ISI Power

Let p 1 , p 2 , p 3 p_{1},p_{2},p_{3} be primes with p 2 p 3 p_{2} \neq p_{3} , such that 4 + p 1 p 2 4+p_{1}p_{2} and 4 + p 1 p 3 4+p_{1}p_{3} are perfect squares.

Find the number of ordered triplets of ( p 1 , p 2 , p 3 ) (p_{1},p_{2},p_{3}) .

If your answer is infinite then type 1 -1 .


The answer is 2.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

1 solution

If p 3 p 2 p_3\ne p_2

Let { p 1 p 2 + 4 = a 2 p 1 p 3 + 4 = b 2 \begin{cases}p_1p_2 + 4 =a^2 \\ p_1p_3 + 4 =b^2 \end{cases}

Thus

p 1 p 2 = a 2 4 = a 2 2 2 = ( a + 2 ) ( a 2 ) p_1p_2 = a^2 - 4 = a^2 - 2^2 = (a+2)(a-2)

p 1 p 3 = b 2 4 = b 2 2 2 = ( b + 2 ) ( b 2 ) p_1p_3 = b^2 - 4 = b^2 - 2^2 = (b+2)(b-2)

c a s e ( i ) case(i) p 1 = a + 2 p_1=a+2 & p 2 = a 2 p_2=a-2

c a s e ( i i ) case(ii) p 1 = a 2 p_1=a-2 & p 2 = a + 2 p_2=a+2

c a s e ( i i i ) case(iii) p 1 = b + 2 p_1=b+2 & p 3 = b 2 p_3=b-2

c a s e ( i v ) case(iv) p 1 = b 2 p_1=b-2 & p 3 = b + 2 p_3=b+2

Now c a s e ( i ) case(i) & c a s e ( i i i ) case(iii) can't be true simultaneously as that would imply p 2 = p 3 p_2 = p_3 .

Similarly c a s e ( i i ) case(ii) & c a s e ( i v ) case(iv) can't be true simultaneously either.

Thus let us consider without any loss of generality b > a p 3 > p 2 b>a \Rightarrow p_3 > p_2 .

That would imply c a s e ( i ) case(i) & c a s e ( i v ) case(iv) are true.

Thus p 1 = a + 2 = b 2 p_1 = a+2 = b-2

p 2 = a 2 = p 1 4 p_2=a-2 = p_1-4

p 3 = b + 2 = p 1 + 4 p_3=b+2 = p_1+4

Now we have to find p 1 p_1 such that p 1 4 , p 1 , p 1 + 4 p_1-4,p_1,p_1+4 all are primes.

Now all primes are of the form 6 k + 1 6k+1 or 6 k 1 6k-1

Thus if p 1 = 6 k 1 p_1 = 6k-1 then

p 2 = 6 k 1 4 = 6 k 5 = 6 ( k 1 ) + 1 p_2 = 6k-1-4 =6k-5=6(k-1)+1 thus it can be a prime.

But p 3 = 6 k 1 + 4 = 6 k + 3 p_3 = 6k-1+4=6k+3 thus it is a multiple of 3 3 and thus not a prime.

Then if p 1 = 6 k + 1 p_1=6k+1 then

p 3 = 6 k + 1 + 4 = 6 k + 5 = 6 ( k + 1 ) 1 p_3 = 6k+1+4 =6k+5=6(k+1)-1 thus it can be a prime.

and p 2 = 6 k + 1 4 = 6 k 3 p_2=6k+1-4=6k-3 thus it is a multiple of three and not a prime, except when k = 1 k=1 then p 2 = 3 p_2=3 which is a prime.

Therefore, If k = 1 k=1 then

p 1 = 6 ( 1 ) + 1 = 7 p_1=6(1)+1=7

p 2 = 6 ( 1 ) 3 = 3 p_2=6(1)-3=3

p 3 = 6 ( 1 + 1 ) 1 = 11 p_3=6(1+1)-1=11

And if p 2 > p 3 p_2>p_3 then

p 1 = 6 ( 1 ) + 1 = 7 p_1=6(1)+1=7

p 2 = 6 ( 1 + 1 ) 1 = 11 p_2=6(1+1)-1=11

p 3 = 6 ( 1 ) 3 = 3 p_3=6(1)-3=3

Thus the possible solutions for ( p 1 , p 2 , p 3 ) (p_1,p_2,p_3) are

( 7 , 3 , 11 ) (7,3,11) & ( 7 , 11 , 3 ) (7,11,3)

Thus number of solutions is 2 \boxed{2}

Note : As I mentioned in the beginning, this proof is based on the fact that p 2 p 3 p_2 \ne p_3

If p 2 = p 3 p_2=p_3 was possible then any pair of cousin primes (i.e. primes with difference of 4) will be a valid solution. And we don't know how many of those exist (yet).

Proof that all primes are of the form 6 k ± 1 6k \pm 1 .

All numbers can be expressed as one of these,

6 k , 6 k + 1 , 6 k + 2 , 6 k + 3 , 6 k + 4 , 6 k + 5 6k,6k+1,6k+2,6k+3,6k+4,6k+5

Out of which 6 k 6k is divisible by 6 6 .

6 k + 2 6k+2 is divisible by 2 2 .

6 k + 3 6k+3 is divisible by 3 3 .

Thus if a number is prime then it must be of form 6 k + 1 6k+1 or 6 k + 5 = 6 ( k + 1 ) 1 = 6 k 1 6k+5=6(k+1)-1=6k'-1

However the converse is not true,

Example:

43 43 is a prime thus 43 = 6 ( 7 ) + 1 43 = 6(7)+1

However, 35 = 6 ( 6 ) 1 = 5 × 7 35 = 6(6)-1 = 5 \times 7 thus 35 35 is not a prime.

There is a pretty significant error that you made in going from p 1 p 2 = ( a 2 ) ( a + 2 ) p_1 p_2 = (a-2)(a+2) to concluding that { p 1 , p 2 } = { a 2 , a + 2 } \{ p_1, p_2 \} = \{ a-2, a+2 \} . Do you see what the error is?

Do you see how the error can be fixed?

Calvin Lin Staff - 4 years ago

Log in to reply

I think that the error might be that if a × b = c × d a\times b = c\times d that doesn't conclude { a , b } = { c , d } \{ a, b \} = \{ c, d \}

But since A = ( a 2 2 2 ) A = (a^2-2^2) is an integer which is product of two primes p 1 × p 2 p_1\times p_2 (as stated in the question) and ( a 2 ) , ( a + 2 ) (a-2),(a+2) are two factors of A A then it must be the case that { p 1 , p 2 } = { a 2 , a + 2 } \{ p_1, p_2 \} = \{ a-2, a+2 \} .

If ( a 2 ) , ( a + 2 ) (a-2),(a+2) could be further factorized into primes then that would mean A A cannot be product of just two primes.(i.e. product of either two or more primes or product of powers of primes).

I hope that was the error you were looking for sir.... (If I got anything wrong then please do correct me.)

Log in to reply

Right. We have to be careful with the factorization, since for example, we could have a × b = 3 × 5 = 1 × 15 a \times b = 3 \times 5 = 1 \times 15 .

So, to patch the hole, you need to explain why if we have 2 factors that are greater than 1, then the only way for the equation to work out is if { a , b } = { c , d } \{ a, b \} = \{ c, d \} . After that, you have to deal with the case when (at least) one of the factors is equal to 1.

Calvin Lin Staff - 4 years ago

Had given it, done in the similar way.

Vraj Mistry - 4 years ago

After you concluded that we need to find prime such that p-4, p and p+4. You don't need to use the fact that all primes are the form 6k+1 or 6k-1. Just try modulo 3

Lau Mark - 4 years ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...