Robotic fabrication of high-quality lamellae for aberration-corrected transmission electron microscopy

Aberration-corrected scanning transmission electron microscopy (STEM) is widely used for atomic-level imaging of materials but severely requires damage-free and thin samples (lamellae). So far, the preparation of the high-quality lamella from a bulk largely depends on manual processes by a skilled operator. This limits the throughput and repeatability of aberration-corrected STEM experiments. Here, inspired by the recent successes of “robot scientists”, we demonstrate robotic fabrication of high-quality lamellae by focused-ion-beam (FIB) with automation software. First, we show that the robotic FIB can prepare lamellae with a high success rate, where the FIB system automatically controls rough-milling, lift-out, and final-thinning processes. Then, we systematically optimized the FIB parameters of the final-thinning process for single crystal Si. The optimized Si lamellae were evaluated by aberration-corrected STEM, showing atomic-level images with 55 pm resolution and quantitative repeatability of the spatial resolution and lamella thickness. We also demonstrate robotic fabrication of high-quality lamellae of SrTiO3 and sapphire, suggesting that the robotic FIB system may be applicable for a wide range of materials. The throughput of the robotic fabrication was typically an hour per lamella. Our robotic FIB will pave the way for the operator-free, high-throughput, and repeatable fabrication of the high-quality lamellae for aberration-corrected STEM.


Supplementary Methods
We explain how AutoTEM 5 software works in detail.

Major parameters that a user can choose and tune
AutoTEM 5 stores many parameters in one recipe. Here, we explain major parameters that have a fundamental impact on fabricating high-quality STEM lamellae. We also note how we set the parameters in this study.

User-chosen parameters in a rough-milling process
To form a protective layer, a user can define deposition material (Pt, C, or W) and the height of the protective layer. AutoTEM 5 automatically calculates the deposition time. Optionally, a user can set an electron-beam deposition before FIB deposition. Then, a user can define the dimension of the chunk (width, depth, and thickness). AutoTEM 5 automatically sets the trench shape whereas the user can change the trench angle and its depth if needed.
To correct the difference in the milling rate of each material, the user can define the correction factor of milling rate that is the relative value to the one of single-crystal Si. This correction factor automatically multiplies the milling durations of all the FIB milling in both the rough-milling and final-thinning processes. For example, we set the correction factor of 3 and 4 in SrTiO and sapphire, respectively.

User-chosen parameters in a lift-out process
This process has a small number of parameters. A user can define the position of the TEM grid where the TEM chunk is transferred. As reported in ref [1], the user can also define whether the attached destination is the side position of a TEM or the top position (see an example in the 3 later part of ref [1]). In this study, we selected the side position so that we can minimize the effect of redeposition from the Cu grid.

User-chosen parameters in a final-thinning process
In AutoTEM 5, the final-thinning process is mainly composed of four steps: step i) thinning of the entire chunk by 30 kV-FIB, step ii) further thinning of the "window" region of the chunk by 30 kV-FIB, step iii) polishing of the window region by low-kV FIB (V1), and iv) final polishing of the window region by lower-kV FIB (V2). In step ii), the user can define the width of the window region. In this study, we set the width of the window as 2.4 μm (see Fig. 1d). In the steps of iii) and iv), the user can define the two values of acceleration voltages used in low-kV polishing, V1 and V2. In this study, we set V1 = 5 kV and V2 = 2 kV.
For every process of FIB thinning / polishing, the user can define FIB current, over-tilt angle, target thickness, and milling offset. For every process of FIB milling, the end-point is determined by the milling time. AutoTEM 5 calculates the milling time of each milling process from the inputs of the target thickness, the width and the depth of the milling region, and the correction of milling rate relative to that of single-crystal Si. See the following section on the definition of the target thickness.

Definition of the "target" thickness in AutoTEM 5
Target thickness in AutoTEM 5 is a nominal value, not measured physically. AutoTEM 5 sets the two milling boxes so that the gap between the two boxes is equal to the sum of the target thickness and the probe size of the FIB at the user-chosen acceleration voltage and current. See also the following schematic.

4
Especially, in the final low-kV polishing, the difference between the "target" thickness and the "actual" thickness has a primary impact on the resulting thickness of the STEM lamella. So, AutoTEM 5 can fine-tune the position of the milling box as an extra parameter of milling offset (see Supplementary Fig. 4).

Image recognition of X-shaped alignment marker
X-shaped makers are initially fabricated by FIB deposition on the bulk material and FIB milling (see Fig. 1a). The bigger marker is for aligning the milling positions in a rough-milling process.
The smaller marker is for aligning the milling positions in the final-thinning process (see Fig.   1d). To correct the position of FIB milling, AutoTEM 5 repeatedly scans the X-shaped alignment marker. Then, AutoTEM 5 calculates the cross-correlation between the original and the present FIB images. In case of the sample is tilted, the effect of tilt is also considered. From the results of cross-correlation, AutoTEM 5 corrects the position and starts FIB milling. The value of cross-correlation is also used when the FIB focus is aligned.  Magenta circles are eye-guides of diffraction spots of 55 pm resolution. Green circles correspond to 60 pm resolution. The scale bar is 1 nm.