Author Topic: easy THF oxidation using KMnO4  (Read 8498 times)

0 Members and 1 Guest are viewing this topic.

armageddon

  • Guest
easy THF oxidation using KMnO4
« on: July 01, 2004, 04:54:00 PM »
Well I tried to oxidize THF to gamma-butyrolactone time ago, and must say I really don't understand why the oxidation is called "strongly exothermic" - I've seen many violent/exotherm things with far more power since then..

This is how I did it (if I remember correctly):

50ml THF were put in a 500ml RBF together with 100ml d.H2O. This was put into a cold water bath (0°C) and heavy magnetic stirring was started. Then were added small portions of KMnO4 (2-3g). After each addition, stirring was continued until all permanganate had dissolved, then the rxn vessel was removed from the ice bath and the temperature was monitored (using a thermometer..). As soon as it started rising to more than 30°C, it was put back into the cooling bath, sometimes a bit of ice water was added (added to rxn, as necessary). This procedure (dissolve KMnO4, wait for temp.rise, then cool down) was repeated until at least 40 grams KMnO4 had been processed (very time consuming).

The color of the reaction was (of course) pink after permanganate addition, but changed to a chocolate-like brown and became viscous as MnO2 was formed. At the end, the volume of the rxn had grown to ~300ml's (due to large amounts of water, needed for cooling and to enable good stirring) and the whole mess was vac filtered to remove MnO2, then it was let stand for ~30 minutes (color was still purple after filtration, but this changed to a slightly milky brown) and filtration was repeated, this time through a thick celite pad, resulting in a crystal-clear colorless solution. The remaining, unreacted THF was then removed with an aspirator (bad for the environment, better distill off) and the remaining water/GBL-mixture (after cooling down) was extracted three times with DCM.

The combined DCM extracts were dried over Na2SO4 and the solvent removed (this time using distillation  :) ) to leave ~5 ml's of GBL, which were diluted with 0.5ml d.H20, heated to boiling (in a test tube) and neutralized with a solution of 2.6g NaOH in 10ml d.H2O. A little bit more heat was applied (initially there were two phases, boiling caused homogenization). After cooling, everything was filtered through a prewetted filter paper, boiled for some 10 minutes (which removed residual GBL and a very disgusting smell) and diluted with enough H2O (and some granules of ascorbic acid) to give a solution containing approx. 0.5g GHB*Na/ml, with a pH of 7.00...

The highest temperature ever noted was 40°C if I remember correctly, and this was after adding ~6g permanganate AT ONCE without external cooling - I really don't know if this can be called "violent"...  ;)

A last comment on ratios used, yields etc.: This experiment was originally tried with 10ml THF, but as NO temperature rise could be noted, the chemist decided to add "a bit" more THF - directly from flask... The result was the accidental addtion of 50ml THF at once, leading to the thought "additional water might be good" and subsequently to the addition of LOTS of d.H2O - in the end the volume was huge, lots of THF remained unchanged and at least 30 grams oxidant were used (at this point, the chemist had become too tired to measure out any amnounts any longer  ;) ) - originally, it was only meant to be a small scale test... ;D  ;D

Greetz A


Rhodium

  • Guest
You haven't even distilled the product lactone
« Reply #1 on: July 01, 2004, 11:26:00 PM »
You haven't even distilled the product lactone and the yield is still only 10% of crude product?


armageddon

  • Guest
yes
« Reply #2 on: July 03, 2004, 05:46:00 PM »
Yes, I didn't distill the lactone, and yield was only 10%.

And I flushed it down the toilet as I just had raw lye to work with back then, no lab grade NaOH or KOH....

As it wasn't meant to be a large-scale production of GBL but rather a test..

Another try with KMnO4/CuSO4 80:20 (0.5mol THF, 200ml H2O, reflux for 6 hours) gave 40% yield... (it seems good to remove formed MnO2 (by filtration) between adding KMnO4 increments, as cooling will be more effective and you have a better visual control of what is goin on)

Would you have been happier if I had distilled the few ml's, getting even less product (maybe 4ml) - of good purity? Or would your flaming become harsher, as the yield would be even worse?  ;D  (I didn't claim the yield is good - I just stated the exotherm is easily controllable)

A similar oxidation was posted here (

Post 304838 (missing)

(Antimony: "Cheap Easy GHB -No Bromate", Newbee Forum)
, but I doubt the usability - ever tried to stir a brown paste of KMnO4/THF/MnO2 magnetically? You will be surprised about how much water is needed to keep it stirrable.. And as GBL is volatile with steam: how does Antimony properly separate GBL, THF and water through distillation?
I dunno..

BTW in Tet.Lett. 23(1), p. 35-38 (1982) (The Oxidation of Alcohols and Ethers Using Calcium Hypochlorite, S.O. Nwaukwa, P.M. Keehn) (a nice article requested by Organikum because it contained info about  benzyl alcohol->benzaldehyde in 98% yield, btw!) the authors also oxidized THF to GBL in 68% yield...

Greetz A


Rhodium

  • Guest
Low yield is better than low purity!
« Reply #3 on: July 03, 2004, 08:22:00 PM »
And I flushed it down the toilet as I just had raw lye to work with back then, no lab grade NaOH or KOH...

I believe the lye would be the least of your health concerns compared to the crude lactone.

Would you have been happier if I had distilled the few ml's, getting even less product (maybe 4ml) - of good purity?

Certainly! If you had purified the product it would be known how much of the crude isolate consisted of butyrolactone, and how much of it was over-oxidized succinic acid, succinic semialdehyde, oxalic acid and other byproducts. Right now we don't know what the actual yield was, nor can the product be used to prepare GHB suitable for human consumption. The literature is full of syntheses where the yield of pure product is awful, but very rarely a crude isolate of unknown purity is reported.

A similar oxidation was posted in

Post 304838 (missing)

(Antimony: "Cheap Easy GHB -No Bromate", Newbee Forum)
, but I doubt the usability - ever tried to stir a brown paste of KMnO4/THF/MnO2 magnetically? You will be surprised about how much water is needed to keep it stirrable.. And as GBL is volatile with steam: how does Antimony properly separate GBL, THF and water through distillation? I dunno..


He said he stirred it with a glass rod, not with a mag-stirrer. I also doubt that he actually distilled his product, he very likely did something similar to your trial, but added those last lines as a kind of suggestion for others to distill their products in case they would repeat his experiment (btw, I have no idea how that post got rated "excellent" - it has now been un-rated).

BTW in Tet.Lett. 23(1), p. 35-38 (1982) (The Oxidation of Alcohols and Ethers Using Calcium Hypochlorite, S.O. Nwaukwa, P.M. Keehn) (a nice article requested by Organikum because it contained info about  benzyl alcohol->benzaldehyde in 98% yield, btw!) the authors also oxidized THF to GBL in 68% yield...

It has been posted in

Post 444048 (missing)

(Aurelius: "Tetrahedron letters 23, 1, 35-38, (1982)", Methods Discourse)



armageddon

  • Guest
succinic acid etc.
« Reply #4 on: July 04, 2004, 06:07:00 PM »
Oops, didn't know that this aspect of said article already had been noticed/discussed beefore... (although its somewhat obvious  ;) )

One word about purity: oxalic acid (toxic) should be insoluble in DCM AFAIK, so it should fall out as a solid upon solvent removal. The other impurities you mention aren't harmfull at all - and succinic acid is sparingly soluble in np solvents (1g/113ml ether!), too...
And as there is a big excess of reductant present, I don't see why overoxidation should be a problem(?).

Right now we don't know what the actual yield was, nor can the product be used to prepare GHB suitable for human consumption.

Yield was 10 percent of theory, and my guinea pigs didn't show strange behaviour nor did they make any complains after the experiment, besides a strong soap taste (tech grade lye)... :P  

(no, I didn't discard it - in retrospect it was bad behaviour to let my poor guinea piggies test it w/o being sure about purity, shame on me  :-[ . And as I didn't want to encourage others to do something like that, I lied in my above post in that no testing on humans was done... :-[  :-[ )

TO OTHERS: DISTILL YOUR LACTONES!!!!!!!!!!!!!!!!!!!!!!!!
(and if you have no vacuum source or are unable to heat to >200°C - steam distill!)

Greetz A


armageddon

  • Guest
improved procedure
« Reply #5 on: July 08, 2004, 06:15:00 AM »
Since our Chief bee wasn't satisfied with the purity of GHB made with above "cheap+easy" method, I thought "why not slightly modify it to give acceptable pure product"? Yields could be somewhat better, but the only hard-to-get chemical needed is THF - the "precursor" in this synth.

So here is the improved procedure:

30 ml THF and 30ml D2O were mixed in a beaker and chilled to below 0° in a fridge. During the time needed to cool sufficiently, ~40 grams KMnO4 were finely ground using a mortar (be careful with the dust). To the prechilled THF/water solution were then added about 4 grams of the permanganate powder. Everything was stirred with a thermometer, watching the temperature. As soon as it hit +7°C, the therm. was rinsed with a bit ice-cold acetone and the beaker put back into the fridge. Every now and then (every 30-45 min.) the mixture was stirred using the thermometer, taking care that the temp. didn't rise to significantly above 10°C (if this happened, some cold acetone was added in little squirts) and put back to cooling.

This was repeated until the reaction had a dark brown appearance and the temp. didn't climb notably higher when being removed from fridge. Then (when NO temp. rise could be seen anymore!), the next few grams of permanganate powder were added and the stirring/cooling/diluting was repeated as necessary (during the last few additions, the reaction became very viscous, so much acetone was added, along with some ~30ml water - end volume was ~200ml), until further additions of permanganate resulted in purple color that didn't fade to brown anymore (in total, adding 34g KMnO4 took more than 14 hours.. :( ). The solution was then allowed to sit some more 2 hours before it was diluted with 50ml EtOH, stirred and again let sit for 2 hours.

The formed MnO2 settled down overnight as a dark brownish solid and on top appeared a slightly milky, off-white solution (containing acetone, water, ethanol, butyrolactone and unreacted THF - in that order). Everything was vacuum filtered through a pad of celite, basified to pH 11 with 20% KOH-solution and the solvents distilled of under "slight" vacuum. the remainder was then acidified to pH 5 by adding H2SO4 dropwise with good stirring, resulting in precipitation of a small amount of K2SO4. The lactone/GHB/water mixture (?) was then steam distilled, the distillate transferred to a 250ml two-necked RBF fitted with a dropping funnel and a reflux condenser and heated to heavy reflux. Next, 2ml of a 20% KOH-solution (I usually prepare standardized solutions of acids and lyes I often use, in this case 20% KOH was what was at hand, maybe other concentrations work better - dunno) was added dropwise via additon funnel and reflux was continued for 2-3 minutes before heat was removed. After having cooled down a bit, the solution was brought to pH 8 with more KOH and the water was evaporated. The hot concentrated solution was then poured into a pyrex bowl and the remaining water slowly evaporated w/careful heating until the GHB potassium salt crystallized (it did so on cooling, together with a small amount of KOH wich solidified too). The total amount of collected solids was 13.4 grams.

This was put in a (clean!) mortar, quickly ground (better don't do this but dissolve the salt directly: it gets real wet real quick when mechanical "crushing" is attempted) and dissolved in some D2O, brought to pH 7.1 by adding some drops of 37% HCl and diluted to a volume of 30 ml with more d.H2O to give a concentration of roughly 0.48g/ml, along with some KCl (probably the best "cut", as K+ ions are exactly what your body  needs when you're on GHB  ;) ).

The so obtained product is a bit salty, but the purity is good enough for human consumption (unless you use impure chemicals of course) - at least I tested it several times until now, and nothing has happened to me yet...

(of course this testing was only done to evaluate the grade of purity and the possible negative effects caused by contaminants - as said, no such effects were observed until now, so there are most likely few enough present to not bother with them)

And remember, the only things you need are:
KOH, H2SO4, acetone, ethanol (NOT denatured alc!! better use vodka (or similar) if u can't get abs.EtOH), THF, KMnO4, a beaker, a thermometer, complete dist. setup (preferably for vacuum dist., but a simple one should suffice), a fridge - and a LOT of patience! I think this is still cheap and easy enough to justify the title of this thread *lol* - good luck.. :)

Greetz A


Rhodium

  • Guest
Yield Calculation: 26%
« Reply #6 on: July 08, 2004, 10:44:00 AM »
KMnO4 - 215 mmol (34g) [158 g/mol]
THF - 370 mmol (30 mL -> 26.66g) [72 g/mol]
K-GHB - 97 mmol [142 g/mol]

Assuming 100% purities and anhydrous products, 0.58 equivalents KMnO4 were used - what is your rationale (out of interest, not disbelief)?

The molar yield gotten was 26% calculated on the amount of THF used - considerably better than last time - congratulations!


armageddon

  • Guest
strange
« Reply #7 on: July 08, 2004, 09:09:00 PM »
Well, purities of water ( :) ), acetone and KMnO4 were 99.99+ percent (analytical grade)...
The (moist) THF ("p. synth.") was present in big excess (at least at the beginning). Maybe I should've added even smaller aliquots of permanganate (or wait even longer between additions) - as I think a good amount of THF evaporated due to heat evolution/long rxn time, thus the rxn ceased already when 0.58 mol KMnO4 were added, as there was no more THF left to react with. At least this came into my mind when realizing that the amount of KMnO4 was less than what was expected...(but then again: I'm sure there was still THF smell present after finishing permanganate addition - maybe I quenched it too early by adding EtOH?)

(maybe some permanganate reacted and oxidized 97mmol THF to GBL, while the remaining 118mmol permanganate did something wierd to part of the remaining 273mol THF, i.e. overoxidation, or were simply used up by the EtOH - and the remaining <200mmol THF partly evaporated, due to long rxn time...)

It probably would be the best to use significantly longer rxn times (like 2 days) and try adding smaller amounts of oxidant (and cooling rxn as strong as possible, maybe experiment with ice/salt bath) and to use only 50% water (percent of volume of THF) to ensure slow rxn. Maybe yields could be improved that way...

(oh, and better consider it being a ~20% yield; the GHB*K was really *very* wet when I weighed it, and the product strenght is indeed slighly below of what was expected (maybe 10% less)... ::) )

(BTW I didn't use deuterium (D2O) but plain distilled water (d.H2O) - "D2O" is just a typo and should read "distilled water"... - So don't go and spend fortunes for deuterated water; it isn't necessary)

Greetz A


Antoncho

  • Guest
Conclusions
« Reply #8 on: July 10, 2004, 03:35:00 PM »
The moral of all these THF oxidations is the fact of their avalanche-like exothermicity: there's no rxn beelow certain temperature, and above certain temperature you've got an xplosion. The acceptable range lies usually within some 15 degrees.

Yet another problem is that the rxn takes a LONG time: in combination with this need of babysitting, it really beecomes a pain in the ass to prepare any useful qtty.

This was seen by SWIM in absolutely all the cases and forms of THF's oxidation he's experienced: Br2/H2O2, Fe/H2O2, Ca(OCl)2...

Sorry, this is a little oftopic here, but i see that you guys have the same problem here. I just want to draw a conclusion: use some form of a thermostat or hire a laborant :) . Or forget about it.

BTW, the highest yield SWIM ever got was around 40%, it was the bromine method, about 12hrs at 30-40 C.


Antoncho

P.S. Once SWIM left a flask with THF/H2O2/H2O/H2SO4/FeSO4/CuSO4 overnight in his bathroom: he thought he was safe since the flask hadn't shown a slightest sign of heating up within two hours.

At 6 o'clock in the morning he and his beeloved wife woke up from the sound of what could bee described as a bathroom's ceiling falling down onto the floor. The (non-closed!) flask exploded so violently that pieces of glass actually were found stuck into the plastic-covered walls.

Take care :o

armageddon

  • Guest
ouch
« Reply #9 on: July 11, 2004, 02:18:00 AM »
Hm, although I don't think that such powerful explosions can occur when you add only small amounts of KMnO4 to a mixture of THF/water/acetone - better be careful... (Ouch! Luckily it happened in an "empty" room  :o )

But the potential destructive power of your rxn is slightly higher, would I say (just from how it sounds  :) ): You have hydrogen peroxide and sulfuric acid present together with water and FeSO4 (BTW isn't Fe(2)-sulfate heptahydrate some reducing agent? Called "iron vitriol"?) - hm, sulfuric and water, oxidizer and a probable reducing agent (does anyone know about this?) together with copper sulfate, and then (as low-boiling "booster") some THF... :-[  :o
But the only thing that will happen if you add too much KMnO4 to aequ. THF is that temp will rise so quick that the THF will boil away; and in a beaker, this should present NO problems.
Even with bigger amounts - the worst thing that can happen (in a WIDE beaker of course  :) ) is a violent boilover = purple mess all around the lab...

But of course, taking all precautions against explosions when working with oxidizers and THF is highly recommended! (Especially when distilling is attempted.. - PEROXIDE DANGER! !!!*REAL*DANGER*!!!)

But, at least, I will never try to perform said reaction after your experience with it, Antoncho! (as a laborant would be too expensive, as is a cooling device with thermostat *lol*)

Greetz A