About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Anti-GMO. Until This Week. | Main | Grad School in Chemistry - The Mental Health Aspects »

January 7, 2013

Oxidizing Alcohols - With Water

Email This Entry

Posted by Derek

This looks like an interesting reaction; let's see what gets made of it. David Milstein's group at the Weizmann Institute in Israel report a new catalytic system to oxidize alcohols to carboxylic acids, with water as the oxygen donor (as shown from labeling experiments). Hydrogen gas bubbles out of the mixture. The catalyst is a ruthenium complex, and although the reaction is not especially fast (18 hour timescale), the turnover numbers seem to be good (0.2% catalyst loading). Interestingly, oxygen actually seems to hurt the catalyst; the system runs better under argon. One possible drawback is that the ruthenium catalyst can serve as a hydrogenation catalyst - alkenes are reduced, what with all the hydrogen around.

Getting rid of (most of) the metals and the high-valent reagents will be worth the trouble industrially, as will getting rid of the need for pressurized oxygen. As it is now, many carboxylic acid compounds are produced on scale via either alkenes (hydroformylation and then oxidation of the aldehydes with a catalyst, or carbonylation), from alkanes via nonselective oxidation in air, or from alcohols via carbonylation.

We're still a long way from ditching the current processes, but if this reaction is robust enough, it could open up some new industrial feedstock routes. (One that I wonder about is replacing the current route to adipic acid, used in Nylon production. It's currently made through a rather foul nitric acid process - if there's enough hexanediol in the world. (Not sure if that's feasible, though - it looks like most of the hexanediol is made instead by reducing adipic acid! Makes you wonder if there's a potential biological route, as there is for butanediol). Edit - fixed this part, due to dropped some carbons between my brain and the keyboard this morning). Someone may also find a nice use for the hydrogen that's given off, and get some sort of two-for-one process. At the very least, this is a reminder of just how much more metal-catalyzed chemistry there is to be discovered. . .

Update: one of the paper's authors has dropped by the comments section, with interesting further details. . .

Comments (24) + TrackBacks (0) | Category: Chemical News


1. Kazoo Chemist on January 7, 2013 8:41 AM writes...

Interesting chemistry, but butane diol to adipic acid?

Permalink to Comment

2. David Borhani on January 7, 2013 9:19 AM writes...

Replace the argon by CO2! ;-)

Permalink to Comment

3. ek on January 7, 2013 10:50 AM writes...

Thanks for highlighting the chemistry Derek. I'm one of the two first authors, but I was stupid enough to make myself second even though I wrote it, since I thought alphabetical order was fair. Now I know you have to be ruthless for career bean-counting purposes :) I'm also a regular reader of this blog, so I was very surprised when I saw this post. Pleasantly surprised.

At first I couldn't believe this chemistry was real, because I thought there must have been trace oxygen from somewhere so there were lots of controls and tests. But it's now been confirmed in related reactions by others in the group as well. I get a couple percent of acid in a related reaction in a closed vessel under pressure. Not that surprising in retrospect.

The Ru catalyst's expensive nature makes it a bit iffy for now... but from a paper I had rejected by JACS on related catalysis, I can tell you that the basic first step of the reaction (dehydrogenation to make the aldehyde) is very, very robust in secondary and benzylic alcohols. The turn over number may be improved dramatically in the future with a bit of tweaking. You can probably already make benzylic acid on huge scales with just this catalyst as of now. Also, for the problem of stoichiometric strong base, the group is working on it, but it probably won't be an article from me if that gets solved.

David Borhani: If you look at some other articles from the group, you can see that CO2 chemistry is being actively pursued. I'll pass on any suggestions to others.

Permalink to Comment

4. LittleGreenPills on January 7, 2013 11:44 AM writes...

I am curious about the unwanted hydrogenation. Is it effective enough that it could be used for intentional hydrogenations?

I assume getting rid of pressurized H2 would be as desirable as getting rid of pressurized O2. I know I never liked charging a reaction vessel with H2.

Permalink to Comment

5. ek on January 7, 2013 12:46 PM writes...

The unwanted hydrogenation turned out to be pretty efficient to a degree that a significant amount of the substrate is hydrogenated that it matters. Even for non activated double bonds... I'm guessing that the concentration of H2 in solution must be pretty low before it escapes, so the catalyst is very efficient at this.

You don't get a build-up of pressure if you have a closed vessel. The alcohol/aldehyde+H2 reaction is very reversible. This was the subject of my other paper which got rejected by JACS. I was waiting for this to get published before re-submitting. I don't remember if I've tried a closed vessel reaction with an internal double bond (cinnamic alcohol) that you want to get rid of on purpose though... seems like a really good idea but for the life of me I can't remember after the 20-30 revisions. But you're generally not looking to destroy functionalities in synthesis...

Permalink to Comment

6. G2 on January 7, 2013 1:09 PM writes...

Hey ek, do you think its fair to bash on the depicted first author? You had the chance to solve this during preparation/submission. At least the corresponding paragraph in the paper is clear:


E.B. made the initial discovery, carried out catalytic experiments and wrote the manuscript. E.K carried out catalytic experiments, stoichiometric experiments, DFT calculations, synthesis and crystallization of complex 10, and wrote the manuscript.

Permalink to Comment

7. ek on January 7, 2013 1:22 PM writes...

I'm only bashing myself since I made up the author order (the paper was written mostly by me), but I realized later that placing of first author matters. :) E.B. is a good friend of mine; we worked on several things together and he made the initial discovery here. Without him this paper would be nowhere.

I'd rather not discuss this and focus on the chemistry. That'll be my last comment pertaining to that issue.

Permalink to Comment

8. ek on January 7, 2013 1:28 PM writes...

Incidentally, here's another good paper by E.B. in Nature Chem involving CO2 hydrogenation with these catalysts.

Permalink to Comment

9. milkshake on January 7, 2013 2:14 PM writes...

ek: have you tried to run the reaction under elevated nitrogen pressure, at 120-150C, to see if you can get NH3 from the produced H2? Ru pincer complexes are extremely interesting for low-pressure Haber process, in 2011 Schneider group reported nitrido-Ru pincer complex hydrogenation even at 50C

Permalink to Comment

10. gippgig on January 7, 2013 2:20 PM writes...

What's the thermodynamics of this reaction (don't have access to the paper)? Offhand, it sounds downright endothermic to me.
This reaction should get a lot of attention from the "hydrogen economy" people. Being able to make hydrogen easily from sugars has significant implications.

Permalink to Comment

11. milkshake on January 7, 2013 4:55 PM writes...

I don't think this catalytic dehydrogenation is overly endothermic - the carboxylate salt formation acts as a thermodynamic sink, and hydrogen partitioning out from the reaction mix also helps to shift the equilibrium. The same catalyst would probably work great for catalytic hydrogenation of carboxylic acid and CO2 to alcohols, using neutral conditions and under H2 baloon. This would be quite valuable, from a process chemistry perspective

Permalink to Comment

12. David Borhani on January 7, 2013 5:59 PM writes...

@ek, #3 --- I was joking about running the reaction under CO2, so it's pretty interesting to hear that it may actually work! Maybe Derek was right on the money with butanediol to adipic acid to nylon...

Permalink to Comment

13. ek on January 7, 2013 6:13 PM writes...


The reaction has been done under nitrogen in a closed Young tube, but not on purpose to get nitrogen splitting. I haven't seen anything to hind at it personally, and I would have recognized the ammonia peak in the NMR for sure... I don't think I've personally had an N2 and H2 mix in a Parr apparatus, but it's worth a try: for someone else at this point.


milkshake is right, the formation of the carboxylate salt with NaOH drives the reaction, and the equilibrium helps. However, so far there have been , uh... snags in converting CO2 to alcohols. The group is working on it and you might see some reports in the future. Obviously, this is one of the holy grails for the boss. As for the carboxylic acid (or salt) to alcohol reaction, I couldn't get it to go under a few different conditions and personally gave up on it a long time ago. I'm sure others will try it with their respective systems.

If you guys are interested in this stuff further, you can check out the work of our biggest competitor/colleague Matthias Beller as well. Also Hansjoerg Gruetzmacher has some similar chemistry, among others.

Permalink to Comment

14. Nick K on January 8, 2013 1:48 AM writes...

#10,11: From heat of combustion data I calculate the first reaction (alcohol to aldehyde) to be endothermic by about 110kJ/mol. Presumably, as Milkshake says, the formation of the carboxylate anion is sufficient to render the reaction possible.

Permalink to Comment

15. pugwash on January 8, 2013 4:52 AM writes...

Drink it?

Permalink to Comment

16. Anonymous on January 8, 2013 11:02 AM writes...

ek, you mentioned in #5 that alcohol -> aldehyde+H2 reaction is very reversible, which is consitent with that the formation of a carboxylate is necessary to favor the equilibrium. At the same time this first stetp is robust in secondary and benzylic alcohols, any thoughts on the reason?

Permalink to Comment

17. ek on January 8, 2013 4:13 PM writes...

Anonymous, if I had to give a justification for it, I would probably get back to chemistry basics and say that the stability of the product favors its formation. The benzylic aldehyde is conjugated to the ring pi system. Secondary alcohols the sp2 conjugation will have more p character and more hyperconjugation because there is an extra beta carbon with it's own sp3 orbitals. That explains the product aldehyde/ketone formation. With a simply primary alcohol, you don't have that stabilization there of what would become an sp2 carbon and a pi bond.

The reverse reaction goes fast because the catalyst is very robust in hydrogenation; it will even go after unactivated CC bonds.

That's all a little hand-waving and based on what I got from reading Anslyn and Dougherty mostly. I'm still standing by the statement that the rate determining step is H transfer from the arm to form H2 since that seems to be the case in others' DFT studies that are linked to in the article... What we have is a putative mechanism with plausible intermediates as I didn't do a full DFT study here; that would be a whole separate project.

It could very well be a bimolecular reaction with an aldehyde trans to that Ru-H, or it could be a trans dihydride complex (where H2 elimination is more favored). What I know from related experiments in organic solvents, is that it's very easy to form the aldehyde even for primary alcohols stoichiometrically. It happens instantly at low temperatures and you see aldehyde, ester, a few complexes, but not H2. Maybe in the case of benzylic aldehyde, the back reaction is slow enough (due to it being more stable), so that water can attack and form the gem-diol, so you see more product.

Maybe my statement that this particular step is robust for secondary and benzylic alcohols is wrong then. The end result is, that you see more product: acid in case of benzylic alcohols and... something else in case of secondary ones (don't think I'm allowed to say until it's published). The mechanism is still a bit of a black box and I'm not completely satisfied with mechanistic studies that have been done by on it, both DFT and experimental.

Permalink to Comment

18. raja on January 9, 2013 12:57 AM writes...

nice and very imp.reaction

Permalink to Comment

19. raja on January 9, 2013 1:05 AM writes...

Hi: did the reaction work for poly hydroxy compounds?

Permalink to Comment

20. ek on January 9, 2013 9:08 AM writes...


If the alcohol functionalities are far away, then yes, there are a few examples in the paper. For ethylene glycol, no. That one doesn't work probably because it forms some sort of stable metallocycle. EB was trying to polymerize it in organic solvents, as it would be good even for hydrogen storing purposes, but it just doesn't seem to behave well. That said, there is someone in the group working on polyalcohols, such as glycerol, etc.. right now. We'll have to wait and see how that pans out.

See the following two related articles:

DOI: 10.1002/pola.25943
DOI: 10.1039/C1CC15778G

Permalink to Comment

21. Raja on January 10, 2013 12:55 AM writes...

Hi ek, Hydrogen storage using EG is very interesting. I feel like Na.formate from MeOH (rather than EtOH) would be of current interest using your method.

Permalink to Comment

22. ek on January 10, 2013 4:41 AM writes...


Yes, that is one of the main reasons why I joined the Milstein lab. I thought if the catalyst could be made to work at r.t. then there is really good potential for hydrogen storage there with Milstein chemistry: 2ROH --> ester + 2H2. This is reversible under hydrogen pressure. This reaction was first discovered by a guy named Jing Zhang when he was a postdoc here. I actually know him from being my roommate in the States after he left the Milstein lab.

Methanol is the best for percentage storage of hydrogen (ethylene glycol is good too though). There are a number of problems. The boiling point (and especially that of methyl formate), is below that of the working temperature of the catalyst. But even with ethanol you get >6% hydrogen storage by molecular weight, which was not the DOE target of 9%, but I calculated it to give the energy of a 50 liter tank of gasoline with ~200K of ethanol. This means a ~270 litre tank, which sucks but is doable if you can generate hydrogen on demand and refill it easily. It's probably best for a stationary location though. The advantages are numerous. Ethanol and ethyl acetate are known liquid compounds and can be stored under air. Plus ethyl acetate can be burned itself when you run out of hydrogen and need power urgently. Ideal for a place where you make H2 from splitting water by solar electricity just like George Olah wants to in his utopian vision for the future in 'The methanol economy', just modified a bit to use H2 as power source and ethanol as storage medium.

Anyways, that's just the idealist picture I had when joining the lab. David liked the idea, but I soon started working on other projects. I modified the ligand, but everything that I made did not give better activity. Finally, me and EB decided that we would collaborate on this thing and try to publish at least something. He would do the catalysis and I would tell him what to do and write the paper, which was fine with me. We didn't have enough toys in the lab to make fancy stuff like a partial vacuum to help H2 escape. The best we got was full conversion of ethyl acetate to ethanol. The back reaction with the same batch went ~60%, and then that ethyl acetate plus H2 didn't go much better. The subsequent cycle took the yields down even further. The catalyst didn't survive so well, and that was at 0.2mol% loading. I guess far away from anything practical. Eventually we sat on it for too long and I think Matthias Beller published something similar so we lost out there. It's an obvious idea for any group working with this chemistry, but a breakthrough in catalyst activity has to occur for it to be even close to practical. This might involve tweaking the ligand further (maybe with the help of DFT), or high-throughput screening which... we don't have.

Permalink to Comment

23. raja on January 11, 2013 2:37 AM writes...

Did the Milstein Cat. (Pyridine based PNN) works in the similar manner to the way bipy. PNN works. I saw several examples using bipy ligand. What is unique nature?, though it may be less liable.

Permalink to Comment

24. ek on January 11, 2013 6:29 AM writes...

The bipy works in a similar manner to the pyridine based PNN. For me, the advantages include easier synthesis of ligand to test a new route, though I did it only once as there is a lab technician who does the synthesis, and it's easier to crystallize stuff in an unrelated project, but also the crystal structure in the current paper. Also you get better NMRs of the complexes and it seems to be more hydrophilic, although both Cl complexes are bricks in water until deprotonated. It should be less labile like you say, but there is the possibility of ring flip there as well.

Most of the group has switched to the bipy as it appears to be easier to make by the route the technician uses... and by my route as well. However, in terms of catalysis, there are differences in performance depending on the reaction and substrate. Hopefully a comprehensive study on various catalysts (pyridine PNN, bipy, etc... and PNP) and product distribution (including some improved conditions) in some reactions that were previously reported like ester and amide synthesis, will be published sometime this year or next.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry