Sciencemadness Discussion Board - Iodomethane prep.

Iodomethane prep.

Pages: 1 2

Arrhenius - 19-6-2009 at 00:10

Iodomethane is widely used as an 'active' methylating agent for the methylation of carbon, nitrogen, oxygen, sulfur and phosphorus. It can also be used to prepare the Corey-Chaykovski reagent. Published procedures generally utilize phosphorus trichloride, which is generally unavailable to the amateur. I have prepared iodomethane on 30mmol scale in 57% yield by direct halogenation of methanol.

Introduction
Alkyl halides can generally be prepared from the corresponding alcohol by treatment with the hydrogen halide.(ref. 1) The hydrogen halide may be prepared in situ from its salt and a strong acid.(ref. 2) Phosphoric acid was employed to produce HI:

H3PO4 + KI → HI + KH2PO3

While this equilibrium does not favor HI, it can be driven by consuming HI in the following dehydration reaction:

HI + ROH → RI + H2O

This reaction is entropically driven by the removal of the alkyl iodide.

Experimental
I combined potassium iodide (5.02g, 30mmol), methanol (24ml, 600mmol) and 85% phosphoric acid (40ml, 690mmol) in a round bottom flask and performed a simple distillation at a relatively slow pace (condensor at -5ºC). At reflux, the salt is entirely dissolved and the reaction mixture is red. I distilled until the pot residue turned clear and no further product came over. BP 50-65ºC. I dilluted the distillate with 10ml of dilute aqueous sodium thiosulfate, removed the red lower layer with a pasteur pipet and dried it with calcium chloride.

Start

Finish

Product (calcium chloride floating on top)

Results
Crude iodomethane (2.42g) was obtained as a dark red liquid in 57% of theory based on potassium iodide.

Discussion
The product should be distilled to improve purity, but the scale presented here is prohibitively small. I plan to scale up the reaction. This same procedure may be used to prepare alkyl bromides in excellent yield; procedures are widely available online.(ref. 3)

References
(1) Sidgwick, NV "The Chemical Elements and Their Compounds" p.1183 (available in sciencemadness library)
(2)Organic Syntheses, Coll. Vol. 4, p.323 (1963); Vol. 31, p.31 (1951). Link
(3) http://www.books-about-california.com/Pages/Experimental_Org...

[Edited on 19-6-2009 by Arrhenius]

Sauron - 19-6-2009 at 05:21

The classical preps of MeI as per Org Syn are

1. The red-P catalyzed elemental iodination of methanol, which has the drawback of requiring a few gams of red P so often unavailable these days

2. Reaction of KI with simethyl sulfate. DMS is nasty stuff.

Your prep sidesteps these neatly so I look forward to your scaleup.

entropy51 - 19-6-2009 at 06:05

Another method which requires neither P nor H3PO4 is as follows:

Dissolve KI in MeOH (solubility 1 gm per 8 mL) and gas with HCl while cooling and stirring. KCl, which is less soluble than KI, will precipiate out.

When saturated with HCl, distill off MeI with ice water in condenser. The MeOH that distills with the MeI can be washed out with small amounts of ice water. The product is normally only slightly tinged with brown, but darkens on storage.

Yield is 50-60%.

basstabone - 19-6-2009 at 09:18

Entropy51: Does the HCl need to be anhydrous?

H3PO4 is very easy to obtain making iodomethane much more available to the home chemist. Thanks a lot Arrehnius!

Nicodem - 19-6-2009 at 11:44

Well done Arrhenius!

I only have two comments. I think you should measure the boiling point (you can use a test tube and a thermometer), just so that your contribution has some analytical data to confirm product identity. I don't really have doubts that what you got is some fairly pure methyl iodide. But you know the rules of science...
The other comment is actually a question. Can I move this thread to the Prepublication section? Alternatively, we can wait for you to do a larger scale experiment with product distillation and all, so you make that one in Prepublication.

Using CaCl2 is a good idea since it removes the last traces of methanol as well. Store your MeI in a glass container with a teflon lined stopper. It rapidly goes trough the usual polypropylene containers and stoppers. Also, keep it away from any residence places - it is after all a carcinogen to treat will all due respect.

Some people had success in using alcoholic solutions of H2SO4 to prepare alkyl iodides from KI or NaI. Substantial oxidation to iodine occurs, but the yields are still satisfactory.

Entropy51, are those your experimental results or do you have a reference for it?

Fleaker - 19-6-2009 at 12:37

Arrhenius, good work! You will notice that your yield will go up when you scale it up. Also, that dark colour is dissolved I2 (shake it with metabisulfite and it will clarify) formed because you didn't purge your system with argon.

This is a much better procedure than using 57% HI (which is hard to get and yields a little worse).

You would be wise to distill it again before use.

benzylchloride1 - 19-6-2009 at 12:41

Great work, Arrhenius, I store my iodomethane over copper wire to prevent it from decomposing. Iodomethane quickly turns from colorless to a brownish iodine like color with out the copper wire. Dont use mercury because iodomethane reacts with it to form methyl mercuric iodide in small quantities; deadly poisonous.

[Edited on 19-6-2009 by benzylchloride1]

Sedit - 19-6-2009 at 12:45

Without a bisulfate or thiosulfate wash is one even able to determine the true yeild of MeI with any accuracy when the I2 is dissolved in it since they only differ in molar weight by a factor of less then 20?

Nice prep Arrhenius can't wait to see more. Most preps using HI seem a little daunting to people such as myself due to the status of Iodine in todays world so its nice to see that side steped abit

entropy51 - 19-6-2009 at 14:01

basstabone, I have not dried the HCl but rather used it directly from the generator containing NaCl + H2SO4.

Nicodem, those are experimental results. I have run the prep several times on a small scale, usually 20 gm of KI. I found a mention of it in one of the 19th century textbooks, but I can't remember which one. I thought it might be Roscoe's Treatise of Chemistry, but it doesn't seem to be in there. I will post a reference if I can find it, but it was not a detailed prep, just a sentence or two. Sorry for the lack of rigor, but it has worked for me and I thought it worth passing along, especially since there seems to be no free iodine in the reaction or product.

chloric1 - 19-6-2009 at 15:10

I remember one time going into the liquor store buying a fifth of everclear and taking it home and nearly gasing myself with ethyl bromide via the alcoholic/H2SO4 sodium bromide process.

Ah! Those youthfull days of dabbling.

Arrhenius - 19-6-2009 at 21:34

Nicodem: I'll report the boiling point when I scale up and re-distill the product. Or I can do a micro boiling point determination. I chose phosphoric acid to avoid oxidizing the HI, but at this point I have no idea if it improves yields. I think a slightly more concentrated phosphoric acid would be useful, but I'm not familiar with the physical properties here. I'll write a 'publication' once i've run it a couple of times. Feel free to move it now if you'd like.

Fleaker: I'll try metabisulfite.

benzylchloride: yes, I added some copper shot to the product.

I shall take an IR to confirm the structure based on the ~1200cm^-1 band. Everyone satisfied with that? I can't seem to find a literature value for this... but I'm exhausted. Maybe someone else can look if they're bored.

Sauron - 19-6-2009 at 23:31

The density is a useful constant and easy to take once you have a reasonable qty. A tared volumetric flask is all that is needed and a scale.

Refractive index is nice.

But you may not have a refractometer at hand.

I do not doubt that you have MeI there, what else can it be from those reactants, and is a heavy highly refractive colatile liquid colored by free I2?

sparkgap - 20-6-2009 at 00:52

There's a liquid film IR spectrum of methyl iodide @ SDBS; you can compare what you get with what's there.

Nice work, but wouldn't hot phosphoric acid react with the glass?

sparky (~_~)

Nicodem - 20-6-2009 at 00:59

I think a constant boiling point taken from the second distillation in your planed upscaling experiment will be more than enough to satisfy the scientific scrutiny. But if you would still want to take an IR there is a iodomethane spectra at SDBS.
I will keep this thread here so that new ideas are discussed, and you will open a full report in the Prepublication section later when you will have the opportunity to do the scale up experiment. I have a suggestion for when you do that: use a distillation column, at least a simple Vigreux one or anything you can get hold of.

The iodine content in the product can't really be relevant since many organic iodo compounds slightly decompose during distillation, but usually only as much as to give them the dark colour. A small piece of copper or silver wire is indeed used to stabilize many organic iodides and bromides in commercial products. Organic iodides also must be kept in dark.

Entropy51, thanks for sharing your experience. Gasing with HCl might not be particularly practical, but it is nevertheless a useful alternative. Another alternative would be to first reflux NaHSO4 in methanol for the time required to form the sodium methylsulfate, then add NaI or KI and fractionate methyl iodide as it forms. Sodium hydrogen sulfate is easily available, probably even more easily than H3PO4.

woelen - 20-6-2009 at 01:51

This looks very interesting and it is a thing I also would like to try. I have almost 1 liter of 85% H3PO4 (somewhat brown, technical grade). I can even add a little P4O10 to remove more of its water.

Before I do so, however, I would like to know whether this does not attack the glassware. I have a micro distillation setup with 25 ml flasks and coolers and I paid a lot for this and don't want to have the glass fogged, because it is attacked by the H3PO4.

Nicodem - 20-6-2009 at 02:14

H3PO4 only starts to seriously etch glass at >150°C (still some etching can be noticed even at 120°C already). Its methanolic solution obviously can not do any damage to the glass due to the low reflux temperature and dillution. Woelen, instead of repeating this experiment, you could prepare MeI by using P4O10 instead of H3PO4 (it is always better to make new alternatives than repeating old stuff). It might give a somewhat better yields, but beware, the quenching of P4O10 by adding it to methanol is highly exothermic and cooling must be used at this stage. But still I would like to see the comparison of results when using H3PO4 vs. P4O10.

Sauron - 20-6-2009 at 04:53

While P4O10 is not particularly dear, it is quitw a bit costlier than tech. orthophosphoric acid. So any increase in yield would be more than offset by cost, and MeI is already going to be a bit pricey due to the cost of KI or (a bit less) NaI.

This renders the use of the anhydride rather academic, wouldn't you agree? At least on a prep scale, and what else are we talking about here?

ItalianXKem - 20-6-2009 at 06:27

you remember acetone and bleach (sodium hypochlorite) = chloroform ?
that synthesis can be simile
iodomethane is CH3I
we have acetone (propan-2-one) CH3COCH3

2 NaIO + CH3COCH3 = Na2CO3 + 2 CH3I

but sodium hypoiodite is rare and to be have lot of money to buy this salt
but , iodium tincture (methanol , ethanol , KI and I2) is very easy to find and buy
first need to separate I to iodium tincture
second make te necessary compound of I
third reaction between compost of I and a alkylating agent
now I don't know

good luck !

Sauron - 20-6-2009 at 06:43

That reaction proces CHI3 not CH3I

I see you have not learned any chemistry since you were last here.

You cannot do a partial haloform reaction, the reaction will proceed as long as there are alpha hydrogens to replace.

Hypoiodite is never bought but is made in situ from Ki and NaOH

But you will never get anything but CHI3 - never CH3I.

[Edited on 20-6-2009 by Sauron]

Nicodem - 20-6-2009 at 07:34

Quote: Originally posted by Sauron

While P4O10 is not particularly dear, it is quitw a bit costlier than tech. orthophosphoric acid. So any increase in yield would be more than offset by cost, and MeI is already going to be a bit pricey due to the cost of KI or (a bit less) NaI.

This renders the use of the anhydride rather academic, wouldn't you agree? At least on a prep scale, and what else are we talking about here?

Well, since H3PO4 is industrially made by quenching P4O10 it would be a bit unusual for it to be any cheaper than its precursor. Actually the major chemicals dealer S***a sells H3PO4 for 2.9 Eur per mol of P and P4O10 for 2.6 Eur. I think this fairly reflects the cost of hydration. But who cares about the price difference? What does it have to do with the topic? Not to even mention that phosphorus pentoxide is an ubiquitous chemical (used as general desiccant for vacuum desiccators) in every lab, while H3PO4 is a bit less common (though in many parts of the world its diluted solutions can at least be bought in stores). But not even its availability is an issue here. The point is in that this forum is about art and science so researching is the goal. And since Woelen has both phosphorus pentoxide and the motivation to do experiments, then why would he care about economics?

Woelen, keep in mind that if using P4O10 you must not use such a huge excess of methanol/acid as in the original post. You should use a 4 : 15 : 1 molar ratio of KI(NaI) vs. MeOH vs. P4O10. One mol of P4O10 can activate 6 mols of MeOH, 4 for sure, the rest depending on the temperature (because of thermodinamic uphill due to the low pKa2 of phosphoric acid). So, using 4 eqivalents of the iodide would assure its complete conversion while using a minimum of 15 equivalents of MeOH would make the reaction mixture less of a slurry and the quenching of P4O10 less dramatic (I would actually suggest to use even more of MeOH to make the slurry easier to stir and the iodide salt more soluble). The reaction should occur at lower temperatures since the methyl (poly)phosphates surely react faster with the iodide.

Arrhenius, next time you do this, if you have a two neck flask, please take temperature measurements of the reaction mixture as well, so that we can see at which point the reaction proceeds optimally with MeI distilling over. If you don't have the required flask, then at least follow the oil bath temperature.

Edit: Corrected the suggested KI(NaI) vs. MeOH vs. P4O10 ratio. Previously I forgot we are dealing with iodides and not HI.

[Edited on 20/6/2009 by Nicodem]

Sauron - 20-6-2009 at 07:58

If you want to make 25 g is matters a little, if you want to make 2,5 Kg it matters a lot. Picking an overpriced supplier like Sigma and then normalizing on price per mol of P is rather arbitrary. Tech grade H3PO4 probably originatess from rock phosphaates industrially, aand you can buy a 20 L pail of it for the price of a Kg of lab grade P4O10. You know as well as I do that it used as rust remover snd is sold fr that in every hardware store. Sigma doubtless is selling ACS grade. So who are you kidding?

I am osten criticized for being a spendthrift who buys reagents rather than make his own. Well, when I do make my own I don't want to spend more than I could buy the same for.

There are lots of members who know how to squeeze a nikle till the buffalo shits, yet here you are telling me that being concerned with cost is somehow cocnstrart to the forum's principles?

Really, that is the least tenable position I have ever seen tou take!

Sauron - 20-6-2009 at 08:27

I just looked at P4O10 on Acros and prices for various grades run from $61 a Kg for lab grade to $125 Kg for ACS analytical. Illogically 2.5 Kg of the lab grade was $197 which is a lot more than 2.5 x $61, go figure.

Anyway what matters is the stoichiometry of the respective reactions and their yields, and only then can the cost of a mol of product per whatever number of mols of the reactant in question were reqiored, be calculated and compared.

The reaction between H3PO4 and KI produces potassium phosphate and 3 moles of HI and the HI then reacts with methanol

The reaction between methanol and P4O10 produces I suspect a mixture of methyl phosphate esters and these esters react with KI to produce MeI and potassium phosphates.

Both reactions appear to me to be sequential, anyway.

I will be surprised if the P4O10 reaction goes cleanly to O=P(OMe)3 asI would predict a sigmificant amount of the mono and di methyl esters so this is going to complicate things isn't it?

[Edited on 20-6-2009 by Sauron]

Nicodem - 20-6-2009 at 09:46

Quote: Originally posted by Sauron

I will be surprised if the P4O10 reaction goes cleanly to O=P(OMe)3 asI would predict a sigmificant amount of the mono and di methyl esters so this is going to complicate things isn't it?

I never said that trimethyl phosphate must form. The only thing that is needed is to make the methyl group in MeOH electrophilic enough for it to react with the iodide anion (which is pretty easy given the strong nucleophilicity of the iodide). Whether this is done by protonation (using H3PO4 as reaction medium) or by phosphate ester formation (like when using P4O5) matter only to the reaction conditions and not in regard to the reaction products. I can't see how the formation of a mixture of mono-, di- and trimethyl phosphates instead of just trimethyl phosphate could ever matter for the reaction outcome. All methyl groups in all the esters are similarly electrophilic and all the esters are in dynamic equilibrium among themselves and H3PO4.

Quote:

The reaction between H3PO4 and KI produces potassium phosphate and 3 moles of HI and the HI then reacts with methanol

You can not obtain 3 mols of MeI from 1 mol H3PO4. The H3PO4 must be used in a huge excess as to be able to protonate methanol and the inorganic product can not be any other phosphate but KH2PO4 (check the pKa1 to pKa3 of H3PO4). Note that Arrhenius used 23 equivalents of H3PO4 vs. KI. That was not a bad idea given he used 85% H3PO4, but even with 100% H3PO4 you can not get beyond the theoretically allowed 1:1 ratio.

[Edited on 20/6/2009 by Nicodem]

Arrhenius - 20-6-2009 at 09:46

Sparkgap: Thanks for the reference. My IR doesn't use salt-plates or pellets, so I doubt I can reproduce that spectrum, but I'll see what I can do. Would be fun, ya?

Nicodem: Thank you for the notes. For the first part of the reaction I had the thermometer extended into the reaction mix. The solution temp is above 75ºC, probably 100ºC if I had to guess at this point. The oil bath was too hot to touch. I'll use two thermometers next time. I had used a large excess of methanol because it does co-distill to an appreciable extent. When the product is worked up from thiosulfate/metabisulfite the methanol is largely removed.

Sauron: I buy my reagents. However, concentrated phosphoric acid is available in some stores in the US. I seriously doubt KI is, but this synthesis should be accessible to most amateurs who are un-paranoid

enough to buy KI. My goal is only to beat the price of $80 for 50g of MeI. This is the price from the only place I could find it (that would sell it to me). I think phosphoric acid can work pretty well here.

Woelen: Try following the OrgSyn prep a little more closely; the only caveat being that the temperatures you have to work with are considerably lower. Let me know how it works. I would suggest atleast a 3x scaleup from what I did above. The amount of phosphoric acid and methanol I used was arbitrary. I used technical phosphoric acid, and I doubt the brown color in yours will cause any problems.

P.S. I particularly like the catalytic hydrogenation used to prepare hexanediol in the OrgSyn prep. Can't wait to have a hydrogenation bomb that can do that!! haha!

[Edited on 20-6-2009 by Arrhenius]

Sauron - 20-6-2009 at 14:30

All right, Nicodem, H3PO4 is cheap enough to employ in such a large excess. Particularly the technical grade is. The lab grade is not so bad, the ACS grade is 2X lab grade and not called for here IMO. So

H3PO4 (excess) + KI -> KH2PO4 + HI

CH3OH + HI -> CH3I + H2O

And we have Arhennius' yield and it is still pretty clear that despite the large excess of acid the major cost is the KI as long as one ddoes not piss money away using ACS acid..

So, can you propose stoichiometry and any excess if required for the reaction using P4O10 instead of H3PO4?

In particular what is molar ratio between P4O10 and MeOH?

Will preformed trimethyl phosphate react with KI to produce MeI and inorganic phosphate?

[Edited on 20-6-2009 by Sauron]

Sauron - 20-6-2009 at 14:44

Arrhenius, I buy reagents whenever possible. Where I live there are a few things restricted, like P. but not I2 and certainly not KI.

I want to make my own MeI to beat the commercial price which is high as you noted and higher here due to shipping and duty. I pay about 2X the European ex-works prices here, so when something is high ex-works it is very high here.

Arrhenius - 20-6-2009 at 16:19

Ya, I'm not really worried about buying stuff. I just need to expediently discover something new so I can look legitimate

maybe publish some home chemistry in Nature or Science.

On another note, here's the IR:

It may be red, but it's the right product. No MeOH in that IR, and it was not re-distilled.

[Edited on 21-6-2009 by Arrhenius]

Sauron - 20-6-2009 at 21:19

Turns out that when quenching P4O10 with methanol, 2 mols each of the monoester and the diester are formed from 1 mol anhydride. This consumes 6 mols MeOH

P4O10 + 6 MeOH -> 2 MeOP(O)(OH)2 + 2 (MeO)2P(O)OH

No water is formed. That equation is balanced, and it is from Ullmann.

Nicodem, will these react with 6 mols KI to give 6 MeI and potassium phosphates?

Is this stoichiometric or are any excesses required?

If no excess needed this could indeed be more economical than using H3PO4.

woelen - 21-6-2009 at 07:45

As preparation for the synth of MeI I did some little experiments.

I tried the quenching of P4O10 with methanol. I took 1 ml of methanol and added a spatula full of P4O10. This gives a strong hissing noise and a turbid liquid with lots of insoluble matter. Not something I expected. I expected a clear liquid and not the slurry I obtained.

Another little experiment I did is dissolving some NaI in methanol (this is amazingly easy, NaI really dissolves well in methanol) and adding this solution to the other test tube with cooled down methanol/P4O10 mix. No apparent reaction occurs (and I did not expect that). I kept the test tube for 10 minutes or so, and still no apparent change, it remained turbid and white.
Then I took it with me in a sunny place and in the sunshine the liquid turns brown quickly. You really can see the color intensifying from colorless through shades of golden yellow to brown.

I have to do the distillation outside (due to the carcinogenic nature of MeI) and now I decided to postpone the experiment to an evening during twilight, when I have the time. Doing it during the daytime with sunlight will lead to very fast decomposition and loss of yield and very impure product.

I decided to do the experiment with 85% H3PO4, to which some P4O10 is added. I do not want to spend too much P4O10 (which is not at all a common chemical for me, while H3PO4 can be purchased easily). It is not a matter of financial things, but a matter of availability. For a real lab other economic rules apply, but for a home chemist there is the additional restriction of availability.

Sauron - 21-6-2009 at 08:12

Woelen, we have the stoichiometry from Ullmann that calls for 1 eq P4O10 and 6 of methanol. We do not yet have Nicodem's sage advice as to whether or not any excess(es) are required, compare the H3PO4 reaction.

Therefore, how do we evaluate 1 mk methanol and "a spatula of" P4O10 to anything? How many g or mmol is a spatula? At least we can calculate how many mmol 1 ml methanol is. FW 32, d 0.791 so you had 40 mmol methanol and therefore, assuming stoichiometry per Ullmann, you required 6.75 mmol P4O10 FW 285 so 1.92 g.

NaI has FW 150 so the stoichiomeytu calls for 4o mmol of that. But till we know how much that spatula was, we are on terra incognita.

[Edited on 21-6-2009 by Sauron]

Fleaker - 21-6-2009 at 09:20

P4O10 + 6 MeOH -> 2 MeOP(O)(OH)2 + 2 (MeO)2P(O)OH
is a very efficient reaction... thanks for taking the time Sauron.

P4O10 is much cheaper than the same amount of oleum or chlorosulfonic acid (P4O10 in quite acceptable grade is something like $25/500g without shipping, but it's available in 20kg sacks for about that much if one cares to look!). It stands to reason that this method should work--it works with the acid anhydride SO3 and methanol but with much greater ferocity. When I was working on DMS and DES synthesis, some of the post-reaction mixes were ink-black from the carbon formed (less of a problem when ether was bubbled through it though).

Sauron, you and I both know it's a trial to get a mass on P4O10 unless one has a dry box or takes elaborate precautions to preclude moisture--how can we fault Woelen on that?

I expected some of Woelen's results as HI is itself light sensitive (as is MeI) but nonetheless I greatly appreciate hearing about the turbidity of the solution--perhaps extra methanol may be required as a reaction medium? Also, if this phosphate ester business parallels that of its sulfate cousins, then perhaps one might actually isolate the ester(s) since at least one of them appears to be a solid (just distill off the MeOH) and then mix with solid NaI and dry distill! Then again, since there seems to be such an exotherm from P4O10 addition, it would be wise to take advantage of that when doing a bulk synthesis, and perhaps regulate the reflux of the mixture through addition).

Here is my primary question for Arrhenius: what's in the literature and what isn't? Have you found any papers or work of a similar nature? This seems like Meat and Potatoes organic chemistry that would've been explored (p)ages ago... if not, a definite contribution can be made: if yields are high, this would be a good alternative, especially amateur chemists who want methylating agents but don't want the hassle and stigma of using P4 and I2.

entropy51 - 21-6-2009 at 10:05

I have tried to attach a literature paper regarding alkyl iodides via the H3PO4 route.

I'm a little suprised that this prep is being discussed as if it's something new. There is quite a bit about it here: https://www.sciencemadness.org/whisper/viewthread.php?tid=16...

Attachment: halides by H3PO4 (Stone).pdf (326kB)
This file has been downloaded 945 times

[Edited on 21-6-2009 by entropy51]

Jor - 21-6-2009 at 10:08

People always told me that MeI is a horrible carcinogen. This keeps me from trying to make some. Are these rumours true?

entropy51 - 21-6-2009 at 10:30

Jor, compared to what? The list of known carcinogens includes Cr(VI), Cd, Ni, benzene, formaldehyde, benzidine, o-toluidine and on and on. The probable carcinogens are an even longer list. And those are only the ones we know about.

http://www.cancer.org/docroot/PED/content/PED_1_3x_Known_and...

I think any amateur (or professional) chemist who works without adequate ventilation and protective gear such as gloves and sometimes a full face shield is playing with fire. I treat MeI with a great deal of respect and carefully minimize my exposure to it, but if you have a reliable hood I don't think you need to fear it anymore than many of the other chemicals you probably use without a second thought.

On the other hand, some of the practices that we've recently seen posted on the forum are certain to result in exposures that will cause health problems if repeated such that the exposure become chronic.

Nicodem - 21-6-2009 at 10:31

Jor, MeI is indeed a carcinogen just like many other stuff. This does not mean that people would stop using it, it just means they will not drink or inhale it. Besides, it depends on your cultural background and education whether you will consider it scary as hell or just a normal reagent to be treated with some additional care. Within a fume hood I have no problems working with it (though I'm nevertheless glad that my job does not often call for its use).

Fleaker, there are quite some literature examples about the reaction of alcohols or ethers with H3PO4/iodides and its preparative use for alkyl iodides, but as far as I remember I never saw any example for the preparation of MeI, just higher alkyl iodides. I never did a thorough search to say for sure, but it could just be possible that that Arrhenius was the first to apply it to methanol.
(BTW, thanks to Entropy for the interesting paper)

Quote: Originally posted by Sauron

Woelen, we have the stoichiometry from Ullmann that calls for 1 eq P4O10 and 6 of methanol. We do not yet have Nicodem's sage advice as to whether or not any excess(es) are required, compare the H3PO4 reaction.

I gave my opinion about the optimal stoichiometry a few posts earlier. You probably missed it.
Formally one equivalent of P4O10 activates 6 equivalents of MeOH, however due to the unfavourable pKa2 of H3PO4 the formation of K2HPO4 is thermodynamically less favoured. Thus 4 activated equivalents of MeOH should react easily (at temperatures of 80°C or less with MeI/MeOH distilling out), but the remaining 2 equivalents might need a bit more heating. Since the goal is to optimize yields toward NaI which is the price limiting element, I suggested to use the 4 NaI : 1 P4O10 ratio. I also suggested at least 15 equivalents of MeOH and explained why.

Jor - 21-6-2009 at 10:44

Yes, but I always thought that alkylating agents are a class of compounds with a high carcinogenic potential. It's not very important whether something is carcinogen or not, it matters how potent it is. And IIRC MeI is very potent, just like hydrazines, dioxins, etc.
But I might be wrong.

I might give it's synthesis a try.
How do you effectively destroy MeI? Is aqeous ammonia sufficient?

DJF90 - 21-6-2009 at 11:15

Well yes methyl iodide is a potent methylating agent. Although I dont think its quite as bad as dimethyl sulfate. Just use adequte precautions and I dont see any hassle.

woelen - 21-6-2009 at 11:46

Quote: Originally posted by Sauron

[...]
NaI has FW 150 so the stoichiomeytu calls for 4o mmol of that. But till we know how much that spatula was, we are on terra incognita.

Please don't give this harsh kind of responses

. What I just did is some testing on a small test tube scale to get a feeling of what I can expect and how things behave. I always do that kind of tests before I use larger quantities.

There is nothing quantitative in these little tests nor did I claim anything in that reaction. In this particular case, I just wanted to share my qualitative observations, because they are not what I expected and some of you might also learn something interesting from it (especially the great sensitivity to light).

If you really want to encourage people to try things then please use another tone! If your tone remains the same like it was in that last post of yours, then this will be my last post in this thread.

-------------------------------------------------------------------------------

The amount of P4O10 relative to the amount of MeOH used in my little experiment is small. My P4O10 is a free flowing powder and I estimate that the volume of powder added to the MeOH is less than 1/10 the volume of MeOH. Considering the free flowing properties of the P4O10, quite some air was in the stuff and then you get an impression how little of this material was added, relative to the MeOH.

I now am setting up things for doing the distillation with 6 grams of NaI (0.04 mol) dissolved in 10 ml of methanol and 3 grams of P4O10 added (just over 0.01 mol) in another 10 ml of methanol. This makes a fairly large excess of MeOH, but otherwise I cannot get all the stuff dissolved and the stuff is too much of a slurry. It is getting darker now here, and the time has come to do dark things Sauron - the apprentice of Morgoth, the Lord of Darkness - likes

Jor - 21-6-2009 at 11:52

I am surprised the P4O10 will not dissolve in MeOH.

I will try this too soon, I have about 150g of the stuff. I also have a free flowing powder, but it is extremely hygroscopic, it gets clumpy after seconds in the air.

If your plain goal is MeI, maybe go the red P route. This will be the route I will take if I ever make some MeI.

And yes Sauron, please be nicer. You are rude to people performing experiments. But we never actually hear a report of an experiment you did. If you don´t like it the way woelen does an experiment, try it yourself. I´m sure you have all chemicals needed at hand. Besides it would be nice to see a report with pictures from you performing an experiment

entropy51 - 21-6-2009 at 12:06

Jor, regarding destruction of MeI, strong NH4OH will neutralize residual traces left on glassware. A 1:1 mixture of EtOH and NH4OH is even better IMO. Finally soak the glass in your base bath of 10% KOH in EtOH.

Obviously don't make more MeI than you need. Its low BP (42.5 C) makes it about as volatile as Br2. Small amounts may be reacted with KOH in EtOH as described here:

http://books.google.com/books?id=geK5gjEn3hkC&pg=PA269&a...

Arrhenius - 21-6-2009 at 12:07

Woelen: Interesting results. I've got things set up to do the scale up, but I'm just waiting until I have enough free time.

Entropy: Don't know why, but I've searched the board numerous times and that page has not come up. A few points are brought up in that thread that are reflected in my prep. e.g. don't need to reflux etc. Honestly, I think people on this board need to be more scientific about their work. If they were, the "publications" section would contain a lot more work. Agree?

Jor: I would just evaporate methyl iodide outside to get rid of it. I'm not really concerned about the carcinogenicity. Most highly reactive chemicals are carcinogens (no surprise). Just be careful not to spill a lot, use cold condensors and have good ventilation. I don't have iodine or red phosphorus, so I won't be taking that route.

woelen - 21-6-2009 at 14:20

I did the experiment but now I already can say that it is a total failure.

I took 10 ml of CH3OH and added 6 grams of NaI (40 mmol). With some stirring this could be dissolved easily and just over 10 ml of clear colorless liquid was obtained.

I took another 10 ml of CH3OH and added 3 grams of very finely powdered P4O10 in very small portions (just over 10 mmol). Each time when a portion was added, a very violent reaction occurred with hissing noise. If accidently some larger granules of P4O10 were added to the methanol, then the granule was covered by a dark brown layer, most likely due to charring of the methanol. After adding all P4O10 the liquid was brown, quite hot and somewhat opalescent. It was not as turbid as I expected, based on the observations in the test tube experiment. It might be that due to the higher temperature of the liquid the P4O10 dissolves more easily.

I let the P4O10/methanol mix cool down and then I mixed both liquids. In this way I obtained appr. 25 ml of brown liquid, which is somewhat opalescent.

This liquid was distilled, using a few boiling stones and a heating mantle for heating. A Liebig cooler was used for cooling and the liquid was collected in a small 25 ml flask, which was put in a beaker of icecold water. The material was heated very carefully and slowly a colorless liquid comes over.
Thermometer readout was steady between 63 and 64 C. This is much higher than the boiling point of CH3I. Actually, it is very close to the boiling point of methanol (64.5 C). Probably my thermometer (which has a range from -10 to + 150 C) is a little bit inaccurate and the main constituent of the liquid must be methanol

The liquid I obtained in the collecting flask was perfectly colorless, not any hint of brown could be observed in this liquid. That also is a bad sign.

Altogether this experiment can be considered a 100% failure. Apparently, P4O10 is not suitable for making CH3I from NaI and CH3OH. I can imagine that the methylphosphate esters are too stable and that they do not easily give up their methyl-group for exchange with a sodium ion and an electron.

The reaction with acid is Na(+)I(-) + HA + CH3OH --> Na(+)A(-) + H2O + CH3I
The methyl ester reaction would be Na(+)I(-) + CH3A --> CH3I + Na(+)A(-)

Probably the latter reaction does not occur that easily. So, if someone want to make CH3I then Arrhenius' method with H3PO4 is much better than the method with P4O10.

I did some testing of the colorless liquid I obtained. It is flammable and burns with a pale blue flame, just like methanol, but it is not as easily ignited as methanol. You have to keep the liquid in a flame for a second or so and then it ignites, while methanol can very easily be lit, just by keeping a flame nearby a drop of it.

I also added water to the colorless liquid (1 volume of water added to 1 volume of liquid). This results in formation of a white milk-like liquid, not at all what can be expected with methanol. When more water is added, then the white material also dissolves in the water.

So, I think that I mainly distilled methanol, given the temperature readout of 63 ... 64 C, but that a small amount of another compound distilled with it as well. This could be some phosphate-based compound, or maybe some traces of a iodide-containing compound?

[Edited on 21-6-09 by woelen]

Arrhenius - 21-6-2009 at 15:31

I would try to use the P4O10 to get to 95% phosphoric acid as in the OrgSyn prep. The distillate temp I collected was from ~50-70ºC. But when water is added to the distillate, there is clearly a dense layer that separates. Don't chalk it up as a total failure, we all learned something. One thing you might consider is starting the reaction at 0ºC.

Paddywhacker - 21-6-2009 at 15:42

Real interesting woelen. Negative results teach us something. Maybe, with just P2O5 + methanol, the mixture was just not acidic enough. A little H3PO4 and a reflux might help.

Some time in the future I want to try toluene sulfonic acid as the acid.

Sauron - 21-6-2009 at 16:30

Woelen, grow an epidermis, will you? I made no personal attack against you, at all. I merely commented that your report was vague on vital details. Clearly you have never given a chemical presentation to a university seminar or a chapter meeting of a professional chemical society. My remarks were not "harsh" or (per Jor) "rude" and I see that they had the intended effect in that you did a much better job with the followup.

My suggestion is that the experiment be repeated with a reflux period prior to distillation. Ude a dewar condenser to preclude loss of MeI

Common sense dictates doing this work in a proper hood or outside. If the latter avoid sunlight.

It is well worth noting that in addition to the H3PO4 method, we have at our disposal the method suggested by entrpy, that is bubbling HCl gas into methanolic NaI or KI; and that of Fleaker using sodium bisulfate (is Sani-Slush still the brand name?)

I am also confident that the P4O10 method will be proven out as experience indicates that Nicodem is about as infallible as Mycroft Holmes.

Polyphosphoric acid reacts with methanol to form orthophosphoric acid and the monomethyl ester exclusively

H6P4O13 + 3 MeOH -> 3 MeOP(O)(OH)2 + H3PO4

The PPA is prepared by mixing 2 mols of 100% H3PO4 with half a mol of P4O10 (one mol "p2O5") and the 100% acid is prepared by adding the calculated amount of P4O10 to react with the water in 85% phosphoric acid.

As Nicodem has stated, the monomethyl ester will not react with KI (HI) at the reflux temp of methanol. Might it react with HI formed by adding KI or NaI to the 3:1 mixture of the ester and H3PO4 above, sans MeOH? Higher temperatures will surely be available! At lest in the 100-120 C range before etching of borosilicate becomes too serious.

[Edited on 22-6-2009 by Sauron]

Nicodem - 22-6-2009 at 00:29

Woelen, thanks for your detailed report, I'm sure many members appreciate it. Have you perhaps followed the reaction mixture temperature as well? Quenching P4O10 without cooling in ice might not have been the best option. The solution turning brown is no good sign.
This week I have a bit less work than usually and was thinking of doing some experiments planed for months (or years), so perhaps I will give a try this one too.

I did some literature search (sorry about doing this too late, but I can't do this efficiently during the weekends or else I would do it earlier). There is almost nothing in Beilstein or SciFinder about the nucleophilic substitutions of alkyl phosphates with iodides (this is pretty understandable since there is nearly no practical use for doing this). Only few examples of which there is a reaction between (EtO)3PO and NaI to give EtI abstracted from a chinese journal, and there is a synthesis of dibenzyl phosphates by their reaction with iodides (JACS, 77 (1955) 5354-5357, attached).

Maybe, later when I'll have more time I will also compile references for other methods of MeI preparations starting from methanol (but I can already now say that the only good one is the one starting from NaHSO4 and methanol).

Quote: Originally posted by Sauron

I am also confident that the P4O10 method will be proven out as experience indicates that Nicodem is about as infallible as Mycroft Holmes.

My experience tells me I'm wrong more often than I'm right. It takes experiments to see what was unexpected, what more needs to be taken into account and what is to do to make things work. Ideas are cheap, it is experimental work that is hard and tedious, but this also the only thing that gives scientific results.

Quote:

As Nicodem has stated, the monomethyl ester will not react with KI (HI) at the reflux temp of methanol.

I never stated that. On the contrary, I said that in my opinion all methyl phosphates would react till monodeprotonation is achieved (with NaH2PO4 and monosodium methyl phosphates being the end products). For dideprotonation (yielding Na2HPO4) more extreme conditions would be needed. That is why I said the ratio of NaI vs. P4O10 should not exceed 4 : 1. The results presented in the attached paper confirm this. Once the alkyl phosphate electrophile is deprotonated its electrophilicity drops sharply. In this JACS paper they reacted tribenzyl phosphates so there is no source of protons to reactivate the electrophile for further reaction, while in a mixture of methyl phosphates resulting from dissolving P4O10 in MeOH there is some. But still, I would regard these additional two methyls as lost in thermodynamic action, and consider NaH2PO4 as the end product.

Attachment: Dealkylation and debenzylation of triesters of phosphoric acid.pdf (518kB)
This file has been downloaded 909 times

[Edited on 22/6/2009 by Nicodem]

Sauron - 22-6-2009 at 01:36

Mea culpa. I misunderstood.

So, the 2 mols of monoester are demethylated completely while the 2 mols of diester are demethylated halfway.

In that light, the reacyion of PPA with methanol, which according to Ullnann gives only monomethyl phosphate (3 mols per mol PPA) and H3PO4 may be worth a try after all, as compared with P4O10 + 6 MeOH only 4 of which can be converted to MeI.

I believe the prep of 1 mol PPA from 85% orthophosphoric, will require less than 1 mol P4O10. Eliminating the 15% water: a mol of 199% H3PO4 weighs 98 g so 15% is 14.5 g water. The stoichiometry of quenching P4O10 with water is

P4O10 + 6 H2O -> 4 H3PO4

So we needn't pull up the calculator to see that less than 1/6th mol P4O10 will be needed to suck up that 14.5 g water. A further 1/2 mol is required to mix with 2 mols 100% H3PO4 so the total expenditure of anhydride to prepare 1 mol PPA of formula H6P4O13 is 833 mmol.

Something to bear in mind pending more definite resulta from P4O10.

Looking through that old thread from 2004 cited upthread, a remark was made that the literature reported better yields with H3PO4 only when 95% acid was used along with 2 eq KI. Maddeningly vague and no no refs. However, the source is the Stone & Shechter paper posted upthread. On 1-propanol as substrate and using 2 eq KI and 2.96 eq 95% H3PO$ and 6 hrs reflux they obtained 90% yield.

Molar ratio alcohol:KI:95% acid 1:2:2.96

[Edited on 22-6-2009 by Sauron]

Nicodem - 22-6-2009 at 05:22

I thought it would be better if I do this experiment today or else I would just postpone it indefinitely. So this is what happened…

Preparation of methyl iodide from methanol using NaI and P4O10

In a two neck flask (50 ml) equipped with a thermometer, there was dissolved 3.1 g P4O10 (10 mmol) in 12 ml methanol (300 mmol) by adding it in small portions while stirring intensively on an ice bath. There was lots of hissing when the reagent gets in contact with methanol, but the addition rate was slow enough to keep the temperature bellow 30°C. To this clear solution was added 5.55 g NaI (35 mmol) and the reaction mixture left stirring at room temperature for one hour (this was not intentionally, but there were other things to do). The NaI dissolved only slowly and a clear solution was obtained. A small Vigreux distillation column (1.5 cm wide, 8 cm long) was set on the flask and the reaction mixture heated and stirred on an oil bath. The first drops of distillate came over at 36.3°C while the reaction mixture temperature was 69°C. A crystalline precipitate begun forming in the reaction mixture. Up to the reaction temperature of 72°C the distillation rate was relatively rapid (two drops per second with the distillate coming over bellow 60°C) but then slowly subsided. Meanwhile, more and more solids formed and the reaction mixture became a hard to stir slurry which caused some minor bumping during the heating. When the reaction mixture reached 100°C the heating was discontinued. When distilling stopped, 10 ml of water was added to the almost clear distillate (6.03 g) having only a very slight violet tinge. The heavier phase was separated (3.06 g) and dried with CaCl2 granules which floated on top. This gave 2.8 g of a colourless, very dense oil (56%). The IR spectra recorded on NaCl confirmed the product identity as methyl iodide:

Sauron - 22-6-2009 at 06:32

Let me be first to express my sincere appreciation of your succesful demonstration of this reaction.

Scaling this ti 1 mol P4O10 basis means

310 g P4O10
555 g NaI
1.2 L methanol

c.300 g MeI product

That is acceptable.

[Edited on 22-6-2009 by Sauron]

DJF90 - 22-6-2009 at 06:57

Fantastic. And when do we get to see pictures?

woelen - 22-6-2009 at 10:54

Interesting to see that Nicodem's results are positive. The only main differences with my experiment are
1) Nicodem only used 12 ml of methanol, while I used 20 ml
2) Nicodem left the solution stirring for 1 hour before doing the distillation
3) The addition of P4O10 was done in an ice bath, while I did not do that and my liquid became quite hot.

Which of these three could be the essential difference? I have the feeling that the 1 hour stir before attempting the distillation is the most important, but the heat of the dissolving of P4O10 also could have had negative impact on the experiment.

Good to see that this reaction is possible with P4O10 and MeOH. Maybe I'll try next weekend again in the late evening (it needs to be fairly dark, because I do this experiment outside and don't want daylight on the reaction mix, it is getting dark here not before 22.30 and good twilight is reached at 23.00).

Sauron - 22-6-2009 at 18:44

N. already told you that the rapid mixing of the P4)10 and MeOH in absence of cooling is the likely culprit, as evidenced by the discolorayion you noted.

I'd advise chilling the methanol and mixing slowly in an ice bath.

The question of lag time is trickier. One way to resolve it is to prepare two identical mpxtures, one you react immediately and the other only after an hou

Yjy not try the method of Stone et al? 90%

32 g MeOH (1 mol) 40 ml
300 g NaI (2 mols) or 333 g KI.
c.300 g 95% H3PO4 (2.96 mols - calculate exactly)

Reflux 6 hours on water bath, I suggest an ice-brine Dewar condenser for reflux and a cold trap to snag any MeI vapor that gets past that.

It ought to be clear that this reaction dies not depend on solvating the iodide in methanol.

[Edited on 23-6-2009 by Sauron]

Nicodem - 22-6-2009 at 23:43

Actually I don't have a clue on why Woelen's experiment failed. For one thing, I do not believe the one hour stirring at room temperature does anything important. The rule of thumb about kinetics says that the rate of most reactions doubles every 10K, which essentially means 4 minutes at 60°C does the same thing as 1 hour at 20°C. I'm also far from being sure that the overheating of methanol during the P4O10 quench is to blame for the failure.

Woelen, in my experiment there was no noteworthy oxidation of HI to iodine. This is also kind of logical given that almost no HI is ever present in the reaction mixture (unlike in the acid catalysed method using H3PO4). Also the reaction apparently starts at a lower temperature, which additionally means less oxidation. Since MeI does not decompose nearly as easily under the influence of light and oxygen as HI does, you don't have to worry about sunlight.
I think the yield could be optimized to something like 70%. Up to 10% were surely lost due to mechanical losses from working on a small scale, while some more could be obtained if I would heat some more.

Here are a few references for other MeOH -> MeI transformations:

de Postis, Comptes Rendus Hebdomadaires des Seances de l'Academie des Sciences, 223 (1946) 82: uses Zn, I2 and HCl at 20°C
Vogel, Journal of the Chemical Society (1943) 643: with HI
Dangjan, Zhurnal Obshchei Khimii, 10 (1940) 1671: uses Fe and I2
Dangjan, Zhurnal Obshchei Khimii, 11 (1941) 617 (CA, 1941, 6925): uses Mg and I2 at 40-50°C
Dangjan, Zhurnal Obshchei Khimii, 11 (1941) 1215 (CA, 1945, 4050): uses Al and I2
Fischer, Beckenkamp, Journal fuer Praktische Chemie/Chemiker-Zeitung, 340 (1998) 58-62: uses SO2 and I2 in MeOH

PS: I only now noticed link posted by Entropy to an old thread, where it appears that other forum members prepared MeI via H3PO4 or H2SO4 years ago. I now remember that old thread, but somehow I did not remember it when I read the report by Arrhenius.

woelen - 23-6-2009 at 00:26

The more I read and think about this, the more I have the feeling that actually I did make quite some MeI. Given the fact that my colorless liquid was much less flammable than methanol and that it became milky when water was added I start to think that I actually had quite some MeI, but mixed with methanol.

I was forced in thinking that the experiment failed, because
1) the thermometer readout of the vapor was between 63 and 64 C.
2) the liquid was perfectly colorless

Quote:

Woelen, in my experiment there was no noteworthy oxidation of HI to iodine. This is also kind of logical given that almost no HI is ever present in the reaction mixture (unlike in the acid catalysed method using H3PO4).

Arrhenius' liquid was brown and I was indeed also expecting a brown liquid in my receiver flask, but now I think of it, I understand why this need not be the case when the distillation is done in almost darkness and without acid present.

Unfortunately I burned away all of my colorless liquid, because I considered it of no value, so I cannot test it anymore

.

I'm certainly going to retry next weekend, I have plenty of NaI and MeOH and sacrificing another 3 grams of P4O10 is worth the experience I gain from this. I had the good luck that I could buy 400 grams of NaI for just EUR 5 + shipping off eBay from some one-time seller who only received a single bid (my bid

).

[Edited on 23-6-09 by woelen]

Sauron - 23-6-2009 at 06:04

I have the Vogel paper from Nicodem's list, this is a usefulgeneral guide to alkyl halides for amassing a list of constants.

Vogel's general procedure for iodides is on a 500 mmol scale and employs constant boiling HI in qty 300 g and a 6 hr reflux. While the yields with ethanol and higher are quite good (85-90+%) that for methanol is 60%.

Unless we are to assume some special mechanistic aspect, I suppose the low yield may be due to evaporative losses.

Compare to the method of H,Stone and H,Shechter (J Org Chem, 15m 491 (1950) posted upthread. They make HI on situ from 2 mols KI and 3 mols 95% H3PO4 per mol ROH and reflux 6 hours just like Vogel but trport 90% yield with many alcohols and glycols (but did not do methanol or ethanol). Their main observation was that 95% H3PO4 worked much better than 85%.

We could prepare HI azeotrope, mix 2 eq of that with 1 eq 95% H3PO4 abd 1 eq methanol and either reflux on a water bath 6 hrs or run the reaction in a Parr pressure bottle in water bath, then let cool and finally chill, transfer and distill. The latter procedure precludes evaporative loss.

This replicated Stone and Schecter without all that KH2PO4 ppt.

[Edited on 23-6-2009 by Sauron]

Attachment: jr9430000636.pdf (1.3MB)
This file has been downloaded 1141 times

Sauron - 23-6-2009 at 23:36

Take note of this elegant prep of anhydrous HI from I2 and tetralin at 200 C, 95% yield and 99.5% purity

Flaky solid iodine of 40 g was dissolved in tetrahydronaphthalene (Tetralin) of 160 g charged in a flask of 500 ml at 40°C to prepare a tetrahydronaphthalene solution of iodine. A flask of 500 ml was charged with tetrahydronaphthalene of 40 g and heated to 200°C. while stirring. The iodine solution prepared above was continuously added thereto over a period of 2 hours while maintaining the above temperature to react them. Crude hydrogen iodide gas generated as the reaction went on was introduced into a 10% sodium hydroxide aqueous solution of 1 liter to absorb the whole amount thereof. A weight change in this aqueous solution was measured with the lapse of time, and the end point of the first reaction was set at the point where the change thereof was not observed. The yield of the crude hydrogen iodide was 94.6%, and the purity thereof was 99.5% or more.

US Patent 5,693,306

The HI could be bubbled into chilled MeOH - I will try to look up the solubility. Refluxing (if necessary in a pressure vessel) then distilling will produce MeI (and water)

I do realize that many of us cannot get I2, but I can and this method appeals to me. More so than H3PO4 + KI. I have a Kg or more I2 on hand so merely need to get some tetralin.

2 I2 + C10H14 -> 4 HI + C10H10

[Edited on 24-6-2009 by Sauron]

Attachment: US5693306.pdf (514kB)
This file has been downloaded 810 times

Sauron - 24-6-2009 at 08:40

Another possible route to hydriodic acid from I2 is the Bunsen reaction which

2 H2O + SO2 + I2 -> H2SO4 + 2 HI

As written above the reaction is reversible, and the product inseperable by distillation but if both excess SO2 and escess I2 are used, two layers form and can be separated easily, the lower layer being HI/I2 or HI3 and higher polyiodides in water. The upper layer is H2SO4.

2 H2O + SO2 + 3 I2 -> H2SO4 + 2 HI3

I assume that the hydrogen polyiodides on heating decompose to HI and I2, an assumption I can check in Mellor.

This reaction is the basis of the sulfur-iodine cycle exploited to generate H2 from H2O but I am looking to instead try to use it to prepare HI, so wish me luck!

[Edited on 24-6-2009 by Sauron]

Attachment: US4089940.pdf (350kB)
This file has been downloaded 784 times

Sandmeyer - 24-6-2009 at 08:59

Of curse, there is also classical preparation in Gatterman's Laboratory Methods of Organic Chemistry on page 95 http://www.sciencemadness.org/library/books/gatterman_1937.p... (IMO, this book, like Vogel's, should be thoroughly read by everyone interested in organic chemistry).

Sauron - 24-6-2009 at 09:59

Sandmeyer, I am sure all participants in this thread are familiar with the red P catalyzed iodination of methanol, which is described not only in Gattermann but Org.Syn., Vogel etc. As you well know most members, including me, cannot readily obtain red P and many others cannot get I2 either. So this method, though timr-honoured, is blocked.

I am looking to avoid the use of H2S if at all possible. I would rather use H3PO4/KI than H2S/I2, and am simply looking for alternatives to both of those.

Clearly the P4O10/methanol/alkali iodide reaction as exemplified by Nicodem is probably the best route to MeI discussed thus far.

That being sais, you won't mind if I look round for other methods?

I2 and tetralin looks good as a facile means of making HI. As described upthread, the excess tetralin can be reused without removal of the napthalene, and when the napthalene builds up, distillation or dilution with more tetralin is all that is needed.

[Edited on 24-6-2009 by Sauron]

PainKilla - 24-6-2009 at 18:24

You can make methyl iodide quite easily and in good yield by slowly adding methylsulfuric acid to a concentrated solution of an iodide (sodium, potassium, ammonium all work) - the original report of this patent was posted on the Hive, where NH4I was used in something like an 80% yield. I posted this post here.

Not as elegant or interesting as the phosphorous chemistry, but by far more accessible, and the yield isn't shabby at all - I'm willing to bet you can get 90% with a good setup and good technique.

Here is a procedure I wrote down a while back when I tested this...

Methyl Iodide Synthesis

100g potassium iodide (.60mol) [1]
80ml methylsulfuric acid (.96mol) [2]
60ml water

In a 500ml flask equipped with magnetic stirrer and prepared for distillation, 100g (.60mol) of potassium iodide is dissolved in 60ml water. This is stirred until the full dissolution of the KI, and then heated on a water bath until the internal temperature measures 80*C. To this solution, methylsulfuric acid is SLOWLY added via an addition funnel, and the methyl iodide distills almost immediately. [3] After about three hours, no more methyl iodide was seen to distill. The receiving flask contained 50ml of saturated NaCl solution, cooled by an ice bath. [4] The methyl iodide is separated from salt solution and dried with magnesium sulfate.

Yield: 23ml (52.4g) (61%) [5]

Notes-
[1]– Any alkali halide salt works well, the ideal salt being NH4I (ammonium iodide), due to the the formed sulfate salts' high solubility in water (this is speculative).

[2]– Methylsulfuric acid was prepared via US Patent 3,047,604. This entailed mixing equimolar ratios of sulfuric acid and methanol, followed by addition of anhydrous sodium sulfate and cooling the mixture to a temperature below 32*C. This was then magnetically stirred for 4 hours. The mixture is allowed to sit until the solids settle, and the methyl sulfuric acid decanted in near quantitative yield. Based on the original post, you can use H2SO4/MeOH 1:1 without going through this sort of purification.

[3]– If added too quickly, hydrogen sulfide gas is generated, and eventually sulfur. Hydrogen Sulfide is toxic. Perform this reaction only under a hood, or outside. IHL-HMN LC50 800 ppm (5 min). Thus, add acid slowly, making careful note of the rate of distillation.

[4]– Storing methyl iodide under water is not necessary, but it makes the purification easier and also reduces losses from evaporation. Note however, that its solubility is 14g/L (it is less in brine solution).

[5]– The yield can very easily be improved by adding methyl sulfuric acid very slowly, as excess acid seems to oxidize the potassium iodide to iodine. Additionally, if a good condenser is employed, the shortest possible distance to receiver is preferable, as the methyl iodide otherwise condenses back into the reaction mixture and eventually decomposes. This was noted to be the cause of lower yields at the time this procedure was written.

References: CS164707 (patent)

Methyl Iodide Physical and Chemical Properties

Molecular formula: CH3I
CAS No: 74-88-4
Appearance: colorless to pale yellow liquid with an ether-like odor
Melting point: -64 C
Boiling point: 42 C
Specific gravity: 2.28
Solubility: 14/g L-H2O (20C)
Stability: Stable. Incompatible with strong bases, strong oxidizing agents. May discolor on exposure to light. Moisture-sensitive.
Toxicology: VERY TOXIC, AND CONTACT MAY BE FATAL. May cause cancer. Possible teratogen. Vesicant. May cause harm to the unborn child. Readily absorbed through the skin. May cause sensitization. Severe irritant. Narcotic. Typical PEL ca. 5 ppm. Possible long term damage, toxicity may be delayed.
Toxicity data:
ORL-RAT LD50 76 mg kg-1
IHL-RAT LC50 1300 mg/m3/4h
Disposal Information: Slowly hydrolyzes to methanol and iodine. Dilute hydroxide solution can be used to speed this process along. Environmental hazard. Note low boiling point when hydrolyzing.

[Edited on 25-6-2009 by PainKilla]

Sauron - 24-6-2009 at 20:41

I believe I posted that patent recently in connection with the preparation of diethyl sulfate, for prurposes of making ethyl centralite. But never mind. Note that it is a USN patent and ocviously they were and are interested in stabilizers for nitrocellulose.

The reaction between methyl hydrogen sulfate (methylsulfuric acid) and an iodide is closely (intimately!) related to the Org.Syn. procedure for methyl iodide from KI and dimethyl sulfate. Methyl hydrogen sulfate is after all the half ester of H2SO4 and methanol, DMS is the full ester.

The Org Syn prep is on a 5 mol scale and gives >90% MeI or about 500 g from 900 g KI. Note that when domethyl suflate reacts methyl hydrogen sulfate is the other product besides MeI; the reaction is between 4.8 mols KI and 5 mols DMS, a 4% excess.

This begs the question of whether methyl hydrogen sulfate can be recovered from the DMS reaction and used used to prepare a further equivalent of MeI.

The other issue is how toxic is the half ester? Obviously it is an alkylating agent, else it would not work in this reaction. The prudent practice is to handle it like DMG, with all the respect a carcinogen deserves. This may have a chilling effect for some members. DMS of course is a human BRAIN carcinogen. Brain tumors are no fun or so my younger sister told me before she died of gliablastoma multiform.

[Edited on 25-6-2009 by Sauron]

Attachment: US3047604.pdf (162kB)
This file has been downloaded 921 times

PainKilla - 24-6-2009 at 20:50

Based on the rates of hydrolysis, methylsulfuric acid is probably as dangerous as sulfuric acid, but no more.

"Dimethyl sulfate hydrolyzes slowly in cold water but rapidly in warm water and acidic solutions. The hydrolysis occurs stepwise, initially forming methyl sulfuric acid, then sulfuric acid and methanol. It can be calculated that DMS hydrolyzes to methyl sulfuric acid with 99.9% completion as follows:

...

Complete conversion to sulfuric acid occurs at a much slower rate. Salts of methyl sulfuric acid are formed when DMS is hydrolyzed in aqueous alkali." source

Strongly suggests methylsulfuric acid is a terrible methylating agent (mainly because the proton will hop off far faster than the methyl).

[Edited on 25-6-2009 by PainKilla]

Sauron - 24-6-2009 at 21:42

That's encouraging.

So

KI + Me2SO4 -> MeI + MeSO4K

Can't we liberate the MeSO4H and isolate it to exploit with more KI?

If not no big deal as MeOH and H2SO4 are almost free (relative to the cost of KI anyway).

PainKilla - 24-6-2009 at 22:04

The reaction occurs pretty much instantly, and at relatively high temperatures with a concentrated ionized solution, so the addition of methylsulfuric acid to this hot mixture probably causes significant destabilization of the solvated iodide ions, possibly allowing for increased nucleophilicity of the iodide towards the methyl group while the acid itself is only 'locally' dissociated during the very brief time during which methyl iodide is formed. Subsequently the very volatile methyl iodide leaves solution, returning some stabilization towards solvation of the remaining sulfate-bisulfate ion... Subsequent addition of acid continues the process. This is conjecture of course, but from what I remember the reaction occurs only at higher temperatures, and with concentrated solutions, maybe suggesting that this reaction occurs due to unique conditions...

This is also roughly the protocol I use for making methyl/ethyl halides, and it works very well for methyl bromide/iodide and ethyl bromide/iodide, although naturally for the bromides the oxidation is a non-issue.

Given the strange behavior of the reaction, I wonder if it can really be described so simply as CH3HSO4 + KI --> CH3I + KHSO4; the fact that both iodine and hydrogen sulfide can be produced if not carefully performed might suggest there is more than meets the eye, and that this reaction is very sensitive to competing reactions (energetically speaking, procedure-wise just requires slow addition). Or rather it is to say, the reaction is only possible because of a very finely controlled set of conditions.

Regardless, I think the acidic conditions are important, which is probably why the dimethyl sulfate / iodide reaction does not go to completion normally...

I agree though, it makes little sense to use DMS to make MeI when this method is more economical and far safer to boot... DMS can handle less reactive substrates than MeI AFAIK, so it's just easier to make it this way and save DMS for those purposes anyway...

[Edited on 25-6-2009 by PainKilla]

Sauron - 24-6-2009 at 23:00

There's no arguing the safety issue, but if one has a hood, gloves, etc and a wash bottle of conc ammonia at hand, DMS is really not very daunting, especially considering the target MeI which shares much of its hazards.

[Edited on 25-6-2009 by Sauron]

Attachment: CV2P0399.pdf (185kB)
This file has been downloaded 1308 times

PainKilla - 24-6-2009 at 23:04

That's 230ml or 525g @ 62% (methyl iodide has a density of 2.28g/ml) for a kilogram of potassium iodide. It makes no difference if you scale up or not - the limiting reagent is KI with either procedure, and the yield is near quantitative (according to the other posts using this general procedure) so... I'd rather take the 'safe' one, unless methanol and sulfuric acid are more expensive for you than iodides are - they aren't for me.

I do agree that safety is relative, but for the average person, making methyl iodide from methanol, sulfuric acid, and an iodide is *far* more accessible than using DMS - which is useful in and of itself and a shame to waste, irrespective of price, on something often inferior and otherwise obtainable via ubiquitious reagents.

And dimethyl sulfate isn't that easy to make (under amateur settings at least) - I've tried extensively and could not get yields anywhere near the claimed industrial 80-90%... 40-50% is the best I was able to get (using a vacuum distillation ~5-10mmHg).

[Edited on 25-6-2009 by PainKilla]

Sauron - 24-6-2009 at 23:15

My mistake.

Obviously, then, methyl hydrogen sulfate is indeed quite competitive with DMS in this reaction.

OT question: will MeSO4H react with KSCN in analogous fashion to give MeSCN?

[Edited on 25-6-2009 by Sauron]

PainKilla - 25-6-2009 at 00:40

I don't know much about the chemistry of that anion but:

http://books.google.com/books?id=zBpJAAAAMAAJ&pg=PA486&a...

(Note that Ca(CH3SO4)2 + KSCN yields CH3SCN upon distillation... However, I recall reading that Ca(CH3SO4)2 liberates (CH3)2SO4 upon distillation, so this might just be an situ way of generating DMS... I could never get the metal salts of methylsulfuric acid to yield DMS though, as patents and old literature claimed.)

Citation in question:

http://www3.interscience.wiley.com/journal/112330280/abstrac...

----

Maybe it's possible, I don't think anyone has ever tried it. If thiocyanates are stable under acidic conditions (doesn't look like it, see attached paper), and those conditions promote S-alkylation over N-alkylation, then you might be OK... I would imagine the reaction would need to be run at a higher temperature however, since methyl thiocyanate is less volatile... It would be interesting to try at least, since iodides would normally be oxidised instantly by straight sulfuric acid, but still manage to react with methylsulfuric acid without extensive oxidation... maybe the thiocyanates will also do so - they certainly appear soluble enough that a concentrated solution would react favorably, if heated to a high enough temperature to allow for the distillation of MeSCN.

Attachment: ACTION OF SULFURIC ACID ON THIOCYANATES.pdf (114kB)
This file has been downloaded 1077 times

[Edited on 25-6-2009 by PainKilla]

Sauron - 25-6-2009 at 01:34

Vogel states that alkyl hslide metathesizes in ethanol with KSCN.

The lit, has barium thiocyanate reacts with DMS to give MeSCN and so that is the basis for my analogy, See the methyl thiocyanare thread for refs and details.

If this does work, I mean the half ester and KSCN, it saves a step and the expense of KI.

If it does not fly, then the half ester makes MeI and MeI then makes MeSCN.

[Edited on 25-6-2009 by Sauron]

[Edited on 25-6-2009 by Sauron]

Nicodem - 25-6-2009 at 04:36

It is interesting that PainKilla posted about his experience with MeI preparation using methylsulfuric acid, because yesterday I just set another experiment based on in situ methylsulfuric acid formation and iodide methylation. Too bad it didn’t live up with the expectations, quite possibly due to the side reaction problem PainKilla described.

Experiment on the synthesis of MeI from MeOH using NaHSO4

In a two neck flask (100 ml) equipped with a thermometer, a suspension of 10.4 g NaHSO4 (80 mmol) and 6.35 g NaI (40 mmol) in 65 ml methanol (1600 mmol) was first purged with argon, closed and left stirring for 18 hours. The color changed to pale brown and HPLC analysis (UV detection at 220 nm) showed the presence of some MeI. A small Vigreux distillation column (1.5 cm wide, 8 cm long) was set on the flask and the reaction mixture heated and stirred on an oil bath. The first drops of distillate came over at 37.5°C while the reaction mixture temperature was 65°C. The temperature of the distillate slowly increased to 58-60°C and remained such for the next hour or so, but the reaction mixture boiling temperature stayed at 66°C. The distillate was collected till 62°C. This colorless liquid (18.7 g) was added to 40 ml of water. A white milky emulsion formed, therefore 20 ml of brine was added. After standing for 30 min the few droplets of heavy oil that settled at the bottom were pipeted out (1 g) and dried with CaCl2 granules which floated on top. This gave 0.95 g of a nearly colorless oil (16%). The IR spectra was identical to that of methyl iodide.
A strong smell of H2S was obvious during the cleaning up of the reaction mixture remains, so apparently some reduction side reaction goes on, though it is strange that no iodine coloration is present.

PS: About MeSCN synthesis via methylation of tiocyanates with methylsulfuric acid... Unlike the iodide anion, the tiocyanate anion is a weak base (pKa of HSCN is 4) while the methylsulfuric acid is a relatively strong acid. Mixing them together means MeOSO3H would get deprotonated (forming a much less electrophilic MeOSO3(-) anion) and the thiocyanate being protonated (forming a much less nucleophilic and unstable HSCN). Thus the reactivity of both reagents would drop severely and a competing HSCN decomposition reaction enabled.
A compromise would be to react a methylsulfate salt with a thiocyanate salt. Methylsulfates are still somewhat reactive methylating agents even though several magnitudes less electrophilic than dimethyl sulfate of methyl sulfuric acid (examples can be found in other forum threads where references about phenoxide and nitrite alkylation with sodium or potassium salts of ethyl- or methylsulfuric acids are given). The lower reactivity of MeOSO3(-) would mean a higher reaction temperature would be required. Thus to a methanolic solution of methylsulfuric acid (prepared by the usual method from NaHSO4) would have to be neutralized with K2CO3, then KSCN added and distilled to dryness to hopefully give a methanolic solution of MeSCN. A relatively simple experiment to verify.

Sauron - 25-6-2009 at 10:01

There's a fair amount of material on methylsulfuric acid in Sartori's book even though it is not a "war gas" because of ots relationship to dimethyl sulfate and the fact that it was present in some French and German "war gas" mixtures as a byproduct of manufacture of DMS from methanol and chlorosulfonic acid.

Heating methylsulfuric acid to 170 C at reduced pressure causes dissociation to H2SO4 and dimethyl sulfate. I guess this was the prep of DMS Painkilla alluded to.

Per sartori the lab and industrial preps of dimethyl sulfate are from methanol and chlorosulfonic acid followed by this thermal disproportionation.

He offers as alternatives, methanil and SO2Cl2 or, methylsulfuric acid + methyl chloroformate.

Painkilla, have you tried the SO2Cl2 route to DMS?

I did a thread on chlorosulfonic acid a while back and will have to review it but I think I found ways to make it other than SO3 + HCl (g).

Of course we know how to make methylsulfuric acid from MeOH and H2SO4 per the US Navy patent as well as from bisulfate.

Anyone working with methylsulfuric acid needs to keep that thermal disproportionation in mind because no one wants to generate DMS vapor accidentalltm do we?

PainKilla - 25-6-2009 at 10:54

Yeah, what's what I figured you meant, I just wanted to see if I could find any mention towards the use of any methylsulfate-based derivative...

Nicodem, are you sure the pKa of HSCN is 4? That was the first thing I looked at, and according to several publications it is a strong acid with a pKa between -1 and 1... This goes along well with the 6-8 pH range cited for potassium thiocyanate solutions, I would expect at least that under hot and concentrated aqueous conditions thiocyanate isn't very much worse a base than iodide.

I would not expect methylsulfuric to retain such a low pKa as plain sulfuric either, so it is bound to be in a similar range (-1 - 1)... In this respect, it might still be worth trying at least, especially if this reaction occurs because of higher-order solution chemistry. Of course, since the pKa's ultimately are much closer, it does seem likely that the yields aren't going to be as good if the reaction does work. On the other hand, if the distillation of the MeSCN occurs fast enough, it might be able to compete with other side reactions, thereby driving the reaction to completion... Although, the bisulfate remaining would also promote formation of HSCN increasingly as the reaction went along (unless the pKa was significantly altered due to the concentrated ionic solutions and temperatures) - I think it's at least worth an experiment.

Interesting to see that NaHSO4/methanol work, albeit it poorly. This further confirms the idea (to me at least) that this is some sort of solution-chemistry phenomenon (with respect to the potentially good yields of the original reaction); at the very least, water appears to be necessary for good yields.

Heating methylsulfuric acid under vacuum was indeed what I referred to Sauron, but the yields on the reaction are low, although they do increase as you consistently increase the vacuum so perhaps a 0.5 - 1mmHg vacuum would give the yields cited in the literature. I've never tried using SO2Cl2; it seemed like a very roundabout way of making DMS - SO2Cl2 seems useful for other things, even if I were to make it...

I don't think the disproportionation would occur under any conditions used in this reaction - I've tried heating plain methylsulfuric acid to high temperature without a vacuum, and IIRC no DMS was obtained - and that was at ~160-200*C, not the 80*C used here. Though naturally, due caution is necessary either way, since methyl iodide is after all, at least as dangerous as DMS with respect to volatility.

I'd try this reaction again in order to improve yield since when I wrote it down initially, but unfortunately I can't do amateur chemistry under my current settings.

Forgot to add:

"In the course of some work on the thiocyanates it became necessary to know the strength of aqueous thiocyanic acid. Ostwald's earlier conductivity experiments indicated that this acid is almost as strong as hydrochloric, but more recently Latimer estimated from thermodynamic calculations that it is a weak acid with an ionization constant of approximately 1x10^-4. The experiments summarized in Table I prove that thiocyanic acid is a strong acid. The third column gives the pH calculated from the relation, pH = -log (H+), where (H+) is the concentration of the hydrogen ion in moles per liter for complete
dissociation of the acids. In a private communication Latimer has informed us that soon after he published his book he realized that thiocyanic acid is a strong acid." (source attached)

Attachment: Strength of Aqueous Thiocyanic Acid.pdf (247kB)
This file has been downloaded 839 times

[Edited on 25-6-2009 by PainKilla]

Sauron - 25-6-2009 at 11:06

The alternative routes to chlorosulfonic acid are:

100% H2SO4 + POCl3 (3:1) gives chlorosulfonic acid and H3PO4, chlorosulfonic acid can be distilled out. POCl3 is made by heating P2O5 and NaCl in an autoclave at 400 C.

Per a US patent, 100% H2SO4 and SO2Cl2 + vatalytic HgSO4 in an autoclave at 160 C 8 hrs gives quantitative chlorosulfonic acid.

I am not suggesting that using chlorosulfonic acid is required to make methylsulfuric acid, but it is well worth keeping in mind.

As I mentioned, Vogel says that alkyl iodides in alcohol react with KSCN (or NaSCN) to give alkyl thiocyanates.

My source for the reaction of dimethyl sulfate and barium thiocyanate is Merck Index, but annoyingly although this is under Literature References, they give no citation. Shit! I suppose the virtue of Ba(SCN)2 is that the resulting BaSO4 will be insoluble. But this salt is made per Inorg.Syn. from NH4SCN which is expensive, also expensive to make since you start with CS2 and NH3, obtain ammonium dithiocarbonate and thermolyze that to NH4SCN, S and H2S (see Ullmann's on thiourea.)

On the other hand since I opened up cheap CS2 from S and acetylene, this is not so costly, just laborious.

Anyway I am hoping that DMS will also work with KSCN and save me the bother.

And if methylsulfuric acid can replace DMS in this so much the better.

[Edited on 25-6-2009 by Sauron]

PainKilla - 25-6-2009 at 11:11

I don't think you even need (anhydrous) methylsulfuric acid for this reaction, as Antoncho demonstrated in the original excavation of this reaction that just plain methanol + sulfuric acid work well; I've use methanol/sulfuric acid this way (without doing the sodium sulfate purification) for other alkyl halides to very good effect, so it seems probable that despite the water, a mixture of methanol/sulfuric acid is already almost entirely methylsulfuric acid.

Sauron - 25-6-2009 at 11:41

Painkilla if you are wanting DMS/DES but having mediocre results with disproportionation of the alkylsulfuric acids, you might try making SO2Cl2 and esterifying that with 2 eq alcohol.

Sources are divided over whether that reaction is slow, or violent, so proceed with caution.

PainKilla - 25-6-2009 at 11:49

I only needed MeI/DMS for methylations, and not on any particularly deactivated substrate either - methyl bromide suffices quite well for most reactions, and is far, far cheaper than iodides are, so I just use that, although it is nevertheless useful to know how to make the whole series of methyl/ethyl halides/sulfates... Thanks!

Sauron - 25-6-2009 at 12:41

Yes, I have maybe 5 Kg NaBr on hand but relately little KI. I could try MeBr and KSCN, easily enough, in a pressure bottle. MeI is easier to handle, just costs more.

Ammonium thiocyanate is actually cheaper (in lab grades) than KSCN and lower MW so more mols/Kg. I thought it was a lot more expensive. NH4SCN is c.$4/mol, 13 mols/Kg and KSCN is c.$6/mol, 10 mols/Kg.

So, I will make Ba(SCN)2 myself and react it with domethyl sulfate. If methylsulfuric acid will do the job, even better.

My interest in MeI as a result just plummeted.

Nicodem, both Merck and Brauer state that HSCN is a very strong acid, certainly not a weak one. So I must cibcurr with Painkilla and recommend you check that datum/

2 NH4SCN + Ba(OH)2 -> Ba(SCN)2 + 2 NH3 + 2 H2O

Ba(SCN)2 + 2 Me2SO4 -> 2 MeSCN + Ba(MeOSO3)2

Attachment: Pages from IS-3.pdf (240kB)
This file has been downloaded 5037 times

[Edited on 26-6-2009 by Sauron]

Nicodem - 26-6-2009 at 00:40

PainKilla, you are correct in assuming that a methanolic solution of H2SO4 at equilibrium is more less methylsulfuric acid in methanol. The equilibrium constant of the H2SO4 + MeOH -> MeOSO3H + H2O reaction is 4.4 (Tetrahedron Letters, 28 (1987) 1627-1628).

Quote: Originally posted by Sauron

Nicodem, both Merck and Brauer state that HSCN is a very strong acid, certainly not a weak one. So I must cibcurr with Painkilla and recommend you check that datum/

I assumed the pKa 4 given in pKa Data Compiled by R. Williams was correct. I now checked Beilstein and there are a few conflicting references with pKa's of 0.85, 4 and 5. If it is truly 0.85 then it is indeed a strong acid, even stronger than methylsulfuric acid of which pKa is from 2 to 3 (various references give a different value, but inside this interval).
Nevertheless, I would never choose to methylate thiocyanates in acidic media. There are simply too many possibilities for side reactions (in acidic in some cases you get the competing N-alkylation as well). Not only is HSCN unstable, even MeSCN can decompose in highly acidic media. When weighting the chances of success between using MeOSO3H and MeOSO3K, I would choose the potassium salt, since even though it is less reactive the reaction is way more predictable. Also, using KSCN gives way better chances of success since it is low melting and you can do the reaction solventless (while ammonium thiocyanate decomposes on heating).

PS: I did not get the point of using Ba(SCN)2. What is all that about?

[Edited on 26/6/2009 by Nicodem]

PainKilla - 26-6-2009 at 00:57

Oh, I certainly agree, considering the paper I posted before that suggested reactions occured even in dilute acid, I wouldn't expect the reaction to work well... But it might be worth trying anyway since after all, the solution is not really highly acidic - bisulfate would be the most acidic species... If the thiocyanate salts are cheap, it's at least worth a try. After all, would you really have expected this way of making methyl iodide to work? I was skeptical until I tried it, and got a pleasant surprise... But the higher temperatures involved in this case may 'break' the reaction. But still, such experimentation is what makes chemistry so interesting!

BTW, do you have a reference for the pKa of methylsulfuric acid? A brief google search didn't yield any experimental calculations, only some theoretical ones (which implied it was ~0); I didn't realize it was actually that weak an acid.

[Edited on 26-6-2009 by PainKilla]

Nicodem - 26-6-2009 at 02:24

Quote: Originally posted by PainKilla

BTW, do you have a reference for the pKa of methylsulfuric acid? A brief google search didn't yield any experimental calculations, only some theoretical ones (which implied it was ~0); I didn't realize it was actually that weak an acid.

I was also surprised, because one would think it would be at least -1 or less, maybe even as strong as H2SO4. I went back to check the abstracts for those references given by Beilstein and it turned out that the pKa from 2 to 3 were actually the acidities of MeOSO3H in various alcohols (for example, in methanol its pKa is 2.16: Ukrainskii Khimicheskii Zhurnal, 44 (1978) 762-763). The only reference that I could find about its aqueous pKa is Canadian Journal of Chemistry, 56 (1978) 2342-2354. Now, this one gives the pKa of MeOSO3H in water as -3.4 which makes more sense, though it is still an "estimated value". Therefore, MeOSO3H appears to be a stronger acid than HSCN.

PainKilla - 26-6-2009 at 03:24

Thanks, it's nice to see some real data finally!

I personally don't like negative pKa's at all, but according to http://www.springerlink.com/content/r3h103ru0l462mr5/ the pKa of thiocyanic acid is -1, elsewhere it's claimed -1 to -2 for thiocyanic acid. Though, it's a very confusing way to think about things in terms of negative pKa, as opposed to thinking of the entire system (configurational/phase space of solvent + ion molecules) and the relativistic measurement of pH. However, I've only recently started getting a better grasp on physics, so I'm still pretty new to thinking this way - still thinking about this sort of thing helps one practice. Studies that examine the water structure and hydration shells of the ions are good places to start:

According to http://www.pnas.org/content/100/8/4557.abstract?ck=nck and http://www.crystalresearch.com/crt/ab39/465_a.pdf which seems to support the previous study, the thiocyanate ion is but weakly hydrated (it is tempting to imagine that iodide is also not very well hydrated), while sulfate and bisulfate on the other hand, are extremely well hydrated... I had thought initially, because of these reasons, that an explanation for this reaction might be derived for the great thermodynamic favorability of solvating the incoming sulfate molecule as it enters into solution, which would simultaneously destabilize weakly hydrated ions (as iodide) and in that transient process increase their nucleophilicity (while the solvent could also increase the electrophilicity of the methyl group as the methylsulfate is solvated) such that the (wanted methylation) reaction occurs...

I found it quite interesting to see the extreme! solubility of potassium thiocyanate - 220g/100ml @ 20*C (and surely this is even higher at higher temperatures, or ~2.5 molecules of water per anion (which goes along well with the neutron diffraction study) - potassium iodide on the other hand, seems to have about 5 water molecules per iodide under the initial conditions used in the reaction...

Based on the crystal structure of the thiocyanate ion, it gives the impression that very slightly 'acidic' conditions (such that the nitrogen coordinates with a water molecule, although coordination with an ion would also work) would in principle promote S-alkylation, but that also assumes such conditions are attainable in the first place, in the sense that they do not cause side reactions to occur, which seems entirely likely if methylsulfuric acid is indeed stronger than thiocyanic acid. But as you said, use of an ion rather than a proton (as in, using a methylsulfate salt) most definitely ensures this is the case... I really doubt the reaction would work well under such hot and acidic conditions.

But it's fun to think about at the very least; I wonder what other sorts of anions could be tested alongside methylsulfuric acid, in order to test it's capacity as a methylating agent...

I wonder then, if methyl iodide could similarly be synthesized from a methylsulfate salt and an iodide... I imagine some sort of solvent would be necessary, since unlike the thiocyanates neither methylsulfates nor iodides have particularly low melting points - maybe a high-boiling point glycol would work, or perhaps the reaction done in a concentrated aqueous solution... In principle the yields would be even better (unless they really can be made consistently to 80-90% with MeHSO4, in which case going through the hassle of making the salts seems pretty pointless), unless this reaction really does only occur because of some sort of strange solution chemistry.

[Edited on 26-6-2009 by PainKilla]

Sauron - 26-6-2009 at 06:08

Nixodem, the point of using Ba(SCN)2 to prepare MeSCN from Me2SO4 is that Merck Index 12th Ed. states this as method for making same - unfortunately without reference.

And ammonium thiocyanate, from which Ba(SCN)2 is made, per Inorg.Sy. attachment above is substantially cheaper than KSCN. Ot's 2/3 the price on a molar basis and 10% less on a 2.5 Kg basis. The discrepency is because MW of NH4SCN is 75 while that of KSCN is 95.

Vogel's brief general statement that alkyl thiocyanates can be made by treating alkyl halide with KSCN in alcohol, is also unreferenced, so I am simply letting the economics drive the decision. I will save MeI for other duties.

Inorg.Syn. cited a number of other preps of Ba(SCN)2, including the original prep by Berzelius. None of them looked as attractive.

I have not found any other lit on prep of MeSCN, so f you run across any, please call them to my attention.

Nicodem - 26-6-2009 at 07:00

Quote: Originally posted by Sauron

I have not found any other lit on prep of MeSCN, so f you run across any, please call them to my attention.

Alkylation of KSCN with alkyl idodides, bromides or tosylates is the most standard method for preparing alkyl thiocyanates. I have no idea why would one want to use the barium salt. Maybe it used to be comercialy available in the ancient times. KSCN is well soluble in acetonitrile as solvent and the reaction with primary alkyl bromides/iodides/tosylates is rapid (with methyl iodide or dimethyl sulfate it should be over in a couple of hours at <40°C). Methanol, ethanol, DMF and other solvents where KSCN is soluble can also be used. There are plenty of references in the literature.
Here are a few outputs from Beilstein regarding the MeSCN synthesis:

Methylation of KSCN with (MeO)2SO2:
Walden, Chemische Berichte, 40 (1907) 3215 (and/or 4301?)
Kaufler, Pomeranz, Monatshefte fuer Chemie, 22 (1901) 495.
Rybin, Zil'berman, Trachenko, J. Gen. Chem. USSR (Engl. Transl.), 50 (1980) 1702-1705.

Methylation of KSCN with MeI:
Hassanein, Akelah, Selim, El-Hamshary, Indian Journal of Chemistry, Section B: Organic Chemistry Including Medicinal Chemistry, 29 (1990) 763-765.
Jaworski, International Journal of Chemical Kinetics, 35 (2003) 61-66 (DOI: 10.1002/kin.10105) nonpreparative, kinetics study

Methylation of KSCN with Ca(MeOSO3)2:
Muspratt, Justus Liebigs Annalen der Chemie, 65 (1848) 253,257.
Cahours, Justus Liebigs Annalen der Chemie, 61 (1847) 96.

MeSCN from S-methylisothiourea hydrochloride thermolysis at 160-200°C:
Arndt, Chemische Berichte, 54 (1921) 2242.

Sauron - 26-6-2009 at 07:23

Barium thiocyanate is still commercially available (S-A, Alfa Aesar) but not cheap, it is however readily prepared from NH4SCN which is very commercial and substantially cheaper than KSCN.

But thanks, the methyl tosylate is a very good idea. Methyl bromide has the disadvantage of low bp obviously.

DJF90 - 26-6-2009 at 08:32

Sauron: You suggested a dewar condenser earlier to ensure no MeI escaped uncondensed during the distillation. Surely this would also provide complete condensation of methyl bromide too? Handling may be difficult but you could always set up the other reactants in the recieving flask to prevent this (would possibly save dirtying more glassware too).

Sauron - 26-6-2009 at 08:48

Sure it can be done, it's just that methyl tosylate has neither the cost of MeI nor the volatility of MeBr.

Perhaps the most interesting of the refs Nicodem posted is the one abour rearrangement of S-methylisothiourea hydrochloride. As it happems I have a Kg of the hemisulfate and this stuff is inexpensive, so if this reaction is efficient it is a short cheap route to MeSCN. Thiocyanates and thioureas have the same relationship as cyanates and ureas.

DJF90 - 26-6-2009 at 11:36

Well it depends on what is available to you. As you have plenty of bromide (5kg right?), methanol and sulfuric acid (I'm assuming here, but its not like you do anything by halfs

) then going by the methyl bromide may be more "convenient". However if you also have equal stock of methyl tosylate (or tosyl chloride) then this would be the most satisfactory way to proceed. The volatility/low bpt of MeBr is easily made negligable as I suggested above, supposing you have dry ice (which you can make from your tank of CO2).

JohnWW - 26-6-2009 at 12:11

Quote: Originally posted by Sauron

Barium thiocyanate is still commercially available (S-A, Alfa Aesar) but not cheap, it is however readily prepared from NH4SCN which is very commercial and substantially cheaper than KSCN.

That is because NH4SCN is made commercially in large quantities (technical grade) as an herbicide, particularly by the Dow Chemical Company. It is commonly mixed with other substances like aminotriazole, and used in weedkillers like Amitrole (which I have used myself on my property) and Radoxone 11, for weeds (especially some broadleaf weeds) which may be Roundup-resistant.
See e.g.:
http://www.freepatentsonline.com/4612034.html
http://www.freepatentsonline.com/WO2007147209.html
http://dictionary.infoplease.com/ammonium-thiocyanate

Sauron - 26-6-2009 at 12:29

True but I also have a Kg of tosyl chloride at hand, and I have the Org Syn prep of methyl tosylate in the notes to the prep attached, which just happens to be on a 1 Kg basis - how serendipitous. The prep calls for only TsCl, methanol and a little NaOH soln and yields 90%. The workup is somewhat simpler than that Klute used in his Me tosylate thread.

As I have some KSCN I can try this reagent, and also the rearrangement of S-methylisothiouronium hemisulfate, the latter once I obtain the Ber. paper by Arndt.

Tosyl chloride is very cheap, 2 Kg <$90, methyl tosylate from same supplier $240 for 2.5 Kg so this is surely a case where it pays to make your own ester.

These reactions do not demand a hood (though some may be run outside due to stench) so I hope to shortly hsve experimental results and photos to post so I can make some of my critics eat crow.

[Edited on 26-6-2009 by Sauron]

Attachment: CV1P0145.pdf (136kB)
This file has been downloaded 1291 times

Sauron - 26-6-2009 at 20:31

While awaiting the Arndt paper from Refs I pulled up my Annalen archive and fortunately have both Bands concerned (61, 65) These two papers deal with alkylation of KSCN with calcium methyl sulfate Ca(MeOSO3)2. Obviously this salt can be easily prepared from methyl hydrogen sulfate (MeOSO3H, methylsulfuric acid) and calcium carbonate or hydroxide.

This reaction does work, but in the manner of this journal at that time (1847-48) no disclosure of yield is made, at least as far as I can tell with my clumsy German comprehension. (I have only looked at the Calhous paper, pertinent pages 95-96.)

Attachment: Pages from Bd61.pdf (431kB)
This file has been downloaded 810 times

A quick look at the second paper also did not show any new details, but ib both cases further study is needed.

[Edited on 27-6-2009 by Sauron]

Attachment: Pages from Bd65.pdf (636kB)
This file has been downloaded 860 times

DJF90 - 27-6-2009 at 01:48

Could it be possible that the calcium salt is used to facilitate an aqueous reaction? I remember reading somewhere that its soluble in aqueus solution, which would provide a good medium as the thiocyanate (as stated before) has excellent solubility in water (remember its an Sn2 reaction (?): rate = k[Ca(MeOSO3)2][SCN-] )

Sauron - 27-6-2009 at 02:16

Liebig earlier obtained MeSCN by mixing alcohol, KSCN and H2SO4 which if I understand correctly worked in poor yield. The use of the Ca salt was Calhous' improvement to the Liebig procedure, avoiding oxidation of the product I suppose.

A decent translation will doubtless clarify things further.

Indian J.Chem. is not available online, and neither is J.General Chem. USSR (English). I have requester the two Ber. papers.

[11] Morton, DE 1 295 549, 1966 (J. T. Venerable, J. Miyashiro, A. W. Seiling).
ï[12] Bayer, DE 1 183 903, 1965 (K. Goliasch, E. Grigat, R. Pütter).

These two German patent citations are from Ullmann's and the accompanying test states that methyl thiocyanate or isothiocyanate can be produced selectively by varying the solvent, temperature and thiocyanate concentration in the reaction of a methyl halide with a thiocyanate salt. The most common solvents are (as Nicodem stated) water, ethanol, and DMF. I am sure that Morton, and Bayer, would have taken out corresponding US patents to these so I will root aroundd for those rather than have to fight with the German patent texts. I think now we are getting down to the nitty gritty.

[Edited on 27-6-2009 by Sauron]

Sauron - 27-6-2009 at 05:06

I've been waltzing with patent SEs for hours and have gotten nowhere. Help!

Walden, Chemische Berichte, 40 (1907) 3215

I haveobtained this paper and in addition to the prep of MeSCN from DMS and KSCN which proceeds in excellent yield the paper also details

DMS + KCN -> 2 MeCN

DMS + KNO2 -> MeNO2 + MeOSO3K

Look again. Dimethyl sulfate reacts with 2 mols of the strong base KCN which implies that potassium methyl sulfate reacts with one mol. You may mot want to make acetobitrile from DME but from methanol and H2SO4 and potassium carbonate + KCN why not?

Potassium nitrite apparently will not react with potassium methyl sulfate but will with DMS. Therefor it may give nitromethane with ,ethylsulfuric acid, methyl tosylate - it is no surprise MeI would work but the others are cheaper.

[Edited on 28-6-2009 by Sauron]

Sauron - 28-6-2009 at 00:57

Sadly the German patents cae to little. The Moton Intl one concerns pressure reactions of MeCl with various inorganic thiocyanates, The Bayer patent, mercaptans amd cyanogen halides.

The Arndt paper does state that S-methylisothiorea hydrochloride thermolyzes efficiently at 160-200 C to MeSCN and NH4Cl, details of preparing the hydrochloride from the hemisulfate are provided.

For Formatik

Sauron - 28-6-2009 at 21:21

As per PM

This post may be deleted once I am sure formatic downloaded the paper.

Attachment: 2236-2242.pdf (393kB)
This file has been downloaded 1245 times

[Edited on 29-6-2009 by Sauron]

Sauron - 29-6-2009 at 01:37

The Arndt paper concerns a new (at that tine) reagent for generation of pure MeSH by reacting S-methylisothiourea hemisulfate with aq NaOH. This I am not interested in per se, but Arndy also observed that the hydrochloride salt thrermolizes betweeen 160 and 200 C to MeSCN and NH4Cl along with a little MeSH minor side product.

There is an Org,Syn. prep vased on Arndt's procedure. Two mols thiourea amd 70 ml water are reacted with 1,1 mol (10% excess) dimethyl sulfate; the first phase of the reaction usually starts without exterbak heating and as it proceeds to relux under its own exotherm may require cooling ro control. When this reaction ceases the methylation is half over and the DMS is now methylsulfuric acid. A futher hour;s reflux with external heating is required to drive methylation to completion. Yield is very high.

This tells me that DMS can be replaced entirely with preformed methyl hydrogen sulfate (methylsulfuric acid) made from methanp; and conc H2SO4. Just use 2 mols thuouewa to 2.2 mols MeOSO3H and reflux from the start, as there will be no exotherm. Otherwise procedure and workup will remain same. Hopefully yiels as well. Cheaper and safer!

To get to MeSCN, Arndt converts the hemisulfate to base and the base ti hydrochloride, I await translation for accurate details. From a kilo of the S-methylthiourea hydrochloride O estimate 500-600 g MeSCN will distill over leaving 600-500 g in the post mostly ammonium chloride, A scrubber is used to absorb the small amoun of methanethiol that forms.

(MeSC(=NH)-NH2).HCl

Easy to see how it breaks into NH4Cl and MeSCN

The S-meythylisothiourea hemisulfate is also used to prepare aminoguanidine bicarbinate without going through nitroguanidine - which ought to make the energetics lot happy.

From Org Syn:

"W. W. Hartman and Ross Philips have submitted a procedure suitable for the preparation of aminoguanidine bicarbonate on a larger scale. The sulfates of methylisothiourea and of hydrazine are allowed to react with the evolution of methyl mercaptan. In a 30-gal. crock are placed 10 l. of water and 5760 g. (20 moles) of methylisothiourea sulfate.2 In a 22-l. flask, 5.2 kg. (40 moles) of hydrazine sulfate3 is stirred with 12 l. of water, and 40% sodium hydroxide is added until all the hydrazine sulfate has dissolved and the solution is just neutral to Congo paper. The exact amount of alkali is noted and a duplicate amount added. The hydrazine solution is then added to the 30-gal. crock with stirring, as fast as possible, without allowing the foam to overflow the crock. The mixing is done out-of-doors, or in an efficient hood, since large volumes of methyl mercaptan are evolved. If the reaction is carried out on a smaller scale in 12- or 22-l. flasks, using appropriate amounts of material, the methyl mercaptan evolved may be absorbed in cold sodium hydroxide solution and isolated if desired. The solution is stirred until evolution of mercaptan stops, and then a few liters of water are distilled off under reduced pressure to free the solution entirely from mercaptan. The residual liquor is chilled in a crock, and a crop of hydrated sodium sulfate is filtered off, washed with ice water, and discarded. The filtrate is warmed to 20–25°, 25 ml. of glacial acetic acid is added, then 4 kg. of sodium bicarbonate, and the solution is stirred vigorously for 5 minutes and thereafter occasionally during an hour, or until the precipitate no longer increases. The precipitate is filtered with suction and washed with ice water and then with methanol, and is dried at a temperature not above 60–70°. The yield is 3760 g. (69% of the theoretical amount). Hydrazine sulfate may be recovered from the final filtrate, if the filtrate is strongly acidified with sulfuric acid and allowed to cool."

Attachment: CV2P0411.pdf (115kB)
This file has been downloaded 807 times

[Edited on 29-6-2009 by Sauron]

Sauron - 29-6-2009 at 23:25

Here is a 1901 paper citrf by Nicodem whicg describes the reactions of MeI as well as DMS with KSCN.

It teaches that DMS reacts even at room temp with KSCN and that cooling may be required to control the reaction. I would expect them yo mean an equimolar reaction, because the reactivity of the methylsulfuric acid potassium salt will be much less, and if it reacts at all will certainly need heating.

Yn any case the yield is quantitative (again I am sure they mean 1:1).

Reactions of the two methylating agets with nitrite and cyanide are also given.

Attachment: mh1901.pdf (202kB)
This file has been downloaded 795 times

Formatik - 30-6-2009 at 14:15

They say the sulfate isn't suitable due to solubility. So when starting from the salts, they convert the sulfate to the chloride and then go from there to get the base.

S-Methyl-isothiourea: For the isolation of the base from its salts, it comes down to liberating them not in water, but in a more easily volatile solvent, namely methyl alcohol. The sulfate is not suitable though due to its insolubility in methyl alcohol, but the aforementioned hydrochloride, which can easily be obtained from the sulfate by conversion with barium chloride; its melting point we have found as in accordance with Dixon to be 59-60º, so conspicuously lower than that of the free base. This hydrochloride is solubilized in a small amount of absolute methyl alcohol, under ice cooling put into a solution of the calculated amount of sodium in a small amount of methyl alcohol, it is then filtered off from sodium chloride, and quickly evaporated in a vacuum desiccator. The residue is taken up in acetone, filtered and then treated as above. One obtains a compound identical to that described above.

0.1758 g subst.: 0.1712 g CO2, 0.1028g H2O, 0.4588 g BaSO4.
C2H6N2S.. Calc.: C 26.6, H 6.7, S 35.6.
Found: » 26.6, » 6.5, » 35.8.

Base "treated as above": One can purify it again by taking it up in acetone at room temperature, filtering off unsolubilized dicyandiamide and cooling with ice. Here usually a part precipitates out in colorless, shiny leaflets, even more if ice cold petroleum ether is gradually added.

[Edited on 30-6-2009 by Formatik]

Formatik's Translation of Arndt (Above)

Sauron - 30-6-2009 at 16:06

Thanks. I was assuming incorrectly thyat one needed to go sulfate -> base -> hydrochloride. Not the case. All I need is to metathesize the sulfate in waterwith BaCl2 in same solvent, BaSO4 will ppt amd be filtered off, water is then rotavaped off.

The "sulfate" is hemisulfate, so I need 1 mol BaCl2 per mol S-methylisothiourea (C2H6N2S)2.H2SO4 to get BaSO4 + 2 C2H6N2S.HCl

That "sodium in methanol amd then filter off NaCl" step is clearly the liberattion of the base from the HCl salt with NaOMe. But I need not do that.

Dicyandiamide is the decomposition product of S-methylsiothiourea from moisture. Also known as cyanoguanidine, and is the dimer of cyanamide.

Thanks, formatik, alles in ordnung!

[Edited on 1-7-2009 by Sauron]

Arrhenius - 5-7-2009 at 19:24

I've finished the scale up to 400mmol. I'll post in the pre-publication section.

Pages: 1 2