In a lab a few weeks ago we determined the transport numbers of Ag+ and NO3- by the Hittorf method (Ag anode & cathode; three apparent compartments in H-cell; net gain in mole of Ag+ in one side, and net loss in other end) and of H+ by the Moving Boundary Method (Cd anode, Pt cathode; HCl(aq); H+moves faster towards cathode than Cd2+ resulting in a boundary). The latter experiment I am wondering about.
In our lab manual (and outside reading: Physical Chemistry, Atkins) I am told the 'solutions' (in this case CdCl2 and HCl, which I initially think of just ions in water) need to have densities such that the slower cation-solution (here, CdCl2) needs to be more dense than the faster cation one (HCl). This is where my question arises. If the different drift speeds cause the boundary, why too do the densities need to be that way - would the solutions mix with different densities? If so, then the drift speeds are all of sudden unimportant? I suspect the two conditions are necessary (more dense on bottom and slower on bottom), but do not know why.
And just a thought (I wish I could do it in a lab): what if we flipped the apparatus around such that the HCl was on the bottom and the CdCl2 on the top (and the anode on the bottom). Would it now be a competition between H+ and Cd2+, but with gravity now on Cd2+'s side?
Thanks for your help
(I had a clash with physical chemistry in the first semester so apologies for the ignorance!)