Sciencemadness Discussion Board - Preparation of Dimethyl Sulfate Discussion

Preparation of Dimethyl Sulfate Discussion

JJay - 8-9-2018 at 05:42

I've been looking at dimethyl sulfate recently. As a methylation agent, it has some attractive properties: It's cheap and effective. While it can't be used to make Grignards directly, dimethyl sulfate is said to react with potassium iodide to make methyl iodide in high yield.

I haven't read many reliable reports where hobbyists said they made dimethyl sulfate successfully. I don't think it's particularly hard to make, but it's extremely toxic and carcinogenic, and most references suggest methods requiring reagents that are hard to obtain. While some old procedures suggest that a fume hood is not required for handling it, Vogel wrote, "The vapour is highly poisonous and the substance should only be used in a fume cupboard with a good draught. The liquid itself is readily
absorbed through the skin, with toxic results."

I recently ran across a simple procedure to make methyl sulfuric acid. It doesn't require anything exotic. Vacuum distillation of methylsulfuric acid produces dimethyl sulfate. I think calcium carbonate would work OK instead of barium carbonate if making dimethyl sulfate.

Quote:

A general method of obtaining methyl and ethyl sulphates is to treat the alcohols with concentrated sulphuric acid or with chlorosulphonic acid. The first method is convenient, but since the reaction is a reversible one it does not give the theoretical yield.

For the methyl hydrogen sulphate, 1 part by weight of methyl alcohol is mixed with 2 parts of strong sulphuric acid, and the mixture kept warm for some hours. It is then diluted with water, and barium carbonate added in excess, to neutralise the free acids. After filtration from the barium sulphate and excess of carbonate, the filtrate, which contains the barium salt of methyl hydrogen sulphate, is evaporated to a conveniently small bulk, and treated with just sufficient sulphuric acid to precipitate all the barium as sulphate, and then again filtered. An aqueous solution of the free methyl sulphuric acid is thus obtained, which can now be evaporated in a vacuum over strong sulphuric acid.

Methyl hydrogen sulphate is an oily liquid, which does not solidify on cooling to - 30°. On distillation, it decomposes into dimethyl sulphate and sulphuric acid.

Dimethyl sulphate, (CH3),SO4, is made by distilling monomethyl sulphate in a vacuum. To obtain a good yield, the monomethyl compound should be free from water and other impurities. It need not, however, be separately prepared. A mixture of 25 kilos. of sulphuric anhydride with 10 kilos. of nearly absolute methyl alcohol is made, the temperature being kept below 0°, and preferably between - 5° and - 10°. The mixture is then distilled in vacuo. An almost theoretical yield of 19 kilos. can be obtained.

Alternatively, 100 parts of chlorosulphonic acid are run into 27 parts of methyl alcohol at a temperature of - 10° to - 15°, care being taken that the liquids are well mixed, and moisture excluded. The mixture is then distilled in a vacuum 4 (20 mm.; the temperature being about 140°). Yield, about 80 per cent. of the theoretical.

I understand that dimethyl sulfate is very toxic and carcinogenic, but it has a fairly high boiling point. Are there any easier ways to make it? Is there any reason it shouldn't be handled with regular lab gear (nitrile gloves, lab coat, etc.) in a fume hood?

Metacelsus - 8-9-2018 at 07:14

I have used dimethyl sulfate before, under the conditions you describe (gloves, lab coat, etc. in a fume hood). I believe it is safer than methyl iodide due to its lower volatility. When using it, I would recommend keeping a beaker of ammonia solution on hand so that any spills can be immediately quenched.

I would be hesitant to prepare dimethyl sulfate on a large scale in a home setting. If there were to be an accident, the results would be quite bad.

Tsjerk - 8-9-2018 at 07:47

Calcium carbonate should work.

Here they do it without either calcium or barium.

https://www.slideshare.net/mobile/NikolayMasin/dimethyl-sulf...

S.C. Wack - 8-9-2018 at 08:58

Get BaCO3 from a ceramics supplier if nothing else (expect sulfide) and don't look back. BaSO4 is heavy, does the job right, and is easily regenerated by boiling with baking soda.

JJay - 8-9-2018 at 09:47

Quote: Originally posted by Tsjerk

Calcium carbonate should work.

Here they do it without either calcium or barium.

https://www.slideshare.net/mobile/NikolayMasin/dimethyl-sulf...

dimethyl-sulfate-synthesis-1-638.jpg - 133kB

That's really interesting. I can't seem to download the presentation, but that procedure looks really easy. I am curious about why the reported yield is so low... but the reagents are cheap. The references given in the presentation don't seem to relate directly to the procedure. Does it work? I will say that I've found that the high yields of US 3047604 are hard to reproduce, but it does work to some extent....

JJay - 8-9-2018 at 09:59

Quote: Originally posted by Metacelsus

I have used dimethyl sulfate before, under the conditions you describe (gloves, lab coat, etc. in a fume hood). I believe it is safer than methyl iodide due to its lower volatility. When using it, I would recommend keeping a beaker of ammonia solution on hand so that any spills can be immediately quenched.

I would be hesitant to prepare dimethyl sulfate on a large scale in a home setting. If there were to be an accident, the results would be quite bad.

Keeping ammonia solution ready for quenching spills is a good idea.

I would be reluctant to prepare it on a large scale at home as well. I don't think it's more dangerous than chlorine or bromine, though....

Loptr - 8-9-2018 at 11:56

Quote: Originally posted by JJay

Quote: Originally posted by Metacelsus

I would have to disagree with the implications of that statement. The halogens smack you across the face to make their presence known, while dimethyl sulfate is just a smell. With chlorine I have to get out of the area because it hurts. I am not sure about the toxicity comparison, but I know that I would likely have more exposure with dimethyl sulfate than chlorine.

JJay - 8-9-2018 at 12:32

Bromine and chlorine have strong and offensive odors, and dimethyl sulfate has little odor. Bromine has an IDLH level of about 1/3 that of dimethyl sulfate and an OSHA PEL at about 1/10. Chlorine has IDLH and PEL levels that are comparable to dimethyl sulfate's. But chlorine is a gas, and bromine is a volatile liquid boiling at 59 C. Dimethyl sulfate boils at 188 C.

I would say that the outlet of the vacuum pump should be inside the fume hood during preparation (along with scrubbers and traps). There should be absolutely no detectable odor. A respirator should be available. People who are not aware of the risks should be kept away when dimethyl sulfate is being prepared or is in use.

AvBaeyer - 11-9-2018 at 07:03

From a safety perspective, I cannot understand why anyone would want to make dimethyl sulfate in a home/amateur laboratory setting using the procedures initially described above. The material is just too dangerous. Methyl iodide is far easier to make and though not the safest of reagents it can be used with appropriate caution. Moreover, when using methyl iodide for a methylation, the iodide can be easily recovered as iodine from the work-up solutions. An alternative is methyl p-toluensulfonate which can be easily prepared following Org Syn procedures. It requires no distillation or follow on purification before use and is relatively non-volatile. It is my go-to methylation reagent, especially when used with a small amount of iodide catalyst. At least in the US, the now ready availability of p-toluenesulfonyl chloride makes methyl p-toluenesulfonate (as well as other esters) the reagent of choice. Just my nickels worth.

AvB

JJay - 11-9-2018 at 10:01

I can't get 85% phosphoric acid, for one. I don't think I can get p-toluenesulfonyl chloride either.

Magpie - 11-9-2018 at 15:28

85% phosphoric acid is available through Duda Diesel.

JJay - 11-9-2018 at 15:30

They won't ship it to my location.

Edit: Actually, they will ship it to my location if I buy it directly off of the company website, but shipping costs far exceed the price of the product.

[Edited on 12-9-2018 by JJay]

Magpie - 11-9-2018 at 17:14

I have seen 85% phosphoric acid sold at brewing supply stores. It is also available through Amazon, some with free shipping:

https://www.amazon.com/s?k=85+phosphoric+acid&hvadid=143...

[Edited on 12-9-2018 by Magpie]

JJay - 11-9-2018 at 18:26

I've looked for it. It's not quite inaccessible, but shipping is expensive. A lot of methylations are lower yielding with methyl iodide than with dimethyl sulfate. Synthesis of methyl iodide with dimethyl sulfate seems to be highest yielding method....

I'm not quite sure why it is regarded as so dangerous. It's dangerous, sure, but I've handled more dangerous things.

zed - 11-9-2018 at 18:28

My buddy Dr.Death, used to work with Dimethylsulfate on occasion.

Claimed it was a lovely way to methylate yer innards, with the bonus of releasing H2SO4 into your delicate lung tissues.

Trimethylphosphate appears to be a nice methylating agent, but the literature doesn't provide a lot of convenient preparations. Not very volatile, and phosphoric acid, just doesn't pack the same scorch factor that H2SO4 does.

P-TolueneSulfonylChloride? I'll look into it. Methyl prep? I assume you mean this one? http://www.orgsyn.org/demo.aspx?prep=CV1P0145

Well, not dirt cheap. But, available on Amazon. https://www.amazon.com/HiMedia-GRM2486-500G-P-Toluenesulfony...

Thank you AvBaeyer.

[Edited on 12-9-2018 by zed]

[Edited on 12-9-2018 by zed]

JJay - 11-9-2018 at 19:08

Quote: Originally posted by zed

But, available on Amazon. https://www.amazon.com/HiMedia-GRM2486-500G-P-Toluenesulfony...

Thank you AvBaeyer.

I could probably get that to work just fine for methylations, but I have no idea what the yields would be. I've never bought from HiMedia before.

JJay - 11-9-2018 at 19:25

After researching the topic a bit, it looks like I could actually make some methyl tosylate with materials I have on hand (I'd use methyl orthoformate and toluenesulfonic acid, both of which I would have to prepare), but I think making and using dimethyl sulfate would be a lot less work, and there is a lot more literature on it. I suspect dimethyl sulfate methylates in higher yields, and I think it's more versatile. Data on the toxicity of methyl tosylate is a bit sparse.

Metacelsus - 11-9-2018 at 20:11

For methylation of phenols, dimethyl sulfate gives quantitative yields. I don't know about other methylating agents (tosylate / phosphate).

If you can get trimethyl orthoformate, you might be able to use that as a weak methylating agent.

AvBaeyer - 12-9-2018 at 00:01

As I mentioned I use methyl p-tosylate as my reagent of choice. For example, I have methylated acetaminophen, vanillin, and ethylvanillin with isolated yields of >90% on a 0.1 mole scale. The reactions are run in acetone using 50% excess potassium carbonate and 10 mole% NaI or KI as catalyst. The reaction is run at 50-55 C in a closed system for 24 hrs. The reaction time is arbitrary as the phenol is largely bound to the potassium carbonate so the disappearance of phenol cannot be conveniently followed by TLC.

I have yet to have a bad product from HiMedia - the tosyl chloride is fine. Also, the free shipping makes HiMedia products excellent bargains.

Trimethyl phospate is a poor alkylation agent by itself. However, when iodide catalysis is used it is quite good since the active methylation agent is methyl iodide.

BTW- Trimethylorthoformate is available from HiMedia if one wants to methylate p-toluenesulfonic acid. There is a least one paper on this that I know of. I am not aware of any direct phenolic methylation with trimethylorthoformate.

AvB

[Edited on 12-9-2018 by AvBaeyer]

Pumukli - 12-9-2018 at 00:08

Would methyl bisulfate work for e.g. methylating phenols? I read that methylations with DMS can be tailored to use both methyls of the reagent so I assume methyl-H-sulfate could be used too.

Its synthesis seems to be fairly simple, it's inexpensive, and probably more forgiving than the other (DMS, MeI) big-gun methylating agents.

Depending on the substrate I'd probably run a few trials to evaluate its effectiveness.

JJay - 12-9-2018 at 00:31

I'm pretty sure that it would work to some extent for methylating phenols, but I don't think it works without heat and an extended reaction time, and the yields look like they would be lousy. Assuming that the first methyl group methylates quantitatively, the second methyl group would only methylate in something like 45% yield on the remaining phenol here: http://www.orgsyn.org/demo.aspx?prep=CV1P0058

I wouldn't recommend it.

Loptr - 12-9-2018 at 02:10

Trimethylsulfonium bromide and its use for the methylation of phenols
http://www.sciencemadness.org/talk/viewthread.php?tid=15704

[Edited on 12-9-2018 by Loptr]

clearly_not_atara - 12-9-2018 at 09:05

Quote:

I am not aware of any direct phenolic methylation with trimethylorthoformate.

After extensive searching one time I did find a paper which methylates phenols with MeO3Me. However, this occurs at several atmospheres and well above 100 C, only on activated phenols. It probably isn't practical.

For phenols tetramethylammonium salts may be worth considering. Trimethylamine hydrochloride is a lot easier to synthesize than you think:

http://www.orgsyn.org/demo.aspx?prep=CV1P0531

The other methylating agent that always sticks in my head are the mixed methyl-alkyl carbonates, which preferentially transfer methyl. Dimethyl carbonate boils too low to be used for methylation at atmospheric pressure, but methyl octyl carbonate should work fine. I wonder if you could make it from methyl carbamate like this:

MeOCONH2 + C8H17OH + [cat. ZnO] >> MeOCO(OC8H17) + NH3 (g)

[Edited on 12-9-2018 by clearly_not_atara]

Pumukli - 12-9-2018 at 11:11

These tetraalkyl-ammoniums seems interesting, but how could you reach tetramethylammonium from trimethylamine? :-)

Wouldn't you need something like MeI first? :-)

clearly_not_atara - 12-9-2018 at 11:24

You can make Me4NBr from MeBr + Me3N; the former compound is much easier to handle and store because MeBr boils at 5 C. If it weren't for the low boiling point of methyl bromide, there would never have been any discussions on methylating agents in the first place -- alkylations with every other group just use the bromides.

You can also make Me4NBr from MeBr + NH3, but this is highly impractical because ammonia boils extremely low, four times as much bromide is required, and you need a catalytic base. If you start with trimethylamine no base is required (because no H atoms are displaced) and the reagents are liquid at 0 C. So having an easy synthesis of trimethylamine makes Me4N+ much more accessible.

EDIT:
Dimethyl sulfone, widely sold as a health supplement), is cleaved to methyl sulfinate by a Birch reduction:

https://pubs.acs.org/doi/abs/10.1021/ja01496a050?journalCode...

Oxidation of the resulting methyl sulfinate with bromine should yield the corresponding sulfonyl bromide, from which the methyl ester is available. Although... I am a little unclear on the safety of this one. Methyl methanesulfonate?

For phenol methylation specifically, I wonder if you can't use sulfonyl halides to perform it "in reverse" by forming the aryl sulfonate and rxn with sodium methoxide.

[Edited on 13-9-2018 by clearly_not_atara]

JJay - 13-9-2018 at 07:00

AvBayer's suggestion of methyl tosylate has considerable appeal. It looks like a fun and interesting chemical to play with, and it's not impractical.

Preparation of dimethyl sulfate is not a great experiment for someone who doesn't have years of experience snuffing out fires and dealing with deadly gas leaks. Even some of the greatest amateur chemists think it is too dangerous to attempt. I'm going to have to work out an appropriate fume hood... my biggest concern is containing possible spills....

Pumukli - 13-9-2018 at 11:03

Tosylates are good methylating agents IF you can make/buy the chlorides. Unfortunately there is no simple route to the methylester from the free acid. If there was, well, it would really help the amateurs. They would be easily recyclable too.

By the way, you can go "fancy" with the sulfonates: electron withdrawing groups in the p-position of their benzene ring makes even better alkylating agents. Once I saw a comparison of reaction rates of various substituted sulfonates and there was a more than tenfold increase of it from p-methyl to p-chloro (or p-nitro?).

Sigmatropic - 13-9-2018 at 11:10

I remember some of the higher tosylates can be made by using a trialkyl borate (specific example was with tosylic acid and tributyl borate). Perhaps the methyl compound can be made that way but alot changes. Most importantly the boiling points of the borate, the alcohol and their possible azeotropes with water. Hence the way in which you can force these reaction changes dramatically. I will try to find references tomorrow.

Loptr - 13-9-2018 at 16:49

Quote: Originally posted by Pumukli

Tosylates are good methylating agents IF you can make/buy the chlorides. Unfortunately there is no simple route to the methylester from the free acid. If there was, well, it would really help the amateurs. They would be easily recyclable too.

By the way, you can go "fancy" with the sulfonates: electron withdrawing groups in the p-position of their benzene ring makes even better alkylating agents. Once I saw a comparison of reaction rates of various substituted sulfonates and there was a more than tenfold increase of it from p-methyl to p-chloro (or p-nitro?).

Cobalt(II) catalyzed tosylation of alcohols with p-toluenesulfonic acid
Subbarayan Velusamy, J. S. Kiran Kumar and T. Punniyamurthy*
Department of Chemistry, Indian Institute of Technology Guwahati, Guwahati 781039, India
Received 23 July 2003; revised 7 October 2003; accepted 17 October 2003

Attachment: Cobalt(II) catalyzed tosylation of alcohols with p-toluenesulfonic acid.pdf (238kB)
This file has been downloaded 362 times

Quote:

Alcohol (3 mmol), p-TsOH (3 mmol) and CoCl2*6H2O (5 mol %) were dissolved in ClCH2CH2Cl (10 mL) and the solution was stirred under reflux (ca. 80 C) for the appropriate time (see Table 2). The reaction mixture was then allowed to cool to ambient temperature and diethyl ether (50 mL) was added. The catalyst was removed by filtration and the filtrate was washed successively with saturated NaHCO3 solution (3 · 10 mL), brine (2 · 10 mL), and water (1 · 10 mL). Drying (Na2SO4) and evaporation of the solvent on a rotary evaporator afforded a residue, which was passed through a short pad of silica gel using a mixture of ethyl acetate and hexane as eluent to afford the analytically pure tosylate.

[Edited on 14-9-2018 by Loptr]

AJKOER - 15-9-2018 at 04:06

Per a source, try following the general method (regrettably reversible) of obtaining methyl sulphate from the action of concentrated sulphuric acid acting on CH3OH but use methanol vapors at low temperature (like -40 C, see http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.530...). To quote:

"The uptake of gas-phase methanol by liquid sulfuric acid has been investigated over the composition range of 40 - 85 wt % H2S04 and between the temperatures of 210 - 235 K.
.....
While reversible uptake was the primary mechanism at low acid concentrations, irreversible reaction between methanol and sulfuric acid at low temperatures, forming methyl hydrogen sulfate and dimethylsulfate, was observed at all concentrations. Above 65 wt % H2S04, more than 90 % of uptake was found to be reactive."

My summary reaction:

CH3OH (g) + H2SO4 (l) --(-40 C)--> CH3HSO4 (l) + H2O (s)

CH3HSO4 (l) + CH3OH (l) --> (CH3)2SO4 (l) + H2O (s) (See Eq 8 )
----------------------------------------------------------------------------

Here is also an experimental photolysis path (to be tested/verified) for those restrained to more ambient temperature ranges. It employs methanol under UV photolysis resulting in the formation of the methyl radical and the hydroxyl radical. As such, I would recommend trying to obtain methyl sulphate from the action of concentrated persulphuric acid acting on methanol, in the further presence of UV light. Expected reactions:

CH3OH + hv --> •CH3 + •OH (and other products, see http://chemistry.emory.edu/faculty/widicusweaver/photolysis.... )

H2S2O8 = H+ + HS2O8-

HS2O8- = H+ + S2O8(2-)

S2O8(2-) + hv --> •SO4- + •SO4- (see, for example, https://www.sciencedirect.com/science/article/pii/S004565350... )

H+ + •SO4- = •HSO4 (pKa < 0)

•CH3 + •HSO4 --> CH3HSO4

CH3HSO4 + CH3OH --> (CH3)2SO4 + H2O

CH3HSO4 + H2O = CH3OH + H2SO4
.....

Hopefully by restricting the presence of water, the last reaction will be limited. Some unwanted reactions, likely detracting from yield, include:

•OH + •HSO4 --> H2O + SO4(2-)

•OH + SO4(2-) = OH- + •SO4-

OH- + H+ = H2O
---------------------------------------

Here is also a photolysis path to methyl iodide. Starting with iodide (from say HI) , the hydroxyl radical forms the monatomic iodine radical as does the direct action of light on iodide. The methyl and the iodine radical could then theoretically combine creating methyl iodide. Reactions including some other products:

HI = H+ + I-

CH3OH + hv --> •CH3 + •OH

I- + hv --> •I + e-

H+ + e- = •H

I- + •OH --> •I + OH-

•H + •OH --> H2O + photon

•H + •I --> HI

•CH3 + •I --> CH3I

•CH3 + •H --> CH4

……..

and other products.

[Edited on 15-9-2018 by AJKOER]

clearly_not_atara - 15-9-2018 at 14:06

Loptr wins the thread. That is incredible. I think you'll want an excess of methanol to compensate for losses. Or maybe you can just perform the rxn in methanol?

Quote:

Once I saw a comparison of reaction rates of various substituted sulfonates and there was a more than tenfold increase of it from p-methyl to p-chloro (or p-nitro?).

Para-bromobenzenesulfonates, affectionately "brosylates", are the most common IIRC.

[Edited on 15-9-2018 by clearly_not_atara]

Loptr - 18-4-2019 at 16:04

Alright, it now looks like dimethyl sulfate is now available in the USA.

I placed an order a few days ago and am still waiting for a shipment notification. And it's cheap, too! Cost me $39.62 in total for 500ml, and that's including shipping.

To be honest, I will be surprised if it arrives.

[Edited on 19-4-2019 by Loptr]