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

« Eight Toxic Foods: The Aftermath | Main | One Conspiracy After Another »

June 24, 2013

How Robust Is That New Reaction, Anyway?

Email This Entry

Posted by Derek

Here's a paper in Nature Chemistry that addresses something that isn't explicitly targeted as often as it should be: the robustness of new reactions. The authors, I think, are right on target with this:

We believe a major hurdle to the application of a new chemical methodology to real synthetic problems is a lack of information regarding its application beyond the idealized conditions of the seminal report. Two major considerations in this respect are the functional group tolerance of a reaction and the stability of specific chemical motifs under reaction conditions. . .

Taking into account the limitations of the current methods, we propose that a lack of understanding regarding the application of a given reaction to non-idealized synthetic problems can result in a reluctance to apply new methodology. Confidence in the utility of a new reaction develops over time—often over a number of years—as the reaction is gradually applied within total syntheses, follow-up methodological papers are published, or personal experience is developed. Unfortunately, even when this information has evolved, it is often widely dispersed, fragmented and difficult to locate. To address this problem, both the tolerance of a reaction to chemical functionality and of the chemical functionality to the reaction conditions must be established when appropriate, and reported in an easily accessible manner, preferably alongside the new methodology.

This is as opposed to the current standard of one or two short tables of different substrates, and then a quick application to some natural product framework. Even those papers, I have to say, are better than some of the stuff in the literature, but we still could be doing better. This paper proposes an additional test: running the reaction in the presence of various added compounds, and reporting the % product that forms under these conditions, the % starting material remaining, and the % additive remaining as well. (The authors suggest using a simple, robust method like GC to get these numbers, which is good advice). This technique will give an idea of the tolerance of the reagents and catalysts to other functional groups, without incorporating them into new substrates, and can tell you if the reaction is just slowed down, or if something about the additive stops everything dead.

Applying this setup to a classic Buchwald amination reaction shows that free aliphatic and aromatic alcohols and amines kill the reaction. Esters and ketones are moderately tolerated. Extraneous heterocycles can slow things down, but not in all cases. But alkynes, nitriles, and amides come through fine: the product forms, and the additives aren't degraded.

I like this idea, and I hope it catches on. But I think that the only way it will is if editors and reviewers start asking for it. Otherwise, it'll be put in the "More work" category, which is easy for authors to ignore. If something like this became the standard, though, all of us synthetic chemists would be better off.

Comments (18) + TrackBacks (0) | Category: Chemical News | The Scientific Literature


1. anon the II on June 24, 2013 11:40 AM writes...

One of the reasons that B. Sharpless deserved and got the Nobel prize in chemistry was the extent to which he provided details on how to make those chiral reactions work, how you could use the products and how you could handle some of the exceptions and problems. He even provided detailed work-up procedures for special cases.

Of course, there are other Nobel prize winning chemists who won for methodology, that chemists are still complaining about trying to reproduce their work with little luck.

It's not a new concept. It's just not been embraced broadly for a variety of reasons.

Permalink to Comment

2. anon the III on June 24, 2013 11:51 AM writes...

"Nature Chemistry is a monthly journal dedicated to publishing high-quality papers that describe the most significant and cutting-edge research in all areas of chemistry."

As anon the II said, this is not a new concept. This is a fancy title for "control experiment". Is it a "robustness screen" to spike a reaction with water to test its water sensitivity? How about to run a reaction in DMSO or THF to test its DMSO or THF sensitivity?

The fact that this was published in a supposedly high-impact journal signals trouble for the field.

Permalink to Comment

3. yoyomama on June 24, 2013 11:53 AM writes...

One of the authors (KDC) was the blogger responsible for "A Retrosynthetic Life".

Permalink to Comment

4. Anonymous on June 24, 2013 11:54 AM writes...

I agree Derek. I think the field would be much better off if every methods paper included a 'Glorius Table' in the supporting information to demonstrate how robust the reaction is.

Permalink to Comment

5. Mad Dog on June 24, 2013 12:08 PM writes...

It is all in the execution. What about a standard set of 12 differing reagents that would always be screened against a new methodology? Alrich et. al. could make a 'kit' where they sell ampules of the 12 control reagents. The price markup would be somewhere north of 1000%; so there is a strong business case for them to ask the journals to require this validation.

Permalink to Comment

6. Anonymous on June 24, 2013 12:18 PM writes...

An experienced, well educated chemist should be able to derive a lot of information from methodology papers. 12-15 substrates with different functionality are a good start, but by no means will a single method published as a communication earn you a name reaction. Perfect conditions that will always work no matter who or what is way out of the scope of such work. If you change from 'bramley' apples to 'granny smith' your cake will taste different (and there is nothing wrong with that). On the other hand, the extend of next to useless tables in high impact publications should be a concern. I often wonder how these make it through peer review. I guess we all tried one or the other infamous Tetrahedron Letters method as undergrad. This is part of the learning process of becoming a chemist. Adding two more tables with low yields in unreasonable solvents or by adding stuff you find on the shelf will not really do much about the underlying issue.
Of course, adding compounds to a reaction that runs neatly under optimised conditions might give you an idea of its scope and limitations, but tolerance of the reaction conditions to chelating heteroatoms or sensitivity of your substrate to heat are not really something spectacular.
Anyway, congrats for making it into Nature Chemistry with that.

Permalink to Comment

7. David Borhani on June 24, 2013 12:28 PM writes...

I agree with anon the II regarding Barry Sharpless. As an undergraduate working in his lab just after the first asymmetric epoxidation paper had come out, with lab focused on "how do we get this to work broadly?", I saw first hand the amazing effort Barry put into getting these reactions to really work --- reproducibly, in high yield and higher %ee, on (sometime very) large scale, with diverse substrates, with alternative reagents or workups.

Truly a model chemist.

Permalink to Comment

8. newnickname on June 24, 2013 12:45 PM writes...

@1 and @7: I believe that KBS already knew the commercial importance of epoxidation chemistry based on his earlier papers and patents (achiral Mo and V catalysis). I think those patents brought in a lot of money to MIT and KBS.

If you know that your discovery or invention stands to make a lot of money you will be more inclined to pin down the details and make sure that you can identify and defend your territory.

If your new method isn't likely to make you rich, you may be less inclined (and LESS WELL FUNDED) to do everything that SHOULD be done as part of "good science".

Maybe Borhani can elaborate but, over the years, the KBS lab had tried tartrates but without success. Katsuki took another stab at it and somehow got it to work well.

That said, I think that KBS is a good scientist.

Permalink to Comment

9. CMCguy on June 24, 2013 2:15 PM writes...

Sure robustness info on new chemistry is always very helpful for people who wish to apply it but what about the poor chemists who has to do all those experiments? Just as medchemists hate making series methyl, ethyl, butyl, futile unless there are available undergrads or first years how many >3rd years grad students want to slog through more dissertation folder? There is only so much one can do with a new reaction and often even less one wants to actually do in the lab once achieved a successful and reasonable demo, especially if can use in a broader project. Leave it up to literature and people who really need to expand to provide the robustness as better means to validate utility.

Permalink to Comment

10. David Borhani on June 24, 2013 3:42 PM writes...

@8: Saying that KBS is a "good scientist" is a gross understatement. His ideas and wealth of knowledge --- to which I, as just a budding chemist (~1980-1981), was fortunately exposed --- are immense. Barry seemed to know all of Houben Weyl, and much more, by heart, and he could suggest, seemingly endlessly relevant examples of things to try in any given situation, all off the tip of his tongue.

Barry also made a habit of touring the (large) lab, usually twice a day, and chatting with everyone about how it was going, what to try next, etc. Morning and afternoon. Many of those conversations, which I overheard (opposite bench, or a bench over) centered on new substrates to test, conditions to try, etc., many aimed at expanding reaction scope or increasing yield. Note the focus in many KBS articles on scale of reaction, cost of reagents, etc. Barry clearly has a strong interest in inventing truly useful chemistry.

I also think much of the driver behind the mechanistic studies of Scott Woodard and MG Finn was to understand the reaction so that it could be made better. Similarly, Bob Hanson's work to lower the catalyst loading (what a simple concept --- add ground up dried molecular sieves!).

Regarding prior tartrate work: I wasn't aware of anything significant. A Scifinder search (pre-1981 KBS publications further filtered by "tartrate") turns up nothing. In the original Katsuki/Sharpless communication (The First Practical Method for Asymmetric Epoxidation JACS 1980, 102:5976), note 2 states: "We had earlier examined the effect of chiral alcohols and chiral diols (including L(+)-diethyl tartrate) on the molybdenum- and vanadium-catalyzed TBHP epoxidation of isolated olefins as well as of allylic alcohols. Small (i.e., <1O% ee) asymmetric inductions were noted, but they were not deemed worth reporting: R. C. Michaelson and K. B. Sharpless, unpublished results, Massachusetts Institute of Technology, 1974."

I don't know how much MIT (or Stanford, or subsequently Scripps), or Barry personally, have profited from the inventions of the Sharpless lab. I'd guess a quite fair amount. I did have the (perhaps mistaken) impression, however, that Barry was already pretty well set financially before money could have been been coming in from licensing the asymmetric epoxidation reaction.

I take issue with the implication of your statement "If you know that your discovery or invention stands to make a lot of money you will be more inclined to pin down the details and make sure that you can identify and defend your territory." I think the statement cheapens Barry's contributions as a exemplary chemist. From all of Barry's actions that I saw, such a consideration was not what drove (and now, I assume, continues to drive) him to do excellent science.

Permalink to Comment

11. barry on June 24, 2013 9:24 PM writes...

surely we should extend "robustness" not just to varied substrates, but to the rest of the reaction. Not so useful if e.g. only the magic bottle of Cr++ (with the nickel contaminant) works.

Permalink to Comment

12. labrat on June 25, 2013 5:56 AM writes...

@9 "Sure robustness info on new chemistry is always very helpful for people who wish to apply it but what about the poor chemists who has to do all those experiments?" - the whole idea of this new approach is that armed with no more than a few simple reagents and a GC, you can generate all these data in a couple of days in the lab rather than waiting for the literature to take 2-3 years to discover it does/doesn't work. Surely a couple of extra days work is not too much to ask when you've probably already spent a year+ developing the reaction?
I'd certainly support methodology manuscripts being obliged to include this kind of data - when the scope is better defined others are much more likely to use (and cite!) the methods.
@1 - is this concept new? not really. Could it have a bigger impact on the field than most of the other papers published in that esteemed journal? Quite possibly!

Permalink to Comment

13. David Borhani on June 25, 2013 9:10 AM writes...

@8: I should add: KBS's first US patent is the Katsuki/Sharpless asymmetric epoxidation: (priority date: Aug. 6, 1980). The V and Mo work was apparently never patented.

Permalink to Comment

14. Quiet Observer on June 25, 2013 10:52 AM writes...

I think another methodology question which isn't asked nearly often enough is "Does that reaction mixture *really* need to be stirred for 16 hours?"

Permalink to Comment

15. a. nonymaus on June 25, 2013 12:24 PM writes...

Re: 13
Peter Seeberger has observed that most reaction times in the literature can be divided into "rapid", "over coffee", "over lunch", and "overnight" when the majority of the reactions are completed "rapidly".

Permalink to Comment

16. GONZALES on June 26, 2013 8:57 PM writes...

How is there still a schnapps called Firewater? That has always baffled me. I mean it's delicious and I use it in hard cider but damn.

Permalink to Comment

17. FeCoNi on June 27, 2013 7:55 AM writes...

This glorius work uses the data from the 2 first tables (2-3 days job) to predict the outcome of the third.
Unfortunately, the most interesting part, the 3rd table, displays only 6 entries out of which only 3 are accurately predicted.
Can we still pretend it's a "robustness screen"?

Permalink to Comment

18. ChristianPFC on July 10, 2013 12:08 AM writes...

Re 14, 15: I often wonder the same and found that some reactions don't require reflux. It might take longer, but I rather stirr at r.t. over night than reflux for two hours.

Another point is exclusion of water and air. Unless there is a compelling reason (Grignard, nBuLi), I start without precautions to exclude water and air. If I am not sure if a reaction will work better in water-free conditions, I just add water and see if it works better (instead of switching to dry solvents).

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