And another condenser on top of the soxhlet. Forgot to add that.

Ethanol is cheap, you can just try to use it in excess.

Eduardo Maurina Morais Ethanol is the solvent in which the reaction takes place and thus is already in excess; and the carbonyl OH is replaced with -OC2H3, leaving water and the HCL as byproducts. So I do indeed understand if you are suggesting that I simply complete the reflux and rotovap the ester and equilibrium products, adding back EtOH several times to remove the water as an azeotrope, and the HCl. But when you are using the solvent as one of the reactants in the Fisher rxn the solvent in this case is already 'pushing' the rxn rightward as best as possible, I would think. If the rxn took place in toluene I could understand using more EtOH? Thank you.

If your ester is one with a relatively high boiling point, the method could be, - make a mixture of the acid, large excess of ethanol and an acidic catalyst like

p-toluenesulphonic acid (PTSA) and cook the mixture under reflux for a required time duration. At the end collect your product by fractional distillation.

Thank you, Ashis Ranjan Bandyopadhyay. I wonder if this would be done because PTSA is a weaker acid c/to HCl and the tosyl leaving group is comparatively more stable? The final product would still be in equilibrium, no? I'm sure the K would likely be different, but in whose favor? (I forgot to state that I used dry-or compressed- HCl from a bottle.) Distillation of the ester (oil in this case) can be done effectively in either catalytic mechanism. No? But I don't know the BP of my ester (for me, just an intermediate for an electrophilic ring closure). Wow! I thought I knew the ways to make an ester. There are so many. But your thinking about PTSA causes me to contemplate it's use at another site, to protect an alcohol group elsewhere in the design. You see, I was going to employ a non-equilibrium approach using SOCl2, to make an acyl halide intermediate, which would then yield an irreversible rxn using EtOH. However, on the other end of the carboxylic acid function is a tertiary alcohol. The tetrahedral carbon OH is also joined by two phenyl groups. So I would likely make a chloro alkane, I think. I need the alcohol there to close to an oxazolidinedione using sodium ethoxide. Therefore, I contemplated using PTSA to protect the OH which could then regenerate. BUT, with the C(C6H6)2 at the same tetrahedral carbon there might be steric hinderance impeding a catalysis; or perhaps the electronic properties of the diphenyl group might help. I should state simply that I am making 5,5-Diphenyl-2,4-oxazolidinedione. I have made it before buy with poor yield. So the bottom line is that I wished to go the acid chloride route. But I was going to seek advice regarding protection of the alcohol with PTSA, and that now might be a useful diversion for me. That is, can I protect the alcohol during the formation of the acid halide from benzillic acid, thus HO-C(C6H6)2COOCl? Well, I simply talk too much; a product of not knowing what I am doing. Really, colleagues, I was simply trying to get some validation for contemplating sieves 3A here to remove a reactant, getting a better yield without having to distil. I was contemplating a 'single pot' synthesis without distilling the ester. This is commonly done. But the real problem and question now becomes, "Can I protect the alcohol? I wonder if the tosyl would protect the alcohol for ring closure later, and catalyze the ester with tosyl at the same time? I have no pride, friends. I am showing how little I know after all these years. Many thanks.

Regarding the end, "I wonder if the tosyl would protect the alcohol for ring closure later, and catalyze the ester with the tosyl at the same time?" Probably not. I forgot, I need to make the ureide prior to closure to the oxazolidinedione. Sorry.

Gary -

This may be easier than you are making it. Sieves are great for some uses but there are several things to be aware of before you start. Between the need for a really hot oven and the apparatus, the sieves/Soxhlett method can be inconvenient unless your lab is equipped for these procedures. Things to consider:

1) As purchased, you can expect mol sieves to be mostly hydrated, so they need to be baked out at high temperature - ca 400C at atmospheric pressure or > 250C under vacuum. Once dried, sieves should be stored in a desiccator. You have already looked into the extra glassware (Soxhlett, etc) so you know what all this entails. Also, my experience with Soxhletts is that a lot of fussing is required to be sure that the reflux rate is stable, insulating the upper parts, making sure that joints don't leak, etc.

2) Often water removal is not necessary for esterification simply because of mass action (lots more alcohol should be present in the solvent than water produced). Other tricks (e.g. slow distillation) can also work to do what you want with the mol sieves to do.

3) Tosic acid and HCl are both sufficiently strong acids for most esterifications. Please specify your carboxylic acid (or at least the class of compound) if you are not sure about this. Remember HCl and TsOH are catalysts, not reactants, so the esterification reaction is the same, so the equilibrium constant is the same regardless of catalyst used. The only differences are in the rates of esterification and the methods of removing acid after reaction is complete. Acid catalysis will not produce tosyl esters from alcohols under the conditions you are interested in, but if there are both alcohol and acid groups present in your substrate, intramolecular reaction to make a lactone may be favored if this would produce a 5-8 membered ring, or it may lead to formation of oligomeric products.

4) Avoid ethanol that contains denaturants - methanol and/or isopropyl alcohol as well as acetone are all commonly present in "denatured alcohol" and this leads to a mixture of products - mostly what you want but the other esters will be hard to separate.

A method I have often used for water removal is slow distillation, where water co-distills with ethanol (or methanol/toluene) as it forms in the reaction mixture. (You can select a solvent combination to do this by looking up azeotropic mixtures. E..g. distillation of mixtures with greater than 96% ethanol will remove all water remaining.)

• In a 1-neck round bottom flask, add absolute ethanol to your carboxylic acid and introduce acid catalyst - solid TsOH.H2O, HCl(g) or one or more drops of HCl(conc) - water in the catalyst will be removed by distillation. Typically, 2-5 mol % should be sufficient. (Note 1 drop of HCl(conc) consists of ~ 0.3 mmol acid and ~ 1.1 mmol water.) If you are not in a hurry, you may want to set this up to stir at room temperature overnight before proceeding. If your acid is a solid with limited solubility in EtOH, this will help you see if dissolution of the solid is a sign of reaction progress. If your mixture is clear at this stage, you should test out your method of detecting unreacted carboxylic acid now.
• Reflux the mixture with a vertical condensor for an hour or so
• Fit the flask with a distillation head and SLOWLY distill - 1-20 drops a minute will be less than 1 mL/min. Monitor carefully to avoid boiling dry.
• You need to monitor the reaction by using TLC or another analytical method which you intend to use later. (Make sure you have this worked out before heating begins so you are sure you know to detect unreacted acid.) If you see that starting acid remains as the liquid volume gets low, stop distillation and add more absolute ethanol before restarting distillation.
• Removal of the catalyst after the reaction is one of the biggest drawbacks of tosic acid, on the other hand HCl will evaporate when you remove excess EtOH. If your product is a straightforward organic ester with no charged groups, you evaporate ethanol and redissolve the residue in organic solvent (e.g. EtOAc) and wash the organic phase with NaHCO3 solution. (Small portions of bicarbonate may be added to the cooled residue before organic extraction - easier to control foaming.) Check the literature for isolation procedures with similar compounds if you have more complex functionalities.
• If you're really not in a hurry, dried strong cation exchange resin (Dowex-50, Amberlite IRC-120, etc.) can substitute for tosic acid, but recognize that this will go MUCH slower than esterification with soluble acid catalysts. The big advantage is that you can remove catalyst by filtration, so you don't need a bicarbonate wash after ethanol evaporation.
• Purify by crystallization, distillation or chromatography as required.

Oops, sorry I didn't read your long answer to Ranjan - this clarifies a lot.

Now my first response is that the answers to your issues must be in the literature, so however clever any of us is, someone out there has done what you need and you could avoid time and trouble reinventing the wheel with a thorough library search. That said, I think you can probably do this in a 2-reaction one-pot procedure by making cyclic a intermediate with phosgene or a phosgene equivalent:

3 Ph2C(OH)CO2H Cl3COC(=O)OCCl3* +6 Et3N --> 3 Ph2C >S(=O)2 before reaction with cyanate.

You should start the literature search trying to find Ph2C

James Kilgore I am sincerely appreciative for all your efforts to help. I thought I wrote too many words, but it seems we are alike on that account. Truly, you generated a trove of useful information and I do thank you, as it has been very helpful. At the same time I am a bit embarrassed as I caused people to work so hard when I simply raised a nice (albeit trivial) little thought about water removal in the Fisher rxn. It's clear that I am new to this forum. But sincere thanks to all.

This synthesis was described just once, as far as I can tell, in the early 1940's (Stoughton, 5,5-dialkyl-2,4-oxazolidinediones, JACS 63;2376;1941). I actually synthesized this, among other compounds over 20 years ago, but started with benzilic acid as that made for an easier synthesis. (Anticonvulsant Potency and Toxicity of 5,5-Diphenyl-2,4-Oxazolidinedione: Drug Development Research 21:19-27, 1990). It was part of an anticonvulsant qsar study, where I compared the respective hydantoin, succinimide, etc. with quantum chemical correlates (in other papers). I gave the rest of the compound to Dr. Codding at the U. of Calgary for crystallography studies, as she had a like interest. Now, I just pulled that paper, and my yields were actually very good, even though I didn't distill. Probably because, as you said, ethanol was in excess as it was the solvent. My comment about the sieves was initially just a thought to apply in acid catalyzed equilibrium ester synthesis in general.

Yes, even though at 73 this is mostly a hobby now, my lab is well equipped, and I keep dry sieves around to keep my absolute EtOH dry. PhD in pharmacology, and have interests in pharmacognosy, so we have lots of Soxhlet equipment and h/o their application. Nevertheless, my thought was trivial and misplaced.

As an aside, this product isn't a dioxolane, as upon making the ethyl ester of benzilic acid I added in same pot sodium ethoxide and refluxed, which closed the oxazolidinedione ring.

So, now you know that only my Bachelor's degree is in chemistry. The problem was that I retained such a love of chemistry that it interfered with my pharmacology training; they didn't know what to do with me. Then along came a neurology practice, and am now in an active retirement.

But if I might push the discussion just to ask one more question: In my limited synthetic career I have found that I make a lot of acid halides. And I did contemplate making the acyl halide of benzilic acid, followed by irreversible esterfication with EtOH. I would then close the ring with Na ethoxide and urea. However, my expectation is that the thionyl chloride I might use would likely replace the alcohol group of benzilic acid with chlorine. Hence, ring closure wouldn't occur. So it was in this context that I considered whether I might protect the alcohol with a tosylate before making the acid chloride. Recall, the structure is

HO-C(C6H5)2COOH

So to make an acid halide without messing with the -OH which is required for closure to an oxazolidinedione.

I am very grateful for everyone's comments. Please don't put much more effort into this, unless you might offer an alternative way to make the acid chloride, keeping the alcohol intact.

I tried to upload a Spartan file of the molecule but I don't think it is going to work.

Who knows, maybe it's dementia time!

Best Regards to All

You're right about my running on and explaining things that aren't asked for, but your original question was extremely general. I also think you really didn't understand my pseudo-SMILES notation - rings are hard to recognize. Structures for the sequence benzylic acid -> diphenyldioxolane -> diphenyloxazolidinedione are attached. The hydroxyl is "protected" by using it as the anchor for the phosgene-derived carbonyl group that's activating the carboxyl from benzylic acid [instead of RC(=O)Cl].

B.t.w. the name I gave before was for the intermediate (see https://pubchem.ncbi.nlm.nih.gov/compound/5_5-Diphenyl-1_3-dioxolane-2_4-dione), not your product. Am pretty sure was a synthetic paper in Tetrahedron or Tet Lett within the past five years that used related chemistry and this is obviously related to very widely used and studied sedatives and anticonvulsants, so it's hard to imagine that there aren't multiple literature preparations.

You're efforts have succeeded. You have taught me new chemistry, and I am much appreciative. This looks like an excellent approach. As a member of the ACS I simply have access to their journals, and Tet/Tet Letters might be one, but I don't recall that being an ACS journal. Searching the specific chemical I couldn't find anything except the old literature. I wish I had a librarian. Doing what I am doing, I don't have access to online university services, e.g. their library. Yes, the discussions did expand quite a bit. Initially it was a simple comment but I let my aspirations loose, seeking more information, and did not formulate the issues as concisely as I should. That, too, has taught me something. "Be more concise." Finally, I see that I simply have to save my documents as a jpg file and they will attach. Thank you for you're excellent contributions. The very best to you.