Sciencemadness Discussion Board - Synthesizing propylene carbonate

Synthesizing propylene carbonate

Upsilon - 20-9-2015 at 15:25

Disclaimer: I know next to nothing about organic chemistry. I am indulging in this experiment for the sole purpose of having propylene carbonate to conduct more inorganic chemistry experiments. I am specifically interested in its use as a polar, aprotic solvent.

Doing research, I have found several scholarly articles about the ability to synthesize propylene carbonate from propylene glycol, urea, and a ZnO catalyst - all relatively cheap and readily available. Methods from article to article vary, but the basic reaction is the same. While I know the stoichiometric amounts of both chemicals I need, I really have no idea how or why this reaction happens. You can explain it in layman's terms if you like, but I can't promise that I'd take terrible interest in it.

Essentially I need to know about all the nuances of this reaction and what factors I can adjust to get higher yield in less time. I don't know how much catalyst to use, what temperature this should take place at, etc.

gdflp - 20-9-2015 at 16:09

It's not as easy as those papers make it seem, see here for a discussion about the synthesis. There are significant problems with the synthesis and some experimentation is needed to optimize it and prevent tar formation.

Upsilon - 20-9-2015 at 16:38

I see. The problems he had, if I understand correctly, were from distillation to separate the PC from the PG? For my use of it as a solvent, would I not necessarily need to separate the two? As long as I can drive the reaction to be dominantly PC, then it should still work as a solvent. The PG still present would be inert for my purposes and not really affect anything other than the amount of the PC/PG mixture needed to dissolve a certain quantity of substance.

gdflp - 20-9-2015 at 16:52

Quote: Originally posted by Upsilon

I see. The problems he had, if I understand correctly, were from distillation to separate the PC from the PG? For my use of it as a solvent, would I not necessarily need to separate the two? As long as I can drive the reaction to be dominantly PC, then it should still work as a solvent. The PG still present would be inert for my purposes and not really affect anything other than the amount of the PC/PG mixture needed to dissolve a certain quantity of substance.

Not if you want an aprotic solvent. Propylene glycol has a very similar pKa to water, and thus the mixture will not be aprotic, rather it will act similar to water in this regard. They will need to be separated if you want to use this as a solvent for any strong bases, or reactive metals, e.g. alkali metals.

Upsilon - 20-9-2015 at 17:03

Hm, I wouldn't have expected PG to be that protic. Looks like I've got some work to do then. Thanks for the info.

byko3y - 21-9-2015 at 04:23

gdflp, could you give us a source of your pka value for propylene carbonate? Because acidic solvents react with alkali metals, as well as water does, so they would be not suitable for electrolysis.

Upsilon, the reaction is pretty simple, it can be compared to amide solvolysis, ammonia just needs to be driven away so the equilibrium
Urea + Propylene glycol <=> propylene carbonate + 2NH3
will be shifted to the right.
I'm sure a lot of metal oxides could catalyze the reaction, ZnO is just more effective.

gdflp - 21-9-2015 at 04:27

I don't have a pKa for propylene carbonate, I was talking about the pKa of 1,2-propanediol which is 14.9.

clearly_not_atara - 21-9-2015 at 08:01

Why not make dioxane? The reaction is rather simple, ethylene glycol/H2SO4, and the product is quite versatile, not quite the same (less polar/lower boiling) but certainly gives you more to work with than the standard water/etoh/chcl2/toluene palette everyone uses. Dioxane can form peroxides in storage, though, so be careful.

gdflp - 21-9-2015 at 08:06

Dioxane is not polar, and thus it is incapable of dissolving many inorganic salts which dissolve readily in propylene carbonate. I believe that this is what the OP was interested in, hence mentioning the polar aprotic solvent.

DCM and dioxane are quite similar in terms of solvent properties, the only major difference is the boiling point.

Oscilllator - 21-9-2015 at 17:35

Quote: Originally posted by Upsilon

Hm, I wouldn't have expected PG to be that protic. Looks like I've got some work to do then. Thanks for the info.

Upsilon unless you have a fractionating column or a novel idea for a separation method it may not be worth your while to try this. IIRC the consensus reached on the other thread was that vacuum fractionation would be the best way to achieve separation. When I attempted this I think I may have pushed the distillation a little hard, so I would recommend adjusting the distillation speed to be quite slow (0.5 drops/sec max) so as to get better separation.
It's also worth pointing out the importance of the preparation of the catayst. I think one reason I may not have achieved full conversion was improper preparation of the catalyst. Note that UnintentionalChaos' prep did not use a calcined Zinc oxide catalyst and perhaps as a result of this he did not isolate any propylene carbonate.
If you can, I would recommend following the patents directions exactly (heating to 500°C for 4 hours) rather than heating the oxide to ~1000° like I did or to heating at 200° like UnintentionalChaos did. Certainly there is a big difference between these two methods, as my ZnO did not dissolve in the reaction mixture whereas UnintentionalChaos' did.

Upsilon - 17-10-2015 at 09:15

Hello again all, apologies for neglecting this thread but I haven't actually had the time to try this out yet (though I will soon). I just had a thought - above it was mentioned that PG will be reactive towards alkali metals. What exactly is the reaction going on when this happens? Because if it forms a solid product (a salt) then, after some simple distillation to drive off a lot of the PG, you could run an NaCl electrolysis in the PC/PG mixture. The sodium metal will react with any remaining PG, and afterwards it would be extremely simple to separate the PC from the salt formed. This idea has gotten me very excited, who'll be the first to burst my bubble??

byko3y - 17-10-2015 at 11:52

Can't you just distill the propylene glycol off?

Upsilon - 17-10-2015 at 12:05

Quote: Originally posted by byko3y

Can't you just distill the propylene glycol off?

Based on what has been posted earlier this is probably much harder than it sounds. Apparently a fractioning column is needed, and that the distillation foams a lot. Not to mention that introducing this high heat can form tar. This electrolysis method would be a lot cleaner, if I can get any confirmation. I'm still about to try it anyway.

gdflp - 17-10-2015 at 12:16

Quote: Originally posted by Upsilon

Hello again all, apologies for neglecting this thread but I haven't actually had the time to try this out yet (though I will soon). I just had a thought - above it was mentioned that PG will be reactive towards alkali metals. What exactly is the reaction going on when this happens? Because if it forms a solid product (a salt) then, after some simple distillation to drive off a lot of the PG, you could run an NaCl electrolysis in the PC/PG mixture. The sodium metal will react with any remaining PG, and afterwards it would be extremely simple to separate the PC from the salt formed. This idea has gotten me very excited, who'll be the first to burst my bubble??

Propylene glycol is an alcohol, and sodium metal is basic. The sodium metal will replace both acidic protons with sodium ions, releasing hydrogen gas and forming disodium propane-1,2-diolate. If the mixture is rich in propylene glycol, you would likely form a slush of the alkoxide in propylene carbonate, which would be difficult to filter and might need to be processed several times to prevent the solution from becoming too full of crystals. You may also encounter troubles with chlorination of the alcohol from the produced chlorine. It's worth a shot, but separating the glycol with a couple vac distillations is likely going to be less troublesome.

BromicAcid - 17-10-2015 at 12:50

Here is a thread where I was working toward ethylene carbonate, same procedure, different diol.

http://www.sciencemadness.org/talk/viewthread.php?tid=2948

Upsilon - 17-10-2015 at 12:59

Quote: Originally posted by gdflp

I'm aware of the possible effect of chlorine so that's why I'd separate it into 2 cells with a salt bridge. I'd use water as the solvent at the anode so not to waste any valuable PC. It should be pretty easy to tell once all of the PG is gone, since visible sodium metal will theoretically start forming.

As for having too high saturation of PG, I don't know. I used the 1.5 PG:urea ratio and 1.5g ZnO catalyst as outlined in the papers. I didn't really calcinate the ZnO like above mentioned, since I'm only doing a test run currently. I put the ZnO in a crucible and held a butane torch to it for several minutes.

I am in the middle of the reaction currently and my ZnO has dissolved which I don't assume is good. The mixture has also turned yellow, which I also assume is not good. It is still bubbling very vigorously though.

UPDATE: The yellowness has indeed turned into the tar described in the other thread.

Note also that I'm not trying to distill it as the reaction progresses. I am simply heating the mixture in open air.

Well, the contents all turned into yellow mass with no liquid left. Pretty safe to say it was a failure.

[Edited on 17-10-2015 by Upsilon]

[Edited on 17-10-2015 by Upsilon]

[Edited on 17-10-2015 by Upsilon]

clearly_not_atara - 17-10-2015 at 16:22

What if you react carbon dioxide with the mono-alkoxide in some inert solvent (acetone?) and isolate the monocarbonate salt, then cyclize that somehow?

halogen - 18-10-2015 at 01:40

The formaldehyde acetal of glycol should be vulnerable to radical oxidation. There's an effect that activates to radicals the carbon next to ether or alcohol. Something about a lone pair. I don't know. But you might look up the oxidation of the acetal, called dioxolane, it is a solvent, with something with lots of oxygen, like permanganate.

Oh.
"Formation of Cyclic Carbonates. In the absence of water, chlorohydrins such as 2-chloroethanol and l-chloro-2-propanol react with an alkah carbonate or bicarbonate to produce cycHc carbonates such as ethylene carbonate [96-49-1] and propylene carbonate [108-32-7] ia yields of up to 80% [c.73]"

Oh; again.
https://books.google.com/books?id=jsMFRVsqtmEC&pg=PA82&a...

These two from the first page of a google search. Told ya so.

[Edited on 18-10-2015 by halogen]

byko3y - 18-10-2015 at 08:41

halogen, Chlorohydrins are toxic and carcinogenic.
Upsilon, what are you doing on the chemistry forum without distillation apparatus with fractioning column? You can't purify any solvent you use, including the propylene carbonate that might have contained some amount of acetic acid or other crap. AFAIK ZnO should not become dissolved during the reaction. Contamination of urea might also be a reason for the problem.

Quote:

I am simply heating the mixture in open air.

You should blow some inert gas through the reaction in this case, because otherwise amides will be formed instead of the propylene carbonate. In fact it's much easier to apply a slight vacuum (300 mmHg; propanediol-1,2 reflux).
Also, you haven't mentioned what procedure exactly were you using. I'm talking about
Synthesis of propylene carbonate from urea and propylene glycol - doi:10.1016/S0167-2991(04)80318-5

[Edited on 18-10-2015 by byko3y]

Upsilon - 18-10-2015 at 11:20

Quote: Originally posted by byko3y

what are you doing on the chemistry forum without distillation apparatus with fractioning column?

Did I ever say I didn't have one? But, true, I do not have a vacuum distillation setup. That's exactly why I asked about using electrolysis to purify the PC, and theoretically it is possible.

Quote: Originally posted by byko3y

You can't purify any solvent you use, including the propylene carbonate that might have contained some amount of acetic acid or other crap.

I'm not sure what you mean? The only contaminants I can forsee are the adverse products of the reaction that I have already witnessed.

Quote: Originally posted by byko3y

Again, this organic stuff is over my head. However if inert gas is involved, then the whole prospect of propylene carbonate is useless to me; I asked about it because I was interested in electrolyzing alkali metal salts in it, as opposed to setting up an inert atmosphere to electrolyze the molten salt in. I suppose there could still be some benefits, but requiring an inert atmosphere for the reaction is a huge blow to its usefulness to me.

Quote: Originally posted by byko3y

Also, you haven't mentioned what procedure exactly were you using. I'm talking about
Synthesis of propylene carbonate from urea and propylene glycol - doi:10.1016/S0167-2991(04)80318-5
]

Though the URL is not working for me, that's the process I'm using - urea + propylene glycol + ZnO catalyst -> propylene carbonate + ammonia

clearly_not_atara - 20-10-2015 at 23:34

Why not just go for acetonitrile instead? It's leagues easier to make than either propylene carbonate or DMF, never mind DMSO.

AcOH + CO(NH2)2 >> AcNH2 + pyrolysis >> MeCN

EtBr + aq NH3 + I2 >> MeCN

IIRC at least the first route has some reports here.

Oscilllator - 21-10-2015 at 15:21

Quote: Originally posted by Upsilon

Did I ever say I didn't have one? But, true, I do not have a vacuum distillation setup.

Upsilon I am afraid this is going to be a big problem. The main reason I stopped attempting synthesis of this compound was that I did not have a proper aspiration pump, only a water aspirator. This was a big problem as it meant I couldn't really run the distillation for very long without wasting hundreds of litres of water.
I would strongly recommend a cheap vacuum pump as your next chemistry purchase as they are tremendously useful.
I would also recommend heating the catalyst for a bit longer - try and get it hot enough to see that cool yellow color.

Ozone - 21-10-2015 at 16:05

When last I checked, dioxane is an ether that is miscible with water. That makes it both polar and aprotic (like THF). It can solvate/complex many inorganic compounds (whether or not it is good for the OP remains to be seen).

I've also used it to solvate polar compounds for GC (to avoid water).

Otherwise, see:

your approach? http://www.sciencedirect.com/science/article/pii/S1001841708... and http://ir.sxicc.ac.cn/bitstream/0/1501/1/573-576.pdf (especially)

From CO2 and 1,2-propylene glycol: http://pubs.acs.org/doi/abs/10.1021/ie070789n

From urea and 1,2-propylene glycol: http://pubs.acs.org/doi/abs/10.1021/ie049948i

It can also be done from propylene oxide and CO2, but that seems to require a supercritical fluid cell.

I'm not bothering to cover the obvious use of phosgene.

Good luck,

O3

Waffles SS - 23-10-2015 at 03:07

Synthesis of Alkylene Carbonates in Ionic Liquid

Quote:

We now propose a new synthesis of alkylene carbonates from urea and alkylene glycols, which requires neither elevated pressure nor high temperature. The reactions are carried out in a two-phase system consisting of a ionic liquid and a chlorinated organic solvent.As ionic liquid we used a mixture of equimolar amounts of a metal chloride, urea, and glycol,which formed a colorless transparent solution on heating above 50°C. The yields attained 80%. The reactions were carried out with a number of glycols and metal chlorides.
V. A. Kuznetsov, M. G. Pervova, and A. V. Pestov

Attachment: kuznetsov2013.pdf (153kB)
This file has been downloaded 384 times

[Edited on 23-10-2015 by Waffles SS]

Upsilon - 27-10-2015 at 17:42

Wait, dimethyl sulfoxide is marked as polar and aprotic, and is readily available for quite cheap. Is there a reason this isn't used?

m1tanker78 - 27-10-2015 at 18:22

Quote: Originally posted by Upsilon

Wait, dimethyl sulfoxide is marked as polar and aprotic, and is readily available for quite cheap. Is there a reason this isn't used?

For alkali metal production? Unfortunately this doesn't work. I've tried and another member (bromic_acid IIRC) has tried and reported failure. A reaction can be observed just dropping a piece of sodium or potassium into DMSO.

I came close to attempting synthesis of ethylene carbonate by reacting ethylene oxide with CO2. I set everything up for a small scale and then chickened out and aborted. Ethylene oxide is nasty and extremely flammable and/or explosive. No thanks!

Upsilon - 30-10-2015 at 18:26

Well I'm not going to give up on making PC. I need to know what exactly is responsible for tar formation - is it due to reaction with the air? Improperly calcinated ZnO? Both? I am willing to build a crude vacuum apparatus from PVC pipe if needed (it would just be a cylinder, NOT a distillation apparatus), but will ammonia gas be harmful to an electric vacuum pump? For separating the PC from unreacted PG I am going to try electrolysis of alkali salts in it to react with the PG.