Page 257 - Read Online

Keeney et al. Microstructures 2023;3:2023041 https://dx.doi.org/10.20517/microstructures.2023.41 Page 5 of 15

the NGO substrate, the B6TFMO sample layer, and an extra oxidation layer on top (which is deemed
reasonable, as the sample had been kept in the air after deposition). The initial density value inputted for the
NGO substrate was taken from the datasheet provided by the substrate manufacturer (Crystal GmBH,
Ostendstraße 25, D-12459 Berlin, Germany), and that for the B6TFMO sample was taken from that reported
by Bartkowska et al. on the m = 5 Aurivillius phase compositions . The deviation between the
[40]
experimental data curve and the fitted curve was calculated to be 2.37%. The thickness values obtained from
the XRR measurements were consistent with measurements obtained by transmission electron microscopy
(TEM) imaging of previous samples prepared under similar DLI-CVD conditions [27,28] . Micro-structural
analysis was performed by high-resolution TEM (HR-TEM) on a JEOL 2100 TEM, operated at 200 kV. To
enable HR-TEM imaging, the samples were coated with gold to counteract charging, followed by coating
with platinum to enable focused ion beam (FIB) cross sectioning. Cross-section lamellae were made using a
FEI Helios Nanolab 600i DualBeam FIB-scanning electron microscope (FIB-SEM) (final thinning settings:
93 pA current at 30 kV, final polish settings: 47 pA current at 5 kV). Thickness analysis of the B6TFMO film
deposited using 100 injections was performed from 55 measurements across a 7.6 µm wide lamella. The
standard deviation (SD) variability of the dataset was ±0.4 nm. An atomic force microscope (AFM), Bruker
Dimension Icon in PeakForce Tapping mode (with Bruker SCANASYST-AIR probes, Al reflex coated,
2 nm tip radius, 70 kHz resonant frequency) was used for topography analysis of the films. Film surface
roughness and surface impurity volume fraction measurements were accomplished by carrying out AFM
measurements (1 µm × 1 µm scan area) over five different areas of the sample surface. Average root mean
square (RMS) roughness (nm) figures are provided, along with the SD variability of the dataset. Average
volume fraction (vol.%) quantities were determined from impurity particle count area as a proportion of the
scan area.

AFM-based nano-machining and PFM experiments
AFM-based nano-machining of the surfaces of the ultrathin B6TFMO films was achieved using an Asylum
Research MFP-3D AFM by applying force from a sufficiently stiff diamond cantilever. NM-RC (single
TM
crystal diamond, Au reflex coated, < 10 nm tip radius, 486 kHz resonant frequency, boron-doped) probes,
commercially available from Adama Innovations (c/o Republic of Work, 12 South Mall, Cork, T12 RD43,
Ireland), were used for the AFM-based nano-machining studies. The normal loading force, F (N), was
calculated using Hookes law, F = kz, where k is the spring constant (N/m), and z (m) is determined by
multiplying the calibrated deflection sensitivity (m/V) by the setpoint (V) (the difference between
photodiode signal when the cantilever is far from the surface and when it is at setpoint). The spring
constant [1.89 × 10 (±8.39%) N/m] was calibrated via the Sader method by performing a thermal tune
2
[41]
(1,000 samples) in free air to measure the resonance frequency and the quality factor of the cantilever.
Loading forces of between 1.86 µN and 5.59 µN were utilized for each scan in this work. Successive scans
were performed over 8 µm × 8 µm areas at a scan frequency of 0.75 Hz to progressively machine
through the sample from the film surface until the substrate was reached. A machining scan angle of
90° was utilized unless otherwise specified. The depths of the nano-machined areas were measured by
performing line section height profiles of imaged areas. The measured RMS roughness values for the
nano-machined areas of the 7.9 nm B6TFMO sample were, in general, rougher [values of 0.52 nm (SD
= ±0.42 nm), 0.46 nm (SD = ±0.38 nm), 0.50 nm (SD = ±0.42 nm), and 0.57 nm (SD = ±0.48 nm) for
depths of 0.7 nm, 2.0 nm, 3.8 nm, and 6.2 nm, respectively] compared with roughness values for
the pristine B6TFMO surface [0.29 nm (SD = ±0.23 nm)], which may be due to fluctuations in
composition, structures, and structural defects in the underlying film. Whereas, the measured RMS
roughness values for the nano-machined areas of the 5.6 nm B6TFMO sample were, as expected, in
general smoother [values of 0.66 nm(SD = ±0.51 nm), 0.09 nm (SD = ±0.07 nm), 0.08 nm (SD = ±0.07
nm), and 0.12 nm (SD = ±0.09 nm) for depths of 0.4 nm, 3.7 nm, 4.6 nm, and 5.3 nm, respectively]
compared with roughness values for the pristine B6TFMO surface [0.37 nm (SD = ±0.27 nm)]. The
roughness values for when nano-machining reached the underlying NGO