(a) From Ti to Zr, with one extra shell and no f electrons in between, the radius grows by
160−147=13 pm
If the same increase occurred from Zr to Hf, the predicted radius of hafnium would be
rHf (predicted)=160+13=173 pm
The real radius is only 159 pm, so the lanthanoid contraction has removed
173−159=14 pm
This is slightly more than the whole expected increase: hafnium ends up 1 pm smaller than zirconium. Between La and Hf the fourteen lanthanoids fill the 4f subshell. Each 4f electron shields the rising nuclear charge poorly, so the outer electrons are pulled steadily closer, and this cumulative shrinkage cancels the effect of the new (n=6) shell.

(b) Zr and Hf have the same valence configuration (d2s2), the same common oxidation state (+4) and, because of the lanthanoid contraction, practically the same atomic and ionic sizes. Their compounds therefore have almost identical chemical and physical properties, much more alike than normal members of a group. In nature one ion easily replaces the other in the same minerals, so they occur together; and since separation methods depend on differences in size and chemical behaviour, they are very hard to separate.