Compositionally complex perovskite oxides: Discovering a new class of solid electrolytes with interface-enabled conductivity improvements

•Discovered compositionally complex perovskite oxides (CCPOs) as solid electrolytes•Compositionally complex designs can outperform conventional doping•Demonstrated grain-boundary-enabled conductivity improvements The recent emergence of high-entropy ceramics and a broader class of compositionally complex ceramics (CCCs) unlocks vast compositional spaces for materials discovery. This study proposes and demonstrates strategies for tailoring CCCs via a combination of non-equimolar compositional designs and control of interfaces and microstructures to discover a new class of compositionally complex solid electrolytes. Consequently, compositionally complex perovskite oxides with enhanced ionic conductivity are unearthed. Notably, this study discovers grain-boundary-enabled conductivity improvements, where compositional designs and processing are utilized to alter microstructures and interfacial structures to improve properties. Promising applications in all-solid-state batteries are envisioned. More generally, these materials discovery strategies can be utilized to discover, design, and tailor a broader range of CCCs for energy storage and many other applications. Compositionally complex ceramics (CCCs), including high-entropy ceramics, offer a vast, unexplored compositional space for materials discovery. Herein, we propose and demonstrate strategies for tailoring CCCs via a combination of non-equimolar compositional designs and control of grain boundaries (GBs) and microstructures. Using oxide solid electrolytes for all-solid-state batteries as an example, we have discovered a class of compositionally complex perovskite oxides (CCPOs) with improved lithium ionic conductivities beyond the limit of conventional doping. For example, we demonstrate that the ionic conductivity can be improved by >60% in (Li0.375Sr0.4375)(Ta0.375Nb0.375Zr0.125Hf0.125)O3-δ compared with the (Li0.375Sr0.4375)(Ta0.75Zr0.25)O3-δ (LSTZ) baseline. Furthermore, the ionic conductivity can be improved by another >70% via quenching, achieving >270% of the LSTZ. Notably, we demonstrate GB-enabled conductivity improvements via both promoting grain growth and altering GB structures through compositional designs and processing. In a broader perspective, this work suggests new routes for discovering and tailoring CCCs for energy storage and many other applications. Compositionally complex ceramics (CCCs), including high-entropy ceramics, offer a vast, unexplored compositional space for materials discovery. Herein, we propose and demonstrate strategies for tailoring CCCs via a combination of non-equimolar compositional designs and control of grain boundaries (GBs) and microstructures. Using oxide solid electrolytes for all-solid-state batteries as an example, we have discovered a class of compositionally complex perovskite oxides (CCPOs) with improved lithium ionic conductivities beyond the limit of conventional doping. For example, we demonstrate that the ionic conductivity can be improved by >60% in (Li0.375Sr0.4375)(Ta0.375Nb0.375Zr0.125Hf0.125)O3-δ compared with the (Li0.375Sr0.4375)(Ta0.75Zr0.25)O3-δ (LSTZ) baseline. Furthermore, the ionic conductivity can be improved by another >70% via quenching, achieving >270% of the LSTZ. Notably, we demonstrate GB-enabled conductivity improvements via both promoting grain growth and altering GB structures through compositional designs and processing. In a broader perspective, this work suggests new routes for discovering and tailoring CCCs for energy storage and many other applications. The classical materials discovery methodology typically relies on searching for new stoichiometric compounds, where small amounts of dopants or additives are often introduced to modify or improve properties. The recent emergence of high-entropy ceramics (HECs) with diversifying crystal structures1Wright A.J. Luo J. A step forward from high-entropy ceramics to compositionally complex ceramics: a new perspective.J. Mater. Sci. 2020; 55: 9812-9827https://doi.org/10.1007/s10853-020-04583-wCrossref Scopus (122) Google Scholar,2Oses C. Toher C. Curtarolo S. High-entropy ceramics.Nat. Rev. Mater. 2020; 5: 295-309https://doi.org/10.1038/s41578-019-0170-8Crossref Scopus (614) Google Scholar,3Rost C.M. Sachet E. Borman T. Moballegh A. Dickey E.C. Hou D. Jones J.L. Curtarolo S. Maria J.-P. Entropy-stabilized oxides.Nat. Commun. 2015; 6: 8485https://doi.org/10.1038/ncomms9485Crossref PubMed Scopus (1183) Google Scholar,4Gild J. Zhang Y. Harrington T. Jiang S. Hu T. Quinn M.C. Mellor W.M. Zhou N. Vecchio K. Luo J. High-entropy metal diborides: a new class of high-entropy materials and a new type of ultrahigh temperature ceramics.Sci. Rep. 2016; 6: 37946https://doi.org/10.1038/srep37946Crossref Scopus (566) Google Scholar,5Jiang S. Hu T. Gild J. Zhou N. Nie J. Qin M. Harrington T. Vecchio K. Luo J. A new class of high-entropy perovskite oxides.Scripta Mater. 2018; 142: 116-120https://doi.org/10.1016/j.scriptamat.2017.08.040Crossref Scopus (451) Google Scholar,6Gild J. Samiee M. Braun J.L. Harrington T. Vega H. Hopkins P.E. Vecchio K. Luo J. High-entropy fluorite oxides.J. Eur. Ceram. Soc. 2018; 38: 3578-3584https://doi.org/10.1016/j.jeurceramsoc.2018.04.010Crossref Scopus (321) Google Scholar,7Gild J. Braun J. Kaufmann K. Marin E. Harrington T. Hopkins P. Vecchio K. Luo J. A high-entropy silicide: (Mo0.2Nb0.2Ta0.2Ti0.2W0.2)Si2.J. Mater. 2019; 5: 337-343https://doi.org/10.1016/j.jmat.2019.03.002Crossref Scopus (197) Google Scholar,8Qin M. Vega H. Zhang D. Adapa S. Wright A.J. Chen R. Luo J. 21-Component compositionally complex ceramics: discovery of ultrahigh-entropy weberite and fergusonite phases and a pyrochlore-weberite transition.J. Adv. Ceram. 2022; 11: 641-655https://doi.org/10.1007/s40145-022-0575-5Crossref Scopus (11) Google Scholar unlocks vast compositional spaces with multi-principal (but no dominant) components. Specifically, high-entropy perovskite oxides (HEPOs), which were first reported in 2018,5Jiang S. Hu T. Gild J. Zhou N. Nie J. Qin M. Harrington T. Vecchio K. Luo J. A new class of high-entropy perovskite oxides.Scripta Mater. 2018; 142: 116-120https://doi.org/10.1016/j.scriptamat.2017.08.040Crossref Scopus (451) Google Scholar attracted great research interest because of their interesting catalytic,9Nguyen T.X. Liao Y. Lin C. Su Y. Ting J. Advanced high entropy perovskite oxide electrocatalyst for oxygen evolution reaction.Adv. Funct. Mater. 2021; 31: 2101632https://doi.org/10.1002/adfm.202101632Crossref Scopus (128) Google Scholar dielectric,10Zhou S. Pu Y. Zhang X. Shi Y. Gao Z. Feng Y. Shen G. Wang X. Wang D. High energy density, temperature stable lead-free ceramics by introducing high entropy perovskite oxide.Chem. Eng. J. 2022; 427: 131684https://doi.org/10.1016/J.CEJ.2021.131684Crossref Google Scholar,11Pu Y. Zhang Q. Li R. Chen M. Du X. Zhou S. Dielectric properties and electrocaloric effect of high-entropy (Na0.2Bi0.2Ba0.2Sr0.2Ca0.2)TiO3 ceramic.Appl. Phys. Lett. 2019; 115: 223901https://doi.org/10.1063/1.5126652Crossref Scopus (90) Google Scholar ferroelectric,12Li X. Ma J. Chen K. Li C. Zhang X. An L. Design and investigate the electrical properties of Pb(Mg0.2Zn0.2Nb0.2Ta0.2W0.2)O3–PbTiO3 high-entropy ferroelectric ceramics.Ceram. Int. 2022; 48: 12848-12855https://doi.org/10.1016/j.ceramint.2022.01.156Crossref Scopus (6) Google Scholar magnetic,13Witte R. Sarkar A. Kruk R. Eggert B. Brand R.A. Wende H. Hahn H. High-entropy oxides: an emerging prospect for magnetic rare-earth transition metal perovskites.Phys. Rev. Mater. 2019; 3: 034406https://doi.org/10.1103/PhysRevMaterials.3.034406Crossref Scopus (120) Google Scholar thermoelectric,14Zheng Y. Yang Y. Wang M. Hu S. Wu J. Yu Z. Lin Y.-H. Electrical and thermal transport behaviours of high-entropy perovskite thermoelectric oxides.BMC Plant Biol. 2021; 21: 377-384https://doi.org/10.1007/s40145-021-0462-5Crossref Scopus (73) Google Scholar magnetocaloric,15Yin Y. Shi F. Liu G.-Q. Tan X. Jiang J. Tiwari A. Li B. Spin-glass behavior and magnetocaloric properties of high-entropy perovskite oxides.Appl. Phys. Lett. 2022; 120: 082404https://doi.org/10.1063/5.0081688Crossref Scopus (4) Google Scholar and electrocaloric11Pu Y. Zhang Q. Li R. Chen M. Du X. Zhou S. Dielectric properties and electrocaloric effect of high-entropy (Na0.2Bi0.2Ba0.2Sr0.2Ca0.2)TiO3 ceramic.Appl. Phys. Lett. 2019; 115: 223901https://doi.org/10.1063/1.5126652Crossref Scopus (90) Google Scholar properties, as well as promising applications as strongly correlated quantum materials16Sharma Y. Musico B.L. Gao X. Hua C. May A.F. Herklotz A. Rastogi A. Mandrus D. Yan J. Lee H.N. et al.Single-crystal high entropy perovskite oxide epitaxial films.Phys. Rev. Mater. 2018; 2: 060404https://doi.org/10.1103/PhysRevMaterials.2.060404Crossref Scopus (118) Google Scholar and in solid oxide fuel cells,17Li Z. Guan B. Xia F. Nie J. Li W. Ma L. Li W. Zhou L. Wang Y. Tian H. et al.High-entropy perovskite as a high-performing chromium-tolerant cathode for solid oxide fuel cells.ACS Appl. Mater. Interfaces. 2022; 14: 24363-24373https://doi.org/10.1021/acsami.2c03657Crossref Scopus (8) Google Scholar solar thermochemical hydrogen generation (water splitting),18Zhang D. De Santiago H.A. Xu B. Liu C. Trindell J.A. Li W. Park J. Rodriguez M.A. Coker E.N. Sugar J.D. et al.Compositionally complex perovskite oxides for solar thermochemical water splitting.Chem. Mater. 2023; 35: 1901-1915https://doi.org/10.1021/acs.chemmater.2c03054Crossref Scopus (4) Google Scholar lithium-ion batteries,19Yan J. Wang D. Zhang X. Li J. Du Q. Liu X. Zhang J. Qi X. A high-entropy perovskite titanate lithium-ion battery anode.J. Mater. Sci. 2020; 55: 6942-6951https://doi.org/10.1007/s10853-020-04482-0Crossref Scopus (45) Google Scholar and supercapacitors.20Guo M. Liu Y. Zhang F. Cheng F. Cheng C. Miao Y. Gao F. Yu J. Inactive Al3+-doped La(CoCrFeMnNiAlx)1/(5+x)O3 high-entropy perovskite oxides as high performance supercapacitor electrodes.J. Adv. Ceram. 2022; 11: 742-753https://doi.org/10.1007/s40145-022-0568-4Crossref Scopus (18) Google Scholar In 2020, Luo and co-workers further proposed to broaden HECs to compositionally complex ceramics (CCCs)1Wright A.J. Luo J. A step forward from high-entropy ceramics to compositionally complex ceramics: a new perspective.J. Mater. Sci. 2020; 55: 9812-9827https://doi.org/10.1007/s10853-020-04583-wCrossref Scopus (122) Google Scholar,21Wright A.J. Wang Q. Huang C. Nieto A. Chen R. Luo J. From high-entropy ceramics to compositionally-complex ceramics: a case study of fluorite oxides.J. Eur. Ceram. Soc. 2020; 40: 2120-2129https://doi.org/10.1016/j.jeurceramsoc.2020.01.015Crossref Scopus (116) Google Scholar to consider non-equimolar compositions and short- and long-range orders, which reduce the configurational entropy but are often essential for optimizing or further improving properties. Moreover, controlling the interfaces and microstructures of HECs and CCCs, along with non-equimolar compositional designs with aliovalent doping, represents additional (unexplored) opportunities that motivated this study. Oxide solid electrolytes are promising candidates for building high-energy-density all-solid-state batteries (ASSBs) owing to their electrochemical, thermal, and structural stability.22Manthiram A. Yu X. Wang S. Lithium battery chemistries enabled by solid-state electrolytes.Nat. Rev. Mater. 2017; 2: 16103https://doi.org/10.1038/natrevmats.2016.103Crossref Scopus (2383) Google Scholar Among the different oxide solid electrolytes, perovskite-type Li0.5La0.5TiO3 (LLTO) drew significant attention owing to its high bulk Li-ion conductivity (σb) on the order of 10−3 S/cm. Its resistive grain boundaries (GBs), however, constrain its total ionic conductivity (σgb ∼10−5 S/cm so that σtotal ∼10−5 S/cm).23Wu J. Chen L. Song T. Zou Z. Gao J. Zhang W. Shi S. A review on structural characteristics, lithium ion diffusion behavior and temperature dependence of conductivity in perovskite-type solid electrolyte Li3xLa2∕3−xTiO3.Funct. Mater. Lett. 2017; 10: 1730002https://doi.org/10.1142/S179360471730002XCrossref Scopus (28) Google Scholar Furthermore, Ti4+ can be reduced to Ti3+ at potential below 1.8 V vs. Li/Li+,24Nakayama M. Usui T. Uchimoto Y. Wakihara M. Yamamoto M. Changes in electronic structure upon lithium insertion into the A-site deficient perovskite type oxides (Li,La)TiO3.J. Phys. Chem. B. 2005; 109: 4135-4143https://doi.org/10.1021/jp046062jCrossref PubMed Scopus (62) Google Scholar,25Chen C. Ionic conductivity, lithium insertion and extraction of lanthanum lithium titanate.Solid State Ionics. 2001; 144: 51-57https://doi.org/10.1016/S0167-2738(01)00884-0Crossref Scopus (141) Google Scholar which transforms LLTO into an electronic conductor that will ultimately short circuit the battery cell.26Wang M.J. Wolfenstine J.B. Sakamoto J. Mixed electronic and ionic conduction properties of lithium lanthanum titanate.Adv. Funct. Mater. 2020; 30: 1909140https://doi.org/10.1002/adfm.201909140Crossref Scopus (42) Google Scholar In an attempt to address the shortcomings of LLTO, Chen et al. reported perovskite Li0.375Sr0.4375Ta0.75Zr0.25O3 (LSTZ) holding one order of magnitude higher GB ionic conductivity (σgb ∼10−4 S/cm) than that of LLTO, a σbulk of 2 × 10−4 S/cm, and a wider electrochemical stability window down to 1.0 V vs. Li/Li+.27Chen C. Stable lithium-ion conducting perovskite lithium–strontium–tantalum–zirconium–oxide system.Solid State Ionics. 2004; 167: 263-272https://doi.org/10.1016/j.ssi.2004.01.008Crossref Scopus (72) Google Scholar LSTZ has shown better performance compared with LLTO, but its ionic conductivity is still lower than other inorganic solid electrolyte candidates,28Bernuy-Lopez C. Manalastas W. Lopez del Amo J.M. Aguadero A. Aguesse F. Kilner J.A. Atmosphere controlled processing of Ga-substituted garnets for high Li-ion conductivity ceramics.Chem. Mater. 2014; 26: 3610-3617https://doi.org/10.1021/cm5008069Crossref Scopus (252) Google Scholar,29Kamaya N. Homma K. Yamakawa Y. Hirayama M. Kanno R. Yonemura M. Kamiyama T. Kato Y. Hama S. Kawamoto K. Mitsui A. A lithium superionic conductor.Nat. Mater. 2011; 10: 682-686https://doi.org/10.1038/nmat3066Crossref PubMed Scopus (3138) Google Scholar calling for innovative strategies to improve. In general, ABO3 perovskite oxides have broad applications. It is known that aliovalent doping can tune structures (e.g., ordering of A sites,30Yao R. Liu Z.-G. Ding Z.-Y. Jin Y.-J. Cao G. Wang Y.-H. Ouyang J.-H. Effect of Sn or Ta doping on the microstructure and total conductivity of perovskite Li0.24La0.587TiO3 solid electrolyte.J. Alloys Compd. 2020; 844: 156023https://doi.org/10.1016/j.jallcom.2020.156023Crossref Scopus (2) Google Scholar,31Teranishi T. Yamamoto M. Hayashi H. Kishimoto A. Lithium ion conductivity of Nd-doped (Li, La)TiO3 ceramics.Solid State Ionics. 2013; 243: 18-21https://doi.org/10.1016/j.ssi.2013.04.014Crossref Scopus (34) Google Scholar concentration of A-site vacancies,32Morata-Orrantia A. García-Martín S. Alario-Franco M.Á. Optimization of lithium conductivity in La/Li titanates.Chem. Mater. 2003; 15: 3991-3995https://doi.org/10.1021/cm0300563Crossref Scopus (48) Google Scholar lattice parameter33Chung H. Kim J.-G. Kim H.-G. Dependence of the lithium ionic conductivity on the B-site ion substitution in (Li0.5La0.5)Ti1−xMxO3 (M=Sn, Zr, Mn, Ge).Solid State Ionics. 1998; 107: 153-160https://doi.org/10.1016/S0167-2738(97)00525-0Crossref Google Scholar,34Okumura T. Ina T. Orikasa Y. Arai H. Uchimoto Y. Ogumi Z. Improvement of lithium ion conductivity for A-site disordered lithium lanthanum titanate perovskite oxides by fluoride ion substitution.J. Mater. Chem. 2011; 21: 10061https://doi.org/10.1039/c0jm04367bCrossref Scopus (11) Google Scholar) and influence properties including Li-ion conductivity. To date, the majority of the studies have been limited to single and co-doping. The amount of dopants is typically below 10 mol % in A- or B-site sublattices to avoid precipitation.33Chung H. Kim J.-G. Kim H.-G. Dependence of the lithium ionic conductivity on the B-site ion substitution in (Li0.5La0.5)Ti1−xMxO3 (M=Sn, Zr, Mn, Ge).Solid State Ionics. 1998; 107: 153-160https://doi.org/10.1016/S0167-2738(97)00525-0Crossref Google Scholar,35Lee S.J. Bae J.J. Son J.T. Structural and electrical effects of Y-doped Li0.33La0.56−xYxTiO3 solid electrolytes on all-solid-state lithium ion batteries.J. Kor. Phys. Soc. 2019; 74: 73-77https://doi.org/10.3938/jkps.74.73Crossref Scopus (12) Google Scholar The perovskite structure can tolerate a wide range of cation dopants following a criterion using Goldschmidt’s tolerance factor,36Goldschmidt V.M. Die gesetze der Krystallochemie.Naturwissenschaften. 1926; 14: 477-485https://doi.org/10.1007/BF01507527Crossref Scopus (2250) Google Scholar which renders it promising as a model system for exploring complex or high-entropy compositions. However, only a single study explored HEPOs as a solid electrolyte, where the reported ionic conductivity was lower than that of baseline material LLTO.37Yazhou K. Zhiren Y. Synthesis, structure and electrochemical properties of Al doped high entropy perovskite Lix(LiLaCaSrBa)Ti1-xAlxO3.Ceram. Int. 2022; 48: 5035-5039https://doi.org/10.1016/j.ceramint.2021.11.041Crossref Scopus (6) Google Scholar In fact, the field of high-entropy solid electrolytes remains largely unexplored, with only few prior reports.37Yazhou K. Zhiren Y. Synthesis, structure and electrochemical properties of Al doped high entropy perovskite Lix(LiLaCaSrBa)Ti1-xAlxO3.Ceram. Int. 2022; 48: 5035-5039https://doi.org/10.1016/j.ceramint.2021.11.041Crossref Scopus (6) Google Scholar,38Fu Z. Ferguson J. Processing and characterization of an Li7La3Zr0.5Nb0.5Ta0.5Hf0.5O12 high-entropy Li–garnet electrolyte.J. Am. Ceram. Soc. 2022; 105: 6175-6183https://doi.org/10.1111/jace.18576Crossref Scopus (3) Google Scholar,39Bérardan D. Franger S. Meena A.K. Dragoe N. Room temperature lithium superionic conductivity in high entropy oxides.J. Mater. Chem. 2016; 4: 9536-9541https://doi.org/10.1039/C6TA03249DCrossref Google Scholar Notably, a recent study demonstrated a high-entropy mechanism to improve ionic conductivity, achieving 2.2 × 10−5 S/cm in Li(Ti,Zr,Sn,Hf)2(PO4)3, which represents substantial improvements from their base materials (thereby being promising).40Zeng Y. Ouyang B. Liu J. Byeon Y.W. Cai Z. Miara L.J. Wang Y. Ceder G. High-entropy mechanism to boost ionic conductivity.Science. 2022; 378: 1320-1324https://doi.org/10.1126/science.abq1346Crossref PubMed Scopus (13) Google Scholar,41Botros M. Janek J. Embracing disorder in solid-state batteries.Science. 2022; 378: 1273-1274https://doi.org/10.1126/science.adf3383Crossref Scopus (1) Google Scholar Herein, we discovered a new class of compositionally complex perovskite oxides (CCPOs) to achieve (an order of magnitude higher) 2.56 × 10−4 S/cm total ionic conductivity with improvements (from the LSTZ baseline) through not only the bulk compositional complexity but also microstructure and GB effects. Specifically, we discovered a class of LSTZ-derived compositionally complex solid electrolytes (CCPOs) in this study. We further revealed the phase-microstructure-property relationship to enable us to achieve improved total ionic conductivity, achieving >270% of the state-of-the-art LSTZ baseline, yet with comparable electrochemical stability. In addition, we showed GB-enabled conductivity improvements, through both promoting grain growth and altering the GB structure and segregation profile via compositional designs and materials processing. To investigate the underlying mechanisms, aberration-corrected advanced microscopy and atomistic simulations using a state-of-the-art active learning moment tensor potential (MTP) and additional density functional theory (DFT) calculations were employed. This work highlights the significance of microstructure and GB controls of CCCs in boosting their ionic conductivities, in addition to non-equimolar compositional designs, which points to a new direction to design and tailor compositionally complex solid electrolytes and potentially many other functional CCCs. In this study, we synthesized 28 different compositions of CCPOs (through high-energy ball milling and sintering at 1,300°C for 12 h in air; see experimental procedures) as a new class of lithium-ion solid electrolytes. The key results of three series of Sn-containing (Li,Sr)(Ta,Nb,Zr,Sn)O3-δ (LSTNZS) CCPOs and two series of Hf-containing (Li,Sr)(Ta,Nb,Zr,Hf)O3-δ (LSTNZH) CCPOs are summarized in supplemental information (Tables S1–S3). δ represents oxygen non-stoichiometry in perovskites. We showed that enhanced ionic conductivities, in comparison with the state-of-art LSTZ baseline, can be attained in these CCPOs via compositional designs and controlling GBs and microstructures, while maintaining comparable electrochemical stability. Figure 1 presents an outline of several generations of CCPOs discovered in this study, in comparison with the LSTZ baseline. In the first generation, the best Sn-containing LSTNZS CCPO (Li0.375Sr0.4375)(Ta0.334Nb0.347Zr0.211Sn0.108)O3-δ shows total ionic conductivity >2.3× higher than that of the LSTZ baseline; however, its electrochemical stability is compromised due to the presence of redox-active Sn (Figure S1). In the second generation, an Hf-containing (and Sn-free) LSTNZH CCPO (Li0.375Sr0.4375)(Ta0.375Nb0.375Zr0.125Hf0.125)O3-δ shows electrochemical stability comparable with LSTZ with slightly lower ionic conductivity (but >60% improvement from the LSTZ baseline), which can be further improved to achieve 0.256 mS/cm total ionic conductivity or 270% of the LSTZ baseline via quenching. A notable discovery of this study is GB-enabled conductivity improvements through both promoting grain growth and altering the GB structure and segregation profile through compositional designs and materials processing. For example, the observed conductivity improvement via quenching is primarily from an increase in specific (true) GB ionic conductivity (Note S1),42Haile S.M. Staneff G. Ryu K.H. Non-stoichiometry, grain boundary transport and chemical stability of proton conducting perovskites.J. Mater. Sci. 2001; 36: 1149-1160https://doi.org/10.1023/A:1004877708871Crossref Scopus (297) Google Scholar resulting from changes in the GB compositional profile and structure, as shown by aberration-corrected (AC) scanning transmission electron microscopy (STEM) presented and discussed in a subsequent section. Figure 2A displays a schematic of the CCPO unit cell of the ABO3 perovskite (A-site-centered view). The atomic-resolution high-angle annular dark-field (HAADF) STEM image of a representative LSTNZH CCPO is shown in Figure 2B and its X-ray diffraction (XRD) pattern is shown in Figure 2D, which confirmed the cubic perovskite structure. Next, we discuss the compositional designs of CCPOs and the resulting phase stability (any secondary phases) and properties (conductivities). Expanding on a simpler compositional design for LSTZ,27Chen C. Stable lithium-ion conducting perovskite lithium–strontium–tantalum–zirconium–oxide system.Solid State Ionics. 2004; 167: 263-272https://doi.org/10.1016/j.ssi.2004.01.008Crossref Scopus (72) Google Scholar we first designed our “x series” CCPOs following a general formula [Li(2/3)xSr1-x(VA″)(1/3)x][(5B)(4/3)x(4B)1-(4/3)x]O3-δ. Here, the A site is occupied by Li+, Sr2+, and A-site vacancies VA″ (in the Kröger–Vink notation), and the B site is occupied by a (presumably random) mixture of 5B (5+ B-site) cations, which are equal moles of Ta5+ and Nb5+ cations, and 4B (4+ B-site) cations, which are equal moles of Zr4+ and Sn4+ (or Hf4+) cations for LSTNZS (or LSTNZH). Omitting VA″ for brevity, the chemical formulas are (Li2/3xSr1-x)(Ta2/3xNb2/3xZr0.5(1-4/3x)Sn0.5(1-4/3x)O3-δ for the x series of LSTNZS and (Li2/3xSr1-x)(Ta2/3xNb2/3xZr0.5(1-4/3x)Hf0.5(1-4/3x)O3-δ for the x series of LSTNZH. Here, we first investigated the (primary vs. secondary) phase formation and stability using XRD. Figures 2C and 2D present the XRD patterns of x series of LSTNZS and LSTNZH, respectively. For x = 8/16, the Sn-containing LSTNZS shows a cubic perovskite structure in the space group Pm 3¯ m, which matches the crystal structure of KTaO3 (PDF #38–1,470) and that of LSTZ. The amounts of secondary phases increase as the x value increases to 9/16 and higher. At x ≥ 9/16, LiNbO3-prototyped rhombohedral structure (space group R3c) and Sr2.83Ta5O15-prototyped with tetragonal structure (space group P4/mbm) formed, which consequently triggered the precipitation of secondary SnO2 and ZrO2 phases. The corresponding phase separation is evident in scanning electron microscopy (SEM) energy-dispersive X-ray spectroscopy (EDS) maps in Figure S2. The Hf-containing LSTNZH system shows a wider single-phase range of up to x = 9/16, with only trace amounts of ZrO2 secondary phase found. At x > 9/16, however, the primary phase becomes LiNbO3-prototyped rhombohedral instead of the cubic perovskite phase. Based on these results, we conclude the primarily single-phase range of CCPOs in the x series is mainly determined by the x value, and the threshold (for the appearance of large amounts of secondary phases) depends on the difference in cation ionic radii. The Sn-containing LSTNZS x series has a lower threshold (x ≥ 9/16) than the Hf-containing LSTNZH x series, which can be attributed to the larger difference in ionic radii between Sn4+ and Zr4+ (∼4.17%) than that between Hf4+ and Zr4+ (1.39%), calculated using Shannon ionic radii (Table S4).43Shannon R.D. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides.Acta Crystallogr. A. 1976; 32: 751-767https://doi.org/10.1107/S0567739476001551Crossref Scopus (54851) Google Scholar This large ionic radius difference makes it more difficult for LSTNZS to form stable cubic perovskite solid solution at x ≥ 9/16 and results in precipitation of the secondary SnO2 and ZrO2 phases. To further test this hypothesis, LSTNZS (x = 9/16) was benchmarked with Nb and Sn co-doped LSTZ counterparts in Figure S3, which again implies that the poor single-phase formation is due to the large ionic radii difference on the B site. To further improve the single-phase formability of the LSTNZS system, we adopted a “natural selection” composition optimization strategy described in the supplemental information (Figures S4–S6; Tables S5 and S6; Note S2) to synthesize the composition of the primary phase guided from SEM-EDS compositional quantification results of the single-phase region of the prior generation, in an iterative procedure. Accordingly, an LSTNZS “y series” with the optimal B-site cation ratio (Li2/3ySr1-y)(Ta0.334Nb0.347Zr0.211Sn0.108)O3-δ was developed. XRD patterns of the LSTNZS y series (Figure 2E) exhibit improved single-phase formation compared with the LSTNZS x series. For LSTNZS (y = 0.6), the main secondary phase was determined to be Li(Ta, Nb)O3. The design workflows of x and y series are illustrated in Figure S7. For subsequent characterizations, we will focus on the Sn-containing LSTNZS y series and Hf-containing LSTNZH x series, which are the two best series obtained in this study. To quantify the primary phase fractions, we performed Rietveld refinements of XRD patterns of LSTNZS y series (Figure S8) and LSTNZH x series (Figure S9). The results are summarized in Table S7. We also quantify the primary phase fractions from SEM-EDS maps (Figure S10) and the results are shown in Figures 3A and 3B for the Sn-containing LSTNZS y series and the Hf-containing LSTNZH x series, respectively. Figure S11 compares results obtained from the two methods, which are largely consistent. The exception is for LSTNZH (x = 10/16), where the refinement accuracy is low due to the presence of multiple secondary phases and peak overlaps and quantification from SEM-EDS maps is likely more accurate. In addition, we measured the Li-ion conductivities of the as-synthesized specimens. Figure 3C illustrates the influence of the compositional variable y (that correlates with the Li+ concentration, 23y, and the VA″ concentration, 13y, in the A site) on the bulk ionic conductivity σbulk (fitted from the measured impedance spectra by the model discussed in Note S1) of the Sn-containing LSTNZS y series. When y = 0.387, σbulkLSTNZS is 0.004 mS/cm (the lowest). As y increases to 0.5, σbulkLSTNZS is improved by two orders of magnitude to 0.118 mS/cm. The bulk ionic conductivity is maximized at y = 9/16, even with the increase of secondary phase fractions. Hence, the change of σbulkLSTNZS is dominant by A-site carrier and vacancy amount (rather than phase purities) at y ≤ 9/16 region. In contrast, the influence of secondary phases becomes dominant when y > 9/16. Although the nominal A-site carrier and vacancy amount are the highest at y = 0.6 in Figure 3C, the existence of Sr2.83Ta5O15, LiTaO3, and LiNbO3 secondary phases indicates the actual Li+ in the main phase is less than the nominal amount. Therefore, the σbulkLSTNZS value reduces to 0.14 mS/cm at y = 0.6. The maximum of σbulkLSTNZS is 0.218 mS/cm at y = 9/16, which is higher than that of optimal LSTNZS x series composition (x = 8/16; 0.11 mS/cm). Given that both samples exhibit similar phase stability (LSTNZS, y = 9/16 vs. x = 8/16), the improvement can be attributed to the higher A-site carrier and vacancy concentration while maintaining cubic perovskite structure. To justify the selection of B-site stoichiometry that gives optimal ionic conductivity, an additional LSTNZS “w series” was fabricated following the formula (Li0.375Sr0.4375)(Ta0.334Nb0.347Zr0.319-wSnw)O3-δ, where w variable dictates the Sn cation fraction on the B site. The results shown in Figures S12A and S12C indicate the improvement of both phase stability and σbulkLSTNZS upon increasing the Sn/Zr fra