[Mobius1988]

Im having a little trouble with conducting practical chemistry on an independent basis and I was hoping some experienced chemists could give me some advice.

In uni we have always been told what reactions to do etc but were rarely told how you go about choosing the conditions. It's easy enough to look up a paper and follow their prep but I want to be a successful career chemist and knowing how to set up a reaction ive never done before is a little difficult at the moment.

For example, when conducting a reaction with a reagent ive never used before, how do I decide what solvent is most appropriate, how many equivalents I will need and what temperature to do the reaction at?

If i need to use a carbonate, how do I decide between sodium, potassium and cesium carbonate? What effect does the counter ion actually have when only the carbonate is going to be acting as a base?

Does anyone know a good book/website that focuses specifically on pka's of a range of groups. I know all the obvious ones but some of the aromatic systems im coming across I have no idea about.

Please help :S

[Arkcon]

That is a pretty daunting missive, I had to read it a few times before I could see where you're coming from.  Shakespeare once had an angry Queen of Denmark say, "More matter, less art!" and you should try to keep that in mind. However, taking it point by point.

Paragraph 1:  Cool

Paragraph 2:  Yes, it is easy.  Its also appropriate.  Its very valid to build on what's been done before.  Its not just common to build on others work to advance, its also the way science and mathematics is taught.  Surely you've noticed.  If you want to be a wacky inventor who creates the totally unheard of invention -- but that's a fictional trope.  Real advancement builds on what we already know.

Paragraph 3:  There are some general concepts to build on here, but again, much of it has been done before, and you can build off the known concepts.  Some more specific questions may help us help you.

Paragraph 4:  OK.  Let me be extra prosaic to shake you up a little.  Sodium.  Why?  'Cause its cheaper.  We always use the sodium salt, unless the application demands something else.  You start with that one first.  Unless your application is biological.  See?  You're over-complicating the problem, to your own detriment.  Nobody else wants to waste money to have random uniqueness.

Paragraph 5:  Yes, we have those.  That should be an easy Google.

Try to form a more specific question, within the framework of what I'd said so far.

[fledarmus]

Most of making a reaction work in a research setting has to do with recognizing the similarities between the reaction you are trying to run and what has been done previously, using that to guide your choice of initial conditions, then paying very careful attention to the results that you get and using that and your analysis of the differences between your reaction and the literature procedure to drive your choice of conditions for your next attempt. For example, you have a reaction that begins with a base pulling a proton from your starting material, but the acidity of that proton in your molecule is lower than the nearest equivalent you could find in the literature; you might need to use a stronger base. Or you may have a second proton in your molecule which is even more acidic - you may need to use an extra equivalent of base and pull both protons before adding your second reagent.

As for your paragraph 4, there are other considerations than what part of the molecule is acting as the base. What happens to the ion that is formed when you have pulled the proton? Usually it is associated with the counterion - your sodium, potassium, or cesium. Cesium is much larger and "softer" than either sodium or potassium, which is frequently helpful in substitution reactions. Lithium, being less electronegative, has slightly more covalent character when it is a counterion, and although any alkali hydroxide can be used to saponify esters, LiOH gives me the best results in the lab for that reaction. And if you really don't want a metal around as a counterion, there are tricks to sequester the metal so it doesn't participate at all - for example, adding a crown ether. Solvent selection and temperature selection also may play a role in optimizing a reaction - if your reagent isn't as active as the literature example you think is closest, you may need to heat yours up more, or for a longer period of time. Getting good at following the progress of your reactions, whether by TLC, HPLC, LC-MS, or even color change is important in making your reactions work.

And sometimes you just don't know for sure what qualities you will need to improve your reaction, and you may have a choice of spending a lot of time doing library work, modeling, and calculations, or you may just go into the lab and set up a dozen small-scale reactions with a dozen different bases to see which one works best. I had one supervisor that used to say "15 minutes in the library would save you a week in the lab" when he thought I was spending too much time running reactions that weren't working, and "15 minutes in the lab would save you a week in the library" when he thought I was over-thinking a reaction.

[Mobius1988]

Thanks for the advice so far, sounds like sticking close to the literature to begin with is the best thing until I have a wider knowledge to apply to my reaction set ups.

As an example of something im not sure about:
I have a HCL salt of an aromatic amine which I want to neutralise to get the free amine. Carbonate is not strong enough and my supervisor suggested sodium hydroxide. However i have an ethyl ester in my molecule and I know these can be hydrolysed by NaOH at room temperature so I suggested using NaOEt instead (to which my supervisor replied that would just be silly, use NaOH). Now fine...Ill use NaOH but I dont know how to carry out this simple reaction without hydrolysing the ester too, do I just use dilute NaOH, or do I add exactly one equivalent and hope it just neutralises the acid?? Its these kind of decisions I have trouble with.

When reactions work I have no problem conducting them but I have a trouble if they wont go to completion. For example im currently doing an SNAr reaction with dibenzylamine (using triethylamine as a base). The reaction has gone half way and seems to have stopped even after I added an extra equivalent of each and raised the temperature. At this point I dont know how to push it to completion?

And something that really confuses me. When I heated the reaction up my supervisor said I may have boiled off the triethylamine and stopped the reaction. But I'm doing this with a reflex condenser going, I thought the whole point of the condenser was that I can have everything boiling and it will condense it all back into a liquid. At uni I always had the solution boiling visibly but on my placement Im basically told not to heat it up too much. Confused.

Sorry for the simple questions but once I get these basics down I dont think Ill have as much trouble with new reactions.

[Mobius1988]

Lithium, being less electronegative, has slightly more covalent character when it is a counterion, and although any alkali hydroxide can be used to saponify esters, LiOH gives me the best results in the lab for that reaction.

Sorry, Its my understanding that electronegativity increases up a group so lithium is more electronegative? Was that just a typo?

[fledarmus]

I'm sorry, yes, that was a typo.

For the HCl salt, how are you planning on removing it? Usually it is done by extraction - you dissolve your salt in water (if it will dissolve), add an aqueous base and extract with ethyl acetate or some other organic solvent. The contact of your amine with the base is minimal - the base is in the water and your acid-free amine is in the organic phase. Any material that does get saponified would also probably be base soluble and would be washed away as well. Your organic phase should contain the pure free base of your compound.

Is this the method you are using? If so, what happens when you dissolve sodium ethoxide in water?

As for the triethylamine boiling off, that depends a lot on the quality of your equipment and the temperature and scale of your reaction. A large scale reaction with normal headspace (say, 50 mL solvent in a 100 mL roundbottom) heated to a gentle reflux with a good-fitting reflux condenser probably won't lose very much. On a smaller scale with more headspace, (say 1 mL in a 10 mL flask) heated to a vigorous boil with a not-quite-tightly fitting reflux condenser might lose all of your base very quickly. Also, if your reflux temperature is significantly higher than the boiling point of TEA, it may all essentially be in the vapor phase - as soon as a drop falls back into solution, it immediately boils away again.

Getting a reaction to go to completion can be a real art form. Adding more starting material is always a good first try, but doesn't always work. Sometimes you have to work up the reaction and rerun the reaction on the crude product. And sometimes you just have to accept that it didn't go to completion, purify any product you can, and recover and rerun the reaction on your remaining starting material.

[Mobius1988]

Yes I am using ethyl acetate and intially I tried by adding the same volume of sodium carbonate but that didnt work at which point I thought I was going to add NaOH but then got worried about the ester. The main reason I was concerned is because I used NaOH to hydrolyde an ester in THF. But I now realise that THF and water are miscible so the hydrolysis will work well whereas if im using ethyl acetate the free base will go into the organic as you have said. So I will do the reaction in water/ethyl acetate and throw in some relatively dilute sodium hydroxide and see what happens.

Dissolving ethoxide into water would give me ethanol since EtO- is more basic than OH-. And ethanol is miscible in water and ethyl acetate so that would just cause problems I believe?

Ah I see what you mean about the triethylamine. So if my reaction hasnt gone to completion could I transfer it to a microwave vial and heat it up in there instead without losing the triethylamine?

[fledarmus]

Dissolving ethoxide into water would give me ethanol since EtO- is more basic than OH-. And ethanol is miscible in water and ethyl acetate so that would just cause problems I believe?

The ethanol isn't really the issue. What else forms? NaOEt + H₂O  EtOH + ?

Ah I see what you mean about the triethylamine. So if my reaction hasnt gone to completion could I transfer it to a microwave vial and heat it up in there instead without losing the triethylamine?

Oh yes - if you have a microwave reactor handy, that is indeed a very useful way of running substitution reactions, even with otherwise volatile reagents. That takes a lot of the work out of selecting conditions!

I've also found that switching to Hunig's base (diisopropylethylamine) instead of triethylamine can help in this circumstance. It is a little higher boiling and a little more soluble in most organic solvents than TEA, and not much more expensive.

[Mobius1988]

Ah sorry was being a bit quick to answer there, obviously you form sodium hydroxide anyway, so might as well just do the reaction in sodium hydroxide ayway.
Ive attempted the reaction now and as you said some of my free ester went into the ethyl acetate layer but most of it stayed in the water. I thought that maybe it just wasnt that soluble in ethyl acetate so I tried extraction with 3:1 CHCl₃:IPA but that didnt work either, even after a second NaOH wash. So I think my product may just be very water soluble anyway.

I trialed my dibenzylamine reaction in the microwave and it seemed to go an extra 10% so ive got the whole reaction mix in there over the weekend and will see how it goes. Thanks for all the help. Ill probably post here in future if I have any more issues. Cheers again.

[Mobius1988]

Actually, I have just thought of another practical issue I had a couple months ago.

I was attempting to place an ethanol chain onto a heteromatic core by using lithium halogen exchange (with an iodine attached to the core) followed by ring opening of ethylene oxide (oxirane) using BF₃.OEt₂ as a lewis acid catalyst. Oxirane is a gas which is the cause of the issues I had.

First time I tried the reaction i pass the oxirane gas over the reacton vessel which was cooled to -78 degrees for the Li exchange. This condensed a huge excess of the oxirane into the reaction vessel and i got ring opening polymerisation and a very poor yield as a result.

Second time I used a piece of kit that allowed me to condense the gas into the reaction vessel dropwise but obviously I couldnt tell how much oxirane was in each drop so again I had difficulty preventing the polymerisation.

I found a paper that managed to do it with a decent yield (70%) and they prepared a solution of the required equivalents of oxirane and BF₃.OEt₂ which they cannula transferred into the reaction vessel. So my question is this: How would I go about condensing a gas into a solution while measuring the mass transfered? Keeping the gas in the solution would require a low temperature and you would get ice on the reaction flask which would completely mess with your measurements surely?

[fledarmus]

The boiling point of ethylene oxide is low, but it isn't incredibly low. If you condense your ethylene oxide under an inert gas into a cooled receiver with a septum, you can purge a syringe through the septum, then suck up and expel the ethylene oxide to cool the syringe enough that the ethylene oxide doesn't immediately boil when it hits the syringe. Finally, suck up the appropriate volume and transfer it through a syringe into your reaction flask, also in an inert gas and containing your solvent. Reasonably dilute solutions of ethylene oxide are stable enough at room temperature that you should be able to weigh your reaction flask and determine more precisely how much ethylene oxide you've added.

Or you can buy a solution of ethylene oxide in either THF, methanol or methylene chloride: http://www.sigmaaldrich.com/catalog/search?interface=All&term=ethylene%20oxide%20solution+&lang=en&region=US&focus=product&N=0+220003048+219853269+219853286

[Mobius1988]

Thanks again, our stores had it available as a gas so we used it as it came without realising you can get it in solution.