Earl of Plasma wrote: It’s like shooting one tiny drop against another and a bit bigger drop. Doesn’t much of the material just pass through. And even if it doesn’t I don’t think much of the kinetic energy will be released. The drop must be slowed down to a much lower speed if the kinetic energy is to be released. The slow down distance is very short. So I suppose that’s why you have to shoot from all direction simultaneously to make that happening. A cylinder of deuterium could if it’s long enough slow down the drop and release the kinetic energy, but the volume is too big and the heat would be spread in too much mass to reach high temperatures enough. So collision is possible but difficult.

tiny pellet hitting other tiny pellet will not pass through since atoms are packed in solids at typical distances of something like 0.3nm. You probably refer to plasma and gasses collisions with much larger inter particle distance. I expect rapid de-acceleration and instant conversion of kinetic energy to heat. I was initially thinking that you need to shot from all directions (4 in tetrahedron configuration at least) to prevent momentum to escape, but then realized that this is a problem only if collision takes place at the speeds comparable to speed of sound in the material (like fission bomb implosion). If however speeds are > 100 times larger then heat escape is much slower then energy delivery. In first approximation, since material temperature goes up and heat moves out faster. Example: leading edge in supersonic plane is not transferring heat to the air ahead of the plane. If the impact area has no time to expand the volume is constrain too.