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

« When Placebos Were All There Were | Main | Publish Your Work The Easy Way »

December 19, 2008

My Compound Goes Where the Wild Goose Goes

Email This Entry

Posted by Derek

A colleague and I got to talking yesterday about something that I'm sure many chemists have noticed. Have you ever chased down some reaction or compound in the literature, only to find yourself wild-goosing back to some obscure journal that no one has ever read - just because no one can be bothered to publish a modern procedure?

Here's how that typically works. You run a SciFinder search on Molecular Structure X. A list with a dozen references comes up. There's a Tet. Lett. from 2002, but what are the chances it'll have any spectral data (or anything useful at all?) Ah, there's one from Tetrahedron in 1995, that should do. So you look over the PDF, search for your compound. . .there it is, number 17. Now to the experimental. . .and you find in the first paragraph that "Compound 17 was prepared according to a published procedure", footnote thirty-eight. And the footnote is to. . .ay, it's to a Chem. Ber. paper from 1932. Ausgezeichnet!

Oh-kay. Back to that SciFinder reference list. How about that Tet. Lett. paper? Nope, on inspection, it turns out to reference the 1995 paper you just looked at. What else? There's a JOC from 1984, let's try that. Good ol' JOC, solid stuff. Well, digging up that PDF, you find that it refers to a 1980 paper from the same group from Synthesis. Hrm. So you chase that one down, there it is, compound 9, and the experimental for it is. . .footnoted to the 1932 paper. Again.

And that's how it goes. Like as not, you can go through the whole list and find that it's made of tissue paper where your compound of interest is concerned. The whole presence of the compound in the literature is, in the end, based on some obscure German university's report from the last days of the Weimar Republic. What's irritating is that while those 1932 folks clearly must have made the compound, it's not always easy to get those papers immediately. And chemistry has, in fact, changed a bit since those days. Papers from that era rely on distillation and crystallization: there are no chromatographic purifications, because there was (by our standards) no such thing as chromatrography. Spectral data? Hah! UV/Vis was cutting edge back then. You'll get a melting point, an adjective-laden description of the appearance of the crystals, and maybe even a note about how the stuff tastes. Great.

You know that the people who re-made the stuff during the last 25 years didn't steam-distill their product or fractionally crystallize it from some mixture of benzene and carbon disulfide or whatever. They ran a quick column and they took an NMR. So why can't they publish that data? The only reason I can usually see is laziness. Why bother? It's a known compound; just reference it and get that manuscript out the door. . .

Comments (27) + TrackBacks (0) | Category: The Scientific Literature


1. Alex on December 19, 2008 9:38 AM writes...

Actually Chemische Berichte as far back as 19 (as I am sure you know) is available online to subscribers or purchase.
In fact it is online all the way back to 1868 (volume 1 issue one) This particular issue has a nice paper by one A. Baeyer that concerns the reduction of Indigo:

Ueber die Reduction des Indigblaus (p 17-18)
A. Baeyer

While the experimental is a bit difficult to read it is probably better than what most papers in Tet Lett and certain papers in Angewandte provides.

Yes that the NMR data is missing is a serious handicap, but if someone states specifically that they made a compound according to the published procedure, The in most if not all cases you will get the compound. In the case of such ancient procedures the compound in question is rarely so complex that you cannot identify the compound from its NMR/HRMS/IR data without a published spectrum to compare with.

Personally I find it more annoying to open a modern Angewandte paper and find zero supporting info. At least in Tet. Lett. you don't expect to find one.

Permalink to Comment

2. MiddleO'Nowhere on December 19, 2008 10:22 AM writes...

That's why I have and plan on making my future students publish full MODERN experimentals and characterizations. My current project involved a large number of wild-goose chases and included a particularly memorable one in which the only note anywhere as to how to do the reaction was the note at the bottom of a scheme giving the reagents used for reaction condition "c". No amounts, times, or other useful information was provided. I also discovered another compound that no one has apparently ever really properly purified. It is reported to be a red-brown liquid/solid (with a low melting point), but actually it is a clear, colorless solid. On that one, I will admit that the red compound passes elemental, gives a sharp melting point, and looks very clean by NMR.

In a similar vein, I absolutely loathe the lax way in which crystallizations are reported.
"Compound X was recrystallized from ether/THF."
Was this a hot mixture, what was the ratio, did they layer it, did they do it by vapor diffusion, was significant solvent evaporation needed, etc.?
How hard is it to add the few details needed to successfully replicate their conditions.
"2g of compound X were recrystallized from 40 mL HOT THF/hexanes. The initial ratio of THF to hexanes before boiling was 1:5.

Permalink to Comment

3. Alex on December 19, 2008 10:45 AM writes...

A good thing about old litterature is that if they actually got the right product (easy to go wrong without a NMR), at least their reactions are reproducible and usually were done at gram scale. Of course sometimes they materials whose origin was important. A classic example is the Woodward/Doering formal total synthesis of quinine which relied on an earlier 1918 procedure by Rabe and Kindler(without details) for converterting their end product, quinotoxine, to quinine.

Due to doubt whether the Rabe/Kindler procedure actually worked, the Williams group tried to reproduce the Rabe procedure.

This work was discussed on the totally synthetic blog a few months ago:

In the last of this sequence they had to examine the quality of aluminum used. Only Aluminium containing enough Al(III) impurities behaved well. The pure Al provided only a trace of product.
Its an interesting read:

Permalink to Comment

4. Anonymous on December 19, 2008 11:07 AM writes...

Frankly I don't understand why people do that - most everyone has access to SciFinder and finding a paper that has a real prep doesn't take more than a few minutes in most cases. And if you had to make a compound you must have found that reference anyway, so why not put it in the paper? My only explanation is that people don't know how to manage references and/or lazy. I am especially perplexed when I see this attitude in papers from academic groups - these are the folks who care about reference count, and publishing procedures is a quick way to increase it. So whenever I get to review a paper, I make it a point to check that the references lead to actual procedures. If they don't - I speak my mind.

Permalink to Comment

5. Maks on December 19, 2008 11:12 AM writes...

I don't know how many times I translated old papers from the GDR or 1960's West German universities for colleagues from other research groups. Also some of the old Tetrahedron stuff was in French, just like one of my favorite papers in French in J. Chem. Res (Miniprint)......

Permalink to Comment

6. eugene on December 19, 2008 11:26 AM writes...

I don't like getting caught up in those chases at all. Even though it gives me a chance to brush up on my German once in a while. That's why I publish modern procedures with as much information as possible in my own papers. If it's in the SI, it's not like a modern journal is going to lose space if you're overly verbose. The SI for a JACS communication is now 50 pages average for me, so there is no excuse not to publish your newer synthetic procedure. We've never had a reviewer tell us it was unnecessary.

With regard to #2,

"Was this a hot mixture, what was the ratio, did they layer it, did they do it by vapor diffusion, was significant solvent evaporation needed, etc.?
How hard is it to add the few details needed to successfully replicate their conditions."

I won't do it. And that's because for a lot of my routine preps the crystallization really depends on the day. Those conditions are way too exact, and I may use anything from 1:1 mixture to 1:5 for the thing and sometimes it's hot and sometimes it isn't. Column chromatography also really depends on diameter and the amount of silica gel, as well as what you had for breakfast that day and how much solvent you have to play around with. Giving more than a 3:1 ratio of Et20 to hexanes is useless and is just wrong and misleading in my opinion. Your students will believe you are insane, because most of the time they'll just add THF until it all dissolves, then add ether until a cloud appears and continue adding until it sticks. They are going to think you there is something really wrong with you for demanding to know any more details and they will be correct. It's not an exact science yet. If you obtained an X-ray, then a bit more detail is necessary, but even then, I'll be satisfied if they just said they put it in the freezer, vapor diffusion, or layered solvent, or whatever. The general directions give you a starting point from which you can modify your separation or crystallization; I think it's different from the actual reaction and characterization part.

Permalink to Comment

7. MiddleO'Nowhere on December 19, 2008 11:55 AM writes...

I'm well aware as are most people of the finicky nature of crystallizations, but that's no excuse not to provide useful direction. Boiling a mixture of solvents is going to change their ratio in unknown ways, but there's a huge difference between a 50:50 or a 10:90 starting mixture. If you recrystallize something as you described (dissolve in THF, add ether till cloudy), you can easily state that in the procedure (and I've seen it done). I routinely write down the amounts of material and how much solvent was required to get it to dissolve. This isn't exact, but gives me a starting point for how much solvent I (or someone else) need to use when the reaction is performed on a different scale.
I don't ever expect the recrystallization info to be an exact recipe (so I doubt my students will think I'm insane), but the more detail the better. As you said, you can be as verbose as you like in the SI. No one will care.

I'm not sure what the Et2O/hexanes bit was referring to, although it might be because I switched solvent systems between the two examples and forgot to change the upper one to THF/hexanes to match.

Permalink to Comment

8. StrangeBrew on December 19, 2008 12:22 PM writes...

Let's face it; an Org Syn worthy procedure is what we should ALL strive for. In regards to modern spectroscopic data, if it's a solid, personally I'd almost always rather have decent crystallization conditions and a good melting point to work from than some 1H data.

Permalink to Comment

9. Hap on December 19, 2008 12:59 PM writes...

Maybe my standards are too high, but I thought the point of a publication was to submit information to the scientfic community. Part of the community's job is to see if it holds up - if it can be replicated. If you don't provide enough data and procedures to replicate your work, then it is impossible for others to substantively assess its reliability. Since most journals have online SI, there's no longer the communication excuse of not having enough room.

I understand that people are interested usually in a specific aspect of their work - the biology, nanostructures, etc.- but those aspects depend on the ways in which you do the workaday stuff of making and preparing them. The fact that people don't want to write full experimentals is irrelevant to the fact that the community needs them to use their work meaningfully.

Permalink to Comment

10. CMC Guy on December 19, 2008 1:37 PM writes...

I have repeated work from pre-WWII and even pre-WWI publications (mostly German and British) so I developed great admiration for skills and ingenuity of those who did chemistry pre-modern instruments and techniques. Assume others know one has to be careful about translations of details as I did a synthesis of an amine twice because the melting point didn't match reported value which turns out was characterized as an acid salt (undescribed genesis) that I didn't catch that till 2nd time (when all modern data confirmed duplication). I converted a sample to the salt and agreement observed.

Some literature/journals can be poor citations to follow however I have argued before that my experience with Patents, especially process ones, seem to suggest often weak on true procedures and data so it appears to be intentional obscuration that is counter to principle that "those skill in the art" could perform.

Permalink to Comment

11. researchfella on December 19, 2008 2:45 PM writes...

I've also had a lit. search for a compound synthesis come to a complete dead end, when you get to an article that says it was ordered from company X, and then you determine that it is discontinued and/or unavailable from any commercial source, and the synthesis method is also not reported anywhere. In this case I have presumed that the compound was produced as a by-product from some manufacturing process that is no longer in use.

Permalink to Comment

12. milkshake on December 19, 2008 4:15 PM writes...

Old pre-war Berichte, Liebigs and J Chem Soc are generally not so bad, the only complain is that sometimes the experimentals are written in shorthand (aka not bothering to explain the routine stuff obvious to anyone skilled in art), and of course you cannot always rely they got the correct isomer - but I have also reproduced some very solid procedures from the beginning of 20th century (one of them from a dude named Emil Fisher). Old French is weirder; I had a friend helping me to translate the original paper on camphor sulfonation to camphorsulfonic acid and it was from circa 1865 by some frog dude with a great literary flair - he wrote a whole charmed novel on how the crystals of CSA slowly formed in the viscous reaction mix over the period of many weeks...

I also remember an old obscure paper on synthesis of hydrastine alkaloids and they had many pages chewing around a key, difficult cyclization that I was interested in, and finally I when red the whole godawful thing i found an experimental procedure that finished with a sentence "The yields of the desired product in this transformation are variable, typically fairly low but sometimes none"

Permalink to Comment

13. eugene on December 19, 2008 4:43 PM writes...

"I'm not sure what the Et2O/hexanes bit was referring to, although it might be because I switched solvent systems between the two examples and forgot to change the upper one to THF/hexanes to match."

Ah sorry, I was referring to column chromatography and I thought I'd put out those two solvents as they are the standard for a lot of low polarity compounds. Usually everyone uses some type of gradient, but it's never mentioned in the procedure.

Permalink to Comment

14. Chemdaddy on December 20, 2008 11:34 AM writes...

Well, as soon as that "key" experimental reference started to have that "this is going to take 40 minutes to track down through decades of scrambled internet references and footnotes, 2 days to receive by mail/fax if I ever even find the right citing, access to an international translator who wants to waste an hour with me and my archaic printout, and $50 bucks or more to order" feeling I'd generally throw in the towel and go into lab and set up a few best guess reactions hoping to simply reinvent that forgotten wheel.

Permalink to Comment

15. Steve on December 20, 2008 11:53 AM writes...

Of course, one of the advantages of working in Cambridge is that it is easy to walk/bike/drive over to the MIT library and look at that article, no matter how obscure or old it might be. They have hard copies of virtually any journal one might need and copying it at $0.10 page it doesn't break the bank.

And at least in our lab, we have people who can translate German/Japanese/French/Chinese/Spanish.

Permalink to Comment

16. charles on December 20, 2008 4:22 PM writes...

To be honest, when I was at UIUC, I enjoyed my goose chase through the hundred-year-old syntheses in falling-apart German and French journals in their chemistry library. On the other hand, I didn't need a lot of help to translate them either. As a bibliophile, it is a really unique feeling to hold such an old publication in your hands and carefully page through to find exactly what you're looking for.

Permalink to Comment

17. Norepi on December 21, 2008 11:12 AM writes...

Oh, gosh. This happened to me two weeks ago. I was looking for a procedure, and all references funneled back to the same Annalen der Chemie paper - 40 minutes later, I'm in the dusty stacks of the library attempting to translate the overly wordy text (I know just enough German to look really stupid), whilst also trying to interpret what the distorted hexagons and pentagons mean (from the pre-Chemdraw era). And lo - although the compound, Obscure Solvent From Which It Was Crystallized, and melting point were reported, no experimental was to be found, and I was, as I have been many times before, forced to make something up.

#3-In addition to not reporting crystallization conditions, it drives me nuts when people don't report boiling points. I hope they can hear me cursing when I end up losing the stuff on the rotovap. #$*&%!

Permalink to Comment

18. Petros on December 22, 2008 7:49 AM writes...

Don't forget early abstracts in CA often have extensive experimental detail (up to ca 1930), or how about Beilstein?, albeit it's in German and the hard copy organisation is eccentric

Permalink to Comment

19. ab on December 22, 2008 3:35 PM writes...

Speaking of older literature, why did the Royal Society change the names of their journals so frequently?

Permalink to Comment

20. S Silverstein on December 22, 2008 6:54 PM writes...

What do you do if you are denied access to SciFinder because some non scientist IT ignorazzi are trying to make their budgetary goals?

See here. I did not make this scenario up.

Permalink to Comment

21. SF on December 22, 2008 8:12 PM writes...

S. Silverstein (#20): "What do you do if you are denied access to SciFinder because some non scientist IT ignorazzi are trying to make their budgetary goals?"

For similar reasons, I was once given a limited-access SciFinder seat, for which we were charged on a per-search basis. They didn't tell me at the time, but this was only cost effective if I made less than something like ten searches a month. After the first month, they quickly found the resources to upgrade my account.

Permalink to Comment

22. SteveM on December 22, 2008 8:14 PM writes...

I got out of the lab 20 years ago (sigh). But what we loved about the German references was that the work was done by Germans. Nothing like Teutonic attention to detail. Everything worked like it was supposed to.

Now replicating a synthesis from a Japanese patent? Like interpreting the scattered entrails of a slaughtered chicken.

Permalink to Comment

23. BCP on December 22, 2008 10:52 PM writes...

I'm with MiddleO'Nowhere here. Closing in on 2 decades in pharma has led me to the point where I'm more and more a believer of the old phrase "if you don't record it, it never happened". I've got tired of vague write ups that can't impart what was original done to another experimenter. Perhaps there was an ingenious, original discovery in there, but if it wasn't properly recorded, it's lost.

There, that's better.

Permalink to Comment

24. jnf on December 24, 2008 11:14 PM writes...

I remember in grad school we tried to reproduce a famous Iowa Satate profs. prep of a umplong reagent which was dry labed completely (Tel Lett, no exp) (BTW this was attempted by 4 grad students) and found another procedure from SRI that listed the MP 120 -127. We went back and found a Sir Robert Robinson procedure that, when modified, gave the desired product and published a full JOC paper.

Permalink to Comment

25. lithiumdiisopropylamide on January 3, 2009 10:39 PM writes...

yes, this is frustrating stuff...i've been on the goose chase i can't count the number...i make it a point to site all the references to some compound we need to make...then write-up a real detailed prep, so that future gens. will have a copy of the article in their file cabs.

Permalink to Comment

26. The Inorganic Gardener on January 4, 2009 12:29 PM writes...

I've spent much time in the last few years doing similar paperchases! I'm not a chemist but a microbiologist working on sulfur metabolism which involves polythionates, which are, as you might be aware, not something you can buy "off the shelf" from Sigma-Aldrich (well, you can get Na or K tetrathionate but none of the others). Most every microbiologist who's worked with them in the last 30 years publishes papers with things like "sodium heptathionate was a generous gift of Prof. X" then you look up Prof. X's work on the subject and find he synthesised it "according to Y (1895)" which then leads to delving into Berichte, as you say.

I was very fortunate as an undergraduate (my first degree was Chemistry) to study in a very old-fashioned (now sadly disbanded) chemistry department where you were taught useful (though we didn't know it at the time) modules like "Translation of German Chemical Texts" and had hours and hours of practical classes on recrystallisation, complete with "how to select the solvent" stuff. Obviously we did NMR and all the modern stuff too but not until we'd been drilled in the chemistry that belonged to The Old School - we even had an assessment where we had to identify (and name uses of) 100 pieces of laboratory glassware!

So now, after many years of digging in the literature and translating polythionate papers (I think potassium trithionate is the oldest of the lot - 1833 in French, if memory serves) I'm about to submit a manuscript detailing their synthesis in full, with detailed instructions for the "lay man" (i.e. for those microbiologists who weren't chemists in a previous lifetime as I was). Alas, my manuscript doesn't have any of the romance of the original papers - one of them had a method which involved passing a solution through "the finest Swedish filtration paper" and "laying the solution down in a corked bottle for some 6 weeks".


Permalink to Comment

27. Anonymous on January 5, 2009 4:08 AM writes...

Beilstein will have saved you from reading all these papers, it would have clearly shown you which paper had reported the preparation and after one click where you could find which spectra.

Honestly, part of your problem is that you think in literature, not in terms of chemistry!

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

How Not to Do It: NMR Magnets
Allergan Escapes Valeant
Vytorin Actually Works
Fatalities at DuPont
The New York TImes on Drug Discovery
How Are Things at Princeton?
Phage-Derived Catalysts
Our Most Snorted-At Papers This Month. . .