SplendidAcylation
Hazard to Others
Posts: 203
Registered: 28-10-2018
Location: Starving in some deep mystery
Member Is Offline
Mood: No one I think is in my tree.
|
|
Some practical questions about Grignard reactions
Hi,
I am planning to make propiophenone by adding ethyl Grignard reagent to benzonitrile.
I have a few questions that I thought I'd better ask before I try it, as my previous attempt at a Grignard reaction was a failure (probably due to
inadequate cooling, the halide escaping out of the condenser).
I will be using homemade ethyl bromide, I also made the benzonitrile, both have been properly dried before being bottled with molecular sieves.
1) How long does the Grignard reagent need to be refluxed with the nitrile?
I haven't seen many practical examples of Grignard addition to a nitrile carbon; I think the reaction will be slower than the addition to a carbonyl
carbon, as the nitrile carbon will be less electrophilic, but I'm not sure how relevant this is.
These are the only two examples I have come across so far:
OrgSyn - Diphenyl ketimine
Rhodium - Synthesis of ketamine
Both of these seem to use rather a long period of reflux (6 hours, and 3 days(!) respectively), is that really necessary?
2) Is it OK to pour the reagents down the condenser?
I will be doing the reaction on a small scale (<5g of halide), and I only have a huge addition funnel and 500mL 3-neck RBFs, so these would be a
bit of overkill, and I would rather not oven-dry my beloved addition funnel for no reason just in case it might crack.
A preferable approach would be to use a smaller RBF, but I only have single-neck ones, so the ethereal solution of ethyl bromide, and subsequently the
ethereal solution of benzonitrile, would both have to be poured down the condenser.
I can see this being a problem, as the condenser will be filled with ice-water, and thus water might condense rather quickly on the inside of the
condenser, as soon as the drying tube is removed, this condensate would then be washed into the flask by the addition of the solutions.
Is this bad practice, or would it be okay, provided the condenser is exposed to the air for a very short time?
3) Can a different halide be added to the ether + magnesium once the Grignard reagent has begun to form?
My first Grignard attempt used ethyl chloride, but due to its low boiling point and the relatively low reactivity of alkyl chlorides in forming
Grignard reagents, the reaction was unsuccessful;
I subsequently made ethyl bromide, which is less volatile, and it forms Grignard reagents more readily, but I have more ethyl chloride so using it
would be preferable.
If I were to form a small quantity of the Grignard reagent using the bromide, could the chloride then be added to the solution to consume the rest of
the magnesium, in essence, using the bromide as a catalyst for the formation of the Grignard reagent from the ethyl chloride?
Thanks so much for your help!
|
|
|
Tsjerk
International Hazard
Posts: 3032
Registered: 20-4-2005
Location: Netherlands
Member Is Offline
Mood: Mood
|
|
The trick with mixed chloride/bromide works for chlorobenzene, which normally doesn't form Grignards, I posted a link around here some time ago, I
will check if whether I can find it.
A large round bottom might not be a problem for smaller reactions, especially when refluxing as the reflux keeps everything flushed down. Instead of
oven drying you could rinse the addition funnel with some acetone and let it air dry, that should make it dry enough.
Edit:
https://www.sciencemadness.org/whisper/viewthread.php?tid=87...
[Edited on 4-5-2022 by Tsjerk]
|
|
|
wxyz
Harmless
Posts: 20
Registered: 13-4-2010
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by SplendidAcylation  |
1) How long does the Grignard reagent need to be refluxed with the nitrile?
Both of these seem to use rather a long period of reflux (6 hours, and 3 days(!) respectively), is that really necessary?
[\rquote]
Yes. Grignard with nitriles can take days. Nitrile group is less reactive than carbonyl to Grignard reagents.
[rquote]
2) Is it OK to pour the reagents down the condenser?
[\rquote]
I don't know if it is bad practice, but I do it all the time, but only for solvents. But for this reaction you should have a moisture trap at the end
of the condenser. If you want to briefly remove it, I suppose you can. But everything, including the condenser, needs to be dry of course.
|
And by the way, I have had poor results with grignard reactions in which the reagents were dried only with 3A sieves (and freshly dried too). Got much
petter results when everything was dried with P4O10
|
|
|
SplendidAcylation
Hazard to Others
Posts: 203
Registered: 28-10-2018
Location: Starving in some deep mystery
Member Is Offline
Mood: No one I think is in my tree.
|
|
@tsjerk
Thanks! That's interesting indeed, I suppose if it works for an unreactive halide like chlorobenzene then it should work with anything (I hope!).
I decided to do it the simpler way and use the 250mL RBF and pour the reagents down the condenser, it seems to have worked so far 
@wxyz
Oops! So the reaction can take a long time... Let's hope I haven't messed up too badly, then.
I tried the reaction yesterday, I'll probably do a write-up on the topic later, but here's a summary:
The 250mL RBF, 3 pipettes (one for the solution of benzonitrile, one for the ether itself, and one for the solution of halide), the Liebig condenser,
stir bar, and a rubber bung with a tube through it, were dried in the oven at 140c for 45 mins.
The magnesium was also dried in the oven, after having been sanded to remove oxide and cleaned with isopropanol.
Once removed from the oven, the magnesium, along with a tiny crystal of iodine, was placed in the flask alongside the stir bar, the condenser was then
attached, and a drying tube containing CaCl2 was attached to the top via the rubber bung (I don't have a QuickFit drying tube).
The flask was then heated with a heat-gun to vapourize the iodine, forming purple vapour in the flask and coating the magnesium.
Without any coolant in the condenser, a small quantity of the ethyl bromide solution was poured down the condenser, and the bung was replaced.
Within seconds, bubbling was observed, evidently the Grignard reagent was forming!
Unfortunately I took a little too long to fill the condenser with cold water, and I think most of the ether escaped.
Once the condenser was connected to a pump recirculating ice water, the bromide solution was added ~1mL at a time, the flask had to be cooled in an
ice-bath for the first few additions, but as the solution became more dilute, this was not necessary, and external heat actually had to be applied at
one point to speed up the reaction.
After all of the bromide was added (~30 mins), most of the Mg had dissolved, and a brown solution had formed.
This was left to stir and reflux for 30 mins or so, whereupon I began adding the benzonitrile solution.
(In each case, dry ether was used to rinse out the bottle)
The addition of the nitrile was surprisingly uneventful, it caused the solution to boil, but no change in appearance was observed.
After all of the nitrile had been added (again, ~30 mins), the solution looked just the same, brown and clear, so I began to heat the flask with warm
water; Quite quickly, a precipitate formed.
The precipitate seems to be white, but there also seems to be an orange oily substance (maybe).
The mixture was allowed to reflux for ~1 hour, but no longer, as unfortunately I was running out of ice to cool the condenser water, and water
straight from the tap is much too warm to condense the ether effectively using a Liebig condenser.
Because of this faux pas, the reaction was suspended, the flask was disconnected from the condenser and stoppered with a rubber bung.
Before I stoppered it, I quickly removed a small quantity of the white precipitate; Once the ether evaporated, it seemed to be odourless, I then added
it to water, and a peculiar smelling compound was formed at once (the solution was cloudy).
I'm not sure if this is what propiophenone should smell like, it's quite unpleasant, kind of acrid...
Now I have a bit of a problem it seems, as I had initially assumed that what looks like a fairly large quantity of precipitate (which I assumed to be
the =NMgBr imine salt) was an indicator that the Grignard addition was complete, but as wxyz says, these reactions can take a long time.
Without a truck-load of ice, I'm going to struggle with keeping the condenser cool for more than a few hours, so my best option at this stage is to
use a different type of condenser.
Unfortunately, the only other condenser I have is a Graham condenser (the kind where the condensate goes through the coil), and these are not
advisable for use in reflux, as the condensate can form a droplet which gets pushed up through the coil.
Still, I have used it before for refluxing chloroform with no problem whatsoever, so it's maybe worth a try.
Thanks for your help everyone, I'll let you know how it goes!
P.S. Does anyone have any ideas what the precipitate might be if it isn't the imine salt?
Edit:
I have swapped the Liebig condenser for my Graham condenser, and the solution is refluxing as we speak.
If the reflux rate gets too vigorous, a column of ether gets pushed up the coils, which is why Graham condensers aren't ideal for refluxing, however,
at a lower reflux rate (2 or 3 drops per second), this problem doesn't seem to arise.
The good news is that even fridge-cooled water is sufficient to condense the ether in just the first two or three coils!
[Edited on 6-5-2022 by SplendidAcylation]
|
|
|
zed
International Hazard
Posts: 2283
Registered: 6-9-2008
Location: Great State of Jefferson, City of Portland
Member Is Offline
Mood: Semi-repentant Sith Lord
|
|
To my recollection, this type synthesis calls for an addition of the Grignard Reagent to the Nitrile.
If you add the Nitrile to the Grignard Reagent, it is possible that 2 moles of Grignard Reagent will react with each mole of Nitrile, producing the
wrong product.
|
|
|
SplendidAcylation
Hazard to Others
Posts: 203
Registered: 28-10-2018
Location: Starving in some deep mystery
Member Is Offline
Mood: No one I think is in my tree.
|
|
| Quote: Originally posted by zed | To my recollection, this type synthesis calls for an addition of the Grignard Reagent to the Nitrile.
If you add the Nitrile to the Grignard Reagent, it is possible that 2 moles of Grignard Reagent will react with each mole of Nitrile, producing the
wrong product. |
Thanks for the reply.
Hmmm, so the Grignard reagent could add to the imine carbon?
I have seen a few examples of Grignard addition to nitriles, and the nitrile always seems to be added to the reagent:
IDENTIFICATION OF NITRILES. PREPARATION OF ALKYL PHENYL KETONES
"The phenylmagnesium bromide is prepared in the usual way from 25 g. of mag-
nesium turnings and 160 g. of bromobenzene in 360 cc. of dry ether. A solution of 0.25 mole of the nitrile in 100 cc. of dry ether is
run in slowly with stirring during a period of fifteen minutes. The solution is stirred for an hour longer and allowed to stand
overnight, The mixture is poured onto 500 g. of ice and 300 cc. of concd. hydrochloric acid, The water layer, which contains the
hydrochloride of the ketimide, is separated from the ether layer and refluxed vigorously for one hour. The solution is cooled and
extracted with four 200-cc. portions of ether. The ether extract is dried over anhydrous calcium chloride and the ether distilled from a
water-bath. The residue is transferred to a small modified Claisen flask and vacuum distilled. "
And OrgSyn - Diphenyl ketimine
"The apparatus consists of a 1-l. three-necked flask equipped with a mechanical stirrer, a 250-ml. dropping funnel, and a Friedrichs reflux condenser
fitted with a calcium chloride drying tube. Magnesium turnings (13.4 g., 0.55 g. atom) and 200 ml. of anhydrous diethyl ether are put in the flask
(Note 1). Slow stirring is started, and 4 ml. of bromobenzene (Note 2) is added from the funnel. After reaction has started (Note 3), the stirring
rate is increased, and moderate reflux is maintained by addition of 80.5 g. of bromobenzene (making a total of 86 g. or 0.55 mole) in 100 ml. of
ether. The solution is refluxed for 30–45 minutes after the addition and is cooled to room temperature. Stirring is continued while a solution of
51.5 g. (0.50 mole) of benzonitrile (Note 2) in 100 ml. of ether is added slowly enough (Note 4) to maintain only a gentle reflux."
But on the other hand, Grignard reagents do add to imines:
N-METHYL-1,2-DIPHENYLETHYLAMINE AND HYDROCHLORIDE
Perhaps a Grignard reagent can add to an imine, but not to an imine salt such as is formed upon the addition of the reagent to a nitrile?
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
