Investigating Multiferroic Phase Change Dynamics at Cryogenic Temperatures Using In-Situ Electron Counted Spectrum Imaging with Synchronized Holder Control

Improper ferroelectrics have strong potential for use in low power domain wall nano-electronic devices, as the formation and motion of conducting domain walls in such materials is governed by strain as opposed to their electric polarization [1,2]. Multimodal STEM spectrum imaging performed in the (scanning) transmission electron microscope (S)TEM is ideal for characterization of the ferroelectric domain dynamics in improper ferroelectrics, as the technique enables correlation of local chemistry and bonding information, with crystallographic and strain information determined from identical specimen regions at micro to (near) atomic scale.<br/><br/>MEMS based heating-biassing and cooling-biassing holders can be used to investigate phase change dynamics in these materials as a function of applied temperature and bias, though manual holder control becomes impractical if high stimuli resolution is required, due to the large number of temperature steps required for meaningful analysis in combination with the large number of individual biassing steps required at each temperature. In order to overcome this challenge, an automation strategy for holder control was recently developed to generalize external stimulus control within the DigitalMicrograph Python scripting framework. In our automation framework, the high-level communication Python library, ZMQ, is used to execute control commands from the embedded <i>in-situ</i> SI data acquisition routine. This generalized modular framework, enables synchronized control of any external device supporting Python and ZMQ.<br/><br/>To validate this framework, ferroic phase change dynamics in the improper ferroelectric: Co-Cl boracite were investigated as a function of temperature and simultaneous applied bias, using a MEMS based heating-biassing holder (Lightning, DENSSolutions). The automation strategy allowed complex holder control patterns to be executed at previously unachieved temperature and bias resolution.<br/><br/>Here we use this automation framework to investigate ferroic phase change behaviour in Fe-I boracite at cryogenic temperatures. <i>In-situ</i> spectrum imaging was performed with a 50-80 pA probe at 300 kV, using a counted mode EELS / energy filter system (GIF Continuum K3, Gatan) and flexible scan control system (Digiscan3). Domain wall dynamics were investigated as a function of applied bias at cryogenic temperatures using a MEMS based <i>in-situ </i>cryogenic cooling-biassing holder (Lightning-Arctic, DENSSolutions). Holder control and synchronization to data capture was performed using Python scripting in the DigitalMicrograph and DENSSolutions Impulse software packages. This scripting allowed multiple pass <i>in-situ</i> spectrum image (SI) data acquisition with all SI passes acquired at fixed holder stimuli conditions. Full voltage sweeps were applied at cryogenic temperatures series, or multiple voltage sweeps performed to investigate cycling effects. All data acquisition and holder control was fully automated. Data processing was performed using a combination of DigitalMicrograph (EELS, 4D STEM) and the Py4DSTEM (4D STEM) software packages.<br/><br/>Custom (non-raster) scanning features available in Gatan DigitalMicrograph 3.62 were used to investigate potential enhancements for atomic resolution imaging, electron energy-loss spectroscopy and ptychography performed on the Fe-I boracite system.<br/><br/>[1] Anisotropic conductance at improper ferroelectric domain walls. Nature materials (2012)<br/>[2] Anomalous Motion of Charged Domain Walls and Associated Negative Capacitance in Copper-Chlorine Boracite, Advanced Materials (2021)<br/>[3] Iliev, M., et al., Acta Physica Polonica A, 116, 2009 p.19-24.<br/>[4] M.C. acknowledges funding from Royal Society Tata University Research Fellowship, EPSRC & Royal Society Enhancement Award.