Challenges of matching the results of 2D axisymmetric and 3D for a cylinder

Hello everyone:
I am modeling the transport of diluted species in a cylinder in 2D axisymmetric and 3D domain. I could not match the simulation results of the 3D model with that of the 2D axisymmetric. The obtained current value (using component coupling: integration) in 2D axisymmetric is higher than that of the 3D model.
Although that modeling the cylinder geometry in 3D sounds very easy and straightforward at first place but it is rather very complicated!
In 3D:
If I solve the problem for the whole cylinder, the current value would be negligible. Then I have to solve the problem for half of the cylinder and considering the symmetry boundary ( no flux) on the other half of the cylinder ( matching the result is not achievable with 2D-axi). And If I solve the problem for a quarter of cylinder, I would get very wrong results as I got the flux in both x and y coordinates axes!!! (with the same values)
Wondering if I am doing something wrong or if it is rather software limitation?
Any response would be greatly appreciated
Thank you so much in advance
Mah