I don't see why not. As long as your elastic modulus is well defined, it should follow the general solution. The only caveats I see are:
1) K is hard to define because the top and bottom "flange" of a soda can are essentially rigid compared to the vertical walls, but there is a curved "connection" between the compression element and the rigid ends. So you could try approximating it as fixed-fixed with radial bending in the cylinder considered hinged?
2) If you axially load the can via the tabbed area, the load on the walls will immediately be eccentric, causing an induced moment, which will lower the available buckling stress. Basically, your calculated F_critical will not be conservative.