# Fileset
[MDR-Pb-Cu-BCSO-Alexandra-text.pdf](https://mdr.nims.go.jp/filesets/8e2431cc-74a7-4a54-b6cc-d7294d5785ae/download)
## Creator
Aleksandra Khanina, Andrei Novitskii, Daria Pashkova, Andrei Voronin, [Takao Mori](https://orcid.org/0000-0003-2682-1846), Vladimir Khovaylo
## Rights
[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)
## Other metadata
[Thermoelectric properties of Bi1−xPb xCu1−xSeO oxyselenides](https://mdr.nims.go.jp/datasets/a13ef739-e4dc-4554-a785-0f29ac11d6cc)
## Fulltext
Thermoelectric Properties of Bi1−xPbxCu1−xSeO Oxyselenides†Aleksandra Khanina,a,∗ Andrei Novitskii,b Daria Pashkova,b Andrei Voronin,a Takao Mori,b,c and Vladimir Khovayloa,d,∗In this work, Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and 0.08) were synthesized by solid-state reaction followed by spark plasmasintering. The effect of simultaneous Bi to Pb substitution and Cu vacancies introduction on thermoelectric properties was investigatedsystematically. The power factor was significantly enhanced, contributing to the increase in the zT value. As a result, the zTmax of0.75 at 773 K was obtained for Bi0.94Pb0.06Cu0.94SeO sample. To reveal the factors constraining the zTmax in Bi1−xPbxCuSeO-basedoxyselenides, a careful analysis of literature data was performed. We highlighted that in Pb-doped oxyselenides, the power factor isalmost independent of the synthesis technique, while the lattice thermal conductivity is the main property determining zTmax and ishighly affected by the synthesis method.IntroductionOver the past few decades, global energy consumption has dra-matically increased, with about two-thirds of produced energybeing lost as waste heat. Therefore, technologies capable of ef-ficiently recuperating this heat are crucial to achieving sustain-able development goals. In this context, thermoelectric materi-als, which enable direct conversion of waste heat into electricity,have attracted increasing attention from the scientific commu-nity.1 The efficiency of thermoelectric energy conversion dependson the material’s transport properties and is determined by thethermoelectric figure of merit zT =α2σT/κtot , where α, σ , T, andκtot represent the Seebeck coefficient, electrical conductivity, ab-solute temperature, and total thermal conductivity, respectively.2Currently, p-type BiCuSeO-based oxyselenides are among themost efficient oxygen-containing thermoelectric materials.3 Bi-CuSeO crystallizes in a tetragonal layered ZrCuSiAs structure typewith a P4/nmm space group and two formula units per unit cell.The crystal structure consists of alternately stacked insulating(Bi2O2)2+ layers (so-called "charge reservoir" layers) and con-ducting (Cu2Se2)2− layers along the c-axis.4,5 These BiCuSeO-based oxyselenides have drawn interest as thermoelectric materi-als, primarily due to their intrinsically low thermal conductivity,not exceeding 1.5 W m−1 K−1 at room temperature, coupled witha relatively high Seebeck coefficient.6 The main focus in enhanc-ing zT of BiCuSeO is optimizing the charge carrier concentrationn, which significantly boosts the power factor (α2σ) by balancingα and σ values. Employing this approach in BiCuSeO-based oxy-selenides has yielded zTmax > 1.2 at T ≥ 773 K achieved throughthe substitution of bismuth with various elements.7,8However, as identified by Ioffe over six decades ago, the effi-cient thermoelectric material requires not only an optimal chargeconcentration nopt but also a maximized ratio of charge carriermobility µ to lattice thermal conductivity κlat , as this determinedthe zTmax achievable at a given temperature and charge carrieraAcademic Research Center for Energy Efficiency, National University of Science andTechnology MISIS, Leninsky Av. 4, Moscow, 119049, Russia.bInternational Center for Materials Nanoarchitectonics (WPI-MANA), National Insti-tute for Materials Science (NIMS), 1-1 Namiki, Tsukuba, Ibaraki, 305-0044, Japan.cGraduate School of Pure and Applied Sciences, University of Tsukuba, 1-1-1 Tenn-odai, Tsukuba, Ibaraki, 305-8573, Japan.dBelgorod State University, Pobedy St. 85, Belgorod, 308015, Russia.∗E-mail: khanina.as@misis.ru, khovaylo@misis.ru.†Electronic supplementary information available.concentration.2 While optimizing charge carrier concentration inBiCuSeO is feasible through heterovalent Bi substitution, meth-ods to maximize µ/κlat are less straightforward. One of the pecu-liar approaches was recently indicated to be beneficial for keep-ing a high n/µ ratio by promoting the carriers’ delocalization be-tween the charge reservoir and conducting layers via introduc-ing Bi and/or Cu vacancies. Although interlayer delocalization ofcarriers is favorable for interlayer charge transfer, conventionalBi substitution is still necessary to provide a sufficient number ofcharge carriers for diffusion.9,10 In this regard, Pb has been iden-tified as the most effective dopant for oxyselenides in terms of thenumber of charge carriers per one Pb atom introduced into thesystem during doping.11 Concurrently, copper vacancies are thepredominant defect in BiCuSeO responsible for its native p-typeself-doping.12,13In our study, we synthesized a series of samples with the nom-inal composition of Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and0.08) to achieve simultaneous optimization of charge carrier con-centration while maintaining high values of their mobility by com-bining the substitution of bismuth with lead (PbBi) and introduc-ing extra copper vacancies (VCu). Bi1−xPbxCuSeO-based oxyse-lenides are among the most studied, and the influence of Bi toPb substitution on transport properties has been thoroughly in-vestigated. We demonstrated that the key factor affecting thethermoelectric performance of Pb-doped BiCuSeO and the vari-ability reported in zTmax values is, in fact, the µ/κlat ratio. Atthe same time, the power factor is predominantly influencedby the PbBi concentration and is weakly dependent on the con-centration of vacancies and/or the second dopant. Ultimately,we observed an almost fourfold increase in the power factor forBi1−xPbxCu1−xSeO with x = 0.06 and 0.08 at 773 K. The µw/κlatratio doubled with an increase from x = 0 to x = 0.02, yet furtherincrease in x did not yield an increase in µw/κlat mainly due tothe increase in κlat for x > 0.02. Ultimately, a notable increase zTby a factor of three was realized, and a zTmax = 0.75 at 773 K forBi0.94Pb0.06Cu0.94SeO was achieved.Materials and MethodsCompounds with the nominal chemical compositions ofBi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and 0.08) were synthesizedas reported elsewhere14 using high purity CuO, Bi, Se, and PbOpowders as rhe raw materials. The obtained products were con-solidated through spark plasma sintering (SPS) at 903 K for 5 minunder an axial pressure of 50 MPa. Phase composition analysis1–5 | 1was carried out through powder X-ray diffraction (PXRD) usinga DRON-3 diffractometer (IC Bourevestnik, Russia) with a Cu-Kαradiation (λ = 1.54178 Å). The PXRD patterns were refined bythe Rietveld method with the self-developed software package.15Scanning electron microscopy (SEM) and energy-dispersive X-rayspectroscopy (EDS) were performed using a Vega 3SB SEM (Tes-can, Czech Republic) equipped with an x-act EDS detector (Ox-ford Instruments, UK). The thermal diffusivity χ was measuredin the axial direction of disk-shaped samples of �10 × 1 mm2using a laser flash analyzer (LFA 467 Hyperflash, Netzsch, Ger-many). The total thermal conductivity κtot was calculated usingthe formula κtot = χCpd, where Cp represents the specific heat ca-pacity which was estimated using the comparison method (withpyroceram-9060 as a reference sample), and is in a good agree-ment with the Cp value calculated by the Debye model. Thedensity d was determined using the Archimedes method. TheSeebeck coefficient α and the electrical conductivity σ were mea-sured in the radial direction of bar-shaped specimens with dimen-sions of 10× 3× 1 mm3 using a commercial apparatus (ZEM-3,Advance Riko Inc., Japan) under a partial He atmosphere from300 to 773 K. The uncertainty in σ and α was 5% and 4%, re-spectively, along with 8% in κtot . The overall uncertainty in zTwas estimated to be less than 17%. The error bars are not shownfor clarity and to enhance the visual presentation of the figures.Results and DiscussionThe PXRD patterns of Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and0.08) samples are presented in Fig. 1a. All major reflections canbe indexed to the tetragonal BiCuSeO phase (PDF#01-076-6689)with a ZrCuSiAs structure type and P4/mmm space group. A mi-nor amount of the Bi2O3 (PDF#01-088-2043) secondary phase(≤ 4.5 vol.%) was detected in all the samples, indicating signifi-cant Cu deficiency as expected.12,13,16,17 In the case of dual dop-ing, the evolution of lattice parameters can be affected by twocompeting factors. On the one hand, Cu deficiency tends to de-crease both lattice parameters a and c,17 while on the other hand,a larger ionic radius of Pb2+ (1.19 Å) as compared to that of Bi3+(1.03 Å)18 should result in an increase of lattice parameters.19,20Our experimental results reveal that, even with a simultaneousincrease in the concentration of PbBi and VCu, both lattice param-eters a and c exhibit a gradual increase (Fig. 1b), resulting inthe expansion of the unit cell volume. However, it is noteworthythat the effect of copper vacancies is noticeable, and for sampleswith the same nominal concentration x, the lattice parameters ofdually doped specimens are slightly smaller than those of singlePb-doped ones.19–21Figure 2 displays an SEM micrograph, providing an overviewof the microstructure representative of all samples. Randomlyarranged platelet grains are stacked densely, which is reflected bythe high relative densities for all samples (≥ 95% of theoreticaldensity). The thickness of the grains ranges from 300 to 600 nm.EDS analysis confirms that the actual compositions of the samplesclosely align with the nominal ones (Table S1).Figure 3a shows the temperature dependence of electricalconductivity σ of Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and0.08) samples. For all the samples, σ gradually decreases with2 0 4 0 6 0 8 0 1 0 0x = 0x = 0 . 0 2x = 0 . 0 6x = 0 . 0 82 � ( d e g r e e )Intensity (arb. units)( a )( b )0 . 0 0 0 . 0 2 0 . 0 4 0 . 0 6 0 . 0 83 . 9 2 53 . 9 3 03 . 9 3 5N o m i n a l c o n t e n t o f P b B i a n d V C u ( x )a (Å)8 . 9 59 . 0 09 . 0 59 . 1 0c (Å)Figure 1 (a) Powder XRD patterns of the Bi1−xPbxCu1−xSeO (x = 0,0.02, 0.06, and 0.08) samples. Bragg’s reflections for the BiCuSeOphase are indicated by gray ticks on the top part of the figure. Diffractionpeak corresponding to the Bi2O3 secondary phase is indicated by ablack solid triangle (▼). (b) Lattice parameters (a and c) as a function ofnominal concentration of PbBi and VCu.5 μmFigure 2 SEM micrograph of the fractured cross-section ofBi0.94Pb0.06Cu0.94SeO sample.the increase of temperature, exhibiting a metal-like behavioras expected for heavily doped semiconductors.22 As the con-tent of PbBi and VCu increases from 0 to 0.08, σ is vastly im-proved from ∼30 Ω−1 cm−1 for the pristine BiCuSeO sample to∼450 Ω−1 cm−1 for Bi0.92Pb0.08Cu0.92SeO at room temperature.Considering only a slight improvement in the weighted mobility2 | 1–502 0 04 0 06 0 0 ( a )� (�-1 cm-1 )3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 01 0 02 0 03 0 0 x = 0 x = 0 . 0 2 x = 0 . 0 6 x = 0 . 0 8( b )T ( K )� (�V K-1 )0 . 60 . 91 . 21 . 5( c )�tot (W m-1 K-1 )3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 00 . 00 . 20 . 40 . 60 . 8zTT ( K )( d )Figure 3 Temperature dependence of the (a) electrical conductivity σ , (b) Seebeck coefficient α, (c) total thermal conductivity κtot , and (d) the figureof merit zT for Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and 0.08) samples.µw (µw ≈ µ(m∗d/me)3/2, where m∗d is the density of states effectivemass, and me is the electron mass) as shown in Fig. S1, the mainreason for such a significant enhancement of the electrical con-ductivity is from the increase in the charge carriers concentration,which is also in agreement with previous reports.21,23 Further-more, the carriers density increase originates mostly from the sub-stitution of Bi3+ with Pb2+, as even introducing only dual vacan-cies VBi and VCu in BiCuSeO led to an order of magnitude lowerconductivity in comparison with Pb-substituted ones.10 Nonethe-less, as discussed in the Introduction section, the introduction ofvacancies plays not a primary but a complementary role, leadingto a more pronounced delocalization of charge carriers betweenthe layers, thereby promoting interlayer charge transfer.9 This,in turn, allows for the retention of higher charge carrier mobil-ity values at the same level of charge carrier concentration (seeFig. 12 in Ref. [11]).Figure 3b illustrates the Seebeck coefficient α as a functionof temperature for Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and0.08) samples. The positive Seebeck coefficient for all samplesin the entire temperature range confirms p-type conduction, in-dicating that holes are the majority charge carriers. The high-est Seebeck coefficient throughout the investigated temperaturerange is observed in nominally undoped BiCuSeO. However, withthe increase in PbBi and VCu concentration, α gradually decreasesconsistently with the changes in σ (Fig. S2). Considering the re-lationship between the Seebeck coefficient and carrier concentra-tion in semiconductors,2 such a simultaneous change in electri-cal conductivity and the Seebeck coefficient further underscoresthat it is primarily driven by a significant increase in charge car-rier concentration, as was discussed previously. Consequently, at773 K, α reduced from 352 µV K−1 for x = 0 to 182 µV K−1 forx = 0.08. The achieved optimal balance between σ and α resultsin a significant enhancement in the power factor (α2σ), showingmore than a fourfold increase at 773 K, from 1.8 µW cm−1 K−2to 7.7 µW cm−1 K−2, as the PbBi and VCu content is raised from x= 0 to 0.08 (Fig. 4a, Fig. S3).Figure 3c shows the temperature-dependent total thermal con-ductivity κtot(T) of Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and0.08) samples. For the pristine BiCuSeO, κtot decreases from1.17 W m−1 K−1 at 300 K to 0.75 W m−1 K−1 at 773 K. Fol-lowing simultaneous Bi for Pb substitution and Cu vacancies in-troduction, κtot initially decreases (x = 0.02) and then rises to0.93 W m−1 K−1 at 773 K (x = 0.08). This can be understoodby considering evolution in both the lattice κlat and electronicκel components of total thermal conductivity. The κlat is calcu-lated from κlat = κtot −κel , where κel can be determined throughthe Wiedemann–Franz law, expressed as κel = LσT, with L rep-resenting the Lorentz number.2 The Lorentz number is estimatedwith sufficient accuracy within the single parabolic band modelwith acoustic phonon scattering from experimental α values asL = 1.5+ exp{−|α|/116}.24 Owing to a significant enhancementin σ , the changes in κtot become more apparent. In pristine Bi-CuSeO, κel contributes to approximately 2% of the total, but thiscontribution rises up to over 30% with an increase in PbBi andVCu concentration (Fig. S4). Analysis of the κlat(T) reveals thatpoint defects are the dominant phonon scattering mechanism, ev-1–5 | 3idenced by a decrease of κlat with temperature as T−0.5 for all thesamples (Fig. S5). However, for the samples with x > 0.02 κlatslightly increases but still remains lower than that of pristine Bi-CuSeO. This increase could be attributed to an increase in thegrain size, sound velocities, and/or Pb–O bond strength due tothe introduction of PbBi and VCu.25Combining the electrical and thermal transport properties, thefigure of merit zT as a function of temperature is calculated anddisplayed in Fig. 3d. For Bi1−xPbxCu1−xSeO samples (x > 0), thezT values are enhanced over the entire test temperature rangeby more than 3 times. This improvement is mainly attributed toa significant boost in the power factor, leading to a zTmax valueof 0.75 at 773 K for Bi0.94Pb0.06Cu0.94SeO. On the one hand, anearly threefold increase in zT was achieved compared to un-doped BiCuSeO. On the other hand, we must admit that this isnot the highest zT value achieved for Bi1−xPbxCuSeO-based oxy-selenides. Furthermore, a detailed analysis of literature data re-veals that reported zT values for Pb-doped BiCuSeO are scatteredbetween 0.56 and 1.05 at T = 773 K and x = 0.06. Consideringthat the power factor is practically independent of the synthesismethod (microstructure: grain size and shape, defects, etc.), asshown in Fig. 4a, the main factor determining zT value is the lat-tice thermal conductivity (Fig. S6). After careful investigation, itcan be concluded that the highest zT values were obtained insamples where lattice thermal conductivity was suppressed byabout twice compared to the glassy limit (≈ 0.59 W m−1 K−1).Interestingly, grain size is not the determining factor in reducingκlat . The lowest κlat values were reported for samples synthesizedby methods like mechanochemical synthesis,26 self-propagatinghigh-temperature synthesis,27,28 and high-energy ball milling,16all of them leading to the formation of various defects, such asnanodots,16 amorphous regions,27,28 etc.ConclusionsThe thermoelectric properties of the Bi1−xPbxCu1−xSeO (x = 0,0.02, 0.06, and 0.08) samples prepared by solid-state reactionroute followed by spark plasma sintering have been studied. Thecombination of Pb for Bi substitution and Cu vacancies introduc-tion increases the electrical conductivity while moderately de-creasing the Seebeck coefficient. The intrinsically low thermalconductivity of BiCuSeO is noticeably decreased by the intro-duction of PbBi and VCu of a small concentration (x = 0.02)and increased with further increase in x. The optimum balancebetween the Seebeck coefficient and electrical conductivity wasachieved for x = 0.06, while for x > 0.02, the µw/κlat ratio es-sentially does not increase and remains at the same level as for x= 0.02, limiting the potentially achievable zTmax. As a result, thevalue of zT was eventually increased by more than 3 times com-pared to the undoped BiCuSeO, and zTmax = 0.75 at 773 K wasachieved for Bi0.94Pb0.06Cu0.94SeO. By careful analysis of the liter-ature data and comparing it with our results, we revealed that forBi1−xPbxCuSeO-based oxyselenides, the main factor determiningzTmax is the κlat . We believe that highlighting the mechanisms de-termining zT value in doped BiCuSeO will help future researchersto form more effective synthesis/doping strategies. x = 0 x = 0 . 0 2 x = 0 . 0 6 x = 0 . 0 8024681 01 . 0 1 . 5 2 . 0 2 . 50246�2 � (�W cm-1 K-2 )( a )�w/�lat (103 cm3 K V-1 J-1 )l n �@ 7 7 3 K( b )Figure 4 (a) Power factor α2σ and (b) µw/κlat ratio as functions of lnσfor the Bi1−xPbxCu1−xSeO (x = 0, 0.02, 0.06, and 0.08) samples.Literature data for other Bi1−xPbxCuSeO-based oxyselenides are alsoshown for comparison (Chen et al., 23 Gu et al., 29,30 Lan et al., 16 Lei etal., 31 Li et al., 32 Liang et al., 33 Liu et al., 26 Pan et al., 34 Ren et al., 27,28Xu et al., 35 Zhu et al. 9). All displayed data points correspond to valuesobtained at 773 K.AcknowledgementsThe study was carried out with financial support from the Rus-sian Science Foundation, project No. 19-79-10282. T.M. andA.N. acknowledge JST Mirai JPMJMI19A1 and JST SPRING JP-MJSP2124.CRediT StatementAleksandra Khanina: Investigation, Formal analysis, Visualiza-tion, Writing – Original draft, Writing – Review & editing; An-drei Novitskii: Conceptualization, Formal analysis, Investigation,Methodology, Visualization, Writing – Original draft, Writing –Review & editing; Daria Pashkova: Data curation, Investigation,Formal analysis, Visualization; Andrei Voronin: Project admin-istration, Funding acquisition; Takao Mori: Resources, Fundingacquisition, Supervision, Writing – Review & editing. VladimirKhovaylo: Resources, Supervision, Writing – Review & editing.Conflicts of InterestThere are no conflicts to declare.Data Availability StatementData will be made available on request.References1 D. Sarma, Essential considerations for reporting thermoelectricproperties, 2021.2 A. F. Ioffe, Semiconductor Thermoelements, and ThermoelectricCooling, Infosearch, 1957.3 R. Freer, D. Ekren, T. Ghosh, K. Biswas, P. Qiu, S. Wan,4 | 1–5L. Chen, S. Han, C. Fu, T. Zhu et al., Journal of Physics: En-ergy, 2022, 4, 022002.4 L. Kholodkovskaya, L. Akselrud, A. Kusainova, V. Dolgikh andB. Popovkin, Materials Science Forum, 1993, pp. 693–696.5 A. Kusainova, P. Berdonosov, L. Akselrud, L. Kholodkovskaya,V. Dolgikh and B. Popovkin, Journal of Solid State Chemistry,1994, 112, 189–191.6 L.-D. Zhao, J. He, D. Berardan, Y. Lin, J.-F. Li, C.-W. Nan andN. Dragoe, Energy & Environmental Science, 2014, 7, 2900–2924.7 Y. Liu, L.-D. Zhao, Y. Zhu, Y. Liu, F. Li, M. Yu, D.-B. Liu, W. Xu,Y.-H. Lin and C.-W. Nan, Advanced Energy Materials, 2016, 6,1502423.8 Y.-L. Pei, H. Wu, D. Wu, F. Zheng and J. He, Journal of theAmerican Chemical Society, 2014, 136, 13902–13908.9 H. Zhu, Z. Li, C. Zhao, X. Li, J. Yang, C. Xiao and Y. Xie,National Science Review, 2021, 8, nwaa085.10 Z. Li, C. Xiao, S. Fan, Y. Deng, W. Zhang, B. Ye and Y. Xie,Journal of the American Chemical Society, 2015, 137, 6587–6593.11 A. Novitskii, V. Khovaylo and T. Mori, Nanobiotechnology Re-ports, 2021, 16, 294–307.12 M. Y. Toriyama, J. Qu, G. J. Snyder and P. Gorai, Journal ofMaterials Chemistry A, 2021, 9, 20685–20694.13 D. Goury, E. Amzallag, D. Bérardan and J. Creuze, The Journalof Physical Chemistry C, 2022, 126, 5960–5969.14 A. Novitskii, I. Serhiienko, S. Novikov, Y. Ashim, M. Zheleznyi,K. Kuskov, D. Pankratova, P. Konstantinov, A. Voronin, O. A.Tretiakov et al., ACS Applied Energy Materials, 2022, 5, 7830–7841.15 E. Shelekhov and T. Sviridova, Metal Science and Heat Treat-ment, 2000, 42, 309–313.16 J.-L. Lan, Y.-C. Liu, B. Zhan, Y.-H. Lin, B. Zhang, X. Yuan,W. Zhang, W. Xu and C.-W. Nan, Advanced Materials, 2013,25, 5086–5090.17 Y. Liu, L.-D. Zhao, Y. Liu, J. Lan, W. Xu, F. Li, B.-P. Zhang,D. Berardan, N. Dragoe, Y.-H. Lin et al., Journal of the Ameri-can Chemical Society, 2011, 133, 20112–20115.18 R. D. Shannon, Acta crystallographica section A: crystal physics,diffraction, theoretical and general crystallography, 1976, 32,751–767.19 D. Yang, X. Su, Y. Yan, T. Hu, H. Xie, J. He, C. Uher, M. G.Kanatzidis and X. Tang, Chemistry of Materials, 2016, 28,4628–4640.20 L. Pan, D. Bérardan, L. Zhao, C. Barreteau and N. Dragoe,Applied Physics Letters, 2013, 102, 023902.21 Y. Lei, H. Yang, J. Qiu, Y. Li, F. Gao, C. Yong, L. Tao, G. Song,N. Wang, S. Peng et al., Journal of the European Ceramic Soci-ety, 2022, 42, 7475–7480.22 V. I. Fistul, Heavily doped semiconductors, Plenum Press, 1969.23 Y.-X. Chen, K.-D. Shi, F. Li, X. Xu, Z.-H. Ge and J. He, Journalof the American Ceramic Society, 2019, 102, 5989–5996.24 H.-S. Kim, Z. M. Gibbs, Y. Tang, H. Wang and G. J. Snyder,APL materials, 2015, 3, 041506.25 B. Feng, G. Li, Z. Pan, Y. Hou, C. Zhang, C. Jiang, J. Hu, Q. Xi-ang, Y. Li, Z. He et al., Materials Letters, 2018, 217, 189–193.26 Y.-c. Liu, J.-L. Lan, B. Zhan, J. Ding, Y. Liu, Y.-H. Lin, B. Zhangand C.-w. Nan, Journal of the American Ceramic Society, 2013,96, 2710–2713.27 G.-K. Ren, S. Butt, K. J. Ventura, Y.-H. Lin, C.-W. Nan et al.,RSC advances, 2015, 5, 69878–69885.28 G.-K. Ren, S. Wang, Z. Zhou, X. Li, J. Yang, W. Zhang, Y.-H.Lin, J. Yang and C.-W. Nan, Nature communications, 2019, 10,2814.29 Y. Gu, X.-L. Shi, L. Pan, W.-D. Liu, Q. Sun, X. Tang, L.-Z. Kou,Q.-F. Liu, Y.-F. Wang and Z.-G. Chen, Advanced Functional Ma-terials, 2021, 31, 2101289.30 Y. Gu, W. Ai, L. Pan, X. Hu, P. Zong, C. Chen, C. Lu, Z. Xu andY. Wang, Materials Today Physics, 2022, 24, 100688.31 Y. Lei, R. Zheng, H. Yang, Y. Li, C. Yong, X. Jiang, R. Liu andR. Wan, Scripta Materialia, 2021, 199, 113885.32 F. Li, Z. Zheng, Y. Chang, M. Ruan, Z. Ge, Y. Chen and P. Fan,ACS applied materials & interfaces, 2019, 11, 45737–45745.33 X. Liang, R. Xu, M. Kong, H. Wan, W. Bai, D. Dong, Q. Li,H. Xu, Z. Li, B. Ge et al., Materials Today Physics, 2023, 34,101084.34 L. Pan, Y. Lang, L. Zhao, D. Berardan, E. Amzallag, C. Xu,Y. Gu, C. Chen, L.-D. Zhao, X. Shen et al., Journal of MaterialsChemistry A, 2018, 6, 13340–13349.35 R. Xu, Z. Chen, Q. Li, X. Yang, H. Wan, M. Kong, W. Bai,N. Zhu, R. Wang, J. Song et al., Research, 2023, 6, 0123.1–5 | 5