Ductile All-Cellulose Nanocomposite Films ... - ACS Publications

Article pubs.acs.org/Biomac

Ductile All-Cellulose Nanocomposite Films Fabricated from Core− Shell Structured Cellulose Nanofibrils Per A. Larsson,*,†,‡ Lars A. Berglund,†,‡,§ and Lars Wågberg†,‡,§ †

KTH Fibre and Polymer Technology, KTH Royal Institute of Technology, SE-100 44 Stockholm, Sweden BiMaC Innovation, KTH Royal Institute of Technology, SE-100 44 Stockholm, Sweden § Wallenberg Wood Science Center, KTH Royal Institute of Technology, SE-100 44 Stockholm, Sweden ‡

S Supporting Information *

ABSTRACT: Cellulosic materials have many desirable properties such as high mechanical strength and low oxygen permeability and will be an important component in a sustainable biomaterial-based society, but unfortunately they often lack the ductility and formability offered by petroleumbased materials. This paper describes the fabrication and characterization of nanocomposite films made of core−shell modified cellulose nanofibrils (CNFs) surrounded by a shell of ductile dialcohol cellulose, created by heterogeneous periodate oxidation followed by borohydride reduction of the native cellulose in the external parts of the individual fibrils. The oxidation with periodate selectively produces dialdehyde cellulose, and the process does not increase the charge density of the material. Yet the modified cellulose fibers could easily be homogenized to CNFs. Prior to film fabrication, the CNF was shown by atomic force microscopy to be 0.5−2 μm long and 4−10 nm wide. The films were fabricated by filtration, and besides uniaxial tensile testing at different relative humidities, they were characterized by scanning electron microscopy and oxygen permeability. The strength-at-break at 23 °C and 50% RH was 175 MPa, and the films could, before rupture, be strained, mainly by plastic deformation, to about 15% and 37% at 50% RH and 90% RH, respectively. This moisture plasticization was further utilized to form a demonstrator consisting of a double-curved structure with a nominal strain of 24% over the curvature. At a relative humidity of 80%, the films still acted as a good oxygen barrier, having an oxygen permeability of 5.5 mL·μL/(m2·24 h·kPa). These properties indicate that this new material has a potential for use as a barrier in complex-shaped structures and hence ultimately reduce the need for petroleum-based plastics.

■

INTRODUCTION The search for and development of novel bio-based packaging and barrier materials that can replace conventional nonrenewable materials is of great importance. Cellulose, being the most abundant biopolymer on Earth, has many advantageous properties such as good mechanical strength and low oxygen permeability and is therefore a desirable material source in a future biomaterial-based and sustainable economy. In recent years, several research groups have fabricated films and coatings from micro- and nanofibrillated cellulose, that is, cellulose where the 4−5 nm wide crystalline fibrils or 10−60 nm wide aggregates of these have been mechanically and chemically separated from each other. This research has been the topic of several review papers.1−3 The crystalline regions of these ordered supramolecular entities have a modulus of about 130 GPa,4,5 but the modulus of films fabricated from cellulose nanofibrils (CNFs), which seldom consist entirely of fully liberated nanoparticles, are still limited to about one tenth of this value, having a typical tensile strength of 130−230 MPa and a strain-at-break in the range of 3−10%.6−9 High-density films made of fibrillated or dissolved celluloses, as well as other polysaccharides, are also known to act as good © 2014 American Chemical Society

oxygen barriers, especially in the absence of moisture, and can possibly be a candidate for the replacement of petroleum-based plastics in, for example, various packaging applications.10−16 If exfoliated clay nanoplatelets are included with the CNFs as a nanocomposite, it is also possible to further lower the gas permeability of the material.17−20 Recent studies also show that CNF films may find use in flexible-electronics applications.21 However, the still relatively low strain-at-break of cellulose films compared with that of plastics limits the possibility to form nonplasticized cellulose films with a complex geometrical shape. An alternative to using a conventional plasticizer to improve the ductility of the cellulose films is to utilize a nanocomposite approach and add a more ductile component to the CNFs.22 This approach may, however, face fabrication difficulties if the ductile component is too small, is noncompatible with cellulose, or for some other reason cannot be homogeneously dispersed together with the cellulose component. In a recent paper from the laboratory of the authors,23 paper sheets with Received: March 7, 2014 Revised: April 15, 2014 Published: April 28, 2014 2218

dx.doi.org/10.1021/bm500360c | Biomacromolecules 2014, 15, 2218−2223

Biomacromolecules

Article

Scheme 1. Periodate (IO4−) Oxidation of One of the Two C2−C3 Bonds in an Anhydroglucose Unit To Form Dialdehyde Cellulose, Followed by a Subsequent Reduction with Borohydride (BH4−) to Dialcohol Cellulose

thorough washing with deionized water. The fibers were then resuspended to 4 g/L, and the dialdehyde cellulose formed was reduced to dialcohol cellulose (right-hand reaction of Scheme 1) by adding 0.5 g of sodium borohydride per gram of fiber. To limit the pH increase on addition of sodium borohydride and thus protect the dialdehyde cellulose from alkaline depolymerization,32,33 the reduction mixture also contained 0.01 M monobasic sodium phosphate. For the reduction step, the suspension was kept under gentle stirring for 4 h, followed by filtration and thorough washing. Fibril Preparation. CNFs were prepared by homogenization (Microfluidics’ Microfluidizer processor M-110 EH) of a 4 g/L fiber suspension at a pressure of 1600 bar. The fibers were homogenized by one pass through 400 and 200 μm chambers connected in series, and three passes through 200 and 100 μm chambers connected in series. The CNF was then stored at 4 °C until further use. Fibril Characterization. To estimate the dimensions of the nanofibrils, a polished silicon wafer (MEMC Electronic Materials, Italy) with a preadsorbed PEI layer (5 min adsorption at 0.1 g/L) was dipped for 10 s in a 20 times diluted CNF suspension, followed by gentle blow-drying with nitrogen and measurement using atomic force microscopy (AFM) in the tapping mode on a Veeco Instruments’ Multi-Mode IIIa, using a model MMP-12100 cantilever. Film Fabrication. CNFs diluted with deionized water to 2 g/L were dispersed for 5 min at 15 000 rpm using an IKA UltraTurrax disperser and 10 min in a sonication bath (VWR Ultrasonic Cleaner) and then vacuum filtered using a 510 × 3600 mesh Twill Dutch Double Weave (BoppUtildi, Sweden). After filtration, a second metal weave was placed on top of the wet film, and the whole assembly was dried for 15 min at 93 °C under a reduced pressure of 95 kPa using the dryer of a Rapid Köthen sheet former (Paper Testing Instruments, Austria). The films were stored at 23 °C and 50% RH until further testing. Scanning Electron Microscopy (SEM). A high-resolution fieldemission scanning microscope (Hitachi S-4800) was used to acquire micrographs of the fabricated films. In order to suppress specimen charging during imaging, the specimens were sputtered for 20−30 s using a Pt−Pd target in a 208 HR Cressington sputter coater. Oxygen Permeability. The oxygen barrier properties of the films were evaluated according to the ASTM D-3985 standard on 5 cm2 samples in a Systech Instruments’ model 8001 oxygen permeation analyser. The measurements were carried out in quadruplicate and were performed symmetrically, that is, with the same relative humidity on both sides of the test specimen. Film Thickness and Density. The apparent film thickness and density were determined at 23 °C and 50% RH by measuring the thickness at ten random locations on each film using a digital Mitutoyo thickness gauge and placing each film of known size on a laboratory balance. Tensile Testing. Tensile tests were performed in an MTS Systems Corporation tensile tester equipped with a 100 N load cell (the actual force measured was in the range of 5−25 N) in a climate-controlled room, with a constant temperature of about 23 °C, at three different relative humidities, 50%, 75%, and 90% RH. The moisture content at each relative humidity was determined by weighing a film prior to each set of tensile tests, followed drying in an oven overnight. The test pieces were 5 mm wide and about 30 μm thick, were clamped with a free span of 20 mm, and were strained at a rate of 2 mm/min. The strain was determined by measuring the grip displacement; the Young’s modulus was calculated as the slope of the stress−strain curve

both high strength and a strain-at-break greater than 10% were produced by modifying the nanofibril structure in a core−shell structure, that is, a schematic structure earlier described by Matsumura et al.,24 of cellulose and dialcohol cellulose by heterogeneous periodate oxidation followed by borohydride reduction (Scheme 1). The formation of the core−shell structure was experimentally shown by X-ray diffraction measurements and a loss in crystallinity corresponding to the separately measured degree of modification. The study of the mechanical properties of the sheets indicated that the ductility was improved not only on the sheet level, affecting the fiber− fiber contact zone, but also on a single fiber level, probably due to a strain-induced restructuring of nanofibrils within each individual fiber. This suggests that this kind of core−shell modification also will affect the ductility of films made from CNFs produced from these fibers. In this context it is worth mentioning that cellulose−dialcohol cellulose films have been made from homogeneously oxidized and reduced cellulose by first dissolving the cellulose.25−27 These films however did not show the same intriguing combination of strength and ductility as paper sheets made from fibers where the nanofibril constituents had been core−shell modified. The present study therefore aims at exploring the possibility of disintegrating these core−shell modified CNFs from the fibers and reassembling them as a strong and ductile all-cellulose composite film, that is, a film composed of only native cellulose and chemically or structurally modified cellulose,28 and to study their film-forming potential, film properties, and potential for use in complex-shaped structures.

■

EXPERIMENTAL SECTION

Materials. Bleached unbeaten softwood kraft fibers (K44), supplied by SCA Forest Products (Ö strand pulp mill, Timrå, Sweden), were used as cellulose source. The chemicals needed during the oxidation and reduction steps, sodium (meta)periodate, sodium borohydride, isopropanol (≥99.8% purity), and sodium phosphate, were all supplied by Sigma-Aldrich. Polyethylenimine (PEI), with a molecular weight of 60 kDa, used as anchoring layer for CNF adsorption prior to AFM imaging, was supplied by Acros Organics. The chemicals were used as received without further purification. Cellulose Modification. To partially oxidize the cellulose to dialdehyde cellulose (left-hand reaction in Scheme 1), 5.4 g of sodium periodate was added per gram of fiber to a gently stirred 4 g/L fiber suspension. Note that the objective is to modify the cellulose chains in the outer part of the cellulose nanofibrils (by which the cellulose fibers are composed), that is, to perform a heterogeneous modification. The suspension also contained 6.3 vol % isopropanol as radical scavenger to prevent side reactions and chain scission.29,30 The fibers were oxidized for 12 h at room temperature, and to further limit side reactions, the reaction was performed in the dark.31 This protocol is known to have a yield of 80−85% and to result in a dialdehyde cellulose content of approximately 3.4 mmol/g of fiber (i.e., about 30% conversion, which was found to maximize the strength of paper sheets) and a charge density of about 17 μequiv/g.23 The oxidation reaction was stopped by filtering off the sodium periodate, followed by 2219


Biomacromolecules

Article

in the region of 0.2−0.6% strain, and the yield point was determined using the offset method (cf. ref 34), using an offset of 0.3%. A total of 12 test pieces were tested at each relative humidity. Formation of Double-Curved Films. A circular film with a diameter of approximately 50 mm was firmly clamped on a female die held by a custom-made rig with four springs that upon loading share some of the applied force to lower the actual pressure on the film (the forming unit is shown in Figure 4a). The rig was placed on the bottom plate of a Fontijne TP400 press (Fontijne Grotnes, The Netherlands), and a corresponding male die was attached by a magnet to the top plate of the press, and a double-curved structure was consequently formed by pressing the two dies together. The press was limited to an operational force of 20 kN, which resulted in a pressure on the film of approximately 300 kPa. The deformed area had a diameter of 30 mm, and the nominal strain along the curvature of the die used was 24%. To plasticize the film prior to formation, it was conditioned overnight in a jar containing a saturated barium chloride solution, that is, at a relative humidity of about 90%, and to maintain the moisture in the film during forming of the double-curved structure, it was placed between two thin polyethylene films.

The opening of the cellulose C2−C3 bond (Scheme 1) does not introduce any significant amount of carboxyl groups,23 and after a homogenization that easily was performed without clogging of the homogenizer at a concentration of about 4 g/L, the material still had a relatively low viscosity and was not particularly prone to form a gel. In fact, during the dewatering stage of the film fabrication, the CNFs formed a rather thin, dense, and weak gel, which presumably contributed to the long dewatering time of about 8 h required to produce an approximately 30 μm thick film after complete drying. Ultimately, after being dried and conditioned at 23 °C and 50% RH, the films had an apparent density of about 1420 kg/m3, which indicates a solid film with few large pores since a cellulose crystal has a density of about 1600 kg/m3.35 This low porosity and good film formation, as well as the CNF dimensions determined by AFM (Figure 1), were further supported by SEM (Figure 2). Altogether, the density and

■

RESULTS AND DISCUSSION CNFs were successfully produced by homogenization of sequentially periodate-oxidized and borohydride-reduced cellulose fibers at 4 g/L. This specific chemistry on solid cellulose is heterogeneous, and an analysis of X-ray diffraction data for this degree of modification has shown a decrease in degree of crystallinity and a decrease from about 3.7 nm to about 3.2 nm in Scherrer crystallite width, which suggests that a fibril core− shell structure has been formed.23 The AFM data in Figure 1

Figure 2. SEM images of the fabricated films at (a) low and (b) high magnification. The pattern seen in panel a is caused by the weave used both as filter and drying support.

Figure 1. AFM height image (5 × 5 μm2) of shell−core structured CNF adsorbed onto a silica surface with an anchoring layer of PEI. The inset graph shows the height profiles along the three lines marked in the AFM image.

physical appearance of both nanofibrils and films made therefrom suggest that the dialcohol cellulose forms a more or less continuous matrix that is reinforced by crystalline CNFs, that is, the film can indeed be considered as an all-cellulose nanocomposite. The distinct diamond-shaped pattern seen in Figure 2a is caused by the weave used for filtration and drying support. This surface structure needs to be remembered in the later discussion on film transparency.

show that the CNFs produced had a width of 4−10 nm and a length ranging from about 0.5 μm to about 2 μm. These are dimensions well in agreement with CNFs produced from other cellulosic fibers, both untreated and chemically modified before homogenization.1−3 The homogenized material also contained some larger fibril aggregates, qualitatively evident in the lack of complete transparency of the CNF dispersion. 2220


Biomacromolecules

Article

The core−shell model developed in prior work on cellulose fibers suggests that the fabricated films should have a high modulus and strength-at-break while having a high ductility, which is further plasticized by moisture.23 Tensile tests were therefore performed at three different relative humidities, 50%, 75%, and 90% RH. Sheets produced from the same fibers, following the modification protocol described above, had a strength-at-break of 93 MPa (at a density of 925 kg/m3) and a strain-at-break of 11%,23 while the films fabricated from shell− core modified CNFs were stronger and more ductile with a strength-at-break of about 175 MPa and an average strain-atbreak of 15%, as shown in Figure 3. This result supports the

Table 1. Mechanical Testing Data of Films Made of Modified and Unmodified CNFsa modified Mechanical Properties at 50% RH strength-at-break (MPa) 175 ± 6 strain-at-break (%) 14.8 ± 0.9 Young’s modulus (GPa) 10.8 ± 0.5 yield strength (MPa) 117 ± 6 yield strain (%) 1.4 ± 0.0 strain hardening modulus (GPa) 0.35 ± 0.01 Mechanical Properties at 75% RH strength-at-break (MPa) 79 ± 6 strain-at-break (%) 23.0 ± 3.2 Young’s modulus (GPa) 4.4 ± 0.4 yield strength (MPa) 39 ± 3 yield strain (%) 1.3 ± 0.1 strain hardening modulus (GPa) 0.16 ± 0.01 Mechanical Properties at 90% RH strength-at-break (MPa) 42 ± n strain-at-break (%) 37.2 ± 3.2 Young’s modulus (GPa) 1.5 ± 0.2 yield strength (MPa) 13 ± 1 yield strain (%) 1.3 ± 0.2 strain hardening modulus (GPa) 0.07 ± 0.00

unmodifiedb 238 ± 20 5.7 ± 1.2 11.5 ± 0.2 117 ± 4 1.3 ± 0.0 2.55 ± 0.28 c c c c c c c c c c c c

a

Values are given with 95% confidence limits. bValues calculated from Larsson et al.36 using the same definitions as in this work. cNot measured. Figure 3. Characteristic stress−strain curves of films made of core− shell modified CNFs. Each curve represents a test piece conditioned at a specific relative humidity. Quantitative data are provided in Table 1.

core−shell modified CNFs form an all-cellulose nanocomposite film with a ductile and continuous dialcohol cellulose phase reinforced by stiff CNF cores. However, pure dialcohol cellulose films have a strain-at-break of several hundred percent,25,27 compared with pure HEC films, which can be strained to about 35%.22 A system with dialcohol cellulose might therefore provide greater ductility and consequently have a larger range of applications where ductility is the critical parameter. Besides being plasticized by water, pure dialcohol cellulose has a glass transition temperature in the range of 80−100 °C,25,26 which presumably allows pure dialcohol cellulose to be heat-processed in various ways. The current all-cellulose composite did not however show any rapid loss in storage modulus when heated in a dynamic mechanical analyzer. The change was of a gradual nature, losing about 80% in storage modulus between 25 and 250 °C, with a small shoulder in the tan δ−temperature plot at about 90 °C and a larger peak at about 220 °C (see Figure S2 in the Supporting Information). In a packaging application, as in the storage and transport of food, a film should provide not only mechanical protection but also a good gas barrier. Cellulose films are in general good oxygen barriers,12,15,17,19 and the present results are no exceptions. At 50% and 80% RH, the films fabricated from CNFs partly converted to dialcohol cellulose had oxygen permeabilities of 0.4 ± 0.2 mL·μm/(m2·24 h·kPa) and 5.5 ± 0.8 mL·μm/(m2·24 h·kPa), respectively. This is on a par with or even slightly better than films of the same density made of the same but unmodified cellulose.36 The current results can also be compared with literature data by, for example, Aulin et al.,17 who fabricated thin films from carboxymethylated CNFs that were significantly more sensitive to moisture, having an average oxygen permeability of 0.5 mL·μm/(m2·24 h·kPa) at

core−shell hypothesis that a highly mobile shell of dialcohol cellulose allows fibril sliding when strained. As can be seen in Figure 3, most of the deformation was of a plastic nature. At 50% RH, the material showed an average yield strength of 117 MPa at about 1.4% strain, followed by a linear strain hardening region with a strain hardening modulus of 0.35 GPa. This can be compared with the unmodified material,36 which had the same yield strength and yield strain but a seven times greater strain hardening modulus, emphasizing that the core−shell modification mainly affects the plastic behavior of the material (Table 1). Both the moisture content in the films and the strain-at-break increased with increasing relative humidity, and at 90% RH, the films could be strained to almost 40%, which, to the best of our knowledge, is the most ductile high-density film made from CNFs without the addition of a separate plasticizer (apart from water). As the relative humidity increased, the ultimate failure mechanism switched from a catastrophic crack growth mode to a mode of extensive local plastic deformation involving fibril or fibril aggregate reorientation, fracture, and pull-out. Initiations of microcracks and some stress whitening could also be seen (see Figure S1 in the Supporting Information). In a study by Sehaqui et al.,22 nanocomposite films with enzyme-pretreated CNFs and hydroxyethyl cellulose (HEC) were fabricated at a volume ratio, CNF:HEC:air, of 54:28:18, that is, basically the same fraction of cellulose derivative as in this work. Their films exhibited yielding followed by strain hardening and an ultimate tensile strength of 200 MPa and a strain-at-break of 14%, comparable to the values shown in Table 1 for the current dialcohol cellulose−cellulose material. This striking similarity in mechanical behavior between the two different systems supports the idea that the 2221


Biomacromolecules

Article

50% RH and 26 mL·μm/(m2·24 h·kPa) at 80% RH. The lower sensitivity to moisture of the current all-cellulose composite may be due to the low charge density of the material and an accompanying lower swelling at high humidity. To explore the possibility of forming complex-shaped structures from the films, such as a double-curved structure, a custom-made mold was used (Figure 4a). A dry film was first

By breaking a bond and introducing two new hydrogen atoms (Scheme 1), that is, making a seemingly small alteration in the molecular structure of the cellulose molecules in the outer part of the CNFs, a combination of high strength and high ductility, and a better formability under moist conditions have been demonstrated. The complete mechanism behind this behavior has not yet been fully clarified and requires further research. It should however also be recognized that this technique of opening the cellulose C2−C3 bond permits many other types of reactions, which are likely to introduce other properties, one being cross-linking to fabricate CNF films with barrier properties that are more or less inert to changes in relative humidity.36 Altogether, significant property changes that strengthen the competitiveness of cellulosic materials compared with those of petroleum-based materials are at hand.

■

CONCLUSIONS Cellulose nanofibrils were successfully produced by homogenization of low-charged cellulose fibers that had been heterogeneously converted to dialcohol cellulose to form a nanoscale core−shell structure. By AFM, the fibrils were estimated to be about 4−10 nm wide and 0.5−2 μm long, and after vacuum filtration and drying under heat and pressure, high-density films could be fabricated. The films acted as good oxygen barriers with an oxygen permeability of about 5.5 mL·μm/(m2·24 h·kPa) at 23 °C and 80% RH. The presence of an adhered shell of dialcohol cellulose surrounding the crystalline nanofibril core led to strong and tough films with a tensile strength of about 175 MPa and a strain-to-failure of about 15% at 23 °C and 50% RH. The films showed an average yield strength of 117 MPa and yield strain of 1.4%. Most of the strain-to-failure was hence of a plastic nature, probably with interfibrillar slippage as the main mechanism, that is, supporting the mechanism suggested in earlier work23 that the high plastic deformation of sheets made of fibers subjected to the same modification is due to plastic deformation within the individual fibers. The core−shell modification also provided a moistureenhanced plastic deformation in the CNF films; at 90% RH, the films could be strained up to about 37% before failure. This suggests that films of core−shell structured CNFs can be used as an alternative to petroleum-based plastics in, for example, packaging requiring a more complex geometrical shape. To demonstrate this, a double-curved structure with a nominal strain of 24% over the curvature was formed.

Figure 4. (a) Forming mold and pressure converter used to form double-curved structures. (b,c) A film pressed into a double-curved structure with 24% nominal stain over the curvature. The formation was performed after conditioning at 23 °C and about 90% RH. The KTH logo is used with permission from KTH Royal Institute of Technology.

conditioned at 90% RH and then quickly sandwiched between two thin polyethylene films and placed in the mold. The sandwich assembly was then placed in a press and a doublecurved film was formed. As can be clearly seen in Figure 4b,c, it was possible to form a double-curved structure with a nominal strain of 24% over the curvature. The almost complete absence of wrinkles in the nonformed part of the film indicates that the film was clamped hard enough to prevent any significant draw, that is, the double-curved structure was formed through plastic deformation of the material. From the results presented in Figure 3 and Table 1, it is plausible to assume that even more complex structures could be formed, but this was experimentally difficult to achieve. There may be several reasons for this, but the most dominant is probably the fact that threedimensional forming requires a multiaxial straining of the film whereas the tensile testing performed was uniaxial (showing significant contraction in the perpendicular direction, see Figure S1 in the Supporting Information). However, considering the ductility and formability of paper board,37−39 a laminate of paper board and the core−shell modified CNF film is not likely to be limited by the formability of the film. Besides formability, Figure 4b shows that the films have a fairly good transparency, which is in agreement with its high density (1420 kg/m3) as well as earlier observations on films made from both pure CNFs and pure dialcohol cellulose.8,25 The observed cloudiness of the film is presumably due to a combination of a small fraction of the earlier mentioned fibril aggregates and the surface roughness seen in Figure 2a. The influence of surface roughness was clearly demonstrated by a distinct difference in optical transparency in air and when placed directly on a surface.

■

ASSOCIATED CONTENT

* Supporting Information S

Macro- and microscopic images of failed test pieces, as well as dynamic mechanical thermal analysis data of the films. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Per A. Larsson. Mailing address: KTH Royal Institute of Technology Department of Fibre and Polymer Technology Teknikringen 56, SE-100 44 Stockholm Sweden. Phone: +46-8790 81 09. E-mail: [email protected]. Author Contributions

The manuscript was written through contributions of all the authors. All the authors have given their approval to the final version of the manuscript. 2222


Biomacromolecules

Article

Notes

(29) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1988, 17, 513−886. (30) Painter, T. J. Carbohydr. Res. 1988, 179, 259−268. (31) Symons, M. C. R. J. Chem. Soc. 1955, 2794−2796. (32) Calvini, P.; Conio, G.; Lorenzoni, M.; Pedemonte, E. Cellulose 2004, 11, 99−107. (33) Sihtola, H. Makromol. Chem. 1960, 35, 250−265. (34) Callister, W. D. Fundamentals of Materials Science and Engineering: An Interactive e.text. Wiley: New York, 2001, Chapter 7. (35) Sugiyama, J.; Vuong, R.; Chanzy, H. Macromolecules 1991, 24, 4168−4175. (36) Larsson, P. A.; Kochumalayil, J. J.; Wågberg, L. In 15th Fundamental Research Symposium: Advanced in Pulp and Paper Research, Cambridge 2013; I’Anson, S., Ed. The Pulp and Paper Fundamental Research Society: Lancashire, U.K., 2013; pp 851−866. (37) Huang, H.; Nygårds, M. Nord. Pulp Pap. Res. J. 2012, 27, 211− 225. (38) Vishtal, A.; Retulainen, E. Bioresources 2012, 7, 4424−4450. (39) Ö stlund, M.; Borodulina, S.; Ö stlund, S. Packag. Technol. Sci. 2011, 24, 331−341.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Dr. Gustav Nyström, KTH Wallenberg Wood Science Center, is acknowledged for skillful assistance in preparing CNFs. Mr. Jonas Sundström, Innventia AB, is acknowledged for assistance in setting up the tensile testing in Innventia’s climate chamber. The financial support from VINNOVA, the Swedish Governmental Agency for Innovation Systems, through the BiMaC Innovation Excellence Centre, is also respectfully recognized.

■

REFERENCES

(1) Klemm, D.; Kramer, F.; Moritz, S.; Lindström, T.; Ankerfors, M.; Gray, D.; Dorris, A. Angew. Chem., Int. Ed. 2011, 50, 5438−5466. (2) Lavoine, N.; Desloges, I.; Dufresne, A.; Bras, J. Carbohydr. Polym. 2012, 90, 735−764. (3) Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J. Chem. Soc. Rev. 2011, 40, 3941−3994. (4) Sakurada, I.; Nukushina, Y.; Ito, T. J. Polym. Sci. 1962, 57, 651− 660. (5) Tashiro, K.; Kobayashi, M. Polymer 1991, 32, 1516−1526. (6) Fukuzumi, H.; Saito, T.; Iwata, T.; Kumamoto, Y.; Isogai, A. Biomacromolecules 2009, 10, 162−165. (7) Henriksson, M.; Berglund, L. A.; Isaksson, P.; Lindström, T.; Nishino, T. Biomacromolecules 2008, 9, 1579−1585. (8) Nogi, M.; Iwamoto, S.; Nakagaito, A. N.; Yano, H. Adv. Mater. 2009, 21, 1595−1598. (9) Syverud, K.; Stenius, P. Cellulose 2009, 16, 75−85. (10) Aulin, C.; Gällstedt, M.; Lindström, T. Cellulose 2010, 17, 559− 574. (11) Edlund, U.; Yu, Y.; Zhu Ryberg, Y.; Krause-Rehberg, R.; Albertsson, A.-C. Anal. Chem. 2012, 84, 3676−3681. (12) Fukuzumi, H.; Saito, T.; Iwamoto, S.; Kumamoto, Y.; Ohdaira, T.; Suzuki, R.; Isogai, A. Biomacromolecules 2011, 12, 4057−4062. (13) Gröndahl, M.; Eriksson, L.; Gatenholm, P. Biomacromolecules 2004, 5, 1528−1535. (14) Johansson, C.; Bras, J.; Mondragon, I.; Nechita, P.; Plackett, D.; Simon, P.; Svetec, D. G.; Virtanen, S.; Baschetti, M. G.; Breen, C.; Clegg, F.; Aucejo, S. BioResources 2012, 7, 2506−2552. (15) Yang, Q.; Fukuzumi, H.; Saito, T.; Isogai, A.; Zhang, L. Biomacromolecules 2011, 12, 2766−2771. (16) Zhu Ryberg, Y. Z.; Edlund, U.; Albertsson, A.-C. Biomacromolecules 2011, 12, 1355−1362. (17) Aulin, C.; Salazar-Alvarez, G.; Lindström, T. Nanoscale 2012, 4, 6622−6628. (18) Kochumalayil, J. J.; Bergenstråhle-Wohlert, M.; Utsel, S.; Wågberg, L.; Zhou, Q.; Berglund, L. A. Biomacromolecules 2013, 14, 84−91. (19) Liu, A.; Berglund, L. A. Carbohydr. Polym. 2012, 87, 53−60. (20) Liu, A.; Walther, A.; Ikkala, O.; Belova, L.; Berglund, L. A. Biomacromolecules 2011, 12, 633−641. (21) Zheng, G.; Cui, Y.; Karabulut, E.; Wågberg, L.; Zhu, H.; Hu, L. MRS Bull. 2013, 38, 320−325. (22) Sehaqui, H.; Zhou, Q.; Berglund, L. A. Soft Matter 2011, 7, 7342−7350. (23) Larsson, P. A.; Berglund, L. A.; Wågberg, L. Cellulose 2014, 21, 323−333. (24) Matsumura, H.; Sugiyama, J.; Glasser, W. G. J. Appl. Polym. Sci. 2000, 78, 2242−2253. (25) Kasai, W.; Morooka, T.; Ek, M. Cellulose 2014, 21, 769−776. (26) Morooka, T.; Norimoto, M. Sen’i Gakkaishi 1991, 47, 328−333. (27) Morooka, T.; Norimoto, M.; Yamada, T. J. Appl. Polym. Sci. 1989, 38, 849−858. (28) Nishino, T.; Matsuda, I.; Hirao, K. Macromolecules 2004, 37, 7683−7687. 2223


Ductile All-Cellulose Nanocomposite Films ... - ACS Publications

Recommend Documents