Gatan oneview camera

The liquid cell is made up of two chips: (1) a heater chip with a 50 nm thick silicon nitride (SiN_x) membrane window (Hummingbird Scientific, Lacey, WA, USA) and (2) bottom chip with a 25 nm SiN_x window (Supplementary Fig. 1) that we fabricated in-house^{50 (link)}. These chips were first plasma treated before experiments to make their SiN_x membrane surfaces hydrophilic, so that liquid can easily wet the interior of the liquid cell. Next, ~500 nL of the nanocube solution was drop casted onto the heater chip and allowed to dry. Then, the liquid cell is assembled within a liquid cell TEM holder (Hummingbird Scientific, Lacey, WA, USA). These procedures are described in detail in the Supplementary Methods.
The holder was loaded into a JEOL 2010FEG TEM (JEOL Ltd., Tokyo, Japan) operated at 200 kV. After heating the Ag nanocubes to the targeted temperature (±10 °C), 1.0 mM of the Au precursor solution was flowed into the cell through the fluid tubing using a syringe pump at a rate of 20 µL per min. Images were recorded with Gatan OneView camera (Gatan Inc, Pleasanton, CA, USA) at a rate of 25 frames per second. The image processing algorithm used to extract the etch rates of nanocubes is described in the Supplementary Methods.

Transmission electron microscopy (TEM) was used to visualize the assembly state of VCC upon association with the insoluble fraction of curcumin in the aqueous buffer. For this, 10 μl from a stock solution of curcumin dissolved in DMSO (20 mg/ml) was added to VCC (2 μM) in a final reaction volume of 1 ml in PBS, and was incubated for 1 h at room temperature. For the control reaction, 20 μl from a stock solution of curcumin dissolved in DMSO (20 mg/ml) was added to the reaction volume of 1 ml in PBS without VCC. The reaction mixtures were subjected to centrifugation at 10,000 rpm for 10 min. The pellets were re-suspended in 200 μl PBS. Samples (4μl) were added to the plasma cleaned 400 mesh copper grids with continuous carbon support (Aritech Chemazone Pvt. Limited), and incubated for 1 min. The excess amount of sample was blotted off and samples were negatively stained with 2% uranyl acetate for 1 min. The excess uranyl acetate was blotted off and the samples were dried at room temperature. The samples were analyzed on a JEM-F200 electron microscope (Jeol Inc.) operating at 200 kV, and images were recorded with a Gatan OneView camera (Gatan Inc.) at a nominal magnification of 60,000×, with a defocus −2.5 μm and the cumulative fluence for the images was limited to ∼60 electrons/pixel.

Two TEMs were used for in situ heating studies: Thermofisher Titan S/TEM (Thermo Fisher Scientific Ltd., Hillsboro, OR, USA) equipped with a Bruker Xflash 6 T | 30 energy-dispersive X-ray (EDX) spectrometer (Bruker, Billerica, MA, USA) and an aberration-corrected JEOL ARM300F (JEOL Ltd., Tokyo, Japan). Both TEMs were operated with an accelerating voltage of 300 kV. The Wildfire heating holders (DENSsolutions, Delft, Netherlands) were used for heating the NPs during in situ observations. Typical electron fluxes used for in situ imaging were in the range of 200−6000 e⁻ Å⁻² s⁻¹. Image series were acquired with two different cameras: a Gatan K2 IS camera (Gatan Inc., Pleasanton, CA, USA) on the Thermofisher Titan S/TEM and a Gatan OneView camera (Gatan Inc., Pleasanton, CA, USA) on the JEOL ARM300F. Inverse FFT images in Fig. 1c were obtained by separately filtering and then combining fcc and B2 spots in FFT patterns in Fig. 1a, and the same processing was used for Supplementary Figs. 3, 4, 10, and 14.