There are no $n$ satisfying $\phi(n)=m$

62 Views Asked by At

This post is a generalisation of a post I read few days ago on this site. The post asks us to find a natural number $n$ such that $\phi(n)=14$. Interestingly there are no numbers satisfying this equation. I have tried to generalize this as follows: Let $m$ be a square free even integer and $m+1$ not a prime of the form $3mod(4)$ then prove that there are no integers $n$ satisfying $$\phi(n)=m$$. I have thought a lot about this but I have no idea where to begin.

Although I do not know whether or not this is true, but I could not find a counterexample to this so I presume the statement is true.Could you please help me in proving this or bring to my notice an obvious counterexample that might have escaped me?

Thank you.

1

There are 1 best solutions below

3
On BEST ANSWER

There are examples, such as $\varphi(11^2)=110=2\times 5\times 11$. (note that $111=3\times 37$ is not a prime).

Let's analyze the possibilities.

Suppose we had a counterexample $\varphi(n)=m$ where $m$ is of the form you desire.

If $n$ is divisible by any prime $p\equiv 1 \pmod 4$ then $4\,|\,p-1\implies 4\,|\,\varphi(n)$ which would tell us that $\varphi(n)$ is not square free.

Thus if $n$ were a solution to your equation we would know that $n$ could only be divisible by $2$ and odd primes $\equiv 3 \pmod 4$.

Now, if $n$ were divisible by two distinct odd primes we'd again have $4\,|\,\varphi(n)$.

Of course, if $p^3\,|\,n$ for some $n$ we'd have $p^2\,|\,\varphi(n)$

It follows from the above that $n$ must have one of the forms $$p,2p,p^2,2p^2$$ for some prime $p\equiv 3\pmod 4$.

We can eliminate the cases $p,2p$: we'd have $$\varphi(p)=\varphi(2p)=p-1$$ But that's not an acceptable $m$ by your criterion (since $m+1$ would clearly be a prime $\equiv 3 \pmod 4$).

That leaves the cases $p^2,2p^2$. There are examples along those lines, such as $m=11^2$ or $2\times 11^2$. similarly $n=19^2$ or $2\times 19^2$ gives $\varphi(n)=342$ and $343$ is not a prime.