m → ∞ lim ∫ 0 ∞ e − m π x sin ( π x ) ln ( e − π m x + 1 ) d x ∫ 0 ∞ e − π m x sin ( π x ) ln ( e − m π x + 1 ) d x = T + 1 2 T ⋅ ζ ( 1 + T ) ln T Find the value of T .
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.
A briiliant solution, thanks
That was pretty hard
Problem Loading...
Note Loading...
Set Loading...
Note first that ∫ 0 ∞ e − a x sin b x d x = a 2 + b 2 b a , b > 0 and so F ( m ) = ∫ 0 ∞ e − π m x sin ( π x ) ln ( 1 + e − m π x ) d x = k = 1 ∑ ∞ k ( − 1 ) k + 1 ∫ 0 ∞ e − ( m π + m k π ) x sin ( π x ) d x = k = 1 ∑ ∞ k ( − 1 ) k + 1 π 2 ( m + m k ) 2 + π 2 π = π m 2 k = 1 ∑ ∞ k [ ( k + m 2 ) 2 + m 2 ] ( − 1 ) k + 1 = π m 2 G ( m ) where G ( m ) = k = 1 ∑ ∞ k [ ( k + m 2 ) 2 + m 2 ] ( − 1 ) k + 1 m > 0 Since ∑ k = 1 ∞ k − 3 is convergent, it is easy to deduce that m → ∞ lim G ( m − 1 ) = m → 0 lim G ( m ) = k = 1 ∑ ∞ k 3 ( − 1 ) k + 1 = k = 1 ∑ ∞ k 3 1 − 2 k = 1 ∑ ∞ ( 2 k ) 3 1 = 4 3 ζ ( 3 ) Now m 4 G ( m ) = m 4 k = 1 ∑ ∞ k [ ( k + m 2 ) 2 + m 2 ] ( − 1 ) k + 1 = k = 1 ∑ ∞ k [ ( 1 + k m − 2 ) 2 + m − 2 ] ( − 1 ) k + 1 and a simplistic argument indicates that m → ∞ lim m 4 G ( m ) = k = 1 ∑ ∞ k ( − 1 ) k + 1 = ln 2 Given that the series for m 4 G ( m ) is not uniformly convergent in m , this deduction is somewhat dodgy. However, a more detailed analysis shows that m 4 G ( m ) = 2 i ( m 2 + 1 ) m 2 [ ( m + i ) [ ψ ( 1 + m 2 − i m ) − ψ ( 1 + 2 1 m 2 − 2 1 i m ) ] − ( m − i ) [ ψ ( 1 + m 2 + i m ) − ψ ( 1 + 2 1 m 2 + 2 1 i m ) ] ] where ψ is the digamma function, and the fact that ψ ( z ) = ln z + O ( z − 1 ) as ∣ z ∣ → ∞ uniformly for ∣ A r g ( z ) ∣ ≤ π − δ for any δ > 0 tells us indeed that m → ∞ lim m 4 G ( m ) = ln 2 Thus m → ∞ lim F ( m − 1 ) F ( m ) = m → ∞ lim G ( m − 1 ) m 4 G ( m ) = 4 3 ζ ( 3 ) ln 2 = 3 ζ ( 3 ) 4 ln 2 which makes T = 2 .