Copper vs silver conductors will not cause a phase shift in you application. Over long lengths of cable the different resistivity of the conductors and, more importantly the different cross sections of the conductors would interact with the I,C, and R characteristics of your speakers to cause a phase shift. However, you are likely not talking about using #6 copper and #28 silver or having runs of 100's of feet. As to an advantage of using silver for the high end and copper for the lo end, the effects of the differences in the R,C, and L values of the two cables are going to be swamped out by the rest of the system. However, you may find it personally valuable in determining the merits of this type of tweaking to obtain your own data by conducting your own experiments. In brief - create your own blind study. Start with generic # 12 cables as one system and the small silver and large copper system. Find a way to hide from yourself which cable system is which - whether hiding the cables in identical unshrinked heat shrink coupled with something to make the diameter look the same - using some type of flexible conduit or pvc or some other method of keeping yourself in the dark as to what cable is being used at any given time. Mark the cables inside the heat shrink. Mark the cables on the outside so that each day you strip off a marking and have another marking underneath. Record your impressions from listening on a day to day basis. At the end of some period of time that gives you a decent number of samples - say 50, remove your controls and see what your results tell you. As to methodology, my suggestions were on the spur of the moment and I would certainly put more time and thought into the experimental setup and how you achieve the necessary controls.
