Getting Better Crystals

George DeTitta detitta at
Thu Aug 3 13:52:20 EST 1995


At the recent ACA conference in Montreal Joe Luft and I described a simple
method to slow down vapor diffusion equilibration that should lead to better
macromolecular crystals.  The method depends on our recent finding that the
rate of equilibration between a hanging drop and its reservoir depends on
the distance between the droplet and the surface of the reservoir solution.

In general the further away a drop is from the reservoir, the more slowly
it equilibrates with the reservoir.  Therefore to slow things down you need to
increase the distance between the drop and reservoir, and to do so you probabably
need to get away from Linbro plates.  In a Linbro, assuming you need at least 
0.25 mL of reservoir solution, the maximum distance between drop and reservoir
surface is about 15 mm.  We use standard lab glass test tubes (127 mm length,
17.5 mL volume, 16 mm OD, 13.5 mm ID) as reservoirs.  To give you an idea of
the slowdown, a 24 microL drop of 1.0M NaCl equilibrating with a reservoir
surface 7.6 mm away reaches 1.92 M NaCl in 48 hours.  The same drop reaches
only 1.67M NaCl after 121 hours when the reservoir is 78 mm away.  

Note that the chemistry of the experiment doesn't vary for these two case.
The same drop (volume, concentration) and same reservoir (concentration)
are used.  Only the kinetics has changed.  And in our experience, slower growth
has usually led to better crystals.

We use `coverslips' made from clear plastic label tape made by Manco (available
at Office Max and Ace Hardware) that has been stuck to a flat common washer that 
just slips over the OD of the test tube.  We put the drop of the adhesive side
(really, it's ok, and even less toxic than Prosil) and then flip the washer
and drop and seal it ove rthe test tube mouth.  Makes a good seal (the tape)
with the test tube as long as the lip of the tube is smooth.  Of course if you're into pain you can always use a glass coverslip.

More information about the Xtal-log mailing list