The power generation potential given is in units which make it almost useless. A kWh is a unit of energy, converted into a frame of reference more useful than the joule. Saying that a device can generate 2 kWh doesn't really tell you much - it could generate that in 10 seconds (good) or 10 years (bad).

I think the quoted figure was meant to be in kW - a unit of power, which is a much more useful piece of information. I looked at the original paper (here: https://arxiv.org/pdf/1907.13170.pdf) but couldn't find any reference to the 2kWh quoted from the researchers.

If the suggestion is correct, and a 10mx10m panel can generate 2kW, then there's some really interesting potential applications. History has shown that the path towards robust technology is normally by making things solid state - vacuum tubes to transistors, solar collectors to solar panels, even helicopters to drones all benefited from a huge decrease in mechanical complexity.