✨ Wild nucleation in liquid-cell TEM ✨

In this video, Liam Spillane from Gatan captures a gold crystal growing wildly for several minutes.

But more interesting things are happening here, other than just the fascinating growth: The lighter area to the left is supposed to contain zero liquid! But clearly, as crystals are growing there, precursers must be present, and therefore, liquid.

This area is where two SiN membranes are bonded together with a thin film of Al2O3 in between—normally, no liquid can enter this tightly bonded region.

But the gold clearly has an affinity for the surfaces. It paves its way into the bonded area, splitting the membranes apart, and grows just as wildly inside this confined region as it does in the deeper liquid on the right.

The "deep" liquid in this case is around 40 nm, but this area in between the bonded regions must be very close to zero thickness!

Details:
- Membranes: 15 nm SiN with ~3 nm Al2O3 coating
- Liquid: 0.5 mM HAuCl4
- Imaging done by Gatan Inc. using their F-200 TEM from JEOL USA and the Metro detector.