[NormalPolymer]

Dear all:

This is quite annoying to me... To remove the water in ethylene glycol, I tried to stir the EG with 3A molecular sieves using a stir bar for an hour. The water content is unexpectedly increased from 1.2% to 5.2%, measured by Karl-Fischer titration. I also tried a short column packed with molecular sieves, the water content was also increased to 2.7%. The molecular sieves should work, because I can feel the heat if a drop of water was added to molecular sieved hold on my palm.
EG looked milky after stir, is it because of the broken pieces of molecular sieves? The filtrate was still milky, could that milky thing interfere with the K-F titrator?

Thanks in advance for your kind *delete me*

[Arkcon]

A quick Google search on the subject points to a reference that does not suggest molecular sieves for drying ethylene glycol.  I don't know if you have a reference that suggests it, or if this is a new occurrence with a previously used procedure, but you should have some source of suitability for this task.

http://delloyd.50megs.com/moreinfo/drying.html

You say there was more moisture after the procedure, perhaps your sieves were old and saturated with moisture?  But you felt the reaction with water, so that must not absolutely be the case.  Still, it could have some water, and still be able to absorb more.

The cloudiness, if it is pronounced, could be the sieves dissolving in the ethylene glycol.  Or it could just be the refractive index of moisture dissolving.  I didn't see what you saw, so I can't differentiate the two.  I don't know if ethylene glycol dissolves molecular sieves, or if ethylene glycol can enter a 3A pore.  If it can, then it will likely drive the moisture out, and ruin the bulk of the solution.

Sometimes old molecular sieves just crumbles, and contaminates your solvent, so maybe that's what happened.

I've performed Karl-Fisher titration of very complex mixtures, so I'd suspect the cloudiness wouldn't cause a false high reading.  But I'm not sure.

Sorry, but that's an awful lot of "I dunno".  Can you try again and see if you get the exact same problem?  Have you done this before so you're sure this works?  Can you try again with fresh reagent to see if this is a fluke?

[fledarmus]

You have to be very careful if you are stirring mixtures containing molecular sieves - depending on the structure, the shape, and the supplier, many are quite fragile. If you are crushing up your sieves, you would make an extremely fine precipitate (cloudiness), and releasing any water that might have already been contained in the sieves. You usually get better results drying solvents if you just leave a layer undisturbed in the bottom of the bottle overnight and let normal dispersion forces do your mixing, or in reactions using an overhead stirring with some space between the stirrer and the bottom of the flask so you don't grind the sieves.

I don't know how to explain the column results, though.

[NormalPolymer]

Hi, Arkcon and fledarmus:

Thanks a lot for your kind *delete me* I asked around and got an explanation: Molecular sieves can be a catalyst for polymerization of EG, and water formed. Does it make sense to you?