Org Prep Daily

July 6, 2008

All experiments that are fit to print

Filed under: lab destruction — milkshake @ 9:32 pm

During thesis defense in Prague, one professor asked me: “The experimental procedure on page 64 is describing an explosion - please would you care to elaborate more on this?”  I found myself explaining that the material, racemic glycidol, was made only once for the project and there was, uh, a mishap as described – and I got enough material to carry forward so I had no urgency to re-do the preparation under more controlled conditions… What I did not tell them then was that the volcano which erupted three years before was set up with blessings of a man working for the said professor.

I needed glycidol and I couldn’t afford to buy it from Aldrich - so I made a whole bunch of chloropropane diol by hydrolysis of epichlorohydrine and I was going to treat it with methoxide. As I couldn’t find a reliable lit procedure for my particular substrate and being lab-naive, I asked a faculty dude about a general chlorohydrine ring-closing conditions. He suggested that I should use chloroform as a solvent for the reaction, and run it at 0C. “But isn’t chloroform supposed to react with methoxide?” I asked him.  “Sure it will - that’s the beauty of the system, the solvent reaction will mop-up any unreacted methoxide so it won’t ruin your epoxide. We in sugar chemistry are using this trick all the time”.

Alrighty, then: I dissolved the chlorohydrine thing - all 136 grams of it -  in chloroform 0.5L and slowly added 1.5 mol of methoxide solution. The addition was not exothermic; in fact the reaction seemed calm like a lamb.  About 3 hours later I removed the cooling bath and as I was walking away suddenly a thermometer flew by. There was a whooshing sound while the geyser commenced - from the flask neck where the addition funnel was just moments before… I slammed the hood sash down and bolted for the door.  Later I found that there was enough left in the flask to proceed with the workup (alongside with the cleanup). The chloroform/methoxide trick did not work as advertised though: there was lot of  methoxypropane diol in my glycidol and I had to do a fractional vacuum distillation on Vigreaux to obtain the pure stuff, in 8.5% yield. 

Few years later I visited the lab and looked into that hood and sure enough, the dried up white cake of salts - round and pretty like a giant camembert - was still there, hanging from the ceiling…


  1. That things happen and you had the balls to report it in your thesis. Bravo!. There are people that don’t even tell the colleague in the next hood if he/she hadn’t notice it. In my opinion this kind of reports can help to prevent future incidents.

    That sort of “tricks” may work for very small scale but on larger, you have to be very very careful.

    Comment by Vasili — July 7, 2008 @ 2:35 am

  2. Chloroform plus strong base give dichlorocarbene. This is often a bad thing. Methylene chloride and azide give CH2(N3)2. Rotovaps are then damaged. Don’t use a 316SS drip pan with hydrazine.

    The world is wildly non-linear. Don’t trust local approximations to be global solutions. Sensitive folk are invariably revealed to be asses after compassionate legislation passes. Talk with your engineers.

    Comment by Uncle Al — July 7, 2008 @ 10:51 am

  3. I’m surprised someone on your committee actually read the experimental section. I’m convinced no one on my committee read more than the first 20 pages of mine. While it was nice to avoid the “fine-toothed comb” treatment, I would have appreciated a little effort on their part. That’s not to say they didn’t tear me apart in my defense. I was grilled like any PhD candidate, it just didn’t have much to do with what was written in my dissertation.

    Of course, I was the first grad student to graduate from my advisor’s group. I’m pretty sure that behind the scenes, her work (i.e. what I had been doing) was being scrutinized significantly. There was a strong desire in our department for her to be successful. She was eventually tenured and has been quite successful.

    Comment by Chemgeek — July 7, 2008 @ 10:59 am

  4. Sorry to ask this irrelevant thing to the topic under discussion. How do academicians get the industrial collaborations? I see a lot of people acknowledging the industrial folks for the support of their research. Any proven tips to attract that kind of funding especially if you are part of poor research group as we were discussing some time ago?

    Comment by Me too — July 7, 2008 @ 6:58 pm

  5. well this is a tough one – and I am definitely not the right person to answer it, as I never had any dealing of this kind.

    To me it appears that industry likes to fund synthetic methodology groups that are already famous, especally if the professor is stil young but shining bright. I suppose MacMillan, for example, has no difficulty attracting industry funding and consultancy offers. (And his people can get an interview for high-profile industry jobs when he picks up the phone and makes some calls, etc.)

    From what I heard the industry funds have often less strings attached than a typical grant but the company that gives out the money wants something – it is the prestige of being associated with a high-profile group, getting graduates from there and supporting a methodology research that might be useful etc – but the first two factors (snobbery of industry folks, and their gaining personal connections to a famous man) seem more decisive.

    It is a lot less common to get industry money for a medchem research project – and if you get it, the company might want to have exclusivity on your inventions or at least have a priority in-licensing your patents (the right of first night). This can be extremely tricky – and I really don’t know much about it. The Chemistry Department in University of Arizona got screwed royaly in this kind of deal, about two decades ago, from what I heard they received something like 500k in VC money and they ended up paying back 20 millions in damages. (Our start-up company got sued too because one U of A professor was added onto some of our papers and patents – and we settled for 300k just to get them off our back; this protracted lawsuit was costing us two millions a year, a major part of our operating budget! It saved our little company from being gobbled up by Pfizer though; Pfizer got scared of the IP lawsuit and pulled out at the last minute from the acquisition. The company stil exists, now as a part of Sanofi and there were no layoffs – had Pfizer taken over we would get axed because they wanted our IP postion, not the company).

    Comment by milkshake — July 7, 2008 @ 7:49 pm

  6. Me too: Alternatively, you could do materials research. :)

    Comment by Ψ*Ψ — July 7, 2008 @ 10:14 pm

  7. Someone should make a movie about your graduate days :)

    Comment by dilutedmagnetics — July 8, 2008 @ 10:16 am

  8. milkshake: Thanks for sharing your thoughts.
    Ψ*Ψ: Thanks for the tip.
    Dilutedmagnetics: Title of it could be ” Lunatics of Science”):

    Comment by Me too — July 8, 2008 @ 5:54 pm

  9. Sorry for skipping the main topic.

    I recall we had a solvent mixture suitable for extracting very polar/water soluble compounds from H2O, but I can’t remember it.

    Do you know any magical solvent mixture or personal trick to get this stuff out of water?.

    Comment by Vasili — July 9, 2008 @ 3:15 am

  10. Vasili: I personally was told to use 7:3 CHCl3/iPrOH and it worked very decently extracting a small MW 1,2-diol (about 50g) from over 4L of H2O and NaOAc. Good luck!

    Comment by HPCC — July 9, 2008 @ 4:18 am

  11. I’ve heard various ratios of the same solvents; 2:1 or 3:1 might work too.

    Comment by Jose — July 9, 2008 @ 11:24 am

  12. This is off topic but I’ll throw it out there to see if I can get any suggestions, which would be greatly appreciated.

    Any way to visualize tetramethoxysilane or its related decompostion products on TLC plate? Obviously UV or phosphomolybdic acid doesn’t work.

    Comment by pc — July 10, 2008 @ 2:16 pm

  13. I never worked with siloxanes but my understanding is that Si(OMe)4 hydrolyses quite rapidly – its gonna be difficult, you know, to detect silica on silica.

    Comment by milkshake — July 10, 2008 @ 2:59 pm

  14. Yes indeed, and that’s why I put it out here.

    BTW, you always respond in a timely manner, very courtesy to your guests. Just want to let you know that I appreciate it very much. Guess others feel the same way too.

    Comment by pc — July 10, 2008 @ 3:35 pm

  15. HPCC and Jose : thanx for the answers.

    I’ve tried DCM/MeOH and DCM/IPA, both 25% and obtained better results with IPA/DCM. I needed 800 mL of volumen to to extract 200 mL of aq. solution but stripped almost all organics from there.

    Comment by Vasili — July 11, 2008 @ 3:15 am

  16. pc: you could maybe use deactivated SiO2 or alumina TLC plates .

    Regarding development I don’t know if you could use KMnO4 solution.

    Comment by Vasili — July 11, 2008 @ 3:20 am

  17. Time for another question, this time regarding Bartoli Indole synthesis! ;)

    Has anyone on this forum ever had success (i.e. a traceable tlc and crude 1H NMR) using 2-alkoxy nitroarenes as substrates? There is a Synth. Comm. paper that mentions the only good substrate is 2-benzhydryloxy nitrobenzene, that the corresponding OBn gave 13% yield. Well, I repeated it, out of skepticism, and indeed, my product was about 15% in the crude NMR. But then, a Bioorg. Med. Chem. Lett. mentioned they made the 7-OMe indole from 2-OMe nitrobenzene using Bartoli’s method, in 65% yield, so all excited I though, let’s go!

    Well, it ain’t much better… Any thoughts??? It seems the only successful examples ever are with halides or alkane substituents on the ring…


    A frustrated indole synthesizer who NEEDS to make bucketloads of 7-alkoxyindole

    Comment by HPCC — July 11, 2008 @ 6:18 am

  18. #17: You probably already know all about the problems associated with making 7-alkoxy indoles. I have a Synth Comm paper here for preparing 7-hydroxy indole (SC 2003 33 507-514) but that might already be the paper you referenced. They made it starting from 3-hydroxy-2-nitrotoluene via a reductive cyclization. There is an OPRD paper that discusses some other methods (OPRD 2007 11 73-80), but I haven’t read through it so you’ll have to figure out if it’s useful. Supposedly the Fischer indole prep with the o-tosyloxy hydrazine works, TETR 1998 54 45-64 (whereas it is well known that the o-methoxy fails). There’s a cute Fischer method using an ethylene linker between the oxygen and nitrogen of the hydrazine, TETR 1997 53 8853-8870, but it might require an electron poor aryl for the removal of the linker (they had a nitro group).

    Comment by Peter — July 11, 2008 @ 9:15 am

  19. I was doing Bartoli only once, with a strange result (it is here under “Bartoli does not like you” about 1 year back, under category “mechanisms”)

    7-methoxyindol is commercial. I think these indoles are usually made by classical Fisher, by cyclizing hydrazones of pyruvic acid. Then you have to decarboxylate the indole-2-carboxylic acid.

    Also if you could access the alkoxyaniline and put bromo or iodo ortho to the NH2 you could do Sonogashira and then cyclise the acetylene with a copper(I) and a base.

    Frankly I dont know what to suggest to solve your problem; I think Peter’s ideas are better.

    Comment by milkshake — July 11, 2008 @ 4:11 pm

  20. Yeah a colleague working on the same project as me said than 7-benzyloxy also miserably fails in Fisher indole synthesis – he tried. I will actually check the tosyloxy paper, because I have not been aware of this one. I think I pretty much tried everything so far that has been published (including everything in the OPRD 2007) and f*ked in my hands.

    Again, thanks for noting the commercial availability of 7-methoxyindole – as well as 7-benzyloxy – but I must remind the avid reader that I am the same guy who needed fuming H2SO4 in a previous thread, and who lives on an island where chemicals are paddled in along with food supplies for Survivor: Singapore. :) :) :)

    I may have to resort to Sonogashira-Larock methods…

    Thanks everyone for the input. Cheers, Peter!

    Comment by HPCC — July 11, 2008 @ 11:27 pm

  21. 7-benzyloxyindole is commercial, and reasonably priced. 7-hydroxyindole is commercial also, though somewhat expensive. Looks like you are trying to re-invent a stuff that’s readily available commercially. I am sure there must be some mainland Chinese company that would sell it to you for a fraction of the price of Aldrich or Alfa.

    Comment by milkshake — July 12, 2008 @ 1:57 am

  22. I agree with Milshake, and I always wondered (maybe HPCC has the answer) why our outsourcing partners in Asia have so much trouble in receiving the starting materials (availability and delivery times) in Asia when almost all commercially available products are made there. E.G Aldrich has its own plants in China

    Comment by Vasili — July 12, 2008 @ 4:17 am

  23. HPCC: How come Dr.David Chen’s ( exetension of Dr. KCN’s Group) functioning there so well?

    Comment by Anonymous — July 12, 2008 @ 1:08 pm

  24. 7-benzyloxyindole is one weird compound. Hydrog it off in EtOAc, concentrate the clear solution, and you get dark purple-red gradeux within minutes.

    Comment by Jose — July 14, 2008 @ 11:21 am

  25. Vasili:

    One of my students was preparing derivatized hydroxyethyl(ethylenediamines), which, despite other intended uses, ended up basically acting as surfactants. The way he got them out of the aqueous phase was to make it very basic, thus denying hydrogen bonding to the compound. Of course, this only works if your material is really stable under such conditions.

    As an aside, does anyone know where to get figures for global bulk chemical production? I’m particularly interested in what are the top five or ten chemicals not produced directly from cracking petroleum. I know acetic acid is up there, as is vinyl chloride, but I’d like to find definite numbers.

    Comment by LibArtsProf — July 16, 2008 @ 10:33 am

  26. LibArtsProf:

    My compound is very aqueous soluble in every pH and I have an aldehyde there, but thanks.
    At pH 7-8 the molecule is neutral and can be extracted wih IPA/CH2Cl2.

    Comment by Vasili — July 17, 2008 @ 2:47 am

  27. Vasili – one solvent (besides the iPrOH/CH2Cl2 or MeOH/CH2Cl2 combinations) that has worked well for me in such a situation is n-butanol. Its bp of 117ºC means it’s not as easy to rotovap off as the CH2Cl2 combos, though.

    Comment by Arno — July 19, 2008 @ 1:33 pm

  28. Arno:

    I used tert-butanol several times as a solvent and although it has a lower bp than n-butanol (82.4 °C) it was difficult to strip out for me (and my pump)or I just had no patience at that time.
    Surprisingly, I just injected the crude solution in t-butanol on top of the flash chrom system and it worked quite well, (just luck probably).

    I maybe should have azeotroped a H2O/t-butanol mixture…

    Comment by vasili — July 20, 2008 @ 2:45 am

  29. Anonymous:
    I have no good answer, except that people typically apply to post-doc for Nicolaou first, then get offered either 2 years in Singapore, or 1+1 (Scrips/Singapore). But people who would apply to work for Nicolaou in the first place are people willing to solve hardcore synthetic problems, so I guess that the limited availability of certain chemicals is not something that’s going to deter them from reaching their goals! Oh, and Chen’s laboratory also being a Nicolaou sub-station… pressure is immense!

    Comment by HPCC — July 24, 2008 @ 1:26 am

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

The Shocking Blue Green Theme Blog at


Get every new post delivered to your Inbox.

Join 137 other followers

%d bloggers like this: