1. Introduction

As color imaging takes on increasing importance in a plethora of applications such as displays, sensors, and communications, the ability to achieve high-quality color with excellent fidelity, high brightness and decent controllability has become an essential but elusive goal. It has been well known that naturally occurring or synthetic pigments can exhibit various colors as a result of the wavelength-selective absorption of light [1,2]. The various colors that appear are therefore dependent on the specific chemical compounds. Alternatively, structural color offers a physical way to realize visual colors, attributing to the direct interaction of visible light with microscopic structures whose feature size is on the wavelength scale [3–8]. One typical example of materials that exhibit structural color is photonic crystals (PCs), which are based on the periodic modulation of the refractive index [9–12]. Such periodic arrangements can Bragg diffract the incident light and stop a certain spectral range of electromagnetic waves from propagating through the PCs. If the stop-bands fall within the visible regime, these materials will exhibit unique structural colors. The most interesting characteristic of PCs is that one can use inherently colorless materials to generate all-spectral colors by simply tailoring the structural features such as dimension, shape or periodic spacing. This unique property allows PCs to play a significant role in a wide variety of visual fields, thereby attracting extensive interest throughout the optical community in the development of design methodologies and fabrication strategies for this important class of periodic structures [13–19].

Colloidal PCs, composed of a periodic arrangement of monodispersed nanoparticles, have emerged as one of the most promising PCs material systems due to the high degree of control over their long-range periodicity, which enables effective light manipulation [20–25]. Moreover, the bottom-up approach of particle self-assembly provides a viable pathway for obtaining high-quality PCs in a very economical way. The beauty of colloidal PCs has been successfully demonstrated by using a variety of colloidal particles that serve as building blocks, such as silica [26,27], polymethylmethacrylate (PMMA) [28,29] and polystyrene (PS) [30,31]. While some good results have been achieved, the photonic visibility of colloidal PCs could be further improved in terms of the brightness and contrast as well as the viewing angle. In fact, commonly used silica and most polymers have intrinsically low refractive indices, which seriously weaken the intensities of the stop-band and, as a result, lead to relatively low reflectivity and narrow bandwidth [32]. Additionally, the appearance of structural color is inevitably affected by the incoherent scattering of incident light, leading to the pale appearance and deterioration in visibility of colloidal PCs [33]. Therefore, to create the appearance of vivid colors, it is of vital importance to intelligently design the material and architecture of colloidal PCs to obtain the optimized photonic properties.

To achieve the above mentioned targets, the research methodologies must be developed to 1) create a more intense and complete photonic stop band, or/and 2) efficiently minimize (absorbing) the incoherent scattered light. Indeed, the intense stop band can be realized by arranging the colloidal particles with long-range structural order and regular periodicity [34], or by employing materials with high refractive index [35]. The incoherent scattering of light, on the other hand, can be diminished by uniformly incorporating light-absorbing materials within the colloidal PCs [36], without affecting the particle assembly. Therefore, it can be expected that visibility will be substantially improved if all of the above technical strategies are effectively combined to offer a synergistic optimization strategy to the design of colloidal PCs. However, it is very challenging to find a fabrication scheme that can provide a good balance between all of the above requirements. First, only a very limited number of materials, which simultaneously possess high refractive-index and transparency in the visible regime, are suitable for the formation of colloidal PCs. Besides, it is very difficult to employ such materials to realize highly monodispersed nanoparticles and highly ordered long-range arrangements. The challenges mainly arise from a variety of aspects, including the difficulty to effectively control the nucleation kinetics of particles, the lack of sufficient surface charges for electrostatic self-assembly of particles, and high particle density causing them to sediment [37]. Thus, the fabrication of high-quality colloidal PCs based on high refractive index materials still remains a challenging problem. Second, although it has been well demonstrated that black materials can absorb scattering light over a broad bandwidth and significantly enhance the structural color, the common methods to incorporate black materials into colloidal PCs are mainly based on physical mixing. However, this causes the random aggregation of carbon materials and their non-uniform distribution within the PCs. Thus, it is necessary to develop a simple and cost-effective process that can not only effectively implant black materials into PCs uniformly, but also have no influence on the formation of colloidal PCs with long-range order.

In this work, we report the creation of high-quality colloidal PCs through the successful formation of ordered long-range self-assembly of high refractive index ZnS@SiO₂ nanospheres. The schematic of the fabrication procedure is shown in Fig. 1. Specifically, to achieve the highly ordered long-range arrangement of nanospheres, the morphological and material characteristics of nanospheres have been fine-tuned through interface engineering, partial surface etching and high temperature calcination. Significantly, the carbon-based materials are successfully incorporated within ZnS@SiO₂ colloidal PCs by using a facile in situ approach, leading to their uniform distribution. Owing to the combination of high-quality colloidal nanospheres with both high refractive-index and excellent long-range order alignment, as well as the black-coated light absorption, the visibility of colloidal PCs has been significantly improved. The maximum reflectivity from the self-assembly PCs can reach 98%. The combined merits in terms of excellent visibility and simple fabrication presented in this work can greatly benefit the scalable production of high-quality colloidal PCs pigments.

 

Fig. 1 Schematic illustrating the procedure for preparing long-range ordered ZnS@SiO₂ colloidal PCs.

Download Full Size | PDF

2. Experimental section

2.1 Synthesis of ZnS colloidal nanospheres

ZnS nanospheres were prepared by modifying the method presented in the literature [38]. Typically, 0.8 mmol of Zn(Ac)₂·2H₂O, 20 mmol of thiourea and 0.2 g of PVP were dissolved in 20 mL of water to form a clear solution. The temperature of solution was increased to 100°C at 700 rpm stirring under refluxing conditions for a certain time. After the mixture was cooled to room temperature, a white product was obtained, which was then washed with water and re-dispersed in 20 mL of water. The reaction time determines the size of the ZnS nanospheres. For example, reaction time of 30 min, 1 h, 1.5 h and 2 h lead to ZnS nanospheres with diameters of 42 nm, 78 nm, 161 nm and 194 nm (See Fig. 2 for details), respectively.

 

Fig. 2 TEM images of as-prepared ZnS monodispersed particles prepared with reaction times of (a) 30 min, (b) 60 min, (c) 90 min, (d) 120 min, (e) 150 min, and (f) 180 min.

Download Full Size | PDF

2.2 Synthesis of ZnS@SiO₂ nanospheres

The as-prepared ZnS nanospheres were then coated with a thin silica layer by the sol-gel method [39,40]. In a typical process, 0.01 g of polyacrylic acid (PAA) was added to the dispersion of ZnS nanospheres and the system was stirred for 20 min. The nanospheres were collected by centrifugation, re-dispersed in 10 mL of ethanol, and then mixed with 4.3 mL of water, 13 mL of ethanol, 0.62 mL of ammonia (28%), and 0.2 mL of tetraethyl orthosilicate (TEOS). After the solution was stirred for 3 h, the ZnS@SiO₂ core/shell nanospheres were collected by centrifugation, washed with ethanol and water several times, and dried on vacuum.

2.3 Calcination and partial etching treatment of ZnS@SiO₂ nanospheres

ZnS@SiO₂ nanospheres were calcined at the desired temperatures (600 °C and 800 °C) under a nitrogen atmosphere for 2 h to crystallize the ZnS@SiO₂ nanospheres, and then etched in a NaOH solution (0.4 M) for 2 h to partially remove the SiO₂ layer, which leads to the nanosphere samples, named as ZnS@SiO₂-600 and ZnS@SiO₂-800, for next characterizations. After treatment, the solution was washed 5 times with water and re-dispersed in water. The resulting solution was centrifuged at 2000 rpm for 3 min to remove any aggregation. The supernatant was then centrifuged at 6000 rpm for another 5 min to separate the relatively smaller spheres. The precipitations obtained by the two separation processes were re-dispersed in water and the solutions were washed by deionized water.

2.4 Fabrication of 3D PC arrays

The 3D PCs were fabricated by using a modified fluidic cells approach [41,42]. In a typical process, a packing cell was constructed from two glass substrates and a square frame (cut with scissors) of Mylar film (32 um thick), and tightened with binder clips. A small hole (~2 mm in diameter) was drilled in the top glass substrate, and a glass tube (~5 mm in diameter) was glued to this hole using epoxy. When an aqueous dispersion of the monodispersed spheres was injected into this cell (through the glass tube), these beads were self-assembled into 3D arrays of ZnS@SiO₂ nanospheres in 5 days.

2.5 Reflectance measurements

An Ocean Optics HR2000CG-UV-NIR spectrometer coupled with a six-around-one reflection/backscattering probe was used to record the reflectance of photonic crystals. The probe was perpendicularly placed in front of the packing cell containing 3D PCs arrays. The incident light from six surrounding fibers illuminated the photonic crystals, and the reflected light was collected by a central fiber. The reflectance spectra were recorded by calibrating the spectra against a STAN-SSH high reflectivity specular reflectance standard.

3. Results and discussion

Zinc sulfide (ZnS), due to its high bulk refractive index (β-ZnS, n₅₈₉ = 2.36) and transparency in the visible regime, is a good candidate as a building block material for colloidal PCs [37,43]. Although the synthesis of ZnS nanospheres has long been studied, their employment in the construction of colloidal PCs with long-range order has seldom been explored. One work on this topic is from Moon Cyu Han et al. [37], who have successfully prepared monodispersed ZnS nanospheres to form colloidal PCs. In this work, we present an alternative and simple procedure for preparing highly monodispersed ZnS colloidal nanospheres as well as the successful incorporation of carbon-based material within highly ordered ZnS colloidal PCs with long-range periodicity. We demonstrate that the visual characteristics of ZnS-based colloidal PCs can be significantly improved by tuning the surface charge and crystallization of colloidal nanospheres, as well as by effectively incorporating (in situ) carbon light-absorption materials within colloidal PCs. In particular, the homogeneous nucleation reaction of the ZnS nanospheres was realized by using zinc acetate dehydrate (Zn(Ac)₂·2H₂O) and thiourea as precursors. Thiourea served as the source of sulphur and the complex reagent by forming Zn²⁺-thiourea ligands. The Zn²⁺-thiourea ligands can be regarded as a reservoir of Zn²⁺ ions, which has a well-balanced rate for releasing ions. Thus, the nucleation process and aggregation of ZnS nanospheres can be effectively controlled, leading to the mass production of monodispersed nanospheres. The surfactant poly(vinylpyrrolidione) (PVP) acted as a structure-directing agent and stabilizer, providing an additional way to facilitate the nucleation process to obtain uniform colloidal nanospheres.

The control of the ZnS particle size was achieved by simply varying the reaction time, offering a broad range of diameters from ~40 nm to over 200 nm, as was shown in Fig. 2. The monodispersion and uniformity of colloidal ZnS nanospheres were also verified by dynamic light scattering (DLS) measurements (standard deviation was <3%). Obviously, longer reaction time leads to the growth of particle size, e.g., the particle diameter is about 40 nm when reaction time is 30 minutes and it grows to ~160 nm when the reaction time is extended to 3 hours.

After the monodispersed ZnS nanospheres were obtained, the next steps are to endow them with some necessary characteristics to benefit the realization of high-quality colloidal PCs. A variety of requirements should be satisfied to achieve this target. First, the particle surfaces should be highly charged to ensure the long-range electrostatic interactions for nanospheres assembly; second, the colloidal nanospheres should be highly crystallized to guarantee the relatively high refractive index; and third, the incorporation of black light-absorption materials should not affect the assembly of the colloidal nanospheres. Notably, the technical routes to address the above issues are not completely independent. It is necessary to design and investigate the appropriate processes to achieve the best balance between these issues for high-quality colloidal PCs.

Here, we used a silica coating to modify the surface charges on the ZnS nanospheres. The Stöber process [44] was used to uniformly coat silica on the ZnS nanospheres through slow hydrolysis of TEOS. Figures 3(a)-3(b) exhibit the as-prepared ZnS (2 hours) and SiO₂–coated ZnS nanospheres. The outer silica layer can provide ionization of surface hydroxyl groups which leads to negative charges on the colloidal nanospheres. Thus, after silica coating, the zeta potential of the colloidal nanospheres was varied from −5.72 mV to −15.34 mV, as shown in Table 1. Moreover, the uniform silica coating can be observed in the TEM image in Fig. 3(b). One can see that the surface has become smoother after the silica coating was applied to the ZnS nanospheres. It should be noted that the silica thickness must be carefully tuned. Given the fact that silica has an intrinsically low refractive-index, silica that is too thick may drastically reduce the effective refractive index of the colloidal nanospheres, while silica that is too thin won’t survive the subsequent surface treatment processes, leading to either an incomplete or even the absence of any silica coating on the ZnS nanospheres. Here, the thickness of the silica coating was tuned to be ~10 nm which exhibits insignificant influence on the refractive index and ensures a complete coating after the subsequent treatments are applied.

 

Fig. 3 (a) Colloidal ZnS nanospheres. (b) Silica-coated ZnS. (c) Silica-coated ZnS treated at 600 °C. (d, f) Silica-coated ZnS treated at 600 °C with surface treatment. (e) Annealed ZnS nanospheres with complete surface etching.

Download Full Size | PDF

Table 1. Zeta-potential and diameter of ZnS and ZnS@SiO₂ nanospheres treated at different calcination temperatures and partial surface etching

View Table

After the silica-coated ZnS nanospheres were obtained, another factor that should be taken into account is crystallization of the colloidal nanospheres, which is a key factor influencing the refractive index of the nanospheres, surface charge and the resulting visibility of the colloidal PCs. In this work, we employed high-temperature calcination to control the crystallization of the prepared ZnS@SiO₂ colloidal nanospheres, guided by the principle that materials synthesized at higher calcination temperatures can lead to better crystallization and a resulting higher refractive index [45]. XRD analysis was used to investigate the crystallite size of the colloidal nanospheres when calcinated at different temperatures, which is shown in Fig. 4(a). Clearly, all of the diffraction peaks can be indexed to the pure hexagonal phase of ZnS, which are in good agreement with values reported in the literature (JCPDS 36-1450). The diffraction peaks locating at 28.5°, 47.5° and 56.3° correspond to the crystal faces of (111), (220) and (311) respectively. There are no other diffraction peaks, implying a high purity of the products. Obviously, the diffraction peak intensities increase and the line width becomes narrower when the calcination temperature is enhanced, which denotes that crystallization is enhanced and that crystallite growth is present. It is worthy to emphasize that the calcination temperature must be carefully tailored since it will not only modify the crystallization, but also vary the characteristics of the colloidal nanospheres in terms of their morphologies and surface charge. As shown in Fig. 3(c), the ZnS nanospheres treated at 600 °C have a larger grain size and slightly reduced particle size compared with ZnS nanospheres without any heat treatment (Table 1).

 

Fig. 4 (a) XRD patterns of ZnS@SiO₂ nanospheres prepared with different calcination temperatures. (b) XPS measurements for the ZnS@SiO₂-600 with Ar ion sputtering (2 keV) for 60 seconds.

Download Full Size | PDF

Besides the high degree of crystallization, another bonus offered by calcination is the ability to incorporate carbon-based materials into the colloidal nanospheres. Researchers have successfully demonstrated that adding carbon materials can effectively enhance visual appearance of structural color due to their excellent ability to control the light scattering [36,46]. However, few works have reported their successful incorporation into long-range ordered colloidal PCs constructed from high refractive index materials. This may be due to the difficulty to simultaneously realize a uniform distribution of carbon in PCs and a highly ordered assembly of nanospheres decorated by carbon materials. In this work, the potential sources for carbon are in the PVP added during the synthesis of the monodispersed ZnS nanospheres. PVP plays as a structure-directing agent and stabilizer in growth of ZnS nanospheres and distributes uniformly in ZnS nanospheres [47]. As shown in Fig. 5, the C and N elements of PVP are uniformly distributed in the ZnS nanospheres. The calcination under a nitrogen atmosphere will carbonize PVP, resulting in the in situ generation and the uniform distribution of carbon in the ZnS nanospheres. Figure 4(b) exhibits the XPS data with sputtering for 60 s obtained from the ZnS@SiO₂ nanospheres treated at 600 °C for 120 minutes. The carbon element can be found with 3.29 wt % loading content, which is consistent with the reported results that describe the desired carbon content for the optimal structural color [36]. In contrast to the traditional approach based on physical mixing, this method directly creates the carbon absorber within the material by in situ carbonization of an organic precursor. If the synthesis conditions are well controlled and the carbon content is precisely tuned, significantly improved visibility of the colloidal PCs can be expected in terms of strong coloration and wide viewing angle. More importantly, this procedure will not affect the assembly of the colloidal nanospheres into long-range ordered arrays since the generation of carbon occurs directly during the formation of the colloidal nanospheres.

 

Fig. 5 (a) HAADF-STEM image of ZnS nanospheres. (b-f)EDS elemental mapping of ZnS nanospheres.

Download Full Size | PDF

Although calcination can provide various benefits to the construction of high-quality colloidal PCs, the temperature must be carefully selected since excessive temperature will substantially destroy the hydroxyl groups on the SiO₂ and therefore reduce the surface charge. Under such circumstances, the surface charge was not sufficient to cause repulsive interactions strong enough for them to self-assemble into long-range ordered colloidal arrays. To address this problem, a surface etching treatment was employed by adding sodium hydroxide (NaOH) in the colloidal solution. NaOH will react with silica to generate sodium silicate, which provides the negatively charged hydroxyl groups. After the surface treatment, the zeta-potential of the colloidal nanospheres reached about −40.24 mV (Table 1), leading to the enhanced surface charge compared to the nanospheres without treatment. The treatment conditions, such as NaOH concentration and treatment time, should be strictly controlled to guarantee the process will not completely remove the silica thin layer. As was shown in Fig. 3(d), the partially etched ZnS@SiO₂ nanospheres have decent monodispersion and uniformity. Importantly, the silica layer still remains a complete coating with a reduced thickness (from 10 nm to 5 nm) after the surface treatment process. For comparison, Fig. 3(e) shows a TEM image of the ZnS nanospheres with the etching process that completely removes the silica layer. Notably, the core-shell structure can be well maintained even after the samples are calcined at high temperatures and with a partial surface etching under an alkaline condition, as shown by the low magnification TEM image in Fig. 3(f). The ability to precisely control the surface etching process, together with the optimized design of particle architectures, allows us simultaneously obtain the highly charged colloidal particles along with the further enhancement of refractive index. Importantly, this paves the way toward the subsequent effective particle assembly and associated creation of high-quality colloidal PCs.

The highly charged ZnS@SiO₂ colloidal nanospheres were then assembled into three-dimensional (3D) face-centered-cubic (fcc) crystals with a long-range periodicity by using the modified fluidic cell method [41,42], enabling PCs with a large area (3.8 cm × 6 cm) to be successfully obtained. SEM images of the PC’s microstructure in Figs. 6(a)-6(b) demonstrate that monodispersed colloidal nanospheres were arranged into a highly ordered photonic array with an fcc structure. This highly ordered periodic structure ensures the strong interference between incident light and the periodic array, which produces a photonic stop-band located in the full visible spectrum regime. Figure 6(c) shows the reflectivity spectra measured from three PCs assembled from ZnS@SiO₂ nanospheres (same diameter) without any post-heat treatment (ZnS@SiO₂), heated at 600 °C (ZnS@SiO₂-600) and 800 °C (ZnS@SiO₂-800). The reflection spectra can be well characterized by the structural color of the assembled array with long-range order. Moreover, the background scattering of light was effectively suppressed by the combined effect of the long-range order and the incorporation of the light-absorbing materials. A very strong reflectivity of 94% is observed from sample ZnS@SiO₂-600, which appears at 472 nm. The PC without heat treatment demonstrates a fair reflectivity of 77% locating at 476 nm, while sample ZnS@SiO₂-800 yields a relatively low reflectivity of 40% at 462 nm. Not surprisingly, ZnS@SiO₂-600 has a stronger reflectivity, which can be attributed to a better balance between particle crystallization, sufficient surface charge and good monodispersion. In contrast, although ZnS@SiO₂ also possesses a good monodispersion of the colloidal nanospheres, the poor crystallization and low surface charge severely influence its photonic performance. With respect to ZnS@SiO₂-800, despite the good crystallization, the overheated temperature results in an irreversible aggregation of colloidal nanospheres (Fig. 7) as well as a relatively low zeta potential (Table 1). The particle aggregation at 800 °C can also be evidenced by DLS analysis, which shows an increased particle size as a result of particle aggregation when they were overheated. These factors limit its effective assembly into highly ordered arrays and thus drastically reduces the corresponding reflectivity. Thus, ZnS@SiO₂ nanospheres heated at 600 °C and treated by an alkaline surface partial etching process are ideally suited to achieve good performance in colloidal PCs.

 

Fig. 6 (a, b) Cross-sectional view of a 3D opaline lattice of 190 nm ZnS@SiO₂-600; (c) Experimental reflectance spectra of 130 nm ZnS@SiO₂ colloidal nanospheres calcined at different temperatures.

Download Full Size | PDF

 

Fig. 7 TEM image of the ZnS@SiO₂-800 sample.

Download Full Size | PDF

Based on the above preparation conditions, red (R), green (G) and blue (B) structural colors were achieved from assembled 3D PCs using nanospheres with different ZnS@SiO₂ particle diameters (130 nm, 160 nm, and 190 nm). As shown in Fig. 8(a), the structural colors exhibit excellent uniformity, high color visibility and saturation at a relatively broad viewing angle even without any external light illumination. The measured reflectivity spectra are shown in Fig. 8(b). All of the 3D PCs have very strong reflection (>90%), with the green color even approaching 100% reflectivity (Fig. 8). The enhancement of visibility and saturation is due to several important properties. The first one is the high refractive index that was achieved by using ZnS. The high refractive index contrast between ZnS and its surrounding medium (water) can lead to a more intense stop band and an enhanced reflectivity, which can greatly improve the color visibility. In addition, provided that the refractive index of ZnS is large enough, the change of incident angle will have small effects on the stop band peak position, thus resulting in relatively low dependence of viewing angle [46,48], as is demonstrated in Fig. 9. In addition, the incorporation of light absorbing carbon material provides a strong absorption over the full band of visible light. As a result, the scattering/background light in the structural color can be greatly absorbed causing the improved coloration and saturation.

 

Fig. 8 (a) Photographs of PCs comprised of calcined ZnS@SiO₂-600 colloids assembled in a cell with different particle diameters of 130 nm (blue), 160 nm (green), and 190 nm (red); (b) the measured reflection spectra from assembled 3D ZnS@SiO₂-600 colloidal PCs exhibiting strong reflectivity red (90%), green (98%) and blue (90%).

Download Full Size | PDF

 

Fig. 9 Photographs of a crystalline array observed from (a) ∼5 ° and (b) ∼30 °.

Download Full Size | PDF

4. Conclusion

In summary, highly monodispersed silica-coated ZnS colloidal nanospheres were successfully prepared and assembled into long-range ordered crystalline colloidal arrays. We explored a variety of technical approaches to improve the coloration and visibility of PCs. First, based on the post-heat treatment, we obtained an enhanced crystallization of colloidal nanospheres and in situ incorporation of carbon-based materials within colloidal PCs, without affecting the long-range ordered arrangement of nanospheres. Second, the highly order assembly of nanospheres was achieved by employing a surface treatment, which offered an enhanced surface charge for strong electrostatic interaction between nanospheres. Owing to the combination of high refractive index and effective control of scattering light, the obtained 3D colloidal PCs have demonstrated very strong brightness, coloration, saturation and low dependence of viewing angle. The present work provides a simple and efficient strategy to create structural color with high visibility, paving the way toward their practical applications in displays, sensors and paints.