216 Microsc. Microanal. 27 (Suppl 1), 2021
doi:10.1017/S1431927621001367 © Microscopy Society of America 2021
Novel thin film lift-off process for in situ TEM tensile characterization
Krishna Kanth Neelisetty
1
, Shyam Kumar CN
2
, Ankush Kashiwar
3
, Torsten Scherer
4
, VS Kiran
Chakravadhanula
5
and Christian Kuebel
6
1
Thermofisher Scientific, Brno, Jihomoravsky kraj, Czech Republic,
2
Institute of Nanotechnology,
Karlsruhe Institute of Technology, United States,
3
Institute of Nanotechnology, Karlsruhe Institute of
Technology, Belgium,
4
Institute of Nanotechnology, Karlsruhe Institute of Technology, Baden-
Wurttemberg, Germany,
5
Institute of Nanotechnology, Karlsruhe Institute of Technology, Telangana,
India,
6
Karlsruhe Institute of Technology, Eggenstein-Leopoldshafen, Baden-Wurttemberg, Germany
Recent advances in MEMS based devices for in situ transmission electron microscopy (TEM) are opening
exciting new directions in nanoscale research. The possibilities to combine various stimuli like biasing,
heating, etc inside the TEM to study various applied materials e.g., solid state batteries, non-volatile
memories, sensors, etc. are continuously increasing, but the challenges to prepare suitable specimens are
increasing at the same time. The majority of samples used for in situ TEM experiments are prepared using
Focused Ion Beam (FIB). Classic lamella preparation has been standardized and almost automated in
previous years. However, unconventional methods are being developed to address the research challenges
in sample preparation for in situ studies, e.g. to reduce dose and contamination. This study presents a
method for successful FIB lift out from a heating membrane of commercially available MEMS based in
situ heating chips onto the push-to-pull (PTP) device (Bruker) for in situ mechanical measurements,
minimizing ion beam damage and contamination of the area of interest.
Thin free-standing polymer films were deposited by spin-coating on a MEMS-based in situ heating device
(Aduro from Protochips) and nanocrystalline graphene (ncg) was formed across the holes of the heating
region as described elsewhere [1,2]. The area of interest for the lift-out is the hole across which the ncg is
formed. The heating region in the Aduro device is a free-standing membrane with 500 micron
2
area and
around 100 nm thick. During the heating cycles to form ncg, this membrane has been strained. The actual
part of interest, the freestanding ncg that formed across the hole is much thinner (around 10 nm), which
necessitates extra care during transfer. A simple crack or deformation of the heating membrane during the
transfer would lead to fracture of the ncg film and thus to failure.
After the region of interest (ROI) is selected, Pt/W is electron beam induced deposited (EBID) at one
corner (20 um from centre in order to avoid shadowing effect and diffusion-induced artefacts). The size
of the lift-out thin film is 50 μm x 50 μm to avoid any redeposition on the ROI during milling and also
acts as safe distance for the ion beam tail. The challenge is not to view the ROI with ion beam as just one
snapshot (~ 300 ns Dwell time, 1540 x 1200 resolution) would damage the ROI in the ncg.
The shape of the thin film lamella and the length of the FIB cut play an important role in the transfer
process. Initially square shaped thin film sections (50 um x 50 um) where cut by making rectangular cuts
of (50μm×1μm). The shape and the cut introduced bending because of strain relaxation in the lamella as
shown in figure 1a-c. This bending is a problem for the PTP as the thin film cannot be fixed properly and
finally compromises the data of the in situ straining measurements. Therefore, a pentagonal/hexagonal
shape (Figure 1d-e) was adapted to successfully reduce the bending. To relieve strain small holes across
the corners were drilled before the edge-cut as shown in figure 1d.
With this approach, bending of the film was minimized and fixing the corners of the film to a PTP was
easy with small amounts of EBID-Pt. The transferred thin film was subsequently strained inside the TEM
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S1431927621001367
Downloaded from https://www.cambridge.org/core. IP address: 3.80.168.101, on 05 Nov 2021 at 00:54:11, subject to the Cambridge Core terms of use, available at