MST43

 Structural study of crystal interfaces using 4D-STEM

The convergent-beam electron diffraction (CBED), a technique using a convergent nano-electron probe based on the transmission electron microscopy, is particularly powerful for investigating local symmetries, crystal structures, and lattice defects of nanometer-sized local areas of specimens. [1,2]  Crystal structural parameters such as atom positions and atomic displacement parameters, and electrostatic potential and electron density distributions can be quantitatively determined by fitting intensities of CBED patterns with dynamical diffraction simulations. [3–6]

Furthermore, the variations of the local crystal structures can be examined by the combined use of scanning transmission electron microscopy (STEM) and CBED (STEM-CBED). This method has recently been termed four-dimensional (4D)-STEM, which refers to the recording of two-dimensional CBED patterns over a two-dimensional grid of electron probe positions. Recent advances in high-sensitivity and high-speed electron detectors and computational methods enabled to record full CBED patterns at different electron probe positions. [7,8] A promising application of 4D-STEM is the structure analysis of crystal interface structures that break periodicity in the crystal. Intriguing physical properties have been increasingly reported to arise from crystal interfaces, such as the ferroelectricity of twin domain walls in the centrosymmetric bulk structure of CaTiO₃. [9]

4D-STEM has been applied to the 90-degree twin domain walls of the ferroelectric BaTiO₃. [10] The structure of the domain walls of BaTiO₃ is still under discussion. For example, large variations in domain wall width were reported depending on the measurement technique. Detailed structural information on the domain walls, especially the distributions of the local electric polarizations, is crucial for the further development of ferroelectric devices. BaTiO₃ doped with 0.3% V to Ti sites was used to suppress the domain wall motion induced by electron beam irradiation. 4D-STEM data, or the 2D set of CBED patterns, were acquired from the areas including the 90-degree twin domain walls using an energy-filter TEM JEM-2010FEF operated at an accelerating voltage of 100 kV at room temperature, where the size of the electron probe was approximately 1 nm, as shown in Fig. 1. A Bloch-wave dynamical diffraction simulation code MBFIT was used to calculate CBED patterns. The effect of the probe position was taken into account using hypothetical supercells with a coherent nano probe. [10]

Figure 2(a) illustrates a crystal structure model of the 90-degree domain wall with the rotation of local polarization caused by the displacements of Ti and O atoms. CBED patterns at probe positions (b)-(e) simulated with this model are shown in Figures 2(b)-(e), respectively. Experimental CBED patterns obtained at probe positions corresponding to the simulated patterns are shown in Figs. 2(g)-(j). It is seen that the qualitative aspects of the mirror symmetry change due to the rotation of the local polarization are well reproduced. In addition, the 4D-STEM observation revealed that intermediate areas exist aside from the area of the rotation of polarization, which are different from the bulk crystal state. [10]

Application of 4D-STEM to the local crystal structure analysis of martensitic transformation in Ti-Ni-Fe alloys [11] will also be discussed in the talk.

References:

[1]  M. Tanaka and K. Tsuda, Microsc. 60, S245 (2011).

[2]  M. Tanaka, in International Tables for Crystallography, edited by U. Shmueli, Vol. B (International Union                of Crystallography, Chester, England, 2010).

[3]  K. Tsuda and M. Tanaka, Acta Cryst. A 55, 939 (1999).

[4]  J. M. Zuo, M. Kim, and J. C. H. Spence, Nature 401, 49 (1999).

[5]  P. N. H. Nakashima, A. E. Smith, J. Etheridge, and B. C. Muddle, Science 331, 1583 (2011).

[6]  K. Tsuda, D. Morikawa, Y. Watanabe, S. Ohtani, and T. Arima, Phys. Rev. B 81, 180102 (2010).

[7] C. Ophus, Microsc. Microanal. 25, 563 (2019).

[8]  C. Ophus, Annu. Rev. Mater. Res. 53, annurev (2023).

[9] D. Morikawa and K. Tsuda, Appl. Phys. Lett. 118, 092901 (2021).

[10] D. Morikawa, Y. Noguchi, and K. Tsuda, Jpn. J. Appl. Phys. 62, SM1003 (2023).

[11]  Y. Kimura, X. Xu, K. Han, K. Niitsu, T. Omori, R. Y. Umetsu, and R. Kainuma, Materials Transactions                64, 1591 (2023).