μ ( n ) = 1 ∑ n 2 1 = B π C A
Let μ ( n ) denote the möbius function , the sum is taken over all positive integers n such that μ ( n ) = 1 , with coprime positive integers A and B . Find A + B + C .
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.
First, rewrite the given sum s . If we add μ ( n ) and μ 2 ( n ) , the terms where μ ( n ) = − 1 cancel out. However the terms where μ ( n ) = 1 we count double. Sums and products over p k denote sums and products over all primes: 2 s = 2 μ ( n ) = 1 ∑ n 2 1 = μ ( n ) = 0 ∑ n 2 μ 2 ( n ) + μ ( n ) ∣ ∣ ∣ ∣ ∣ ζ ( k ) = n ∈ N ∑ n k 1 = p 1 ∏ 1 − p 1 k 1 1 , k ∈ N , k > 1
Both sums on the left converge absolutely, so we may reorder them in any way we want. The first reordering combines all n with the same number of prime factors. The second reordering takes only the first few summands of the first few sums to convert them into a growing product over all primes that resembles the zeta-function above: μ ( n ) = 0 ∑ n 2 μ 2 ( n ) μ ( n ) = 0 ∑ n 2 μ ( n ) = 1 + p 1 ∑ p 1 2 1 + p 1 < p 2 ∑ p 1 2 p 2 2 1 + p 1 < p 2 < p 3 ∑ p 1 2 p 2 2 p 3 2 1 + … = p 1 ∏ ( 1 + p 1 2 1 ) = p 1 ∏ 1 − p 1 2 1 1 − p 1 4 1 = ζ ( 4 ) ζ ( 2 ) = 1 − p 1 ∑ p 1 2 1 + p 1 < p 2 ∑ p 1 2 p 2 2 1 − p 1 < p 2 < p 3 ∑ p 1 2 p 2 2 p 3 2 1 ± … = p 1 ∏ ( 1 − p 1 2 1 ) = ζ ( 2 ) 1
Adding both together, the red terms cancel out and we are left with 2 s . We remember ζ ( 2 ) = 6 π 2 , ζ ( 4 ) = 9 0 π 4 to get 2 s = π 4 ⋅ 6 π 2 ⋅ 9 0 + π 2 6 = π 2 2 1 ⇒ s = 2 π 2 2 1 , A + B + C = 2 1 + 2 + 2 = 2 5
Problem Loading...
Note Loading...
Set Loading...
We can rewrite this as n = 1 ∑ ∞ n 2 f ( n ) where f ( n ) = ⎩ ⎪ ⎨ ⎪ ⎧ 1 0 0 , μ ( n ) = 1 , μ ( n ) = 0 , μ ( n ) = − 1 . for it to be zero when μ is, μ must be one of its factor. For it to be zero when μ = − 1 → μ + 1 = 0 , μ + 1 must be its factor. But when μ = 1 , the second part will be 2, so we need to divide by 2 to make it one. We find f ( n ) = 2 μ ( n ) ( μ ( n ) + 1 ) = 2 ∣ μ ( n ) ∣ + 2 μ ( n ) Hence the summation becomes 2 1 ( n = 1 ∑ ∞ n 2 ∣ μ ( n ) ∣ + n = 1 ∑ ∞ n 2 μ ( n ) ) Using dirichlet series and dirichlet convolution , n = 1 ∑ ∞ n s μ ( n ) = ζ ( s ) 1 ∣ μ ∣ ∗ λ = I ⟹ n = 1 ∑ ∞ n s ∣ μ ( n ) ∣ = ∑ n = 1 ∞ n s λ ( n ) 1 = ζ ( 2 s ) ζ ( s )
From the riemann zeta function , we know ζ ( 2 ) = 6 π 2 , ζ ( 4 ) = 9 0 π 4 . Substituting in these values gives 2 π 2 2 1 , so A + B + C = 2 1 + 2 + 2 = 2 5 .