# Fileset

[JApplPhys139_1014102.pdf](https://mdr.nims.go.jp/filesets/3524c2e6-99ab-407d-9ff8-f63b44f627a4/download)

## Creator

[Suzuka Udagawa](https://orcid.org/0009-0004-4163-9406), [Takanori Mimura](https://orcid.org/0009-0009-3818-5905), [Tetsuhiro Katsumata](https://orcid.org/0000-0002-8430-935X), [Yoshitaka Matsushita](https://orcid.org/0000-0002-4968-8905), [Takashi Mochiku](https://orcid.org/0000-0003-2208-4279), [Yoshiyuki Inaguma](https://orcid.org/0000-0002-1174-8407)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Formation process and composition-dependent properties in non-centrosymmetric Ta2O5 phase with Zr substitution](https://mdr.nims.go.jp/datasets/c06e79c3-70e2-4a17-a58f-7a0085afbb9b)

## Fulltext

Formation process and composition-dependent properties in non-centrosymmetric Ta2O5 phase with Zr substitutionViewOnlineExportCitationRESEARCH ARTICLE |  MARCH 10 2026Formation process and composition-dependent propertiesin non-centrosymmetric Ta2O5 phase with Zr substitutionSuzuka Udagawa  ; Takanori Mimura   ; Tetsuhiro Katsumata  ; Yoshitaka Matsushita  ;Takashi Mochiku  ; Yoshiyuki Inaguma J. Appl. Phys. 139, 104102 (2026)https://doi.org/10.1063/5.0312634Articles You May Be Interested InA dielectric material, Zr0.10Ta0.90O2.45, with a noncentrosymmetric L-Ta2O5-related structureAppl. Phys. Lett. (April 2025)Plasmonic spectral tunability of conductive ternary nitridesAppl. Phys. Lett. (June 2016)Growth and dielectric properties of Ta2O5 single crystal fibersPICALO2008  12 March 2026 00:17:58https://pubs.aip.org/aip/jap/article/139/10/104102/3383021/Formation-process-and-composition-dependenthttps://pubs.aip.org/aip/jap/article/139/10/104102/3383021/Formation-process-and-composition-dependent?pdfCoverIconEvent=citejavascript:;https://orcid.org/0009-0004-4163-9406javascript:;https://orcid.org/0009-0009-3818-5905javascript:;https://orcid.org/0000-0002-8430-935Xjavascript:;https://orcid.org/0000-0002-4968-8905javascript:;https://orcid.org/0000-0003-2208-4279javascript:;https://orcid.org/0000-0002-1174-8407https://crossmark.crossref.org/dialog/?doi=10.1063/5.0312634&domain=pdf&date_stamp=2026-03-10https://doi.org/10.1063/5.0312634https://pubs.aip.org/aip/apl/article/126/15/152902/3344122/A-dielectric-material-Zr0-10Ta0-90O2-45-with-ahttps://pubs.aip.org/aip/apl/article/108/26/263110/30820/Plasmonic-spectral-tunability-of-conductivehttps://pubs.aip.org/lia/picalo/proceedings/PICALO2008/2008/609/945708https://servedbyadbutler.com/redirect.spark?MID=188841&plid=3541191&setID=1044475&channelID=0&CID=1697306&banID=524364881&PID=0&textadID=0&tc=1&rnd=2756114879&scheduleID=3737490&adSize=1640x440&data_keys=%7B%22%22%3A%22%22%7D&mt=1773274678368717&spr=1&referrer=http%3A%2F%2Fpubs.aip.org%2Faip%2Fjap%2Farticle-pdf%2Fdoi%2F10.1063%2F5.0312634%2F20935226%2F104102_1_5.0312634.pdf&request_uuid=c165b6c5-3f03-41ce-848e-89590005436c&hc=2c611718007aee16b0fdf1bc019102bc58f7e10e&location=Formation process and composition-dependentproperties in non-centrosymmetric Ta2O5 phasewith Zr substitutionCite as: J. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634View Online Export Citation CrossMarkSubmitted: 18 November 2025 · Accepted: 23 February 2026 ·Published Online: 10 March 2026Suzuka Udagawa,1 Takanori Mimura,1,a) Tetsuhiro Katsumata,2 Yoshitaka Matsushita,3Takashi Mochiku,3 and Yoshiyuki Inaguma1,b)AFFILIATIONS1Department of Chemistry, Faculty of Science, Gakushuin University, 1-5-1 Mejiro, Toshima-ku, Tokyo 171-8588, Japan2Department of Chemistry, School of Science, Tokai University, 4-1-1 Kitakaname, Hiratsuka, Kanagawa 259-1292, Japan3Research Network and Facility Services Division, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba,Ibaraki 305-0047, Japana)Author to whom correspondence should be addressed: 20220185@gakushuin.ac.jpb)Email: yoshiyuki.inaguma@gakushuin.ac.jpABSTRACT(ZrxTa1−x)2O5−x was prepared by a solid-state reaction of ZrO2 and Ta2O5, and the L0-Ta2O5 phase was obtained by cooling the H-Ta2O5phase. High-temperature x-ray diffraction measurements showed that the starting materials, ZrO2 and low-temperature L-Ta2O5, formedthe high-temperature H-Ta2O5 phase when heated above 1360 °C. Upon cooling, this phase sequentially transformed into L00-Ta2O5, thehigh-temperature L0-Ta2O5 phase, and L0-Ta2O5 phases. As the Zr content, x, decreased, the transition from the H-Ta2O5 phase to theL00-Ta2O5 phase slowed. The temperature dependence of the dielectric constant revealed a maximum value, which is attributed to the phasetransition from L0-Ta2O5 to L00-Ta2O5. This transition temperature decreases by approximately 50 °C for every 0.01 increase in the x value.The L0-Ta2O5 phase exhibited negative volumetric thermal expansion (NTE) behavior near the phase transition temperature. As x decreased,the NTE coefficient increased from −1.09 × 10−6/K (77–127 °C) for x = 0.10 to −2.06 × 10−5/K (327–427 °C) for x = 0.05. The substitutionof Zr into Ta2O5 stabilized the non-centrosymmetric L0-Ta2O5 phase and controlled the phase transition temperature and thermalexpansion behavior.© 2026 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license(https://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1063/5.0312634I. INTRODUCTIONDielectric materials are essential to a wide variety of technolo-gies, including microelectronics, medical devices, and energystorage systems.1–6 Ta2O5 is a significant dielectric material recog-nized for its high dielectric constant, nonlinear refractive index,and wide bandgap.7 Therefore, it is used in dynamic random-accessmemory (DRAM) applications and antireflection coatings.8,9Ta2O5 exists primarily in two different phases based on thetemperature: the high-temperature H-Ta2O5 phase and the low-temperature L-Ta2O5 phase, with a phase transformation occurringat T = 1360 °C.10 This transformation is reversible and occursslowly; annealing below 1360 °C and/or slow cooling effectivelyfacilitates the transition from the H-Ta2O5 phase to the L-Ta2O5phase.10–12 These two phases have complex structures with variousreported crystal structures. For the H-Ta2O5 phase, the proposedspace groups include I2, Imma, C2/m, and I41/amd, while for theL-Ta2O5 phase, Pmm2, Pccm, and P2mm have been suggested. Thetetragonal I41/amd structure (a = 3.86 Å and c = 36.18 Å) is widelyaccepted for the H-Ta2O5 phase, whereas the orthorhombic P2mmstructure (a = 6.198 Å, b = 40.290 Å, and c = 3.888 Å) is for theL-Ta2O5 phase.12,13Cation substitution can change the crystal structure andphysical properties in Ta2O5-based materials. The compounds(SixTa1−x)2O5−x form the L-Ta2O5 phase for x = 0.00–0.11, whichJournal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-1© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.1063/5.0312634https://doi.org/10.1063/5.0312634https://pubs.aip.org/action/showCitFormats?type=show&doi=10.1063/5.0312634http://crossmark.crossref.org/dialog/?doi=10.1063/5.0312634&domain=pdf&date_stamp=2026-03-10https://orcid.org/0009-0004-4163-9406https://orcid.org/0009-0009-3818-5905https://orcid.org/0000-0002-8430-935Xhttps://orcid.org/0000-0002-4968-8905https://orcid.org/0000-0003-2208-4279https://orcid.org/0000-0002-1174-8407mailto:20220185@gakushuin.ac.jpmailto:yoshiyuki.inaguma@gakushuin.ac.jphttps://creativecommons.org/licenses/by/4.0/https://creativecommons.org/licenses/by/4.0/https://doi.org/10.1063/5.0312634https://pubs.aip.org/aip/japhas a dielectric constant of ∼30.14 By contrast, (ZrxTa1−x)2O5−x canform either the L-Ta2O5 or H-Ta2O5 phase for x = 0.00–0.14,depending on heating conditions, exhibiting dielectric constants ofroughly 30 and 50, respectively.15 For (TixTa1−x)2O5−x, theH-Ta2O5 phase forms within the composition range of x = 0.0–0.25, corresponding to dielectric constants of 20–126 depending onsubstitution concentrations.16,17 Notably, (Ti0.04Ta0.96)2O4.96 showsa high dielectric constant of 126 owing to structural distortion.17,18Recently, we identified a new dielectric material, refined as theL0-Ta2O5 phase, in (Zr0.10Ta0.90)2O4.90, which was synthesized byheating a mixture of Ta2O5 and ZrO2 powders at 1700 °C for10 h.19 This material has a C-centered orthorhombic structure(a = 6.3717 Å, b = 10.8003 Å, and c = 3.87058 Å) related to theL-Ta2O5 structure. The L0-Ta2O5 phase exhibits a strong second-harmonic generation (SHG) signal, which is not present in the con-ventional L-Ta2O5 phase. It also shows a higher dielectric constantof ∼55, compared with 33 for the L-Ta2O5 phase fired at 1200 °Cfor 10 h. Moreover, high-temperature x-ray diffraction (HTXRD)measurements revealed the high-temperature L00-Ta2O5 phase,which has a pseudo-hexagonal structure. The dielectric constantdepends on temperature and exhibits a maximum value of 60 at87 °C, corresponding to this phase transition. These findings indi-cate the potential of L0-Ta2O5 as a new non-centrosymmetricdielectric material. As it stands, the L0-Ta2O5 phase has only beensynthesized for the specific composition of (Zr0.10Ta0.90)2O4.90; noresearch has been conducted on the synthesis and characterizationof other compositions within the (ZrxTa1−x)2O5−x family.Moreover, why heating at 1700 °C results in the formation of theL0-Ta2O5 phase instead of the previously observed L-Ta2O5 orH-Ta2O5 phases in (ZrxTa1−x)2O5−x systems remains unclear.In this work, we investigated the composition range that leadsto the formation of the L0-Ta2O5 phase within the (ZrxTa1−x)2O5−xsystem. We found that a pure L0-Ta2O5 phase formed in the com-position range of x = 0.05–0.11 by optimizing the heat-treatmentconditions. To investigate the formation pathway of the L0-Ta2O5phase from the starting materials, ZrO2 and L-Ta2O5, we con-ducted HTXRD measurements. We report the formation processand composition-dependent properties of non-centrosymmetricL0-Ta2O5 type (ZrxTa1−x)2O5−x.II. EXPERIMENTAL(ZrxTa1−x)2O5−x (x = 0.04–0.13) was synthesized via a solid-state reaction using ZrO2 (rare metallic, >99.99% purity, <2% Hf)and Ta2O5 (rare metallic, >99.99% purity, low-temperatureL-Ta2O5) powders as starting materials in a stoichiometric ratio.The powders were mixed with ethanol using an agate mortar, thendried at 120 °C, and pressed into f 7 mm or f 10 mm pellets.These pellets were sintered at 1700 °C for either 10 or 5 h at aheating rate of 2.6 °C/min and cooling rate of 1.2 °C/min.Subsequently, the pellets were heated at 1200 °C for 10 h. In bothheating processes, the pellets were heated in air. XRD measure-ments were performed for phase identification using a PANalyticalX’Pert3 powder diffractometer (Cu Kα, λ = 0.15418 nm). To investi-gate the formation process of the L0-Ta2O5 phase, in situ HTXRDmeasurements were performed over a temperature range of 28–1450 °C using a Bruker AXS D8 ADVANCE diffractometerequipped with a high-temperature stage (Cu Kα, λ = 0.15418 nm).The starting material consisted of a mixture of ZrO2 and L-Ta2O5powders in a stoichiometric ratio corresponding to(Zr0.10Ta0.90)2O4.90. The sample was placed on a Pt heater anddirectly heated under a nitrogen atmosphere. Synchrotron powderXRD was conducted for the sample with x = 0.05 from −173 to800 °C using Debye–Scherrer-type powder diffractometer atSPring-8 (BL13XU, λ = 0.375695 nm). The sample powder waspacked into a quartz glass capillary with an outer diameter of0.1 mm and a wall thickness of 0.01 mm.SHG measurements were performed at room temperature toconfirm the non-centrosymmetry. A modified Kurtz nonlinearoptical (NLO) system using 1064 nm light (Nd:YAG laser) wasused for the SHG measurement.20 To measure the dielectric cons-tant, f 7 mm gold electrodes were deposited on both sides of thepellets using a direct-current sputtering method. The dielectricconstant was measured using an inductance–capacitance–resistance(LCR) meter (4284A; Agilent, Palo Alto, CA, USA) at frequenciesof 1 kHz, 10 kHz, 50 kHz, 100 kHz, 200 kHz, 500 kHz, and 1MHzwith an applied signal of 1 V in the temperature range of 30–370 °C in air.III. OPTIMIZING THE HEATING CONDITIONSTO FORM THE L0-Ta2O5 PHASE IN (ZrxTa1−x)2O5−xFOR x = 0.05–0.11Figure 1 shows the XRD patterns of (ZrxTa1−x)2O5−x(x = 0.04–0.13) obtained by heating at 1700 °C. The referencepattern for the L-Ta2O5 phase (P2mm) is included for compari-son.13 For the compositions with x = 0.06–0.13, the two prominentpeaks around 2θ = 28° were inverted in relation to the peak intensi-ties of the L-Ta2O5 phase. This inversion indicates the formation ofthe L0-Ta2O5 phase. Among these compositions, the samples withx = 0.09–0.11 formed a single L0-Ta2O5 phase without any impurityphases. However, the Zr6Ta2O17 phase was observed as an impurityin the samples with x = 0.12–0.13. Note that the Zr6Ta2O17 phaseremained even after re-heating at 1700 °C for 10 h (not shownhere). Therefore, the solid-solution limit of ZrO2 to Ta2O5 was esti-mated to be x = 0.11 in (ZrxTa1−x)2O5−x. For compositions withx = 0.05–0.08, the H-Ta2O5 phase was formed. The weight fractionof the H-Ta2O5 phase relative to the L0-Ta2O5 phase increased withdecreasing ZrO2 substitution, and only the H-Ta2O5 phase wasobtained for the samples with x = 0.04–0.05.Based on the binary phase diagram of Ta2O5 and ZrO2, theH-Ta2O5 phase is formed at 1700 °C during heating.15,21 This indi-cates that the L0-Ta2O5 phase is formed by cooling the H-Ta2O5phase. Additionally, the L0-Ta2O5 phase has a high-temperatureL00-Ta2O5 phase, as mentioned in a previous report.19 Therefore,the phase transitions during cooling are expected to proceed asH-Ta2O5→ L00-Ta2O5→ L0-Ta2O5, as illustrated in Fig. S1(a) in thesupplementary material.The transformation from the H-Ta2O5 phase to the conven-tional L-Ta2O5 phase is sluggish; therefore, the phase transition tothe L00-Ta2O5 phase is also expected to be slow.10–12 For composi-tions with x = 0.04–0.08, the H-Ta2O5 phase could be fully or par-tially quenched to room temperature, even with a slow furnacecooling rate of 1.2 °C/min. This indicates that the activation energyJournal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-2© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.60893/figshare.jap.c.8322903https://pubs.aip.org/aip/japbarrier required for the transition from the H-Ta2O5 phase to theL00-Ta2O5 phase is relatively high in this composition region. Topromote this phase transition, either slowing the cooling ratefurther or implementing an additional annealing process is neces-sary, as shown in Figs. S1(a) and S1(b) in the supplementarymaterial.To produce the L0-Ta2O5 phase, samples of x = 0.04–0.08,which initially formed the H-Ta2O5 phase when heated at 1700 °Cfor 10 h, were subsequently heated at 1200 °C for 10 h. This tem-perature was chosen because it is slightly lower than the stabiliza-tion point of the H-Ta2O5 phase at 1360 °C and is expected topromote the phase transition to the L00-Ta2O5 phase. Figure 2 dis-plays the XRD patterns of (ZrxTa1−x)2O5−x (x = 0.04–0.08) afterre-heating. The two prominent peaks around 2θ = 28° wereinverted, indicating the formation of the L0-Ta2O5 phase. A pureL0-Ta2O5 phase was obtained for samples with x = 0.05–0.08,whereas an additional impurity L-Ta2O5 phase was formed forx = 0.04. Based on Figs. 1 and 2, a single L0-Ta2O5 phase was con-firmed for the samples with x = 0.05–0.11. Moreover, the SHGresponses observed for these compositions support the formationof the L0-Ta2O5 phase, as shown in Fig. S2 in the supplementarymaterial.IV. FORMATION PROCESS OF THE L0-Ta2O5 PHASECONFIRMED BY HTXRD MEASUREMENTThe expected formation sequence of the L0-Ta2O5 phase, illus-trated in Figs. S1(a) and S1(b) in the supplementary material,begins with a mixture of ZrO2 and Ta2O5 as the starting materials.Upon heating, this mixture transforms into a solid solution of thehigh-temperature H-Ta2O5 phase.15,21 As it cools, it transformsinto the L00-Ta2O5 phase, followed by a transition to the L0-Ta2O5phase.19To verify this expected formation process, HTXRD analysiswas conducted according to the pathway outlined in Fig. S1(b) inthe supplementary material. ZrO2 and Ta2O5 powders mixed in astoichiometric ratio corresponding to (Zr0.10Ta0.90)2O4.90 were usedas the starting materials. The whole time series of the HTXRDresults is shown in Fig. 3 as a 3D diagram. The horizontal and ver-tical axes represent the 2θ angle and time, respectively, while thetemperature at each time point is drawn on the right. Yellowdashed lines separate the obtained phases at each time scale.Reference patterns for the L-Ta2O5 phase (P2mm) and H-Ta2O5phase (I41/amd) are also included in the figure above. Figure S3 inthe supplementary material presents the raw XRD data collected ateach measurement step.FIG. 1. XRD patterns of (ZrxTa1−x)2O5−x heated at 1700 °C for 10 h, with areference pattern of L-Ta2O5 (P2mm, ICSD No. 9112). The filled circles and tri-angles represent reflections from the H-Ta2O5 phase and Zr6Ta2O17,respectively.FIG. 2. XRD patterns of (ZrxTa1−x)2O5−x heated at 1700 °C for 10 h, followedby additional heating at 1200 °C for 10 h. Reference pattern is L-Ta2O5 with aP2mm structure (ICSD No. 9112). The filled squares indicate reflections fromthe L-Ta2O5 phase.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-3© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://pubs.aip.org/aip/japAt room temperature, the observed XRD peaks were those ofstarting materials of ZrO2 and Ta2O5. With an increase in tempera-ture, the peak from ZrO2 observed at 2θ = 31.5° disappeared at1200 °C, indicating that ZrO2 was dissolved in Ta2O5 as shown inFig. S4 in the supplementary material. This dissolution resulted ina change in the diffraction peak positions of the L-Ta2O5 phase.Additionally, a small amount of the impurity Zr6Ta2O17 phaseformed, as shown in Fig. 3. This Zr6Ta2O17 phase remained evenafter the HTXRD measurements were completed.The binary phase diagram of (ZrxTa1−x)2O5−x at x = 0.10 and1200 °C reveals the phase segregation of L-Ta2O5 (Zr-substitutedL-Ta2O5) and Zr6Ta2O17,21 which is consistent with our results.Although this phase diagram describes the L-Ta2O5 phase, theXRD peak at 1200 °C closely resembles the L00-Ta2O5 phase, as thegap between the two diffraction peaks around 2θ = 28° narrowedand merged into a single peak. However, the L0-Ta2O5 phase didnot form after 1200 °C heat treatment for 10 h, confirming its iden-tity as the L-Ta2O5 phase. The transition from the L-Ta2O5 toH-Ta2O5 phase was confirmed at 1400 °C, aligning with a previousreport that this phase forms above 1360 °C.10 The H-Ta2O5 phaseremained the main phase when the temperature was held at1450 °C and subsequently decreased to 1150 °C.During the holding process at 1150 °C, peaks correspondingto the L00-Ta2O5 phase emerged. With increasing holding time, thepeak intensity increased, while the peak intensity of the H-Ta2O5phase decreased. This observation indicates that the H-Ta2O5phase gradually changed into the L00-Ta2O5 phase. After holding at1150 °C for 5 h, the L00-Ta2O5 phase became the main phase.However, the H-Ta2O5 phase remained, indicating that the phasetransition was incomplete owing to the limited holding time. Uponcooling to 30 °C, peaks corresponding to the L0-Ta2O5 phaseemerged at 200 °C, indicating a transition from the L00-Ta2O5 phaseto the L0-Ta2O5 phase. Notably, the phase transition temperatureobserved here differs from that reported previously (ca. 400 K,127 °C), estimated from the observable change in the XRD peakpattern of HTXRD measurement.19 This discrepancy can be attrib-uted to the fact that the amount of Zr in the composition ofL0-Ta2O5 is lower than in (Zr0.10Ta0.90)2O4.90 owing to the forma-tion of Zr6Ta2O17, as shown in Fig. 3. The correlation between Zrsubstitution content and phase transition temperature will be dis-cussed later.For compositions with x = 0.09–0.11, the single L0-Ta2O5phase was achieved by heating at 1700 °C for 10 h, while for com-positions with x = 0.05–0.08, both the L0-Ta2O5 phase and impurityFIG. 3. HTXRD patterns of a sample consisting of mixed Ta2O5 and ZrO2 powders as a function of the temperature in the range of 28–1450 °C. The lower-right figure indi-cates the temperature profile as a function of time during the measurement. The upper figure is the reference diffraction patterns of L-Ta2O5 (ICSD No. 9112), H-Ta2O5(ICSD No. 157683), L0-Ta2O5[(Zr0.10Ta0.90)2O4.90 measured at 27 °C19], and L00-Ta2O5 [(Zr0.10Ta0.90)2O4.90 measured at 800 °C19]. The diffraction peaks observed near2θ = 40° and 45° originate from the Pt heater used during the measurement.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-4© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.60893/figshare.jap.c.8322903https://pubs.aip.org/aip/japH-Ta2O5 phase were obtained, as shown in Fig. 1. According to theHTXRD measurement, the L0-Ta2O5 phase forms through asequence of phase transitions during cooling: first from theH-Ta2O5 phase to the L00-Ta2O5 phase and then from the L00-Ta2O5phase to the L0-Ta2O5 phase. Notably, if the H-Ta2O5 phase cannottransition into the L00-Ta2O5 phase at high temperatures, it isquenched to room temperature. Therefore, the formation of theH-Ta2O5 phase for x = 0.05–0.08 might be derived from the incom-plete phase transition to the L00-Ta2O5 phase. This phase transitionis kinetically controlled, as indicated by the requirement for a cons-tant holding time of 1150 °C to complete the transition. The peakintensity of the H-Ta2O5 phase increases with decreasing ZrO2content, indicating that the transition of the H-Ta2O5 phase intothe L00-Ta2O5 phase slows with decreasing Zr content. In otherwords, a lower Zr content leads to a higher activation barrier forthis phase transition. Therefore, additional heat treatment isrequired to overcome this activation barrier to obtain the L0-Ta2O5phase for the compositions with x = 0.05–0.08, as shown in Fig. 2.V. COMPOSITIONAL DEPENDENCE OF THE PHYSICALPROPERTIESA. Lattice parametersThe lattice parameters of the L0-Ta2O5 phase for x = 0.05–0.13were refined using the Le Bail method.22 Figure 4 shows the com-positional dependence of the lattice parameters. All data valuesincluding errors are shown in Table S1 in the supplementarymaterial. The lattice parameters exhibit an increase in a and adecrease in b as the amount of ZrO2 substitution increases withinthe range x = 0.05–0.11, where the L0 -Ta2O5 phase is observed as asingle phase. Moreover, c decreases in the composition range ofx = 0.05–0.08, while it remains nearly constant between x = 0.08and 0.11. The unit-cell volume also decreased upon increasing xfrom 0.05 to 0.06, followed by an increase when x was raised fromFIG. 5. (a) Compositional and temperature dependence of real part of dielectric constant of (ZrxTa1−x)2O5−x (x = 0.05–0.10) measured at 1 MHz. (b) Compositional depen-dence of phase transition temperature between L0-Ta2O5 and L00-Ta2O5 for (ZrxTa1−x)2O5−x (x = 0.05–0.10).FIG. 4. Compositional dependence of the unit-cell volume and lattice parame-ters for (ZrxTa1−x)2O5−x (x = 0.05–0.13). Filled red circles represent dataobtained from samples subjected to sequential heating at 1700 °C followed by1200 °C. Filled black squares represent data from samples heated only at1700 °C.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-5© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://pubs.aip.org/aip/jap0.08 to 0.11. At x = 0.12–0.13, where the impurity Zr6Ta2O17 isproduced as shown in Fig. 1, a and b remain constant, while cincreases. These variations in lattice parameters would be derivedfrom not only the Zr substitution effect but also the formation ofoxygen vacancies. However, a comprehensive structural analysis ofeach composition has not been conducted. A detailed structuraldetermination by neutron diffraction measurements will revealthese specific changes in the future.B. Dielectric constantFigure 5(a) presents the temperature dependence of the realpart of the dielectric constant measured at 1 MHz for x = 0.05–0.10in (ZrxTa1−x)2O5−x. The dielectric constant and tanδ measured atvarying frequencies are shown in Fig. S5 in the supplementarymaterial. With increasing ZrO2 content, the dielectric constant at30 °C rises from 42 for x = 0.05 to 50 for x = 0.10. The dielectricconstant exhibits a temperature dependence, reaching a maximumvalue at the specific temperature associated with the phase transi-tion from L0-Ta2O5 to L00-Ta2O5.19 Figure 5(b) shows the composi-tional dependence of the phase transition temperatures determinedfrom the maximum value of the dielectric constant displayed inFig. 5(a). The phase transition temperature decreased linearly from370 °C for x = 0.05 to 100 °C for x = 0.10 with increasing ZrO2content, corresponding to a reduction of ∼50 °C for each 0.01increment in x.C. Negative thermal expansionSynchrotron HTXRD measurements were conducted atSPring-8 (BL13XU, λ = 0.375 695 nm) to investigate the tempera-ture evolution of the L0-Ta2O5-type phase (x = 0.05). This studyexamined a temperature range of −173 to 800 °C, as shown inFig. S6 in the supplementary material. The phase transition tem-perature for the transformation of the L0-Ta2O5 into the L00-Ta2O5phase, determined from the HTXRD data, is ∼427 °C for x = 0.05.By contrast, a previous study reported a transition temperature of127 °C for x = 0.10.19 This indicates that samples with a lowerZrO2 substitution exhibit a higher phase transition temperature,consistent with the findings from dielectric constantmeasurements.The obtained lattice parameters are shown in Fig. 6 and datavalues including errors are shown in Table S2 in thesupplementary material. The data for x = 0.05 are represented byred circles, while the data for x = 0.10 are plotted as black squaresfor reference.19 For the samples with x = 0.05, the parameter bincreases, while the parameters a and c decrease up to 400 °C.Above this temperature, all parameters gradually increase.Therefore, the phase transition temperature is estimated to beapproximately 427 °C based on HTXRD measurements. Theobserved difference of about 50 °C in phase transition temperaturebetween dielectric constant and HTXRD measurements arisesfrom the distinct phenomena each technique detects. Dielectricconstant measurements are highly sensitive to polarization fluctu-ations, whereas XRD measurements detect changes in the averagecrystal structure. Since polarization fluctuations become apparentbefore structural changes, the phase transition temperaturedetermined by dielectric constant measurements tends to be lowerthan that determined by XRD.The lattice volume decreases between 327 and 427 °C in thevicinity of the phase transition temperature, indicating negative vol-umetric thermal expansion (NTE). Here, the volumetric thermalexpansion coefficient, αv , was calculated as αv ¼ 1V0� �V�V0T�T0� �,where T represents the final temperature, T0 the initial temperature,and the corresponding volumes are V and V0, respectively. Atx = 0.05, αv was calculated to be −2.06 × 10–5/K from the latticevolumes V0 and V at 327 and 427 °C in Table S2 in thesupplementary material. This value is comparable to that ofZrW2O8, a well-known NTE material with αv =− 2.7 × 10–5/K.23FIG. 6. Temperature dependence of the unit-cell volume and lattice parametersfor (ZrxTa1−x)2O5−x (x = 0.05 and 0.10). The filled red circles and black squaresrepresent the data for compositions of x = 0.05 and 0.10, respectively. Theerrors are within the circle and square data points.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-6© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.60893/figshare.jap.c.8322903https://pubs.aip.org/aip/japFor x = 0.10, the thermal expansion coefficient αv was −1.09 × 10–6/Kin the temperature range of 77–127 °C. The sample with x = 0.05had a wider NTE temperature range than x = 0.10 along with alarger thermal expansion coefficient.19 The unit-call volume sig-nificantly shrinks before the phase transition from L0-Ta2O5 toL00-Ta2O5 when the amount of ZrO2 substitution is low. Uniquely,the NTE temperature range of the L0-Ta2O5 type (ZrxTa1−x)2O5−xvaried by ∼200 °C (∼473 K) through composition control, makingit a promising candidate material for negative thermal expansionapplications.V. CONCLUSIONIn this study, we investigated the formation process andcomposition-dependent properties of L0-Ta2O5 type (ZrxTa1−x)2O5−x.HTXRD measurements indicated that the mixture of starting materi-als, ZrO2 and Ta2O5, formed the H-Ta2O5 phase upon heating above1360 °C. As the H-Ta2O5 phase cooled, it changed into the L00-Ta2O5phase, followed by the formation of the L0-Ta2O5 phase. A single-phase L0-Ta2O5 was obtained within the composition range ofx = 0.05–0.11. The material properties depended on the compositionin the following ways:• The lower the amount of ZrO2 substitution, the slower the phasetransition from the H-Ta2O5 phase to the L00-Ta2O5 phase,making the H-Ta2O5 phase more likely to be quenched.• With increasing extent of ZrO2 substitution, the lattice parame-ters a and c increase, while b decreases.• The dielectric constant at 30 °C increases with increasing extentof ZrO2 substitution, ranging from εr = 42 to 50. The tempera-ture dependence of the dielectric constant revealed a maximumthat corresponds to the phase transition from L0-Ta2O5 toL00-Ta2O5. This transition temperature decreases by ∼50 °C forevery 0.01 increase in x.• The L0-Ta2O5 phase shows NTE behavior near the phase transi-tion temperature. Specifically, as the extent of ZrO2 substitutiondecreases, the negative volume expansion coefficient increases.For the x = 0.10 sample, the thermal expansion coefficient, αv, inthe 77–127 °C range is −1.09 × 10−6/K, and for the x = 0.05sample, αv is −2.06 × 10−5/K in the 327–427 °C range.In conclusion, (ZrxTa1−x)2O5−x (x = 0.05–0.10), which formsthe L0-Ta2O5 phase, is a material that allows for control over thedielectric constant, phase transition temperature, and thermalexpansion behavior based on its composition.SUPPLEMENTARY MATERIALSee the supplementary material for the predicted phase transi-tion pathway to the L0-Ta2O5 phase from the starting materials ofZrO2 and L-Ta2O5, SHG signals for (ZrxTa1−x)2O5−x (x = 0.05–0.11), temperature dependence of HTXRD profiles of(Zr0.10Ta0.90)2O4.90, lattice parameters with error bars for sampleswith x values ranging from 0.05 to 0.13 as well as for x = 0.05 and0.10 during HTXRD measurements, and temperature dependenceof real part of the dielectric constant for (ZrxTa1−x)2O5−x (x = 0.05–0.10) and raw synchrotron XRD data of (Zr0.05Ta0.95)2O4.95.ACKNOWLEDGMENTSThe authors thank Dr. Shintaro Kobayashi, Dr. ShogoKawaguchi, and facility members at Japan Synchrotron RadiationResearch Institute (JASRI) for the help with the SXRD experiments.The SXRD experiments were conducted at the BL13XU beamline atSPring-8 with the approval of JASRI (Proposal No. 2024B1693). Thisstudy was partly supported by Japan Society for the Promotion ofScience (JSPS) KAKENHI, Grant Nos. JP23H02050, JP23K26743, andJP25K17643, research granted from Murata Science and EducationFoundation, and from the Yoshishige Abe Memorial Fund.AUTHOR DECLARATIONSConflict of InterestThe authors have no conflicts to disclose.Author ContributionsSuzuka Udagawa: Conceptualization (lead); Investigation (lead);Visualization (lead); Writing – original draft (lead); Writing –review & editing (equal). Takanori Mimura: Conceptualization(equal); Funding acquisition (equal); Investigation (equal);Supervision (equal); Writing – original draft (equal); Writing –review & editing (equal). Tetsuhiro Katsumata: Investigation(equal); Writing – review & editing (supporting). YoshitakaMatsushita: Investigation (supporting); Writing – review & editing(supporting). Takashi Mochiku: Investigation (supporting);Writing – review & editing (supporting). Yoshiyuki Inaguma:Funding acquisition (equal); Investigation (equal); Supervision(equal); Writing – original draft (equal); Writing – review &editing (equal).DATA AVAILABILITYThe data that support the findings of this study are availablefrom the corresponding author upon reasonable request.REFERENCES1R. J. Cava, J. Mater. Chem. 11(1), 54–62 (2001).2R. M. Wallace and G. D. Wilk, Crit. Rev. Solid State Mater. Sci. 28(4), 231–285(2003).3M. Ghevondyan, M. Davtyan, and M. Aghayan, Discov. Mater. 5, 43 (2025).4K. Zhang, L. Chao, and J. Zhou, Materials 11(8), 1332 (2018).5Q. Li, F.-Z. Yao, Y. Liu, G. Zhang, H. Wang, and Q. Wang, Annu. Rev. Mater.Res. 48(1), 219–243 (2018).6J. P. B. Silva, K. C. Sekhar, H. Pan, J. L. MacManus-Driscoll, and M. Pereira,ACS Energy Lett. 6(6), 2208–2217 (2021).7H.-F. Zhang, B.-Y. Ning, T.-C. Weng, and X.-J. Ning, J. Am. Ceram. Soc.104(12), 6413–6423 (2021).8M. Hiratani, S. Kimura, T. Hamada, S. Iijima, and N. Nakanishi, Appl. Phys.Lett. 81(13), 2433–2435 (2002).9F. Rubio, J. Denis, J. M. Albella, and J. M. Martinez-Duart, Thin Solid Films90(4), 405–408 (1982).10J. L. Waring and R. S. Roth, J. Res. Natl. Bur. Stand. A Phys. Chem. 72A,175–186 (1967).11G. L. Brennecka, D. A. Payne, P. Sarin, J. M. Zuo, W. M. Kriven, andH. Hellwig, J. Am. Ceram. Soc. 90(9), 2947–2953 (2007).12X. Q. Liu, X. D. Han, Z. Zhang, L. F. Ji, and Y. J. Jiang, Acta Mater. 55(7),2385–2396 (2007).Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-7© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.60893/figshare.jap.c.8322903https://doi.org/10.1039/b003681lhttps://doi.org/10.1080/714037708https://doi.org/10.1007/s43939-025-00225-7https://doi.org/10.3390/ma11081332https://doi.org/10.1146/annurev-matsci-070317-124435https://doi.org/10.1146/annurev-matsci-070317-124435https://doi.org/10.1021/acsenergylett.1c00313https://doi.org/10.1111/jace.18026https://doi.org/10.1063/1.1509861https://doi.org/10.1063/1.1509861https://doi.org/10.1016/0040-6090(82)90545-4https://doi.org/10.6028/jres.072a.018https://doi.org/10.1111/j.1551-2916.2007.01790.xhttps://doi.org/10.1016/j.actamat.2006.11.031https://pubs.aip.org/aip/jap13N. C. Stephenson and R. S. Roth, Acta Crystallogr. B 27(5), 1037–1044 (1971).14R. J. Cava, J. J. Krajewski, W. F. Peck, Jr., and G. L. Roberts, J. Appl. Phys.80(4), 2346–2348 (1996).15R. J. Cava and J. J. Krajewski, J. Appl. Phys. 83(3), 1613–1616 (1998).16Y. Wang and Y.-J. Jiang, Mater. Sci. Eng. B 99(1–3), 221–225 (2003).17R. F. Cava, W. F. Peck, Jr., and J. J. Krajewski, Nature 377(6546), 215–217(1995).18D. Makovec, J. M. Zuo, R. Twesten, and D. A. Payne, J. Solid State Chem.179(6), 1782–1791 (2006).19T. Mimura, S. Udagawa, and Y. Inaguma, Appl. Phys. Lett. 126(15), 151902 (2025).20Y. Inaguma, A. Aimi, D. Mori, T. Katsumata, M. Ohtake, M. Nakayama, andM. Yonemura, Inorg. Chem. 57(24), 15462–15473 (2018).21P. Luo, X. Wu, W. Xiao, F. Zhang, Y. Wang, D. Huang, and Y. Du, J. Am.Ceram. Soc. 105(1), 668–686 (2022).22A. Le Bail, H. Duroy, and J. L. Fourquet, Mater. Res. Bull. 23(3), 447–452(1988).23J. S. O. Evans, T. A. Mary, T. Vogt, M. A. Subramanian, and A. W. Sleight,Chem. Mater. 8(12), 2809–2823 (1996).Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 104102 (2026); doi: 10.1063/5.0312634 139, 104102-8© Author(s) 2026 12 March 2026 00:17:58https://doi.org/10.1107/S056774087100342Xhttps://doi.org/10.1063/1.363068https://doi.org/10.1063/1.366873https://doi.org/10.1016/S0921-5107(02)00539-1https://doi.org/10.1038/377215a0https://doi.org/10.1016/j.jssc.2006.03.014https://doi.org/10.1063/5.0259303https://doi.org/10.1021/acs.inorgchem.8b02767https://doi.org/10.1111/jace.18079https://doi.org/10.1111/jace.18079https://doi.org/10.1016/0025-5408(88)90019-0https://doi.org/10.1021/cm9602959https://pubs.aip.org/aip/jap