ARTICLE pubs.acs.org/Biomac
Nanocomposite Films Based on Xylan-Rich Hemicelluloses and Cellulose Nanofibers with Enhanced Mechanical Properties Xin-wen Peng,† Jun-li Ren,*,† Lin-xin Zhong,‡ and Run-cang Sun*,†,‡ † ‡
State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou, 510640, China Institute of Biomass Chemistry and Utilization, Beijing Forestry University, Beijing 100083, China ABSTRACT:
Interest in xylan-rich hemicelluloses (XH) film is growing, and efforts have been made to prepare XH films with improved mechanical properties. This work described an effective approach to produce nanocomposite films with enhanced mechanical properties by incorporation of cellulose nanofibers (CNFs) into XH. Aqueous dispersions of XH (6475 wt %), sorbitol (1625 wt %), and CNF (020 wt %) were cast at a temperature of 23 °C and 50% relative humidity. The surface morphology of the films was revealed by scanning electron microscopy (SEM) and atomic force microscopy (AFM). The thermal properties and crystal structure of the films were evaluated by thermal analysis (TG) and X-ray diffraction (XRD). The surface of XH films with and without CNF was composed primarily of nanonodules, and CNFs were embedded in the XH matrix. Freeze-dried XH powder was amorphous, whereas the films with and without CNF showed a distinct peak at around 2θ = 18°, which suggested that XH molecules aggregated or reordered in the casting solution or during water evaporation. Furthermore, the nanocomposite films had improved thermal stability. XH film with 25 wt % plasticizer (sorbitol, based on dry XH weight) showed poor mechanical properties, whereas incorporation of CNF (520 wt %, based on the total dry mixture) into the film resulted in enhanced mechanical properties due to the high aspect ratio and mechanical strength of CNF and strong interactions between CNF and XH matrix. This effective method makes it possible to produce hemicellulose-based biomaterials of high quality.
’ INTRODUCTION Natural biopolymers have many advantages over synthetic polymers, such as low cost, biocompatibility, nontoxicity, and biodegradability. Hemicelluloses, comprising the noncellulose cell-wall polysaccharides of agricultural and forest plants, are considered to be inexhaustible and renewable resources for the production of functional biopolymers and biomaterials.1 Recently, considerable interest has been directed to hemicellulosesbased biomaterials because of their bioactive, biocompatible, and oxygen barrier properties, which give them potential in numerous applications, such as drug delivery, tissue engineering, and food packaging.24 Among these research activities, hemicellulose-based films have received ever-increasing interest. Konjac glucomannan (KGM) and galactoglucomannan (GGM) films were developed by the use of an external plasticizer like glycerol or sorbitol and exhibited good barriers against grease and oxygen. This makes them candidates for food-packaging materials.58 The mechanical properties of hemicellulose-based films could be r 2011 American Chemical Society
enhanced by blending GGM and KGM or poly(vinyl alcohol) (PVOH).9 Xylan-type hemicelluloses, which are the primary hemicelluloses in hardwood and annuals such as barley and wheat straw, are considered to be suitable materials for film preparation. Edible arabinoxylan-based films were prepared by emulsifying or coating.10,11 Film-forming properties of pure xylan but self-supporting films were produced in the presence of lignin or glycerol.4 Plasticized films made of aspen glucuronoxylan had low oxygen permeability and thus may be used as food packaging materials.12 Composite films were developed by incorporating up to 40% xylan in wheat gluten.13 Chemical modifications (e.g., fluorination, benzylation) were also carried out to lower the moisture sensitivity of hemicellulose-based films.14,15 Significantly, the film formation, structure, mechanical properties, and moisture Received: June 27, 2011 Revised: August 3, 2011 Published: August 05, 2011 3321
dx.doi.org/10.1021/bm2008795 | Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules content were affected by the molecular structure of hemicelluloses.8,16,17 The presence of side groups (e.g., arabinose, galactose) forced the xylan or mannan chains apart, so the most interactive hemicellulose was that in which the backbone was least substituted with arabinose or galactose.8,16,17 Although extensive work has been done to prepare the commercial potential of hemicelluloses-based biodegradable films, poor mechanical behavior is still a key problem that restricts their uses in a wide range of applications. In recent years, incorporation of biodegradable nano reinforcements such as cellulose whiskers (CNW), CNF, and bacterial cellulose into other polymers has already proven to be an important strategy for obtaining nanocomposites with high mechanical performance.1820 These biodegradable nanoreinforcements also offer great possibilities for the development of novel hemicellulose-based nanocomposite materials. Recently, an attempt has been made to prepare hemicellulose-based nanocomposite films by using CNW as nanoreinforcement. The addition of sulfated CNW led to substantial improvements in mechanical properties, oxygen permeability, and reduction in water transmission of xylan or KGM films.2127 CNF, a new class of nanoreinforcement with a wide variety of properties and functionalities for a wide range of applications in green nanocomposites, has tremendous attraction because of its unique characteristics such as good mechanical properties and high aspect ratio.19,28 CNF-reinforced composites showed higher mechanical properties than CNW-reinforced (except for the CNW isolated from tunicate) composites because of the higher aspect ratio of CNF.29,30 Additionally, the flexibility of CNF could produce more flexible composites as compared with the stiff CNW.31 In this study, in an attempt to improve the mechanical properties of hemicellulose-based films and thus facilitate their applications, CNF used as nanoreinforcement was incorporated into xylan films in the presence of plasticizers. To the best knowledge of the authors, there is little information available in the literature about the microstructure and mechanical properties of CNF-resinforced nanocomposite films based on glucomannan or xylan-rich hemicelluloses. This study provides a novel nanocomposite film based on xylan-rich hemicelluloses and offers an effective method to produce hemicelluloses-based biomaterials with enhanced mechanical properties.
’ EXPERIMENTAL SECTION Materials. Xylan-rich hemicelluloses (XHs) were isolated from holocellulose of bamboo (Dendrocalamus membranaceus Munro, DmM) using 10% KOH at 23 °C for 10 h with a solid-to-liquid ratio of 1:20 (g/mL). The holocellulose was obtained by delignification of the extractivefree DmM (4060 mesh) with sodium chlorite in acidic solution (pH 3.7 to 4.0, adjusted by 10% acetic acid) at 75 °C for 2 h. The XH obtained from DmM was freeze-dried. The sugar composition (relative weight percent) by the sugar analysis is: 89.38% xylose, 5.75% arabinose, 1.87% glucose, 0.66% galactose, 1.78% glucuronic acid, and 0.55% galacturonic acid. Sorbitol and bleached sisal pulp fibers were obtained from Shanghai Bio Life Science and Technology and Dongfang Sisal Group. Preparation of CNF. The bleached sisal pulp was first immersed in distilled water for 8 h and beaten in a PFI mill with 5000 revolutions at concentration of 10 wt %. The pretreated pulp was then subjected to the homogenizing action of a slit homogenizer (Laboratory Homogenizer 15 M, Gaulin, Everett, MA.). A 1% pulp suspension was passed through the slit repeatedly, and CNFs were obtained after 30 times, as shown in Figure 1.
ARTICLE
Figure 1. Optical images of sisal pulp fiber and CNF suspension, from left to right: before homogenization, passed through the slit 15 times, and passed through the slit 30 times.
Preparation of Nanocomposite Films. Films were prepared by mixing XH, sorbitol, CNF, and deionized water (35 mL) during magnetic stirring at 80 °C for 60 min. The total amount of dry substance (XH, sorbitol, and CNF) in each film was kept constantly at 1 g. The CNF contents in the nanocomposite films were 0.0, 5.0, 10.0, 15.0, and 20.0 wt % of the total dry mixture, respectively. The sorbitol content was kept 25.0 wt % (based on the dry weight of XH). The solutions were poured onto polystyrene dishes (12.5 12.5 cm) and allowed to dry in a temperature of 23 °C and a relative humidity (RH) of 50% for 10 days. Scanning Electron Microscopy Analysis. For microsurface analysis, the films were observed by SEM. The films were fixed on a metal stub using carbon tape and coated with gold using an Agar HR sputter coater. A Hitachi s-4300 scanning electron microscope (SEM) operating at 5 kV was used to obtain the secondary electron images. Atomic Force Microscopy Analysis. Nanomorphology of film surface was studied by atomic force microscopy (AFM) (Nanoscope III, Veeco). In AFM scanning, two to four interest locations on each sample were tested. Small pieces of films were glued onto metal disks and attached to a magnetic sample holder located on the top of the scanner tube. Topographic (height) and phase images were recorded under ambient air conditions. All of the images were recorded in tapping mode in air using silicon cantilevers with a resonance frequency between 250 and 300 kHz and a scan angle of 0°. Moreover, the AFM image analysis software allows a roughness analysis for film surface. The root-mean-square surface roughness (Rrms) was determined as the standard deviation of Z (height) values within the given area and calculated using the following equation
∑
Rrms ¼ RðF ðZi Zave Þ, NÞ where Zave is the average of the Z values within the given area, Zi is the current Z value, and N is the number of points within the given area. All of the surface roughness parameters were calculated from the AFM images with an AFM software program. Each surface roughness value presented was averaged from five separate measurements from different areas. Measurement of Thickness. Film thickness was measured using a micrometer (Lorentzen & Wettre, precision 1 μm). Measurement was taken at five different locations on each film, and the mean value was used in the calculations to determine the mechanical test measurement. Equilibrium Moisture Content. The equilibrium moisture content of the samples at the given constant temperature and RH and was measured. Three pieces of each film were conditioned at 23 °C and RH of 50% for each sample and were continuously weighted until equilibrium was reached. Finally, the samples were dried in an oven (105 °C) to reach constant mass and weighed again. The equilibrium moisture content was measured gravimetrically and calculated as the weight of water in the sample compared with the total weight. X-ray Diffraction. Diffractograms were recorded in reflection mode in the angular range of 440° (2θ) by steps of 0.02° (2θ). The measurements were done with a diffractometer (Bruker, model D8 advance). The Cu KR 3322
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules
ARTICLE
radiation generated at 40 kV and 40 mA was monochromatized using a 20 μm Ni filter. The scanning speed was 0.0385° s1. Samples were tested without being grinded. Thermal Analysis. Thermal analysis was performed by using thermogravimetric analysis (TGA) and differential thermal analysis (DTA) on a simultaneous thermal analyzer (Pyris Diamond TG/DTA, PE Instrument). The apparatus was continually flushed with nitrogen. The sample weighed between 9 and 11 mg and heated from room temperature to 500 °C at a heating rate of 10 °C/min. Tensile Strength Testing. The tensile strength of the films was measured with a tensile testing machine (Instron Universal testing machine 5565) fitted with a 200 N load cell. The samples were cut in the rectangular specimens with a width of 15 mm and length of 70 mm, and eight replicate specimens were tested from each film type. The initial distance between the grips was 25 mm, and the separation rate of the grips was kept constantly at 4 mm/min. The stressstrain curve was recorded for each sample. The measurements were performed at 23 °C and RH of 50%, and mechanical tensile data were averaged over eight specimens.
’ RESULTS AND DISCUSSION Film Formation. Recently, increasing efforts have been devoted to developing hemicellulose-based films. The mechanical performance of the films prepared from xylan alone is low because the films are rather brittle and too fragile to handle, which is a well-known issue for obtaining films from pure hemicelluloses.4,12,32 The poor ability of film forming is attributed to the insufficient chain length of the polymer, high glasstransition temperature, or poor solubility in water.12 For improvement of the film performance, it is necessary to introduce a second component into the films, for instance, xylitol, sorbitol, chitosan, carboxymethyl cellulose, glycerol, or even lignin.4,12,32 The addition of plasticizers (e.g., xylitol, sorbitol) in polymeric materials leads to modification in the molecular 3D organization and therefore to mechanical properties.33 In this study, sorbitol was used as plasticizer to confer the workability and flexibility of
Figure 2. Optical images of films: (a) controlled, (b) 5 wt % CNF reinforced, and (c) 20 wt % CNF reinforced.
XH films, as shown in Figure 2. Cracks occurred to the transparent XH film even in the presence of 25 wt % sorbitol (based on the dry XH weight, Figure 2a). Film formation involves a drying step, known to cause a collapse of film to some degree as the last layer of solvent evaporates.34 However, the parts of the film without cracks were flexible and free-standing and were used as control sample for testing. Addition of CNF, however, significantly improved the film formation, as indicated by the continuous nanocomposite films (Figure 2b,c). A continuous film could be obtained by incorporation of 5 wt % CNF into XH matrix, and no cracks were observed, being indicative of stronger mechanical strength. Cellulose nanofibers played an important role in preventing crack formation and growth, which resulted in composites with higher mechanical performance.3537 So the presence of CNF in the XH matrix could effectively prevent the formation and growth of cracks during drying, producing continuous nanocomposite films. CNFs also allowed chitosan to form continuous films of various shapes easily while reducing the number of creases and deformation of wet film.38 These results are encouraging because a continuous and self-supporting film can be obtained from hemicelluloses by simply incorporating CNF into XH matrix. Equilibrium Moisture Content. Table 1 shows the equilibrium moisture content of films. The control sample (sample 1) and CNF film (sample 6) exhibited the highest (12.55 wt %) and lowest (9.81 wt %) moisture content. The moisture uptake decreased as the CNF content increased. Because the moisture content of sorbitol is only 2.38 wt % and the drying process did not result in the loss of sorbitol (data not shown), the influence of sorbitol on the moisture content results can be negligible. Hemicelluloses are generally randomly branched molecules, whereas cellulose is unbranched. The disordered regions are likely to be the preferred sorption sites, and thus an increase in CNF content in the nanocomposite film leads to the decrease in the moisture content. Topography of Films. The topography of produced films was analyzed by SEM and AFM. AFM images with higher resolution were recorded to determine the surface structural information of the films, whereas the height images were generated for the additional morphology and surface roughness of the films. Representative images observed by SEM and AFM are shown in Figures 3 and 4. SEM image (Figure 3a) shows that the surface of the control XH film was not smooth but is instead composed of small nodules. The higher resolution AFM image (Figure 4a) reveals that the nodules with diameters of 1070 nm were tightly connected to each other, forming a nodular structure. Similar nodular structure was also reported for other hemicellulose-based films21,39 as well as films from other polymers such as cellulose,40
Table 1. Thickness, Moisture Content, and Surface Roughness of XH Films samples
XH + sorbitol (wt %)a
CNF (wt %)
thickness (μm)
moisture content (wt %)
Rrms (nm)b
2.38 ( 0.27
sorbitol 1
100
0
52 ( 2
12.55 ( 0.37
2.03
2
95
5
52 ( 3
12.26 ( 0.32
2.09
3
90
10
53 ( 2
11.59 ( 0.21
2.24
4
85
15
55 ( 2
11.30 ( 0.24
2.36
5
80
20
56 ( 3
11.13 ( 0.25
2.59
6
0
100
62 ( 3
9.81 ( 0.29
9.51
a Total mass of XH and sorbitol. The sorbitol content was 25 wt % of the XH for all samples. b These values were determined as averages from several images taken from topographically similar areas free of manufacturing defects.
3323
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules
ARTICLE
Figure 3. SEM images of films: (a) controlled, sample 1; (b) 100 wt % CNF, sample 6; (c) 10% CNF reinforced, sample 3; and (d) 20 wt % CNF reinforced, sample 5.
cellulose acetate,41 and starch.42 Great efforts have been made to investigate the nodule formation mechanism, and several theories have been developed. Some authors attribute nodule formation to the aggregates or micelles that were initially present in the casting solution.43 It was believed that nodules were the result of liquidliquid demixing by nucleation and growth of a polymer-rich phase.44 A surface phenomenon as a cause for nodule formation was also proposed.45 However, details of the formation mechanism of the nodular structure has not been agreed upon: no experimental evidence is available to support any of the proposed mechanisms. Kesting46 suggested that the skin layer of films mainly consisted of nodules, and the structure was rather compact. The functional pores were primarily the interstitial voids between closely packed nodules. Therefore, the internodular spaces can be considered to be defects that allow gas to permeate and cracks to develop under load. Interestingly, films obtained from xylan and chitosan showed a very smooth surface with few nodules,47 which indicates a strong interaction between XH and chitosan.
Precipitation of hemicelluloses occurred below an arabinose/ xylose ratio of 0.1.17 Aggregates began to form in an interval of the arabinose/xylose ratio between 0.23 and 0.31 in an aqueous solution. Therefore, removal of arabinose substituents will result in a gradual association of unsubstituted xylan chains. These results indicate that the intermolecular interaction can be tailored by controlling the substituted groups or incorporation of a secondary component that would interact with the molecular chains of hemicelluloses. The surface of CNF film (Figure 3b) appeared as an interconnected web structure. AFM images show (Figure 4b) that the diameter of CNF was 20 ( 10 nm, and the nanofiber length was estimated to be >1000 nm, resulting in a high aspect ratio 50 to 100. The surfaces of the nanocomposite films reinforced with CNF were also primarily composed of nodules, as shown in Figures 3c,d and 4c,d. It was difficult to identify the CNFs when their content was no more than 10 wt % (Figures 3c and 4d) because most of the CNFs were embedded in the XH matrix. As the CNF content increased up to 20 wt %, however, 3324
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules
ARTICLE
Figure 4. AFM images of films: (a) controlled, sample 1; (b) 100 wt % CNF, sample 6; (c) 10 wt % CNF reinforced, sample 3; and (d) 20 wt % CNF reinforced, sample 5. The left pictures are phase images and right pictures are 3D images from height phase. The scanning scale is 1 1 μm.
the randomly orientated cellulose nanofibers were detected (encircled in Figure 3d and arrowed in Figure 4d). This may indicate that a CNF network began to form within the XH matrix.
In this case, CNF were in contact with each other and surrounded by XH matrix, leading to a continuous CNF network in the matrix. 3325
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules Additionally, surface roughness (Rrms) obtained from AFM height images shows that surface roughness of XH film (2.03 nm) was much lower than that of CNF film (9.51 nm), as shown in Figure 4a,b and Table 1. This was due to the entanglement of CNF, also indicated by the thickness (62 μm). For this reason, an increase in CNF content from 5 to 20 wt % led to surface roughness increasing from 2.09 to 2.59 nm. X-ray Diffraction. XRD patterns were collected from freezedried XH powder (sample 0), XH film with 25 wt % sorbitol (sample 1, Table 1), nanocomposite film (sample 5, Table 1), and CNF film (sample 6, Table 1). The fractions of XH with and without sorbitol obtained by the water casting method were also analyzed to investigate further the crystal structure of XH
Figure 5. X-ray diffractions of films, from bottom to top: freeze-dried XH powder (sample 0), unplasticized XH film fractions (XHF) and plasticized XH film (sample 1), nanocomposite film (sample 5), and CNF film (sample 6).
ARTICLE
(Figure 5). The diffraction of the freeze-dried XH powder showed no distinct crystalline peak, whereas the unplasticized XH film (XHF) and plasticized XH film (sample 1) displayed a sharp distinct peak at 2θ = ∼18°. This result indicates that the freeze-dried XH powder is amorphous, whereas the XH films obtained by the water-casting method are semicrystalline. Therefore, XH chains aggregate or reorder in the casting solution or during water evaporation. It has been reported that hemicellulose-based films, for instance, xylan, KGM, and spruce GGM films, were semicrystalline.12,23 Arabinoxylan-based films had a small crystalline peak in the region of 2θ = 1721°.48,49 Glucuronoxylan-based films were also found to have a crystalline peak at 2θ = ∼18° because of the strong hydrogen bonds among hemicelluloses chains. Interestingly, the film obtained from xylan with high arabinose/xylose ratio (0.5) was found to be an amorphous material, whereas the arabinoxylan film with relatively low arabinose/xylose ratio (0.20 to 0.37) showed distinct crystallinity.16 The unsubstituted sequences of the xylan backbone may approach each other and form stable interchain associations and consequently form crystalline regions.50 The hindrance of crystallization by substituents was also observed in acetyl-substituted hemicelluloses.51 In this study, the arabinose/ xylose ratio was 0.064. The insufficient arabinose substitution in xylan chains led to easy association or aggregation of xylan chains in aqueous solution or during the film formation, which is supposed to contribute to the brittle XH film.12 Therefore, a crystalline peak at 2θ = 18° in the XRD of XH films indicates that some randomly oriented XH chains reorganized into crystals during casting, resulting in crystalline structure of XH films. The presence of plasticizer does not change the crystalline structure; the plasticized XH film (sample 1) also exhibits a crystalline peak at 2θ = 18°. CNF displayed the typical XRD pattern of native cellulose (sample 6), with characteristic diffraction peaks at 2θ = 22.7°.52 For nanocomposite film (sample 5) in
Figure 6. TG/DTA curves of freeze-dried XH powder (sample 0), unplasticized XH film fractions (XHF), and plasticized XH film (sample 1), nanocomposite film (sample 5), and CNF film (sample 6). 3326
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules
ARTICLE
Table 2. Tensile Testing Results of the Films Produced samplesa
a
tensile strength
tensile strain
Young’s modulus
(MPa)
(%)
(MPa)
1
11.9 ( 0.9
3.4 ( 0.2
735 ( 87
2
15.5 ( 1.2
2.9 ( 0.3
1322 ( 98
3
20.2 ( 2.3
2.6 ( 0.3
1578 ( 128
4
28.9 ( 1.9
1.8 ( 0.2
2355 ( 121
5
39.5 ( 2.2
1.4 ( 0.1
3404 ( 138
6
150.4 ( 5.7
1.1 ( 0.1
17212 ( 298
Corresponding to the samples in Table 1.
Figure 5, the sharp peak at 2θ = ∼22.7° is a characteristic of cellulose I.53 The unplasticized XH film (XHF) and plasticized XH film (sample 1) show a weak peak at 2θ = ∼22.7°. It is interesting because there is no cellulose present. (See the sugar analysis results in the Materials section.) The semicrystalline rye arabinoxylan and hardwood xylan (unplasticized and plasticized by sorbitol or xylitol) also showed a weak peak at 2θ = ∼22.7°.12,16 The identification of the underlying mechanisms is currently complicated to elucidate. One possible reason is that there are some similar intramolecular hydrogen bonds that exist in the crystal structure of xylan and cellulose.12 Thermogravimetric Analysis. Because of the differences in chemical structure between hemicelluloses and cellulose, they usually decompose at different temperatures.54 Figure 6 shows the typical TGA/DTA curves of the freeze-dried XH powder (sample 0), XH film fractions with and without sorbitol (sample 1 and XHF), nanocomposite film (sample 5), and CNF film (sample 6). All samples started to decompose at ∼200 °C, but the maximum decomposition rates differed, which is indicated by the Tmax (the decomposition temperature corresponding to the maximum weight loss rate) in Figure 6. The freeze-dried XH powder (sample 0) and plasticized XH film (sample 1) showed their Tmax at 278 and 285 °C, suggesting that the plasticized XH film had higher thermal stability mainly because of the presence of crystal structure. The unplasticized XH film also showed the Tmax at 285 °C, which is also attributed to the crystal structure of XH. These observations agree well with the results obtained from XRD. Nanocomposite film (sample 5) and CNF film (sample 6) had higher Tmax (301 °C). XH, which is extensively branched and has many amorphous regions, is very easy to degrade to volatiles CO, CO2, and some hydrocarbons, and so on. Because cellulose consists of a long polymer of glucose without branches, its structure is in a good order and is very strong, thus possessing higher thermal stability. The results indicated that the thermal stability of the nanocomposite film was improved. Tensile Testing. The average values of tensile stress, Young’s modulus, and tensile strain at break of the prepared films in Table 2 and the typical stressstrain curves in Figure 7 show the effect of CNF content on the mechanical properties of XH film. XH film with sorbitol (sample 1) was relatively weak, which was indicated by the maximum stress of 11.9 MPa. Incorporation of CNF could result in a significant improvement in the tensile strength of the nanocomposite films. The addition of 5 wt % CNF increased both the tensile stress and Young’s modulus of XH film by 30 and 80% (sample 2). At the CNF content of 20 wt % (sample 5), the tensile stress and Young’s modulus of the nanocomposite film increased up to 39.5 and 3404 MPa, which were much higher than those of XH film without CNF (sample
Figure 7. Tensilestrain curves of the films. The CNF content is indicated in the Figure. The inset is an expanded view of the films reinforced with 0, 5, 10, 15, and 20 wt % CNF (based on the total mixture weight).
1). This indicates the high load-bearing capacity. Neat CNF film (sample 6) exhibits excellent tensile strength and a rigid feature, which were suggested by the highest tensile stress (150.4 MPa) and Young’s modulus (17 212 MPa). In addition, as compared with the nanocomposite films obtained with 5 and 10 wt % CNF, films with 15 and 20 wt % CNF showed much higher Young’s modulus and more rigid feature (Figure 7, the inset). This may suggest the initial formation of CNF network within the XH matrix when the CNF content increased to 15 or 20 wt %, leading to interaction among CNFs. These results indicate that CNF can be used as a good nanoreinforcement to improve the mechanical properties of the hemicellulose-based films. Similar results were also reported for other nature macromolecules-based films reinforced with CNF. The tensile stress of amylopectin film increased from 0.35 to 15.0 MPa, and Young’s modulus increased from 1.6 to 780 MPa by incorporation of 20 wt % CNF into the film.53 High-performance nanocomposite films based on polyvinyl alcohol and CNF could be obtained from their mixture.55 The nanocomposite film showed an increase in tensile stress from 65 (pure polyvinyl alcohol film) to 103 MPa (composite film with 5 wt % CNF).55 Polylactic acid nanocomposites reinforced with CNF also showed improved mechanical properties.5658 Insufficiency of arabinose substituents will cause more insoluble fractions, which may act as stress concentrators and lead to a material with lower tensile properties.16 Therefore, cracks occurred during film formation, and the tensile stress and Young’s modulus strength were relatively low. Improvement in the tensile strength of the nanocomposite film as a consequence of the addition of CNF could be ascribed to the high aspect ratio, good mechanical strength of CNF, and the strong interactions between CNF and XH matrix could restrict the segmental mobility of the polymer chains in the vicinity of the nanoreinforcements and interface destroying.30,59 The interface between the matrix and reinforcement plays a critical role in determining the external load transfer within the films. The strong hydrogen bonding (interactions between hydroxyl groups) and van der Waals forces between the CNF and the matrix (XH) allow good interfacial adhesion between XH matrix and the reinforcement (CNF).30,55,60 The interactions 3327
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules (hydrogen bonding and van der Waals forces) between XH and cellulose were used to develop molecular anchors, which gives new chemical functions to cellulose crystals.61 Cerclier et al.62 also took advantage of the interactions to produce multilayered films composed of xyloglucan and cellulose nanocrystals using a layer-by-layer construction approach.62,63 For the nanocomposites films, external load can be effectively transferred between matrix and the CNFs due to the strong interfacial interactions, which results in the reduction in stress concentration and crack formation within the nanocomposite films. These results show the possibility of forming hemicellulose-based films with high mechanical performance.
’ CONCLUSIONS The first attempt to introduce cellulose nanofibers into XH to produce nanocomposite films with significantly enhanced mechanical properties was developed in this study. SEM and AFM images showed that the surface of the XH film and the nanocomposite film reinforced with CNF were primarily composed of nodules that had a diameter of 1070 nm, and CNFs were embedded in the XH matrix. The XH film was semicrystalline, which was indicated by the distinct peak at 2θ = ∼18°. The nanocomposite film exhibited improved thermal stability as compared with the XH film without CNF. Incorporation of CNF resulted in better film formation and a significant improvement in the tensile strength that could be ascribed to the high aspect ratio and strong interactions between CNF and the XH matrix. These results suggest an effective and simple method to produce hemicelluloses-based film of high quality. ’ AUTHOR INFORMATION Corresponding Authors
*E-mail:
[email protected] (J.-l.R.),
[email protected] (R.-c.S.). Phone/Fax: +86-20-87111861.
’ ACKNOWLEDGMENT This work was supported by the grants from National Natural Science Foundation of China (nos. 31070530 and 30930073), Ministry of Science and Technology (973-2010CB732201/4), Guangdong Natural Science Foundation (no. 07118057), Foundation for Distinguished Young Talents in Higher Education of Guangdong, China (LYM09017), and the Fundamental Research Funds for the Central Universities (2009ZM0153), SCUT. ’ REFERENCES (1) Ebringerova, A.; Hromadkova, Z.; Heinze, T. Adv. Polym. Sci. 2005, 186, 1–67. (2) Hansen, N. M. L.; Plackett, D. Biomacromolecules 2008, 9, 1493– 1505. (3) Lindblad, M. S.; Ranucci, E.; Albertsson, A. C. Macromol. Rapid Commun. 2001, 22, 962–967. (4) Goksu, E. I.; Karamanlioglu, M.; Bakir, U.; Yilmaz, L.; Yilmazer, U. J. Agric. Food Chem. 2007, 55, 10685–10691. (5) Mikkonen, K. S.; Heikkila, M. I.; Helen, H.; Hyvonen, L.; Tenkanen, M. Carbohydr. Polym. 2010, 79, 1107–1112. (6) Hartman, J.; Albertsson, A. C.; Lindblad, M. S.; Sjoberg, J. J. Agric. Food Chem. 2006, 100, 2985–2991. (7) Cheng, L. H.; Karim, A. A.; Seow, C. C. Food Chem. 2008, 107, 411–418. (8) Mikkonen, K. S.; Rita, H.; Helen, H.; Talja, R. A.; Hyvonen, L.; Tenkanen, M. Biomacromolecules 2007, 8, 3198–3205.
ARTICLE
(9) Mikkonen, K. S.; Yadav, M. P.; Cooke, P.; Willfor, S.; Hicks, K. B.; Tenkanen, M. Bioresources 2008, 3, 178–191. (10) Peroval, C.; Debeaufort, F.; Despre, D.; Voilley, A. J. Agric. Food Chem. 2002, 50, 3977–3983. (11) Phan The, D.; Debeaufort, F.; Peroval, C.; Despre, D.; Courthaudon, J. L.; Voilley, A. J. Agric. Food Chem. 2002, 50, 2423–2428. (12) Grondahl, M.; Eriksson, L.; Gatenholm, P. Biomacromolecules 2004, 5, 1528–1535. (13) Kayseriliolu, B.; Bakir, U.; Yilmaz, L.; Akkas-, N. Bioresour. Technol. 2003, 87, 239–246. (14) Grondahl, M.; Gustafsson, A.; Gatenholm., P. Macromolecules 2006, 39, 2718–2721. (15) Hartman, J.; Albertsson, A. C.; Sjoberg, J. Biomacromolecules 2006, 7, 1983–1989. (16) Hoije, A.; Sternemalm, E.; Heikkinen, S.; Tenkanen, M.; Gatenholm, P. Biomacromolecules 2008, 9, 2042–2047. (17) Sternemalm, E.; H€ oije, A.; Gatenholm, P. Carbohydr. Res. 2008, 343, 753–757. (18) Habibi, Y.; Lucia, L. A.; Rojas, O. J. Chem. Rev. 2010, 110, 3479–3500. (19) Eichhorn, S. J.; Dufresne, A.; Aranguren, M.; Marcovich, N. E.; Capadona, J. R.; Rowan, S. J.; Weder, C.; Thielemans, W.; Roman, M.; Renneckar, S.; Gindl, W.; Veigel, S.; Keckes, J.; Yano, H.; Abe, K.; Nogi, M.; Nakagaito, A. N.; Mangalam, A.; Simonsen, J.; Benight, A. S.; Bismarck, A.; Berglund, L. A.; Peijs, T. J. Mater. Sci. 2010, 45, 1–33. (20) Nakagaito, A. N.; Iwamoto, S.; Yano, H. Appl. Phys. A: Mater. Sci. Process. 2005, 80, 93–97. (21) Saxena, A.; Elder, T. J.; Pan, S.; Ragauskas, A. J. Composites, Part B 2009, 40, 727–730. (22) Petersson, L.; Oksman, K. Compos. Sci. Technol. 2006, 66, 87–96. (23) Mikkonen, K. S.; Mathew, A. P.; Pirkkalainen, K.; Serimaa, R.; Xu, C. L.; Willfor, S.; Oksman, K.; Tenkanen, M. Cellulose 2010, 17, 69–81. (24) Rusli, R.; Shanmuganathan, K.; Rowan, S. J.; Weder, C.; Eichhorn, S. J. Biomacromolecules 2011, 12, 1363–1369. (25) Saxena, A.; Ragauskas, A. J. Carbohydr. Polym. 2009, 78, 357–360. (26) Saxena, A.; Elder, T. J.; Ragauskas, A. J. Carbohydr. Polym. 2011, 84, 1371–1377. (27) Saxena, A.; Elder, T. J.; Kenvin, J.; Ragauskas, A. J. Nano-Micro Lett. 2010, 2, 235–241. (28) Samir, M. A. S. A.; Alloin, F.; Paillet, M.; Dufresne, A. Macromolecules 2004, 37, 4313–4316. (29) Cheng, Q. Z.; Wang, S. Q.; Rials, T. G. Composites, Part A 2009, 40, 218–224. (30) Siqueira, G.; Bras, J.; Dufresne, A. Biomacromolecules 2008, 10, 425–432. (31) Fujisawa, S.; Okita, Y.; Fukuzumi, H.; Saito, T.; Isogai, A. Carbohydr. Polym. 2011, 84, 579–583. (32) Edlund, U.; Ryberg, Y. Z.; Albertsson, A. C. Biomacromolecules 2010, 11, 2532–2538. (33) Cuq, B.; Gontard, N.; Guilbert, S. Polymer 1997, 38, 2399–2405. (34) Stokke, B. T.; Elgsaeter, A.; Brant, D. A.; Kitamura, S. Macromolecules 1991, 24, 6349–6351. (35) Nakagaito, A. N.; Yano, H. Appl. Phys. A: Mater. Sci. Process. 2004, 78, 547–552. (36) Iwamoto, S.; Nakagaito, A. N.; Yano, H. Appl. Phys. A: Mater. Sci. Process. 2007, 89, 461–466. (37) Okubo, K.; Fujii, T.; Thostenson, E. T. Composites, Part A 2009, 40, 469–475. (38) Nordqvist, D.; Idermark, J.; Hedenqvist, M. S. Biomacromolecules 2007, 8, 2398–2403. (39) Simkovic, I.; Gedeon, O.; Uhliarikova, I.; Mendichic, R.; Kirschnerova, S. Carbohydr. Polym. 2011, 83, 769–775. (40) Zhang, Y.; Shao, H.; Hu, X. J. Appl. Polym. Sci. 2002, 86, 3389–3395. (41) Stamatialis, D. F.; Dias, C. R.; de Pinho, M. N. J. Membr. Sci. 1999, 160, 235–242. (42) Thire, R. M. S. M.; Simao, R. A.; Andrade, C. T. Carbohydr. Polym. 2003, 54, 149–158. 3328
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
Biomacromolecules
ARTICLE
(43) Panar, M.; Hoehn, H. H.; Hebert, R. R. Macromolecules 1973, 6, 777–780. (44) Kamide, K.; Iijima, H.; Matsuda, S. Polym. J. 1993, 25, 1113–1131. (45) Wienk, I. M.; van den Boomgaard, T.; Smolders, C. A. J. Appl. Polym. Sci. 1994, 53, 1011–1023. (46) Kesting, R. E. J. Appl. Polym. Sci. 1990, 41, 2739–2752. (47) Gabrielii, I.; Gatenholm, P. J. Appl. Polym. Sci. 1998, 69, 1661–1667. (48) Mikkonen, K. S.; Heikkinen, S.; Soovre, A.; Peura, M.; Serimaa, R.; Talja, R A.; Helen, H.; Hyvonen, L.; Tenkane, M. J. Appl. Polym. Sci. 2009, 114, 457–466. (49) Hoije, A.; Grondahl, M.; Tømmeraas, K.; Gatenholm, P. Carbohydr. Polym. 2005, 61, 266–275. (50) Andrewartha, K. A.; Phillips, D. R.; Stone, B. A. Carbohydr. Res. 1979, 77, 191–204. (51) Gr€ondahl, M.; Gatenholm, P. In Polysaccharides: Structural Diversity and Functional Versatility, 2nd ed.; Dumitriu, S., Eds.; Marcel Dekker: New York, 2005; pp 509514. (52) Tingaut, P.; Zimmermann, T.; Lopez-Suevos, F. Biomacromolecules 2010, 11, 454–464. (53) Svagan, A. J.; Samir, M. A. S. A.; Berglund, L. A. Biomacromolecules 2007, 8, 2556–2563. (54) Yang, H.; Yan, R.; Chen, H.; Dong, H. L.; Zheng, C. Fuel 2007, 86, 1781–1788. (55) Wang, B.; Sain, M. Compos. Sci. Technol. 2007, 67, 2521–2527. (56) Iwatake, A.; Nogi, M.; Yano, H. Compos. Sci. Technol. 2008, 68, 2103–2106. (57) Suryanegara, L.; Nakagaito, A. N.; Yano, H. Compos. Sci. Technol. 2009, 69, 1187–1192. (58) Jonoobi, M.; Harun, J.; Mathew, A. P.; Oksman, K. Compos. Sci. Technol. 2010, 70, 1742–1747. (59) Bondeson, D.; Syre, P.; Nisk,a, K. O. J. Biobased Mater. Bioenergy 2007, 1, 367–371. (60) Hanus, J.; Mazeau, K. Biopolymers 2006, 82, 59–73. (61) Zhou, Q.; Rutland, M. W.; Teeri, T. T.; Brumer, H. Cellulose 2007, 14, 625–641. (62) Cerclier, C.; Cousin, F.; Bizot, H.; Moreau, C.; Cathala, B. Langmuir 2010, 26, 17248–17255. (63) Jean, B.; Heux, L.; Dubreuil, F.; Chambat, G.; Cousin, F. Langmuir 2009, 25, 3920–3923.
3329
dx.doi.org/10.1021/bm2008795 |Biomacromolecules 2011, 12, 3321–3329
