Arts & Architecture

Position of defects with respect to domain walls in Fe3+-doped Pb [Zr0. 52Ti0. 48] O3 piezoelectric ceramics

Description
Position of defects with respect to domain walls in Fe3+-doped Pb [Zr0. 52Ti0. 48] O3 piezoelectric ceramics
Published
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/49460595 Position of Defects With Respect to DomainWalls in Fe3+-Doped Pb[Zr0.52Ti0.48]O3Piezoelectric Ceramics  Article   in  Applied Physics Letters · February 2011 DOI: 10.1063/1.3555465 · Source: OAI CITATIONS 36 READS 183 5 authors , including: Some of the authors of this publication are also working on these related projects: Processing, structural and functional charcterization of ferroelectric ceramics based on BiFeO3, PZTand PMN-PT   View projectPeter JakesForschungszentrum Jülich 61   PUBLICATIONS   601   CITATIONS   SEE PROFILE Li JinXi'an Jiaotong University 49   PUBLICATIONS   718   CITATIONS   SEE PROFILE All content following this page was uploaded by Li Jin on 09 January 2014. The user has requested enhancement of the downloaded file.  Position of defects with respect to domain walls in Fe 3+ -dopedPb † Zr 0.52 Ti 0.48 ‡ O 3  piezoelectric ceramics Peter Jakes, 1 Emre Erdem, 1 Rüdiger-A. Eichel, 1,a  Li Jin, 2 and Dragan Damjanovic 2 1  Institut für Physikalische Chemie I, Universität Freiburg, Albertstr. 21, D-79104 Freiburg, Germany 2 Ceramics Laboratory, Swiss Federal Institute of Technology (EPFL), Lausanne 1015, Switzerland   Received 30 December 2010; accepted 27 January 2011; published online 18 February 2011  The position of    Fe Zr,Ti   − V  O••  • defect complexes in Pb  Zr 0.52 Ti 0.48  O 3   PZT   piezoelectric ceramicswas investigated by means of electron paramagnetic resonance   EPR   spectroscopy. The method of analysis pursued to obtain information on the   Fe Zr,Ti   − V  O••  • position is to compare the EPR spectraof Fe 3+ -doped PZT specimen at different states, i.e., a powder that is representative for a systemwith considerably reduced amount of non-180° domain walls and a sintered ceramic of identicalcomposition but with markedly developed domain structure. By considering the local site symmetryfor the Fe 3+ -functional center, indirect evidence is obtained that the   Fe Zr,Ti   − V  O••  • defect complexesare located within domains and not at domain walls. ©  2011 American Institute of Physics .  doi:10.1063/1.3555465  Lead zirconate titanate   Pb  Zr 1−  x  Ti  x   O 3 , PZT   is widelyused as piezoelectric material for sensing, transducing andactuating devices. 1 Tailored properties may be obtained by aliovalent   doping with transition-metal or rare-earth ions. 2 Ferroelectrically  hard   materials are obtained by acceptor-doping, where the hardening mechanism is attributed to theformation of defect dipoles between the acceptor ions andcharge compensating oxygen vacancies in the first coordina- tion sphere of the acceptor ion. 3–8 In the case of Fe 3+ -dopedPZT, the formation of    Fe Zr,Ti   − V  O••  • defect complexes hasbeen observed, 9,10 where by doping over the solubility limit,Fe-doping may result in formation of magnetic secondaryphases. 11 Contrary to the situation with acceptor doping, in soft  , donor-doped PZT the donor ions and the charge com-pensating lead vacancies occur as “isolated” defects that arelocated at remote coordination spheres. 12 The srcin of themarkedly different behavior of soft and hard ferroelectricsprobably lies in this difference in proximity between dopingions and charge compensating vacancies. In hard materialsthe energy potential of each domain can be described by adeep “V”-shaped potential. 13 Indirect evidence shows thatless well-defined defect complexes in soft materials create arandom potential in which motion of domain walls is lessrestricted than in acceptor doped compounds. 14 Because the performance of a piezoelectric device islargely determined by the development of a domain structureand the movement of non-180° domain walls during appli-cation of an external field, 14,15 the impact of the defect struc-ture on the domains configuration and the interaction of de-fects with domain walls define important issues. A number of models have been proposed interpreting both the  stabiliza-tion  of a certain domain pattern 3–5,16–20 and the  clamping  of domain walls by defect dipoles. 9,13,21–23 Concerning the placement of defects with respect to do-mains and ensuing domain configuration, three basic mecha-nisms are discussed for acceptor doped “bulk” ferroelectrics.  i   Volume effect  : owing to the polarization  P  D  associatedwith the defect dipole   Fe Zr,Ti   − V  O••  • , an orientation of spon-taneous polarization of the surrounding lattice  P S   parallel to P  D  is energetically favored. 4,5,16–19  ii   Domain-wall effect  : toneutralize internal stresses and charges, defects may migrateto domain walls, stabilizing their position. 20  iii   Grain-boundary effect  : due to the f ormation of secondary phases athigh dopant concentrations, 11 surface charges at the grainboundaries are generated that stabilize a certain domainconfiguration. 3 Mechanisms   i   and   ii   are directly related toan interaction between defect structure and domain walls.It has been proposed that the clamping of domain walls isdue to restoring f orces that limit the domain-walldisplacements. 9,13,21–23 On the other hand, the defect dipoleswere shown to follow the change in polarization directionafter application of an external field, provided the corre-sponding thermal and electric energies are high enough. 16,17 However, during fast field cycling, the defect-dipole orienta-tion remains unperturbed leading to reversible change inmacroscopic polarization with electric field.Recently, the origin of the fine domain structure inFe 3+ -doped PZT has been discussed in terms of oxygenvacancies disrupting the oxygen octahedral network andthus the spontaneous polarization. 24 Defect dipoles couldthus be located preferentially at domain walls imprintinga  nanodomain structure  to the material. 18,25,26 It is,therefore, of a main interest to characterize the location of   Fe Zr,Ti   − V  O••  • defect dipoles either within a domain or at adomain wall. In order to define this location electron para-magnetic resonance   EPR   has been applied. 27 In particular,the corresponding local symmetry at the Fe 3+ -site is moni-tored by means of the second-rank fine-structure interaction.The X-band   9.8 GHz   continuous wave EPR measure-ments were performed using a continuous-wave EPR spec-trometer   EMX, Bruker  , equipped with a cylindrical TE 112 resonator. All EPR measurements were performed at 20 K.The free Fe 3+ -ion possesses five unpaired electrons in ahalf-filled 3 d  -shell and can be described as orbital singlet.Its ground-state configuration is  6 S  5 / 2  with electron spin S  =5 / 2. The sixfold spin degeneracy can be lifted by thefine-structure interaction and an external magnetic field. Thecorresponding spin Hamiltonian for this high-spin systemcan be written as a  Electronic mail: ruediger.eichel@physchem.uni-freiburg.de. APPLIED PHYSICS LETTERS  98 , 072907   2011  0003-6951/2011/98  7   /072907/3/$30.00 © 2011 American Institute of Physics 98 , 072907-1  H =  g e   e B 0  ·  S  +   k  =2,...,5− k   q  k   B k q O k q  S   x  , S   y , S   z   1  in which the  g -matrix is taken as isotropic with g e =2.0023—the free electron  g -value,    e  denotes the Bohrmagneton,  B 0  the external magnetic field,  B k q are the fine-structure spin-Hamiltonian parameters, and  O k q are the ex-tended Stevens spin operators. 28 Experimentally, only the  B 2 q parameters can be determined, which are several orders of magnitude larger than the  B k   2 q terms for the   Fe Zr,Ti   − V  O••  • center. Correspondingly, the relevant site-symmetry is em-bodied in the second-rank symmetric and traceless fine-structure tensor that has diagonal form in its eigensystem  −    B 20 −  B 22   0 00 −    B 20 +  B 22   00 0 2   B 20   .   2  By convention,  B 20 is taken to as the principal value with thelargest absolute magnitude    B 20     B 22  . Correspondingly, itmay be distinguished between  cubic    B 20 =  B 22 =0  ,  axial   B 20  0,  B 22 =0  , and  rhombic    B 20  0,  B 22  0   symmetry atthe Fe 3+ -site. The thus determined  local  site symmetry asdetermined by EPR need not necessarily coincide with the global  crystal symmetry of the material. This situation isschematically illustrated in Fig. 1.A cubic site symmetry for the Fe 3+ -functional centerinvokes a complete oxygen octahedron, such that no  Fe Zr,Ti   − V  O••  • defect dipole is present. Such a situation ispresent at the paraelectric state above  T  C    see Fig. 1  a  . Thecharge compensating oxygen vacancy will then be a mobiledefect that contributes to the ionic conductivity. In the ferro-electric tetragonal phase of PZT, an axial site symmetry isonly observed if the orientation of the   Fe Zr,Ti   − V  O••   • defectdipole is collinear to the orientation of spontaneous polariza-tion in the corresponding unit cell   see Fig. 1  b  . In case theorientation of the   Fe Zr,Ti   − V  O••   • defect dipole is perpendicu-lar to the direction of spontaneous polarization, the localsymmetry is reduced to rhombic as compared to the tetrago-nal global symmetry 10  see Fig. 1  c  . The same local sym-metry, but generally with a different size of   B 22 , results forany orientation of the   Fe Zr,Ti   − V  O••  • defect dipole in therhombohedral PZT phase, as shown in Fig. 1  d  . This situa-tion holds also for a location of the   Fe Zr,Ti   − V  O••  • defect di-pole at a  thin  domain wall, as depicted in Fig. 1  e  , where theunit cells in the domain wall have low symmetry. However,if a  thicker   domain wall is considered, the unit cells mayhave cubic symmetry. In that situation the site symmetry atthe Fe 3+ -site is determined by the oxygen vacancy, such thatthe Fe 3+ experiences axial site symmetry   see Fig. 1  f   .The schematic illustration of non-180° domain walls inFigs. 1  e   and 1  f    considers domain walls along the  001  -direction. Although for PZT of tetragonal symmetrynon-180° domain walls commonly are considered to bealong the   101   direction, the argument of reduced localsymmetry for a Fe 3+ -site at a domain wall is not affected.This holds also for a situation when “nanodomains” havedeveloped 25 that do not possess a definite crystallographicrelationship to each other. 24 The corresponding X-band EPR spectra of Fe 3+ -dopedPZT 52/48 are depicted in Fig. 2  a  . Analogously to the re-cently reported EPR analysis of  Fe 3+ -doped hard and  La 3+ ,Fe 3+  -codoped soft PZT, 10,29 the observed X-bandspectra are representative for the so-termed  low-frequency FIG. 1. Schemmatical representation of the different structural arrangementsin Fe 3+ -doped PZT that result in  cubic   a  ,  axial   b   and   f   , and  rhombic  c  –  e   local symmetry at the Fe 3+ -site. Cubic site symmetry exclusivelyresults for  free  Fe 3+ centers above  T  C   in the paraelectric state. In the ferro-electric phases   Fe Zr,Ti   − V  O••  • defect dipoles are formed.   e   Thin  domainwall and   f    thick   domain wall. The domain-wall thickness is indicated by d  w .FIG. 2.   Color online   X-band   9.8 GHz   EPR spectra of Fe 3+ -doped PZT52/48 recorded at 20 K.   a   Experimental spectra for calcined powder   top  and sintered ceramic   bottom  .   b   Detailed view of the relevant resonancescompared to numerical spectrum simulations for the pure axial   top   andrhombic   bottom   Fe Zr,Ti   − V  O••  • site symmetries. An F-center in the ceramiccompound is indicated by an asterisk. 072907-2 Jakes  et al.  Appl. Phys. Lett.  98 , 072907   2011   regime   3  B q 2  h mw   with two main resonances at lowfields. First, they are characteristic for the formationof    Fe Zr,Ti   − V  O••  • defect dipoles. Second, the resonance at110 mT is representative for a center of axial site symmetry  Figs. 2  b  , top  , while the resonance at 160 mT is due to acenter of rhombic site symmetry   Fig. 2  b  , bottom  .In order to distinguish if    part of the   Fe Zr,Ti   − V  O••  • defect complexes are located at domain walls, twodifferent samples were investigated. A powder of Fe 3+ -dopedPZT 52/48, as representative for a mechanically stress-freesystem with considerably reduced amount of non-180° do-main walls, is compared with a sintered dense ceramic of identical composition but owing to the release of internalstress with markedly developed domain structure. Corre-sponding 1 at. % Fe 3+ -doped PZT 52/48 ceramics were syn-thesized by a conventional solid state process using standardmixed oxide route. 30 Samples were sintered at 1200 °C for2 h. The powder used in this experiment was obtained bycrashing one of the sintered samples and milling the powderin a planetary mill. The particle size after the milling wasverified by a scanning electron microscope. Most particlesappeared as individual grains with an average size of around1–2    m, which is comparable to the grain size in the sin-tered samples. The ceramic specimen is characterized by afine domain structure. 24 In order to rule out any effect of poling, both samples were annealed above  T  C   previous to theEPR experiments   600 °C for 1 h  . In case some of the de-fect dipoles would be located at a domain wall, the reso-nances indicative for axial and rhombic site symmetryshould show a variation in relative intensity for the ceramicas compared to the powder.The striking observation is that both ceramic and powderspectra display an identical intensity ratio of resonances fromaxial and rhombic   Fe Zr,Ti   − V  O••  • centers. Correspondingly,the amount of    Fe Zr,Ti   − V  O••  • defect dipoles located at a do-main wall has to be below the detection limit for EPR. Ac-cordingly, the amount of defects at domain walls should be atleast by nine orders of magnitude smaller than the amountlocated within domains.The conclusion may be drawn that in an unpoled ce-ramic the defect dipoles rather not locate at domain walls.This situation has already been proposed earlier 3,4 and hasbeen supported by a recent  in situ  optical microscopy studythat reported domain-pattern conservation after polarizationswitching with cyclic electric field. 23 While giving strongevidence that defect dipoles are concentrated within domains  between domain walls  , the present study leaves open thequestion whether defects are preferentially distributed closeto the grain boundaries or uniformly through the grains.Placement of defects near grain boundaries would be consis-tent with recent theoretical models that report three orders of magnitude stronger clamping pressure on domain walls fromcharged defects at the grain boundary than from the defectdipoles between domain walls aligned with polarization. 31 This research has been financially supported by the DFGcenter of excellence 595 “Electrical Fatigue in FunctionalMaterials” and FNS No. 200020-124498. 1 N. Setter,  Piezoelectric Materials in Devices   Swiss Federal Institute of Technology, Lausanne, 2002  . 2 M. J. Hoffmann and H. Kungl, Curr. Opin. Solid State Mater. Sci.  8 , 51  2004  . 3 K. Carl and K. H. Härdtl, Ferroelectrics  17 , 473   1978  . 4 P. V. Lambeck and G. H. Jonker, Ferroelectrics  22 , 729   1978  . 5 H. Meštri ć , R.-A. Eichel, T. Kloss, K. P. Dinse, S. Laubach, P. C. Schmidt,K. A. Schönau, M. Knapp, and H. Ehrenberg, Phys. Rev. B  71 , 134109  2005  . 6 G. Arlt and H. Neumann, Ferroelectrics  87 , 109   1988  . 7 R.-A. Eichel, P. Erhart, P. Träskelin, K. Albe, H. Kungl, and M. J. Hoff-mann, Phys. Rev. Lett.  100 , 095504   2008  . 8 M. Morozov and D. Damjanovic, J. Appl. Phys.  107 , 034106   2010  . 9 W. L. Warren, D. Dimos, B. A. Tuttle, and D. M. Smyth, J. Am. Ceram.Soc.  77 , 2753   1994  . 10 E. Erdem, R.-A. Eichel, H. Kungl, M. J. Hoffmann, A. Ozarowski, J. vanTol, and L. C. Brunel, Phys. Scr., T  T129 , 12   2007   ; IEEE Trans.Ultrason. Ferroelectr. Freq. Control  55 , 1061   2008  . 11 H. J. Kleebe, S. Lauterbach, L. Silvestroni, H. Kungl, M. J. Hoffmann, E.Erdem, and R.-A. Eichel, Appl. Phys. Lett.  94 , 142901   2009  . 12 R.-A. Eichel, H. Meštri ć , H. Kungl, and M. J. Hoffmann, Appl. Phys. Lett. 88 , 122506   2006  . 13 U. Robels and G. Arlt, J. Appl. Phys.  73 , 3454   1993  . 14 D. Damjanovic, in  Science of Hysteresis , edited by G. Bertotti and I.Mayergoyz   Academic, Oxford, 2006  , Vol. III, pp. 337–465. 15 S. C. Hwang and R. Waser, Acta Mater.  48 , 3271   2000  . 16 W. L. Warren, G. E. Pike, K. Vanheusden, D. Dimos, B. A. Tuttle, and J.Robertson, J. Appl. Phys.  79 , 9250   1996  . 17 L. X. Zhang, E. Erdem, X. Ren, and R.-A. Eichel, Appl. Phys. Lett.  93 ,202901   2008  . 18 C. H. Park and D. J. Chadi, Phys. Rev. B  57 , R13961   1998  . 19 S. Pöykkö and D. J. Chadi, Phys. Rev. Lett.  83 , 1231   1999  . 20 V. S. Postnikov, V. S. Pavlov, and S. K. Turkov, J. Phys. Chem. Solids  31 ,1785   1970  . 21 H. J. Hagemann, J. Phys. C  11 , 3333   1978  . 22 J. O. Gentner, P. Gerthsen, N. A. Schmidt, and R. E. Send, J. Appl. Phys. 49 , 4485   1978  . 23 L. X. Zhang and X. B. Ren, Phys. Rev. B  71 , 174108   2005  . 24 L. Jin, Z. He, and D. Damjanovic, Appl. Phys. Lett.  95 , 012905   2009  . 25 K. A. Schönau, L. A. Schmitt, M. Knapp, H. Fuess, R.-A. Eichel, H.Kungl, and M. J. Hoffmann, Phys. Rev. B  75 , 184117   2007  . 26 Q. Tan, J. F. Li, and D. Viehland, Philos. Mag. B  76 , 59   1997  . 27 R.-A. Eichel, J. Electroceram.  19 , 9   2007  ; Phys. Chem. Chem. Phys.  13 ,368   2011  . 28 A. Abragam and B. Bleaney,  Electron Paramagnetic Resonance of Tran-sition Ions   Clarendon, Oxford, 1970  . 29 W. L. Warren, B. A. Tuttle, B. N. Sun, Y. Huang, and D. A. Payne, Appl.Phys. Lett.  62 , 146   1993  . 30 M. I. Morozov and D. Damjanovic, J. Appl. Phys.  104 , 034107   2008  . 31 Y. A. Genenko, Phys. Rev. B  78 , 214103   2008  . 072907-3 Jakes  et al.  Appl. Phys. Lett.  98 , 072907   2011  View publication statsView publication stats
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks