Communication pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. 2017, 139, 14865-14868
Colossal Volume Contraction in Strong Polar Perovskites of Pb(Ti,V)O3 Zhao Pan,†,‡,§ Jun Chen,*,†,§ Xingxing Jiang,∥ Lei Hu,† Runze Yu,§ Hajime Yamamoto,§ Takahiro Ogata,§ Yuichiro Hattori,§ Fangmin Guo,⊥ Xi’an Fan,‡ Yawei Li,‡ Guangqiang Li,‡ Huazhi Gu,‡ Yang Ren,⊥ Zheshuai Lin,∥ Masaki Azuma,§ and Xianran Xing† †
Department of Physical Chemistry, University of Science and Technology Beijing, Beijing 100083, China State Key Laboratory of Refractories and Metallurgy, Wuhan University of Science and Technology, Wuhan 430081, China § Materials and Structures Laboratory, Tokyo Institute of Technology, 4259 Nagatsuta, Midori, Yokohama 226-8503, Japan ∥ Center for Crystal R&D, Key Laboratory of Functional Crystals and Laser Technology, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, China ⊥ X-Ray Science Division, Argonne National Laboratory, Argonne, Illinois 60439, United States ‡
S Supporting Information *
electronic transitions (e.g., BiNiO3 and LaCu3Fe4O12),4 and spontaneous volume ferroelectrostriction in ferroelectrics.5 Recent studies have focused on the discovery of large NTE over a wide temperature range. It is known that chemical modification normally reduces the NTE property,5 in examples such as Sn-doped ZrMo2O8,6 Ga/Fe-substituted ScF3,2 Codoped La(Fe, Si)13,7 and Si-doped Mn3GaN.8 One promising compound for large NTE is the ABO3 perovskite-type ferroelectric of PbTiO3 (PT). In the past half century, PT-based ceramics have been successfully developed to exhibit high piezoelectric performances by formation of a morphotropic phase boundary (MPB) or monoclinic phase,9 such as Pb(Zr,Tix)O3, and PT-Pb(Mg1/3Nb2/3)O3. The origin of the high performance of electromechanical properties at the MPB can be well elaborated by the existence of a monoclinic phase.10 Apart from the well-known piezoelectricity and ferroelectricity, PT exhibits unusual NTE from room temperature (RT) to its Curie temperature (TC = 490 °C), with an average volumetric CTE of −1.99 × 10−5/°C.5 Studies on crystal structure and first-principles calculations have proved a close relationship between ferroelectricity and NTE, that is, the ferroelectrovolume effect (FVE).5 By taking advantage of the flexible structure of PT, it is possible to achieve enhanced NTE by means of modulating its ferroelectricity. Herein, we have successfully realized NTE enhancement over a wide temperature range in a binary system of Pb(Ti1−xVx)O3. Much enhanced tetragonality (c/a) has been observed after the V chemical substitution. Both NTE and TC have been largely enhanced in the Pb(Ti1−xVx)O3 solid solutions. Most intriguingly, a notable volume contraction as large as 3.7% is obtained in Pb(Ti0.7V0.3)O3 during the ferroelectric-to-paraelectric (FE-toPE) phase transition, which is rarely reported in previous studies.5 A series of Pb(Ti1−xVx)O3 compounds (x = 0.1−0.6) were prepared by a high-pressure and high-temperature reaction (see Supporting Information (SI)). The X-ray diffraction (XRD)
ABSTRACT: The unique physical property of negative thermal expansion (NTE) is not only interesting for scientific research but also important for practical applications. Chemical modification generally tends to weaken NTE. It remains a challenge to obtain enhanced NTE from currently available materials. Herein, we successfully achieve enhanced NTE in Pb(Ti1−xVx)O3 by improving its ferroelectricity. With the chemical substitution of vanadium, lattice tetragonality (c/a) is highly promoted, which is attributed to strong spontaneous polarization, evidenced by the enhanced covalent interaction in the V/Ti−O and Pb−O2 bonds from firstprinciples calculations. As a consequence, Pb(Ti0.9V0.1)O3 exhibits a nonlinear and much stronger NTE over a wide temperature range with a volumetric coefficient of thermal expansion αV = −3.76 × 10−5/°C (25−550 °C). Interestingly, an intrinsic giant volume contraction (∼3.7%) was obtained at the composition of Pb(Ti0.7V0.3)O3 during the ferroelectric-to-paraelectric phase transition, which represents the highest value ever reported. Such volume contraction is well correlated to the effect of spontaneous volume ferroelectrostriction. The present study extends the scope of the NTE family and provides an effective approach to explore new materials with large NTE, such as through adjusting the NTE-related ferroelectric property in the family of ferroelectrics.
M
ost natural materials expand upon heating, a phenomenon known as positive thermal expansion (PTE). In contrast, some materials exhibit abnormal contraction on heating, referred to as negative thermal expansion (NTE). The discovery of NTE provides an opportunity to tailor the coefficient of thermal expansion (CTE) of materials that usually suffer from mechanical degradation and structural instability under thermal shock. So far, the known NTE mechanisms include phonon-related transverse vibration in flexible framework structures (e.g., ZrW2O8, Ag3[Co(CN)6], and ScF3),1−14 the magnetovolume effect (e.g., Invar Fe−Ni alloys and Mn3Cu1−xGexN),3 valence state or © 2017 American Chemical Society
Received: August 14, 2017 Published: October 10, 2017 14865
DOI: 10.1021/jacs.7b08625 J. Am. Chem. Soc. 2017, 139, 14865−14868
Communication
Journal of the American Chemical Society
Figure 2. Calculated spontaneous polarization of Pb(Ti1−xVx)O3 (x = 0.1−0.6). The inset is the crystal structure.
by assuming a point-charge model based on the refinement results.14 The calculated PS value shows an increase tendency from 59 μC/cm2 for pure PT15 to 61, 68, 73, 83, 90, and 92 μC/ cm2 for x = 0.1, 0.2, 0.3, 0.4, 0.5, and 0.6, respectively (Figure 2). The values are much larger than for typical high-performance piezoelectric materials such as PbZr1−xTixO3 (54 μC/cm2 at x = 0.42).16 Furthermore, according to Landau theory with the relationship of TC = αPS2,15 the present Pb(Ti1−xVx)O3 could be expected to be high-TC ferroelectric materials. Taking Pb(Ti0.5V0.5)O3 for instance, its TC is expected to be as high as 1140 °C by reference to PT (PS = 59 μC/cm2, TC = 490 °C).15 It is proposed that the large polarization in PT-based ferroelectrics can give rise to enhanced NTE.5 The enhancement of NTE can be expected in the present Pb(Ti 1−x V x)O 3 compositions. In order to precisely investigate the thermal expansion property of the Pb(Ti1−xVx)O3 system, the temperature dependence of the unit cell volume was extracted by means of structure refinement of SXRD data (SI Figures S7−S9). Pb(Ti1−xVx)O3 exhibit nonlinear and strong NTE over a wide temperature from RT to near TC (Figure 3). The unit cell volumes are not sensitive to temperature variation before the FEto-PE phase transition. However, they dramatically contract during the FE-to-PE phase transition, i.e., 1.4% for x = 0.1 (Figure 3a). The average volumetric CTE values of Pb(Ti1−xVx)O3 are −3.76 × 10−5/°C (RT to 550 °C), −4.97 × 10−5/°C (RT to 600 °C), and −7.36 × 10−5/°C (RT to 600 °C) for x = 0.1, 0.2, and 0.3, respectively. According to the measurements of hightemperature SXRD, the TC for x = 0.1 is about 550 °C, and it is between 550 and 600 °C for x = 0.2 and 0.3. In the present study, a relatively large temperature interval was used to measure the temperature dependence of SXRD, with 40 °C for x = 0.2 and 25 °C for x = 0.3. TC between x = 0.2 and x = 0.3 cannot be well distinguished. In principle, TC should be higher for x = 0.3 due to the larger lattice distortion. Note that the present NTEs are much stronger than that of PT (αV = −1.99 × 10−5/°C, RT to 490 °C) and even stronger that for 0.4PT-0.6BiFeO3 (αV = −3.92 × 10−5/ °C), which was reported to exhibit the strongest NTE in Pbbased ferroelectrics.13 This suggests that the present Pb(Ti1−xVx)O3 solid solution could be a promising candidate to adjust CTEs of materials. Specially, with increasing vanadium content, the volume contraction is much increased, to as large as 2.4% and 3.7% for Pb(Ti0.8V0.2)O3 and Pb(Ti0.7V0.3)O3, respectively (Figure 3b,c). It is noteworthy that such large volume contractions are rarely observed in the currently available NTE materials, such as giant NTE materials of double perovskite LaCu3Fe4O12 (−1.0%),4b CuO nanoparticles (−1.1%),3c typical antiperovskite Mn3AN (−1.3%),3b and BiNiO3 (−2.5 to −3.4%).4a It also should be mentioned that the TC of
Figure 1. (a) XRD patterns and (b) lattice parameters of Pb(Ti1−xVx)O3 (x = 0.1−0.6) at room temperature.
patterns of Pb(Ti1−xVx)O3 compounds are presented in Figure 1a. The samples are of high quality with negligible impurity. All investigated samples can be well indexed into tetragonal symmetry. It is known that PbVO3 exhibits a larger tetragonality (c/a) than PT.11 With increasing content of x, the (001) peak exhibits an apparent shift to a lower angle region, indicating expansion of the c axis. The (100) peak shows an opposite trend. It shifts slightly to a higher angle region, demonstrating contraction of the a(b) axis. The detailed lattice parameters were refined and are plotted in Figure 1b. The lattice parameter c increases almost linearly, whereas a decreases continuously as a function of x. As a result, an unusually enhanced c/a is observed. The enhanced tetragonality should be closely related to the strong hybridization between the Pb 6s and O 2p orbitals, due to the special stereochemically active 6s lone-pair electronic configuration of Pb. It needs to be mentioned that most chemically modified PT-based ferroelectrics, such as (1-x)PTxBi(Ni1/2Ti1/2)O3 and (1-x)PT-xBi(Mg1/2Ti1/2)O3, exhibit weakened c/a.12 Few of them show abnormally enhanced tetragonality, such as (1-x)PT-xBiFeO3 and (1-x)PT-xBi(Zn1/2Ti1/2)O3.13 Here, the maximum tetragonality (c/a = 1.17 for x = 0.6) is even larger than those of 0.5PT-0.5BiFeO3 (c/a = 1.14) and 0.6PT-0.4Bi(Zn1/2Ti1/2)O3 (c/a = 1.11).13 In the PTbased piezoelectric materials, large tetragonality is usually accompanied by interesting physical properties, such as large spontaneous polarization (PS), high TC, and enhanced NTE.5 The refined crystal structures of Pb(Ti1−xVx)O3 (x = 0.1, 0.2, 0.3, 0.4, 0.5, and 0.6) are closely correlated with the parent compound PT (SI Figures S1−S6). Note that after the introduction of vanadium the enhanced tetragonality leads to pyramidal rather than octahedral coordination (see inset of Figure 2). The large tetragonality can be attributed to the large PS displacements at the A site of Pb (δzA) and the B site of Ti/V (δzB). The PS value of all investigated samples can be calculated 14866
DOI: 10.1021/jacs.7b08625 J. Am. Chem. Soc. 2017, 139, 14865−14868
Communication
Journal of the American Chemical Society
Therefore, the volume difference Vexp − Vnm represents the quantitative contribution from ferroelectricity. A large value of ωS means a stronger NTE, while a smaller one means a weaker NTE. Here, the RT ωS values for Pb(Ti1−xVx)O3 (x = 0.1, 0.2, and 0.3) are 4.1%, 6.6%, and 8.5%, respectively. As a NTE material, the volume contraction (ΔV) is very important, since it can be directly related to the ability for thermal expansion compensation. It is 1.35%, 2.34%, and 3.59% for x = 0.1, 0.2, and 0.3, respectively, during the FE-to-PE phase transition (Figure 3). Note that the present ωS is much higher than that of PT (3.1%),5 and even higher than for 0.5PT-0.5BiFeO3 (ωS = 5% at RT),13a which was previously reported as the strongest NTE in Pb-based ferroelectrics. Similarly, spontaneous volume magnetostriction (SVMS, ωS = ωexp − ωnm) has been adopted to quantitatively describe the magnetovolume effect on NTE. The present ωS value is higher than those of typical magnetic NTE materials, such as the Fe0.65Ni0.35 Invar alloy (5.8% at 0 K),5 the intermetallic LaFe11.2Si1.8 (ωS = 2.2% at 0 K),18 and the perovskite-type ferromagnetic SrRuO3 (ωS = 0.33% at 40 K).19 The strong SVFS effect is well ascribed to the large PS,5 which can be interpreted by the good relationship analogous to the Landau theory,
ωS ∼ αδzA 2
(2)
where α is the coupling coefficient between ω S and PS displacement (δzA). We can see that ωS shows a good linear correlation with the square of PS displacement, δzA2 (Figure S11). The enhanced ferroelectricity induces large ωS and therefore produces giant NTE like in Pb(Ti0.7V0.3)O3. To intuitively study the effect of V/Ti substitution on hybridization with oxygen atoms, the electron density of Pb(Ti1−xVx)O3 (x = 0, 0.25, and 0.5) was calculated by firstprinciples calculations (Figure 4). Accordingly, in pure PbTiO3
Figure 3. Temperature dependence of unit cell volume for Pb(Ti1−xVx)O3 (x = 0.1, 0.2, and 0.3). The SVFS and volume change are indicated.
Pb(Ti1−xVx)O3 solid solutions has been increased, which is attributed to the improved polarization according to the empirical relationship of TC = αPS2.15 For further increasing content of x (x > 0.3), the compounds decompose before the FEto-PE phase transition temperature can be reached due to the increased lattice distortion and weakened thermal stability of perovskite structure (SI, Figure S10).17 The studies of crystal structure and first-principles calculation have revealed the close relationship between ferroelectricity and NTE in PT-based ferroelectrics.5 Recently, a new concept of spontaneous volume ferroelectrostriction (SVFS) was proposed to quantitatively describe the FVE, in which the baseline of the description of SVFS is estimated for the contribution purely by the thermal expansion of phonon vibration.13a The SVFS is calculated by ωS =
Vexp − Vnm Vnm
× 100%
Figure 4. Electron density map around (a) Ti−O1 bond in PbTiO3, (b) V−O1 bond in Pb(Ti0.75V0.25)O3, (c) V−O1 bond in Pb(Ti0.5V0.5)O3, (d) Pb−O2 bond in PbTiO3, (e) Pb−O2 bond in Pb(Ti0.75V0.25)O3, and (f) Pb−O2 bond in Pb(Ti0.5V0.5)O3.
(x = 0), the highest value of electron density around the Ti−O bond along the c-axis is 1.25 Å−3 (Figure 4a). In Pb(Ti0.75V0.25)O3 and Pb(Ti0.5V0.5)O3, more electron cloud overlap occurs around the counterpart V−O bond. As a result, stronger hybridization occurs between V−O, which is evidenced by the maximum electron densities of 1.56 and 1.72 Å−3 for Pb(Ti0.75V0.25)O3 and Pb(Ti0.5V0.5)O3, respectively (Figure 4b,c). Moreover, the overlapping areas of Pb−O2 bonds around the vanadium atom in Pb(Ti0.75V0.25)O3 and Pb(Ti0.5V0.5)O3 are more prominent than those in PbTiO3 (Figure 4d−f). All these observations
(1)
where Vexp and Vnm are experimental and nominal volumes, respectively (Table S1). Vnm is calculated by extrapolation from paraelectric to ferroelectric phase and can be basically considered as the volume component from lattice thermal vibration. 14867
DOI: 10.1021/jacs.7b08625 J. Am. Chem. Soc. 2017, 139, 14865−14868
Communication
Journal of the American Chemical Society
User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC0206CH11357. The high-temperature synchrotron radiation experiments were performed at the BL44B2 and BL19B2 of SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2015B1127, 2016A1060, and 2016B1850). We acknowledge the help from Kanagawa Institute of Industrial Science and Technology.
indicate that vanadium substitution enhances the covalent interaction in not only V/Ti−O bonds but also Pb−O2 bonds. Therefore, stronger ferroelectricity occurs in Pb(Ti1−xVx)O3, which simultaneously gives rise to larger NTE behavior. The present work provides a route for exploring enhanced NTE materials by introducing the strong polarity perovskite PbVO3 into PT-based ferroelectrics. From this, we anticipate that other unreported PT-based solid solutions that contain polar perovskites, such as BiCoO3 and Bi(Zn1/2V1/2)O3,14,20 should also exhibit enhanced tetragonality and giant NTE, analogous to the behavior in the present Pb(Ti1−xVx)O3 system. Furthermore, since the present Pb(Ti1−xVx)O3 solid solutions exhibit large PS, good ferroelectric properties could be foreseeable if the perovskite structure can be stabilized via strain engineering by growing epitaxial thin films. Additionally, it needs to be noted that the present NTE materials exhibit a large anisotropy, which will cause difficulty in sintering ceramics. However, powders of the present giant NTE materials could be good candidates as high-performance thermal expansion compensators in composites to control thermal expansion.21 In summary, the tetragonality of PT has been abnormally improved with the replacement of titanium with vanadium. The unusually enhanced tetragonality and large PS, which are attributed to the enhanced covalent interaction in V/Ti−O and Pb−O2 bonds from first-principles simulation, are presumably associated closely with the enhanced NTE. Correspondingly, enhanced NTE over a wide temperature range has been achieved in Pb(Ti1−xVx)O3 solid solutions. Intriguingly, the highest volume contraction (3.7%) has been observed for the composition of Pb(Ti0.7V0.3)O3. The present work extends the scope of giant volume contraction candidates and highlights a route to achieve enhanced NTE in PbTiO3based ferroelectrics by substituting isostructural perovskites with large distortion to adjust the NTE-related physical property.
■
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b08625. Sample preparation, experimental methods, and firstprinciples simulation (PDF)
■
REFERENCES
(1) (a) Mary, T. A.; Evans, J. S. O.; Vogt, T.; Sleight, A. W. Science 1996, 272, 90. (b) Goodwin, A. L.; Calleja, M.; Conterio, M. J.; Dove, M. T.; Evans, J. S. O.; Keen, D. A.; Peters, L.; Tucker, M. G. Science 2008, 319, 794. (2) Hu, L.; Chen, J.; Fan, L. L.; Ren, Y.; Rong, Y. C.; Pan, Z.; Deng, J. X.; Yu, R. B.; Xing, X. R. J. Am. Chem. Soc. 2014, 136, 13566. (3) (a) Van Schilfgaarde, M.; Abrikosov, I. A.; Johansson, B. Nature 1999, 400, 46. (b) Takenaka, K.; Takagi, H. Appl. Phys. Lett. 2005, 87, 261902. (c) Mohn, P. Nature 1999, 400, 18. (4) (a) Azuma, M.; Chen, W. T.; Seki, H.; Czapski, M.; Olga, S.; Oka, K.; Mizumaki, M.; Watanuki, T.; Ishimatsu, N.; Kawamura, N.; Ishiwata, S.; Tucker, M. G.; Shimakawa, Y.; Attfield, J. P. Nat. Commun. 2011, 2, 347. (b) Long, Y. W.; Hayashi, N.; Saito, T.; Azuma, M.; Shimakawa, Y. Nature 2009, 458, 60. (5) Chen, J.; Hu, L.; Deng, J. X.; Xing, X. R. Chem. Soc. Rev. 2015, 44, 3522. (6) Tallentire, S. E.; Child, F.; Fall; Vella-Zarb, L.; Evans, I. R.; Tucker, M. G.; Keen, D. A.; Wilson, C.; Evans, J. S. O. J. Am. Chem. Soc. 2013, 135, 12849. (7) Huang, R. J.; Liu, Y. Y.; Fan, W.; Tan, J.; Xiao, F. R.; Qian, L. H.; Li, L. F. J. Am. Chem. Soc. 2013, 135, 11469. (8) Sun, Y.; Wang, C.; Wen, Y. C.; Chu, L. H.; Nie, M. J. Am. Ceram. Soc. 2010, 93, 650. (9) (a) Haertling, G. H. J. Am. Ceram. Soc. 1999, 82, 797. (b) Eitel, R. E.; Randall, C. A.; Shrout, T. R.; Park, S. E. Jpn. J. Appl. Phys. 2002, 41, 2099. (10) (a) Noheda, B.; Cox, D. E.; Shirane, G.; Gonzalo, J. A.; Cross, L. E.; Park, S.-E. Appl. Phys. Lett. 1999, 74, 2059. (b) Fan, L.; Chen, J.; Ren, Y.; Pan, Z.; Zhang, L.; Xing, X. Phys. Rev. Lett. 2016, 116, 027601. (c) Liu, H.; Chen, J.; Fan, L.; Ren, Y.; Pan, Z.; Lalitha, K. V.; Rödel, J.; Xing, X. Phys. Rev. Lett. 2017, 119, 017601. (11) Shpanchenko, R. V.; Chernaya, V. V.; Tsirlin, A. A.; Chizhov, P. S.; Sklovsky, D. E.; Antipov, E. V.; Khlybov, E. P.; Pomjakushin, V.; Balagurov, A. M.; Medvedeva, J. E.; Kaul, E. E.; Geibel, C. Chem. Mater. 2004, 16, 3267. (12) (a) Choi, S. M.; Stringer, C. J.; Shrout, T. R.; Randall, C. A. J. Appl. Phys. 2005, 98, 034108. (b) Randall, C. A.; Eitel, R. E.; Jones, B.; Shrout, T. R. J. Appl. Phys. 2004, 95, 3633. (13) (a) Chen, J.; Wang, F. F.; Huang, Q. Z.; Hu, L.; Song, X. P.; Deng, J. X.; Yu, R. B.; Xing, X. R. Sci. Rep. 2013, 3, 2458. (b) Suchomel, M. R.; Davies, P. K. Appl. Phys. Lett. 2005, 86, 262905. (14) Oka, K.; Azuma, M.; Chen, W. T.; Yusa, H.; Belik, A. A.; Takayama-Muromachi, E.; Mizumaki, M.; Ishimatsu, N.; Hiraoka, N.; Tsujimoto, M.; Tucker, M. G.; Attfield, J. P.; Shimakawa, Y. J. Am. Chem. Soc. 2010, 132, 9438. (15) Abrahams, S. C.; Kurtz, S. K.; Jamieson, P. B. Phys. Rev. 1968, 172, 551. (16) Noheda, B.; Gonzalo, J. A.; Cross, L. E.; Guo, R.; Park, S. E.; Cox, D. E.; Shirane, G. Phys. Rev. B 2000, 61, 8687. (17) Belik, A. A. J. Solid State Chem. 2012, 195, 32. (18) Jia, L.; Sun, J. R.; Zhang, H. W.; Hu, F. X.; Dong, C.; Shen, B. G. J. Phys.: Condens. Matter 2006, 18, 9999. (19) Kiyama, T.; Yoshimura, K.; Kosuge, K.; Ikeda, Y.; Bando, Y. Phys. Rev. B 1996, 54, R756. (20) Yu, R. Z.; Hojo, H.; Oka, K.; Watanuki, T.; Machida, A.; Shimizu, K.; Nakano, K.; Azuma, M. Chem. Mater. 2015, 27, 2012. (21) Takenaka, K.; Okamoto, Y.; Shinoda, T.; Katayama, N.; Sakai, Y. Nat. Commun. 2017, 8, 14102.
AUTHOR INFORMATION
Corresponding Author
*
[email protected] ORCID
Jun Chen: 0000-0002-7330-8976 Zheshuai Lin: 0000-0002-9829-9893 Masaki Azuma: 0000-0002-8378-321X Xianran Xing: 0000-0003-0704-8886 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Grant Nos. 91422301, 21231001, and 21590793), National Program for Support of Top-notch Young Professionals, the Program for Chang Jiang Young Scholars, and the Fundamental Research Funds for the Central Universities, China. This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science 14868
DOI: 10.1021/jacs.7b08625 J. Am. Chem. Soc. 2017, 139, 14865−14868
