Sciencemadness Discussion Board - Reagents for the preparation of acid chlorides: pros and cons

Reagents for the preparation of acid chlorides: pros and cons

Melgar - 22-6-2016 at 11:45

I've done a large amount of research on what reagents are used to prepare acid chlorides, and they mostly seem to be:

Phosphorus chlorides
Thionyl chloride
Phosgene and derivatives

All of these chemicals are somewhat dangerous to synthesize, and phosphorus chlorides are pretty much out of the question. However, diphosgene seems to have some really interesting properties that make it seem like an incredibly useful reagent, while being somewhat less dangerous and easier to store than regular phosgene. Thionyl chloride seems to be the gold standard, and I've made an impure form of it before (that I never got to use) so I know what the process is like. But I'm not sure what, if any advantages there are to thionyl chloride over diphosgene. Obviously, both would necessitate the use of a fume hood. My guess is that thionyl chloride is more selective, and so is used more, which would make sense.

As far as syntheses, the reactions I would use to obtain them are both similar to other reactions I have extensive experience with, and I'd estimate them to be similarly difficult and dangerous. Any reasons I should attempt one over the other?

Thionyl chloride synthesis I used:
Gas elemental sulfur with chlorine gas until enough of it is liquid to allow bubbling. Keep doing this until the color changes to indicate presence of SCl2. Continue bubbling Cl2, but now also bubble SO2 by adding HCl to sodium metabisulfite. Alternately, use thermal decomposition of iron sulfate.

This was all done outdoors, the first time, incidentally, and any gases were bubbled into a solution of potassium iodide and sodium bicarbonate. (tests for acids and chlorine) Never got about to purifying it or determining yield, but it seemed to work.

Proposed solution for diphosgene:
Prepare methyl formate from calcium formate and HCl-saturated methanol. Distill to purify. Dissolve chlorine in methyl formate, and add a catalytic amount of bromide salt to generate bromine monochloride in situ. (Bromine monochloride will form radicals in visible light, as opposed to chlorine, which requires UV). Possibly run the HCl vapors into a hydrogen peroxide solution to generate more chlorine gas for the reaction. Again: fume hood not optional.

Another benefit of diphosgene is that it can be used to generate thionyl chloride, should I ever need it. Any major downsides I've overlooked?

[Edited on 6/22/16 by Melgar]

ziqquratu - 22-6-2016 at 15:58

Thionyl chloride is popular because it's cheap, relatively stable, fairly safe to handle, and gives only gaseous byproducts. As you note, it is pretty selective in what it reacts with, which is nice (it means you can really push the system - using neat thionyl chloride and heating it, if need be - to persuade unreactive functional groups to cooperate, without too much risk to the rest of your molecule).

Phosphorous chlorides certainly work, although they're rarely used in the literature - not least because other reagents are generally front of mind for the reaction (I reckon I could ask anyone in my lab, and phosphorous chlorides would be a long way down the list for making acid chlorides - and we use them every day for other purposes!).

The phosgenes are also effective, but they tend to be very expensive and thus little used. Also, the synthesis necessarily generates phosgene itself, which is of course a potentially significant hazard. They also decompose into phosgene, which can be a greater hazard - accidentally breathing in a lungful when you open your container! I've used di- and triphosgene a couple of times in in the past, for very specific purposes (to prepare unsymmetrical carbonates and carbamates), but have never turned to them for acid chlorides.

Personally, I've never prepared any of the above, but I'd lean towards thionyl chloride for a couple of reasons. 1) It's the more commonly used reagent, which means that you're more likely to find a procedure which uses it for the product you're looking to obtain (instead of having to adapt procedures which use other methods). 2) The synthesis is a marginally less unpleasant (no risk of phosgene - admittedly less of a concern given you're flowing large amounts of chlorine through the system either way!), but the storage is much less hazardous (HCl and SO₂ vs. phosgene when you open the bottle). 3) To my memory, the preparation of thionyl chloride has been documented in a hobbyist environment and works pretty well, whilst I've not come seen a similar synthesis of diphosgene. That could be a sign of my ignorance, of course, but if not you'd potentially be breaking new ground. If that's your goal, great - new preparations of potentially useful reagents which are adapted to non-professional settings are great! But, if you just want to use it, you may not be interested in spending time to troubleshoot all the problems that could arise, when you can just go with a relatively tried-and-true method.

Dr.Bob - 22-6-2016 at 17:34

I have never tried to make any of them, but have used all of them. You missed oxalyl chloride, which is one of the best as it often works at rooom temp, unlike POCl3 and thionyl chloride, which sometimes or often require refluxing. It is also fairly volatile, so most of the time, you can work up the reactions by merely evaporating them, which is also true of thionyl choride, but not true of POCl3, which can make a big mess upon workup. I don't know if there is a good OTC prep for oxalyl chloride, but at least oxalic acid is easy to find as wood bleach. Wikipedia claims that "It can be prepared by treating oxalic acid with phosphorus pentachloride.", so no worse to make than most other nasties.

The key to a good acid chloride synthesis is dissolve the acid in DCM, add the oxalyl chloride, and then one drop of a 1% soln. of DMF in DCM to it. That generates a catalytic cycle which makes the reaction go much faster at RT, but don't ad too much DMF or you can make too many by-products. This is a classic trick, but really works well.

Melgar - 22-6-2016 at 21:32

Quote: Originally posted by ziqquratu

Thionyl chloride is popular because it's cheap, relatively stable, fairly safe to handle, and gives only gaseous byproducts. As you note, it is pretty selective in what it reacts with, which is nice (it means you can really push the system - using neat thionyl chloride and heating it, if need be - to persuade unreactive functional groups to cooperate, without too much risk to the rest of your molecule).

Phosphorous chlorides certainly work, although they're rarely used in the literature - not least because other reagents are generally front of mind for the reaction (I reckon I could ask anyone in my lab, and phosphorous chlorides would be a long way down the list for making acid chlorides - and we use them every day for other purposes!).

I've gassed S2Cl2 with chlorine, and I've gassed esters and hydrocarbons with chlorine. S2Cl2 was definitely worse. Not to mention, it's next to impossible to get out all of the S2Cl2 unless you add the nastiest of all the nasty chemicals to work with: OLEUM! That's right, to really do this synthesis right, you really need to decompose a sulfate salt, (preferably one that gives a mix of SO2 and SO3), and believe it or not, that's better than the alternative, which is overgassing with Cl2.

According to my research, in the presence of dissolved chlorine, SO3 reacts with S2Cl2 to give every sulfur its own oxygen, and would really cut down on the SCl2 and Cl2 vapors.

Quote:

The phosgenes are also effective, but they tend to be very expensive and thus little used.

Ironic, given that burning a mixture of methanol and chloroform/MeCl2 in the right proportions can generate it so easily.

Quote:

Also, the synthesis necessarily generates phosgene itself, which is of course a potentially significant hazard. They also decompose into phosgene, which can be a greater hazard - accidentally breathing in a lungful when you open your container! I've used di- and triphosgene a couple of times in in the past, for very specific purposes (to prepare unsymmetrical carbonates and carbamates), but have never turned to them for acid chlorides.

My plan is to chlorinate methyl formate as opposed to methyl chloroformate. If nothing else, then if I can't synthesize the ester properly, then I'll know I certainly have no business messing around with phosgene. Better to start with the least dangerous reaction and work up from there.

Quote:

Personally, I've never prepared any of the above, but I'd lean towards thionyl chloride for a couple of reasons. 1) It's the more commonly used reagent, which means that you're more likely to find a procedure which uses it for the product you're looking to obtain (instead of having to adapt procedures which use other methods). 2) The synthesis is a marginally less unpleasant (no risk of phosgene - admittedly less of a concern given you're flowing large amounts of chlorine through the system either way!), but the storage is much less hazardous (HCl and SO₂ vs. phosgene when you open the bottle). 3) To my memory, the preparation of thionyl chloride has been documented in a hobbyist environment and works pretty well, whilst I've not come seen a similar synthesis of diphosgene. That could be a sign of my ignorance, of course, but if not you'd potentially be breaking new ground. If that's your goal, great - new preparations of potentially useful reagents which are adapted to non-professional settings are great! But, if you just want to use it, you may not be interested in spending time to troubleshoot all the problems that could arise, when you can just go with a relatively tried-and-true method.

Actually, I kinda want to break new ground. I actually discovered two reactions already that I posted here: bromide-catalyzed chlorination (of toluene to benzyl chloride at least) and the fact that erythritol tetranitrate will detonate when heated rapidly in the presence of aluminum. Both reactions were confirmed by other members, and I've seen both threads referenced extensively. That makes me really happy. I feel like I've contributed my part to overall scientific knowledge.

Sure, one of my threads explains how to make NPEDs, but I also proved that ETN can be reliably detonated in the presence of metal via thermal shock, something that anyone using ETN should DEFINITELY know. To my knowledge, this had not been documented before.

Quote:

I have never tried to make any of them, but have used all of them. You missed oxalyl chloride, which is one of the best as it often works at rooom temp, unlike POCl3 and thionyl chloride, which sometimes or often require refluxing. It is also fairly volatile, so most of the time, you can work up the reactions by merely evaporating them, which is also true of thionyl choride, but not true of POCl3, which can make a big mess upon workup. I don't know if there is a good OTC prep for oxalyl chloride, but at least oxalic acid is easy to find as wood bleach. Wikipedia claims that "It can be prepared by treating oxalic acid with phosphorus pentachloride.", so no worse to make than most other nasties.

You may be forgetting how hard it is for the amateur chemist to obtain elemental phosphorus.

One really nice thing about phosgene is that it can be used to MAKE thionyl chloride and oxalyl chloride. I'm half-tempted to make a methylene chloride burner, run it through a dryer, and bubble the vapors into liquid SO2 just to avoid combining all the nastiest chemicals I have experience together.

Incidentally, since you mention oxalyl chloride, I've read that the electron withdrawing effect of the benzene ring also makes benzoyl chloride a decent, mild reagent for making other acid chlorides. If this is the case, I could theoretically chlorinate toluene to benzotrichloride, then combine this with an equimolar amount of benzoic acid to generate two units of benzoyl chloride for each unit of benzotrichloride I painstakingly made; the benzoic acid could be obtained on eBay for a song.

I wonder if this could be done by chlorinating benzyl alcohol? The resulting hydrogen chloride would combine with more benzyl alcohol to make benzyl chloride, which could then be further chlorinated. I like the benzotrichloride + benzoic acid route though, if I'm going to generate an acid chloride from another acid chloride. It seems like it works on all the carboxylic acids I'm interested in, at least according to an organic chemistry text I read recently, which I'm unable to locate for some reason.

[Edited on 6/23/16 by Melgar]

byko3y - 22-6-2016 at 23:08

Quote: Originally posted by Melgar

Gas elemental sulfur with chlorine gas until enough of it is liquid to allow bubbling. Keep doing this until the color changes to indicate presence of SCl2. Continue bubbling Cl2, but now also bubble SO2 by adding HCl to sodium metabisulfite. Alternately, use thermal decomposition of iron sulfate

I don't believe it works. The reaction is too slow at normal conditions to be usefull, and also gives a complex mixture of many different products.

clearly_not_atara - 22-6-2016 at 23:18

I know SO2Cl2/HOBt (benzotriazole) works for alcohols, but has it ever been tried for carboxylates? It seems like the mechanism would generalize: it forms a 1-benzotriazolesulfonate ester, which then attacks chloride to release a stable zwitterion.

https://www.researchgate.net/profile/Dr_Chandra_Azad/publica...

Oh! Also, I should point out, that during the last thread where someone was dumb enough to make phosgene, IIRC we found a reference saying something to the effect that if you want to make diphosgene from methyl formate you have to do the first chlorination selectively to methyl chloroformate; apparently just replacing methyl chloroformate with the formate in the OrgSyn tek can cause a "boom". Also, IIRC, the OP of that thread was never heard from again.

[Edited on 23-6-2016 by clearly_not_atara]

[Edited on 23-6-2016 by clearly_not_atara]

Eddygp - 23-6-2016 at 00:05

As others have already clearly stated, thionyl chloride is the way to go: safer than the alternatives, more commonly used, etc.

Praxichys - 23-6-2016 at 04:44

Cyanuric trichloride is a good one that nobody has mentioned. It is almost as versatile as oxalyl chloride, as long as you don't mind filtering cyanuric acid.

Thionyl chloride isn't too hard to make if you can get your hands on P2O5. Some pyrotechnic suppliers are selling it now, and I can attest that it is the legitimate stuff.

Here is my video on S2Cl2:
https://www.youtube.com/watch?v=oYw89ieffa4

After this, just gas it with more Cl2 until a prescribed weight gain. Set up a short-path distillation rig with a 2 neck boiling flask that you no longer like, as it will be etched. Drip cold concentrated H2SO4 onto powdered P2O5 in the boiling flask and then gently heat, making sure the distilling SO3 doesn't clog the short path. Once the SO3 is collected and weighed, the correct (excess SO3) amount of SCl2 is weighed out and the two flasks are connected with a 105-deg double male elbow and a claisen adapter with a condenser attached vertically. The SO3 is heated gently to drive it over to the SCl2 flask, forming thionyl chloride. The excess SO3 prevents S2Cl2/SCl2 contamination in the final SOCl2, because their close BPs make them hard to separate. After driving the SO3 over, the SO3 flask is removed and its claisen port is stoppered, leaving only the condenser. After bringing to reflux briefly, the mix is cooled and the vertical condenser is replaced by a vigreux column and the SOCl2 is fractionated at its BP.

Surprisingly simple, not to mention that pretty much all the popular carboxylic acid chlorides (incl. oxalyl-, acetyl-) can be made from SOCl2.

EDIT: Of course, in practice the "simple" method I describe is really a pain in the ass - it's a shitstorm of drying tubes made incredibly corrosive by nearby reactions and small burns from flame-drying glassware, followed by the nightmare-inducing reek of sulfur chlorides and the suffocating process of tearing down two massive chlorine generators, topped by billowing, acrid clouds of sulfuric acid mist and the inevitable fine coating of powdered sulfur and SO3-scorched grease all over everything... your keck clips will be destroyed, your lab stands will be corroded. Your gas scrubbers will have sulfur stuck in them and something you didn't know was there will probably burn you through a nitrile glove at some point. The SO3 generation flask will be etched, and the lab will have a lingering stench for weeks... and all for 150ml of a rotten liquid that doesn't store well.

To put it simply, it's fun as hell.

[Edited on 23-6-2016 by Praxichys]

Melgar - 23-6-2016 at 05:54

Quote: Originally posted by byko3y

I don't believe it works. The reaction is too slow at normal conditions to be usefull, and also gives a complex mixture of many different products.

It certainly did something, although now I suspect I probably made SO2Cl2. I was never able to purify it, because I cracked the glass cylinder when I was moving it indoors when it started to rain. I suppose it would have worked if I had used sulfur trioxide instead of dioxide. And this is supposed to be the SAFE route!

Quote: Originally posted by clearly_not_atara

Oh! Also, I should point out, that during the last thread where someone was dumb enough to make phosgene, IIRC we found a reference saying something to the effect that if you want to make diphosgene from methyl formate you have to do the first chlorination selectively to methyl chloroformate; apparently just replacing methyl chloroformate with the formate in the OrgSyn tek can cause a "boom". Also, IIRC, the OP of that thread was never heard from again.

I was going to attempt to use a chlorination method I've used before, in which a small amount of bromide salt is introduced, which generates bromine monochloride in situ. Bromine monochloride can then be selectively activated by incandescent light, producing the halogenation at lower temperatures, and without any significant risk of a runaway reaction. The danger you speak of, is that the heat generated will raise the temperature to the point where chlorine radical generation is spontaneous, which is of course, quite... exothermically unfavorable.

If that doesn't work, then no loss on my part. I'll try making benzoyl chloride instead.

clearly_not_atara - 23-6-2016 at 09:18

Actually, you could make cyanuric chloride. It's a little more dangerous than making SOCl2, but the final product is much safer to work with (and NCCl is still less scary than phosgene). You essentially make cyanogen chloride by passing Cl2 over sodium cyanide, although maybe ferricyanide would work (chlorine is pretty reactive after all). Then NCCl trimerizes to trichlorotriazine.

You may be able to make use of this:

http://www.sciencedirect.com/science/article/pii/S0043135498...

[Edited on 23-6-2016 by clearly_not_atara]

Melgar - 23-6-2016 at 13:03

Quote: Originally posted by clearly_not_atara

Actually, you could make cyanuric chloride. It's a little more dangerous than making SOCl2, but the final product is much safer to work with (and NCCl is still less scary than phosgene). You essentially make cyanogen chloride by passing Cl2 over sodium cyanide, although maybe ferricyanide would work (chlorine is pretty reactive after all). Then NCCl trimerizes to trichlorotriazine.

You may be able to make use of this:

http://www.sciencedirect.com/science/article/pii/S0043135498...

[Edited on 23-6-2016 by clearly_not_atara]

Nice. I can get potassium and sodium cyanide for less than $20 a pound at the jewelers' supply store. I don't suppose TCT substitutes for cyanogen halides in reactions that call for it very frequently though? I'm still kind of interested in trying the diphosgene synthesis, even if it's just a few grams, just to test out my theory. After I stumbled on that milder free-radical chlorination reaction a few years back, I never really attempted any of the other syntheses that this reaction could potentially be used for.

clearly_not_atara - 23-6-2016 at 23:51

Just make sure you do your research before making cyanogen chloride. It can be done safely, but it is still a deadly war gas. Trimerization occurs by passing NCCl over a catalyst.

TCT can be used to make most acyl halides, I'm not sure if it'll convert amides to chlorimines but I think it will convert phthalic acid to phthaloyl chloride, and phthaloyl chloride will totally convert amides to chlorimines. It might even convert oxalic acid to its halide, but I'm less sure of that. I also don't know what the reaction products with sulfonic acids look like.

I think TCT is a pretty good replacement for SOCl2 overall. However, NCCl itself is a different beast: it dealkylates amines and undergoes various complex condensation reactions, as it is a very reactive electrophile and the reaction products tend to be unstable.

Magpie - 24-6-2016 at 09:03

Acetyl chloride can be made from acetic anhydride and HCl. What specific acid chloride do you need?

[Edited on 24-6-2016 by Magpie]

Chemi Pharma - 24-6-2016 at 09:52

Magpie,

Acetic anhydride and CL2 does not make acetyl chloride, but a mixture of Chloroacetyl chloride and Chloroacetic acid. See the patent below:

To the other guys, specifically "Praxichys" or Doug's Lab at you tube: I sent you a long time ago, at you tube discussion board, a patent to make thionyl chloride from your S2CL2 pre made, but it seems you didn't pay enough attencion.

Of course it's possible to sinthesize thionyl chloride from SCL2 (or S2Cl2), Chlorine and SO2, using ferric chloride as a catalist, withouth the use of nasty, corrosive and lethal SO3. Even wikypedia say the same.

When my Lab get plenty ready, what i hope be soon, it will be my first great movie demonstration, and it will be posted here and at You Tube.

See everybody the patent below:

Attachment: thionyl chloride.pdf (707kB)
This file has been downloaded 535 times

Attachment: CHLOROACETYL CHLORIDE.pdf (380kB)
This file has been downloaded 539 times

Magpie - 24-6-2016 at 10:25

Quote: Originally posted by Chemi Pharma

Magpie,

Acetic anhydride and CL2 does not make acetyl chloride...

I meant to say HCl instead of Cl2. I corrected the post. Thanks for the note.

Praxichys - 24-6-2016 at 11:35

@Chemi Pharma - I am very interested to see your results. I abandoned this method early on since I lacked a good way to regulate the two gases. It makes me wonder if dripping H2SO4 onto a mixture of powderd Ca(OCl)2 and Na2S2O5 would generate the gases in the proportion in which the powders were mixed...

With your permission, I can reexamine this with better glassware and methods. I will not make a video about it if you would like to do that first.

My biggest problem is that any SO2Cl2 contamination is basically impossible to remove since its BP is so close to SOCl2.

Chemi Pharma - 24-6-2016 at 12:11

Of course i want to see you trying that first.

Science first, the father of the child goes further.

Read all the patent and you will see it's easy avoid SO2CL2 formation, in favor of SOCL2 just keeping the temperature of the reflux solution above 200ºC.

Your experiment at You Tube went very well. Improving your experience, I think we can produce Thionyl Chloride in one pot instead destilating the S2CL2/SCL2 mixture, as you did, just keeping refluxing it, adding more Chlorine at one side and also dry SO2 toghether at the other side of a three neck baloon, inside the mixture on refluxing.

The SO2 could be generated almost dry dripping conc. H2SO4 over sodium thiosulfate or bisulfite.

I Thought use a gas meter to measure the volume of chlorine and SO2 bubbled into the refluxing mixture.

You know: 22,4 liters of a gas is equal an one mol of the gas. One mol of Chlorine weight 71 grs and one mol of SO2 weight 64 grs.

It's just a stechiometry matter, but i believe an excess of gases will increase the yield.

When the reaction has done, cease the gas inlet, diminish the temperature, add a few grams of ANHNIDROUS ferric chloride as a catalizer and distill the thionyl chloride with a fracionated column, cause the boiling point of SCL2 and SOCL2 are too close.

Try this experiment Praxichys. I trust in your knowledgment cause iI saw everithing what you did as experiments at You Tube. If not I will do it soon.

Everybody here at sciencemadness is waiting for an OTC way to sinthesize Thionyl Chloride, using just raw materials, cause it's a hard to get chemical to buy on the market, although so useful in organic chemistry, thanks the creators of the mustard gas.

lucky!

clearly_not_atara - 24-6-2016 at 12:22

Since we already know that SO2 + Cl2 makes SO2Cl2 in the presence of a catalyst -- and it's famous as a reaction that almost anything will catalyse, including carbon, water, and visible light -- so it stands to reason that ferric chloride will do this as well. The equilibrium must therefore involve the comproportionation of SO2Cl2 and SCl2, and because the reaction succeeds, we know that this comproportionation must be kinetically viable and thermodynamically favorable under the reaction conditions.

So the direct comproportionation of SCl2 and SO2Cl2 is another way to carry out this reaction, most likely, and the instability of SCl2 explains where the free energy comes from. This is useful because the reactants are now both condensed at STP, and it's a dangerous reaction so it doesn't need to be any harder to contain than it already is...

Also, using excess SCl2 may be a way to avoid any SO2Cl2 in the reaction yield.

[Edited on 24-6-2016 by clearly_not_atara]

Chemi Pharma - 24-6-2016 at 12:35

Clearly,

I would like you read first the entire arguments contained at the patent i had post. The patent seems to bring too many solid arguments, enough to make me feel it's realible;

Posting again to everybody download and read:

Attachment: thionyl chloride.pdf (707kB)
This file has been downloaded 490 times

clearly_not_atara - 24-6-2016 at 12:41

Chemi, I think you misunderstand. I assumed the patent was correct; I just suggested on mechanistic grounds an easier way to perform the same reaction. In particular, we know that SCl2 does not react with Cl2 above 250 K or so, because SCl4 is unstable at these temperatures. SCl2 also does not react with SO2. But SO2 reacts rapidly with Cl2.

Note that S2Cl2 does not work in my modification, because we have no reason to believe it will react with SO2Cl2. In this case we know that it does react with Cl2 directly.

[Edited on 24-6-2016 by clearly_not_atara]

Chemi Pharma - 24-6-2016 at 12:44

Sorry Clearly,

I read your comment too fast. It was a misunderstood.

I apologize.

S.C. Wack - 24-6-2016 at 17:48

Quote: Originally posted by Melgar

Incidentally, since you mention oxalyl chloride, I've read that the electron withdrawing effect of the benzene ring also makes benzoyl chloride a decent, mild reagent for making other acid chlorides. If this is the case, I could theoretically chlorinate toluene to benzotrichloride, then combine this with an equimolar amount of benzoic acid to generate two units of benzoyl chloride for each unit of benzotrichloride I painstakingly made; the benzoic acid could be obtained on eBay for a song.

I wonder if this could be done by chlorinating benzyl alcohol? The resulting hydrogen chloride would combine with more benzyl alcohol to make benzyl chloride, which could then be further chlorinated. I like the benzotrichloride + benzoic acid route though, if I'm going to generate an acid chloride from another acid chloride. It seems like it works on all the carboxylic acids I'm interested in, at least according to an organic chemistry text I read recently, which I'm unable to locate for some reason.

BTW...benzaldehyde is easily chlorinated if one wants to use it that way...actual reference, JACS 60, 1325 (1938) aka A Convenient Preparation of Volatile Acid Chlorides...and as noted there, phthalyl chloride came first (JACS 59, 206 (1937), also sort of a SOCl2 preparation), which that benzotrichloride will give with phthalic anhydride...

Pumukli - 25-6-2016 at 09:52

Depending on what sort of acid chloride do you have in mind there may be another route which employs sulfuryl-chloride as chlorinating agent:
E.g.: 2 Sodium-benzoate + SO₂Cl₂ -> Na₂SO₄ + 2 benzoyl-chloride
(or something like that, depending on the exact reagents and stochiometry)

I faintly remember reading a paper about sulfuryl-chloride a few months ago and the above equation is from that paper. I also remember that water is a catalyst poison in making sulfuryl-chloride (at least when activated carbon (Norit) is used). Thoroughly dried carbon gives best results.

MeshPL - 26-6-2016 at 11:46

What concerns PCl3, which no one cares about, since it is "impossible" to produce from red phosphorus, I know a guy, who is also interested in chemistry, and he claims, that it can be made (and he did made it) from red phosphorus, but:
-temperature must be kept low, even below 100C.
-chlorine must be diluted with inert gas, down to 10%-20%.
-silica should be added as a catalyst.
I'm not sure if that actually works, but that's what he said. I have no way to test this, but maybe somebody over here will do - eventually.

Also wikipedia mentions PCl3 can be oxidised with potassium chlorate:
PCl3 + KClO3 ---> 3POCl3 + KCl
I wonder, if simmilar thing works for SCl2:
3SCl2 + KClO3 ---> 3SOCl2 + KCl
Or:
3SCl2 +KClO3 ---> 3SO2Cl2 + 2KCl

zed - 26-6-2016 at 13:38

Well, one of you clever monkeys came up with this.....AcetylChloride from Acetonitrile, AceticAcid, and HCL gas. Even made a video demo!

https://www.youtube.com/watch?v=ExDI9KUW-9Q

Yield is about 30%. Not bad.... considering the cheapness of the starting materials, and the current difficulties with obtaining Acetyl Chloride.

[Edited on 26-6-2016 by zed]

S.C. Wack - 26-6-2016 at 14:54

PS hexachloro p-xylene and SO2 gives terephthaloyl chloride and thionyl chloride according to US3411886...

Quote: Originally posted by MeshPL

What concerns PCl3, which no one cares about, since it is "impossible" to produce from red phosphorus

I'm doubtful of anyone saying whatever can't be done, and here for this Thorpe claimed it was easy I've mentioned before in the relevant thread, also there's that part of Inorganic Syntheses (2, 145) that goes "Two hundred grams of dry red phosphorus is placed in the flask. Excess phosphorus must be present to inhibit the formation of phosphorus(V) chloride." They found a way and it's probably not the only one.

Quote: Originally posted by zed

Well, one of you clever monkeys came up with this.....AcetylChloride from Acetonitrile, AceticAcid, and HCL gas.

Mr. Colson. Benzotrichloride and acetic acid gives a mixture of acetyl chloride and benzoyl chloride says US1965556.

Chemi Pharma - 26-6-2016 at 17:06

MeshPL,

you can sinthesize PCL5 easily from red phosphorous in chloroform suspension, bubbling into dry chlorine.

To make PCL3 you just need to reflux PCL5 with stoichmetric amount of red phosphorous at a dry and moisture free apparatus full of a inert gas, like nitrogen or helium.

See the recipe below:

Attachment: Phosphorus pentachloride.doc (23kB)
This file has been downloaded 452 times

clearly_not_atara - 28-6-2016 at 13:18

Quote:

Well, one of you clever monkeys came up with this.....AcetylChloride from Acetonitrile, AceticAcid, and HCL gas. Even made a video demo!

https://www.thieme-connect.com/products/ejournals/html/10.10...

EtOH (1 mmol) + TCCA (0.75 mmol) + NH3 (45 mmol) (aq) >> MeCN (aq)

It's not practical to make acetonitrile for solvent use -- you'd need a 45x molar excess of ammonia -- but it's fine if you're just gonna make acetyl chloride. Since it would appear that the formed acetamide is inert to the reaction conditions, you may be able to use other nitriles as well (if acetonitrile proves to be too hard to separate from aqueous ammonia...).

[Edited on 28-6-2016 by clearly_not_atara]

Cryolite. - 28-6-2016 at 16:16

(forgot my password, waiting for reset email)

There's also www.sciencemadness.org/talk/files.php?pid=69963&aid=1539 , which provides a way to oxidize alpha-amino acids to nitriles using TCCA and sodium hydroxide. The reaction conditions are mild, the excesses small, and the extractions are easy, and most notably this should provide a way to oxidize L-alanine (sold as a nutritional supplement) to acetonitrile. Although I haven't seen this particular substrate tested in any of the papers on this oxidation, the paper does show oxidation of aspartic acid to cyanoacetic acid in 99% yield. (OTC malonic acid maybe?) I don't see any reason for this to fail for alanine.

zed - 28-6-2016 at 17:09

Some of the guys are inundated with Acetonitrile. Seems it is used as solvent in certain analytical procedures. Can also be produced by the dehydration of Acetamide.

Still, the point is....Acid Chlorides are hard to come by. Acetonitrile, not so much.

Melgar - 6-7-2016 at 01:34

I think I may go for the thionyl chloride route, since I'll be traveling to a location where I can use every dangerous reagent there is (rich guy with a home lab, fume hood, the works). So the idea is that I'll buy a bucket of TCCA tablets, drip HCL on them, gas elemental sulfur until it stops absorbing chlorine and changes color, then set that aside. Next, get a flask full of sodium bisulfate, and heat it until steam stops coming off and it starts to melt, around 400C. Then send the tube into the sulfur dichloride from earlier, with appropriate suck-back prevention set up, and gradually heat the sodium pyrosulfate to decomposition (460C or so). Run that through my distillation apparatus just to cool it down and measure its temperature (SO3 has a boiling point of 46C) then bubble that through SCl2.

A few questions: do SO2, S2Cl2, or Cl2 interfere with the last step of the reaction? According to the wiki page , there are reactions that use chlorine, consume SO2, and generate thionyl chloride, so it would seem to be beneficial to continue bubbling Cl2 into the mixture after the SO3 bubbling started, no? Would Cl2 have decent solubility in this mixture? Would there be much chance of contamination with SO2Cl2? Supposedly that needs a catalyst to form, but carbon will do, and SO3 will turn quite a lot of things into carbon. In any case, I'll have to fractionally distill the end product. Hopefully the fact that their boiling points are only 5K different won't be a problem. In any case, sulfuryl chloride can apparently turn alcohols into alkyl chlorides among other things, so I wouldn't be disappointed if any came over.

I did find this excerpt from a book on World War I, about diphosgene:

https://books.google.com/booksid=u2U7AwAAQBAJ&lpg=PT166&...

Supposedly they used it to fill shells, because then they didn't have to refrigerate the factories where it was produced. The heat from the shells exploding would release the phosgene. Anyway, chlorination of methyl formate was the dominant process used by the Germans in 1917, so it can't be that disaster-prone, right? The fact that you don't have actual phosgene gas at any point is quite attractive.

careysub - 6-7-2016 at 07:30

Quote: Originally posted by Melgar

I did find this excerpt from a book on World War I, about diphosgene:

https://books.google.com/booksid=u2U7AwAAQBAJ&lpg=PT166&...

Supposedly they used it to fill shells, because then they didn't have to refrigerate the factories where it was produced. The heat from the shells exploding would release the phosgene. Anyway, chlorination of methyl formate was the dominant process used by the Germans in 1917, so it can't be that disaster-prone, right? The fact that you don't have actual phosgene gas at any point is quite attractive.

Although diphosgene is a liquid at RT, with a boiling point at 128 C, unlike phosgene which is a gas at RT, it is NOT SIGNIFICANTLY SAFER THAN PHOSGENE!!!!!!

Its volatility at RT is 111000 mg/m^3, which is so high it might as well be phosgene. The CUMULATIVE lethal exposure is 3200 mg-min/m^3 so exposure to a concentration of only 0.15% of its saturation limit in air will sentence you to death in 20 minutes (no treatment exists). And remember you can accumulate a lethal exposure over the course of a week, in little bits.

It was not the heat from shells that caused the diphosgene to evaporate, it was simply the fact that the explosion disperses it as an aerosol. Ordinary evaporation turns it into a vapor in seconds.

You have to go to triphosgene to get any significant safety improvement. This form is both actually less toxic that the other two (about 1/6 as toxic as phosgene) and has a volatility of only 2400 mg/m^3.

[Edited on 6-7-2016 by careysub]

Melgar - 6-7-2016 at 12:04

Quote: Originally posted by careysub

Quote: Originally posted by Melgar

Oh, I wouldn't touch the stuff without a fume hood, and containers of 28% NH4OH sitting out until the room reeked of ammonia. I'd be able to see the vapors before any of it could get near my airways, not to mention the fact that it'd be neutralized. And of course, it'd all be set up for use in a closed system. Also, any reaction vessels would be flushed then rinsed with ammonia, and basically I'd always have ammonia handy to quench anything that started to look bad.

Some people hate the smell of ammonia, but to me, it smells like (relative) safety. It's nice being able to see halogenated nasty shit and acid vapors before I can smell it.

Regardless, I'd probably only try making it once, and dump anything leftover into (you guessed it!) ammonia once I'm done with whatever synthesis I'd attempt to use it for.

[Edited on 7/6/16 by Melgar]

careysub - 6-7-2016 at 14:10

Triphosgene can be made by chlorinating dimethyl carbonate (available from Elemental Scientific LLC) with a UV lamp.

That might be safer.

Attachment: eckert1987.pdf (248kB)
This file has been downloaded 442 times

Dope Amine - 18-7-2016 at 22:35

Volatile acyl chlorides can be made by distilling a mixture of benzoyl chloride and a given acid. This is my favorite way to make propionyl chloride.

Paraphrasing the experimental procedure - Set up for simple distillation (with moisture protection) a mixture of 2 moles benzoyl chloride for every 1 mole of propionic acid. Aggressively heat (high heat setting) the mixture and a distillate will come over at the b.p. of propionyl chloride. Stop distilling when the head temperature drops. The article recommends then re-distilling the distillate but I don't bother anymore after having tried that once. Rather than using a drying tube on the vac intake of the distillation setup, I prefer to run a hose into an IPA solution that I have stirring. This is because HCl-gassed IPA is a favorite crystallization tool of mine (dissolve freebase in toluene and then add HCl/IPA). The only drawback of the HCl gassing route is that you'll need to keep an eye on it toward the end of the distillation in order to prevent the eventual suckback which would ruin your distillate product.

J. Am. Chem. Soc., 1938, 60 (6), pp 1325–1328

[Edited on 19-7-2016 by Dope Amine]

Cryolite. - 19-7-2016 at 14:09

That method works, but the issue is obtaining benzoyl chloride. It can be made by the partial hydrolysis of benzotrichloride, which in turn can be made by free radical chlorination of toluene, but the benzotrichloride intermediate is extremely lachrymatory and toxic. Even worse, producing the benzotrichloride must be done by free radical chlorination with a radical initiator (UV light works well), and thefore the synthesis must not be done very carefully-- free radical reactions can run out of control very quickly.

Despite this, this seems like an otherwise good way to make volatile acid chlorides. I'm pretty sure that benzotrichloride itself can be used to form acid chlorides equivalently, which will increase efficiency.

Now, the question is: how do we make nonvolatile acid chlorides?

clearly_not_atara - 20-7-2016 at 11:28

I assume the acetonitrile method generalizes to arbitrary nitriles:

RCN + HCl >> RC(NH)Cl

RC(NH)Cl + R'COOH >> RC(NH)OC(O)R'

"" + R'COOH >> RCONH2 + (R'CO)2O

(R'CO)2O + HCl >> R'COCl + R'COOH

So benzonitrile + HCl might work for acyl chlorides which are more volatile than benzonitrile.

Cryolite. - 20-7-2016 at 13:46

That would work (I believe benzonitrile itself can be made from benzoic acid, urea, and sulfamic acid as a dehydrator), but the issue is obtaining nonvolatile acid chlorides is still off-limits for this procedure, unless a way to separate the acid chloride from the benzamide byproduct is found. Maybe a solvent in which acid chlorides are soluble, but amides are not?

Texium - 20-7-2016 at 14:56

Now, I don't know exactly what any of you want acid chlorides for, but have you thought about making acid bromides instead? They react in the same way as acid chlorides in many reactions, and are easier to prepare in a home lab. I would assume that the main reason that acid chlorides are used more frequently in industrial and commercial labs is because on the commercial scale they are easier and less expensive to prepare.

While PCl₃ is very difficult to prepare in the home lab, PBr₃ is actually quite easy if you already have bromine on hand. I put up a video showing the simple synthesis of it a couple days ago: Synthesis of Phosphorus Tribromide

I'm planning to make either acetyl bromide or benzoyl bromide using the PBr₃ soon.

Melgar - 20-7-2016 at 15:47

Quote: Originally posted by Cryolite.

That method works, but the issue is obtaining benzoyl chloride. It can be made by the partial hydrolysis of benzotrichloride, which in turn can be made by free radical chlorination of toluene, but the benzotrichloride intermediate is extremely lachrymatory and toxic. Even worse, producing the benzotrichloride must be done by free radical chlorination with a radical initiator (UV light works well), and thefore the synthesis must not be done very carefully-- free radical reactions can run out of control very quickly.

I've done that reaction, and it can be done with visible light and a catalytic amount of bromine, which can be added in the form of a bromide salt, and forms bromine monochloride in situ. This allows for lower temperatures and much more even penetration of the light, rather than having it all get absorbed in the first few millimeters. Thermal runaways are virtually impossible, as long as the vessel stays lit.

Quote:

Despite this, this seems like an otherwise good way to make volatile acid chlorides. I'm pretty sure that benzotrichloride itself can be used to form acid chlorides equivalently, which will increase efficiency.

Can't benzotrichloride generate two equivalents of acid chloride? That would make sense, since mixing one mole of benzotrichloride with one mole of benzoic acid results in two moles (more or less) of benzoyl chloride. Since water will hydrolyze benzotrichloride to benzoic acid, then even if benzotrichloride is totally nonreactive to anything but benzoic acid and water, it would generate two equivalents of acid chloride. The only thing you'd have to worry about is side reactions.

Quote:

That would work (I believe benzonitrile itself can be made from benzoic acid, urea, and sulfamic acid as a dehydrator), but the issue is obtaining nonvolatile acid chlorides is still off-limits for this procedure, unless a way to separate the acid chloride from the benzamide byproduct is found. Maybe a solvent in which acid chlorides are soluble, but amides are not?

In that case, it's probably worth the effort to just make thionyl chloride instead. Either phosphorus pentoxide (order from Firefox Pyrotechnics in the US, make phosphorus halides outside the US) added to sulfuric acid, or heating sodium bisulfate to 460C can generate the SO3, chlorine reacting with elemental sulfur can generate sulfur dichoride and disulfur dichloride, both of which react with SO3 to generate thionyl chloride.

Quote:

What magical land is this where people speak with American accents and have easy access to red phosphorus? I assumed you must be South African or something until I watched the video.

[Edited on 7/21/16 by Melgar]

zed - 20-7-2016 at 16:14

Nice.

Though..... some of the guys at home, have real problems obtaining red-phosphorus.

Well, any kind of elemental phosphorus...actually.

A useful chlorinating agent, can be synthesized via P2O5, when available.

Cryolite. - 21-7-2016 at 13:17

Quote:

I remember reading that thread-- thanks for the reminder. From what I remember, the evidence Nicodem provided suggested that a very pure benzyl chloride product was obtained via Br-catalyzed TCCA chlorination was obtained, but no higher-chlorinated products were obtained. I guess this was due to the massive excess of toluene used. It might be worth a shot-- I'll try it out if I'm at all free during the weekend.

Quote:

That was what I meant: using benzotrichloride directly as the chlorinator instead of quenching with water to benzoyl chloride. This gives you twice the overall chlorinating power, improving efficiency. However, going by the intermediacy of benzoyl chloride by reaction with benzoic acid seems like a good idea, as it is a lot nicer to handle and store than the BzCl-on-steroids benzotrichloride.

Quote:

I am well aware of this route, but unfortunately I am slightly terrified of SO3, and so producing thionyl chloride in this way is not for me

[Edited on 21-7-2016 by Cryolite.]

clearly_not_atara - 21-7-2016 at 13:40

Quote: Originally posted by zts16

Now, I don't know exactly what any of you want acid chlorides for, but have you thought about making acid bromides instead? They react in the same way as acid chlorides in many reactions, and are easier to prepare in a home lab. I would assume that the main reason that acid chlorides are used more frequently in industrial and commercial labs is because on the commercial scale they are easier and less expensive to prepare.

While PCl₃ is very difficult to prepare in the home lab, PBr₃ is actually quite easy if you already have bromine on hand. I put up a video showing the simple synthesis of it a couple days ago: Synthesis of Phosphorus Tribromide

I'm planning to make either acetyl bromide or benzoyl bromide using the PBr₃ soon.

PBr3 is awesome. It is perhaps the reagent of choice when available. There are, however, two important considerations:

* byproduct. PBr3 hydrolyses to phosphorus acid. While harmless at ordinary temperatures, it disproportionates upon boiling to release phosphine gas, which is highly toxic, and pyrophoric in high concentrations. Reaction mixtures containing spent PBr3 should never be distilled at atmospheric pressure. This is particularly important since the acyl bromides produced are not usually very volatile.

* raw materials. Phosphorus is illegal.

Melgar - 22-7-2016 at 20:29

Quote: Originally posted by Cryolite.

I think it also had to do with the fact that the reaction was taking place throughout the whole vessel, rather than just in the edges. This would tend to make the reaction products more homogeneous.

Incidentally, here is the link to the thread we're talking about, for anyone interested:

http://www.sciencemadness.org/talk/viewthread.php?tid=14063

When I was researching this reaction to see if anyone else had discovered it first, the only reference I found to the use of bromine in the chlorination of toluene was in Ullman's Encyclopedia of Industrial Chemistry, in the benzotrichloride section, where it mentions that addition of bromine would assist in the completion of the reaction. For everyone complaining about how difficult it is to fully chlorinate the toluene alpha methyl, this could potentially make things a lot easier.

Quote:

Good to know that about benzotrichloride. I don't have experience with either of them, so I'm glad to know ahead of time which one of them is least pleasant to work with.

[Edited on 7/23/16 by Melgar]

Texium - 22-7-2016 at 20:53

Quote: Originally posted by clearly_not_atara

* raw materials. Phosphorus is illegal.

Well, if I'm to assume you're in the US, it's not illegal, it's just that the sale of it within the country is regulated to the point of it being unavailable.

Melgar - 26-7-2016 at 03:50

Quote: Originally posted by Cryolite.

I'm hearing references that phosphorus oxychloride can be prepared fairly easily by heating P2O5 in the presence of a large excess of sodium chloride. That's essentially the acid chloride of phosphoric acid, correct? If so, that would be ideal for anyone who could get ahold of P2O5, which is perfectly legal to buy and sell in the US.

Cryolite. - 26-7-2016 at 12:38

That prep is actually in len1's book-- intimately mix dry table salt and phosphorus pentoxide and dry distill at 300+ C. There is etching of the glass, but without water present it is minor. However, I was under the impression that most acid chlorides cannot be directly made from phosphoryl chloride. After all, the use of phosphorus pentachloride stops at the oxychloride stage on most substrates.

If you are proposing the use of phosphoryl chloride as a solvent, it would work to dissolve acid chlorides, but just like phosphorus pentoxide it reacts with amides, forming nitriles. However, this does proceed through the imidoyl chloride stage, just like the use of HCl on nitriles. Maybe in an excess of the carboxylic acid with a catalytic amount of the amide and a stream of dry HCl gas, phosphoryl chloride could be used to make acid chlorides. Hmmm...

[Edited on 26-7-2016 by Cryolite.]

clearly_not_atara - 26-7-2016 at 17:07

The preparation of benzonitrile is much easier than the preparation of phosphorus pentoxide or sulfur trioxide (which in terms of handling alone is already one of the more treacherous things anyone will attempt) since benzonitrile can be had with benzaldehyde + ammonia + TCCA according to the paper previously posted. This should be particularly high-yielding since benzaldehyde is particularly good at forming imines. Anisaldehyde and piperonal will also generate usable nitriles, AFAICT.

EDIT: one interesting possibility is to use m-chlorobenzaldehyde to make m-chlorobenzonitrile and m-chlorobenzoic acid and ultimately mCPBA.

[Edited on 27-7-2016 by clearly_not_atara]

Chemi Pharma - 12-2-2017 at 09:34

Clearly,

the link you've posted before goes to nowhere.

I think the paper you're talking about is the one i attached below.

Attachment: alcohols, aldehydes, amines and benzyl halides to nitriles with TCCA + NH4OH.pdf (115kB)
This file has been downloaded 722 times

clearly_not_atara - 18-2-2017 at 22:29

^Thanks for catching that. Sometimes links stop working, especially when I find the paper on an academic blog or something and they realize it's being downloaded

Maybe the nitrile route could extend to oxalyl chloride? I suppose the necessary assumption is that oxalyl chloride won't react with benzamide:

2PhCN + (CO2H)2 + 2 HCl >> PhCONH2 + (COCl)2

Oxalyl chloride is a pretty versatile substance IIRC and not easy to make other ways

[Edited on 19-2-2017 by clearly_not_atara]

JJay - 18-2-2017 at 22:46

Trifluoroacetic mixed anhydrides can often substitute for acyl chlorides. Trifluoroacetic anhydride is a little exotic but has many uses and isn't out of reach of the advanced amateur.

Melgar - 23-2-2017 at 12:37

I ended up taking a second look at phosphorus halides. If I had to choose between already-prepared phosphorus pentachloride, and phosphorus tribromide, and I mainly wanted to prepare acid chlorides, which should I use and why? I'm leaning towards phosphorus pentachloride, since it's cheaper, and turns into phosphoryl chloride when it loses two chlorine atoms. Apparently, phosphoryl chloride can turn some, but not all carboxylic acids into their acid chlorides, depending on which chlorine atom is active and whether the carboxylic acid is stronger than the corresponding phosphoric-chloride acid? Does phosphorus tribromide have a better atom efficiency? Are acid bromides that much easier to work with than chlorides, assuming most of the acids I'll be working with will have three or more carbons?

Still, there's a lot more literature on acid chlorides, so I'm kind of biased towards that approach anyway.

Kapitan - 6-3-2017 at 18:04

Thionyl chloride is one of my favorite reagents. It always seems to always work really well, and the workup is a snap: Just distill off the excess reagent. Plus, once I use SOCl2 I can't smell any of the other bad-smelling reagents that I use for the rest of the day

Oxalyl chloride is nice because it's like the magical self-purifying reagent (all the decomposition products are gaseous), but in my experience the results are more inconsistent than with SOCl2. I guess it's supposed to be like the "greener" SOCl2, but it just doesn't seem to work as well.

I try to avoid POCl3 and PCl5, both because they're super nasty reagents, and also because I was sprayed in the face with them once and can't stand the smell.

[Edited on 7-3-2017 by Kapitan]

[Edited on 7-3-2017 by Kapitan]

Elemental Phosphorus - 27-3-2017 at 06:12

I think that for almost all uses, it does not matter whether an acid bromide or chloride is used. The chlorine atom will form a stronger bond with alkali ions, like in the creation of anhydrides, but it is a worse leaving group owing to the increased strength of the carbon-chlorine bond. The bromine atom forms a weaker bond with sodium/potassium ions and hydrogen but is a better leaving group.

Melgar - 27-3-2017 at 10:41

Quote: Originally posted by Elemental Phosphorus

I think that for almost all uses, it does not matter whether an acid bromide or chloride is used. The chlorine atom will form a stronger bond with alkali ions, like in the creation of anhydrides, but it is a worse leaving group owing to the increased strength of the carbon-chlorine bond. The bromine atom forms a weaker bond with sodium/potassium ions and hydrogen but is a better leaving group.

Probably. Assume phosphorus is the limiting reagent though. Wouldn't PCl5 be better, if only because it has more Cl groups? After all, after it gives up the first two, it's still the acid chloride of phosphoric acid, and can continue to react with carboxylic acids to produce acid chlorides.

Also, I've since learned that if Cl2 and SO2 are combined in the presence of phosphorus, thionyl chloride and phosphorus oxychloride are both formed together, and can be separated by distillation. That seems like far less of a pain in the ass than screwing around with SO3, and you get two useful reagents instead of just one. You can use POCl3 when possible then, and SOCl2 when it's not possible.

[Edited on 3/27/17 by Melgar]

Elemental Phosphorus - 27-3-2017 at 11:50

If the PCl5 is already prepared and cheaper it is probably the better way. Otherwise, preparation of PBr3 is easier to do. I would rather make the thionyl chloride to be honest, since it is not too hard to prepare SO3 (think a thick steel retort with FeSO4 on a coal fire) and the steel will be passivated by an FeSO4 layer. Mole for mole, gram for gram PCl5 is the better way (with phosphorus as limiting reagent), if you don't mind chlorine gas, or just buy the PCl5.