Secondary and benzylic alcohols oxydations using various oxydants

It doesent exist. I read it somewhere in the last 2 weeks... the ferric ion oxidises iodide to iodine

Edit- somehow i missed woelen saying exactly the same thing


[Edited on 26-10-2007 by Antwain]

I suspected this would be the case. I'm glad you seem convinced of it. For the first few trials I wanted to perform the reaction with a procedure as close as possible as the paper, with only minor differences (FeBr3 aq solution instead of salt). Seeing that sulfate ions are there as a possible contamination anyway, it's true it didn't seem like a huge difference. I still want to give it a try this way until I obtain a isolated yield, and then compare with FeCl3 and KBr, saving me a lot of work. We will see if they really could be interchangeable. I guess the argentin laboratory, and in alot of others, they just needed to open the "anhydrous FeBr3" vial. Opening a "KBr" vial and a "Fe salt" vial would have taken twice the effort
I'm glad we shared the same thought, see the thread in the general chem forum.

I doubted FeI3 would be stable. It's pity, could have been another path of expolration though.

One thing I'm concerned of would be possible ether cleavage when using ether-substitued alcohols, such as 2-methoxybenzylalcohol. FeBr3 is a lewis acid, I'm not sure if it would be powerfull enough in a aq. solution to cause problems on that side. This has just appeared to me, I haven't looked into it yet. I remebered concerns with using AlCl3 and various rearrangements, but I don't recall if it was in aq. solutions.


EDIT: Nicodem, so you think the mechanism would be similar to the one proposed in the KBr/TCCA paper? Then why bother with Fe salts in the first place? They hardly mention hypervalent Fe species in the ref. I'd be interested in your interpretation of this mechanism, as I can't acces the library at the moment.

[Edited on 26-10-2007 by Klute]

Oxydation of benzylic alcohols with TCCA catalyzed by KBr and wet SiO2

M. A. Zolfigol et al. Syn., 12 (2006) 2043–2046

The reaction was performed on a 50mmol scale, with benzyl alcohol.

In a 250mL 4-neck RBF with condenser, thermometer and magnetic stirring, 5.41g (50mmol) of benzylalcohol were diluted in 75mL DCM, under slight agitation. 0.72g (6mmol) of dry KBr were then added, wich evidently didn't dissolve.
5g of granular, amorphous silica gel (made from sodium silicate solution and aq. HCl, washed, dried and activated months ago) were grounded into a fine powder with mortar and pestle. 5g (278mmol) of water were than added dropwise and left to absorb in the silica. The mixture was than grounded to a fluid paste with the pestle. Another ~1g of silica was added and grounded until the paste turned to a powdery solid, easily "aired" with the spatula (1). This solid was then added to the reaction mixture with steady stirring, and formed a snowflake-like suspension, looked like those little decoration objects you shake to make some snow (who hasn't seen one of those?). Once all the hydrated silica was added, 5.05 g (20mmol) of 92% TCCA powder(2), previously grounded, was added in 1/2g portions over 30 min. The first portion made the clear solution turn yellowish. This color darkened to an light orange over the addition, and cleared up to a very light green at the end of the addition. A solid started to stick to the bootom of the flask, even with the vigorous stirring, and was broken down with a glass rod regulary. It often agglomerated into lumps that were broken down with the rod. The rod smelled nicely of benzaldehyde when smelled, with a chlorine-like smell. The temp increased very gently from 14°c to 18°c over the 30min, the TCCA could have been added more quickly.

All in all, it sounds very promising, much cleaner reaction, shorter, and more gentle. A pleasure to work on.

According to another paper on secondary alcohols oxydations with TCCA in acetone that Sauron posted IIRC, I might give a try at addding the TCCA as a solution acetone, with a base to catch the formed HCL, such as sodium acetate (cf US5821374) or triethylamine. Would be much more pratical and avoid the clumping of the cyanuric acid formed (possibly trapping the unreacted TCCA).


Rest of the reaction later on!

(1) Getting the right consistency is essential, as otherwise the silica will stay in lumps even with stirring, which is surely detrimental to the reagent's contact. When it turned from a paste to a powdery solid, it was easily broken down to fine, light flakes with the mag stirring. The suspension is very easily stirred, even with 75ml DCM. Solvant amount could perhaps be cut to 50mL for 50mmol of substrate. On the other hand I'm scarred the silica might damadge the inside of the flask, scrapping it. (Edit: it doesn't, the flask was as new after 12H in the base bath)

(2) Powdered "long-lasting" pool tablets were used, as they contained 92% TCCA, and 100% TCCA was much (twice) more expensive. No reaction with dilute H2SO4, so no carbonates,possibly some cyanuric acid to stabilise or inert materials from production.


EDIT:

After 2h30 reaction time, TLC indicated presenc eof BzOH.. Another 0.5g of TCCA was added in 2 portions, care taken to crash the lumps that formed. After another hour, BzOH was still present in fair quantites, so another 0.5g TCCA was added, as previously. After 4 h total reaction time, the piss yellow DCM was decanted (strong smell of chlorine present), the white solids vacuum filtered (some solids appeared in the filteration flask, dissolved cyanuric acid? BzOOH?), extracted with 2x20mL DCM followed by vac. filteration. The yellow organic phase was washed with 2x100mL sat. Na2CO3, no bubbling (were did that HCl go? All turned to Cl2?)), the washes cleared to color greatly to a faint green/yellow, washed with 3x50ml dH2O until neutral/nearly acidic (BzOOH), dried over Na2So4, and placed in the fridge for dist. of the solvant tomorow. Decided on removing solvant before bisulfite extraction.

A variation with acetone as cosolvant, or solvant entirely, and a base to capture the HCl, which seems to quantitatively turn to chlorine, possibly chlorinating and/or oxidizing the product, seems to be better approach, although this system appears to be much more efficient and clean thatn the Fe3+/H2O2.
The TCCA clumping up, and the HCl turning to Cl2 are the major drawbacks. A base seems necessay when using acetone as solvant or chloriantion of the solvant to lachrymatory chloroacetones could happen from what I've read in related threads. We will see how effective this system is tomorow, with the isolated yields.

And now the deserved and awaited bedtime

[Edited on 29-10-2007 by Klute]

[Edited on 31-10-2007 by Klute]

BzOOH crystals were present in the erlenmeyer containing the solution, although it was kept under Argon. It was decanted from the Na2SO4 in a 250mL flask, and the dessicant washed with a little fresh DCM. The DCM was removed on a 50°c water bath, under argon, to leave a clear oil, very slightly yellow, smelling nicely of benzaldehyde, still containg a little DCM. It was transfered to a 100mL beaker, and a fresh sat solution of metabisulfite added (~15mL), and everything stirred after the addition of 20mL denat EtOH. A beautiful white precipitate formed after a few minutes, making a thick slurry. More EtOH was added to thine up the slurry, and everything left to stir for 15min, then filtered, washed with water, EtOH, and ether. After being left to dry by suction of the buchner for 20min, and dried in a 40°c stove, the crystals weighed 4.28g (20.38mmol), 40.75% yield.
Some yellow oil droplets remained at the bottom of the aq./EtOH filtrate, surely unreacted BzOH contamined with impurities.

Not too bad for a first try. Leaving the solution over night caused aerial oxidation, and not trapping the formed HCl must have formed BzOOH by oxidation, aswell as soom BzOCl, wich of course turned to more BzOOH upon aq. workup.

A trial with acetone as solvant and using a weak base will be tried tomorow, hopefully a workable yield will be obtained.

BTW, FeBr3/H2O2 gave 79% yield of butanone, as a bisulfite adduct. Works well with secondary alcohol. Cyclohexanol must work out well, although I do not have some on hand, only cyclohexanone. And I'm not good to reduce it just to try oxydizing it back again


The TCCA/acetone/pyridine paper I mentionned, thanks to Sauron for posting this.
TCCA in acetone in presence of pyrindine oxydation of secondary alcohol
[Edited on 31-10-2007 by Klute]

[Edited on 31-10-2007 by Klute]

So I guess adding Et3N isn't as efficient I supposed it to be. I will just stick to the TCCA/KBr protocole then. But having a certain amount of unreacted BnOH (thank you for that reactification! ) is pretty annoying, especially when using solid subsitued BnOH's, seperating the two aren(t that easy wihtout bisulfite adduct formation. A few recrystallizations might do the trick.

During the oxidations, the temp rarely goes above 10°c, and I use only neon light to work, this might be less of of the problem if the photodegradation is dependant on the spectrum of the light. I think I will cover the reaction flask with aluminium.

What do you suggest to avoid consumption of the TCCA by the HCl? I just can't understand how they supposbly get so high yields in the paper, while the oxidation seems uncomplete even when using bigger excess and longer reaction time. At the same time, they simply evaporate the solvant and collect the residu. It would seem pretty obvious that they used GC to determine the purity of the residu, and correct the yield, but it 's isn't explicitly mentionned anywhere.

I'm thinking of trying the reaction on 2-MeOBnOH, and 5-bromo-2-MeOBnOH, aswell as either 4-BrBnOH or 4-NO2BnOH (although they have bad yeilds on this one, can't imagine how bad it will be when I try it ) because I have then on hand. If the reaction give correct yields without byproducts, even if uncomplete, I think we can safely say it's a valid protocole, necessitating some recylcing in order to avoid waste of substrates. I will still send a few samples to GC/MS to see if any chlorinated product are present, as they could possibly be hard to differentiate with TLC.

Again, any suggestions are welcome.

I suppose the result would be the same as using Na acetate, ie a thick slurry, needing huge amounts of solvants. As there's already the silica and the KBr in solid phase, having too much solids will surely prevent good contact between all the reagents...

The TCCA in EtOAC gave a little under 35% yield of the bisulfite (minium EtOH washes), and some chloroacetone was formed as I cried my eyes out when the solvant was removed and I dismantled the distn setup... So some chloroacetone is present as an adduct in the white powder I guess...

I really don't know what else to do to get this working correctly. I will try adding Et3N anyway, to see if that really changes something or not. I really feeling disapointed.

EDIT: I'm going to try solid Ca(OCl)2 with BnOH in DCM in presence of Bu4NBR... We will see what that gives... I might also try a ACOH/HOCl biphasic oxidation... I'll finish by finding one procedure working if I try all of them, no?

[Edited on 3-11-2007 by Klute]

"Please tryt it out ourself if you have the possibility, on any seconday or benzylic alcohols."

The only secondary alcohol I have is IPA, and making acetone is not exactly exiting, it being three times cheaper than the IPA over here.
That is unless it gets banned due to teenies and terrorists peroxidising it...

The only benzylic alcohol I have is much less boring, and stands right next to the nitroethane and butylamine. But I would never break laws just to verify a concept!
Oh btw it is 3,4-methylenedioxy-benzylalcohol....

Thank you trilobite, for the article, I wonder why I didn't stumble on them before. A few years ago, I did alot of experimenting with theis Cu2+/Fe2+/S2O82- system, both with toluene and benzyl alcohol. Not once did i manadged to get a decent yield of the aldehdye, although it smelled strongly of almonds. The brown/red organic phase after the 2hours of reaction had alot of impurities, benzoic acid was present at all times, and was some times the major product (very large cristals on cooling). I had put alot of time and effort in the method (and alot of reagents), and just finished by forgetting about it.
The only information I had back then was the patent, which obviously sounded very good. I used Cu(OAc)2 and FeSO4.7H2O as salts, and 99.9% lab grade ammonium peroxodisulfate (more soluble than the K persulfate).

Have you tried out this method? (not the Mn persulfate) What where your results if any? I'd love to hear some succes stories about it, the reaction itself was quite simple and easy (never got any runaway, although I used 600mL max total solution).

I think I'm going to try freshly precipitated MnO2, at a small scale, although from what I heard it's a bitch to work with...

BTW, benzaldehyde itself isn't really a major concern to me, I've got a bottle of lab grade and not much uses to it, it's more about finding generally oxidation method, that will equally work with substitued benzaldehydes, but I'm not going to waste expensive/rare substrates before hand!

Very interesting! One of my questions back when I tried to find whatever research that had been done with free radical oxidations like these was whether you could use (NH₄₂S₂O₈. I myself have never tried the reaction, but have been planning to try it out when I get the chance. The reason for my enthusiasm is that the mechanism is quite well established by several authors, but the patent is where these principles are applied to practice. You may have the references if you wish so. It is easy to understand why you may not be as enthusiastic, though. Here is what I think could have been the problem.

First toluene is oxidized to a radical cation by a sulfate radical, the radical cation loses a proton and becomes a benzyl radical. And now the important part: Cu²⁺ ion oxidises the benzyl radical further in a one electron oxidation step, and one of ligands coordinated to the metal end up on the benzylic carbon. Usually you'd expect to have a benzyl carbocation from such an oxidation, which would then react with eg. water or some other nucleophile swimming nearby. But in this case one of the ligands coordinated to the copper ion acts as the nucleophile.

So if you have a lot of chloride ions in your solution, you get some benzyl chloride, high concentrations of acetate give some benzyl acetate and so on. In the presence of ammonium ions opper ammine complexes can form, especially since relatively small amounts of copper are used. These complexes could then cause formation of benzylamine as product. Benzylamine might be further oxidized to benzonitrile, or it might form an imine with benzaldehyde which might react further some way. I guess this sort of reaction pathways could be possible also for benzyl alcohol. Of course they depend on the pH of the reaction mixture, which decreases as the reaction proceeds.

Lower pH should also decrease the concentration of copper ammines. Actually ammonium peroxydisulfate has been used in some of the papers by other researchers where this sort of reactions have been studied, but typically in an acetic acid or acetonitrile solution, not aqueous methanol. Usually these articles focus on mechanistic studies, such problems might not show up when there is something like 0.2 or 1 equivalents of copper used in relation to the substrate or oxidant. In the patent several solvents are mentioned, including acetic acid and acetonitrile, but no example is given where these solvents or ammonium peroxydisulfate were used, so I guess they are there to claim as much as possible. The downside of using potassium peroxydisulfate would be that it should either be added as a slurry (stirrer in the addition funnel?) or portionwise as powder. This would be a rational explanation for why these guys seem to use the sodium salt all the time So, maybe this could be behind your troubles.

Here are abstracts on similar reactions which give nitriles. They are all done in basic conditions, as they obviously should be.



<b>New synthesis of nitriles.</b>
Brackman, W.; Smit, P. J.
Recueil des Travaux Chimiques des Pays-Bas, 82(8), 757-62 (1963).
Journal written in English. CAN 59:75021 ISSN 0165-0513

<u>Abstract</u>
Nitriles were prepd. by stirring a methanolic soln. of an aldehyde, NH3, a complexed Cu salt, and a strong base under O. Under these conditions, MeOH was slowly oxidized to give low yields of cyanide and cyanate. With MeOH as the only substrate, substitution of NH3 by a primary amine led to the formation of small amts. of isocyanides. Thus, 100 ml. soln. of 4 millimoles CuCl2.2H2O, 400 millimoles NH3, 30 millimoles NaOMe, and 50 millimoles BzH in MeOH at 30 Deg stirred 6 hrs. with O gave 66 PhCN. The yield was 79% when the reaction time was 23.5 hrs. Other nitriles prepd. were the following (compd. and % yield detd. by mass spectroscopy given): p-MeOC6H4CN, 82; 3,4-methylenedioxybenzonitrile, 82; 4,3-HO(MeO)C6H3CN, >40; o-O2NC6H4CN, 51; 2,6-ClC6H3CN, >76.5; 2-cyanofuran, <10; CH2:CHCN, trace; EtCN, 18; PrCN, 28; Me(CH2)5CN, 63, cinnamonitrile, 55. Aromatic alcs. were also employed, but the yields were low. H2O retarded the reaction. A mechanism involving the oxidn. of aldimine to an imine radical by the Cu(II) complex as the rate-detg. step was proposed.

See also patents DE1169937 and GB920987


<b>A catalytic synthesis of nitriles from aldehydes and alcohols in the presence of aqueous ammonia by oxidation with NiSO4-K2S2O8.</b>
Yamazaki, Shigekazu; Yamazaki, Yasuyuki.
Chemistry Letters, (4), 571-4 (1990).
Journal written in English. CAN 113:58637 ISSN 0366-7022

<u>Abstract</u>
Arom. and conjugated aldehydes and alcs. were converted to nitriles by nickel-catalyzed oxidn. in the presence of aq. NH3 with K2S2O8 under the basic aq. conditions. Thus, PhCHO was treated with K2S2O8 in the presence of NiSO4 in H2O-NH3 to give 76% PhCN. Similar treatment of PhCH2OH gave 82% PhCN.


<b>Tetrabutylammonium peroxydisulfate in organic synthesis. Part 8. An efficient and convenient nickel-catalyzed oxidation of primary amines to nitriles with tetrabutylammonium peroxydisulfate.</b>
Chen, Fen-Er; Peng, Zuo-Zhong; Fu, Han; Liu, Ji-Dong; Shao, Lan-Ying.
Journal of Chemical Research, Synopses, (12), 726-727 (1999).
Journal written in English. CAN 132:92845 ISSN 0308-2342

<u>Abstract</u>
Primary amines were oxidized to the corresponding nitriles in excellent yields with tetrabutylammonium peroxydisulfate catalyzed by nickel copper formate under basic aq. conditions. Thus, treatment of MeCH2CH2CH2NH2 in ClCH2CH2Cl with (Bu4N)S2O8, aq. Cu(HCO2)2/Ni(HCO2)2, and aq. KOH at room temp. for 10 h gave 92% MeCH2CH2C.tplbond.N.


<b>Tetrabutylammonium peroxydisulfate in organic synthesis. Part X. An efficient nickel-catalyzed one-pot synthesis of nitriles from aldehydes by oxidation with tetrabutylammonium peroxydisulfate.</b>
Chen, Fen-Er; Fu, Han; Meng, Ge; Cheng, Yu; Lu, Yin-Xiang.
Synthesis, (11), 1519-1520 (2000).
Journal written in English. CAN 134:29005 ISSN 0039-7881

<u>Abstract</u>
Various aliph., arom. and heterocyclic aldehydes were efficiently transformed to the corresponding nitriles in a 1-pot procedure by Ni-catalyzed oxidn. with (Bu4N)2S2O8 in the presence of NH4HCO3 under basic aq. conditions. The process affords excellent yields of pure nitriles.


<b>Tetrabutylammonium peroxydisulfate in organic synthesis. XIII. A simple and high
ly efficient one-pot synthesis of nitriles by nickel-catalyzed oxidation of primary alcohols with tetrabutylammonium peroxydisulfate.</b>
Chen, Fen-Er; Li, Yong-Ye; Xu, Mei; Jia, Hui-Qing.
Synthesis, (13), 1804-1806 (2002).
Journal written in English. CAN 138:204807 ISSN 0039-7881

<u>Abstract</u>
A facile 1-pot method is presented for the synthesis of nitriles from the corresponding primary alcs. by Ni-catalyzed oxidn. with tetrabutylammonium peroxydisulfate in the presence of ammonium hydrogen carbonate under basic aq. conditions. This convenient synthetic method provides an easy and simple access to various aliph., arom. and heterocyclic nitriles in excellent yields with very high purity.


I also stumbled upon this article that discusses the oxidation of alcohols with MnO₂ prepared by thermal decomposition of manganese carbonate or oxalate. Usually it is assumed that the oxide needs to be prepared by reduction of permanganate The reference is Journal of Organic Chemistry, 19, 1608-1616 (1954).

Keep up the good work!

Attachment: JOC_1954_1608-1616.pdf (566kB)
This file has been downloaded 1288 times

GC confirmed 99,4% 3,4-dimethoxybenzaldehyde, straight after recrystallization! The minor impurity is the starting substarte, wich seems to pass in very small amounts during steam distillation. The last batch, the less pur crysatls, contained a bit more alcohol, and a very small amount of colored impurity that doesn't elute with DCM. After recrystalization, it was impossible to differentiate between the two samples..
The total dried wait of the isolated solids thus sums up to 6.75g (40.62mmol) 81.24%!
I haven't distilled the DCM extracts of the aq. distillate yet, I hope I can get another 0.1-0.3g out of it...

Can't wait to try the reaction on another substarte, but unfortunaly, it's not until a little moment...

Now that's what I wanted to hear! If you ever want to test the limits of this method, then 2-hydroxybenzyl alcohol, vanillyl alcohol, 3,4-dihydroxybenzyl alcohol and maybe 2,5-dihydroxybenzyl alcohol should be simple but tricky substrates with activated rings, possibility for acid-catalyzed polymerization and also oxidation to the quinone (dihydroxylated species). Interesting curiosities, but maybe not the most practical substrates that one would try first.

Preparation of Triacetoneamine(4-Oxo-2, 2, 6, 6-tetramethylpiperidine), an Improved Method
G SOSNOVSKY, M KONIECZNY - Synthesis, 1976 - thieme-connect.com
Thieme-connect / Abstract, Contact Us. communication, Synthesis 1976; 1976:
735-736 DOI: 10.1055/s-1976-24178.

Sommelet reaction on the supposed 3,4-dimethoxybenzylchloride

Update: I've finally filtered the recrysatllized material, giving small beautifull white shiny needles. A rough mp seems to indicate a value close to the starting material (supposed chloride), a little on the low side though a rustic method was used as I'm not at work. It also has the same odor when melted. So it's definatively not the aldehyde. The sommelet hydrolysis went exactly like the different describtions I've read. So maybe it wasn't the chloride after all. The mp is still too high to be the alcohol. I'll do a TLC on it when I'll have time, to compare it with the crystals that were used for the sommelet, and the alcohol. I don't get it. I really don't see what else the oxidation could have produced. The starting material is of unquestionnable purity/identity.
Any ideas?

When I'll get back onto the 4-oxoTEMPO project, I'll try a oxydation on the same material and compare the products. I can't figure out what to do otherwise, not being able to access either NMR or IR in near futur.