Forward osmosis or FO, is supercool. You’ll find a document in the reference archive with “elimelech-jms” in the filename that has some interesting info about it, and in the membrane bioreactors folder, some docs about MBRs that use it.
There is a company called HTI that makes products like the “seakpak” and |”hydrowell” that uses FO. Apparently (this info partly from the elimelech doc) the membranes used by HTI are fiber reinforced cellulose triacetate (CTA). If you look in the reference archive, you can see there is an MBR using a CTA membrane. I would have thought the CTA would get digested by the bacteria, but apparently not, or maybe it does, but they chose to ignore that for the time being. Indeed, the HTI hydrowell lasts for only 60 days (at $400 per membrane), with the failure mechanism being some kind of bacterial fouling, or something bacteria related anyway, apparently. The Seapak has an even shorter life for some reason. In reverse osmosis systems that use CTA, they need to be cleaned and/or to stay exposed to chlorine, or mold grows on the CTA.
The other common material used on osmotic systems is polyimide, which can’t stand chlorine, apparently, but might be a better match for an MBR.
But wait, I forgot to tell you how cool a forward osmosis MBR would be – You see, remember, the higher the MLSS is, the more compact the MBR can be. The limiting factor on how high it can be here, is the fouling of the membrane, remember. Yeah, exactly. Because the forwards osmosis process is not pressure driven, you don’t have the problem of particles being pushed against the membrane, fouling it. You could have a super duper high MLSS!
The next awesome thing is that is leaves the dissolved contaminants in the digestion chamber.
Fo separates the dissolved stuff too, so the dissolved concentration can be quite high in the digestion chamber, which speeds up the rate of digestion even more.
Thirdly, you could potentially get rid of a lot of the polishing stuff, all (or a lot of) that activated charcoal. Fourthly, the water could be really clean, even cleaner than you might expect to get with activated charcoal polishing.
There are 2 main barriers: One, getting a good membrane. There are some membranes made for FO. One is mentioned in one of the documents (“forwardosmosismembranesuggestion”). That’s a CTA membrane and worse, it’s spiral wound. There is the HTI stuff, again, CTA, I doubt it is expensive, more likely they are just charging what they can get for the hydrowell filter stuff. In the elimelech doc they talk about using flat plate RO membranes, which could be polyimide, and while definitely not ideal, maybe those could be used. Maybe you could just get a film of a suitable polyimide 50 microns thick from somewhere? Hm. A good internet hunt might turn up something better.
The other problem is the draw solution. You need to get the draw solute back out of the water somehow, of course. There are quite a few options here because you choose the draw solute. Electrosorption might be doable at home, and electrodialysis, and ion exchange, which could be regenerated with those methods. The obvious one is RO, but that’s pretty expensive and power hungry. Another solution is magnetic nanoparticles, which sounds fancy, but some types can be produced at home, including iron nanoparticles (which might leave you with a lot of iron in the water, though). They are removed, of course, with a magnetic field. The most interesting, I think, for a do it yourselfer, could be just using really big molecules, like polyethylene glycol, and then remove them using cross flow ultrafiltration. That could be pretty sweet, and cheap, too, if it would work. There is a doc in the Ref archive about using PEG.
Lastly, I don’t know how practical this would be, but I have read about ammonia and carbon dioxide being combined to form the draw solute, under pressure. Then reduce the pressure, the gasses come back out of solution and you have your water back. The problem is you have to get all or mostly all the co2 and nh4 out, or it gets lost to the outgoing water stream. That takes some slightly more sophisticated machinery.
Anyway, if the details could be figured out, it would make an awesome water treatment system for a tinyhouse, very compact, inexpensive, low power consumption, very simple to maintain, environmentally friendly.