Sciencemadness Discussion Board
Not logged in [Login ]
Go To Bottom

Printable Version  
Author: Subject: Alkoxide formation - Sodium hydride vs Sodium metal
laserlisa
Harmless
*




Posts: 48
Registered: 5-2-2016
Member Is Offline

Mood: No Mood

[*] posted on 25-2-2016 at 11:47
Alkoxide formation - Sodium hydride vs Sodium metal


Hi members of SM!

Im planning on performing a nucleophilic aromatic substitution on a fluorobenzene using sodium methoxide.

As I have been reading up on how to run this reaction, I have noticed that most of the references uses NaH and alcohol to form the alkoxide, instead of sodium metal dissolution in alcohol.

From what Ive gathered NaH seems quite dangerous to handle, and its commercially available in an oil dispersion, meaning one would have to wash off the oil prior to use, and Ive read that these waste oil washings will contain NaH traces so care must be taken handeling it. Perhaps Im wrong?

Sodium metal on the other hand is quite easy to work with in my experience, and does not seem nearly as dangerous.

I suppose one example of where NaH would be preferred is if you want to run the reaction at a higher temperature than the boiling point of the alcohol, for instance I want to run my reaction at around 100C in NMP with sodium methoxide in methanol. In order to achieve this temperature I figured that the excess methanol could slowly be distilled off as the reaction proceeds.

With NaH I guess the distillation would not be needed since you would add an equimolar amount of alchol, so no excess alcohol would be present in the reaction.
However the distillation does not seem cumbersome at all, atleast not compared to the hassle of handeling NaH.

My questions are:

1. What are the advantages of using NaH over sodium metal when forming the alkoxide?

2. Would removal of excess methanol by distillation during the reaction work as I have forseen? Is it a method that is actually employed? I have not yet found any article doing it.


Sorry for the poor grammar, english is not my native language.
Thanks!
View user's profile View All Posts By User
chemrox
International Hazard
*****




Posts: 2928
Registered: 18-1-2007
Location: UTM
Member Is Offline

Mood: psychedelic

[*] posted on 25-2-2016 at 12:26


Your English is better than a lot of "English speakers" here.
Could you post a ref? (We usually do that) When I make NaMeO I use Na metal and alcohol. Metal hydrides are very reactive. More so than alkali metals. Be careful! In general it's best to make it from methanol and Na rather than buy the commercial stuff which can be contaminated with NaOH. NaH is a lot more costly than Na which can be obtained on ebay.




"Ignorance is the Mother of Devotion." — Robert Burton.
View user's profile View All Posts By User
UC235
International Hazard
*****




Posts: 532
Registered: 28-12-2014
Member Is Offline

Mood: No Mood

[*] posted on 25-2-2016 at 12:26


For cheap simple alkoxides, sodium metal is an excellent choice. You can bring the alcohol to reflux if it's hindered and is reacting sluggishly, but alcohols like methanol are quite vigorous.

NaH is a much more selective reagent, however, and is greatly preferred in stepwise organic synthesis where the substrate is expensive, functionalized, or thermally sensitive. Sodium metal does a whole slew of other reactions and will react with other parts of the molecule, acting as a single-electron reducing agent. See Birch reduction, Bouvealt-Blanc reduction, alkyne reduction with sodium, wurtz reaction, acyloin condensation, and probably quite a few more that I missed.

With only a few exceptions, NaH does one thing in organic chemistry. It removes acidic hydrogens as hydrogen gas. You don't need to heat much, and it's finely divided and reacts much faster than sodium would where reaction is limited to the surface of the chunk of metal.


[Edited on 25-2-2016 by UC235]
View user's profile View All Posts By User
ziqquratu
National Hazard
****




Posts: 385
Registered: 15-11-2002
Member Is Offline

Mood: No Mood

[*] posted on 25-2-2016 at 15:13


I would say that NaH is preferred when you want to minimise NaOH contamination (assuming you're using decent quality NaH, there will likely be less NaOH present than when you're cutting and weighing sodium metal). In addition, NaH is easier to handle (it's a clumpy but not particularly sticky powder which tends to be fairly benign as far as reactivity during handling, as opposed to having to cut lumps of metal which react rapidly with atmospheric moisture) and easier to weigh out accurately (again, powder versus slices). It also tends to be readily available on the shelf in most professional labs, and more widely used in the literature, so going to the metal tends to be given little thought.

On the other hand, sodium metal reacts more slowly (due to smaller surface area) than does NaH, and is quite a bit cheaper, so if you're making a large batch would often be preferred on those grounds.

In either case you have oil to deal with; obviously easier to remove this from the metal, but in many cases there's little advantage to removing the oil from the NaH anyway (since it's highly non-polar, it's usually easy to remove when you're cleaning up your product - very easy if you're doing chromatography, or you can often wash it away from a solid or by recrystallisation). In either case, the washings are rarely a significant hazard - they're unlikely to carry over too much of the reactive species in either case - but naked NaH can be very reactive if you're not careful in handling it.

Ultimately, for small-scale work where I want reasonable accuracy (without going to the extreme of titrating my base) or where my alcohol was valuable (i.e. the product of some earlier synthesis, as opposed to an off-the-shelf solvent), I would always prefer NaH; for large scale work, sodium metal would be a better option.

As for your question regarding removal of methanol - I see no reason why it shouldn't work, assuming there's no azeotrope of methanol with your other solvent. Even using NaH to prepare your base, I'd use an excess of methanol, so you would be left removing it anyway.

There is one other alternative, though - you could perhaps use (or prepare) solid sodium methoxide, instead of preparing a solution of the base.
View user's profile View All Posts By User
alexchim
Harmless
*




Posts: 8
Registered: 23-12-2015
Location: France
Member Is Offline

Mood: No Mood

[*] posted on 26-2-2016 at 10:29


It is not bad to use sodium hydride, even if it is coated with paraffin oil, it does not interfere with the reaction.
That's the same for metallic sodium, it contains a little of oil. It could be washed with cyclohexane prior to use.

[Edited on 26-2-2016 by alexchim]
View user's profile View All Posts By User
DrMethyl
Harmless
*




Posts: 34
Registered: 23-11-2015
Member Is Offline

Mood: No Mood

[*] posted on 26-2-2016 at 14:42


When you use NaH you don't need to wash it up with solvent before use unless the mineral oil fuck up your work up. If you plan to do an acid-base extraction you don't need the washing step. I prefer using NaH this way because it's much less hazardous.

I suggest using Na metal. Be careful. add it only a little piece at the time because it is more reactive in MeOH than EtOH. In MeOH the reactivity is just between water and EtOH.
View user's profile View All Posts By User
zed
International Hazard
*****




Posts: 2230
Registered: 6-9-2008
Location: Great State of Jefferson, City of Portland
Member Is Offline

Mood: Semi-repentant Sith Lord

[*] posted on 26-2-2016 at 17:50


The kids in my neighborhood, don't make alkoxides that way.
View user's profile View All Posts By User
Dr.Bob
International Hazard
*****




Posts: 2334
Registered: 26-1-2011
Location: USA - NC
Member Is Offline

Mood: No Mood

[*] posted on 26-2-2016 at 17:51


If the NaH was being used to deprotonate something, then it would likely be easier and cleaner than NaOMe. But for performing a nucleophilic aromatic substitution on a fluorobenzene using sodium methoxide, I don;t see that NaH is needed, but your idea of making the methoxide first and then putting it in NMP and heating to 100C might allow removal of the excess methanol, then add the fluoroaromatic and heat.

If the only reactive group is the fluoro, then using an excess amount of methoxide will drive the reaction much better. Often displacing a fluoro with an alkoxide is much harder than with an amine. Also, DMSO seems to work better than many other solvents for these, NMP might work quite well, but I have not used it for that, and NMP is a little harder to remove than DMSO. Both require a lot of water washing to remove the solvent from the product in ether, EtOAc, or hexanes.
View user's profile View All Posts By User
laserlisa
Harmless
*




Posts: 48
Registered: 5-2-2016
Member Is Offline

Mood: No Mood

[*] posted on 29-2-2016 at 11:42


Thank you all for your great answers. Im very grateful that you took the time to give me such detailed answers. Ive read them with great interest.

I will go with the methoxide and distill off excess methanol. Would be great if I could run the reaction in methanol so I will try that first, if not I will resort to NMP or DMSO.

Big thanks.
View user's profile View All Posts By User
zed
International Hazard
*****




Posts: 2230
Registered: 6-9-2008
Location: Great State of Jefferson, City of Portland
Member Is Offline

Mood: Semi-repentant Sith Lord

[*] posted on 3-3-2016 at 16:12


Again, some lab manuals may make Na-Alkoxides with Sodium Metal, but many folks here would use safer, more practical, and less expensive methods.

Nicodem has posted expertly on the topic. Knows much more about it than I do. I'll try to find some of his posts.

Or, perhaps Dr. Bob. Yup! Dr.Bob..... NaOMe from NaOH and Methanol.


"It is very difficult to "purify" strong base solutions, whether alkoxides or lithium bases. That is why it is best to make them pure from the start. It will be difficult to purify any alkoxide solution containing 5% water to get pure sodium ethoxide. For methanol, it is easier to get close, since the initial methanol can be obtained nearly anhydrous, so there is only 1 molar equivalent of water in the solution, whereas ethanol containing 5% water has a large amount more water than that. You can do the math is you want to check this, I have done it before and don't care to repeat it. That (the 5% water in most ethanol) is one of several reasons that making ethanol based biodiesel via transesterification is very difficult compared to methanol based biodiesel.

The same way, it is hard to purify say, butyl lithium, as it is too reactive to really isolate-if you evaporate it, you get a complex aggregate of material which may never redissolve well again, and the material may react with the remaining solvent (especially THF) to form new compounds."

So? Would the methods used to make NaOMe, for biodiesel synthesis, be suitable for you. The guys have answered your query as you posted it, Na vs NaH..... but the question is: Is there a better way to make methoxide, suitable for your purposes?

[Edited on 4-3-2016 by zed]

[Edited on 4-3-2016 by zed]
View user's profile View All Posts By User
zed
International Hazard
*****




Posts: 2230
Registered: 6-9-2008
Location: Great State of Jefferson, City of Portland
Member Is Offline

Mood: Semi-repentant Sith Lord

[*] posted on 6-3-2016 at 16:05


Well, I've lost the ability to edit my last post.

There is another question to be addressed. Can this substitution actually be made?

NaOMe can be reacted with some aryl-halides, containing an -OH group, with the assist of a copper catalyst. Iodides and Bromides are fairly reactive. When a Chloride or a Fluoride is involved....Maybe not! Such materials can be quite recalcitrant about reacting. Which is why, Bromo-Benzene is usually favored, over Chloro-Benzene....In the preparation of Grignard Reagents. Hard to get Chloro-Benzene started. Haven't really considered the prep of Phenyl-Magnesium Fluoride.

I'll cruise the net.

https://books.google.com/books?id=kEIrAAAAYAAJ&pg=PA212&...

Maybe. What else you got?

[PDF]Nucleophilic Aromatic Substitution - ChemConnections

Won't let me make a direct link, but the above will take you there.





[Edited on 7-3-2016 by zed]
View user's profile View All Posts By User
laserlisa
Harmless
*




Posts: 48
Registered: 5-2-2016
Member Is Offline

Mood: No Mood

[*] posted on 6-3-2016 at 23:21


Zed I think you are talking about an Ullmann reaction, Im not sure though.

With nucleophilic aromatic substitution however fluoride is a very good leaving group as long as the benzene ring is deactivated, even more so than the heavier halides.
I should have mentioned in my original post that my substrate aromatic is highly deactivated.

Have a look at the attached reference Practical synthesis of aromatic ethers by SNAr of
fluorobenzenes with alkoxides


[Edited on 7-3-2016 by laserlisa]

Attachment: Practical synthesis of aromatic ethers by SNAr of.pdf (87kB)
This file has been downloaded 514 times

[Edited on 7-3-2016 by laserlisa]
View user's profile View All Posts By User
zed
International Hazard
*****




Posts: 2230
Registered: 6-9-2008
Location: Great State of Jefferson, City of Portland
Member Is Offline

Mood: Semi-repentant Sith Lord

[*] posted on 7-3-2016 at 14:00


Thank you for the link!

Since your quest seems feasible. Perhaps it would be nice if you could produce your Alkoxides by less demanding means. Certainly, Na Metal should work, and cleanly too, but other approaches do exist and there should be many references in the archives. Again, to my recollection, Nicodem has posted expertly on the subject.

Seems like this patent might be relavant. I'll check.

US Patent #2,278,550:

[Edited on 7-3-2016 by zed]
View user's profile View All Posts By User
zed
International Hazard
*****




Posts: 2230
Registered: 6-9-2008
Location: Great State of Jefferson, City of Portland
Member Is Offline

Mood: Semi-repentant Sith Lord

[*] posted on 8-3-2016 at 16:44


Well, once again, I've lost edit function.

I'll try to post a direct link.

http://pdfpiw.uspto.gov/.piw?Docid=2278550&idkey=NONE&am...
View user's profile View All Posts By User
clearly_not_atara
International Hazard
*****




Posts: 2153
Registered: 3-11-2013
Member Is Offline

Mood: Big

[*] posted on 9-3-2016 at 14:11


I think in a few cases the alkoxide can be synthed by dehydrating an alcoholic solution of hydroxides with a powerful desiccant like K2CO3. In the case of sodium ethoxide the desiccant can be NaOH itself when an excess of dry sodium hydroxide is used.

Alkoxide solutions however are not as stable as their facile preparation might suggest: atmospheric oxygen can (slowly) oxidize them to form aldehydes (which then polymerize), and atmospheric CO2 adds to produce the monoalkyl carbonate. A solution of NaOEt in EtOH turns brown if left uncovered.
View user's profile View All Posts By User
zed
International Hazard
*****




Posts: 2230
Registered: 6-9-2008
Location: Great State of Jefferson, City of Portland
Member Is Offline

Mood: Semi-repentant Sith Lord

[*] posted on 10-3-2016 at 17:53


Well, Na would prefer to perform a menage' with CO2&H2O...versus some ol' alcohol. We shift the equilibrium, by ejecting CO2 and H2O from our reaction mix.....

And, of course, once we have achieved our goal, we have to protect our desired product, by slamming the door. We gotta seal it up, to protect it from standard atmosphere. Otherwise, the magic will reverse itself.

Which is why in days of yore, some posters complained, "The Reaction Didn't Work!"

"I threw the chemicals in a Mason Jar, and I boiled the shit out of it, and I didn't get NaOMe!". Fraud!

My friends in go Go-Cart racing niche, tell me that when you are running on Methanol, if you leave the cap off of your fuel tank over-night, a 3/4 full tank, will miraculously become A FULL tank, in the morning.

I don't know for certain, but ethanol may be worse. It is hard to keep water out of it.





[Edited on 12-3-2016 by zed]
View user's profile View All Posts By User

  Go To Top