EP2895545A2 - Polymer nanocomposite having switchable mechanical properties - Google Patents

Abstract

A polymer nanocomposite (1) having switchable mechanical properties comprises a matrix polymer (2) and a nanoparticle network, wherein the nanoparticle network is formed by a formation of a substantially three-dimensional network of nanoparticles (3) which are incorporated in the matrix polymer (2) and interact with each other and/or with the matrix polymer (2). The polymer nanocomposite (1) in a first switching state comprises a first stiffness characterized by a first tensile storage modulus (Ε') greater than 6 GPa, in a second switching state comprises a second stiffness characterized by a second tensile storage modulus (E') of less than 1 GPa and is switchable between the first switching state and the second switching state by exposing the polymer nanocomposite (1) to a stimulus that influences interactions among the nanoparticles (3) and/or between the nanoparticles (3) and the matrix polymer (2). In this way a polymer nanocomposite is provided which in a beneficial way may be employed within medical applications

Description

Polymer nanocomposite having switchable mechanical properties

Description The invention relates to a polymer nanocomposite having switchable mechanical properties, a method for producing a polymer nanocomposite and a method for inducing a stiffness change in a polymer nanocomposite having switchable mechanical properties.

Such nanocomposite comprises a matrix polymer and a nanoparticle network. The nanoparticle network is formed by a formation of a substantially three-dimensional network of nanoparticles which at least partly are dispersed in the matrix polymer and interact with each other and/or with the matrix polymer. The polymer nanocomposite in a first switching state comprises a first stiffness characterized by a first tensile storage modulus and in a second switching state comprises a smaller, second stiffness characterized by a second tensile storage modulus. The nanocomposite is switchable between the first switching state and the second switching state by exposing the polymer nanocomposite to a stimulus that influences interactions among the nanoparticles and/or between the nanoparticles and the matrix polymer. A nanocomposite of this kind is for example known from US 2009/0318590 A1 , which shall be incorporated herein by reference.

Materials that selectively respond to external stimuli are often referred to as "smart", "intelligent", or "adaptive" due to their intrinsic ability to change their physical or chemical properties on command. Among many other uses, they have garnered significant attention due to their potential applications in biomedical and biotechnological fields, including their use as transient implants, drug delivery carriers, tissue-engineering scaffolds, thermoresponsive hydrogels, self-healing materials, cell cultures, bioseparation membranes, sensors and actuators. One intriguing feature exhibited by some of these materials is their ability to change mechanical properties "on command ". One recently demonstrated approach for the design of such materials relies on the preparation of polymer-based nanocomposites, which are comprised of a polymeric matrix and reinforcing nanoparticles, whose interactions are stimuli-responsive and regulate the mechanical properties of the bulk material (compare for example Shanmuganathan, K. et al., Prog Polym Sci, 2010, 35, 212, which shall be incorporated herein by reference).

Amongst many potential nanofillers, crystalline cellulose nanoparticles, referred to as "nanowhiskers", have been investigated. Cellulose nanowhiskers (NWs) can be isolated from a variety of sources, including plants (e.g. wood, cotton, or wheat straw), marine animals (tunicates), as well as bacterial sources, such as algae, fungi, and amoeba (protozoa). Using NWs isolated from tunicates, the first NW-reinforced polymer nanocomposites were reported in 1995 (compare Favier V. et al., Macromolecules, 1995, 28, 6365 and Favier V. et al., Polym. Adv. Technol. , 1995, 6, 351 ). These materials displayed substantially enhanced mechanical properties, which were explained with the formation of a percolating, hydrogen-bonded network of NWs within the polymer matrix. The widespread interest in NW-based nanocomposites is explained by the low cost, outstanding mechanical properties, availability, sustainability, biodegradability, and low density of NWs. It was demonstrated that the stiffness of NW-based nanocomposites can be reversibly changed by controlling the degree of interactions between the rigid filler (compare for example Capadona, J. R. et al., Science. 2008, 319, 1370, to be incorporated herein by reference). The NWs form a percolating network within the matrix which is - in the absence of a competitive hydrogen bonding agent - held together by hydrogen bonds among the surface hydroxyl groups. This causes a significant reinforcement of the polymer matrix. Upon exposure to chemicals that can competitively hydrogen-bond to the NWs, e.g. water, the interactions between individual NWs are reduced and the nanocomposite softens considerably, as predicted by mechanical models. Mechanically-adaptive nanocomposites described in US 2009/0318590 A1 are based on a rubbery ethylene oxide-epichlorohydrin copolymer (EO-EPI) and cellulose nanowhiskers isolated from tunicates (TNWs). The dry EO-EPI/TNW nanocomposites exhibit an increase of the storage modulus E' from 3.7 MPa (neat EO-EPI) to 800 MPa at a NW content of 19% v/v at 25 °C. These materials undergo a pronounced and reversible modulus reduction from 800 to 20 MPa upon exposure to water. Plasticization of the polymer matrix upon aqueous swelling has been shown to reinforce the effect, for example in nanocomposites based on polyvinyl acetate) (PVAc) and TNWs or CNWs. PVAc TNW-based nanocomposites may show an increase of the E^' from for example 1 .8 GPa (neat PVAc) to 5.2 GPa for a nanocomposite containing 16.5% v/v TNWs. Due to the glassy nature of the PVAc matrix, these materials exhibit a much higher initial stiffness than the EO-EPI based systems, but soften greatly (5.2 GPa to 12 MPa) upon exposure to water due to matrix plasticization and TNW decoupling.

Such mechanically-adaptive NW-containing nanocomposites are potentially useful as substrates for intracortical microelectrodes. Neural prosthetic devices, which connect the brain with the outside world, promise to be useful for many clinical applications, but it has proven difficult to achieve long-term connectivity, presumably on account of the mechanical mismatch between current electrode materials and the cortical tissue. Initial in-vivo experiments with PVAc TNW nanocomposites suggest that mechanically-adaptive intracortical neural prosthetics can more rapidly stabilize neural cell populations at the interface than rigid systems, which may serve for improving the functionality of intracortical devices (compare for example Harris, J. P. et al., J. Neural Eng., 201 1 , 8, 040610).

It is an object of the instant invention to provide a polymer nanocomposite, a method for producing a polymer nanocomposite and a method for inducing a stiffness change in a polymer nanocomposite which in a beneficial way may be employed within medical applications.

The object is achieved by means of a polymer nanocomposite having switchable mechanical properties and comprising a matrix polymer and a nanoparticle network. The nanoparticle network is formed by a formation of a substantially three-dimensional network of nanoparticles which are incorporated, for example dispersed, in the matrix polymer and interact with each other and/or with the matrix polymer. Herein, the polymer nanocomposite in a first switching state comprises a first stiffness characterized by a first tensile storage modulus greater than 6 GPa, in a second switching state comprises a second stiffness characterized by a second tensile storage modulus of less than 1 GPa and is switchable between the first switching state and the second switching state by exposing the polymer nanocomposite to a stimulus that influences interactions among the nanoparticles and/or between the nanoparticles and the matrix polymer.

It has been found that the realization of devices for biomedical applications, such as intracortical electrodes and other devices, benefits if mechanically-adaptive materials with a stiffness characterized by a tensile storage modulus (Ε') of greater than 6 GPa, preferably more than 8 GPa, and most preferably more than 9 GPa, are provided. Hence, a polymer nanocomposite is provided which, in its first switching state, comprises an enhanced stiffness - as compared to previously known switchable nanocomposites - characterized by a tensile storage modulus greater than 6 GPa. By exposing the polymer nanocomposite to a stimulus, for example a composition containing water such as body tissue or a body fluid, the polymer nanocomposite can be switched to a second switching state in which the stiffness is reduced to a tensile storage modulus of less than 1 GPa. Hence, in the second switching state the polymer nanocomposite is significantly softer than in the first switching state.

The stimulus serves to influence interactions among the nanoparticles and/or between the nanoparticles and the matrix polymer. By means of the stimulus, hence, the mechanical properties of the polymer nanocomposite are altered by influencing the interactions of the constituents of the polymer nanocomposites, i.e. the interactions between nanoparticles between themselves and between the nanoparticles and the matrix polymer. Due to the effect of the stimulus on the interactions between the nanoparticles among each other and/or with the matrix polymer the stiffness of the polymer nanocomposite is changed, allowing for a switching between the first switching state and the second switching state.

The nanocomposite exhibits a tensile storage modulus of less than 1 GPa, preferably less than 500 MPa, more preferably less than 200 MPa, most preferably less than 100 MPa, in particular 20 MPa, in the second switching state.

The switching herein is preferably reversible. Hence, the polymer nanocomposite may be switched from the first switching state to the second switching state and vice versa.

Although the switching preferably is reversible, it shall be noted that in principle this is not necessary. It also is conceivable that the nanocomposite can be switched only in one direction for example from the first switching state to the second switching state, hence from rigid to soft. The first tensile storage modulus in the first switching state of the polymer nanocomposite is preferably measured in a dry state at room temperature, i.e. 25 C. The second tensile storage modulus is preferably measured when the polymer nanocomposite is subjected to a stimulus, for example a body fluid, at the temperature of the stimulus, for example 37 C.

As discussed more thoroughly below, it has been shown in many scientific studies, which are incorporated here by reference, that the tensile storage modulus of polymer nanocomposites in which a nanoparticle network is formed by a formation of a substantially three-dimensional network of nanoparticles which are incorporated, for example dispersed, in the matrix polymer and interact with each other through non- covalent interactions, can be predicted by a widely used theoretical model that is referred to as the percolation model (see for example: Capadona, J.R; van den Berg, O.; Capadona, L; Schroeter, M.; Tyler, D.; Rowan, S.J.; Weder, C; "A versatile approach for the processing of polymer nanocomposites with self-assembled nanofibre templates "; Nature Nanotechnology, 2007, 2, 765-769; Capadona, J.R; Shanmuganathan K.; Tyler, D.; Rowan, S.J.; Weder, C; "Bio-inspired chemo-mechanical polymer nanocomposites that mimic the sea cucumber dermis"; Science, 2008, 319, 1370-1374. Capadona, J.R; Shanmuganathan K.; Trittschuh, S.; Seidel, S.; Rowan, S.J.; Weder, C; "Polymer Nanocomposites with Microcrystalline Cellulose"; Biomacromolecules, 2009, 10, 712- 716; Shanmuganathan K.; Capadona, J.R.; Rowan, S.J.; Weder, C; "Biomimetic Mechanically Adaptive Nanocomposites"; Prog. Polym. Sci., 2010, 35, 212- 222. Shanmuganathan K.; Capadona, J.R.; Rowan, S.J.; Weder, C; "Bio-inspired mechanically-adaptive nanocomposites derived from cotton cellulose whiskers"; J. Mater. Chem., 2010, 20, 180-186). The model relies on several parameters, which are discussed below, but it does not feature a specific parameter that takes into account interactions between the nanofiller and the matrix polymer. Consequently, a given filler is assumed to induce a similar reinforcing effect in different polymers. This was also experimentally proven for many systems. It now was surprisingly found that the reinforcing effect, and therewith the tensile storage modulus (Ε') in the first switching state, is much higher than predicted by the percolation model if a matrix polymer is employed which can display strong interactions with the nanofiller. In one embodiment of the present invention, for example, such an effect was achieved by providing polar poly(vinylalcohol) as matrix polymer and cellulose nanowhiskers as nanoparticles, allowing for hydrogen bonding between them. The polymer nanocomposite preferably is obtainable applying a process comprising the steps of:

- providing the matrix polymer,

- incorporating the nanoparticle network into the matrix polymer to obtain an intermediate mixture,

- subjecting the intermediate mixture to a heat treatment step involving exposure to a temperature equal to or greater than 100 C. For example, in one embodiment, a matrix polymer may be provided dissolved in a fluid such as deionized water. Nanoparticles may in one embodiment be provided for example in the shape of powder in a lyophilized or spray-dried state and may be dispersed in a fluid such as deionized water. Both the solution of the matrix polymer and the dispersion of the nanoparticles may be obtained by techniques known to the person skilled in the art, for example by a stirring and/or sonicating for a predetermined amount of time at a predetermined temperature possibly greater than room temperature. Also, the mixture of the matrix polymer and the nanoparticles may be obtained by a suitable mixing technique known to the person skilled in the art, such as stirring or sonication or both, resulting in a homogeneous mixture.

The heat treatment step may for example include a compression molding of the polymer nanocomposite at a temperature equal to or greater than 100 C The compression molding may for example take place at a temperature between 100 C and 160 C, for example at 120 C or 150 Ce. During compression molding a pressure, for example between 500 psi and 5000 psi, for example 1000 psi or 2000 psi, is applied to the intermediate mixture of the matrix polymer and the nanoparticles for a predetermined amount of time. For example, in a first step a pressure of 1000 psi may be applied for a duration of 1 to 20 minutes, for example 2 minutes, followed by a second step in which a pressure of 2000 psi for a duration of 5 to 45 minutes, for example 15 minutes, is applied, hence applying different pressures for different amounts of time.

After the heat treatment step the resulting polymer nanocomposite may be allowed to rest for a predetermined amount of time, for example 30 to 150 minutes, for example 90 minutes, possibly at an elevated temperature smaller than the temperature applied during the heat treatment step, for example 70 C. Prior to the heat treatment step, the intermediate mixture obtained by incorporating the nanoparticles into the matrix polymer is preferably subjected to a drying step involving exposure to an elevated temperature of 30 C to 80 °C over an elongated period of time of for example one day to 10 days. For example, the drying step may include a first drying step including exposure to a first drying temperature of 30 °C to 50 °C, for example 35 °C, over a first period of time of one day to 10 days, for example five days. The drying step may further include a second drying step including exposure to a second drying temperature of 50 C to 80 C, for example 70 C, over a second period of time of five hours to 48 hours, for example 24 hours. Such drying steps serve to evaporate water from the mixture of the matrix polymer and the nanoparticles and may be carried out in a suitable oven into which the mixture of the matrix polymer and the nanoparticles is placed.

The exposure of the mixture of the matrix polymer and the nanoparticles to an elevated temperature of equal to or greater than 100 C during the heat treatment step has a twofold effect. In this regard it was surprisingly found that a heat treatment of the mixture of the matrix polymer and the nanoparticles dispersed therein may serve to increase the stiffness of the resulting polymer nanocomposite in the first switching state, hence yielding a material which exhibits a large stiffness in its first, initial switching state. Furthermore, the heat treatment step may also serve to adjust the stiffness in the soft, second switching state, as well as the swelling characteristics of the polymer nanocomposite when subjected to a stimulus in the shape of a water-containing fluid.

The matrix polymer, in a preferred embodiment, comprises a - preferably highly - polar polymer capable of forming non-covalent interactions, in particular hydrogen bonds, with the nanoparticles. For example, the matrix polymer may comprise vinyl polymers such as polyvinyl alcohol (PVOH), poly(acrylic acid), poly(acryl amide)s, polyvinyl pyridine), copolymers of these respective monomers and other monomers, polycondensates such as polyamides and polyesters, polysaccharides such as cellulose, starch, alginates, pectins, hyaluronane, chitin, chitosan and their derivatives, proteins, and other polar polymers. By forming non-covalent interactions such as hydrogen bonds with the nanoparticles, the matrix polymer in the first switching state may interact with the nanoparticles yielding an increased stiffness in the first switching state. Such bonds are effected by a stimulus and are at least partly released when subjecting the polymer nanocomposite to the stimulus such that the stiffness of the polymer nanocomposite is decreased yielding a significantly reduced stiffness in the second switching state. The nanoparticles are advantageously capable of forming non-covalent interactions, such as hydrogen bonds, with each other and/or with the matrix polymer.

In an advantageous embodiment, nanofibers having a length which is large with respect to their width are used as nanoparticles. In one embodiment, cellulose nanowhiskers such as nanowhiskers derived from tunicates (tunicate nanowhiskers, TNW) or from cotton (cotton nanowhiskers, CNW) may be used as nanofibers.

TNWs are advantageously isolated from tunicates by preparing a cellulose pulp and by applying a hydrolysis, for example a sulphuric acid hydrolysis, to the cellulose pulp.

CNWs may for example be obtained by isolation employing for example a hydrolysis, for example a sulphuric acid hydrolysis, and dialysis treatment, followed for example by a sonication for a predetermined amount of time, for example three hours, and a settlement period of for example 18 hours. The resulting supernatant may then be decanted off, and the resulting CNW dispersion may be dried, for example spray-dried to yield a dried CNW powder.

Advantageously, nanofibers comprising an aspect ratio (ratio between length and diameter) greater than 5, preferably greater than 10, more preferably greater than 50, most preferably greater than 80 are employed. For example, CNWs exhibit an aspect ratio of for example about 10 (having a length of about 200 ± 70 nm and a diameter (width) of about 22 ± 6 nm). TNWs exhibit an aspect ratio of for example about 83 (having a length of for example about 2500 ± 1000 nm and a diameter (width) of about 30 ± 5 nm).

The polymer nanocomposite, as substantial constituents, comprises a matrix polymer, for example PVOH, and nanoparticles. In this regard, the polymer nanocomposite may comprise in one embodiment a content of 1 % to 30 %, preferably 2 % to 25 %, more preferably 3 % to 20 % v/v of nanoparticles (v/v indicates the volume fraction defined as the volume of a constituent divided by the volume of all constituents of a mixture prior to mixing, in contrast to w/w which indicates the weight fraction). Dependent on the exact content fraction of the nanoparticles in the matrix polymer the mechanical properties of the polymer nanocomposite in the first switching state and in the second switching state may fine-tuned. In general, a larger fraction of nanoparticles in the polymer nanocomposite yields a larger stiffness both in the first switching state and in the second switching state. The polymer nanocomposite may - besides the matrix polymer and the nanoparticles - comprise further constituents such as - but not limited to - plasticizers, process agents, stabilizers and/or other filling material.

As mentioned already above, the switching between the switching states is preferably reversible. Hence, for switching the polymer nanocomposite from the first switching state to the second switching state a stimulus reduces the interactions among the nanoparticles and/or between the nanoparticles and the matrix polymer, thus reducing the stiffness of the polymer nanocomposite. Vice versa, for switching the polymer nanocomposite from the second switching state to the first switching state a stimulus enhances the interactions among the nanoparticles and/or between the nanoparticles and the matrix polymer, thus increasing the stiffness of the polymer nanocomposite. Herein, the stimulus for switching the polymer nanocomposite from the first switching state to the second switching state may, in one embodiment, include subjecting the polymer nanocomposite to a water-containing composition, in particular a water- containing fluid, for example a body fluid. The water-containing composition herein may have an elevated temperature of for example 30 °C to 50 ' C, preferably 35 ' C to 40 C, for example 37 °C.

The switching between the first switching state and the second switching state takes place over time and hence must be considered as a process. It can be imagined that the switching processed may be stopped after some time prior to reaching a final switching state, the polymer nanocomposite then adopting an intermediate switching state in between the first switching state and the second switching state. In this regard, the first switching state and the second switching state represent extreme states prior or after switching which the polymer nanocomposite may adopt. The process may be reversed by simply removing the stimulus which is applied for switching form the first to the second switching state. For example, the polymer nanocomposite may be switched from the second switching state to the first switching state by drying the water-containing nanocomposite, either at ambient conditions, or under vacuum and/or at an elevated temperature, or under other suitable conditions

The ratio of the storage moduli of the two switching states may, in one embodiment, be greater than 20, preferably greater than 50, more preferably greater than 100, and most preferably greater then 500. For example, a polymer nanocomposite using PVOH as matrix polymer and nanoparticles in the shape of tunicate nanowhiskers (TNWs) or cotton nanowhiskers (CNWs) may exhibit a change of storage moduli from 6.8-13.7 GPa in the first switching state to 2-160 MPa in the second switching state.

The nanoparticle network formed by the formation of nanoparticles is substantially three- dimensional, hence yielding essentially isotropic mechanical properties of the polymer nanocomposite. In particular, the stiffness characterized by the tensile storage modulus in the first switching state and in the second switching state is substantially equal in all three spatial directions, such that no preferred direction with an increased tensile storage modulus exists.

Those skilled in the art will appreciate that other embodiments may rely on the same principle. For example, a nanocomposite may be produced from a matrix polymer that has suitable groups for pi-pi-interactions and nanoparticles which are suitably functionalized so that they can exhibit pi-pi interactions with the polymer and/or other nanoparticles. Exposure to an agent that can form pi-pi interactions with the polymer and/or the nanoparticles can then serve to switch between said two switching states. Those skilled in the art will also know that other embodiments may rely on appropriately functionalized nanoparticles other than cellulose nanowhiskers, for example, but not limited to, boemite whiskers, carbon nanotubes, and nanoparticles made from synthetic or natural polymers. The object is also achieved by a method for producing a polymer nanocomposite having switchable mechanical properties, comprising the steps of:

- providing a matrix polymer and

- incorporating a nanoparticle network into the matrix polymer, wherein the nanoparticle network is formed by a formation of a substantially three-dimensional network of nanoparticles which are incorporated, for example dispersed, in the matrix polymer and interact with each other and/or with the matrix polymer, to obtain an intermediate mixture,

- subjecting the intermediate mixture to a heat treatment step involving exposure to a temperature equal to or greater than 100 °C.

As a result, advantageously n a polymer nanocomposite which in a first switching state comprises a first stiffness characterized by a first tensile storage modulus greater than 6 GPa, in a second switching state comprises a second stiffness characterized by a second tensile storage modulus of less than 1 GPa and is switchable between the first switching state and the second switching state by exposing the polymer nanocomposite to a stimulus that influences interactions among the nanoparticles and/or between the nanoparticles and the matrix polymer is obtained.

The object is furthermore achieved by a method for inducing a stiffness change in a polymer nanocomposite having switchable mechanical properties, the method comprising the steps of:

- providing a matrix polymer incorporating a nanoparticle network, wherein the nanoparticle network is formed by a formation of a substantially three-dimensional network of nanoparticles which are incorporated, for example dispersed, in the matrix polymer and interact with each other and/or with the matrix polymer,

- switching the polymer nanocomposite between a first switching state, in which the polymer nanocomposite comprises a first stiffness characterized by a first tensile storage modulus greater than 6 GPa, and a second switching state, in which the polymer nanocomposite comprises a second stiffness characterized by a second tensile storage modulus of less than 1 GPa, by exposing the polymer nanocomposite to a stimulus that influences interactions among the nanoparticles and/or between the nanoparticles and the matrix polymer.

The advantages and advantageous embodiments described above for the polymer nanocomposite are applicable to such methods in an analogous fashion such that it shall be referred to the above explanations.

The idea underlying the invention shall subsequently be described in more detail with regard to the embodiments shown in the figures. Herein,

Fig. 1A, B show AFM height images for lyophilized TNWs (Fig. 1A) and spray- dried CNWs (Fig. 1 B) deposited from aqueous dispersions (0.1 mg/mL) onto freshly cleaved mica surfaces;

Fig. 2A-D show dynamic mechanical analysis (DMA) data of dry PVOH and dry PVOH/NW nanocomposites as a function of temperature and NW content: Tensile storage moduli E' (Fig. 2A) and loss tangent tan δ (Fig. 2B) of PVOH TNW nanocomposites; tensile storage moduli E' (Fig. 2C) and loss tangent tan δ (Fig. 2D) of PVOH/CNW nanocomposites; shows tensile storage moduli (Ε') of neat PVOH and PVOH/TNW nanocomposites as a function of NW content in the dry state at 100 C (■), re-dried after swelling with ACSF for 1 week (□), ACSF- swollen after immersion in ACSF at 37 °C for 1 week (A ) and 1 month (Δ) (the solid line shows values predicted by the percolation model for the dry state (E'_r= 80 GPa), the dotted line shows values predicted by the Halpin-Kardos model for samples conditioned in ACSF at 37 °C (E'_lr = 130 GPa, E'_tr = 5 GPa, E'_s = 1 1 Pa, G'_r = 1.77 GPa, G^'s = 3.9 MPa, v, = 0.3, v_s = 0.44)); shows tensile storage moduli (Ε') of neat PVOH and PVOH/CNW nanocomposites as a function of NW content in the dry state at 100 °C (■), and ACSF-swollen after immersion in ACSF at 37 °C for 1 week (▲) (the solid line shows values predicted by the percolation model (E'_r = 10 GPa), the dotted line shows the prediction by the Halpin-Kardos model for samples conditioned in ACSF at 37 °C (EY = 130 GPa, E',_r = 5 GPa, E'_s = 1.44 MPa, G _r = 1 .77 GPa, G'_s = 0.5 MPa, v_r = 0.3, v_s = 0.44); because of solvent uptake the volume fraction of NWs in ACSF-swollen samples is lower than in the dry state; Data points represent averages of N = 3-6 measurements ± s.d.); shows data indicating the swelling of PVOH/TNW nanocomposites compression-molded at 150 °C (■), PVOH/CNW nanocomposites compression-molded at 120 C ( ·), a PVOH/TNW nanocomposite compression-molded at 120 C (o), and a PVOH/CNW nanocomposite compression-molded at 150 °C (□) as a function of NW content, after the samples were immersed in ACSF at 37 °C for 1 day (data represent averages of N = 3 measurements ± s.d.); show dynamic mechanical analysis (DMA) data of ACSF-swollen films of neat PVOH and PVOH/TNW nanocomposites (Fig. 5A) and PVOH/CNW nanocomposites (Fig. 5B) as a function of temperature and NW content after immersion in ACSF at 37 °C for 1 week; shows a schematic drawing illustrating the interactions between PVOH and NWs; shows an enlarged view of the drawing of Fig. 6 A in the region A; show conductometric titration curves of NWs: Lyophilized TNWs (Fig. 7A), spray-dried CNWs (Fig. 7B), and lyophilized CNWs (Fig. 7C); show representative scanning electron microscopy images of NWs (SEM, Fig. 8A) and transition electron microscopy images of NWs (TEM, Fig. 8B-D), wherein Fig. 8 A and 8B show lyophilized TNWs, Fig. 8C shows spray-dried CNWs and Fig. 8D shows lyophilized CNWs, all deposited from aqueous dispersions (0.1 mg/mL); show DSC thermograms (second heating) of PVOH TNW nanocomposites (Fig. 9A) and PVOH/CNW nanocomposites (Fig. 9B), together with thermograms of neat PVOH films compression- molded at 150 °C (Fig. 9A) and 120 °C (Fig. 9B); shows DMA data of ACSF-swollen films of neat PVOH and PVOH TNW nanocomposites after immersion in ACSF at 37 °C for 1 month; shows DMA data of ACSF-swollen films of 16% v/v PVOH/TNW nanocomposites after immersion in ACSF at 37 °C for 1 day (the processing temperature is indicated in the figure); shows DMA data of ACSF-swollen films of 16% v/v PVOH/CNW nanocomposites after immersion in ACSF at 37 °C for 1 day (the processing temperature is indicated in the figure); states thermal properties of neat PVOH and PVOH/NW nanocomposites as a function of NW type and content; Table 2 states tensile storage moduli (Ε') of dry and ACSF-swollen films of neat PVOH and PVOH/NW nanocomposites determined by DMA (data represent averages (N = 4-7));

Table 3 gives a comparison of tensile storage moduli (Ε') of current materials with previous mechanically-adaptive nanocomposites comprising -16% v/v of NWs; and

Table 4 states swelling data of neat PVOH and PVOH TNW nanocomposites at 37 °C in ACSF as a function of TNW content (data represent averages (N = 3) ± s.d.).

The present invention shall subsequently be described in terms of mechanically-adaptive nanocomposites comprising polyvinyl alcohol) (PVOH) and cellulose nanowhiskers (NWs), whose mechanical properties change significantly upon exposure to physiological conditions. It shall be noted, however, that the invention is not limited to such particular composites described herein, but may also be implemented using other matrix polymer materials and/or nanoparticles.

PVOH/NW nanocomposites described herein are produced using NWs derived from tunicates (TNWs) and cotton (CNWs) having different aspect ratios (TNWs = 83 and CNWs = 10), surface charge densities (TNWs = 90 mmol/kg and CNWs = 40 mmol/kg), and filler contents (0-16% v/v). Dynamic mechanical analysis results are presented in the following which show an enhancement of the tensile storage moduli (E ) of the dry nanocomposites in comparison to the neat PVOH.

For example, as will be described in detail below, below the glass transition temperature (T_g -70 C) the tensile storage modulus E^' of the dry PVOH TNW nanocomposites increases from 7.3 GPa for the neat polymer to 13.7 GPa for the nanocomposite with 16% v/v TNWs. The stiffness of the materials is greatly reduced upon exposure to aqueous conditions, on account of reduced NW-NW and NW-matrix interactions. Immersing the nanocomposites into artificial cerebrospinal fluid (ACSF) at body temperature (37 °C) causes a drastic drop in E', for example from 13.7 GPa to 160 MPa in the case of the nanocomposites containing 16% v/v TNWs. Nanocomposites comprising 16% v/v CNWs exhibit an E^' of 9.0 GPa in the dry state, which dropped to -1 MPa upon exposure to artificial cerebrospinal fluid (ACSF) used for simulating physiological conditions. As will be described, the swelling characteristics of the materials, and therewith the extent of mechanical switching, is also influenced via the processing conditions. The isolation, modification, characterization and application of cellulose nanofibers from plants (including wood, coconut husks, sisal, tunicates, cotton, ramie, straw, sugar beet, and many others) and animals (e.g. tunicates) is currently attracting significant attention, driven by the abundance and renewable nature of the biological sources and the attractive mechanical properties of nanocellulose. The latter originate from the hierarchical, uniaxially oriented structure of native cellulosic materials. Extended and aligned cellulose macromolecules (D-glucose units, which are condensed through β(1→4) glycosidic bonds) assemble into microfibrils, in which they are stabilized through hydrogen bonds. The microfibrils are largely crystalline, but they also contain regions that are less well ordered (i.e., largely amorphous). The cross-sectional dimension of the microfibrils ranges from 2-20 nm, depending on the origin of the cellulose. The elementary fibrils aggregate to form microfibrils with diameters of -15-20 nm, which further aggregate into larger bundles and finally, with "binders^'' consisting of amorphous lignin and hemicelluloses, into cellulosic fibers. These hierarchical structures can be deconstructed by mechanical and chemical processes, which yield highly crystalline cellulose nanofibers, collectively referred to as nanocelluloses or cellulose nanofibers.

Two main types of nanocelluloses can be distinguished - microfibrillated cellulose (MFC) and cellulose nanocrystals (CNC) - although it should be noted that differences in the isolation processes and the nature of the source also cause some variation within these families.

The isolation of MFC (also referred to as cellulose microfibrils, nanofibrillated cellulose, and cellulose nanofibrils) involves milling the native cellulose, followed by subsequent alkali and bleaching treatments to eliminate lignin and hemicelluloses. Mechanical disintegration of bleached wood pulp, which has traditionally been carried out in high- pressure homogenizers, separates the microfibrils and affords MFC in the form of fibrils with a diameter of -20 nm and a length of several micrometers. The ' amorphous defects^'' present in the native microfibrils are retained and the fibrous nature gives rise to entangled, network-like structures.

CNCs (also referred to as cellulose whiskers or nanowhiskers (NWs), nanocrystalline cellulose (NCC), cellulose microcrystals, and rod-like cellulose crystals) are produced by the hydrolysis of the bleached cellulose pulp with mineral acids. The hydrolysis serves to remove the amorphous cellulose domains, thereby also reducing the molecular weight. The remaining particles are usually separated by ultrasonic treatment. The dimensions of the resulting rod-like CNC particles depend primarily on the structure in the native material (i.e. the source) and to a lesser extent also on the hydrolysis conditions employed. The diameter of CNCs is in the range of 5 - 25 nm and their length varies between 100 - 300 nm (in the case of wood and cotton) and several pm (in the case of tunicin). Since CNCs lack the amorphous domains present in the original microfibrils (and also in MFC), they display extremely high elastic moduli (100-150 GPa) and a high tensile strength (-10 GPa) in their long direction. A discussion of nanocellulose production and properties has been thoroughly treated in several reviews, such as Siqueira, G. et al., Polymers 2010, 2, 728 and Klemm, D. et al., Angew. Chem. Int. Ed. 201 1 , 50, 5438, incorporated herein by reference. For the embodiments described below, nanowhiskers derived from tunicates (TNWs) or cotton (CNWs) have been employed, both representing types of cellulose nanocrystals (CNCs), although in principle also other types of nanofibers may be used.

Subsequently, in a first section experiments conducted shall be described. In a second section, then, the results of the experiments shall be discussed to illustrate the effects of nanocomposites embodying the invention.

EXPERIMENTS Materials. Polyvinyl alcohol) (PVOH) 99% hydrolyzed (Mw = 85,000-124,000 g/mol; v = 1.26 g/mL) and all reagents were purchased from Sigma-Aldrich and used without further purification. Artificial cerebrospinal fluid (ACSF) was prepared by dissolving the following materials in 1 L of deionized water: NaCI = 7.25 g, KCI = 0.22 g, NaHC03 = 2.18 g, CaCI2-2H20 = 0.29 g, KH2P04 = 0.17 g, MgS04-7H20 = 0.25 g, and D-glucose = 1.80 g. A literature value of 1.46 g/mL was used for the density of the NWs.

Isolation of Cellulose Nanowhiskers from Tunicates. TNWs were isolated from tunicates (Styela clava) collected from floating docks in Point View Marina (Narragansett, Rl). The TNWs were prepared by sulfuric acid hydrolysis of the cellulose pulp. The protocol was based on the method described in Favier et al., Macromolecules 1995, 28, 6365, utilized modifications as described in Shanmuganathan et al., ACS Appl. Mater. Interfaces 2010, 2, 165. The bleached tunicate mantles were blended at high speed, yielding a fine cellulose pulp. Sulfuric acid (95-97%, 600 mL) was slowly (over the course of 2 h) added under vigorous mechanical stirring to an ice-cooled suspension of tunicate cellulose pulp in deionized water (6 g in 600 mL, 20 C). After 500 mL of the acid had been added, the dispersion was removed from the ice bath and was heated to 40 °C during the addition of the final 100 mL of acid. After the acid addition was complete, the dispersion was heated to 60 °C and was kept at this temperature for 1 h under continuous stirring. The mixture was then cooled to room temperature, centrifuged (30 min at 3300 rpm), and the supernatant solution was decanted. Deionized water was added and the centrifugation step was repeated until the pH of the dispersion reached about 5. After the last centrifugation the resulting NWs were dialyzed in three successive 24 h treatments against deionized water to remove the last residues of the sulfuric acid. The suspension was diluted with deionized water (total volume 1 L) and sonicated for 18 h, before it was filtered through a No. 1 glass filter in order to remove any remaining aggregates. The concentration of the NWs in the final dispersion was determined gravimetrically to be ~3 mg/mL. This dispersion was freeze-dried using a VirTis BenchTop 2K XL lyophilizer with an initial temperature of 25 °C and a condenser temperature of -78 °C. The TNWs aerogel those produced was stored and used as needed.

Isolation of Cellulose Nanowhiskers from Cotton. CNWs were isolated from Whatman filter paper with minor modifications to a procedure published in Capadona, J. R. et al., Biomacromolecules. 2009, 10, 712. After sulfuric acid hydrolysis and dialysis treatment, the resulting dispersion was sonicated for 3 h and left to settle at room temperature for 18 h. The supernatant was then decanted off and the CNW dispersion was spray-dried using a Buchi Mini Spray Dryer (Model B-191 ) to yield dried CNWs as a white powder. The drying parameters were an inlet temperature of 1 10 C a flow rate of 4 mL/min, a nozzle airflow of 700 mL/min, an aspiration rate of 70%, and an outlet temperature of 60 C.

Preparation of PVOH/NW Nanocomposites. Lyophilized TNWs or spray-dried CNWs were dispersed in deionized water at a concentration of 5 mg/mL by sonicating for 10 h and 7 h, respectively. PVOH was dissolved in deionized water at a concentration of 50 mg/mL by stirring for 2 h at 90 °C. Nanocomposites comprising 4-16% v/v NWs were prepared by combining the appropriate amounts of the NW dispersion and PVOH solution to cast a film weighing 1 g. This mixture was stirred at room temperature for 30 min, followed by sonication for 30 min, and the resulting homogeneous mixtures were cast into Teflon Petri dishes of a diameter of 100 mm. The dishes were placed into an oven at 35 °C for 5 days to evaporate the water, and the resulting films were then further dried in the oven at 70 °C for 24 h. The films were compression-molded between spacers in a Carver laboratory press (1000 psi for 2 min, followed by an increase of pressure to 2000 psi for 15 min). Unless otherwise stated, PVOH/TNW films were compression- molded at 150 °C, and PVOH/CNW films were compression molded at 120 °C. Both types of films were allowed to cool to -70 C over ca. 90 min under the applied pressure to yield 70-1 10 pm thin nanocomposite films. For reference purposes, neat PVOH films were prepared in a similar manner by solution-casting and subsequent compression- molding at 120 C and 150 C.

Scanning Electron Microscopy (SEM). The morphology of the NWs was examined by scanning electron microscopy (SEM) using a FEI XL 30 SIRON FEG microscope. A droplet of dilute aqueous NW dispersions (0.1 mg/mL) were deposited on a silicon wafer (TEDPELLA, Inc.) and allowed to dry. Then the samples were coated with a layer of gold (~5 nm), and observed with an accelerating voltage of 5 kV.

Transmission Electron Microscopy (TEM). For TEM studies, 3 pL of dilute aqueous NW dispersions (0.1 mg/mL) were deposited on carbon-coated grids (Electron Microscopy Sciences) and allowed to dry. The samples were examined with a Philips CM100 Biomicroscope operated at an accelerating voltage of 80 kV. NW dimensions were determined by analyzing 10 TEM images of NWs with a total of more than 100 individual NWs of which length and width were measured. The dimensions thus determined are reported as average values ± standard error.

Atomic Force Microscopy (AFM). Atomic force microscopy was carried out on a NanoWizard II (JPK Instruments) microscope. 10 pL of dilute aqueous NW dispersions (0.1 mg/mL) were deposited onto freshly cleaved mica (SPI Supplies Division of Structure Probe, Inc.) and allowed to dry. The scans were performed in tapping mode in air using silicon cantilevers (NANO WORLD, TESPA-50) with a scan rate of 1 line/sec.

Conductometric Titration. Conductometric titrations were performed to quantify the surface charges of NWs. 50 mg of the NWs were suspended into 10-15 mL of aqueous 0.01 M hydrochloric acid. After 5 min of stirring and 30 min of sonication, the suspensions were titrated with 0.01 M NaOH. The titration curves show the presence of a strong acid, corresponding to the excess of HCI, and a weak acid corresponding to the sulfate-ester surface groups (see Fig. 7 A to 7C). Swelling Behavior. Prior to mechanical testing of ACSF-swollen samples, the degree of swelling was determined by measuring the weight of the samples pre- and post-swelling:

Mass of wet sample - Mass of dry sample

Desree of swelling (%) = ^χ 100

" ^' Mass or dry sample ^ ^

To minimize the error in measuring the degree of swelling, once the wet samples were taken out of the ACSF, they were placed on paper tissue to wick any excess ACSF from the surface; the samples were then immediately weighed.

Dynamic Mechanical Analysis (DMA). Mechanical properties of the PVOH/NW nanocomposites were characterized by dynamic mechanical analysis (DMA) using a TA instruments Model Q800. Tests were conducted in tensile mode using a temperature sweep method (0-140 C) at a fixed frequency of 1 Hz, a strain amplitude of 30 m, a heating rate of 5 ^c C/min, and a gap distance between jaws of -10 mm. The samples were prepared by cutting strips from the films with a width of ~6 mm. To determine the mechanical properties of the films in the wet state, the samples were swelled in ACSF at 37 C for periods of 1 week and 1 month. After the degree of swelling had been measured, DMA experiments were conducted in tensile mode with a submersion clamp, which allowed measurements while the samples were immersed in ACSF. In this case, the temperature sweeps were done in the range of 23-75 C with a heating rate of 1 °C/min, a constant frequency of 1 Hz, a strain amplitude of 30 pm, and a fixed gap distance between jaws of 15 mm. Differential Scanning Calorimetry (DSC). Differential scanning calorimetry experiments were carried out with a Mettler Toledo STAR instrument under N2 atmosphere. The typical procedure included heating and cooling cycles of approximately 10 mg sample in a DSC pan from -50 to 250 °C using a heating rate of 10 C/min. The glass transition temperature (T_g) was determined from the midpoint of the specific heat increment at the glass-rubber transition, while the melting temperature (T_m) was taken as the peak temperature of the melting endotherm. RESULTS AND DISCUSSION

Isolation and Physical Properties of Cellulose Nanowhiskers The NWs used in this study were isolated from tunicates (TNWs) and cotton (CNWs) by sulfuric acid hydrolysis, using protocols that represent modified versions of well- established methods. In the case of CNWs, spray-drying was used to isolate the dry NWs (see Experimental Section). Polymer nanocomposites with TNWs have consistently been shown to exhibit superior mechanical properties than those with CNWs, a fact that is mainly credited to their higher aspect ratio (-70 vs. -10) and on-axis stiffness (tensile modulus -143 vs. -105 GPa). CNWs, on the other hand, are more viable for commercial exploitation because they are isolated from an abundant and sustainable bio-source. Due to their high density of strongly interacting surface hydroxyl groups, NWs have a strong tendency for self-association.30.43, 46 Atomic force and electron microscopy of the NWs confirm that re-dispersion of the dried materials in water is readily possible (Fig. 1 , and Fig. 8A to 8D). The dimensions of the TNWs, determined from TEM micrographs, were an average length and width of 2500 ± 1000 nm and 30 ± 5 nm, respectively. The average aspect ratio (A, defined as length to diameter (width) ratio, l/w) of the TNWs is therefore 83. The charge density of negatively charged sulphate esters on the NW surface that are introduced during hydrolysis has been suggested to modulate NW-NW interactions and to affect their dispersability. By conductometric titration, the density of sulfate groups of the present TNWs was determined to be -90 mmol/kg (Fig. 7A).

The CNWs used here were measured to have an average length and width of 220 ± 70 nm and 22 ± 6 nm, respectively, resulting in an aspect ratio of about 10. In addition to exhibiting a lower aspect ratio than the TNWs, the charge density on the surface of CNWs (-40 mmol/kg, see Fig. 7B and 7C), is significantly lower than that of TNWs. While TNWs were dried and isolated by lyophilization, spray-drying was used for CNWs. In order to assess any influence of the drying process on the physical properties of the CNWs, one batch of as-prepared CNWs was, after dialysis and sonication, split into two portions, which were dried by lyophilization and spray drying, respectively. TEM and conductometric titration data suggest that the drying method has no influence on the physical dimensions of the CNWs or on their surface charge density and morphology (Fig. 7 A to 7C and Fig. 8A to 8D).

Nanocomposite Processing PVOH solutions and NW dispersions were combined, and after solution-casting and evaporation of solvent, the resulting films were re-shaped by compression-molding to result films of the nanocomposite with 4-16% v/v NWs and a thickness of 70-100 iim. Due to the limited thermal stability of the nanocomposites above the melting temperature (T_m) of PVOH (-220 C), the films were compressed at a temperature much below T_m. PVOH/TNW nanocomposites were compression-molded at 150 °C without any visible color changes, while PVOH/CNW nanocomposites yellowed, when processed at this temperature (Fig. 9A, 9B). As a consequence, PVOH/CNW nanocomposites were processed at 120 ' C, unless otherwise noted. Indeed, several explanations have been proposed in the literature regarding the degradation of NWs. The differences in thermal degradation may possibly arise from a combination of effects that including differences in surface chemistry and charge density, which are related to the source of NWs.

Thermal Properties

The thermal properties of PVOH/NW nanocomposites were determined using differential scanning calorimetry (DSC, Table 1 ). The DSC curves (Fig. 10) show that the T_g (68 and 71 °C) and T_m (216 and 206 °C) of the neat PVOH only slightly depends on the temperature at which the films were compression-molded. In both cases, the incorporation of NWs led to an increase of T_g by approximately 10 ' C, which interestingly was independent of the NW content. Upon introduction of NWs the width of the melting peak increases, and the degree of crystallinity (χ₀) increases slightly, perhaps on account of a small nucleation effect of the NWs. Also this effect was largely independent of the NW content.

Mechanical Properties of Dry PVOH/NW Nanocomposites

The mechanical properties of the nanocomposites were established using dynamic mechanical analysis (DMA, Table 2). Fig. 2A shows the tensile storage moduli (Ε') of the PVOH/TNW nanocomposites and a neat PVOH reference film in the dry state as a function of temperature. &AI room temperature (25 C), the neat PVOH matrix, processed at 150 °C exhibits an E^' of 7.3 GPa. Upon increasing the temperature, E' drastically decreases to 840 MPa at 100 °C (~T_g + 30 °C) due to a transition from the glassy to the rubbery regime at -70 °C, which is seen as a maximum in the tan δ curves (Fig. 2B). PVOH TNW nanocomposites containing 4 to 16 % v/v TNWs showed a significant increase in E^' compared to the neat matrix below and above the T_g. At 25 C, E^' increased from 7.3 GPa (neat PVOH) to 13.7 GPa for the nanocomposite containing 16% v/v TNWs (Fig. 2 A and Table 2). A more significant reinforcement was observed above T_g. At 100 °C, the nanocomposite containing 16% v/v TNWs shows an E^' of 5.4 GPa, which represents a seven-fold increase over the stiffness of the neat PVOH (840 MPa) at this temperature.

The E^' of nanocomposites prepared with CNWs exhibits a similar trend as observed for the TNW nanocomposites, although the stiffness increase was more modest. As discussed above, PVOH/CNW nanocomposite films yellowed upon compression molding at 150 C and were thus processed at 120 °C. At all temperatures, the E^' of dry PVOH reference films, processed at 120 °C, was found to be slightly lower than that of the neat PVOH processed at 150 °C (Fig. 2A, 2C, and Table 2). For example, at 25 C E' values of 7.3 and 7.0 GPa were measured. PVOH/CNW nanocomposite with 16% v/v CNWs exhibited a E^! of 9.0 GPa at 25 °C, which is higher than that of the neat PVOH (7.0 GPa), but lower than the E^' of 13.7 GPa of the TNW nanocomposite with the same NW content. Above T_g (at 100 °C) E' of this nanocomposite was 1 .4 GPa, which represent a two-fold enhancement over the stiffness of the neat polymer (Fig. 2C and Table 2). The lower reinforcement displayed by the CNWs compared to the TNWs is consistent with previous findings and is attributed at least in part to lower aspect ratio and stiffness of CNWs. Fig. 2B and 2D display the loss factor (tan δ) curves of neat PVOH and the two series of nanocomposites as a function of temperature. Tan δ is the ratio of loss modulus to storage modulus E^'VE^' of the material and is indicative of its damping behavior. All curves show a single relaxation peak centered at T,„ which corresponds to the T_g determined by DSC. The introduction of NWs led to a reduction in peak intensity and a shift of T„_. to higher temperature compared to the neat PVOH films. This trend is attributed to the reduced mobility of PVOH chains in the amorphous phase due to the presence of the NWs.

Table 3 shows a comparison of the E' values of the dry PVOH/NW nanocomposites studied here with previously reported mechanically-adaptive nanocomposites based on a range of polymer matrices. The data are quoted for a filler content of -16% v/v. The comparison shows that the stiffness of the present PVOH/TNW nanocomposites is, at 25 °C as well as T_g + 30 °C, more than three times higher than that of the stiff est mechanically-adaptive nanocomposites reported to date. This implies that a good dispersion of the NWs has been achieved in the PVOH matrix and supports the conclusion that strong H-bonding interactions between the NWs and the polymer matrix indeed increase the reinforcing effect of the cellulose. This is illustrated in Fig. 6 A and 6B. As shown in Fig. 6A, a polymer nanocomposite 1 comprises a matrix polymer 2, in the particular embodiment described herein PVOH, and a nanofiber network formed by a substantially three-dimensional network of nanofibers 3, in the particular embodiment described herein tunicate nanowhiskers (TNWs) or cotton nanowhiskers (CNWs). The matrix polymer comprises crystalline regions 20 and amorphous regions 21. As shown in Fig. 6B, the matrix polymer 2, in this case PVOH, is capable of forming hydrogen bonds with the nanofibers 3 (cellulose nanowhiskers NWs, in this case tunicate nanowhiskers TNWs or cotton nanowhiskers CNWs). Such hydrogen bonding (H-bonding) interactions between the NWs and the matrix polymer 2 are believed to have a reinforcing effect of the polymer nanocomposite 1 at least in the dry state of the polymer nanocomposite 1 (first switching state). Upon subjection to a stimulus, for example upon exposure to a water-containing composition, such hydrogen bonds at least partially are released, yielding a reduction in stiffness in the soft state (second switching state) of the nanocomposite 1 .

The mechanical reinforcement in optimally assembled NW nanocomposites is caused by the formation of a percolating NW network, in which stress transfer is facilitated by hydrogen bonding between the NWs. The stiffness of these materials can be described by a percolation model that has been successfully used to predict the mechanical behavior of heterogeneous materials, such as polymer blends and nanocomposites. Detailed information about the percolation model and its use for modeling NWs-based nanocomposites has been reported for example in Samir, M. A. et al., Biomacromolecules, 2005, 6, 612. Fig. 3A and 3B show the predictions for the two nanocomposites series made by the percolation model, along with experimentally determined values of dry PVOH/NW nanocomposites at 100 C, i.e., at ~T_g + 30 °C. For the calculations, aspect ratios (A) of 83 and 10 (as determined by TEM) were used for TNWs and CNWs, respectively, and storage moduli E'_s of 840 MPa (for TNW nanocomposites processed at 150 C) and 700 MPa (for CNW nanocomposites processed at 120 C) were employed for the neat polymer matrix at 100 C (as determined by DMA). The tensile storage modulus of the NW phase, E^' _r, was in previous studies derived by either measuring the stiffness of a neat TNW or CNW film or by fitting the model against the experimentally determined properties of the nanocomposites and using E^' _r as a fit parameter. While the morphology (and therewith the stiffness) of a neat NW film depends strongly on the processing conditions and has little resemblance to that of a NW network within a polymer matrix, the E'_r values of 5 - 24 GPa for TNW-based, and 0.6 - 5 GPa for CNW-based nanocomposites determined by these approaches appeared to roughly match. Interestingly, E^' _r values of 80 GPa and 10 GPa are required to fit the model to the data for the TNW-based and CNW-based nanocomposites with PVOH studied here (Fig. 3A and 3B). A comparison of the data for several other TNW- based nanocomposites shows that for a given NW content, E' increases with the polarity of the polymer matrix (PS<PVAc<Epoxy), suggesting that systems with pronounced NW- polymer interactions may exhibit larger reinforcement due to factors that are not explicitly accounted for in the percolation model.

Swelling Behavior

The swelling behavior of the nanocomposites in physiological conditions was investigated by immersing the materials into artificial cerebrospinal fluid (ACSF) at 37 C to mimic physiological conditions. It is known that heat-treated PVOH is no longer water soluble, and that the processing temperature of PVOH affects the permeability of the material and thereby the potential for water uptake. Indeed, the swelling characteristics of the materials studied here were found to be strongly dependent on the temperature used for compression molding, but not the type or content of NWs. Neat PVOH and PVOH/TNW nanocomposite films processed at 150 °C exhibit -40% w/w swelling, whereas neat PVOH films and PVOH/CNW nanocomposites processed at 120 °C exhibit approximately -120% w/w swelling (Fig. 4), regardless of the NW content. The conclusion that the processing temperature is the primary factor for the swelling behavior was further confirmed by swelling PVOH/CNW nanocomposites, which were processed at 150 C, and PVOH TNW nanocomposites, which were processed at 120 °C (Fig. 4). Compared to the other nanocomposites, these samples exhibited swelling that was consistent with their processing temperature rather than the type of NW. Swelling experiments in ACSF were extended over the course of two months to investigate the possible changes that might occur during prolonged biological implantation of the material (Table 4). The data show that for the PVOH/TNW nanocomposites processed at 150 C equilibrium swelling is reached within 24 hours and that these materials do not degrade over the course of two months. Similarly, PVOH/CNW nanocomposite films reached an equilibrated swelling within 24 hours and maintained their integrity for at least one week. A comparison of the swelling data of the present PVOH/TNW and the previously investigated PVAc/TNW nanocomposites29 shows that the PVOH-based nanocomposites swell much less than their PVAc-based counterparts. For instance, the PVAc TNW nanocomposites comprising 16.5% v/v TNWs displayed a degree of swelling of -80%, while the PVOH/TNW nanocomposites shows -40% w/w of swelling with 16% v/v NWs. Mechanical Properties of ACSF-Swo!ien Nanocomposites

The mechanical properties of ACSF-swollen PVOH/NW nanocomposites were determined by DMA using a submersion clamp set-up, which allowed the samples to be immersed in ACSF during the measurements. Neat PVOH films softened substantially upon submersion in ACSF for one week and exhibited mechanical properties that appear to be correlated with their swelling behavior. Neat PVOH films processed at 150 °C displayed a change in E^' from -7.3 GPa (dry) to -1 1 MPa (ACSF-swollen), whereas E^' for neat PVOH films processed at 120 °C changed from 7.0 GPa to ~1 MPa. The ACSF- swollen PVOH TNW nanocomposites display an E' that is higher than of the neat PVOH films (Fig. 5 and Fig. 10), but considerably lower than that of the corresponding materials in the dry state. For example, E^' of the material comprising 16% v/v TNWs dropped from 13.7 GPa (dry, RT) to -160 MPa (ACSF-swollen at 37 °C). The data in Fig. 3 and Table 2 show that the switching is reversible and that ACSF exposure for 1 week and 1 month has the same effect. Moreover, the data in Fig. 3A show that the relation between E' of the ACSF-swollen PVOH TNW nanocomposites and the TNW content is fairly well described by the Halpin-Kardos model, whose application to the modeling of NWs-based nanocomposites has been reported for example by Shanmuganathan, K. et al., J. Mater. Chem., 2010, 20, 180. Parameters used for the modeling were taken from Hajji, P. et al., Polym. Compos., 1996, 17, 612, except E^' _s of 1 1 MPa and 1 .44 MPa for materials processed at 150 and 120 °C, respectively, which were measured by DMA. The fact that the model underestimates E' suggests that the NW-NW and perhaps also NW-matrix interactions are reduced upon swelling with ACSF, but - perhaps on account to the abundance of hydroxyl groups on the matrix polymer and the NWs that can interact with water - not entirely switched off. All of the ACSF-swollen samples show a significant drop of E^! at -60 °C. Since this temperature is below the T_g of the plasticized PVOH and far below the T_m of the matrix (-220 C), this transition may relate to the dissolution of the matrix. The ACSF-swollen PVOH and PVOH/CNW films, which were processed at 120 °C and display considerably more swelling than the above-discussed TNW-based materials, exhibit an E' of 1 - 4 MPa at 37 C (Fig. 3B and 5B, Table 2). Due to the rather low modulus, these measurements feature a significant error, and it is therefore not possible to draw a firm conclusion about how the nanocomposite composition affects the modulus of the ACSF-swollen materials. The fact that the ACSF-induced mechanical switching was very significant and the finding that similar values were observed for both the neat PVOH and the nanocomposites suggest that the mechanical properties of these materials were largely governed by the swollen polymer matrix, supporting the notion that the water effectively reduced hydrogen bonding between the NWs.

In order to further probe the influence of the processing temperature on the mechanical switching, the tensile storage moduli (Ε^') of nanocomposites with 16% v/v TNWs that had been compression-molded at 120 C and 150 C were compared (Table 2). It was found that the E' of PVOH TNW films processed at 150 C, switched from 13.7 GPa (dry, RT) to 164 MPa (ACSF-swollen at 37 C), whereas the material processed at 120 °C switched from 12.3 GPa to 60 MPa (Fig. 1 1 ). A similar result, was observed for the ACSF-swollen PVOH/CNW films, which if processed at 150 °C exhibited an E^' of 13 MPa at 37 °C, while the same composition processed at 120 °C, showed an E' of 2 MPa at 37 °C (Fig. 12). A lower processing temperature therefore not only influences the mechanical properties of the materials in the dry state, but also reduces E^' of the ACSF- swollen state considerably. While the PVOH/CNW films processed at 150 °C show some yellowing, their dry-state mechanical properties are largely comparable to those of the material processed at 120 °C, and (on account of less swelling) a higher modulus in the soft state is achieved, suggesting that whatever effect is responsible for the yellowing, it is not negatively impacting the material^'s mechanical characteristics. Thus, it appears that if PVOH is used as a matrix, not only the NW type and content but also the processing temperature can be used to tailor the mechanical contrast of cellulose NWs- based, water-responsive, mechanically adaptive nanocomposites.

In summary of the experimental results and discussion given above, water-responsive, mechanically-adaptive nanocomposites based on polymer nanocomposites described herein, in particular such nanocomposites using PVOH as matrix polymer and TNWs or CNWs as nanofibers, may offer an initial stiffness that is significantly higher than that of previous generations of such responsive materials.

In particular the use of PVOH as a matrix polymer into which nanofibers such as NWs are incorporated may yield tensile storage moduli of PVOH/NW nanocomposites which - in both the glassy and rubbery regime - in the first switching state are significantly higher than those of comparable, other nanocomposites. It appears that in addition to NW-NW interactions, polymer-NW interactions, which are promoted by the strong propensity of PVOH to form hydrogen bonds, may be a factor in this context. Another factor is the possibility of controlling the swelling characteristics of the PVOH matrix, and therewith the properties of water- or ACSF-swollen nanocomposites, via the processing conditions. By such means, the change in stiffness upon switching between the switching states of the TNW-based nanocomposites upon exposure to ACSF could be varied between a 90- fold to a 200-fold modulus change. By fine-tuning the process conditions a desired modulus change upon exposure to a suitable stimulus may be set.

Although not as stiff initially in the dry state compared to PVOH/TNW, PVOH/CNW nanocomposites exhibit a larger mechanical contrast (up to 900-fold), as they soften more than TNW-based nanocomposites. This effect possibly is related to the lower reinforcing power of CNWs. Despite the fact that decomposition of the CNW-containing materials starts around 150 C, the nanocomposites maintain useful mechanical properties. Therefore, one could envision employing a higher processing temperature for PVOH/CNW nanocomposites to further increase the stiffness of the water-swollen state.

The invention is not limited to the embodiments described above but may be carried out also by entirely different embodiments. In particular, other matrix polymer materials such as polyamide may be used rather than PVOH. Further, different nanofibers other than tunicate nanowhiskers (TNW) or cotton nanowhiskers (CNW), for example cellulose nanowhiskers not derived from tunicates or cotton, may be employed.

List of Reference Numerals

1 Composite

2 Matrix polymer

20 Crystalline region

21 Amorphous region

3 Nanofibers (cellulose nanowhiskers)

Claims

Claims:

A polymer nanocomposite (1 ) having switchable mechanical properties, comprising: a matrix polymer (2) and

a nanoparticle network, wherein the nanoparticle network is formed by a formation of a substantially three-dimensional network of nanoparticles (3) which are incorporated in the matrix polymer (2) and interact with each other and/or with the matrix polymer (2),

wherein the polymer nanocomposite (1 ) in a first switching state comprises a first stiffness characterized by a first tensile storage modulus (E ) greater than 6 GPa, in a second switching state comprises a second stiffness characterized by a second tensile storage modulus (Ε') of less than 1 GPa and is switchable between the first switching state and the second switching state by exposing the polymer nanocomposite (1 ) to a stimulus that influences interactions among the nanoparticles (3) and/or between the nanoparticles (3) and the matrix polymer (2).

The polymer nanocomposite (1 ) of claim 1 , obtainable by applying a process comprising the steps of :

providing the matrix polymer (2),

incorporating the nanoparticle network into the matrix polymer (2) to obtain an intermediate mixture,

subjecting the intermediate mixture to a heat treatment step involving exposure to a temperature equal to or greater than 100 C.

The polymer nanocomposite (1 ) of claim 2, wherein the heat treatment step includes compression molding the polymer nanocomposite (1 ) at a temperature equal to or greater than 100 °C.

The polymer nanocomposite (1 ) of claim 2 or 3,

wherein, prior to the heat treatment step, the intermediate mixture is subjected to a drying step involving exposure to an elevated temperature of 30 °C to 80 C over an elongated period of time of 1 day to 10 days and/or

wherein the drying step includes a first drying step including exposure to a first drying temperature of 30 C to 50 C for example 35 °C, over a first period of time of 1 day to 10 days, for example 5 days, and

a second drying step including exposure to a second drying temperature of 50 °C to 80 °C, for example 70 °C, over a second period of time of 5h to 48h, for example 24h.

The polymer nanocomposite (1 ) of one of the preceding claims, wherein the matrix polymer (2) comprises a polymer capable of forming non-covalent interactions, in particular hydrogen bonds, with the nanoparticles (3).

The polymer nanocomposite (1 ) of one of the preceding claims, wherein said matrix polymer (2) comprises at least one of vinyl polymer such as polyvinyl alcohol, poly(acrylic acid), poly(acryl amide), polyvinyl pyridine), a copolymer of at least one of such monomers, a polycondensate such as polyamide or polyester, a polysaccharide such as cellulose, starch, alginates, pectins, hyaluronane, chitin, chitosan or at least one of its derivatives, and/or a protein.

The polymer nanocomposite (1 ) of one of the preceding claims, wherein the nanoparticles (3) are capable of forming non-covalent interactions, in particular hydrogen bonds, with each other and/or with the matrix polymer (2).

The polymer nanocomposite (1 ) of one of the preceding claims, wherein the nanoparticles comprise nanofibers.

The polymer nanocomposite (1 ) of claim 8, wherein the nanofibers (3) comprise cellulose nanowhiskers, preferably wherein the nanofibers (3) comprise cellulose nanowhiskers derived from tunicates or cotton.

10. The polymer nanocomposite (1 ) of one of claims 8 to 9, wherein the nanofibers (3) comprise an aspect ratio greater than 5, preferably greater than 10, more preferably greater than 50, most preferably greater than 80.

The polymer nanocomposite (1 ) of one of the preceding claims, wherein the polymer nanocomposite (1 ) comprises a content of 1 % to 30 %, preferably 2 % to 25 %, more preferably 3 % to 20 % v/v of nanoparticles (3).

12. The polymer nanocomposite (1 ) of one of the preceding claims, wherein the polymer nanocomposite (1 ) is switchable from the first switching state to the second switching state by a stimulus which reduces the interactions among the nanoparticles (3) and/or between the nanoparticles (3) and the matrix polymer (2), and the polymer nanocomposite (1 ) is switchable from the second switching state to the first switching state by a stimulus which enhances the interactions among the nanoparticles (3) and/or between the nanoparticles (3) and the matrix polymer (2).

13. The polymer nanocomposite (1 ) of one of the preceding claims, wherein the polymer nanocomposite (1 ) is switchable from the first switching state to the second switching state by subjecting the polymer nanocomposite (1 ) to a water-containing composition and/or

wherein the polymer nanocomposite (1 ) is switchable by subjecting the polymer nanocomposite (1 ) to a composition at an elevated temperature between 30

°C and 50 C, preferably 35 °C to 40 °C.

14. The polymer nanocomposite (1 ) of one of the preceding claims, wherein the ratio of the storage moduli of the two switching states is more than 20, preferably more than 50, more preferably more than 100, and most preferably more than 500 and/or

wherein the polymer nanocomposite (3) displays essentially isotropic mechanical properties.

15. A method for producing a polymer nanocomposite (1 ) having switchable mechanical properties, comprising the steps of:

providing a matrix polymer (2) and

incorporating a nanoparticle network into the matrix polymer (2), wherein the nanoparticle network is formed by a formation of a substantially three- dimensional network of nanoparticles (3) which are incorporated in the matrix polymer (2) and interact with each other and/or with the matrix polymer (2), to obtain an intermediate mixture,

subjecting the intermediate mixture to a heat treatment step involving exposure to a temperature equal to or greater than 100 °C.

16. The method according to claim 18, wherein a polymer nanocomposite (1 ) which in a first switching state comprises a first stiffness characterized by a first tensile storage modulus (E ) greater than 6 GPa, in a second switching state comprises a second stiffness characterized by a second tensile storage modulus (Ε') of less than 1 GPa and is switchable between the first switching state and the second switching state by exposing the polymer nanocomposite (1 ) to a stimulus that influences interactions among the nanoparticles (3) and/or between the nanoparticles (3) and the matrix polymer (2) is obtained.

17. A method for inducing a stiffness change in a polymer nanocomposite having switchable mechanical properties, the method comprising the steps of:

providing a matrix polymer (2) incorporating a nanoparticle network, wherein the nanoparticle network is formed by a formation of a substantially three- dimensional network of nanoparticles (3) which are incorporated in the matrix polymer (2) and interact with each other and/or with the matrix polymer (2), switching the polymer nanocomposite (1 ) between a first switching state, in which the polymer nanocomposite (1 ) comprises a first stiffness characterized by a first tensile storage modulus (Ε') greater than 6 GPa, and a second switching state, in which the polymer nanocomposite (1 ) comprises a second stiffness characterized by a second tensile storage modulus (Ε') of less than 1 GPa, by exposing the polymer nanocomposite (1 ) to a stimulus that influences interactions among the nanoparticles (3) and/or between the nanoparticles (3) and the matrix polymer (2).