On some equations concerning quantum electrodynamics coupled to quantum gravity, the

gravitational contributions to the gauge couplings and quantum effects in the theory of
gravitation: mathematical connections with some sector of String Theory and Number Theory

Michele Nardelli 1, 2

1
Dipartimento di Scienze della Terra
Università degli Studi di Napoli Federico II, Largo S. Marcellino, 10
80138 Napoli, Italy

2
Dipartimento di Matematica ed Applicazioni “R. Caccioppoli”
Università degli Studi di Napoli “Federico II” – Polo delle Scienze e delle Tecnologie
Monte S. Angelo, Via Cintia (Fuorigrotta), 80126 Napoli, Italy

Abstract

This paper is principally a review, a thesis, of principal results obtained from various authoritative
theoretical physicists and mathematicians in some sectors of theoretical physics and mathematics.
In this paper in the Section 1, we have described some equations concerning the quantum
electrodynamics coupled to quantum gravity. In the Section 2, we have described some equations
concerning the gravitational contributions to the running of gauge couplings. In the Section 3, we
have described some equations concerning some quantum effects in the theory of gravitation. In the
Section 4, we have described some equations concerning the supersymmetric Yang-Mills theory
applied in string theory and some lemmas and equations concerning various gauge fields in any
non-trivial quantum field theory for the pure Yang-Mills Lagrangian. Furthermore, in conclusion, in
the Section 5, we have described various possible mathematical connections between the argument
above mentioned and some sectors of Number Theory and String Theory, principally with some
equations concerning the Ramanujan’s modular equations that are related to the physical vibrations
of the bosonic strings and of the superstrings, some Ramanujan’s identities concerning π and the
zeta strings.

1. On some equations concerning the quantum electrodynamics coupled to quantum

gravity. [1]

The key equations that govern the behaviour of the coupling constants in quantum field theory are
the renormalisation group Callan-Symanzik equations. If we let g denote a generic coupling
constant, then the value of g at energy scale E , the running coupling constant g (E ) , is determined
by
dg (E )
E = β (E , g ) , (1)
dE

1
where β (E, g ) is the renormalisation group β -function. Asymptotic freedom is signaled by
g (E ) → 0 as E → ∞ , requiring β < 0 in this limit.
With regard the Einstein gravity with a cosmological constant coupled to quantum electrodynamics
in four spacetime dimensions, a standard calculation shows that the effective action to one-loop
order is given by

Γ (1) =
1
2
(
ln det ∆i j − ln det Qαβ − ln det iγ µ ∂ µ + eγ µ Aµ − im . ) (2)

The last term (with Aµ the background gauge field) is the result of performing a functional integral
over the Dirac field. The middle term is the contribution from the ghost fields required to remove
the unphysical degrees of freedom of the gravity and electromagnetic fields. The first term is the
result of integrating over the spacetime metric and electromagnetic fields. For operator ∆i j the heat
kernel K ij ( x, x';τ ) is defined by

∂ i
− K j ( x, x' ;τ ) = ∆ik K kj ( x, x' ;τ ) (3)
∂τ

with boundary condition K ij ( x, x';τ = 0) = δ ijδ ( x, x') . τ is called the proper time. The Green
function G ij ( x, x') for the operator ∆i j is

∆ik G kj ( x, x') = δ ijδ ( x, x') . (4)

It follows that the Green function and heat kernel are related by

∞
G ( x, x') = ∫ dτK ij ( x, x' ;τ ) . (5)
i
j
0

The importance of the heat kernel for quantum field theory arises from the existence of an
asymptotic expansion as τ → 0 :

∞
K ij (x, x;τ ) ≈ (4πτ ) ∑τ Er j ( x )
−n / 2 r i
(6)
r =0

where n is the spacetime dimension (chosen as 4 here) and the heat kernel coefficients Er j ( x )
i

depend only locally on the details of coefficients of the differential operator ∆i j . The divergent part
of the effective action, as well as the Green function, can be related to the heat kernel coefficients.
Formally
∞
dτ
L∆ = ln det ∆i j = − ∫ d n x ∫ trK ij ( x, x;τ ) . (7)
1 1
2 2 0
τ

The one-loop effective action (2) is then given by

Γ (1) = L∆ − 2 LQ − 2 LDirac . (8)

2
As with the Green function (4) the divergent part of (7) comes from the τ ≈ 0 limit of the proper
time integral. The divergent part of L∆ is

1 
divp (L∆ ) = −
1
2 ∫
d 4 x  Ec4trE0 + Ec2trE1 + trE2 ln Ec2  . (9)
32π 2 

The lower limit on the proper time integration can be kept as τ = 0 and the divergent part of the
effective action L∆ contains a simple pole as ε → 0 given by

divp (L∆ ) =
1
∫d
4
xtrE2 . (10)
16π 2ε

The general form of ∆i j is

( )
∆i j = Aαβ j ∂α ∂ β + Bα j ∂α + (C ) j
i
( ) i i
(11)

( ) ( )
for coefficients Aαβ j , Bα
i i
j and (C ) j that depend on the spacetime coordinates through the
i

background field. Normal coordinates are introduced at x' with x µ = x'µ + y µ and all of the
coefficients in (11) are expanded about y µ = 0 . This gives

(A ) = ( A ) + ∑ (A )y
∞
αβ i αβ i αβ µ1
... y µ n
i
j 0 j µ1 ...µ n j (12)
n =1

with similar expansions for Bα ( ) i

j and (C ) j . The Green function is Fourier expanded as usual,
i

G ij ( x, x') = ∫ d n peip ⋅ y G ij ( p ) ,
1
(13)
(2π ) n

except that the Fourier coefficient G ij ( p ) can also have a dependence on the origin of the coordinate
system x' that is not indicated explicitly. If

G ij ( p ) = G0 j ( p ) + G1 j ( p ) + G2 j ( p ) + ...
i i i
(14)

where Gr j ( p ) is of order p −2 − r as p → ∞ it is easy to see that to calculate the pole part of G ij (x, x )
i

as n → 4 only terms up to and including G2 j ( p ) are needed.

i

The gravity and gauge field contributions result in

3 3 1 3 1 3 1 1 
tr E1 = κ 2  − ω + ω 2 + ξ − ωξ + ωζ − ωξζ + ω 2ζ  F 2 + 12 + 8ξ 2 + 3v + vζ 2 Λ . (15) ( )
8 4 8 8 2 8 32 32 

The overall result for the quadratically divergent part of the complete one-loop effective action (8)
that involves F 2 is

3
 (1) κ 2 Ec2  3 3 5 3 1 3 1 1 
Γquad =− 2 
− ω + ω 2 + ξ − ωξ + ωζ − ωξζ + ω 2ζ  ∫ d 4 xF 2 . (16)
32π  8 4 8 8 2 8 32 32 
(1) 1 κ Ec  3 3
2 2
5 3 1 3 1 1 
Γquad =− 2 
− ω + ω 2 + ξ − ωξ + ωζ − ωξζ + ω 2ζ  ∫ d 4 xF 2 . (16b)
16 2π  8 4 8 8 2 8 32 32 

If ξ → 0 , ζ → 0 , ω → 1 are taken to obtain the gauge condition independent result, the non-zero
result
(1) κ 2 Ec2
2 ∫
Γquad =− d 4 xF 2 (17)
128π

is found.
The net result for the divergent part of the effective action that involves F 2 and therefore
contributes to charge renormalization is

 κ 2 Ec2 3κ 2 Λ
( )
divp Γ (1)
= −
 − ln Ec +
2 e2 
ln Ec2  ∫ d 4 xF 2 . (18)
 128π 256π 48π
2 2 2

1  κ Ec 3κ Λ
( ) 
2 2 2 2
e
divp Γ (1) =  − − ln Ec2 + 2 ln Ec2  ∫ d 4 xF 2 . (18b)
16  8π 2
16π 2
3π 

From this the renormalization group function in (1) that governs the running electric charge to be
calculated to be
e3 κ2  2 3 
β (E , e ) = −  E + Λ e . (19)
12π 2 32π 2  2 

The first term on the right hand side of (19) is that present in the absence of gravity and results in
the electric charge increasing with the energy. The second term on the right hand side of (19)
represents the correction due to quantum gravity. For pure gravity with no cosmological constant, or
for small cosmological constant Λ , the quantum gravity contribution to the renormalization group
β -function is negative and therefore tends to result in asymptotic freedom.

2. On some equations concerning the gravitational contributions to the running of gauge

couplings. [2]

The action of Einstein-Yang-Mills theory is

1 1 
S = ∫ d 4 x − g  2 R − g µα gνβ Fµνa Fαβa  (20)
κ 4 

where R is Ricci scalar and Fµνa is the Yang-Mills fields strength Fµν = ∇ µ Αν − ∇ν Α µ − ig [Α µ , Αν ].
It is hard to quantize this lagrangian because of gravity-part’s non-linearity and minus-dimension
coupling constant κ = 16πG . Usually, one expands the metric tensor around a background metric

4
g µν and treats graviton field as quantum fluctuation hµν propagating on the background space-time
determined by g µν ,
g µν = g µν + κhµν . (21)

Furthermore, we can rewrite the lagrangian (20) also as follows:

 1 1 
S = ∫ d 4x − g  R − g µα gνβ Fµνa Fαβa  . (20b)
16πG 4 

Let us set g µν = η µν , where η µν is the Minkowski metric. hµν is interpreted as graviton field,
fluctuating in flat space-time. The lagrangian can be arranged to different orders of hµν or κ . The
free part of gravitation is of order unit and gives the graviton propagator

PGµνρσ (k ) = [
i νρ µσ
k2
g g + g µρ gνσ − g µν g ρσ ] (21b)

1
in the harmonic gauge C µ = ∂ν h µν − ∂ µ hνν = 0 . For simplicity, the metric g µν is understood as
2
η . The interactions of gauge field and gravity field are determined by expanding the second term
µν

of the lagrangian (20). We can compute the β function by evaluating two and three point functions
of gauge fields. These Green functions are general divergent, so counter-terms are needed to cancel
these divergences. The relevant counter-terms to the β function are

T µν = iδ abQ µν δ 2 , T µνρ = gf abcVqkp

µνρ
δ1 , Q µν ≡ q µ qν − q 2 g µν ,
≡ gνρ (q − k ) + g ρµ (k − p ) + g µν ( p − q ) .
µνρ µ ν ρ
Vqkp (22)

∂ 3 
The β function is defined as β ( g ) = gµ  δ 2 − δ1  . With the consistency condition
∂µ  2 
1
I 2Rµν = g µν I 2R , we obtain for two and three point functions
2

TLR(a + b )µν = 2κ 2Q µν ∫ dx − I 2R (0 ) + 2 I 2R (Mq2 ) + q 2 (3 x 2 − x )I 0R (Mq2 )

 3 
(23)
 2 

I 2 (0 ) + (g µν q ρ − q µ gνρ )I 2R (Mq2 ) + (gνρ k µ − kν g ρµ )I 2R (Mk2 ) +

 1 µνρ R
TLR(d + e )µνρ = 2igκ 2 ∫ dx  Vqkp
 2
( ) ( )]
+ g ρµ pν − p ρ g µν I 2R Mp2 (24)

from which we can directly read off the two-point and three-point counter-terms δ 2κ and δ1κ
respectively
1  2  M c2  µ s2  
δ 2κ = κ 2  M − µ 2
s
 ln − γ + 1 + y  2   ;
2 
16π 2  µ
c 2 w
 s  M c  

5
 1  2  M c2  µ s2  
δ1κ = κ 2  cM − µ 2
s
 ln − γ + 1 + y  2   .
2  (25)
16π 2 µ 2 w
  s  M c  

Putting δ1κ and δ 2κ in the following equation

∂ 3 κ κ 
∆β κ = gµ  δ 2 − δ1  , (26)
∂µ  2 

we obtain the gravitational corrections to the gauge β function

− µ s2  M c2  µ s2 
∆β κ = gκ 2  ln + 1 − γ + y 0
 M 2  (26b)
16π 2  µ s2
w
 c 

In general, the total β function of gauge field theories including the gravitational effects may be
written as follows

2 
1 2 − µs M c2  µ s2 
βκ = − b g 3
+ g κ  ln + 1 − γ + y  2  . (27)
16π 2 16π 2  µ s2
0 w 0
 M c 

The interesting feature of gauge theory interactions is the possible gauge couplings unification at
ultra-high energy scale when the gravitational effects are absent. Where the running of gauge
coupling in the Model Standard Super Symmetric (MSSM) without gravitational contributions is
known to be

α e−1 (µ ) = α e−1 (M ) + ; α w−1 (µ ) = α w−1 (M ) + ln ; α s−1 (µ ) = α s−1 (M ) −

33 M 1 M 3 M
ln ln (28)
10π µ 2π µ 2π µ

with experimental input at M Z

α e−1 (M Z ) = 58.97 ± 0.05 ; α w−1 (M Z ) = 29.61 ± 0.05 ; α s−1 (M Z ) = 8.47 ± 0.22 (29)

We note that these values are equal to the following values connected with the aurea ratio:

38,12461180 + 20,56230590 = 58,68691770; 29,12461180; 8,49844719.

Indeed, we have:

[(Φ ) + (Φ ) ]× 3 = (11,09016994 + 1,61803399) × 3 = 12,70820393 × 3 ≅ 38,12461180 ;

35 / 7 7/7

[(Φ ) ]× 3 = 6,85410197 × 3 ≅ 20,56230590 ;

28 / 7

[(Φ ) + (Φ ) + (Φ ) ]× 3 = (6,85410197 + 2,61803399 + 0,23606798) × 3 =

28 / 7 14 / 7 −21 / 7

= 9,70820394 × 3 ≅ 29,12461180 ;

[(Φ )14 / 7
+ (Φ )
−28 / 7
+ (Φ )
−42 / 7
+ (Φ )
−63 / 7
]× 3 = (2,61803399 + 0,145898033 + 0,055728089 +
6
 + 0,013155617 ) × 3 = 2,832815729 × 3 ≅ 8,49844719 .

5 +1
As usual, Φ = ≅ 1,61803399 , i.e. the value of the aurea ratio.
2

3. On various equations concerning some quantum effects in the theory of gravitation

[3]

The general spherical static metric is given by

ds 2 = − f (r )dt 2 + h(r ) dr 2 + r 2 dΩ 2 ,
−1
(30)

where f (r ) and h(r ) are arbitrary functions of the coordinate r . The angular part of the metric is
diagonal and given by
d −2  
( )
d −2
dΩ 2 = ∑  dθ i ∏ sin 2 θ j  . (31)
2

i =1  j = i +1 

Consider an interesting classical scalar field φ ( x ) living in a spacetime with the metric (30). This
field has an action given by:

∞
 
S = ∫ d d x − g  − φ∇ 2φ − ∑ λnφ n  . (32)
 n =2 

The λn are a set of arbitrary coupling constants. In particular, λ2 ≡ m 2 gives the mass of a weakly
coupled excitation of this field. We will expand φ in the eigenmodes of the free, classical wave
equation such that
φ ( x ) = ∫ dµ p a pφ p ( x ) , (33)
∇ 2φ p = m 2φ p . (34)

A sufficiently large set of quantum numbers p label the eigenbasis. The abstract formal expression
dµ p simply represents an appropriate measure over the modes under which

∫ dµ φ (x )φ ( y ) = δ (x − y ) ,
d
p p p

and
∫d
d
x − g φ p ( x )φq ( x ) = δ pq .

Now we want to consider the energy density, ρ , of a massless scalar field in a infinitely large
hypercubic blackbody cavity at temperature T . Consider a real scalar field φ I ( x, t ) , where I is
some kind of p -dimensional polarization index representing p internal degrees of freedom (for
example, in a well-chosen gauge, the transverse polarization of an Abelian vector field behaves
essentially like an internal index on a scalar field with p = d − 2 ). Further assume that the field is
sufficiently weakly coupled that each polarization component can be treated as an independent field

7
obeying an action similar to eq. (32) with all interaction coefficients higher than λ2 ≡ M 2 set equal
to zero. Then each field component obeys the classical equation of motion

∇ 2φ I ( x ) = M 2φ I ( x ) . (35)

The solutions to equation (35) may be expressed as a sum over modes labelled by a wave vector ka
obeying k 2 = − M 2 :
φkI = AkI sin (ka x a ) + BkI cos(ka x a ) , (36)

for arbitrary real coefficients AkI and BkI . We take the state to be labelled by the d − 1 spatial
components of ka and fix the frequency of each mode by ωk2 ≡ k02 = M 2 + ki k i . We now confine
the field to live in a cubic box of side length L by demanding Dirichlet boundary conditions at
x i = 0, L for i = 1...d − 1 . This demands BkI = 0 and

π
ki = mi , (37)
L

where mi is a spatial vector whose components are non-negative integers. The total energy in the
hypercube is given by
p ∞ ∞
U = ∑ ∑ ... ∑ω n k mI , (38)
I =1 m1 = 0 m d −1 = 0

where we understand that ωk is given by

π2 d −1
ωk = M 2 +
L2
∑m
i =1
2
i . (39)

The set of integers nmI defining the quantum state must obey the appropriate statistics for the field
φ I . We have described a pure quantum state of the theory. At a finite temperature T and zero
chemical potential, the system will be in a mixed state governed by the partition function

Q= ∑
{ }
e βω
n mI
− k
, (40)

where β ≡ 1 / T . This can be evaluated to give

∑ ln(1 − ξe βω ),
∞ ∞
ln Q = − pξ ∑ ... − k
(41)
m1 = 0 m d −1 = 0

where ξ = 1 for bosons and ξ = −1 for fermions. The overall factor of p occurs because the energy
is independent of p , so each polarization mode contributes equally. The average occupation
number of a given momentum mode in the thermal state is then given by

8
 p
1 ∂ p
nm = ∑ nmI = − ln Q = βω k . (42)
I =1 β ∂ωk e −ξ

The total energy in the hypercube can now be found by combining the expressions (38) and (42), or
by
∂ ∞ ∞ ∞ ∞
pω
U =− ln Q = ∑ ... ∑ ωk nm = ∑ ... ∑ βω k k . (43)
∂β m1 = 0 m d −1 = 0 m1 = 0 m d −1 = 0 e −ξ

We should pass from a state labelling in terms of quantum numbers mi to a labelling in terms of
physical momenta ki , with a mode density determined by the differential limit of equation (37).
The sums over mi then become integrals over ki as

L →∞ Ld −1 pω k
U → ∫ d d −1k , (44)
(2π ) e − ξ
d −1 βω k

where ωk is understood as M 2 + ki k i . The factors of 2 in the denominator of the measure arise

because the integrals over the ki run over both positive and negative values, whereas the mi were
only summed over non-negative values. Equation (44) scales properly with the volume, so that even
in the infinite volume limit we can define the spectral energy density over ki modes. Using the
spherical symmetry of the infinite volume limit and defining k = ki k i = ki , eq. (44) becomes

U
=ρ= p
Vol S d − 2 ( ) ∞ k d −2 M 2 + k 2
V (2π )d −1 ∫
0
dk
eβ M 2 +k 2
−ξ
. (45)

This defines the spectral energy density over the magnitude of the spatial momentum, via
ρ ≡ ∫ dkuk (k ) , as

uk (k ) = p
( )
Vol S d − 2 k d − 2 M 2 + k 2
. (46)
(2π )d −1 e β M 2 + k 2 − ξ

Similarly, we can define the spectral energy density over the frequency as

uω (ω ) = p
( ) (
Vol S d − 2 ω 2 ω 2 − M 2 )( d −3) / 2

, (47)
(2π )d −1 e βω − ξ

where ω runs over [M , ∞ ] .

The total energy density can now be evaluated using either eq. (46) or eq. (47) match. Simple
analytic results can be found for the case M = 0 , which will also apply when T >> M . In this case,
ω = k and equations (46) and (47) match. They give

ρ = pT d
(
Vol S d − 2 ) ∞ x d −1
(2π )d −1 ∫0
dx
ex − ξ
. (48)

9
The integral can be evaluated by pulling the exponential into the numerator, performing a Taylor
series in e − x , doing the integral, and resumming the Taylor series. The result for the general definite
integral is given by

(1−ξ ) / 2
x d −1dx  2  a ne− a − b ne−b
d −1
ζ (d )(d − 1)!∑
b
∫a e x − ξ = 1 − 2d  n=0 n!
, (49)

where ζ (s ) is the Riemann zeta function, which can be defined for real s > 1 as

∞
ζ (s ) = ∑
1
s
. (50)
n =1 n

This series arises in the evaluation of the integral with the ξ = 1 , for bosons. For fermions, ξ = −1 ,
the corresponding series is
∞
R (s ) ≡ ∑ (− 1)
n −1 1
. (51)
n =1 ns

This series can be evaluated using

ζ (s ) − R (s ) = + s + s + ... = s ζ (s ) ,
2 2 2 2
s
(52)
2 4 6 2

so that
 2
R (s ) = 1 − s ζ (s ) , (53)
 2 

which is the origin of this factor in eq. (49). So, eq. (48) becomes

(1−ξ ) / 2
 2 ζ (d )(d − 1)!
ρ = p 1 − d  T d . (54)
 2   d − 1  (d −1) / 2
2 d − 2 Γ π
 2 

Furthermore, we can rewrite the eq. (48) also as follows:

ρ = pT
(
Vol S d − 2 ) ∞ x d −1  2
(1−ξ ) / 2
ζ (d )(d − 1)!
∫ = p 1 − d 
d
dx x T d . (54b)
(2π )d −1 0 e −ξ  2   d − 1  (d −1) / 2
2 d − 2 Γ π
 2 

Before, we have considered the energy density, ρ , of a massless scalar field in a infinitely large
hypercubic cavity at temperature T . We now want to calculate the energy flux, Φ , emitted from a
blackbody with this same temperature. For the flux calculation, instead of Vol S d − 2 we will ( )
encounter

π
Ω cos(θ d − 2 )Θ − θ d − 2  =

∫d
d −2

2 

10
 d −3
2π (d − 2 ) / 2
2π π π
= ∫ dθ 1 ∫ dθ 2 ...∫ dθ d − 3 ∫
π /2 
dθ d − 2 cos θ d − 2( )∏ (sin (θ )) i i −1 
 = =
1
( )
Vol S d − 3 =
0 0 0 0
 i =2  (d − 2 )Γ d − 2  d − 2
 2 
 d −1
Γ 
(
= Vol B d − 2 ) =
1  2 
Vol S d − 2 , ( ) (55)
π (d − 2) Γ d − 2 
 
 2 

where Θ( x ) is the step function and B n is the n-dimensional unit ball: the compact subspace of R n
bounded by S n −1 (for example, B 3 is the unit 3-ball of volume 4π / 3 bounded by the sphere S 2 of
area 4π ). The fact that the expression for the energy density becomes that for the flux when
( ) ( )
Vol S d − 2 is replaced by Vol B d − 2 makes physical sense, since B n is the projection of S n onto R n .
We are left with the relationship of flux to energy density as

 d −1
Γ
Φ=  2 

ρ=
Vol B d − 2
ρ.
( )
d −2
π (d − 2 )Γ 
Vol S d − 2 ( ) (56)

 2 

Thus, the d-dimensional Stefan-Boltzmann law is given by

(1−ξ ) / 2
 2 ζ (d )(d − 1)!
Φ = p 1 − d  Td . (57)
 2   d − 2  d /2
2d − 2 (d − 2)Γ π
 2 

4. On some equations concerning the supersymmetric Yang-Mills theory applied in string

theory and some lemmas and equations concerning various gauge fields in any non-trivial
quantum field theory for the pure Yang-Mills Lagrangian. [4] [5]

The fields of the minimal N = 2 supersymmetric Yang-Mills theory are the following: a gauge field
Am , fermions λαi and λα&i transforming as (1 / 2,0,1 / 2) and (0,1 / 2,1 / 2) under
SU (2 )L × SU (2 )R × SU (2 )I , and a complex scalar B - all in the adjoint representation of the gauge
group. Covariant derivatives are defined by

Dm Φ = (∂ m + iAm )Φ , (58)

and the Yang-Mills field strength is

Fmn = ∂ m An − ∂ n Am + [ Am , An ] . (59)

The supersymmetry generators transform as (1 / 2,0,1 / 2) ⊕ (0,1 / 2,1 / 2) ; introducing infinitesimal

parameters ηαi and ηα&i , furthermore D = [B, B ] . The minimal Lagrangian is

11
L=
1
2 ∫
 −1
d 4 xTr  Fmn F mn − iλiα&σ αmα& Dm λαi − Dm B D m B − [B , B ] −
1 2 i
B ε ij λαi , λαj +
i
[ 
Bε ij λα&i , λiα&  ] [ ]
e M  4 2 2 2 
(60)
Here Tr is an invariant quadratic form on the Lie algebra which for G = SU ( N ) we can
conveniently take to be the trace in the N dimensional representation.
It is possible to realize a mass term for the N = 1 matter multiplet (which consists of B and
ψ α = λα2 ) by adding to the Lagrangian a term of the form I (ω ) + {Q1 ,...}, where Q1 is the charge
&

corresponding to the ρ1 transformation. Furthermore, Q1 is the only essential symmetry. We

obtain:

2M
& &
( )
Lˆ = L + I (ω ) + {Q1 ,...} = L − ∫ d 4 xTr mλα2 λα 2 + m λ2&σ& λσ& 2& − e 2 ∫ d 4 x 2m mTrB B ,
1
(61)
M

with
m = σ mn 2& 2& ωklε mnkl . (62)

The mass is proportional to the holomorphic two-form ω . The N = 1 gauge multiplet, consisting of
the gauge field Am and the gluino λα = λαi , remains massless.

With regard the Λ16 -amplitude in the type IIB description, the classical action for the operator (Λ )
16

in the AdS5 × S 5 supergravity action is

 1 

( )
− 2π  iχ 0  16  
S Λ [J ] = e  gs 
g s−12VS 5 ∫ d 4 x0 dρ 0
1
t ∏ 
ρ 05 p =1 
K (ρ , x ; x ) 1
ρ γ 5̂
+ (x − x )µ
γ µ J Λ (x ) . (63)
ρ0
16 4 0 0 p 0 0 p ˆ p


This result is in agrees with the following expression obtained in the Yang-Mills calculation:

8π 2
16  
ρ 04
(ρ η + (x )
+ iθ YM
G16 (x p ) = gYM − x0 )µ σ αµpα& p ξ
−
1 1
∏
α& p A p
∫ d x0dρ0 ρ 05 ∫
η ξ 
2 Ap
8 g YM 4 8 8

[ ]
e d d
p =1  ρ + (x − x )
0 αp
ρ0
2 4 p
2
 0 p 0 
(64)
i.e. the correlation function in the super Yang-Mills description.

The low energy effective action for type IIB superstring theory in ten dimensions includes the
interaction (in the string frame)

 
S ≈ ls4 ∫ d 10 x − g  ζ (3) e − 2φ + 2ζ (3)ζ (2 ) + 6ζ (4 )e 2φ + ζ (6 )e 4φ + ...  D 6 R 4 ,
2 2
(65)
 9 

where the … involve contributions form D-instantons.

We consider the perturbative contributions to the D 6 R 4 interaction. The sum of the contributions to
the four graviton amplitude at tree level and at one loop in type II string theory compactified on T 2
is proportional to

− − 2φ ( )(
Γ − ls2 s / 4 Γ − ls2t / 4 Γ − ls2u / 4 )(
+ 2πI  R 4 , (66)
)
 2

V e
( )(
Γ 1 + ls s / 4 Γ 1 + ls t / 4 Γ 1 + ls u / 4
2 2 2
)( )
12
where V2 is the volume of T 2 in the string frame, s, t , u are the Mandelstam variables, and I is
obtained from the one loop amplitude. The amplitude is the same for type IIA and type IIB string
theories. Now I is given by
d 2Ω
I =∫ Z lat F (Ω, Ω ) (67)
F Ω2
2

where F is the fundamental domain of SL(2, Z ) , and d 2Ω = dΩd Ω / 2 . In the above expression, the
lattice factor Z lat which depends on the moduli is given by

 πT2 Ω 
2

Z lat = V2 ∑ exp − 2πiT (det A) − (1 U )A   . (68)

A∈Mat ( 2× 2 , Z )  Ω 2U 2  1  

Expanding eq. (67) to sixth order in the momenta, we get that

I=
3
(
ls6 3 3
)[
s + t + u 3 Iˆ1 + Iˆ2 , ] (69)

where
d 2Ω 3
d 2ν i
Iˆ1 = 4 ∫ Z lat ∫ ∏ ln χˆ (ν 1 − ν 2 ; Ω )ln χˆ (ν 1 − ν 3 ; Ω )χˆ (ν 2 − ν 3 ; Ω ) , (70)
FL Ω 2
i =1 Ω 2
T
2

and
d 2Ω 3
d 2ν i
Iˆ2 = ∫FL Ω22 lat ∫T ∏
Z [ln χˆ (ν 1 −ν 2 ; Ω )]3 . (71)
i =1 Ω 2

We have also that

(4π )3 Iˆ d 2Ω
Z lat E3 (Ω, Ω )
SL ( 2 , Z )
1 = ∫ = Iˆ11 + Iˆ12 + Iˆ33 , (72)
4 FL Ω22

where Iˆ11 , Iˆ12 , and Iˆ13 are the contributions from the zero orbit, the non-degenerate orbits and the
degenerate orbits of SL(2, Z ) respectively. In the eq. (72), we have used the expression

5/2
3π
E3 (Ω, Ω ) K 5 / 2 (2πΩ 2 m1m2 )e 2πim1m2 Ω1 .
SL ( 2 , Z )
= 2ζ (6 )Ω + ζ (5) + π 3 Ω 2 ∑
3 m1
(73)
4Ω 2
2 2
m1 ≠ 0 , m 2 ≠ 0 m2

Thence, we can rewrite the eq. (72) also as follows:

(4π )3 Iˆ
5/ 2
d 2Ω 3π
Z lat × 2ζ (6 )Ω32 + ζ (5) + π 3 Ω 2 ∑ K 5 / 2 (2πΩ 2 m1m2 )e 2πim1m2 Ω1 . (74)
m1
=∫
Ω2 4Ω 2
1 2 2
4 FL
m1 ≠ 0 , m 2 ≠ 0 m2

The contribution from the zero orbit gives

13
 d 2Ω
E3 (Ω, Ω )
SL ( 2 , Z )
Iˆ11 = V2 ∫ =0. (75)
FL Ω 2
2

The contribution from the non-degenerate orbits gives

πT2
∞ dΩ − 2πiTkp −
2
kΩ + j + pU
E3 (Ω, Ω )
∞ SL ( 2 , Z )
Iˆ12 = 2V2 ∫ dΩ1 ∫
−∞ 0 Ω2
2
∑e
k > j ≥ 0, p ≠ 0
Ω 2U 2
=
2
5/ 2

= 2 T2 E3 (U ,U ) K 5 / 2 (2πT2 pk )e 2πipkT1 . (76)

SL ( 2 , Z ) p
∑
p ≠ 0, k ≠ 0 k

Furthermore, the contribution from the degenerate orbits gives

3πζ (5) 
πT2
∞ dΩ 2 
2
− j + pU
E3 (Ω, Ω )  E (U ,U )
SL ( 2 , Z ) SL ( 2 , Z )
 2ζ (6 )T23 +
1/ 2
Iˆ13 = V2 ∫ dΩ1 ∫
−1 / 2 0 Ω 2
2
∑ e
( j , p )≠ (0 , 0 )
Ω 2U 2
= 3 
π  2  3
4T2 
.
2

(77)

Now we want to show two lemmas and equations concerning the gauge fields as described in the
Jormakka’s paper “Solutions to Yang-Mills equations [5]”. Thence, in the next Section, we
describe some possible mathematical connections between some equations concerning this
interesting argument and some equations concerning the Ramanujan’s modular equations that are
related to the physical vibrations of the bosonic strings and of the superstrings, and some
Ramanujan’s identities concerning π.

Lemma 1

Let the gauge field satisfy

∑ j =1 ρ 2j
3

Aµa , R (0, x1 , x2 , x3 ) = sa cµ e
−β 2
(78)

where ρ j and cµ are as in the following expressions:

1
r j = ρ j + iσ j , ρ j , σ j ∈ R ; ρ1 = x1 − x2 + 2x3 ; ρ 2 = x1 − x2 − 2x3 ; ρ 3 = − x1 + x3 ;
2
h(rj ) = u (ρ j ,σ j ) + iv (ρ j ,σ j )
1
σ1 = σ 2 = 0 ; σ3 = x0 ; (79)
2

c0 = 2 ; c1 = c2 = 1 ; c3 = 0 ; e0 = − 2 ; e1 = e2 = 1 ; e3 = 0 , (80)

and β , sa ∈ R . Then
π 
3/ 2

∫ d x(A (0, x , x , x ))
2 1
3 a
=s c  
2 2
. (81)
k 1 2 3 a k
2 β3

14
The expression (79) describes a localized gauge field, gauge boson, which moves in the x1 , x2
direction with the speed of light as a function of x0 . We select a concrete case that gives easy
calculations. Let f (ρ j ) = − β 2 ρ 2j and extend it to h(rj ) = − β 2 rj2 . The real and imaginary parts of
d µ = cµ + ieµ are described from (80). We evaluate the gauge potential at x0 = 0 and take the real
part.

We change the variables to y1 , y2 , y3

2 1 2 1
y1 = 3 x1 − x2 − x3 ; y2 = x2 − x3 ; y3 = 2 x3 (82)
3 6 3 3

Then
3

∑ρ
j =1
2
j = y12 + y22 + y32 (83)

As y2 and y3 are not functions of x1 we can change the order of integration

− ( )2 y12 − ( )2 y12
1 1
= ∫ d 2 xe − β ( y 2 + y 3 ) ∫ dx1e 2 = ∫ d 2 xe − β ( y 2 + y3 )
∑ j =1 ρ 2j ∑ j =1 y 2j
3 3
−β 2 −β 2 2β 1 2β
∫ d xe = ∫ d 3 xe ∫ =
2 2 2 2 2 2
3
dy1e 2
3
= ∫ d 2 xe − β ( y 2 + y 3 )
2 2 2 1
3
2π ( 2β )
−1
(84)

As y3 is not a function of x2 we can change the order of integration

( )2 y12
= ∫ d 2 xe − β ( y 2 + y 3 ) ( )
1
(
− β 2 y 22 + y 32 ) 1 − 2β 1
2π 2β
−1
∫d 3∫
=
2 2 2
2 2
xe dy e 1
3
( )2 y 22 ( )2 y 22
(
) ∫ dx e ∫ dx e ( ) ∫ dx e
1 1
1 −1 − 2β 1 3 −1 − 2β
− β 2 y 32 − β 2 y 32
= 2π 2β 3 2
2
= 2π 2β 3 ∫ dy e 2
2
=
3 3 2

(2π )( 2 β ) ∫ dx e (2π ) 2 β ( ) ∫ dy e (2π )2 ( )

1 3 1 −2 − β 2 y 32 31 −2 − β 2 y 32 1 31 3 −3
= = 3 3 = 2β . (85)
3 2 3 22 3 22

Thus
3
 π 2 1 π  1
3
−2β 2 ∑ ρ 2j − 2 β 2 ∑ ρ 2j
∫d =   3 ⇒ ∫ d 3 xe =   3⇒
3
xe
2 β 2 β
π 
3/ 2

(
⇒ ∫ d 3 x Aka (0, x1 , x2 , x3 ) = sa2ck2  
2
) 1
β3
. (86)
2

Thence, we can conclude that the eq. (81) is true.

We can rewrite the eq. (85) also as follows:

− ( )2 y12
− β 2 (y 2 + y 2 ) 1
= ∫ d 2 xe − β ( y 2 + y 3 ) ( ) ( )
1
2β
(2π )2
1 −1 1 31 3 −3
∫ d xe 2 3 3 ∫ dy1e 2 2π 2β = 2β =
2 2 2
2

3 3 22

15
 ( ) π3
2

=  ∫ d 2 xe − β ( y 2 + y 3 )
 1 −1  1
2π 2β
2 2 2
= ×
( )
 8 6
. (86b)
 3  4 2β

Lemma 2

Let the gauge field satisfy

∑ j =1 ρ 2j
3

Aµa , R (0, x1 , x2 , x3 ) = sa cµ e
−β 2
(87)

where ρ j and cµ are as in (79), (80) and β , sa ∈ R . Then

1 π3
∫ d LR = −
3

β 16
∑s B
a
2
a (88)

where in Minkowski’s metric at x0 = 0 , we have B = 0 . In the negative definite metric

−1 0 0 0
 
 0 −1 0
(g )µν µ ,ν =
0
0
0 −1 0 
, (88b)
 
0 − 
 0 0 1 

13 2
we have that B = + + 4.
3 3
From (79) and (82) follows that

1 1 1 1 1 1 1 2
ρ1 = y1 − y2 + y3 ; ρ2 = y1 − y2 − y3 ; ρ3 = − y1 − y2 . (89)
3 6 2 3 6 2 3 3

For Minkowski’s metric

P (ρ ) = 0 = B1 y12 + B2 y22 + B3 y32 + B4 y1 y2 + B5 y1 y3 + B6 y2 y3 (90)

where Bk = 0 for all k . For the metric (88b)

P (ρ ) = 4 ρ12 + 4 ρ 22 + ρ 32 − 4 ρ 2 ρ 3 = B1 y12 + B2 y22 + B3 y32 + B4 y1 y2 + B5 y1 y3 + B6 y2 y3 (91)

where
13 2 4 4 4
B1 = ; B2 = ; B3 = 4 ; B4 = − 2; B5 = − ; B6 = − . (92)
3 3 3 6 3

We do the integration with generic parameters B j . Then

16
 −
1
(
( 2 β )2 ρ12 + ρ 22 + ρ 32 ) −
1
(
( 2 β )2 y12 + y 22 + y 32 )
∫ d xe
3 2
P(ρ ) = ∫ d xe 3 2
( B1 y12 + B2 y22 + B3 y32 + B4 y1 y2 + B5 y1 y3 + B6 y2 y3 ) .
(93)

As y2 and y3 are not functions of x1 we can change the order of integration and change the
integration parameter x1 to y1

−
1
(
( 2 β )2 ρ12 + ρ 22 + ρ 32 ) −
1
(
( 2 β )2 y12 + y 22 + y 32 )
∫ d xe
3 2
P (ρ ) = ∫ d 3 xe 2
( B1 y12 + B2 y22 + B3 y32 + B4 y1 y2 + B5 y1 y3 + B6 y2 y3 ) =
( )
(2 β )2 y 22 + y 32 ( 2 β )2 y12
( ) (2 β )2 y12
1 1 1
− − −
= ∫ d 2 xe 2  B1 ∫ dx1 y12e

2
+ B2 y22 + B3 y32 + B6 y2 y3 ∫ dx1e 2
+

−
1
(2 β )2 y12 
+ (B4 y2 + B5 y3 )∫ dx1 y1e 2 =


− ( 2 β ) 2 ( y 22 + y 32 ) − ( 2 β )2 y12
( − ( 2 β )2 y12
)
1 1 1
1
= ∫  1∫ + + + 6 2 3 ∫ +
2 2 2 2 2 2 2
d xe  B dy y
1 1 e B y
2 2 B y
3 3 B y y dy1e
3 
−
1
(2 β )2 y12 
+ (B4 y2 + B5 y3 )∫ dy1 y1e 2 =


− ( 2 β ) 2 ( y 22 + y 32 )
( ) 
1
1 1 1
= ∫ d 2
xe 2  B1 2π + B2 y22 + B3 y32 + B6 y2 y3 2π  . (94)
3  (2 β )3
2β 

As y3 is not a function of x2 we can change the order of integration and change the integration
parameter x2 to y2 :

−
1
(
( 2 β )2 ρ12 + ρ 22 + ρ 32 ) −
1
(
( 2 β )2 y12 + y 22 + y 32 )
∫ d xe
3 2
P (ρ ) = ∫ d 3 xe 2
( B1 y12 + B2 y22 + B3 y32 + B4 y1 y2 + B5 y1 y3 + B6 y2 y3 ) =
− ( 2 β ) 2 ( y 22 + y 32 )
( ) 
1
1 1 1
= ∫ d 2
xe 2  B1 2π + B2 y22 + B3 y32 + B6 y2 y3 2π  =
3  (2β )3
2β 
− ( 2 β )2 y 32  − ( 2 β ) 2 y 22 − ( 2 β )2 y 22
1 1 1
1 3 1 1
= 2π ∫ dx3e 2  B1 3 ∫ dy e 2
+ B ∫ dy y 2
e 2
+
 (2 β )
3 2  2 2
2β
2 2

− ( 2 β )2 y 22 − ( 2 β )2 y 22 
1 1
1 1
+ B3 y32
2β ∫ dy 2 e 2
+ B y
6 3
2β ∫ dy y
2 2 e 2 =


− ( 2 β ) 2 y 32  1 
1
1 3 1 1 1 1 1
= 2π ∫ dx3e 2  B1 2π + B2 2π + B3 y32 2π =
3 2  (2β )
3
2β 2β (2β )3
2β 2β 
− ( 2 β )2 y 32  1 
1
1 31 1 1
= 2π ∫ dy3e 2  B1 + B2 + B3 y32 =
2 
3 22  (2 β )4
(2 β )4
(2 β ) 
3
1 31
(2π )2 (B1 + B2 + B3 ) 1 5 = π 2 1 5 (B1 + B2 + B3 ) .
3
= (95)
3 22 (2β ) (2 β )
We can rewrite this equation also as follows:

17
  3 − 12 (2 β )2 (y12 + y 22 + y 32 )
2
 π3
∫ d xe B y 2
+ (
B y 2
+ B y 2
+ B y y + B y y + B6 2 3  =
y y ×
81
. ) (95b)

1 1 2 2 3 3 4 1 2 5 1 3
 4 16 β 5 2
( )
Thence, we can conclude that the eq. (88) is true. Indeed, we have that:

−
1
(
( 2 β )2 ρ12 + ρ 22 + ρ 32 ) −
1
(
( 2 β )2 y12 + y 22 + y 32 )
∫ d xe
3 2
P(ρ ) = ∫ d xe3 2
( B1 y12 + B2 y22 + B3 y32 + B4 y1 y2 + B5 y1 y3 + B6 y2 y3 ) =
1 π3
3
=π2
1
( B + B + B ) = ∫ d 3
L = − ∑ sa2 B . (96)
(2β )5 1 2 3 R
β 16 a

5. Ramanujan’s equations, zeta strings and mathematical connections

Now we describe some mathematical connections with some sectors of String Theory and Number
Theory, principally with some equations concerning the Ramanujan’s modular equations that are
related to the physical vibrations of the bosonic strings and of the superstrings, the Ramanujan’s
identities concerning π and the zeta strings.

3.1 Ramanujan’s equations [6] [7]

With regard the Ramanujan’s modular functions, we note that the number 8, and thence the
numbers 64 = 8 and 32 = 2 × 8 , are connected with the “modes” that correspond to the physical
2 2

vibrations of a superstring by the following Ramanujan function:

 ∞ cos πtxw' 
− πx 2 w '
 ∫0 cosh πx e dx  142
4 anti log πt 2 ⋅ 2
− t w'
 φw' (itw') 
w'
4
1  e
8=
3   10 + 11 2   10 + 7 2  
log   + 
 


  4   4 
 . (97)

Furthermore, with regard the number 24 (12 = 24 / 2 and 32 = 24 + 8) this is related to the physical
vibrations of the bosonic strings by the following Ramanujan function:

18
  ∞ cos πtxw' 
− πx 2 w '
 ∫0 cosh πx e dx  142
4 anti log πt 2 ⋅ 2
 e
−
4
w'
φ (itw ' )  t w'
24 =  
w '

  10 + 11 2   10 + 7 2  
log   + 
 


  4   4 
 . (98)

It is well-known that the series of Fibonacci’s numbers exhibits a fractal character, where the forms
5 −1
repeat their similarity starting from the reduction factor 1 / φ = 0,618033 = 2 (Peitgen et al.
1986). Such a factor appears also in the famous fractal Ramanujan identity (Hardy 1927):

5 −1 5
0,618033 = 1 / φ = = R(q) +
2 3+ 5  1 q f 5 (−t ) dt 
1+
2
exp ∫ 1/ 5

4/5 
 5 0 f ( −t ) t  , (99)

 
 
3  5 
π = 2Φ − R(q) +
20  3+ 5  1 q f (−t ) dt  
5

 1+ exp ∫ 
4/5  
 5 0 f (−t ) t   ,
1/ 5
and  2
(100)

5 +1
Φ=
where 2 .

Furthermore, we remember that π arises also from the following identities (Ramanujan’s paper:
“Modular equations and approximations to π” Quarterly Journal of Mathematics, 45 (1914), 350-
372.):

π=
12
log 
( )(
 2 + 5 3 + 13 

) π=
24 
log 
 10 + 11 2 


+

 10 + 7 2  




130  2 ,
142   4   4 
(100a) and  .
(100b)

From (100b), we have that

π 142
24 =
  10 + 11 2   10 + 7 2  
log   +  
  4   4 
    
. (100c)

Let u (q ) denote the Rogers-Ramanujan continued fraction, defined by the following equation

19
 q1 / 5 q q 2 q 3
u := u (q ) := , q <1 (101)
1 + 1 + 1 + 1 + ...

( )
and set v = u q 2 . Recall that ψ (q ) is defined by the following equation

ψ (q ) := f (q, q 3 ) = ∑ q n (n +1) / 2 =
∞
(q ; q )
2 2
∞

n =0 (q; q )2
∞
. (102)

Then
8 ψ 5 (q ) dq
( 1+
= log u 2v 3 + 5 log ) ( 5 − 2 uv 2  )
.
∫
5 ψ q q5
( ) 1− ( 5 + 2 uv 2 ) (103)

(
We note that 1 + 5 − 2 = 2 ⋅ 0,61803398 ) and that 1 − 5 + 2 = 2 ⋅ 1,61803398 , where ( )
φ = 0,61803398 and Φ = 1,61803398 are the aurea section and the aurea ratio respectively. Let
k := k (q ) := uv 2 . Then from page 326 of Ramanujan’s second notebook, we have

1− k 
2
21+ k 
u5 = k  and v = k 
5
. (104)
1+ k  1− k 
It follows that
1  1− k 
( )
log u 2v 3 = log k 8
5  1+ k 
. (105)

If we set ε = ( )
5 + 1 / 2 = 1,61803398 , i.e. the aurea ratio, we readily find that ε 3 = 5 + 2 and
ε −3
= 5 − 2 . Then, with the use of (105), we see that (103) is equivalent to the equality

8 ψ 5 (q ) dq 1  8 1 − k   1 + ε −3k 
5 ∫ ψ q5 q 5  1 + k 
= +   . (106)
( ) log k  5 log
 1− ε k 
3

Now from Entry 9 (vi) in Chapter 19 of Ramanujan’s second notebook,

ψ 5 (q ) f (q 2 , q 3 )
ψ (q 5 )
= 25q 2
ψ (q )
ψ 3
(q 5
) + 1 − 5q
d
dq
log
f (q, q 4 )
. (107)

By the Jacobi triple product identity

f (a, b ) = (− a; ab )∞ (− b; ab )∞ (ab; ab )∞ , (108)

we have

(
f q 2 , q3) (
=
)(
− q 2 ; q5 ∞ − q3 ; q5 ∞
=
) ( )( )(
q; q 5 ∞ q 4 ; q 5 ∞ q 4 ; q10 ∞ q 6 ; q10 ∞
= q1 / 5
)(
u (q ) )
(
f q, q ) (
4 5
)(
− q; q ∞ − q ; q ∞
4 5
) 2 5
(
3 5
)(
2
q ;q ∞ q ;q ∞ q ;q ∞ q ;q ∞ 10 8
)( 10
)(
v(q )
,
) (109)

by the following expression

u (q ) = q1 / 5
(q; q ) (q ; q )
5
∞
4 5
∞
(q ; q ) (q ; q )
2 5
∞
3 5
∞
. (110)

20
Using (109) in (107), we find that

8 ψ 5 (q ) dq
∫
5 ψ q q 5
( ) 5q
( )
= 40∫ qψ (q )ψ 3 q 5 dq + ∫ dq − 8∫ log q1 / 5u / v dq =
8 d
dq
( )
1− k
= 40 ∫ qψ (q )ψ 3 (q 5 )dq − 8 log(u / v ) = 40 ∫ qψ (q )ψ 3 (q 5 )dq + log k − log
8 24
, (111)
5 5 1+ k

where (104) has been employed. We note that we can rewrite the eq. (111) also as follows:

8 ψ 5 (q ) dq 1− k
∫ ( )
5 ψ q q5
5
( )
=40 ∫ qψ (q )ψ 3 q 5 dq + log k − log
8 24
5 1+ k
. (112)

5
Multiplying both sides for , we obtain the following identical expression:
64

1 ψ 5 (q ) dq 1− k
∫
8 ψ q q5
( ) 8
1
8
( )
= 25 ⋅ ∫ qψ (q )ψ 3 q 5 dq + log k − 3 ⋅ log
1 1
8 1+ k
, (112b)

or
1 ψ 5 (q ) dq 1  1− k 
∫ ( )
8 ψ q q 8
5
( )
=  25∫ qψ (q )ψ 3 q 5 dq + log k − 3 log  . (112c)
1+ k 

In the Ramanujan’s notebook part IV in the Section “Integrals” are examined various results on
integrals appearing in the 100 pages at the end of the second notebook, and in the 33 pages of the
third notebook. Here, we have showed some integrals that can be related with some arguments
above described.

∞ ∞ ∞
e − 2 a nψ (n )dn =
1 k xdx
∫ + ∑ − 4a 2 ∫ 2πx
2

( )( ) ( )( )
2
4 a ; (113)
0 8πa 2
k =1 e
2πk
− 1 4a + k
4 4 0 e − 1 4a 4 + x 4

∞ xdx 1 π ∞
1
4a 2 ∫
0 (e 2πx
)(
− 1 4a + x
4 4
=
) −
4a 4
+ a ∑
k =1 a + (a + k )
2 2
. (114)

Multiplying both sides for π 2 , we obtain the equivalent expression:

∞ xdx π2 π3 ∞
1
π 2 ⋅ 4a 2 ∫
0 (e 2πx
)(
− 1 4a + x
4 4
=
) −
4a 4
+ a π 2
∑
k =1 a + (a + k )
2 2
. (114b)

Let n ≥ 0 . Then
sin (2nx )dx π ∞
(− 1) e − (2 k +1)n cos{(2k + 1)n}
k
∞
∫ = − 2∑ . (115)
0 x(cosh (πx ) + cos(πx )) 4 k = 0 (2 k + 1) cosh{(2 k + 1)π / 2}

Also here, multiplying both sides for π 2 , we obtain the equivalent expression:

21
 sin (2nx )dx π3 ∞
(− 1) e −(2k +1)n cos{(2k + 1)n} .
k
∞
π 2
∫ = − 2π 2 ∑ (115b)
0 x(cosh (πx ) + cos(πx )) 4 k = 0 (2 k + 1) cosh{(2k + 1)π / 2}

Now we analyze the following integral:

 1 + 1 + 4x 
log 

 dx = π ;
2
1
 2
I := ∫ (116)
0 x 15

( )
Let u = 1 + 1 + 4 x / 2 , so that x = u 2 − u . Then integrating by parts, setting u = 1 / v and using the
z log(1 − w )
following expression Li2 ( z ) = − ∫ dw , z ∈ C , and employing the value
0 w
 5 −1 π 2  5 −1
Li2  =
 − log 2   , we find that

 2  10  2 

( 5 +1 / 2) ( ) log(u 2 − u )
5 +1 / 2 ( )  log u log(u − 1) 
5 +1 / 2
I =∫
log u
(2u − 1)du = − ∫1 du = − ∫  + du =
1 u −u
2
u 1
 u u 
1  5 + 1  ( 5 +1)/ 2 log(1 − v ) − log v
= − log 2  +∫
 1 dv =
2  2  v
1  5 + 1  5 −1 1  5 +1
 + Li2 (1) − log 2 
π2 π2 π2
= − log 2   − Li2  =− + = . (117)
2  2   2 
  2  2 
  10 6 15

Thence, we obtain the following equation:

 1 + 1 + 4x 
log 

I =∫
1
 2  dx = ( )
5 +1 / 2 log u
( ) π2
0 x ∫1 u2 − u
2u − 1 du =
15
. (118)

15
Multiplying both sides for π , we obtain the equivalent expression:
4

 1 + 1 + 4x 
log 

15 1 
I = π∫
2  dx = 15 π ( )
5 +1 / 2 log u
( ) 15 π 2 π 3
π⋅
4 ∫1
2u − 1 du = = ;
4 0 x u2 − u 4 15 4
 1 + 1 + 4x 
log 

15 1 
I = π∫
2  dx = 15 π ( )
5 +1 / 2 log u
( ) π3
4 ∫1
2u − 1 du = . (118b)
4 0 x u2 − u 4

In the Ramanujan’s paper: “Some definite integrals connected with Gauss’s sums” (Ramanujan,
1915 – Messanger of Mathematics, XLIV, 75-85), are described the following integrals:

22
 ∞ cos πx 2 1 1 3  ∞ sin πx 2 1
∫ 0
x3
sinh πx
dx =  − 2  , (119)
16  4 π  ∫
0
x3
sinh πx
dx =
16π
, (120)

∞ x cos 2πx 1 1 3 5  ∞ x 2 cos 2πx 1  5 5 

∫0
e 2π x
−1
dx =  − + 2  ,
64  2 π π 
(121) ∫
0
e 2π x
−1
dx = 1 − + 2  . (122)
256  π π 

Now, we sum (119) and (120) obtaining:

∞ cos πx 2 ∞ sin πx 2 1 1 3  1 1 3 1
∫
0
x3
sinh πx
dx + ∫0
x3
sinh πx
dx =  − 2  + = −
16  4 π  16π 64 16π 2
+
16π
.

Now we multiply both the sides for π 3 and obtain:

∞ cos πx 2 ∞ sin πx 2 π 3 3π 3 π 3 π 3 3π π 2 1  π 3 
π (∫
3
x3
dx + ∫ x 3
dx) = − + = − + =  − 3π + π 2  ,
0 sinh πx 0 sinh πx 64 16π 2
16π 64 16 16 16  4 

thence
 ∞ cos πx 2 ∞ sin πx 2  1  π 3 
π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  . (123)
 0 sinh πx 0 sinh πx  16  4 

Now, we sum (121) and (122) obtaining:

∞ x cos 2πx ∞ x 2 cos 2πx  1 3 5 1 5 5 

∫0
e 2π x − 1
dx + ∫
0
e 2π x
−1
dx =  − +
 128 64π 64π
2
+ − + .
256 256π 256π 2 

Now we multiply both the sides for π 3 and obtain the following equivalent equation:

 ∞ x cos 2πx ∞ x cos 2πx

2
 1 1π3 3 2 5 1 5 5 
π  ∫ 2π x
 3
dx + ∫ 2π x dx  = ⋅  − π + π + π 3 − π 2 + π  . (124)
 e0
−1 0
e − 1  16 2  4 2 2 8 8 8 

In the work of Ramanujan, [i.e. the modular functions,] the number 24 (8 x 3) appears repeatedly.
This is an example of what mathematicians call magic numbers, which continually appear where we
least expect them, for reasons that no one understands. Ramanujan‘s function also appears in string
theory. Modular functions are used in the mathematical analysis of Riemann surfaces. Riemann
surface theory is relevant to describing the behavior of strings as they move through space-time.
When strings move they maintain a kind of symmetry called "conformal invariance". Conformal
invariance (including "scale invariance") is related to the fact that points on the surface of a string's
world sheet need not be evaluated in a particular order. As long as all points on the surface are taken
into account in any consistent way, the physics should not change. Equations of how strings must
behave when moving involve the Ramanujan function. When a string moves in space-time by
splitting and recombining a large number of mathematical identities must be satisfied. These are the
identities of Ramanujan's modular function. The KSV loop diagrams of interacting strings can be
described using modular functions. The "Ramanujan function" (an elliptic modular function that
satisfies the need for "conformal symmetry") has 24 "modes" that correspond to the physical
vibrations of a bosonic string. When the Ramanujan function is generalized, 24 is replaced by 8 (8 +
2 = 10) for fermionic strings.
23
With regard the Section 1, we have the following mathematical connections between the eqs. (16b)
and (18b) and the eqs. (112c), (123) and (124):

(1) 1 κ 2 Ec2  3 3 5 3 1 3 1 1 
Γquad =− 2 
− ω + ω 2 + ξ − ωξ + ωζ − ωξζ + ω 2ζ  ∫ d 4 xF 2 ⇒
8 4π  8 4 8 8 2 8 32 32 
1 ψ (q ) dq 1  1− k 
( )
5
⇒ ∫ =  25∫ qψ (q )ψ 3 q 5 dq + log k − 3 log
( )
8 ψ q q 85  ; (125)
1+ k 

(1) 1 κ 2 Ec2  3 3 5 3 1 3 1 1 
Γquad =− 2 
− ω + ω 2 + ξ − ωξ + ωζ − ωξζ + ω 2ζ  ∫ d 4 xF 2 ⇒
16 2π  8 4 8 8 2 8 32 32 
 ∞ cos πx 2
∞ sin πx 2
 1 π 3

⇒ π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  ; (126)
 0 sinh πx 0 sinh πx  16  4 

(1) 1 κ 2 Ec2  3 3 5 3 1 3 1 1 
Γquad =− 2 
− ω + ω 2 + ξ − ωξ + ωζ − ωξζ + ω 2ζ  ∫ d 4 xF 2 ⇒
16 2π  8 4 8 8 2 8 32 32 
 ∞ x cos 2πx ∞ x cos 2πx
2
 1 1π 3
3 5 1 5 5 
⇒ π 3  ∫ 2π x dx + ∫ 2π x dx  = ⋅  − π 2 + π + π 3 − π 2 + π  ; (127)
 0 e −1 0
e − 1  16 2  4 2 2 8 8 8 

1 1  κ 2 Ec2 3κ 2 Λ
( )
divp Γ (1) = ⋅  −
8 2  8π 2
−
16π 2
ln E 2
c +
e2
3π 2

ln Ec2  ∫ d 4 xF 2 ⇒

1 ψ (q ) dq 1  1− k 
( )
5
⇒ ∫ =  25∫ qψ (q )ψ 3 q 5 dq + log k − 3 log
( )
8 ψ q q 8
5
1+ k 
 ; (128)

1  κ 2 Ec2 3κ 2 Λ
divp Γ( )(1)
= −
16  8π 2
−
16π 2
2 e2
3π

ln Ec + 2 ln Ec2  ∫ d 4 xF 2 ⇒

 ∞ cos πx 2 ∞ sin πx 2  1  π 3 
⇒ π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  ; (129)
 0 sinh πx 0 sinh πx  16  4 

1  κ 2 Ec2 3κ 2 Λ
( )
divp Γ (1) =  −
16  8π 2
−
16π 2
ln Ec
2
+
e2
3π 2

ln Ec2  ∫ d 4 xF 2 ⇒

 ∞ x cos 2πx ∞ x cos 2πx
2
 1 1π3 3 2 5 1 5 5 
⇒ π  ∫ 2π x
3
dx + ∫ 2π x dx  = ⋅  − π + π + π 3 − π 2 + π  . (130)
 0 e −1 0
e − 1  16 2  4 2 2 8 8 8 

With regard the Section 2, we have the following mathematical connections between the eqs. (24)
and (27) and the eqs. (123) and (124):

 1 µνρ R
TLR(d + e )µνρ = 2ig (16πG )∫ dx  Vqkp ( ) ( ) ( ) ( )
I 2 (0 ) + g µν q ρ − q µ gνρ I 2R Mq2 + gνρ k µ − kν g ρµ I 2R Mk2 +
2
( ) ( )]
+ g ρµ pν − p ρ g µν I 2R Mp2 ⇒
 ∞ cos πx 2 ∞ sin πx 2  1  π 3 
⇒ π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  ; (131)
 0 sinh πx 0 sinh πx  16  4 

24
  1 µνρ R
TLR(d + e )µνρ = 2ig (16πG )∫ dx  Vqkp (
I 2 (0 ) + g µν q ρ − q µ gνρ I 2R Mq2 + gνρ k µ − kν g ρµ I 2R Mk2 + ) ( ) ( ) ( )
2
(
+ g ρµ pν − p ρ g µν I 2R Mp2 ⇒ ) ( )]
 ∞ x cos 2πx ∞ x cos 2πx
2
 1 1π3 3 2 5 1 5 5 
⇒ π  ∫ 2π x
3
dx + ∫ 2π x dx  = ⋅  − π + π + π 3 − π 2 + π  ; (132)
 e0
−1 0
e − 1  16 2  4 2 2 8 8 8 

2 
1 2 − µs M c2  µ s2 
βκ = − b g 3
+ g κ  ln + 1 − γ + y0
 M 2  ⇒
16π 2 16π 2  µ s2
0 w
 c 
 ∞ cos πx 2
∞ sin πx 2
 1 π 3

⇒ π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  ; (133)
 0 sinh πx 0 sinh πx  16  4 

2 
1 2 − µs M c2  µ s2 
βκ = −
b g 3
+ g κ  ln + 1 − γ + y0
 M 2  ⇒
16π 2 16π 2  µ s2
0 w
 c 
 ∞ x cos 2πx ∞ x cos 2πx
2
 1 1π 3
3 5 1 5 5 
⇒ π 3  ∫ 2π x dx + ∫ 2π x dx  = ⋅  − π 2 + π + π 3 − π 2 + π  . (134)
 e0
−1 0
e − 1  16 2  4 2 2 8 8 8 

Thence, mathematical connections with the Ramanujan’s modular equations that are related to the
physical vibrations of the superstrings and with the Ramanujan’s identities concerning π.

With regard the Section 4, we have the following mathematical connections between the eqs. (85)
and (95b), that derive from the (95), and the eqs. (114b), (115b), (118b), (123) and (124):

− ( 2 β ) y12
− β 2 ( y 22 + y 32 ) 1
= ∫ d 2 xe − β ( y 2 + y 3 ) ( )
1 2
1 −1
∫ ∫ 2π 2 β =
2 2 2
2 2
d xe dy1e
3 3
− ( 2β ) y 2 − ( )2 y 22
( ) ( )
1 2 2 1
1 −1 1 3 −1 2β
= 2π 2 β ∫ dx3e − β y 3 ∫ dx2e 2 = 2π 2 β ∫ dx3e − β y 3 ∫ dy2e 2 =
2 2 2 2

3 3 2

=
1
3
3
2
(
(2π ) 2 β ) ∫ dx e
−2
3
− β 2 y 32
=
1
3
31
22
(2π ) 2 β ( ) ∫ dy e
−2
3
− β 2 y 32
=
1
3
31
22
(2π )2
3
( 2β )
−3

( ) π3
2

=  ∫ d 2 xe− β ( y 2 + y 3 )
 1 −1  1
2π 2β
2 2 2
= × ⇒
( )
 8 6
 3  4 2β
∞ xdx π π 1 2 3 ∞
⇒ π 2 ⋅ 4a 2 ∫ = − + aπ 2 ∑ 2
0 (e 2πx
)(
− 1 4a + x
4 4
4a 4 )
k =1 a + (a + k )
2
; (135)

( )2 y12
= ∫ d 2 xe − β ( y 2 + y 3 ) ( )
1
(
− β 2 y 22 + y 32 ) 1 − 2β 1
2π 2β
−1
∫ d xe ∫ dy1e =
2 2 2
2 2
3 3
( )2 y 22 ( )2 y 22
( ) ∫ dx e ∫ dx e ( ) ∫ dx e
1 1
1 −1 − 2β 1 3 −1 − 2β
− β 2 y 32 − β 2 y 32
= 2π 2β 3 2
2
= 2π 2β 3 ∫ dy e 2
2
=
3 3 2

(2π )( 2 β ) ∫ dx e (2π ) 2 β ( ) ∫ dy e (2π )2 ( )

1 3 −2 − β 2 y 32 1 31 −2 − β 2 y 32 1 31 3 −3
= = 3 3 = 2β
3 2 3 22 3 22

25
 ( ) π3
2

=  ∫ d 2 xe− β ( y 2 + y 3 )
 1 −1  1
2π 2β
2 2 2
= × ⇒
( )
 8 6
 3  4 2β
sin (2nx )dx π3 ∞
(− 1) e− (2 k +1)n cos{(2k + 1)n} ; k
∞
⇒π 2
∫ = − 2π 2 ∑ (136)
0 x(cosh (πx ) + cos(πx )) 4 k = 0 (2k + 1) cosh{(2k + 1)π / 2}

− ( 2 β ) y12
− β 2 ( y 22 + y 32 ) 1
= ∫ d 2 xe − β ( y 2 + y 3 ) ( )
1 2
1 −1
∫ ∫ 2π 2 β =
2 2 2
2 2
d xe dy1e
3 3
− ( 2β ) y 2 − ( )2 y 22
( ) ( )
1 2 2 1
1 −1 1 3 −1 2β
= 2π 2 β ∫ dx3e − β y 3 ∫ dx2e 2 = 2π 2 β ∫ dx3e − β y 3 ∫ dy2e 2 =
2 2 2 2

3 3 2

=
1
3
3
2
(2π ) 2 β ( ) ∫ dx e
−2
3
− β 2 y 32
=
1
3
31
22
(2π ) 2 β ( ) ∫ dy e
−2
3
− β 2 y 32
=
1
3
31
22
(2π )2
3
( 2β ) −3

( ) π3
2

=  ∫ d 2 xe− β ( y 2 + y 3 )
 1 −1  1
2π 2β
2 2 2
= × ⇒
( )
 8 6
 3  4 2β
 1 + 1 + 4x 
log 

15 1 
⇒ I = π∫
2  dx = 15 π ( 5 +1 / 2) log u
( ) π3
4 ∫1
2u − 1 du = ; (137)
4 0 x u2 − u 4

( )2 y12
= ∫ d 2 xe − β ( y 2 + y 3 ) ( )
1
(
− β 2 y 22 + y 32 ) 1 − 2β 1
2π 2β
−1
∫d 3∫
=
2 2 2
2 2
xe dy e 1
3
( )2 y 22 ( )2 y 22
( ) ∫ dx e ∫ dx e ( ) ∫ dx e
1 1
1 −1 − 2β 1 3 −1 − 2β
− β 2 y 32 − β 2 y 32
= 2π 2β 3 2
2
= 2π 2β 3 ∫ dy e 2
2
=
3 3 2

(2π )( 2 β ) ∫ dx e (2π ) 2 β ( ) ∫ dy e (2π )2 ( )

1 3 −2 − β 2 y 32 1 31 −2 − β 2 y 32 1 31 3 −3
= = 3 3 = 2β
3 2 3 22 3 22

( ) π3
2

=  ∫ d 2 xe− β ( y 2 + y 3 )
 1 −1  1
2π 2β
2 2 2
= × ⇒
( )
 8 6
 3  4 2β
 ∞ cos πx 2 ∞ sin πx 2  1  π 3 
⇒ π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  ; (138)
 0 sinh πx 0 sinh πx  16  4 

( )2 y12
= ∫ d 2 xe − β ( y 2 + y 3 ) ( )
1
(
− β 2 y 22 + y 32 ) 1 − 2β 1
2π 2β
−1
∫d 3∫
=
2 2 2
2 2
xe dy e 1
3
( )2 y 22 ( )2 y 22
( ) ∫ dx e ∫ dx e ( ) ∫ dx e
1 1
1 −1 − 2β 1 3 −1 − 2β
− β 2 y 32 − β 2 y 32
= 2π 2β 3 2
2
= 2π 2β 3 ∫ dy e 2
2
=
3 3 2

(2π )( 2 β ) ∫ dx e (2π ) 2 β ( ) ∫ dy e (2π )2 ( )

1 3 −2 − β 2 y 32 1 31 −2 − β 2 y 32 1 31 3 −3
= = 3 3 = 2β
3 2 3 22 3 22

( ) π3
2

=  ∫ d 2 xe− β ( y 2 + y 3 )
 1 −1  1
2π 2β
2 2 2
= × ⇒
( )
 8 6
 3  4 2β
 ∞ x cos 2πx ∞ x cos 2πx
2
 1 1π3 3 5 1 5 5 
⇒ π 3  ∫ 2π x dx + ∫ 2π x dx  = ⋅  − π 2 + π + π 3 − π 2 + π  . (139)
 e0
−1 0
e − 1  16 2  4 2 2 8 8 8 

26
  3 − 12 (2 β )2 (y12 + y 22 + y 32 )
2
 π3
∫ d xe ( B y 2
+ B y 2
+ B y 2
+ B y y + B y y + B)
6 2 3 
y y = ×
81
⇒

1 1 2 2 3 3 4 1 2 5 1 3
 (
4 16 β 5 )2

∞ xdx π2 π3 ∞
1
⇒ π 2 ⋅ 4a 2 ∫ 2πx
( 0 e )( )
− 1 4a + x
4 4
= −
4a 4
+ a π 2
∑
k =1 a + (a + k )
2 2
; (140)

 3 − 12 (2 β )2 (y12 + y 22 + y 32 )
2
 π3
∫ d xe ( B y 2
+ B y 2
+ B y 2
+ B y y + B y y + B6 2 3 
y y )= ×
81
⇒

1 1 2 2 3 3 4 1 2 5 1 3
 4 16 β 5 ( )2

sin (2nx )dx π3 ∞

(− 1) e− (2 k +1)n cos{(2k + 1)n} ;
k
∞
⇒π 2∫ = − 2π 2 ∑ (141)
0 x(cosh (πx ) + cos(πx )) 4 k = 0 (2k + 1) cosh{(2k + 1)π / 2}

 3 − 12 (2 β )2 (y12 + y 22 + y 32 )
2
 π3
∫ d xe ( B y 2
+ B y 2
+ B y 2
+ B y y + B y y + B y)
6 2 3 
y = ×
81
⇒

1 1 2 2 3 3 4 1 2 5 1 3
 (
4 16 β 5 )2

 1 + 1 + 4x 
log 

 dx = 15 π ( 5 +1)/ 2 log u (2u − 1)du = π ; (142)
3
15 1  2
⇒ I = π∫
4 0 x 4 ∫1 u2 − u 4

 3 − 12 (2 β )2 (y12 + y 22 + y 32 )
2
 π3
 ∫ d xe ( 2 2 2
)
B1 y1 + B2 y2 + B3 y3 + B4 y1 y2 + B5 y1 y3 + B6 y2 y3  = ×
81
⇒
  (
4 16 β 5 )2

 ∞ cos πx 2 ∞ sin πx 2  1  π 3 
⇒ π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  ; (143)
 0 sinh πx 0 sinh πx  16  4 

 3 − 12 (2 β )2 (y12 + y 22 + y 32 )
2
 π3
∫ d xe ( B y 2
+ B y 2
+ B y 2
+ B y y + B y y + B y y )
6 2 3  = ×
81
⇒

1 1 2 2 3 3 4 1 2 5 1 3
 (
4 16 β 5 2 )
 ∞ x cos 2πx ∞ x cos 2πx
2
 1 1π3 3 5 1 5 5 
⇒ π 3  ∫ 2π x dx + ∫ 2π x dx  = ⋅  − π 2 + π + π 3 − π 2 + π  . (144)
 e0
−1 0
e − 1  16 2  4 2 2 8 8 8 

π3
Thence, mathematical connections with the Ramanujan’s identities concerning π, precisely .
4

3.2 Zeta Strings [8]

The exact tree-level Lagrangian for effective scalar field ϕ which describes open p-adic string
tachyon is
1 p2  1 − 2 


1
Lp = 2  − ϕp ϕ+ ϕ p +1 
g p −1  2 p +1  , (145)

= −∂ t2 + ∇ 2
where p is any prime number, is the D-dimensional d’Alambertian and we adopt
metric with signature (− +... + ) . Now, we want to show a model which incorporates the p-adic
string Lagrangians in a restricted adelic way. Let us take the following Lagrangian

27
 n −1 1  1 1 n +1 


−
L = ∑ Cn Ln = ∑ L = 2 
− φ ∑ n 2
φ +∑ φ 
n ≥1 n + 1
2 n
n ≥1 n ≥1 n g  2 n ≥1 . (146)

Recall that the Riemann zeta function is defined as

ζ (s ) = ∑
1 1
=∏ −s
p 1− p
s
n ≥1 n
, s = σ + iτ , σ > 1 . (147)

Employing usual expansion for the logarithmic function and definition (147) we can rewrite (146)
in the form
1 1 
 
L = − 2  φζ  φ + φ + ln (1 − φ )
g 2  2   , (148)



ζ 
φ <1  2  acts as pseudodifferential operator in the following way:
where .


 ixk  k 2 ~
ζ  φ (x ) = e ζ  − φ (k )dk
1
(2π )D ∫
 r
2  2 ,
− k 2 = k02 − k 2 > 2 + ε , (149)

φ (k ) = ∫ e(−ikx )φ ( x )dx
~
where is the Fourier transform of φ ( x ) .
Dynamics of this field φ is encoded in the (pseudo)differential form of the Riemann zeta function.
When the d’Alambertian is an argument of the Riemann zeta function we shall call such string a
“zeta string”. Consequently, the above φ is an open scalar zeta string. The equation of motion for
the zeta string φ is


  k2 ~ φ
eixk ζ  − φ (k )dk =
1
ζ  φ =
(2π )D ∫
r
2 k 02 − k 2 > 2 + ε
 2 1−φ
(150)

which has an evident solution φ = 0 .

For the case of time dependent spatially homogeneous solutions, we have the following equation of
motion
 − ∂2  − ik 0 t  k0  ~
2
φ (t )
ζ  t φ (t ) = φ (k0 )dk0 =
1
∫ e ζ 
 2  (2π ) k 0 > 2 +ε 
 2 1 − φ (t )
. (151)

With regard the open and closed scalar zeta strings, the equations of motion are

n ( n −1)

 ixk  k2 ~
e ζ  − φ (k )dk = ∑θ 2 φ n
1
ζ  φ = D ∫
2 (2π )  2 n ≥1
, (152)

28
  n 2 n(n − 1) n (n2−1) −1 n +1 


ζ  θ =
1 ixk 
D ∫e ζ −

k2 ~
θ (k )dk = ∑ θ +
2(n + 1)
θ φ −1  ( )
4 (2π )  4 n ≥1   , (153)

and one can easily see trivial solution φ = θ = 0 .

With regard the Section 3, we have the following mathematical connection between the eq. (54b)
and the eq. (150):


  k2 ~ φ
eixk ζ  − φ (k )dk =
1
ζ  φ =
(2π )D ∫
r
2 k 02 − k 2 > 2 + ε
 2 1−φ
⇒

⇒ ρ = pT
Vol S d − 2 ( ) ∞ x d −1  2
(1−ξ ) / 2
ζ (d )(d − 1)!
∫ = p 1 − d 
d
dx x Td . (154)
(2π )d −1 0 e −ξ  2  2 d −2  d − 1  (d −1) / 2
Γ π
 2 

With regard the Section 4, we have the following mathematical connections between the eqs. (74)
and (77) and the eqs. (150):

(4π )3 Iˆ
5/ 2
d 2Ω 3π
Z lat × 2ζ (6 )Ω 32 + ζ (5) + π 3 Ω 2 ∑ K 5 / 2 (2πΩ 2 m1m2 )e 2πim1 m2 Ω1 ⇒
m1
=∫
Ω2 4Ω 2
1 2 2
4 FL
m1 ≠ 0 , m 2 ≠ 0 m2


  k2 ~ φ
∫k02 − kr 2 > 2 +ε e ζ  − 2 φ (k )dk = 1 − φ
1
ζ  φ = ixk

⇒  2 (2π )
D
; (155)

3πζ (5) 
πT2
∞ dΩ 2 
2
− j + pU
E3 (Ω, Ω )  E (U ,U )
SL ( 2 , Z ) SL ( 2 , Z )
∑ e  2ζ (6 )T23 +
1/ 2 Ω 2U 2
Iˆ13 = V2 ∫ dΩ1 ∫ 2
= 3  2  3
⇒
−1 / 2 0 Ω2
2 ( j , p )≠ ( 0 , 0 ) π  4T2 

  k2 ~ φ
eixk ζ  − φ (k )dk =
1
ζ  φ =
(2π )D ∫
r
k 02 − k 2 > 2 + ε 1−φ
⇒    2
2
. (156)

1 6
In conclusion, multiplying for π the eq. (77), we obtain the following equivalent equation:
8
3πζ (5) 
πT
∞ dΩ π3 
2
− j + pU
( )  E3 (U ,U )
2
SL ( 2 , Z ) SL ( 2 , Z )
 2ζ (6 )T23 +
1
∑
1/ 2
Iˆ13 = π 6 × V2 ∫ dΩ1 ∫ 2
E Ω , Ω e Ω 2U 2
=
8 −1 / 2 0 Ω 2 3
4  2
4T2 
2 ( j , p )≠ ( 0 , 0 )
(157)
This equation can be related with the expression (138) and with the eq. (150), giving the following
interesting complete mathematical connections:

3πζ (5) 
πT2
∞ dΩ π3 
2
− j + pU
E3 (Ω, Ω )  E (U ,U )
SL ( 2 , Z ) SL ( 2 , Z )
 2ζ (6 )T23 +
1
∑ e
1/ 2
Iˆ13 = π 6 × V2 ∫ dΩ1 ∫ 2 Ω 2U 2
= 2  3
⇒
8 −1 / 2 0 Ω 2
4  4T2 
2 ( j , p )≠ ( 0 , 0 )
( )2 y12
( ) 1 = ∫ d 2 xe − β ( y 2 + y 3 ) ( )
1
− 2β 1 −1
− β 2 y 22 + y 32
⇒ ∫ d xe ∫ dy1e 2π 2β =
2 2 2
2 2
3 3
( )2 y 22 ( )2 y 22
( ) ∫ dx e ( ) ∫ dx e
1 1
1 −1 − 2β 1 3 −1 − 2β
− β 2 y 32 − β 2 y 32
= 2π 2β 3 ∫ dx e 2
2
= 2π 2β 3 ∫ dy e
2
2
=
3 3 2

29
 =
1
3
3
2
(
(2π ) 2 β ) ∫ dx e
−2
3
− β 2 y 32
=
1
3
31
22
(
(2π ) 2 β ) ∫ dy e
−2
3
− β 2 y 32
=
1
3
31
22
(2π )2
3
( 2β )
−3

( ) π3
2

=  ∫ d 2 xe− β ( y 2 + y 3 )
 1 −1  1
2π 2β
2 2 2
= × ⇒
( )
 8 6
 3  4 2β
 ∞ cos πx 2 ∞ sin πx 2  1  π 3 
⇒ π 3  ∫ x 3 dx + ∫ x 3 dx  =  − 3π + π 2  ⇒
 0 sinh πx 0 sinh πx  16  4 

  k ~
2
φ
eixk ζ  − φ (k )dk =
1
ζ  φ = D ∫k 2 − kr 2 > 2 + ε
⇒  
2 (2π ) 0  2 1−φ
. (158)

Acknowledgments

I would like to thank Prof. Branko Dragovich of Institute of Physics of Belgrade (Serbia) for his
availability and friendship. I’ld like to thank Prof. Jorma Jormakka of Department of Military
Technology, National Defence University, Helsinki, for his availability. In conclusion, I’ld like to
thank also Dr. Pasquale Cutolo for his important and useful work of review and his kind advices.

References

[1] David J. Toms – “Quantum gravitational contributions to quantum electrodynamics” – arXiv:

1010.0793v1 [hep-th] – 5 Oct 2010

[2] Yong Tang and Yue-Liang Wu – “Gravitational Contributions to the Running of Gauge
Couplings” – arXiv:0807.0331v2 [hep-ph] – 10 Dec 2008

[3] Sean Patrick Robinson – “Two Quantum Effects in the Theory of Gravitation” – Massachusetts
Institute of Technology – June 2005

[4] Edward Witten – “Supersymmetric Yang-Mills Theory on a four-manifold” – arXiv:hep-th/

9403195v1 – 31 Mar 1994;
Massimo Bianchi and Stefano Kovacs – “Yang-Mills Instantons vs. type IIB D-instantons” –
arXiv:hep-th/9811060v2 – 27 Nov 1998;
Anirbau Basu – “The D 6 R 4 term in type IIB string theory on T 2 and U-duality” – arXiv:0712.
1252v1 [hep-th] – 8 Dec 2007

[5] Jorma Jormakka – “Solutions to Yang-Mills Equations” – arXiv:1011.3962v2 [math.GM]

15 Nov 2010 - http://arxiv.org/PS_cache/arxiv/pdf/1011/1011.3962v2.pdf

[6] Bruce C. Berndt, Heng Huat Chan, Sen-Shan Huang – “Incomplete Elliptic Integrals in
Ramanujan’s Lost Notebook” - American Mathematical Society, Providence, RI, 2000, pp. 79-
126.

[7] Bruce C. Berndt – “Ramanujan’s Notebook” – Part IV – Springer-Verlag (1994).

[8] Branko Dragovich – “Zeta Strings” – arXiv:hep-th/0703008v1 – 1 Mar 2007.

30