h a l f b a k e r y
On the one hand, true. On the other hand, bollocks.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
In the food industry, reverse osmosis is used to concentrate a variety of liquids, ranging from fruit juices, to whey (to make whey protein concentrates), to maple sap (to make syrup).
It should be easy to design a counter-current osmotic filtering device, with two fluid inputs, and two fluid outputs.
As each fluid passes by the other, water moves from one fluid to the other, through the membrane, pushed by the difference in water potential.
If one fluid is, for example, fruit juice, and the other is a concentrated salt brine, water will move from the juice to the brine, resulting in concentrated juice, and dilute saltwater. No high pressure pumps are needed, so this is a very low energy consuming process.
If you have a cheap source of concentrated brine, then this will surely be more economical than reverse osmosis... You'd probably have to build your factory next to a desalination plant for this, but it could happen.
Even if you don't have a cheap external source of brine, it still might be more economical to re-concentrate the dilute saltwater into brine, than to run the high pressure pump needed for reverse osmosis. For example, if you have a solar thermal collector that can produce the necessary temperatures to boil down the salt water.
Brine used to concentrate tomato juice
... seems direct osmosis was thought of before, but now the materials are ready [loonquawl, Sep 10 2009]
||The bit that always seems glossed over in discussion about osmosis (and reverse osmosis) is where the magical membranes come from - they have to be thin, so they can't (presumably) take a great deal of pressure - which (I guess - bit of a leap here) means that they'd have to look something like stacked car-radiators (a tube snaking backwards and forwards within a frame, stacked x number of times to create a square module) in turn stacked together to provide a massively parallel processing unit - (no pressure means parallel processing in order to get the throughput) - The thin membranes also mean that the inputs have to be carefully filtered (to avoid particles of anything that might rip, tear or clog the delicate membranes) - combine parallel processing, trickle-feed, pre-filtered and cleaning requirements and all in all, you've got a pretty tricky thing to maintain - imagine if one of your modules springs an (inevitable) leak, how do you know which one (assuming it's one of thousands) and how do you repair it without taking down your whole array and checking each module Christmas-tree style? I'm not saying it's a bad idea (in fact, anything that's osmosis related gets a thumbs up from me) I'm just trying to imagine the practical difficulties of running an osmotic processing plant.