2

I have proven that the sum of any two irrational numbers is not always irrational and that a rational plus an irrational is irrational, however I am not sure how to prove this specific case.

I was thinking along the lines of: $\sqrt{2}+\sqrt{3}=n$ where $n$ is an element of rationals. By squaring both sides

$$(\sqrt{2}+\sqrt{3})^2 = n^2$$ $$7 + 2\sqrt{2}\sqrt{3} = n^2$$ I have proven the sum of a rational and irrational is irrational yet it is too long and complicated to prove that root two and root three make root six. Moreover, it will be an additionally difficult to prove that two times root six is irrational so I was wondering if there was a quicker less complex way to finish this proof?

Jack D'Aurizio
  • 353,855
Sadee Luongo Lamothe
  • 33
  • 1
    Also see https://math.stackexchange.com/questions/1516482/proving-sqrt-2-sqrt-3-is-irrational – user236182 Oct 16 '17 at 18:56
  • You should have in your tool box the proof that $\sqrt 2$ is irrational. And it is pretty easy to generalize that proof to show that the square root of a non-perfect square is irrational. So proving the irrationality of $\sqrt 6$ should not be your problem. – Doug M Oct 16 '17 at 18:57

6 Answers6

5

The novelty in this question is the request for a proof that doesn't involve knowing (or proving) the irrationality of $\sqrt6$. So here's a proof that only assumes the irrationality of $\sqrt2$ and/or $\sqrt3$.

If $\sqrt2+\sqrt3$ is rational, then so is $\sqrt3-\sqrt2$, since $(\sqrt2+\sqrt3)(\sqrt3-\sqrt2)=3-2=1$. But then so is

$$\sqrt2={(\sqrt2+\sqrt3)-(\sqrt3-\sqrt2)\over2}$$

(and also $\sqrt3=((\sqrt2+\sqrt3)+(\sqrt3-\sqrt2))/2$).

Barry Cipra
  • 79,832
2

Assume that, $$(\sqrt{2}+\sqrt{3})=\frac{p}{q}$$ Then we have $$(\sqrt{2}+\sqrt{3})=\frac{p}{q}\in\Bbb Q \implies 5+2\sqrt{6} =(\sqrt{2}+\sqrt{3})^2 =\frac{p^2}{q^2}\in\Bbb Q\\\implies \sqrt{6} =\frac 52+\frac{p^2}{2q^2}\in\Bbb Q ~~(Impossible )$$

since $\sqrt 6\not \in\Bbb Q$

Hence $$\sqrt{2}+\sqrt{3}\not \in\Bbb Q$$

Guy Fsone
  • 23,903
2

You can show that $\sqrt{2}+\sqrt{3}$ is a zero of the polynomial $x^4-10x^2+1$. By the Rational Root Theorem, the only possible rational roots of this polynomial are $1$ and $-1$.

So $\sqrt{2}+\sqrt{3}$ is irrational.

paw88789
  • 40,402
1

It is enough to generalize a little one proof of $\sqrt{2}\not\in\mathbb{Q}$, namely the following one.

Given a prime $p$ and a positive integer $n$, we may denote as $$\nu_p(n) = \max\{m\in\mathbb{N}: p^m|n\}.$$

Claim: if $\nu_p(n)$ is odd, then $\sqrt{n}\not\in\mathbb{Q}$.

Proof: The assumption $\sqrt{n}=\frac{a}{b}$ with $a,b\in\mathbb{N}^+$ leads to $$ a^2 = n b^2 $$ which is impossible, since $\nu_p(a^2)$ is even while $\nu_p(nb^2)=\nu_p(n)+\nu_p(b^2)$ is odd.

Corollary: if $n$ is a positive integer, $\sqrt{n}\in\mathbb{Q}$ only if $n$ is an integer square.

Corollary: $\sqrt{6}\not\in\mathbb{Q}$.

Corollary: $\sqrt{2}+\sqrt{3}\not\in\mathbb{Q}$, since the assumption $\sqrt{2}+\sqrt{3}\in\mathbb{Q}$ leads to $\sqrt{6}\in\mathbb{Q}$, contradicting the previous point.

Community
  • 1
Jack D'Aurizio
  • 353,855
0

after squaring we get $$2\sqrt{6}=\frac{m^2}{n^2}-5$$ the right-hand side is rational, and the left-hand side irrational.

Dr. Sonnhard Graubner
  • 95,283
0

Suppose $\sqrt{2}+\sqrt{3} = q$ for some $q\in \mathbb{Q}$. Then $$2+2\sqrt{6}+3 =q^2$$

and so $$\sqrt{6} = {q^2-5\over 2}$$

Since all elementary operation $(x,-,\cdot, :)$ are closed in $\mathbb{Q}$, the right side is rationa and the left is not. A contradiction.

nonuser
  • 90,026