Bionanocomposites are novel class of nanomaterials that consist of a constituent of biological origin and inorganic nanoparticles with at least one of dimensions in the range of 1–100 nm [1–3]. Nowadays they receive more and more attention because of an opportunity to be an alternative to the synthetic polymers. Plastics that are made of them are manufactured from fossil resources. Petroleum, natural gas and coal are not renewable and exhaustible, being finite in quantity. A problem is that about 40 % of plastics are used for packaging [4, 5]. Because they are poor (bio)degradable, the humankind is now being faced with a wealth of waste increasing every year.
Biopolymers, as believed, can help to solve the problems of fossil resources shortening and man-made plastics pollutants. There are their sustainable, renewable sources and, what is also important, they are biocompatible and biodegradable. Nevertheless, substitution of biopolymers for synthetic polymers meets with their properties inadequate for many applications. They have lower mechanical strength, heat degradation temperature, higher gas permeability, etc. than plastics. Therefore, it is much required to reinforce and improve properties of biopolymer materials. Inorganic fillers are frequently added. Researchers from the Toyota company demonstrated by using natural clay montmorillonite in the early 1990s that it reinforced notably, improved dimensional stability, water and gas barrier properties of naylon-6 at amount of few percentages when was properly exfoliated into individual nanoparticles [6–8]. The same effects occur, as shown by many examples in reviews [1, 2], for composites made of biopolymers. According to the classification suggested in , composites consisting of a constituent of biological origin and inorganic nanoparticles were classed as bionanocomposites, of synthetic polymers and nanoparticles, as nanocomposites, while biocomposites and composites, respectively, include filler of which dimension stands out above the nanosized particles.
Chitosan is an extraordinary biopolymer of significant versatility and promise. This cationic polysaccharide is produced by deacetylation of chitin (Fig. 1) that is at the second place among the abundant organic compounds on earth after the cellulose [9–13]. Although there are as much as 100 billion tones of chitin renewed annually in living nature that is a little smaller than cellulose [13, 14], it is still almost underexploited [12, 13, 15–18]. Serious efforts are made to develop bionanocomposites with chitosan as a building block. Because it is mainly performed through aqueous solutions, one meets with a severe problem of high sensitivity of this cationic polysaccharide to the presence of anionic substances and nanoparticles in its solutions. Even their trace amount can cause its precipitation and formation of heterogeneous material with inadequate properties [2, 19–22].
This disadvantage was turned into virtue in our recent novel approach in which chitosan electrostatic interactions with anionic substances and nanoparticles were regulated via its charging [21, 23, 24]. As a weak polyelectrolyte, it can be in charged and non-charged forms (Fig. 1) [10, 14, 18]. It means that its charging can be manipulated by a straightforward change of the pH value.
Our first studies have been performed with synthetic saponite belonging to the smectite family of clays [23, 25, 26]. It was demonstrated that the formation of bionanocomposites proceeded in a self-organizing manner [2, 26]. When it took place in a solution, one could see its jellification owing to a three-dimensional fibrillar network generated by chitosan with saponite [23, 25], whereas films had two-level hierarchical stratified structure from micrometer-sized plates made up of alternating layers of nanoplatelets and polysaccharide macromolecules .
Our further investigation with various types of nanoparticles has demonstrated that the developed approach may be considered as a general method applicable for the fabrication of bionanocomposites where they are made from chitosan. Here these results are presented.
Materials and methods
High-molecular-weight chitosan with a degree of deacetylation of ca. 85 % was obtained from Sigma-Aldrich, xanthan, from Fluka and D-glucono-δ-lactone (GL), from Merk. Glycerol (Russia) was of analytical grade. Pristine sepiolite with a 99 % content of pure silicate from Yunclillos (Toledo, Spain) was gifted by Tolsa S. A. (Spain), synthetic saponite SKS-20 with a composition of (Si4−xAlx)IV(Mg3)VIO10(OH)2xM nH2O, where M is Na+ or К+, from Clariant (Germany). Multiwalled carbon nanotubes CM-95 (MWCNTs) of 95 % purity were purchased from Il Jin Nanotech (Korea). Distilled or deionized water was used to prepare solutions.
Synthetic saponite was easily dispersed into individual clay nanoparticles in water. Their transparent solution was formed shortly after adding a weighted amount in water and stirring in magnetic stirrer. A representative image two of them may be seen in Fig. 2A. One may see two platelets overlapping one another only partially when a solution drop was dried on the grid. Saponite nanoparticles were of oval or oblong shapes of which size was in the range of 20–30 nm, while thickness was ca. 1 nm. Each of them consists of two continuous tetrahedral sheets separated by an octahedral one. This 2:1 layer structure of saponite nanoplatelet is shown in Fig. 2B.
Dispersion of sepiolite needed great efforts. It was made without the addition of surfactants that could adsorb on clay surface, modifying it. To disintegrate bundles of fibers, water with weighted amount of pristine sepiolite was initially treated in an ultrasonic bath for 40 min. In the next day, the swollen precipitate was dispersed by means of an ultrasonic dispenser Sonopuls HD 2070 (Bandelin, Germany). Examination under the SEM showed the presence of rather well disaggregated sepiolite nanofibrils (Fig. 2C). One may see long, narrow strips of rectangular cuboids that are typical of this kind of clay mineral. Sepiolite (Si12O30Mg8(OH,F)4(H2O)4×8H2O [27, 28]) is also the 2:1 phyllosilicate like saponite but structurally differs from it. Pristine sepiolite is built, as shown in Fig. 2D, by continuous two-dimensional tetrahedral silica sheets separated an octahedral magnesium oxide/hydrohyde sheets that is not continuous. These three-layer blocks with tetrahedral apices pointing in the opposite directions are stacked in such a manner that voids are formed. They are called tunnels of ca. 0.4 × 1.1 nm2. Sepiolite fibrils looks like ribbons (Fig. 2C), of which cross-section has a cuboid shape, being in size ranging from 10 to 30 nm in width, 5–10 nm in thickness and 2–5 μm long [27, 28].
MWCNTs were oxidized through a one-step hydrothermal treatment in 2 % nitric acid. Their weighted 200 mg in 200 g of water with HNO3 were sealed into a Teflon equipped stainless steel autoclave, which was placed in a drying oven at 180 °C for 12 h. To separate the oxidized MWCNTs, reaction mixture was centrifuged at Supra 30 K centrifuge. The precipitate was washed 3× with water through centrifugation/redispersion cycles, then with methanol and finally acetone. After drying at 40 °C, the MWCNTs were dispersed in water, giving a stable dispersion. A TEM image is presented in Fig. 2E. The oxidation was confirmed by Fourier transform infrared spectroscopy (Fig. 2F). It was inferred from an increased vibrational mode of O–H stretching found at 3440 cm−1 that hydroxyl groups were mainly introduced.
Hydrogel preparation was followed the procedure developed in [23, 25]. Briefly, ground fine microparticles of chitosan were added into an initially prepared solution of nanoparticle dispersion. To distribute them over the entire solution volume, the mixer was vigorously stirred. The pH was ≥6.3 of the pK value of chitosan to have polysaccharide in the uncharged state. The solution was acidified by an appropriate weighted amount of GL introduced under the vigorous stirring. After the GL solubilization, it was left without stirring. The mixture was shortly stirred once again at the first signs of settlement of chitosan microparticles. It was repeated as just needed to hold the solution in homogeneous state.
In order to construct a phase diagram for the chitosan-saponite-water system, these three components were combined with broad variation of their weighted amount in mixtures. GL was added to transfer polysaccharide into the charged state. After the preparation of samples in accordance with the procedure described in the above section, vials were left for 7–10 days at ambient conditions. They were checked regularly for the phase separation over this time period. This observations served to determine the phase boundaries.
To fabricate free-standing film, mixture with GL and homogeneously distributed nanoparticles and chitosan microparticles was prepared as described above. The difference was only in 2 wt% of glycerol added as a plasticizer. When the viscosity increased quite sufficiently to impede the phase separation, mixture was poured into a Teflon container. It was enclosed by Parafilm containing holes that regulated the solvent evaporation. The container was left in a thermostat at 40 °C up to complete drying.
To optimize the bionanocomposite formulation, concentrations of chitosan and nanoparticles were varied in broad range. They were started from an amount close to the zero. The most quantities were mainly around 2.5 wt%.
Scanning electron microscopy
Hydrogel samples for examination were prepared by means of supercritical drying in a Critical Point Dryer (BalTec CPD 030, UK). Before that, water was exchanged for ethanol and finally for acetone. Small pieces of dried hydrogel and films were placed on a copper objective table. Their upper parts were cut and a surface thus freshly prepared was coated with a nanometer thick gold layer. Observations were performed by a scanning electron microscope EVO 40 SEM (Carl Zeiss, Germany).
Transmission electron microscopy
Micrographs were taken using a Libra 200 (Carl Zeiss, Germany) or JEM-2010 (JEOL, Japan) transmission electron microscopes. They both provided the high resolution at the accelerating voltage of 200 kV. To prepare samples, a tiny piece was dispersed in ethanol by means of an ultrasonic dispenser. Its droplet was placed onto a copper grid coated with an amorphous polymer or carbon film and dried under a warm air stream.
Small-angle X-ray scattering
SAXS measurements were carried out with a S3-MicroPIX SAXS spectrometer (Hecus X-Ray Systems GmbH, Austria) equipped with a X-ray generator (XenoX GeniX, France) operating at 50 kV and 1 mA with CuKα radiation (λ = 0.154 nm). Argentum behenate CH3(CH2)20COOAg served as calibration standard.
For tensile measurements, film stripes were cut from specimens in the size of 5 mm × 25 mm. Both their ends were clipped with a gripper for soft materials a working distance between which was 10 mm. The tensile strength – strain dependencies were measured by using a Haake Mars III (Thermo Scientific) equipped with a cell for tensile measurements. Tensile loading was applied at the rate of 100 % extension per 60 min at 25.0 ± 0.5 °C. Average value and standard deviation out of three measurements were calculated for each sample.
Results and discussion
The developed approach shares a basic principle of the electrostatic association of oppositely charged counterparts. Charged nanoparticles in such associates can be considered as macromolecules . In that case, they may be treated in the context of a polyelectrolyte complexes theory [22, 30].
General procedure for preparing bionanocomposites lies in the mixing of chitosan solution with nanoparticle dispersion. It seems simple but rapid cooperative electrostatic interactions, as a rule, result in the formation of a heterogeneous precipitate (see, e.g., [31, 32]). For this reason, an additional treatment is used like an intensive ultrasonic irradiation. It provides only partial improvement of homogeneity along with deteriorative effects including the disruption of macromolecules.
Our method suggested in [21, 23] obviates the above-mentioned problems by relying on the property of weak polyelectrolytes to be in a charged or non-charged state. It depends on the pH of an aqueous solution . Chitosan macromolecule has numerous amino groups (Fig. 1) the pK value of which is equal to 6.3 . The preparation of bionanocomposites proceeds in two main stages: (1) dispersing of chitosan as microparticles in a solution of a dispersion of nanoparticles at pH slightly higher than 6.3 and (2) acidification of the mixture. Of prime importance in the method is the manner in which the pH is shifted. A simple addition of an acid, e.g., acetic one results in a rapid charging of a macromolecule, yielding a heterogeneous material like a precipitate. If one takes D-glucono-δ-lactone (GL) applied widely in the food industry , there is its slow hydrolysis with the generation of D-gluconic acid after the contact with water (Fig. 1). The used GL provides gradually the acidification of solution and the chitosan charging. Only then did we find out the formation of a uniform bionanocomposite. The reason has to do with the process proceeding in the self-organization regime [2, 26].
Monolithic bionanocomposite hydrogels were formed as soon as a critical concentration of chitosan and nanoparticles was achieved. Their area is indicated in a phase diagram of chitosan–saponite–water system in Fig. 3A. Area marked as “Phase Separation” means that precipitation was observed in a case of small concentrations of oppositely charged counterparts or the formation of hydrogel phase being in contact with water took place when composition was closer to the phase boundary shown by the orange line.
When examining an effect of nanoparticle shape on the formation of chitosan hydrogels, we could not prepare them after the exchange of saponite for the sepiolite in formulations. Precipitation was observed at any concentration ratio as polysaccharide became more and more charged. The absence of jellification may be ascribed to a difference in the surface charge density of nanoparticles. Cationic exchange capacity is only ca. 0.15 meq/g for sepiolite  that is much less smaller than that of saponite equal to about 0.9 meq/g . This result make it obvious that the charge density is very important factor in the formation of bionanocomposites by the suggested method.
The oxidized MWCNTs were not highly charging as well. Their oxidation resulted mainly in the appearance of hydroxyl groups (see Fig. 2). To provide the stabilization of films with chitosan via the electrostatic interactions, xanthan was additionally introduced. As shown in our article , they could be mixed homogeneously together with the help of our approach.
Worthy of mention is the versatility of the method. We have made good use of it to combine chitosan with various negatively charged polysaccharides [21, 24], clay nanoparticles [23, 25, 26], hydroxyapatite microparticles  and recently with graphen oxide and carbon nanotubes first considered here. It can be applied to fabricate hydrogels, freestanding films and coatings.
Structural features of chitosan bionanocomposites were strongly dependable on a hydrogel or a film was prepared . Jellification was caused by the formation of a three-dimensional network from fibrils. It was built in the media with large amount of water exceeding 95 wt%. Films are waterless materials. Their formation in the course of drying proceeded with stratification resulting in a layered structure.
An insight into the structural organization in chitosan hydrogel with saponite may be gained from Fig. 3B and C. Their compositions are marked by the red circles in the phase diagram 3a. One may see a three dimensional network in both the systems. It consists of fibrillary loops of various size. The structure looks like foam although any corresponding technique was not used. An increase of the density of three-dimensional network is apparent with increasing the concentration of jellifying components, which follows from comparison of the images in Fig. 3B and C. The network formation by chitosan with saponite is responsible for the jellification and generation of mechanically strong hydrogel. Details concerning rheological properties of bionanocomposites can be found in our previous works [23, 25]. Sepiolite, as mentioned above, did not form hydrogels owing to the small density of surface charges.
This type of bionanocomposites does not contain water. It was removed via the evaporation when a pre-jellified mixture was left for drying. This way of preparation led to conspicuous differences in the film structure. Fibrils were not found. Films, as seen from representative images in Fig. 4, composed of stacked layers. For sepiolite (Fig. 4B), one can easily recognize its nanofibrils aligned parallel to the layers. Individual saponite nanoplatelets are not an easy matter for the identification because of too small dimensions (Fig. 4A). Our previous study  showed that they had a similar alignment. Stratified structure of films composed of alternating nanoparticles and chitosan layers was related with a self-organizing process in systems. It was embodied also in the formation of a novel flower-like hierarchical structure found on the chitosan-saponite film surface. Their formation was not mentioned before. A set of three SEM pictures presented in Fig. 5 was taken at various magnification to have an impression about the morphology at various length scales. Flowers with the diameters about 10 μm (Fig. 5A) consist of plates of a rather uniform thickness of ca. 100 nm (Fig. 5C), whereas their length is as large as several microns (Fig. 5B).
An insight into the structural organization of bionanocomposite films was gained by means of a SAXS. One may see a plot in Fig. 6A presenting the logarithm of normalized intensity (I) versus the scattering vector (q) for three samples (curves 2–4) with the constant saponite concentration of 1.5 wt% and chitosan content varied in broad range from 0.02 to 2.4 wt%. SAXS pattern of a film made from only saponite is also shown by curve 1. When clay nanoparticles were combined with chitosan, there is an appearance of Bragg reflex at the same q ≈ 0.35 Å−1 in the three presented curves 2–4. Its position did not depend on the varied polysaccharide concentration in forming solutions. The Bragg reflections correspond to a d-spacing of 1.75 nm. This parameter characterizing a periodicity in the examined bionanocomposites is accounted for by stacks of clay nanoplatelets associated face-to-face with interlaying chitosan macromolecules. Their presence was confirmed with the help of TEM observations. An image of stack is given in Fig. 6B. An assessment of the distance between clay nanoparticles, which are only visible, gave it between 1.5 and 2.0 nm. This result is in a rather good agreement with the SAXS measurements. Detailed examination by using SAXS, wide angle X-ray diffraction, TEM, SEM and dynamic mechanical thermal analysis (article is under preparation) suggested a coplanar alignment of saponite nanoplatelets with two monolayers of chitosan macromolecules in the gap.
Chitosan possesses a good film-forming ability but its films are not sufficiently strong [2, 38, 39]. Their mechanical strength was improved drastically when nanoparticles were introduced by the developed method. Results obtained with sepiolite and oxidized MWCNT are presented in Fig. 7. There are the Young’s modulus (Fig. 7A and E) and the elongation εbreak, at which a film break was observed, plotted against their concentrations. Examples of how the E and εbreak values were determined for films made from chitosan alone and 1 wt% chitosan with 3.2 wt% sepiolite are shown in inserted Fig. 7C and D, respectively.
Films prepared from only chitosan in the regime of its gradual charging are rather elastic. The elongation at break approached to 100 % (Fig. 7C). With introducing the nanoparticles, there were changes in the mechanical behavior. Films could be assigned to materials with limited plasticity (Fig. 7D). This transformation led to drastic increase in the Young’s modulus running up to two orders of magnitude (Fig. 7A) but the elongation at break decreased notably (Fig. 7B) that was indicative of sacrificed toughness.
The drastic change of mechanical properties of chitosan films by introduced nanoparticles is caused by electrostatic interactions between them. They provide strong salt-like linkages of oppositely charged counterparts that make such structures highly stable [22, 40]. Chitosan macromolecules can associate with one another only through hydrogen bonding, whereas electrostatic repulsion between like charged groups brings about a destabilizing effect. Negatively charged nanoparticles not only compensate it, but introduce additional strong electrostatic linkages. Furthermore, clay nanoparticles bearing silanol groups on their surface can also form hydrogen bonds with chitosan macromolecules [28, 41]. Their combination with the electrostatic interactions reinforced the system, imparting stability and mechanical strength obvious from increased Young’s modulus (Fig. 7A). At the same time the strong association with nanoparticles hinder markedly the mobility of macromolecules , thus resulting in a decrease of the film elasticity as followed from the data for the elongation at break (Fig. 7B).
It should be mentioned that bionanocomposites included glycerol as plasticizer. They were formed rather brittle without its addition. Some details concerning its plasticizing effect may be found in our previous article .
Introduced MWCNTs had also significant impact on the mechanical properties seen from Fig. 7E and F. There is a similar increase of E and a decrease of εbreak. It should be pointed out that the MWCNTs influenced the measured parameters at smaller concentration than sepiolite (Fig. 7A and B) a difference between which run into an order of magnitude. The effect can be related to the well-known extraordinary properties of CNTs, first of all to their remarkable tensile strength . Oxidized MWCNTs could integrate into polysaccharide films owing to hydroxyl groups introduced in the course of hydrothermal treatment (Fig. 2F), thus causing an increase of the Young’s modulus (Fig. 7E) and a decrease of the elongation at break (Fig. 7F). The compensation of electrostatic charge was provided through the addition of xanthan. Its mixtures with chitosan were characterized in .
This work was partially supported by grants 14-03-91700 and 14-03-31350 from the Russian Foundation for Basic Research, 44/2014/75-Dvo from the OPTEC and a grant 2013K2A1A707627 from the Ministry of Science, ICT, and Future Planning of Republic of Korea.
E. Ruiz-Hitzky, M. Darder, P. Aranda. “An introduction to bio-nanohybrid materials”, in Bio-Inorganic Hybrid Nanomaterials, E. Ruiz-Hitzky, K. Ariga, Y. M. Lvov (Eds.), pp. 1–40, Wiley-VCH, Weinheim (2008).Google Scholar
V. P. Ananikov, E. A. Khokhlova, M. P. Egorov, A. M. Sakharov, S. G. Zlotin, A. V. Kucherov, L. M. Kustov, M. L. Gening, N. E. Nifantiev. Mendeleev Commun.25, 75 (2015).Google Scholar
V. A. Fomin, V. V. Guseev. Prog. Rubb. Plastics Tech.17, 186 (2001).Google Scholar
S. Pilla. “Engineering applications of bioplastics and biocomposites – An overview”, in Handbook of Bioplastics and Biocomposites Engineering Applications, S. Pilla (Ed.), pp. 1–15, Scrivener, Salem, Massachusetts (2011).Google Scholar
Y. Kojima, A. Usuki, M. Kawasumi, A. Okada, T. Kurauchi, O. Kamigaito. J. Polym. Sci. A -Polym. Chem.31, 1755 (1993).Google Scholar
R. A. A. Muzzarelli. Chitin, Pergamon Press, Oxford (1977).Google Scholar
G. A. F. Roberts. Chitin Chemistry, MacMillan Education, Basingstoke, Hampshire, UK (1992).Google Scholar
J. G. Winterowd, P. A. Sandford. “Chitin and chitosan”, in Food Polysaccharides and Their Applications, A. M. Stephen (Ed.), pp. 441–462, Marcel Dekker, New York (1995).Google Scholar
E. Khor. Chitin: Fulfilling a Biomaterials Promise, Elsevier Science, Amsterdam (2001).Google Scholar
K. Kurita. Marine Biotechnol.8, 203 (2006).Google Scholar
S. Hirano. “Chitin and chitosan”, in Ullmann’s Encyclopedia of Industrial Chemistry, pp. 471–481, Wiley-VCH, Weinheim (2000).Google Scholar
P. A. Sandford. “Chitosan: Commercial uses and potential applications”, in Chitin and Chitosan. Sources, Chemistry, Biochemistry, Physical Properties and Applications, G. Skjak-Braek, T. Anthonsen, P. A. Sandford (Eds.), pp. 51–69, Elsevier Applied Science, London (1989).Google Scholar
E. Khor. Curr. Opin. Solid. State. Mat. Sci.6, 313 (2002).Google Scholar
R. N. Tharanathan, F. S. Kittur. Crit. Rev. Food Sci. Nutrition. 43, 61 (2003).Google Scholar
Y. A. Shchipunov, N. A. Ivanova, S. A. Sarin. Mendeleev. Commun.19, 149 (2009).Google Scholar
Y. A. Shchipunov, I. V. Postnova. Composite Interf.16, 251 (2009).Google Scholar
Y. A. Shchipunov, N. Ivanova, V. Silant’ev. Green Chem.11, 1758 (2009).Google Scholar
Y. Shchipunov, S. Sarin, I. Kim, C. S. Ha. Green Chem.12, 1187 (2010).Google Scholar
M. F. Brigatti, E. Galan, B. K. G. Theng. “Structure and mineralogy of clay minerals”, in Handbook of Clay Science. Developments in Clay Science, F. Bergaya, B. K. G. Theng, G. Lagaly (Eds.), pp. 19–86, Elsevier, Amsterdam (2006).Google Scholar
M. Darder, M. Lopez-Blanco, P. Aranda, A. J. Aznar, J. Bravo, E. Ruiz-Hitzky. Chem. Mater.18, 1602 (2006).Google Scholar
G. Lagaly. “Smectic clays as ionic macromolecules”, in Developments in Ionic Polymers, A. D. Wilson, H. J. Prosser (Eds.), pp. 77–140, Elsevier, London (1986).Google Scholar
H. Dautzenberg, W. Jaeger, J. Kotz, B. Philipp, C. Seidel. Polyelectrolytes: Formation, Characterization and Applications, Hanser, Munich (1994).Google Scholar
L. Wang, A. Wang. J. Hazardous Mater.147, 979 (2007).Google Scholar
J. D. Dziezak. “Natural acids and acidulants”, in Encyclopedia of Food Science and Nutrition, B. Caballero (Ed.), pp. 12–17, Academic Press, New York (2004).Google Scholar
E. Ruiz-Hitzky. J. Mater. Chem.11, 86 (2001).Google Scholar
Y. Yamaguchi, H. Hoffmann. Colloid. Surf. A121, 67 (1997).Google Scholar
Y. A. Shchipunov, I. V. Postnova. Colloid J.73, 565 (2011).Google Scholar
J. W. Rhim, H. M. Park, C. S. Ha. Prog. Polym. Sci.38, 1629 (2013).Google Scholar
A. B. Zezin, V. A. Kabanov. Russ. Chem. Rev.51, 1447 (1982).Google Scholar
A. Krueger. Carbon Materials and Nanotechnology, Wiley-VCH, Weinheim (2010).Google Scholar