Sciencemadness Discussion Board - Issues Synthesizing Saccharin from Anthranilic Acid

Issues Synthesizing Saccharin from Anthranilic Acid

SaccharinSlayer157 - 12-1-2025 at 23:56

So I’ve been attempting to synthesize saccharin from anthranilic acid using the reaction outlined in this video for the past few weeks, but I’ve run into some persistent issues that I can’t seem to resolve. Here’s the slightly scaled-down reaction scheme I’m working with:

Diazonation

Dissolve 1 g of anthranilic acid in 5 mL of 20% HCl, cool the mixture to 0 °C, and add 0.625 g of sodium nitrite (dissolved in 5 mL water) dropwise while stirring.
I keep the temperature strictly below 3 °C throughout this step. The reaction produces a pale yellow solution, which seems consistent with diazonium salt formation.

Sulfonation?

Add a solution of 1 g sodium metabisulfite in 25 mL of 5% HCl dropwise at 0 °C.
This is where things seem to go off track. Nitrogen gas evolution is supposed to occur here, but I’ve never observed any bubbling. The only noticeable change is a slight yellowing of the solution.
According to the video, the sulfonyl chloride is formed directly in this step via an aryl sulfonate intermediate. However, this source suggests that sulfonic acid is formed first and must be chlorinated separately to convert it into the sulfonyl chloride. I’m not sure which pathway is correct—or if I’m failing at this step entirely.

Cyclization

Add 1.25 g of ammonium chloride to the solution, then gradually increase the pH using NaOH to strongly basic conditions. This is supposed to induce cyclization to saccharin.
I’ve also tried directly adding aqueous ammonia to bring the solution to basic pH, but nothing seems to work.

To troubleshoot, I’ve experimented with both chlorine gas and sodium hypochlorite to convert the presumed sulfonic acid into a sulfonyl chloride after the SO₂ reaction step. However, these runs resulted in the same failure. All my attempts consistently produce red or orange solutions, which makes me think the diazonation is working, but I’m stuck after that point. I suspect the key problem lies in this step since I don’t observe nitrogen gas evolution, and I’m not sure if the sulfonyl chloride or sulfonic acid is being formed at all.

I’d appreciate any advice or insights anyone might have with this type of synthesis. Are my conditions off? Is the SO₂ addition step inherently problematic when done this way? Or am I missing something ridiculously obvious? I don't have a ton of experience with organic syntheses like this, so that's very possible

Let me know if more details or photos of my setup would help clarify things. I’ve included photos of:

The anthranilic acid solution right after diazonation and "sulfonation".
My anthranilic acid (from methyl anthranilate, twice recrystallized)
The reaction mixture after chlorination with chlorine gas.

Thanks in advance for any help!

[Edited on 13-1-2025 by SaccharinSlayer157]

bnull - 13-1-2025 at 03:45

The "improved method" is the one shown in the Wikipedia article for saccharin. Here's the figure:

650px-Maumee_synthesis_of_saccharin.png - 27kB

The method is described in reference 43, "Sweeteners" in Ullmann's Encyclopedia. The entry in Ullmann's has the following:

Notice that the reaction with sulfur dioxide is made in the presence of copper, which is absent in the Wikipedia figure. The process is given in reference 98 of Ullmann's entry. It is the patent US 2667503, "Method of Preparing Ortho Sulfonyl Chloride Benzoic Acid Esters", by Oliver F. Senn. Column 1, lines 19-25:

Quote:

The present invention is based upon a different series of reactions, which have the advantage of being cleancut, involving no difficultly separable isomers or side reactions. The resultant saccharin does not have the bitter after taste, which has previously retarded its acceptance as a flavoring material and a sugar substitute.

Sounds good, eh? Column 1, lines 26-30:

Quote:

In the practice of the invention, an ester of 2,2'-dicarboxy-diphenyl-disulfide (dithiosalicylic acid), which may be prepared as described in the application for U. S. Letters Patent by George F. Schlaudecker, Serial No. 239,007, filed concurrently herewith, is oxidized by suspending it in cold water and passing in chlorine under controlled conditions, while agitating the suspension.

Uh-oh. Here's the reaction in US 2667503:

Blimey, it's not the same as in Ullmann's or Wikipedia. Schlaudecker gave up on that application and filed another, which became US2705242, "Esterification of Dithiosalycilic Acid". Cor blimey, it has nothing on the preparation of dithiosalycilic acid. The whole thing is useless.

It is a series of errors: by Gert Lipinski, who edited the entry in Ullmann's and didn't bother to revise the references^Note; by the Wikipedia editors, who didn't bother to verify the patent given in Ullmann's and committed an error in the figure; by the guy in the video, who apparently didn't check either Ullmann's or the patent, otherwise he'd have seen it's the wrong patent or at least that there is a mistake in the figure.

Your diazotization seems fine, it is the rest that is messed up. My advice is: check the references. If you see a process and they give a reference, check it. If they don't, try to find where it came from and check it.

Note: I think the correct process is explained by one of the references between 90 to 104 of Ullmann's entry (https://onlinelibrary.wiley.com/doi/10.1002/14356007.a26_023). No, I didn't bother to check them, except for number 98. It may also be that Lipinski misplaced the correct reference. Anyone can forget where they put a paper.

Edit: Typo. Oops.

[Edited on 13-1-2025 by bnull]

chempyre235 - 13-1-2025 at 06:19

Quote:

Shouldn't mild oxidation of thiosalicylic acid yield the disulfide? If so, then you could use thiosalicylic acid, which is prepared from anthranilic acid via diazotization, followed by reaction with sodium sulfide, according to Wikipedia.

Edit: I see now that part of the synthesis referenced calls for reduction with zinc. Maybe it forms the disulfide without oxidation?

[Edited on 1/13/2025 by chempyre235]

bnull - 13-1-2025 at 06:58

The issue here is that the "improved method" is developed nowhere in either Ullmann's or the patents above, and these are associated with the wrong picture, or rather the wrong patents are associated with the correct picture (from Ullmann's). It's like having the picture of a Lancashire hotpot illustrating an apple pie recipe in a cookbook; if you make the pie and expect it to look like the hotpot, you'll be disappointed.

I found the correct patent: US4464537, "Preparation of Saccharin". From the Abstract:

Quote:

A process for the preparation of saccharin by reacting an aqueous hydrochloric acid solution of o-methylcarbonylbenzenediazonium chloride with sulfur dioxide, wherein

the aqueous diazonium salt solution is reacted with sulfur dioxide at from 0 °C to 100 °C, in the presence of a water-immiscible or only partially water-miscible inert organic solvent,
in order to decompose the diazonium salt, the reaction mixture is treated simultaneously or subsequently with a diazonium salt decomposition catalyst,
the aqueous organic reaction mixture, or the organic phase obtained after removing the aqueous phase, is treated with an oxidizing agent at from 0 °C to 100 °C, and
the organic phase is reacted with aqueous ammonia at from 0 °C to 50 °C, and the saccharin is isolated from the aqueous phase in a conventional manner by acidifying with a strong acid.

Attachment: US4464537 - Preparation of Saccharin.pdf (525kB)
This file has been downloaded 80 times

Edit: One more thing: you can use your methyl anthranilate as it is instead of making anthranilic acid from it. Methyl probably acts as a protecting group for the -COOH. If the reaction proceeds the same way with anthranilic acid is a good question.

[Edited on 13-1-2025 by bnull]

SaccharinSlayer157 - 13-1-2025 at 17:20

Wow, that's incredibly helpful! I actually enjoy digging through patents and references so I'll have no trouble finishing up what you started. I think I'll try the biphasic/copper catalyst next and report what I find, that patent looks hopeful.

Thanks again for the insights!

SaccharinSlayer157 - 25-2-2025 at 11:34

Finally, I've earned my name! Here's the write-up:

Synthesis of Saccharin (25 mmol Scale)

Reagents:
Hydrochloric Acid (31.45 wt%) – Sourced from hardware store
Methyl Anthranilate – Purchased from Mystic Moments on Amazon
Potassium Nitrite – Obtained from Wood-Finishing Enterprises (most challenging to source)
Sulfur Dioxide Source – Generated using sodium metabisulfite and H₂SO₄
Perchloroethylene (PCE) – Super Tech Parts Cleaner
Copper(II) Chloride – Prepared from old pennies
Copper(I) Chloride – Made by reducing aqueous copper(II) chloride with SO₂
Chlorine Source – Generated using TCCA and HCl
Ammonia (5 wt%) – Standard grocery store strength

1. Diazonation
In a 100 mL round-bottom flask, 25 mL of 31.45% HCl was combined with 3.8 g (3.25 mL) of methyl anthranilate. The flask was then placed in an ice bath and cooled to 0°C.

Separately, 2.15 g of potassium nitrite was dissolved in 10 mL of water. This solution was added dropwise to the cold methyl anthranilate mixture with vigorous stirring. The temperature was carefully maintained between 0–5°C during the addition, which took approximately 30 minutes.

2. Sandmeyer Sulfonation
10 mL of chilled PCE was saturated with SO₂ and then added to the cold diazonation mixture. The reaction was gradually heated to 50°C using a water bath while maintaining strong stirring.

Once the bath temperature reached 20°C, a catalyst mixture containing 200 mg of copper(II) chloride (CuCl₂) dissolved in a minimal amount of water with 50 mg of copper(I) chloride (CuCl) suspended in it was added in small portions to prevent excessive bubbling from nitrogen evolution. SO₂ was continuously bubbled through the mixture for an additional 10 minutes. After all reagents were added, the mixture was kept on heat and stirred for an additional 30 minutes.

3. Chorine oxidation
After completion of the sulfonation step, the mixture was allowed to cool to 40°C. Chlorine gas was then bubbled through the reaction mixture for 5 minutes under strong stirring, followed by an additional 5 minutes of stirring to ensure complete oxidation.

4. Final Ammonia Treatment and Precipitation
Ammonia was added to the reaction mixture under strong stirring until the aqueous layer became basic. The organic layer was separated off, and HCl was added to re-acidify the aqueous phase.

Saccharin precipitated out of the solution and was then filtered, washed thoroughly with cold water, and recrystallized twice from ethanol for purification

My yield was 1g, which is only 22% of theoretical. I’m confident that a large portion of that came from transfer losses and the two recrystallizations, I got a bit sloppy near the end of the synthesis. The tiny scale obviously didn’t help either. I didn't really bother with being very exact with the amount of gas I used since they're cheap and the scale was so small, but I imagine this would become necessary if I was to scale this reaction up.

One other thing I noted was that almost immediately upon mixing the HCl and methyl anthranilate I got a thick white precipitate. I'm assuming this was anthranilic acid from hydrolysis but I was surprised by how quickly it proceeded at RT. If this is the case then the reaction should proceed the same from AA, as whatever protection the methyl group could have provided is gone from the start.

Also, I followed this exact procedure twice before without adding copper(I) chloride to the catalyst mixture without success, because EVERY SOURCE I’ve read, including the patent bnull provided (which was beyond helpful by the way), said that copper(II) chloride would be sufficient to catalyze the diazonium decomposition reaction. I finally came across this article which states in a footnote that,

Quote:

“The original literature suggests that copper(II) chloride dihydrate can be used as a catalyst, since it is reduced by the sulfur dioxide to copper(I). It has been noted on several occasions that catalytically inactive mixtures result. If copper(II) chloride dihydrate is used, it is expedient to add copper(I) chloride (1 g) to ensure efficient catalysis in the early stages of reaction.”

I could have saved SO much time had I known this sooner :mad:

Mixture immediately after diazonation

Nitrogen evolution after addition of the copper catalyst

Mixture after chlorination

Separation after ammonia addition

Final Product

[Edited on 25-2-2025 by SaccharinSlayer157]

bnull - 25-2-2025 at 14:34

Quote: Originally posted by SaccharinSlayer157

Also, I followed this exact procedure twice before without adding copper(I) chloride to the catalyst mixture without success, because EVERY SOURCE I’ve read, including the patent bnull provided (which was beyond helpful by the way), said that copper(II) chloride would be sufficient to catalyze the diazonium decomposition reaction.

It may depend on the scale.

Good to see that it worked after all.