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

« Wyeth v. Levine: Pre-emption Goes Away | Main | More on Wyeth v. Levine and Preemption »

March 5, 2009

Your Temperamental Diva Reactions

Email This Entry

Posted by Derek

Since I was talking the other day about getting published procedures to work (or not!), I thought I should mention that most chemists have, at one time or another, had reactions of their own that not even they can get to work right every time. Most chemical reactions are reasonably robust, within limits (see here for a proposal to establish some!) But every so often, you come across one that has a narrow tolerance, sometimes for things that you can’t even put your finger on.

I’ve seen this happen particularly in low-temperature carbanion reactions. Some of these anions don’t particularly want to form in the first place, and they can be quite sensitive to concentration, the presence of different amounts of salts and counterions, variations in temperature, and so on. The rates and efficiencies of cooling and stirring can affect some of these factors, as can the age and handling of the reagents, and the rates at which they’re added into the reaction mixture. If you’ve got a system that just barely works, a lot of things can push it over the edge.

My personal experience with this first came in grad school, when I had a cyanocuprate reagent opening an epoxide. As I mentioned on the blog a few years ago, I tried that system out, after several other reagents had given not-so-great yields, and it worked really well. So I tried it again – same results! I scaled it up (at the time, “scaled it up” meant running it on about a gram), and it worked again. Problem solved! Little did I realize that the reaction would never work again. It failed the next time, and the next, and the next. I tried everything I could think of. I made everything cleaner, I made everything fresh: no product. I made everything sloppy, with no particular care, the way I’d done it in the beginning. No product. Nothing ever worked. I never did sort out what was going wrong; it was easier, in the end, to find another reaction.

Scaling up such a reaction is especially difficult – even relatively laid-back reactions have to be looked at closely when moved up to larger scales, much less a jumpy, skittish one that gets the vapours and passes out at the first sign of trouble. It’s the job of the process chemists to avoid such narrow-window chemistry whenever possible. The idea process reaction is one that provides the same yield, with the same purity profile, under a wide range of conditions: foolproof, in other words. Naturally, nothing is really foolproof (fools are too tricky), but you do what you can.

Comments (30) + TrackBacks (0) | Category: Life in the Drug Labs


1. MedChem on March 5, 2009 10:37 AM writes...

That brought back some painful memories of my own grad school experiences. I once had a 20 some plus step sequence to access an advanced intermediate. It worked fine the first few rounds but all of a sudden I'd get complete epimerization of an alpha center to an aldehyde during a Corey-Fuches reaction. We never figured out what happened, and I got called all kinds of names by my advisor for being unable to reproduce my own reactions.

Permalink to Comment

2. MedChem on March 5, 2009 10:40 AM writes...

That brought back some painful memories of my own grad school experiences. I once had a 20 some plus step sequence to access an advanced intermediate. It worked fine the first few rounds but all of a sudden I'd get complete epimerization of an alpha center to an aldehyde during a Corey-Fuches reaction. We never figured out what happened, and I got called all kinds of names by my advisor for being unable to reproduce my own reactions. I felt so guilty, like I was the only synthetic chemist this sort of thing happend to. Of course now I know better.

Permalink to Comment

3. CMC Guy on March 5, 2009 10:56 AM writes...

Derek sounds like tripped in the "magic reagent bottle" black hole. The phenomenon occurs when a lab receives or makes a catalyst that works great but the next bottle(s) or prep(s) do not follow suit. I have experienced this once or twice myself and have heard anecdotal tales from others (even of a prof who keep a bottle under lock & key for special use because it gave outstanding results compared to other lots of "same" catalyst). Does seem to often involve metal reagents with copper a particular poltergeist. Sometimes the Inorganic/Organicmetallic/Materials chemists can sort it out with serious effort (usually find a extremely low "contaminant") but it makes one both fearsome and awe inspired by the nature of chemistry.

Permalink to Comment

4. CC on March 5, 2009 10:56 AM writes...

Witchcraft, I belive, is the term for this type of reaction.

Permalink to Comment

5. Hap on March 5, 2009 11:09 AM writes...

Hudlicky describes a magic bottle of Pearlman's catalyst in his book. I assume metals are difficult to free from low levels of contamination of related metals and so there is always the chance that something else could be present that would give better (or different) results (the Heck reactions referenced in comments to the "Hot Chemistry" post, or the development of Nozaki-Hiyama-Kishi couplings).

Someone complained that the IBX-mediated oxidations of ketones to unsatd. ketones aren't reliable - I've seen one in a total synthesis but don't know if they're reliable in general.

Permalink to Comment

6. processchemist on March 5, 2009 11:37 AM writes...

Temperamental chemistry is not only catalyst related.
We work since some years on a temperamental acylation. The product comes out in 10 Kg batches. It took something like 1 year to reach a robust process at this scale. Then one year passed away without troubles. Then the supplier of a key reagent switched source (we're talking about an ISO certified company). The product is identical by COA of the supplier, identical to our analyses. New product tested ok in small scale, then *bang*. On scale: -30, -50,- 25% in yeld.
We got back to bench, and needed a fast DOE to obtain new conditions to have the old standard yelds. Lucky us that the product is not involved in API manufacturing.

Permalink to Comment

7. Norepi on March 5, 2009 12:29 PM writes...

This reminds me of my labmate's magical bottle of 3-methoxycatechol. He had some crazy reaction he was doing with it that worked quite well. He ordered a new bottle, which was, unfortunately, a liquid (3-MC is a solid that melts at ca. 45C); we never found out what made it a liquid, but the crazy reaction didn't work though the NMR, etc of the starting material was identical. He tried it a zillion times; alas, no avail...He then called them on the phone angrily demanding they send him a new bottle of the stuff, and they did (this time, a solid). He tried the crazy reaction again, and lo, it worked.

Only this time, the product (which should've been quite stable) randomly burst into flames on the bench and polymerized into boiling black goo. We never figured that one out.

This is the same guy who had an -OMe (he really liked his methoxies) that he just couldn't cleave using BBr3. He tried the reaction about 20 times, and on the 21st time, he somehow bothered the molecule enough to actually evict the offending methyl group.

Permalink to Comment

8. Harry on March 5, 2009 12:45 PM writes...

ANY protection/deprotection sequence is fraught with perils IMHO. Usually, if you want it to stay on, it falls off at a harsh look, and when you want to take it off, you need thermonuclear conditions, no matter what the literature says.

This seems particularly likely to occur on scaleup (at least in my hands).

Permalink to Comment

9. road kill still alive on March 5, 2009 12:58 PM writes...

In my experiences, it's been reactive boron reagents like Norepi mentions and metal-catalyzed reactions accounting for the majority of finicky reactions.

Derek may or may not remember the large magic BF3-etherate bottle that everyone used in grad school, that was ancient, was black, and had a hole in the Sure-Seal cap that was punctured so frequently with 12-gauge needle that a small truck could drive through it. Worked great every time; just wouldn't use it for an Org Syn prep.

One time we had a hydrogenolysis to do for someone, and it just never worked. Turns out they finally admitted they had trouble getting it to work unless they used a particular bottle of palladium catalyst. Alas, such details never get into papers or patents.

And I typically wouldn't consider any reaction being scaled up as tempermental based on a smaller scale. There are too many variables. That's why there are med chem departments and process chem departments.

Permalink to Comment

10. Sili on March 5, 2009 1:19 PM writes...

Not being a woo-woo, I won't invoke the positions of the planets (Rudolf Steiner/biodynamic chemistry?) and just call this the Jekyll and Hyde effect.

As I recall the novel an impure reagent was what made the elixir work in the first place.

Permalink to Comment

11. Chemjobber on March 5, 2009 1:56 PM writes...

I believe this is how the Nozaki-Hiyama-Kishi reaction was found. (I will never forget that one of the authors on the original Kishi papers is named William J. Christ. Wow -- imagine having that as a last name.)

I've heard tell that the Carreira zinc-catalyzed alkynylation reaction is one that is quite dependent on the source of the zinc triflate, although I'm willing to be corrected on that point.

Permalink to Comment

12. Petros on March 5, 2009 2:54 PM writes...

A double bond isomerisation was mine.

Having rua Wittig reaction under 50 to 100 diferent conditions to maximize he desired Z isomer, always 80 to 90%, we wanted the E isomer for biologicla comparison.

My first attempt at photochemical isomerization want well by HPLC gave about 75% E by HPLC when a family crisis meant I had to be away from work for about a week. Thereafter no one could get the isomerization to work!

Permalink to Comment

13. BOB on March 5, 2009 2:57 PM writes...

It had to be the Cu(I) (I'm guessing that you used some Cu(I) salt as a cuprate precursor) - I had some CuBrSMe2 that only worked the first time. After that there was some green Cu(II) and no pdt.

Temperamental chemistry is the worst.

Permalink to Comment

14. Chrispy on March 5, 2009 3:17 PM writes...

Chemists aren't the only ones with fluky reactions! All molecular biology, from PCR to ligation, has an element of randomness to how well it works...

Permalink to Comment

15. Chris C on March 5, 2009 5:44 PM writes...

I second Chrispy on this...practically everything we do in molecular cell biology is subject to the most unpredictable variation.

Permalink to Comment

16. philip on March 5, 2009 7:21 PM writes...

As far as I know, cell culturing is generally much more finicky than anything we chemists do.

For instance, I've heard that J&J had a hell of a time successfully reproducing Amgen's original process for erythropoietin alfa (Epogen/Procrit).

Permalink to Comment

17. InfMP on March 5, 2009 9:35 PM writes...

I've been lucky enough so far...but seeing a colleague struggling to reproduce something is something I will never forget.

The worst problem for me was months of grease in my spectra, couldn't publish any of what I did until I started using HPLC grade everything and working on larger scale.

For those who haven't read about this, an interesting story about Woodward's formal Quinine synthesis in the 40's was recently clarified when William's showed that the Aluminum needed to be old and oxidized to make it work. Seeman has one of the most interesting stories about this ive ever read (Angew for both articles).

Permalink to Comment

18. Robert on March 6, 2009 4:27 AM writes...

One of the guys who was a senior PhD student when I was a junior was mapping out a synthesis in the usual way: work it out on a small scale, then bring a bigger batch through the same scale, then hit the next step etc. Near the end of his PhD he needed to make more material, and ran into a brick wall. Can't remember the details of the reaction, but he found out earlier that it needed dry pyridine. Now, later, he couldn't get it to work... He was a fantastic and determined chemist (tragically died in a car crash around 15 years ago), so methodically went through everything, including moisture content of the pyridine. Nothing worked.

Finally, he cracked it. He had dried his "old" pyridine with mol sieves (as he had with his "new" pyridine as well). The difference was that in the first case the sieves were from an old bottle and contained lots of fines. The second time he used nice new sieves. Didn't affect the drying, but he found that addition of powdered sieves to the reaction worked _perfectly_....!

Permalink to Comment

19. farma on March 6, 2009 5:38 AM writes...

No.16 has it on the nail, cells can be awkward little blighters as can whole-tissues, and whole animals even more so.

I find myself getting into a nice routine and feeling a great affinity for the little creature/section of creature I'm working on any particular day.

Permalink to Comment

20. Harry on March 6, 2009 8:44 AM writes...

The comment about the seive fines reminds me of a colleague that ran a reaction- can't quite recall what it was, but he inadvertantly put selenium impregnated boiling chips intended for the Kjeldhal test in his reaction.

It ran perfectly, but the next time he tried it, he picked up the correct boilng chips, and it didn't work at all. Took lots of creative thinking to crack that one.

Permalink to Comment

21. cookingwithsolvents on March 6, 2009 9:07 AM writes...

mol sieve dust in reactions also pops up in transition metal chemistry as a source of reproducibility problems. Water, too.

I had a reaction which specifically said something like 'extremely dry conditions required' and it just didn't work. Used some wet solvent (by Na/benzophenone dianion soln in THF. . .1 drop should stay purple in 10ml of solvent) and it worked beautifully.

I had a HUGE amount of trouble trying to reproduce some preps where you condense an specific aryl aldehydes with diethyl malonate (and other similar carbanions, often in LARGE excess). I finally gave up after making tons of gunk and have always suspected that the temperature control is KEY for this reaction. One report even stirred it in a cold room overnight. All my temperature modifications were for nothing, though.

Permalink to Comment

22. milkshake on March 6, 2009 12:52 PM writes...

My former boss was developing a drug process route and he had there a benzylic alcohol oxidation to ketone in the presence of unprotected secondary amine so things were pretty touchy. He found out that he could do it with aqueous KBrO3 but sometimes the reaction was slow and crapped up. The harder he worked on it the more variable results he was getting. Then he realized most failed rections were set up in evenings and nights - the oxidation worked best on a sunny day - the thing was apparently a radical chain process. The industry reactor ended up with glass cover, and illuminated with halogen floodlights - an impressive sight in the plant.

Permalink to Comment

23. CMC Guy on March 6, 2009 3:36 PM writes...

milkshake your mention of radicals triggered memories of some Polymer Chemists I used to work with many years ago (and some of their stories seem close to your (ad)ventures). Many industrial scale polymerizations are radical chain processes which are typically well defined and controlled, except of course when they aren't. Although they usually learned sufficiently during development; blowing out valves, ejected reactor tops from run-away reactions seemed to be part of that development process and "close-calls" seemed numerous (and I wondered about their Insurance/wife's nerves). Likewise heard the adage about if you see an old operator running away best not to ask why until both far off. Sounds like there were/are many Drums and even a few RR Tanker cars of polymerized monomers that never made it to the reactors buried in old plant sites. They talked about a few lessons that took place at scale which can make it the hard kind (although more in terms of facility and equipment not people).

Most related to this post topic they came across a reaction where small scale bench work looked good but the midsized (1-2 gal) did not follow suit. As it goes the midsized reactor, a metal canister (bomb) was brand new whereas the small bombs had been around for years. Turns out just needed to add a touch of "rust" and things went as expected (Not sure if the chemistry/material ever went to production). Think I have heard noted technique used elsewhere also (this blog or milkshakes?).

Permalink to Comment

24. milkshake on March 6, 2009 4:54 PM writes...

no it was not me but I remember a story about an inorganic chemical plant in commie Czechoslovakia that had sporadically produced exceptionally beautiful-looking batches of CuSO4.5H20 - large deep blue gem-like crystals, and they were sending all these for export and their western customers demanded this particular grade from them. The problem was that the darker-blue export crystals were happening only rarely and it was very important for the company to earn the hard currency. So they were tracking all factors that led to the formation of gorgeous crystals, and eventually figured out that it happened preferentially in one out of four crystallization vats in the corner of the plant floor, and mostly on a night shifts and also in the winter when there was lots of snow they had several these batches all at once. Then it transpired that the key factor was the location of men's WC in the courtyard far from the factory floor and some employees found it convenient no to wad through the snow and relieved themselves right into the vat... Ammonia complexes of copper have deep blue color and co-crystallize with the pentahydrate.

Permalink to Comment

25. HOMO-LUMO on March 6, 2009 5:59 PM writes...

I can recall a somewhat tricky alkylation using haloacetates as electrophiles with LDA. At the end the trick was on figuring out the halogen impurities on the commercial batches from sigma. Zn made things even sweeter, what a metal.

Permalink to Comment

26. Cyan on March 6, 2009 10:51 PM writes...

When I was in undergrad biochem, I heard an anecdote about difficulty replicating a protein crystallization protocol for for X-ray crystallography. The difference turned out to be that the original researchers had carried the raw material, pig blood, in a brass bucket, and the group trying to duplicate the crystallization was using a steel one -- the zinc turned out to be key.

Permalink to Comment

27. anniechem on March 7, 2009 12:00 AM writes...

I love these stories! My MSc. supervisor once recounted to me the story of one of his colleagues during his post-doc. The colleague had an unfortunate accident, and dropped a flask on the floor that contained a large amount of an advanced intermediate. My supervisor and his colleague frantically began the 'floor extraction protocol' and actually managed to recover a large portion. The colleague continued on with his synthesis, and to his delight, the next step was a success. However, when he went back to make more material, that next step never worked again. My supervisor and his colleague referred to it as the 'Linoleum Effect'.

Permalink to Comment

28. Jose on March 7, 2009 12:04 AM writes...

I heard a story out of HC Brown's lab about a borohydride (Li? maybe something fancier?) that wasn't doing chemistry due to horrible solubility. Finally one postdoc got it to go. HC said, well, repeat it, and I'll believe it! Tried and failed for weeks, until finally the variable was isolated- egg shaped stirbar vs. simple oval, meant grinding to small enough particle size to allow (some) solubility.

Permalink to Comment

29. Maks on March 7, 2009 3:11 PM writes...

I remember a talk about peroxides and the temperature sensitivity:

"Optimal results were obtained performing the reaction at a narrow slot of temperature (−10 °C ± 2 °C). "
DOI: 10.1021/jm020584a
Not so robust......

Permalink to Comment

30. alex on March 9, 2009 5:12 AM writes...

That was my PHD work :-)
My coworker and I arrived at that temperature after multiple tries.
At lower temperatures nothing happened at at temperature above -10 or you would have reduction of the peroxide function.

I our total synthesis of Yingzhaosu A, we had to keep the temperature strictly at -11. In that work we also had an aldehyde in the molecule in addition to the endoperoxide. at -11 it worked like a charm. perfect diastereoselectivity and yield a but variable but always 70-99%. However we had to monitor hydrogen consumption very carefully.

No not robust at all, but quite efficient on even on 2 gram scale

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