Sciencemadness Discussion Board

Drying with molecular seives - how fast?

JefferyH - 11-9-2014 at 13:59

I've never worked with molecular sieves before but found out how affordable they are and was hoping to pick up some 3a sieves to dry some solvents. Most of what I found said it takes 2 days to dry a solvent when the solvent is stored with the sieves... how much faster could the solvent be dried if I just stirred the sieves with the solvent for a a reasonable time?

aga - 11-9-2014 at 15:09

Oh dear.

I was under the impression that you just poured liquids through them to magically dry liquids, in a Molecular Sieve kinda way.

Tin, Label, Error ?

If anybody actually knows how to use Molecular Sieves, please tell.

Specifically, how Long the contact, which Size etc.
(i got 100g of 4a)

JefferyH - 11-9-2014 at 15:22

aga that could also work, depending on how much surface area contact there is. You sound like you have done that before, have you had any bad experiences with reactions using solvents dried that way?

I can't tell if you are being sarcastic because you used the word "magically", but the sieves dry by manner of only certain sized particles being able to fit into the porous interior, with other larger organics sliding right off.

An idea I had for drying solvents was construct a make-shift vacuum filter thing, having a long cylindrical tube filled with something a lot of calcium chloride or other drying agent, and just pour the solvent through. After most drips down I could apply the vacuum. Have to say I haven't tried it yet, so I am unsure of the effectiveness or whether or not any solvent would be lost.


[Edited on 11-9-2014 by JefferyH]

Metacelsus - 11-9-2014 at 16:17

Doesn't heating work to dehydrate them?

Amos - 11-9-2014 at 16:58

Quote: Originally posted by Little_Ghost  
Quote: Originally posted by Cheddite Cheese  
Doesn't heating work to dehydrate them?

Apparently yes, but talking mum into letting me put ANYTHING remotely related to chemistry in her precious cooker just isnt going to happen. So for me it would be easier to freeze dry :D and a lot quicker


This is where a cheap food dehydrator(which I mentioned in an earlier post) would really do the job, even if it is a laughable notion. You get a nice hot stream of air constantly passed over the interior. But the reception to me posting about that was less than great, so maybe not.

smaerd - 11-9-2014 at 17:25

They require some pretty serious conditions for regeneration. Bunsen burner constant shaking under high-vacuum works(until the pressure gets too high remember PV=nRT then cool and repeat). Kugehl rohr would be infinitely better. I can't be certain but I don't see how a pressure cooker would remove moisture from them, atleast efficiently. Food dehydrator might work, but again it might take ages. It takes several hours to regenerate maybe 250mL of sieves by hand. Granted that is a lot.

Typically solvents are stirred with molecular sieves while sealed for a prolonged period of time(24 hours, etc). The pore size determines what can be sequestered and from what solvent medium. There is plenty of information about this on google, and some scientific papers specifically relating to their efficacy. Of course sequestration is a function of time and available sites/concentration.

Being more specific can help a lot. IE: What is being dried or removed and from what, how dry does it have to be, etc.

JefferyH - 11-9-2014 at 17:35

I'm thinking of removing water from solvents like acetone, methanol, ethanol, and so on. Lets say we assume a water content of less than 2%, probably 1% or less. How long of a drying time are we talking about here?

smaerd - 11-9-2014 at 17:46

About a day with the appropriate amount of sieves to be damn sure it's 'dry'.

ziqquratu - 11-9-2014 at 18:22

To get extremely dry solvents (e.g. suitable for organometallic reactions) starting from reasonably dry solvents, 10%w/v of 3A sieves (so, 100g of sieves per litre of solvent). Allow it to stand for 48h and that should be quite suitable.

As for regeneration - first get rid of most of the solvent (allow the sieves to stand in air for a while, or flush some compressed air through a beaker full of them). This is important to prevent the vapours catching fire - which can be a problem even if you don't use a flame (Et2O autoignites at around 400*C). You can then easily dry them by heating - I think about 400*C overnight works (never done it that way myself), but it's quite a bit more practical to heat under high(ish) vacuum (say <10mbar) - I used a vacuum oven which allowed me to heat them at 150*C for 4h, but you can also use a Bunsen burner (or whatever) and get it done much faster (minutes, I believe).

One last point - do not attempt to dry acetone with molecular sieves. Sieves are mildly acidic, and are respectable catalysts for the aldol condensation - which means not only do you fill your acetone with gunky aldol polymers, you also generate water in the process... This can be a noticeable problem after less than 12h. For acetone, stirring with MgSO4 works quite well enough for most purposes.

I'm fairly sure I've posted this reference before, but I've also attached it here again.

Attachment: Efficient drying of organic solvents using molecular sieves or activated alumina.pdf (765kB)
This file has been downloaded 2343 times

Magpie - 11-9-2014 at 19:26

ziqquratu:

Thanks for posting that JOC article - much useful information there.

400°C seems unnecessarily high for regeneration. This might be tough to reach for many. Most of what I read says 250°C will do.

I had an autoignition event when regenerating my 3A sieves in my kitchen oven (sorry, Little Ghost). I set the oven for 400°F which is 204°C. I definitely recommend getting those sieves solvent free before regenerating!

Your 400° for autoignition of Et2O seemed way too high - Wiki says 160°C.

TheChemiKid - 12-9-2014 at 03:53

Can sieves be destroyed from too high a heat, or could I just use my oxy-propane torch?

ziqquratu - 12-9-2014 at 04:56

Oh my goodness, I must have been asleep writing that!

Yes, 160*C for the autoignition temperature of diethyl ether is much more reasonable. As for the 400*C for activation of sieves, perhaps I was recalling drying something else.

Thank you, Magpie, for the important corrections!

TCK, I don't know about destroying them with heat, but I suspect your glass would fail before the sieves...

Bikemaster - 12-9-2014 at 10:43

Quote: Originally posted by TheChemiKid  
Can sieves be destroyed from too high a heat, or could I just use my oxy-propane torch?


Yes, it can be destroyed. You need to stay under 500°C to prevent degradation.

aga - 12-9-2014 at 13:43

So no Magic then. Bugger.

I just stuffed 100g 4a balls into a sep funnel added to the end of my ethanol disty rig in the hope that they would sieve out some water molecules.

'Regeneration' was done by sticking the balls in a tin can on the hotplate until they looked dry again (they looked wet after 'use')

Molecular Sieves seemed like a Magic Bullet, so once again, if it Seems too good to be true ...

Thanks for the responses pointing out that it can take DAYS for the molecular sieves to do their work.

SM once again has handed out the Real Deal.

Edit :

For practice i should (properly) regenerate the 4a sieves i have, and do a quantitative comparison with dry NaCl as a dessicator, as that's much more easily available.

Yes. That sounds likely in the near future.

[Edited on 12-9-2014 by aga]

ziqquratu - 13-9-2014 at 04:01

Although other desiccants may appear to work faster, the ultimate level of dryness may well be nowhere near as good. However, measuring the residual water in order to compare and contrast may very well be outside the abilities of a home chemist (and a large proportion of professional ones - myself definitely included!)

One other thing, when regenerating sieves you can (qualitatively) check that they've been activated - if you take 2-3 pellets in the palm of your hand and add a drop of water, activated sieves will get really hot really fast (I'm not talking mild warmth, you can most definitely feel it!!)

DJF90 - 13-9-2014 at 05:31

You'd use Karl Fischer titration to measure water conte,t of solvents. I do so on a daily basis in my line of work. I don't quite understand how people can complain that solvents like THF come "wet" - I've had readings as low as 50 ppm from a drum (that had been perviously opened) and I suspect,much of the problem stems from academia when they've not even bothered to check (I don't think I ever saw a KF titrator at university)