# Fileset

[2_BAS_YSZ_GO.docx](https://mdr.nims.go.jp/filesets/de05d1b1-b785-4999-a4e1-ffe23fbabe40/download)

## Creator

Shingo Machida, Masaya Suzuki, Toshimichi Shibue, [Gaku Okuma](https://orcid.org/0000-0002-2997-9166), Daiki Takeda

## Rights

This document is the Accepted Manuscript version of a Published Work that appeared in final form in Inorganic Chemistry, copyright © 2025 American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see https://doi.org/10.1021/acs.inorgchem.5c03060.[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Mutual phase stability between kaolinite-derived hexacelsian and zirconia in a mixed system](https://mdr.nims.go.jp/datasets/8bd10283-c2bb-47a1-99cb-a13d4773c955)

## Fulltext

Mutual phase stability of a two-dimensional aluminosilicate and zirconia in a mixed systemShingo Machida1*, Masaya Suzuki1, Toshimichi Shibue2, Gaku Okuma3, and Daiki Takeda41Materials Research and Development Laboratory, Japan Fine Ceramics Center, 2-4-1, Mutsuno, Atsuta-ku, Nagoya, Aichi, 456-8587, Japan2Materials Characterization Central Laboratory, School of Science and Engineering, Waseda University, 3-4-1, Okubo, Shinjuku-ku, Tokyo, 169-8585, Japan3Research Center for Structural Materials, National Institute for Materials Science, 1‑2‑1 Sengen, Tsukuba, Ibaraki 305‑0047, Japan4Chemical and Food Department, Industrial Research Institute of Ishikawa, 2-1, Kuratsuki, Kanazawa, Ishikawa, 920-8203, Japan.*E-mail: shingo.machida@jfcc.or.jp Abstract:This study demonstrates that the phase stability of a two-dimensional aluminosilicate can be achieved through the presence of an unreacted coexisting component. Specifically, the mutual phase stability and chemical inertness of hexacelsian—a metastable phase of BaAl2Si2O8 with a layered structure—and yttria-stabilized zirconia (YSZ) were systematically investigated across a wide temperature range up to 1600°C, using solid-state reactions of kaolinite and BaCO3 with or without YSZ. Despite its metastable nature, hexacelsian remained intact without transforming into celsian—the stable BaAl2Si2O8 phase without a layered structure—under prolonged heating when co-present with ZrO2. Additionally, while YSZ alone shows both tetragonal and monoclinic ZrO2 phases upon calcination, only the tetragonal phase is retained when calcined in the presence of hexacelsian. This mutual stability is supported by complementary analyses—including X-ray diffraction, scanning electron microscopy, and thermal analysis—which revealed the formation of low-crystalline hexacelsian at 800°C, preceding the appearance of fully crystalline hexacelsian and the formation of calcined products in the kaolinite–YSZ and BaCO3–YSZ mixtures. These results offer insights into the design and functionality of thermally stable two-dimensional aluminosilicates in multiphase systems.IntroductionTwo-dimensional materials, such as layered inorganic solids composed of stacked nanolayers, exhibit unique functionalities driven by their surface, structural, morphological properties, including their affinity for and stability under ambient conditions.1-7 These properties are governed by factors such as composition, structure, size, and the number of layers—down to single-layer nanosheets.1-7 Such characteristics highlight the importance of two-dimensional materials, from critical components to functional additives, in both fundamental research and practical applications.1-7 Regarding functional additives6,7, optimizing the properties of two-dimensional materials often requires that they remain stable under ambient conditions8,9, emphasizing the need to suppress reactions with surrounding compounds. Additionally, if the two-dimensional structures are metastable phases, their transformation into stable phases should be delayed or suppressed, highlighting the importance of ensuring phase stability in metastable layered structures. Under conditions favorable for transforming metastable phases into stable ones8.9, two-dimensional materials are typically exposed to elevated temperatures. However, most two-dimensional materials—such as layered aluminosilicate clay minerals, which are cost-effective and suitable for practical applications—tend to lose their layered structures through thermal decomposition.10 By contrast, hexacelsian (BaAl2Si2O8), a member of the RAl2SI2O8 series (where R represents alkaline earth elements)11, consists of alternately stacked tetrahedral aluminosilicate layers and alkaline earth ions (Scheme 1). It typically forms at relatively high temperatures (above 900°C) through solid-state reactions involving kaolinite (Al2Si2O5(OH)4)12-18, a layered aluminosilicate clay mineral. Notably, each kaolinite layer comprises an AlO2(OH)4 sheet and a SiO4 sheet19,20, making it a homogeneous nanoscale mixer of Si and Al. As calcination temperature increases, hexacelsian—the metastable phase of BaAl2Si2O8 with hexagonal symmetry—generally tends to transform into celsian, the stable monoclinic phase that lacks a layered structure (Scheme 1), at around 1000 °C.12-16 However, a recent study showed that hexacelsian’s layered structure becomes dominant at temperatures above 1500°C when using a purified kaolinite system.18 Although the phase transformation typically occurs above 1500°C over 48 h, increasing kaolinite purity and reducing the particle size of other raw materials enables the preservation of hexacelsian’s structure even at 1600°C for 72 h.18 These findings suggest that hexacelsian has potential as a functional additive at relatively higher temperature region. To demonstrate this, its phase stability warrants further investigation, and its chemical inertness with other compounds must be evaluated. Zirconia (ZrO2) is the best candidate due to its single-component nature and thermal stability at elevated temperatures. Additionally, since Zr is not a constituent of hexacelsian, a mixed system of hexacelsian and ZrO2 provides a promising model for investigating the phase stability and chemical inertness of hexacelsian at elevated temperatures in the presence of another component. In this study, the thermal behavior of hexacelsian in the presence of ZrO2 was examined. As a result, mutual phase stability between hexacelsian and tetragonal ZrO2 was maintained after calcination at 1600°C. Notably, tetragonal ZrO2, which forms around 1200 °C during heating, typically reverts to monoclinic ZrO2 during cooling in the 1000–800 °C range.21 Additionally, no solid-state reactions occurred between the two phases. Based on these findings, possible mechanisms are discussed. In this context, insights from the reactions between ZrO2 and the hexacelsian precursors—kaolinite and barium carbonate (BaCO3)—are particularly informative. Furthermore, the reaction pathway that yields low-crystalline hexacelsian at 800 °C, which precedes the formation of fully crystalline hexacelsian at higher temperatures14, also provides valuable clues. For ZrO2, a commercially available, easily sintered grade was selected to facilitate both mechanistic interpretation and future practical applications.8,9 It is also worth noting that the absence of solid-state reactions, as mentioned above, indicates that kaolinite can be strongly milled using a zirconia pot and balls without triggering unintended reactions. This implies that ZrO2 contamination from the milling apparatus does not affect the solid-state reaction between kaolinite and BaCO3. Therefore, in this study, a product obtained by strongly grinding kaolinite with BaCO3 was briefly compared with one obtained using conventional grinding. ExperimentalGeneral informationThe crystalline phases of the raw materials and the calcined specimens were characterized by X-ray diffraction (XRD) using a diffractometer (Empyrean, PANalytical) operated at 40 mA and 45 kV with monochromated Cu Kα radiation. The step size and scan time were set to 0.01° (2θ) and 1.0 s, respectively. For clarity and ease of comparison, identical XRD profiles are presented in multiple figures throughout this study. Specimen and raw material morphologies were examined by field-emission scanning electron microscopy (FE-SEM; SU8000, HITACHI) at acceleration voltages of 2.0 or 15 kV, with electron beam energies of 10 or 30 µV and a working distance of 8 cm for secondary and backscattered electron imaging (SE or BE), respectively. Prior to observation, all samples were sputter-coated with a 3 nm osmium layer. For simplicity, FE-SEM images obtained using SE and BE modes are hereafter referred to as SEM (SE) or SEM (BE) images, respectively. The thermal behavior of powder specimens passed through a 150 µm sieve (to ensure visual uniformity) was examined by thermogravimetric (TG) and differential thermal analysis (DTA), recorded using a Rigaku TG 8120 at a heating rate of 10 °C/min under nitrogen. Solid-state 27Al and 29Si nuclear magnetic resonance (NMR) spectra were recorded on a Bruker AVANCE400 NEO spectrometer at 79.49 and 104.26 MHz, respectively, using the magic angle spinning (MAS) technique with pulse delays of 60 s and 1 s for 29Si and 27Al, respectively. The samples were packed in 4 mm zirconia rotors and spun at 8 kHz. 29Si and 27Al chemical shifts were externally referenced to polydimethylsiloxane (−33.8 ppm) and an aqueous aluminum nitrate solution (0 ppm), respectively. Elemental analysis of kaolinite was conducted by inductively coupled plasma optical emission spectrometry (ICP-OES; SPS-3100, Seiko Instruments Inc.) by the Tokai Technology Center. Prior to analysis, the kaolinite specimens were thermally decomposed in mixed acids including hydrofluoric acid and then dissolved in hydrochloric acid.MaterialsKaolinite used in this study—Kanpaku kaolinite—was a reference clay sample provided by the Clay Science Society of Japan (JSCC-1101c). Notably, this kaolinite is originally white, and its XRD pattern shows weak reflections attributed to quartz, indicating minor by-product contamination.22 The Si/Al molar ratio was 1.0318, and the Fe content was 160 ppm. Barium carbonate (BaCO3) and ethanol were obtained from Wako Pure Chemical. Yttria-stabilized zirconia with ease sintering properties, containing 3 mol% yttria (considered a trace amount), was obtained from TOSHO Co. (TZ-3Y-E). For simplicity, this yttria-stabilized zirconia is hereafter referred to as YSZ in this study. Sample preparation Notably, solid-state reactions of kaolinite with carbonates typically yield single crystalline phases of RAl2Si2O812-18; however, off-stoichiometric compounds form if the raw materials—particularly during the grinding process—are not sufficiently mixed.22 Moreover, achieving uniform mixing and promoting accelerated solid-state reactions require disruption of kaolinite’s stacking order.23 A previous study reported that the stacking order of pristine kaolinite is only slightly disrupted under weak grinding conditions, in contrast to kaolinite with expanded layers.23 To minimize this variability and unify the degree of stacking disorder under various experimental conditions as much as possible, this study employed a two-step grinding process: the first step involved relatively strong grinding, while the second, following previous studies16-18,23, applied milder grinding. It is worth noting that kaolinite with expanded layers—i.e., purified kaolinite used in earlier studies16-18,23—accelerated solid-state reactions and stabilized metastable phases at relatively high temperatures, but this required multiple experimental steps. In contrast, pristine kaolinite was employed in this study to better reflect practical applications. For clarity, the XRD patterns and SEM images of the raw materials are presented ahead of the results and discussion on calcined specimens. For the first grinding step, kaolinite (20.64 g) and BaCO3 (8.0 g) were mixed in water (150 mL) in a planetary ball mill at 500 rpm for 30 min (repeated 4 times) using a resin vessel (300 mL) containing 1,000 alumina balls (5 mm in diameter). The molar ratio of 1:1 corresponds to the stoichiometric composition of BaAl2Si2O8. After milling, the slurry was sieved to remove the alumina balls and then dried at 100°C for a day. The dried solid was passed through a 150 µm sieve to ensure visible uniformity.For the second grinding step, based on previous studies16-18,23, the dried powder (916 mg) and YSZ (246 mg) were wet-milled in ethanol (8 mL) at 250 rpm for 12 h using a planetary ball mill in a resin vessel (12.5 mL) containing 120 alumina balls (3 mm in diameter). The resulting mixture was centrifuged at 5000 rpm for 5 min, dried at 100 °C for 1 h, and passed through a 150 µm mesh sieve to ensure uniformity. These powders were mixed to achieve a tentative 1:1 molar ratio of hexacelsian (BaAl2Si2O8) to ZrO2. For comparison, another batch was prepared with a lower hexacelsian content, adjusting the molar ratio to 1:8. Additionally, the second grinding procedure was applied to mixtures of kaolinite and YSZ (1:1 molar ratio), BaCO3 and YSZ (1:1 molar ratio), as well as to BaCO3 alone.All resultant mixtures were calcined at 600, 800, 1000, 1300, or 1600°C for 12 h at a heating rate of 10°C/min, followed by a furnace cooling. These calcination temperatures were selected based on specific thermal events associated with the solid-state reaction of BaAl2Si2O8 using kaolinite and BaCO3 used as raw materials, as reported previously.18 For comparison, individual raw materials were also calcined at 600, 800, 900, 950, 1000, 1300 or 1600°C for 12 h. For reference, the XRD patterns of selected calcined raw materials—including the calcined ground specimens—are presented prior to those of the mixed products containing hexacelsian and YSZ.To investigate the effects of strong grinding of kaolinite as mentioned in the Introduction, kaolinite was dry-milled in a planetary ball mill at 400 rpm for 60 min per cycle, repeated 36, 72, or 192 times. The milling was performed in a ZrO2 vessel (250 mL) containing 50 ZrO2 balls (10 mm in diameter). After 192 cycles, the resulting solid was mixed with BaCO3 using the second grinding procedure. Both the final mixture and the kaolinite–BaCO3 mixture obtained through the first and second grinding steps were calcined at 1600 °C for 48 h. This calcination condition was adopted from a previous study for comparison.18 The specimens are hereafter referred to as BAS-milled-1600 and BAS-1600, respectively.It should be noted that abbreviations for most of the ground raw materials and calcined specimens are introduced in the Results and Discussion section to avoid unnecessary complexity. Since there are multiple ground raw materials and their calcined forms at various temperatures, using abbreviations before identifying their crystalline phases would be premature.Results and Discussion  As mentioned above (see experimental), the degree of disruption in the stacking order of kaolinite in the ground raw materials governs the extent to which the solid-state reaction proceeds.23 This extent is evaluated based on the intensity ratio of the (001) reflection, associated with stacking order, to the (020) reflection, which corresponds to the lateral atomic arrangement.23 In the XRD patterns of the ground raw materials (Figure S1), no significant differences are observed in the intensity of the (001) reflection relative to the (020) reflection between the kaolinite–BaCO3 mixtures and those also containing YSZ (Figure S1d,e). These results indicate that the efficiency of the solid-state reactions in this study is not significantly affected by the extent of stacking disorder in kaolinite. For reference, the XRD patterns of the raw materials are shown in Figure S1a–c.  As shown in Figure S2, the sizes of the hexagonal platy kaolinite particles and the rod- to round-like BaCO3 particles are reduced after grinding, regardless of the presence of YSZ (Figure S2a, b, e, f). The YSZ particles also lose their original spherical and aggregated morphologies after grinding (Figure S2c, d, f). The mixing state of the ground materials is complex; BaCO3 and YSZ particles cannot be clearly distinguished in the SEM (BE) images (Figure S2e’, f’), although their overall appearance remains visually similar. Previous studies have shown that the milder second-step grinding (see experimental) does not significantly affect particle size.16-18,23 Therefore, the particle sizes of kaolinite and BaCO3, as well as their mixing states with or without YSZ, are unlikely to be influenced by the second grinding step.   Figure 1 shows XRD patterns of calcined specimens prepared from ground kaolinite and BaCO3, as well as those of calcined YSZ. Notably, crystalline hexacelsian forms via a low-crystalline phase, which exhibits a halo profile in XRD patterns, as reported previously.14 Other characteristics of this low-crystalline hexacelsian are discussed later. In this study, the XRD patterns of the ground kaolinite-BaCO3 specimens calcined at 800 and 900°C exhibit mostly broad halo profiles, while those calcined at 950 and 1000°C show distinct reflections of crystalline hexacelsian (Figure1Aa-d). As the calcination temperature increases from 1000 to 1300°C, reflections of celsian appear, indicating the well-known phase transformation from hexacelsian to celsian (Figure 1Ae). At 1600°C, reflections of celsian disappear, and hexacelsian―metastable BaAl2Si2O8―remains the dominant phase (Figure 1Af). This pseudo-reversal phase transformation is consistent with a previous report indicating that the delayed phase transformation from hexacelsian to celsian is governed by the particle growth of hexacelsian in purified kaolinite systems, leading to the retention of hexacelsian at relatively higher temperatures and shorter calcination times.18 Based on these results, the specimens obtained by calcining the ground raw materials containing kaolinite and BaCO₃ at 1000, 1300, or 1600°C—where hexacelsian forms clearly, with or without celsian—are hereafter referred to as hexacelsian at 1000°C, hexacelsian at 1300°C (containing celsian), and hexacelsian at 1600°C, respectively, for simplicity. Although hexacelsian also forms at 950 °C, hexacelsian at 1000 °C is subjected in this study for comparison with previous reports.16-18 Meanwhile, the XRD pattern of the YSZ used in this study primarily shows reflections of tetragonal ZrO2, with minor reflections attributed to monoclinic ZrO2 (Figure1Ba). These weaker reflections disappear in the XRD patterns of YSZ calcined at 1000 and 1300 °C, but become relatively strong and sharp again after calcination at 1600 °C (Figure 1Ba-d). This is likely because, under furnace cooling, the specimens calcined at 1000 and 1300°C were not held at high temperatures long enough during cooling to allow sufficient transformation from the tetragonal to the monoclinic phase, unlike the specimen calcined at 1600°C. For simplicity, these calcined YSZ samples are hereafter referred to as YSZ at 1000 °C, YSZ at 1300 °C, or YSZ at 1600 °C, respectively.  Figure 2 shows the XRD patterns of ground materials containing kaolinite, BaCO3, and YSZ, along with patterns of those without YSZ or of YSZ alone, calcined at 1600 or 1300°C for ease of comparison. As shown, reflections attributed to calcian and monoclinic ZrO2 are absent in the XRD patterns of the ground materials containing all three components calcined at 1600 or 1300°C (Figures 2Ac and Bc). Compared to the XRD patterns of hexacelsian at 1600°C and 1300°C, the reflection near 30° (2θ) is relatively strong in the co-calcined specimens, which is attributed to the overlap of the strong reflection of tetragonal ZrO2 with hexacelsian’s profile. These results indicate that the phase transformations of both hexacelsian and ZrO2 are suppressed in their co-presence.  Figure 3 shows SEM images of ground materials containing kaolinite, BaCO3, and YSZ calcined at 1600 °C, alongside those of hexacelsian at 1600 °C and YSZ at 1600 °C. Notably, since the ground materials were passed through a 150 µm sieve prior to calcination (see experimental), the resulting calcined specimens can be consisted of agglomerates smaller than 150 µm (Figure 3a–c). While the surface morphologies appear largely similar in SEM (SE) images regardless of YSZ inclusion (Figure 3b, d), the SEM (BE) images reveal the presence of two distinct phases, as evidenced by particles with different brightness levels (Figure 3b’, d’). The brighter particles are attributed to YSZ, based on the atomic numbers of Ba and Zr, the contrast observed in the SEM (BE) image of pure YSZ calcined at 1600 °C (Figure 3f’), and the enhanced reflection intensity near 30° (2θ) in the XRD pattern, which corresponds to tetragonal ZrO2, overlapping with hexacelsian (Figure 2A). Notably, compared with the SEM (SE) images of YSZ before calcination (Figure S2c, d), the images of YSZ at 1600 °C show larger grains (Figure 3f’), indicating grain growth during calcination. In contrast, the grains attributed to YSZ in the SEM (BE) image of the specimen containing all three raw materials calcined at 1600 °C are relatively smaller (Figure 3d’). Furthermore, the fine grains observed in the SEM (SE) image of hexacelsian at 1600°C seems to be less prominent in the corresponding image of the specimen containing the three raw materials (Figure 3b’, d’).  Based on the above results, the mutual phase stability of hexacelsian and tetragonal ZrO2 is evident from their co-existence in the 1300–1600°C range. Although the specimens examined in this study are not bulk materials, the calcined agglomerates—derived from powders that passed through a 150 µm sieve—can be regarded as relatively small sintered bodies. As such, the grain growth behavior of hexacelsian and YSZ in the mixed system differs from that observed when each is present alone. Therefore, the plausible mechanisms underlying their phase stability, including microstructural considerations, are discussed below. In advance of this discussion, the chemical inertness between hexacelsian and YSZ is further examined. For simplicity, the calcined specimens containing hexacelsian and YSZ are hereafter referred to as hexacelsian–YSZ at X°C, where X represents the calcination temperature.  Figure 4 shows the X-ray diffraction (XRD) patterns of ground raw materials composed of kaolinite and ZrO2, as well as BaCO3 and ZrO2, calcined at 1600 °C. For comparison, the pattern of the hexacelsian–YSZ specimen calcined at 1600°C is also shown (Figure 4a). The XRD pattern of the kaolinite–ZrO2 mixture reveals reflections corresponding to mullite and zircon (ZrSiO4) (Figure 4b). Notably, kaolinite typically transforms into mullite and cristobalite above 1200°C.24-27 Cristobalite is formed via the crystallization of amorphous silica, which separates during the conversion of metakaolinite (Al2O3·2SiO2)—a dehydrated, amorphous phase generated between 400 and 600°C—into mullite (2Al2O3·SiO2 to 3Al2O3·2SiO2) above 1000°C.24-28 Because cristobalite reflections are absent in the XRD pattern of the kaolinite–ZrO2 specimen calcined at 1600°C (Figure 4b), this suggests that the amorphous silica formed during mullite crystallization reacted with ZrO2 to form zircon ZrSiO4. Meanwhile, the BaCO3–ZrO2 specimen calcined at 1600°C indicates the formation of barium zirconate (BaZrO3), consistent with a well-known solid-state reaction.29 Unfortunately, the strongest reflections in the XRD patterns of tetragonal ZrO2 (PDF No. 01-070-8758) and BaZrO3 (PDF No. 00-006-0399) appear at nearly identical 2θ positions, making phase identification difficult. Additionally, the second strongest reflection of BaZrO3 overlaps with a reflection appearing just below 50° in the XRD pattern of hexacelsian (PDF No. 00-028-0124), further complicating the analysis. Although the third strongest reflection of BaZrO3 is absent in the XRD pattern of the hexacelsian–YSZ specimen calcined at 1600°C (Figure 4a), and the morphologies observed in the SEM (BE) images of the kaolinite–YSZ and BaCO3–YSZ mixtures calcined at 1600 °C (Figure S3) are not present in the corresponding image of the hexacelsian–YSZ specimen (Figure 3d’), these findings are discussed in more detail below.  Figure 5 shows the XRD patterns of various ground raw materials calcined at 800 and 1000 °C. Notably, metakaolinite—formed by calcining kaolinite in the 400–1000 °C range—exhibits broad halo profiles.28 Therefore, the XRD patterns of the kaolinite–YSZ mixtures calcined at both 800 and 1000 °C primarily reflect the presence of ZrO2. The BaCO3–YSZ mixture calcined at 800 °C shows only reflections corresponding to tetragonal ZrO2 and unreacted BaCO3 (Figures 2B and S1b), indicating that the solid-state reaction forming BaZrO3 does not proceed at this temperature (Figure 5Ae). In contrast, reflections attributed to BaZrO3 are clearly observed in the XRD pattern of the BaCO3–YSZ mixture calcined at 1000 °C, indicting that the solid-state reaction occurs between 800 and 1000°C. Meanwhile, the XRD pattern of the mixture containing kaolinite, BaCO3, and YSZ calcined at 1000 °C (Figure 5Bc) closely resembles that obtained at 1600 °C (Figure 2Bc), indicating the formation of crystalline hexacelsian at 1000°C. This observation is consistent with the XRD pattern of hexacelsian at 1000 °C (Figure 1Ad). Notably, the third strongest reflection of BaZrO3 is also absent in this pattern, as previously seen in the hexacelsian–YSZ specimen calcined at 1600°C (Figure 4a). However, the XRD patterns do not clearly distinguish whether hexacelsian or BaZrO3 preferentially forms. In this context, the halo profile of the XRD pattern for the kaolinite–BaCO3 mixture calcined at 800 °C, along with that of the mixture containing all three components at 800 °C (Figure 5Ba, c), provides useful insight in addition to the TG-DTA curves, as discussed below. Notably, the XRD pattern of the kaolinite–BaCO3 mixture calcined at 600°C is dominated by reflections from BaCO3 (Figures S4 and S1b), indicating that hexacelsian does not form at this temperature.  The XRD results described above are supported by TG-DTA curves. The TG-DTA profiles of the kaolinite–BaCO3 mixtures, with and without YSZ (Figure 6a, b), are essentially identical. In both, mass losses and endothermic peaks in the 400–600°C range correspond to kaolinite dehydroxylation, while those in the 600–800°C range are attributed to BaCO3 decarbonation, as previously reported.18 Additionally, an exothermic peak with a plateau near 1000 °C is associated with the formation of crystalline hexacelsian.18 Similarly, the TG-DTA curve of the kaolinite–YSZ mixture shows dehydroxylation in the same temperature range, and the exothermic plateau near 1000 °C indicates the onset of mullitization (Figure 6c), consistent with earlier studies.27 Notably, a mass loss accompanied by an endothermic peak is observed in the TG-DTA curve of the BaCO3 mixture (Figure 6d), similar to that of pure BaCO3 (Figure S5a). The earlier two endothermic peaks with plateaus correspond to phase transformations of BaCO3 crystal phases, while the later ones with mass loss indicate BaCO3 decarbonation.30,31 A similar profile is also observed in the TG-DTA curve of ground BaCO3 (Figure S5b), which exhibits reduced particle size and slightly broadened XRD peaks (Figures S6 and S7). Given the XRD patterns (Figure 5), these results indicate that hexacelsian forms prior to BaZrO3.  As mentioned above, crystalline hexacelsian forms via a low-crystalline phase, which exhibits a halo in the XRD pattern.14 This is supported by previously reported solid-state ²⁹Si and ²⁷Al NMR results of the RAl2Si2O8 series. Both crystalline and amorphous phases show similar NMR profiles.14,22 For example, in the reaction between kaolinite and CaCO3, the ²⁷Al MAS NMR spectrum of metastable CaAl2Si2O8 displays a 58 ppm signal, attributed to tetrahedrally coordinated Al in its layered calcium aluminosilicate framework (composed of alternately stacked tetrahedral aluminosilicate layers and interlayer Ca ions), along with a 13 ppm signal.22 Although the origin of the 13 ppm signal remains uncertain14,22, it is likely associated with six-coordinated Al, possibly corresponding to unreacted portions in the structure, as suggested by NMR studies of zeolitic materials.32 The ²⁹Si MAS NMR spectrum of metastable CaAl2Si2O8 also shows a −86 ppm signal for tetrahedral Si, along with broader upfield features attributed to unreacted Si species.22 Hexacelsian (metastable BaAl2Si2O8) exhibits similar NMR features.14 In this study, solid-state MAS NMR measurements were conducted because the XRD patterns of kaolinite–BaCO3 mixtures with or without YSZ calcined at 800°C exhibited either a halo profile or reflections from tetragonal ZrO2. The ²⁷Al MAS NMR spectra of both 800 °C specimens (with and without YSZ) are identical, showing signals at 58 and 13 ppm (Figure 7b,c). Their ²⁹Si MAS NMR spectra also show identical −86 ppm signals. These NMR profiles are consistent with those reported previously14,22, indicating the presence of low-crystalline hexacelsian. Notably, the ²⁷Al and ²⁹Si MAS NMR spectra do not exhibit signals characteristic of pristine kaolinite25,26,28 (Figure 7a)—namely, the 0 ppm signal from octahedrally coordinated Al in the ²⁷Al MAS NMR and the Q³ doublet at −90.3 and −90.7 ppm in the ²⁹Si MAS NMR (which appears as a single peak due to the figure’s scale). Therefore, a hexacelsian-like structure is formed at 800 °C regardless of the presence of YSZ. Given the XRD patterns and TG-DTA curves (Figures 5 and 6), these results indicate that hexacelsian forms preferentially over BaZrO3. The rapid formation of hexacelsian is further supported by the XRD patterns of the kaolinite–BaCO3–YSZ mixture with a molar ratio of 1:8 (hexacelsian : ZrO2) (Figure S8), which closely resemble those obtained with a 1:1 ratio (Figures 2Ac and 5Bc). This indicates hexacelsian forms rapidly even when the contact ratio between kaolinite and BaCO₃ is reduced in the presence of YSZ.  Based on the above results, it is evident that hexacelsian does not react with YSZ, even at the relatively high temperature of 1600 °C. Additionally, the relatively rapid solid-state reaction of kaolinite with BaCO3 to form hexacelsian prevents these raw materials from reacting with ZrO2. Notably, BaCO3 decarbonation occurs above 800°C in both the presence of ZrO2 and upon particle size reduction (Figures 7d and S5). However, in the presence of dehydrated kaolinite (metakaolinite), decarbonation occurs below 800°C (Figure 7a,b). Given that low-crystalline hexacelsian forms at 800°C (Figures 1a, 5a, 5c, 8b, and 8c), this indicates that metakaolinite accelerates BaCO3 decarbonation. This suggests that the reaction between metakaolinite and Ba proceeds first and promotes decarbonation, which warrants further investigation.26 These findings indicate that hexacelsian is a promising additive for ZrO2-based materials. Therefore, their sintering behavior is briefly discussed in terms of grain size within tens- to hundreds-of-micrometer-scale agglomerates, regarded as relatively small sintered bodies. Based on the grain growth behavior of YSZ from room temperature to 1600 °C (Figures S1c, S1d, 2e, and 2f’), and the relatively small bright regions observed in the hexacelsian-YSZ at 1600 °C (Figure 2d’), it can be inferred that the presence of hexacelsian suppresses YSZ grain growth. Meanwhile, the fine grains of hexacelsian observed at 1600 °C appear to be reduced in the presence of YSZ (Figures 2b’ and 2d’). The relatively larger grains of hexacelsian at 1600 °C also appear to decrease in size under these conditions (Figures 2b and 2d), indicating that hexacelsian grain growth is not uniform. Furthermore, the presence of YSZ inhibits the development of large hexacelsian grains. Grains that would otherwise grow into larger ones in the absence of YSZ instead grow together with relatively fine grains, leading to the suppression of distinct fine grains in the presence of YSZ. However, the detailed mechanisms behind these grain growth behaviors and their uniformity likely depend on the initial packing structure of the raw materials. To clarify and control these processes, further studies incorporating both raw materials with various shapes18 and various calcined specimens33,34 are needed. Nevertheless, the mutual phase stability of hexacelsian and tetragonal ZrO2 is evident in this study. In general, phase transformations from metastable to stable phases occur at the surfaces of metastable phases or at structural defects.35,36 According to the SEM (BE) image (Figure 2d’), YSZ grains appear to be located between the hexagonal-shaped grains of hexacelsian, indicating that viscous flow or surface diffusion34,35,37 between the two compounds is delayed in their co-presence. This implies that viscos flow or surface diffusion are hindered when unreacted materials are in direct contact. Regarding hexacelsian, larger grain sizes have been reported to exhibit sluggish phase transformation, as discussed in previous studies.18,38 However, verifying these hypotheses will require future studies39,40 on morphological changes in the specimens. In this context, the reaction inertness of hexacelsian with ZrO2 offers insights for future investigations.  When kaolinite is strongly dry-milled using ZrO2 balls and a ZrO2 pot (see experimental), the kaolinite reflections disappear from the XRD patterns of the ground specimens, leaving only reflections associated with tetragonal ZrO2 (Figure 8). The color change of kaolinite from white to brown (Figure 8, left inset) indicates contamination by ZrO2 from the pot and balls, a well-known phenomenon (Figure 8, right inset). After grinding, the hexagonal platy morphology of kaolinite (Figure S1a) disappears, as observed in SEM (SE) images of the ground specimens, where smaller or aggregated platy particles are present (Figure 9). In the SEM (BE) image, the absence of contrast differences among particles (Figure 9c’) indicates uniform dispersion of ZrO2 contamination. Notably, previous studies have reported that prolonged grinding of kaolinite converts its morphology into aggregated spherical particles observed via transmission electron microscopy (TEM).41,42 In this study, the ground kaolinite obtained here is beneficial for delaying phase transformations, as evidenced by the XRD pattern and SEM image of BAS-milled-1600 (Figures 10b and 11a’). In this XRD pattern, only hexacelsian reflections are observed. By contrast, the XRD pattern of BAS-1600 shows both hexacelsian and weak celsian reflections. In a previous study, no celsian reflections were observed in specimens prepared with purified kaolinite.18 In this study, kaolinite was used without further purification. Although the Si/Al ratio of 1.03 is identical to that of the purified kaolinite used previously18, the presence of 160 ppm Fe likely influences the phase transformation. Clarifying this effect using spectroscopic techniques43 would be a topic for future investigation. SEM images (Figure 11) reveal that the grain size of BAS-milled-1600 is significantly larger than that of BAS-1600. As discussed in previous studies18,38, larger grain sizes tend to suppress phase transformation, explaining the absence of celsian in BAS-milled-1600. Furthermore, earlier research indicated that reducing the particle size of raw materials promotes the growth of large hexacelsian grains.18 Therefore, the sintering behavior of kaolinite with different particle shapes and crystallinities—as well as those of BaCO3 and low-crystalline hexacelsian—warrants further investigation.34,35 These strategies also shed light on aluminosilicate design approaches42-50, positioning hexacelsian as a functional additive for bulk materials.8,9 Such studies will be reported in due course. ConclusionsThis study demonstrates that hexacelsian―a metastable BaAl2Si2O8 phase prepared by the solid-state reaction of kaolinite and BaCO3―maintains its layered structure and does not undergo a solid-state reaction with YSZ, even after prolonged calcination at 1600°C. Additionally, the tetragonal ZrO2 phase is preserved after calcination at 1600°C in the presence of hexacelsian, whereas calcination of YSZ alone results in a mixture of tetragonal and monoclinic phases. Similar trends are observed at 1300 °C, a temperature at which hexacelsian typically transforms into celsian, the stable phase of BaAl2Si2O8. Given the preserved grain boundaries between hexacelsian and tetragonal ZrO2, their mutual stability is evident in the mixed system. In this context, the rapid formation of low-crystalline hexacelsian at 800 °C—which precedes the formation of mullite, ZrSiO4, and BaZrO3 observed in the kaolinite–YSZ and BaCO3–YSZ mixtures—indicates a favorable reaction pathway initiated by early-stage reactions between metakaolinite and BaCO3. These results indicate that the distinctive phase stability and chemical inertness of hexacelsian and YSZ provide compelling benefits for multiphase oxide systems, especially in cases where layered inorganic solids acted as main components or functional additives. Future studies may further explore their potential in tailoring the microstructure and enhancing the performance of aluminosilicate-based materials.Acknowledgement Special thanks to Mr. Toshiaki Ogasawara and Ms. Mikiko Kuroda, staffs in Tokai Technol. Center, for ICP analyses and their sample preparations. This research was supported by Japan Fine Ceramics Center (JFCC) Grant for Advanced Technology Development Research (T25A4560). The solid-state MAS NMR experiments was conducted in the Materials Characterization Central Laboratory of Waseda University, which was shared in the MEXT Project for promoting public utilization of advanced research infrastructure (program for supporting the construction of core facilities, Grant No. JPMXS0440500021).  Author contributionShingo Machida: conceptualization, data curation, investigation, writing—original draft, project administration, supervision. Masaya Suzuki: investigation. Toshimichi Shibue: writing—review and editing. Gaku Okuma: writing—review and editing. Daiki Takeda—conceptualization, writing—review and editing.Conflicts of Interest The authors declare no conflict of interest.References1. Saito, K.; Morita, M.; Okada, T.; Wijitwongwan, R.P.; Ogawa, M. Designed Functions of Oxide/Hydroxide Nanosheet via Elemental Replacement/Doping, Chem. Soc. Rev., 2024, 53, 10523-10574.2. Ogawa, M.; Saito, K.; Sohmiya M. A Controlled Spatial Distribution of Functional Units in the Two　Dimensional Nanospace of Layered Silicates and Titanates, Dalton Trans., 2014, 42, 10340-10354.3. C. Detellier, Functional Kaolinite, Chem. Rec., 2018, 18, 868-877.4. Sugahara, Y. Chemical Process Employing Inorganic Layered Compounds for Inorganic and Inorganic-Organic Hybrid Materials, J. Ceram. Soc. Jpn., 2014, 122, 523-529.5. Miyamoto, N.; Nakato, T. Liquid Crystalline Inorganic Nanosheet Colloids Derived from Layered Materials, Isr. J. Chem., 2012, 52, 881-894.6. Lawn, B. R. Indentation of Ceramics with Spheres: A Century after Hertz, J. Am. Ceram. Soc., 1998, 81, 1977-1994.7. Höland, W.; Göts, W.; Garl, G.; Vogel, M. Microstructure of Mica Glass-Ceramics and Interface Reactions between Mica Glass-Ceramics and Bone, Cells Mater., 1992, 2, 105-122.8. Huang, J.; Cao, X.; Chen, W.; Guo, X.; Li, M.; Wang, W.; Dong, S.; Liu, L.; Chen, M. A Comprehensive Review of Thermally Sprayed Abradable Sealing Coatings: Focusing on Abradability, Chinese J. Aeronaut., 2024, 37, 1-25.9. Sun, X.; Liu, Z.; Du, L.; Lan, H.; Zhang, H.; Huang, C.; Ziang, M.; Yue, F.; Zhang, W. A Study on YSZ Abradable Seal Coatings Prepared by Atmospheric Plasma Spray and Mixed Solution Precursor Plasma Spray, J. Therm. Spray Tech., 2021, 30, 1199-1212.10. Heller-Kallai, L. Thermally Modified Clay Minerals, In Developments in Clay Science Volume5A Handbook of Clay Science, 2nd ed; Handbook of Clay Science 5A; Bergaya, F.; Lagaly, G., Eds.; Elsevier: Oxford UK, 2013; pp 411-432.11. Töpel-Schadt, J.; Müller, W. F.; Pentinghaus, H. Transmission Electron Microscopy of SrAl2Si2O8: Feldspar and hexacelsian polymorphs, J. Mater. Sci., 1978, 13, 1809-1816.12. Kobayashi, Y.; Sumi, K.; Asai, T.; Kato, E. Fabrication of Dense Celsian Ceramics by Reaction of Kaolin with SrCO3 and BaCO3, J. Ceram. Soc. Jpn., 1999, 197, 657-661.13. Kobayashi, Y. Transformation Kinetics From Hexacelsian to Celsian for Powders Having Uniform Particle Size, Ceram. Int., 2001, 27, 179-184.14. Okada, K.; Arai, H.; Kameshima, Y.; Yasumori, A.; MacKenzie, K. J. D. Uptake of various cations by amorphous BaAl2Si2O8 prepared by solid-state reaction of kaolinite with BaCO3, Mater. Lett., 2003, 57, 3554-3559.15. Kobayashi, Y.; Pee, J.-H.; Murabayashi, Y.; Katayama, M. Thermal Expansion of Sintered Polycrystals of Hexacelsian Solid Solution (Sr, Ba)Al2Si2O8, J. Ceram. Soc. Jpn., 2016, 126, 881-885.16. Machida, S.; Katsumata, K.; Yasumori. A. A Stable Layered Inorganic Solid at High Temperature: Heat Treatment of Eu-doped Hexacelsian without Phase Transformation. J. Solid-State Chem., 2003, 322, 123945.17. Machida, S.; Katsumata, K.; Yasumori. A. Promotion of Oxidation in the Powder Compact: Effect of Nitrate as Raw Material on the Solid-State Reaction Forming Hexacelsian Probed with Europium Ion. J. Solid-State Chem., 2023, 15, 124218.18. Machida, S. Pseudo-Reversal Phase Transformation of Hexacelsian to Celsian in Purified Kaolinite System. J. Solid-State Chem., 2025, 348, 125379.19. Brigatti, M. F.; Galán, E.; Theng, B. K. G. Structure and Mineralogy of Clay Minerals, In Developments in Clay Science, 2nd ed. Handbook of Clay Science 5A; Bergaya, F.; Lagaly, G., Eds.; Elsevier: Oxford UK, 2013; pp. 21–29.20. Weiss, A. A secret of chinese porcelain manufacture. Angew. Chem. Int. Ed., 1963, 12, 697–703.21. Chevalier, J.; Gremillard, L.; Virkar, A. V.; Clarke, D. R. The Tetragonal-Monoclinic Transformation in Zriconia: Lessons Learned and Future Trends. J. Am. Ceram. Soc., 2009, 92, 1901-1920.22. Okada, K.; Watanabe, N.; Jha, K. V.; Kameshima, Y.; Yasumori, A.; MacKenzie, K. J. D. Effects of Grinding and Firing Conditions on CaAl2Si2O8 Phase Formation by Solid-State Reaction of Kaolinite with CaCO3. Appl. Clay Sci., 2003, 23, 329–336.23. Machida, S.; Katsumata, K.; Yasumori, A. Effects of Kaolinite Layer expansion and Impurities on the Solid-State Reaction of Kaolinite. 2021, RSC Adv., 11, 38473–38477.24. Rocha, J.; Klinowski, J. 29Si and 27Al Magic-Angle-Spinning NMR Studies of the Thermal Transformation of Kaolinite. Phys. Chem. Mater. 1990, 17, 179-186.25. Rocha, J.; Klinowski, J.; Adams, J. M. Solid-State NMR Elucidation of the Role of Mineralizers in the Thermal Stability and Phase Transformation of Kaolinite. J. Mater. Chem., 1991, 26, 3009-3018.26. Bergaya, F.; Dion, P.; Alcover, J.-F.; Clinard, C.; Tchoubar, D.; TEM Study of Kaolinite Thermal Decomposition by Controlled-Rate Thermal Analysis. J. Mater. Chem., 1996, 31, 5609-5075.27. Chakraborty, A. K. Mullite Phase, In Phase Transformation of Kaolinite Clay; Chakraborty, A. K., Eds.; Springer India: New Delhi, India, 2014; pp. 235–272.28. Machida, S.; Katsumata, K.; Yasumori, A. Expansion of Orderly Stacked Metakaolinite Layers and Order Destruction Using a Kaolinite-Tetraphenylphosphonium Chloride Intercalation Compound. RSC Adv., 2021, 11, 23090-23094.29. Dunyushkina, L. A. Densification of Proton-Conducting ABO3 (A =Ca, Sr, Ba; B = Zr, Hf) Based Ceramics: Review of Sintering Technologies and Their Impact of Charge Transport Properties. J. Euro. Ceram. Soc., 2025, 45, 116908.30. Ouyang, D.; Zhang, Q.-T.; Chen, X.; Guo, Z.; Yang, C. Preparation and the Thermal Properties of Porous Barium Nitrate. Sci. Technol. Energ. Mater., 2021, 82, 89-94.31. Bardwell, C. J.; Bickley, R. I.; Poulston, S.; Twigg, M. V. Thermal Decomposition of Bulk and Supported Barium Nitrate, Thermochim. Acta, 2015, 613, 94-99.32. Ravi, M.; Sushkevich, V. L.; van Bokhoven, A. Towards a Better Understanding of Lewis Acidic Aluminum in Zeolites. Nature Mater., 2020, 19, 1047-1056.33. Okuma, G. 3D Microstructural Evolution in Sintering Observed by X-ray Microtomography. J. Ceram. Soc. Jpn., 2023, 131, 522-532.34. Okuma, G.; Wakai, F. Synchrotron X-ray Multiscale Tomography: Visualization of Heterogeneous Microstructures and Defects in Ceramics. J. Am. Cream. Soc., 2024, 107, 1706-1724.35. El-Meliegy, E. A. M.; Abdel-Hameed, S. A. M. Enstatite-Celsian Glass Ceramics, Mater. Charact., 2005, 55, 28-34.36. Lee, K. T.; Aswath, P. B. Enhanced Production of Cesian Brium Aluminosilicates by a Three-Step Firing Technique, Mater. Chem. Phys., 2001, 71, 47-52.37. Wakai, F.; Katsuta, K.; Kanchika, S.; Shinoda, Y.; Akatsu T.; Shinagawa, K. Sintering Force behind the Viscos Sintering of Two Particles. Acta Materialia, 2016, 109, 292-299.38. Bahat, D. Kinetic Study on the Hexacelsian-Celsian Phase Transformation, J. Mater. Sci., 1970, 5, 805-810.39. Yao, Y.; Wakimoto, D.; Miyamoto. H.; Sasaki. K.; Kuramata, A.; Hirano, K.; Sugawara, Y.; Ishikawa, Y. X-ray Topographic Observation of Dislocations in β-Ga2O3 Schottky Barrier Diodes and Their Glide and Multiplication under Reverse Bias. Scripta Materialia, 2023, 226, 115216.40. Yao, Y.; Hirano, K.; Kuramata, H.Sugawara, Y.; Ishikawa, Y. Lattice Misorientation at Domain Boundaries in β-Ga2O3 Single-Crystal Substrates Observed via Synchrotron Radiation X-ray Diffraction Imaging and X-ray Reticulograph. J. Am. Ceram. Soc., 2023, 106, 5500.41. Takahashi, H. Effects of Dry Grinding on Kaolin Minerals. Ⅰ. Kaolinite. Bull. Chem. Soc. Jpn., 1959, 32, 235-245.42. Takahashi, H. Effects of Dry Grinding on Kaolin Minerals. Ⅱ. Halloysite. Bull. Chem. Soc. Jpn., 1959, 32, 252-263.43. Kéri, A.; Dähn, R.; Krack, M.; Churakov. S. V. Characterization of Structural Ion in Smectites－An Ab Initio Based X-ray Adsorption Spectroscopy Study. Environ. Sci. Technol., 2019, 53, 6877-6886.44. Miyawaki, R.; Tomura, S.; Inukai, K.; Shibasaki, Y.; Okazaki, M.; Samejima, S.; Satokawa, S. Formation Process of Kaolinite from the Amorphous Mixture of Silica and Alumina. Clay Sci., 1992, 8, 272-284.45. Satokawa, S.; Osaki, Y.; Samejima, S.; Miyawaki, R.; Tomura, S.; Shibasaki, Y.; Sugahara, Y. Effects of the Structure of Silica-Alumina Gel on the Hydrothermal Synthesis of Kaolinite. Clays. Clay Miner., 1994, 42, 288-297.46. Satokawa, S.; Miyasaki, R.; Osaki, Y.; Tomura, S.; Shibasaki, Y. Effects of Acidity on the Hydrothermal Synthesis of Kaolinite from Silica-Gel and Gibbsite. Clays Clay Miner., 1996, 44, 417-42347. Táborosi, A.; Szilagyi, R. K.; Zsirka, B.; Fónagy, O.; Horváth, E.; Kristóf, J. Molecular Treatment of Nano-Kaolinite Generations. 2018, Inorg. Chem. 57, 7151–7167.48. Machida, S.; Gotoh T.; Katsumata, K.; Yasumori. A. A Facile Method of Generating Aluminosilicate Nanoparticles: Complete Fragmentation of Kaolinite Nanoscrolls via Manual Grinding. Appl. Clay Sci., 2021, 214, 106295.49. Hashimoto, M.; Takahashi, S.; Kawahara, K.; Ogawa, T.; Kawashita, M. Effect of Heating Conditions on the Magnetic Properties of Micron-Sized Carboxyl Modified-Magnetic Particles Synthesized by a Spray Pyrolysis and Heating Process. Adv. Powder Technol., 2022, 33, 1034112.50. Nakamura, T.; Wada, M.; Hayashi, K.; Kitaoka, S.; Nagai, T.; Yano, J.; Muto, N.; Nakahira, A. Development of Rapid Debinding Treatement Using Superheated Steam and Debinding Behavior for Alumina Molded Bodies. J. Jpn. Soc. Powder Powder Metallurgy, 2019, 66, 275-281.Figures and SchemeScheme 1. The crystal structures of hexacelsian designed by VESTA. Orenge spheres indicate Ba, and pink and light blue spheres indicate Al and Si. Black solid lines indicate the unit cell.Figure 1. (A) XRD patterns of the kaolinite-BaCO3 mixture calcined at (a) 800°C, (b) 900°C, (c) 950°C, (d) 1000°C, (e) 1300°C, and (f) 1600°C. (B) XRD patterns of YSZ: (a) as-received, and calcined at (b) 1000°C, (c) 1300°C, and (d) 1600°C. Figure 2. (A) XRD patterns of (a) the kaolinite-BaCO3 mixture, (b) YSZ, (c) the kaolinite-BaCO3-YSZ mixture calcined at 1600°C. (B) XRD patterns of the corresponding specimens calcined at 1300°C. closed squares indicate reflections due to tetragonal ZrO2Figure 3. SEM (SE) images of the specimens calcined at 1600°C: (a) the kaolinite-BaCO3 mixture, (b) its enlarged view, (c) the kaolinite-BaCO3-YSZ mixture, (d) its enlarged view, (e) YSZ, and (f) its enlarged view. The corresponding SEM (BE) images of (b), (d), and (f) are shown in (b’), (d’), and (f’), respectively.Figure 4. XRD patterns of the specimens calcined at 1600°C: (a) the kaolinite-BaCO3-YSZ mixture, (b) the kaolinite-YSZ mixture, and (c) the BaCO3-YSZ mixture.Figure 5. (A) XRD patterns of the specimens calcined at 800°C: (a) the kaolinite-BaCO3 mixture, (b) YSZ, (c) the kaolinite-BaCO3-YSZ mixture, (d) the kaolinite-YSZ mixture, and (e) the BaCO3-YSZ mixture. (B) XRD patterns of the corresponding specimens calcined at 1000°C.Figure 6. TG-DTA curves of (a) the kaolinite-BaCO3 mixture, (b) the kaolinite-BaCO3-YSZ mixture, (c) the kaolinite-YSZ mixture, and (d) the BaCO3-YSZ mixture.Figure 7. 27Al and 29Si MAS NMR spectra of (a) kaolinite, (b) the kaolinite-BaCO3 mixture calcined at 1600°C, and (c) the kaolinite-BaCO3-YSZ mixture calcined at 1600°C.Figure 8. XRD patterns of kaolinite subjected to relatively strong dry grinding: (a) 36, (b) 72, and (c) 192 cycles. Closed squares indicate reflections from tetragonal ZrO2. Insets show photographs of kaolinite and its ground form after 192 cycles (left, from left to right), and the ZrO2 balls and pot.Figure 9. SEM (SE) images of kaolinite subjected to relatively strong dry grinding: (a) 36, (b) 72, and (c) 192 cycles. The corresponding SEM (BE) image of (c) is shown in (c’).Figure 10. XRD patterns of (a) BAS-1600 and (b) BAS-milled-1600.Figure 11. SEM (SE) images of (a) BAS-1600 and (b) BAS-milled-1600. The corresponding SEM (BE) images are shown in (a’) and (b’), respectively. TOC2image3.tiffimage4.jpegimage5.tiffimage6.tiffimage7.tiffimage8.tiffimage9.tiffimage10.jpegimage11.tiffimage12.jpegimage13.tiffimage1.tiffimage2.tiff