7
$\begingroup$

Let $a$ be a perfect square number whose decimal representation is the concatenation of two perfect squares, for example $49$ (from $4$ and $9$), $169$ (from $16$ and $9$), ... and $4900$, $490000$ that are divisible by $10$ (simple). (What is called in mathematics such numbers?)

How we can prove there are infinitely many such perfect squares with above property that not divisible by $10$.

$\endgroup$
  • $\begingroup$ I might be being dumb but I am really struggling to understand what the property you want is. $\endgroup$ – Jack Yoon Sep 18 '14 at 8:46
  • $\begingroup$ I am not sure what you are exactly asking. Does the sequence $p^2$ with prime $p$ meet your requirement? $\endgroup$ – gammatester Sep 18 '14 at 8:46
  • $\begingroup$ Whatever your question exactly is, the answer is probably yes and you can presumably construct such a sequence using Euclid's formula for Pythagorean triples. $\endgroup$ – J.R. Sep 18 '14 at 8:50
  • $\begingroup$ As I think @MathMan want to ask about those perfect squares which can be obtained after juxtaposition of some perfect squares. e.g. $49$ is obtained after juxtaposition of $4$ and $9$. $\endgroup$ – Shodharthi Sep 18 '14 at 9:01
  • 1
    $\begingroup$ See the Online Encyclopedia of Integer Sequences, which lists this as A039686 though it gives remarkably little information oeis.org/A039686 $\endgroup$ – Mark Bennet Sep 18 '14 at 9:48
6
$\begingroup$

Let $$x_k = 2^{2k-1} 5^{2k} - 10^k + 1 = 4\underbrace{99...9}_{k-1}\underbrace{00...0}_{k-1}1.$$ Then $x_k^2$ has the desired property, as it is the concatenation of the squares of $$2^{k-1}5^k-1=4\underbrace{99...9}_{k-1} \quad \mbox{ and }\quad 10^k-1 = \underbrace{99...9}_{k}.$$

For small $k$, we get the following:

$$ k=1: \quad x_1^2 = 41^2 = \underbrace{16}_{4^2}\,\underbrace{81}_{9^2}. $$ $$ k=2: \quad x_2^2 = 4901^2 = \underbrace{2401}_{49^2}\,\underbrace{9801}_{99^2}. $$ $$ k=3: \quad x_3^2 = 499001^2 = \underbrace{249001}_{499^2}\,\underbrace{998001}_{999^2}. $$ $$ \vdots $$

It is an easy computation to check that this works for every $k$. Note that this method does not find all solutions, but it does show that there are infinitely many of them. Probably there will be other infinite families of solutions.

$\endgroup$

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.