[3dem] Re: Freezing hydrophobic grid (Chuck Sindelar)

[Chuck Sindelar]

[Note to the administrator: though the name of my 3dem account is  
"sindelar at alum.mit.edu", I am actually sending/receiving 3dem  
messages from my work account here at lbnl-- hope this works]


On Apr 7, 2008, at 8:10 PM, Puey Ounjai wrote:


> I am trying to make thin ice on hydrophobic carbon surface
> (continuous). I am trying to make ice equally spreaded on the surface.
> Usually, it breaks and form droplet on the carbon?
> Any  suggestion would be highly appreciated!
>

Hi Puey,
I have come up with a relatively simple solution to your problem,  
which I faced because my protein complexes dissociated when place on  
glow-discharged holey carbon grids.

The solution:

(1) add ~2.5-3 µL of sample to an ordinary, NON-glow-discharged holey  
carbon grid.  NB: This required some care to ensure that the entire  
carbon surface was fully covered by sample.

(2) use the edge of a piece of filter paper to gently "wick" away  
most of the sample from the surface of the grid.  This will leave a  
very thin meniscus on the carbon surface which should just barely be  
visible by eye, but will still cover the entire area of carbon.

(3) mount the grid in a plunge freezing apparatus (we use a homemade  
device).  Now, carefully approach the grid with a piece of filter  
paper (held by both hands and flexed slightly convex towards the grid  
works best).  When contact is made to the grid, indicated by a wet  
spot on the filter paper, IMMEDIATELY (within 1/2 second) plunge the  
grid into liquid ethanol.


The results I get from this procedure are generally not as uniform as  
with glow-discharged grids, but with a little practice I found that I  
can reproducibly get large areas of ice of the proper thickness.   
Unlike the glow-discharged method, however, ice in my method tends to  
not to have the pretty symmetry of being thinnest in the middle of  
grid squares and gradually getting thicker towards the edge of the  
squares.  Instead, you tend to get irregular, patchy areas of good  
ice often separated by "veins" of thicker ice.  Also, many squares  
tend to have ice that is practically opaque, it is so thick (i.e.  
useless).

With these caveats, however, I can definitely say that I have had  
great success once I mastered the method-- on a good grid on a good  
day, working with microtubule filaments, I can collect enough data  
for an 8Å structure!


good luck

Chuck