CRYOCRYSTALLOGRAPHY: SMOOTHING THE PATH TO SUCCESS. Håkon Hope, Department of Chemistry, University of California, Davis, Davis, CA 95616, USA.

Over the past twenty years biological cryocrystallography has changed from a rarely used, esoteric specialty to a near mainstream technique. There are important forces driving this development: (1) As projects become more and more ambitious, the need to protect precious crystals from the ravages of radiation damage becomes more important, (2) cryo-techniques have become simpler, (3) cooling apparatus has improved, and (4) there is better understanding of the processes involved in the prevention of ice formation. However, we are still some distance away from total control.

Although the design principles of gas-stream cooling equipment are well understood, there is a discrepancy between what has been established and what is being practiced. It is possible by simple means to keep a crystal frost free for an indefinite length of time, without special shielding, even in a humid environment. Even so, many laboratories struggle with ice on their samples. Details of nozzle design, stream control, and interaction between cold stream and crystal mount will be presented.

With equipment design under control, the most difficult part of a cooling experiment is the cooling process itself. Early methods for prevention of ice formation in the crystal focused on modifying the internal water structure. We now think that in the great majority of cases, ice for-mation starts at the surface of the crystal. The problem then becomes one of modifying the surface solvent so that ice nucleation is prevented. A concep-tually simple way of doing this is to remove all solvent from the surface. In well over one-half of all cases this is possible. For the remaining crystals the current practice is to add an antifreeze ("cryoprotectant") to a portion of the solvent mixture, typically 15-25%. Antifreeze agents are small-molecule compounds that are readily water-soluble, and that diffuse rapidly through aqueous solutions. Examples are glycerol, ethylene glycol or MPD. Because just the surface layer needs to be treated, only brief rinses are required, often less than 10 s. Most crystals do not tolerate significant changes in their solvent environment, so the short rinses make it easier to avoid crystal damage prior to cool-down.

Cooling in liquid propane is popular in some laboratories. We have developed tools and techniques for liq. N₂ cool-down, transfer, storage and transport that are very reliable, and simpler than propane-based techniques. Contrary to popular belief, liq. N₂ affords the more rapid cooling.