Euler’s Integrals and Multiple Sine Functions

Shin-ya Koyama
Nobushige Kurokawa

Running title. Euler’s Integrals

Abstract. We show that Euler’s famous integrals whose integrands contain the logarithm
of the sine function are expressed via multiple sine functions.
2000 Mathematics Subject Classification: 11M06

1 Introduction
Rπ
Euler studied the definite integrals 0
2
xn log(sin x)dx for n = 0 and 1. In [E1] (1769), he
proved the famous result
Z π
2 π
log(sin x)dx = − log 2, (1.1)
0 2
which is frequently explained as an example of tricky integrals in analysis courses. A bit
later, Euler [E2] (1772) stated that
 
Z π X∞ 2
2 1 1 π
x log(sin x)dx = 3
− log 2
0 2 n=1 n 4
n:odd

7 π2
= ζ(3) − log 2. (1.2)
16 8
Euler proved (1.1) by using
∞
X cos(2nx)
log(sin x) = − − log 2 (1.3)
n=1
n

for 0 < x < π. The actual integration is obvious since

Z π
2
cos(2nx)dx = 0
0

1
for n = 1, 2, 3, · · · . Hence, the original proof of Euler is not tricky contrary to the usual
explanation. In the case of (1.2), Euler primary wanted to calculate the value ζ(3). He
started from the divergent series expression

ζ(3) = −4π 2 ζ 0 (−2)

∞
X
= 4π 2 n2 log n
n=1

and by investigating it he reached (1.2). Thus his arguments are difficult to follow and
literally invalid. Moreover Euler did not prove the functional equation ζ(3) = −4π 2 ζ 0 (−2)
conjectured by himself in [E3]. We notice that when we use (1.3) we can give a secure
calculation for (1.2):
Z π
X∞ Z π
2 1 2 π2
x log(sin x)dx = − x cos(2nx)dx − log 2
0 n=1
n 0 8

with integration by parts


Z  1
π  − 2 if n is odd,
2
x cos(2nx)dx = 2n
0 

0 if n is even.

It might be remarkable that Euler missed this way.

In this paper we investigate definite integrals
Z x
θr−2 log(sin θ)dθ
0

for r = 2, 3, 4, · · · containing the Euler’s case x = π/2 from the point of view of multiple
sine functions. Let

xr−1
∞
Y  x nr−1
Sr (x) = e r−1 Pr
n=−∞
n
n6=0

∞ 
Y x  x (−1)r−1 nr−1
xr−1
= e r−1 Pr Pr −
n=1
n n

be the multiple sine function studied in [K1, K2, KK1, KK2, KOW, KW], where
 
u2 ur
Pr (u) = (1 − u) exp u + + ··· + .
2 r

2
For example
∞ 
Y  
x 1− x nn 2x
S2 (x) = e x e ,
n=1
1+ n

Y∞   n2 !
x2 x2 2
S3 (x) = e 2 1− 2 ex
n=1
n
and   3 !
Y∞
x3 1 − nx n 2n2 x+ 2 x3
S4 (x) = e 3
x e 3 .
n=1
1 + n

Then we show the following result.

Theorem 1 For 0 ≤ x < π and for r = 2, 3, 4, ..., we have

Z x x
r−2 xr−1 π r−1
θ log(sin θ)dθ = log(sin x) − log Sr .
0 r−1 r−1 π
In particular we have:

Theorem 2 For r = 2, 3, 4, ...,

Z π  
2
r−2 π r−1 1
θ log(sin θ)dθ = − log Sr .
0 r−1 2
Examples.
Z π  
2 1
log(sin θ)dθ = −π log S2
0 2
Y∞  n !
1 2n − 1
= −π log e 2 e , (1.4)
n=1
2n + 1

Z π  
2 π2 1
θ log(sin θ)dθ = − log S3
0 2 2
∞   n2 !!
π2 1
Y 1 1
= − log e 8 1− 2 e 4 , (1.5)
2 n=1
4n

Z π  
2
2 π3 1
θ log(sin θ)dθ = − log S4
0 3 2
Y∞  n3  !!
π3 1 2n − 1 2 1
= − log e 24 exp n + . (1.6)
3 n=1
2n + 1 12

3

√
1
 
We notice that we have S2 12 = 2 and S3 12 = 2 4 exp − 7ζ(3) 8π 2
from Euler’s results
(1.1) and (1.2)

combined with (1.4) and (1.5). We demonstrate a calculation of the special
value S4 12 from the product expression directly as follows.

Theorem 3    
1 1 9ζ(3)
S4 = 2 8 exp −
2 16π 2
and Z π
2 3π π3
θ2 log(sin θ)dθ = ζ(3) − log 2.
0 16 24
In our calculation a generalization of the Stirling formula is crucial.
Acknowledgement. We thank the referees for their suggestions to refine the presentation.

2 Multiple Sine Functions

To make this paper self-contained we prove some basic properties of multiple sine functions.
For general background we refer to [KK1, KK2, KOW, M].

Proposition 1 For r = 2, 3, 4, ..., Sr (x) is a meromorphic function in x ∈ C and it satisfies

Sr0 (x)
= πxr−1 cot(πx).
Sr (x)
Proof. The fact that Sr (x) is meromorphic function in x ∈ C (and its order as a meromor-
phic function being r) is seen from the product expression defining Sr (x). Let us calculate
the logarithmic derivative. From
∞   x   x (−1)r−1 nr−1
xr−1 Y
Sr (x) = e r−1 Pr Pr −
n=1
n n

we have
xr−1 X  ∞ x  x 
r−1 r−1
log Sr (x) = + n log Pr + (−1) log Pr −
r − 1 n=1 n n
xr−1
∞
X  x  x
r−1 r−1
= + n log 1 − + (−1) log 1 +
r − 1 n=1 n n
     
x 1 x 2 1 x r
+ + + ··· +
n 2 n r n
   2  r !!
r−1 −x 1 −x 1 −x
+(−1) + + ··· + .
n 2 n r n

4
Hence

X∞   
Sr0 (x) r−2 r−1 1 (−1)r−1 1 x xr−1
=x + n + + + + ··· + r
Sr (x) n=1
x−n x+n n n2 n
 r−1

r−1 1 x rx
+(−1) − + 2 + · · · + (−1) .
n n nr

Here
1 x xr−1 ( nx )r − 1
+ + ··· + r =
n n2 n x−n
and
1 x rx
r−1 (−1)r ( nx )r − 1
− + 2 + · · · + (−1) = .
n n nr x+n
Thus
X∞  x r 
Sr0 (x) r−2 r−1 (n) ( nx )r
= x + n −
Sr (x) n=1
x−n x+n
∞
X
r−2 2xr
= x +
n=1
x2 − n2
r−1
= πx cot(πx),

where we used ∞
1 X 2x
π cot(πx) = + .
x n=1 x2 − n2

Proposition 2 For 0 ≤ x < 1 and for r = 2, 3, 4, ...,

Z x
log Sr (x) = πtr−1 cot(πt)dt.
0

Proof. Since Sr (0) = 1, both sides are 0 at x = 0. Hence it is sufficient to remark that
the differentiations of both sides are πxr−1 cot(πx) from Proposition 1.

3 Euler’s Integrals
Using Proposition 2 we show Theorems 1 and 2.
Proof of Theorems 1 and 2: By integration by parts in
Z x
log Sr (x) = πtr−1 cot(πt)dt
0

5
we have
Z x
 r−1
x
log Sr (x) = t (r − 1)tr−2 log(sin πt)dt
log(sin πt) 0 −
0
Z x
r−1
= x log(sin πx) − (r − 1) tr−2 log(sin πt)dt.
0

Hence changing the variable to θ = πt in the integral, we have

Z
r−1 r − 1 πx r−2
log Sr (x) = x log(sin πx) − r−1 θ log(sin θ)dθ.
π 0

This gives Theorem 1. Then, letting x = 1/2 we have Theorem 2.

Examples:  
Z π
4 π 1
log(sin θ)dθ = − log 2 − π log S2 ,
0 8 4
Z π √  
3 π 3 1
log(sin θ)dθ = log − π log S2 ,
0 3 2 3
Z π  
4 π2 π2 1
θ log(sin θ)dθ = − log 2 − log S3 ,
0 64 2 4
Z π √  
3 π2 3 π2 1
θ log(sin θ)dθ = log − log S3 .
0 18 2 2 3

4 A calculation of the special value

Proof of Theorem 3: Since
  Y∞  n3  !
1 1 2n − 1 1
S4 = e 24 exp n2 + ,
2 n=1
2n + 1 12

we put
N
Y   n3  !
1 2n − 1 1
AN = e 24 exp n2 +
n=1
2n + 1 12
  Y N
1 2 2 N 3 3
= exp + (1 + · · · + N ) + × (2n − 1)n −(n−1)
24 12 n=1
3
×(2N + 1)−N

6
and show    
1 9 0 1 9ζ(3)
lim AN = 2 exp 8 ζ (−2) = 2 exp −
8 .
N →∞ 4 16π 2
Then the value of the integral follows from (1.6). We use the Stirling formula
N
Y √ 1
N! = n∼ 2πN N + 2 e−N
n=1

and its generalization

N
Y N3 2 N3
2 + N2 + N N
nn ∼ exp (−ζ 0 (−2)) N 3 6 e− 9
+ 12
,
n=1

which follow from the Euler-Maclaurin summation formula for ζ 0 (s):

N
X N 1−s log N N 1−s
ζ 0 (s) = lim − n−s log n + −
N →∞
n=1
1−s (1 − s)2

1 s 1
+ N −s log N − N −s−1 log N + N −s−1
2 12 12

valid in Re(s) > −3. We refer to Hardy [H] Chap. XIII for the Euler-Maclaurin summation
formula and its applications. Then, letting s = 0 and −2 we see

XN   !
√ 1
log 2π = −ζ 0 (0) = lim log n − N+ log N − N
N →∞
n=1
2
 
N!
= lim log 1
N →∞ N N + 2 e−N
and
N
X   !
N3 N2 N N3 N
−ζ 0 (−2) = lim n2 log n − + + log N − +
N →∞
n=1
3 2 6 9 12
QN 2
!
n=1 nn
= lim log N3 2 N3
.
N →∞ + N2 + N N
N 3 6 e− 9
+ 12

7
Using
N
Y N
Y
n3 −(n−1)3 2 −3n+1
(2n − 1) = (2n − 1)3n
n=1 n=1
N
! 34 N
! 14
Y Y
(2n−1)2
= (2n − 1) × (2n − 1)
n=1 n=1
Q2N ! 43 Q2N ! 41
n2
n=1 n n=1 n
= QN × QN
(2n)2
n=1 (2n) n=1 (2n)

and the (generalized) Stirling formulas

2N
Y 8N 3 8N 3
2 0 +2N 2 + N +N
nn ∼ e−ζ (−2) (2N ) 3 3 e− 9 6 ,
n=1

N N
!4
Y 2 2 +···+N 2 )
Y 2
(2n)(2n) = 24(1 nn
n=1 n=1
2
 N3 2 N3
4
N (N +1)(2N +1) 0 + N2 + N N
∼ 2 3 e−ζ (−2) N 3 6 e− 9
+ 12
,

2N
Y √ 1
n∼ 2π(2N )2N + 2 e−2N ,
n=1
N
Y N
Y
N
√ 1
(2n) = 2 n∼ 2πN N + 2 e−N 2N ,
n=1 n=1

we have
   
1 N3 N2 N N
AN ∼ exp + + + +
24 3 2 6 12
9 0 3 N 3 N N3 N
×e 4 ζ (−2) N N − 4 2N − 4 e− 3 − 8
N N 1 N 3
×N 4 2 4 + 8 e− 4 × (2N + 1)−N .

8
Hence, combining with
 −N 3
−N 3 −N 3 1
(2N + 1) = (2N ) 1+
2N
  
−N 3 3 1
= (2N ) exp −N log 1 +
2N
 2  3 !!
3 1 1 1 1 1
∼ (2N )−N exp −N 3 − +
2N 2 2N 3 2N
 
3 N2 N 1
= (2N )−N exp − + − ,
2 8 24

we obtain the desired result

 
1 9 0
lim AN = 2 exp
8 ζ (−2) .
N →∞ 4

Remarks
√
1. The fact S2 ( 12 ) = 2 is also proved as Theorem 3 and we get (1.1) again from (1.4).
In fact:
  Y∞  1 n 
1 1 1 − 2n
S2 = e 2
1 e
2 n=1
1 + 2n
YN  n 
1 2n − 1
= lim e 2 e
N →∞
n=1
2n + 1
 1  2  3  N !
1 1 3 5 2N − 1
= lim e 2 ··· eN
N →∞ 3 5 7 2N + 1
 
1 3 · 5 · · · (2N − 1)
N
= lim e 2 e
N →∞ (2N + 1)N
 
1 (2N )! N
= lim e 2 1 Ne
N →∞ N !22N N N (1 + 2N )
√
= 2

by the (usual) Stirling formula.

9
 2. The case of S3 (1/2) is similar by using the generalized Stirling’s formula:
  Y∞   n2 !
1 1 1 1
S3 = e8 1− 2 e4
2 n=1
4n
YN   n2 !
1 1 1
= lim e 8 1− 2 e4
N →∞
n=1
4n
YN  n2
N
+ 1 (2n − 1)(2n + 1)
= lim e 4 8
N →∞
n=1
(2n)2
Q 1/2
2N
n n2  1 2
N
+ 1 n=1 (2N )! 2 (2N + 1)N
= lim e 4 8 Q 4 × × 4(12 +···+N 2 )
N →∞ N n 2 2N N ! 2
n=1 n
 
1 7 0
= 2 4 exp ζ (−2) .
2

Thus we obtain Euler’s formula (1.2) via (1.5).

√
3. Except for S2 (1/2) = 2 we do not know the algebraicity of Sr (1/2) for r ≥ 2. In
fact we cannot deny even the optimistic expectation Sr (1/2) ∈ 2Q .

References
[E1] L. Euler “De summis serierum numeros Bernoullianos involventium” Novi commentarii
academiae scientiarum Petropolitanae 14 (1769) 129-167 [Opera Omnia I-15, pp.91-
130].

[E2] L. Euler “Exercitationes analyticae” Novi commentarii academiae scientiarum

Petropolitanae 17 (1772) 173-204 [Opera Omnia I-15, pp.131-167].

[E3] L. Euler “Remarques sur un beau rapport entre les series des puissances tant directes
que reciproques” Mémoires de l’académie des sciences de Berlin 17 (1761) 83-106 (Lu
en 1749) [Opera Omnia I-15, pp. 70-90].

[H] G.H. Hardy “Divergent Series” Oxford Univ. Press 1949.

[K1] N. Kurokawa “Multiple sine functions and Selberg zeta functions” Proc. Japan Acad.
67A (1991) 61-64.

[K2] N. Kurokawa “Multiple zeta functions: an example” Adv. Stud. in Pure Math. 21
(1992) 219-226.

10
[KK1] N. Kurokawa and S. Koyama “Multiple sine functions” Forum Math. 15 (2003) 839-
876.

[KK2] S. Koyama and N. Kurokawa “Kummer’s formula for multiple gamma functions” J.
Ramanujan Math.Soc. 18 (2003) 87-107.

[KOW] N. Kurokawa, H. Ochiai and M. Wakayama “Multiple trigonometry and zeta func-
tions” J. Ramanujan Math. Soc. 17 (2002) 101-113.

[KW] N. Kurokawa and M. Wakayama “On ζ(3)” J. Ramanujan Math. Soc. 16 (2001) 205-
214.

[M] Yu. I. Manin “Lectures on zeta functions and motives (according to Deninger and
Kurokawa)” Asterisque 228 (1995) 121-163.

2-5-27 Hayabuchi, Tsuzuki-ku, Yokohama 224-0025, Japan

koyama@tmtv.ne.jp
Department of Mathematics, Tokyo Institute of Technology, 152-8551, Japan
kurokawa@math.titech.ac.jp

11