Most o-ring manufactures provide leak data under typical circumstances. Most vacuum applications use o-rings with a material diameter (minor diameter of toroid shaped o-rings) of 1/16″ to 1/4″ but they can be smaller and larger. The choice of minor diameter depends on mechanical tolerances and how well you adhered to standard o-ring groove rules. Again, manufacturers typically provide those rules for their o-rings in both circular and rectangular cross section. In most cases you can design o-ring systems that fully compress so they have zero space between the two pieces. Using an 1/8″ minor diameter o-ring with a 2″ major diameter (again toroidal geometry), I’ve completed experiments holding hydrogen at 5 Torr for 8-10 hours with a pressure change of less than 10 mTorr.
To make the seal last, I suggest a silicone vacuum grease applied extremely lightly. When done well, a viton o-ring will look like shiny and smooth. The hardness of the o-ring material is also important. I prefer hard o-rings for most applications. Soft o-rings tend not to do well in vacuum situations. Viton is the best choice in my opinion but some people like teflon or other common materials. If you are sealing metal to ceramic take great care to clean both pieces. O-ring sealing to rough ceramics is challenging.
Another thought that I haven’t tried yet but I intend to try in the next month or so is metal o-rings. Vendors can supply reusable steel o-rings that are completely sealed tubes with extremely thin walls. The o-rings can be coated in silver or tin to improve the seal. They should work wonderfully for situations when plasma or high temperature gas could corrode an o-ring but I have no experience. The ceramic could be coated in a thin ring of silver or tin where the o-ring mounts to improve the seal. Best of luck.
.
