euler kronecker constants from ramanujan to ihara
play

Euler-Kronecker constants: from Ramanujan to Ihara Pieter Moree - PowerPoint PPT Presentation

Euler-Kronecker constants: from Ramanujan to Ihara Pieter Moree (MPIM, Bonn) Amsterdam, CWI December 2, 2011 Workshop Herman te Riele (Partly) joint work with Florian Luca (Morelia, Mexico) Kevin Ford (Urbana-Champaign, Illinois) Values of


  1. Euler-Kronecker constants: from Ramanujan to Ihara Pieter Moree (MPIM, Bonn) Amsterdam, CWI December 2, 2011 Workshop Herman te Riele

  2. (Partly) joint work with Florian Luca (Morelia, Mexico) Kevin Ford (Urbana-Champaign, Illinois) Values of the Euler phi-function not divisible by a given odd prime, and the distribution of Euler-Kronecker constants for cyclotomic fields, arXiv:1108.3805.

  3. Definition K , number field. 1 � ζ K ( s ) = ( N a ) s , Re ( s ) > 1 . a

  4. Definition K , number field. 1 � ζ K ( s ) = ( N a ) s , Re ( s ) > 1 . a Laurent series: ζ K ( s ) = c − 1 s − 1 + c 0 + O ( s − 1 ) . c 0 Euler-Kronecker constant of K : EK K := c − 1

  5. Definition K , number field. 1 � ζ K ( s ) = ( N a ) s , Re ( s ) > 1 . a Laurent series: ζ K ( s ) = c − 1 s − 1 + c 0 + O ( s − 1 ) . c 0 Euler-Kronecker constant of K : EK K := c − 1 � ζ ′ K ( s ) 1 � lim ζ K ( s ) + = EK K , s − 1 s → 1 EK K is constant in logarithmic derivative of ζ K ( s ) at s = 1.

  6. Definition K , number field. 1 � ζ K ( s ) = ( N a ) s , Re ( s ) > 1 . a Laurent series: ζ K ( s ) = c − 1 s − 1 + c 0 + O ( s − 1 ) . c 0 Euler-Kronecker constant of K : EK K := c − 1 � ζ ′ K ( s ) 1 � lim ζ K ( s ) + = EK K , s − 1 s → 1 EK K is constant in logarithmic derivative of ζ K ( s ) at s = 1. Example. ζ ( s ) = � n − s = 1 / ( s − 1 ) + γ + O ( s − 1 ) .

  7. Definition K , number field. 1 � ζ K ( s ) = ( N a ) s , Re ( s ) > 1 . a Laurent series: ζ K ( s ) = c − 1 s − 1 + c 0 + O ( s − 1 ) . c 0 Euler-Kronecker constant of K : EK K := c − 1 � ζ ′ K ( s ) 1 � lim ζ K ( s ) + = EK K , s − 1 s → 1 EK K is constant in logarithmic derivative of ζ K ( s ) at s = 1. Example. ζ ( s ) = � n − s = 1 / ( s − 1 ) + γ + O ( s − 1 ) . EK Q = γ/ 1 = γ = 0 . 577 . . . ... Euler-Mascheroni constant

  8. Historical background Sums of two squares Landau (1908) x � B ( x ) = 1 ∼ K . � log x n ≤ x , n = a 2 + b 2

  9. Historical background Sums of two squares Landau (1908) x � B ( x ) = 1 ∼ K . � log x n ≤ x , n = a 2 + b 2 Ramanujan (1913) � x dt x � � B ( x ) = K + O , log r x � log t 2 where r > 0 is arbitrary.

  10. Historical background Sums of two squares Landau (1908) x � B ( x ) = 1 ∼ K . � log x n ≤ x , n = a 2 + b 2 Ramanujan (1913) � x dt x � � B ( x ) = K + O , log r x � log t 2 where r > 0 is arbitrary. K = 0 . 764223653 ... : Landau-Ramanujan constant.

  11. Historical background Sums of two squares Landau (1908) x � B ( x ) = 1 ∼ K . � log x n ≤ x , n = a 2 + b 2 Ramanujan (1913) � x dt x � � B ( x ) = K + O , log r x � log t 2 where r > 0 is arbitrary. K = 0 . 764223653 ... : Landau-Ramanujan constant. Shanks (1964): Ramanujan’s claim is false for every r > 3 / 2.

  12. Non-divisibility of Ramanujan’s τ ∞ ∞ ( 1 − q m ) 24 = � � τ ( n ) q n . ∆ := q m = 1 n = 1 After setting q = e 2 π iz , the function ∆( z ) is the unique normalized cusp form of weight 12 for the full modular group SL 2 ( Z ) .

  13. Non-divisibility of Ramanujan’s τ ∞ ∞ ( 1 − q m ) 24 = � � τ ( n ) q n . ∆ := q m = 1 n = 1 After setting q = e 2 π iz , the function ∆( z ) is the unique normalized cusp form of weight 12 for the full modular group SL 2 ( Z ) . Fix a prime q ∈ { 3 , 5 , 7 , 23 , 691 } .

  14. Non-divisibility of Ramanujan’s τ ∞ ∞ ( 1 − q m ) 24 = � � τ ( n ) q n . ∆ := q m = 1 n = 1 After setting q = e 2 π iz , the function ∆( z ) is the unique normalized cusp form of weight 12 for the full modular group SL 2 ( Z ) . Fix a prime q ∈ { 3 , 5 , 7 , 23 , 691 } . For these primes τ ( n ) satisfies an easy congruence, e.g., : d 11 ( mod 691 ) . � τ ( n ) ≡ d | n Put t n = 1 if q ∤ τ ( n ) and t n = 0 otherwise.

  15. A further claim of Ramanujan Ramanujan in last letter to Hardy (1920):

  16. A further claim of Ramanujan Ramanujan in last letter to Hardy (1920): “It is easy to prove by quite elementary methods that � n k = 1 t k = o ( n ) .

  17. A further claim of Ramanujan Ramanujan in last letter to Hardy (1920): “It is easy to prove by quite elementary methods that � n k = 1 t k = o ( n ) . It can be shown by transcendental methods that n C q n � t k ∼ ; (1) log δ q n k = 1 and � n n dx n � � � t k = C q + O , (2) log r n log δ q x 2 k = 1 where r is any positive number’.

  18. A further claim of Ramanujan Ramanujan in last letter to Hardy (1920): “It is easy to prove by quite elementary methods that � n k = 1 t k = o ( n ) . It can be shown by transcendental methods that n C q n � t k ∼ ; (1) log δ q n k = 1 and � n n dx n � � � t k = C q + O , (2) log r n log δ q x 2 k = 1 where r is any positive number’. Rushforth, Rankin: Estimate (1) holds true.

  19. A further claim of Ramanujan Ramanujan in last letter to Hardy (1920): “It is easy to prove by quite elementary methods that � n k = 1 t k = o ( n ) . It can be shown by transcendental methods that n C q n � t k ∼ ; (1) log δ q n k = 1 and � n n dx n � � � t k = C q + O , (2) log r n log δ q x 2 k = 1 where r is any positive number’. Rushforth, Rankin: Estimate (1) holds true. M. (2004): All estimates (2) are false for r > 1 + δ q

  20. Non-divisibility of Euler’s ϕ -function (Spearman-Williams, 2006). Put � E q ( x ) = 1 . n ≤ x , q ∤ ϕ ( n ) Question � x x dt E q ( x ) ∼ c q or E q ( x ) ∼ c q ? log 1 / ( q − 1 ) x log 1 / ( q − 1 ) t 2 That is, is the Landau approximation or Ramanujan approximation better?

  21. Non-divisibility of Euler’s ϕ -function (Spearman-Williams, 2006). Put � E q ( x ) = 1 . n ≤ x , q ∤ ϕ ( n ) Question � x x dt E q ( x ) ∼ c q or E q ( x ) ∼ c q ? log 1 / ( q − 1 ) x log 1 / ( q − 1 ) t 2 That is, is the Landau approximation or Ramanujan approximation better? Assume ( q , n ) = 1. We have q ∤ ϕ ( n ) iff n does not have a prime divisor p that splits completely in Q ( ζ q ) .

  22. Euler-Kronecker constants of multiplicative sets We say that S is multiplicative if m and n are coprime integers then mn is in S iff both m and n are in S .

  23. Euler-Kronecker constants of multiplicative sets We say that S is multiplicative if m and n are coprime integers then mn is in S iff both m and n are in S . Common example is where S is a multiplicative semigroup generated by q i , i = 1 , 2 , . . . , with every q i a prime power and ( q i , q j ) = 1.

  24. Euler-Kronecker constants of multiplicative sets We say that S is multiplicative if m and n are coprime integers then mn is in S iff both m and n are in S . Common example is where S is a multiplicative semigroup generated by q i , i = 1 , 2 , . . . , with every q i a prime power and ( q i , q j ) = 1. Example I . n = X 2 + Y 2 . Example II . If q is a prime and f a multiplicative function, then { n : q ∤ f ( n ) } is multplicative.

  25. Euler-Kronecker constants of multiplicative sets We say that S is multiplicative if m and n are coprime integers then mn is in S iff both m and n are in S . Common example is where S is a multiplicative semigroup generated by q i , i = 1 , 2 , . . . , with every q i a prime power and ( q i , q j ) = 1. Example I . n = X 2 + Y 2 . Example II . If q is a prime and f a multiplicative function, then { n : q ∤ f ( n ) } is multplicative. If ( m , n ) = 1, then q ∤ f ( mn ) ⇐ ⇒ q ∤ f ( m ) f ( n ) ⇐ ⇒ q ∤ f ( n ) and q ∤ f ( m )

  26. Euler-Kronecker constant of a multiplicative set Assumption. There are some fixed δ, ρ > 0 such that asymptotically x � � π S ( x ) = δπ ( x ) + O . log 2 + ρ x

  27. Euler-Kronecker constant of a multiplicative set Assumption. There are some fixed δ, ρ > 0 such that asymptotically x � � π S ( x ) = δπ ( x ) + O . log 2 + ρ x We put ∞ � n − s . L S ( s ) := n = 1 , n ∈ S Can show that, Euler-Kronecker constant � L ′ S ( s ) δ � γ S := lim L S ( s ) + s − 1 s → 1 + 0 exists.

  28. The second order term and γ S We have 1 +( 1 + o ( 1 )) C 1 ( S ) S ( x ) = C 0 ( S ) x log δ − 1 x � � , as x → ∞ , log x where C 1 ( S ) = ( 1 − δ )( 1 − γ S ) .

  29. The second order term and γ S We have 1 +( 1 + o ( 1 )) C 1 ( S ) S ( x ) = C 0 ( S ) x log δ − 1 x � � , as x → ∞ , log x where C 1 ( S ) = ( 1 − δ )( 1 − γ S ) . Theorem . Suppose that δ < 1 . If γ S < 1 / 2 , the Ramanujan approximation is asymptotically better than the Landau one. If γ S > 1 / 2 it is the other way around.

  30. The second order term and γ S We have 1 +( 1 + o ( 1 )) C 1 ( S ) S ( x ) = C 0 ( S ) x log δ − 1 x � � , as x → ∞ , log x where C 1 ( S ) = ( 1 − δ )( 1 − γ S ) . Theorem . Suppose that δ < 1 . If γ S < 1 / 2 , the Ramanujan approximation is asymptotically better than the Landau one. If γ S > 1 / 2 it is the other way around. Follows on noting that by partial integration we have � x 1 + 1 − δ 1 log δ − 1 dt = x log δ − 1 x � � �� log x + O . log 2 x 2 A Ramanujan type claim, if true, implies γ S = 0.

  31. Landau versus Ramanujan for q ∤ ϕ Theorem . (M., 2006, unpublished). Assume ERH. For q ≤ 67 we have γ ϕ ; q < 1 / 2 and Ramanujan’s approximation is better.

Recommend


More recommend