Thursday, 17 February 2011

The Story of the Neverending Column

I draw ever closer to something that resembles my target molecule. Today’s attempt at chemistry comes in the form of adding a TBDPS group to a primary alcohol I spent yesterday making. The reaction in question was put on last night and worked up this morning. TBDPSCl, imidazole, and my substrate at room temperature in DMF was stirred overnight and quenched and worked up in the normal way this morning. 

 The literature reference for this compound told me it required purification by column chromatography (a phrase that always makes my blood run cold) eluting with 2:1 petrol and ethyl acetate. This compound has taken me a few days to prepare and I learnt very quickly as a PhD student not to trust the literature on stuff like this so I thought I would play it safe and use 4:1 instead, just to increase the separation of the three spots. “This should be a nice easy column, I’ll be done in an hour or so” I thought to myself and set about taking fractions. The first two spots came off within the first ten fractions along with the first signs of the major product which I can only assume to be the one I want.  NMR of the first two fractions showed me that one was assorted crap and one was what I assume to be unreacted TBDPSCl. I then increased the amount of ethyl acetate in my solvent mixture and continued collecting fraction after fraction after fraction...

Eighty fractions later following a particularly interesting regimen of “take fraction, TLC, stain, cry, repeat” my patience was starting to wane somewhat as this final product showed absolutely no signs of slowing down. As I’m sure I have mentioned previously, I will do almost anything to avoid running a column but this one really takes the cake. I assume what is coming off is pure (I’m removing the solvent as I write this) but cannot describe to the non-chemists amongst those of you who read this how boring it is collecting fractions that come out the bottom of a column. I don’t make anything which is a nice colour, I don’t make anything fluorescent and I don’t make anything which smells nice so I just have to sit there and hope my column is working OK before I check at the end.  This purgatory can and does go on for extended periods of time and takes up a solid proportion of my time in the lab. Anyway, if this stuff is not completely pure and in at least 95% yield I’m going to be dramatically unimpressed (I joke of course, anything remotely positive will do for me at the moment).

Anyway, I have training tonight so I’m going to spend the next two hours running around in the cold to hopefully let off some steam so I can get back “in the game” so to speak in the morning. Tonight I can only pray to the chromatography gods in the hope that tomorrow will provide me with a slightly more fruitful way to spend my time, like doing some chemistry that might actually be publishable at some point in the future.


  1. Haa haa the life of an organic chemist, so many man hours! I spent the day attending a seminar, doing some washing up, going to the powder xray diffraction lab of the earth sciences department to pick up data then I went to the pub. Ahhh the way of a physical chemist, think for a week, experiment for a day model/work out what the hell you did for a week, done! During this time copious coffee is consumed,

  2. I guess it depends on your substrate, but I always used methylene chloride for silyl protections- DMF is a terrible solvent for lots of reasons. But I remember stupid columns like this taking forever at first, too- eventually for silyl protections I'd get to the point where I'd either just run the reaction and carry on without further purification or run the fastest column possible- and they should (hopefully) get a lot faster for you as time goes on and you figure out better ways to run them.

  3. I was literally following a procedure straight from the literature, that's the only reason I used DMF! I can't stand the stuff either and I will have a go with DCM next time (if I have to do this reaction again).

    Ended up finishing off the column this morning running it with 100% EtOAc. 75% yield could be better but it definitely beats the 35% quoted!

  4. Silylation: DBU in MeCN works really well if you don't have any base-sensitive functionality. Otherwise NEt3+DMAP in MeCN is the way to go.

  5. In the future when you do this you'll know that when you've eluted the first two spots and you're just waiting for the third to come off you can simply crank up the % of ethyl acetate.

    For silyl protections DCM tends to be the better choice in my experience.

    Nothing says "welcome to grad school" like a large scale column.

  6. Ah, handmade columns, TLC... those were the days :) Mass-trigged prep HPLC (reverse phase) or UV-trigged flash using disposable, prepacked columns (straight phase) is how we do it in the industry (referred to as "the real world" by some) today. Machines make life easier, and stupider.

    Found your blog just now via Reddit. I like what I see! Added you to my blogroll, hope you don't mind. Check out my page if you have the time. We're in the same business - synthesis

    DrFreddy -

  7. Some of the happiest TBS protections I've ever done were in DMF. The trick is to run the reactions really concentrated. After the reaction, pour onto ice-cold HCl and filter if solid or extract with ether otherwise, chromatography wasn't required. Never used TBDPS though.

  8. Ah you've expressed my feelings about running columns! The very last sentence, and the “take fraction, TLC, stain, cry, repeat”.. i couldn't agree more!
    I've been taking a look at your blog every now and then. But this entry really got me. I'll be checking more often!
    I pray to the chemistry god by the way, no specific ones like the chromatography gods. No need to narrow down the chances. But the chemistry god totally hates me anyway!

  9. This blog is really nice and helpful for readers to understanding better about the Preparative HPLC Columns.