10
$\begingroup$

Let $p,q$ are irrational numbers such that $pq=1$. Then what can we say about character of $p+q$? i.e, $p+q$ is rational or irrational? Probably, I believe it is irrational but could not prove.

$\endgroup$
24
$\begingroup$

Take $$p=\sqrt { 2 } ,q=\frac { 1 }{ \sqrt { 2 } } \\ \\ p=2-\sqrt { 3 } ,q=2+\sqrt { 3 } \\ $$

$\endgroup$
  • $\begingroup$ very nice !! ok I get it $\endgroup$ – Jamal Gadirov May 21 '17 at 19:10
13
$\begingroup$

The question then becomes, if $p$ is irrational, then is $p+\frac{1}{p}$ irrational?

In other words, you want to know if $$ \frac{p^2+1}{p}=\frac{a}{b} $$ for some integers $a$ and $b$.

Let's just try a value for $\frac{a}{b}$. Let's try $\frac{a}{b}=3$. Therefore, we are trying to solve:

$$ \frac{p^2+1}{p}=3. $$ This can turned into $$ p^2-3p+1=0 $$ and solved with the quadratic formula: In this case, $$ p=\frac{3\pm\sqrt{9-4}}{2}=\frac{3}{2}\pm\frac{\sqrt{5}}{2}, $$ which is irrational. Therefore, the answer to your question is that $p+q$ can be rational.

In general, we can find a $p$ that satisfies the conditions if $\frac{p^2+1}{p}\geq 2$. Most of the results from this are irrational. We can certainly replace $\frac{a}{b}$ with an irrational number greater than $2$ and get possibilities for $p$ and $\frac{1}{p}$ where their sum is irrational.

$\endgroup$
3
$\begingroup$

Hint: consider for example:

  • $\;p=\sqrt{2}, q= 1 / \sqrt{2}$

  • $\;p=3+2\sqrt{2}, q= 3 - 2 \sqrt{2}$


[ EDIT ] For more examples of irrationals with product $1$ and:

  • irrational sum: let $p$ be a transcendental number (such as $\pi$ for example) and $q=1/p\,$, then $p+q$ will necessarily be transcendental, thus irrational; the same goes for $p$ being the sum between a transcendental number and an algebraic irrational (such as $\pi+\sqrt{2}$ for example);

  • rational sum: let $p,q$ be the roots of the equation $x^2 - n \,x + 1 = 0\,$ for integer $n \gt 2$. Both $p,q$ are irrational by the rational root theorem, and $pq=1\,$, $p+q=n\,$ by Vieta's relations.

$\endgroup$
  • $\begingroup$ ok that means both of them is possible $\endgroup$ – Jamal Gadirov May 21 '17 at 19:10
1
$\begingroup$

Let consider $S=p+q$, $P=pq=1$ and $p>q$

The roots of the polynom $P[X]=X²-SX+P=X²-SX+1$ are $p$ and $q$.

$discriminant = \delta = S²-4 > 0$ given that $P[X]$ has two roots and they are irrational numbers. Then, $ p = \frac{S + \sqrt{S²-4}}{2}$ and $ q = \frac{S - \sqrt{S²-4}}{2}$

As soon as $|S| > 2$ and $S²-4$ is not a perfect square, you can choose $S$ as you want, rational or irrational. $p$ and $q$ will be irrational and the sum will be what you have chosen.

$\endgroup$
0
$\begingroup$

There is another case, namely $p$ and $q=\frac{1}{p}$ are transcendental irrationals. In that case $p+q$ is necessarily transcendental, hence irrational, because $p+\frac{1}{p}=r$ with $r$ rational would imply $p^2-pr+1=0$ contradicting $p$ is transcendental.

$\endgroup$

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.