7

The question is to find all natural numbers $n$, such that $(n+1)(n+12)$ can be written as a product of $4$ consecutive numbers.

My approach was the following: starting from the remark that if you add 1 to a product of four consecutive integers you receive a perfect square, I can write: $(n+1)(n+12)+1=k^2$, with $k \in \mathbb{N}$. Then, $n^2+13n+13=k^2 \iff 13(n+1)=k^2-n^2=(k-n)(k+n)$. Now, because $13$ is a prime, could I perhaps use the following logic: we might have two cases, either $k-n$ or $k+n$ should be 13(or 1?), and the other should be $n+1$. Solving for $n$ will give us $n=4$(I have however only considered the case where both $k-n, k+n$ are equal to 13, not 1. Is this approach correct, or could it be turned into a correct one?

J. W. Tanner
  • 63,683
  • 4
  • 43
  • 88
  • 1
    @ThomasAndrews $n(n+1)(n+2)(n+3)+1 = (n^2+3n+1)^2$, so yes, the product of four consecutive integers plus $1$ is always a square. The constant term is $1$, not $7$. – jjagmath Oct 30 '24 at 22:55
  • Actually, it is, check https://math.stackexchange.com/questions/155040/prove-that-the-product-of-four-consecutive-positive-integers-plus-one-is-a-perfe?rq=1 for proof. – Dante Ricciagli Oct 30 '24 at 22:56
  • Ah, yes, messed up. – Thomas Andrews Oct 30 '24 at 23:01
  • 1
    From $13a = b c$ you cannot conclude that $13=b$ and $a=c$. For example, $13\times 20 = 26\times 10$, but $13\ne 26$ and $13\ne 10$. – jjagmath Oct 30 '24 at 23:09
  • 4
    If you use completion of the square on $n^2+13n+13$, you eventually get to $(2n+13)^2 - (2k)^2 = 117$ I think, so $(2n+2k+13) (2n-2k+13) = 117$. From here, if you use the prime factorization of 117 that should lead you to all solutions. – Daniel Schepler Oct 30 '24 at 23:22
  • @DanielSchepler And the number of cases (factorizations) to check can be reduced by noticing that $n+k\equiv0\pmod{13}$, see also my answer. – Servaes Oct 31 '24 at 18:57
  • Duplicate of this. Please don't reopen duplicates - it destroys site organization. – Bill Dubuque Oct 31 '24 at 22:47
  • 3
    @BillDubuque This question isn't asking about proving that the product of four consecutive integers is one less than a perfect square; it's about using that fact to solve a different problem regarding such a product. – Carmeister Oct 31 '24 at 23:08
  • 3
    @BillDubuque: How exactly is this question a duplicate? It seems to me that the asker was already aware of the previous question (citing it in a comment) and asked a different question that builds upon the number-theoretical properties of a product of four consecutive integers. – Dan Oct 31 '24 at 23:08
  • @Carmeister I had a good dupe link for that part too but strangely it seems it didn't get posted (network issue?) Let me see if I can find it again (we already have tens if not hundreds of answers showing how to solve such diophantine equations). – Bill Dubuque Oct 31 '24 at 23:53
  • 3
    @BillDubuque To clarify, should (almost) all "solve diophantine equations by bounding between squres" simply be closed as a dupe of some other? It doesn't matter that the expression is different? $\quad$ Extending that logic, should all questions that involve a technique (say denesting radicals, applying power of a point, etc) likewise be marked as dupes? – Calvin Lin Nov 01 '24 at 10:54
  • @CalvinLin See abstract duplicates. Do you seriously think that the method in your answer has never been posted here over the last decade and a half? What is novel in your answer? – Bill Dubuque Nov 01 '24 at 19:19
  • @BillDubuque Of course I think it's a standard problem. $\quad$ TBH This is my first time seeing that "abstract duplicate", I'm not on meta much. Noted that this is the preferred approach. I will refrain from answering in future. – Calvin Lin Nov 03 '24 at 13:26

3 Answers3

9

Jjagmath indicated what is wrong with your approach. All that you have is $ 13 \mid k-n$ or $13 \mid k+n$.

If $ 13 \mid k-n$, then $n+1 \geq k+n$, which is a contradiction since $ k \geq \sqrt{ 2 \times 13 } > 5$. So $ 13 \mid k+n$.
However, I can't push through from here.


Alternative solution: Observe that

$$ (n+3)^2 = n^2 + 6n + 9 < n^2 + 13 n + 13 < n^2 + 14n + 49 = (n+7)^2$$

Then, check $(n+j) ^2 = n^2 + 13n + 13 $ for $ j = 4$ to 6, so we just need to check these 3 cases.

  • Checking $ j = 4$ yields $n = \frac{ 3}{5}$
  • Checking $j = 5$ yields $n = 4$ .
    • This gives $k = j+n = 9$ and $ 13 \mid k+n = 13$.
    • $(n+1)(n+12) = 80$ is not a multiple of 3, hence cannot be written as the product of 4 consecutive integers.
  • Checking $j = 6$ yields $n = 23$.
    • This gives $k = j+n = 29$ and $ 13 \mid k+n = 52$.
    • $(n+1)(n+12) = 4 \times 5 \times 6 \times 7$

Hence, the only solution is $n = 23$.

Note: We could tighten up the first inequality slightly. Since $ n \geq 1$, we also have $(n+4) = n^2 + 8n + 16 < n^2 + 13n + 13$, so we could reject $j = 4$.

Calvin Lin
  • 77,541
  • 1
    @JohnOmielan Oh good point! I overlooked that. – Calvin Lin Oct 30 '24 at 23:40
  • You could close the argument regarding the number of cases faster by noticing that $(n+3)^2 < n^2 + 13n + 13 < (n+7)^2$. Thank you for your answer, it was really helpful! – Dante Ricciagli Oct 31 '24 at 10:08
  • @DanteRicciagli OMG That's a silly arithmetic error. Thanks for letting me know, fixed. – Calvin Lin Oct 31 '24 at 10:12
  • Please strive not to post more (dupe) answers to dupes of FAQs. This is enforced site policy, see here. If you have no interest in helping to organize the site, then please delay answering likely duplicates till after others have processed the question. – Bill Dubuque Oct 31 '24 at 20:19
2

One way continue your line of thought is as follows: If $(n+1)(n+12)$ is the product of four consecutive integers for some natural number $n$, then there is a natural number $k$ such that $$(n+1)(n+12)+1=k^2,\tag{1}$$ and because $n\geq0$ we have $k^2\geq13$ and so $k\geq4$. Rearranging $(1)$ shows that $$13(n+1)=k^2-n^2=(k+n)(k-n),$$ where $k+n>n+1$ and so $k-n<13$. In particular $13\mid k+n$, say $k+n=13m$. Then $$13(n+1)=(k+n)(k-n)=13m(13m-2n),$$ and again rearranging yields $$(2m+1)(26m-4n-13)=-9.$$ Of course $2m+1>0$ and so $2m+1\in\{1,3,9\}$, and $m\in\{0,1,4\}$ respectively. Then for the other factor we correspondingly find \begin{eqnarray} 26m-4n-13=-9\qquad&\text{ and }\qquad m=0\qquad&\text{ so }\qquad n=-1,\\ 26m-4n-13=-3\qquad&\text{ and }\qquad m=1\qquad&\text{ so }\qquad n=4,\\ 26m-4n-13=-1\qquad&\text{ and }\qquad m=4\qquad&\text{ so }\qquad n=23,\\ \end{eqnarray} Of course $n=-1$ is not a natural number, so this is not a solution. For $n=4$ we get $(n+1)(n+12)=5\times16=80$, but $$1\times2\times3\times4=24<80<120=2\times3\times4\times5,$$ so this is also not a solution. For $n=23$ we get $$(n+1)(n+12)=24\times35=2^3\times3\times5\times7=4\times5\times6\times7,$$ and so the unique solution is $n=23$.

Servaes
  • 67,306
  • 8
  • 82
  • 171
  • 4
    You could also quickly rule out 80 as being a product of 4 consecutive numbers by the fact that it's not even divisible by 3. – Daniel Schepler Oct 31 '24 at 19:09
0

First question: why $(n+1) \cdot (n+12)$? I would simply write this as $n \cdot (n+11)$ :-)

So, the question is: how to write $n \cdot (n+11)$ as $k \cdot (k+1) \cdot (k+2) \cdot (k+3)$.

I would go for three obvious possibilities:

  1. $n = k \cdot (k+1)$ and $n+11 = (k+2) \cdot (k+3)$
  2. $n = k \cdot (k+2)$ and $n+11 = (k+1) \cdot (k+3)$
  3. $n = k \cdot (k+3)$ and $n+11 = (k+1) \cdot (k+2)$

The first equation becomes:
$(k+2) \cdot (k+3) - k \cdot (k+1) = 11$
$k^2+5k+6 - (k^2+k) = 11$
$4k+6 = 11$, which has no integer solutions.

The second equation becomes:
$(k+1) \cdot (k+3) - k \cdot (k+2) = 11$
$k^2+4k+3 - (k^2+2k) = 11$
$2k+3 = 11$, which has $k=4$ as an answer, leading to $n=24$ (or $n=23$ in your definition of $n$).
Indeed $(23+1) \cdot (23+12) = 4 \cdot 5 \cdot 6 \cdot 7$

The third equation becomes:
$(k+1) \cdot (k+2) - k \cdot (k+3) = 11$
$k^2 + 3k + 2 - (k^2 + 3k) = 11$
$2 = 11$, which also has no solutions :-)

Now, obviously: when the product of two first numbers equals the product of four second numbers, then it would be great if each of those two first numbers could be written as a product of two of those four numbers (which is the case I have elaborated), but this is not always the case, so my answer serves the purpose "Your question does have a solution." but not the purpose "Give me all the solutions.".

Dominique
  • 3,267
  • 3
    Unfortunately, you may not just assume that's how the factoring works. – Calvin Lin Oct 31 '24 at 15:25
  • @CalvinLin: that's the exact meaning of my last sentence. – Dominique Oct 31 '24 at 15:42
  • 4
    It's really buried and so it gives the impression that you have a correct solution (hence leading to upvotes). My personal preference would be that you're upfront with the gap in the solution, before presenting your idea. – Calvin Lin Oct 31 '24 at 15:50