Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: n does not divide abc
Replies: 4   Last Post: Jul 15, 2014 2:18 AM

 Messages: [ Previous | Next ]
 robersi Posts: 2,410 From: NNY Registered: 8/18/09
n does not divide abc
Posted: Jul 15, 2014 12:28 AM

Assume a^n + b^n = c^n where a,b,c are non zero integers and n is an odd prime.

and the gcd(a,b,c)=1

also assume

a^n-a=0 (mod n^2)
b^n-b=0 (mod n^2)
c^n-c=0 (mod n^2)

then there exists an x,y, and z such that

x+y-z=0
where 0<(x,y,z)<n
x^n+y^n=z^n mod(n^3)

also abc!=0 (mod n)

====================================================
let

a = x + nI_a
b = y + nI_b
c = z + nI_c

where 0<(x,y,z)<=n

a = -x' + n(I_a+1)
b = -y' + n(I_b+1)
c = -z + n(I_c+1)

-n<(-x',-y',-z')<=0 ;
0<=(x',y',z')<n
===================
x+x'=n
y+y'=n
z+z'=n
===================

also assume

a^n-a=0 (mod n^2)
b^n-b=0 (mod n^2)
c^n-c=0 (mod n^2)

therefore a+b-c=0 (mod n^2)

and then
because the gcd(a,b,c)=1

x+y-z=0 (mod n)
x'+y'-z'=0 (mod n)

but because

0<(x,y,z)<=n
and
0<=(x',y',z')<n

we have

x+y-z=0 or =n
x'+y'-z'=n or =0
==========================================

assume, for the time being, that

x+y-z=0

and

x'+y'-z'=n

then

nI_a + nI_b - nI_c = 0 (mod n^2)

a^n = x^n + (n^2)I_a(x^(n-1)) +n^3(I_a')
b^n = y^n + (n^2)I_b(y^(n-1)) +n^3(I_b')
c^n = z^n + (n^2)I_c(z^(n-1)) +n^3(I_c')

a^n = x^n + (n^2)I_a +n^3(I_a'')
b^n = y^n + (n^2)I_b +n^3(I_b'')
c^n = z^n + (n^2)I_c +n^3(I_c'')

a^n+b^n-c^n = 0 = x^n+y^n-z^n +(n^2)(I_a+I_b-I_c) +n^3(I')

x^n+y^n-z^n = -(n^2)(I_a+I_b-I_c)-n^3(I')

x^n+y^n=z^n mod(n^3)
---------------------------------------------------

OR we have (in the other case) simalarly

x'+y'-z'=0
where -n<(x',y',z')<=0
x'^n + y'^n - z'^n = 0 mod(n^3)
---------------------------------------------------------------------------------------------------------

Assume a, b, and c are non-zero integers and n is an odd prime.
a^n + b^n = c^n can be rewritten as A^n + B^n = C^n where a = A, b = B, and c = C.
a^n + (-c^n) = -b^n can be rewritten as A^n + B^n = C^n where a = A, -c = B, and -b = C.
b^n +(-c^n) = -a^n can be rewritten as A^n + B^n = C^n where b = B, -c = A, and -a = C.
So because the GCD(A,B,C) = 1 then it can can be said without a loss of generality that AB!= 0 mod n since
A^n + B^n = C^n can take anyone of these three forms.

------------------------------------------------------------------------------------------
therefore in either case if n|c then z=0 mod n

so either z=0 or z=n.

----------------------------------------------------------------------------------------

because gcd(a,b,c)=1 and we assumed in the first case (w.o.l.o.g.) that 0<(x,y,z)<=n

then only z can = n. ()

assume z = n

x^n + y^n = n^n mod (n^3)

x^n = -y^n mod (n^3)

therfore

x=-y mod (n^2)

z!=n

abc!=0 mod n
---------------------------------------------------------------------------------------
recall
0<=(x',y',z')<n

OR we have (in the other case)

x'+y'-z'=0

x'^n + y'^n - z'^n = 0 mod(n^3)

----------------------------------------------------------------------------------------
--------------------------------------------------------------------------------------
So assume C=0 (mod n) ===> z' = 0 ====> since x'+y'=z' and -n<(x',y',z')<=0
and the gcd(a,b,c)=1
then z!=0 ==========> abc!=0 (mod n)

----------------------------------------------------------------------------

Date Subject Author
7/15/14 robersi
7/15/14 quasi
7/15/14 robersi
7/15/14 quasi
7/15/14 quasi