5/27/2007

surprise!

Sorry, Excimer. I was going to do a post on something else, but then I got to the lab today...and found these!

Please pray to the crystal gods for Ψ*Ψ.
Surprise crystals are maybe the best kind of surprise ever. The only thing that really comes close is surprise money...if it's more than $20.[1] Kyle did a post a while back on growing crystals, but he failed to mention the only method that has seemed to work for me thus far. See, whenever I make a conscious effort to get something x-rayable, I think my compounds realize it. And then they laugh at me and aggregate or powder out or gellate. It's a little depressing. Almost all of my intermediates crystallize if I so much as look at them funny. The things I want x-ray structures from? No cooperation whatsoever...unless I am not seriously trying.
I advocate the accidental method of crystal growing. The last thing that worked was left after a mess of solvent evaporated--it was solvent I had used to rinse the glassware I was working with. Still can't remember exactly what was in there. Then there are the pretty red things in the picture (which, since I am excited about them now, will probably give lousy data). Those came out of the first solvent I tried. I didn't think it would work. I like being wrong sometimes.
[1] "Surprise money" refers to cash you forgot about and later found in the couch, your freshly washed pants, etc.

12 comments:

孙尉翔 said...

Try XRD a bit...

Anonymous said...

I've been trying to grow crystals half-heartedly (and unsuccessfully) for the last few months. Maybe I should try the not-try approach.

Anonymous said...

I'd been trying to crystallize a compound for about six months without any success - I just kept getting powder. One day I made up a concentrated sample so that I could get a decent 13C NMR spectrum. I knew the compound was soluble in chloroform, so I was surprised when not all of my sample went into CDCl3... but I carried on and added a few drops of CD3CN and it all dissolved. Any samples for automated overnight runs had to be set up on the sample changer before the doors were locked at 6 pm, so I raced downstairs and made it just in time. I got back to the lab and noticed that the very distinctive 'CDCl3' bottle was actually one I'd recycled and contained a TLC solvent mixture of 1:1 EtOAc/hexanes... unfortunately, I couldn't go and rescue my sample from the now locked NMR facility... The next morning, I confessed my stupidity to my colleagues before venturing downstairs to retrieve my sample, whereupon I noticed crystals the size of bricks in the bottom of the NMR tube!!! So, there you go, it needed a ~10:10:1 mixture of EtOAc/hexanes/CD3CN to crystallize - who knew?! After repeating the process in the lab and getting crystals, it was found that the NMR machine's strong magnetic field wasn't essential for crystallization...! - neither was spinning at 20 Hz, or deuterated solvent!

Excimer said...

Andrew, XRD (I'm assuming you mean powder XRD) won't help you if you have a polycrystalline structure. In order to get the unambigous single-crystal structure, you need a single crystal.

Liberal Arts Chemist said...

The ignored NMR tube is quite possibly the best method for getting crystals. The little plastic cap is a weak enough seal to slowly bleed out solvent. The two issues are getting the crystals out intact and confessing your method.

Anonymous said...

Yeah - I think I grew most of my crystals (serendipitously or otherwise!) in NMR tubes, much more success with that method than vial in vial...

Anonymous said...

The stuff likes to crystallize in the magnet because it is few degrees colder there...

It has been my experience things suddenly crystallize when you don't want them to: NMR tubes, sep funnels, in the needle and the syringe.

Anonymous said...

The last resort is to homogeneously dimerize the crap with formaldehyde or 1,4-bis(bromomethyl)benzene or something. Adding a C2-axis almost always gives phat crystals, even with the most uncooperative proteins.

Ψ*Ψ said...

I've had a little luck with NMR tubes, but never for the things I most desperately wanted to crystallize.
Proteins? Meh. My current group does crystal engineering. If we don't get nice sparkly things, probably we fucked up somewhere.

Georg-Martin Krapper said...

In grad student days, I managed to crystallise some stuff for one of the other group members totally by accident. She had it sitting in the freezer in a round bottomed flask on a cork ring (do you still use these?). It must have shifted because when I opened the door it popped out. It was a little early in the morning for dynamic fielding and so the flask followed Newton's directives and transferred it's kinetic energy to floor. We used to do some C13 work and I figured that I should rescue what could be rescued before it's owner got in. The glassware had broken into a few large pieces and the viscous product was sticking well. I gathered the bits into a dish and left it on her bench. I spent a nervous (she was the year ahead of me) 45 minutes waiting for her to get in. The material was indeed C13 enriched but it's owner was delighted because this long forgotten sample had started to crystallise. We were doing C13 NMR and so we never got to find out whether the crystals were good enough for crystallography. As an aside, I too have an intense dislike of centipedes amd would love to enroll one in a 'physiological effects of liquid nitrogen' experiment.

Anonymous said...

What's wrong with the "pretty pictures" post? I can't comment.

How fluffy is the most fluffiest compound you've seen? 15 g of my "certain heptacyclic thiphene" hardly make it in a 125 mL Erlenmeyer.

mercury said...

as an inorganic synthesis guy (new main group structure and bonding) the importance of an x-ray structure was paramount. we had lots of techniques, like vapour diffusion, slow solvent evaporation, liquid-liquid diffusion...but the one that just seemed to work the best was "just stick it in the freezer and check on it a year later".

NMR tube method also a fave.