OK, why don't we try solving this problem by using a much lower-energy plasma with increased density? We could try to solve the buoyancy problem by making it colder (say, 1 eV, or 8000K, which is a bit hotter than the surface of the Sun), thus necessitating a thousand times more ions in the same volume, but its density would still be much too low to push it through the atmosphere on momentum alone. It wouldn't necessarily float up, but try throwing a balloon at someone and you can see how well an object with atmospheric density would fly if hurled at the target.
No, if you want it to push its way through atmosphere on momentum, it must be either much denser than air or moving at extreme velocity, which sci-fi plasma weapons generally do not (and which would make it more of a particle beam than a traditional sci-fi "plasma weapon"). So what if we decrease the volume to make it as dense as a solid projectile? Well, that takes care of the "can't push its way through atmosphere" problem, but now you have to make it tiny, and in order to do that, you need to squeeze it with immense pressure. If we squeeze our 1MJ plasmoid into a 1cc volume and apply the ideal gas law (which is a good model for plasmas), we find that the pressure is in the range of 700 GPa! When you consider the fact that this is a thousand times greater than the yield strength of high-grade steel, you can begin to see the problem.
How many problems arise when you need a containment field a thousand times stronger than steel just to hold your plasmoid together? Some questions leap to mind, such as "if they can create such a strong containment field which somehow supports itself and doesn't even need a projector device, then why can't they make personal shields as strong or even stronger?" One would also have to ask why it doesn't glow like the Sun, since it would be hotter than the Sun's photosphere and denser than steel. And finally, one would have to ask what the point is of this whole speculation, since our plasma "bullet" is now denser than aluminum and should act like a real bullet now, which means it should drop in gravity. While that may not be an insurmountable hurdle for a hypothetical sci-fi weapon, it certainly doesn't match the sci-fi weapons we know, which do not arc noticeably in gravity.
In conclusion, the idea of a slowly moving self-contained plasmoid weapon simply doesn't make any sense. Your "bolt" is constantly trying to blow itself apart on the way to the target, you must invent some kind of ridiculously strong yet easy-to-run containment field to make it hold together (thus raising obvious questions about why this super containment technology is not used to effortlessly protect against these bolts), and when it finally does hit the target and the mythical "containment field" shatters, the barely-contained ions within will promptly scatter in all directions, thus wasting the majority of their energy by dissipating harmlessly into space. Even those ions that do strike the surface of the target will achieve poor penetration; most of their kinetic energy is randomized rather than being directed forward, and the gas cloud lacks the characteristics which would allow it to push through solid armour rather than simply heating its surface. And after all that, the plasmoid won't move in a straight line the way they're invariably shown in sci-fi; it should arc downward in gravity.
