Sciencemadness Discussion Board

Preparation of Thalidomide

karlos³ - 6-3-2020 at 19:51

I want to present you guys my first attempt at the preparation of this infamous medication, a known teratogenic which caused one of the largest scandals in the pharmaceutical industry ever.
It causes horrible birth defects, but is still in use nowadays against leprosy, certain types of cancer, and even a harmless but ugly skin condition called psoriasis.
Its intended use was originally as a sedative and then it was found to be effective against pregnancy induced morning sickness, which lead to increased prescription especially for pregnant women, and thus the whole misery had started.
Interestingly, it is actually even a strong sedative, as it is structually somewhat related to barbiturates.

A. Reagents
- L-Glutamic acid
- Phthalic acid anhydride
- DMF
- Pyridine
- PEG-400
- Ammonium acetate
- HCl N6
- aqueous Ammonia 25%

B. Equipment
- heated Magneticstirplate
- Oilbath
- Erlenmeyer flask
- 30cm Liebig condenser with moderately cold water running through
- Vacuum filtration setup

C. Procedure

Preparation of N-Phthaloyl-DL-glutamic Acid

- 14,95g of phthalic anhydride and
- 14,74g of L-glutamic acid were given together into
- 40ml of pyridine, then put into a preheated oilbath and the temperature increased until at 115°C.
It was stirred at this temperature for 100min, then the heat was turned off and the oilbath removed.
When the temperature fell to ca. 70°C, 150ml of cold water were added to the stirred mixture in the flask.
Enough N6 HCl was added until the mixture was strongly acidic, then the clear solution was left on the stirrer for a few hours, standing in a cold water bath.
When the mixture was checked the next time, circa 3-4h later, a very fine powdery precipitate was visible in the lower half of the flask.
This precipitate was vacuum filtered and washed with 3x20ml of water, then dried with moderate heat.
A sample of the dried powder melted at 192-194°C.
The total weight was 20,5g, which corresponds to 73,9%(also, 73,9mmol of the theory).

Synthesis of Thalidomide

- 10,25g of N-Phthaloyl-DL-glutamic Acid(37mmol) and
- 10g of anhydrous ammonium acetate(129,5mmol, 3,5eq) were heated in the oil bath on the stirrer, in
- 25ml PEG-400.
That is were my thinking went wrong though.
The reagents still hadn't satisfyingly dissolved in my chosen solvent when around 100°C, so I pondered a bit and tried to make at least up for some yield opposed to no yield.
I added 25ml of DMF to the mixture, which is a known working solvent for this reaction, just not very effective.
So, it was heated further until the mixture reached 152-155°C and was kept there for 50min, then the oilbath was removed and the heat turned off.
At around 95°C, 200ml of cold water were added to the mixture, which immediately resulted in a few very tiny precipitates.
The mixture was then put into the ice-bath and left there for an hour.
More precipitated, and I had the suspicion some of it could be unreacted substrate, so I added a bit of 25% ammonia solution to keep any unreacted substrate in solution and to have only product as the precipitate.
The solid was filtered out and washed with 3x50ml of cold water and 2x20ml of cold ethanol, then dried using moderate heat.
The dried substance was weighed and it was not surprising that it amounted to only 1,1g of an off-white powder.
Then it was recrystallised, dissolved in DMF and precipitated with ethanol and the next day filtered, washed with cold ethanol(2x20ml), and dried again.
The resulting substance, now really white, amounts now to 930mg, or 3,6mmol, which corresponds to 10,27% of the theory.
A melting point test resulted in a temperature of 268-270°C.


D. Discussion

The second reaction to form the gluthethimide ring was totally crap.
I had the idea that the reaction could be done in PEG-400, since it is so high boiling(ideally, diphenyl ether at a temperature of 175°C showed the best results).
But the solubility issues crushed my hope about that.
So I had to use the much lower yield giving DMF(it results in around half as much product as the diphenyl ether).
The advantage I had with the DMF/PEG-400 mixture though was, that I had a simpler workup, since the solvents are soluble in water, unlike diphenyl ether.
Possible yield with diphenyl ether is around 65%, using DMF they haven't even reached 40%, merely 38% of the theory.
This was the reason why I thought PEG-400 would be a good choice, table S7. on page, that I was under the impression the temperature of the reaction correlates with the yield, and diphenyl ether was both the highest boiling and yielding of the ones they tried, with the yield somewhat increasing together with the temperature.

As I still have some precursor, and as this one was easily made, I will try the second reaction again without making such stupid mistakes and then trying to save everything with a not very smart decision.
But at least I got still a bit out of it instead of nothing :)

E. References

Vu, B. D., Ho Ba, N. M., & Phan, D. C. (2019). A FACILE SYNTHESIS OF THALIDOMIDE.
https://sci-hub.se/10.1021/acs.oprd.9b00122

SUPPORTING INFORMATION
A FACILE SYNTHESIS OF THALIDOMIDE
Binh Duong Vu, Ngoc Minh HoBa1, Dinh Chau Phan
https://pubs.acs.org/doi/suppl/10.1021/acs.oprd.9b00122/supp...

unionised - 7-3-2020 at 06:32

Interestingly, it sometimes causes psoriasis.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4078429/

morganbw - 7-3-2020 at 07:57

Carl sir, respect from me.
A very nice write up.

Tsjerk - 8-3-2020 at 04:25

Very nice write up! Thalidomide is nice stuff as long as it is not used by pregnant women.

Maybe you could try this with urea in a microwave? As soon as the water is gone no more radiation is absorbed, so once the reaction is done the reaction stops. No need for a solvent nor temperature control. I did this with a different subtrate.

Source:
https://www.mdpi.org/ecsoc/ecsoc-5/Papers/e0008/e0008.htm

AvBaeyer - 8-3-2020 at 18:53

Karlos:

Interesting report. Finally some real chemistry being posted - thank you. Here are a couple of comments.

If you use L-glutamine in your first reaction in place of L-glutamic acid, the product (N-phthaloyl-L-glutamine) can be cyclized directly to thalidomide using acetic anhydride.

See: F.E King et al., J. Chem. Soc., 1959, 873-880.

Carbonyldiimidazole (CDI) and cat DMAP in THF with N-phthaloyl-L-glutamine can also be used for the final step in place of acetic anhydride.

See: G.W. Muller et al., Org. Process Res. Dev., 3, 139-140 (1999).

AvB

[Edited on 9-3-2020 by AvBaeyer]

woelen - 9-3-2020 at 00:09

This is a good example of nice amateur organic chemistry at a good level. Nice to see this kind of results.
Doing organic chemistry in a home lab at a decent level is quite hard, but as this demonstrates, it can be done!
Thumbs up!

karlos³ - 26-3-2020 at 16:44

Thank you all for the replies, but lets be honest, despite working this first trial was really half-assed!
Usually I do better, this was just something really unusual to me.

@Tsjerk, Thanks for the suggestion with the microwave method, I just haven't chosen this route because they used a sealed vessel, this was the reason not to choose this method.
I assumed it is something special similar to a bomb for high pressurized reactions, able to withstand quite some pressure and this was not something I would like to do in the homelab.
But you have simply made a premixed solution of the reagents, evaporated that dry and nuked it in the microwave and nothing more? That is impressive and something I would have chosen too.
I am sure that the method of your baclofen synthesis will be better than what I did here...
Just like this: http://www.sciencemadness.org/talk/viewthread.php?tid=104964...
Or do you any better and more specific suggestions?
If you could give me some tips besides that I would be very grateful!

@AvBaeyer, thank you for the suggestion of alternative routes, I've seen them actually beforehand already and my choice was the glutamic acid route because of the reagents on hand already.
While it may not be the best or highest yielding method, the advantage was for me that I only needed to purchase two cheap reagents, and no need to use my last little bit of Ac2O, or CDI which I do not have(sadly).
Specifically, I chose this route because it doesn't uses the nasty 4-DMAP but simply cheap pyridine for the synthesis of the intermediate.

But regarding one-pot conditions, there is also a way directly to the product with glutamic acid, under microwave conditions, with phthalic anhydride and the glutamic acid, and preferably thiourea as nitrogen source, giving somewhat around 50-60% yield this way.
Ammonium acetate, urea, were tried by them too, but this increases the quantity of the side product phthalimide, while thiourea seemingly favors the glutarimide formation.
While interesting, the use of microwave conditions, specifically these ominous special-microwave-able and sealed vessels were what made it unattractive to me, as I have almost no experience with reactions in the microwave.
Also the thiourea had to be bought, only for that one reaction, while the ammonium acetate was already there.

Anyways, I'll try with urea the next time, still have some intermediate and plan to make some more too, so I can try the one or other glutarimide preparation, hopefully with better results :)

Tsjerk - 27-3-2020 at 06:20

Quote: Originally posted by karlos³  
@Tsjerk, Thanks for the suggestion with the microwave method, I just haven't chosen this route because they used a sealed vessel, this was the reason not to choose this method.
I assumed it is something special similar to a bomb for high pressurized reactions, able to withstand quite some pressure and this was not something I would like to do in the homelab.
But you have simply made a premixed solution of the reagents, evaporated that dry and nuked it in the microwave and nothing more? That is impressive and something I would have chosen too.
I am sure that the method of your baclofen synthesis will be better than what I did here...
Just like this: http://www.sciencemadness.org/talk/viewthread.php?tid=104964...
Or do you any better and more specific suggestions?
If you could give me some tips besides that I would be very grateful!


I never noticed they used any kind of pressure vessel. I just put everything in a beaker, with a bigger beaker upside down over it, and nuke it as you say. It takes a while to get started if the material is too dry, maybe a drop of water would help. After that, you will get a disgusting brown boiling mass, wait until it stops boiling. As in my earlier write up linked above I used hot 33% ethanol to dissolve everything before it solidifies... If it solidifies you will have to break the beaker or have a lot of patience. In my case the product dissolves in hot diluted ethanol, while the gummy side-product doesn't. The 33% ethanol will still boil and splash as you add it but it will make a sort of a foam of your mess. Now stir hard, hot filter and you will have a clear solution that crystallizes upon cooling.

The microwave step is a bit dirty, it fumes of NH3 like mad and the up-side down beaker is definitely a recommendation if you don't want to clean all the inside of your microwave. Now you only have to clean the bottom plate.

I think this reaction is quite endothermic, it doesn't go very fast and seems to require quite some microwave power, possibly because of the water that has to evaporate. It is just a sticky mass that splashes. I got far better results when first dissolving the stoichiometric amounts (little excess urea) of reactants in water and then boil that to dryness than by mixing the two and use that directly.


Edit: and don't put more than 10 grams in a 100ml beaker.


[Edited on 27-3-2020 by Tsjerk]

karlos³ - 29-9-2020 at 10:23

I think I actually managed to ruin the microwave glutarimide formation :o
I ran that at 20mmol substrate, with 24mmol urea, and formed the mono-urea salt just as you describe.
Then I reacted that at a lower wattage though(600W) and I did so for almost six minutes.
It had stopped foaming mostly after 4min(circa) and at 5min I thought it did again, and close at 6min I realised that this isn't the reaction anymore.
Well, what can I say, I think I carbonised most of the product.

I tried then immediately to dissolve whatever is possible, and added 12ml of MeOH carefully under stirring.
But the complete bottom of the flask is full with tarry soot that didn't wanted to dissolve.
And the beaker cracked due to the temperatures, but luckily did not break.
So I filtered that red-brown MeOH and added some water to it, it is now being cooled and will hopefully show a little bit of product anyways....

But this synthesis, I will not do that again, thats for sure!
The remaining precursor will be reacted in PEG-400 with urea, and then it is over with me and the glutarimide ring formations!

I found a variation of the alternative route using glutamine, one that uses tosyl chloride for the cyclisation, and thats a perfect alternative for me, I will attempt this.
AvBaeyer was right, and I should have turned my brain on earlier to think what suitable reagents can catalyse the cyclisation else.
Here is it: https://patents.google.com/patent/US20050272934A1/en
Nice how they are running that one-pot, in example 2&3, thats definitely an attractive way for me.
I will just carefully add the tosyl chloride in little portions to the pyridine solution of the glutamine+phthalic anhdyride reaction mixture.

karlos³ - 1-11-2020 at 04:14

I've tried a little bit around, also attempted the one-pot synthesis.

What I tried was the amidation in PEG-400, using urea as nitrogen source, that worked fine and fast, the yield was after purification 28,59%.
Could have worked better, but since I pushed the temperature rising pretty strong, I accidentally got over 200° and the PEG decided to get discolored together with my product.
Nonetheless, the advantage is that its a pretty fast process.

Tried the one-pot synthesis twice, far from being optimised though.
I used to little pyridine as solvent, and both times issues when I cooled the reaction to add the tosyl chloride, it quickly became thick and viscous and resulted in hot-spots causing discoloration.
The first time I was more careful and so it had worked better, but also wasted product because I didn't heated well enough after the tosyl chloride was in the reaction, only kept it somewhat warm,
In turn, only 11,5% yield, although it was directly of an impressive white and did not require much purification, only washed it twice in ethanol and then recrystallised it from EtOH/DMF with a fitting melting point.
The second trial one-pot was even worse, nothing to say about it, discolored and I haven't even calculated the yield, not to add further frustration.

A friend has attempted this method likewise, but he used acetic anhydride instead of tosyl chloride, with a similar outcome(including discoloration and not really an optimal yield).

Another attempt according to the attached document and the following patent: https://patents.google.com/patent/WO2009083724A1/en
has been made.

The attached paper claims a good yield of N-phthaloyl L-glutamine by the reaction of phthalic anhydride and glutamine in DMF, it sure looks good.
Although I couldn't replicate that outcome, actually, it was even far from it... and I wasted my last 17g of phthalic anhydride on it, yet still don't know what happened.

But the good news is, that I tried the cyclisation with Ac2O nonetheless on the, and yes that was all of it, 1,9g of the phthaloyl precursor.
I reacted it in a solution of 10ml DMF(5 would have sufficed) with 2ml of acetic anhydride, a 3,2eq. to the N-phthaloylglutamine, for 3h and kept it at 110-120°C the whole time.

When it had cooled down it was put in a bath of cold water, then I've used a 25% solution of ethanol in water and added it to the stirred mixture, at RT, in total 40ml but already after half of it had been added it became turbid and the first thalidomide crystals show up on the walls from the stirring.
After all of it had been added, I stirred an hour further and cooled it then down much.
Few hours later it was filtered and washed in the filter with lots of cold water, and then dried.
The next day 1,372mg were the result, or 5,31mmol corresponding to a yield of 77,3% of the theory.
The appearance was very impressive I had to show a picture to some friends because it looked was just so clean and white.
Today I took a melting point and it came out at 269-270°, and that without even doing a single recrystallisation, unbelievable that only water was used to purify and wash everything unwanted out of the product :o
Although, for a substance that barely fits a half gram into a liter of water at RT, one can probably clean this stuff up very well with only water if the purity was high enough to begin with.

There is still room for improvement left, but so far it is quite a lot of fun.
Luckily, I could enjoy that a friend was joining in with a likewise large interest in this substance and its preparation :)
Together we put much research and experiments in it and tried several approaches using both glutamine and glutamic acid with phthalic anhydride from various patents and papers.
And I can proudly say, so far we probably found out almost everything that does not work as claimed :D
But not only because of this, but also because of the occasional satisfying result it was a very entertaining experience.
I really enjoyed the exchange of ideas and findings, then getting back to the project and later on present each other the results followed by speculating how that came to be :D

Attachment: 37(234-237) JMCT11.pdf (175kB)
This file has been downloaded 632 times

Metacelsus - 1-11-2020 at 04:51

Thalidomide is pretty interesting biologically, since its effects are due to altering the specificity of E3 ubiquitin ligases, thus causing certain proteins to be degraded. This can be exploited: if you link a thalidomide derivative with a compound that binds a protein, the target protein will be ubiquitylated and degraded. These chimeric compounds are called PROTACs.

karlos³ - 1-11-2020 at 17:18

I can just say that the sheer quantity of articles about this substance is really impressively large.
That has definitely been very helpful to prepare it.
Its pharmacology is most interesting as well, again, making it probably one of the best researched substances existing.
At least, as I have read, it is the one holding the record for the most publications published about it.

But I have to admit my interest in it stems from another pharmacologically unique property of it, as in the attached paper.
In that respect it turned out to be outstanding, and I'm not repeating what that paper claims but confirming it.

It is in general a rather strange substance with its effects, sharing most of the properties of related substances, while on the other hand being seemingly free from the disadvantages that are ubiquitous in almost every other compound with the same, or better, a similar mode of action.
Definitely another remarkable and interesting property of it.

Attachment: 90dd4916f4a4d6eb3d2a9116a8a2f939.pdf (52kB)
This file has been downloaded 652 times

Fery - 1-11-2020 at 20:57

Hi karlos³, thx for sharing the doc. These sleep phases are very important. A lot of psychiatric substances damage them even those used for curing sleeping disorders. Agomelatine is an exception as it is similar to melatonin, but agomelatine could damage liver. There is very likely no remedy without adverse effect.

Tsjerk - 14-1-2021 at 09:39

I prepared N-Phthaloyl-DL-glutamic acid without solvent. Only water and HCl were used in the workup. Yield: 54%

Also I got the microwave preperation of thalidomide to work, but I'm still working on optimizing this. For now I screwed up one run by burning it like karlos, and in a second run I got a (crude) 28% yield. When I was preparing new phthaloyl glutamate I screwed up because my glutamic acid was not dry enough.

Billman and Harting showed phthaloyl glutamate could be made without a solvent at 185oC in 1948.

Attachment: billman1948.pdf (223kB)
This file has been downloaded 556 times

A couple of years later, Sheehan and Bolhofer showed this reaction also proceeds at 145oC, but they do not isolate phthaloyl glutamate and go straight for the anhydride.

Attachment: sheehan1950.pdf (509kB)
This file has been downloaded 541 times

In 1956 Tipson shows the reaction run at 145oC (although run in a Dean Stark with xylene) gives a separable mixture.

Attachment: tipson1956.pdf (484kB)
This file has been downloaded 557 times


In 2001 there was a publication on the microwave assisted synthesis of thalidomide with phthaloyl glutamate and urea or thiourea. Thiourea is supposed to give yields up to 85%. In a one-pot reaction phthalic anhydride, glutamic acid and thiourea are supposed to give a 60% yield of thalidomide.

Attachment: seijas2001.pdf (31kB)
This file has been downloaded 711 times


Phthaloyl glutamate: Phthalic anhydride (10 gr, 68 mmol) and glutamic acid (10 gr, 68 mmol) were mixed and finely ground with mortar and pestle. The powder was loaded in a 100 ml beaker, covered with aluminium foil, and placed in a 145oC oil bath. After five minutes the reaction mixture begins to melt and after a few more minutes starts to bubble. The bubbles are water which is generated in the reaction. Magnetic stirring was used to keep the foam down to prevent the mixture from boiling over. Manual stirring would be impractical as the phthalic anhydride sublimes.
After 25 minutes the last white solid particles react, and the bubbling ceases. At this point the oil bath is removed and while stirring, water and 5 ml of 30% HCl* is added to a total volume of 75 ml. The reaction is heated until all solids dissolve. The mixture is cooled to 4oC and the precipitated product is filtered and rinsed with cold water. After recrystallization from water (50 ml) the product is filtered over vacuum and dried in an oven at 150oC. When almost dry, the powder is cooled and powdered with mortar and pestle and dried at 150oC for another hour. This final step makes there is no residual phthalic acid left, as this would sublime. No weight loss was noticed, but maybe the phthalic acid already sublimed in the first round. Yield: 10.17 gr, 54%.

*Glutamic acid left in the reaction mixture is very soluble as the hydrochloride.


Preliminary:
Thalidomide was made by dissolving 8,5 gr phthaloyl glutamate and 1.93 gr (1.05 eq.) in 25 ml of hot water and boiling to almost dryness (monitored by weighing the beaker) in a 100 ml beaker. The beaker was placed in an oven at 100o until dry. A 400 ml beaker was then placed upside down over the smaller beaker and the microwave function was then turned on at 500 Watt. The phthaloyl glutamate mono urea salt starts to boil gives off water and ammonia fumes. Care must be taken not to let the mixture get too hot. When it becomes orange it is enough. This was done by doing short microwave bursts of 10-30 seconds. Methanol was added to the still hot mixture which makes it stirrable. The reaction was heated to a boil and stirred until the crystals are almost colorless. This is filtered and rinsed with a little cold methanol. Yield: 2.2 gr, 28%

[Edited on 14-1-2021 by Tsjerk]

DocX - 1-1-2022 at 06:01

Quote: Originally posted by Tsjerk  

Thalidomide was made by dissolving 8,5 gr phthaloyl glutamate and 1.93 gr (1.05 eq.) in 25 ml of hot water
[Edited on 14-1-2021 by Tsjerk]


Sorry, 1.93 gr what? Is there a word missing there?

SuperOxide - 1-1-2022 at 08:17

Quote: Originally posted by DocX  
Quote: Originally posted by Tsjerk  

Thalidomide was made by dissolving 8,5 gr phthaloyl glutamate and 1.93 gr (1.05 eq.) in 25 ml of hot water
[Edited on 14-1-2021 by Tsjerk]


Sorry, 1.93 gr what? Is there a word missing there?


I think he was saying:
Quote:
... was made by dissolving 8,5 grams phthaloyl glutamate and 1.93 grams (1.05 eq.) in 25 ml of hot water...


Maybe the 8,5 was supposed to be 8.5? Still not sure it makes sense, lol.

Tsjerk - 1-1-2022 at 08:54

Ah, yes. 8,5 gr isn't correct English. It is Dutch, 1 gr is 1 gram as 8,5 gram is 8.5 grams. The 1.05 eq is urea. Thiourea is supposed to work better.

The comma/dot difference is a bitch in Excel, making for example queries and CSV files just that little bit more complicated to use.

SuperOxide - 1-1-2022 at 08:56

Quote: Originally posted by Tsjerk  

The comma/dot difference is a bitch in Excel, making for example queries and CSV files just that little bit more complicated to use.
I had a feeling that's what it was, lol. No worries. Thanks for clarifying.

Syn the Sizer - 2-1-2022 at 00:55

This is a reaction I could safely do in my current living situation. I have been minimizing my chemistry, but this reaction is simple and I have all the reagents on hand to do it.

This actually sounds like a fun synth.

Syn'

DocX - 8-1-2022 at 01:47

Quote: Originally posted by SuperOxide  
Quote: Originally posted by DocX  
Quote: Originally posted by Tsjerk  

Thalidomide was made by dissolving 8,5 gr phthaloyl glutamate and 1.93 gr (1.05 eq.) in 25 ml of hot water
[Edited on 14-1-2021 by Tsjerk]


Sorry, 1.93 gr what? Is there a word missing there?


I think he was saying:
Quote:
... was made by dissolving 8,5 grams phthaloyl glutamate and 1.93 grams (1.05 eq.) in 25 ml of hot water...


Maybe the 8,5 was supposed to be 8.5? Still not sure it makes sense, lol.


No, the abbreviation gr is not a problem, neither is the use of commas instead of periods: the left-out reagent is more of a problem :). Apparently it's urea, but it's not mentioned in the text which is kind of strange. Maybe want to edit that in? Or am I missing something obvious where you are supposed to know intuitively that it's urea?

Tsjerk - 8-1-2022 at 08:53

I know you were just asking about what the 1,93 gr was, as I indeed forgot to put it in the text.

I would edit, but that is only possible for a day or so, and the post is almost a year old.

DocX - 8-1-2022 at 14:56



[Edited on 20222222/1/8 by DocX]

DocX - 8-1-2022 at 14:58

Quote: Originally posted by Tsjerk  
I know you were just asking about what the 1,93 gr was, as I indeed forgot to put it in the text.

I would edit, but that is only possible for a day or so, and the post is almost a year old.


Haha, ok, sorry. Yes, and you clarified in in the later post.

First off though: thank you and Karlos for the excellent write-ups and interesting literature!

This thread leaves me, as a noob, a little confused though. I have quite a bit of phtalic anhydride around, and I'd love to try this out. I've read all the patents and write-ups, and I must confess to being a little bit bewildered about the starting materials.
In some l-glutamic acid is the precursor, in some others l-glutamine. Glutamine is much more accessible to me, but I can't fully wrap my head around the consequence of using that instead of glutamate. What reagents are required in the formation of thalidomide if one uses glutamine?

Thiourea and urea are both accessible to me, but acetic anhydride or thionyl chloride (there's another source of confusion for me: Karlos keep mentioning tosyl chloride as the reagent, but in the reference he gives the reagent is thionyl chloride) are not.

I hope someone a lot more savvy than me can help me out by a bit of ELI5.

[Edited on 20222222/1/9 by DocX]

karlos³ - 8-1-2022 at 19:00

With glutamic acid, you will get the intermediate N-phthaloylglutamic acid, which is cyclized thermally and in lower yield and purity with urea or thiourea.
It will simpyl be not as clean, but it is a very robust reaction that works fine.

With glutamine, you get N-phthaloylglutamine which already has a nitrogen where you want to put one at last in the former case.
So you just need a reagent that is able to close the ring, and for that an acid chloride, or anhydride like the acetic anhydride, is used.
Lately I started to use acetyl chloride instead, you want to keep the temperature lower for that and keep it an hour or so at that before you start heating it above the bp of the stuff, but then it will have reacted and you can continue and close the ring by heating.
One can also, although costly, use peptide coupling reagents like CDI for this reaction.

So in short, the former is pretty robust but as not clean and comparably low yield.
The second suffers from their reagents availability for the last step.
But, it provides one with very clean, snow-white product in good yields.
If done right, that is.
It does not even need further purification.

Tosyl chloride is mentioned as a possible alternative in these usual useless patent listings at the start, among their claims I believe.
Thats why I used that.
It did not work so well though, but I believe that has other reasons, for example my idea to try it one-pot.

ELI5? Ok, searching.... ah!
Was that sufficiently ELI5?

It helps if you simply draw the molecules out next to each other.

DocX - 9-1-2022 at 02:56

Quote: Originally posted by karlos³  
With glutamic acid, you will get the intermediate N-phthaloylglutamic acid, which is cyclized thermally and in lower yield and purity with urea or thiourea.
It will simpyl be not as clean, but it is a very robust reaction that works fine.

With glutamine, you get N-phthaloylglutamine which already has a nitrogen where you want to put one at last in the former case.
So you just need a reagent that is able to close the ring, and for that an acid chloride, or anhydride like the acetic anhydride, is used.
Lately I started to use acetyl chloride instead, you want to keep the temperature lower for that and keep it an hour or so at that before you start heating it above the bp of the stuff, but then it will have reacted and you can continue and close the ring by heating.
One can also, although costly, use peptide coupling reagents like CDI for this reaction.

So in short, the former is pretty robust but as not clean and comparably low yield.
The second suffers from their reagents availability for the last step.
But, it provides one with very clean, snow-white product in good yields.
If done right, that is.
It does not even need further purification.

Tosyl chloride is mentioned as a possible alternative in these usual useless patent listings at the start, among their claims I believe.
Thats why I used that.
It did not work so well though, but I believe that has other reasons, for example my idea to try it one-pot.

ELI5? Ok, searching.... ah!
Was that sufficiently ELI5?

It helps if you simply draw the molecules out next to each other.


That was EXEMPLARY! Naturally, the information was there all the time in your previous posts, and this was what I had more or less figured already. But I got lost in the plethora of patents and trials, being a Bear of Very Little Brain. Or at least trying out another new field of knowledge. In which case "ELI5" is a fantastic search term.

So I'll think I need to go for the suboptimal first version, due to, as you say, the availability and safety of the reagents. Phtalic anhydride is the least problem, so I can live with a lower yield. I found a great vendor for it based in Ukraine who sells it dirt cheap.

Thank you for taking the time to set me straight!

Tsjerk - 9-1-2022 at 10:49

Definitely get some DMF to recrystallize when you go the glutamic acid route.

[Edited on 9-1-2022 by Tsjerk]

DocX - 9-1-2022 at 13:43

Yes, there is an ongoing research for the needed reagents. I live in one of the most repressive countries on the planet when it comes to odd hobbies of suspicious nature, so it will take a while.

DocX - 22-10-2022 at 13:22

So finally I got around to do this.

I thought I'd combine the two methods in the write-ups above with the best yields: pyridine as solvent in the first step (with glutamic acid and phtalic anhydride), and thiourea + DMF in the second.

I purchased the first reagent, phtalic anhydride, from a seller in Ukraine before the war. The glutamic acid I made myself from commercial Monosodium Glutamate. I also made the pyridine through decarboxylation of niacin with copper carbonate as catalyst. The DMF and the thiourea were both purchased from one of the excellent vendors on this site.

1. 14.95 grams of phtalic anhydride in the form of flaky crystals was mixed with the same amount of glutamic acid in a 3-necked RB fitted with a thermometer and a reflux column. 100 ml of pyridine was added.
And already here I messed up. I had measured up 100 ml of conc HCl and 150 ml of dH2O to use later in the reaction, and in a moment of confusion I added them both to the flask. I don't want to talk about it.
White smoke was generated when the HCl was added, and the flask was put into a heating mantle in a fume hood.
Temp was raised to 115 C and kept there for 100 mins. When the heat was turned off and the temperature began to decrease, white precipitate was formed in the solution. PH was checked: 2.
The flask was then kept on an ice bath with stirring for 2 hours, before the precipitate was vacuum filtered and dried.
Yield: 17.3 g of white, sparkly crystals, which should be like 64% of theory. Melting point 192-194, to my utter joy.

2. The intermediate N-phtaloyl glutamic acid from above was mixed with 15.3 grams of thiourea in the same 3-necked flask.
50 ml of DMF was added.
The reaction mixture was heated to 167 c and refluxed at this temperature for 1 hour. The smell of sulphur from the thiourea mixed with the smell of DMF degrading to dimethylamine was really something.
Heating was turned off, and at 110 C 60 ml:s of dH2O was added and refluxed for another 30 minutes.
After cooling, the flask was transferred to an ice bath and kept there under stirring for another 30 minutes, after which the flask was refrigirated for about an hour.
The greyish-pink precipitate was vacuum filtered and washed with 60 ml of dH2O, then refluxed with 60 ml ethanol for about 20 minutes.
Filtered again, wsshed again with ethanol and water, and then dried. It was now perfectly white and sparkly.
Yield: 12.5 grams.

So I ran the first reaction in a strongly acidic solution by mistake and I was afraid I had ruined it, but the melting point checked out. The second reaction was textbook though, except for not recrystallising from DMF. The resulting sparkly, free-flowing pristine white powder seemed clean enough though.

Bioassay was made with 400 mg:s but gave no noticeable effect. So maybe I messed it up anyway?

DocX - 24-10-2022 at 07:55

My take on the failure to produce a bioactive end product is that I'm actually looking at a crystal mix predominately consisting of phtalamide. I performed a recrystallisation from DMF which gave me 9.5 g:s in return, but bioassay failed again.

The melting point of phtlamide is like for thalidomide a good bit over 200 C (238 for phtalamide, 270 for thalidomide). I haven't tried the melting point yet. It's a bit of a hassle for me to do that with these temperatures since I don't have a melting point apparatus yet. But I'll get around to it.

I'm planning a second recrystallisation from DMF, and this time basify the solution even more with ammonia to dissolve the phtalamide. We'll see what my yield will be from that. Probably catastrophic, but right now I just want to get something functional out of this.

If that doesn't work, I have to run with the suspicion that my self-synthesised glutamic acid is faulty somehow. Possibly also the pyridine. In that case, I will revisit this synth with confirmed, lab grade reagents.

I will make this work.

Also, out of curiosity: how many of you who tried this reaction did a bioassay or a melting point determination of your final yield?

[Edited on 20222222/10/25 by DocX]

Mateo_swe - 15-11-2022 at 09:26

I will eventually try making some thalidomide, i have all needed reagents except the glutamic acid or glutamine.
But i do have some food-grade monosodium glutamate from the asian store.
I could try either the lower yielding first method or the second method using acetic anhydride or thionyl chloride.

I like DocX´s method above, did you try it again and got it to work?

DocX - 18-11-2022 at 12:21

No, I haven't gotten round to it yet. I'm waiting for lab grade reagents to try and exclude all the possible impure or faulty ones.

I did try to clean the product I have several times through recrystallisation from DMF, but to no avail.

Also, next time I'll go back to the original patent method using ammonium acetate as a nitrogen donor. I figure it's best to go to the absolute basic and then work my way from there.

Pyridine + phtalic anhydride + L-glutamic acid in 115 C.
Then Phtaloyl Glutamate + ammonium acetate in PEG400 and a bit of DMF at 160-180 C to get the product. I'll probably run that reaction really long also, to get good conversion.

As soon as I get something functional out of this I can start elaborating for yield.

This would be an excellent point in life to start meddling with TLC, I guess.

[Edited on 20222222/11/19 by DocX]

Preparation of N,N-phthaloylglutamic acid

SplendidAcylation - 21-1-2023 at 05:48

I have attempted the solvent-free procedure detailed by Tsjerk, I have been successful in the preparation of thalidomide, although my yields were relatively low.

I have completed a write-up detailing the preparation of the N,N-phthaloylglutamic acid intermediate, it is attached as a PDF.

My yields were consistent with the 54% reported by Tsjerk.

The write-up for the preparation of thalidomide is on its way :)

Please let me know if I've made any mistakes, or anything, all comments welcome!

Thanks for reading :)

Attachment: Preparation of N,N-phthaloylglutamic acid by the solvent-free reaction of phthalic anhydride and L-glutamic acid.pdf (52kB)
This file has been downloaded 280 times

[Edited on 21-1-2023 by SplendidAcylation]

Mateo_swe - 23-1-2023 at 12:50

Very nicely done, SplendidAcylation, good work!

And a good writeup as well with repeated experiments.
Only thing i missed was the final step, you should have included it in the writeup.
But anyway, very nice.

Did you use thiourea in the microwave for the last step as in the paper "Microwave Promoted Synthesis of a Rehabilitated Drug: Thalidomide"?
Also in the paper they use this workup:
The crude RM after microwave irradiation was dissolved in THF and purified by column chromatography on silica gel (THF-hexane, 1:1) to afford a solid product of (±)-thalidomide (Yield 85%).

Did you use same workup with the column or some other way?
Also, are a fumehood required at this small scale?

I will try repeat your experiment and maybe increase the scale somewhat.
I really like to try a microwave for a reaction and this is the perfect experiment.

Attachment: Microwave Promoted Synthesis of a Rehabilitated Drug - Thalidomide - seijas2001.pdf (31kB)
This file has been downloaded 229 times

[Edited on 2023-1-23 by Mateo_swe]

SplendidAcylation - 24-1-2023 at 03:36

Quote: Originally posted by Mateo_swe  
Very nicely done, SplendidAcylation, good work!

And a good writeup as well with repeated experiments.
Only thing i missed was the final step, you should have included it in the writeup.
But anyway, very nice.

Did you use thiourea in the microwave for the last step as in the paper "Microwave Promoted Synthesis of a Rehabilitated Drug: Thalidomide"?
Also in the paper they use this workup:
The crude RM after microwave irradiation was dissolved in THF and purified by column chromatography on silica gel (THF-hexane, 1:1) to afford a solid product of (±)-thalidomide (Yield 85%).

Did you use same workup with the column or some other way?
Also, are a fumehood required at this small scale?

I will try repeat your experiment and maybe increase the scale somewhat.
I really like to try a microwave for a reaction and this is the perfect experiment.



[Edited on 2023-1-23 by Mateo_swe]


Note: Since I wrote this message, I have completed the write-up, however, for continuity, I shall leave the message here anyway.

Thanks!
By " final step" do you mean the microwave synthesis of thalidomide?
I am doing a separate write-up on that, it isn't quite finished yet!

I shall just answer your questions quickly because you will see all the details in the upcoming write-up:

I used urea.
I didn't use the column, I think the product I obtained is fairly pure though.

You definitely don't need a fume-hood, you won't even notice the smell of ammonia at this scale, unless you put your nose up to the test-tube.

Edit: The thalidomide write-up is now complete! It is attached as a PDF.



[Edited on 24-1-2023 by SplendidAcylation]

Attachment: Microwave synthesis of thalidomide.pdf (61kB)
This file has been downloaded 292 times

Mateo_swe - 25-1-2023 at 12:54

Very nice!

OK, you used urea and not thiourea.
Thiourea is said to give better yields, if you can get some.
Regarding the question if a fumehood is needed, i was thinking about thiourea possible could make it smell a bit, especially if pyridine solvent were to be tried.
I think i can try this without a fumehood even if thiourea is used, its so small scale.
And i can make a quickfix fume extractor.

I wonder a bit about your yields though, i dont get how you calculated.
In the paper you write 35.5%, 20.0% and 26.7% yields of thalidomide for experiment 1, 2 and 3.
Are those yields based on the N-phthaloylglutamic acid used?
You get more final product in last experiment 3 using less N-phthaloylglutamic acid than in experiment 1 but still get lower yield?

I get
Exp 1
3.9mmol N-phthaloylglutamic acid used, 0.31g(1.2mmol) thalidomide, 1.2/3.9=30.8% yield

Exp 2
3.7mmol N-phthaloylglutamic acid used, 0.25g(0.97mmol) thalidomide, 0.97/3.7=26.2% yield

Exp 1
3.4mmol N-phthaloylglutamic acid used, 0.45g(1.7mmol) thalidomide, 1.7/3.4=51.25% yield

Or do i calculate wrong?
Im a beginner at chemistry so that is a likely explanation.
I will add my experiment here also when its done.
Good work SA

Edit:
Ok i see i calculated on the crude yield, thats why i got better yields.
However, searching for thalidomide solubility i find
Sparingly soluble in methanol, ethanol, acetone, ethyl acetate, butyl acetate, glacial acetic acid.
Very soluble in DMF, dioxane, pyridine.
Practically insoluble in ether, chloroform, benzene.

Sparringly soluble in other alcoholes (no mention of iPrOH though) makes me wonder if another solvent would have been better to use for the purification step.
Maybee you got better yields than you think.

[Edited on 2023-1-25 by Mateo_swe]

[Edited on 2023-1-25 by Mateo_swe]

SplendidAcylation - 2-2-2023 at 03:48

Quote: Originally posted by Mateo_swe  
Very nice!

OK, you used urea and not thiourea.
Thiourea is said to give better yields, if you can get some.
Regarding the question if a fumehood is needed, i was thinking about thiourea possible could make it smell a bit, especially if pyridine solvent were to be tried.
I think i can try this without a fumehood even if thiourea is used, its so small scale.
And i can make a quickfix fume extractor.

I wonder a bit about your yields though, i dont get how you calculated.
In the paper you write 35.5%, 20.0% and 26.7% yields of thalidomide for experiment 1, 2 and 3.
Are those yields based on the N-phthaloylglutamic acid used?
You get more final product in last experiment 3 using less N-phthaloylglutamic acid than in experiment 1 but still get lower yield?

I get
Exp 1
3.9mmol N-phthaloylglutamic acid used, 0.31g(1.2mmol) thalidomide, 1.2/3.9=30.8% yield

Exp 2
3.7mmol N-phthaloylglutamic acid used, 0.25g(0.97mmol) thalidomide, 0.97/3.7=26.2% yield

Exp 1
3.4mmol N-phthaloylglutamic acid used, 0.45g(1.7mmol) thalidomide, 1.7/3.4=51.25% yield

Or do i calculate wrong?
Im a beginner at chemistry so that is a likely explanation.
I will add my experiment here also when its done.
Good work SA

Edit:
Ok i see i calculated on the crude yield, thats why i got better yields.
However, searching for thalidomide solubility i find
Sparingly soluble in methanol, ethanol, acetone, ethyl acetate, butyl acetate, glacial acetic acid.
Very soluble in DMF, dioxane, pyridine.
Practically insoluble in ether, chloroform, benzene.

Sparringly soluble in other alcoholes (no mention of iPrOH though) makes me wonder if another solvent would have been better to use for the purification step.
Maybee you got better yields than you think.

[Edited on 2023-1-25 by Mateo_swe]

[Edited on 2023-1-25 by Mateo_swe]



:)

Yes, although the yields using urea are supposed to be 63% which is still six times better than my yields! :(

Regarding the solubility in alcohol, yeah, there is a possibility that the isopropanol dissolved some of the product, however, I don't think so, because:

"These were filtered from suspension, washed twice with 2mL cold water, and dried, yielding 0.25g of off-white powder.
The crude product was extracted with 4g of boiling isopropanol; Once dried, the undissolved product weighed only ~50mg.
A second extraction with boiling isopropanol removed no further soluble impurities, as the undissolved product after the second extraction, upon drying, still weighed ~50mg."


If the product were significantly soluble in isopropanol, the weight would have decreased from 50mg rather than remain the same.

Of course, my scales aren't 100% accurate, so it's possible the weight might have dropped slightly! :)

I have been thinking a bit about ways to improve the yield...

It occurred to me that maybe a catalyst could be used to improve the yields; Boric acid is known to catalyze the amidation of carboxylic acids in a urea melt, which is essentially what we are doing here in this reaction, so maybe boric acid would act as a catalyst here too?

Who knows.

dicyanin - 28-2-2024 at 16:44

Attached is a paper that reports a facile small size (~2-5 grams scale depending on substrate) microwave synthesis of a variety of N-phtaloylamino acids in good yields.

They prepare an intimate mixture of 10 mmol amino acid and 10 mmol phtalic anhydride in a 100 ml beaker and wet it by adding a small amount of glacial acetic acid. They cover the beaker with a watch glass and proceed to heat at 1000W for 15-20 minutes in an ordinary household microwave oven (they used a Panasonic model NN-CD987W). After cooling and adding some methanol, the crude product was filtered, dried and recrystallised from ethanol. They report yields ranging from 81-100%. Substrates used were glycine, alanine, beta-alanine, L- and DL-valine, 2-aminoisobutyric acid, S-methyl-L-cysteine, phenylalanine, tyrosine, anthranilic acid, PABA and 4-aminophenylacetic acid.

Given what we know already, glutamic acid should perform in a similar fashion.

Attachment: facile microwave method for preparation of N-phtaloylamino acids__al-hazimi2012.pdf (519kB)
This file has been downloaded 131 times

recrystallisation and resolution of thalidomide polymorphs

dicyanin - 13-3-2024 at 03:10

In crystallography, polymorphism describes the phenomenon where a compound or element can crystallize into more than one crystal structure. Racemic thalidomide is dimorphic, the existence of two polymorphic phases is discussed in the following papers :

(The existance of a third form was alluded to in the Caira et al. 1994 paper, but later retracted.)

The two forms are characterised in terms of their different X-ray diffraction patterns, infrared spectra and intrinsic dissolution properties. The more recent of these three papers, "Solid state evaluation of some thalidomide raw materials" (Carini et al. 2009) investigated six commercially available sources of thalidomide and concluded that there was a lack of homogeneity due to differences related to crystalline constitution, crystal habit and intrinsic dissolution rate of the samples analysed, suggesting that none of the known processes used to produce thalidomide are capable of producing a pure polymorph.
polymorph_packing.png - 95kB

It is well known that physical properties such as dissolution behavior can affect bioavailability, which can affect the required dose of a formulation. It is important that pure, stable, crystalline forms are used in pharmaceutical preparations as conversion from a form showing greater dissolution and better bioavailability to a less soluble form can potentially have disastrous consequences. The phenomenon is well-known with racemic modafinil, where no less than six different anhydrous polymorphic phases exist (of which four are stable), in addition to a monohydrate.

The attached patent WO2011154739 includes two purification methods that can resolve the crystalline phases of racemic thalidomide in either its alpha- or beta-form with high polymorphic purity through a solvation-desolvation process.

α-thalidomide
10 g of crude thalidomide was dissolved in 50 ml DMSO with gentle heat. This solution was slowly added, with good stirring, to 170 ml methanol at a temperature of 45-50°C. The mixture was kept at that temperature with stirring for a further 50 minutes, then slowly cooled to 40°C and filtered. The solids were washed with methanol and collected through vacuum filtration. Finally the solid pure product was dried to a constant weight under mild vacuum (~100 mmHg) at a temperature of 50-55°C to give the alpha-form of thalidomide as white to off-white solid.
Yield = 8-8.5 g (~85%)

β-thalidomide
10 g crude thalidomide was dissolved in 60 ml DMF and heated at 70-75°C for 2 hours. The solvent was removed by distillation under reduced pressure (80-100 mmHg) at a temperature of 65-70°C. To the residual mass, acetone was added to produce a slurry, which was stirred for 2 hours. The slurry was then filtered and washed with additional fresh acetone. The solid was then dried to a constant weight under mild vacuum (80-100 mmHg) at 55-60°C to give the beta-form of thalidomide as a white to off-white solid.
Yield = 9.5 g (95%)

Attachment: thalidomide polymorphisms_WO2011154739.pdf (2MB)
This file has been downloaded 104 times

my results

dicyanin - 3-4-2024 at 11:47

L-Glutamic acid was prepared from Ajinomoto MSG, according to the procedure outlined on the OrgSyn page (note 4) with the use of activated carbon for high purity.
N-phthaloyl-DL-glutamic acid was prepared by heating an intimate mixture of phthalic anhydride and L-glutamic acid in an oil bath as outlined in the post by Tsjerk earlier in this thread.

Experimental
5.25 g N-phthaloyl-DL-glutamic acid and 1.55 g thiourea was dissolved in 20 ml warm methanol. This solution was poured into a porcelain dish and the methanol evaporated in a well-ventilated area with the aid of a hairdryer. The dish was then further dried in a stream of hot air (70°C). The solids were finely ground using mortar and pestle, and placed in a large (25 x 200 mm) borosilicate test tube, the top of which was plugged with a ball of toilet paper.

The bottom sides of the test tube were gently heated with a hot air gun until the powder at the side just started to melt. The tube was then placed in a porcelain cup in the microwave oven (600W) and heated in short bursts of 10-30 seconds. The contents started to melt, bubble, water vapour and gases evolved, which judging from the smell included ammonium sulfide. For that reason in between the short bursts of heating the test tube was taken outside, the paper wad removed, and the foul smelling volatiles released. This was repeated until the liquified contents became orange in colour.

The tube was allowed to cool down and the contents inside solidified. When still warm to the touch (but not hot) 30 ml methanol was added, which caused the solids to break up easily. After filtration, washing with cold methanol and air drying 1.60 g (33%) thalidomide was obtained.

Thoughts & remarks
The yield is indeed slightly better than when urea is used, but I don't know if that is worth dealing with the stench evolved. It did bring back pleasant memories of childhood, when we used to buy 'stink bombs' which were glass ampoules with an ammonium sulfide solution inside. Once we shattered one under our feet inside the driving bus on a school trip, I distinctly remember a raging busdriver and inadequate ventilation on the bus, we had to suffer the foul atmosphere equally as our classmates but thought that was worth the joke. :D
Anyway it's the same substance, and I suspect in about the same amounts as in such a stink bomb ampoule. On this scale I don't think it's dangerous with adequate ventilation, but it's not very pleasant either and such odor could certainly draw attention. Something to consider.

mixture_porcelain_dish.png - 741kB thalidomide_crude.png - 292kB

Recrystallisation
1.60 g crude thalidomide was dissolved in 8 ml DMSO with gentle heating. In a roundbottom flask fit with condensor, placed in a warm water bath on a hotplate/magnetic stirer, was placed 30 ml methanol and the contents were heated to 45-50°C. With good stirring the thalidomide in DMSO solution was added dropwise to the methanol, and after addition the solution obtained was left stirring at that temperature for 50 minutes.
Contents were then poured into a beaker, flask washed with 10 ml methanol and added to the beaker. On cooling to room temperature crystals started to appear. The beaker was then placed into the freezer for 2 hours, and a mass of fine needle-shaped crystals had formed, which were filtrated and washed with cold methanol.
after air-drying 0.91 g thalidomide (presumably as the alpha form) were obtained, 57% yield based on the crude thalidomide


thalidomide_recrystallised_alpha-form.png - 1.1MB


A preparation of the recrystallised thalidomide crystals was made for viewing under the microscope. The following pictures were taken at 40x, 100x and 400x magnification respectively.



thalidomide_recrystallised_alpha-form_40x.png - 674kB thalidomide_recrystallised_alpha-form_100x.png - 1.5MB thalidomide_recrystallised_alpha-form_400x.png - 1.6MB

[Edited on 3-4-2024 by dicyanin]

dicyanin - 10-4-2024 at 05:07

An intimate mixture of 5 grams phthalic anhydride and 5 grams L-glutamic acid (finely ground together in a mortar with pestle) in a large (23 x 200 mm) test-tube was heated in an oil bath at 145-150°C for 30 minutes. Here's a picture taken from the post-reaction mixture cooled to room temperature :

IMG_20240410_075517.jpg - 97kB

At the bottom the melt turned into a glassy solid, this is chiefly our product N-phthaloyl-DL-glutamic acid. There are still some white particles present through the mix, consisting of the precursor products.
Same for the white crust on the sides of the tube above that, this is because the volume of the powder mixture is about twice that of the melt.
Then above that you see large pretty needles of phthalic anhydride that sublimated from the mixture, and water droplets formed during the reaction.