Wednesday, July 22, 2009

Lithium Aluminium Hydride Reductions - Rochelle's Salt

Haha I'm still (barely) alive. Thanks for sticking around. Been busier than usual sorting my private and professional life out. Wrote a ton of grant proposals, published some papers (here, here, here and here), writing a book chapter, trying to be productive in the lab (fat chance) as well as having a life after work and some time off. So Curly Arrow got down prioritised for a while. Hopefully that is changing now.
Last week I did a lithium aluminium hydride reduction on large scale (see picture). This reminded me of the first time I had to work a reaction of this type up. My first experience (sometime last century) was a DIBAL reduction and if you haven't tried this stuff yet I can tell you that all these aluminium hydride reagents end up forming massive aluminium emulsions that are impossible to work with. The first time round I ended up making an utter mess and getting a very low yield. Realising that I couldn't possibly be the first chemist to encounter this problem I looked into things. The trick is obviously to break the emulsion up. There is a number of ways to do this. My favourite method is to use a saturated aqueous solution of Rochelle's salt (sodium potassium tartrate). Rochelle's salt is an excellent ligand for aluminium and breaks the aluminium emulsion. The procedure is simple. Cool your finished reduction down to 0 degrees C, or lower depending on the situation (For my large scale reduction I cooled it with acetone/dry ice) and quench excess reducing agent with something non-protic. For example ethyl acetate or acetone works well. Just remember to use something you can easily evaporate off when things are done. Don't be impatient and add it dropwise with vigorous stirring. Use a addition funnel for larger scale reactions. When the quench is complete remove the cooling bath.
I find it convenient to have a saturated aqueous solution of Rochelle's salt standing around. Please note that Rochelle's salt has a ridiculously high solubility in water so when preparing the aqueous solution go easy on the water and pick a small flask. When my reaction is quenched and everything looks like jelly I add some Rochelle's salt solution. Often I'll add it as a half saturated (or even more dilute) solution (a larger aqueous layer sometimes makes separation of the phases at the end easier). After pouring Rochelle's into your flask get the mixture stirring vigorously, have a cup of coffee and check your email. The better stirring and the more Rochelle's you use the faster it'll break up the emulsion. Ultimately you end up with two nice and clear phases that are simple to separate in a separatory funnel. D!


Ψ*Ψ said...

i hate hate hate working up DIBALH reactions, but lithal gave me no problem. of course, i've only ever done one of each ;)

Tot. Syn. said...

Sometimes it only take a coffee to allow the Rochelle's salt to do it's thing - other times it can take overnight. I stick with the semi-aqueous workup most times - and I've done some pretty huge LAH reduction. The real bastard was a reduction of butyne-1,4-diol to the alkene. I'd normally go through an entire 1L bottle of LAH solution.

Anonymous said...

I just wanted to let you know that your posts are greatly appreciated. They may be necessarily infrequent, but I have learned several quite useful things from Curly Arrow and look forward eagerly to each one.

Your dry column vacuum chromatography technique alone has saved me many hours of tedium!

gyges said...

Couple of comments about your set-up, as shown in the pic.

You need a lab jack and a cork ring.

Dunno if the comment is helpful or irritating: trying to be the former; apologies if I'm being the latter.

Daniel Sejer said...

Are you German by any chance? Germans appear to be very keen on lab jacks and cork rings for reaction set-ups. Why a lab jack for a non-heated reaction? D!

gyges said...


German? - No.

Why lab jack? To be able to quickly apply either coolant and hence hold exotherm, or empty bowl to stop the the liquors spilling onto the f/c. Standard practice wherever I've worked.

Daniel Sejer said...

Gyges, I was also taught to do chemistry like that. I do tend to have a crystallisation dish under my flask to collect my compound in case of breakage. However, with the possible exception of removing a heating source I generally haven't found lab jacks particularly useful under my stirrer. I don't think there is anything fast about a lab jack. The models I have encountered tend to be rather slow and involve a fair bit of twisting. I find readjusting the clamp to be a much faster process. Anyway, enough of this. I am a firm believer in "whatever works for you" as long as it doesn't pose any danger to your surroundings. D!

gyg3s said...


"Anyway, enough of this. I am a firm believer in "whatever works for you" as long as it doesn't pose any danger to your surroundings. D!"


krest17 said...

I also did reductions with LAH on kilo-scale. I use prep from Org.Syn. (don’t have reference on it – but can find if you wish). It was for x g of LAH x g of water, then 3x g of 15% KOH, then x g of water – something like this (I am not sure 100% - please check). It works always amazing – you get precipitate, really nice powder, easy to filtrate, wash with ether, and evaporate – done. No any phase separation, funnels and etc.

krest17 said...

I also did reductions with LAH on kilo-scale. I use prep from Org.Syn. (don’t have reference on it – but can find if you wish). It was for x g of LAH x g of water, then 3x g of 15% KOH, then x g of water – something like this (I am not sure 100% - please check). It works always amazing – you get precipitate, really nice powder, easy to filtrate, wash with ether, and evaporate – done. No any phase separation, funnels and etc.

Daniel Sejer said...

krest17, yes this is one of the alternative methods that I referred to that works really well too. I also can't remember the exact conditions but I know that this work-up procedure can be found in the following paper: Skouroumounis, G.; Winter, B. Helv. Chim. Acta 1996,
79, 1095.
I'll look it up when I'm back at work on Monday, if someone doesn't beat me to it. Have a nice weekend chemists. D!

Heiko said...

hm... that's new to me with Rochelle's salt. i did some low valent titanium reduction (TiCl4 with Zn dust) which give during work up also a mess of black/grey stuff. i broke this suspension with a mixture of citric acid/ammonium acetate which worked very well too.
currently i have some problems with braking up a hydrolised ZnI2 suspension... tried it with citric acid too but it didn't work... maybe you have some ideas?

Daniel Sejer said...

This is the LiAlH4 procedure (slightly edited) from the Helvetica paper I mentioned earlier:
"....this ester (15.4 g, 75 mmol) in Et2O (100 ml) was added dropwise to a stirred suspension of LiAIH4 (2.5 g, 65 mmol) in Et20 (150 ml), while allowing the temp. to rise to 35 oC (reflux). After 10 min, the mixture was cooled to 4-5 oC, acetone (3 ml) was added and then 1 N aq. NaOH (12.5 ml). The mixture was allowed to
reach r.t., Na2S04 was added and the mixture filtered and evaporated...." D!

Anonymous said...

The water/NaOH/water workup recipe is also found in Fieser and Fieser #1 under LAH.

LittleMissProcess said...

Another neat trick to quench your LiAlH4 reactions is to cool it down and add some sodium sulphate decahydrate. Warm to rt, filter, concentrate. Bingo!


I will try my reaction with LAH next time.

Anonymous said...

I've done a few LAH reductions and always use the Fieiser workup.
And for DiBAl, quench with EtOAc, then add saturated tartaric acid solution, works better than Rochelle's.
Can also use citric acid to solubilise aluminium salts.

Anonymous said...

@heiko, try aqueous disodium EDTA

Anonymous said...

In several posts, there is a LAH workup procedure using sequential addition of water, KOH (or NaOH) solution and water. Daniel Sejer cited a HCA paper as a source. However, during my studies in the 60', I was tought that this procedure was introduced in 1955 by V. M. Micovic and M. L. Mihailovic (“Lithium Aluminum Hydride in Organic Synthesis”)

The Curious Chemistry Grad said...

I did DIBALH reaction on few analogues and I had problems with it too:

On the other hand, I reduced ester and amide on the same compound with classic rx with DIBALH, but only all amide was reduced to amine and very little ester reduced to alcohol.

Why is that?

At the end of the day, I collected the esters and did transesterification with methoxide to get desired alcohol as the side product.

peach said...

Never owned a lab jack, but could see how it'd help. I tend to do a lot of heating using tin foil tipi's as opposed to oil baths. That way, if things get over excited, I can flick the plate off and whip the foil away and it'll cool in a controllable manner (particularly if I'm not roasting the plate to full in the first place). Also, no spills, no fires, no slippery glass, doesn't smell like an old chippy that's failed all of it's last health inspections etc.

I use a ball and socket if I need to support heavy flasks or stretches of glass, so the run can be 'bent' to be clamped (without fiddling for ages for a semi-decent results) or rest in the right place (minus a jack). Even with the clamp tight, there is still enough flexibility to save the glass in some accidental knocks, as it can move a little.

I would certainly have a cork ring under that though. But then, I bet that rail behind the glass is bolted to the hood and not going anywhere fast.

My most annoying glassware problems as of late was watching all the keck clips disintegrate and then ping off my glass when porting reactive gases around (despite them being QuickFit and lined with Krytox).


Anonymous said...

just q quick clarification for myseld. reducing ester with LIAlH4 to alcohol. You guys did not mention adding acid as the last part of the procedure! why is that??

Anonymous said...

I did LAH reduction many times, I quenched it different procedures sometimes EtOAc slow addition at 0oC, sometimes sat aq NH4Cl solution at 0oC, sometimes sodium sulfate paste, sometimes water-15% KOH-water –
In most of all cases I did extraction, but not in last case just filtration and concentration
why it so, what is the Mechanism involved in this at basic conditions where all other cases using acidic work up to quench basic LAH.
And one more important thing while I used this method water-15% KOH-water – I got fire
is it may be using of water?
generally water is not suitable to quench these LAH, nBuLi, DIBAL etc

Nalajala said...

ok. you have eplained nicely about ur reaction, but always Rochel's is not suitable based on sustrate we have to use quenching reagents like sodium sulphate, ammonium chloride, NaOH etc.,

Anonymous said...

Quenching with EtOAc can be risky. If your alcohol is unhindered enough, you can transesterify the EtOAc to get some acyl protected alcohol. I quenched with EtOAc for the reduction of undecanoic acid and ended up with 44% acyl protected alcohol.

Anonymous said...

Hello everyone. I am currently working on a reduction using LiAlH4 and AlCl3. Does anyone have experience which workup works best for this combo? So far I've tried: 1. quenching with HCl, then extraction.
2. Fieser workup.
I feel like both aren't perfect. Thanks a lot in advance!

Daniel Sejer said...

@anon: try Rochelle's salt like described in the post. It works wonders but be patient and leave it to stir until you see clear phase separation. D!

Stephanie Valenzuela said...


I will be doing my first LiAlH4 reduction pretty soon, fingers crossed this week. I have been trying to do some reading to prep and have in my head the procedures so I don't fumble in the midst of it all. Do you happen to know of any good videos or readings I can look up? I will be using a manifold. Any help would be Appreciated!! THANK YOU! :)

Jamil Gregory said...

from the Merlic Group at UCLA, The Organic Companion

To work up a reaction containing x g lithium aluminum hydride
Dilute with ether and cool to 0°C
Slowly add x mL water
Add x mL 15 % aqueous sodium hydroxide
Add 3x mL water
Warm to RT and stir 15 min
Add some anhydrous magnesium sulfate
Stir 15 min and filter to remove salts

works perfectly for me.

Sameer Vyasamudri said...

Hi All,

Has any one done LAH to reduce double bond and Nitro to amine in one go..
I found several procedures and followed the same, but not successful with yield.
Your suggestions are highly appreciated..


Anonymous said...

Hello all,

I used LiAlH4 to reduce an epoxide. Has any one done this ? and did it work? If it worked, can you please give me some details , like the amount that you used, the time it had taken to form the products... etc


Mister B said...

Thank you for this post, Curly, I found some nice advice reading it !

I am struggling with one specific LAH reduction.
N-Boc tyramine is reduced by LAH in 1,4-dioxane. Work in a real pain.

N-methyl tyramine combine high solubility in water, basic center (NHMe) and a phenol... I've tried almost every work-up I could find and none give me the desired product, pure enough. (I can't run a chromatography on it, it's way to polar... )

Any advice from you or any reader please ?