This device "could" produce 1.5 gallons per day, per square metre of solar collecting area. The researchers estimate a system suitable for a family might be built for $100, so double or triple that after profits etc.
I assume that the 1.5 gallons was the highest yield they would have had from a single day, had they had a 1 square metre solar panel, and the expected average yield would be less (accounting for cloud cover etc.)
But even with reduced yield, for very small communities of a few dozen people, this definitely could work out.
I'm a little confused then, later in the article it says "The team estimates that a system with a roughly 1-square-meter solar collecting area could meet the daily drinking water needs of one person." So something doesn't add up.
[Edit: the paper is available at https://pubs.rsc.org/en/content/articlehtml/2020/ee/c9ee0412.... It says, "To meet the average daily water intake for one adult (≈3.2 L),49 100 TMSS devices can be placed into a 10 × 10 array, filling an 1 m2 area, which would provide approximately 10–20 L of clean water every day depending on the weather condition."
The lower bound of 10l is outdoor performance on a partly sunny day. So I think they are just being conservative by saying it would meet the requirements of one person - coastal areas are frequently cloudy, and in some locations, there might be little sun for extended periods of time.]
'drinking water' often doesn't mean the water a human literally drinks, but all the clean water necessary for living everyday life (bathing, flushing toilets, washing dishes, cleaning clothes, cooking etc.).
I am pretty sure if you had to get your water from one of these devices you would not be using it to flush toilets or other non-essential use.
And actually it does seem like 'drinking water' is actually drinking water and not for all those other uses you mentioned. As 3.7 litres is the amount required for a human male each day.
Offhand I'd say that you are doubling the plumbing system plus you are using a much more corrosive fluid.
It would be an interesting experiment, since toilets use a great deal of water. Take a beach town, build a separate pressure water and sewer system, see if it's worth it. It's certainly more complicated than a sailboat's set up.
Sounded astonishingly high, but a quick search from a few sources (one cited [1]) checks out. Now to go down a rabbit hole of how this 82 gallons breaks down and what I can do to reduce my consumption...
I assume that the 1.5 gallons was the highest yield they would have had from a single day, had they had a 1 square metre solar panel, and the expected average yield would be less (accounting for cloud cover etc.)
But even with reduced yield, for very small communities of a few dozen people, this definitely could work out.