Direct Patterning of CsPbBr3 Nanocrystals via Electron-Beam Lithography

Lead-halide perovskite (LHP) nanocrystals have proven themselves as an interesting material platform due to their easy synthesis and compositional versatility, allowing for a tunable band gap, strong absorption, and high photoluminescence quantum yield (PLQY). This tunability and performance make LHP nanocrystals interesting for optoelectronic applications. Patterning active materials like these is a useful way to expand their tunability and applicability as it may allow more intricate designs that can improve efficiencies or increase functionality. Based on a technique for II–VI quantum dots, here we pattern colloidal LHP nanocrystals using electron-beam lithography (EBL). We create patterns of LHP nanocrystals on the order of 100s of nanometers to several microns and use these patterns to form intricate designs. The patterning mechanism is induced by ligand cross-linking, which binds adjacent nanocrystals together. We find that the luminescent properties are somewhat diminished after exposure, but that the structures are nonetheless still emissive. We believe that this is an interesting step toward patterning LHP nanocrystals at the nanoscale for device fabrication.

S2 S1. EUV exposed and developed CsPbBr 3 nanocrystals We exposed some samples to extreme UV light (13.5 nm, 92 eV) in the Swiss Light Source of the Paul Scherrer Institute in Villigen, Switzerland. Using interference lithography 1 we wrote lines with 100 nm pitch. Although we were not able to optimize the system for this type of exposure, initial results point towards the fact that patterning perovskite nanocrystals with this type of highenergy radiation is possible and an interesting route to explore in the future. Figure S1 shows some examples of EUV patterned lines (100 nm pitch) at different doses. From doses of about 475 mJ cm -2 we start to see contrast and interrupted lines of CsPbBr3 left on the substrate after development. From 600 mJ cm -2 and up we observe well defined lines of nanocrystals. The size dispersion in the nanocrystal size, however, disrupts smooth line edges. to right 5 μm, 1 μm, 500 nm, 200 nm, 100 nm and 50 nm) exposed to different doses (top to bottom 8000, 4000, 3000, 2000, 1500, 1000, 500 and 200 μC cm -2 ; text is exposed to 10 000 μC cm -2 ). It is clear that the increasing dose leads to increasing contrast between the substrate and the exposed patterns. However, at higher doses (from 2000, but more clearly from 4000 µC cm -2 ) there are also more nanocrystals staying behind on the substrate due to overexposure or blurring of the intended patterns. This is an effect of electrons scattering outside of the intended region and causing cross-linking reactions outside the intended area. (b) Zoom-in of the top of 5 μm lines exposed to 2000 μC cm -2 and text exposed to 10 000 µC cm -2 . (c) Zoom-in of top of 200 nm, 100 nm and 50 nm lines exposed to 8000 μC cm -2 . (d) Zoom-in of top of 100 nm exposed to 8000 μC cm -2 .

S4
S3. Raw PL spectra CsPbBr 3 before and after exposure Figure S3. Raw PL spectra of exposed (2000 µC cm -2 ) and unexposed CsPbBr 3 . After exposure we observe a reduction of PL intensity as well as a broadening on the high energy side. The normalized data can be found in Figure 3a of the main text. Data is taken from PL map shown in Figure 4c of the main text.

S5
S4. Optical microscope images of e-beam exposed perovskite nanocrystal films Figure S4. Optical microscope images of 3 different e-beam exposed perovskite nanocrystal films. Exposure to the ebeam alters the absorbance of the films, as also observed in Figure 4.

S6
S5. Absorption and reflectance maps of exposed films Figure S5. Absorption (a-c) and reflectance (d-f) maps of exposed CsPbBr 3 films. Exposure doses are the same for all samples. White labels indicate the dose factor, which should be multiplied with 2000 μC cm -2 . Individual exposure fields are 50 μm x 50 μm. Absorption and reflectance was recorded at the same time. Overall summary of these measurements can be found in the main text (Figure 4b).
S7 S6. AFM of exposed but undeveloped perovskite films Figure S6. AFM of perovskite nanocrystal film exposed to EUV (92 eV). From previous work we know that EUV induces a similar patterning mechanism to e-beam. The edge of the exposed area is measured here. Although a change in roughness can be observed after exposure, the overall thickness of the film does not change significantly.

S7. Transfer matrix optical modeling
We use the python script provided by the McGehee group 2,3 to calculate how changes in refractive index may affect the measured absorbance and reflectance for our system. We     With a thinner layer of 50 nm ( Figure S10c) the influence of the substrate becomes more substantial and we again see more backscattering. With a thicker perovskite layer of 200 nm ( Figure S10d) there is more elastic scattering inside the perovskite film, causing a broadening of the cone of electrons. In both cases this may have a small influence on the blurring, but the vast majority of the electrons still penetrates the film fully in a narrow cone. It should also be noted