“multifiltration” Is a term NASA uses to describe a filtration bed used for removing dissolved contaminants from water in their water recycling system. I don’t know if they invented the term or what, though. They seem to have a brand name for it for some reason,”Unibeds,” though I don’t know why they do that. I wasn’t able to find any more into on unibeds.
There is a document in the references archive that I posted with “urc-1998” in the filename. It describes how the filter beds are composed of several different filtration medias; The first is MCV resin, this either releases iodine into the wate, killing bacteria and inactivating viruses, or absorbs it, depending whether or not it has been soaked in an iodine solution beforehand. A similar iodinated resin stuff is used the the “lifestraw” drinking water filter. There is a complimentary resin stuff that I think nasa developed, to remove iodine, too. The russian space agency used silver somehow, instead, in this role.
Next is the mixed bed ion exchange resin bed. There is a whole folder on ion exchange in the reference file. Basically this would be ion exchange resin of the hydrogen and hydroxyl form like used to produce demineralized water, and I think removes pretty much all the ionic contaminants. I don’t know why they put it first in line, I thought it could be damaged by organic material, but maybe that is because it adsorbs the material. Ion exchange is sort of like a special type of adsorption, wherein all the adsorption sites are the same, and quite strong. Activated carbon can remove ions, too, to some degree. You can see there is some non-mixed ion exchange resin there, too. I don’t know what that’s for. I can only assume it adsorbs something the other ion exchange resin does not, somehow, or it might be there to provide some advanced warning that the filter bed is exhausted, by releasing certain stuff into the water when the filter bed is almost used up (though there is already another mechanism to detect that, though.) EDIT: I think this might be intended to do some remineralizing, to reduce the corrosivity of the water. It may be releasing sodium ions and bicarbonate, or something.
Next is the activated carbon. There are many, many different kinds of activated carbon, and they all have different abilities to adsorb different materials to different degrees. What you want to look up here is the “isotherm” for a specific adsorbate-adsorbent pair. For example, iodine and activated carbon (the material surrounding the carbon also has some effect, in this case water, and also other contaminants.) For detail, I will simply refer you to the wikipedia article on adsorption. For some reason freundlich isotherms seem to be the most commonly used. However I have read many times that it is a good idea to do a small scale test before building you big system, to make sure you’ve got the right kind and amount of carbon. However, from what I have read there are 6 broad types, in terms of their adsorbate-adsorbent chracteristics, this is a good site to read up on that : http://www.tigg.com/granular-activated-carbon.html .
There are also kinetics to be concerned about. You can never have to good kinetics. Basically the poorer the kinetics, the slower the water must flow through the activated carbon bed. With perfect kinetics you could flow it through as fast as you wanted to. But it takes a time for the concentration in the surrounding water vs. the amount of material adsorbed to the carbon referred to in the isotherm to be approached. The numbers in the isotherm represent a situation that is achieved only after a period of time (how long depends on how the isotherm was measured), i.e. the equilibrium levels. That is, it takes time for the contaminants to diffuse to the adsorption sites and stuff. The smaller the particle size, the better the kinetics. I never found any really good info on kinetics, but apparently usually you just use the smallest particle size you can get away with based on other factors like cost, and pressure drop through the activated carbon filter bed etc. and do that small scale test to check things out and make sure it’s working good enough to get the job done. There is this handy thing called the zero length column test, I don’t have any handy links, but I’m sure you could read up on it somewhere, that can be used to get some hard data on isotherms and kinetics for the particular contaminants and carbon you are thinking of using. Actually, I suspect the kinetics are mush the same for the different contaminants, because it’s basically a brownian motion diffusion process, with some turbulence in the ware and stuff thrown in, which should be more or less the same for different contaminants, I think, but that’s just a guess. There is a folder on some stuff I happen to have collected on adsorption in the reference stuff, too.
You can see there are 2 activated carbon beds. Presumably these are different types of activated carbon, intended to adsorb different stuff.
The polystyrene-divinylbenzene stuff is more adsorption media. Again, made to adsorb some contaminants that activated carbon doesn’t do a very good job on. There are many different kind of adsorption media, zeolite, activated alumina, but this stuff is actually the same plastic ion exchange beads are made from. The production process shares a lot in common with ion exchange resin, but this stuff has no functional groups in it at all.
Then they have some more mcv resin. I think this is absorbing iodine, because as you can see elsewhere in the doc, iodine poisons the catalyst downstream.
Read the docs on the nasa system to see how they handle stuff the Multifitration beds can’t get. There is a link to the abstract of a paper I came across which supposedly describes the stuff that accumulates in a system like this, but I couldn’t get ahold of the full document.