For a HS level project, I don't think this is a good idea... sorry, but it seems way too complicated for HS. I'm currently a MechE major in college and I've already found out the hard way that simple projects executed cleanly with good scientific method following is the way to go.
Try dumbing down your project ALOT and forcus on the steps of the scientific method. A simple project would be something more along the lines of "which gas shoots a paintball fastest"? Your hypothesis could involve a challenge on the current compressed air standard of paintball.
-Next you want to develope a simple, yet efficient and consistant method to propel a single paintball, where you can control all variables, leaving only the gas to be the test variable.
To do this you could develop a very simple PVC spud gun, with a smallish volume chamber of around 40 cubic inches (or 2 feet of 2" PVC) and a electronic 1" operated solenoid sprinkler valve (you can get them for around $20 on ebay, look for rainbird brand, orbit is also good, look on spudfiles for more info on this)
-Now you want your barrel to be about 1/2 the volume of the chamber, and .75" sch 80 pvc is a size to go with because it is about .05" larger than a paintball, so a little paper towel as wadding will fill that gap, and to help with consistancy from test to test, precut all wadding ahead of time so you have the same amount. You barrel of .75" SCH 80 pvc should be about 45" long if you do the math.
-To finish off this test launcher, you must now make an easy way to inject gas into the cylinder, and a gauge which measures from 0-100 PSI (available at home improvement store) and use the SAME AMOUNT of gas every time. remember you must keep all variables constant except for the test variable - the type of gas.
-Now that you have a consistant, and SIMPLE testing launcher, you must obtain a few common compressed gasses. here is where cleverness and safety come into play.
--Since you will need less than 100PSI to propell the ball at standard paintball game speeds (300fps) you can get away with a gas like propane. (propane will not explode if used as the propelling gas, don't worry about that)
--Also you can use standard compressed air from a compressor.
--You can also use REGULATED CO2, non regulated is a nono unless you know what your doing, as more than 400PSI (CO2 comes out at 800PSI) will cause your PVC test launcher to burst.
--You will want a lite gas, so get a little party favor Helium canister.
--Find one more gas if you can, perhaps nitrogen if you can get ahold of it.
Now that you have a test - run it. You will get different velocity's for each gas, so test then and compare. The fastest is the best perhaps, but you can also bring up stuff in your conclusion like cost/shot and ****.
good luck. remember doing good on a physics project is not always about what you make (unless it's revolutionary, a spoolie IS NOT!), but how you make it. The professionalism you use in your experiment is what a teacher will see and be impressed by.
Sorry if this is not exactly what your project asks you to do - but I'm just assuming it is since it's the standard project, and you really didn't explain it well