This seems like a solid advance. There aren't any odd nanostructures or high-maintenance components that I can see.<p><i>Why can't we use these to solve drought in Australia/California/etc.?</i><p>There are a couple of reasons.<p>One is that vapor based water purification is very energy-intensive compared to reverse osmosis (where impure water is mechanically forced through membrane pores small enough to exclude salts/viruses/bacteria). It takes at least 2258 kJ/liter to vaporize water. Reverse osmosis desalination of seawater currently requires in the neighborhood of 10.8 kJ/L (3 kWh/m^3). There's a modest factor-of-5 advantage for using solar thermal energy without first transforming it to electricity that can drive pumps, but it's swamped by the factor-of-200 higher energy requirements for evaporation. (Incidentally, this is why even "cheap" nuclear heat sources can't beat electrically driven RO plants when it comes to large scale desalination.)<p>The second reason simple thermal desalination units can't solve California's problems is a consequence of the first reason: you just can't get enough water from these units. They're fine for providing a bare minimum of clean drinking water. They are nowhere near capable of providing enough fresh water for everyone to water lawns, bathe, and wash things -- much less supply the real 800 pound Californian gorilla, agricultural irrigation.<p><i>If this system is so inefficient, why would anyone use it?</i><p>The advantages are downward scalability and low infrastructure requirements. There are a lot of parts of the world where there's enough water to irrigate crops but it still might make people sick to drink it untreated. Solar distillation can remove arsenic/bacteria/viruses from water and make it safe for direct human consumption. Reverse osmosis units are advantageous only at larger scale, they require more maintenance, and they need electricity. Solar distillation can serve locations that lack reliable electricity and/or maintenance workers. (Solar-PV electricity might be an option in the future, but that would still be more complex and would require further design iterations on RO; current units are designed to operate near continuously.)