[PhantomGremlin]

What I'm more impressed with, though, is the scientific journalism.

Actually I was disappointed with the article.

The image at top shows two output streams: waste, and filtered water. I was expecting an explanation of this while reading the article. But as I read the words, the build-up wasn't matching the image.

Finally we come to: In a working system it would simply be a question of splitting the water stream into three as it left the processor, with the two outer branches being recycled and the inner one tapped and piped to consumers.

So, no, pretty sloppy.

[microcolonel]

The sentence before the one you mention explains it; the only problem with the top diagram in the header image is that it doesn't make it clear that opposite the CO₂ side of the stream, there is an air side. Seems like the diagram represents their experimental model (two exit streams), not the practical application (three exit streams).

> As the team hoped, this arrangement caused suspended particles with positive surface charges to concentrate towards the CO₂ side of the water stream, and those with negative surface charges to concentrate towards the air side, leaving the centre of the stream more or less particle-free.

[jerf]

I just checked this after reading the article, and that diagram appears as-is in the original paper, per jesseaustin's link to https://www.nature.com/articles/ncomms15181 . So I'm actually going to have to give the journalist a pass; I hardly expect them to produce more accurate diagrams than what is in the original paper.