Ordered macroporous titania photonic balls by micrometer-scale spherical assembly templating

Abstract

Highly ordered macroporous photonic balls (i.e. inverse opaline structure) composed of titania frameworks were fabricated by using a titania precursor templated around polystyrene spheres which had been assembled into polymer photonic balls (i.e. opaline structure). Narrow disperse polymer photonic balls consisting of monodisperse surface-modified polystyrene (PS) latex particles were prepared by utilizing a suspension system. The diameters of the opaline balls can be controlled in a range of a few or a few tens of micrometers. The macroporous titania structure made by this method was well-defined because the PS spheres making up the polymer photonic balls were close-packed and ordered in three-dimensions. Furthermore, crystalline types of titania (anatase or rutile) were readily adjusted through tuning the calcination temperatures, so the macroporous titania inverse opaline balls composed of anatase or rutile can be used for various applications.

---

Instruction

Colloidal crystals (e.g., opal materials) are extensively used as templates to synthesize ordered macroporous (e.g., inverse opal) materials with varied composition. They have broad potential applications, such as in catalysis, separation, chemical sensors, and photonic band-gap materials. The morphology and substrate of the macroporous materials can be tuned by using different shape stacked opal templates and precursors. A major barrier to technological application of these materials is the lack of simple, easily controlled methods for mounting or shaping the templates into usable solid objects. The simplest examples of such templates are face-centered cubic (fcc) colloidal crystals, formed spontaneously in all-identical spherical colloids, such as polymer latexes or silica suspensions. Recently, more complex structures have been made by a variety of clever techniques, such as altering the shape of the colloidal particles,¹ directing their assembly with patterned surfaces,² or tuning the interaction between particles.³ Furthermore, photonic crystals in the form of colloidal clusters (uniform spherical colloidal aggregates), that is so-called “opaline photonic balls”, have received more and more attention because of their unique photonic properties resulting from the arrangement of colloids with spherical shape, for example, as light scatterers, light diffusers, and pigments for electronic paper and electronic displays.⁴ Several research groups have reported some methods to fabricate photonic balls. Velev and coworkers synthesized spherical assemblies from polystyrene latex particles by growing colloidal crystals in aqueous droplets suspended on fluorinated oil.⁵ Yang and coworkers did similar work to assemble spherical colloidal templates through colloidal crystallization of suspended polystyrene latex sphere particles in aqueous droplets straddling an air–oil interface.⁶ These two kinds of photonic balls are from a few hundreds of micrometers to a few millimeters in size, and the size distribution of these balls can not be controlled easily. However, one of the important issues in the design of photonic balls is the control of their uniformity in size and shape, so as to meet the requirements of practical use of these materials. Recently, Yang and coworkers synthesized uniform photonic balls and their inverse structures by injecting an aqueous suspension of polymer latex spheres into a surfactant-laden oil phase at an oil/water junction of capillary tubes or soft-microfluidic devices^7,8 and by electrospraying an aqueous colloidal suspension.⁹ The corresponding inverse opaline balls being composed of close-packed hollow particles consisting of a low-refractive-index air core and a high-refractive-index metal oxide shell (such as titania) are suitable building blocks for photonic crystals, so-called “inverse opaline photonic balls”.

Here we describe a new method to generate narrow disperse opaline photonic balls which are in the range of a few or a few tens of micrometers in size, and then, to template the closely packed colloidal spherical assemblies into macroporous titania (titanium dioxide) photonic balls (inverse opaline photonic balls) possessing ordered macropores having pore diameters in the range of a few hundreds of nanometers comparable to optical wavelengths. Our synthetic route for making the macroporous titania photonic balls is a three-step template-assisted fabrication process, illustrated in Scheme 1.

Scheme 1 Schematic diagram of the preparation method for photonic balls composed of titania building blocks.

First, the polymer opaline photonic balls were fabricated by the colloidal crystallization of aqueous emulsion droplets in a suspension system similar to a suspension polymerization system. A hydrophobic silicone liquid was selected to be a dispersing medium (i.e. the continuous phase) of the suspension system. The building blocks of the crystal balls were monodisperse polymer latex spheres. The size of the balls could be controlled by tuning the concentration of the aqueous latex, position of the stirrer in the system, rotation speed of stirring of the suspension system and the volume ratio of the silicone liquid to the aqueous latex. Second, the polymer photonic balls were infused with the titania precursor solution which was converted into titania in the void spaces between the polymer spheres. Finally, the polymer spheres were removed by calcination, which left macropores at their sites. The pore diameters were in the range of a few tens of nanometers to a few hundreds of nanometers and could be adjusted by using different sizes of polymer latex spheres. These macroporous titania photonic balls possessed a narrow size distribution and an ordered internal lattice of pores. Furthermore, titania macroporous photonic balls consisting exclusively of crystalline rutile or anatase could be obtained by tuning the calcination temperature. The crystalline rutile titania has a sufficiently high refractive index to lead to a complete photonic band gap of the corresponding colloid crystals¹⁰ and the crystalline anatase is of practical significance for potential applications in photocatalysis.¹¹

Experimental

Synthesis of monodisperse latex particles

To ensure the structure stabilization of our photonic balls during the preparation procedure, the latex particles used in our study were hydrophilic and cross-linked polystyrene. Monodisperse cross-linked polystyrene (PS) latex particles with carboxyl groups on their surface were synthesized by emulsifier-free emulsion copolymerization using potassium persulfate and Na₂HSO₃ as a redox initiating system, tri(ethylene glycol) diacrylate (TEGDA) as cross-linker, and methacrylic acid as functional monomer to provide carboxyl groups on the surface of particles. During polymerization, a definite amount of NaHCO₃ was added to partially ionize the carboxyl groups. The resulting emulsion was purified by dialyzing in deionized water using a dialysis tube. Fig. 1 is the TEM image of the monodisperse cross-linked polystyrene (PS) latex particles with carboxyl groups on their surface.

Fig. 1 TEM image of the monodisperse cross-linked polystyrene (PS) latex particles with carboxyl groups on their surface.

Preparation of polymer spherical clusters (i.e. opaline photonic balls)

Photonic balls consisting of monodisperse cross-linked polystyrene (PS) latex particles with carboxyl groups on their surface were prepared by using a suspension dispersed system. Scheme 2 is the schematic of this system.

Scheme 2 Schematic diagram of the synthetic method for opaline balls. The emulsion droplets are suspended in the silicone oil and colloidal crystals form during the drying process, as described in the text.

A 3000 ml three-neck flask with 2300 ml hydrophobic silicone oil (viscosity 50 cSt, as a continuous phase, purchased from Beijing Chemical Factory) was fixed in a controllable temperature water cell. The silicone oil was rotated at constant angular velocity (∼480 rpm) using a stirrer. The aqueous emulsion (115 ml, concentration of 1.5–5.0%) synthesized as above was uniformly injected into the stabilized stream silicone oil using a constant air or N₂ flow pressure through a tapered container, which has a capillary tube with an inner diameter of 10–20 µm at the top. Narrow size distribution emulsion droplets were formed in the suspension system. Then colloidal crystallization of the emulsion droplets gradually occurred when water in the droplets was evaporated under an constant temperature (here, we selected 55 °C). About six hours later, colloidal spherical assemblies were formed and the suspension system was cured at 80 °C for another six hours to enhance the structure. The polymer photonic balls (opaline structure) were obtained after the colloidal spherical assemblies in the suspension system were filtered and washed with n-hexane, isopropanal, and ethanol, respectively, and dried in the air. Subsequently, they were treated with stainless steel sieves with different pore sizes (purchased from Shang Yu Instrument Company, ZeJiang province, China) to remove some asymmetric balls. Fig. 2 shows the optical micrographs and SEM images of the polymer photonic balls prepared using the above process, and the polystyrene latex spheres that formed the balls were ∼200 and ∼400 nm in diameter, respectively.

Fig. 2 (a), (c) Optical micrographs of opaline photonic balls formed by polymer latex spheres with two different diameters, ∼200 and ∼400 nm respectively. (b), (d) SEM images of these two balls.

Preparation of ordered macroporous titania spherical aggregates (i.e. the inverse opaline photonic balls)

The opaline balls obtained as mentioned above were immersed in a titania precursor for a week to allow the precursor permeation sufficiently. The precursor was composed of tetrabutyl titanate (TBT) and ethanol (the TBT : ethanol ratio was 3 : 4). The precursor-soaked opaline balls were washed carefully with ethanol to rinse off the precursor on the surface of the balls, and then exposed to the air for several days to undertake a hydrolyzation process with moisture from the atmosphere to form polymer/titania composite balls. Finally, the composite balls were calcined at 500 °C or 700 °C for 8–10 hours (the heating rate was fixed at 1 °C min⁻¹), resulting in polymer removal and anatase or rutile titania crystallite formation, respectively.

Characteristics

Transmission electron microscopy (TEM) images of the monodisperse cross-linked polystyrene (PS) latex particles with carboxyl groups on their surface were obtained by TEM (H-800, Electron Microscope, Hitachi). The morphology of the photonic balls and their inverse ones was observed by scanning electron microscopy (SEM, Stereoscan 250 MK3, Cambridge Instruments and JSM-6360LV, Jeol Ltd, Japan) and optical microscopy (XJZ-1A, ChongQing, China). Polydispersity of the photonic balls was measured using a Zetasizer (3000HS, Malver Instruments). Thermogravimetric analysis of the inorganic–organic composite balls was determined by using a thermoanalyzer (O₂ atmosphere, heating rate 10 K min⁻¹, Perkin Elmer, TGS-2, America). Crystallite types of titania were determined by X-ray diffraction (XRD) using a X-ray diffractometer (D/Max 2500 VB2t/PC, Rigaku, Japan).

Results and discussion

Narrow size distribution opaline photonic balls

In this study, narrow size distribution opaline photonic balls were prepared by a suspension system in which a silicone oil (viscosity, 50 cSt)¹² was the continuous phase, and polymer emulsion droplets were the disperse phase. As we know, the disperse phase droplets that were suspended in the continuous phase should be monodisperse or narrow-dispersed in size as the oil : water volume ratio was sufficiently high and the stirring rate was constant, this is based on the fact that the suspension droplet destruction caused by stirring should be omitted in this case. According to this idea, a set of devices was designed to obtain narrow-dispersed aqueous emulsion droplets in a suspension system using synthetic conditions of 20 : 1 volume ratio (silicone liquid : latex suspension) and a constant stirring rate. Then, the polymer spheres in the aqueous emulsion droplets which were suspended in the silicone liquid were concentrated by evaporating the water under evaporation temperature of 55 °C. This evaporation temperature of 55 °C during crystallization was selected based on a report of the optimum crystallization temperature for opal fabrication.¹³ The droplet size decreased monotonically with time during evaporation of water and the polymer sphere concentration increased at the same time. As the concentration exceeded a certain transition value, the polymer latex spheres began to order into a face-centered cubic (fcc) structure to form close-packed opaline photonic balls. It should be noted here that the hydrophobic silicone liquid can cover the hydrophilic polymer emulsion droplets to form a stabilized suspension system because it is lighter than water and polymer microspheres (specific gravity values: 0.97 g ml⁻¹, 1.00 g ml⁻¹ and >1.00 g ml⁻¹, respectively). The vapor pressure of the silicone liquid is important, because low-viscosity silicon liquid will evaporate before completion of the crystallization process (viscosity 50 cSt). After crystallization, the opaline balls were obtained through filtration, soaking and rinsing with n-hexane, isopropanol, and ethanol, respectively. Finally, the balls were dried and cured at 80 °C for several hours to increase the degree of close-packing of the crystal balls and evaporate substantially any residual silicone oil.

The size of the photonic balls obtained by using the above method can be tuned through changing the initial polymer latex concentration. Three different latex concentrations (1.62%, 3.24% and 4.86%) were used, and three distinct balls with average diameters (∼40, ∼80, and ∼100 µm, respectively) were obtained (based on the optical microscopy images) (generation conditions: volume ratio 20 : 1, a constant stirring rate ∼480 rpm).

The balls with average diameters ∼40 µm were treated with stainless steel sieves with pore sizes of 50 and 30 µm. From the mass ratios of photonic balls of various diameters, it can be seen that the photonic balls are narrowly dispersed (the percentage of balls with diameters larger than 50 µm and smaller than 30 µm is totally less than 10%, however the balls in the size range 30–50 µm account for over 90%). The polydispersity of the balls with diameters of 30–50 µm is around 8% based on the Zetasizer.

We found that in these systems the diameters of the photonic balls will increase very little when the air flow pressure rises and other parameters, such as latex concentrations, volume ratios of oil/emulsion and stirring rate, are fixed. Also, the diameters decrease with increasing stirring rate while the other parameters are fixed. Besides, when the volume ratio of silicone oil to emulsion decreases to 10 : 1, the monodispersity of the photonic balls decreases greatly. For instance, when we prepared photonic balls with diameters of around 40 µm, the percentage of 30–50 µm photonic balls is only 65%, which is much less than 90% obtained from the preparation with the volume ratio 20 : 1. However, if the volume ratio increased to 30 : 1, the percentage of photonic balls in the 30∼50 µm range has not changed greatly, compared with that obtained at 20 : 1. Meanwhile, the yield of photonic balls between 30 and 50 µm decreased dramatically. Therefore, it is suitable to prepare photonic balls in the range 30–50 µm at a volume ratio of 20 : 1.

From all the results mentioned above, we can conclude that the optimal conditions for preparing photonic balls in the range 10–100 µm are: volume ratio of silicon oil to emulsion 20 : 1, stirring rate 480 rpm, air or nitrogen flow pressure to push the emulsion into the silicon system by using a capillary with a mouth diameter of 10–20 µm is kept stable. Fig. 2–4 show the SEM and optical microscope images of photonic balls obtained under these conditions. Fig. 2(b, d), Fig. 3 and Fig. 4(a, b) indicate that the surface of the photonic ball has a highly ordered hexagonal structure while the interior has a close-packed fcc structure.

Fig. 3 SEM image of the fracture section of broken organic–inorganic composite balls.

Fig. 4 SEM images of the opaline balls and inverse macroporous balls. (a) Surface image of the opaline balls. (b) Fractional section image of broken opaline balls. (c) Image of an inverse opaline ball. (d) Surface image of the inverse opaline balls.

Macroporous titania inverse opaline photonic balls

The opaline balls prepared above were soaked in a solution of titania precursor (TBT : ethanol = 3 : 4), which went through the void spaces between the polymer latex spheres by capillary force. The precursor concentration is an important factor in this process. When TBT was in excess, the precursor could not fully infiltrate into the interstices, while when ethanol was in excess, a solid oxide ceramic gel block should not be produced. The precursor concentration of TBT : ethanol = 3 : 4 in this study was selected based on a series of systematically conditional experimental results that we will report elsewhere. After infiltration, the balls which were fully infiltrated by the precursor were pulled out and exposed to the air. Finally, the titania precursor was hydrolyzed into the oxide ceramic gel as a result of its exposure to the moisture of the ambient air, and organic–inorganic composite balls were obtained. Fig. 3 is the SEM image of the fractional section morphology of a broken composite ball. It clearly shows that the void spaces of the opaline balls were filled up by inorganic material (titania).

It is especially noteworthy that before the hydrolysis reaction proceeded, the residual precursor remaining on the surface of the balls should be removed. Otherwise, a thick skin of titania would be formed on the surface of the macroporous balls. To avoid this titania skin formation, we rinsed the balls with ethanol or isopropanol carefully in a moisture-free environment. Finally, the organic–inorganic composite balls were calcined at 500 °C for crystalline anatase titania or at 700 °C for crystalline rutile titania macroporous microstructure inverse opaline photonic balls, respectively. Fig. 4 is the SEM images of the opaline photonic balls (a, surface and b, inner structures) and its inverse opaline balls (c and d). Fig. 4(a) and (b) show that the opaline balls were close-packed and highly ordered in three-dimensions. From the surface of the macroporous balls (Fig. 4(c) and (d)), we can see that the void spaces whose sizes were about 130–150 nm in diameter were interconnected in three dimensions through windows which were 25–30% smaller than the diameters of the original latex beads (∼200 nm), indicating shrinkage during the calcination process. The wall thickness of the windows was about 40–60 nm.

Thermogravimetric analysis of the organic–inorganic composite balls shows that the organic polymer spheres could be completely decomposed at high temperatures above 450 °C. According to the TG results, the calcinations of the composite balls were performed at 500 or 700 °C for preparation of anatase or rutile type titania, respectively. When the composite balls were calcined at 500 °C, macroporous balls consisting exclusively of crystalline anatase titania was obtained, while calcinations at 700 °C yielded macroporous balls consisting exclusively of crystalline rutile titania (Fig. 5). It should be noted that the phase transformation from anatase to rutile occurs in a narrow temperature range between 550–650 °C. The anatase–rutile ratio changes with increasing calcination temperature. As mentioned above, the crystalline type (anatase or rutile) of the titania inverse opaline balls can be fully controllable for various applications by adjusting the calcination temperature.

Fig. 5 X-Ray diffraction patterns of the macroporous titania photonic balls at different calcination temperatures.

Conclusion

Opaline photonic balls and inverse macroporous balls were fabricated in this study. The opaline photonic balls were assembled by utilizing monodisperse cross-linked polystyrene (PS) latex particles with carboxyl groups on their surface and using a suspension dispersed system. The size of the balls can be adjusted by tuning the following factors: the concentration of the aqueous latex, the position of the stirrer in the system, the rotation speed of stirring and the volume ratio of silicone oil to aqueous latex. The modified monodisperse polystyrene spheres used for the building blocks of opaline balls were synthesized by conventional emulsifier-free emulsion copolymerization. Inverse opaline macroporous balls consisting of titania frameworks were obtained from titania precursor templated around polystyrene spheres making up close-packed opaline balls. The cavity diameter of the macroporous balls could be tuned by using different sized PS spheres. Furthermore, the crystalline types of titania consisting of macroporous balls were also controllable by adjusting the calcination temperature. So, the macroporous titania photonic balls obtained by this method can be used for various fields, such as photonic band-gap materials and photocatalysts, through controlling the titania crystalline types, anatase or rutile.

Acknowledgements

This work was financially supported by the Beijing Foundation of Natural Science, P. R. China (Nos.Z012013)

References

1. (a) E. Snoeks, A. van Blaaderen, T. van Dillen, C. M. van Kata, M. L. Bronersma and A. Polman, Adv. Mater., 2000, 12, 1511–1514; (b) Y. Lu, Y. Yin and Y. Xia, Adv. Mater., 2001, 13, 415–420; (c) P. Jiang, J. F. Bertone and V. K. Colvin, Science, 2001, 291, 453–457; (d) Z. Zhong, Y. Yin and Y. Xia, Adv. Mater., 2000, 12, 206–209.
2. (a) Y. Yin, Y. Lu and Y. Xia, J. Am. Chem. Soc., 2001, 123, 771–772; (b) Y. Yin and Y. Xia, Adv. Mater., 2001, 13, 267–271.
3. (a) M. S. Fleming, T. K. Mandal and D. R. Walt, Chem. Mater., 2001, 13, 2210–2216; (b) A. L. Hiddessen, S. D. Rofgers, D. A. Weitz and D. A. Hammer, Langmuir, 2000, 16, 9744–9753; (c) K. Furusawa and O. D. Velev, Colloids Surf. A, 1999, 159, 359–371.
4. (a) G.-R. Yi, J. H. Moon and S.-M. Yang, Adv. Mater., 2001, 13, 1185–1188; (b) F. Iskandar, Mikrajuddin and K. Okuyama, Nano Lett., 2001, 1, 231–234; (c) B. Comiskey, J. D. Albert, H. Yoshizawa and J. Jacobson, Nature, 1998, 394, 253–255.
5. O. D. Velev, A. M. Lenhoff and E. W. Kelar, Science, 2000, 287, 2240–2243.
6. G.-R. Yi, J. H. Moon and S.-M. Yang, Chem. Mater., 2001, 13, 2613–2618.
7. G.-R. Yi, S.-J. Jeon, T. Thorsen, V. N. Manoharan, S. R. Quake, D. J. Pine and S.-M. Yang, Synth. Met., 2003, 139, 803–806.
8. G.-R. Yi, S.-J. Jeon, T. Thorsen, V. N. Manoharan, S. R. Quake, D. J. Pine and S.-M. Yang, Adv. Mater., 2003, 15, 1300–1304.
9. J. H. Moon, G.-R. Yi, S.-M. Yang, D. J. Pine and S. B. Park, Adv. Mater., 2004, 16, 605–609.
10. V. N. Manoharan, A. Imhof, J. D. Thorne and D. J. Pine, Adv. Mater., 2001, 13, 447–450.
11. A. Mills and S. Le Hunte, J. Photochem. Photobiol. A: Chem., 1997, 108, 1–35.
12. H. Fudouzi, J. Colloid Interface Sci., 2004, 275, 227–230.
13. Y. H. Ye, F. LeBlanc, A. Hache and V. V. Truong, Appl. Phys. Lett., 2001, 78, 52–54.

---