Electronic Materials Research Laboratory, Key Laboratory of the Ministry of Education & International Center for Dielectric Research, School of Electronic Science and Engineering, Xi’an Jiaotong University, Xi’an 710049, China
Abstract:Ferroelectric/piezoelectric perovskites are an important class of functional material and have broad application prospects in commercial, industrial, military and other areas because of their high dielectric constants, high piezoelectric coefficients, and high electromechanical coupling coefficients. Their structures, applications, and physical mechanisms have been intensively studied in condensed matter physics and material science. The piezoelectric properties of ferroelectric materials mainly originate from the intrinsic field-induced lattice distortion and extrinsic domain inversion and domain wall motion. Therefore, the understanding of and the distinguishing between these mechanisms are important for ascertaining the origin of the high-piezoelectric properties and developing new functional materials. In this article, we review the research progress of technical means and methodology of analyzing the changes of crystal lattices and magnetic domains of materials under the action of an externally applied electric field through the high-energy synchrotron X-ray diffraction experiments. The techniques and analysis methods involved in the review cover the time-resolved X-ray diffraction, single/double-peak analysis, full-pattern refinement, center-of-mass calculation, and field-induced phase transformation analysis, which are used to study the intrinsic and extrinsic contributions to sample’s macroscopic properties. It is expected to provide the research methods, which fulfill the individual experimental requirements, and the technical support for the mechanism analysis of various piezoelectric materials through the introduction and review of various methods. Keywords:piezoelectricity/ X-ray diffraction/ intrinsic contribution/ extrinsic contribution
由于时间分辨衍射测量需要反映材料结构与外部刺激间的关系, 采集衍射数据的设备(如探测器)触发信号与控制外部刺激的控制信号间必须要保持同步[39]. 如在施加交变电场的时间分辨测量中, 各个时间点的衍射谱必须对应该时间点的准确电压, 而不是错误匹配到其他时间点的电压上. 对于数据获取与控制信号的同步, 不同研究者往往有着不同的数据获取方案. Choe等[40]的数据采集系统如图2(a)所示, 系统的信号同步过程如图2(b)所示. 该系统可以收集不同电压条件下样品的二维衍射谱. 系统保持同步的思路是: 图 2 (a) Choe等[40]的数据采集系统; (b) Choe等[40]的系统中信号同步过程; (c) Daniels等[41]的数据采集系统; (d) Daniels等的系统中数据采集的时间序列; (e)频闪技术中样品所施加电场与时间的关系, 以及相关衍射强度随电场变化趋势[40] Figure2. (a) Data acquisition system by Choe et al.[40]; (b) signal synchronization process in the system of Choe et al.[40]; (c) data acquisition system by Daniels et al.[41]; (d) timing sequences for data acquisition processes in the system of Daniels et al.; (e) time dependence of the AC electric field and the collected intensity of diffraction wings, showing the field-induced intensity exchange between the two wings[40]. (a) (b) (e) Copyright ? 2017 International Union of Crystallography. Reproduced with permission of the International Union of Crystallography.
在铁电材料中, 非180°畴壁运动对宏观应变的贡献主要来自于畴体积分数变化导致特定晶体学方向上的宏观应变. 以四方相的镧掺杂PZT陶瓷为例, 在二维衍射谱中, 对德拜-谢勒环特定角度上的衍射强度进行积分, 可以得到与电场具有不同夹角的陶瓷晶粒的一维衍射谱. 从图6可以看出, 与电场不同夹角的一维衍射谱中(002)衍射峰与(200)衍射峰的衍射强度不同, 即反映了四方相PZT陶瓷中002畴与200畴的体积分数不同. 通过单双峰拟合的方式, 可以得到这两个衍射峰的衍射强度$ {I}_{002} $和${I}_{200}$, 则002畴的体积分数为 图 6 La掺杂PZT陶瓷中002畴体积分数与电场不同夹角的关系(底图分别显示与电场呈0°与90°条件下(002)与(200)衍射峰体积分数的变化)[49] Figure6.η002 as a function of the field amplitude as well as orientation with respect to the direction of applied field, for an unpoled La-doped tetragonal PZT ceramic under the application of static electric fields. The measured and fitted (002)-type diffraction peaks corresponding to the particular values of η002 (marked by circles and indicated by arrows) are shown in the bottom section of the figure. For the fitted diffraction patterns, the deconvoluted (200) and (002) peaks are shown in black solid lines. The integration of individual (002) and (200) peaks are terminated beyond the peak position of the adjacent peak, as indicated by the color-shaded areas[49] (Copyright ? 2011 John Wiley and Sons).
根据本征贡献与非本征贡献的计算方法, 研究者们对于包括含铅压电体、无铅压电体和弛豫铁电体等在内的多晶压电体系中具有代表性的材料进行了探索与分析. 在传统含铅压电材料中, PZT由于发现年代较早、性能优越、应用广泛, 成为多种新型实验方法首选的研究对象. Pramanick等[49]对不同组分镧/铁掺杂的PZT陶瓷进行了原位、高能、时间分辨的X射线衍射测试, 分别得到了不同电场强度下晶格应变与畴壁运动对压电常数$ {d}_{33} $的贡献, 如图8所示. Zhao等[61]报道了MPB组分范围的Nb掺杂的PZT陶瓷中电场诱导的本征应变与非本征应变对宏观应变都有着较大贡献, 在靠近MPB组分的陶瓷中本征应变的贡献更高, 而远离MPB组分的陶瓷中非本征应变的贡献更高. 图 8 La掺杂的PbZr0.52Ti0.48O3陶瓷中晶格应变与畴壁运动对宏观压电常数及非线性压电常数的贡献[49] Figure8. Contributions of lattice strain and domain wall motion to macroscopic piezoelectric coefficient and non-linear piezoelectric coefficient in La-doped PbZr0.52Ti0.48O3 ceramics[49] (Copyright ? 2011 John Wiley and Sons)
在无铅压电材料中, KNN基材料因其较高的压电系数与优异的温度性, 具有广阔的研究前景. Fu等[62]报道了 (K0.48–xNa0.52)(Nb0.92–xSb0.08)O3-xLiTaO3陶瓷在多晶型相界(polymorphic phase boundary, PPB), 即正交相-四方相相界附近组分的压电响应来源, 材料富正交相组分中的应变以本征晶格应变为主, 而在富四方相组分中以非本征的可逆畴翻转应变为主. 同时非本征应变的贡献不仅取决于晶格畸变, 还取决于极化取向、PPB组分的相比例和畴类型. Ochoa等[63]报道了KNN基四方相(K0.44Na0.52Li0.04)-(Nb0.86Ta0.10Sb0.04)O3 (KNL-NTS)陶瓷中, 非180°畴壁运动对电场诱导宏观应变的贡献极大, 占到了宏观应变的约80%. Zheng等[64]报道了一种KNN基压电陶瓷$(1\!-\!x)({\rm K}_{1-y}{\rm Na}_y)({\rm Nb}_{1-z}{\rm Sb}_z){\rm O}_3 \text{-}x{\rm Bi}_{0.5}({\rm Na}_{1-w}{\rm K}_w)_{0.5}$HfO3 的新相界(三方-四方相相界). 在施加电场过程中, 陶瓷的(100)和(220)衍射峰的峰强度比发生了明显变化, 而这则与极化矢量在三方相-四方相之间的翻转密切相关, 如图9所示. 图 9 (1–x)(K1–yNay)(Nb1–zSbz)O3-xBi0.5(Na1–wKw)0.5HfO3 (x = 0.035, y = 0.52, z = 0.05, w = 0.18)陶瓷 (a), (b) (100)和(220)衍射峰随电场的演变过程; (c) (100)和(220)衍射峰中低角度衍射峰与高角度衍射峰的强度之比(I1/I2)随电场的变化[64] Figure9. (1–x)(K1–yNay)(Nb1–zSbz)O3-xBi0.5(Na1–wKw)0.5HfO3 ceramic with x = 0.035, y = 0.52, z = 0.05 and w = 0.18: (a), (b) Evolution of the (100) and (220) pseudocubic reflections as a function of the electric field; (c) ratio of low angle peak intensity to high angle intensity (I1/I2) for (100) and (220) pseudocubic reflections as a function of the electric field[64] (Copyright ? 2017 The Royal Society of Chemistry)
在其他无铅压电材料方面, Ba(Zr, Ti)O3-(Ba, Ca)TiO3 (BZT-BCT)也表现出了能与PZT相媲美的压电性能. Tutuncu等[65]对材料MPB附近组分的陶瓷进行了电场原位X射线衍射分析. 他们报道了在机电耦合响应中宏观应变主要由90°畴壁运动贡献导致; 随着组分越靠近MPB边界, 90°畴壁运动对宏观应变的贡献越大, 从而压电系数$ {d}_{33} $也随之增大. 在BiFeO3体系中, Khansur等[66]报道了三方相BiFeO3陶瓷中电场诱导应变主要来自非180°畴翻转, 并且观察到了非180°畴翻转的强弛豫行为. Li等[67]发现非遍历性(non-ergodic, NR)弛豫铁电体 0.57BiFeO3-0.21(K0.5Bi0.5)TiO3-0.22PbTiO3 (BF-KBF-PT)在电场诱导下会不可逆地从非遍历性赝立方相$ Pm\bar{3}m $转变为三方铁电相$ R3 c $. 在交变电场下, BF-KBF-PT陶瓷中本征应变对宏观应变$ {\varepsilon }_{33} $的贡献明显大于非本征应变. 在NaNbO3-BaTiO3 (NN-BT)体系中, Zuo等通过同步辐射X射线衍射、X射线吸收精细结构谱、拉曼光谱研究了NN-BT弛豫铁电体在外场作用下的结构变化[68]. 图10(a)—(g)显示了样品在外加电场的作用下, 极性纳米微区(polar nanoregions, PNRs)逐渐演化成铁电微畴, 且当电场大于0.8 kV/mm时, 观察到畴翻转现象. 电场诱导的PNR生长过程主导了场致应变的最快增长和最大应变滞后的形成, 而不是随后的畴翻转. 如图10(h)所示, 当外加电场释放时, 所有的变化又都被完全恢复, 表明电场强迫PNR生长和畴翻转的过程是可逆的. 图 10 NN-BT陶瓷{200}衍射峰在电场作用下的重新分布现象[68] Figure10. {200} reflections and their redistributions under electric field for NN-BT[68] (Copyright ? 2017 AIP Publishing)
24.4.原位单晶衍射 -->
4.4.原位单晶衍射
相对于陶瓷, 单晶纯度高、缺陷少、不存在晶界、机电性能更好. 同时, 单晶X射线衍射能够将倒易空间中的三维方向进行区分, 避免了同一族{hkl}衍射峰的相互重叠, 从而能够得到具有不同取向畴结构所生成的衍射强度分布. 受限于分辨率, 陶瓷衍射实验多针对具有较高对称性(如三方相和四方相)的材料进行分析, 或将MPB结构简单等同于三方和四方的两相共存结构, 然而高分辨单晶衍射数据可清晰区分低对称性结构(如单斜相)中的较小衍射峰分峰, 具有更大的通用性. Choe等[33]通过时间分辨原位单晶X衍射实验研究了NBT单晶在电场作用下的极化矢量旋转现象. 未施加电场时(004)和(400)衍射峰不重合, 显示材料为单斜相, 施加正向电场时, (004)和(400)峰在散射矢量方向上的距离加大, 随着电场逐步减小, 两峰间距进一步减小, 在施加最小电场–14 kV/cm下, 两峰距离最近, 但仍不重合. 通过对(004)和(400)峰三维倒易空间坐标随电场的变化进行分析和拟合, 他们还原了该材料在电场诱导下, 极化矢量在{110}镜面上的旋转过程. Hu等[69]生长出了具有超高电致应变的KNN单晶, 并通过原位单晶同步辐射X射线衍射实验观察到施加1.2 kV/mm电场时, {302}和{310}峰处产生新的衍射斑点, 显示了新的铁弹畴的产生, 同时表明该单晶的高电致应变主要来源于畴的重新分布. Zhang等[70]提出了一种通过高分辨单晶X射线衍射实验收集三维倒易空间衍射强度随交变电场变化, 从而分离MPB组分PZT单晶压电效应中本征贡献与非本征贡献的方法. 该实验中可观察到一个衍射峰中往往包含着多个独立衍射强度分布, 分别代表一类铁电/铁弹畴. 在数据分析过程中, 将三维衍射强度的坐标系规定为: X平行于散射矢量方向, Y和Z分别为选定的垂直于X的一个方向; 每个衍射峰的分量在YZ二维衍射强度分布图(沿X方向积分)中用一个YZ Box标记出来, 如图11(b)和图11(c)所示. 假设两组布拉格衍射峰的质心位置分别为$ {X}_{1}, {X}_{2} $, 对应曲线下的面积为$ {I}_{1}, {I}_{2} $, 同时假设两个衍射峰组的强度互换主要是由于畴翻转导致, 即$ \Delta{I}_{1}\approx -\Delta{I}_{2} $, 则电场诱导下的质心位置偏移可以表达为 图 11 (a) {111}衍射峰的衍射强度(沿YZ方向积分)与X的关系曲线, 垂直的红蓝线分别对应$ {E}_{+}\backslash {E}_{-} $状态下的质心位置; (b), (c)沿不同X范围积分的二维衍射强度分布图, 分别对应图(a)中的Group 1和Group 2; (d), (e)一个YZ Box范围内积分的衍射强度与X的关系曲线, 其中(d), (e)分别对应Group 1中的Box 2和Group 2中的Box 2 [70] Figure11. (a) The X dependence of the diffraction intensity around {111} reflections, integrated within the full YZ range. The vertical red and blue lines mark the center of mass positions corresponding to the E+ and ${E_ - } $ states. (b), (c) YZ dependence of the diffraction intensity integrated within two ranges of X, corresponding to Group 1 and Group 2 in panel (a). Several boxes are marked to show the positions of Bragg peak sub-components. (d), (e) Integrated intensities within one YZ box against X under four states of field. (d) Corresponds to Box 2 in Group 1 and (e) to Box 2 in Group 2[70] (Copyright ? 2018 International Union of Crystallography. Reproduced with permission of the International Union of Crystallography)