DOI: 10.1002/chem.201301929
Mesoporous Silica and Organosilica Films Templated by Nanocrystalline Chitin Thanh-Dinh Nguyen, Kevin E. Shopsowitz, and Mark J. MacLachlan*[a] Abstract: Liquid crystalline phases can be used to impart order into inorganic solids, creating materials that mimic natural architectures. Herein, mesoporous silica and organosilica films with layered structures and high surface areas have been templated by nanocrystalline chitin. Aqueous suspensions of spindle-shaped chitin nanocrystals were prepared by sequential deacetyla-
tion and hydrolysis of chitin fibrils isolated from king crab shells. The nanocrystalline chitin self-assembles into a nematic liquid-crystalline phase that Keywords: liquid crystals · mesoporous organosilica · mesoporous silica · nanocrystalline chitin · nematic ordering
Introduction Liquid crystal (LC) templating is a powerful paradigm for the construction of periodic mesoporous materials.[1] First developed for the preparation of silica,[2] LC templating has been extended to organosilica[3] and diverse compositions.[4] By varying the charge and nature of the template (e.g., molecular surfactants vs. triblock copolymers), it has proven possible to tune the pore diameters, the organization, and the properties of the mesoporous materials.[5] Sophisticated biomaterials with tailor-made properties are normally formed in nature with hierarchical organization of simple nanoscale elements.[6] LC templating offers a biomimetic approach to prepare materials with intricate organization that mimics the remarkable structures found in nature.[7] Using LC templating, artificial nematic structures have been prepared with control over morphology and structural orientation.[5] Layered materials that were reported with aligned mesochannel structures have photonic properties, giving them potential opportunities in applications as optical elements,[8a] catalyst supports,[8b] and adsorbents.[8c] Nematic and chiral nematic LC phases are observed for many biomacromolecules, such as polysaccharides, collagen, proteins, glycoproteins, and DNA, which can serve as biotemplates to prepare solid replicas.[9] Chitin, which is found in the exoskeletons of crabs, shrimps, beetles, and other insects, is the second most abundant polysaccharide on the [a] Dr. T.-D. Nguyen, Dr. K. E. Shopsowitz, Prof. Dr. M. J. MacLachlan Department of Chemistry University of British Columbia 2036 Main Mall, Vancouver, British Columbia, V6T 1Z1 (Canada) Fax: (+ 1) 604-822-2847 E-mail: [email protected] Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201301929.
15148
has been used to template silica and organosilica composites. Removal of the chitin template by either calcination or sulfuric-acid-catalyzed hydrolysis gave mesoporous silica and ethylene-bridged organosilica films. The large, crack-free mesoporous films have layered structures with features that originate from the nematic organization of the nanocrystalline chitin.
planet (after cellulose).[10] Remarkably, the cuticles of the king crab have a BouACHTUNGREliACHTUNGREgand structure of the layered chitin fibrils combined with proteins and minerals (mainly calcium carbonate).[11] Jewel beetle shells have a hierarchical chiral nematic organization of the chitin fibrils, resulting in green colors with a metallic sheen.[12] Recently, we used lyotropic LC nanocrystalline cellulose (NCC) to template free-standing mesoporous-silica glasses with tunable chiral nematic structures.[13a,b] The helical pitch of the chiral nematic order can be controlled by changing the concentration ratio of silica sol–gel precursor to NCC used in the preparation, thus tuning the wavelength of the reflected light from the ultraviolet to near-infrared region. By employing an organosilica precursor, flexible films of mesoporous organosilica with chiral nematic ordering were obtained.[13c,d] Spindle-shaped nanocrystalline chitin (NCh) can be prepared by acid-catalyzed hydrolysis of the fibril samples isolated from arthropod exoskeletons.[14a] Aqueous NCh suspensions can be stabilized in acidic media (pH 3.5–4.0), where the protonation of the amino groups exposed on the crystallites provides a positive surface charge and disrupts their aggregation. The chitin rods are known to self-assemble into a nematic liquid-crystalline phase. The extent of the deacetylation of NCh changes their surface charge density and was found to affect the critical concentration for the nematic phase formation.[14b] Drying the NCh suspensions could give solid films with hierarchical organization. It is apparent that the LC phase of NCh might potentially be used as a template to construct porous solid-state materials. Mesoporous silica and organosilica materials prepared by LC templating with polymeric surfactants have been the subject of widespread research in materials chemistry.[15] The use of the LC phase of NCh to template these materials may give films with pore structures that replicate the nemat-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15148 – 15154
FULL PAPER ic order of the LC phase. Incorporation of ethylene bridges into the mesoporous organosilica film is expected to improve its mechanical flexibility compared with the pure silica analogue. The pioneering work of Alonso and Belamie[16] demonstrated the first examples of mesostructured silicas formed through NCh template-directed sol–gel condensation of SiACHTUNGRE(OC2H5)4 precursors; however, only poor order of these materials was observed. In addition, the method was not extended to organosilicas, which could lead to distinct materials with improved properties. Finding a new pathway to prepare replicas of the layered NCh structure in mesoporous silica and organosilica films remains a challenge. In this paper, we report the first experimental evidence for the replication of a nematic structure in mesoporous silica and ethylene-bridged organosilica films templated by NCh. Spindle-shaped chitin nanocrystals prepared by sequential deacetylation and hydrolysis of chitin fibrils isolated from king crab shells organized into layered nematic structures that functioned as templates for silica/NCh and organosilica/NCh composites. After removal of the chitin template, large, crack-free mesoporous silica and organosilica films with sizes of several centimeters were obtained. These mesoporous materials with layered organization and high specific surface area were replicated from NCh.
Figure 1. Nematic order of the NCh liquid crystals (NCh-4). a) TEM image of individual NCh fibers (scale bar, 2 mm). b) Left side, top to bottom: Vials of aqueous NCh suspensions with different concentrations at pH 4 and a NCh film prepared by drying the NCh suspension (6.6 wt %, pH 4; scale bar = 1 cm). Right side: An illustration of the layered structure of NCh. c) POM image of the NCh suspension showing birefringent textures characteristic of the nematic organization (scale bar = 100 mm). d) SEM image viewed along the edge of the NCh film at fractures (scale bar = 500 nm).
Results and Discussion King crab shells were treated with base and acid to remove protein and minerals, then bleached with a dilute H2O2 solution to remove pigment (Figure S1 a,b in the Supporting Information). The isolated chitin samples were then partially hydrolyzed in acid to produce nanocrystalline fragments.[14a] The acid-hydrolysis process also led to deacetylation of the chitin, resulting in free primary amine groups. The positive surface-charge density of the chitin crystallites in acidic media, which is related to the degree of deacetylation (DDA), is crucial in determining the organization of the ordered nematic LC phase. In this work, we set out to find new reaction conditions, using either hydrolysis or sequential deacetylation and hydrolysis, to prepare the nematic LC phase of the nanocrystalline chitin. NCh samples were prepared by hydrochloric-acid-catalyzed hydrolysis of the isolated chitin fibrils at 104 8C for different lengths of time (2, 10, 18 h). For comparison, a NCh sample was prepared by deacetylation of the chitin fibrils at 90 8C for 2 h in alkaline solution followed by hydrochloric-acid-catalyzed hydrolysis at 104 8C for 18 h (see Table S1 and Figure S1 c in the Supporting Information). The NCh samples prepared by onestep and two-step processes have a similar chitin weight loss ( 12 wt. %), suggesting that little degradation of the chitin fibrils occurred during deacetylation. The NCh samples show an improved stability as suspensions in water with increasing DDA. The NCh suspensions typically formed gels at pH < 2.5 and at high concentrations ( 8.0 wt %), as shown in Figure 1 b. Transmission electron
Chem. Eur. J. 2013, 19, 15148 – 15154
microscopy (TEM) of the dilute NCh solution (0.0005 wt %) reveals that the chitin crystallites are discrete spindleshaped nanocrystals with diameters of 10–18 nm and lengths of 300–500 nm (Figure 1 a). The acid and/or base treatments for the different heating durations did not cause a significant change in the crystallite size or shape. The nanocrystalline chitin structure was confirmed by Fourier-transform infrared (FTIR) spectroscopy (Figure S2 a in the Supporting Information) and powder X-ray diffraction (PXRD) analyses (Figure S2 b in the Supporting Information), indicating that the deacetylation and hydrolysis did not modify the original crystalline structure of a-chitin. The resulting chitin nanocrystals show good crystallinity after removal of non-crystalline fractions during hydrolysis. These data confirm that the spindle-shaped nanocrystals with partial deacetylation are nanoscale fragments of the chitin microfibrils. A nematic LC phase of NCh can form when the critical concentration and appropriate pH of the NCh suspension are reached. We found that the nematic organization of the prepared nanocrystalline chitin forms at a concentration of 6.6 wt % and pH 4. Figure 1 and Figures S3 and S4 in the Supporting Information show photographs, polarized optical microscopy (POM) images, and scanning electron microscopy (SEM) images of the aqueous NCh suspensions and selfassembled NCh films. The acid-catalyzed hydrolysis of the chitin fibrils gave NCh-1 at 2 h and NCh-2 at 10 h, which after drying at ambient conditions, gave NCh films consisting only of randomly oriented chitin nanocrystals (no evidence for long-range order); consequently, poor birefrin-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15149
M. J. MacLachlan et al.
gence was observed (Figure S3 a,b in the Supporting Information). This is in contrast to literature reports of NCh suspensions that showed fingerprint textures observed under crossed polarizers, characteristic of chiral nematic organization.[14a] As the hydrolysis time was extended to 18 h, birefringence of NCh-3 was enhanced, providing evidence for the alignment of NCh (Figure S3 c,d in the Supporting Information). The absence of fingerprint textures observed under POM indicates that nematic organization rather than chiral nematic order was present. In addition, we did not observe iridescence that is often associated with chiral nematic order. A NCh sample (NCh-4) prepared by sequential deacetylation (90 8C for 2 h) and hydrolysis (104 8C for 18 h) exhibits strong birefringence when observed under crossed polarizers (Figure 1 c and Figure S4 b,c in the Supporting Information). SEM images (Figure 1 d and Figure S4 d,e in the Supporting Information) show the morphology of the NCh films prepared from an aqueous NCh-4 suspension. A nematic organization is apparent from the image rather than the BouACHTUNGREliACHTUNGREgand structure of the chitin fibrils in the cuticles of the king crabs.[11] In the image, the chitin nanocrystals are oriented parallel to each other, which is likely directed by surface charge interactions of the crystallites. Edge and cross-sectional views of the NCh film show a layered structure. We found the nematic structure of the LC chitin produced from the king crab shells was more ordered than from other crab specimens, and chitin nanocrystals from shrimp shells gave especially poor ordering in our hands. The nematic order of the NCh suspension is related to the protonation of the exposed surface amino groups. The surface potential for the dilute NCh suspension (0.01 wt %, pH 4) was determined by zeta potential measurements (see Table S1 in the Supporting Information). The zeta potential reflects the dependence of the ionic strength of the suspension on the formation of an electrical double layer at the NCh surface. This value is considered as a proportional function of the charge on the NCh surface.[17] The respective zeta potentials of NCh-1, NCh-2, and NCh-3 prepared by acid hydrolysis were found to be 0.6, 2.3, and 10.3 mV, respectively. There was a slight increase of the zeta-potential value when the hydrolysis time was extended, probably due to an increase of DDA that would give more exposed surface amino groups, favoring protonation. This additional surface charge appears to result in an improved alignment of the chitin nanocrystals. The one-step acid hydrolysis of the chitin fibers results in two reactions: depolymerization and deacetylation.[18] The depolymerization occurred preferentially over deacetylation, thus frequently generating chitin nanocrystals with fewer exposed amino groups and corresponding low surface-charge density. This could inhibit the organization of the chitin nanocrystals into an ordered structure even with extended hydrolysis time. This poor ordering was overcome by performing the two-step process, deacetylation followed by hydrolysis, to produce NCh-4 that has a zeta potential of 43.1 mV, much higher than those of the samples prepared by
15150
www.chemeurj.org
a one-step process. This is consistent with an increasing number of exposed surface amino groups. The highly positive surface-charge density of NCh-4 derived from the protonation of the surface amino groups resulted in an improved organization in the films. However, the layered ordering was only found to arise under a narrow range of reaction conditions tested (deacetylation at 90 8C for 2 h followed by hydrolysis at 104 8C for 18 h yielded NCh-4). We varied both the duration of the deacetylation and the alkaline concentration, but found that other conditions led only to poor organization. The LC phases of NCh were used to template silica and organosilica materials through evaporation-induced self-assembly of NCh with corresponding tetramethylorthosilicate (TMOS) and 1,2-bis(trimethoxysilyl)ethane (BTMSE) precursors. The silica-based precursors were mixed with the NCh suspensions (6.6 wt %, pH 4) with stirring at room temperature and underwent hydrolysis and condensation to form homogeneous suspensions within 1 h. The composite suspensions were transferred to 60 mm diameter Petri dishes and dried under ambient conditions for 48 h to yield silica/NCh and organosilica/NCh composite films. We observed that the hierarchical organization of the LC NCh template employed is transferred to the composites. Thus, when the suspensions of NCh-1, NCh-2, or NCh-3 having poor ordering were used as templates, the silica films obtained after removal of the chitin by calcination have inhomogeneous layered structures with poor orientation and exhibit low optical anisotropy under POM (Figure S5 in the Supporting Information). Even though NCh-3 exhibits birefringence, the resulting silica films still have a disordered layered structure (Figure S6 in the Supporting Information). The ordered nematic organization of the NCh-4 template is expected to produce silica-based materials with better ordering than NCh-1, NCh-2, and NCh-3. Composites were prepared by LC templating with NCh-4 by using different ratios of the silica-based precursors to NCh (see Table S2 in the Supporting Information). POM images (Figure S7 a,b in the Supporting Information) of the homogeneous suspensions of silica/NCh and organosilica/NCh composites still exhibit birefringence, revealing that the precursors do not disrupt the alignment of NCh. The birefringent textures of these composite suspensions are analogous to those of the aqueous liquid-crystalline suspension of smectite clay.[19] Drying the composite suspensions gave large, crack-free silica/NCh and organosilica/NCh composite films with sizes of several centimeters (Figure 2 a,d) that show birefringence under POM, similar to that observed in the NCh-4 films (Figure 2 e and Figure S7 c in the Supporting Information). Layered structures originating from the alignment of NCh were observed by SEM images of the composites (Figures S8 and S9 in the Supporting Information). In contrast to the pure NCh films, the composites could not be dispersed in water. Calcination of the silica/NCh composites at 540 8C under air and treatment of the organosilica/NCh composites in H2SO4 (6 m) at 90 8C for 18 h successfully removed the chitin
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15148 – 15154
Mesoporous Silica and Organosilica Films
Figure 2. Silica/NCh (S7) and organosilica/NCh (OS2) composites before and after removal of the chitin template. Photographs of a) the silica/ NCh composites and b) the calcined silica film. c) POM image of the calcined silica films. d) Photograph and e) POM image of the organosilica/ NCh composites (UBC logo is shown to demonstrate transparency of the film). f) Photograph of the large, crack-free organosilica film after sulfuric acid hydrolysis of the chitin.
template, yielding the corresponding mesoporous silica and organosilica films. We previously developed an acid-hydrolysis approach to remove nanocrystalline cellulose from chiral nematic organosilica/cellulose composites to generate iridescent mesoporous organosilica films.[13c] This same procedure was used to remove the chitin template in the organosilica/ NCh composites. We observed that the decomposition of the chitin in H2SO4 (6 m) at 90 8C produces a byproduct that stained the resulting films brown. These byproducts were eliminated by treatment with a H2O2/H2SO4 mixture to give colorless films. After chitin removal, large, crack-free films were recovered with shapes similar to the original composites and dimensions of several centimeters (Figure 2 f). The silica films are transparent (Figure 2 b), but the organosilica films appear opaque after chitin removal. Complete removal of the chitin template in the silica and organosilica films was confirmed using a variety of techniques. For the silica/NCh composites, IR bands associated with vibrational modes of NCh (e.g., n˜ sACHTUNGRE(N H)) are absent in the silica after chitin removal (Figure S10 a,b in the Supporting Information). Thermogravimetric analyses (TGA) show that the decomposition of the chitin ( 75 wt %) in the composites occurred at ~ 150–600 8C compared to 100 wt % loss for the pure NCh film (Figure S10 c,d in the Supporting In-
Chem. Eur. J. 2013, 19, 15148 – 15154
FULL PAPER formation). PXRD patterns show that the silica/NCh composites have peaks characteristic of the crystalline chitin, while the calcined silica obtained after chitin removal has only a broad peak at 228 2q attributed to amorphous silica (Figure S10 e,f in the Supporting Information). Elemental analysis indicated that the silica/NCh composites contain 37 wt % carbon, but that carbon was absent in the corresponding calcined silica (referred to as S7). These results are consistent with the complete removal of the chitin in the composites to yield pure silica films. For the organosilica films (OS2), solid-state 13C cross-polarization/magic-angle spinning (CP/MAS) NMR spectra (Figure 6 a) of the organosilica/NCh composites before and after sulfuric acid hydrolysis of the chitin show that the chitin was removed without damaging the organosilica backbone. The composites have peaks at 20–180 ppm assigned to the chitin and a single peak at 5 ppm assigned to ethylene (SiCH2CH2Si) carbons of the organosilica.[13c] After acid hydrolysis treatment, only the single peak of the organosilica remains, demonstrating that the chitin was completely removed. Solid-state 29Si CP/MAS NMR spectra (Figure S11 in the Supporting Information) of the composites before and after chitin removal show peaks at 65 and 57 ppm assigned to corresponding T3 (CSiACHTUNGRE(OSi)3) and T2 (CSiACHTUNGRE(OSi)2OH) Si atoms of the ethylene-bridged organosilica.[13c] Signals expected for SiO4 species between 90 and 120 ppm are not observed, demonstrating that no cleavage of Si C bonds occurred during acid hydrolysis. TGA analysis (Figure S12 in the Supporting Information) shows that the organosilica is stable up to 400 8C, above which the ethylene bridges (10–15 wt %) decomposed. In contrast to the calcined silica, elemental analysis indicated that the organosilica contains 14.7 wt % carbon and 4.6 wt % hydrogen. In comparison, TGA and elemental analysis of a silica sample prepared by removal of the chitin in the silica/NCh composites using the exact same procedure of the organosilica preparation (heating the composites in H2SO4 (6 m) at 90 8C for 18 h) show that no residual organic species were present. This further confirms the acid hydrolysis conditions employed to be sufficient for the complete degradation and removal of the chitin, while preserving the ethylene bridges in the organosilica films. POM images (Figure 2 c) of the silica films show a significant difference in birefringence compared with the silica/ NCh composites. After chitin removal, the birefringent textures of the silica films shift from sepia to sepia–violet colors. The organosilica films exhibit low optical anisotropy due to their opacity. In contrast with the brittle silica films, the ethylene-bridged organosilica films are quite flexible. The layered organization in the silica and organosilica films is evident in SEM images shown in Figures 3 and 4, respectively. Cross-sectional views perpendicular to the flat surfaces indicate that the silica and organosilica films both have layered structures throughout their entire thickness (Figures 3 a and 4 a). Electron microscopy images of edges at high magnification reveal that there are spindle-like features in each layer (Figures 3 b,c and 4 b,c). These spindle-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15151
M. J. MacLachlan et al.
During templating of the silica-based precursors with NCh, the organization of the LC chitin was transferred to the silica and organosilica, and the nanometer-scale order was maintained during calcination and chitin hydrolysis. The porosity of the silica and organosilica films was determined by nitrogen adsorption/desorption studies. Figures 5 a,b and 6 b show that the silica and organosilica films have
Figure 3. SEM images of the calcined silica films with layered ordering (S7). a) Edge view of the films at fractures showing a relatively smooth top surface (scale bar = 50 mm). b) Cross-sectional view of the film at fractures showing a layered structure (scale bar = 5 mm). c) Higher magnification showing each layer composed of numerous spindle-shaped pores (scale bar = 1 mm). d) Top view showing a flat surface constituted of the alignment of the spindlelike features (scale bar = 2 mm). Figure 5. Nitrogen and solvent adsorption of layered mesoporous silica films (S7). a) Typical type-IV adsorption/desorption isotherms characteristic of the mesoporous structure. b) BJH adsorption of pore size distribution. c) TEM image of S7 (scale bar = 100 nm). d) Photograph of S7 after placing a water droplet (scale bar = 1 cm). e) POM image of S7 after adsorbing water showing a disappearance of birefringence in the wet part of the film at the right side compared with the dried part at the left side (scale bar = 100 mm).
Figure 4. SEM images of the organosilica films with layered ordering (OS2) after sulfuric acid hydrolysis of the chitin. a) Edge views of the films at fractures at a) low (scale bar = 50 mm) and b) high (scale bar = 5 mm) magnifications showing relatively smooth top surfaces and a layered structure. c) Cross-sectional view showing spindlelike features in each layer (scale bar = 2 mm). d) Top view showing a flat surface with aligned spindlelike features (scale bar = 2 mm).
like features are organized with long-range ordering, as observed by viewing the flat top surface of the films (Figures 3 d and 4 d). These images reveal that the silica and ethylene-bridged organosilica films both accurately replicated the nematic organization of NCh-4 in a layered structure.
15152
www.chemeurj.org
typical type-IV isotherms with type-H2 hysteresis, which indicate the introduction of mesoporosity into the silica-based solids. The porosity of these materials can be varied by changing the concentration ratios of the silica and organosilica sol–gel precursors to NCh used in the preparations (see Table S2 in the Supporting Information). Brunauer– Emmett–Teller (BET) surface areas and pore volumes were found to be 420–650 m2 g 1 and 0.48–0.70 cm3 g 1, respectively, for the silica, and 690–800 m2 g 1 and 0.32–0.40 cm3 g 1, respectively, for the organosilica samples. A decrease in the pore wall thickness for the silica and organosilica prepared using the lower precursor loadings could result in the partial collapse of the network upon calcination and hydrolysis. The use of higher precursor loadings gave thicker walls, probably preventing connectivity between pores (Figure S13 in the Supporting Information). The pore size distribution of the mesoporous organosilicas determined by BJH (Barrett–Joyner–Halenda) analysis is broad with peaks at 9–14 nm (inset of Figure 6 b), which is considerably larger than the 4–7 nm pore diameters of the mesoporous silicas generated by calcination (Figure 5 b). The pore size distribution of the organosilica is close to the diameter range ( 10–18 nm) of the chitin template observed by TEM. Pore diameters of the organosilicas are
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15148 – 15154
Mesoporous Silica and Organosilica Films
FULL PAPER of birefringence. After drying the films, they returned to their original states and the structural birefringence resulting from alignment of the pores fully recovered. In contrast, the silica/NCh and organosilica/NCh composites, which are nonporous, showed no significant change in their birefringence upon addition of water. Other solvents, such as ethanol and acetone, were also effectively absorbed by the layered mesoporous silica and organosilica films.
Conclusion
Figure 6. Nitrogen and solvent adsorption of layered mesoporous organosilica films (OS2). a) Solid-state 13C NMR spectra of the organosilica/ NCh composites before and after sulfuric acid hydrolysis of the chitin. b) Nitrogen adsorption/desorption isotherms of OS2; inset of BJH adsorption of pore size distribution. c) TEM image of OS2. d) Photograph of OS2 after placing a water droplet in the center of the film, leading the wet part of the film to be transparent.
larger than those of the calcined silica films, indicating that there is less pore shrinkage from the acid-catalyzed hydrolysis of the chitin template than from the calcination. Thus, the pore sizes of the mesoporous silica and organosilica can be tuned by choosing the conditions to remove the chitin template. It is worth noting that the pore sizes of the mesoporous materials prepared using NCh as a template are larger than those prepared using NCC as a template,[13a,c] consistent with their relative diameters. This highlights the nanoscale control over porosity afforded by using biologically derived, nanocrystalline templates. TEM images of the mesoporous silica (Figure 5 c) and organosilica (Figure 6 c) confirmed the formation of the aligned porous networks within the films, where the mesoporous channels run parallel to a preferred orientation within each layer. Spindle-shaped pores were observed consistent with the layered organization of the mesoporous structures. The pore diameters observed from the TEM images are around 4–8 nm for the silica and 10–15 nm for the organosilica in accordance with the nitrogen adsorption data. The layered films with partially aligned pore channels yielded strong birefringence in the case of the silica. The order and mesoporosity of the silica and organosilica replicas were further elucidated by solvent absorption into the films. When water droplets were placed on the silica and organosilica films, water was rapidly absorbed and both films immediately appeared transparent, as shown in Figures 5 d and 6 d. Examination of these films under POM (Figure 5 e) shows a disappearance of birefringence. These results show that water fills the pores of the silica and organosilica, not only the surfaces, and its matching of the refractive index of the silica and organosilica leads to nullification
Chem. Eur. J. 2013, 19, 15148 – 15154
In conclusion, mesoporous silica and organosilica films with layered structures originating from the nematic LC phase of NCh have been prepared. The nematic LC phase of chitin prepared using the two-step process of sequential deacetylation and hydrolysis was used to template silica/NCh and organosilica/NCh composites. The chitin template can be removed by calcination in the case of the silica/NCh composites, or by acid-catalyzed hydrolysis in the case of the organosilica/NCh composites, giving mesoporous silica and organosilica films, respectively. Large, crack-free silica and organosilica films retained the shapes of the precursor composites and had sizes of several centimeters. The silica and organosilica replicas show layered nematic organization and mesoporosity arising from a transfer of the LC phase of the chitin. The use of renewable nanocrystalline chitin as a template allows for the large-scale preparation of the ordered mesoporous materials. A further research direction could be to use these materials as a hard template to replicate the pore organization into other materials. These novel materials might open an opportunity for useful applications in optical elements, catalyst supports, and adsorbents.
Experimental Section Chitin purification: Dried king crab shells (25 g) were soaked in an aqueous solution of NaOH (5 wt %, 500 mL) at 80 8C for 6 h to remove protein. The products were cooled to room temperature and washed thoroughly with water. Next, the shells were treated with an aqueous solution of HCl (7 vol %, 500 mL) for 4 d at room temperature to remove calcium carbonate minerals. After collecting the products and rinsing with abundant water, the samples were treated with a NaOH solution (5 wt %, 500 mL) at 80 8C for 8 h to completely remove residual protein. The ratio of the shells to the NaOH (5 wt %) or HCl (7 vol %) solution was 1 g per 20 mL. The products were then washed thoroughly with water and remaining pigments were oxidized with an aqueous solution of H2O2 (5 vol %, 500 mL) at 90 8C for 1 h. The fully bleached samples were collected and washed several times with distilled water followed by drying in an oven at 40 8C overnight to obtain white chitin. The extraction yield of the chitin from the king crab shells was about 13 wt % following the above procedure. Preparation of aqueous NCh suspensions: Acid-catalyzed hydrolysis: Dried chitin samples (2.5 g) were cut into small flakes ( 1 cm) and treated with a HCl solution (4 m, 50 mL) in a round-bottom flask equipped with a reflux condenser. The reaction mixture was heated at 104 8C with vigorous stirring for different lengths of time (2, 10, 18 h) to obtain corresponding samples NCh-1, NCh-2, and NCh-3. Dark-brown products that formed during the process were oxi-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15153
M. J. MacLachlan et al.
dized by injecting H2O2 (2.5 mL, 30 vol %) into the hot reaction mixture at the end of the hydrolysis and keeping them at the same temperature for 30 min. After acid hydrolysis, the reaction mixture was diluted with 100 mL of distilled water followed by centrifugation (at 2200 rpm for 10 min) to remove soluble chitin species. This step was repeated two more times to produce an aqueous chitin suspension with pH 2. The suspension was transferred into dialysis bags and dialyzed against distilled water for 48 h until the pH of the suspension increased to about 4. The dialysis bags were then immersed in polyethylene glycol 400 for 3 h to obtain an appropriately concentrated sample that would favor the formation of a nematic LC phase. To increase the stabilization of the nanocrystalline chitin colloids, the suspension (40 mL) was dispersed with ultrasound treatment for 1 h using a Biosonic ultrasonic (50 Hz, 100 W). Alkaline-treated deacetylation and acid-catalyzed hydrolysis: The stabilization of the aqueous NCh suspension, which is affected by protonation of the exposed surface amino groups, is responsible for the organization of the individual chitin nanocrystals into a nematic LC phase. Hence, a two-step process of sequential deacetylation and hydrolysis of the isolated chitin was carried out. Typically, the isolated chitin flakes (2.5 g) were treated with NaOH solution (25 mL, 33 wt %) at 90 8C for 2 h with vigorous stirring. The reaction mixture was cooled to room temperature, centrifuged, and neutralized by washing with abundant water to obtain nanocrystalline chitin sample with partial deacetylation. The sample was then treated with HCl solution (50 mL, 4 m) at 104 8C with vigorous stirring for 18 h to obtain NCh-4. The sample was treated with H2O2 and isolated as described above. Preparation of mesoporous silica and organosilica films from the nematic NCh template: For the preparation of the mesoporous silicas, the aqueous NCh suspension (15 mL, 6.6 wt % NCh, pH 4) was sonicated for 10 min using an Aquasonic 50T sonic cleaner. An appropriate amount of the silica precursor (TMOS) was mixed with the NCh suspension (15 mL, 6.6 wt %) with stirring at room temperature for 1 h. The silica sol–gel concentration was varied corresponding to samples (S1 to S9), as presented in Table S2 in the Supporting Information. Portions (3.6 mL) were then transferred to 60 mm diameter polystyrene Petri dishes and dried at ambient conditions within 48 h to give silica/NCh composites. For removal of the chitin, the silica/NCh composites were calcined under flowing air at 540 8C for 6 h to recover silica films (47 mg of S7 was obtained from 230 mg of silica/NCh composites). For the preparation of the mesoporous organosilicas, the organosilica precursor (BTMS) was used instead of TMOS to produce organosilica/NCh composites under the exact same procedure. The respective amounts of BTMS used for the preparation of samples was 120 mg (OS1) and 240 mg (OS2). For removal of the chitin, the organosilica/NCh composites were placed in H2SO4 (900 mL, 6 m) and heated to 90 8C for 18 h. The products were then filtered and soaked in a piranha solution (20 mL 30 % H2O2/100 mL H2SO4) for 1 h to give colorless films. The films were then washed copiously with distilled water and dried at ambient conditions to recover organosilica films (91 mg of OS2 was obtained from 230 mg of organosilica/NCh composites).
Acknowledgements We are grateful to the Natural Sciences and Engineering Research Council (NSERC) of Canada for funding.
[1] a) C. G. Goltner, M. Antonietti, Adv. Mater. 1997, 9, 431 – 436; b) M. E. Raimondi, J. M. Seddon, Liquid Crystals 1999, 26, 305 – 339; c) M. Antonietti, Phil. Trans. R. Soc. A 2006, 364, 2817 – 2840. [2] a) C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli, J. S. Beck, Nature 1992, 359, 710 – 712; b) Y. Wan, D. Zhao, Chem. Rev. 2007, 107, 2821 – 2860; c) A. Firouzi, F. Atef, A. G. Oertli, G. D. Stucky, B. F. Chmelka, J. Am. Chem. Soc. 1997, 119, 3596 – 3610.
15154
www.chemeurj.org
[3] a) H. Zhu, D. J. Jones, J. Zajac, J. Roziere, R. Dutartre, Chem. Commun. 2001, 2568 – 2569; b) F. Hoffmann, M. Cornelius, J. Morell, M. Froba, Angew. Chem. 2006, 118, 3290 – 3328; Angew. Chem. Int. Ed. 2006, 45, 3216 – 3251. [4] a) G. S. Attard, P. N. Bartlett, N. R. B. Coleman, J. M. Elliott, J. R. Owen, J. H. Wang, Science 1997, 278, 838 – 840; b) S. R. J. Oliver, G. A. Ozin, J. Mater. Chem. 1998, 8, 1081 – 1085; c) P. V. Braun, P. Osenar, V. Tohver, S. B. Kennedy, S. I. Stupp, J. Am. Chem. Soc. 1999, 121, 7302 – 7309; d) G. S. Armatas, M. G. Kanatzidis, Science 2006, 313, 817 – 820. [5] a) V. Percec, Phil. Trans. R. Soc. A 2006, 364, 2709 – 2719; b) S. T. Tolbert, A. Firouzi, G. D. Stucky, B. F. Chmelka, Science 1997, 278, 264 – 268; c) P. Feng, X. Bu, D. J. Pine, Langmuir 2000, 16, 5304 – 5310. [6] a) C. Sanchez, H. Arribart, M. M. G. Guille, Nat. Mater. 2005, 4, 277 – 288; b) J. Aizenberg, J. C. Weaver, M. S. Thanawala, V. C. Sundar, D. E. Morse, P. Fratzl, Science 2005, 309, 275 – 278. [7] a) W. J. Chung, J. W. Oh, K. Kwak, B. Y. Lee, J. Meyer, E. Wang, A. Hexemer, S. W. Lee, Nature 2011, 478, 364 – 368; b) H. Zhou, T. Fan, D. Zhang, ChemSusChem 2011, 4, 1344 – 1387. [8] a) F. Castles, F. V. Day, S. M. Morris, D. H. Ko, D. J. Gardiner, M. M. Qasim, S. Nosheen, P. J. W. Hands, S. S. Choi, R. H. Friend, H. J. Coles, Nat. Mater. 2012, 11, 599 – 603; b) B. F. G. Johnson, S. A. Raynor, D. S. Shephard, T. Mashmeyer, J. M. Thomas, G. W. J. Chung, J. W. Oh, K. Kwak, B. Y. Lee, J. Meyer, E. Wang, A. Hexeme Sankar, S. Bromley, R. Oldroyd, L. Gladden, M. D. Mantle, Chem. Commun. 1999, 1167 – 1168; c) K. Kosuge, S. Kubo, N. Kikukawa, M. Takemori, Langmuir 2007, 23, 3095 – 3102. [9] a) A. C. Neville, Biology of the Arthropod Cuticle, Springer, New York, 1975, pp. 159 – 234; b) S. J. Woltman, G. D. Jay, G. P. Crawford, Nat. Mater. 2007, 6, 929 – 938. [10] H. Merzendorfer, J. Comp. Physiol. B 2006, 176, 1 – 15. [11] a) Y. BouACHTUNGREliACHTUNGREgand, Tissue Cell 1972, 4, 189 – 217; b) P. Y. Chen, A. Y. M. Lin, J. McKittrick, M. A. Meyers, Acta Biomater. 2008, 4, 587 – 596. [12] a) T. Lenau, M. Barfoed, Adv. Eng. Mater. 2008, 10, 299 – 314; b) V. Sharma, M. Crne, J. O. Park, M. Srinivasarao, Science 2009, 325, 449 – 451. [13] a) K. E. Shopsowitz, H. Qi, W. Y. Hamad, M. J. MacLachlan, Nature 2010, 468, 422 – 425; b) J. A. Kelly, M. Yu, W. Y. Hamad, M. J. MacACHTUNGRELachlan, Adv. Opt. Mater. 2013, 1, 295 – 299; c) K. E. Shopsowitz, W. Y. Hamad, M. J. MacLachlan, J. Am. Chem. Soc. 2012, 134, 867 – 870; d) A. S. Terpstra, K. E. Shopsowitz, C. F. Gregory, A. P. Manning, C. A. Michal, W. Y. Hamad, J. Yang, M. J. MacLachlan, Chem. Commun. 2013, 49, 1645 – 1647. [14] a) J. F. Revol, R. H. Marchessault, Int. J. Biol. Macromol. 1993, 15, 329 – 335; b) J. Li, J. F. Revol, R. H. Marchessault, J. Appl. Polym. Sci. 1997, 65, 373 – 380. [15] a) H. Yang, N. Coombs, I. Sokolov, G. A. Ozin, Nature 1996, 381, 589 – 592; b) G. S. Attard, J. C. Glyde, C. G. Goltner, Nature 1995, 378, 366 – 368. [16] a) B. Alonso, E. Belamie, Angew. Chem. 2010, 122, 8377 – 8380; Angew. Chem. Int. Ed. 2010, 49, 8201 – 8204; b) E. Belamie, M. Y. Boltoeva, K. Yang, T. Cacciaguerra, B. Alonso, J. Mater. Chem. 2011, 21, 16997 – 17006. [17] J. Li, J. F. Revol, E. Naranjo, R. H. Marchessault, Int. J. Biol. Macromol. 1996, 18, 177 – 187. [18] A. Einbu, K. M. Varum, Biomacromolecules 2008, 9, 1870 – 1875. [19] L. J. Michot, I. Bihannic, S. Maddi, S. S. Funari, C. Baravian, P. Levitz, P. Davidson, Proc. Natl. Acad. Sci. USA 2006, 103, 16101 – 16104.
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Received: May 20, 2013 Published online: September 25, 2013
Chem. Eur. J. 2013, 19, 15148 – 15154
Mesoporous Silica and Organosilica Films Templated by Nanocrystalline Chitin Thanh-Dinh Nguyen, Kevin E. Shopsowitz, and Mark J. MacLachlan*[a] Abstract: Liquid crystalline phases can be used to impart order into inorganic solids, creating materials that mimic natural architectures. Herein, mesoporous silica and organosilica films with layered structures and high surface areas have been templated by nanocrystalline chitin. Aqueous suspensions of spindle-shaped chitin nanocrystals were prepared by sequential deacetyla-
tion and hydrolysis of chitin fibrils isolated from king crab shells. The nanocrystalline chitin self-assembles into a nematic liquid-crystalline phase that Keywords: liquid crystals · mesoporous organosilica · mesoporous silica · nanocrystalline chitin · nematic ordering
Introduction Liquid crystal (LC) templating is a powerful paradigm for the construction of periodic mesoporous materials.[1] First developed for the preparation of silica,[2] LC templating has been extended to organosilica[3] and diverse compositions.[4] By varying the charge and nature of the template (e.g., molecular surfactants vs. triblock copolymers), it has proven possible to tune the pore diameters, the organization, and the properties of the mesoporous materials.[5] Sophisticated biomaterials with tailor-made properties are normally formed in nature with hierarchical organization of simple nanoscale elements.[6] LC templating offers a biomimetic approach to prepare materials with intricate organization that mimics the remarkable structures found in nature.[7] Using LC templating, artificial nematic structures have been prepared with control over morphology and structural orientation.[5] Layered materials that were reported with aligned mesochannel structures have photonic properties, giving them potential opportunities in applications as optical elements,[8a] catalyst supports,[8b] and adsorbents.[8c] Nematic and chiral nematic LC phases are observed for many biomacromolecules, such as polysaccharides, collagen, proteins, glycoproteins, and DNA, which can serve as biotemplates to prepare solid replicas.[9] Chitin, which is found in the exoskeletons of crabs, shrimps, beetles, and other insects, is the second most abundant polysaccharide on the [a] Dr. T.-D. Nguyen, Dr. K. E. Shopsowitz, Prof. Dr. M. J. MacLachlan Department of Chemistry University of British Columbia 2036 Main Mall, Vancouver, British Columbia, V6T 1Z1 (Canada) Fax: (+ 1) 604-822-2847 E-mail: [email protected] Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201301929.
15148
has been used to template silica and organosilica composites. Removal of the chitin template by either calcination or sulfuric-acid-catalyzed hydrolysis gave mesoporous silica and ethylene-bridged organosilica films. The large, crack-free mesoporous films have layered structures with features that originate from the nematic organization of the nanocrystalline chitin.
planet (after cellulose).[10] Remarkably, the cuticles of the king crab have a BouACHTUNGREliACHTUNGREgand structure of the layered chitin fibrils combined with proteins and minerals (mainly calcium carbonate).[11] Jewel beetle shells have a hierarchical chiral nematic organization of the chitin fibrils, resulting in green colors with a metallic sheen.[12] Recently, we used lyotropic LC nanocrystalline cellulose (NCC) to template free-standing mesoporous-silica glasses with tunable chiral nematic structures.[13a,b] The helical pitch of the chiral nematic order can be controlled by changing the concentration ratio of silica sol–gel precursor to NCC used in the preparation, thus tuning the wavelength of the reflected light from the ultraviolet to near-infrared region. By employing an organosilica precursor, flexible films of mesoporous organosilica with chiral nematic ordering were obtained.[13c,d] Spindle-shaped nanocrystalline chitin (NCh) can be prepared by acid-catalyzed hydrolysis of the fibril samples isolated from arthropod exoskeletons.[14a] Aqueous NCh suspensions can be stabilized in acidic media (pH 3.5–4.0), where the protonation of the amino groups exposed on the crystallites provides a positive surface charge and disrupts their aggregation. The chitin rods are known to self-assemble into a nematic liquid-crystalline phase. The extent of the deacetylation of NCh changes their surface charge density and was found to affect the critical concentration for the nematic phase formation.[14b] Drying the NCh suspensions could give solid films with hierarchical organization. It is apparent that the LC phase of NCh might potentially be used as a template to construct porous solid-state materials. Mesoporous silica and organosilica materials prepared by LC templating with polymeric surfactants have been the subject of widespread research in materials chemistry.[15] The use of the LC phase of NCh to template these materials may give films with pore structures that replicate the nemat-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15148 – 15154
FULL PAPER ic order of the LC phase. Incorporation of ethylene bridges into the mesoporous organosilica film is expected to improve its mechanical flexibility compared with the pure silica analogue. The pioneering work of Alonso and Belamie[16] demonstrated the first examples of mesostructured silicas formed through NCh template-directed sol–gel condensation of SiACHTUNGRE(OC2H5)4 precursors; however, only poor order of these materials was observed. In addition, the method was not extended to organosilicas, which could lead to distinct materials with improved properties. Finding a new pathway to prepare replicas of the layered NCh structure in mesoporous silica and organosilica films remains a challenge. In this paper, we report the first experimental evidence for the replication of a nematic structure in mesoporous silica and ethylene-bridged organosilica films templated by NCh. Spindle-shaped chitin nanocrystals prepared by sequential deacetylation and hydrolysis of chitin fibrils isolated from king crab shells organized into layered nematic structures that functioned as templates for silica/NCh and organosilica/NCh composites. After removal of the chitin template, large, crack-free mesoporous silica and organosilica films with sizes of several centimeters were obtained. These mesoporous materials with layered organization and high specific surface area were replicated from NCh.
Figure 1. Nematic order of the NCh liquid crystals (NCh-4). a) TEM image of individual NCh fibers (scale bar, 2 mm). b) Left side, top to bottom: Vials of aqueous NCh suspensions with different concentrations at pH 4 and a NCh film prepared by drying the NCh suspension (6.6 wt %, pH 4; scale bar = 1 cm). Right side: An illustration of the layered structure of NCh. c) POM image of the NCh suspension showing birefringent textures characteristic of the nematic organization (scale bar = 100 mm). d) SEM image viewed along the edge of the NCh film at fractures (scale bar = 500 nm).
Results and Discussion King crab shells were treated with base and acid to remove protein and minerals, then bleached with a dilute H2O2 solution to remove pigment (Figure S1 a,b in the Supporting Information). The isolated chitin samples were then partially hydrolyzed in acid to produce nanocrystalline fragments.[14a] The acid-hydrolysis process also led to deacetylation of the chitin, resulting in free primary amine groups. The positive surface-charge density of the chitin crystallites in acidic media, which is related to the degree of deacetylation (DDA), is crucial in determining the organization of the ordered nematic LC phase. In this work, we set out to find new reaction conditions, using either hydrolysis or sequential deacetylation and hydrolysis, to prepare the nematic LC phase of the nanocrystalline chitin. NCh samples were prepared by hydrochloric-acid-catalyzed hydrolysis of the isolated chitin fibrils at 104 8C for different lengths of time (2, 10, 18 h). For comparison, a NCh sample was prepared by deacetylation of the chitin fibrils at 90 8C for 2 h in alkaline solution followed by hydrochloric-acid-catalyzed hydrolysis at 104 8C for 18 h (see Table S1 and Figure S1 c in the Supporting Information). The NCh samples prepared by onestep and two-step processes have a similar chitin weight loss ( 12 wt. %), suggesting that little degradation of the chitin fibrils occurred during deacetylation. The NCh samples show an improved stability as suspensions in water with increasing DDA. The NCh suspensions typically formed gels at pH < 2.5 and at high concentrations ( 8.0 wt %), as shown in Figure 1 b. Transmission electron
Chem. Eur. J. 2013, 19, 15148 – 15154
microscopy (TEM) of the dilute NCh solution (0.0005 wt %) reveals that the chitin crystallites are discrete spindleshaped nanocrystals with diameters of 10–18 nm and lengths of 300–500 nm (Figure 1 a). The acid and/or base treatments for the different heating durations did not cause a significant change in the crystallite size or shape. The nanocrystalline chitin structure was confirmed by Fourier-transform infrared (FTIR) spectroscopy (Figure S2 a in the Supporting Information) and powder X-ray diffraction (PXRD) analyses (Figure S2 b in the Supporting Information), indicating that the deacetylation and hydrolysis did not modify the original crystalline structure of a-chitin. The resulting chitin nanocrystals show good crystallinity after removal of non-crystalline fractions during hydrolysis. These data confirm that the spindle-shaped nanocrystals with partial deacetylation are nanoscale fragments of the chitin microfibrils. A nematic LC phase of NCh can form when the critical concentration and appropriate pH of the NCh suspension are reached. We found that the nematic organization of the prepared nanocrystalline chitin forms at a concentration of 6.6 wt % and pH 4. Figure 1 and Figures S3 and S4 in the Supporting Information show photographs, polarized optical microscopy (POM) images, and scanning electron microscopy (SEM) images of the aqueous NCh suspensions and selfassembled NCh films. The acid-catalyzed hydrolysis of the chitin fibrils gave NCh-1 at 2 h and NCh-2 at 10 h, which after drying at ambient conditions, gave NCh films consisting only of randomly oriented chitin nanocrystals (no evidence for long-range order); consequently, poor birefrin-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15149
M. J. MacLachlan et al.
gence was observed (Figure S3 a,b in the Supporting Information). This is in contrast to literature reports of NCh suspensions that showed fingerprint textures observed under crossed polarizers, characteristic of chiral nematic organization.[14a] As the hydrolysis time was extended to 18 h, birefringence of NCh-3 was enhanced, providing evidence for the alignment of NCh (Figure S3 c,d in the Supporting Information). The absence of fingerprint textures observed under POM indicates that nematic organization rather than chiral nematic order was present. In addition, we did not observe iridescence that is often associated with chiral nematic order. A NCh sample (NCh-4) prepared by sequential deacetylation (90 8C for 2 h) and hydrolysis (104 8C for 18 h) exhibits strong birefringence when observed under crossed polarizers (Figure 1 c and Figure S4 b,c in the Supporting Information). SEM images (Figure 1 d and Figure S4 d,e in the Supporting Information) show the morphology of the NCh films prepared from an aqueous NCh-4 suspension. A nematic organization is apparent from the image rather than the BouACHTUNGREliACHTUNGREgand structure of the chitin fibrils in the cuticles of the king crabs.[11] In the image, the chitin nanocrystals are oriented parallel to each other, which is likely directed by surface charge interactions of the crystallites. Edge and cross-sectional views of the NCh film show a layered structure. We found the nematic structure of the LC chitin produced from the king crab shells was more ordered than from other crab specimens, and chitin nanocrystals from shrimp shells gave especially poor ordering in our hands. The nematic order of the NCh suspension is related to the protonation of the exposed surface amino groups. The surface potential for the dilute NCh suspension (0.01 wt %, pH 4) was determined by zeta potential measurements (see Table S1 in the Supporting Information). The zeta potential reflects the dependence of the ionic strength of the suspension on the formation of an electrical double layer at the NCh surface. This value is considered as a proportional function of the charge on the NCh surface.[17] The respective zeta potentials of NCh-1, NCh-2, and NCh-3 prepared by acid hydrolysis were found to be 0.6, 2.3, and 10.3 mV, respectively. There was a slight increase of the zeta-potential value when the hydrolysis time was extended, probably due to an increase of DDA that would give more exposed surface amino groups, favoring protonation. This additional surface charge appears to result in an improved alignment of the chitin nanocrystals. The one-step acid hydrolysis of the chitin fibers results in two reactions: depolymerization and deacetylation.[18] The depolymerization occurred preferentially over deacetylation, thus frequently generating chitin nanocrystals with fewer exposed amino groups and corresponding low surface-charge density. This could inhibit the organization of the chitin nanocrystals into an ordered structure even with extended hydrolysis time. This poor ordering was overcome by performing the two-step process, deacetylation followed by hydrolysis, to produce NCh-4 that has a zeta potential of 43.1 mV, much higher than those of the samples prepared by
15150
www.chemeurj.org
a one-step process. This is consistent with an increasing number of exposed surface amino groups. The highly positive surface-charge density of NCh-4 derived from the protonation of the surface amino groups resulted in an improved organization in the films. However, the layered ordering was only found to arise under a narrow range of reaction conditions tested (deacetylation at 90 8C for 2 h followed by hydrolysis at 104 8C for 18 h yielded NCh-4). We varied both the duration of the deacetylation and the alkaline concentration, but found that other conditions led only to poor organization. The LC phases of NCh were used to template silica and organosilica materials through evaporation-induced self-assembly of NCh with corresponding tetramethylorthosilicate (TMOS) and 1,2-bis(trimethoxysilyl)ethane (BTMSE) precursors. The silica-based precursors were mixed with the NCh suspensions (6.6 wt %, pH 4) with stirring at room temperature and underwent hydrolysis and condensation to form homogeneous suspensions within 1 h. The composite suspensions were transferred to 60 mm diameter Petri dishes and dried under ambient conditions for 48 h to yield silica/NCh and organosilica/NCh composite films. We observed that the hierarchical organization of the LC NCh template employed is transferred to the composites. Thus, when the suspensions of NCh-1, NCh-2, or NCh-3 having poor ordering were used as templates, the silica films obtained after removal of the chitin by calcination have inhomogeneous layered structures with poor orientation and exhibit low optical anisotropy under POM (Figure S5 in the Supporting Information). Even though NCh-3 exhibits birefringence, the resulting silica films still have a disordered layered structure (Figure S6 in the Supporting Information). The ordered nematic organization of the NCh-4 template is expected to produce silica-based materials with better ordering than NCh-1, NCh-2, and NCh-3. Composites were prepared by LC templating with NCh-4 by using different ratios of the silica-based precursors to NCh (see Table S2 in the Supporting Information). POM images (Figure S7 a,b in the Supporting Information) of the homogeneous suspensions of silica/NCh and organosilica/NCh composites still exhibit birefringence, revealing that the precursors do not disrupt the alignment of NCh. The birefringent textures of these composite suspensions are analogous to those of the aqueous liquid-crystalline suspension of smectite clay.[19] Drying the composite suspensions gave large, crack-free silica/NCh and organosilica/NCh composite films with sizes of several centimeters (Figure 2 a,d) that show birefringence under POM, similar to that observed in the NCh-4 films (Figure 2 e and Figure S7 c in the Supporting Information). Layered structures originating from the alignment of NCh were observed by SEM images of the composites (Figures S8 and S9 in the Supporting Information). In contrast to the pure NCh films, the composites could not be dispersed in water. Calcination of the silica/NCh composites at 540 8C under air and treatment of the organosilica/NCh composites in H2SO4 (6 m) at 90 8C for 18 h successfully removed the chitin
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15148 – 15154
Mesoporous Silica and Organosilica Films
Figure 2. Silica/NCh (S7) and organosilica/NCh (OS2) composites before and after removal of the chitin template. Photographs of a) the silica/ NCh composites and b) the calcined silica film. c) POM image of the calcined silica films. d) Photograph and e) POM image of the organosilica/ NCh composites (UBC logo is shown to demonstrate transparency of the film). f) Photograph of the large, crack-free organosilica film after sulfuric acid hydrolysis of the chitin.
template, yielding the corresponding mesoporous silica and organosilica films. We previously developed an acid-hydrolysis approach to remove nanocrystalline cellulose from chiral nematic organosilica/cellulose composites to generate iridescent mesoporous organosilica films.[13c] This same procedure was used to remove the chitin template in the organosilica/ NCh composites. We observed that the decomposition of the chitin in H2SO4 (6 m) at 90 8C produces a byproduct that stained the resulting films brown. These byproducts were eliminated by treatment with a H2O2/H2SO4 mixture to give colorless films. After chitin removal, large, crack-free films were recovered with shapes similar to the original composites and dimensions of several centimeters (Figure 2 f). The silica films are transparent (Figure 2 b), but the organosilica films appear opaque after chitin removal. Complete removal of the chitin template in the silica and organosilica films was confirmed using a variety of techniques. For the silica/NCh composites, IR bands associated with vibrational modes of NCh (e.g., n˜ sACHTUNGRE(N H)) are absent in the silica after chitin removal (Figure S10 a,b in the Supporting Information). Thermogravimetric analyses (TGA) show that the decomposition of the chitin ( 75 wt %) in the composites occurred at ~ 150–600 8C compared to 100 wt % loss for the pure NCh film (Figure S10 c,d in the Supporting In-
Chem. Eur. J. 2013, 19, 15148 – 15154
FULL PAPER formation). PXRD patterns show that the silica/NCh composites have peaks characteristic of the crystalline chitin, while the calcined silica obtained after chitin removal has only a broad peak at 228 2q attributed to amorphous silica (Figure S10 e,f in the Supporting Information). Elemental analysis indicated that the silica/NCh composites contain 37 wt % carbon, but that carbon was absent in the corresponding calcined silica (referred to as S7). These results are consistent with the complete removal of the chitin in the composites to yield pure silica films. For the organosilica films (OS2), solid-state 13C cross-polarization/magic-angle spinning (CP/MAS) NMR spectra (Figure 6 a) of the organosilica/NCh composites before and after sulfuric acid hydrolysis of the chitin show that the chitin was removed without damaging the organosilica backbone. The composites have peaks at 20–180 ppm assigned to the chitin and a single peak at 5 ppm assigned to ethylene (SiCH2CH2Si) carbons of the organosilica.[13c] After acid hydrolysis treatment, only the single peak of the organosilica remains, demonstrating that the chitin was completely removed. Solid-state 29Si CP/MAS NMR spectra (Figure S11 in the Supporting Information) of the composites before and after chitin removal show peaks at 65 and 57 ppm assigned to corresponding T3 (CSiACHTUNGRE(OSi)3) and T2 (CSiACHTUNGRE(OSi)2OH) Si atoms of the ethylene-bridged organosilica.[13c] Signals expected for SiO4 species between 90 and 120 ppm are not observed, demonstrating that no cleavage of Si C bonds occurred during acid hydrolysis. TGA analysis (Figure S12 in the Supporting Information) shows that the organosilica is stable up to 400 8C, above which the ethylene bridges (10–15 wt %) decomposed. In contrast to the calcined silica, elemental analysis indicated that the organosilica contains 14.7 wt % carbon and 4.6 wt % hydrogen. In comparison, TGA and elemental analysis of a silica sample prepared by removal of the chitin in the silica/NCh composites using the exact same procedure of the organosilica preparation (heating the composites in H2SO4 (6 m) at 90 8C for 18 h) show that no residual organic species were present. This further confirms the acid hydrolysis conditions employed to be sufficient for the complete degradation and removal of the chitin, while preserving the ethylene bridges in the organosilica films. POM images (Figure 2 c) of the silica films show a significant difference in birefringence compared with the silica/ NCh composites. After chitin removal, the birefringent textures of the silica films shift from sepia to sepia–violet colors. The organosilica films exhibit low optical anisotropy due to their opacity. In contrast with the brittle silica films, the ethylene-bridged organosilica films are quite flexible. The layered organization in the silica and organosilica films is evident in SEM images shown in Figures 3 and 4, respectively. Cross-sectional views perpendicular to the flat surfaces indicate that the silica and organosilica films both have layered structures throughout their entire thickness (Figures 3 a and 4 a). Electron microscopy images of edges at high magnification reveal that there are spindle-like features in each layer (Figures 3 b,c and 4 b,c). These spindle-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15151
M. J. MacLachlan et al.
During templating of the silica-based precursors with NCh, the organization of the LC chitin was transferred to the silica and organosilica, and the nanometer-scale order was maintained during calcination and chitin hydrolysis. The porosity of the silica and organosilica films was determined by nitrogen adsorption/desorption studies. Figures 5 a,b and 6 b show that the silica and organosilica films have
Figure 3. SEM images of the calcined silica films with layered ordering (S7). a) Edge view of the films at fractures showing a relatively smooth top surface (scale bar = 50 mm). b) Cross-sectional view of the film at fractures showing a layered structure (scale bar = 5 mm). c) Higher magnification showing each layer composed of numerous spindle-shaped pores (scale bar = 1 mm). d) Top view showing a flat surface constituted of the alignment of the spindlelike features (scale bar = 2 mm). Figure 5. Nitrogen and solvent adsorption of layered mesoporous silica films (S7). a) Typical type-IV adsorption/desorption isotherms characteristic of the mesoporous structure. b) BJH adsorption of pore size distribution. c) TEM image of S7 (scale bar = 100 nm). d) Photograph of S7 after placing a water droplet (scale bar = 1 cm). e) POM image of S7 after adsorbing water showing a disappearance of birefringence in the wet part of the film at the right side compared with the dried part at the left side (scale bar = 100 mm).
Figure 4. SEM images of the organosilica films with layered ordering (OS2) after sulfuric acid hydrolysis of the chitin. a) Edge views of the films at fractures at a) low (scale bar = 50 mm) and b) high (scale bar = 5 mm) magnifications showing relatively smooth top surfaces and a layered structure. c) Cross-sectional view showing spindlelike features in each layer (scale bar = 2 mm). d) Top view showing a flat surface with aligned spindlelike features (scale bar = 2 mm).
like features are organized with long-range ordering, as observed by viewing the flat top surface of the films (Figures 3 d and 4 d). These images reveal that the silica and ethylene-bridged organosilica films both accurately replicated the nematic organization of NCh-4 in a layered structure.
15152
www.chemeurj.org
typical type-IV isotherms with type-H2 hysteresis, which indicate the introduction of mesoporosity into the silica-based solids. The porosity of these materials can be varied by changing the concentration ratios of the silica and organosilica sol–gel precursors to NCh used in the preparations (see Table S2 in the Supporting Information). Brunauer– Emmett–Teller (BET) surface areas and pore volumes were found to be 420–650 m2 g 1 and 0.48–0.70 cm3 g 1, respectively, for the silica, and 690–800 m2 g 1 and 0.32–0.40 cm3 g 1, respectively, for the organosilica samples. A decrease in the pore wall thickness for the silica and organosilica prepared using the lower precursor loadings could result in the partial collapse of the network upon calcination and hydrolysis. The use of higher precursor loadings gave thicker walls, probably preventing connectivity between pores (Figure S13 in the Supporting Information). The pore size distribution of the mesoporous organosilicas determined by BJH (Barrett–Joyner–Halenda) analysis is broad with peaks at 9–14 nm (inset of Figure 6 b), which is considerably larger than the 4–7 nm pore diameters of the mesoporous silicas generated by calcination (Figure 5 b). The pore size distribution of the organosilica is close to the diameter range ( 10–18 nm) of the chitin template observed by TEM. Pore diameters of the organosilicas are
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 15148 – 15154
Mesoporous Silica and Organosilica Films
FULL PAPER of birefringence. After drying the films, they returned to their original states and the structural birefringence resulting from alignment of the pores fully recovered. In contrast, the silica/NCh and organosilica/NCh composites, which are nonporous, showed no significant change in their birefringence upon addition of water. Other solvents, such as ethanol and acetone, were also effectively absorbed by the layered mesoporous silica and organosilica films.
Conclusion
Figure 6. Nitrogen and solvent adsorption of layered mesoporous organosilica films (OS2). a) Solid-state 13C NMR spectra of the organosilica/ NCh composites before and after sulfuric acid hydrolysis of the chitin. b) Nitrogen adsorption/desorption isotherms of OS2; inset of BJH adsorption of pore size distribution. c) TEM image of OS2. d) Photograph of OS2 after placing a water droplet in the center of the film, leading the wet part of the film to be transparent.
larger than those of the calcined silica films, indicating that there is less pore shrinkage from the acid-catalyzed hydrolysis of the chitin template than from the calcination. Thus, the pore sizes of the mesoporous silica and organosilica can be tuned by choosing the conditions to remove the chitin template. It is worth noting that the pore sizes of the mesoporous materials prepared using NCh as a template are larger than those prepared using NCC as a template,[13a,c] consistent with their relative diameters. This highlights the nanoscale control over porosity afforded by using biologically derived, nanocrystalline templates. TEM images of the mesoporous silica (Figure 5 c) and organosilica (Figure 6 c) confirmed the formation of the aligned porous networks within the films, where the mesoporous channels run parallel to a preferred orientation within each layer. Spindle-shaped pores were observed consistent with the layered organization of the mesoporous structures. The pore diameters observed from the TEM images are around 4–8 nm for the silica and 10–15 nm for the organosilica in accordance with the nitrogen adsorption data. The layered films with partially aligned pore channels yielded strong birefringence in the case of the silica. The order and mesoporosity of the silica and organosilica replicas were further elucidated by solvent absorption into the films. When water droplets were placed on the silica and organosilica films, water was rapidly absorbed and both films immediately appeared transparent, as shown in Figures 5 d and 6 d. Examination of these films under POM (Figure 5 e) shows a disappearance of birefringence. These results show that water fills the pores of the silica and organosilica, not only the surfaces, and its matching of the refractive index of the silica and organosilica leads to nullification
Chem. Eur. J. 2013, 19, 15148 – 15154
In conclusion, mesoporous silica and organosilica films with layered structures originating from the nematic LC phase of NCh have been prepared. The nematic LC phase of chitin prepared using the two-step process of sequential deacetylation and hydrolysis was used to template silica/NCh and organosilica/NCh composites. The chitin template can be removed by calcination in the case of the silica/NCh composites, or by acid-catalyzed hydrolysis in the case of the organosilica/NCh composites, giving mesoporous silica and organosilica films, respectively. Large, crack-free silica and organosilica films retained the shapes of the precursor composites and had sizes of several centimeters. The silica and organosilica replicas show layered nematic organization and mesoporosity arising from a transfer of the LC phase of the chitin. The use of renewable nanocrystalline chitin as a template allows for the large-scale preparation of the ordered mesoporous materials. A further research direction could be to use these materials as a hard template to replicate the pore organization into other materials. These novel materials might open an opportunity for useful applications in optical elements, catalyst supports, and adsorbents.
Experimental Section Chitin purification: Dried king crab shells (25 g) were soaked in an aqueous solution of NaOH (5 wt %, 500 mL) at 80 8C for 6 h to remove protein. The products were cooled to room temperature and washed thoroughly with water. Next, the shells were treated with an aqueous solution of HCl (7 vol %, 500 mL) for 4 d at room temperature to remove calcium carbonate minerals. After collecting the products and rinsing with abundant water, the samples were treated with a NaOH solution (5 wt %, 500 mL) at 80 8C for 8 h to completely remove residual protein. The ratio of the shells to the NaOH (5 wt %) or HCl (7 vol %) solution was 1 g per 20 mL. The products were then washed thoroughly with water and remaining pigments were oxidized with an aqueous solution of H2O2 (5 vol %, 500 mL) at 90 8C for 1 h. The fully bleached samples were collected and washed several times with distilled water followed by drying in an oven at 40 8C overnight to obtain white chitin. The extraction yield of the chitin from the king crab shells was about 13 wt % following the above procedure. Preparation of aqueous NCh suspensions: Acid-catalyzed hydrolysis: Dried chitin samples (2.5 g) were cut into small flakes ( 1 cm) and treated with a HCl solution (4 m, 50 mL) in a round-bottom flask equipped with a reflux condenser. The reaction mixture was heated at 104 8C with vigorous stirring for different lengths of time (2, 10, 18 h) to obtain corresponding samples NCh-1, NCh-2, and NCh-3. Dark-brown products that formed during the process were oxi-
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
15153
M. J. MacLachlan et al.
dized by injecting H2O2 (2.5 mL, 30 vol %) into the hot reaction mixture at the end of the hydrolysis and keeping them at the same temperature for 30 min. After acid hydrolysis, the reaction mixture was diluted with 100 mL of distilled water followed by centrifugation (at 2200 rpm for 10 min) to remove soluble chitin species. This step was repeated two more times to produce an aqueous chitin suspension with pH 2. The suspension was transferred into dialysis bags and dialyzed against distilled water for 48 h until the pH of the suspension increased to about 4. The dialysis bags were then immersed in polyethylene glycol 400 for 3 h to obtain an appropriately concentrated sample that would favor the formation of a nematic LC phase. To increase the stabilization of the nanocrystalline chitin colloids, the suspension (40 mL) was dispersed with ultrasound treatment for 1 h using a Biosonic ultrasonic (50 Hz, 100 W). Alkaline-treated deacetylation and acid-catalyzed hydrolysis: The stabilization of the aqueous NCh suspension, which is affected by protonation of the exposed surface amino groups, is responsible for the organization of the individual chitin nanocrystals into a nematic LC phase. Hence, a two-step process of sequential deacetylation and hydrolysis of the isolated chitin was carried out. Typically, the isolated chitin flakes (2.5 g) were treated with NaOH solution (25 mL, 33 wt %) at 90 8C for 2 h with vigorous stirring. The reaction mixture was cooled to room temperature, centrifuged, and neutralized by washing with abundant water to obtain nanocrystalline chitin sample with partial deacetylation. The sample was then treated with HCl solution (50 mL, 4 m) at 104 8C with vigorous stirring for 18 h to obtain NCh-4. The sample was treated with H2O2 and isolated as described above. Preparation of mesoporous silica and organosilica films from the nematic NCh template: For the preparation of the mesoporous silicas, the aqueous NCh suspension (15 mL, 6.6 wt % NCh, pH 4) was sonicated for 10 min using an Aquasonic 50T sonic cleaner. An appropriate amount of the silica precursor (TMOS) was mixed with the NCh suspension (15 mL, 6.6 wt %) with stirring at room temperature for 1 h. The silica sol–gel concentration was varied corresponding to samples (S1 to S9), as presented in Table S2 in the Supporting Information. Portions (3.6 mL) were then transferred to 60 mm diameter polystyrene Petri dishes and dried at ambient conditions within 48 h to give silica/NCh composites. For removal of the chitin, the silica/NCh composites were calcined under flowing air at 540 8C for 6 h to recover silica films (47 mg of S7 was obtained from 230 mg of silica/NCh composites). For the preparation of the mesoporous organosilicas, the organosilica precursor (BTMS) was used instead of TMOS to produce organosilica/NCh composites under the exact same procedure. The respective amounts of BTMS used for the preparation of samples was 120 mg (OS1) and 240 mg (OS2). For removal of the chitin, the organosilica/NCh composites were placed in H2SO4 (900 mL, 6 m) and heated to 90 8C for 18 h. The products were then filtered and soaked in a piranha solution (20 mL 30 % H2O2/100 mL H2SO4) for 1 h to give colorless films. The films were then washed copiously with distilled water and dried at ambient conditions to recover organosilica films (91 mg of OS2 was obtained from 230 mg of organosilica/NCh composites).
Acknowledgements We are grateful to the Natural Sciences and Engineering Research Council (NSERC) of Canada for funding.
[1] a) C. G. Goltner, M. Antonietti, Adv. Mater. 1997, 9, 431 – 436; b) M. E. Raimondi, J. M. Seddon, Liquid Crystals 1999, 26, 305 – 339; c) M. Antonietti, Phil. Trans. R. Soc. A 2006, 364, 2817 – 2840. [2] a) C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli, J. S. Beck, Nature 1992, 359, 710 – 712; b) Y. Wan, D. Zhao, Chem. Rev. 2007, 107, 2821 – 2860; c) A. Firouzi, F. Atef, A. G. Oertli, G. D. Stucky, B. F. Chmelka, J. Am. Chem. Soc. 1997, 119, 3596 – 3610.
15154
www.chemeurj.org
[3] a) H. Zhu, D. J. Jones, J. Zajac, J. Roziere, R. Dutartre, Chem. Commun. 2001, 2568 – 2569; b) F. Hoffmann, M. Cornelius, J. Morell, M. Froba, Angew. Chem. 2006, 118, 3290 – 3328; Angew. Chem. Int. Ed. 2006, 45, 3216 – 3251. [4] a) G. S. Attard, P. N. Bartlett, N. R. B. Coleman, J. M. Elliott, J. R. Owen, J. H. Wang, Science 1997, 278, 838 – 840; b) S. R. J. Oliver, G. A. Ozin, J. Mater. Chem. 1998, 8, 1081 – 1085; c) P. V. Braun, P. Osenar, V. Tohver, S. B. Kennedy, S. I. Stupp, J. Am. Chem. Soc. 1999, 121, 7302 – 7309; d) G. S. Armatas, M. G. Kanatzidis, Science 2006, 313, 817 – 820. [5] a) V. Percec, Phil. Trans. R. Soc. A 2006, 364, 2709 – 2719; b) S. T. Tolbert, A. Firouzi, G. D. Stucky, B. F. Chmelka, Science 1997, 278, 264 – 268; c) P. Feng, X. Bu, D. J. Pine, Langmuir 2000, 16, 5304 – 5310. [6] a) C. Sanchez, H. Arribart, M. M. G. Guille, Nat. Mater. 2005, 4, 277 – 288; b) J. Aizenberg, J. C. Weaver, M. S. Thanawala, V. C. Sundar, D. E. Morse, P. Fratzl, Science 2005, 309, 275 – 278. [7] a) W. J. Chung, J. W. Oh, K. Kwak, B. Y. Lee, J. Meyer, E. Wang, A. Hexemer, S. W. Lee, Nature 2011, 478, 364 – 368; b) H. Zhou, T. Fan, D. Zhang, ChemSusChem 2011, 4, 1344 – 1387. [8] a) F. Castles, F. V. Day, S. M. Morris, D. H. Ko, D. J. Gardiner, M. M. Qasim, S. Nosheen, P. J. W. Hands, S. S. Choi, R. H. Friend, H. J. Coles, Nat. Mater. 2012, 11, 599 – 603; b) B. F. G. Johnson, S. A. Raynor, D. S. Shephard, T. Mashmeyer, J. M. Thomas, G. W. J. Chung, J. W. Oh, K. Kwak, B. Y. Lee, J. Meyer, E. Wang, A. Hexeme Sankar, S. Bromley, R. Oldroyd, L. Gladden, M. D. Mantle, Chem. Commun. 1999, 1167 – 1168; c) K. Kosuge, S. Kubo, N. Kikukawa, M. Takemori, Langmuir 2007, 23, 3095 – 3102. [9] a) A. C. Neville, Biology of the Arthropod Cuticle, Springer, New York, 1975, pp. 159 – 234; b) S. J. Woltman, G. D. Jay, G. P. Crawford, Nat. Mater. 2007, 6, 929 – 938. [10] H. Merzendorfer, J. Comp. Physiol. B 2006, 176, 1 – 15. [11] a) Y. BouACHTUNGREliACHTUNGREgand, Tissue Cell 1972, 4, 189 – 217; b) P. Y. Chen, A. Y. M. Lin, J. McKittrick, M. A. Meyers, Acta Biomater. 2008, 4, 587 – 596. [12] a) T. Lenau, M. Barfoed, Adv. Eng. Mater. 2008, 10, 299 – 314; b) V. Sharma, M. Crne, J. O. Park, M. Srinivasarao, Science 2009, 325, 449 – 451. [13] a) K. E. Shopsowitz, H. Qi, W. Y. Hamad, M. J. MacLachlan, Nature 2010, 468, 422 – 425; b) J. A. Kelly, M. Yu, W. Y. Hamad, M. J. MacACHTUNGRELachlan, Adv. Opt. Mater. 2013, 1, 295 – 299; c) K. E. Shopsowitz, W. Y. Hamad, M. J. MacLachlan, J. Am. Chem. Soc. 2012, 134, 867 – 870; d) A. S. Terpstra, K. E. Shopsowitz, C. F. Gregory, A. P. Manning, C. A. Michal, W. Y. Hamad, J. Yang, M. J. MacLachlan, Chem. Commun. 2013, 49, 1645 – 1647. [14] a) J. F. Revol, R. H. Marchessault, Int. J. Biol. Macromol. 1993, 15, 329 – 335; b) J. Li, J. F. Revol, R. H. Marchessault, J. Appl. Polym. Sci. 1997, 65, 373 – 380. [15] a) H. Yang, N. Coombs, I. Sokolov, G. A. Ozin, Nature 1996, 381, 589 – 592; b) G. S. Attard, J. C. Glyde, C. G. Goltner, Nature 1995, 378, 366 – 368. [16] a) B. Alonso, E. Belamie, Angew. Chem. 2010, 122, 8377 – 8380; Angew. Chem. Int. Ed. 2010, 49, 8201 – 8204; b) E. Belamie, M. Y. Boltoeva, K. Yang, T. Cacciaguerra, B. Alonso, J. Mater. Chem. 2011, 21, 16997 – 17006. [17] J. Li, J. F. Revol, E. Naranjo, R. H. Marchessault, Int. J. Biol. Macromol. 1996, 18, 177 – 187. [18] A. Einbu, K. M. Varum, Biomacromolecules 2008, 9, 1870 – 1875. [19] L. J. Michot, I. Bihannic, S. Maddi, S. S. Funari, C. Baravian, P. Levitz, P. Davidson, Proc. Natl. Acad. Sci. USA 2006, 103, 16101 – 16104.
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Received: May 20, 2013 Published online: September 25, 2013
Chem. Eur. J. 2013, 19, 15148 – 15154
