Find the largest integer n such that 2 n divides 3 1 0 2 4 − 1 .
Source: 104 Number Theory Problems Book
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.
If I'm not mistaken this is an exact copy of a competition question. First evidence is that I have this exact question on a handout I got for a Olympiad math class, and second is that I got the correct answer without even doing the problem.
However, since I can't find the source, I'll let this one go.
Log in to reply
Here is proof that 2 ∣ 3 2 n + 1 but 4 ∣ 3 2 n + 1 .
It's obvious that 2 ∣ 3 2 n + 1 . To prove that 4 ∣ 3 2 n + 1 , we take m o d 4 : 3 2 n + 1 ≡ ( − 1 ) 2 n + 1 ≡ 1 n + 1 ≡ 2 ( m o d 4 ) , and we are done. □
Does that mean one should not share questions that are not his/her own?
Log in to reply
At least put the source down in italics at the bottom.
This problem is example 1.10 from the book '104 Number Theory Problems' by Andreescu, Andrica and Feng.
Wait... Wasn't ϕ ( 2 n ) a primitive root mod 3?That means that the smallest value k which satisfies 3 k c o n g t o 1 ( m o d 2 n ) is ϕ ( 2 n ) . That means that the smallest number k for which 3 k − 1 is divisible by 2 1 2 is 2 1 1 , contradicting to the answer!?Where did I go wrong?
Log in to reply
I believe you forgot the extra power of two you receive from 3 1 + 1 .
and note 3 2 n + 1 is also not divisible by 4, but only 2.
3 2 1 − 1 i s d i v i d e d b y 2 3 , 3 2 2 − 1 i s d i v i d e d b y 2 4 , 3 2 3 − 1 i s d i v i d e d b y 2 5 , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 n − 1 i s d i v i d e d b y 2 n + 2 S o , 3 1 0 2 4 − 1 = 3 2 1 0 − 1 i s d i v i d e d b y 2 1 0 + 2 = 2 1 2
how can 2^n divide evenly into 3^1024 -1? the units digit of 3^1024 is one. 3^1024 -1 would have a units digit of zero then. 2^n can never have a units digit of zero.
We use Theorem 4 in this article:
v 2 ( 3 1 0 2 4 − 1 1 0 2 4 ) = v 2 ( 3 − 1 ) + v 2 ( 3 + 1 ) + v 2 ( 1 0 2 4 ) − 1 = 1 + 2 + 1 0 − 1 = 1 2
Great solution, I never actually have seen the LTE Lemma be used until now.
An awesome solution
If you don't understand, go to this link: http://activities.tjhsst.edu/vmt/wiki/images/f/fe/LTEFinal.pdf
Hey bro I think every thing is right but one thing how you find
v2 (3-1) = 1
v2 (3+1)=2
v2 (1024)=10
Log in to reply
v_2(x) is the number of factors of 2 in x.
1024 = 2^10, 3-1 = 2, and 3+1 = 4.
wew
Notice that 3 1 0 2 4 − 1 = ( 3 5 1 2 ) 2 − 1 = ( 3 5 1 2 − 1 ) ( 3 5 1 2 + 1 ) .
Now we do the same with the new difference.
( 3 5 1 2 − 1 ) ( 3 5 1 2 + 1 ) = ( 3 2 5 6 − 1 ) ( 3 2 5 6 + 1 ) ( 3 5 1 2 + 1 )
Continuing this, we get:
( 3 − 1 ) ( 3 + 1 ) ( 3 2 + 1 ) ( 3 4 + 1 ) ( 3 8 + 1 ) ( 3 1 6 + 1 ) … ( 3 5 1 2 + 1 )
Now we will find the largest power of two in each of these factors. We will first do all of the ones where the exponent on the 3 is greater than 1 .
By Euler's Theorem we have that, 3 2 n + 1 ≡ 1 + 1 ≡ 2 ( m o d 4 )
And so each factor where the exponent on the 3 is greater than one is divisible by 2 but not by 4 . This gives a factor of 2 9 thus far, and now we will incorporate the final two terms. ( 3 − 1 ) ( 3 + 1 ) = 2 ⋅ 4 = 2 3 , and so the answer is 3 + 9 = 1 2
What I thought:)
Okay, let's be careful, 'cause this solution is tricky. First of all, observe that decomposing 3 1 0 2 4 − 1 using a 2 q − b 2 = ( a q + b ) × ( a q − b ) 10 times, we have
3 1 0 2 4 − 1 = ( 3 5 1 2 + 1 ) × ( 3 2 5 6 + 1 ) × ( 3 1 2 8 + 1 ) × ( 3 6 4 + 1 ) × ( 3 3 2 + 1 ) × ( 3 1 6 + 1 ) × ( 3 8 + 1 ) × ( 3 4 + 1 ) × ( 3 2 + 1 ) × 2 3
But we may write
3 x = ( 2 + 1 ) x = n = 0 ∑ x ( n x ) 2 x − n
For some natural x . Remember that we want only even x 's so that x = 2 m implies that
( n 2 m ) = n ! ( 2 m − n ) ! ( 2 m ! )
Is always even (unless for n = 2 m in which chase the binomial becomes 1 ). In order to get rid of the only non-even case we may write the sum in this way:
3 x = ( 2 + 1 ) x = 1 + n = 0 ∑ x − 1 ( n x ) 2 x − n
And, therefore, the sum may be written as 2 k θ with k ≥ 2 , since the binomial is even and 2 x − n is also even. Then we are able to write
3 x + 1 = 2 + n = 0 ∑ x − 1 ( n x ) 2 x − n = 2 + 2 k 1 θ 1 + ⋯ + 2 k x θ x = 2 × ( 1 + 2 k 1 − 1 θ 1 + ⋯ + 2 k x − 1 θ x )
But you may notice that 2 k 1 − 1 θ 1 + ⋯ + 2 k x − 1 θ x is still an even number, because each k i is ≥ 2 . Then 1 + 2 k 1 − 1 θ 1 + ⋯ + 2 k x − 1 θ x is odd, which implies that
2 × ( 1 + 2 k 1 − 1 θ 1 + ⋯ + 2 k x − 1 θ x ) = 3 x + 1
Has only ONE prime factor 2 . Therefore, by the first decomposition, each of the parentheses have only ONE prime factor 2 . We have 9 parentheses and a factor 2 3 multiplying them, so that we have a total of 1 2 prime factors 2 . There is our answer.
Just use difference of squares to obtain
( 3 − 1 ) ( 3 + 1 ) ( 3 2 + 1 ) ( 3 4 + 1 ) ( 3 8 + 1 ) ( 3 1 6 + 1 ) ( 3 3 2 + 1 ) ( 3 6 4 + 1 ) ( 3 1 2 8 + 1 ) ( 3 2 5 6 + 1 ) ( 3 5 1 2 + 1 )
And since 3 k + 1 ≡ 2 ( m o d 4 ) , for k = 2 x , x ≥ 1 , we cannot have a factor of 2 y , y ≥ 2 ; thus, we have
one 2 as a factor for 3 k + 1 , for k = 2 x , 1 ≤ x < 1 0 . This gives us 2 9
and
one 2 in 3 − 1
and
two 2 in 3 + 1 The largest value of n must be
n = 1 + 2 + 9 = 1 2
This can be solved by induction.
Consider the largest integer n such that 2 n divides 3 2 m − 1 . After trying small cases, we conjecture that n=m+2.
Base case: When m=1, 3 2 m − 1 = 3 2 − 1 = 8 , so n=3=1+2. -> Claim true for m=1.
Inductive step: Assume it is true for m=k.
3 2 k + 1 − 1 = ( 3 2 k − 1 ) ( 3 2 k + 1 )
We know that 3 2 k − 1 is divisible by at most k+2, so all we have to do is show that ( 3 2 k + 1 ) is divisible by at most 2.
Working modulus 4, we see that the powers of 3 have residues 3,1,3,1... . That is, even powers of 3 are all congruent to 1 mod 4. So ( 3 2 m + 1 ) is congruent to 2 mod 4, and so the highest power of 2 that divides it is 2.
Then we have shown that for m=k+1, the largest integer n is k+3. (Assuming the claim is true for m=k.) And since it is true for m=1, by induction, it is true for all positive integers m.
1 0 2 4 = 2 1 0 , so the answer is 10+2 = 1 2 .
3 1 0 2 4 − 1 = ( 3 5 1 2 − 1 ) ( 3 5 1 2 + 1 ) = ( 3 2 5 6 − 1 ) ( 3 2 5 6 + 1 ) ( 3 5 1 2 + 1 ) = . . . = ( 3 − 1 ) ( 3 + 1 ) ( 3 2 + 1 ) ( 3 4 + 1 ) . . . ( 3 5 1 2 + 1 ) The largest integer m such that 2 m divides 3 2 a − 1 for any positive integer a is 1. This is because 3 2 a − 1 is even but 3 2 a + 1 = 9 n + 1 is congruent to 1 a + 1 = 2 modulo 4. Also, ( 3 − 1 ) ( 3 + 1 ) = 2 3 . Thus the largest value of n such that 2 n divides 3 1 0 2 4 − 1 is 3 + 1 × 9 = 1 2
Sorry, there is a typo error. The two ( 3 2 a − 1 )s should be 3 2 a + 1 .
Nice!
The result simply follows from a particular case of Lifting the Exponent Lemma: if $$2\mid x-y$$ and n is even, then: $$\upsilon 2(x^n-y^n)=\upsilon 2(x-y)+\upsilon 2(x+y)+\upsilon 2(n)-1$$ In this case, we have: $$\upsilon 2(3^{1024}-1^{1024})=\upsilon 2(2)+\upsilon 2(4)+\upsilon 2(1024)-1=1+2+10-1=12$$ So 12 is the correct result.
Better than binomial expansion
I wrote a program...
a = (3**1024)- 1
for n in range (1,10000000):
b = 2 ** n
if(a % b == 0):
print n
We can write 3 1 0 2 4 − 1 as ( 3 5 1 2 ) 2 − 1 2
Thus, by using ( a 2 − b 2 ) = ( a + b ) ( a − b ) ,
3 1 0 2 4 − 1 becomes ( 3 5 1 2 + 1 ) ( 3 5 1 2 − 1 )
We can further decompose ( 3 5 1 2 − 1 ) as ( 3 2 5 6 + 1 ) ( 3 2 5 6 − 1 ) .
Again, we can decompose ( 3 2 5 6 − 1 ) as ( 3 1 2 8 + 1 ) ( 3 1 2 8 − 1 )
Iterating again and again in the same manner, finally we get
( 3 5 1 2 + 1 ) ( 3 2 5 6 + 1 ) ( 3 1 2 8 + 1 ) ( 3 6 4 + 1 ) ( 3 3 2 + 1 ) ( 3 1 6 + 1 ) ( 3 8 + 1 ) ( 3 4 + 1 ) ( 3 2 + 1 ) ( 3 2 − 1 )
This becomes
( 3 5 1 2 + 1 ) ( 3 2 5 6 + 1 ) ( 3 1 2 8 + 1 ) ( 3 6 4 + 1 ) ( 3 3 2 + 1 ) ( 3 1 6 + 1 ) ( 3 8 + 1 ) ( 3 4 + 1 ) ( 3 2 + 1 ) ( 2 3 )
Now, a number of the form ( 3 b + 1 ) where b is a power of 2 has only one 2 as a prime factor. I guessed this by expanding a few terms.
Therefore, there are a total of 1 2 two's in the prime factorization of this number, counting the 3 two's in 2 3 . And as a result, 2 1 2 will be the highest power of two that will divide 3 1 0 2 4 − 1 .
Thus, our answer is 1 2
3^1024 - 1 = (3^512 +1)(3^256 +1)(3^128 +1)(3^64 +1)(3^32 +1)(3^16 +1)(3^8 +1)(3^4 +1)(3^2 +1)(3+1)(3-1) therefore, in each factor except (3+1) we have even number ie.. one two is available and in (3+1),we have two two's =2^2 therefore the answer is (1x10)+(2x1)=10+2=12,the required answer
Trivial by lifting exponent Lemma.
This lemma states that v(x^n - y^n) = v(x-y)+v(x+y)+v(n)-1 if n is even and x,y are odd. Note that v(m) denotes the greatest k such that 2^k | m. Note also that k could be 0 for an odd m.
So applying the lemma we get the answer 1+2+10-1=12.
3^1024-1=(3^512+1)(3^256+1)(3^128+1)(3^64+1(3^32+1)(3^16+1)(3^8+1)(3^4+1)(3^2+1)(3+1)(3-1). The last digits of its factors from 3^512+1 to 3^4+1) are 2 and if they are divided by 2 their last digit will become 1 and it won't get further divided by 2 so 7 is the higest power 2 till factors 3^4+1 and the remaining factors yield 5 as the maximum power of 2. Hence, 12 = n. Note: I factorized the given number by using a^2-b^2 = (a+b)(a-b).
3 1 0 2 4 − 1 = ( 3 5 1 2 + 1 ) ( 3 5 1 2 − 1 ) using the identity a 2 − b 2 = ( a + b ) ( a − b ) So, further simplifying the equation, we get: 3 1 0 2 4 − 1 = ( 3 5 1 2 + 1 ) ( 3 2 5 6 + 1 ) ( 3 1 2 8 + 1 ) ( 3 6 4 + 1 ) ( 3 3 2 + 1 ) ( 3 1 6 + 1 ) ( 3 8 + 1 ) ( 3 4 + 1 ) ( 3 2 + 1 ) ( 3 + 1 ) ( 3 − 1 ) Note that, All factors of 3 1 0 2 4 − 1 are even numbers, and thus multiples of 2 To check if they are multiples of 2 2 we know, 3 = − 1 ( m o d 4 ) 3 e v e n p o w e r = 1 ( m o d 4 ) 3 e v e n p o w e r + 1 = 1 + 1 = 2 ( m o d 4 ) Again note that the factor 3 + 1 = 4 in which there are two powers of 2 Therefore n = 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 2 = 1 2 note that all factors have even powers of 3 except 3+1
Since all factors except 3+1 have 2 1 as factor...I tried to solve the problem from scratch, without the knowledge of any thoerem. Please feel free to mention any problem with my solution, as it looks a bit untidy. :)
Use Lifting The Exponent Lemma (LTE) that is . Let x and y be two odd integers and let n be an even positive integer. Then v 2 ( x n − y n ) = v 2 ( x − y ) + v 2 ( x + y ) + v 2 ( n ) − 1 . Substitute the stuff!
Or
First, factor 3 1 0 2 4 − 1 as ( 3 5 1 2 + 1 ) ( 3 2 5 6 + 1 ) . . . ( 3 + 1 ) ( 3 − 1 ) . The problem is asking for the number of factors of two in this product. Clearly, 3 − 1 = 2 has one factor of two, and 3 + 1 = 4 has two factors of two. Notice that 3 2 + 1 = 1 0 is congruent to 2 ( m o d 4 ) , so it only has one factor of two. The same applies to 3 3 + 1 = 2 8 . In fact, all the factors from ( 3 2 + 1 ) and above are congruent to 2 ( m o d 4 ) .
Assume 3 k + 1 ≡ 2 ( m o d 4 ) . Then: 3 k ≡ 1 ( m o d 4 ) 3 2 k ≡ 1 ( m o d 4 ) 3 2 k ≡ 2 ( m o d 4 )
In total, there are 1 + 2 + 9 = 1 2 factors of 2 , so n = 1 2 .
Sorry that's supposed to be 3 1 0 2 4 not 3 1 0 2 4 . I also meant 3 2 k and not 3 2 k .
Log in to reply
By the way, for modulus you should use \pmod (and without the parenthesis) and then it won't look weird.
Problem Loading...
Note Loading...
Set Loading...
Note that 2 1 0 = 1 0 2 4 and x 2 − y 2 = ( x + y ) ( x − y )
3 2 1 0 − 1
= ( 3 2 9 + 1 ) ( 3 2 9 − 1 )
= ( 3 2 9 + 1 ) ( 3 2 8 + 1 ) ( 3 2 8 − 1 )
= ( 3 2 9 + 1 ) ( 3 2 8 + 1 ) . . . . . . . . . ( 3 2 1 + 1 ) ( 3 2 0 + 1 ) ( 3 − 1 )
We know 3 2 n + 1 is divisible by 2 , but not divisible by 4 .
So ( 3 2 9 + 1 ) ( 3 2 8 + 1 ) . . . . . . . . . ( 3 2 1 + 1 ) is divisible by 2 9 and ( 3 2 0 + 1 ) ( 3 − 1 ) = 4 ∗ 2 = 8 is divisible by 2 3 .
Hence, 3 1 0 2 4 − 1 is divisible by 2 9 + 3 = 2 1 2 and n = 1 2 .