Proof (x + y + xy = 36) -> x,y are not N

  • Thread starter Thread starter Ronnie1303
  • Start date Start date
  • Tags Tags
    Proof
AI Thread Summary
The discussion revolves around proving that the equation x + y + xy = 36 has no solutions in natural numbers. Participants agree that both x and y must be even, which leads to the conclusion that the sum must also be even. A key insight involves grouping the equation and analyzing the implications of dividing by (x + y), revealing contradictions when assuming x and y are even. The debate also touches on whether zero is considered a natural number, which complicates the proof, as including zero would yield trivial solutions. Ultimately, the consensus is that the proof demonstrates no valid natural number solutions exist for the equation.
Ronnie1303
Messages
2
Reaction score
0
Hi guys, I've been struggling with this proof for some quite time now. I have to prove that

x + y + xy = 36; has no solution in Natural numbers.

I've started with proving that both x and y have to be even numbers (which is easy to prove) but now I'm going in circles. The problem is, that there are even numbers which are solution to this equation ( -2 and -38, if I recall correctly). Any ideas here?
 
Physics news on Phys.org
You could always brute-force it, though that seems inelegant. Let x and y each run from 1:36 (or let y run from 1:x, since the expression is symmetric).
 
If they are both even, x'+y'+x'y'=18, and there are less cases to consider when using brute force approach.

And the story doesn't end here.
 
Ronnie1303 said:
Hi guys, I've been struggling with this proof for some quite time now. I have to prove that

x + y + xy = 36; has no solution in Natural numbers.

I've started with proving that both x and y have to be even numbers (which is easy to prove) but now I'm going in circles. The problem is, that there are even numbers which are solution to this equation ( -2 and -38, if I recall correctly). Any ideas here?

Consider a rectangle with sides of length x + 1 and y + 1, which are both natural numbers if x and y are.

What's the area of this rectangle when x + y + xy = 36?
 
  • Like
Likes 2 people
pasmith said:
Consider a rectangle with sides of length x + 1 and y + 1, which are both natural numbers if x and y are.

What's the area of this rectangle when x + y + xy = 36?

Very neat. :approve:
 
pasmith said:
Consider a rectangle with sides of length x + 1 and y + 1, which are both natural numbers if x and y are.

Nice! I did not expect to be so profoundly enlightened in such a short sentence :smile:
 
But if so (and if I understand it correctly), answer depends on whether we consider zero to be a natural number?
 
Borek said:
But if so (and if I understand it correctly), answer depends on whether we consider zero to be a natural number?

I think the majority of people would agree that 0 is not a natural number. Some wouldn't. But if 0 were a natural number that makes the problem dead easy. x=0, y=36. Hardly worth asking.
 
Dick said:
I think the majority of people would agree that 0 is not a natural number.
Actually, I would say that this is a 50/50 question. I myself don't agree that N does not include 0. I know that in many cases it makes no sense to say it does, but N* should be used to represent {1, 2, 3, ...}.
Dick said:
But if 0 were a natural number that makes the problem dead easy.
Agreed.
But the way it is written "Proof (x + y + xy = 36) -> x,y are not N" sounds really imperative. I wouldn't even expect myself to be able to prove that x, y are N*.
Dick said:
x=0, y=36
Or y = 0, x = 36.
 
  • #10
Let's start with your (correct) realization that both x and y must be even. This is easily shown.

Group the equation as follows:

(X + Y) + XY = 36

Dividing each term by (X + Y) gives

1 + XY/(X+Y) = 36/(X+Y)

Both XY/(X + Y) and 36/(X + Y) must be even.

But adding 1 to an even number cannot yield an even sum.

This contradicts the fact that X and Y must be even.

Check my algebra and logic.
 
  • #11
FermiAged said:
Dividing each term by (X + Y) gives
1 + XY/(X+Y) = 36/(X+Y)
(...)
This contradicts the fact that X and Y must be even.

I really do like this solution.
 
  • #12
FermiAged said:
Let's start with your (correct) realization that both x and y must be even. This is easily shown.

Group the equation as follows:

(X + Y) + XY = 36

Dividing each term by (X + Y) gives

1 + XY/(X+Y) = 36/(X+Y)

Both XY/(X + Y) and 36/(X + Y) must be even.

But adding 1 to an even number cannot yield an even sum.

This contradicts the fact that X and Y must be even.

Check my algebra and logic.

I'm a little slow today. Why must XY/(X + Y) and 36/(X + Y) be even? Why indeed must they be natural numbers at all?

I hope you can show me, because the proof is otherwise delightful.
 
  • #13
Assuming x and y are even, xy is divisible by 4, and x+y is even.
 
  • #14
Right, but that doesn't mean they're even natural numbers at all!

I mean, if X=6 and Y=4, they are both even, but neither XY/(X + Y) nor 36/(X + Y) is even an integer (they are 2.4 and 3.6 respectively).

Also, an even number divided by an even number isn't necessarily even.

Am I missing something, or does the proof fall apart at that step?
 
  • #15
Chogg is correct. My proof stinks. Let me take advantage of pasmith's suggestion.

(X + 1)(Y + 1) = X + Y + XY + 1 = 36 + 1 = 37

Since 37 is prime, (X + 1) and (Y + 1) must be 1 and 37 (or vice versa)

That means X and Y must be 0 and 36 (or vice versa)

Back to the argument about zero being a natural number!
 
  • #16
What happened to the rule to not provide complete solutions?
 
  • #17
Acoording to the definition it's very elusive whether 0 is a natural number or not. Neither is incorrect, however, if 0 were a natural number, the solution of this assignment is painfully obvious.

Since the assignment states that proof is needed such that there is no solution among natural numbers, we kind of have to assume right off the bat that 0 is Not a natural number.

It is relatively easy to prove, one can show that the values for x and y are fractions, fractions that at no time of the day can be natural values.
 
Last edited:
  • #18
FermiAged said:
Let's start with your (correct) realization that both x and y must be even. This is easily shown.

Group the equation as follows:

(X + Y) + XY = 36

Dividing each term by (X + Y) gives

1 + XY/(X+Y) = 36/(X+Y)

Both XY/(X + Y) and 36/(X + Y) must be even.

But adding 1 to an even number cannot yield an even sum.

This contradicts the fact that X and Y must be even.

Check my algebra and logic.

You don't have to do any division - if you do, you have to justify that the resulting quotients are all integers, as you yourself have realized.

Actually, it's very easy to see why x and y must both be even. Just use these well-worn properties:

odd + odd = even
even + even = even
odd + even = odd
odd*even = even
even*even = even

Those should be obvious, but they're trivially proven.

There are only 3 possibilites: x and y are both odd, exactly one of x and y is odd, or both x and y are even.

First case. You get odd + odd + odd = even + odd = odd. That can't equal 36.

Second case. Say x is odd, y is even. The relation is symmetric, so it doesn't matter which you choose to be odd. So you get odd + even + even = odd + even = odd. Again it can't equal 36.

The final case is the only one that is admissible. Even + even + even = even. There's no contraindication equating that to 36.

Done.
 
  • #19
LCKurtz said:
What happened to the rule to not provide complete solutions?

I agree that enforcement of this has been fairly arbitrary. I've been issued an infraction before for providing an alternate solution *after* the original poster had solved the problem.

Then again, I don't know if what was provided in this thread constitutes a full solution. Pasmith's post provides a very neat solution and the large insight has already been revealed. But the small insight of recognising that 37 is prime is still required.

As to the question of whether 0 is a natural number, there is no standard convention. That makes this a dumb question. It could easily have been rectified by saying there are no solutions in the positive integers.
 
  • #20
FermiAged said:
Back to the argument about zero being a natural number!

lendav_rott said:
Acoording to the definition it's very elusive whether 0 is a natural number or not

I say that 0 is not a natural number! That settles it, doesn't it? :smile:
 
  • #21
LCKurtz said:
I say that 0 is not a natural number! That settles it, doesn't it? :smile:
Obviously.
 
  • #22
LCKurtz said:
I say that 0 is not a natural number! That settles it, doesn't it? :smile:

Only for those that agree :-p
 
  • #23
LCKurtz said:
What happened to the rule to not provide complete solutions?

The rule is still there. It's just too bad I haven't found this thread earlier, otherwise I would have deleted most of the posts. Now it's too late for that.

Pasmith gave a pretty large hint, but not too large. So that was not a problem.

Please feel free to report these kind of threads next time.
 
  • #24
hey guys... the question says x, y should not be N. then why are u repeatedly saying that the answers are 0 and 36? 36 is a natural number as far as I know:D
 
  • #25
Borek said:
If they are both even, x'+y'+x'y'=18, and there are less cases to consider when using brute force approach.

And the story doesn't end here.

Actually, that should be ##x' + y' + 2 x'y' = 18##.
 
Back
Top