Dual Mechanism of Dry Strength Improvement of Cellulose Nanofibril

Oct 17, 2016 - *E-mail: [email protected]. ... cross-linked with wet strength cross-linker polyamide-epichlorohydrin resin (PAE) were investi...
2 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/IECR

Dual Mechanism of Dry Strength Improvement of Cellulose Nanofibril Films by Polyamide-epichlorohydrin Resin Cross-Linking Sudhir Sharma and Yulin Deng* School of Chemical & Biomolecular Engineering and RBI, Georgia Institute of Technology, 500 10th Street Northwest, Atlanta, Georgia 30332, United States ABSTRACT: The chemical and physical properties of cellulose nanofibril (CNF) films cross-linked with wet strength cross-linker polyamide-epichlorohydrin resin (PAE) were investigated. In addition to the expected wet strength improvement, a unique stress−strain behavior for dry cross-linked films was observed. This stress−strain behavior indicates that varying stress-bearing mechanisms arise from the formation of new bonds caused by cross-linking at different levels of cross-linker addition. Crosslinked films showed an approximate 150% increase in dry strength, when compared to non-cross-linked CNF films. The cross-linked films showed a transition of both surface and bulk properties from hydrophilic to hydrophobic. The water contact angle increased from 50° to 110°, while the water retention value decreased by approximately 25% for the cross-linked films. Moreover, water vapor permeability also showed a 2-fold decrease. Possible mechanisms for the changes in behavior are discussed in detail. performance of cross-linked CNF films. Some studies have shown that CNFs have a high adsorbing capacity for PAE because CNFs have a high content of anionic carboxyl sites.8,11−15 Even though PAE cross-linkers enhance wet strength, there is a trade-off with biodegradability and recyclability. If too much PAE cross-linker is used, the fibers are rendered nonbiodegradable and nonrecyclable.12 Thus, it is imperative to use as small an amount of PAE as possible. In this study, we propose the cross-linking of mechanically produced CNFs with very low concentrations of PAE cross-linker, typified by Kymene 557H wet-strength resin (1% by fiber weight), so as not to significantly affect the biodegradability of the CNF films. We propose taking advantage of a longer curing period to promote not only hetero-cross-linking between carboxyl sites and PAE azetidinium groups, but also the formation of PAE networks.8,11,14 Additionally, we expect the longer curing period to produce a degree of hornification in cellulose; we have previously demonstrated that this changes the CNF film surface and bulk from hydrophilic to hydrophobic.16 In this study we will mainly explore the mechanism of mechanical strength enhancement and water interaction that results from the crosslinking of CNF films.

1. INTRODUCTION Polymer films are used in a variety of packaging applications. However, in recent years, the significant environmental impact from waste petroleum-based packaging materials has become a great concern. Barriers made from microfibrillated cellulose (MFC) or cellulose nanofibrils (CNFs) have garnered significant attention because of their excellent biodegradability, renewability, and barrier and mechanical properties.1−3 Even though films made from CNFs possess these excellent properties, their mechanical and barrier properties begin to degrade in humid environments because of their high affinity to water.4−8 The hydrophilic nature of cellulose limits its applications in packaging applications. Therefore, it is imperative to develop solutions that can stop or limit the degradation of the mechanical and barrier properties of CNF membranes in humid environments.1−3,9,10 Using a cross-linker to increase the wet strength of paper products such as tissue paper, paper towels, and liquid packaging is very common. Polyamide-epichlorohydrin (PAE) resin is one effective papermaking cross-linking agent, and it is commercially available in a PAE water solution. The increase in wet strength offered by the PAE’s cross-linking of cellulose has already been studied extensively and is well-understood.8,11 The mechanisms of wet strength enhancement have been attributed to the formation of a covalent ester bond between carboxyl groups of cellulose and azetidinium groups of PAE. There is also a secondary mechanism of self-cross-linking PAE groups that form a water insoluble network during the curing process.11−14 Previously, we have demonstrated the excellent properties of PAE cross-linked CNF aerogels in wet environments.8 However, no studies have been made to date on the mechanical © 2016 American Chemical Society

Received: Revised: Accepted: Published: 11467

July 29, 2016 October 2, 2016 October 17, 2016 October 17, 2016 DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474

Article

Industrial & Engineering Chemistry Research

Figure 1. (A and B) As-obtained glass slide dried CNF slurry.

Figure 2. (A, B) Non-cross-linked CNF films; (C, D) PAE cross-linked CNF films.

After drying, the films with cross-linker were heated in an oven at 120 °C for 3 h to complete the cross-linking process. Subsequently all the films were conditioned at 23 °C and 50% RH for 24 h. Multiple films were made for each type of sample. 2.3. Characterization. The films were characterized by physical dimensions, water retention, contact angle, thermal properties, stress−strain analysis, Fourier transform infrared (FTIR) analysis, and scanning electron microscopy (SEM) imaging of the surface and cross section of the fracture surface. First, films were cut into discs with a diameter of 42 mm using a disc cutter and then weighed. Three discs were cut, and for each disc, five measurements of thickness were made with a Mitutoyo micrometer, so that the film’s thickness and grammage could be assured. Water retention was measured by soaking the film discs in water for 30 min and then measuring the difference in weight, to determine absorbed water. Water vapor permeability was measured by a modified ASTM cup method. A centrifuge tube with a diameter of 1.52 cm was three-quarters filled with water, and then the top was sealed with the CNF film to be tested. After being sealed, the tubes were weighed and then placed in a vacuum oven at a temperature of 37 °C. After 6 h the tubes were weighed again to measure the weight loss. The water vapor permeability is expressed as shown in eq 1:

2. EXPERIMENTAL SECTION 2.1. Materials. The CNFs were obtained from the University of Maine, as an approximately 3% solids slurry in water. We obtained the commercially available cross-linker Kymene from Ashland Inc. (Covington, KY) as a 12.5% solid suspension. 2.2. Fabrication of Films. The CNF slurry was first diluted to 1%, and then a weighed amount of the slurry was vigorously stirred for 30 min. Cross-linked films were prepared by adding 1 wt % Kymene, with respect to dry fiber weight, to the slurry while stirring. Subsequently, films were fabricated via an ultrafiltration process followed by hot press drying and conditioning. Millipore polyvinylidene difluoride (PVDF) membranes with a diameter of 142 mm and a pore size of 0.22 μm were used for the ultrafiltration process. A grammage of approximately 100 g/m2 was targeted in order to get thick, dense, and robust films. The slurries were dewatered until no more water could be removed. The dewatered mat was placed between smooth metal caul plates and four sheets of blotter paper and then compressed at a load of 415 psi, to remove any excess water that remained after dewatering. After being pressed, the films were dried in a flatbed dryer under approximately a 345−415 psi load, at 50 °C, for 24 h. 11468

DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474

Article

Industrial & Engineering Chemistry Research WVP =

Δm ·T ⎛ g·μm ⎞ ⎟ ⎜ Δt ·A ·ΔP ⎝ day·m 2·kPa ⎠

(1)

Here, Δm is the weight loss, T the thickness of the films, Δt the time of the experiment, A the cross-sectional area of the film, and ΔP the pressure difference between the inside of the tube and the outside. The pressure inside the tube is the saturation pressure at 37 °C while the oven outside is under vacuum, and the pressure is assumed to be zero. The contact angle was measured with a First Ten Angstrom goniometer, and FTA32 software was used to perform image analysis. The FTIR spectra for the films were collected using the direct film method. We used a Bruker Vector 80v for the FTIR analysis. For mechanical testing, we used a standard ASTM dog bone D-1708 and tested three samples of each type of film, in an Instron Bluehill II machine, to measure the mechanical strength of the films. The dog-bone samples are 38 mm × 15 mm in overall size and 5 mm × 22 mm in the gauge or test-zone area. Mechanical testing was carried out for dry and wet samples. In the case of the wet strength test, the samples were soaked in deionized water for 30 min prior to testing, and excess water was removed by blotters following TAPPI test methods. The films’ morphology was analyzed by SEM (LEO 1530 SEM, Carl Zeiss) at 3−15 kV. The samples were placed on SEM stages and gold sputtered (Quorum 150ES) for 60 s. Thermal analysis was performed in a PerkinElmer TGA from 25 to 500 °C at a rate of 10 °C/min.

Figure 3. FTIR analysis.

primary difference between the two is the presence of a new peak in the 1550 cm−1 region. This peak can be attributed to the amide II (−NH) groups of PAE cross-linker, and it confirms the presence of PAE in the cross-linked films. The presence of a shoulder in the cross-linked film spectrum, at 1728 cm−1, confirms cross-linking between PAE and cellulose which, in turn, can be attributed to the stretching of the CO ester bonds that were formed between the PAE azetidinium groups and the cellulose carboxyl groups. Additionally, a 1260 cm−1 area of the cross-linked film spectrum also shows a reduction in the valley; this is thought to be due to the formation of additional C−O bonds between the PAE and the cellulose.8,15 After the cross-linking was confirmed by FTIR analysis, our primary interest concerned the interaction with water. The severe degradation of mechanical and barrier properties in humid conditions has been extensively studied in the literature.19 Because of PAE cross-linking, it is expected that the degradation of CNF film properties, as a result of water absorption, should be significantly reduced. Because crosslinking is carried out by heating at 120 °C for 3 h, there should also be some hornification of the films which would limit their interaction with water.4,11−13,16 We measured the contact angle and the water retention values to observe the surface and bulk interaction. While it is impossible to obtain highly accurate measurements of the contact angle because of the partial absorption of water into the pores of CNF films and interaction with free surface −OH groups, the best possible measurements showed a significant difference between the non-cross-linked CNF films and the cross-linked films. Figure 4 shows typical observations of water droplets on the surface of the films 5 min after the droplets were placed. Table 1 shows the measured contact angle, water retention value, and water vapor permeability for the membranes. Non-cross-linked CNF films displayed the typical hydrophilic characteristics of the films, whereas cross-linked CNF films showed a transition to hydrophobic in both surface and bulk properties. The contact angle transitioned from 50° for noncross-linked CNF films to 110 °C for cross-linked films, indicating a hydrophobic surface, while the water retention decreased by approximately 25% after cross-linking. The crosslinking of PAE/CNF films was made by heating at 120 °C for 3

3. RESULTS AND DISCUSSION To prepare the films, we used commercially available CNFs, obtained as a 3% slurry in water. The SEM micrographs of the obtained material are shown in Figure 1. These were taken by drying a droplet of the slurry on a glass slide, followed by the standard SEM procedure described previously. It was important to produce films with physical uniformity in order to compare the non-cross-linked and cross-linked films in a consistent manner. The slurry was diluted to 1% with and without PAE cross-linker, and films were fabricated via an ultrafiltration, compression, and hot press method. Highly uniform films were obtained, with each having a thickness in the range of 290 ± 15 μm and grammage of 95 ± 4 g/m2. After being dried, the films containing PAE were heated at 120 °C for 3 h to complete the PAE cross-linking, and all films were subsequently conditioned. SEM images of the surface of the films are shown in Figure 2. While both films are of a high grammage that is close to the intended target and visually uniform, the surface of the noncross-linked CNF films is very rough with a large number of free fibrils on the surface. Conversely, the cross-linked films show high smoothing and densification of the surface. This is to be expected because cross-linking forms extra bonds within the CNF film. It is noted that SEM images in Figure 2 are not at the same voltage; however, this was necessary to obtain clear images at comparable magnification. We note this may hinder ability for consistent comparison because of different depth of penetration. On the basis of our previous work with thermal exposure of CNF films and a qualitative observation of the SEM images, we can contend that a significant loss of porosity and densification was observed in the film structure because of the thermal exposure.16−18 FTIR analysis is shown in Figure 3 and was performed to confirm cross-linking between CNF films and PAE. The 11469

DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474

Article

Industrial & Engineering Chemistry Research

Figure 4. Non-cross-linked CNF contact angle (left) and PAE cross-linked CNF contact angle (Right).

Table 1. Water Interaction Properties sample

non-cross-linked CNF

cross-linked CNF

contact angle (deg) water retention value (g/m2) water retention value (g water/g dry film) water vapor permeability (ml·μm/(day·m2·kPa))

50.2 ± 0.6 77.4 ± 3.9 0.81 ± 0.04 202 ± 10.1

110.5 ± 0.7 52.1 ± 2.6 0.59 ± 0.03 103.2 ± 5.1

Figure 5. TGA and DTG analysis.

Figure 6. Stress−strain analysis (left, dry tests; right, wet tests).

closer together, thus causing a loss of porosity in the films and a reduction in the pore space for water to be absorbed.18,20,22 The concurrent loss of hydroxyls and porosity supports a significant increase in water contact angle. Furthermore, PAE cross-linking could bring the fibrils closer together because it forms covalent bonds with the carboxyl groups of the cellulose,

h. Some degree of cellulose hornification was induced at this temperature, as previously reported. Typically hornification that is produced by thermal exposure causes the cellulose hydroxyl groups to hydrogen bond irreversibly with each other, reducing sites for interaction with water.18,20,21 Additionally, because of this interfibril hydroxyl hydrogen bonding, the fibrils are pulled 11470

DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474

Article

Industrial & Engineering Chemistry Research

Figure 7. Fracture site SEM images: (A, B) non-cross-linked CNF (dry); (C, D) non-cross-linked CNF (wet).

Figure 8. Fracture site SEM images: (A, B) PAE cross-linked CNF (dry); (C, D) PAE cross-linked CNF (wet).

cross-linked films show significantly less weight loss overall, indicating that the amount of absorbed water is much lower in this case. The cross-linked films, as observed from the water retention value, were expected to show less water retention because of interhydroxyl hydrogen bonding, loss of porosity, and cross-linking.16,18,20,21 Figure 6 shows typical tensile stress−strain/extension curves that were obtained for dry and wet, non-cross-linked and crosslinked CNF films. The dry tests refer to the samples that were tested after the conditioning process, whereas the wet samples refer to the samples that were soaked in water for 30 min prior to mechanical testing. For SEM analysis the wet samples had to be dried beforehand at ambient conditions (23 °C and 50% RH) for 24 h. Figures 7A,B and 8A,B show the fracture surface

thus creating bonds between the fibrils.11,14,15 Additionally, the 50% decrease in water vapor permeability is significant. When combined with the reduction in water retention and the increase in contact angle, this could possibly result in a degradation of gas permeability in humid environments. The thermal analysis curves for non-cross-linked CNF film and the films, shown in Figure 5, demonstrate clear differences between the thermal stability of the two materials. The crosslinked films showed a delayed onset (298 °C for cross-linked as compared to 279 °C), a lower maximum rate of degradation, and a reduction in weight loss before the maximum degradation temperature was reached. After conditioning at 50% RH for 24 h and before analysis, the amounts of water absorbed by the different films were determined with thermal analysis. The 11471

DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474

Article

Industrial & Engineering Chemistry Research

native cellulose sheets using FTIR analysis and have showed that ester bond formation, between the azetidinium groups of PAE and the carboxyl groups of the cellulose, is the main factor for wet strength development.11,14,30 Ahola et al.13 studied the adsorption of PAE on cellulose nanofibrils and then used that as a wet strength additive for paper. They showed that a significant increase occurs in wet strength when both PAE and cellulose nanofibrils were absorbed onto native cellulose fibers. The SEM images of the wet tested film fracture surface with cross-linker (Figure 8C,D) show a structure that is very different from that of the non-cross-linked wet tested CNF films (Figure 8A,B). The wet tested cross-linked films do not show a tear behavior because of the protection mechanism, whereas the non-cross-linked CNF films do. The difference between the stress−strain behavior of the wet non-cross-linked and cross-linked CNF films clearly shows that, in the case of cross-linked films, the stress-bearing mechanism is different and that most of the stress is borne by the new bonds that are created because of cross-linking, because no tearing behavior is observed.11−15 On the other hand, the dry cross-linked samples showed a complex stress−strain behavior never before observed in CNF films, with multiple distinct regions, which implies different load bearing mechanisms at different stress levels. The curve can be viewed in two distinct phases, one before the plastic deformation region, with constant stress observed at approximately 67.5 MPa, and one after. The region before the plastic deformation is initially very similar in behavior to non-cross-linked CNF films, albeit with a much higher slope, beginning in a pseudo elastic region followed by a knee and then followed by a strain hardening region. After the plastic deformation region, there is another knee that leads into a secondary strain hardening region, after which failure occurs. The SEM micrographs of the fracture surface (Figure 7A,B) show not only broken fibrils and bundles of fibrils but also highly ordered sheetlike structures. The broken fibrils indicate that the concluding mechanism of film failure is the breakages of fibrils and of interfibril bonds, just as observed in non-cross-linked CNF films. Therefore, we can reasonably assume that the stress−strain behavior after the plastic deformation is due to load bearing of the cellulose fibrils and interhydroxyl hydrogen bonds, as in non-cross-linked CNF films. In the plastic deformation region there is no demonstrated increase in film stress; we postulate that this is because of the breaking of the PAE networks that are formed during the cross-linking process and that none of the CNF bonds are broken in this region. The highly ordered sheetlike fibril structures observed in the fracture surface SEM images (Figure 7A,B) indicate that the reorientation of the fibers is much more prolonged in the strain hardening region than in non-cross-linked CNF films. It is possible that fibril reorientation also occurs during the plastic deformation and not just the strain hardening regions. The stress−strain behavior that occurs prior to plastic deformation indicates that there is a different load-bearing mechanism in that region than for non-cross-linked CNF films. This can only be attributed to the additional bonds that were formed during cross-linking, because the non-cross-linked CNF films did not show this behavior. It can be assumed that the initial pseudo elastic behavior is because of PAE−carboxyl bonds, followed by the knee which is representative of interfibril debonding. This is followed by some strain hardening for a very short strain range because PAE networks take over

SEM micrographs of the dry and wet, non-cross-linked and cross-linked CNF films, respectively, and after mechanical testing and drying in the case of the wet test samples. The noncross-linked dry CNF films displayed the typical behavior characteristics of CNF films, whereas the wet CNF films displayed a “tearing” behavior. The dry cross-linked CNF films showed a very distinct, and to our knowledge never before observed, CNF films stress−strain behavior: multiple knees, plastic deformation, and multiple strain hardening regions. Conversely, wet cross-linked CNF films showed a singular elastic region before break, atypical from the expected tear behavior for non-cross-linked CNF films. The dry non-cross-linked CNF films showed a linear behavior up to approximately 25 MPa of tensile stress. After this pseudoelastic region, there is a knee in the curve which is thought to occur because of some degree of fibril−fibril debonding. The knee is followed by a linear strain hardening region to approximately 40 MPa, where failure occurs. In the strain hardening region, there is a significant reinforcement effect due to the straightening and reorientation of the nanofibrils and interfibril slippage. This reinforcement behavior results from the alignment of the nanofibrils in the direction of the applied strain. Failure of the film follows the failure of the fibrils and breakage between their existing bonds.23−29 This is also observed in the SEM micrographs of the fracture surface of the non-cross-linked CNF films (shown in Figure 7A,B), where broken individual fibrils and bundles of aggregated broken fibrils are visible. Some fibrils, oriented toward the direction of the applied strain, are also visible. This behavior is typical of CNF films and has already been well-documented in the literature.24−29 The wet non-cross-linked CNF films do not show a linear behavior in their stress−strain behavior. This is expected because CNF films absorb a significant amount of water and swell. Absorption of water causes the interfibril bonding to be disrupted by the presence of water and causes the films to lose strength. Therefore, instead of typical CNF film stress−strain behavior, the films display a tearing behavior, in which the water-swollen fibrils are simply pulling apart from each other and partially weakened bonds are easily broken. There is an elastic region followed by a loss of stress with increasing strain; this is where the fibrils start coming apart and some necking is observed. Subsequently, as the strain increased, the film tears apart completely. This is also clearly observed in the SEM images of the fracture surface of the wet films (Figure 7C,D). Fibrils that have been pulled apart, and agglomerated bundles of fibrils can be seen clearly, instead of the broken fibrils that are observed in the case of dry CNFs. Cross-linked films did not display tearing behavior in the mechanical analysis after being soaked; like the non-crosslinked CNF films they are protected by the PAE cross-linking and show failure at a stress level just slightly lower than that of dry non-cross-linked CNF films. Obokata and Isogai and various research groups have studied this mechanism in detail.11 The mechanism of interaction between PAE and cellulose has been described as a dual mechanism, that is, it consists of a reinforcing and protecting mechanism. The reinforcement mechanism is a result of the reaction between the carboxyl functional groups of the cellulose and the azetidinium groups of the PAE. Conversely, the protection mechanism occurs because of the inhibition of interfiber detachment in water that is caused by the formation of water−insoluble PAE networks. Other authors have characterized PAE cross-linked carboxymethylated 11472

DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474

Article

Industrial & Engineering Chemistry Research

(5) Gebald, C.; Wurzbacher, J. A.; Tingaut, P.; Steinfeld, A. Stability of Amine-Functionalized Cellulose during Temperature-VacuumSwing Cycling for CO2 Capture from Air. Environ. Sci. Technol. 2013, 47, 10063−10070. (6) Lee, J.; Deng, Y. The morphology and mechanical properties of layer structured cellulose microfibril foams from ice-templating methods. Soft Matter 2011, 7, 6034−6040. (7) Yang, X.; Cranston, E. D. Chemically Cross-Linked Cellulose Nanocrystal Aerogels with Shape Recovery and Superabsorbent Properties. Chem. Mater. 2014, 26, 6016−6025. (8) Zhang, W.; Zhang, Y.; Lu, C.; Deng, Y. Aerogels from crosslinked cellulose nano/micro-fibrils and their fast shape recovery property in water. J. Mater. Chem. 2012, 22, 11642−11650. (9) Roy, D.; Semsarilar, M.; Guthrie, J. T.; Perrier, S. Cellulose modification by polymer grafting: a review. Chem. Soc. Rev. 2009, 38, 2046−2064. (10) Hussain, F.; Hojjati, M.; Okamoto, M.; Gorga, R. E. Review article: Polymer-matrix Nanocomposites, Processing, Manufacturing, and Application: An Overview. J. Compos. Mater. 2006, 40, 1511− 1575. (11) Obokata, T.; Isogai, A. The mechanism of wet-strength development of cellulose sheets prepared with polyamideamineepichlorohydrin (PAE) resin. Colloids Surf., A 2007, 302, 525−531. (12) Su, J.; Mosse, W. K. J.; Sharman, S.; Batchelor, W.; Garnier, G. Paper strength development and recyclability with polyamideamineepichlorohydrin (PAE). BioResources 2012, 7, 913−924. (13) Ahola, S.; Ö sterberg, M.; Laine, J. Cellulose nanofibrils adsorption with poly(amideamine) epichlorohydrin studied by QCMD and application as a paper strength additive. Cellulose 2008, 15, 303−314. (14) Obokata, T.; Yanagisawa, M.; Isogai, A. Characterization of polyamideamine-epichlorohydrin (PAE) resin: Roles of azetidinium groups and molecular mass of PAE in wet strength development of paper prepared with PAE. J. Appl. Polym. Sci. 2005, 97, 2249−2255. (15) Geng, Y.; Li, K.; Simonsen, J. Further investigation of polyaminoamide-epichlorohydrin/stearic anhydride compatibilizer system for wood-polyethylene composites. J. Appl. Polym. Sci. 2006, 99, 712−718. (16) Sharma, S.; Zhang, X.; Nair, S. S.; Ragauskas, A.; Zhu, J.; Deng, Y. Thermally enhanced high performance cellulose nano fibril barrier membranes. RSC Adv. 2014, 4, 45136−45142. (17) Chen, Y.; Wang, Y.; Wan, J.; Ma, Y. Crystal and pore structure of wheat straw cellulose fiber during recycling. Cellulose 2010, 17, 329−338. (18) Brancato, A.; Walsh, F. L.; Sabo, R.; Banerjee, S. Effect of Recycling on the Properties of Paper Surfaces. Ind. Eng. Chem. Res. 2007, 46, 9103−9106. (19) Aulin, C.; Gällstedt, M.; Lindström, T. Oxygen and oil barrier properties of microfibrillated cellulose films and coatings. Cellulose 2010, 17, 559−574. (20) Vieira, M. G. A.; Rocha, S. C. S. Drying conditions influence on physical properties of recycled paper. Chem. Eng. Process. 2007, 46, 955−963. (21) Luo, X.; Zhu, J. Y. Effects of drying-induced fiber hornification on enzymatic saccharification of lignocelluloses. Enzyme Microb. Technol. 2011, 48, 92−99. (22) Park, S.; Venditti, R. A.; Jameel, H.; Pawlak, J. J. Changes in pore size distribution during the drying of cellulose fibers as measured by differential scanning calorimetry. Carbohydr. Polym. 2006, 66, 97−103. (23) Henriksson, M.; Berglund, L. A.; Isaksson, P.; Lindström, T.; Nishino, T. Cellulose Nanopaper Structures of High Toughness. Biomacromolecules 2008, 9, 1579−1585. (24) Syverud, K.; Stenius, P. Strength and barrier properties of MFC films. Cellulose 2009, 16, 75−85. (25) Singh, S.; Mohanty, A. K. Wood fiber reinforced bacterial Bioplastic composites: Fabrication and performance evaluation. Compos. Sci. Technol. 2007, 67, 1753−1763.

the load bearing right after debonding. The plastic deformation seems to occur because of the breakage of bonds in the PAE networks; in this region there is virtually no increase observed in the stress of the films. Additionally, the observed highly ordered sheets seem to occur because of the “sliding” of ordered fibril networks over each other without load bearing. Moreover, the strain hardening region that was observed for the crosslinked films, after plastic deformation, is larger in range when compared to the non-cross-linked CNF films, providing an increased opportunity for ordering in the fibril networks. Eventually, breakage seems to occur as a result of the typical breakage behavior of fibril bonds.

4. CONCLUSIONS PAE resin-based, wet strength enhancing agents are commonly used in the paper industry. Previously, significant study has elucidated the mechanism of PAE cross-linking with cellulose by heat-curing the resin. Other studies have also shown a significantly better retention of PAE resins in CNF materials when compared to native cellulose fibers. We observed that the CNF film surface transitions from hydrophilic to hydrophobic after the cross-linking process and displayed a contact angle increase from 50° to 110°. Additionally, the water retention value also displayed roughly a 25% decrease, which implies an increase in the bulk hydrophobicity of the material. Water vapor permeability decreased 2-fold, from 202 to 103 ml·um/ (day·m2·kPa). As was expected, the wet strength of the crosslinked material showed a significant improvement. The noncross-linked CNF film showed the simple tear behavior that was expected from wet and swollen CNFs, whereas the cross-linked films showed a single elastic region, ending in an eventual failure at a stress level similar to that of dry CNF film. Most importantly, we demonstrated never before observed complex stress−strain behavior and increases in the strength of dry cross-linked CNF films. These were made up of multiple different phase behaviors, indicating stress bearing via different mechanisms in the cross-linked material at different stress levels.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +01-404-894-5759. Fax: 01-404-894-4778. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS S.S. acknowledges the PSE fellowship provided by Renewable Bioproducts Institute at Georgia Tech. REFERENCES

(1) Siró, I.; Plackett, D. Microfibrillated cellulose and new nanocomposite materials: a review. Cellulose 2010, 17, 459−494. (2) Hansen, N. M. L.; Plackett, D. Sustainable Films and Coatings from Hemicelluloses: A Review. Biomacromolecules 2008, 9, 1493− 1505. (3) Lange, J.; Wyser, Y. Recent innovations in barrier technologies for plastic packaginga review. Packag. Technol. Sci. 2003, 16, 149− 158. (4) Cai, H.; Sharma, S.; Liu, W.; Mu, W.; Liu, W.; Zhang, X.; Deng, Y. Aerogel Microspheres from Natural Cellulose Nanofibrils and Their Application as Cell Culture Scaffold. Biomacromolecules 2014, 15, 2540−2547. 11473

DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474

Article

Industrial & Engineering Chemistry Research (26) Sehaqui, H.; Zhou, Q.; Ikkala, O.; Berglund, L. A. Strong and Tough Cellulose Nanopaper with High Specific Surface Area and Porosity. Biomacromolecules 2011, 12, 3638−3644. (27) Sehaqui, H.; Morimune, S.; Nishino, T.; Berglund, L. A. Stretchable and Strong Cellulose Nanopaper Structures Based on Polymer-Coated Nanofiber Networks: An Alternative to Nonwoven Porous Membranes from Electrospinning. Biomacromolecules 2012, 13, 3661−3667. (28) Sehaqui, H.; Liu, A.; Zhou, Q.; Berglund, L. A. Fast Preparation Procedure for Large, Flat Cellulose and Cellulose/Inorganic Nanopaper Structures. Biomacromolecules 2010, 11, 2195−2198. (29) Sehaqui, H.; Allais, M.; Zhou, Q.; Berglund, L. A. Wood cellulose biocomposites with fibrous structures at micro- and nanoscale. Compos. Sci. Technol. 2011, 71, 382−387. (30) Haggkvist, M.; Solberg, D.; Wagberg, L.; Odberg, L. The influence of two wet strength agents on pore size and swelling of pulp fibres and on tensile strength properties. Nord. Pulp Pap. Res. J. 1998, 13, 292−298.

11474

DOI: 10.1021/acs.iecr.6b02910 Ind. Eng. Chem. Res. 2016, 55, 11467−11474