For the preparation of PS,^14-17) we used the micro-emulsion [PS] process and surfactant-free [SFree-PS] emulsion process.^13,22) The emulsion polymerization starts with an emulsion incorporating water, monomer, and surfactants. In order to make an oil-in-water emulsion, droplets of monomer (the oil) are emulsified (with surfactants) in a continuous phase of water. A conventional emulsion system with styrene 7.84 wt %, SDS 1.96 wt %, water 90.20 wt %, and KPS 3.60 mM is polymerized at 70°C.

PMMA spheres of uniform diameter were prepared by surfactant-free emulsion polymerization, as reported.^8,14,21) For the preparation process of PMMA latex beads, the chemicals were methyl methacrylate monomer [M55909, Sigma Aldrich, USA] and granular 2,2′-Azobis(2-methylpropionamidine) dihydrochloride, an AMPD initiator [M440914, Sigma Aldrich, USA]. In order to make 0.5-1.5 g of macroporous material, we used 10 g of PMMA. Polymerization was initiated with the AMPD initiator. The colloidal suspension was centrifuged at 1500 rpm for 24 h; water was decanted and the solid was allowed to dry for 3 days to sediment the spheres. The five neck round-bottom flask [Sigma Aldrich] was clearly cleaned before the start of the polymer reaction of PMMA. The capacity of the five-neck flask was 3,000 mL with center Joint: ST/NS 24/40. The cleaning of PMMA chunks was performed using acetone, tetrahydrofuran [THF] [Sigma Aldrich], a flow meter, and baking soda, this last of which was used to remove Cl gas (Fig. 1(a)).

Cleaning of the 5 neck reactor for the PMMA template was performed in order to obtain good cakes. The dry cake chunk material, a mixture of unreacted polymer of PMMA and spheres, was softened using acetone and/or THF; the wet chunk material was cut and taken out using a spoon or pipet. This serious cleaning took 1.5 h; later, the utilized solvent was reused. For the formation of the 3DOM structure of the TiO₂ skeleton, we used PMMA beads as one of the porogen materials. The dry body of the centrifuged PMMA was crushed; a three gram PMMA block sample was obtained as wet-like lumps.

Surfactant-free PS or PMMA latex spheres were synthesized using an emulsifier-free emulsion polymerization technique.^{8,13,14,21,22)} The diameters of the mono-disperse PMMA or PS latex spheres were approximately 500 nm or 750 nm, respectively. We used pieces of broken polymer pallets in the flask inner surface to prepare the colloidal crystal template of the PS spheres in the centrifuge at 1000 rpm. The diameters of the pieces of broken polymer pellets that were taken out as types of latex chunks were between 0.5 cm and 1.5 cm after the polymerization in the five neck flask of the PS template. Dry bodies of the centrifuged PS were crushed and came to look like wet lumps of PS chunks, as shown in the left picture of Fig. 1.

For the synthesis of 3DOM TiO₂, 3.3 g of titanium butoxide [Ti(CH₂OH)₄, Sigma Aldrich] was mixed with 2.25 mL of ethanol and stirred for 3 minutes, as shown in Fig. 1. A few drops of 0.07 g H₂SO₄ were added for pH adjustment. The mixture solution was stirred for 25 min. The solution started to become a transparent and uniform sol. As can be seen in Fig. 1, a chunk from the dry body of the PMMA or PS on the surface of the beaker was put into the mixture solution. We were able to observe bubbling from the chunk.

If the bubbling stopped, the wet chunk was taken out and the sample was dried for a day. For drying, the wet body was put on the culture beaker and covered with wax paper for slow drying. For the calcination, the cakes of dry powder were poured into an alumina crucible and put into a furnace, as shown in the last part of Fig. 1. According to the firing schedule, the cakes were calcined in a temperature range of 450°C to 750°C for two hours; they were then cooled to 30°C. During the calcination process, N₂ gas was passed through baking soda in a plastic box and then was supplied into the quartz tube in the furnace. The amount of flowing N₂ gas was controlled by counting the number of gas bubbles in the attached flask.

XRD [Bruker, M18XCE] was used for the characterization of the TiO₂ ceramic skeleton after the firing process. The sample body was pulverized into TiO₂ powder for the XRD measurement using a Cu target. The microstructure of the sintered TiO₂ body was investigated using FE-SEM [Hitachi S4800, Japan].

After Dynosphere^® was first fabricated at the University of Michigan (under the support of Dow Chemical), PS latex has been typically been composed of monodispersed micro/nano-particles. In this emulsion-free polymerization process, the diameters of the PS beads are known to be between 100 nm and 600 nm. Using emulsion polymerization, we can obtain nano-beads with diameters above 50 nm; in micro-emulsion polymerization, nano-beads with diameters below 50 nm are obtained. For the semi-continuous micro-emulsion polymerization of styrene, we used SDS [Sodium dodecylsulfate] emulsifier [MICOLIN, Miwon]. Additionally we prepared surfactant-free PS beads. From the FE-SEM microstructure shown in Fig. 3(a), it can be seen that the diameters of the PS beads is between 10 nm and 50 nm. Fig. 4(a) shows the microstructure of SFree-PS-TiBu; in this sample, the PS beads were prepared by surfactant-free polymerization; the diameter of the mono-disperse PS beads is estimated to be ~700 nm.

PMMA latex spheres with diameters of 500 nm were obtained, as shown in Fig. 4(b). The synthesized raw compounds were stirred and centrifuged in a five neck flask reactor at 1000 rpm. As a colloidal crystal template, pellets were formed from the polymer. Pieces of broken polymer pellets on the surface of flask were taken out; the pellet sizes were from roughly 0.5 cm to 1.5 cm. PMMA is less toxic than polystyrene. For the preparation of the template cakes, the mixture of unreacted PMMA and spheres was softened using acetone and/or THF. Using a spoon, the chunk was pulled out and cut. This case was serious and took 1.5 hrs. The obtained PMMA powder aggregates can be ground simply using a finger to push the aggregates between the weighing covers.

Figure 3(b) shows the 3DOM structure of the PS-TiBu-550 sample calcined at 550°C. In this sample, the PS beads were prepared through micro-emulsion polymerization; the sizes were between 10 nm and 50 nm, as can be seen in Fig. 3(a). In the 3DOM structure of the PS-TiBu-550 sample, shown in Fig. 3(b), we can observe the crystal growth of TiO₂ particles during the firing process. Fig. 3(c) provides enlarged pictures of the rectangular positions of P1, P2, P3, and P4 in the left micrograph of Fig. 3(b). Several-nanometer particles of TiO₂ were grown and agglomerated to several tens-of-nanometer granules; the abnormal grain growth turned into a several-hundred-nm single crystal (P4). Fig. 3(d) shows the results of an EDX taken for a single crystal region (1) and for the agglomerated particle region (2). X-ray spectra analysis shows similar patterns in the two regions, indicating that there are only slight differences in the composition and intensity of the anatase TiO₂ phase. Considering these results, it seems that condensation via solid state reaction among TiO₂ nano-particles occurred at 550°C. During the firing process, several-nm pore channels were infiltrated with TiO₂ nano-particles by capillary pressure; the 3DOM architecture became dense through the solid-state condensation of TiO₂ nano-particles. Fig. 3(e) shows the 3DOM structure of the PS-TiBu-620 sample, for which PS was prepared using the micro-emulsion process and Tibutoxide solution was infiltrated with the monodispersed PS particles. The dry mixture of Ti-butoxide and PS beads was calcined at 620°C. The particle size of TiO₂ is several nm and the diameter of the PS beads is in the range of 10 to 50 nm.

In Fig. 4(a), showing the 3DOM microstructure of the SFree-PS-TiBu, monodispersed PS beads with diameters of ~ 700 nm were impregnated in the Ti-butoxide ethanol solution and the resulting mixture was dried at room temperature. Fig. 4(b) shows the 3DOM microstructure of the PMMA-TiBu compound at room temperature. The size of the PMMA mono-spheres is ~ 500 nm; we can observe agglomerates of several-nm monospheres of TiO₂. In the experimental process, PMMA beads were impregnated in the Ti-butoxide ethanol solution. From the information presented in Fig. 4 and Fig. 5, it can be postulated that the uniform distribution of PMMA changes according to the volumetric mixing ratio of PMMA beads in the Tibutoxide solution. During the FE-SEM observation, we were able to observe a uniform distribution, as shown in Fig. 5(b)(c) (d)(e). On the other hand, it was occasionally possible to observe the non-uniform distribution that is seen in Fig. 5(a) in most of the samples. In order to obtain a uniform 3DOM bulk structure, the process should be more finely controlled. Fig. 5 shows the 3DOM structure of the PMMA-TiBu samples calcined at 450°C, 550°C, 600°C, 650°C, and 750°C for PMMA-TiBu-450, PMMA-TiBu-550, PMMA-TiBu-600, PMMA-TiBu-650, and PMMA-TiBu-750, respectively. In Fig. 5(f), the monodispersed TiO₂ granules show opalescent behavior^25-29) according to the direction of light for the PMMA-TiBu-750 sample from Fig. 5(e). The boundary layer Ti-OH compound between the PMMA beads was calcined and formed into an agglomerate of TiO₂ granules and the 3DOMm structure of TiO₂. Monodispersed beads of PMMA or PS, shown in Fig. 4(a) (b), exhibit opalescent color emission according to the light direction, which is the typical behavior of mono-dispersed beads.^8,24-29) The color emission varies with the dimensions of monodispersed beads and with the type of added solution, such as water, alcohol, or toluene. The 3DOM skeleton of the TiO₂ granules with several-nm pores was filled with the solvent; increasing the pore size or filling the pores with solvent resulted in an increase in the wavelength of light reflected by the photonic crystal.^28-32)

In Fig. 5(c), the pore size and the boundary wall thickness of PMMA-TiBu-600 were estimated to be ~ 500 nm and ~ 10 nm, respectively. In the PMMA/Ti-butoxide sample, PMMA was burned out through the slow firing schedule; the formed TiO₂ nano-powders were condensed at 450°C, as shown in Fig. 5. From the number of powder grains in the boundary, it is supposed that the diameter of TiO₂ powder at 450°C may be estimated to be ~5 nm. The dimensions of the pores formed by the burn-out of the 500 nm PMMA beads decreased to 480 nm and 270 nm for PMMA-TiBu-550 and PMMA-TiBu-750, respectively. During the initial stage of the firing process, the PMMA beads were burned out and TiO₂ nano-powders were formed through the evaporation of alcoholic groups attached to Ti-butoxide. With the increase of the firing temperature, the formed TiO₂ nanoparticles agglomerated to granules and the 3DOM skeleton of TiO₂ was built through the grain growth of TiO₂ granules in the boundary wall. In the sample calcined at 750°C, as shown in Fig. 5(e), crystallites of TiO₂ were condensed to larger grains in the network boundary region of the PMMA beads. In the magnified pictures, the boundary regions among the pores, which were formed by the burn-out of PMMA beads, became larger in the PMMA-TiBu-750 sample than these boundary regions did in the PMMA-TiBu-450 sample. The densification of the TiO₂ powders resulted in a larger grain size in the boundary region and in a smaller pore size of PMMA-TiBu- 750.

Figure 6 shows the 3DOM structure of the Ti-butoxide mixture of PS; this mixture was prepared by surfactant-free polymerization, and the dry mixture was calcined at 450°C, 600°C, and 750°C for the samples of SFree-PS-TiBu-450, SFree-PS-TiBu-600, and SFree-PS-TiBu-750, respectively. Fig. 6(a) shows monodispersed uniform PS beads of 700 nm. The 3DOM structure developed with the firing temperature increase. In the high temperature samples shown in Fig. 6(c) and Fig. 6(d), we can observe exaggerated grain growth at 600°C and 750°C, respectively. The grain growth induced a larger 3DOM wall structure for the higher temperature samples.

The next subject of research will be to determine the progress for controlling the morphologies of 3DOM materials at different length scales. To prepare periodic pore structures, void spaces between spheres in a colloidal crystal template are infiltrated by any type of fluid that can penetrate the template and can be converted into a solid. Removal of the template spheres leaves a solid skeleton that surrounds the air holes left in the original locations of the spheres.