I was wondering what number of cells is best for a pellet for TEM sample processing. I believe the best pellets are ~ 1mm^3 in literature, but I wanted to know what worked best for everyone else.
Your question is a bit confusing to me. You mean the recommended particle size to see by TEM? or do you mean the recommended grids to see this type of sample ???
Well, it depends on your sample. When working with individual cells like bacteria, yeasts, etc., you should always see the pellet at the bottom of the tube to control the fixation process (aldehyde, +- osmium tetroxide). There is enough for standard bacterial cultures like E. coli or Bacillus subtilis to take 0.8 mL of exponential culture, mix it with an equal volume of double strength fixative and then spin it down in a 2 mL Eppendorf tube. The swing-out rotor is an advantage. However, if you have a lower cells concentration, you can use PCR tubes 0.2 mL with the proper adapter for the final cell's pelleting. Those tubes have a smaller bottom size, and you get the pellet in a smaller volume.
I would say the more the best! But sometimes you don't have the choice for the (big) size of the sample.
But if the pellet is too big, there will be a problem with resin polymerization because the good resin mixture will not reach the center of the pellet. Same for tissues you need to cut them into small pieces. So the 1mm3 volume is a good start for the embedding of pellets or tissues.
On the opposite, because I am working at an EM facility, people come sometimes with small samples for diverse reasons. It can be subpopulation of cells isolated from a mouse organ for example. Even with cells in culture, I may start with a small sample because cells are treated with very expensive chemical compounds or cytokines. So I don't ask people to treat cells in a culture flask, I can start with a well plate (6, 24-12, even 96).
For small samples, you need to adapt the recipe. For examples: avoid tremendous number of centrifugations where you may lose sample. You can embed the pellet in gelatin/low-melting-point agarose at the start. For polymerization it exists sharp conical molds to concentrate the sample however difficult to manipulate with resin... and gloves and to centrifuge. You can also think about flat embedding on plastic/glass for adherent cells, which avoid all centrifugations.