Controlling the Color of Lead-Free Red Overglaze Enamels and a

Apr 19, 2016 - empirical experience. Inspired by a recent particle-design method, we also developed a practical facile process to prepare red paints t...
0 downloads 10 Views 2MB Size
Download PDF
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Research Article www.acsami.org

Controlling the Color of Lead-Free Red Overglaze Enamels and a Process for Preparing High-Quality Red Paints Hideki Hashimoto,*,†,‡ Hirofumi Inada,§ Yuki Okazaki,§ Taigo Takaishi,§ Tatsuo Fujii,‡ and Jun Takada‡ †

Department of Applied Chemistry, School of Advanced Engineering, Kogakuin University, Hachioji 1982-0015, Japan Graduate School of Natural Science and Technology, Okayama University, Okayama 700-8530, Japan § Kyoto Municipal Institute of Industrial Technology and Culture, Kyoto 600-8815, Japan ‡

S Supporting Information *

ABSTRACT: Akae porcelain, an artistic Japanese traditional overglaze ceramic typically known for Kakiemon-style ware, has fascinated porcelain lovers around the world for over 400 years because of the graceful red color displayed by akae that matches so well with white porcelain bodies. In this work, we clarified the factors that control the color of akae and those that are conventionally controlled by artisans based on empirical experience. Inspired by a recent particle-design method, we also developed a practical facile process to prepare red paints that yields high-quality akae. Various akae samples were prepared from a combination of lead-free alkali borosilicate glass frits with different particle sizes and hematite powders with differing dispersibilities. Polarized light microscopy, scanning electron microscopy, and transmission electron microscopy analyses indicate that considering only the dispersibility of hematite powders is not sufficient, but the frit-particle size must be controlled to obtain high-quality akae with a high reflectance value for ≥580 nm visible light. In addition, we developed a process for preparing high-quality red paints that uses a large-particle frit powder and a strongly aggregated-hematite powder, both of which are easily obtainable. The red paint composed of frit, hematite, and the solvent is mixed until the paint is drying. By adding more solvent and repeating this process three times, we obtained high-quality akae with a higher reflectance value than for the akae prepared from a frit with submicron-sized particles and weakly aggregated-hematite powder. On the basis of transmission electron microscopic observations, we consider the red paint to consist of a core/shell-like composite structure of frit and hematite, forming a three-dimensional network in the akae glass layer. The good dispersibility of these particles leads to high-quality akae. KEYWORDS: red overglaze enamels, lead-free frit, hematite, particle size, core/shell composite particles

■

INTRODUCTION

found, both for daily use and for artistic purposes. Here we define “high quality”, “beautiful”, or “vivid” akae as akae with high L*, a*, and b* values on the CIE 1976 color space.3 For example, Fukiya bengala,4−9 a red pigment produced in Japan in the 18th century, has values of L* = 41.6, a* = 34.6, and b* = 27.5.9 Then, similar and slightly higher values are considered to be “high quality”. The traditional Japanese akae is generally prepared as follows:10 the red paints are prepared by mixing and crushing for a long time (e.g., 100 h) a low-melting-point glass-frit powder, 10−25 wt % iron-oxide red pigment (α-Fe2O3: hematite), and a solvent in a wet process. The red paint is applied with a paint brush to a calcined porcelain body with a white glaze, and the painted porcelain is dried and then heat treated at 700−900 °C in air. The result is an akae work, which

Japanese Traditional Red Overglaze Ceramics, Akae. In the early 17th century, the Japanese artistic red overglaze ceramic ware akae was first created by Sakaida Kakiemon, who was instructed in this new skill, by hearsay, by a Chinese person living in Nagasaki, Japan. Since then, this overglaze has become extremely popular with porcelain lovers all over the world. In particular, Kakiemon-style ware was exported to Europe in the 17th and 18th centuries, enthralling the royalty and aristocracy of the time.1 The hallmark of Kakiemon-style ware is a graceful, clear akae with a simplified design inspired by nature that depicts soft, warm, well-spaced figures on a milky white background. Kakiemon-style ware strongly influenced pottery producers throughout Europe from the 17th century on, resulting in the production of numerous replicas called “Kakiemon Utsushi” by various pottery producers: Meissen in Germany, Sevres in France, Worcester in England, Delft in Holland, etc.2 The akae technique is now used by pottery producers around the world and numerous akae works may be © 2016 American Chemical Society

Received: February 4, 2016 Accepted: April 19, 2016 Published: April 19, 2016 10918

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

and vivid when the hematite-particle size is small and welldispersed, when the glass layer is thin, and when the solubility of hematite is low. Takada also developed stable leaded frits with 3−7 wt % Fe2O3, which inhibits the dissolution of hematite into the frit, and obtained bright and vivid red akae. Although Takada’s report constitutes a pioneering study that tried to scientifically elucidate the essence of akae, the effect of particle dispersibility of hematite and its microstructure on the color of akae remains insufficiently understood. Recently, although some detailed investigations have been done into akae of the old Imari and Chinese porcelain from the historical viewpoint by using a state-of-the-art analytical method, the essential coloring mechanism of akae is still not clear.7,8,35−41 The Aim of the Study. We believe that scientific studies on traditional technologies can provide artists with new inspirations and chemists with new concepts for creating novel functional materials. A comprehensive study of akae, including hematite red pigments, glass frits, the method of preparing red paints, and the mechanism behind the color of akae, is warranted because akae is a composite made of glass layers and hematite particles. Such a study should describe a method for easily controlling the color tone of akae. However, no such study on akae has been reported since Takada in 1958,10 and, to the best of our knowledge, no study exists detailing how to prepare akae with lead-free frits. In the present work, we prepared akae samples with lead-free frits that, due to concerns for human health and the environment, will soon be widely used in the porcelain industry. We also discuss the factors that control the color of akae and develop a simple method to prepare red paints for high-quality akae. In particular, we focus on particle size within the frit and the particle size and dispersibility of hematite powders. The akae samples thus prepared were analyzed mainly by light and electron microscopy. Based on this analysis we designed a facile and practical method to prepare red paints and composite particles for hematite and frits, from which we obtain high-quality akae.

is a composite in which hematite particles are dispersed in a glass layer.10 Hematite Red Pigments. Hematite has been widely used as a red pigment since prehistoric times. Recently, hematite has become an important material in nanotechnology because of its potential use as a catalyst, gas-sensing material, electrode material for lithium-ion batteries, and photoanode material for photoelectrochemical water-splitting reactions.11−18 Even today, research into hematite red pigments is widely conducted and novel red pigments are developed, with their color and thermostability precisely tuned by controlled nanostructure.19−29,9 The coloring mechanism at work in hematite powder was described by Takada in 1958, and the effect on color of particle size and dispersibility of hematite was clearly revealed.30 The color tone of hematite powder depends on its particle size and dispersibility from small and/or well-dispersed particles with a bright yellowish-red color to large and/or strongly aggregated particles that appear dark gray.30 In the early 18th century, a high-quality red pigment (Fukiya bengala) was developed in Japan and used by pottery producers throughout Japan, such as in Arita, Kutani, and Kyoto.4−8 However, manufacturing of Fukiya bengala was prohibited by an antipollution law in 1970, eliminating this pigment from the market. Recently, we analyzed existing Fukiya bengala in detail and found that its main component is fine hematite particles containing 1 mol % Al.5 We also synthesized hematitecorundum nanocomposites inspired by Fukiya bengala. The synthesized powders are expected to see use as red colorants for akae porcelain, because they exhibit a high-quality, thermally stable red color.9 Lead-Free Frits. The use of red paints commonly composed of glass frits that contain lead and have low melting points, is strictly regulated because lead is harmful to human health and to the environment. Therefore, many areas in Japan have worked on developing lead-free glass frits to produce porcelain.31−34 Along these lines, we developed lead-free glass frits for Kyo and Kiyomizu wares and provided the frits to other porcelain producers. However, the current lead-free frits do not yield high-quality akae, and overcoming this problem to obtain improved lead-free frits is strongly desired. In addition, the color of akae depends significantly on the particular porcelain producer, despite different producers using the same lead-free frits. In fact, each porcelain producer uses their own techniques developed by master potters, and these are passed down from generation to generation through an apprenticeship system. With their special skills acquired after so many years of experience, as well as a great deal of trial and error, the master potters create artistic works. Unfortunately, their techniques and skills are poorly documented in most cases, making it very important for the porcelain industry to clarify the coloring mechanism of akae and the underlying chemical reactions that take place during its production. Coloring Mechanism for Akae Created by Leaded Frits. In one of the few studies on the coloring mechanism in akae, Takada focused on particle size and dispersibility of hematite powders and revealed the following facts by making and analyzing akae samples which are prepared by using a leaded frit and hematite powders of various particle sizes and dispersibilities.10 Akae is a composite in which hematite particles are dispersed in a glass layer. The red color depends on the hematite-particle size, the dispersibility of the hematite particles, the thickness of the glass layer, and the solubility of hematite with respect to glass. The red color becomes bright

■

EXPERIMENTAL SECTION

Preparation of Hematite and Frit Powders. We prepared two types of hematite powders with the same primary particle size but with different particle-size distributions. Based on our previous report,9 we synthesized hematite powder using the polymer complex method. Fe(NO3)3·9H2O (Nacalai Tesque, 99%), citric acid (Wako Pure Chemical Industries, 98%), and ethylene glycol (Tokyo Chemical Industry, 99.5%) were used as reagents. The mixture of (Fe(NO3)· 9H2O):(citric acid):(ethylene glycol):(distilled water) = 1:5:15:50 (molar ratio) and was heated at ∼80 °C for 7 h to form metal-citrate complexes and subsequently at 170 °C for 12 h in ambient air. To remove organic components, the solids thus obtained were pyrolyzed at 445 °C for 24 h in ambient air. The powder obtained is called “medium-sized hematite” (M-hematite). M-hematite was ball-milled (ANZ-51S, Nitto Kagaku) in ethanol at 80 rpm for 65 h in a highdensity polyethylene container (325 mL) and 540 g zirconia balls with 5 mm-diameter. The slurry was adjusted to 1.5 vol %. The volume ratio of air, zirconia balls, and slurry was 5:3:2. After ball milling, the slurry was dried in ambient air at 60 °C overnight. The powder thus obtained is called “small-sized hematite” (S-hematite). Three kinds of frit powders with different primary particle size were prepared as follows: We used a commercially available alkali borosilicate glass frit (Kyo Muen Daigusuri, Kyoto Iwasaki) containing small amounts of ZnO. The chemical composition of the frit is listed in Table S1 in the Supporting Information. The as-obtained frit is called a “large-sized frit” (L-frit) and was crushed for 100 h by using an automatic mixing alumina mortar (Nittoh Kagaku) in a dry process. This powder is called a “medium-sized frit” (M-frit). L-frit was ball10919

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

Figure 1. Micrographs and particle-size distributions for frits and hematite. BSE-COMPO images for the (a) L-, (b) M-, and (c) S-frit powders. TEM images for (d) M- and (e) S-hematite powders. The large white contrasts seen in (d) and (e) are holes of carbon thin film for sample support. Particle-size distribution for (f) L-, M-, and S-frit powders and (g) M- and S-hematite powders. milled in ethanol at 80 rpm for 90 h in a zirconia container (1 L) with 1.8 kg of 5 mm-diameter zirconia balls. The slurry was adjusted to 20 vol %. The volume ratio of air, zirconia balls, and slurry was 5:3:2. After ball milling, the slurry was dried in air at 60 °C for 2 days. The powder thus obtained is called a small-sized frit (S-frit). The hematite and frit powders obtained were characterized by using scanning electron microscopy (SEM) (JIB-4500, JEOL), transmission electron microscopy (TEM) (JEM-2100F, JEOL), and laser particle size analyzer (LA-910, HORIBA). SEM images for frit powders were obtained as compositional images in backscattered-electron (BSE) mode. For TEM measurements, a powder sample was suspended in ethyl alcohol. The suspension was dropped onto a copper grid coated with a carbon thin film having holes with several micrometers diameter (Nisshin EM) and then dried. Preparation of Akae Samples. The general procedure for preparing akae samples is as follows: By using a porcelain mortar (101, Ishikawa Kojo), a hematite powder, a frit powder, and a green tea solution were mixed at a weight ratio of 4:0.4:2.5 for ∼10 min to obtain red paints. We used a green tea solution as a dispersing agent because the tannic acid in green tea is known to be a good dispersant.42 The red paints are denoted as, for example, “LM red paint” when made from the combination of L-frit and M-hematite. By using a writing brush, the red paint was applied to a white porcelain substrate, which was calcined with a transparent glaze under a reducing atmosphere at 1250 °C, and dried in air at room temperature. Finally the painted porcelain piece was heated at 780 °C for 5 min in ambient air at a heating rate of 2.5 °C/min and then cooled in the furnace. Various combinations of hematite and frit powders were examined. The samples thus obtained are called, for example, the “LM sample” when made from the combination of L-frit and M-hematite. We also developed a precombination treatment, which we apply once or three times for combining L-frit and M-hematite. The resultant samples are called “LM-P1” and “LM-P3”, respectively. The mixture of hematite, frit, and tea solution was mixed by using a porcelain mortar for ∼1 h until the water vaporized completely, following which an additional ∼2.5 mL of green tea solution was

poured onto the dried mixture. This operation was repeated either once or three times. The slurry was mixed again by the porcelain mortar for 10 min. The resultant red paints are called “LM-P1” and “LM-P3”. Subsequent operations are the same as described above in the general procedure. Color tone and reflection spectra were obtained by spectrophotometry (CM-2600d, Konica Minolta Sensing) by using a CIE Standard Illuminant D65. The color measurements were done at least three times on the red overglaze enamels in places where the substrate was not visible. In the CIE 1976 L*a*b* color space, L* indicates lightness and a* and b* indicate chromaticity (i.e., color directions). Positive and negative values of a* indicate reddish and greenish colors, respectively, whereas positive and negative values of b* indicate yellowish and bluish colors, respectively. The measurement window is round with a diameter of 3 mm. Constituent phases were determined by X-ray diffraction (Ultima IV, Rigaku) with monochromatic Cu−Kα radiation. The microstructure was examined by using an polarized light microscope (Optiphoto, Nikon), SEM compositional BSE images (BSE-COMPO), and TEM. For TEM measurements, ultrathin samples were prepared by using a focused-ion-beam (FIB) system (JIB-4500, JEOL) operated at an acceleration voltage of 30 kV to reduce thickness of the sample to less than 100 nm.

■

RESULTS AND DISCUSSION Characterization of Raw Material Powders. Figures 1a− c show SEM images for as-obtained and crushed frit powders. Particle sizes for L, M, and L-frit were 2−100, 2−20, and ∼2 μm, respectively. Figures 1d and e show TEM images for asprepared and crushed hematite (M-hematite and S-hematite) powders, respectively. The two powders both had primary particle sizes in the range of 40−140 nm, and the primary particles were aggregated to form secondary particles with diameters of ∼6 and ∼1 μm in M- and S-hematite, respectively. Particle-size distributions of the frit and hematite powders are shown in Figures 1f and g. L-frit shows broad left-skewed 10920

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces frequency distribution with a maximum frequency at 42.2 μm, whereas M- and S-frit have sharp monomodal distributions with maximum frequencies at 4.2 and 1.2 μm, respectively. Mhematite has a very broad left-skewed distribution with a maximum frequency at 27.6 μm, and S-hematite has a monomodal distribution with a maximum frequency at 0.82 μm. The particle sizes measured by various methods are summarized in Table S2 in the Supporting Information. Microscopic observations gave results consistent with these particle-size distributions. Three types of frits and two types of hematite powders with different particle sizes were thus obtained. We used the traditional empirical method to crush frit particles, which calls for using an alumina mortar for a long time (e.g., ∼100 h)10 to obtain single-micron-sized M-frit particles, as opposed to modern crushing technology that uses ball milling to obtain sub- to single-micron-sized particles of Sfrit. Ball milling also drastically downsized agglomeratehematite particles, but obtaining monodispersed hematite particles by this method remains difficult. Characterization of Dried Red Paints. SEM images and particle-size distributions for dried red paints are shown in Figure S1. At first glance, the mixing does not seem to change the size of the frit particles. For mixed powders of M-hematite of each frit, strongly aggregated large white contrasts (aggregated hematite particles) were observed (Figures S1a− c). For mixed powders of S-hematite of each frit, white contrasts are distributed over the entire image (Figures S1d−f). These results indicate that S-hematite particles are easily dispersed and that they distribute evenly within the spaces between frit particles. The particle-size distributions shown in Figures S1g and h should reflect the particle-size distributions within frits because the total amount of frit in the red paints was very high (frit:hematite = 22:1 in volume). For the combination of S-hematite with each frit, the particle-size distributions for MS and SS red paints were slightly broadened, and the distribution for LS red paint shifted toward smaller particles (Figure S1h). For the combination of M-hematite with each frit, the same trend was observed, although more prominently (Figure S1g) than for the combination of S-hematite with each frit. These results suggest that large particles in L-frit were crushed during mixing and that the aggregated state of hematite affects the particle size-distribution of frit powders after mixing: the strongly aggregated hematite powders agglomerate frit particles, and well-dispersed hematite powders only slightly affect the dispersibility of frit particles. The XRD patterns of dried red paints are shown in Figure S2. The broad bulge centered around 2θ = 23°, which corresponds to frit, and the sharp Bragg peaks of hematite are observed in all dried red paints, indicating that the mixing operation does not affect the atomistic structure of hematite and frit powders. The crystallite size calculated from the halfmaximum full-width of the hematite (104) plane located at 2θ = 33.15° is 50 nm, which is closely consistent with results based on TEM and particle-size distributions. XRD Measurements of Akae. Figure 2 shows XRD patterns for akae samples. The broad bulge of frit and the peaks of hematite are observed, as for the dried red paints. Crystallite sizes obtained for the (104) plane increased from 50 nm for dried red paints to 70−80 nm for akae samples, which indicates that the hematite particles grew during heating. Bragg peaks corresponding to cristobalite (SiO2), quartz (SiO2), and zinc ferrite (ZnFe2O4) were also observed from all akae samples. SiO2 could form by crystallization of the silicon oxide contained

Figure 2. XRD patterns for the various akae samples.

in the frit, and zinc ferrite could form by a reaction between the zinc in the frit and the iron dissolved from the hematite. The formation of SiO2 crystals does not affect the color of akae because they are transparent crystals. However, the formation of large quantities of zinc ferrite could affect the color of akae because zinc ferrite is a brown-colored crystal. In the present study, the XRD peaks corresponding to zinc ferrite are very weak, so the quantity produced is considered to be extremely small. Therefore, the effect of zinc ferrite on the color of akae is considered to be negligible. However, the quantity of zinc ferrite produced changes upon varying the condition under which akae is prepared. In terms of color control, new compositional glass frits that do not produce colored crystals are preferable for akae production. The development of optimal frits for akae would make an excellent topic for future research. Measurements of Akae Color. Figure 3a shows photographs of akae samples. The LM, MM, and SM samples appear from upper left to upper right and LS, MS, and SS samples appear from lower left to lower right. The color of the samples systematically changes along this progression. The reflectance curves and L*, a*, and b* values of samples appear in Figures 3b and c. The reflectance edge for all samples appears near 580 nm, and small bulges appear near 630 and 720 nm. The spectra are similar to those for hematite powders,9 which indicates that the akae color comes from the hematite particles. The reflectance increases as frit-particle size decreases for constant hematite-aggregated-particle size. Comparing reflectance values which link to L*, a*, and b* values, we obtain LM ≃ MM < LS < SM ≃ MS < SS. For samples made from M-hematite, the results for L*, a*, and b* from the samples using L- and M-frits are nearly the same, whereas these parameters increase for samples made from S-frit. For samples made from S-hematite, the L*, a*, and b* parameters increase as the frit-particle size decreases, with the SS sample having the highest values. The color of akae is thus found to depend on frit-particle size when the hematite-aggregated-particle size is constant: the akae color gets better as the frit-particle size decreases. The L*, a*, and b* parameters for LM, MM, and LS samples have almost the same low values, indicating that high L*, a*, and b* values cannot be obtained for excessively large fritparticle size. Comparting now the SM and LS or MS samples, we see that the SM sample has greater L*, a*, and b* values than the LS sample and nearly the same a* and b* values as the MS sample, although a slightly smaller L* value. This result indicates that high L*, a*, and b* values can be obtained by using S-frit even if it is combined with strongly aggregated Mhematite. These results indicate that frit-particle size is more 10921

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

was large, and the hematite was distributed as aggregated particles ∼20 μm in size (Figure 4a). For MM and SM samples, the glass region and aggregate-hematite particles gradually decreased in size (Figures 4b and c, left), and relatively large aggregate-hematite particles appeared (Figures 4b and c, right). The polarized light microscopy images for the LS and MS samples (Figures 4d and e) are analogous to those for the LM and SM samples, respectively (Figures 4a and c). For the SS sample, the glass region and aggregate-hematite particles are small, and the hematite particles are evenly distributed (Figure 4f). These results show that the size of the glass region correlates with the size of the aggregate-hematite particles: when the one is large (small), the other is large (small). We also find a correlation between the results of color measurements (Figure 3) and polarized light microscopy images (Figure 4a−f). The parameters L*, a*, and b* tend to have high (low) values when the glass region and aggregatehematite particles are small (large). For the LM (Figure 4a) and LS (Figure 4d) samples, the glass region and aggregatehematite particles are large. For the MM (Figure 4b) sample, although the glass region is smaller than for the LM (Figure 4a) and LS (Figure 4d) samples, large, strongly aggregated hematite particles are sometimes observed. Therefore, the L*, a*, and b* values for these three samples (LM, LS, and MM (Figure 4a, d, and b)) could have similar low values. For the same reason, the L*, a*, and b* values for the SM (Figure 4c) and MS (Figure 4e) samples could have similar values. The SS (Figure 4f) sample has the smallest glass region and the smallest aggregatehematite particles, resulting in the highest values for L*, a*, and b*. These results are consistent with the accepted understanding that the dispersibility of hematite particles contributes strongly to the color of akae. However, we also find that, for controlling the color of akae, not only particle size and dispersibility of hematite powders must be considered but also the particle size of frit powders. SEM Images of Akae. Although polarized light microscopy gives us general information concerning the dispersibility of hematite particles in the akae glass layer, it does not give detailed information on the distribution of hematite particles. BSE-COMPO images, whose image contrast depends on atomic number, were acquired by SEM (Figures 4g−j). Hematite particles, whose main component is iron (with high atomic number), appear as bright white, whereas glass regions, whose main component is silicon (with low atomic number), appear dark. As with the results of polarized light microscopy, LM and LS samples have large glass regions and large aggregate-hematite particles (Figures 4g and i, left), and the SM and SS samples have small glass regions and small aggregate-hematite particles (Figures 4h and j, left). Although M-hematite contains large aggregate-hematite particles with a size of 27.6 μm, the size decreases to ∼10 μm and ∼1 μm for LM and SM samples (Figures 4g and h), respectively, after making akae. The aggregate-hematite particles could be crushed by using a porcelain mortar when the red paints were prepared, presumably because S-frit works as a crushing medium, resulting in drastically decreasing the size of the aggregatehematite particles from 27.6 μm to ∼1 μm in the SM sample. Focusing first on the LS and SS samples made from S-hematite with an aggregate size of 0.82 μm, we find that hematite forms large aggregate particles ∼20 μm in size in the LS sample (Figure 4i), whereas highly dispersed hematite preserves the original size of aggregates in the SS sample (Figure 4j). The dispersibility of aggregate-hematite particles in the SS sample is

Figure 3. (a) Photographs, (b) reflectance curves, and (c) plots of a*, b*, and L* for the various samples. Error bars are indicated on the plots of L*, a*, and b*.

important than the size of aggregated hematite particles. The SS sample has the highest L*, a*, and b* values of all samples, indicating that the combination of small frit and well-dispersed hematite drastically enhances the color of akae. The fact that the frit-particle size strongly contributes to the color of akae is a new insight, because particle size and hematite dispersibility were heretofore thought to be the main factors controlling the color of akae. Polarized Light Microscopy of Akae. Next, we used light microscopy to investigate the dispersibility of hematite particles in the akae glass layer. Figures 4a−f show polarized light microscopy images of various samples. Reddish-brown particles are aggregate-hematite particles, and the transparent moiety whose inner depth is dark is the glass region in which hematite particles are not present. For the LM sample, the glass region 10922

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

Figure 4. Micrographs for the various samples. Polarized light microscopy images for (a) LM, (b) MM, (c), SM, (d) LS, (e) MS, and (f) SS samples. Right-side images in panels (b) and (c) are enlarged images. BSE-COMPO images for (g) LM, (h) SM, (i) LS, and (j) SS samples. Right-side images are enlarged images. TEM images for (k) LM, (l) SM, (m) LS, and (n) SS samples.

good, and the glass region that is free from hematite particles is narrow. The reason that strongly aggregated-hematite particles form is as follows: A red paint prepared by mixing frit, hematite, and a green tea solution is painted onto the porcelain sample, which is then dried and heat treated at high temperature to obtain an akae sample. The size of aggregate-hematite particles is basically less than that of frit particles, and the amount of hematite is less than that of frit (frit:hematite = 22:1 in volume). After applying a red paint to the porcelain sample, the red paint slurry loses its fluidity, and the frit particles in the slurry simultaneously rearrange and are fixed in place by the decrease in volume caused by vaporizing water. After the frit particles are fixed, the surface tension of the water moves aggregate-hematite particles

to the spaces between frit particles during further water vaporization, until finally the hematite particles become fixed in the interspaces and the water is completely vaporized. When the frit particles are sufficiently small, the hematite particles assemble in the interspaces to form small aggregates, because the interspace volume is small. In contrast, when the frit particles are large, the interspace volume between frit particles expands, resulting in the formation of larger aggregate-hematite particles. For this reason, large aggregate-hematite particles form in the LS sample. The state of dispersion of frit particles and aggregate-hematite particles in dried red paint may thus reveal the dispersibility of hematite particles in the final akae glass layer. 10923

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

interspaces with diameters of 0.2 and 0.5 μm, respectively. Although these values are relatively less than experimental results for aggregate-hematite-particle sizes for LS (∼10 μm) and SS (∼1 μm) samples, their size scale is roughly consistent with experimental results. This indicates that the reaggregation of hematite particles in the interspaces of frit particles during the drying process of red paint is quite reasonable. The optimal particle size of frit suitable for 100 nm hematite particles is calculated from triangle to square lattices to be 0.2−0.6 μm. Thus, to obtain high-quality akae, we should calculate the optimal combination of particle sizes of hematite and frit and use raw materials with particle sizes close to the calculated values. In addition, it is possible to delicately tune the color by shifting particle size from the optimal size. However, this method contains fatal disadvantages. Two general methods exist for akae painting: the first is the traditional handwriting method, and the second is the transferprinting method, which is superior to mass production. We used the former method in this study. For small frit (e.g., S-frit of 1.2 μm diameter), the total surface area of the frit particles is large, and the appropriate fluidity cannot be obtained unless a massive amount of solvent is used. Upon painting by handwriting with the red paint prepared from the combination of S-frit and S-hematite with the same amount of solvent for Lor M-frit and M- or S-hematite, the rheological property and drawing taste differ from the conventional ones. A drastically varying drawing taste is not preferred, because akae porcelain is drawn by artisans. Controlling the slurry fluidity of red paints is very important in obtaining a drawing taste that is the same as the conventional one. For transfer-printing, which is superior to mass production, red paint slurry is prepared in the same way as for the handwriting method, and transfer-printing papers are prepared from the red paint. Obtaining an appropriate fluidity requires a large amount of solvent in the slurry because the concentration of solid components becomes low, and the transfer-printing paper could crack because of significant shrinkage due to drying. In addition, a lot of energy and time are necessary for crushing frit particles and aggregate-hematite particles. Precombination Treatment. We now propose a simple preparation process for red paints that solves all above problems. The raw materials used are L-frit and M-hematite which are easily obtainable. We find that frit-particle size is an extremely important factor for akae coloring, which indicates that the combination of frit- and hematite-particle sizes must be optimized. Thus, for obtaining high-quality akae, having hematite particles dispersed in the interspaces between frit particles, making a three-dimensional network, is preferable over having aggregate-hematite particles isolated in the interspaces between frit particles. Here we focus on a mechanofusion process,43 which is a well-known method for preparing functional particles for lithium-ion battery electrodes and pharmaceutical agents. The particle design targeted is a core−shell-like structure which hematite primary particles are fixed to the surface of frit particles to form a single-particle layer. To create such a structure, the reaggregation of hematite particles during the drying of red paint should be prevented. To this end, we tried hybridizing hematite and frit particles by using the following simple method that focuses on the similarity between commercially available mechanofusion devices and mixing by using mortar and pestle. L-frit, M-hematite, and the green tea solution were mixed until the slurry dried, after which 2.5 mL of the green tea

Here, if using excessively large frit particles, a good color cannot be obtained even if well-dispersed hematite particles are used, because the reaggregation of hematite may occur while the red paint dries. For small frit particles, a good color can be obtained even if strongly aggregated-hematite particles are used, because the frit particles may work as a crushing medium. These results and analysis suggest that finding the optimal combination of frit-particle size and aggregate-hematite particle size is vital to obtaining high-quality akae. Cross-Sectional TEM Imaging of Akae. To obtain further information concerning the state of dispersion of hematite in the akae glass layer, we acquired cross-sectional TEM images (Figures 4k−n). In akae, the primary particle size varies from 40−140 nm for M- and S-hematite to 40−300 nm for hematite. This particle growth could be caused by heating. By applying the energy-dispersive X-ray microanalysis in glass regions with no hematite particles, iron and zinc are detected in all samples. This iron and zinc may come from dissolution of hematite particles and original components in frit, respectively. Takada explained the particle-growth mechanism in leaded glass frit as follows:10 When a mixed hematite-frit powder is heated above the melting point of frit, a portion of the hematite primary particle dissolves in the melted frit. The smaller the size of the hematite particles is, the faster the hematite particles dissolve. Hematite has certain particle size distribution from small (20− 50 nm) to large (110−600 nm). Although the smaller particles dissolve in the melted frit and completely disappear, only the surfaces of the large particles dissolve. Continuing the heat treatment causes saturation of the dissolved iron in the glass around the large particles and hematite precipitates on the large particles, resulting in particle growth. The broader the particlesize distribution, and the smaller particles are present, the faster the particle growth occurs. In our system of lead-free frit and hematite, crystal growth of hematite particles may occur based on a mechanism analogous to that proposed by Takada. Determining the mechanism more precisely will require a detailed study of the crystal growth mechanism and thermal behavior of the frits, pigments, and mixtures, which we plan in future work. We next focus on the state of dispersion of hematite particles. For samples made from M-hematite (LM and SM; Figures 4k and l), hematite primary particles strongly aggregate to each other, resulting in narrow interspaces between each particle. In contrast, for samples made from S-hematite (LS and SS; Figures 4m and n), hematite primary particles only weakly aggregate to each other, so the interspaces between each particle are relatively broad. These results suggest that reaggregation forces generated by water vaporization in the red paints for LS and SS samples suffice to gather aggregatehematite particles into the interspaces of frit particles but are insufficient in forcing primary particles into intimate contact with each other. Optimal Combination of Frit- and Hematite-Particle Sizes. Next, we consider the optimal combination of frit- and hematite-particle sizes based on the results and discussion above. The median diameters derived from the particle-size distributions for L- and S-frit (Figure 1f) are 21.6 and 1.2 μm, respectively. Assuming that frit particles are spheres, we calculate the diameter of spheres that would fit into the interspaces made by two-dimensional triangle and square lattices of frit spheres (Figure S3). For L-frit, triangle and square lattices provide interspaces with diameters of 3.4 and 8.9 μm, respectively. For S-frit, triangle and square lattices provide 10924

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

values than those for the SS sample, which has the most striking color, as mentioned above. Interestingly, the LM-P3 sample has better color than the SS sample, although frit-particle size in LM-P3 red paint is larger than that of S-frit particles. The dispersion of hematite in LM-P1 and LM-P3 samples is confirmed by polarized light microscopy, BSE-COMPO, and TEM images, as shown in Figure 6. The polarized light microscopy images for samples prepared by precombination treatment resemble those of the MS samples, suggesting a downsizing of L-frit particles. As can be seen in the lowmagnification BSE-COMPO image (Figure 6c, left), dispersion of hematite particles for LM-P1 is intermediate between that of LM (Figure 4g, left) and SS (Fi.g 4j, left) samples. In the higher-magnification image of LM-P1 (Figure 6c, right), the moiety that dispersion of hematite primary particles is better (Figure 6c, right, red dotted circles) than in that of the SS sample (Figure 4j, right) is observed, although the aggregatehematite particles are larger than in the SS sample. In the lowmagnification BSE-COMPO image for the LM-P3 sample (Figure 6d, left), although the hematite-particle-free glass region is larger than in the SS sample (Figure 4j, left), the total dispersion of hematite particles seems to be better than in the SS sample. In the higher-magnification image of the LM-P3 sample (Figure 6d, right), very small aggregate-hematite particles close to monodispersed particles are highly dispersed. We find the distinctive dispersion of LM-P1 and LM-P3 samples in the TEM images of Figures 6e and f: hematite particles surround the glass region that was originally considered to be a frit particle (Figures 6e and f, red arrows). Small aggregate-hematite particles surround the large glass region in the LM-P1 sample, and hematite particles close to primary particles surrounded the large and small glass regions. Considering these results, we propose in Figure 7 schematic drawings of the distribution of frit and hematite particles in dried red paints. For the combination of S-frit and S-hematite, hematite particles lead to aggregates in the interspaces between frit particles, and aggregate-hematite particles are not continuously distributed (Figure 7a). However, in singleprecombination treatment red paint, frit particles are covered with small aggregate-hematite particles and hematite particles are distributed in a continuous manner (Figure 7b). In tripleprecombination treatment red paint, frit particles are covered with hematite particles close to primary particles and hematite particles are continuously distributed (Figure 7c). In the present akae system, the precombination treatment could create the desired frit and hematite core−shell-like structures. By fixing the composite to akae, hematite particles form a threedimensional network structure in the glass layer, realizing an ideal dispersibility and resulting in the most striking red color for the LM-P3 sample. The reason the core−shell-like structure occurs after the precombination treatment is not clear, but many intricately related factors, such as shearing force generated from mortar-and-pestle impacts and binder effects of organic components contained in the green tea solution, could be at work. This proposed process of the precombination treatment allows red paints to be fabricated for high-quality akae even if frit with large particles and hematite with strongly aggregated particles are used as raw materials.

solution was added again to the dried powder. The resulting product was mixed for 10 min before painting. This operation is called a “precombination treatment”. This treatment was repeated two additional times. We expected the following three effects: (i) downsizing of large frit particles by mortar and pestle; (ii) downsizing of aggregate-hematite particles by frit particles; (iii) immobilization of hematite particles on the frit particles by condensation and shearing forces by mortar-andpestle impact. The third effect could be prominent when the concentration of solid components increases because of water vaporization. SEM images and particle-size distributions of dried red paints prepared by the single- and tripleprecombination treatments are shown in Figure S4. The Lfrit particles are crushed to almost the same size as the M-frit particles in both samples. Figure 5 shows the results of color

Figure 5. (a) Reflectance curves for LM, SS, LM-P1, and LM-P3 samples. Insets show photographs for LM-P1 (left) and LM-P3 (right) samples. (b) Plots of a*, b*, and L* for LM, MS, SS, LM-P1, and LMP3 samples. MS is shown as the red diamond. Error bars are indicated on the plots of L*, a*, and b*.

measurement of LM-P1 and LM-P3 samples made by applying the single- and triple-precombination treatments, respectively. For comparison, LM, MS, and SS samples are also shown. The reflectance curve is similar for all samples, and the reflectance values are in the following order: LM < MS < LM-P1 < SS < LM-P3. The LM-P1 sample has drastically higher L*, a*, and b* values than those for the LM sample but nearly the same values as the MS sample. This result suggests that L-frit particle size decreases to become the same size as M-frit particles. Surprisingly, the LM-P3 sample has higher L*, a*, and b*

■

CONCLUSIONS We studied the factors that control the color of akae by observing the particle size of glass frit and hematite and, 10925

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

Figure 6. Polarized light microscopy images for (a) LM-P1 and (b) LM-P3 samples. BSE-COMPO images for (c) LM-P1 and (d) LM-P3. Right images of panels (c) and (d) are enlarged images. TEM images for (e) LM-P1 and (f) LM-P3 samples.

Figure 7. Schematic drawings showing distributions of frit and hematite particles (blue and red circles, respectively) in red paints. (a) SS, (b) LM-P1, and (c) LM-P3 red paints.

inspired by a recent particle-design method, developed a facile and practical process for preparing red paints. For a constant aggregate-hematite particle size, L*, a*, and b* values for an akae sample increase as the frit-particle size decreases, presumably because hematite particles became highly dispersed in the glass layer for decreasing frit-particle size. However, for an excessively large frit-particle size, the dispersion of hematite does not improve, even upon using weakly aggregate-hematite powder. These results indicate that considering particle size and dispersibility of hematite powder is not sufficientcontrolling frit-particle size is extremely important for obtaining highquality akae, suggesting the existence of an optimal combination of frit- and hematite-particle sizes. We calculated the optimal frit- and hematite-particle sizes by regulating frit- or hematite-particle sizes (e.g., the optimal frit-particle size is 0.2− 0.6 μm in 100 nm hematite). However, from a practical viewpoint, using submicron-sized frit is difficult. We thus developed a simple process for preparing red paints that leads to high-quality akae and that use large frit particles and strongly aggregate-hematite particles, which are easy to obtain. This method is very simple: red paint is mixed in a mortar until it dries, at which point solvent is added again, and the result is further mixed. Repeating this precombination treatment three times, we obtain akae samples with higher L*, a*, and b* values than for those prepared using submicron-sized frit and weakly aggregate-hematite particles. The red paints prepared by this

precombination method may have a frit-hematite core−shelllike structure. Hematite particles close to primary particles could become highly dispersed in a three-dimensional network, resulting in high-quality akae. Because the mortar and pestle used in the proposed precombination treatment are commonly used instruments by porcelain artisans, the method is very easy to implement. We thus believe that this method will become widely used in the porcelain industry to prepare red paints.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b01549. Chemical composition of the frit, particle sizes of frit and hematite powders, and the SEM images, the particle-size distributions, and the XRD patterns for dried red paints (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone/Fax: 81 426284537. E-mail: [email protected]. jp. Notes

The authors declare no competing financial interest. 10926

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces

■

(19) Katsuki, H.; Komarneni, S. Microwave-Hydrothermal Synthesis of Monodispersed Nanophase α-Fe2O3. J. Am. Ceram. Soc. 2001, 84, 2313−2317. (20) Katsuki, H.; Komarneni, S. Role of α-Fe2O3 Morphology on the Color of Red Pigment for Porcelain. J. Am. Ceram. Soc. 2003, 86, 183− 185. (21) Bondioli, F.; Ferrari, A. M.; Leonelli, C.; Manfredini, T. Synthesis of Fe2O3/Silica Red Inorganic Inclusion Pigments for Ceramic Applications. Mater. Res. Bull. 1998, 33, 723−729. (22) Hosseini-Zori, M.; Bondioli, F.; Manfredini, T.; Taherinassaj, E. Effect of Synthesis Parameters on a Hematite−Silica Red Pigment Obtained Using a Coprecipitation Route. Dyes Pigm. 2008, 77, 53−58. (23) Hosseini-Zori, M.; Taheri-Nassaj, E. Nano Encapsulation of Hematite into Silica Matrix as a Red Inclusion Ceramic Pigment. J. Alloys Compd. 2012, 510, 83−86. (24) Zhang, Y.; Rao, P.; Lü, M.; Zeng, D.; Wu, J. Synthesis and Color Evolution of Silica-Coated Hematite Nanoparticles. J. Am. Ceram. Soc. 2009, 92, 1877−1880. (25) Zhao, F.; Li, W.; Luo, H. Sol−Gel Modified Method for Obtention of Gray and Pink Ceramic Pigments in Zircon Matrix. J. SolGel Sci. Technol. 2009, 49, 247−252. (26) Zhao, F.; Gao, Y.; Luo, H. Reactive Formation of Zircon Inclusion Pigments by Deposition and Subsequent Annealing of a Zirconia and Silica Double Shell. Langmuir 2009, 25, 13295−13297. (27) Kim, Y.; Pee, J.-H.; Chang, J. H.; Choi, K.; Kim, K. J.; Jung, D.Y. Silica Effect on Coloration of Hematite Nanoparticles for Red Pigments. Chem. Lett. 2009, 38, 842−843. (28) Yu, R.; Kim, Y.; Pee, J. H.; Kim, K. J.; Kim, W. Thermal Behavior and Coloration Study of Silica-Coated-Fe2O3 and-FeOOH Nanocapsules. J. Nanosci. Nanotechnol. 2011, 11, 6283−6286. (29) Hashimoto, H.; Asaoka, H.; Nakano, T.; Kusano, Y.; Ishihara, H.; Ikeda, Y.; Nakanishi, M.; Fujii, T.; Yokoyama, T.; Horiishi, N.; Nanba, T.; Takada, J. Preparation, Microstructure, and Color Tone of Microtubule Material Composed of Hematite/Amorphous-Silicate Nanocomposite from Iron Oxide of Bacterial Origin. Dyes Pigm. 2012, 95, 639−643. (30) Takada, T. On the Effects of Particle Size and Shape on the Colour of Ferric Oxide Powders. Funtai oyobi Funmatsu Yakin 1958, 4, 160−168. (31) Yoshida, S.; Tsutsumi, Y. Properties of Lead-Free Frits for Glazes. Interceram 1992, 41, 173−175. (32) Tomura, S. Regulations of Lead Release from Potteries and Its Countermeasures. Ceram. Jpn. 1997, 32, 159−163. (33) Yoshida, S.; Hukunaga, A. Countermeasures for Lead Release from Japanese Colors of Arita Ware and Hasami Ware. Ceram. Jpn. 1997, 32, 167−169. (34) Imai, K.; Yano, H.; Kawauchi, T. Countermeasures for Lead Release from Japanese Colors of Kyo Ware and Kutani Ware. Ceram. Jpn. 1997, 32, 170−172. (35) Hidaka, M.; Horiuchi, H.; Ohashi, K.; Wijesundera, R. P.; Kumara, L. S. R.; Sung, N. E. Local Structures and Electronic Band States of α-Fe2O3 Polycrystalline Particles Included in the Red-Color Overglazes and the Transparent Glazes of the Kakiemon-Style Porcelains by Means of X-ray Absorption Spectra (II). Ceramica 2009, 55, 223−232. (36) Hidaka, M.; Ohashi, K.; Kajihara, S.; Wijesundera, R.; Kumara, L.; Choi, J.; Sung, N. Structural Properties of the Red-Color Overglaze for the HIZEN Porcelains Produced in the early Edo Period of Japan. Ceram. Int. 2009, 35, 875−886. (37) Hidaka, M.; Ohashi, K.; Wijesundera, R. P.; Kumara, L. S. R.; Sugihara, S.; Momoshima, N.; Kubuki, S.; Sung, N. E. Local Sructures and Electronic Band States of α-Fe2O3 Polycrystalline Particles in the Glazes of the HIZEN Celadons Produced in the Edo Period of Japan, by Means of X-ray Absorption Spectra (II). Ceramica 2011, 57, 155− 165. (38) Hidaka, M.; Takeuchi, K.; Wijesundera, R. P.; Kumara, L. S. R.; Watanabe, M.; Choi, J. Y.; Sung, N. E. Correlation Between the Greenlike Coloration and the Structural and Electronic Properties of Celadon Glazes (I). Ceramica 2012, 58, 328−337.

ACKNOWLEDGMENTS We thank Mr. Tadanori Yokoyama and Mr. Yuya Arakawa for helpful discussions. This study was financially supported by the Special Funds for Education and Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan and by the Kazuchika Okura Memorial Foundation.

■

REFERENCES

(1) Yabe, Y.; Saito, T.; Itoh, Y.; Sasaki, H.; Oka, Y.; Arakawa, M.; Karasawa, M.; Kaneko, K. The Concise History of Japanese Ceramics, 5th ed.; Bijutsu Shuppan-Sha: Tokyo, Japan, 2007. (2) Sasaki, T. Toujiki, Umi wo Yuku; Zoshinkai Publishers Inc.: Sizuoka, Japan, 1999; p 238. (3) Ohno, Y. CIE Fundamentals for color measurements. Paper for IS&T NIP16 Conference, Canada, Oct. 2000, 16−20, 1−6. (4) Nagao, T. Fukiya no Hanashi ∼ Dentou to Bunka no Machi Dai San Syu∼; The Sanyo Shimbun: Okayama, Japan, 1976; p 193. (5) Asaoka, H.; Kusano, Y.; Nakanishi, M.; Fujii, T.; Takada, J. Characterization and Reproduction of “Fukiya Bengara”, Noble Red Color Pigment, by Fe2O3-Al2O3 System. Funtai oyobi Funmatsu Yakin 2003, 50, 1062−1067. (6) Kondo, Y.; Yoshida, A. Zusetsu Okayama-ken no Rekishi; Kawade Shobo Shinsha: Tokyo, Japan, 1990. (7) Kajihara, S.; Hidaka, M.; Wijesundera, R. P.; Kumara, L. S. R.; Koga, M.; Kobayashi, S.; Tsuru, T.; Koga, K.; Shimomura, K.; Choi, J.Y.; Sung, N. E.; Park, Y. J. Correlation Between the Izumiyama Porcelain Ceramics and the Red-Overglaze Enamels of the KakiemonStyle Porcelains. Ceram. Int. 2008, 34, 1681−1689. (8) Hidaka, M.; Horiuchi, H.; Ohashi, K.; Wijesundera, R.; Kumara, L.; Choi, J. Y.; Park, Y. J. Structural Properties of the Red-Color Overglazes on the Kakiemon-Style Porcelains Produced in the Later 17th Century by Means of X-ray Diffraction (I). Ceramica 2009, 55, 120−127. (9) Hashimoto, H.; Nakanishi, M.; Asaoka, H.; Maeda, T.; Kusano, Y.; Fujii, T.; Takada, J. Preparation of Yellowish-Red Al-Substituted αFe2O3 Powders and Their Thermostability in Color. ACS Appl. Mater. Interfaces 2014, 6, 20282−20289. (10) Takada, T. Studies on Iron Red Glazes. Funtai oyobi Funmatsu Yakin 1958, 4, 169−186. (11) Wagloehner, S.; Reichert, D.; Leon-Sorzano, D.; Balle, P.; Geiger, B.; Kureti, S. Kinetic Modeling of the Oxidation of CO on Fe2O3 Catalyst in Excess of O2. J. Catal. 2008, 260, 305−314. (12) Huo, L.; Li, W.; Lu, L.; Cui, H.; Xi, S.; Wang, J.; Zhao, B.; Shen, Y.; Lu, Z. Preparation, Structure, and Properties of Three-Dimensional Ordered α-Fe2O3 Nanoparticulate Film. Chem. Mater. 2000, 12, 790− 794. (13) Chen, J.; Xu, L.; Li, W.; Gou, X. α-Fe2O3 Nanotubes in Gas Sensor and Lithium-Ion Battery Applications. Adv. Mater. 2005, 17, 582−586. (14) Cabana, J.; Monconduit, L.; Larcher, D.; Palacin, M. R. Beyond Intercalation-Based Li-Ion Batteries: the State of the Art and Challenges of Electrode Materials Reacting Through Conversion Reactions. Adv. Mater. 2010, 22, E170−92. (15) Ji, L.; Lin, Z.; Alcoutlabi, M.; Zhang, X. Recent Developments in Nanostructured Anode Materials for Rechargeable Lithium-Ion Batteries. Energy Environ. Sci. 2011, 4 (8), 2682−2699. (16) Björkstén, U.; Moser, J.; Grätzel, M. Photoelectrochemical Studies on Nanocrystalline Hematite Films. Chem. Mater. 1994, 6, 858−863. (17) Seki, M.; Yamahara, H.; Tabata, H. Enhanced Photocurrent in Rh-Substituted α-Fe2O3 Thin Films Grown by Pulsed Laser Deposition. Appl. Phys. Express 2012, 5, 115801. (18) Yang, T.-Y.; Kang, H.-Y.; Jin, K.; Park, S.; Lee, J.-H.; Sim, U.; Jeong, H.-Y.; Joo, Y.-C.; Nam, K. T. An Iron Oxide Photoanode with Hierarchical Nanostructure for Efficient Water Oxidation. J. Mater. Chem. A 2014, 2, 2297. 10927

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928

Research Article

ACS Applied Materials & Interfaces (39) Hidaka, M.; Takeuchi, K.; Wijesundera, R. P.; Kumara, L. S. R.; Sugihara, S.; Momoshima, N.; Kubuki, S.; Sung, N. E. Structural and Electronic Properties of Iron Oxides in the Celadon Glazes (II). Ceramica 2012, 58, 534−541. (40) Wang, L.; Zhu, J.; Yan, Y.; Xie, Y.; Wang, C. Micro-Structural Characterization of Red Decorations of Red and Green Color Porcelain (Honglvcai) in China. J. Raman Spectrosc. 2009, 40, 998− 1003. (41) Hao, W.; Luo, W.; Chen, Y.; Wang, C. Micro-Analysis on Chinese Over-glaze Red Decoration of Linshui Kiln from Jin Dynasty (1115−1234ad). J. Raman Spectrosc. 2014, 45, 224−227. (42) Ohnishi, S. Tougei no Dentou Gihou, 9th ed.; Rikogakusha Publishing Co., Ltd.: Tokyo, Japan, 1986; p 291. (43) Yokoyama, T.; Urayama, K.; Naito, M.; Kato, M.; Yokoyama, T. The Angmill Mecanofusion System and its Applications. KONA 1987, 5, 59−68.

10928

DOI: 10.1021/acsami.6b01549 ACS Appl. Mater. Interfaces 2016, 8, 10918−10928