# Fileset

[Liu_JCSJ_main.pdf](https://mdr.nims.go.jp/filesets/f3df35c9-fb11-400b-b2df-97e26cb2df1e/download)

## Creator

Xun Liu, [Takeo Ohsawa](https://orcid.org/0000-0001-7528-8940), [Noriko Saito](https://orcid.org/0000-0002-8104-0172), [Kohsei Takahashi](https://orcid.org/0000-0002-6443-1534), [Takashi Takeda](https://orcid.org/0000-0003-2510-4562), Kenzo Deguchi, [Shinobu Ohki](https://orcid.org/0000-0002-7357-3833), Tetsuo Kishi, Tetsuji Yano, [Hiroyo Segawa](https://orcid.org/0000-0002-7198-8410), [Naoki Ohashi](https://orcid.org/0000-0002-4011-0031)

## Rights

© 2026 The Ceramic Society of Japan 

Abstract

Europium-doped oxynitride glass powder (Sr–Si–Al–O–N) was synthesized by a sol–gel method followed by ammonolysis to investigate the effect of the ammonolysis conditions on the structure and properties of the glass powder. In particular, the effect of the ammonia gas flow rate during nitridation was studied. The effective nitrogen concentration (Neff) in the obtained powder, analyzed by X-ray fluorescence, increased with an increase in the flow rate, and the results of X-ray photoemission, nuclear magnetic resonance, and Fourier transform infrared (FT-IR) spectroscopy measurements indicated that the population of Si–N bonds increased with an increase in Neff. However, the presence of hydrogen-terminated structures, such as –NHn, in the powder with high Neff was confirmed by FT-IR measurements. The presence of hydrogen-terminated structures, such as –Si–NHn, and the thermal stability of these hydrogen-related structures were further investigated by thermal analyses, including thermal desorption measurements, which suggested that hydrogen-terminated structures can be easily formed during the nitridation of the gel and that the formation of hydrogen-terminated structures inhibits the polymerization of the glass structures.
References (73)

    1) K. M. Youssef, A. J. Zaddach, C. Niu, D. L. Irving and C. C. Koch, Mater. Res. Lett. 3, 95 (2015). https://doi.org/10.1080/21663831.2014.985855.

    2) D. Kumar, R. Seetharam and K. Ponappa, J. Alloy. Compd. 972, 172732 (2024). https://doi.org/10.1016/j.jallcom.2023.172732.

    3) K. Maeda, F. Takeiri, G. Kobayashi, S. Matsuishi, H. Ogino, S. Ida, T. Mori, Y. Uchimoto, S. Tanabe, T. Hasegawa, N. Imanaka and H. Kageyama, B. Chem. Soc. Jpn. 95, 26 (2022). https://doi.org/10.1246/bcsj.20210351.

    4) J. J. Xu and K. Wu, Coordin. Chem. Rev. 486, 215139 (2023). https://doi.org/10.1016/j.ccr.2023.215139.

    5) X. Chen and K. M. Ok, Chem. Sci. 13, 3942 (2022). https://doi.org/10.1039/D1SC07121A.

    6) Y.-F. Shi, W.-B. Wei, X.-T. Wu, H. Lin and Q.-L. Zhu, Dalton T. 50, 4112 (2021). https://doi.org/10.1039/D1DT00222H.

    7) X. Chen and K. M. Ok, Chem. Sci. 13, 3942 (2022). https://doi.org/10.1039/D1SC07121A.

    8) R. Xie, N. Hirosaki, M. Mitomo, Y. Yamamoto, T. Suehiro and N. Ohashi, J. Am. Ceram. Soc. 87, 1368 (2004). https://doi.org/10.1111/j.1151-2916.2004.tb07738.x.

    9) N. Hirosaki, R.-J. Xie, K. Kimoto, T. Sekiguchi, Y. Yamamoto, T. Suehiro and M. Mitomo, Appl. Phys. Lett. 86, 1 (2005). https://doi.org/10.1063/1.1935027.

    10) D. Wen, H. Kato and M. Kakihana, ACS Sustain. Chem. Eng. 8, 12286 (2020). https://doi.org/10.1021/acssuschemeng.0c04389.

    11) D. Wen, H. Liu, Z. Ma, L. Zhou, J. Li, Y. Guo, Q. Zeng, P. A. Tanner and M. Wu, Angew. Chem. Int. Edit. 62, e202307868 (2023). https://doi.org/10.1002/anie.202307868.

    12) J. I. Yamaura, S. Maki, T. Honda, Y. Matsui, A. Noviyanto, T. Otomo, H. Abe, Y. Murakami and N. Ohashi, Chem. Commun. 56, 1385 (2020). https://doi.org/10.1039/c9cc07029j.

    13) F. Oehler, H. T. Langhammer and S. G. Ebbinghaus, J. Eur. Ceram. Soc. 37, 2129 (2017). https://doi.org/10.1016/j.jeurceramsoc.2016.12.030.

    14) A. Hosono, Y. Masubuchi, T. Endo and S. Kikkawa, Dalton T. 46, 16837 (2017). https://doi.org/10.1039/C7DT03765A.

    15) L. Mao, X. Cai, H. Gao, X. Diao and J. Zhang, Nano Energy 39, 172 (2017). https://doi.org/10.1016/j.nanoen.2017.06.026.

    16) H. Banno, S. Funahashi, T. Asaka, N. Hirosaki and K. Fukuda, J. Solid State Chem. 230, 149 (2015). https://doi.org/10.1016/j.jssc.2015.06.043.

    17) J. Podwórny, A. Pawełek and J. Czechowski, Powder Diffr. 32, S2 (2017). https://doi.org/10.1017/S088571561700001X.

    18) K. Fukui, S. Iimura, A. Iskandarov, T. Tada and H. Hosono, J. Am. Chem. Soc. 144, 1523 (2022). https://doi.org/10.1021/jacs.1c11353.

    19) S. Yamamoto, Y. Ohashi, Y. Masubuchi, T. Takeda, T. Motohashi and S. Kikkawa, J. Alloy. Compd. 482, 160 (2009). https://doi.org/10.1016/j.jallcom.2009.03.151.

    20) J. Li, Y. Jiang, Z. Zhang, M. Tsuji, M. Miyazaki, M. Kitano and H. Hosono, Adv. Energy Mater. 13, 2302424 (2023). https://doi.org/10.1002/aenm.202302424.

    21) A. Noviyanto, T. Nishimura, M. Kitano and N. Ohashi, J. Eur. Ceram. Soc. 36, 4083 (2016). https://doi.org/10.1016/j.jeurceramsoc.2016.06.031.

    22) M. Namba, H. Takatsu, H.-B. Li, K. Murayama, R. Terada, Q. Yang, T. Terashima, H. Ohta and H. Kageyama, Chem. Mater. 36, 2076 (2024). https://doi.org/10.1021/acs.chemmater.3c03178.

    23) J. Lee, Q. Zhang and F. Saito, Ind. Eng. Chem. Res. 40, 4785 (2001). https://doi.org/10.1021/ie010086k.

    24) N. Arumugam, A. Hönnerscheid and M. Jansen, Z. Anorg. Allg. Chem. 629, 939 (2003). https://doi.org/10.1002/zaac.200200452.

    25) Y. Hinuma, A. Grüneis, G. Kresse and F. Oba, Phys. Rev. B 90, 155405 (2014). https://doi.org/10.1103/PhysRevB.90.155405.

    26) A. Duval, P. Houizot and T. Rouxel, J. Am. Ceram. Soc. 106, 1611 (2023). https://doi.org/10.2109/jcersj2.18097.

    27) H. Hyuga, J. Ceram. Soc. Jpn. 126, 968 (2018). https://doi.org/10.2109/jcersj2.18097.

    28) K. Watari, H. J. Hwang, M. Toriyama and S. Kanzaki, J. Mater. Res. 14, 1409 (1999). https://doi.org/10.1557/JMR.1999.0191.

    29) S. Watanabe, Y. Osawa, S. Machida, K. Katsumata, A. Yasumori, K. Takahashi, K. Deguchi, S. Ohki and H. Segawa, J. Non-Cryst. Solids 573, 121107 (2021). https://doi.org/10.1016/j.jnoncrysol.2021.121107.

    30) X. Liu, T. Ohsawa, J. Xu, M. Yanagida, K. Takahashi, T. Takeda, T. Kishi, T. Yano, H. Segawa and N. Ohashi, J. Mater. Chem. C 13, 18286 (2025). https://doi.org/10.1039/D5TC01527H.

    31) H. Segawa, N. A. Wójcik, K. Takahashi, T. Takeda and S. Ali, J. Am. Ceram. Soc. 107, 2930 (2024). https://doi.org/10.1111/JACE.19615.

    32) H. Segawa and N. Hirosaki, Ceram. Int. 44, 4783 (2018). https://doi.org/10.1016/j.ceramint.2017.12.063.

    33) S. Li, L. Wang, D. Tang, Y. Cho, X. Liu, X. Zhou, L. Lu, L. Zhang, T. Takeda, N. Hirosaki and R.-J. Xie, Chem. Mater. 30, 494 (2018). https://doi.org/10.1021/acs.chemmater.7b04605.

    34) R. E. Loehman, J. Non-Cryst. Solids 42, 433 (1980). https://doi.org/10.1016/0022-3093(80)90042-3.

    35) H. Schmidt, M. Gupta and M. Bruns, Phys. Rev. Lett. 96, 055901 (2006). https://doi.org/10.1103/PhysRevLett.96.055901.

    36) H. Haneda, T. Ohgaki, I. Sakaguchi, H. Ryoken, N. Ohashi and A. Yasumori, Appl. Surf. Sci. 252, 7265 (2006). https://doi.org/10.1016/j.apsusc.2006.02.179.

    37) A. Hosono, Y. Masubuchi and S. Kikkawa, Ceram. Int. 43, 2737 (2017). https://doi.org/10.1016/j.ceramint.2016.11.101.

    38) Y.-R. Zhang, T. Motohashi, Y. Masubuchi and S. Kikkawa, J. Eur. Ceram. Soc. 32, 1269 (2012). https://doi.org/10.1016/j.jeurceramsoc.2011.12.001.

    39) S. Kikkawa, S. Sun, Y. Masubuchi, Y. Nagamine and T. Shibahara, Chem. Mater. 28, 1312 (2016). https://doi.org/10.1021/acs.chemmater.5b04149.

    40) A. Hosono, S.-K. Sun, Y. Masubuchi and S. Kikkawa, J. Eur. Ceram. Soc. 36, 3341 (2016). https://doi.org/10.1016/j.jeurceramsoc.2016.05.006.

    41) Y. Masubuchi, S. Ohtaki, K. Fujii, M. Yashima, M. Higuchi and S. Kikkawa, J. Eur. Ceram. Soc. 40, 6288 (2020). https://doi.org/10.1016/j.jeurceramsoc.2019.10.004.

    42) A. Hosono, M. Inoguchi, Y. Masubuchi, K. Murayama, M. Iha, M. Higuchi and S. Kikkawa, J. Eur. Ceram. Soc. 40, 2317 (2020). https://doi.org/10.1016/J.JEURCERAMSOC.2020.02.008.

    43) Y. Osawa, K. Iwasaki, T. Nakanishi, A. Yasumori, Y. Matsui, T. Nishimura, T. Ohsawa, H. Segawa and N. Ohashi, J. Am. Ceram. Soc. 104, 4420 (2021). https://doi.org/10.1111/JACE.17836.

    44) R. K. Brow, C. G. Pantano and D. C. Boyd, J. Am. Ceram. Soc. 67, c72 (1984). https://doi.org/10.1111/J.1151-2916.1984.TB18834.X.

    45) K. Szaniawska, L. Murawski, R. Pastuszak, M. Walewski and G. Fantozzi, J. Non-Cryst. Solids 286, 58 (2001). https://doi.org/10.1016/S0022-3093(01)00478-1.

    46) B. E. Warren, Chem. Rev. 26, 237 (1940). https://doi.org/10.1021/cr60084a007.

    47) H. W. Nesbitt, G. M. Bancroft, G. S. Henderson, R. Ho, K. N. Dalby, Y. Huang and Z. Yan, J. Non-Cryst. Solids 357, 170 (2011). https://doi.org/10.1016/J.JNONCRYSOL.2010.09.031.

    48) G. Pintori and E. Cattaruzza, Opt. Mater. 13, 100108 (2022). https://doi.org/10.1016/J.OMX.2021.100108.

    49) J. F. Moulder and J. Chastain, “Handbook of X-ray Photoelectron Spectroscopy: A Reference Book of Standard Spectra for Identification and Interpretation of XPS Data”, Physical Electronics Division, Perkin-Elmer Corp. (1992).

    50) K. Kawashima, M. Hojamberdiev, O. Mabayoje, B. R. Wygant, K. Yubuta, C. B. Mullins, K. Domen and K. Teshima, CrystEngComm 19, 5532 (2017). https://doi.org/10.1039/C7CE00614D.

    51) B. Li, Y. Adachi, J. Li, H. Okushi, I. Sakaguchi, S. Ueda, H. Yoshikawa, Y. Yamashita, S. Senju, K. Kobayashi, M. Sumiya, H. Haneda and N. Ohashi, Appl. Phys. Lett. 98, 082101 (2011). https://doi.org/10.1063/1.3556440.

    52) S. Hirose, S. Ueda and N. Ohashi, J. Appl. Phys. 125, 95301 (2019). https://doi.org/10.1063/1.5066574/156311.

    53) K. R. Carduner, R. O. Carter, M. E. Milberg and G. M. Crosbie, Anal. Chem. 59, 2794 (1987). https://doi.org/10.1021/ac00150a015.

    54) C. Cozzan, K. J. Griffith, G. Laurita, J. G. Hu, C. P. Grey and R. Seshadri, Inorg. Chem. 56, 2153 (2017). https://doi.org/10.1021/acs.inorgchem.6b02780.

    55) R. van Weeren, E. A. Leone, S. Curran, L. C. Klein and S. C. Danforth, J. Am. Ceram. Soc. 77, 2699 (1994). https://doi.org/10.1111/j.1151-2916.1994.tb04664.x.

    56) C. I. Merzbacher, B. L. Sherriff, J. S. Hartman and W. B. White, J. Non-Cryst. Solids 124, 194 (1990). https://doi.org/10.1016/0022-3093(90)90263-L.

    57) S. Watanabe, Y. Osawa, S. Machida, K. Katsumata, A. Yasumori, K. Takahashi, K. Deguchi, S. Ohki and H. Segawa, J. Non-Cryst. Solids 573, 121107 (2021). https://doi.org/10.1016/J.JNONCRYSOL.2021.121107.

    58) Z. Zhao, D. Xiao, K. Chen, R. Wang, L. Liang, Z. Liu, I. Hung, Z. Gan and G. Hou, ACS Central Sci. 8, 795 (2022). https://doi.org/10.1021/acscentsci.1c01497.

    59) S. Sakka, J. Non-Cryst. Solids 181, 215 (1995). https://doi.org/10.1016/S0022-3093(94)00514-1.

    60) N. Hirosaki, C. Kocer, S. Ogata and K. Tatsumi, Phys. Rev. B 71, 104105 (2005). https://doi.org/10.1103/PhysRevB.71.104105.

    61) E. Leonova, A. S. Hakeem, K. Jansson, B. Stevensson, Z. Shen, J. Grins, S. Esmaeilzadeh and M. Edén, J. Non-Cryst. Solids 354, 49 (2008). https://doi.org/10.1016/j.jnoncrysol.2007.07.027.

    62) A. Bertoluzza, C. Fagnano, M. Antonietta Morelli, V. Gottardi and M. Guglielmi, J. Non-Cryst. Solids 48, 117 (1982). https://doi.org/10.1016/0022-3093(82)90250-2.

    63) H. Aguiar, J. Serra, P. González and B. León, J. Non-Cryst. Solids 355, 475 (2009). https://doi.org/10.1016/J.JNONCRYSOL.2009.01.010.

    64) A. Bertoluzza, C. Fagnano, M. Antonietta Morelli, V. Gottardi and M. Guglielmi, J. Non-Cryst. Solids 48, 117 (1982). https://doi.org/10.1016/0022-3093(82)90250-2.

    65) B. T. Poe, P. F. McMillan, C. A. Angell and R. K. Sato, Chem. Geol. 96, 333 (1992). https://doi.org/10.1016/0009-2541(92)90063-B.

    66) H. Ono, T. Ikarashi, Y. Miura, E. Hasegawa, K. Ando and T. Kitano, Appl. Phys. Lett. 74, 203 (1999). https://doi.org/10.1063/1.123293.

    67) M. Klanjšek Gunde and M. Maček, Phys. Status Solidi A 183, 439 (2001). https://doi.org/10.1002/1521-396X(200102)183:2<439::AID-PSSA439>3.0.CO;2-B.

    68) K. Kamiya, M. Ohya and T. Yoko, J. Non-Cryst. Solids 83, 208 (1986). https://doi.org/10.1016/0022-3093(86)90068-2.

    69) J. Szépvölgyi, I. Mohai and J. Gubicza, J. Mater. Chem. 11, 859 (2001). https://doi.org/10.1039/b008429h.

    70) H. Mizoguchi, Y. Osawa, M. Sasase, N. Ohashi, M. Kitano and H. Hosono, J. Phys. Chem. Lett. 14, 9516 (2023). https://doi.org/10.1021/ACS.JPCLETT.3C02446.

    71) J. Bouquiaux, S. Poncé, Y. Jia, A. Miglio, M. Mikami and X. Gonze, Chem. Mater. 35, 5353 (2023). https://doi.org/10.1021/acs.chemmater.3c00537.

    72) R. Shafei, P. J. Strobel, P. J. Schmidt, D. Maganas, W. Schnick and F. Neese, Phys. Chem. Chem. Phys. 26, 6277 (2024). https://doi.org/10.1039/D3CP06039J.

    73) R. Shafei, P. J. Strobel, P. J. Schmidt, D. Maganas, W. Schnick and F. Neese, J. Phys. Chem. C 129, 1495 (2025). https://doi.org/10.1021/acs.jpcc.4c06912.

Content from these authors
Supplementary material (1)
Supporting Information  Download (PDF)
© 2026 The Ceramic Society of Japan

この記事はクリエイティブ・コモンズ [表示 4.0 国際]ライセンスの下に提供されています。
https://creativecommons.org/licenses/by/4.0/deed.ja
[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Tracing nitridation reaction toward efficient production of oxynitride glasses as hosts for bright luminescence centers](https://mdr.nims.go.jp/datasets/c56e3e67-be15-426d-94b4-fb80ef29fcc4)

## Fulltext

Tracing nitridation reaction toward efficient production of oxynitride glasses as hosts for bright luminescence centersFULL PAPERTracing nitridation reaction toward efficient productionof oxynitride glasses as hosts for bright luminescence centersXun Liu1,2, Takeo Ohsawa1, Noriko Saito1, Kohsei Takahashi1, Takashi Takeda1, Kenzo Deguchi1,Shinobu Ohki1, Tetsuo Kishi3, Tetsuji Yano3, Hiroyo Segawa1,3,³ and Naoki Ohashi1,2,41National Institute for Materials Science (NIMS), 1–1 Namiki, Tsukuba, Ibaraki 305–0044, Japan2Interdisciplinary Graduate School of Engineering Sciences, Kyusyu University, 6–1 Kasugakoen, Kasuga, Fukuoka 816–8580, Japan3Department of Chemistry and Materials Science, Graduate School of Science and Engineering, Institute of Science Tokyo,2–12–1 Ookayama, Meguro-ku, Tokyo 152–8552, Japan4Materials DX Research Center for Element Strategy, Institute of Science Tokyo,4259 Nagatsuta, Midori-ku, Yokohama 226–8503, JapanEuropium-doped oxynitride glass powder (Sr–Si–Al–O–N) was synthesized by a sol–gel method followed byammonolysis to investigate the effect of the ammonolysis conditions on the structure and properties of the glasspowder. In particular, the effect of the ammonia gas flow rate during nitridation was studied. The effectivenitrogen concentration (Neff) in the obtained powder, analyzed by X-ray fluorescence, increased with an increasein the flow rate, and the results of X-ray photoemission, nuclear magnetic resonance, and Fourier transforminfrared (FT-IR) spectroscopy measurements indicated that the population of Si–N bonds increased with anincrease in Neff. However, the presence of hydrogen-terminated structures, such as –NHn, in the powder withhigh Neff was confirmed by FT-IR measurements. The presence of hydrogen-terminated structures, such as –Si–NHn, and the thermal stability of these hydrogen-related structures were further investigated by thermalanalyses, including thermal desorption measurements, which suggested that hydrogen-terminated structures canbe easily formed during the nitridation of the gel and that the formation of hydrogen-terminated structuresinhibits the polymerization of the glass structures.Key-words : Oxynitride glass, Nitridation, Ammonolysis, Sol–gel method, Europium, Luminescence[Received October 29, 2025; Accepted December 6, 2025; Published online January 16, 2026]1. IntroductionIn recent years, a major focus in materials science hasbeen the exploration of compounds with complex chem-ical compositions to realize further higher materials per-formance. The term “high-entropy alloys”1,2) is high-lighted in metallurgy, and “mixed-anion compounds”3–5) isanother encouraging term for inorganic functional materi-als. Because atomic numbers are integers, alloying andmaking solid solutions are the most common approachesfor achieving precise control of material properties. Manyexamples of mixed-anion compounds are available. Forinstance, oxychalcogenides6) and oxyhalides7) have beenconsidered for electro-optical coupling because anionmixing is a possible way to introduce the lattice distortionsassociated with electric dipoles. Similarly, many phos-phors are produced by anion mixing to form specificcoordination structures around luminescent centers.8–11)Two categories of mixed-anion compounds exist: thoseinvolving isovalent anion mixing, such as oxychalcoge-nides, and those involving heterovalent anion mixing, suchas oxyhalides. Charge compensation is an issue in hetero-valent anion mixing. Simultaneous cation substitution tomaintain charge neutrality and crystal structure,12–15) ormodification of the crystal structure to allow changes in thecation/anion ratio,16–18) is necessary to form heterovalentanion exchanges, both of which are required to formheterovalent mixed-anion compounds.19) Because the va-lence band in inorganic compounds is usually formed withthe valence electrons of anions, heterovalent anion mixingdirectly impacts the electronic structure of the valenceband.15,20) Hence, state-of-the-art techniques are sometimesrequired for the synthesis of heterovalent anion mix-tures.19,21–24) Despite the difficulties in synthesis, hetero-valent anion mixing extends the ability of material design.Mixed-anion compounds in electrochemical systems areexpected to modify the electronic and chemical propertiesof the materials. Anion exchange affects the most funda-mental parameters, for example, work-function and elec-tron affinity,20,25) which are essential in electrochemistry.This study focuses on oxynitride glasses, a family ofmixed-anion systems. The incorporation of nitrogen intooxide glasses is a feasible approach in glass technology.26)³ Corresponding author: H. Segawa; E-mail: SEGAWA.Hiroyo@nims.go.jpJournal of the Ceramic Society of Japan (2026), Advance Publication by J-stageDOI https://doi.org/10.2109/jcersj2.25148 JCS-Japan©2025 The Ceramic Society of JapanThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by-nc/4.0/),which permits use, distribution, and reproduction in any medium for non-commercial purposes, provided the original work is properly cited.https://doi.org/10.2109/jcersj2.25148https://creativecommons.org/licenses/by-nc/4.0/As nitrides are expected to exhibit higher mechanicalstrength and melting temperatures than oxides, oxynitrideshave been highlighted as potential structural materials forhigh-temperature applications.27,28) Hence, the enhance-ment of the mechanical and chemical properties is the pur-pose of oxynitride glass development. We are interested inthe development of host materials for luminescent lantha-nide centers. Oxynitride glasses are expected to becomeoutstanding host materials for luminescent lanthanide cen-ters,29–31) assuming an analogy to crystalline oxynitridephosphors such as MxSi12¹(m+n)Alm+nOnN16¹n (M = lan-thanide), conventionally called SiAlON.8,32,33) However,ordinary glass processes, including melting and casting,are unlikely to be appropriate for oxynitride glass produc-tion, because stable melts are rarely formed.26,34) Thus,nitridation of conventional oxide glass has been consid-ered a conventional route to form oxynitride glasses.As mentioned above, the high-temperature stability ofnitrides and oxynitrides is an evident advantage for theirapplication. However, their very slow ion diffusivity, whichis the origin of their high-temperature stability, causesdifficulties in their fabrication. To the best of our knowl-edge, reports on the nitrogen ion diffusion coefficient innitrides are rare because extremely low diffusivity causestechnical difficulties.35,36) Melting, followed by casting, isthe most conventional method for fabricating glass. Shap-ing by casting is the most obvious advantage of glassmaterials. However, formation of molten oxynitride glassesis very difficult.37–42)To overcome such difficulties, gels and porous nano-granular oxides have been considered as source materialsfor the synthesis of oxynitride glasses.43–45) The smallgrain size and large surface area of the oxide grains shouldenhance nitridation with a very large surface area andshorten the diffusion path for the completion of nitrida-tion.21) However, our previous demonstration suggestedthat a relatively long heat treatment (e.g., 24 h) is neces-sary to obtain a stable oxynitride glass structure and thatthe very slow structural relaxation in oxynitride glasses isstill an issue for their preparation of oxynitride glasses.Hence, we were motivated to trace the nitridation behaviorduring ammonolysis of oxide gels with the aim of pro-posing an efficient nitridation process to produce oxy-nitride glass powders with relatively short reaction times.In particular, we focused on the effect of ammonia gasflow rate during ammonolysis to elucidate the reactionkinetics.In this study, Eu-doped Sr–Si–Al–O–N glass was inves-tigated as a case study to elucidate the nitridation behaviorof xerogels. The ammonolysis conditions were varied toachieve variations in the nitrogen concentration in theresultant materials, and their chemical states were charac-terized by spectroscopic techniques, including magic-angle-spin nuclear magnetic resonance (MAS-NMR), Fou-rier transform infrared spectroscopy (FT-IR), and X-rayphotoelectron spectroscopy (XPS), in addition to thermalanalysis, that is, thermal gravimetry (TG) and thermaldesorption spectrometry (TDS). Luminescence originatingin the Eu centers was also studied as a probe to observevariations in glass structures.2. ExperimentGlass powder samples, (Si0.85Al0.15Sr0.10Eu0.03)–O–N,were synthesized through a two-step process, that is, thepreparation of xerogels by a sol–gel process and subse-quent nitridation by ammonolysis. The sol–gel processwas initiated by the hydrolysis of tetraethoxysilane(TEOS, >95%, Fujifilm) using an aqueous solution ofnitric acid, and the TEOS:H2O:HNO3 ratio was set at1:4.7:0.001 in molar ratio. After stirring the obtained silicasol for 1 h, a solution mixed with ethanol and N,N-dimethylformamide (DMF, >99.5%, Wako), in whichAl(NO3)3·9H2O (99.9%, Fujifilm) and Eu(NO3)3·6H2O(99.9%, Wako) were dissolved, was added to the silica soland the mixture was stirred for 15min.42) Subsequently, anaqueous solution containing Sr(NO3)2 (>98%, Wako) wasadded to the mixture to set the TEOS:H2O = 1:9.4 andwas stirred for 15min. Finally, the nominal composition ofthe sol was TEOS:EtOH:DMF:H2O = 1:4.7:4.7:14.1 inmolar fraction, and the nominal composition of themetallic elements in the sol was Si:Al:Sr:Eu = 85:15:10:3.The obtained sol was dried at 80 °C for approximately 3days to form transparent solidified gels. The dry gel wascrushed in an agate mortar for several minutes.The nitridation of the gel powder was examined byammonolysis using a tube furnace with a pure silica glasstube. Highly purity ammonia gas (NH3, 99.999%, ShowaDenko K.K.) was supplied at flow rates ()AM) of 20, 100,300, and 500 cm3/min to evaluate the effect of gas flow onthe nitridation behavior. Before introducing the NH3 gas,the furnace tube was evacuated to reduce the oxygenpartial pressure to the lowest possible level. NH3 gas wasintroduced into the furnace prior to heating and flowedcontinuously through the high-temperature process. Thefurnace was heated to 1000 °C at the ramping rate of330 °C/h unless otherwise noted and held at 1000 °C for12 h. Hereafter, the samples are identified by the codesP020, P100, P300, and P500 with )AM during ammonol-ysis of 20, 100, 300, and 500 cm3/min, respectively. Inaddition, a subset of P500 samples was prepared at aslower ramping rate of 110 °C/h to elucidate the role of thecalcination process. Because organic decomposition anddehydrogenation occur during heating, the ramping rate isexpected to influence the chemical states of the resultantsamples. These samples are denoted as P500SR. Notably,the exhaust gas from the furnace during ammonolysis wasanalyzed using gas chromatography to estimate the actualgas composition in the furnace.In addition to the powders synthesized via ammonol-ysis, two reference powders were prepared. One referencepowder, identified by PwO, was prepared by oxidizing theprecursor gel at 1000 °C for 6 h in air atmosphere. Theother one, identified by PwH, was prepared in two steps asfollows: First, the gel was heated in air atmosphere at1000 °C for 4 h and subsequently heated in H2 gas streamat 1000 °C for 2 h.Liu et al.: Tracing nitridation reaction toward efficient production of oxynitride glasses as hosts for bright luminescence centersJCS-JapanAbsence of crystalline phases in the obtained powderwas confirmed by powder X-ray diffraction (XRD) withCuK¡ radiation using a MiniFlex diffractometer (RigakuCo. Ltd.). The average chemical composition was ana-lyzed via X-ray fluorescence (XRF) using a ZSX Primus IIspectrometer (Rigaku Co., Ltd.) with semi-qualitativeanalysis functions. Spectroscopic characterizations wereperformed to elucidate the coordination structures formedin the samples. The vibrational modes of the samples werecharacterized using a Nicolet iS50FT-IR spectrometer(Thermo Fisher Scientific Inc.). For most FT-IR measure-ments, the powder samples were pelletized with KBrpowder and the pellets were dried by heating at 150 °C for15min in order to avoid the contribution of extrinsic waterto the vibrational spectra. In addition, the attenuated totalreflection (ATR) technique was adopted for the FT-IRmeasurements, and a diamond prism was utilized to avoidthe effects of water adsorption on the KBr powder. Thechemical state of every element in the resultant glasspowders was characterized by XPS, using a PHI-Quantesspectrometer (ULVAC-PHI, Inc.) equipped with a mono-chromatic AlK¡ X-ray generator. The pressure in the XPSchamber was maintained at 1 © 10¹8 Torr during measure-ments and a charge neutralizer was used to enable mea-surements for the insulating powders.The local structures around Si and Al were also inves-tigated using MAS-NMR. The MAS-NMR signal of 27Alat 208.49MHz was studied with a JNM-ECZ 800MHzspectrometer (JEOL Ltd.) and the samples were set in a3.2mm double-tuned broadband probe spinning at 20 kHz.An aqueous solution of AlCl3 (1mol%) was used as areference for the 27Al MAS-NMR measurements. Thepulse width of 0.6¯s was used, which was one-third of³/2 (= 1.8¯s) in the solution, the relaxation delay wasset at 1.0 s, and the chemical shift was calibrated by set-ting the chemical shift of the solution at 0 ppm. Further,high-resolution 29Si signal was collected by the dipolar-decoupling-magic-angle-spinning (DDMAS) procedure at99.38MHz with a JNM-ECA 500MHz spectrometer(JEOL Ltd.), and tetramethylsilane was used as an externalstandard. The samples were packed in a 4mm zirconiarotor, the spin frequency was set at 10 kHz, and the pulsewidth and relaxation delay were set at 3¯s of ³/2 pulseand 120 s, respectively.Thermal analysis was performed by TDS and differ-ential thermal analysis (DTA). TG–DTA measurementswere performed with a conventional instrument (TG–DTA2000SA, Bruker Japan K.K.) under N2 gas stream at 150cm3/min, using Al2O3 powder as a reference. For TG–DTA measurements, approximately 20mg powder sam-ples were set in a Pt pan and heated at a ramping rate of10 °C/min. TDS spectra were measured using an EMD-WA1000S system (ESCO, Ltd.). The powdered sampleswere packed with Pt foil to avoid reaction with the cruciblemade of silicon carbide. The base pressure in the analyticalchamber was maintained at 10¹9 Torr or lower prior toheating. The sample temperature was elevated to 1000 °Cat a ramping rate of 3 °C/min, held at 1000 °C for 3 h, andcooled down to room temperature. This temperature pro-gram was repeated twice to ensure complete desorptionduring the first heating cycle. The P500SR sample afterthe TDS measurement was analyzed using XPS again toexamine the changes in the chemical state during themeasurement.The photoluminescence (PL) and photoluminescenceexcitation (PLE) spectra were obtained using an F-7100fluorescence meter (Hitachi High-Tech Corp.). The PLspectra were excited by UV light at ­ = 300 nm, and theintensity of the PLE spectra was monitored at the wave-length of maximum PL intensity. In addition, the quantumefficiency (QE) of the luminescence was measured usinga QE-2100 spectrometer (Otsuka Electronics Co., Ltd.)according to the literature.9) All optical measurementswere performed at room temperature.3. Results and discussionThe XRD patterns of the samples after ammonolysis areshown in Fig. S1. Except for P020, all samples remainedamorphous after ammonolysis. The XRD pattern of P020exhibited sharp diffraction peaks assignable to crystallinecristobalite, indicating insufficient NH3 flow, that is, resid-ual oxygen gas in the furnace, resulting in crystallization.The actual atmosphere in the furnace during ammonolysis,shown in Fig. S2 (Supporting Information), indicated thatthe Po2 in the furnace with low )AM was relatively high. Itis also noteworthy that the halo pattern in the XRD profilesshowed shift and broadening with )AM. Because the halopattern correlates with the short-range arrangement ofatoms in glass,46) it is presumed that glass structures areresponsible for )AM.The effective nitrogen concentration (Neff) values, asdefined by Eqs. (1) and (2),31) were evaluated from theresults of the XRF analysis and are shown in Fig. 1.Assuming stoichiometry (formal charge of the constituentelements, particularly Eu2+) in the resulting samples, thesample composition can be expressed by Eq. (1), and,hence, Neff was calculated using Eq. (2) from the exper-imentally observed nitrogen concentration, [N], and oxy-gen concentration, [O].504030201005004003002001000330°C/h110°C/hRamping rateP020P100P300P500SRP500Neff (%) NH3 flow rate (cm3/min)Fig. 1. Nitrogen contents in the synthesized powder versus theammonia flow rate ()AM) on preparation. The powders are iden-tified by Pn where n corresponds to )AM. Blue and red symbolsdistinguish the ramping rate in the nitridation process.Journal of the Ceramic Society of Japan (2026), Advance Publication by J-stage JCS-JapanSi0:85Al0:15Sr0:10EuðIIÞ0:03O2:055 þ1:370Neff2N2! ð1�NeffÞðSi0:85Al0:15Sr0:10Eu0:03O2:055Þþ NeffðSi0:85Al0:15Sr0:10Eu0:03N1:370Þþ 2:055Neff2O2 ð1ÞNeff ð%Þ ¼ ½N� � 3½N� � 3þ ½O� � 2� 100 ð2ÞThe monotonic increase of Neff with )AM looks reasonable,as )AM represents nitridation activity. As presented inFig. S2 in Supporting Information, the actual ammoniapartial pressure in the furnace increased with the increasein )AM.Notably, the charge compensation assumed in Eq. (1) isnot satisfied with the obtained powder. In the PwO pow-der prepared by firing in air, the XRF analysis resultssuggested a slight anion (oxygen) excess under the as-sumption of an ordinary formal charge, as presented inTable S1. This indicates that some cations that cannot bedetected by XRF exist in the obtained powder to satisfycharge neutrality. Because the instrument has no sensitivityto very light elements, that is, H–Be, we could not analyzethe hydrogen concentration with the current setup. Thisimplies that the presence of hydrogen-related componentsis a possible explanation for the observed anion excess.Hence, the splitting of the –Si–O–Si bond in the manner ofEq. (3), for instance, is a reasonable understanding of theexcess oxygen in PwO.½­Si­O­Si­� þ H2O ! ½­Si­O­H� þ ½H­O­Si­�ð3ÞThe degree of anion excess was more obvious in thesample prepared with higher )AM, for example, P500SR.As can be observed in Table S2, the tendency of anionexcess was more obvious in the powders prepared withhigher )AM, which was confirmed from the results of XRFanalyses. Hence, it is also presumed that the breaking ofbonds in the glass structure, in the manner of Eq. (4), is apossible consideration.½­Si­N­Si­� þ NH3! ½­Si­N­H� þ ½H­N­Si­� þ 12H2 ð4ÞThe possibility of these terminal, nonbridging structureswill be discussed further, along with the results of othermeasurements.Because the presence of –NH and –OH groups was notassumed in Eqs. (1) and (2), the Neff values determinedfrom the total oxygen [O] and nitrogen [N] concentrationsmay differ from those in the original definition. However,for convenience, we will still use the Neff value determinedfrom total [O] and [N] as a measure for the degree ofnitridation.The results of the XPS measurements are shown inFig. 2. As a charge neutralizer was used to enable XPSmeasurement by suppressing charging due to photoioniza-tion, calibration of the electron binding energy (Eb) wasnecessary for sample-to-sample comparison. In this study,the binding energies were normalized by referencing theO1s core-level peaks of each sample instead of Eb of theC1s core-level peaks. As shown in Fig. 2(a), the spectralprofile for the C1s core-level region is broad and asym-metric, which indicates that the composition of carbon-related impurities is very complex and that these peaks arenot suitable for Eb calibration. Hence, Ebs were calibratedby locating the O1s peak at 532 eV, which is close to thetypical Eb for silicate glass reported in some studies.47,48)As shown in Fig. 2(b), which presents the O1s spectraafter calibration, the spectral shape of O1s does not changewith )AM. As neither broadening nor splitting of the O1speak due to differences in preparation conditions wasobserved, it seems that the chemical state of oxygen is288 286 284 282 280 278 276C1sSr3p1/2(a)538 536 534 532 530 528 526O1s(b)404 402 400 398 396 394 392N1s(c)108 106 104 102 100 98 96(d) Si2p10210180 78 76 74 72 70 68(e)Al2p76.076.5 75.5Binding energy (eV)P020 P100 P300P500 P500SRIntensity (a.u.)Fig. 2. XPS spectra of the samples after ammonolysis withdifferent ammonia flow rates. The sample identifier, Pn, indicatespowder prepared with ammonia flow rate at n cm3/min. Thebinding energy is normalized by locating the O1s peak at 532 eV.Part of the spectra are magnified in the insets.Liu et al.: Tracing nitridation reaction toward efficient production of oxynitride glasses as hosts for bright luminescence centersJCS-Japancommon to all powders in this study. Hence, usage of O1speaks for calibration seems more reliable than usage ofC1s peaks, and the chemical state of the other elements ischaracterized by Eb difference between the focused ele-ment and oxygen.Next, we focus on the N1s core-level spectra. As shownin Fig. 1, Neff in P020 was very low; consequently, the N1speak intensity of P020 was very weak and was not avail-able for discussion. Thus, the N1s spectra of P020 wasomitted from Fig. 2(c). For the other samples, the N1speak is located at approximately 397.5 eV and exhibits avery similar profile, indicating that the chemical states ofnitrogen in these samples are close to each other. If weassume that energy calibration locating the O1s peak at532 eV is quite reasonable, the observed N1s XPS signal isvery similar to that of crystalline silicon nitride, whichnormally appears at Eb µ 398 eV.49) Regardless of the Ebcalibration issues, it is evident that the shapes of N1s andO1s peaks merely changed with )AM, and that the differ-ence between Eb of O1s and N1s is constant in all samples.Notably, the energy difference between N1s and O1s (134eV) observed in this study is constant against variation of)AM and is the same as that in other oxynitrides, such asSrNbO2N.50) In general, the XPS peak energies show anobvious shift not only due to changes in the chemical statebut also due to a shift in the Fermi energy.51,52) Thus, theenergy difference between O1s and N1s is a good measurefor the chemical state analyses of oxynitrides. Hence, wecan safely conclude that the chemical states of oxygen andnitrogen are unchanged with )AM from the viewpoint of aphotoemission study. The value of Neff determined fromXPS measurements is shown in Supplemental Information(Fig. S3). It was similar to Neff obtained from XRF mea-surements, showing the increase of Neff with )AM. Thedifference in the Neff behavior shown in Figs. 1 and S3 waslikely due to probing depth, high surface sensitivity ofXPS and choice of the relative sensitivity factors forconvertion of XPS peak intensity to composition.As described above, the XRF analyses indicated anexcess of anions compared with those in the assumedcomposition expressed by Eq. (1). Notably, the variationof Neff with )AM determined by XPS (Fig. S3) was con-sistent with that analyzed by XRF. Hence, the excess ofanions is a feasible characteristic of the powder preparedwith high )AM. Assuming that Eqs. (3) and (4) are appli-cable, two or more types of oxygen and nitrogen arepresent, for example, bridging oxygen, such as –Si–O–Si–,and terminal oxygen, such as –Si–O–H. However, the O1sand N1s core-level spectra appeared to be of a singleorigin. Tentatively, we assume that anions at differentcoordination positions, such as the terminal and bridgingpositions, cannot be resolved by the XPS core-level spec-tra obtained in this study, and that the intensity of thosepeaks represents the total concentration of the correspond-ing elements.While the Eb difference between O1s and N1s wasmerely changed with )AM, the XPS peak for some cationicelements showed shift with )AM. Indeed, Eb of the Si2ppeaks shifted with )AM as shown in Fig. 2(d). Corre-sponding to the monotonical increase in Neff with increas-ing )AM, the Eb at Si2p peak showed monotonical shift tothe lower Eb direction with the increase in )AM. This resultsuggests that the chemical state of Si is strongly affectedby Neff, which reproduces the effect of Neff on the XPSprofile of oxynitride glasses, as shown in our previousstudy on Si–Al–Sr–Eu–O–N glass films.30) As the presentstudy employs the O1s core-level for energy calibration,the correct expression for the Si2p XPS peak is that theenergy difference between Si2p and O1s is parallel to thatbetween Si2p and N1s. This indicates that the ionicities ofSi–O and Si–N were enhanced with an increase in Neff.In contrast to the behavior of the Si2p XPS peaks, theAl2p peak profile [Fig. 2(e)] was almost the same for allthe samples, regardless of )AM. When we carefully com-pared the Al2p peak profiles, the effect of )AM onto thepeak profile of Al2p was found at the tail at a higher Ebside, as shown in the inset. It appears that the spectral taildevelops with an increase in )AM. Because the presence ofa spectral tail in the Al2p region due to ammonolysis wasmore evident in our previous study on Si–Al–Sr–Eu–O–Nglass films,30) the behavior of the spectral tail of Al2pshould be attributed to the effect of nitridation. Notably,the effect of nitridation on the peak shape of Al2p, de-velopment of tail at higher Eb direction with )AM, wasopposite to that of Si2p shift to lower Eb direction with theincrease in )AM. The effect of )AM on the peak profile ofSr3p at approximately Eb = 280.5 eV [Fig. 2(a)] seemedto be rather similar to that of Al2p. The Sr3p peak ex-hibited a slight shift to a higher Eb direction with increas-ing )AM. As the magnitude and direction of the peak shiftdue to )AM were different from element to element, neithercharging nor the shift of Fermi level should be the reasonsfor the XPS peak shift. Hence, the results suggested thatthe chemical state of cations systematically changed withthe increase in Neff resulting from the increase in )AM.Figure 3(a) shows 29Si NMR spectra of the powderprepared with different )AM. As shown in this figure, thespectral profile changed drastically with )AM. An excep-tionally sharp peak found in the spectrum of P020 atapproximately ¹110 ppm was assigned to the crystallinephase, cristobalite, formed during ammonolysis. As thenitrogen content increases due to a large )AM, signals athigher chemical shift become obvious and the wholespectra becomes broad. Previous NMR studies of severaloxynitride and nitride compounds indicated that the 29Sispectra of the [SiOxNy] coordination structures showed alarger chemical shift with an increase in the fraction ofnitrogen involved.53–55) According to those previous stud-ies, the increase in Neff with the increase in )AM can beattributed to the increase in both the population of[SiOxNy] and y in [SiOxNy].On the other hand, the profile of 27Al NMR spectrashows less obvious changes with Neff, as can be observedin Fig. 3(b). As the peak at 52–57 ppm has been identifiedas tetrahedral [AlO4] coordination56,57) most of the alumi-num is presumed to occupy tetrahedral sites coordinatedJournal of the Ceramic Society of Japan (2026), Advance Publication by J-stage JCS-Japanwith the four oxide ions. It is also notable that octahedral[AlO6] and hexahedral [AlO5] are rarely formed, asaluminum-oxygen polyhedra with higher coordinationnumbers should cause peaks in the range of lower chemi-cal shifts.58)The XPS and NMR spectra of the samples are consis-tent. In fact, the Al2p XPS spectra and 27Al NMR spectradid not show an obvious shift resulting from the increasein Neff, whereas the Si2p core-level spectra and 29Si NMRspectra showed a shift with an increase in Neff. Hence, itis evident that the chemical state of silicon is sensitive toNeff, whereas that of aluminum rarely responds to changesin Neff. This tendency has been commonly observed inprevious studies on oxynitrides, including silicon andaluminum.59–61)Figure 4(a) presents the FT-IR spectra of the Si–O- andSi–N-related modes. As shown in this figure, the spectraappear to be composed of several peaks, denoted by bandsI–V. Peaks centered at approximately 1200 cm¹1 (band I)and 1080 cm¹1 (band II) were assigned to the asymmetricstretching mode of –Si–O–Si bonds,62,63) and the peakcentered at 800 cm¹1 (band V) was also identified as asymmetric bond stretching mode associated with the –Si–O–Si-forming network structure in the glass.64) A compar-ison between the FT-IR profiles of PwO and silica glassindicated that bands II and V in the current samples pres-ented obvious broadening owing to the addition of Sr andAl. The peak denoted by band-III at approximately 960cm¹1, assignable to the stretching vibration of Al–O–Sibonds,65) could appear in the current samples. However,the presence of band III was not evident because of thebroadening of band II. In addition to the aforementionedvibrational modes, an additional mode, band IV, was ob-served in these spectra. According to the literature, thepeak at approximately 900–950 cm¹1, that is, band IV, canbe assigned to the Si–N stretching mode.66,67) As the inten-sity of band-IV seems consistent with Neff, the develop-ment of the spectral intensity at approximately 900–950cm¹1 can be reasonably attributed to the increase in thepopulation of the Si–N bond. These results are consistentwith the results of the XPS and NMR measurements,which indicate that nitride ions preferentially form Si–Nbonds. Apparently, the broadening of the NMR signal dueto the appearance of the shoulder on the small chemicalshift side, the shift of the Si2p XPS peak to a lower Ebdirection, and the shift of the FT-IR peak from 1080 cm¹1to a smaller wavenumber direction arise from the sameorigin, the increase in nitrogen sitting as the nearestneighbor to silicon.In addition to the Si–O and Si–N related modes, a peakat approximately 1400 cm¹1 was found in the FT-IR spec-tra of P500 and P500SR as shown in Figs. 4(a) and 4(b).Furthermore, broad but obvious peaks at approximately3200–3600 cm¹1 were observed in the FT-IR spectra ofP500 and P500SR, as shown in Fig. 4(b). These peaks canbe assigned to hydrogen-related vibrational modes68,69)originating from C–H, N–H, and O–H, as indicated in thefigure. As mentioned, actual oxygen partial pressure (Po2)100 0 -100 -200 -300Intensity( a.u.)29Si chemical shift (ppm)* *(a)P020 P100 P300P500 P500SR[SiO4][SiNO3][SiN2O2][SiN3O][SiN4]27Al chemical shift (ppm)150 100 50 0 -50Intensity( a.u.)* *(b) [AlO4]Fig. 3. NMR signal of 29Si (a) and 27Al (b) of the synthesizedpowder prepared under different ammonolysis conditions. Thepowder was identified by Pn, where n is the flow rate of ammonia(cm3/min) during ammonolysis.-CH-OH15002000250030003500Absorbance(a.u.) -NH(b) DiamondP020 P100 P300P500 P500SRWavenumber (cm-1)1500 1000 500Absorbance(a.u.)(a)-CHIII IIIIVVPwOSilica glassWavenumber (cm-1)Fig. 4. FT-IR spectra of the powder samples after ammonolysiswith different flow rates of the ammonia stream. The sampleidentifier, Pn, indicates the powder prepared by ammonolysiswith flow rate of n cm3/min. Lower wavenumber range measuredwith the conventional transmission mode (a), and higher wave-number range measured with the ATR mode with a diamondprism (b).Liu et al.: Tracing nitridation reaction toward efficient production of oxynitride glasses as hosts for bright luminescence centersJCS-Japanduring ammonolysis was relatively high for smaller )AMconditions, for example, )AM = 100. As shown in Fig. S2,the concentrations of the residual hydrocarbons in P100and P300 were reduced by combustion. Meanwhile, verylow Po2 achieved by increased )AM should be a reason forthe significant amount of residual hydrocarbon. The pres-ence of these C–H, N–H, and O–H-related modes suggeststhe modification of the glass structure by the formation ofhydrogen-terminated structures. This is consistent with theXRF results obtained by assuming the splitting of the glassnetwork structure, as expressed by Eqs. (3) and (4). Nota-bly, our previous study30) also indicated the breakage ofthe glass network structure by excess nitrogen.Figure 5 shows the TDS spectra of the selected sam-ples, where the recorded signals were labeled by the mass/charge (m/Z) ratio as the output from the quadrupole massspectrometer. First, we focused on the behaviors of m/Z =17 (violet) and 18 (red). Assuming that these two signalsare dominated by 16O1H and 1H216O, they must be parallelto each other. As shown in Fig. 5(a), for P100, the profilesof m/Z = 17 and 18 were parallel to each other, and theirintensity ratio was close to 1:2 over the entire tempera-ture range. This indicated that these two signals could beattributed to water desorption. However, in Fig. 5(c) forP500, the profiles of m/Z = 17 and 18 are not parallel toeach other, and the intensity of m/Z = 17 is higher thanthat of m/Z = 18 in most temperature ranges. This sug-gested that the signal for m/Z = 17 from P500 was notdominated by 1H16O.By contrast, the profiles of m/Z = 14 (yellow) and 15(green) provide hints for identifying the TDS profiles. Forthe P100 powder shown in Figs. 5(a) and 5(b), the spectralfeatures of m/Z = 14 and 15, such as the peaks denoted byarrows, are also observed in the signal for m/Z = 13, 29,and 31. Because m/Z = 13, 29, and 31 can be attributed to12C1H, 12C21H5, and 12C1H316O, respectively, it is pre-sumed that the signals for m/Z = 14 and 15 in Fig. 5(a)are dominated by hydrocarbons, that is, CH2 and CH3. Bycontrast, in the P500 shown in Figs. 5(c) and 5(d), thebehavior of m/Z = 14 and 15 looks similar to that ofm/Z = 17 but dissimilar to that in m/Z = 13, 29, and 31.This indicated that the signals for m/Z = 14 and 15 inFig. 5(c) were not dominated by hydrocarbons. It is rea-sonable to conclude that the signals for m/Z = 14, 15, and17 from P500 originate from 14N, 14N1H, and 14N1H3,respectively. Hence, both P100 and P500 contain –CHngroups, and P500 specifically contains –NHm groups.The intensities of the nitrogen-related signals (m/Z =14, 15 and 17) in Fig. 5(c) were more than one order ofmagnitude higher than those of P100 over the entire tem-perature range, but did not show any obvious peaks (tem-perature dependency). As most of the nitrogen in P500forms Si–N bonds, as indicated by the FT-IR and NMRmeasurements, the very high nitrogen desorption signalcan be attributed to the breaking of Si–N bonds. As theintensity of the nitrogen-related signals from P500 wasrelatively high, even at relatively low temperatures, it ispresumed that some of the Si–N bonds formed in P500 arethermally unstable. As nitrogen-related desorption contin-ued at higher temperatures and no obvious peaks wereobserved, it also indicates that the nitrogen desorption ratefrom P500 was limited by the rate of out-diffusion in P500.It is likely that various types of Si–N bonds exist in P500.Subsequent XPS measurements of the P500SR powderafter the TDS measurements showed that the nitrogenconcentration decreased during the TDS measurements.The Neff value of P500SR after TDS measurement wasdetermined to be 34% by XPS, whereas that before TDSmeasurement was 38%. Although obvious desorption ofnitrogen-related fragments was observed in the TDS pro-file, as described above, the amount of desorption duringTDS was limited. A characteristic feature of the powder isthat an excess of anions remained in the powder after TDS.The TG/DTA measurements (not shown) also indicatedweight loss of P500 and P500SR. The magnitude of theweight loss was only a few percentages of the initialweight. Such a weight change was not sufficiently large tosuppress excess anions in the powders. Hence, it wasconfirmed that excess anions remained even after thermalanalysis at high temperatures. The quantity of residualorganic components is also an important parameter forcharacterizing the gel ammonolysis process. Although the131314141515161628 293117181718181828293110-1110-1010-910-1210-1110-1010-910-1110-1010-91328 29 3114 15 16 1718m/ZSignal intensity (a.u.)(b) P100200 400 600 80010-12(d) P500(c) P500(a) P100Temperature (ºC)10-1110-1010-9Fig. 5. Thermal desorption spectra of P100 (a and b) and P500(c and d). The powder was identified by Pn, where n (cm3/min) isthe flow rate of ammonia during ammonolysis. The arrows areguides to eyes.Journal of the Ceramic Society of Japan (2026), Advance Publication by J-stage JCS-Japandesorption of hydrocarbons during TDS measurementsis evident and the FT-IR signal originating from hydro-carbons was observed, we do not have sufficient results todiscuss the amount of residual hydrocarbons with reason-able accuracy.Summarizing the above presented results, it is evidentthat nitrogen was successfully incorporated into the pow-der by ammonolysis and that P100 and P300 were likelystable oxynitride glasses. Although TDS indicated thepresence of reaction residues in these three powders, themagnitude of the weight change measured by TG wasnegligible, and the 29Si NMR signal was relatively sharpfor these three powders. The P020 powder also appearedto be a stable oxynitride, but crystallization during ammo-nolysis and its low Neff are not very suitable for quanti-tative discussion. Meanwhile, a tread-off in ammonolysisprocess was highlighted in the sample prepared with high)AM. Incorporation of nitride ion into the gel was evi-dently enhanced by employing high )AM conditions, but itwas indicated that obvious anion-excess was achieved byemploying high )AM.The very high Neff of P500SR, which was prepared witha very slow ramping rate during ammonolysis, must alsobe highlighted. In our previous study,30) the effect of theammonolysis duration on oxynitride glass formation wasinvestigated, and it was found that a shorter ammonolysisperiod resulted in a higher Neff. In the present study, theslow ramping rate at the initiation of ammonolysis dras-tically increased Neff. Those results indicate that the incor-poration of nitride ions was a rapid reaction in the gel.However, it appears that nitride ions introduced in theearly stages of ammonolysis inhibit the formation of stableoxynitride glass structures. We speculate the followingreaction:½­Si­ðNHmÞ� þ ½ðNHnÞ­Si­�! ½­Si­N­Si­� þ ¤NHx: ð5ÞThe reaction expressed by Eq. (5) assumes that struc-tures terminated by the N–H groups, such as –Si–(NH2),were formed at the early stage of nitridation, and that theN–H groups are decomposed to form bridging nitrogen(–Si–N–Si–) for polymerization at higher temperatures.The results of XRF measurement for P500SR (Table S2)indicated presence of hydrogen in the powder after ammo-nolysis as mentioned above. We considered very slowdiffusivity in nitrides and oxynitrides compared to oxides.That should be the case for bulk solid glasses. The currentstudy applies ammonolysis to the gel before polymeriza-tion. This is a key for understanding the very quick in-crease of nitrogen concentration in the present study. Themicrostructure of the gel, very fine grains with relativelyhigh specific surface area (Fig. S4 in the Supporting Infor-mation) should also be a reason for the very fast nitrogeninsertion behavior. Those should be the causes for forma-tion of the structure terminated by –Si–(NHn) at the earlystage of ammonolysis.As the spectral profile of XPS [Fig. 2(d)], NMR[Fig. 3(a)], and FT-IR [Fig. 4(a)] indicated monotonicalincrease of Si–N bond density with Neff, but it seems thosemethod did not distinguish if Si–N bonds were involvedin –Si–(NHn) or –Si–N–Si–. Hence, it is a very importantindication that not only XPS, NMR, and FT-IR but alsosome additional characterizations are needed to character-ize oxynitride glasses in detail.From these considerations, we speculate that a suffi-ciently long thermal relaxation time is necessary for com-pleting polymerization to form stable oxynitride glassstructures. Indeed, Eq. (5) assumes the presence of twonear-terminal structures. When a single terminal structureexists alone and is surrounded by relatively stable bridgingstructures, the simple removal of the –NHn groups resultsin the formation of dangling bonds. It is likely that a large-scale reconstruction of the glass structure is necessary toremove the terminal structures when the distance betweenthem is relatively large. Hence, the present results suggestthat a nitridation strategy is important for the efficientproduction of oxynitride glasses with highly stable struc-tures by polymerization, as it seems that strong nitridationresults in the formation of many terminal structures, suchas Si–NHn, which are not easily removed once formed.Hence, the introduction of nitride ions without the forma-tion of a metastable terminal structure is essential for theefficient production of oxynitride glasses by ammonolysis.In terms of process optimization for efficient oxynitrideformation, it is useful to use a gas mixture for ammonol-ysis. As can be observed in Fig. S2, the decompositionrate of ammonia strongly depends on )AM. Ammoniacracking is known to be a slow reaction.70) As previouslymentioned, it is possible that the residual oxygen gas in thefurnace plays an important role in nitridation and polymer-ization. The hydrogen gas concentrations are not shown inFig. S2 because of instrumental limitations. However, thepresence of hydrogen due to the decomposition of ammo-nia must affect the formation of hydrogen-terminatedstructures. Hence, consideration of the dynamics of ammo-nia cracking, not only due to the reacting power but alsodue to furnace materials, is needed for designing efficientnitridation without forming hydrogen-terminated struc-tures and for the polymerization of the oxynitride glassstructures.Regarding the optical properties, all the powdersshowed broadband green luminescence upon excitationwith ultraviolet light after ammonolysis, as shown inFig. 6(a). The observed PL and PLE profiles indicate thatdivalent Eu ions (Eu2+) act as luminescence centers. Inter-estingly, the PLE profile of the powder not prepared byammonolysis but by heating in a hydrogen atmosphereshowed a PLE profile almost equal to that of the othernitrogen-containing powders. However, the shape of its PLspectrum was different from that of the other powders. Infact, the PL bandwidth of PwH was much narrower thanthat of the powder after ammonolysis.These results provide two indications: the excited statesof Eu2+ ion in oxide glass that contribute to radiativerecombination are very similar to those in oxynitride glass,and the radiative recombination process of Eu2+ in oxideLiu et al.: Tracing nitridation reaction toward efficient production of oxynitride glasses as hosts for bright luminescence centersJCS-Japanglass is different from that in oxynitride glasses. As the tailof the PL spectra at the larger wavelength side is mostlycaused by interactions with the vibrational mode,71–73) thedifference in the PL bandwidth between the PwH andoxynitride glasses was likely due to changes in the coor-dination structure around Eu2+. Although the Neff in P020was considerably lower than that in P300, the PL peakwavelengths and bandwidths of P020 and P300 werenearly identical. This result indicated that the coordinationstructure of the radiative Eu2+ center was highly specific.Because of the dissimilarity between the PL profiles ofPwH and P020, the radiative Eu2+ center formed in P020should have different coordination structures. The PL andPLE spectra of P020 and P300 were similar, although theXPS, NMR, and FT-IR spectra of the two powders exhib-ited dissimilar profiles. It means that Si–N bonds werechanged according to )AM but the luminescence profileremained unchanged. These results indicate that only Euions with specific coordination structures involving nitrideions become radiative Eu2+ centers. Notably, the PL pro-file of P500SR was much broader than those of the otheroxynitride glass powders. This may indicate that the radi-ative recombination process at the Eu2+ ion was affectedby the presence of thermal structures, such as –Si–N–H.Figure 6(b) shows the IQE of the Eu2+ luminescence inthe powders prepared under different ammonolysis con-ditions. Notably, the IQE seems to reach a maximum in theNeff range of 10–20%. As mentioned above, the chemicalstate of Si exhibits systematic changes corresponding toNeff. However, the luminescence efficiency exhibits a dif-ferent trend from that of Neff. Hence, the degree of nitri-dation and modification of the chemical state of Si are notsimple measures for enhancing the radiative recombina-tion at the Eu2+ center. In addition, the IQE of P500SRwas significantly lower than those of the other powders.Furthermore, the PL spectrum of P500SR was significantlybroader than those of the other powders. It can also bepresumed that a high concentration of –Si–NHm causesluminescence quenching of the Eu2+ centers.Before concluding the discussion, the prospective strat-egy for efficient luminescence with Eu2+ centers should bementioned. Because the maximum IQE in the previousstudy30) was obtained after long-term thermal treatment ofthe film with a relatively low Neff, we can expect that alonger ammonolysis period with suitable nitridation activ-ity, likely a condition for P100 or P300, may enable us toobtain a higher IQE of Eu2+ luminescence.4. ConclusionWe examined the synthesis of Si–Al–Sr–Eu–O–N glasspowder using a sol–gel process followed by ammonolysis.It was indicated that a relatively high ammonia flow rate,)AM, in ammonolysis and ammonolysis at a relatively lowtemperature induced higher Neff in the resultant glasses.The results of NMR measurements indicated that thenitride ion incorporated in the oxynitride glass was thenearest neighbor to silicon, and XPS results suggested thatthe chemical state of silicon is highly sensitive to Neff.However, the chemical composition analyzed by XRFshowed an obvious anion excess, indicating the formationof terminal structures in the form of –Si–CHn and –Si–NHm. The presence of those terminal structures was con-firmed by FT-IR measurements. Subsequent thermal anal-ysis indicated that glass with a high Neff is thermally unsta-ble and undergoes thermal decomposition, and that com-plete removal of these terminal structures takes time. Theobtained results suggest a strategy for enhancing the Eu2+luminescence in oxynitride glass, that is, the removal ofexcess anions, which is likely due to the presence of N–Hcomponents. For the utilization of oxynitride glasses inpractical applications, it has been suggested that a nitrida-tion process contributing to polymerization involvingnitride ions without the formation of terminal structuressuch as –Si–NHm, is highly desirable. We expect that fur-ther consideration for controlling the nitridation atmos-phere, such as the use of a gas mixture for simultaneouscontrol of the oxygen partial pressure and nitridation activ-ity, will enable the efficient production of highly function-alized oxynitride glasses.Acknowledgements Part of this study will be incorpo-rated into the Ph.D. thesis of LX. Part of this study wasconducted at the Tokodai Institute for Elemental Strategieswith support from Grant Number JPMXP0112101001 and theData-Driven Materials Research Institute for Electronics withsupport from JPMXP1122683430. The XPS and NMR mea-surements were performed at Research Network and FacilityServices Division, NIMS, under support from the ARIM pro-gram, Nos. JPMXP1225NM5319 and JPMXP1223NM5260,respectively.300 400 500 600 700NormalizedintensityWavelength (nm)(a)PwHP500SRPLEPL0 10 20 30 40 5001020304050IQE(%)Neff (%)PwHP300P500P500SRP100P020(b)PwHP300 P500 P500SRP100P020Fig. 6. Photoluminescence (PL) excited by UV light (­ex =300 nm), and photoluminescence excitation (PLE) spectra moni-tored at luminescence intensity maximum (a). Internal quantumefficiency as a function of the effective nitrogen concentration(Neff) (b).Journal of the Ceramic Society of Japan (2026), Advance Publication by J-stage JCS-JapanReferences1) K. M. Youssef, A. J. Zaddach, C. Niu, D. L. Irving andC. C. Koch, Mater. Res. Lett. 3, 95 (2015). https://doi.org/10.1080/21663831.2014.985855.2) D. Kumar, R. Seetharam and K. Ponappa, J. Alloy.Compd. 972, 172732 (2024). https://doi.org/10.1016/j.jallcom.2023.172732.3) K. Maeda, F. Takeiri, G. Kobayashi, S. Matsuishi, H.Ogino, S. Ida, T. Mori, Y. Uchimoto, S. Tanabe, T.Hasegawa, N. Imanaka and H. Kageyama, B. Chem.Soc. Jpn. 95, 26 (2022). https://doi.org/10.1246/bcsj.20210351.4) J. J. Xu and K. Wu, Coordin. Chem. Rev. 486, 215139(2023). https://doi.org/10.1016/j.ccr.2023.215139.5) X. Chen and K. M. Ok, Chem. Sci. 13, 3942 (2022).https://doi.org/10.1039/D1SC07121A.6) Y.-F. Shi, W.-B. Wei, X.-T. Wu, H. Lin and Q.-L. Zhu,Dalton T. 50, 4112 (2021). https://doi.org/10.1039/D1DT00222H.7) X. Chen and K. M. Ok, Chem. Sci. 13, 3942 (2022).https://doi.org/10.1039/D1SC07121A.8) R. Xie, N. Hirosaki, M. Mitomo, Y. Yamamoto, T.Suehiro and N. Ohashi, J. Am. Ceram. Soc. 87, 1368(2004). https://doi.org/10.1111/j.1151-2916.2004.tb07738.x.9) N. Hirosaki, R.-J. Xie, K. Kimoto, T. Sekiguchi, Y.Yamamoto, T. Suehiro and M. Mitomo, Appl. Phys.Lett. 86, 1 (2005). https://doi.org/10.1063/1.1935027.10) D. Wen, H. Kato and M. Kakihana, ACS Sustain.Chem. Eng. 8, 12286 (2020). https://doi.org/10.1021/acssuschemeng.0c04389.11) D. Wen, H. Liu, Z. Ma, L. Zhou, J. Li, Y. Guo, Q. Zeng,P. A. Tanner and M. Wu, Angew. Chem. Int. Edit.62, e202307868 (2023). https://doi.org/10.1002/anie.202307868.12) J. I. Yamaura, S. Maki, T. Honda, Y. Matsui, A.Noviyanto, T. Otomo, H. Abe, Y. Murakami and N.Ohashi, Chem. Commun. 56, 1385 (2020). https://doi.org/10.1039/c9cc07029j.13) F. Oehler, H. T. Langhammer and S. G. Ebbinghaus,J. Eur. Ceram. Soc. 37, 2129 (2017). https://doi.org/10.1016/j.jeurceramsoc.2016.12.030.14) A. Hosono, Y. Masubuchi, T. Endo and S. Kikkawa,Dalton T. 46, 16837 (2017). https://doi.org/10.1039/C7DT03765A.15) L. Mao, X. Cai, H. Gao, X. Diao and J. Zhang,Nano Energy 39, 172 (2017). https://doi.org/10.1016/j.nanoen.2017.06.026.16) H. Banno, S. Funahashi, T. Asaka, N. Hirosaki and K.Fukuda, J. Solid State Chem. 230, 149 (2015). https://doi.org/10.1016/j.jssc.2015.06.043.17) J. Podwórny, A. Pawełek and J. Czechowski,Powder Diffr. 32, S2 (2017). https://doi.org/10.1017/S088571561700001X.18) K. Fukui, S. Iimura, A. Iskandarov, T. Tada and H.Hosono, J. Am. Chem. Soc. 144, 1523 (2022). https://doi.org/10.1021/jacs.1c11353.19) S. Yamamoto, Y. Ohashi, Y. Masubuchi, T. Takeda, T.Motohashi and S. Kikkawa, J. Alloy. Compd. 482, 160(2009). https://doi.org/10.1016/j.jallcom.2009.03.151.20) J. Li, Y. Jiang, Z. Zhang, M. Tsuji, M. Miyazaki, M.Kitano and H. Hosono, Adv. Energy Mater. 13, 2302424(2023). https://doi.org/10.1002/aenm.202302424.21) A. Noviyanto, T. Nishimura, M. Kitano and N. Ohashi,J. Eur. Ceram. Soc. 36, 4083 (2016). https://doi.org/10.1016/j.jeurceramsoc.2016.06.031.22) M. Namba, H. Takatsu, H.-B. Li, K. Murayama, R.Terada, Q. Yang, T. Terashima, H. Ohta and H.Kageyama, Chem. Mater. 36, 2076 (2024). https://doi.org/10.1021/acs.chemmater.3c03178.23) J. Lee, Q. Zhang and F. Saito, Ind. Eng. Chem. Res. 40,4785 (2001). https://doi.org/10.1021/ie010086k.24) N. Arumugam, A. Hönnerscheid and M. Jansen, Z.Anorg. Allg. Chem. 629, 939 (2003). https://doi.org/10.1002/zaac.200200452.25) Y. Hinuma, A. Grüneis, G. Kresse and F. Oba, Phys.Rev. B 90, 155405 (2014). https://doi.org/10.1103/PhysRevB.90.155405.26) A. Duval, P. Houizot and T. Rouxel, J. Am. Ceram. Soc.106, 1611 (2023). https://doi.org/10.2109/jcersj2.18097.27) H. Hyuga, J. Ceram. Soc. Jpn. 126, 968 (2018). https://doi.org/10.2109/jcersj2.18097.28) K. Watari, H. J. Hwang, M. Toriyama and S. Kanzaki,J. Mater. Res. 14, 1409 (1999). https://doi.org/10.1557/JMR.1999.0191.29) S. Watanabe, Y. Osawa, S. Machida, K. Katsumata, A.Yasumori, K. Takahashi, K. Deguchi, S. Ohki and H.Segawa, J. Non-Cryst. Solids 573, 121107 (2021).https://doi.org/10.1016/j.jnoncrysol.2021.121107.30) X. Liu, T. Ohsawa, J. Xu, M. Yanagida, K. Takahashi, T.Takeda, T. Kishi, T. Yano, H. Segawa and N. Ohashi,J. Mater. Chem. C 13, 18286 (2025). https://doi.org/10.1039/D5TC01527H.31) H. Segawa, N. A. Wójcik, K. Takahashi, T. Takeda andS. Ali, J. Am. Ceram. Soc. 107, 2930 (2024). https://doi.org/10.1111/JACE.19615.32) H. Segawa and N. Hirosaki, Ceram. Int. 44, 4783(2018). https://doi.org/10.1016/j.ceramint.2017.12.063.33) S. Li, L. Wang, D. Tang, Y. Cho, X. Liu, X. Zhou, L.Lu, L. Zhang, T. Takeda, N. Hirosaki and R.-J. Xie,Chem. Mater. 30, 494 (2018). https://doi.org/10.1021/acs.chemmater.7b04605.34) R. E. Loehman, J. Non-Cryst. Solids 42, 433 (1980).https://doi.org/10.1016/0022-3093(80)90042-3.35) H. Schmidt, M. Gupta and M. Bruns, Phys. Rev. Lett. 96,055901 (2006). https://doi.org/10.1103/PhysRevLett.96.055901.36) H. Haneda, T. Ohgaki, I. Sakaguchi, H. Ryoken, N.Ohashi and A. Yasumori, Appl. Surf. Sci. 252, 7265(2006). https://doi.org/10.1016/j.apsusc.2006.02.179.37) A. Hosono, Y. Masubuchi and S. Kikkawa, Ceram. Int.43, 2737 (2017). https://doi.org/10.1016/j.ceramint.2016.11.101.38) Y.-R. Zhang, T. Motohashi, Y. Masubuchi and S.Kikkawa, J. Eur. Ceram. Soc. 32, 1269 (2012).https://doi.org/10.1016/j.jeurceramsoc.2011.12.001.39) S. Kikkawa, S. Sun, Y. Masubuchi, Y. Nagamine and T.Shibahara, Chem. Mater. 28, 1312 (2016). https://doi.org/10.1021/acs.chemmater.5b04149.40) A. Hosono, S.-K. Sun, Y. Masubuchi and S. Kikkawa,J. Eur. Ceram. Soc. 36, 3341 (2016). https://doi.org/10.1016/j.jeurceramsoc.2016.05.006.41) Y. Masubuchi, S. Ohtaki, K. Fujii, M. Yashima, M.Higuchi and S. Kikkawa, J. Eur. Ceram. Soc. 40, 6288(2020). https://doi.org/10.1016/j.jeurceramsoc.2019.10.004.Liu et al.: Tracing nitridation reaction toward efficient production of oxynitride glasses as hosts for bright luminescence centersJCS-Japanhttps://doi.org/10.1080/21663831.2014.985855https://doi.org/10.1080/21663831.2014.985855https://doi.org/10.1016/j.jallcom.2023.172732https://doi.org/10.1016/j.jallcom.2023.172732https://doi.org/10.1246/bcsj.20210351https://doi.org/10.1246/bcsj.20210351https://doi.org/10.1016/j.ccr.2023.215139https://doi.org/10.1039/D1SC07121Ahttps://doi.org/10.1039/D1DT00222Hhttps://doi.org/10.1039/D1DT00222Hhttps://doi.org/10.1039/D1SC07121Ahttps://doi.org/10.1111/j.1151-2916.2004.tb07738.xhttps://doi.org/10.1111/j.1151-2916.2004.tb07738.xhttps://doi.org/10.1063/1.1935027https://doi.org/10.1021/acssuschemeng.0c04389https://doi.org/10.1021/acssuschemeng.0c04389https://doi.org/10.1002/anie.202307868https://doi.org/10.1002/anie.202307868https://doi.org/10.1039/c9cc07029jhttps://doi.org/10.1039/c9cc07029jhttps://doi.org/10.1016/j.jeurceramsoc.2016.12.030https://doi.org/10.1016/j.jeurceramsoc.2016.12.030https://doi.org/10.1039/C7DT03765Ahttps://doi.org/10.1039/C7DT03765Ahttps://doi.org/10.1016/j.nanoen.2017.06.026https://doi.org/10.1016/j.nanoen.2017.06.026https://doi.org/10.1016/j.jssc.2015.06.043https://doi.org/10.1016/j.jssc.2015.06.043https://doi.org/10.1017/S088571561700001Xhttps://doi.org/10.1017/S088571561700001Xhttps://doi.org/10.1021/jacs.1c11353https://doi.org/10.1021/jacs.1c11353https://doi.org/10.1016/j.jallcom.2009.03.151https://doi.org/10.1002/aenm.202302424https://doi.org/10.1016/j.jeurceramsoc.2016.06.031https://doi.org/10.1016/j.jeurceramsoc.2016.06.031https://doi.org/10.1021/acs.chemmater.3c03178https://doi.org/10.1021/acs.chemmater.3c03178https://doi.org/10.1021/ie010086khttps://doi.org/10.1002/zaac.200200452https://doi.org/10.1002/zaac.200200452https://doi.org/10.1103/PhysRevB.90.155405https://doi.org/10.1103/PhysRevB.90.155405https://doi.org/10.2109/jcersj2.18097https://doi.org/10.2109/jcersj2.18097https://doi.org/10.2109/jcersj2.18097https://doi.org/10.1557/JMR.1999.0191https://doi.org/10.1557/JMR.1999.0191https://doi.org/10.1016/j.jnoncrysol.2021.121107https://doi.org/10.1039/D5TC01527Hhttps://doi.org/10.1039/D5TC01527Hhttps://doi.org/10.1111/JACE.19615https://doi.org/10.1111/JACE.19615https://doi.org/10.1016/j.ceramint.2017.12.063https://doi.org/10.1021/acs.chemmater.7b04605https://doi.org/10.1021/acs.chemmater.7b04605https://doi.org/10.1016/0022-3093(80)90042-3https://doi.org/10.1103/PhysRevLett.96.055901https://doi.org/10.1103/PhysRevLett.96.055901https://doi.org/10.1016/j.apsusc.2006.02.179https://doi.org/10.1016/j.ceramint.2016.11.101https://doi.org/10.1016/j.ceramint.2016.11.101https://doi.org/10.1016/j.jeurceramsoc.2011.12.001https://doi.org/10.1021/acs.chemmater.5b04149https://doi.org/10.1021/acs.chemmater.5b04149https://doi.org/10.1016/j.jeurceramsoc.2016.05.006https://doi.org/10.1016/j.jeurceramsoc.2016.05.006https://doi.org/10.1016/j.jeurceramsoc.2019.10.004https://doi.org/10.1016/j.jeurceramsoc.2019.10.00442) A. Hosono, M. Inoguchi, Y. Masubuchi, K. Murayama,M. Iha, M. Higuchi and S. Kikkawa, J. Eur. Ceram.Soc. 40, 2317 (2020). https://doi.org/10.1016/J.JEURCERAMSOC.2020.02.008.43) Y. Osawa, K. Iwasaki, T. Nakanishi, A. Yasumori, Y.Matsui, T. Nishimura, T. Ohsawa, H. Segawa and N.Ohashi, J. Am. Ceram. Soc. 104, 4420 (2021). https://doi.org/10.1111/JACE.17836.44) R. K. Brow, C. G. Pantano and D. C. Boyd, J. Am.Ceram. Soc. 67, c72 (1984). https://doi.org/10.1111/J.1151-2916.1984.TB18834.X.45) K. Szaniawska, L. Murawski, R. Pastuszak, M.Walewski and G. Fantozzi, J. Non-Cryst. Solids286, 58 (2001). https://doi.org/10.1016/S0022-3093(01)00478-1.46) B. E. Warren, Chem. Rev. 26, 237 (1940). https://doi.org/10.1021/cr60084a007.47) H. W. Nesbitt, G. M. Bancroft, G. S. Henderson, R. Ho,K. N. Dalby, Y. Huang and Z. Yan, J. Non-Cryst.Solids 357, 170 (2011). https://doi.org/10.1016/J.JNONCRYSOL.2010.09.031.48) G. Pintori and E. Cattaruzza, Opt. Mater. 13, 100108(2022). https://doi.org/10.1016/J.OMX.2021.100108.49) J. F. Moulder and J. Chastain, “Handbook of X-rayPhotoelectron Spectroscopy: A Reference Book ofStandard Spectra for Identification and Interpretationof XPS Data”, Physical Electronics Division, Perkin-Elmer Corp. (1992).50) K. Kawashima, M. Hojamberdiev, O. Mabayoje, B. R.Wygant, K. Yubuta, C. B. Mullins, K. Domen and K.Teshima, CrystEngComm 19, 5532 (2017). https://doi.org/10.1039/C7CE00614D.51) B. Li, Y. Adachi, J. Li, H. Okushi, I. Sakaguchi, S.Ueda, H. Yoshikawa, Y. Yamashita, S. Senju, K.Kobayashi, M. Sumiya, H. Haneda and N. Ohashi,Appl. Phys. Lett. 98, 082101 (2011). https://doi.org/10.1063/1.3556440.52) S. Hirose, S. Ueda and N. Ohashi, J. Appl. Phys. 125,95301 (2019). https://doi.org/10.1063/1.5066574/156311.53) K. R. Carduner, R. O. Carter, M. E. Milberg and G. M.Crosbie, Anal. Chem. 59, 2794 (1987). https://doi.org/10.1021/ac00150a015.54) C. Cozzan, K. J. Griffith, G. Laurita, J. G. Hu, C. P.Grey and R. Seshadri, Inorg. Chem. 56, 2153 (2017).https://doi.org/10.1021/acs.inorgchem.6b02780.55) R. van Weeren, E. A. Leone, S. Curran, L. C. Klein andS. C. Danforth, J. Am. Ceram. Soc. 77, 2699 (1994).https://doi.org/10.1111/j.1151-2916.1994.tb04664.x.56) C. I. Merzbacher, B. L. Sherriff, J. S. Hartman and W. B.White, J. Non-Cryst. Solids 124, 194 (1990). https://doi.org/10.1016/0022-3093(90)90263-L.57) S. Watanabe, Y. Osawa, S. Machida, K. Katsumata, A.Yasumori, K. Takahashi, K. Deguchi, S. Ohki and H.Segawa, J. Non-Cryst. Solids 573, 121107 (2021).https://doi.org/10.1016/J.JNONCRYSOL.2021.121107.58) Z. Zhao, D. Xiao, K. Chen, R. Wang, L. Liang, Z. Liu, I.Hung, Z. Gan and G. Hou, ACS Central Sci. 8, 795(2022). https://doi.org/10.1021/acscentsci.1c01497.59) S. Sakka, J. Non-Cryst. Solids 181, 215 (1995). https://doi.org/10.1016/S0022-3093(94)00514-1.60) N. Hirosaki, C. Kocer, S. Ogata and K. Tatsumi, Phys.Rev. B 71, 104105 (2005). https://doi.org/10.1103/PhysRevB.71.104105.61) E. Leonova, A. S. Hakeem, K. Jansson, B. Stevensson,Z. Shen, J. Grins, S. Esmaeilzadeh and M. Edén, J. Non-Cryst. Solids 354, 49 (2008). https://doi.org/10.1016/j.jnoncrysol.2007.07.027.62) A. Bertoluzza, C. Fagnano, M. Antonietta Morelli, V.Gottardi and M. Guglielmi, J. Non-Cryst. Solids 48, 117(1982). https://doi.org/10.1016/0022-3093(82)90250-2.63) H. Aguiar, J. Serra, P. González and B. León, J. Non-Cryst. Solids 355, 475 (2009). https://doi.org/10.1016/J.JNONCRYSOL.2009.01.010.64) A. Bertoluzza, C. Fagnano, M. Antonietta Morelli, V.Gottardi and M. Guglielmi, J. Non-Cryst. Solids 48, 117(1982). https://doi.org/10.1016/0022-3093(82)90250-2.65) B. T. Poe, P. F. McMillan, C. A. Angell and R. K. Sato,Chem. Geol. 96, 333 (1992). https://doi.org/10.1016/0009-2541(92)90063-B.66) H. Ono, T. Ikarashi, Y. Miura, E. Hasegawa, K. Andoand T. Kitano, Appl. Phys. Lett. 74, 203 (1999). https://doi.org/10.1063/1.123293.67) M. Klanjšek Gunde and M. Maček, Phys. Status SolidiA 183, 439 (2001). https://doi.org/10.1002/1521-396X(200102)183:2<439::AID-PSSA439>3.0.CO;2-B.68) K. Kamiya, M. Ohya and T. Yoko, J. Non-Cryst.Solids 83, 208 (1986). https://doi.org/10.1016/0022-3093(86)90068-2.69) J. Szépvölgyi, I. Mohai and J. Gubicza, J. Mater. Chem.11, 859 (2001). https://doi.org/10.1039/b008429h.70) H. Mizoguchi, Y. Osawa, M. Sasase, N. Ohashi, M.Kitano and H. Hosono, J. Phys. Chem. Lett. 14,9516 (2023). https://doi.org/10.1021/ACS.JPCLETT.3C02446.71) J. Bouquiaux, S. Poncé, Y. Jia, A. Miglio, M. Mikamiand X. Gonze, Chem. Mater. 35, 5353 (2023). https://doi.org/10.1021/acs.chemmater.3c00537.72) R. Shafei, P. J. Strobel, P. J. Schmidt, D. Maganas, W.Schnick and F. Neese, Phys. Chem. Chem. Phys. 26,6277 (2024). https://doi.org/10.1039/D3CP06039J.73) R. Shafei, P. J. Strobel, P. J. Schmidt, D. Maganas, W.Schnick and F. Neese, J. Phys. Chem. C 129, 1495(2025). https://doi.org/10.1021/acs.jpcc.4c06912.Journal of the Ceramic Society of Japan (2026), Advance Publication by J-stage JCS-Japanhttps://doi.org/10.1016/J.JEURCERAMSOC.2020.02.008https://doi.org/10.1016/J.JEURCERAMSOC.2020.02.008https://doi.org/10.1111/JACE.17836https://doi.org/10.1111/JACE.17836https://doi.org/10.1111/J.1151-2916.1984.TB18834.Xhttps://doi.org/10.1111/J.1151-2916.1984.TB18834.Xhttps://doi.org/10.1016/S0022-3093(01)00478-1https://doi.org/10.1016/S0022-3093(01)00478-1https://doi.org/10.1021/cr60084a007https://doi.org/10.1021/cr60084a007https://doi.org/10.1016/J.JNONCRYSOL.2010.09.031https://doi.org/10.1016/J.JNONCRYSOL.2010.09.031https://doi.org/10.1016/J.OMX.2021.100108https://doi.org/10.1039/C7CE00614Dhttps://doi.org/10.1039/C7CE00614Dhttps://doi.org/10.1063/1.3556440https://doi.org/10.1063/1.3556440https://doi.org/10.1063/1.5066574/156311https://doi.org/10.1063/1.5066574/156311https://doi.org/10.1021/ac00150a015https://doi.org/10.1021/ac00150a015https://doi.org/10.1021/acs.inorgchem.6b02780https://doi.org/10.1111/j.1151-2916.1994.tb04664.xhttps://doi.org/10.1016/0022-3093(90)90263-Lhttps://doi.org/10.1016/0022-3093(90)90263-Lhttps://doi.org/10.1016/J.JNONCRYSOL.2021.121107https://doi.org/10.1021/acscentsci.1c01497https://doi.org/10.1016/S0022-3093(94)00514-1https://doi.org/10.1016/S0022-3093(94)00514-1https://doi.org/10.1103/PhysRevB.71.104105https://doi.org/10.1103/PhysRevB.71.104105https://doi.org/10.1016/j.jnoncrysol.2007.07.027https://doi.org/10.1016/j.jnoncrysol.2007.07.027https://doi.org/10.1016/0022-3093(82)90250-2https://doi.org/10.1016/J.JNONCRYSOL.2009.01.010https://doi.org/10.1016/J.JNONCRYSOL.2009.01.010https://doi.org/10.1016/0022-3093(82)90250-2https://doi.org/10.1016/0009-2541(92)90063-Bhttps://doi.org/10.1016/0009-2541(92)90063-Bhttps://doi.org/10.1063/1.123293https://doi.org/10.1063/1.123293https://doi.org/10.1002/1521-396X(200102)183:2%3C439::AID-PSSA439%3E3.0.CO;2-Bhttps://doi.org/10.1002/1521-396X(200102)183:2%3C439::AID-PSSA439%3E3.0.CO;2-Bhttps://doi.org/10.1016/0022-3093(86)90068-2https://doi.org/10.1016/0022-3093(86)90068-2https://doi.org/10.1039/b008429hhttps://doi.org/10.1021/ACS.JPCLETT.3C02446https://doi.org/10.1021/ACS.JPCLETT.3C02446https://doi.org/10.1021/acs.chemmater.3c00537https://doi.org/10.1021/acs.chemmater.3c00537https://doi.org/10.1039/D3CP06039Jhttps://doi.org/10.1021/acs.jpcc.4c06912