I can't give you a straight answer, but I can give you some real thermodynamic run-around! The energy for the expansion comes from the chemical combustion of the explosive, or in the case of nukes, from nuclear processes. But it is very difficult to predict ahead of time what is happening because it is so dependent on conditions.

For example, an expanding gas tends to cool, which causes it to contract, (or not expand as fast). If you can keep the heat from escaping, all the energy can go into the expansion. This is called an isothermal process. But most expansions are at least partly adiabatic which means some energy is lost to cooling as the gas expands.

In the case of bullet ballistics, the situation is very complex, and it depends on a lot of things, such as how long the bullet is affected by friction in the barrel, how fast the powder burns, how fast the air in the barrel (which opposes the bullet) can get out of the way, and a host of other factors, such as the residual temperature and pressure of the gases after the bullet is on its way, and how much of that pressure is siphoned off to operate the mechanism (an M-14 is semi-automatic). Usually the manufacturers select the combustion rate of the powder so it is all burned up before the bullet exits the barrel. They usually tune the mix for the barrel length the round is most likely to encounter in the market, or in combat. The acceleration of the projectile is determined by how much powder has burned at every point along the way and how much room there is in the expanding cylindrical volume between bullet and chamber at every point along the way.

I remember one of the first problems I had in my Fortran class was to calculate bullet speed, location, and barrel pressure in the barrel of an M-14 rifle during the entire path of the bullet down the barrel. I was provided with the combustion properties of the powder. The speed reached a maximum at the muzzle, for obvious reasons, but the accleration was maximum in just the first tiny fraction of the journey (i.e., most of the work was done long before the powder was used up, and long before the bullet left the muzzle). I remember the curve for barrel pressure peaked when the bullet was about halfway out, but speed climbed quickly to near-max then leveled off long prior to exiting the muzzle.
Of course, the ammo was tuned to that piece, in another weapon, the performance might be very different. Ballistics weenies are always experimenting with different "loads" to optimize the performance of different rounds for different weapons and applications.

However, it is possible to calculate the volume of gas produced quite easily, IF you know the chemistry involved. Gunpowder is made up up carbon, which does the burning, and some compound like saltpeter (KNO3)which provides the oxygens needed to burn the carbon to CO2(all the other stuff is just traces of material used to help catalyze and moderate the reaction). If you can find out the chemical composition of the gunpowder, and how much carbon and saltpeter there are in each round, you can calculate the volume of CO2 produced. Of course, you will have to reduce everything to standard temperature and pressure (one Atmosphere, 0 deg C). Gases expand to fit any volume available for them.

If the manufacturers did their engineering right, most of the carbon will be converted to carbon dioxide and just enough oxygen will be left over to combine with the other substances in the propellant. I assume the other materials take up only a small proportion of the propellant, by mass. You will need the chemical formula for the combusion process to determine how much KNO3 was needed, and how the sulfur and other reactants interacted to produce the burn and to determine the oxides of the waste products.

Most commercial rounds are designed for the typical weapon they are likely to be used in, with a safety factor built in to guard against an old or damaged barrel from bursting. I would imagine a .22 or .38 would have as much of the chemical energy available in the propellant utilized with as little left over as possible, provided it was fired from a standard piece..

Hint: if you fire either of those rounds in a short-barreled weapon, note the muzzle flash. Pistol (.22) or snub-nose (.38) shooters can afford to go with a lighter load to reduce recoil, flash, smoke and report without affecting muzzle velocity. This is why putting a .357 Magnum round in a snub-nosed revolver is pointless. Most of the energy is wasted.