNWs content within the nanocomposites. For the PLA-1, PLA-3 and PLA-5 nanocomposites, *8.7, 11.2 and 13.9% void content was calculated, whereas around only 1.2% void content was calculated for the solvent cast pure PLA film. This was suggested to be due to entrapped air within the CNWs in addition to the use of a volatile solvent during the
The functional groups of acid hydrolysed CNWs, PLA and nanocomposites were identified using FTIR–ATR spectroscopy (see Fig. 3). For the CNWs alone, the bands at 3336 and 3290 cm-1 were attributed to free O–H stretching vibration and H-bond, respectively. The bands at 2922 and 1429 cm-1 indicated the C–H stretching and bending of –CH2 groups, respectively. The characteristic peak at 1644 cm-1 represents the absorbed O–H bending in the nanowhiskers structure [55]. The peak at 1162 cm-1 was attributed to the C–O–C stretching bridge of glucose ring structure of cellulose. The bands at 1057 and 1033 cm-1 represented the C–O stretching at position C-6 and C-3 in the saccharide structure, respectively [2]. For pure PLA, the peak at 2920, 1748 and 1080 cm-1 represented the symmetrical stretching of C–H, C=O and –C–O– groups of ester bonds, respectively, and the bands at 1450, 1381 and 885 cm-1 were assigned to asymmetrical stretching of –CH3 group, symmetrical deformation of –CH3 and C–H group in the polymer, respectively [56]. It was seen that the peak for free hydroxyl groups at 3336 cm-1 in PLA-1 and PLA-3 disappeared, which was suggested to be due to the presence of a small amount of CNWs in the nanocomposites, however, this peak was observed for PLA-5 which contained a higher percentage of CNWs. The band at 1748 cm-1 in all the nanocomposites represented the C=O group of PLA. The intensity of the peak at 1033–1057 cm-1 for the C–O group of the saccharide structure in the nanowhiskers increased with an increasing loading of nanowhiskers in the nanocomposites, as expected. Mechanical properties The mechanical properties of the nanocomposites compared to the pure PLA are depicted in Fig. 4. PLA-1 showed a significant percentage increase (34%) in tensile strength, whilst PLA-3 and PLA-5 showed little increase (8%) in comparison with the pure PLA (tensile strength *21.2 MPa). This was suggested to be due to better dispersion of the CNWs within the polymer matrix for PLA-1 compared to PLA-3 and PLA-5. At higher concentrations of CNWs in the nanocomposites, poor dispersion was observed due to the strong self-aggregating nature of CNWs leading to a less pronounced increase in tensile strength. However, the tensile properties of pure PLA investigated in this study found lower in comparison to expectation and this was due to the presence of lots of void created within the film during the processing into thin film and also probably due to the solvent effect on polymer.
123
2680
J Mater Sci (2012) 47:2675–2686
Fig. 1 Electron microscopy images: a TEM image of freeze-dried CNWs, b TEM image of nanocomposite (PLA-1), c SEM image of the surface topography of pure PLA film, d SEM image of the surface topography of the nanocomposite (PLA-1) film
Fig. 2 Representative image of the solvent cast pure PLA and nanocomposites films
Similar mechanical properties for pure PLA film was also reported by Liu et al. [39] who investigated solvent cast PLA film (Natureworks, 2002D) and found the tensile strength *19.4 MPa and modulus *1.4 GPa. The tensile modulus of PLA-1, PLA-3 and PLA-5 also showed an improvement of 31, 32 and 47%, respectively, as compared to PLA alone. This significant increase in tensile modulus for all the nanocomposites investigated was expected due
123
to the high theoretical modulus (130 GPa) of the CNWs [19]. The tensile test data calculated in this study was found to be statistically significant comparing PLA alone against PLA-1 (P \ 0.05), PLA-3 (P \ 0.05) and PLA-5 (P \ 0.05). Comparing PLA-3 and PLA-5 against PLA-1 the tensile strength data was also found to be statistically significant (P \ 0.05). However, comparing PLA-3 and
J Mater Sci (2012) 47:2675–2686
2681
Fig. 5 Elongation at break properties for the PLA and nanocomposite films
Fig. 3 FTIR–ATR spectrum for the CNWs, pure PLA and nanocomposites investigated in this study
Fig. 4 Tensile strength and modulus properties for the PLA and nanocomposite films
PLA-5 the tensile strength data was considered not to be statistically significant (P [ 0.05). In comparison with the pure PLA film, a dramatic reduction in elongation at break was seen for all the nanocomposites investigated (see Fig. 5). A reduction in elongation at break was seen for PLA-1, PLA-3 and PLA-5 by *85, 88 and 90% (respectively) compared to PLA alone, suggesting that incorporation of CNWs made the nanocomposites very brittle [57]. Thermomechanical properties The thermomechanical properties of the nanocomposite films were investigated via DMA to explain the mechanical
behaviour of the nanocomposites with change in temperature, especially in the plastic region of polymer. The DMA results revealed the tensile storage modulus of the nanocomposites as a function of temperature as seen in Fig. 6a. The incorporation of CNWs in the PLA matrix showed an improvement in storage modulus for all the nanocomposites in the visco-elastic temperature region of polymer. The amount of CNWs incorporated in PLA matrix played a significant role in improving the storage modulus of the nanocomposites at the temperature regions investigated in this study. An increase in storage modulus 30% was observed at 25 °C in PLA-5 compared to PLA alone, whereas, at 37 and 50 °C the storage modulus of PLA-5 was seen to increase 70 and 182%, respectively, in comparison with pure PLA film. The tan delta peaks obtained for the nanocomposites investigated were seen to shift to the right, i.e. towards the higher temperature regions as compared to the tan delta peak for PLA alone (see Fig. 6b). The tan delta peaks for PLA-1, PLA-3 and PLA-5 were seen at 58, 59 and 62 °C compared to 51 °C for pure PLA suggesting that addition of CNWs significantly improved the storage modulus in the plastic region of the polymer. The right shift of tan delta peaks also suggested that the reinforcement established in the nanocomposites had significant influential effect on the segmental motion within the PLA matrix. It was also seen that the width of the tan delta peaks in the nanocomposites was wider as compared to pure PLA which indicated the increasing temperature span required for the transition, which was also reported by Petersson et al. [1]. This was attributed to be due to higher cross-linking density of the nanocomposites and surface induced crystallisation on the nanowhiskers with an increase in CNW concentration within the nanocomposites.
123
2682
J Mater Sci (2012) 47:2675–2686
Fig. 7 TGA thermogram of the CNWs, pure PLA and nanocomposites investigated in this study
The melting thermogram of the pure PLA and nanocomposite films is depicted in Fig. 8. The melting peak of pure PLA film appeared around 170 °C, whilst a slight shift of melting peaks to the right at around 172 °C was observed in all the nanocomposites and this was suggested a little effect of CNWs in increasing the melting temperature of the nanocomposites. Crystallisation properties
Fig. 6 a Tensile storage modulus curves from DMA data for pure PLA and nanocomposites, b tan delta curves from DMA data for pure PLA and nanocomposites
Thermal properties The thermal stability of the CNWs and nanocomposites are shown in Fig. 7, which plots residual weight versus temperature. All the materials were found to be thermally stable in the temperature region between 20 and 210 °C as compared to a very similar literature value 25 to 220 °C for nanocomposites produced using PLA and nanowhiskers isolated from microcrystalline cellulose [36]. The CNWs started to lose their weight after 210 °C, whilst the nanocomposites showed decomposition above 320 °C. Similar thermal behaviour of CNWs was also reported in the literature [58, 59] and this was suggested to be due to the use of acid hydrolysed cotton as the reinforcing agent and probably due to the presence of sulphate groups on the surface of the CNWs. At higher temperature (*400 °C) the residual weight percent of the nanocomposites was seen to increase with increasing CNWs content.
123
The crystallisation behaviour of PLA in the nanocomposites was investigated using DSC analysis. The enthalpy of fusion of pure PLA and nanocomposites was obtained from the second heating run of DSC data (see Fig. 8). The percentage crystallisation of PLA and the nanocomposites (Table 2) showed that the crystallinity of solvent cast PLA (around 38.1%) was comparable to the literature values of 40.7% [45]. However, an increase in enthalpy of fusion was seen in the nanocomposites, which was due to the incorporation of crystalline CNWs in the nanocomposites. As a result, the crystallisation of PLA in the nanocomposites found increasing as expected probably due to the crystal nucleation of PLA polymer influenced by the presence of CNWs crystallites [39]. However, the crystallisation property of CNWs produced and their effect on the crystallisation of PLA in the nanocomposites were investigated further using XRD. Figure 9 shows the XRD traces of the CNWs and the nanocomposites. From the diffraction pattern of CNWs the highest peak was observed at 2h = 22.8° and the double peak signal at 2h = 14.9° and 16.5° was consistent with the diffraction pattern of reference cotton-based cellulose [60] identified from the ICDD patent PDF-2database (File no. 00-050-2241). An approximate crystallinity of 89.1% was calculated for the freeze-dried CNWs, which was very close to the result stated by other studies [48, 50]. The main diffraction peaks for PLA alone were seen at 2h = 16.5°
J Mater Sci (2012) 47:2675–2686
2683
Fig. 8 DSC thermogram of the pure PLA and nanocomposites from the second heating run
Table 2 Crystallinity index of pure PLA and nanocomposite films investigated using DSC data Materials
Enthalpy of fusiona (J g-1) (±Std. dev.)
PLA
35.6 (±0.5)
38.1
PLA-1
37.4 (±0.3)
40.4
PLA-3
37.9 (±0.2)
41.7
PLA-5
38.1 (±0.1)
42.9
a
Crystallisation (%)
Instrument’s cell constant 0.9965
and 18.9° and a weaker peak at around 22.5° which was also consistent with the literature values [61]. The XRD traces for the nanocomposites showed clear retention of the cellulose crystallites with increasing peak intensity at 2h = 14.9°, 16.5° and 22.5°, and this was again suggested to be due to an increase in the crystallinity of the nanocomposites, which was also well consistent with the aforementioned DSC results. DSC analysis was used only to quantify the crystallinity of PLA in the nanocomposites, whilst the degree of crystallinity for CNWs was calculated using XRD analysis. Degradation and swelling properties The degradation behaviour of the nanocomposites after a week in deionised water at varying temperatures (25, 37 and 50 °C) can be seen in Fig. 10a. The percentage mass loss of PLA alone increased significantly with increasing temperature (up to *5.2% at 50 °C) and this loss was attributed to hydrolysis of the polymer influenced by higher temperature. At room temperature the mass loss of the nanocomposites was seen to increase with increasing CNWs content compared to PLA alone. However, at 37 °C for all nanocomposites the percentage mass loss was slightly lower compared with room temperature studies and this was suggested to be due to hydrolysis of the
Fig. 9 XRD patterns of the CNWs, pure PLA and nanocomposites investigated in this study. Closed diamond Reference cellulose peaks identified from the ICDD patent PDF-2 database (File no. 00-0502241)
amorphous domain of the polymer at higher temperature and subsequent interfacial reinforcement of polymer by nanocrystals hindering the rapid degradation of these nanocomposites [56]. A significant percentage mass loss was observed at 50 °C for all the nanocomposites and it was suggested that this increase was due to the leaching of CNWs and a continuous breakdown of the interface between the CNWs and the polymer matrix [62]. The pH of the degradation media for PLA alone remained relatively neutral for the duration of the study along with PLA-1 as seen in Fig. 10b. However, a decrease in pH was observed for PLA-3 and PLA-5 with increasing temperature (from 37 to 50 °C) and this was suggested to be due to the leaching of acid-hydrolysed CNWs into the deionised water. Figure 11a shows the swelling behaviour of the nanocomposites immersed in deionised water at alternate temperatures (25, 37 and 50 °C) for 1 week and it was observed that the percentage of swelling ratio increased with increasing CNW content at all the temperatures that were investigated. This effect was attributed to the increased hydrophilicity of the nanocomposites due to the presence of CNWs [63] in the PLA matrix. The threedimensional network between the nanowhiskers formed by hydrogen bonding was suggested to have a significant influence on the swelling behaviour of the nanocomposites, which was also reported by Garcia de Rodriguez et al. [64]. A significant decrease in swelling ratio of PLA alone at higher temperature was observed due to the continuous degradation of PLA via hydrolysis. However, the presence of CNWs in the nanocomposites increased the swelling ratio at 37 °C, which was suggested to be due to an increase in surface area within the nanocomposites caused by the addition of CNWs and also water accumulation at the PLA–CNW interface [64]. At 50 °C, the swelling ratio was seen to decrease and this was suggested to be due to
123
2684
J Mater Sci (2012) 47:2675–2686
Fig. 10 a Mass loss of nanocomposite films in deionised water at different temperatures for 1 week, b pH of deionised water containing degraded sample of nanocomposites at different temperatures
the degradation of PLA and disruption of the PLA–CNW interface within the nanocomposites. The percentage water uptake profile of the nanocomposites at different temperatures over a 1-week period is shown in Fig. 11b. The water uptake for PLA alone and the nanocomposites was seen to increase continuously with an increase in temperature as expected due to the hydrolysis of PLA and leaching of CNWs from the nanocomposites suggesting a rapid breakdown of the PLA–CNW interface and formation of microvoids in the samples [65]. Moreover, with an increase in CNWs content the water uptake was also seen to increase at the temperatures investigated. This was again suggested to be due to the hydrophilic nature of the cotton-derived CNWs incorporated and similar effects were also reported by Angle`s and Dufresne [66], who investigated starch/tunicate whisker nanocomposites, and by Garcia de Rodriguez et al. [64], who examined polyvinyl acetate/sisal cellulose whiskers nanocomposites. The above study showed that incorporating 1–5 wt% of CNWs increased the mechanical, thermomechanical properties as well as crystallinity of the nanocomposites
123
Fig. 11 a Swelling ratio and b water uptake of nanocomposite films at different temperatures for 1 week in deionised water
compared to PLA alone. Future study will investigate on improving the dispersion issues of CNWs in the matrix.
Conclusions The influence of the CNWs (1–5 wt%) on the improvement of the mechanical, thermomechanical, crystallisation and hydrolytic degradation properties of nanocomposites based on PLA was investigated in this study. Dispersion of the rod-like nanowhiskers within the PLA matrix was limited and seen to be largely aggregated. The presence of voids identified in the nanocomposites played an important role on the mechanical and water absorption properties of the nanocomposites. The incorporation of CNWs in the PLA matrix showed significant improvement in tensile strength, tensile modulus, storage modulus and crystallinity of the nanocomposites investigated. In the plastic temperature region for PLA the tan delta curves confirmed that the reinforcement and surface-induced crystallisation was
J Mater Sci (2012) 47:2675–2686
established in the nanocomposites due to the addition of CNWs. The presence of CNWs in the nanocomposites had a significant influence on their degradation at varying temperatures. The swelling ratio and water uptake of the nanocomposites presented here showed increasing with the amount of CNWs in all temperature regions investigated due to the addition of hydrophilic CNWs and formation of microvoids during hydrolysis of PLA. This study demonstrated that incorporation of CNWs within a PLA matrix can significantly improve the thermomechanical, crystallisation and degradation properties of the PLA. This opens up further possible applications for the use of these materials within the biomedical and food packaging field. Acknowledgements The authors would like to acknowledge the financial support provided by the University of Nottingham, Faculty of Engineering (via the Dean of Engineering Research Scholarship for International Excellence).
References 1. Petersson L, Oksman K (2006) Compos Sci Technol 66:2187. doi:10.1016/j.compscitech.2005.12.010 2. Li Q, Zhou J, Zhang L (2009) J Polym Sci Part B 47:1069. doi: 10.1002/polb.21711 3. Azami M, Orang F, Moztarzadeh F (2006) International conference on biomedical and pharmaceutical engineering, p 259 4. Ma ZW, Kotaki M, Inai R, Ramakrishna S (2005) Tissue Eng 11:101 5. McCullen SD, Ramaswamy S, Clarke LI, Gorga RE (2009) Wiley Interdiscipl Revi Nanomed Nanobiotechnol 1:369. doi: 10.1002/wnan.39 6. Updegraff DM (1969) Anal Biochem 32:420. doi:10.1016/ s0003-2697(69)80009-6 7. Glasser WG, McCartney BK, Samaranayake G (1994) Biotechnol Prog 10:214. doi:10.1021/bp00026a011 8. Park CH, Kang YK, Im SS (2004) J Appl Polym Sci 94:248. doi: 10.1002/app.20879 9. Ku¨mmerer K, Menz J, Schubert T, Thielemans W (2011) Chemosphere 82:1387. doi:10.1016/j.chemosphere.2010.11.069 10. Miyamoto T, Takahashi S, Ito H, Inagaki H, Noishiki Y (1989) J Biomed Mater Res 23:125. doi:10.1002/jbm.820230110 11. Luo LH, Wang XM, Zhang YF, Liu YM, Chang PR, Wang Y, Chen Y (2008) J Biomater Sci Polym Ed 19:479. doi: 10.1163/156856208783719491 12. Dufresne A, Kellerhals MB, Witholt B (1999) Macromolecules 32:7396. doi:10.1021/ma990564r 13. Favier V, Chanzy H, Cavaille JY (1995) Macromolecules 28:6365. doi:10.1021/ma00122a053 14. Habibi Y, Chanzy H, Vignon M (2006) Cellulose 13:679. doi: 10.1007/s10570-006-9075-y 15. Morin A, Dufresne A (2002) Macromolecules 35:2190. doi: 10.1021/ma011493a 16. Zuluaga R, Putaux J-L, Restrepo A, Mondragon I, Gan˜a´n P (2007) Cellulose 14:585. doi:10.1007/s10570-007-9118-z 17. Favier V, Canova GR, Cavaille JY, Chanzy H, Dufresne A, Gauthier C (1995) Polym Adv Technol 6:351 18. Eichhorn S, Dufresne A, Aranguren M et al (2010) J Mater Sci 45:1. doi:10.1007/s10853-009-3874-0
2685 19. Hepworth DG, Bruce DM (2000) J Mater Sci 35:5861. doi: 10.1023/a:1026716710498 20. Henton DE, Gruber PE, Lunt J, Randall J (2005) In: Mohanty AK, Mishra M, Drzal LT (eds) Natural fibers, biopolymers, and biocomposites. Taylor & Francis, Boca Raton, p 524 21. Dorgan JR, Lehermeier H, Mang M (2000) J Environ Polym Degrad 8:1. doi:10.1023/a:1010185910301 22. Drumright RE, Gruber PR et al (2000) Adv Mater 12:1841 23. Gupta B, Revagade N, Hilborn J (2007) Prog Polym Sci 32:455. doi:10.1016/j.progpolymsci.2007.01.005 24. Ahmed I, Jones I, Parsons A, Bernard J, Farmer J, Scotchford C, Walker G, Rudd C (2011) J Mater Sci Mater Med 22:1825. doi: 10.1007/s10856-011-4361-0 25. Buschle-Diller G, Hawkins A, Copper J (2006) Modified fibers with medical and specialty applications. Springer, The Netherlands, p 67 26. Felfel RM, Ahmed I, Parsons AJ, Haque P, Walker GS, Rudd CD (2010) J Biomater Appl. doi:10.1177/0885328210384532 27. Parsons AJ, Ahmed I, Haque P, Fitzpatrick B, Niazi MIK, Walker GS, Rudd CD (2009) J Bionic Eng 6:318. doi:10.1016/s16726529(08)60132-8 28. Oksman K, Skrifvars M, Selin JF (2003) Compos Sci Technol 63:1317. doi:10.1016/s0266-3538(03)00103-9 29. Ahmed I, Cronin PS, Abou Neel EA, Parsons AJ, Knowles JC, Rudd CD (2009) J Biomed Mater Res B 89:18. doi:10.1002/ jbm.b.31182 30. Gattin R, Copinet A, Bertrand C, Couturier Y (2002) Int Biodeterior Biodegrad 50:25. doi:10.1016/s0964-8305(02)00039-2 31. Wollerdorfer M, Bader H (1998) Ind Crop Prod 8:105. doi: 10.1016/s0926-6690(97)10015-2 32. Martin O, Ave´rous L (2001) Polymer 42:6209. doi:10.1016/ s0032-3861(01)00086-6 33. Sriupayo J, Supaphol P, Blackwell J, Rujiravanit R (2005) Carbohydr Polym 62:130. doi:10.1016/j.carbpol.2005.07.013 34. de Mesquita JOP, Donnici CL, Pereira FV (2010) Biomacromolecules 11:473. doi:10.1021/bm9011985 35. Bleach NC, Nazhat SN, Tanner KE, Kelloma¨ki M, To¨rma¨la¨ P (2002) Biomaterials 23:1579. doi:10.1016/s0142-9612(01)002836 36. Petersson L, Kvien I, Oksman K (2007) Compos Sci Technol 67:2535. doi:10.1016/j.compscitech.2006.12.012 37. Bondeson D, Oksman K (2007) Compos Part A Appl Sci Manuf 38:2486. doi:10.1016/j.compositesa.2007.08.001 38. Jonoobi M, Harun J, Mathew AP, Oksman K (2010) Compos Sci Technol 70:1742. doi:10.1016/j.compscitech.2010.07.005 39. Liu DY, Yuan XW, Bhattacharyya D, Easteal AJ (2010) Express Polym Lett 4:26 40. Sanchez-Garcia M, Lagaron J (2010) Cellulose 17:987. doi: 10.1007/s10570-010-9430-x 41. Revol JF, Bradford H, Giasson J, Marchessault RH, Gray DG (1992) Int J Biol Macromol 14:170 42. Labet M, Thielemans W (2011) Cellulose 18:607. doi: 10.1007/s10570-011-9527-x 43. ASTM D 2734-94 (1997) Annual book of ASTM standards 8(2) 44. Fischer EW, Sterzel HJ, Wegner G (1973) Kolloid Z Z Polym 51:980 45. Khurma JR, Rohindra DR, Devi R (2005) South Pac J Nat Sci 23:22 46. Segal SCJJ, Martin AE, Conrad CM (1959) Text Res J 29:786 47. Weimer PJ, Hackney JM, French AD (1995) Biotechnol Bioeng 48:169. doi:10.1002/bit.260480211 48. Elazzouzi-Hafraoui S, Nishiyama Y, Putaux JL, Heux L, Dubreuil F, Rochas C (2008) Biomacromolecules 9:57. doi: 10.1021/bm700769p 49. Thielemans W, Warbey CR, Walsh DA (2009) Green Chem 11:531
123
2686 50. Heath L, Thielemans W (2010) Green Chem 12:1448 51. Azizi Samir MAS, Alloin F, Dufresne A (2005) Biomacromolecules 6:612. doi:10.1021/bm0493685 52. Chua P, Dai S, Piggott M (1992) J Mater Sci 27:919. doi: 10.1007/bf01197642 53. Jawaid M, Khalil HPSA, Bakar AA, Khanam PN (2011) Mater Des 32:1014. doi:10.1016/j.matdes.2010.07.033 54. Mohan R, Kishore A (1985) J Reinf Plast Compos 4:186. doi: 10.1177/073168448500400203 55. Qua E, Hornsby P, Sharma H, Lyons G (2011) J Mater Sci 46:6029. doi:10.1007/s10853-011-5565-x 56. Ndazi BS, Karlsson S (2011) Express Polym Lett 5:119. doi: 10.3144/expresspolymlett.2011.13 57. Shumigin D, Tarasova E, Krumme A, Meier P (2011) Mater Sci (Medzˇiagotyra) 17:32 58. Oksman K, Etang JA, Mathew AP, Jonoobi M (2011) Biomass Bioenergy 35:146. doi:10.1016/j.biombioe.2010.08.021
123
J Mater Sci (2012) 47:2675–2686 59. Roman M, Winter WT (2004) Biomacromolecules 5:1048 60. Abou-Sekkina MM, Sakran MA, Saafan AA (1986) Ind Eng Chem Prod Res Dev 25:676. doi:10.1021/i300024a021 61. Yasuniwa M, Tsubakihara S, Iura K, Ono Y, Dan Y, Takahashi K (2006) Polymer 47:7554. doi:10.1016/j.polymer.2006.08.054 62. Shen L, Yang H, Ying J, Qiao F, Peng M (2009) J Mater Sci Mater Med 20:2259. doi:10.1007/s10856-009-3785-2 63. Shanmuganathan K, Capadona JR, Rowan SJ, Weder C (2010) Prog Polym Sci 35:212. doi:10.1016/j.progpolymsci.2009.10.005 64. Garcia de Rodriguez NL, Thielemans W, Dufresne A (2006) Cellulose 13:261. doi:10.1007/s10570-005-9039-7 65. Yew GH, Mohd Yusof AM, Mohd Ishak ZA, Ishiaku US (2005) Polym Degrad Stab 90:488. doi:10.1016/j.polymdegradstab.2005. 04.006 66. Angle`s MN, Dufresne A (2000) Macromolecules 33:8344. doi: 10.1021/ma0008701