For any k∈N∗ we denote Pk(A)={ak∣a∈A}. The condition CP(k) is then equivalent with
x−y∈Pk(A),for any x,y∈Pk(A),
meaning that Pk(A) is a subgroup of the additive group (A,+).
a) For A=Z4, we have that P2k(A)={0^,1^}, respectively P2k+1(A)={0^,1^,3^}, for any k∈N∗. These are not subgroups of the group (Z4,+). Hence, the ring (Z4,+,⋅) does not have the property CP(k) for any k∈N, with k≥2.
b) For n∈N, n≥3, we consider the ring (Zn,+,⋅). Because 1^∈Pk(Zn) for any k∈N, k≥2, and (Zn,+) is cyclic, generated by 1^, it follows that CP(k)⟺Pk(Zn)=Zn. Equivalently, CP(k)⟺ the function pk:Zn→Zn, defined by pk(x)=xk for any x∈Zn, is bijective.
Then we can rewrite M(n)={m∈N∗∣pm is bijective\}.
Because for even k we have that pk(1^)=pk(−1^), and 1^=−1^, it follows that any m∈M(n) is odd, so that M(n)⊆2⋅N+1.
Because p1=idZn is bijective, we have that 1∈M(n).
Let m1,m2∈M(n) be arbitrary. Since the functions pm1 and pm2 are bijective, the function pm1m2=pm1∘pm2 is also bijective, as the composition of two bijective functions. We deduce that m1⋅m2∈M(n).
It follows that M(n) is a submonoid of the monoid (N∗,⋅).