Sciencemadness Discussion Board

Drying ether for Grignard

Stibnut - 29-8-2016 at 09:09

I'm going to be doing my first Grignard pretty soon - I made some methyl iodide over the weekend and I now have everything but the drying tube for CaCl2. I'm going to do it with methyl iodide because Chem Player considers methylmagnesium iodide the least fussy of the Grignards, as evidenced by his low-yield but successful "Ghetto Grignard" reaction. The goal will be to make tert-butyl alcohol from acetone.

I have about 250 mL of reagent-grade ether left from a 500 mL flask of it I purchased about three weeks ago; the rest was used earlier as a low-boiling solvent. I want to make sure it's dry, but I don't have sodium metal. I do have 5 g of lithium, though, unopened and packaged with argon. Could I use that as my desiccant? Other desiccants I have include CaCl2 and MgSO4; are these useful or would lithium be the best choice?

Melgar - 29-8-2016 at 13:12

First off, the easiest Grignard reagent I ever made was benzylmagnesium chloride. You can bubble chlorine into toluene in sunlight, and add a pinch of a bromide salt to generate non-ionic bromine species (via displacement by chlorine), which promote a more even distribution of halogen radicals, and a much faster reaction. Once you've chlorinated a third to a half of it, just heat it at reflux to drive off any HCl vapors. The toluene doesn't even need to be removed, since it's neutral to the reaction conditions as long as you use some extra ether.

As for the ether, drying it over sodium metal is standard practice, but sodium hydroxide works too, it's just that you have to store it with the NaOH for at least a week. I just put maybe 1 cm of NaOH prills in the bottom of a liter bottle, that always seemed to work. Maybe four days or so for KOH, but I'd definitely use a base, and not something that could somehow release an acid, or has a conjugate acid, since Grignard reagents are really strong bases. Lithium doesn't work as well, mostly just because it's so light it floats in everything, and water would sink to the bottom.

As far the Mg, I always had good luck just drilling holes in a block of metal and letting the shavings collect in ether. It was just a small block of Mg I got as part of a firestarter kit at Harbor Freight. Adding a tiny bit of iodine helped too, but don't stir it up after you add it. You want the iodine to diffuse such that it hits the Mg at a high concentration. Granted, that last bit it my own theory, but it's worked pretty well for me. Definitely start stirring once the reaction gets underway though.

Stibnut - 29-8-2016 at 18:40

Interesting - I didn't know you could chlorinate toluene that easily. I'm also surprised you had good luck with a chloride - I was under the impression that chlorine Grignards were more fussy than the bromine and iodine ones. As for magnesium, I've got a bunch of magnesium in shaving form already, so I'm good there.

The problem with using NaOH is that the ether container is aluminum, so it would be rapidly corroded. I'd be looking for something that isn't caustic. If I just dumped in some anhydrous CaCl2, would that be at all helpful? Or maybe I should think about getting molecular sieves?

[Edited on 30-8-2016 by Stibnut]

MeshPL - 30-8-2016 at 05:54

Chlorine Grignards are more fussy than bromine and iodine, but benzyl grignards are less fussy than aliphatic and especially vinyl and phenyl. In fact you can make iodine phenyl grignards much, much more easily than chlorine phenyl.

Why don't you transfer your ether into a different container?

I'm not sure how good is CaCl2, but if you keep molecular sieves long enough in ether, they will dry it really well. In fact I read that when drying THF, molecular sieves are faster and more effective than sodium + benzophenone, so you can definitely go and buy them. I'm not sure if they are so good fot Et2O, but they can't be that bad... Still CaCl2 may also be a good solution, but you must wait for a better expert...

Magpie - 30-8-2016 at 06:27

3A mole sieves are my "go to" drying agent for ether and ethanol with <1% H2O.

Stibnut - 30-8-2016 at 10:17

I could transfer it to another container - I was just afraid I'd expose it to more atmospheric moisture that way, this being the summer in the Midwest.

I went ahead and ordered 3A molecular sieves. It looks like they'll serve me very well as drying agents for all sorts of reactions, making them more versatile than sodium metal. I searched and found a thread from 6 years ago with the exact same title, where the poster did successfully dry his ether enough to run a Grignard with a combination of molecular sieves and CaCl2. I'm not even sure if he really needed the CaCl2.

chemrox - 30-8-2016 at 10:28

The ether should be fine. I always add the halide to a refluxing mixture of ether and Mg turnings (fresh is best).
(where the hell is the dictionary file for the site?) The refluxing ether will exclude the atmospheric water and O2 as much as is needed. The myth that Grignards are difficult is hard to suppress. Lab instructors and text book writers are to blame. An I2 cryx is sometimes helpful in initiating these.

Dan Vizine - 30-8-2016 at 13:46

For drying, molecular sieves are simply the best agent there is. The residual water is less than that left by Mg perchlorate, sodium wire, or benzophenone ketyl by orders of magnitude. They are applicable to nearly all lab solvents. They are superb for ethers. And they don't touch aluminum, so your nice light-proof container is fine.

zed - 30-8-2016 at 14:08

Hunh? You have a semi-freshly opened flask of ether? It should be dry enough.

Drop a crystal of Iodine, through the top of your condenser, onto your mound of warm Magnesium Metal, run in a few drops of Ether/Alkyl Halide mix. It should start to fizz. Thereafter, you are off to the races.

Starting a Grignard, with recycled ether, can be a challenge. Such ether has often been saturated with water, and it is sloppy wet. So, you don't start the drying process with an expensive reagent like Sodium. You start your drying with anhydrous Sodium Sulfate, or Magnesium Sulfate.

When you get yer Ether fairly dry, resort to other means.

As for drying truly wet Ether, with Sodium Metal......It doesn't work very well. The surface of your Sodium Metal, becomes encrusted with NaOH and H2 bubbles, blocking access to the Sodium's surface, and thus blocking further drying. Perhaps refluxing is helpful. Dunno.

Back in the old days Ether was cheap, I could buy it by the drum load, if I wanted to. So when drying wet Ether, proved to be a pain in the ass.... Well, I just bought fresh Ether.

[Edited on 30-8-2016 by zed]

Melgar - 30-8-2016 at 16:34

The reason I used NaOH and not CaCl2 or Na2SO4 is that NaOH also neutralizes a whole bunch of other things that can interfere with a Grignard, particularly CO2. CaCl2 can also neutralize CO2, but does so by releasing HCl, which is much worse. Store CaCl2 in an aluminum container exposed to air sometime if you don't believe me. If you do Grignards on a regular basis, get a wine jug, put it in the oven cold, then preheat the oven to like 300 with the jug in it, and then let it cool off when it gets to that temperature. When the jug gets to a temperature where it's not uncomfortable to hold your hand to it, you can transfer room-temperature fluids to it without any risk of it breaking due to thermal gradients.
Quote: Originally posted by zed  
As for drying truly wet Ether, with Sodium Metal......It doesn't work very well. The surface of your Sodium Metal, becomes encrusted with NaOH and H2 bubbles, blocking access to the Sodium's surface, and thus blocking further drying. Perhaps refluxing is helpful. Dunno.


There's a reason they use sodium metal though: sodium hydroxide is extremely hygroscopic, and forms a slush that the interior reacts with. And H2, being nonpolar, should be pretty soluble in ether, no? I mean, it would probably take several days to dry wet ether to the point where you could do a Grignard with it, but as long as you have a little patience, and possibly also magnetic stirring to keep the fluid circulating, it'd get there. Molecular sieves work the fastest due to their extremely high surface area, but I'd prefer using sodium on impure ether just because of how many other contaminants it takes care of.
Quote: Originally posted by Stibnut  
Interesting - I didn't know you could chlorinate toluene that easily. I'm also surprised you had good luck with a chloride - I was under the impression that chlorine Grignards were more fussy than the bromine and iodine ones. As for magnesium, I've got a bunch of magnesium in shaving form already, so I'm good there.

The problem with using NaOH is that the ether container is aluminum, so it would be rapidly corroded. I'd be looking for something that isn't caustic. If I just dumped in some anhydrous CaCl2, would that be at all helpful? Or maybe I should think about getting molecular sieves?

[Edited on 30-8-2016 by Stibnut]

I may have discovered this particular variant on the toluene chlorination reaction; here's the thread where I posted about it: http://www.sciencemadness.org/talk/viewthread.php?tid=14063

Nicodem was nice enough to replicate the experiment and do a proper analysis of the reaction products with NMR, and results were very positive.

As far as magnesium, it forms a passivating oxide layer like aluminum does, only not as quickly, which is why I drilled it myself and got it into ether as quickly as possible. Hence, less oxide layer. Once the reaction gets started though, you don't really need to worry about it anymore. It helps to break up the turnings while they're immersed in ether too, just so you're sure to get some raw metal exposed. As far as containers though, couldn't you just use a different one? I used a glass wine jug, back when I was doing Grignards using ether distilled from starting fluid. It never gave me any problems. Molecular sieves work fastest due to their high surface area, but Na and NaOH neutralize more potential contaminants, so I preferred them for that reason.

As far as why benzyl chloride works so well, it's because that methyl group on toluene forms particularly weak bonds, both to hydrogen and to chlorine. So there isn't even any need to use a heavier halide (which would also form weaker bonds).

You shouldn't use CaCl2 with either aluminum or ether that'll be used for Grignards, because it forms an equilibrium with CO2 and HCl. So if there's ever any CO2 present, there will also be some HCl present, and that will eventually eat through your aluminum. If you have the patience, and plan to recycle your ether, storing it in glass over sodium or NaOH is the way to go, because they'll neutralize so many more things, particularly acidic things that like to react with strongly basic Grignards.

Stibnut - 31-8-2016 at 11:36

Thanks for the advice and info, everyone! I have filled up a 100 mL reagent bottle up 1/3 of the way with NaOH and filled the rest with ether. It sounds like the Grignard shouldn't be as difficult as I thought it would be, provided I oven-dry everything. I'll wait a few days for the ether to dry over the NaOH and then give it a try. Once the sieves come in I'll put them in the main bottle.

One quick question: people are suggesting a little trace of I2 to start the reaction. I'm using methyl iodide, which even though I'm storing it with copper in a freezer, is still slightly yellow from traces of iodine. Does adding a little crystal help anyway?

In the future, I'll probably distill ether from starting fluid and getting it that way. Chemsavers' $52 for 500 mL is truly a rip-off, and from the thread I found on getting ether from starting fluid, it seems that it is easy to distill it from heptane in a water bath. I assume I could just add a little trace of BHT to keep peroxides from forming and then dry with a mix of both NaOH (mostly to get dissolved CO2) and sieves.

Metacelsus - 31-8-2016 at 13:48

I've heard that BHT-stabilized ether isn't very good for Grignard reactions, but I've never tried it myself. Peroxides aren't a problem if you use the ether promptly. If you do decide to store it, try and find the smallest possible (to minimize headspace) opaque container with a tight-sealing cap.

Adding iodine to get the reaction started probably won't be necessary if your magnesium is clean.

[Edited on 8-31-2016 by Metacelsus]

Stibnut - 31-8-2016 at 14:54

The ether I have now is BHT-stabilized too, but it's a tiny amount: they list a minimum of 5 ppm. Does anyone know if this is enough to prevent the reaction from happening? I don't mind if it adds a bit to the initiation time as long as the reaction actually works.

Melgar - 1-9-2016 at 17:19

Quote: Originally posted by Stibnut  
Thanks for the advice and info, everyone! I have filled up a 100 mL reagent bottle up 1/3 of the way with NaOH and filled the rest with ether. It sounds like the Grignard shouldn't be as difficult as I thought it would be, provided I oven-dry everything. I'll wait a few days for the ether to dry over the NaOH and then give it a try. Once the sieves come in I'll put them in the main bottle.

One quick question: people are suggesting a little trace of I2 to start the reaction. I'm using methyl iodide, which even though I'm storing it with copper in a freezer, is still slightly yellow from traces of iodine. Does adding a little crystal help anyway?

In the future, I'll probably distill ether from starting fluid and getting it that way. Chemsavers' $52 for 500 mL is truly a rip-off, and from the thread I found on getting ether from starting fluid, it seems that it is easy to distill it from heptane in a water bath. I assume I could just add a little trace of BHT to keep peroxides from forming and then dry with a mix of both NaOH (mostly to get dissolved CO2) and sieves.


If you reuse ether, the NaOH will also neutralize any acids formed as a result of the grignard reacting with CO2. IIRC, the best agent to use is sodium metal, since it reacts with oxygen as well. This prevents peroxides from forming, (I think, since the sodium would react with the oxygen first) and may neutralize peroxides too, although the sodium would have to be broken up periodically so that fresh metal is exposed. Otherwise it gets covered with NaOH and the metal can't react with oxygen.

Stibnut - 2-9-2016 at 21:46

I see - that does explain why people use Na rather than NaOH, even though NaOH is much cheaper and easier to handle.

I've read that molecular sieves aren't a good idea for acetone because they're prone to causing aldol reactions. Would MgSO4 be a good choice? If not, what would be the best drying agent(s) to use for the acetone?

Melgar - 4-9-2016 at 17:42

I've used CaCl2 and MgSO4 before without any trouble. Of course, that was with acetone that was with acetone that was supposed to be dry already, but the CaCl2 balls stuck together, meaning they were able to extract at least some water. (CaCl2 is deliquescent, and its surface will get damp when it's absorbed some water.) I think the trouble with molecular sieves is that they're TOO good. So they actually change an equilibrium that would normally be highly biased towards regular acetone, to one in which the aldol condensation product is a significant component of the solution.

stoichiometric_steve - 5-9-2016 at 03:45

Quote: Originally posted by Melgar  
that was with acetone that was with acetone that was supposed to be dry already, but the CaCl2 balls stuck together, meaning they were able to extract at least some water


what you describe was the reaction of CaCl2 with acetone, the two of which are obviously incompatible for drying purposes.

Alice - 6-9-2016 at 01:09

Is calcium oxide suitable for drying ether? I didn't find any reference for it, but I can't find a reason why it would not work either.