Grignard reactions are famous for being highly water-sensitive.

We wanted to find out exactly how sensitive they are and if it was viable to perform them with this sort of dry conditions available to the amateur chemist.

So we attempted to make a Grignard reagent using methyl iodide but using some extreme conditions.

As a control we dried the 100 ml flask in an oven at 150 degrees C for an hour,

then removed it and stoppered it while it dried.

While it cooled, our magnesium turnings are from a metal shop.

They've got black specks on them and visible white oxide.

We have, however, dried them in the oven.

We weighed out 1 gram of the magnesium turnings and then placed these in the flask.

We didn't attempt to dry them in any way.

We then covered them with 10 ml of diethyl ether, which was straight out of a new reagent bottle.

Again, we didn't dry this in any way.

We then measured out another 15 ml of ether into a small beaker which had been oven dried and the same was as the flask.

We added 1 gram of methyl iodide to this.

The methyl iodide was homemade and stored in a bottle containing some anhydrous calcium chloride.

Notice how the calcium chloride granules float due to the high density of the methyl iodide.

Then we got a tiny crystal of dry iodine.

There's a crystal.

There's a peanut here so you can see the scale of this.

We put the iodine into the flask with the ether and magnesium turnings.

Then we added a small amount of the methyl iodide in ether solution.

We've got a cold water bath set up just in case we need it.

At this point we started the stopwatch.

We waited a couple of minutes but there was no sign of a reaction.

So we added all of the methyl iodide.

We added the methyl iodide solution to the flask and continued to wait.

We're fast forwarding and this is 12 minutes into the reaction.

There's a slight turbidity and some bubbles.

Now we're 15 minutes in.

There's some white precipitate in.

A definite reaction occurring.

Just for your information we didn't use a reflux condenser on the flask but we used a stopper very lightly inserted in order to keep air and moisture out as much as possible.

Now we're 20 minutes in.

At 23 minutes the yellow color of the iodine disappeared and the reaction accelerated.

Here we are 26 minutes in and the flask is starting to get slightly warm.

If we were going to use our Grignard reagent for another reaction we'd probably want to put a condenser on at this point and stop the ether from evaporating off.

The total reaction time took about 45 minutes before the bubbling died right down and we were left with a dirty grey product looking like dish water.

So this works but it takes rather a long time.

Next we did a control experiment where we thoroughly dried the ether using a few small slivers of sodium metal.

You can see from the hydrogen bubbles that the ether does contain some water.

We then did everything we could.

We did everything in exactly the same way as before.

1 gram magnesium, 10 ml dry ether, crystal of iodine.

Then added a small amount of the methyl iodide solution in sodium dried ether.

The reaction followed almost exactly the same course as before.

We added the rest of the solution after a few minutes and it was only around the 10 minute mark that we saw the reaction really starting.

Again.

15 minutes was needed for complete reaction.

Our conclusion is that the reaction needs to be dry but past a certain point the quality of the magnesium you use has a greater impact on the reaction speed.

But you can get away with quite a lot.

We then did a third experiment.

We wanted to see if it was possible to use a different solvent than ether.

So we took some dichloromethane and dried it in a

small amount of phosphorus pentoxide.

Do note that you cannot dry chlorinated solvents using sodium metal.

This is extremely dangerous and can cause an unpredictable explosion.

After leaving for a few hours we filtered this into a small oven dried conical flask.

The liquid was very slightly milky due to some very fine phosphorus pentoxide particles.

We added a gram of magnesium metal exactly as before.

And then a tiny crystal of iodine which turned the dichloromethane solution an orange red color as opposed to the purple color in ether.

We added a small amount of a methyl iodide solution in dry dichloromethane using the same proportions as before.

Once again after a few minutes we saw no reaction.

And then waited.

And waited.

If anything the solution turned slightly orange over time.

But two hours later there was no sign of a reaction.

So it looks like Grignards don't easily work in DCM in the same way.

This was an interesting experiment.

It proves that if you have access to reagent quality ether and calcium chlorides.

A drying agent.

There's no reason you can't perform a Grignard reaction.

Having access to high quality magnesium is great.

But you can make do with low quality turnings provided you have some patience.

Just take precautions with ether and use ventilation.