Let $x \in \mathbb{N}$, the set of positive integers. The sum of the divisors of $x$ is denoted by $\sigma(x)$. Denote the deficiency of $x$ by $D(x):=2x-\sigma(x)$, and the sum of the aliquot parts of $x$ by $s(x):=\sigma(x)-x$. Finally, denote the abundancy index of $x$ by $I(x):=\sigma(x)/x$.

If $m$ is odd and $\sigma(m)=2m$, then $m$ is called an odd perfect number. Euler proved that an odd perfect number (if one exists) must have the form $m=q^k n^2$ where $q$ is the special / Euler prime satisfying $q \equiv k \equiv 1 \pmod 4$ and $\gcd(q,n)=1$.

Here is my question:

Is ${n^2}/D(n^2) \in \mathbb{N}$, if $q^k n^2$ is an odd perfect number?


From the fundamental equality $$\frac{\sigma(n^2)}{q^k}=\frac{2n^2}{\sigma(q^k)}$$ one can derive $$\frac{\sigma(n^2)}{q^k}=\frac{2n^2}{\sigma(q^k)}=\gcd(n^2,\sigma(n^2))$$ so that we ultimately have $$\frac{D(n^2)}{s(q^k)}=\frac{2s(n^2)}{D(q^k)}=\gcd(n^2,\sigma(n^2)).$$

Thus, we have $$y:=\frac{n^2}{D(n^2)}=\frac{{n^2}}{s(q^k)\gcd(n^2,\sigma(n^2))}=\frac{{n^2}D(q^k)}{2s(q^k)s(n^2)}.$$

In particular, we obtain $$\frac{n^2}{D(n^2)}=\frac{{n^2}}{s(q^k)\gcd(n^2,\sigma(n^2))}=\frac{\sigma(q^k)}{2s(q^k)}.$$

Using the reasoning in this blog post, we obtain $$\frac{q}{2} + \frac{1}{2{q^{k-1}}} - \frac{1}{2q^k} < y \leq \frac{q}{2} + \frac{1}{2{q^{k-1}}}.$$

Equality holds if and only if $k=1$.

That is, it appears that $y \in \mathbb{N}$ if and only if $k=1$ (i.e., the Descartes-Frenicle-Sorli conjecture holds).

So the question "Is ${n^2}/D(n^2) \in \mathbb{N}$?" is equivalent to asking whether $k=1$, for $m=q^k n^2$ an odd perfect number with special / Euler prime $q$.

  • $\begingroup$ Unless n is prime is your first line necessarily true? $\endgroup$ – Steve B Jul 20 '18 at 3:02
  • $\begingroup$ @SteveB, the first line $$\frac{\sigma(n^2)}{q^k}=\frac{2n^2}{\sigma(q^k)}$$ is necessarily true since $m=q^k n^2$ is perfect implies that $\sigma(m)=2m$. Now use the fact that $\sigma$ is (weakly) multiplicative and that $\gcd(q,n)=1$. You will also need to use the fact that $\gcd(q^k, \sigma(q^k))=1$. $\endgroup$ – Jose Arnaldo Bebita-Dris Jul 20 '18 at 7:00
  • $\begingroup$ @SteveB, by the way, it is known (by work of Nielsen, 2015) that $\omega(n) \geq 9$, where $\omega(x)$ is the number of distinct prime factors of $x$. In other words, it is known that $n$ must be composite. $\endgroup$ – Jose Arnaldo Bebita-Dris Jul 20 '18 at 13:57

We have $$\frac{n^2}{D(n^2)}\in\mathbb N\iff k=1$$

Proof :

If $\frac{n^2}{D(n^2)}\in\mathbb N$, then we have $$\frac{2n^2}{D(n^2)}=\frac{2n^2}{2n^2-\sigma(n^2)}=\frac{2n^2}{2n^2-\frac{2n^2q^k}{\sigma(q^k)}}=\frac{2n^2}{2n^2-\frac{2n^2q^k(q-1)}{q^{k+1}-1}}=q+\frac{q-1}{q^k-1}\in\mathbb N$$ from which we have to have $$\frac{q-1}{q^k-1}\in\mathbb N$$ from which we have to have $$\frac{q-1}{q^k-1}\ge 1\implies q-1\ge q^k-1\implies q^{k-1}\le 1$$ from which $k=1$ follows.

If $k=1$, then $$\frac{n^2}{D(n^2)}=\frac{q^{2}-1}{2(q-1)}=\frac{q+1}{2}\in\mathbb N\quad\square$$

  • $\begingroup$ Thank you for this splendid answer, @mathlove! +1! =) $\endgroup$ – Jose Arnaldo Bebita-Dris Oct 10 '18 at 5:16
  • $\begingroup$ @Jose Arnaldo Bebita Dris: Just curious. Are there any reasons you have not accepted only this answer? :) $\endgroup$ – mathlove Oct 11 '18 at 15:29
  • 1
    $\begingroup$ My apologies, I missed this one! Gladly accepting your answer now. =) $\endgroup$ – Jose Arnaldo Bebita-Dris Oct 12 '18 at 3:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.