1. Introduction

Hydrogel is a typical soft material that has a unique, threedimensionally cross-linked network structure swollen with a large amount of water. Basically, cross-linkings may be divided into two different types, i.e., chemical (covalent bonds) and physical (ionic bonds, hydrogen bonds, etc.). In the past decades, hydrogels have received increasing attention owing to a wide range of potential applications, including tissue engineering, drug delivery, membrane separation, electrolytes and soft robot [1-15]. However, most of the existing natural and synthetic hydrogels have poor mechanical properties, which unfortunately limit their applications in many fields where tough and flexible hydrogels are necessary. In this sense, a surge of research has been reported on the development of nanocomposite hydrogels reinforced with various nanofillers, including clay nanosheets [16], silica nanoparticles [6, 9, 17], graphene oxide (GO) nanosheets [7, 18-22] and carbon nanotubes [23]. Recently, we have prepared a high-strength GO/PAA nanocomposite hydrogels composed of a single network with dual crosslinking points through dynamic ionic interactions [7]. GO nanosheets are not only nanofiller with extraordinary mechanical strength and large surface area, but also serve as analogous crosslinking points [4-9, 15] in the gel network, which cooperate with reversible ionic cross-linking points among the polymer chains to endow the gel with super toughness and stretchability.

However, like most of nanofillers, GO is optically opaque, sacrificing the transparency of the resulting nanocomposite hydrogels and thus limiting its application range. In this regard, hexagonal boron nitride (h-BN), as an inorganic analogue of graphene, is an ideal alternative since it has comparable mechanical properties yet better optical properties than GO and most of other nanofillers [18, 24]. To the best of our knowledge, however, the preparation of hBN reinforced hydrogels remains untouched, probably owing to the stronghydrophobicity of h-BNnanosheetsconventionally exfoliated in organic solvents by sonication.

In our previous work, we have successfully developed a simple and green method to prepare transparent h-BN/PVA nanocomposites in aqueous solutions [25]. Direct sonication of h-BN powder in water can produce hydrophilic h-BN nanosheets functionalized with -OH groups [26], which lays a basis for the interactions between h-BN nanosheets and the hydrophilic polymer chains, such as polyacrylamide (PAM). Therefore, h-BN nanosheets can take the place of GO nanosheets as the analogous crosslinking points in the nanocomposite hydrogels. Inspired by this pioneer work, here we propose a new trial to prepare high-strength, transparent h-BN/PAM nanocomposite hydrogels. The h-BN/PAM nanocomposite hydrogels, in contrast to the dark GO/PAM nanocomposite hydrogels [19, 21, 22], are highly transparent. Moreover, the transparent h-BN/PAM nanocomposite hydrogels have significantly enhanced tensile and compressive properties compared to neat PAM hydrogels, thus opening up new opportunities for developing next-generation transparent, tough and flexible hydrogels.

2. Experimental

Typically, h-BN nanosheets were exfoliated from h-BN powder by direct sonication in deionzied water (initial concentration=2 mg mL^-1) at 300 W for 8 h, followed by centrifugation at 9000 rpm for 40 min. The top 2/3 supernatant was collected, whose concentration was determined to be 0.36 mg mL^-1 by drying a fixed volume and weighing the remaining solid. Then, 9 mg of N, N0-methylenebisacrylamide (BIS) was added to 30 mL of supernatant. The mixed solution was subjected to gentle sonication for 30 min and cooled to ambient temperature, to which 9 g of acryl amide and 45 mg of ammonium persulfate (APS) were added. The mixture was stirred at ambient temperature for 30 min, evacuated, and left for reactions at 28±1 ℃ under N₂ protection for 24 h to obtain h-BN/PAM nanocomposite hydrogels where the h-BN content was 0.12 wt%. To obtain h-BN/PAM nanocomposite hydrogels with different h-BN contents, the supernatant was diluted accordingly while the other procedures were kept unchanged. The chemically cross-linked PAM hydrogel is fabricated in similar procedure but without h-BN.

UV-vis spectra was recorded by a Pgeneral TU-1810 twin-beam spectrophotometer from 300 nm to 900 nm. DSC was recorded by a Shimadzu DSC-60 from ambient temperature to 140 ℃ at a heating rate of 10 ℃ min^-1 in nitrogen (40 mL min^-1). SEM was carried out by a Tescan VEGA3 operated at an accelerating voltage of 10 kV. HRTEM was carried out by a JEOL JEM-2010 operated at an accelerating voltage of 120 kV. The tensile and compressive tests were performed by a Zwick 005 Materials tester. Fourier transform infrared (FT-IR) spectra was performed by a Thermo Scientific Nicolet 6700 by the KBr pellet method in the range of 500-3500 cm^-1 with a resolution of 4 cm^-1.

3. Results and discussion

Fig. 1a shows a TEM image of the obtained few-layer h-BN nanosheets. It can be clearly seen that the edge is curled up, indicating the h-BN nanosheets are very thin and flexible. Fig. 1b shows an AFM image of the h-BN nanosheets spin-coated on mica, which reveals individual nanosheets with lateral sizes of~500 nm and thickness of≤1.4 nm, also indicating the presence of few-layer h-BN nanosheets.

As reported previously [25, 26], h-BN powder can be directly exfoliated in water by hydrolysis with the assistance of sonication, resulting in hydrophilic h-BN nanosheets with -OH groups. Thus the nanosheets can be easily dispersed in water. The FT-IR spectra in Fig. 2 give an evidence that a peak at~3400 cm^-1 appears clearly, which could be attributed to the stretching signal of O-H. This result is in good agreement with the previous papers [26, 27].

A comparison between the h-BN aqueous solution and deionized water under laser irradiation is shown in Fig. S1 in Supporting information, which displays an obvious Tyndall effect of the former proving that the h-BN nanosheets are uniformly dispersed in deionized water to form a colloidal solution. The superior transparency and mechanical properties of the obtained h-BN/ PAM nanocomposite hydrogels are revealed vividly by taking photos under different conditions, as shown in Figs. 3 and S2. As seen from Fig. S2 in Supporting information, the h-BN/PAM nanocomposite hydrogels are highly transparent even at the highest h-BN content (0.12 wt%), which is a huge advantage over the dark GO/PAM nanocomposite hydrogels [19, 21, 22]. It is found that the h-BN/PAM nanocomposite hydrogels (h-BN content=0.12 wt%) can easily be bent (Fig. 3a), knotted (Fig. 3b) and stretched (Fig. 3c). Here it should be noted that even in a knotted state, the h-BN/PAM nanocomposite hydrogels can still be stretched by more than 10 times (Fig. 3d), further demonstrating their ability to withstand high-level deformation resulting from their superior mechanical properties.

The transparency of the h-BN/PAM nanocomposite hydrogels with various h-BN contents is evaluated quantitatively by UV-vis spectroscopy, as shown in Fig. 4a. For comparison, the UV-vis spectrum of neat PAM hydrogels is also given. The transmittance values of these hydrogels at a wavelength of 520 nm are presented in Fig. 4b. It can be seen that at a wavelength of 520 nm, the PAM hydrogels have a transmittance as high as 97.5%. Though the introduction of the h-BN nanosheets inevitably sacrifices part of the transmittance, all the transmittance values of these hydrogels are higher than 85%, implying it is enough transparent for applications. As shown in Fig. 4, at a h-BN content of 0.06 wt%, the transmittance of the h-BN/PAM nanocomposite hydrogels is still above 90%. Even at the highest h-BN content (0.12 wt%), the h-BN/PAM nanocomposite hydrogels are still highly transparent with a transmittance of 85%. This merit, which is absent in the dark GO/PAM nanocomposite hydrogels [19, 21, 22], is fairly plausible since it may endow our h-BN/PAM nanocomposite hydrogels with promising potentials in light responsive soft robot [28], and liquid microlenses [29].

To illustrate the superior mechanical properties of the h-BN/ PAM nanocomposite hydrogels over neat PAM hydrogels, their stress-strain curves under tensile and compressive tests are showninFig. 5.FromFig. 5a, the neat PAM hydrogels have a tensile strength of only 112 kPa with an elongation at break of 850%. After the addition of the h-BN nanosheets, the elongation at break of all h-BN/PAM nanocomposite hydrogels increases to 1000% or so, directlyprovingthatthe h-BN nanosheets are dispersed in the PAM matrix in a uniform way with strong interfacial interactions. It could be ascribed to hydrogen bonding formed among -OH group on the end face of h-BN nanosheets and -NH2 group of PAM chains and the resulting h-BN nanosheets as analogous cross-linkings in the hydrogels. At a h-BN content of only 0.04 wt%, the tensile strength increases by~20% to 134 kPa. When the h-BN content is 0.12 wt%, the tensile strength is up to 160 kPa, an increase of~43% from that of the PAM hydrogels. Moreover, the h-BN/PAM nanocomposite hydrogels also significantly excel the PAM hydrogels in the compressive tests. It is obvious from Fig. 5b that the neat PAM hydrogels have a relatively low compressive strength of 361 kPa. When the h-BN nanosheets are introduced, the compressive strength increases to 422 kPa (0.04 wt%), and further to 600 kPa (0.12 wt%). These results are consistent with our previous reports [4-9, 15] that the nanomaterials with quantities of functional groups can work as analogous crosslinking points to sustain the increased stress and maintain the configuration of the gel network. Compared with the hydrogels composited with GO and nanoclay, the relatively limited improvement in mechanical properties of the gel by h-BN in this paper is mainly because the maximum content of the h-BN (0.12 wt%) is much lower than that of GO or nanoclay in GO/PAM hydrogels [21, 30] or nanoclay/PAM hydrogels [31, 32]. On the other hand, compared with GO or nanoclay, the lower density of -OH groups of h-BN nanosheets will result in a lower density of the analogous cross-linkings in the hydrogels and thus cause a relatively lower improvement in mechanical properties. However, compared with neat PAM hydrogels, the tensile strength of h-BN/PAM hydrogels is improved by 43% at maximum with only 0.12 wt% h-BN content. This result means a relatively high effect of enhancement with the introduction of h-BN nanosheets into the neat PAM hydrogels.

The morphological information of the h-BN/PAM nanocomposite hydrogels (h-BN content=0.12 wt%) and neat PAM hydrogels is provided by SEM observation, as shown in Fig. 6. As seen from this figure, introducing a small quantity of h-BN nanosheets does not significantly alter the network structure of the PAM matrix, in other words, the two types of hydrogels exhibit a similar network structure without distinguishable differences. The insert in Fig. 6b shows a high-magnification SEM image of the h-BN/PAM nanocomposite hydrogels, which clearly reveals a high porosity characteristic of the PAM hydrogels. The similar network structure results in considerable water absorbency, as shown in Fig. S4 in Supporting information. The swelling ratio of either hydrogels can reach an equilibrium of~12 times after 75 h.

4. Conclusion

In conclusion, transparent, tough and flexible dually crosslinked PAM hydrogels have successfully been prepared through chemical crosslinking of BIS and analogous crosslinking of hydrophilic h-BN nanosheets. The resulting h-BN/PAM nanocomposite hydrogels are highly transparent and have significantly enhanced tensile and compressive properties. Thus it provides new opportunities for achieving next-generation transparent, tough and flexible hydrogels that hold great promise in such important applications as light responsive soft robot and liquid microlenses etc.

Acknowledgments

This work was financially supported by NSFC (Nos. 21474058 and 21274079), the State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, Donghua University (Project No. LK1404) and Tsinghua University Scientific Research Project (No. 2014Z22069).

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cclet.2016.04.002.