Thermoresponsive cryogels reinforced with cellulose nanocrystals

Abstract

Herein, we report the first study of thermoresponsive cryogels with cellulose nanocrystals (CNCs) incorporated into the structure. Free radical polymerization was utilized to synthesize cryogels of poly(N-isopropylacrylamide) (PNIPAAm), resulting in thermoresponsive gels after the cryo-polymerization. Two types of CNCs were investigated: one which had reactive vinyl groups on the surface, enabling covalent incorporation and crosslinking with the cryogel network; and one which had no reactive groups on the surface, rendering it physically embedded in the network. The degree of crosslinking of the cryogels was controlled by varying the addition of N,N′-methylenebisacrylamide (MBAm). The cryogels were analyzed by FE-SEM and were all found to be macroporous. The morphology of the gels was largely dependent on the reaction conditions and the presence of CNC. The swelling properties of the freeze-dried gels were investigated and all gels exhibited a thermoresponsive behavior. Our study showed that the incorporation of CNCs is an effective method to alter both the morphologies and the mechanical properties of a cryogel, although the final properties of the cryogels depend on several different parameters. Due to the complexity of the system, a clear trend regarding the CNC incorporation is difficult to conclude, but compression testing showed that a cryogel having 1 wt% of crosslinkable CNC was far superior to the other gels in terms of mechanical properties, exhibiting that the presence of crosslinkable groups on the surface of CNCs could have a large influence over the final properties.

---

Introduction

Hydrogels are crosslinked, three-dimensional polymer networks, swollen by a large amount of water. They can be divided into different categories based on their origin, type of crosslinking functionality (covalent or physical) or based on their response to the surrounding medium, as conventional- or responsive hydrogels.^1,2 Hydrogels have many current and potential application areas such as; contact lenses, drug delivery systems, tissue engineering, and sensors.^3–7 Traditionally, hydrogels derived from synthetic polymers exhibit poor mechanical properties; being soft, brittle and/or poor at withstanding deformations,^8,9 imposing significant limitations on their applicability. More recently, hydrogels with improved mechanical properties have been obtained, and shown to be much tougher.^9–15 Several different methods for the synthesis of tough hydrogels were developed during the first decade of the 21^st century, such as the double network gel and the nanocomposite gel.^16–18 In nanocomposite hydrogels, inorganic/organic nanoparticles are incorporated into a polymer network during synthesis and acts as physical or chemical crosslinkers in the gel.^9,18

Cellulose nanocrystals (CNCs), with diameters up to 15–20 nm and typical aspect ratios of 50 and above,¹⁹ were first incorporated into hydrogels as recent as 2010.¹⁰ Prior to this, CNCs had been extensively investigated as reinforcing fillers in polymer matrixes due to their high modulus of elasticity and high strength.^19–21 Following the initial study on the application of CNC in hydrogels,¹⁰ numerous studies have been performed, incorporating CNC into a variety of matrices such as polyacrylamide, poly(acrylic acid), poly(N,N-dimethylacrylamide) and poly(N-isopropylacrylamide) (PNIPAAm).^{10–15,22–24} The incorporation of CNC into hydrogels has shown to improve the mechanical strength and flexibility.^{10–13,22–24}

A cryogels is a type of hydrogel, synthesized below the freezing point of the solvent used in the reaction system. Upon freezing, the major part of the solvent forms crystals leaving only a liquid microphase in which the reactants are concentrated.^25–28 The polymerization occurs in the liquid microphase, creating a gel with an interconnected, macroporous, structure after thawing. Depending on the reaction conditions and choice of monomer and other reactants, the size of the pores (within a cryogel) can be varied in a large range, from 0.1–200 μm.^25–28 The most commonly used solvent for the synthesis of cryogels is water, but DMSO and water mixtures with other solvents have also been utilized.^25–29 A variety of monomers has been employed for the synthesis of cryogels by free radical polymerization (FRP), the most commonly used being acrylamide.^25–28

Cryogels have attracted interest due to their potential applications within the biomedical field,^25–28 but they are also of interest within other fields e.g. filters and sorbents.^25–28 By producing cryogels based on thermoresponsive polymers, such as PNIPAAm, the macroporous structure is combined with the responsive properties of the polymer. PNIPAAm is a well-known thermoresponsive polymer, with a lower critical solution temperature (LCST) of around 32–33 °C.^2,30 Typically, hydrogels of PNIPAAm are swollen and hydrated below the LCST, while they shrink above the LCST where they become hydrophobic and expel water.^2,29–32 The larger pores of a PNIPAAm cryogel, enables a much faster swelling and de-swelling behavior in comparison to the corresponding conventional hydrogels.^2,29,30,32 It has also been shown that the deformation during the de-swelling of a cryogel can release adsorbed bioparticles (such as cells and viruses), creating materials possible to utilize as reusable separation gels for bioparticles.^30,32 In a few studies cellulose particles in the form of microcrystalline cellulose and bacterial cellulose, have been incorporated into cryogels, forming gels with improved mechanical properties.^33–36 The incorporation of CNCs into cryogels are only found in one previous study in the literature.³⁵

In this study, CNC is added to the reaction mixture during the cryogel production, with the aim of creating gels with altered mechanical properties in comparison to neat PNIPAAm cryogels. CNC was chosen as the reinforcing element due to its well-known ability to drastically improve the mechanical properties of both hydrogels and polymer matrices.^{10–13,22–24} Two different types of CNCs were investigated: unmodified CNC and CNC modified with a polymerizable group, an acrylate. NIPAAm was chosen as the monomer for the cryo-polymerization (i.e. polymerization below the freezing point of water) due to its thermoresponsive properties and its previous demonstrated use in cryogels.^2,29,30,32 The incorporation of cellulose into a thermoresponsive cryogel is only found in one previous study by Syverud et al., where they created cryogels based on cellulose nanofibrils (CNF) where CNF formed the main part of the network and PNIPAAm was utilized as a crosslinker.³⁷ To the best knowledge of the authors, the present study is the first of its kind, where CNC has been incorporated in a thermoresponsive cryogel, where the matrix is polymerized in the presence of the CNCs. It is also the first study comparing the incorporation of unmodified CNCs and CNCs modified with polymerizable groups on the surface.

Experimental

Materials

Acrylate-functional cellulose nanocrystals (abbreviated as CNC-AA, 0.62 wt% in water suspension) were synthesized according to Carlmark and co-workers³⁴ and HCl-cellulose nanocrystals (abbreviated as CNC-HCl, 0.38 wt% in water suspension) were synthesized as reported elsewhere.^38,39 N-Isopropylacrylamide (NIPAAm, 97%), N,N,N′,N′-tetramethylethylenediamine (TEMED, 99%), and ammonium persulfate (APS, 98%) were purchased from Sigma-Aldrich. NIPAAm was purified by recrystallization in n-heptane twice before use. N-Heptane (99%) was purchased from Merck KGaA. N,N′-Methylenebis(acrylamide) (MBAm, 98%) was purchased from Fluka. All materials were used as received unless stated otherwise.

Instrumentation

Field emission scanning electron microscopy (FE-SEM) images were recorded on a Hitachi S-4800 FE-SEM. The samples were mounted on a substrate with carbon tape and coated with 5 nm of platinum/palladium with a sputter coater (Cressington 208HR).

Fourier transform infrared spectroscopy (FT-IR) was performed using a Perkin-Elmer Spectrum 2000 FT-IR equipped with a MKII Golden Gate, single reflection ATR System from Specac Ltd., (London, U.K.). The ATR-crystal used was a MKII heated Diamond 45° ATR Top Plate. For each spectrum 16 scans were recorded.

Thermogravimetric analysis (TGA) was performed on a TA Instruments Hi-Res TGA 2950 analyzer under a N₂ flow of 30 mL min⁻¹, at a heating rate of 10 °C min⁻¹. The samples were heated from 30 to 700 °C.

The thermo-responsive behavior of the gels was analyzed by storing the gels at a temperature of 50 °C (above the LCST) in deionized water until a stable mass was reached, determined by removing the samples from the water and measuring the mass of the samples at 1 h intervals. The gels were then placed in deionized water at room temperature (below the LCST) and weight gain was recorded at pre-determined intervals by weighing. The surfaces of the gels were carefully wiped free from water before each mass determination.

The swelling of gels from the dry state was performed on freeze-dried gels. The gels were placed in deionized water at room temperature (below the LCST) and the weight gain was measured at pre-determined time intervals by weighing. The surfaces of the gels were carefully wiped free from water before each mass determination. The dimensions of the water swollen gels were ca.: 10 mm in diameter and 8 mm in height.

Compression testing was performed on an Instron-5944 equipped with a 50 N load cell. Compressions were performed at a compression rate of 20% min⁻¹, until break or a maximum compression strain of 90%. The tests were performed on water swollen cylindrical samples (ca. 10 mm in diameter and 8 mm in height) synthesized in test tubes. The compression properties of each sample are based on the average value of at least five measurements.

Synthesis of cryogels – typical procedure

The typical procedure for the synthesis of cryogels was as follows: NIPAAm, MBAm (NIPAAm + MBAm = 0.5 g, molar ratios 15 : 1 or 25 : 1), APS (5.9 × 10⁻² mmol, 13.2 mg), and CNC in water (CNC-HCl or CNC-AA 0, 1, 2 or 5 wt%, dry content of CNC out of the total dry content of the gels) were added to a 25 mL round bottom flask, equipped with a magnetic stirrer. Deionized H₂O was added to adjust the final volume in all samples to 10 mL. The flask was placed in a water ice bath and purged with argon for 10 min. TEMED (5.9 × 10⁻² mmol, 6.8 mg) was added and the reaction mixture was vigorously stirred for 20 s. The reaction mixture was transferred into 4 glass test tubes with an approximate diameter of 10 mm. The test tubes were quickly moved into a standard laboratory freezer (−20 °C), and left to react overnight. The cryogels were thawed at room temperature before removal of the gels from the test tubes. The cryogels were purified by extraction in deionized water for 4 days, replacing the water twice a day.

Results and discussion

FRP was employed to synthesize cryogels of PNIPAAm. CNCs were successfully incorporated into the gels by the addition of CNC to the reaction mixture, thereby modifying the properties of the gels. The incorporation of CNC when producing gels in cryo-state is highly interesting, as it enables the possibility to tune the mechanical properties, as described above.

Herein, two different kinds of CNCs at three different concentrations (1, 2, and 5 wt%) and two different molar ratios of NIPAAm to MBAm (15 : 1 or 25 : 1, Table 1), Scheme 1, were used to synthesize CNC-containing cryogels. The CNCs were either unmodified with hydroxyl groups at the surface (CNC-HCl), or modified with acrylic acid creating acrylic, polymerizable, groups at the surface of the CNCs (CNC-AA). Previous work has shown that acrylic groups on the surface of nanocellulose are polymerizable.³⁸ Hence, the CNCs with acrylate groups have the possibility to be covalently incorporated into cryogel and also act as crosslinker whereas the unmodified CNC is only physically mixed. Reference cryogels were synthesized containing the same molar ratio of NIPAAm to MBAm as the CNC containing cryogels, but with no CNC added.

Table 1 Sample name and composition of synthesized cryogels

Sample name	CNC type and dry content in cryogels [wt%]	Ratio of MBAm : NIPAAm
PNIPAAm15CNC-HCl1%	CNC-HCl 1%	1 : 15
PNIPAAm15CNC-HCl2%	CNC-HCl 2%	1 : 15
PNIPAAm15CNC-HCl5%	CNC-HCl 5%	1 : 15
PNIPAAm25CNC-HCl1%	CNC-HCl 1%	1 : 25
PNIPAAm25CNC-HCl2%	CNC-HCl 2%	1 : 25
PNIPAAm25CNC-HCl5%	CNC-HCl 5%	1 : 25
PNIPAAm15CNC-AA1%	CNC-AA 1%	1 : 15
PNIPAAm15CNC-AA2%	CNC-AA 2%	1 : 15
PNIPAAm15CNC-AA5%	CNC-AA 5%	1 : 15
PNIPAAm25CNC-AA1%	CNC-AA 1%	1 : 25
PNIPAAm25CNC-AA2%	CNC-AA 2%	1 : 25
PNIPAAm25CNC-AA5%	CNC-AA 5%	1 : 25
PNIPAAm15	No CNC added	1 : 15
PNIPAAm25	No CNC added	1 : 25

---

Scheme 1 Schematic illustration of the structure of the synthesized cryogels, (a) no CNC (b) CNC-HCl and (c) CNC-AA. Blue spheres represent MBAm crosslinking points and red spheres CNC crosslinking points.

The successful incorporation of both CNC-HCl and CNC-AA into the cryogels was verified by FT-IR (Fig. 1 and ESI Fig. S1 and S2†). In the finger print region of cellulose (∼3400–3200 cm⁻¹ originating from the OH-stretching),⁴⁰ no difference between the different samples modified with cellulose and the unmodified sample could be seen due to the overlap with the secondary amide N–H stretching occurring around 3300 cm⁻¹ from PNIPAAm. However, in the spectra of the CNC modified cryogels, a band appears in the region of 1160–1130 cm⁻¹ which is ascribed to the –C–O–C– stretch of the β-1,4-glycosidic linkage from the cellulose backbone,^41,42 which is not visible in the spectrum of the neat PNIPAAm gel. There is a difference in the intensity of the –C–O–C– band depending on the amount of CNC added, with a lower intensity of the band when less CNC was added. This shows that it is possible to vary the final concentration of CNC in the gels by varying the amount of CNC added to the reaction mixture. Seemingly, irrespectively of which type of CNC that is used, i.e. CNC-HCl or CNC-AA, there is no difference in the intensity of the CNC related bands indicating that the CNC-HCl is embedded into the polymer network and not removed during purification. It is not possible to differentiate which type of CNC that had been added from the FT-IR spectra of the gels, since the spectra are strongly dominated by the large content of PNIPAAm, coupled with the fact that there is only a small difference in the FT-IR spectra of the CNCs (ESI Fig. S3†).

Fig. 1 FT-IR spectra of PNIPAAm cryogels, with (dotted lines in red and green) and without CNC (black solid line).

To further verify the successful incorporation of CNC into the cryogel, TGA was performed. The TGA measurement showed that the degradation of the CNC containing gels was different from that of the neat gel, supporting the FT-IR observation that the CNC was present.

For the CNC-HCl containing cryogels, the amount of CNC added (1, 2 or 5 wt%) did not affect the degradation behavior; however, the thermal stability was observed to increase as compared to the neat (Fig. 2a). For the CNC-AA gels on the other hand, the thermal stability decreased with higher loadings of CNC-AA (Fig. 2b). The highest thermal stability of CNC-AA cryogels was observed for the gel containing 1 wt% of CNC-AA. The difference in degradation behaviors of the cryogels containing 5 wt% CNC-AA and those of the neat gels were very small. Only a minor increase in the degradation onset temperature of the modified gels, as well as small differences in the shapes of the degradation curves, could be seen.

Fig. 2 TGA thermograms of PNIPAAm cryogels containing CNC-HCl (left) and CNC-AA (right) of three different concentrations, as well as a neat PNIPAAm cryogel as a reference.

The difference in degradation behavior between the cryogels modified with CNC-HCl and those modified with CNC-AA showed that the surface functionalization of CNCs indeed affect the degradation behavior. A possible explanation for the lower onset temperature for the thermal degradation of CNC-AA containing samples could be that the CNCs formed aggregates in the gels, due to increased hydrophobicity of these CNCs. A poor dispersion of the CNCs in the cryogels would likely give rise to degradation behavior similar to those of the neat PNIPAAm gels, which have a lower onset temperature for degradation than neat CNCs.³⁸ This behavior during thermal degradation was seen for the samples containing 2 and 5 wt% of CNC-AA, which would indicate that these CNC-AAs form unstable NIPAAm/water suspensions already at very low CNC concentration (≥0.1 wt% in dispersion), causing aggregation of the CNCs. Our finding corroborated results by Gray and co-workers where CNCs where incorporated into a PVAc matrix, and where it was shown that a larger loading than 1.5 wt% disrupted rather than reinforced the hydrogel structure.³⁵ Even though the PNIPAAm cryogels containing CNC-AA at only 1 wt% of total dry content are the most stable, it is important to note that this depend on the choice of monomer and solvent used for the cryogel synthesis, as this would affect the aggregation behavior. Furthermore, the incorporation of CNC-AA in larger amounts also increases the crosslinking density. This increase could cause an overall more heterogeneous network structure. A heterogeneous structure could have a negative influence on the mechanical properties as previous studies have shown that a good dispersion of the CNCs in the matrix polymer is important.³⁵

All synthesized cryogels had an elastic, spongy, structure. To determine if the gels had the macroporous morphology expected for cryogels, FE-SEM of fractured surfaces was performed. Samples were fractured under liquid N₂, followed by lyophilization. The FE-SEM images clearly show that the morphology of the gels was affected by the addition of CNC (Fig. 3 and 4 and ESI Fig. S4†).

Fig. 3 FE-SEM of PNIPAAM cryogels with a crosslinking ratio of 1 : 25 (MBAm : NIPAAm) at ×100 magnification.

Fig. 4 FE-SEM of PNIPAAM cryogels with a cross-linking ratio of 1 : 15 (MBAm : NIPAAm) at ×100 magnification.

The pure PNIPAAm cryogels, independently of cross-linking density (1 : 15 or 1 : 25), exhibited large interconnected pores (10–200 μm diameter), forming a network of porous channels in the gels. The large pores were surrounded by porous polymer walls consisting of numerous interconnected, but significantly smaller, pores (approx. 1–10 μm in diameter) with thin walls.

The cryogels modified with 1 wt% of CNC-AA displayed clear morphological differences from the structures of the neat gels. The PNIPAAm₂₅CNC-AA_1% gels had a structure containing only large interconnected pores separated by thin walls. The PNIPAAm₁₅CNC-AA_1% gels, however, had structures of large interconnected pores separated by relatively thick polymer walls. The difference between these samples most likely depends on the higher crosslinking density of the PNIPAAm₁₅CNC-AA_1% gel. The cryogels with 5 wt% of CNC-AA were more similar in structure to the pure cryogels, but exhibited a denser structure of seemingly larger heterogeneity, most likely caused by aggregation of the CNCs and the higher crosslinking density of these samples.

The CNC-HCl-containing cryogels all displayed morphologies different from those of the corresponding neat PNIPAAm cryogels. The PNIPAAm gels containing 1 wt% CNC-HCl had large interconnected pores (10–200 μm diameter) surrounded by thick polymer walls containing only a few small pores (1–10 μm in diameter). The PNIPAAm gels containing 5 wt% of CNC-HCl contained pores of sizes 1–200 μm in diameter that did not appear to be interconnected. The PNIPAAm₁₅CNC-HCl_5% cryogels contained pores that appeared to be broken. This was likely caused by an initiation of the polymerization before the samples had been properly frozen, which then caused the polymer network to break during the ice-crystal formation. The thicker walls, and the broken structure of one of the CNC-HCl containing gels, suggested that the polymerizations either occurred at a faster rate, or that the freezing rate of water was slower in these polymerization systems than in those containing CNC-AA or in the neat cryogels.

The swelling ratios of freeze dried gels were monitored by placing the gels in deionized water at a temperature below the LCST of PNIPAAm and measuring weight gain over time (Fig. 5 and 6 and ESI Fig. S5†). Similar systems studied previously, mainly PVA gels reinforced with cellulose, showed a clear trend that the incorporation of cellulose improved the gel's swelling capacity and water uptake.^15,33 Herein, such a clear trend was not visible, as the swelling largely depended on the pore size and also due to the fact the PNIPAAm in itself has an exceptional swelling ratio in water. It was found that all gels, apart from PNIPAAm₂₅CNC-AA_1% and PNIPAAm₁₅CNC-HCl_5%, showed similar swelling behavior, swelling to approximately three times their dry weight in 10 minutes, but not reaching their final swelling ratio, of approximately fifteen times their dry weight, until more than 24 hours had passed. This indicated that the small pores and relatively thick polymer walls in these gels substantially increased the diffusion time of water within the gels. However, the PNIPAAm₂₅CNC-AA_1% cryogels reached their final swelling ratio within ten minutes, displaying a much faster swelling of the gels that consisted of large pores separated by thin polymer walls, which enabled fast diffusion of water molecules through the cryogels. PNIPAAm₁₅CNC-HCl_5% also displayed a faster swelling ratio than most of the gels, but a lower final swelling ratio. The swelling behavior of these cryogels could be explained by their morphology which enabled a fast swelling due to the large broken pores but that also lowered the highest possible water retention of these gels.

Fig. 5 Swelling ratio of freeze dried PNIPAAm cryogels with a cross-linking ratio of 1 : 25 (MBAm : NIPAAm) when placed in deionized water below the phase transition temperature.

Fig. 6 Swelling ratio of freeze dried PNIPAAm cryogels with a cross-linking ratio of 1 : 15 (MBAm : NIPAAm) when placed in deionized water below the phase transition temperature.

Thermoresponsive gels of PNIPAAm are known to collapse around the temperature for the LCST of PNIPAAm,⁴³ hence the produced cryogels are expected to have a phase transition temperature close to the LCST of the PNIPAAm. The thermoresponsive behavior of the cryogels was therefore determined by de-swelling the gels in deionized water at a temperature above LCST for PNIPAAm, i.e. well above 32 °C. When no further weight loss of the gels was registered, the gels were placed in deionized water below the phase transition temperature and the gain in mass was measured as a function of time (Fig. 7 and 8 and ESI Fig. S6†). The swelling was determined in weight% of the original mass of the gels.

Fig. 7 Mass percentage out of original mass, for cryogels placed in water below the phase transition temperature with a cross-linking ratio of 1 : 25 (MBAm : NIPAAm).

Fig. 8 Mass percentage out of original mass, for cryogels placed in water below the phase transition temperature with a cross-linking ratio of 1 : 15 (MBAm : NIPAAm).

A majority of the gels de-swelled to around 50% of the original mass when heated above the phase transition temperature, suggesting that the walls of these cryogels were too rigid to collapse fully upon heating. The re-swelling of these gels, after being placed in deionized water below the phase transition temperature, did not reach 100% and was not complete even after one week, possible due to irreversible collapse of some of the pores during de-swelling. However, the never dried gels exhibited a similar re-swelling behavior as that of the freeze-dried samples, indicating that the morphology of the gels remained unaltered during freeze-drying.

Three of the gels exhibited de- and re-swelling behaviors clearly different from the others. The PNIPAAm₂₅CNC-AA_1% gels de-swelled to 13% of their original mass, and then re-swelled to a swelling degree above 100% in less than ten minutes. The large de-swelling and rapid re-swelling could be attributed to the large pores which desorbed/adsorbed water quickly, and the thin walls which rendered the gel less rigid than those with thicker walls. Furthermore, the de-swelling of these gels occurred at a fast rate, drastically reducing the size of the gels after only one minute in deionized water above 32 °C (Fig. 9).

Fig. 9 Photographs of a PNIPAAm₂₅CNC-AA_1% cryogel after storage in deionized water below (left) and above (right) the phase transition temperature.

The PNIPAAm₁₅CNC-HCl_5% cryogels displayed a de-swelling of down to 10% of their original mass, and they re-swelled to 100% of the original mass within 60 minutes. This swelling behavior can most probably be attributed to the broken morphological structures in these gels which removed rigidness in the structures and facilitated the polymer chain collapse within these gels compared to undamaged gels. The fractured structure of these gels most likely also gave rise to the large pores which enabled the de-swelling/re-swelling to occur at a relatively fast rate.

The PNIPAAm₁₅CNC-HCl_1% gel de-swelled to less than 30% of their original mass, indicating a lower rigidity of the walls within these cryogels than for most of the other gels.

To investigate the mechanical properties of the cryogels, the compressive strain at break was determined for all samples (Fig. 10 and ESI Fig. S7†). Most of the gels had a compressive strain at break of around 50%, and did, up to this point, regain their original shape when the compressive force was released. This finding supports the explanation for the de-swelling properties upon heating of the gels; that the polymer structures of these cryogels are too rigid to undergo any further reversible collapse at approximately 50% deformation. As a consequence, this would cause the cryogels to break under further compression. In an investigation by Shibayama, where nanoparticle modified hydrogels were studied, it was shown that the incorporation of the nanoparticles resulted in hydrogels with extraordinary deformability and other studies into cellulose modified hydrogels have shown similar results.^11,24,37 In our study, the incorporation of the CNCs resulted in cryogels that were either similar to the neat PNIPAAm cryogels or were more rigid. However, the sample PNIPAAm₂₅CNC-AA_1% exhibited a compressive strain at break which was remarkably higher than the neat cryogel, above 90%, and the gel regained its shape upon release of the compressive force, showing that these cryogels were highly elastic (Fig. 11). An increase in the compressive stress at break, for the CNC containing samples in comparison to the neat gels, was seen for the gels with a cross-linking ratio of 1 : 15. This shows that the compressive strength of materials can be improved by the addition of CNC. This phenomena was, however, not evident for all the gels with a cross-linking ratio of 1 : 25.

Fig. 10 Compressive strain and compressive stress at break for the synthesized cryogels. *For sample PNIPAAm₂₅CNC-AA_1% only two out of five samples broke at the maximum compressive strain of 90%, thus indicating that the true compressive strain at break was even higher than the reported value, as the data for the non-broken gels were excluded from the data presented.

Fig. 11 Photographs of (a) PNIPAAm₂₅CNC-AA_1% cryogel and (b) PNIPAAm₂₅cryogel before (left), during (middle), and directly after (right), compression.

Conclusions

CNCs have successfully been incorporated into thermoresponsive PNIPAAm cryogels, by the addition of CNCs to the reaction mixtures prior to cryo-polymerization, as shown by FT-IR and TGA. CNCs modified with polymerizable groups that can be covalently incorporated into the network and unmodified CNCs physically adsorbed to the cryogels were both successfully incorporated into the cryogel structure. The addition of CNCs altered both the morphology and the mechanical properties of the gels. All gels showed a thermoresponsive behavior with de-swelling ratios of up to 87% when stored in deionized water above the phase transition temperature of the gels. When compressed below the compression of break, the gels immediately recovered to their original shape after removal of the compressive force. The PNIPAAm₂₅CNC-AA_1% cryogel showed both the largest response in swelling due to changes in temperature and could withstand the highest compressive strain before break. This shows that the presence of polymerizable groups on the surface of the CNCs affected the cryogel formation, and that the CNCs most probably had been covalently attached to the cryogel in this case. Surprisingly, this positive effect was not evident for the gels containing 2 and 5 wt% of the acrylate-functionalized CNCs compared to the gels with unmodified CNC. In this case, the fact that the CNC-AA is more hydrophobic than CNC-HCl could affect its incorporation, causing agglomeration of the CNCs in the reaction mixture at higher concentration of CNC. Furthermore, increasing the content of CNC-AA will also increase the overall crosslinking density of the system, causing a more heterogeneous system. Overall, the addition of CNCs to cryogels has shown to be a facile and effective way of altering the cryogel structure and to tune mechanical properties and swelling behavior.

Acknowledgements

The Vinnova funded excellence center BiMaC Innovation and Wilhelm Beckers Jubileumsfond are greatly acknowledged for financial support. Joakim Engström is greatly acknowledged for providing the CNC-HCl.

Notes and references

1. C. Chang and L. Zhang, Carbohydr. Polym., 2011, 84, 40–53.
2. A. Srivastava, E. Jain and A. Kumar, Mater. Sci. Eng., A, 2007, 464, 93–100.
3. R. Yoshida, Curr. Org. Chem., 2005, 9, 1617–1641.
4. J. E. Meiring, M. J. Schmid, S. M. Grayson, B. M. Rathsack, D. M. Johnson, R. Kirby, R. Kannappan, K. Manthiram, B. Hsia, Z. L. Hogan, A. D. Ellington, M. V. Pishko and C. G. Willson, Chem. Mater., 2004, 16, 5574–5580.
5. N. A. Peppas, J. Z. Hilt, A. Khademhosseini and R. Langer, Adv. Mater., 2006, 18, 1345–1360.
6. R. J. Russell, A. C. Axel, K. L. Shields and M. V. Pishko, Polymer, 2001, 42, 4893–4901.
7. I. Tokarev and S. Minko, Soft Matter, 2009, 5, 511–524.
8. K. S. Anseth, C. N. Bowman and L. Brannon-Peppas, Biomaterials, 1996, 17, 1647–1657.
9. M. Shibayama, Soft Matter, 2012, 8, 8030–8038.
10. C. Zhou, Q. Wu, Y. Yue and Q. Zhang, J. Colloid Interface Sci., 2010, 353, 116–123.
11. D. Yang, X. Peng, L. Zhong, X. Cao, W. Chen, S. Wang, C. Liu and R. Sun, RSC Adv., 2015, 5, 13878–13885.
12. S. Atifi, S. Su and W. Y. Hamad, Nord. Pulp Pap. Res. J., 2014, 29, 95–104.
13. J. Yang, J.-J. Zhao, F. Xu and R.-C. Sun, ACS Appl. Mater. Interfaces, 2013, 5, 12960–12967.
14. J. Yang, C.-R. Han, F. Xu and R.-C. Sun, Nanoscale, 2014, 6, 5934–5943.
15. J. Yang, C.-R. Han, J.-F. Duan, M.-G. Ma, X.-M. Zhang, F. Xu and R.-C. Sun, Cellulose, 2013, 20, 227–237.
16. K. Haraguchi and T. Takehisa, Adv. Mater., 2002, 14, 1120–1124.
17. J. P. Gong, Y. Katsuyama, T. Kurokawa and Y. Osada, Adv. Mater., 2003, 15, 1155–1158.
18. K. Haraguchi, Curr. Opin. Solid State Mater. Sci., 2008, 11, 47–54.
19. M. A. S. A. Samir, F. Alloin and A. Dufresne, Biomacromolecules, 2005, 6, 612–626.
20. J. Yang, C.-R. Han, X.-M. Zhang, F. Xu and R.-C. Sun, Macromolecules, 2014, 47, 4077–4086.
21. V. Favier, H. Chanzy and J. Y. Cavaille, Macromolecules, 1995, 28, 6365–6367.
22. J. Yang, J.-J. Zhao and X.-M. Zhang, Cellulose, 2014, 21, 3487–3496.
23. R. Cha, Z. He and Y. Ni, Carbohydr. Polym., 2012, 88, 713–718.
24. J. Yang, C.-R. Han, J.-F. Duan, M.-G. Ma, X.-M. Zhang, F. Xu, R.-C. Sun and X.-M. Xie, J. Mater. Chem., 2012, 22, 22467–22480.
25. B. M. A. Carvalho, S. L. Da Silva, L. H. M. Da Silva, V. P. R. Minim, M. C. H. Da Silva, L. M. Carvalho and L. A. Minim, Sep. Purif. Rev., 2014, 43, 241–262.
26. V. I. Lozinsky, Russ. Chem. Rev., 2002, 71, 489–511.
27. V. I. Lozinsky, I. Y. Galaev, F. M. Plieva, I. N. Savina, H. Jungvid and B. Mattiasson, Trends Biotechnol., 2003, 21, 445–451.
28. F. M. Plieva, I. Y. Galaev and B. Mattiasson, Macroporous Polymers Production Properties and Biotechnological/Biomedical Applications, ed. B. Mattiasson, A. Kumar and I. Y. Galaev, CRC Press, 2009, pp. 23–47.
29. X.-Z. Zhang and C.-C. Chu, J. Mater. Chem., 2003, 13, 2457–2464.
30. P. Perez, F. Plieva, A. Gallardo, J. San Roman, M. R. Aguilar, I. Morfin, F. Ehrburger-Dolle, F. Bley, S. Mikhalovsky, I. Y. Galaev and B. Mattiasson, Biomacromolecules, 2008, 9, 66–74.
31. S. Hajizadeh, H. Kirsebom and B. Mattiasson, Soft Matter, 2010, 6, 5562–5569.
32. I. Y. Galaev, M. B. Dainiak, F. Plieva and B. Mattiasson, Langmuir, 2007, 23, 35–40.
33. O. M. Paduraru, D. Ciolacu, R. N. Darie and C. Vasile, Mater. Sci. Eng., C, 2012, 32, 2508–2515.
34. L. E. Millon, G. Guhados and W. Wan, J. Biomed. Mater. Res., Part B, 2008, 86, 444–452.
35. T. Abitbol, T. Johnstone, T. M. Quinn and D. G. Gray, Soft Matter, 2011, 7, 2373–2379.
36. A. Mihranyan, Cellulose, 2013, 20, 1369–1376.
37. K. Syverud, H. Kirsebom, S. Hajizadeh and G. Chinga-Carrasco, Nanoscale Res. Lett., 2011, 6, 626.
38. A. Boujemaoui, S. Mongkhontreerat, E. Malmström and A. Carlmark, Carbohydr. Polym., 2015, 115, 457–464.
39. O. van den Berg, M. Schroeter, J. R. Capadona and C. Weder, J. Mater. Chem., 2007, 17, 2746–2753.
40. T. Kondo, Cellulose, 1997, 4, 281–292.
41. M. Kac̆uráková, P. Capek, V. Sasinková, N. Wellner and A. Ebringerová, Carbohydr. Polym., 2000, 43, 195–203.
42. M. Sekkal, V. Dincq, P. Legrand and J. P. Huvenne, J. Mol. Struct., 1995, 349, 349–352.
43. G. A. Komarova, S. G. Starodubtsev, V. I. Lozinsky, E. V. Kalinina, K. Landfester and A. R. Khokhlov, Langmuir, 2008, 24, 4467–4469.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra12603g

---