Bessel equation

Jan Dereziński

Department of Mathematical Methods in Physics

Faculty of Physics
University of Warsaw
Pasteura 5, 02-093 Warszawa, Poland
June 8, 2020

1 Introduction
Consider the Laplace operator on R2

∆2 = ∂x2 + ∂y2

and the Helmholtz

−∆2 F (x, y) = EF (x, y). (1.1)
Introduce the polar coordinates

x = r cos φ, y = r sin φ,
p y
r = x2 + y 2 , φ = arctan .
x
The 2-dimensional Laplacian in polar coordinates is
1 1
−∆2 = −∂r2 − ∂r − 2 ∂φ2 . (1.2)
r r
We make an ansatz
F (x, y) = v(r)u(φ).
The equation (1.1) becomes
 1 1 
− ∂r2 − ∂r − 2 ∂φ2 v(r)u(φ) = Ev(r)u(φ). (1.3)
r r
We divide both sides by v(r)u(φ) and multiply by r2 . We put functions depend-
ing on r on one side and on φ on the other side. We obtain

∂φ2 u(φ)


r2 ∂r2 + r∂r + E v(r)
=− = C, (1.4)
v(r) u(φ)

1
where C does not depend on r or φ. We obtain an equation for u:

∂φ2 u(φ) = Cu(φ), (1.5)

solved by linear combinations of eimφ and e−imφ with C = m2 . Condition

u(φ) = u(φ + 2π) implies that m ∈ Z. Thus we obtain

r2 ∂r2 + r∂r − m2 + Er2 v(r) = 0.

(1.6)
√
Now if E < 0, then we change the variables z = −Er obtaining the modified
Bessel equation

z 2 ∂z2 + z∂z − m2 − z 2 v = 0.


(1.7)
√
Now if E > 0, then we change the variables z = Er obtaining the (standard)
Bessel equation

z 2 ∂z2 + z∂z − m2 + z 2 v = 0.


(1.8)

Certain distinguished solutions of (1.7) are denoted Im , Km , and of (1.8) are

±
denoted Jm and Hm . We call them jointly the Bessel family. They are probably
the best known and the most widely used special functions in mathematics and
its applications.
The parameter m in the above analysis was an integer. However, if we
consider the Laplacian on the surface of the cone described by 0 ≤ φ < α, then
the condition u(φ) = u(φ + α) leads to mα 2π ∈ Z. Hence non-integer values of
the parameter m are also relevant in applications.
We will see that the Helmholtz equation in any dimension d

−∆d F = EF (1.9)

leads in spherical coordinates to the Bessel equation. The d-dimensional Lapla-

cian in spherical coordinates is given by
d−1 1
−∆d = −∂r2 − ∂r − 2 ∆Sd−1 ,
r r
where r is the radial coordinate and ∆Sd−1 is the Laplace-Beltrami operator on
the sphere Sd−1 . Eigenvalues of −∆Sd−1 for d = 2, 3, . . . are

l(l + d − 2), l = 0, 1, 2, . . . (1.10)

where l corresponds to the order of spherical harmonics. Setting F = v(r)u(Ω),

where Ω are the angular coordinates, we obtain the radial part of the Helmholtz
equation

r2 ∂r2 + (d − 1)r∂r − l(l + d − 2) + Er2 v(r).

(1.11)

By the same scaling argument as above, we can reduce ourselves to the case
E = ±1. We will see that (1.11) is equivalent to (1.7) or (1.8).

2
 The operator in (1.7)/(1.8) can be transformed as
d d
r1− 2 r2 ∂r2 + r∂r ∓ r2 − m2 r−1+ 2


(1.12)
 d  d 
= r2 ∂r2 + (d − 1)r∂r ∓ r2 − m − + 1 m + − 1 . (1.13)
2 2
Setting
d
m := l + − 1, (1.14)
2
we can rewrite (1.13) as

r2 ∂r2 + (d − 1)r∂r ∓ r2 − l(l + d − 2), (1.15)

which is the radial part of the Helmholtz equation in dimension d for spherical
harmonics of order l, see (1.11). (1.15) is sometimes called the d-dimensional
Bessel equation.
Note that in even dimensions the parameter m takes integer values and in
odd dimensions it takes half-integer values, see (1.14).
Note special cases of (1.13):

r−m r2 ∂r2 + r∂r ∓ r2 − m2 rm

= r2 ∂r2 + (1 + 2m)r∂r ∓ r2 , (1.16)

1
1
r 2 r2 ∂r2 + r∂r ∓ r2 − m2 r− 2
 

1
= r2 ∂r2 ∓ 1 + 1/4 − m2 2 .
r

Here are some other operators related to the Bessel equation: Set r := tδ ,
so that ∂r = 1δ t1−δ ∂t . Then
1 1
r 2δ r2 ∂r2 + r∂r ∓ r2 − m2 r− 2δ


(1.17)
1
  1
= t 2 δ −2 t2 ∂t2 + δ −2 t∂t ∓ t2δ − m2 t− 2 (1.18)
 1 1
= δ −2 t2 ∂t2 ∓ (δtδ−1 )2 + − m2 δ 2 2 . (1.19)
4 t
If we set r = et , so that r∂r = ∂t , then

r2 ∂r2 + r∂r ∓ r2 − m2 = ∂t2 ∓ e2t − m2 . (1.20)

2 Modified Bessel equation

2.1 Integral representations
The modified Bessel equation is given by the operator

Im (z, ∂z ) := z 2 ∂z2 + z∂z − z 2 − m2 .

3
 γ
Theorem 2.1 Bessel–Schläfli type representations Let ]0, 1[3 τ 7→ t(τ ) be a
contour such that
z  z  t(1)
(t − t−1 ) + m exp (t + t−1 ) t−m  = 0, (2.21)

2 2 t(0)

Then Z z 
exp (t + t−1 ) t−m−1 dt (2.22)
γ 2
is a solution of the modified Bessel equation

Proof. We differentiate the integral with respect to the parameter z:

Z z 
(z 2 ∂z2 + z∂z − z 2 − m2 ) exp (t + t−1 ) t−m−1 dt
γ 2
Z   2 
z 2 z z 
t + t−1 + t + t−1 − z 2 − m2 exp (t + t−1 ) t−m−1 dt



=
γ 2 2 2
Z  2 
z 2 z  z 
t − t−1 + t + t−1 − m2 exp (t + t−1 ) t−m−1 dt



=
γ 2 2 2
Z  
z   z  
= ∂t (t − t−1 ) + m exp (t + t−1 ) t−m dt
γ 2 2
z  z  t(1)
= (t − t−1 ) + m exp (t + t−1 ) t−m  = 0.

2 2 t(0)

γ
Theorem 2.2 Poisson type representations. Let ]0, 1[3 τ 7→ t(τ ) be a contour
such that t(1)
1
(1 − t2 )m+ 2 ezt  = 0.

t(0)

Then Z
1
m
z (1 − t2 )m− 2 ezt dt
γ

is a solution of the modified Bessel equation.

Proof. We use (1.16).

Z
1
z∂z2 + (1 + 2m)∂z − z (1 − t2 )m− 2 ezt dt


γ
Z
1
= (1 − t2 )m− 2 (zt2 + (1 + 2m)t − z)ezt dt
γ
Z  
1
= − ∂t (1 − t2 )m+ 2 ezt dt = 0.
γ

4
2.2 Modified Bessel function
The modified Bessel equation has a regular-singular point at 0 with the indicial
equation
λ(λ − 1) + λ − m2 = 0.
Its indices at 0 are equal to ±m.
Therefore, we should look for a solution of the modified Bessel equation in
the form
 ∞
X
0 = z 2 ∂z2 + z∂z − z 2 − m2 cn z m+n (2.23)
n=0
∞
X   ∞
X
= cn (m + n)2 (m + n − 1) + m + n − m2 z n − cn z n+2 . (2.24)
n=0 n=0

This leads to the recurrence relation

cn (2m + n)n = cn−2 . (2.25)

The initial condition c−1 = 0 together with (2.25) implies that cn = 0 for n
odd. For even subscripts, we can rewrite (2.25) in the form

c2n (2m + 2n)2n = c2(n−1) . (2.26)

With c0 = 1, this is solved by

1
c2n = .
22n (m + 1) · · · (m + n)n!
1
Multiplying this with Γ(m+1) , we obtain

1
c2n = .
22n Γ(m + n + 1)n!
The resulting function is called the modified Bessel function:
∞ z 2n+m


X
2
Im (z) = .
n=0
n!Γ(m + n + 1)

It is a solution of the modified Bessel equation with the parameter ±m. Note
1
that Γ(m+1) 6= 0 for m 6= −1, −2, . . . For m 6= −1, −2, . . . the function Im is
the unique solution of the modified Bessel equation satisfying
 z m 1
Im (z) ∼ , z ∼ 0,
2 Γ(m + 1)
which can be treated as a definition of the modified Bessel function. (By f (z) ∼
g(z), z ∼ 0, we understand that fg(z)
(z)
is analytic around zero and at zero equals
1).

5
 If m 6∈ Z, then I−m (z) and Im (z) are linearly independent and span the
space of solutions of the modified Bessel equation.
We have

Im (e±iπ z) = e±iπm Im (z), (2.27)

Im (z) = Im (z). (2.28)

In particular, Im (x) is real for x > 0, m ∈ R.

2.3 Integral representations of modified Bessel function

Theorem 2.3 (Bessel-Schläfli-type representations.) Let Rez > 0. Then
Z
1 z 
Im (z) = exp (t + t−1 ) t−m−1 dt (2.29)
2πi ]−∞,0+ ,−∞[ 2
z2
Z  
1  z m
= exp s + s−m−1 ds. (2.30)
2πi 2 ]−∞,0+ ,−∞[ 4s

We also have
Z
1 z 
I−m (z) = exp (t + t−1 ) t−m−1 dt. (2.31)
2πi [(0−0)+ ] 2

Here, the contour starts at 0 from the negative side on the lower sheet, encircling
0 in the positive direction and ends at 0 from the negative side on the upper sheet.
One of concrete realizations of (2.29) is the Schläfli representation:
Z π Z ∞
1 1
Im (z) = e z cos φ
cos(mφ)dφ − sin(mπ) e−z cosh β−mβ dβ. (2.32)
2π −π π 0

Proof. To see this note that by Thm 2.1, the RHS of (2.29) is a solution of the
modified Bessel equation. Besides,
z2
Z  
1  z m
exp s + s−m−1 ds
2πi 2 ]−∞,0+ ,−∞[ 4s

is holomorphic around zero. By the Hankel identity

Z
1 1
= exp (s) s−m−1 ds.
Γ(m + 1) 2πi ]−∞,0+ ,−∞[

m
Therefore, the RHS of (2.29) behaves at zero as ∼ z2 1
Γ(m+1) . Therefore, it
coincides with Im (z), at least for z 6∈ {. . . , −2, −1}. By continuity, it is Im also
for z ∈ {. . . , −2, −1}.
To see (2.31), we make a substitution t = s−1 in (2.29) noting that dt =
−2
−s ds, and then we change the orientation of the contour.
To see (2.32), we take a contour consisting of three pieces: −e−β : β ∈
] − ∞, 0], eiφ : φ ∈ [−π, π], −e−β : β ∈ [0, −∞[. 2

6
Theorem 2.4 (Poisson-type representations.) We have the Poisson rep-
resentation
1  z m Z 1 1 1
Im (z) = √   (1 − t2 )m− 2 ezt dt, m>− . (2.33)
πΓ m + 1 2 −1 2
2

The following representation is due to Hankel:

Γ( 21 − m)  z m
Z
1 1
Im (z) = √ (t − 1)m− 2 (t + 1)m− 2 ezt dt. (2.34)
2πi π 2
[1,−1− ,1+ ]

Proof. By Thm 2.2, the RHS of (2.33) satisfies the modified Bessel equation.
Then we use the fact that
√
Γ(m + 21 ) π
Z 1
1
(1 − t2 )m− 2 dt = , (2.35)
−1 Γ(m + 1)

m
to see that the RHS of (2.33) behaves at zero as ∼ z2 1
Γ(m+1) .
Similarly, to show (2.34) we use
Z √
1 1 1 π
(t − 1)m− 2 (t + 1)m− 2 dt = . (2.36)
2πi [1,−1− ,1+ ] Γ( 21 − m)Γ(m + 1)
2

2.4 Modified Bessel function for integral parameters

For m ∈ Z the Bessel-type integrals (2.29), (2.30) and (2.32) simplify:

Theorem 2.5 For m ∈ Z we have

Im (z) = I−m (z).

Z
1 z  dt
Im (z) = exp (t + t−1 ) m+1 (2.37)
2πi [0+ ] 2 t
z2
Z  
1 z m
  ds
= exp s + m+1
(2.38)
2πi 2 [0+ ] 4s s
Z π
1
= ez cos φ cos(mφ)dφ. (2.39)
2π −π

Proof. We will give two proofs.

7
 The first is based on the power series. It is enough to assume that m =
0, 1, . . . .
∞ z 2n+m


X
2
Im (z) = (2.40)
n=0
n!(n + m)!

∞ z 2(n+m)−m


X
2
= (2.41)
n=0
(n + m)!(n + m − m)!

∞ z 2n−m


X
2
= (2.42)
n=m
n!(n − m)!

∞ z 2n−m


X
2
= = I−m (z). (2.43)
n=0
n!Γ(n − m + 1)

The second uses the contour integrals (2.29) and (2.31): we note that if
m ∈ Z, one can be deformed into the other. 2

Theorem 2.6 (Generating function.)

z  ∞
X
exp (t + t−1 ) = tm Im (z). (2.44)
2 m=−∞

Proof. Again, we will give two proofs.

The first uses power series:
∞
X X tm (z/2)m+2n
tm Im (z) =
m=−∞
n!(n + m)!
n≥0, n+m≥0
X X (z/2t)n (tz/2)m+n
=
n!(n + m)!
n≥0 n+m≥0

= ez/2t etz/2 .
The second notes that (2.44) is the Laurent series in t of a function holomor-
phic in C\{0}, and (2.37) is just the formula for the coefficient in the Laurent
series. 2

2.5 MacDonald function

We define the Macdonald function:
1 ∞
Z  z 
Km (z) := exp − (s + s−1 ) s±m−1 ds. (2.45)
2 0 2
Other names: the Basset function or the modified Bessel function of the second
kind. The integral (2.45) is absolutely convergent. Substitution s = t−1 shows
that m can be replaced by −m (and thus Km = K−m ).

8
Theorem 2.7 Km solves the modified Bessel equation. We have

Km (z) = Km (z).

Km (x) is real for x > 0, and m ∈ R or m ∈ iR.

Proof. We can write

e−iπm
Z z 
Km (z) = exp (t + t−1 ) t−m−1 dt (2.46)
2 ]−∞−i0,0] 2
iπm Z
e  z 
= exp (t + t−1 ) t−m−1 dt, (2.47)
2 ]−∞+i0,0] 2

which are integrals satisfying Theorem 2.1. (Note that the contours lie on
the boundary of the Riemann surface of the principal branch of t−m−1 , each
projecting onto ] − ∞, 0], the first is on the lower sheet and the second on the
upper sheet). 2

Theorem 2.8
π
K−m (z) = Km (z) = (I−m (z) − Im (z)). (2.48)
2 sin πm
Proof. We add the appropriate multiples of (2.46) and (2.47):

−4i sin(πm)Km (z) = −2eiπm Km (z) + 2e−iπm Km (z)

Z z 
= exp (t + t−1 ) t−m−1 dt
]−∞−i0,0] 2
Z z 
+ exp (t + t−1 ) t−m−1 dt
]0,−∞−i0] 2
Z z 
= exp (t + t−1 ) t−m−1 dt
]−∞,0+ ,−∞[ 2
Z z 
− exp (t + t−1 ) t−m−1 dt
(0−0)+ 2
= 2πi(Im (z) − I−m (z)).

where, as we recall, (0 − 0)+ is the contour starting at 0 from the negative side
on the lower sheet, encircling 0 in the positive direction and ending at 0 from
the negative side on the upper sheet. This proves (2.48). 2
It is useful to note that for positive z the contour in (2.45) can be turned by
π
2, losing however its absolute convergence at 0 or ∞. We thus obtain
π ∞
e±i 2 m
Z  
zi
Km (z) := exp − (s − s−1 ) s±m−1 ds. (2.49)
2 0 2

9
Theorem 2.9 Setting s = eθ in (2.45) and t = eφ in 2.49) we obtain

1 ∞ −z cosh θ −mθ
Z
Km (z) = e e dθ, (2.50)
2 −∞
π
e∓i 2 m ∞ ∓iz sinh φ −mφ
Z
= e e dφ. (2.51)
2 −∞

2.6 Analytic continuation of the MacDonald function

Theorem 2.10
1
∓iKm (e∓iπ z) ± ieiπm Km (z) .


Im (z) = (2.52)
π
Proof. Write
π
K(e−iπ z) = eiπm I−m (z) − e−iπm Im (z)


2 sin πm
We subtract from this eiπm times (2.48) obtaining
π
K(e−iπ z) − eiπm K(z) −e−iπm Im (z) + eiπm Im (z) = iπIm (z).


=
2 sin πm
This proves (2.52). 2
The function Km (z) is exponentially decaying whereas ∓iKm (e∓iπ z) are
exponentially exploding. It will be useful to have such “exploding partners”
for Km (z). Unfortunately, the natural domain for ∓iKm (e∓iπ z) is C\[0, ∞[,
which is inconvenient if we are primarily interested in [0, ∞[. It seems useful to
introduce the following function (which has no name and apparently does not
appear in the literature):

i
Km (eiπ z) − Km (e−iπ z) .


Xm (z) := (2.53)
2
Theorem 2.11 We have the identities
1

Im (z) = Xm (z) − sin(πm)Km (z) , (2.54)
π
π

Xm (z) = I−m (z) + I−m (z) . (2.55)
2

2.7 Asymptotics of the MacDonald function

Theorem 2.12 For | arg z| < π − ,

Km (z)
lim −z √ = 1.
|z|→∞ e √ π
2z

10
Proof. We use the steepest descent method. Set φ(t) := − 12 (t + t−1 ). We
compute
1
φ0 (t) = − (1 − t−2 ), φ00 (t) = −t−3 .
2
Hence φ has a critical point at t0 = 1 with φ(t0 ) = −1 and φ00 (t0 ) = −1. Thus
1 ∞ −m−1
Z
Km (z) = t exp(zφ(t))dt
2 0
Z ∞
φ00 (t0 )
 
1 2
' exp zφ(t0 ) + z (t − t0 ) dt
2 −∞ 2
√
1 −z ∞
Z z
2
 1 −z 2π
= e exp (t − 1) dt = e √ .
2 −∞ 2 2 z
2

Corollary 2.13 As x → ∞ we have

1
Im (x) ∼ ex . √ (2.56)
2πx
Next we derive the precise asymptotics of the MacDonald function
Theorem 2.14
∞
√ −z
X ( 21 − m)n ( 12 + m)n
Km (z) ∼ πe (−1)n 1 . (2.57)
n=0 n!(2z)n+ 2
Proof. To derive (2.57), at least formally, we first we transform the equation:
1
1
ez z 2 z 2 ∂z2 + z∂z − z 2 − m2 z − 2 e−z (2.58)
1
= z 2 ∂z2 − 2z 2 ∂z − m2 + . (2.59)
4
Acting with (2.59) on
X∞
cn z −n (2.60)
n=0
we obtain
∞ 
X  1  −n 
n(n + 1)cn z −n + 2ncn z −n+1 − m2 − cn z . (2.61)
n=0
4
This yields the recurrence relation
 1
2ncn = − (n − 1)n − m2 + cn−1 (2.62)
4
 1  1 
= − n − − m n − + m cn−1 . (2.63)
2 2
Therefore,
( 1 − m)n ( 12 + m)n
cn = (−1)n 2 . (2.64)
n!2n
2

11
2.8 More integral representations
Theorem 2.15 (Poisson-type representations.)
Km (z)
 z ±m √πΓ(∓m + 1 ) Z 1
= 2
ezt (1 − t2 )±m− 2 dt. (2.65)
2 2πi
]−∞,−1+ ,−∞[
 z m √ Z ∞
π 1 1
= e−sz (s2 − 1)m− 2 ds, m>− ; (2.66)
2 Γ(m + 12 ) 1 2
 z −m Γ(m + 1 ) Z ∞ 1
= √ 2 e−isz (s2 + 1)−m− 2 ds, m > 0. (2.67)
2 2 π −∞

Proof. The RHS of (2.65) solves the modified Bessel equation by Thm 2.2. Let
us check its behavior for Rez → ∞. We set t = zs − 1:
Z
1 1 1
ezt (t + 1)±m− 2 (−t + 1)±m− 2 dt (2.68)
2πi
]−∞,−1+ ,−∞[
 2 m− 12 e−z 1 Z
1
∼ e−s s±m− 2 ds (2.69)
z z 2πi
]−∞,−1+ ,−∞[
 2 ±m −z
e
= √ . (2.70)
z 2zΓ( 21 ∓ m)
−z √
Therefore, the RHS of (2.65) behaves as e √2z π . Therefore, it coincides with
Km (z).
(2.67) follows from (2.65) by setting t = is. 2

2.9 MacDonald function for integer parameters

Pn
Theorem 2.16 Set Hn := k=1 k1 . Then for m = 0, 1, 2, . . .
 z 
Km (z) = (−1)m+1 log + γ Im (z)
2
m−1 ∞
 z 2k−m (m − k − 1)! (−1)m X
1 X Hk + Hm+k  z 2k+m
+ (−1)k + .
2 2 k! 2 k!(m + k)! 2
k=0 k=0

Proof. Set
d 1 1
φ(z) := =− ∂z log Γ(z).
dz Γ(z) Γ(z)
Then
φ(−n) = (−1)n n!, n = 0, 1, 2, . . . ,

γ − Hn
φ(n + 1) = , n = 0, 1, 2, . . .
n!

12
Besides,
∞
z  X φ(m + k + 1)  z m+2k
∂m Im (z) = log Im (z) + .
2 k! 2
k=0

Hence for m = 0, 1, 2, . . .
∞
X Hm+k  z m+2k
  z 
∂n In (z) = log + γ Im (z) − , (2.71)

n=m 2 (m + k)!k! 2
k=0

m−1
  z  X (m − k − 1)!  z 2k−m
∂n In (z) = log + γ I−m (z) + (−1)m−k−1 (2.72)

n=−m 2 k! 2
k=0

∞
X H−m+k  z −m+2k
− . (2.73)
(−m + k)!k! 2
k=m

The last sum can be written as

∞  z m+2k
X Hk
− .
k!(k + m)! 2
k=m

We use the De L’Hopital rule:

d
π dm (I−m − Im (z))
Km (z) = d
2 dm sin πm
m
 
(−1) d  d 
= − In (z) − In (z)
 
2 dn n=−m dn n=m

Corollary 2.17 As x → 0, we have

1  x m
Im (x) ∼ , m 6= −1, −2, . . . ; (2.74)
Γ(m + 1) 2

− ln x2 − γ if



 m = 0,

Γ(m) 2
m
Km (x) ∼ 2 x if Rem ≥ 0, m 6= 0; (2.75)

 Γ(−m) x

m
2 2 if Rem ≤ 0, m 6= 0.

Clearly, for integer m

1
Im (z) = Zm (z). (2.76)
π

13
2.10 Relationship to hypergeometric type functions
The modified Bessel function and the hypergeometric functions 0 F1 and 1 F1 are
closely related:

1  z m  z2 
Im (z) = 0 F1 1 + m;
Γ(m + 1) 2 4
1  z  m  1 
= e−z 1 F1 m + ; 2m + 1; 2z .
Γ(m + 1) 2 2

The MacDonald function is closely related to the hypergeometric function

2 F0 : √
2π 1 1 1
Km (z) = √ e−z 2 F0 + m, − m; −; − .
z 2 2 2z

2.11 Recurrence relations

Theorem 2.18

2∂z Im (z) = Im−1 (z) + Im+1 (z), (2.77)

2mIm (z) = zIm−1 (z) − zIm+1 (z). (2.78)

2∂z Km (z) = −Km−1 (z) − Km+1 (z), (2.79)

2mKm (z) = −zKm−1 (z) + zKm+1 (z). (2.80)

Proof. Recall that

Z z 
2∂z Im (z) = exp (t + t−1 ) (t−m + t−m−2 )dt.
γ 2

Z z 
0 = 2 ∂t exp (t + t−1 ) t−m dt
γ 2
Z z 
= −2m exp (t + t−1 ) t−m−1 dt
γ 2
Z z  Z z 
+z exp (t + t−1 ) t−m dt − z exp (t + t−1 ) t−m−2 dt.
γ 2 γ 2

By (2.46), eiπm Km satisfy the same recurrence relations. 2

Corollary 2.19
m 
∂z (z m Im (z)) = z m Im−1 (z), or ∂z +
Im (z) = Im−1 (z),
z
 m
z −m Im (z) = z −m Im+1 (z), or ∂z −


∂z Im (z) = Im+1 (z).
z

14
Hence
 n
1
∂z z m Im (z) = z m−n Im−n (z), (2.81)
z
 n
1
∂z z −m Im (z) = z −m−n Im+n (z). (2.82)
z

Analoguous identities hold for Km (z).

2.12 Half-integral parameters

1
For m = 2 we have  
1 1
z 2 I 21 z − 2 = z 2 ∂z2 − 1 .
1
Hence, z − 2 e±z is annihilated by I 12 .

 z  12  1
1 sinh z 2 2
I 12 (z) = = sinh z (2.83)
2 Γ(1 + 12 ) z πz
 z − 12   12
1 2
I− 21 (z) = cosh z = cosh z, (2.84)
2 Γ(1 − 21 ) πz
 π  12
K± 21 (z) = e−z , (2.85)
2z
 π  12
X± 21 (z) = ez . (2.86)
2z
One way to derive (2.83) and (2.84) is to use
r r
2n+ 21 π n n+1 π
2 n!Γ(1/2 + n + 1) = 2 n!2 (1/2)n+1 = (2n + 1)!,
2 2
r r
1 π n n π
22n− 2 n!Γ(−1/2 + n + 1) = 2 n!2 (1/2)n = (2n)!.
2 2

15
 For k = 0, 1, 2, . . . we have
 2  21 1
1 k sinh z
I 12 +k (z) = z 2 +k ∂z (2.87)
π z z
 2  21 ∞
X 1 (−j)k
= (−2)k z 2 +2j−k
π j=0
(2j + 1)!
1  
= Z 12 +k (z) − (−1)k K 21 +k (z) ,
π
 2  21 1  1 k cosh z
I− 12 −k (z) = z 2 +k ∂z (2.88)
π z z
∞
 2  21 X 1 ( 1 − j)k
= (−2)k z − 2 +2j−k 2
π j=0
(2j)!
1  
= Z 12 +k (z) + (−1)k K 21 +k (z) ,
π
 π  12 1  1 k e−z
K±( 12 +k) (z) = z 2 +k − ∂z (2.89)
2 z z
∞
 π  12 X 1 ( 1 − n2 )k
= 2k z − 2 −k+n (−1)n 2 ,
2 n=0
n!
π 
= (−1)k I− 12 −k (z) − I 12 +k (z) ,
2
 π  12 1  1 k ez
X±( 12 +k) (z) = z 2 +k ∂z (2.90)
2 z z
∞
 π  12 X 1 ( 1 − n2 )k
= (−2)k z − 2 −k+n 2
2 n=0
n!
π  
= I 1 (z) + I 12 +k (z) .
2 − 2 −k
Note that in the sum for (2.87) the terms with j = 0, . . . , k − 1 vanish.

2.13 Wronskians
Recall that the Wronskian is defined as W (f, g) := f g 0 − f 0 g. The Wronskian
of two solutions of the modified Bessel equation satisfies
 1
∂z + W (z) = 0.
z
Hence W (z) is proportional to z1 . Using

1  z ±m 1  z ±m−1
0
I±m (z) ∼ , I±m (z) ∼ ,
Γ(±m + 1) 2 2Γ(±m) 2

16
we can compute the Wronskian of Im and I−m , and then other Wronskians:
2 sin πm
W (Im , I−m ) = − , (2.91)
πz
1
W (Km , Im ) = , (2.92)
z
2
W (Km , Xm ) = . (2.93)
z

3 Standard Bessel equation

3.1 Bessel equation
Replacing z with ±iz in the modified Bessel equation leads to the standard
Bessel equation, given by the operator
Jm (z, ∂z ) := z 2 ∂z2 + z∂z + z 2 − m2 .

3.2 Integral representations

Theorem 3.1 Bessel–Schläfli representations Let γ be a contour satisfying
z  z  1 γ(1)
−1 −1
(t + t ) + m exp (t − t ) m  = 0, (3.94)
2 2 t γ(0)
Then Z z  dt
C exp (t − t−1 ) m+1 (3.95)
γ 2 t
is a solution of the Bessel equation.
Theorem 3.2 Poisson type representations Let γ be a contour satisfying
1
γ(1)
(1 − t2 )m+ 2 eizt  = 0.

γ(0)

Then Z
1
zm (1 − t2 )m− 2 eizt dt
γ
is a solution of the Bessel equation.

3.3 Bessel function

The Bessel function is defined as
m
Jm (z) = e±iπ 2 Im (∓iz)
∞
2n+m
X (−1)n z 2
=
n=0
n!Γ(m + n + 1)
1 −iπ m m
2 K(−iz) − eiπ 2 K(iz)


= e
iπ

17
 Jm (e±iπ z) = e±imπ Jm (z).

Theorem 3.3 Let Rez > 0. (Bessel-Schläfli-type representations)

Z
1 z  dt
Jm (z) = exp (t − t−1 ) m+1
2πi ]−∞,0+ ,−∞[ 2 t
2
Z  
1 z
  m z ds
= exp s − m+1
.
2πi 2 +
]−∞,0 ,−∞[ 4s s

(Poisson-type representations)
 z m Z 1
1 1 1
Jm (z) = √   (1 − t2 )m− 2 eizt dt, m>− ,
πΓ m + 1 −1 2 2
2
Z
1  1  z m
 1 1
Jm (z) = √ Γ −m (t − 1)m− 2 (t + 1)m− 2 .
2πi π 2 2 [1,−1− ,1+ ]

3.4 Relationship to hypergeometric type functions

1  z m  z2 
Jm (z) = F
0 1 1 + m; −
Γ(m + 1) 2 4
1  z m −iz
  1 
= e 1 F1 m + ; 2m + 1; 2iz .
Γ(m + 1) 2 2

3.5 Bessel function for integer parameters

Let m ∈ Z.
Theorem 3.4
∞
2n+m
X (−1)n z2
Jm (z) = (−1)m J−m (z) = .
n=0
n!(n + m)!

Theorem 3.5
Z
1 z  dt
Jm (z) = exp (t − t−1 ) m+1
2πi [0+ ] 2 t
z2
Z  
1 z m
  ds
= exp s − m+1
2πi 2 [0+ ] 4s s
Z π
1
= cos(z sin φ − mφ)dφ.
π 0

18
3.6 Hankel functions
There are two Hankel functions. Both are analytic continuations of the Mac-
Donald function – one to the lower and the other to the upper part of the
complex plane:

± 2 ∓i π (m+1)
Hm (z) = e 2 Km (∓iz).
π
In the literature the usual (and less natural) notation for Hankel functions is
(1) + (2) −
Hm (z) = Hm (z), Hm (z) = Hm (z). (3.96)

Note the identities

π ±i π (m+1) ±
Km (z) = e 2 H (±iz), (3.97)
2
±
H−m (z) = e±mπi Hm±
(z), (3.98)
1 + −


Jm (z) = Hm (z) + Hm (z) , (3.99)
2
1 mπi +
e Hm (z) + e−mπi Hm −


J−m (z) = (z) , (3.100)
2
± ie∓mπi Jm (z) − iJ−m (z)
Hm (z) = ± . (3.101)
sin mπ
.
π
Theorem 3.6 The following asymptotic formulas are true for −π ± 2 +δ <
arg z < π ± π2 − δ, δ > 0:
±
Hm (z)
lim
12 = 1,
z→∞ 2 imπ iπ
πz e±iz e∓ 2 ∓ 4

Here is a more precise asymptotics:

√ ∞
2 πm π
X ( 12 − m)n ( 21 + m)n
Hm (z) ∼ √ e±iz∓i 2 ∓i 4
±
. (3.102)
πz n=0
n!(±2iz)n

3.7 Integral representations of Hankel functions

Let us first consider Bessel-Schläfli-type representations. For Rez > 0,
Z
1 z  dt
+
Hm (z) = − exp (t − t−1 ) m+1 ,
πi ]−∞,(0+1·0)− [ 2 t
Z
− 1  z  dt
Hm (z) = exp (t − t−1 ) m+1 .
πi ]−∞,(0+1·0)+ [ 2 t

By ] − ∞, (0 + 1 · 0)− [ we understand the contour starting at −∞, encircling 0

clockwise and reaching zero from the positive direction. The contour is located
on the upper halfplane.

19
 Similarly, by ] − ∞, (0 + 1 · 0)+ [ we understand the contour starting at −∞,
encircling 0 counterclockwise and reaching zero from the positive direction. The
contour is located on the lower halfplane.
Note that
z  z  1
lim (t + t−1 ) + m exp (t − t−1 ) m = 0,
t→0+1·0 2 2 t
where by t → 0 + 1 · 0 we denote the convergence to zero through positive values
of t (sometimes denoted by t → 0+ ). Hence the contours ] − ∞, (0 + 1 · 0)+ [ and
] − ∞, (0 + 1 · 0)− [ satisfy (3.94).
+
If 0 < arg z < π, then a good contour in the representation of Hm is [i∞, 0].
−
If −π < arg < 0, then for Hm one can use [−i∞, 0]. This leads to the represen-
tations valid for ±Imz ≥ 0:
1 ∞
Z  z  ds
±i π m ± ∓i π m ±
e 2 Hm (z) = e 2 H−m (z) = ± exp ±i (s + s−1 ) m+1 (3.103)
πi 0 2 s
1 ∞
Z
=± exp (±iz cosh(t) − mt) dt. (3.104)
πi −∞
Poisson type integral representations valid for Imz ≥ 0 and all m:
Γ( 21 − m)  z m
Z
1 1
±
Hm (z) = √ eizt (t − 1)m− 2 (t + 1)m− 2 dt.
πi π 2 ]i∞,∓1∓ ,i∞[

Poisson type representations valid for m ≥ − 12 .

 z m Z ∞
±i π m ± 2 1
e 2 Hm (z) = ± √ e±isz (s2 − 1)m− 2 ds.
2 i πΓ( 12 + m) 1

3.8 Hankel functions for integer parameters

± i2  z 
Hm (z) = Jm (z) ± log + γ Jm (z) (3.105)
π 2
m−1 ∞
i X  z 2k−m (m − k − 1)! i X (−1)k (Hk + Hm+k )  z 2k+m
∓ ∓ . (3.106)
π 2 k! π k!(m + k)! 2
k=0 k=0

3.9 Neumann function

Neumann function is defined as
1 −
+


Ym (z) = 2i Hm (z) − Hm (z)
cos πmJm (z)−J−m (z)
= sin πm .
We then have
+ −
Hm (z) = Jm (z) + iYm (z), Hm (z) = Jm (z) − iYm (z).

20
Theorem 3.7 For m ∈ Z we have
 
Ym (z) = π2 log( z2 ) + γ Jm (z)
m−1 ∞
(m−k−1)! z 2k−m (−1)k
− π1 1 z m+2k
P P

k! (2) − π k!(m+k)! ( 2 ) Hk + Hm+k .
k=0 k=0

3.10 Recurrence relations

Theorem 3.8 We have the identities

2∂z Jm (z) = Jm−1 (z) − Jm+1 (z),

2mJm (z) = zJm−1 (z) + zJm+1 (z).

Sometimes, more convenient are the following forms of the recurrence relations:
Corollary 3.9
m 
∂z (z m Jm (z)) = z m Jm−1 (z), or Jm (z) = Jm−1 (z),
∂z +
z
 m
z −m Jm (z) = z −m Jm+1 (z), or −∂z +


−∂z Jm (z) = Jm+1 (z).
z
Besides,  n
1
∂z z m Jm (z) = z m−n Jm−n (z),
z
 n
1
− ∂z z −m Jm (z) = z −m−n Jm+n (z).
z

±
Analogous identities hold for Hm (z), and Ym (z).

3.11 Half-integer parameters

1
2 2
J 21 (z) = = sin z,
πz
  21
2
J− 12 (z) = cos z,
πz
  12
2
H±
1 (z) = ∓i e±iz ,
2 πz
  21
± 2
H− 1 (z) = e±iz
2 πz

21
 For k = 0, 1, 2, . . . we have
 2  12
1
 1 k sin z
J 12 +k (z) = z 2 +k − ∂z (3.107)
π z z
 2  12 X ∞
1 (−j)k
= 2k (−1)j z 2 +2j−k ,
π (2j + 1)!
j=k
 2  12 1
1 k cos z
J− 12 −k (z) = z 2 +k ∂z (3.108)
π z z
∞
 2  12 X 1 ( 1 − j)k
= (−2)k (−1)j z − 2 +2j−k 2 ,
π j=0
(2j)!
 2  121
 1 k e±iz
H±
1
+k
(z) = ∓i z 2 +k − ∂z (3.109)
2 π z z
∞
 2  12 X 1 ( 1 − n2 )k
= ∓i 2k (±i)n z − 2 −k+n 2 ,
π n=0
n!
 2  12 1  1 k e±iz
±
H− 1 (z) = z 2 +k ∂z (3.110)
2 −k π z z
∞
 2  12 X 1 ( 1 − n2 )k
= (−2)k (±i)n z − 2 −k+n 2 .
π n=0
n!

3.12 Wronskians of solutions of the Bessel equation

The Wronskian of two solutions of the Bessel equation satisfies
1

∂z + W (z) = 0.
z
Hence W (z) is proportional to z1 . Using
1 z ±m 0 1 z ±m−1
J±m (z) ∼ Γ(±m+1) ( 2 ) , J±m (z) ∼ Γ(±m) ( 2 ) ,

we can compute the Wronskian of Jm (z) and J−m (z):

2 sin πm − + 4i 2
W (Jm , J−m ) = − , W (Hm , Hm )=− , W (Jm , Ym ) = .
πz πz πz

3.13 Putting together the Hankel and Macdonald func-

tion
±
Km and Hm are naturally analytic functions on the Riemann’s surface of the
logarithm.
√ However,√ it is sometimes convenient to treat them together, as func-
tions of w, resp. −w, where w ∈ R. It is important to indicate precisely how
the analytic continuation of the square root is performed—whether we bypass
the branch point at zero from above or from below. This is encoded by adding

22
±i0 to the variable. More precisely, we will use the following notation:
( √

√
Km w , w > 0,
Km w ∓ i0 := √ √ (3.111a)
Km (∓i −w = ±i π2 e±iπm Hm ±



−w , w < 0;
( √
√

±
√
Hm±
± i w = ∓i π2 e∓iπm Km w , w > 0,
Hm −w ± i0 := ±
√
(3.111b)
Hm −w , w < 0.

Note the functions (3.111a) and (3.111b) should be understood as distributions,

possibly with a singularity at 0.

4 Helmholtz equation in 2 dimensions

4.1 Polar coordinates
Introduce in R2 polar coordinates
x = r cos φ, y = r sin φ,
p y
r = x2 + y 2 , φ = arctan .
x
Remark 4.1 Change from Cartesian coordinates to polar cordinates can be in-
terpreted as a unitary transformation U : L2 (R2 ) → L2 ([0, ∞[×[0, 2π], rdrdφ)
defined as
(U f )(r, φ) := f (r cos φ, r sin φ).
We have
∆ = ∂r2 + r−1 ∂r + r−2 ∂φ2 .
It is also convenient to introduce another variety of polar coordinates, setting
w = eiφ :
x = 2r (w + w−1 ),
y = r
2i (w − w−1 ).
We have ∂φ = iw∂w , and so
∆ = ∂r2 + r−1 ∂r − r−2 (w∂w )2 . (4.112)

4.2 Plane waves

We look for solutions of the Helmholtz equation
(∆ + 1)F = 0,
where F is a distribution in S 0 (R2 ). We can assume that
Z Z
1
F (x, y) = F̂ (ξ, η)ei(xξ+yη) dξdη.
2π

23
We obtain

(−ξ 2 − η 2 + 1)F̂ (ξ, η) = 0,

so that suppF̂ ⊂ {(x, y) | x2 + y 2 = 1}. Thus the Helmholtz equation is solved

by Z
1
Ψg (x, y) = ei(x cos ψ+y sin ψ) g(ψ)dψ. (4.113)
2π
g is an arbitrary distribution on the circle S 1 .
In particular, if we set g = 2πδψ , then the corresponding solutions are the
plane waves
fψ (x, y) := ei(x cos ψ+y sin ψ) ,
or in polar coordinates
fψ (r, φ) = eir cos(φ−ψ) .

4.3 Circular waves

We will denote by H the Hilbert space of functions on R2 of the form (4.113)
with g ∈ L2 (S 1 ). We will treat them as “nice” solutions of the Helmholtz
equation. We have the scalar product
Z 2π
(Ψg1 |Ψg2 ) := g1 (ψ)g2 (ψ)dψ.
0

Note that if we set gm (ψ) = i−m eimψ , then √1 gm

2π
is an orthonormal basis in
L2 (S 1 ). We define the mth circular wave:
Z 2π
1 −m
fm := Ψgm = i ei(x cos ψ+y sin ψ) eimψ dψ. (4.114)
2π 0
√
2πfm form an orthonormal basis of H.

Proposition 4.2 In polar coordinates we have

fm (r, φ) = Jm (r)eimφ . (4.115)

Proof. Recall that

Z
1 r 
Jm (r) = exp (t − t−1 ) t−m−1 dt (4.116)
2πi [0+ ] 2
Z 2π
1
= eir sin φ e−imφ dφ. (4.117)
2π 0

24
Hence
Z 2π
1 π
fm (r, φ) = eir cos(φ−ψ) eimψ e−im 2 dψ (4.118)
2π 0
Z 2π
1
= eir sin ξ e−imξ eimφ dξ (4.119)
2π 0
= Jm (r)eimφ , (4.120)

where we set ξ = φ − ψ − π2 . 2
Clearly, by the inverse Fourier transformation, a plane wave can be decom-
posed into circular waves:
∞
X
fψ = fm e−imψ im . (4.121)
m=−∞

Note also that (4.121) corresponds to the formula for the generating function.
∞
r −1 X
e 2 (t−t )
= tm Jm (r). (4.122)
m=−∞

π
Indeed, when we insert t = ei(φ−ψ+ 2 ) in (4.122), we obtain (4.121).

4.4 Helmholtz equation on a disk

Consider the disc K = {x2 + y 2 < 1} and the Helmholtz equation on K with
the Dirichlet
∆D F + k 2 F = 0, F (x, y) = 0, (4.123)
and Neumann boundary conditions

∆N F + k 2 F = 0, (x∂x + y∂y )F (x, y) = 0. (4.124)

We are looking for F ∈ L2 (K) and k ≥ 0 that solve these eigenvalue problems
Recall that
1 ∂φ2
∆ = ∂r2 + ∂r + 2 . (4.125)
r r
±
Solutions of the radial equation are spanned by Jm (kr) and Hm (kr), m =
0, 1, 2, . . . . We check which solutions are square integrable. We have Jm (kr) ∼
rm , m = 0, 1, . . . , therefore they are all square integrable. On the othere hand,
±
Hm (kr) ∼ r−m , m = 1, 2, . . . , therefore they are not square integrable.
The function H0± (kr) ∼ log(r), is square integrable. We however discard
it as well—it corresponds to a “Dirac delta charge” at the origin. In fact, if
we apply the Gauss law and compute the flux through the circle around 0 we
obtain Z 2π
1
rdφ∂r ln(r) = 2πr = 2π.
0 r

25
In ther words
∆ ln(r) = 2πδ0 .
Hence this solution corresponds to a “charge” equal to 2π at zero.
Hence acceptable solutions are given by

F (r, φ) = eimφ Jm (kr), m ∈ Z. (4.126)

The boundary conditions give

Jm (k) = 0, (4.127)
0
Jm (k) = 0. (4.128)

Thus the frequencies are the zeros of the Bessel function Jm , respectively of the
0
derivative of the Bessel function Jm .

4.5 Translation operator

For a function f on R and t ∈ R we define

(Ut f )(x) := f (x + t).

Ut can be understood as a family of operators acting on various spaces of func-

tions on R. It satisfies Ut Us = Ut+s , Ut = 1. For instance, Ut can be interpreted
as operator on L2 (R), in which case they are unitary.
We also have the operator ∂x . We have
d
Ut f = ∂x Ut f.
dt
Therefore, we write
Ut = et∂x
and call ∂x the generator of translations. In quantum mechanics customarily
instead of ∂x one uses the momentum operator p = 1i ∂x , which is Hermitian in
the sense of L2 (R).

4.6 Rotation operator on L2 (R2 )

For a function f on R2 and for θ ∈ R we define

(Rθ f )(x, y) := f (cos θx − sin θy, sin θx + cos θy).

We obtain a family of operators satisfying Rθ1 Rθ2 = Rθ1 +θ2 , R0 = 1. In partic-

ular, understood as operators on L2 (R2 ), they are unitary.
Define
L = x∂y − y∂x .
We will show that
d
Rθ f = LRθ f. (4.129)
dθ

26
Introduce notation

x̃ := x cos θ − y sin θ, ỹ := x sin θ + y cos θ.

d
Rθ f (x, y) = (x sin θ + y cos θ)∂x̃ f (x̃, ỹ)
dθ
+(−x cos θ + y sin θ)∂ỹ f (x̃, ỹ),
LRθ f (x, y) = x(sin θ∂x̃ + cos θ∂ỹ )f (x̃, ỹ)
−y(cos θ∂x̃ − sin θ∂ỹ )f (x̃, ỹ),

which shows (4.129). Therefore, we write

Rθ = eθL .

In quantum mechanics customarily one uses the angular momentum operator

1 2 2
i L, which is Hermitian in the sense of L (R ).
In polar coordinates the operator Rθ acts as

(Rθ f )(r, φ) = f (r, φ + θ), Rθ f (r, w) = f (r, weiθ ).

Here is L in the polar coordinates:

L = ∂φ = iw∂w .

4.7 The group R2 o SO(2)

Consider translations and rotations on R2 :

u(x1 ,y1 ) (x, y) = (x − x1 , y − y1 ), (4.130)

rθ (x, y) = (x cos θ + y sin θ, −x sin θ + y cos θ). (4.131)

Transformations of the form (x1 , y1 , θ) := u(x1 ,y1 ) rθ form a group:

(x2 , y2 , θ2 )(x1 , y1 , θ1 )
=(x2 + cos θ2 x1 + sin θ2 y1 , y2 − sin θ2 x1 + cos θ2 y1 , θ2 + θ1 ) (4.132)

We have an obvious complex form of this group, where we write zi = xi + iyi ,

wi = e−iθi . Then (4.132) corresponds to

(z2 , w2 )(z1 , w1 ) = (z2 + w2 z1 , w2 w1 ). (4.133)

For a function f on R2 set

 
U(x1 ,y1 ) f (x) = f u−1
(x1 ,y1 ) (x, y) , (4.134)
−1


Rθ f (x, y) = f rθ (x, y) (4.135)

27
We have
(U(x1 ,y1 ) f )(x) := f (x + x1 , y + y1 ), (4.136)
(Rθ f )(x, y) := f (cos θx − sin θy, sin θx + cos θy). (4.137)
These operators can be understood on various spaces of functions on R2 , e.g.
on L2 (R2 ) We have a representation
R2 o SO(2) 3 (x, y, θ) 7→ U (x, y, θ) := U (x, y)R(θ) ∈ U (L2 (R) ).
The Laplacian ∆ is an invariant operator:
U (x, y, θ)∆U (x, y, θ)−1 = ∆. (4.138)

4.8 Lie algebra R2 o so(2)

The operators ∂x , ∂y and L form a Lie algebra commuting with ∆:
[∂x , L] = ∂y , (4.139)
[∂y , L] = −∂x . (4.140)
We have
∂x = cos φ∂r − r−1 sin φ∂φ ,
∂y = sin φ∂r + r−1 cos φ∂φ . (4.141)
Therefore,
A+ := ∂x + i∂y = eiφ (∂r + ir−1 ∂φ ) = w(∂r − r−1 w∂w ),
−
A := ∂x − i∂y = e−iφ (∂r − ir−1 ∂φ ) = w−1 (∂r + r−1 w∂w ),

Note the relations

[A+ , A− ] = 0, (4.142)
[A± , L] = ±A± , (4.143)
+ − − +
∆ = A A =A A , (4.144)
±
[∆, L] = [∆, A ] = 0. (4.145)

4.9 Action of translations and rotations on solutions of

the Helmholtz equation
The operators U(x,y) preserve H and are unitary:
U(x0 ,y0 ) Ψg = ΨÛ(x g, (4.146)
0 ,y0 )

Û(x0 ,y0 ) g(ψ) = ei cos ψx0 +i sin ψy0 g(ψ), (4.147)

∂x Ψg = Ψi cos ψg , (4.148)
∂y Ψg = Ψi sin ψg (4.149)

28
 In other words, plane waves diagonalize translations:

Ux0 ,y0 fψ = ei(x0 cos ψ+y0 sin ψ) fψ .

∂x fψ = i cos ψfψ , ∂y fψ = i sin ψfψ . (4.150)

The operators Rθ preserve the space H and are unitary:

Rθ Ψg = ΨÛθ g , (4.151)
R̂θ g(ψ) = g(ψ + θ), (4.152)
LΨg = Ψ∂ψ g . (4.153)

Thus circular waves diagonalize rotations:

Rθ fm = eimθ fm .

Lfm = imfm .
By (4.143) A± raises/lowers L by 1. This is expressed in the following
proposition:

Proposition 4.3

A+ fm = fm+1 , (4.154)
A− fm = fm−1 . (4.155)

Proof.

A± f m = w±1 (∂r ∓ r−1 w∂w )Jm (r)wm

= w±1 (∂r Jm (r) ∓ r−1 mJm (r))wm
= Jm±1 (r)wm±1 = fm±1 .

We used
 m
∂z + Jm (z) = Jm−1 (z), (4.156)
z
 m
−∂z + Jm (z) = Jm+1 (z). (4.157)
z
2

4.10 Graf addition formula

Theorem 4.4 Assume that R, r, ρ and Φ, φ, ψ are related as
s
reiφ + ρeiψ
q
R = (reiφ + ρeiψ )(re−iφ + ρe−iψ ), eiΦ = .
re−iφ + ρe−iψ

29
Then
∞
X
imΦ
Jm (R)e = Jn (ρ)einψ Jm−n (r)ei(m−n)φ .
n=−∞

If m ∈ Z, then there are no restrictions on the parameters in the formula. If

m is nonintegral and all variables real, then one has to assume that ρ < r (or,
equivalently, |Φ − φ| < π2 ). We can then replace the Bessel function in Jm (R)
(i)
and Jm−n (r) with Hm or Ym

Proof. We put ψ̃ = ψ − φ, Φ̃ = Φ − φ. Then the problem is reduced to the

case φ = 0.
∞
X
Jn (ρ)einψ Jm−n (r)
n=−∞
∞ Z
1 X r
exp (t − t−1 ) t−m−1 Jn (ρ)(teiψ )n


=
2πi n=−∞ γ 2
Z
1 r
ρ

= exp (t − t−1 ) + teiψ − (teiψ )−1 t−m−1 dt
2πi γ 2 2
Z
1 R
(s − s−1 ) s−m−1 dseimΦ = eimΦ Jm (R).


= exp
2πi γ 2

In the first step we used the integral representation of Jm−n (r), in the second
step we used the generating function to sum up Jn (ρ), finally, in the last step
we used r + ρeiψ = ReiΦ and turned the contour. 2
Substituting
x1 = r cos φ, y1 = r sin φ,

x2 = ρ cos ψ, y2 = ρ sin ψ,

x = R cos Φ, y = R sin Φ,
we obtain
(x1 , y1 ) + (x2 , y2 ) = (x + y)
and the addition formula can be rewritten as
p  x + iy m
Jm ( x2 + y 2 ) p
x2 + y 2
q  x + iy n q  x + iy m−n
2 2 1 1
X
= Jn ( x22 + y22 ) p 2 2
Jm−n ( x 2 + y2 ) p
1 1 .
n∈Z
x2 + y2 x1 + y12
2

Writing U (x, y) for U(x,y) , we can rewrite the Graf addition formula as
  X p  x + iy m−n
U (x1 , y1 )fm (x2 , y2 ) = fn (x2 , y2 )Jm−n ( x2 + y 2 ) p .
n∈Z
x2 + y 2
(4.158)

30
Let Unm (x, y) denote the matrix of U (x, y) in the basis of circular waves. By
(4.158),
p  x + iy m−n
Unm (x, y) = Jm−n ( x2 + y 2 ) p . (4.159)
x2 + y 2
The matrix of the operator U (x, y, θ) in the basis of the circular waves is
p  x + iy m−n
Unm (x, y, θ) := Jm−n ( x2 + y 2 ) p eimθ . (4.160)
x2 + y 2
We have the representation property:
  ∞
X
Uk,m (x2 , y2 , θ2 )(x1 , y1 , θ1 ) = Uk,n (x2 , y2 , θ2 )Un,m (x1 , y1 , θ1 ).
m=−∞

5 Distributions in d = 1
5.1 Homogeneous distributions of order −1 and 0
The function x1 is not in L1loc , therefore it does not define a regular distribution.
However, it can be naturally interpreted as a distribution as follows
Z Z a Z ∞ Z a
1 1 1

P φ(x)dx := + φ(x)dx + φ(x) − φ(0) dx.
x −∞ a x −a x

(We are writing P to indicate that it is not the usual integral). Equivalently,
1 1 1 1 
:= + .
x 2 (x + i0) (x − i0)
The Sochocki formula is relationship between three kinds of order −1 distribu-
tions:
1 1
= ∓ iπδ(x).
x ± i0 x

Z
e−ixk dx = 2πδ(k), (5.161)
∓i
Z
θ(±x)e−ixk dx = , (5.162)
k ∓ i0
Z
1
sgn(x)e−ixk dx = −2i , (5.163)
k
Z
δ(x)e−ixk dx = 1, (5.164)
Z −ikx
e
dx = ∓2πiθ(±k), (5.165)
x ± i0
Z −ikx
e
P dx = −πisgn(k), (5.166)
x

31
 (
−2πiθ(ξ)e−iλξ
Z
−iξs −1 Imλ < 0,
e (s − λ) ds = (5.167)
2πiθ(−ξ)e−iλξ Imλ > 0;
Z
P e−iξs (s − λ)−1 d s = −πisgn(ξ)e−iλξ , Imλ = 0. (5.168)

5.2 Homogeneous distributions of integral order

Define for n = 0, 1, 2, . . .
1 1 1 1 
:= + .
xn+1 2 (x + i0)n+1 (x − i0)n+1

Clearly
1 1
∂xn = (−1)n n! n+1 .
x x

Z
xn e−ixk dx = 2πin δ (n) (k), (5.169)

(−i)n+1 n!
Z
xn θ(±x)e−ixk dx = ± , (5.170)
(k ∓ i0)n+1
Z
1
xn sgn(x)e−ixk dx = 2(−i)n+1 n! n+1 , (5.171)
k
Z
δ (n) (x)e−ixk dx = in k n , (5.172)

e−ikx 2π(−i)n+1 n
Z
dx = ± k θ(±k), (5.173)
(x ± i0)n+1 n!
Z −ikx
e π(−i)n+1 n
P dx = k sgn(k). (5.174)
xn+1 n!

5.3 Homogeneous distributions of arbitrary order I

For any λ ∈ C
(±ix + 0)λ := lim (±ix + )λ .
→0

is a tempered distribution. If Reλ > −1, then it is simply the distribution given
by the locally integrable function
π
e±isgn(x) 2 λ |x|λ . (5.175)

The functions
xλ± := (±x)λ θ(±x) (5.176)

32
define distributions only for Reλ > −1. We can extend them to λ ∈ C except
for λ = −1, −2, . . . by putting
1  π π

xλ± := − e−i 2 λ (∓ix + 0)λ + ei 2 λ (±ix + 0)λ . (5.177)
2i sin πλ
Instead of xλ± it is often more convenient to consider

xλ±
ρλ± (x) := (5.178)
Γ(λ + 1)
Γ(−λ)  −i π λ π

= e 2 (∓ix + 0)λ − ei 2 λ (±ix + 0)λ . (5.179)
2πi
Note that using (5.178) and (5.179) we have defined ρλ± for all λ ∈ C.
Theorem 5.1 The distributions ρλ± satisfy the recurrence relations

∂x ρλ± (x) = ±ρλ−1

± (x).

At integers we have
xn±
ρn± = , n = 0, 1, . . . ; (5.180)
n!
ρ−n−1
± = (±1)n δ n (x), n = 0, 1, . . . . (5.181)
Their Fourier transforms are below:
Z
e−iξx ρλ± (x)dx = (±iξ + 0)−λ−1 ,
Z
−λ−1
e−iξx (∓iξ + 0)λ dξ = 2πρ± (x).

Proof. (5.181) follows from

(∓1)n+1 n!  
ρ−n−1
± (x) = (x ± i0)−n−1 − (x ∓ i0)−n−1 (5.182)
2πi
(±1)n n  
=− ∂x (x ± i0)−1 − (x ∓ i0)−1 . (5.183)
2πi
2

Theorem 5.2 Let −n − 1 < Reλ, λ 6∈ {. . . , −2, −1}. Then for any a > 0,
Z Z ∞
P xλ+ φ(x)dx = xλ φ(x)dx
a
Z a  n−1
X xj 
+ xλ φ(x) − φ(j) (0) dx
0 j=0
j!
n−1 j
X X (−1)l
+ aλ+j+1 φ(j) (0) . (5.184)
j=0
(j − l)!(λ + 1) · · · (λ + 1 + l)
l=0

33
If −n − 1 < Reλ < −n, we can even go with a to infinity
Z Z ∞  n−1
X xj 
P xλ+ φ(x)dx = xλ φ(x) − φ(j) (0) dx. (5.185)
0 j=0
j!

Proof. We use induction. Suppose that the formula is true for λ

Z Z
λ−1
−λP x+ φ(x)dx = P xλ+ ∂x φ(x)dx (5.186)
Z ∞
= xλ ∂x φ(x)dx (5.187)
a
Z a n
λ
 X xj 
+ x ∂x φ(x) − φ(j) (0) dx
0 j=0
j!
n−1 j
X
λ+j+1 (j+1)
X (−1)l
+ a φ (0) .
j=0
(j − l)!(λ + 1) · · · (λ + 1 + l)
l=0

Then we integrate by parts, obtaining the identity for λ − 1. 2

5.4 Homogeneous distributions of arbitrary order II

We also can define
1 
|x|λ = (−ix + 0)λ + (ix + 0)λ ,


π (5.188)
2 cos( 2 λ)
1 
|x|λ sgn(x) = − (−ix + 0)λ + (ix + 0)λ .


π (5.189)
2i sin( 2 λ)

The Fourier transforms:

λ+1  −λ−1
1 Γ(
2 )x
Z
|k|λ e−ixk dk = π 2 , (5.190)
Γ(− λ2 ) 2
 
λ+2  −λ−1
1 Γ(
2 )x
Z
|k|λ sgn(k)e−ixk dk = −iπ 2 1−λ sgn(x), (5.191)
Γ( 2 ) 2
 

Especially symmetric expressions for Fourier transforms are obtained if we

34
introduce
 λ 1 −1  x2  λ2
λ
ηev (x) := Γ + (5.192)
2 2 2
 λ 1 λ  
−1
= (2π) Γ − + 2− 2 (ix + 0)λ + (−ix + 0)λ (5.193)
2 2
λ
22  λ  λ 
= √ Γ 1+ ρ+ (x) + ρλ− (x) , (5.194)
π 2
λ −1  x2  λ+1
2 1
λ
ηodd (x) := Γ +1 (5.195)
2 2 x
 λ λ 1  
= i(2π)−1 Γ − 2− 2 − 2 (ix + 0)λ − (−ix + 0)λ (5.196)
2
λ
− 21 
2 2 1 λ  λ 
= √ Γ + ρ+ (x) − ρλ− (x) . (5.197)
π 2 2
We then have the following relations:
λ λ−1 λ λ−1
∂x ηev = ληodd , ∂x ηodd = ηev , (5.198)
λ λ+1 λ λ+1
xηev (x) = (λ + 1)ηodd (x), xηodd (x) = ηev (x); (5.199)
λ −λ−1 λ −λ−1
Fηev = ηev , Fηodd = −iηodd ; (5.200)
√
−1−2m (−1)m 2 (2m)
ηev (x) = m 1
δ (x), m = 0, 1, . . . ; (5.201)
2 2 m
√
−2m (−1)m 2 (2m−1)
ηodd (x) = m 1
δ (x), m = 1, 2, . . . . (5.202)
2 2 m

5.5 Anomalous distributions of degree −1

We introduce the distributions θ(±k)
k = ±k±−1
:
Z Z ∞
−1
k+ φ(k)dk := − log(k)φ(1) (k)dk (5.203)
0
Z a Z ∞
φ(k) − φ(0) φ(k)
= dk + log(a)φ(0) + dk (5.204)
0 k a k
 ∞ φ(k)
Z 
= lim + φ(0) ln() , (5.205)
&0  k
Z Z 0
−1
k− φ(k)dk := − log(−k)φ(1) dk (5.206)
−∞
0 −a
φ(k) − φ(0)
Z Z
φ(k)
=− dk + log(a)φ(0) − dk (5.207)
−a k −∞ k
 Z − φ(k) 
= lim − + φ(0) ln() . (5.208)
&0 −∞ k

35
We have
1 −1 −1
= −k− + k+ ,
k
We also define
1 −1 −1
= k− + k+ .
|k|
1
Proposition 5.3 Here are the Fourier transform of various forms of |k| and
the logarthm:
Z
−1 −ixk
k± e dk = − log(±ix + 0) − γ (5.209)
iπ
= − log |x| ∓ sgn(x) − γ, (5.210)
Z 2
1 −ixk
P e dk = −2 log |x| − 2γ, (5.211)
|k|
Z
1
log |x|e−ixk dx = −πP − 2πγδ(k), (5.212)
|k|
Z
θ(∓k)
log(±ix + 0)e−ixk dx = −2π − 2πγδ(k), (5.213)
|k|
Z
−1
log(x ∓ i0)e−ixk dx = −2πk∓ + (−2πγ ∓ iπ)δ(k), (5.214)
Z  
−1
log(x − λ)e−ixk dx = e−iλk − 2πk∓ + (−2πγ ∓ iπ)δ(k) , ±Imλ > 0.
(5.215)

Proof. We start from one of the formulas for the Euler constant. We change
the variable from k to yk, with y > 0:
Z ∞
−γ = e−k log(k)dk
0
Z ∞
= e−yk log(ky)d(ky)
0
Z ∞ Z ∞
= y log(y) e−ky dk + y e−ky log(k)dk
0 0
Z 1 Z ∞
1
∂k (e−ky − 1) log(k)dk − ∂k e−ky log(k)dk



= y log(y) −
y 0 1
Z 1 −ky Z ∞ −ky
e −1 e
= log(y) + dk + dk.
0 k 1 k

The rhs is analytic in y on the right halfplane. It is constant on the positive

halfline. So it is constant on the whole halfplane. Therefore, we can replace y
with ix. This proves (5.209), which implies (5.210) and (5.211).

36
 By inverting the Fourier transform we obtain (5.212) and (5.214). We can
also get (5.214) from (5.212):
Z Z
log(x ∓ i0)e−ixk dx = log |x| ∓ iπθ(−x) e−ixk dx


(5.216)
1 1
= −πP − 2πγδ(k) ± π (5.217)
|k| (k + i0)
1
= −2πP + (−2πγ ∓ iπ)δ(k). (5.218)
k∓
2
Here is an alternative approach:
1  1 1 
= lim − δ(k) , (5.219)
k± ν&0 k 1+ν ν
±
1  1 2 
= lim − δ(k) (5.220)
|k| ν&0 |k|1+ν ν

Here is the computation of the Fourier transform by this method:

Z −ikx Z −ikx
e e 1
P dk ≈ 1+ν dk − (5.221)
k± k± ν
1
= Γ(ν)(±ik + 0)−ν − , (5.222)
ν
1 
1
≈ − γ 1 − ν log(±ik + 0) − , (5.223)
ν ν
≈ − log(±ik + 0) − γ. (5.224)

5.6 Anomalous distributions of integral degree

Define

−n−1 (∓1)n n −1
k± := ∂ k k± , (5.225)
n!
sgn(k) −n−1 −n−1
:= k+ + (−1)n k− . (5.226)
k n+1

37
Theorem 5.4 Using Hn defined in (.449), we have

1 δ (n) (x)
+ (∓1)n Hn (5.227)
x1+n
± n!
 1 (∓1)n δ (n) (x) 
= lim − , (5.228)
ν→0 x1+n−ν
± ν n!
1 n

δ (n) (x)
+ (−1) + 1 Hn (5.229)
|x|1+n n!


 1 (−1)n + 1 δ (n) (x) 
= lim − , (5.230)
ν→0 |x|1+n−ν ν n!
sgn(x)
δ (n) (x)
1+n
+ (−1)n − 1 Hn (5.231)
|x| n!


 sgn(x) (−1)n − 1 δ (n) (x) 
= lim − . (5.232)
ν→0 |x|1+n−ν ν n!

Proof. It is enough to consider only x−n−1

+ .
∞


∂xn+1 xν φ(x)
Z Z
P x−n−1+ν
+ φ(x)dx =P dx (5.233)
0 ν(ν − 1) · · · (ν − n)
∞
xν+ φ(n+1) (x)
Z
= dx (5.234)
0 (−ν)(1 − ν) · · · (n − ν)
∞
(xν+ − 1)φ(n+1) (x)
Z
= dx (5.235)
0 (−ν)(1 − ν) · · · (n − ν)
∞
φ(n+1) (x)
Z
+ dx (5.236)
0 (−ν)(1 − ν) · · · (n − ν)
∞
log(x)φ(n+1) (x)
Z
=− dx (5.237)
0 n!
1 φ(n) (0)
+ (5.238)
ν (1 − ν) · · · (n − ν)
Z
−n−1
= P x+ φ(x)dx (5.239)

1 φ(n) (0) φ(n) (0)

+ + Hn + O(ν). (5.240)
ν n! n!
2
The above proof is taken from Hörmander, sect. 3.2. Note that Hörmander
treats (5.228) as the standard regularization of x−n−1
± . We prefer the definition
(5.225).

38
Theorem 5.5
∞ n−1
xj (j) xn (n) 
Z Z
1  X
P x−n−1
+ φ(x)dx = φ(x) − φ (0) − φ (x) dx
0 xn+1 j=0
j! n!
1 Z ∞
1  φ(n) (x) φ(n) (0)  1 φ(n) (x)
Z
+ − dx + dx
0 x n! n! 1 x n!
φ(n) (0)
− Hn . (5.241)
n!
Proof. Let a > 0. If we assume that Reν > −1, then we can use (5.184) with
n replaced with n + 1:
Z
1
P n+1−ν φ(x)dx (5.242)
x+
Z ∞
1
= n+1−ν
φ(x)dx
a x
Z a n
1  X xj (j) 
+ n+1−ν
φ(x) − φ (0) dx
0 x j=0
j!
n j
X
−n+j+ν (j)
X 1
− a φ (0) . (5.243)
j=0
(j − l)!(n − l − ν) · · · (n − ν)
l=0

The last term of (5.243) is

n−1
X 1
− aν φ(n) (0) (5.244)
(n − l)!(n − l − ν) · · · (n − ν)
l=0
1
− aν φ(n) (0) (5.245)
(−ν) · · · (n − ν)
Hn
= − φ(n) (0) (5.246)
n!
1 (n) 1 1
+ φ (0) + log(a)φ(n) (0) (5.247)
ν n! n!
(n) Hn
+ φ (0) + O(ν) (5.248)
n!
1 1 1
= φ(n) (0) + log(a)φ(n) (0) + O(ν). (5.249)
ν n! n!

39
Thus we have proven that
Z Z ∞
1
P x−n−1
+ φ(x)dx = φ(x)dx
a xn+1
Z a n
1  X xj (j) 
+ φ(x) − φ (0) dx
0 xn+1 j=0
j!
n−1 j
X X 1
− a−n+j φ(j) (0) (5.250)
j=0
(j − l)!(n − l) · · · n
l=0

φ(n) (0) φ(n) (0)

− Hn + log(a) (5.251)
n! n!
Then we take a → ∞, noting that
Z a
x−1 dx = log(a).
1
2

Proposition 5.6 The Fourier transform:

(∓ix)n 
Z 
−n−1 −ixk
P k± e dk = − log(±ix + 0) − γ (5.252)
n!
(∓ix)n  iπ 
= − log |x| ∓ sgn(x) − γ (5.253)
n! 2
Proof. We use (5.228):
Z ∞ −ixk  Z ∞ e−ixk
e (∓ix)n (∓ik)n 
dk + Hn = lim dk − (5.254)
0 k n+1 n! ν&0 0 k 1+n−ν νn!
 (∓ik)n 
= lim Γ(−n + ν)(±ix + 0)n−ν − (5.255)
ν&0 νn!
!
(−1)n  1 
n

(∓ik)n
= lim − γ + Hn (±ix) 1 − ν log(±ix + 0) −
ν&0 n! ν νn!
(∓ix)n

= − γ + Hn − log(±ix + 0) . (5.256)
n!
Note that the terms containg Hn cancel. 2

5.7 Infrared regularized distributions

Theorem 5.7 Let n + 1 > 2α > n. Then
θ(k)  θ(k)
= lim (5.257)
k 2α m→0 (k 2 + m2 )α
n−1
X Γ(α − 2j − 21 )Γ( 2j + 21 ) 
− 2α−j−1
(−1)j δ (j) (k) . (5.258)
j=0
2m Γ(α)j!

40
Proof. Clearly,
Z Z ∞ n−1
X kj
θ(k)φ(k) 1  (j)

P dk = φ(k) − φ (0) dk (5.259)
k 2α 0 k 2α j=0
j!

is the limit as m → 0 of
Z ∞ n−1
X kj
1 
(j)

φ(k) − φ (0) dk. (5.260)
0 (k 2 + m2 )α j=0
j!

Now
∞
kj Γ(α − 2j − 12 )Γ( 2j + 12 )
Z
dk = (5.261)
0 (k 2 + m2 )α 2m2α−j−1 Γ(α)
2

Theorem 5.8 Let 2p + 1 > 2α > 2p − 1, p = 1, 2, . . . . Then

1  1
= lim (5.262)
|k|2α m→0 (k 2 + m2 )α
p−1 3
X π 2 m−2α+2l+1 (−1)l (2l)

− 1

3 δ (k) . (5.263)
l=0
Γ(α) sin π(α − 2 ) 22l l!Γ( 2 − α + l)

Proof. Clearly,
p−1
k 2l (2l) 
Z Z
φ(k) 1  X
P dk = φ(k) − φ (0) dk (5.264)
k 2α k 2α (2l)!
l=0

is the limit as m → 0 of
p−1
k 2l (2l) 
Z
1  X
φ(k) − φ (0) dk. (5.265)
(k 2 + m2 )α (2l)!
l=0

Now

k 2l m−2α+2l+1 Γ(α − l − 12 )Γ(l + 12 )

Z
1
dk = (5.266)
(2l)! (k 2 + m2 )α Γ(α)(2l)!
3
π 2 m−2α+2l+1 (−1)l
= . (5.267)
Γ(α) sin π(α − 21 ) 22l l!Γ( 23 − α + l)

41
Theorem 5.9 Let n = 0, 1, . . . . Then

θ(k)  θ(k)
n+1
= lim n 1 (5.268)
k m→0 (k + m2 ) 2 + 2
2

n−1
Xm−n+j Γ( n2 − 2j )Γ( 2j + 21 ) 
− (−1)j δ (j) (k) (5.269)
j=0
2Γ(α)j!
 
 1 Hp+1 ( 1 ) + log(m) 1
(−1)δ (2p+1) (k), n = 2p + 1;
2 2 (2p+1)!
−   (5.270)
 1 Hp − log(2) + log(m) 1 δ (2p) (k), n = 2p.
2 (2p)!

Proof. Clearly,
Z
θ(k)φ(k)
P dk (5.271)
k 2α
is the limit as m → 0 of
Z ∞ n−1
X kj
1 
(j)

φ(k) − φ (0) dk (5.272)
0 (k 2 + m2 )n+1 j=0
j!
1
kn φ(n) (0) φ(n) (0)
Z
− n 1 − Hn . (5.273)
0 (k 2 + m2 ) 2 + 2 n! n!

Now for n = 2p + 1 we have

Z 1
k 2p+1
2 2 p+1
dk (5.274)
0 (k + m )
p 1 Z 1
X k 2j k
=−  + dk (5.275)

2
2j(k + m ) 02 j 2 2
j=1 0 k +m
p
X 1 1 t
=− + arctan (5.276)
j=1
2j(1 + m2 )j 2 m
1
= − Hp − log(m) + o(m0 ). (5.277)
2
Then we use
1 1 1
H2p+1 − Hp = Hp+1 . (5.278)
2 2 2

42
 For n = 2p we compute
Z 1
k 2p
dk (5.279)
2 2 p+ 12
0 (k + m )
p−1 1
X k 2j+1 1 Z k
=−  + dk (5.280)

(2j + 1)(k 2 + m2 ) j+ 1 2 0 (k 2
1
+ m2 ) 2
j=0 0
p
X 1 p

=− j+ 12
+ log 1 + 1 + m2 − log(m) (5.281)
(2j + 1)(1 + m2 )
j=1
1 1
= − Hp + log(2) − log(m) + o(m0 ). (5.282)
2 2
Then we use
1 1 1
H2p − Hp = Hp . (5.283)
2 2 2
2
Using (2.67) we derive
1
e−ikx 2π 2 m−2α+1  m|x| α− 21
Z
dk = Kα− 12 (m|x|). (5.284)
(k 2 + m2 )α Γ(α) 2
Note that (5.284) is bounded if α > 21 , has a logarithmic singularity at zero
if α = 12 , and has a singularity |x|2α−1 if α < 12 . Therefore, it is no longer a
regular distribution if α < 0. However, by applying (−∂x2 +m2 ) sufficiently many
times to (5.284) we can interpret it as a distribution for all α. (For α = −n,
n = 0, −1, −2, . . . we simply obtain (−∂x2 + m2 )n ).
Suppose now that the assumptions of Thm 5.8 are satisfied. Let us compute
the Fourier transform of the linear combination of the deltas:
p−1 3
π 2 m−2α+2l+1 (−1)l
Z X
1


2l l!Γ( 3 − α + l)
δ (2l) (k)e−ikx dk (5.285)
l=0
Γ(α) sin π(α − 2 ) 2 2
p−1 3
X π 2 m−2α+2l+1
= x2l . (5.286)
Γ(α) sin π(α − 12 ) 22l l!Γ( 32 − α + l)


l=0

Now the rhs of (5.284), using the identity (2.48), can be written as
3
π 2 m−2α+1  m|x| α− 21
1

I−α+ 12 (m|x|) (5.287)
Γ(α) sin π(α − 2 ) 2
3
π 2 m−2α+1  m|x| α− 12
− Iα− 12 (m|x|) (5.288)
Γ(α) sin π(α − 21 )


2
Now (5.287) is equal to (5.286) modulo O(m−2α+2p+1 ). (5.287) is equal to
1
π 2 Γ( 21 − α)  |x| 2α−1
(5.289)
Γ(α) 2
modulo O(m2 ). This is a confirmation of the correctness of Thm 5.8.

43
5.8 Distributions on halfline
We will denote by C ∞ [0, ∞[ smooth function having all right-sided derivatives
at 0. We set
∞
CN [0, ∞[ := {φ ∈ C ∞ [0, ∞[ : φ(2m+1) (0) = 0, m = 0, 1, . . . }, (5.290)
∞ ∞ (2m)
CD [0, ∞[ := {φ ∈ C [0, ∞[ : φ (0) = 0, m = 0, 1, . . . }. (5.291)
0 0
SN [0, ∞[, SD [0, ∞[ have obvious definitions. We set SN [0, ∞[, SD [0, ∞[ to be
their duals.
Note that ∂x and the multiplication by x map SN [0, ∞[ into SD [0, ∞[ and
0 0
vice versa, as well as SN [0, ∞[ into SD [0, ∞[ and vice vera.
The cosine transformation with the kernel
r
2
FN (x, k) := cos(xk)
π
0
maps SN [0, ∞[ into into itself. We have
Likewise, the sine transformation with the kernel
r
2
FN (x, k) := sin(xk)
π
0
maps SD [0, ∞[ into into itself.
Let Iφ(x) := φ(−x). I maps S(R), as well as extends to a map of S 0 (R) into
itself. We will write

Sev (R) := {φ ∈ S(R) : Iφ = φ}, (5.292)

0 0
Sev (R) := {λ ∈ S (R) : Iλ = λ}, (5.293)
Sodd (R) := {φ ∈ S(R) : Iφ = −φ}, (5.294)
0 0
Sodd (R) := {λ ∈ S (R) : Iλ = −λ}. (5.295)

If φ ∈ SN [0, ∞[, we set

(
φ(x) x ≥ 0;
φev (x) :=
φ(−x) x ≤ 0.

Note that φev ∈ Sev (R).

If λev is an even distribution in S 0 (R), then we can associate with it a
0
distribution in SN [0, ∞[ by
Z ∞ Z
1
λN (x)φ(x)dx := λev (x)φev (x)dx.
0 2
Similarly, if φ ∈ SD [0, ∞[, we set
(
odd φ(x) x ≥ 0;
φ (x) :=
−φ(−x) x ≤ 0.

44
Note that φodd ∈ Sodd (R).
If λodd is an odd distribution in S 0 (R), then we can associate with it a
0
distribution in SD [0, ∞[ by We set
Z ∞ Z
1
λD (x)φ(x)dx := λodd (x)φodd (x)dx.
0 2
The usual Fourier transform F preserves Sev (R) and Sodd (R). The Fourier
transform on even distributions is closely related to the cosine transform and
on odd distributions to the sine transform:
0
FN λN = (Fλ)N , λ ∈ Sev (R), (5.296)
0
FD λD = i(Fλ)D , λ∈ Sodd (R). (5.297)

An example of an even distribution is ηev . Let ηN denote the corresponding

0
distributtion in SN [0, ∞[.
Likewise, an example of an odd distribution is ηodd . Let ηD denote the
0
corresponding distributtion in SD [0, ∞[.
We have
λ −λ−1 λ −λ−1
F N ηN = ηN , F D ηD = ηD ; (5.298)

6 Distributions in arbitrary dimension

6.1 Sphere S d−1
Consider the Euclidean space Rd . Introduce two varieties of spherical coordi-
nates on a d − 1-dimensional sphere

(θd−2 , . . . , θ1 , φ) ∈ [0, π] × · · · × [0, π] × [0, 2π[,

(wd−2 , . . . , w1 , φ) ∈ [0, π] × · · · × [0, π] × [0, 2π[,

with wj = cos θj , The spherical measure on Sd−1 is

sind−2 θd−2 dθd−2 · · · sin θ1 dθ1 dφ

2
= (1 − wd−2 )(d−3)/2 dwd−2 · · · dw1 dφ.

Theorem 6.1 The area of the d − 1-dimensional sphere is

d
2π 2
Sd−1 = ,
Γ( d2 )

or, in a more elementary form,

2π m+1
S2m+1 = , m = 0, 1, . . . ; (6.299)
m!
2π m
S2m = 1 , m = 0, 1, . . . . (6.300)
( 2 )m

45
Proof. Method I. We compute in two ways the Gaussian integral: in the
Cartesian coordinates
Z
2 2 d
e−x1 −···−xd dx1 · · · dxd = π 2 ,

and in spherical coordinates:

Z ∞
2 1 d
Sd−1 e−r rd−1 dr = Γ . (6.301)
0 2 2
Method II. We compute the area of the sphere in the spherical coordinates:
Z π Z π Z 2π
d−2
Sd−1 = sin φd−1 dφd−1 · · · sin φ2 dφ2 dφ1
0 0 0
Then we use
√
πΓ( k−1
Z π 2π
2 )
Z
k−1
sin φk dφk = k
, k = 2, . . . , d − 1; dφ1 = 2π.
0 Γ( 2 ) 0
2

6.2 Homogeneous functions in arbitrary dimension

Theorem 6.2 Let −d < λ < 0. Then on Rd
λ+d  −λ−d
d Γ(
2 )ξ 
Z
|x|λ e−ixξ dx = π 2 . (6.302)
Γ(− λ2 ) 2
 

Proof. We use the spherical coordinates:

Z
|x|λ e−ixξ dx (6.303)
Z ∞ Z π
= dr dφd−1 rλ+d−1 e−ir|ξ| cos φd−1 rλ+d−1 sind−2 φd−1 Sd−2 (6.304)
0 0
Z π
2  
= Γ(λ + d) (i|ξ| cos φd−1 + 0)−λ−d + (−i|ξ| cos φd−1 + 0)−λ−d sind−2 φd−1 dφd−1 Sd−2
0
λ + d  Z π2
= Γ(λ + d)2 cos π |ξ|−λ−d cos−λ−d φd−1 sind−2 φd−1 dφd−1 Sd−2 .
2 0
Then we apply
d−1
2π 2
Sd−2 = ,
Γ( d−1
2 )
π
1 Γ( −λ−d+1 )Γ( d−1
2 )
Z 2
cos−λ−d φd−1 sind−2 φd−1 dφd−1 = 2
λ
,
0 2 Γ(− 2 )
1
λ + d λ + d + 1
Γ(λ + d) = π − 2 2λ+d−1 Γ Γ
2 2
λ + d  π
cos π = ,
2 Γ( λ+d+1
2 )Γ( −λ−d+1
2 )

46
and we obtain (6.302) 2
In order to express (6.302) in a more symmetric way, define

1  x2  λ2
η λ (x) := , λ > −d.
Γ( λ+d
2 )
2

We extend it to λ ≤ −d by setting
(−2)m m λ
η λ−2m (x) :=
∆ η (x). (6.305)
− λ2 m

Then

Fη λ = η −λ−d , (6.306)
2 λ λ+2
x η = (λ + d)η , (6.307)
λ λ−2
∆η = λη . (6.308)

6.3 Renormalizing the |k|−d function

Define the distribution |k|−d on Rd :
Z Z Z
P |k|−d φ(k)dk := |k|−d φ(k) − φ(0) dk + |k|−d φ(k)dk..


|k|<1 |k|>1

Theorem 6.3 We have an alternative definition of |k|−d :

d
−d

−d+ν 2π 2 
|k| = lim |k| − δ(k) . (6.309)
ν&0 νΓ( d2 )

Here is its Fourier transform:

Γ( d2 )
Z r  1 d 1
d P |k|−d e−ikx dk = − log + ψ − γ
2π 2 2 2 2 2
= − log r − γ, d = 1;
r
= − log − γ, d = 2;
2
1 1
= − log r − γ + Hm , d = 2m + 1;
2 2
r 1
= − log − γ + Hm , d = 2(m + 1).
2 2
Proof.
Z Z d
2π 2
|k|−d+ν dk = |k|−1+ν d|k|Sd−1 = . (6.310)
|k|<1 |k|<1 νΓ( d2 )

This proves (6.309).

47
 Z
|k|−d+ν e−ikx dk (6.311)
 r −ν π d2 Γ( ν )
2
= (6.312)
2 Γ( d2 − ν2 )
d
 r
 π 2  ν d
 2 
≈ 1 − ν log d
1+ ψ −γ
2 Γ( 2 ) 2 2 ν
d 
2π 2 1  r  1  d  1 
≈ − log + ψ − γ . (6.313)
Γ( d2 ) ν 2 2 2 2
2

7 Bessel potentials
In this section we analyze some distinguished solutions of

(1 + 2)α D(x) = δ(x), (7.314)

where 2 is the Laplacian on the pseudo-Euclidean space Rq,d−q (q minuses and

d − q pluses). Formally, D is given by the inverse Fourier transform

eipx
Z
dp
D(x) = , (7.315)
(1 + p2 )α (2π)d

where p2 is the square of p wrt the scalar product of Rq,d−q . However usually
we need to regularize the integrand of (7.315)
Note that if D solves (7.314), and m > 0, then

(m2 + 2)α md−2α D(mx) = δ(x). (7.316)

The following identities will be useful:

Z ∞
1 1
= e−sA sα−1 ds, (7.317)
Aα Γ(α) 0
π
e∓i 2 α ∞ ±itA α−1
Z
1
= e t dt. (7.318)
(A ± i0)α Γ(α) 0
We will also need the Fourier transform of the Gaussian function on the Eu-
clidean space Rd , and of the Fresnel function on the pseudo-Euclidean space
Rq,d−q (with q minuses):
Z
sp2
 2π  d2 x2
dpe− 2 eipx = e− 2s , (7.319)
s
Z
tp2
 2π  d2 π x2
dpe±i 2 eipx = (∓i)q e±i 4 d e∓i 2t . (7.320)
t

48
7.1 Euclidean and anti-Euclidean signature
Consider the Euclidean space Rd , with |x| denoting the Euclidean norm.
Theorem 7.1
eipx  |x| α− d2
Z
dp 2
= d Kα− d (|x|), (7.321)
(p2 + 1)α (2π)d Γ(α)(4π) 2 2 2

ipx d−1  d
|x| α− 2 ±
Z
e dp π(∓i) 
2 α d
= d Hα− d (|x|). (7.322)
(−p + 1 ± i0) (2π) Γ(α)(4π) 2 2 2

Proof. By (7.317) and then by (2.45),

eipx dp
Z

(1 + p2 )α
 |x| α 1 Z ∞ Z 2 |x|s
= ds dpsα−1 e−(1+p ) 2 eipx
2 Γ(α) 0
 |x| α− d2 π d2 Z ∞ d 1 |x|
= dssα− 2 −1 e−(s+ s ) 2
2 Γ(α) 0
d 
2π 2 |x| α− d2

= Kα− d (|x|).
Γ(α) 2 2

By (7.318), and then by (3.103),

eipx dp
Z

(1 − p2 ± i0)α
 |x| α e∓i π2 α Z ∞ Z 2 |x|
= dt dpe±it(1−p ) 2 tα−1 eipx
2 Γ(α) 0
 |x| (α− d2 ) e∓i π2 (α+ d2 ) π d2 Z ∞ |x| 1 d
= dte±i 2 (t+ t ) tα− 2 −1
2 Γ(α) 0
d−1 d d
π(∓i) π 2
 |x|  α− 2 ±
= Hα− d (|x|).
Γ(α) 2 2

7.2 General signature

Consider now a pseudo-Euclidean space of general signature Rq,d−q .
Theorem 7.2
eipx
Z
dp
(7.323)
(1 + p ± i0) (2π)d
2 α
√
2(∓i)q  x2 ∓ i0 α− d2 p

= d Kα− d x2 ∓ i0 (7.324)
Γ(α)(4π) 2m
2 2

q 
√ d
πi(∓i) −x2 ± i0 α− 2 ±
 p

=± d Hα− d −x2 ± i0 . (7.325)
Γ(α)(4π) 2 2 2

49
Remark 7.3 In (7.324) and (7.325) we use the notation explained in (3.111a)
and (3.111b). Note that (7.324) works best for x2 > 0, because then we can
ignore ∓i0. Likewise, (7.325) is best suited for x2 < 0, because then we can
ignore ±i0.

Proof of Thm 7.2. using (7.318) and (2.49) we obtain

eipx dp
Z

(1 + p2 ± i0)α
∓i π Z ∞ Z
e 2α i 2
= α dt dpe± 2 t(1+p ) tα−1 eipx
2 Γ(α) 0
π d d Z
(∓i)q e∓i 2 (α− 2 ) π 2 ∞ i x2 d
= d dte± 2 (t− t ) tα− 2 −1 . (7.326)
2α− 2 Γ(α) 0
√
For x2 ≥ 0, we change the variable t = ±is x2 , so that (7.326) becomes
 √x2 (α− d2 ) (∓i)q π d2 Z ∞ √
2
− 2x (s+ 1s ) α− d
dse s 2 −1
2 Γ(α) 0
√
d
2(∓i)q π 2  x2 α− d2 √

= Kα− d x2 ,
Γ(α) 2 2

√
For x2 ≤ 0, we change the variable t = s −x2 . By (7.318) and (3.103) we
transform (7.326) into
 √−x2 (α− d2 ) (∓i)q e∓i π2 (α− d2 ) π d2 Z ∞ √
−x2 1 d
dse±i 2 (s+ s ) sα− 2 −1
2 Γ(α) 0
q d
√ d
2(∓i) π 2
 −x 2  α− 2
p

= ∓i Kα− d ∓ i −x2
Γ(α) 2 2

q d
√ d
πi(∓i) π 2  2
−x α− 2 ±
 p

=± Hα− d −x2 .
Γ(α) 2 2

√ √ √
Then we notice that x2 and ∓i −x2 can be joined in x2 ∓ i0. 2

7.3 The Minkowski signature

1,d−1
R is called the Minkowski space. (We use the signature “mostly pluses”).
This case is especially important and rich.
Let us state the special case of Thm 7.2 for the Minkowski signature as a
separate theorem. We also introduce special notation.

50
Theorem 7.4
eipx
Z
dp
DαF/F (x) := (7.327)
(1 + p2 ± i0)α (2π)d
2i  √x2 ∓ i0 α− d2 p

=∓ d Kα− d x2 ∓ i0 (7.328)
Γ(α)(4π) 2 2 2

√
π  −x ± i0 α− d2
2
±
p

= d Hα− d −x2 ± i0 . (7.329)
Γ(α)(4π) 2 2 2

Suppose that x0 denotes the first coordinate of R1,d−1 , which we assume to

be timelike (having a negative coefficient in the scalar product). The remaining,
spacelike coordinates will be denoted ~x. Then we set

J ∨ := {x ∈ R1,d−1 : x2 ≤ 0, x0 ≥ 0},
J ∧ := {x ∈ R1,d−1 : x2 ≤ 0, x0 ≤ 0}.

Theorem 7.5
eipx
Z
dp
Dα∨/∧ (x) := (7.330)
(1 + p ∓ i0sgnp ) (2π)d
2 0 α

π  √−x2 + i0 α− d2 p
0 +


=θ(±x ) d Hα− d −x2 + i0
Γ(α)(4π) 2 2m 2

 √−x2 − i0 α− d2
!
p
−


+ Hα− d 2
−x − i0 (7.331)
2m 2

is a distribution whose support is contained in J ∨/∧ . Inside J ∨/∧ we have the

identity

2π  √−x2 α− d2 p
∨/∧


Dα (x) = d Jα− d −x2 . (7.332)
Γ(α)(4π) 2 2 2

We also have
DαF (x) + DαF (x) = Dα∨ (x) + Dα∧ (x). (7.333)

Proof. Let us now prove that the support of (7.330) with the minus signed is
contained in J ∨ . By the Lorentz invariance it suffices to prove that it is zero
for x0 < 0. We write
0 0
eipx dp e−ip x +i~p~x dp0 d~
Z Z
p
2 0 α
=
α
(p + 1 − i0sgnp ) p~ + 1 − (p + i0)2
2 0

Next we continuously deform the contour of integration, replacing p0 + i0 by

p0 + iR, where R ∈ [0, ∞[. We do not cross any singularities of the integrand
0 0
and note that e−ix (p +iR) goes to zero (remember that x0 < 0).

51
 Next we note that
Z 
1 1 
2 α
+ 2 α
eipx dp (7.334)
(1 + p + i0) (1 + p − i0)
d √
ππ 2  −x2 + i0 α− d2 + p

= Hα− d −x2 + i0
Γ(α) 2 2

√
 −x2 − i0 α− d2
!
p
−


+ Hα− d −x2 − i0 . (7.335)
2 2

Taking into account the support properties, we obtain (7.331). Finally, using
(3.99) we obtain (7.332). 2
We also introduce special notation for some solutions of

(1 + 2)α D(x) = 0. (7.336)

DαPJ (x) := Dα∨ (x) − Dα∧ (x), (7.337)

Dα(+)/(−) (x) := −iDαF (x) + iDα∨/∧ (x) (7.338)
= iDαF (x) − iDα∧/∨ (x). (7.339)

Here are expicit formulas for these solutions:

π  √−x2 + i0 α− d2 p
+
DαPJ (x) 0


=sgn(x ) d Hα− d −x2 + i0
Γ(α)(4π) 2 2m 2

√
 −x2 − i0 α− 2 d
!
p
−


+ Hα− d −x2 − i0 (7.340)
2m 2

2  px2 ± isgnx0 0 α− d2 p

Dα(±)


= d Kα− d x2 ± isgnx0 0 . (7.341)
Γ(α)(4π) 2 2 2

Note the identities

DαPJ (x) = −iDα(+) (x) + iDα(−) (x), (7.342)

DαF (x) − DαF (x) = iDα(+) (x) + iDα(−) (x). (7.343)

The support of DαPJ is contained in J ∨ ∪ J ∧ . In the interior of J ∨ ∪ J ∧ we have

the identity

2π  √−x2 α− d2 p
PJ 0


Dα (x) =sgn(x ) d Jα− d −x2 . (7.344)
Γ(α)(4π) 2 2 2

52
7.4 Fourier transforms and Bessel type functions
We will have two generic notations for elements of Rd : for k ∈ Rd , p := |k|, and
for x ∈ Rd , r := |x|.

Theorem 7.6 Let Reα > d2 . Then

Γ(α − d2 )
Z
d
(k 2 + m2 )−α dk = π 2 md−2α , (7.345)
Γ(α)
2  mr α− d2
Z
d
e−ikx (k 2 + m2 )−α dk = π 2 md−2α Kα− d (mr). (7.346)
Γ(α) 2 2

Proof. (7.345) is
∞
Γ( d2 )Γ(α − d2 )
Z
Sd−1 (p2 + m2 )−α pd−1 dp = Sd−1 2−1 md−2α .
0 Γ(α)

To prove (7.346) we use

Z ∞
1
A−α = e−tA tα−1 dt.
Γ(α) 0

We obtain
Z
(1 + k 2 )−α e−ikx dk
 r α 1 Z ∞ Z 2 rt
= dt dktα−1 e−(1+k ) 2 e−ikx
2 Γ(α) 0
 r α− d2 π d2 Z ∞ d −1 r
= dttα− 2 −1 e−(t+t ) 2 .
2 Γ(α) 0
2
Note that the rhs of (7.346) is locally integrable for α > 0. Therefore, (7.346)
is true also for Reα > 0, if the Fourier transform is interpreted appropriately.
We can actually extend (7.346) to all α ∈ C, in the sense of distributions.
In the range 0 < Reα < d2 both (7.346) and (7.347) are regular distributions
Using the asymptotics of the MacDonald function we easily see that as m & 0,
the rhs of (7.346) converges to

d Γ( d2 − α)  r 2α−d
π2 . (7.347)
Γ(α) 2

In the distributional sense this convergence is true for all α ∈ C except for
α = d2 + n, n = 0, 1, . . . , because then k2α
1 1
= kd+2n , which is the anomalous case
and an additional renormalization is needed.

53
7.5 Averages of plane waves on sphere
Consider the Euclidean space Rd . Let us take the average of plane waves over
the unit sphere Sd−1 . Let dΩ denote the standard measure on the sphere.

Z
e−ikx dΩ(k) (7.348)
Z π
= e−ipr cos θ sind−2 θdθSd−1 (7.349)
−π
Z 1
d−3
= e−iprw (1 − w2 ) 2 dwSd−1 (7.350)
−1
d d
= (2π) 2 J d −1 (pr)(pr)− 2 −1 . (7.351)
2

The Fourier transform of a radial function is radial and we have the identity
Z Z
d d
f (|k|)e−ikx dk = (2π) 2 f (p)J d −1 (rp)(rp)− 2 +1 pd−1 dp.
2

Here are the low dimensional cases:

Z
f (|k|)e−ikx dk
Z ∞
= 2 f (p) cos(pr)dp, d = 1;
0
Z ∞
= 2π f (p)pJ0 (pr)dp, d = 2;
Z0 ∞
sin(pr)
= 4π f (p)p2 dp, d = 3.
0 pr

In particular, in dimension 1 we obtain

Z ∞
2π 1/2  r α− 12
2 (1 + p2 )−α cos(pr)dp = Kα− 12 (r). (7.352)
0 Γ(α) 2

Setting m = α − 21 , we obtain the Poisson representation (2.65).

In dimension d = 2 we obtain
Z ∞
2π  r α−1
2π (1 + p2 )−α J0 (pr)pdp = Kα−1 (r).
0 Γ(α) 2

In dimension d = 3 we obtain
Z ∞ 3
sin(pr) 2π 2  r α− 32
4π (1 + p2 )−α pdp = Kα− 23 (r),
0 pr Γ(α) 2

which could be also deduced from (7.352) by differentiating wrt r.

54
7.6 General signature
Suppose that the scalar product on Rd has a signature with q minuses.
Theorem 7.7
Z
e−ikx (m2 + k 2 ± i0)−α dk (7.353)
 √ α− d2 √


q
π d2 md−2α 2(∓i) m x2

Kα− d m x2 x2 ≥ 0;
Γ(α) 2 2
=  √ α− d2 √ (7.354)
±π d2 md−2α πi(∓i)q m −x2 ±


Hα− d m −x2 x2 ≤ 0.

Γ(α) 2 2

Proof. We use π ∞
e∓i 2 α
Z
(A ± i0)−α = e±itA tα−1 dt. (7.355)
Γ(α) 0

(7.353) for x2 ≥ 0 is
 √x2 α e∓i π2 α Z ∞ Z 2
√
x2
dt dke±it(1+k ) 2 tα−1 e−ikx (7.356)
2 Γ(α) 0
 √x2 (α− d2 ) (∓i)q e∓i π2 (α− d2 ) π d2 Z ∞ √
x2 1 d
= dte±i 2 (t− t ) tα− 2 −1 (7.357)
2 Γ(α) 0
 √x2 (α− d2 ) (∓i)q π d2 Z ∞ √
x2 1 d
= dte− 2 (s+ s ) sα− 2 −1 , (7.358)
2 Γ(α) 0

which is the first case of (7.354).

(7.353) for x2 ≤ 0 is
 √−x2 α e∓i π2 α Z ∞ Z 2
√
−x2
dt dke±it(1+k ) 2 tα−1 e−ikx (7.359)
2 Γ(α) 0
 √−x2 (α− d2 ) (∓i)q e∓i π2 (α− d2 ) π d2 Z ∞ √
−x2
±i 2 (t+ 1t ) α− d
= dte t 2 −1 , (7.360)
2 Γ(α) 0

which is the second case of (7.354). 2

7.7 Averages of plane waves on hyperboloid

Consider the Minkowski space R1,d−1 . Let us take the average of plane waves
d−1
over the unit future hyperboloid H+ . Let dΩ denote the standard measure on
Hd−1
+ . Let x be a future oriented vector.

55
 Z
e−ikx dΩ(k) (7.361)
Z π
= e−ipr cosh θ sinhd−2 θdθSd−1 (7.362)
−π
Z 1
d−3
= e−iprw (w2 − 1) 2 dwSd−1 (7.363)
−1
1 d d
= e−iπ(m+ 2 ) (2π) 2 H −
d
−1
(pr)(pr)− 2 −1 . (7.364)
2

8 Integrals of Bessel functions

8.1 Scalar products
Theorem 8.1 We have the following indefinite integrals:
Z ∞
y  
xKm (ax)Km (bx)dx = aK m−1 (ay)K m (by) − bKm (ay)Km−1 (by) ,
y a 2 − b2
Re(a + b) > 0,
∞
y2 y2
Z
my
xKm (ax)2 dx = Km (ay)2 + Km (ay)Km−1 (ay) − Km−1 (ay)2 ,
y 2 a 2
Rea > 0.

Proof. Let us prove the first identity. Using Km = K−m , we write

 
y aKm−1 (ay)Km (by) − bKm (ay)Km−1 (by)
= ay −m+1 K−m+1 (ay)y m Km (by) − by m Km (ay)y −m+1 K−m+1 (by).

We differentiate using the recurrence relations. We obtain

a2 y −m+1 K−m (ay)y m Km (by) + aby −m+1 K−m+1 (ay)y m Km−1 (by)
−aby m Km−1 (ay)y −m+1 K−m+1 (by) − b2 y m Km (ay)y −m+1 K−m (by)
= (a2 − b2 )yKm (ay)Km (by).

56
Theorem 8.2 We have the following definite integrals:
Z ∞
π(am b−m − a−m bm )
xKm (ax)Km (bx)dx = ,
0 2 sin mπ(a2 − b2 )
m 6= 0, |Rem| < 1, Re(a + b) > 0;
Z ∞
ln a − ln b
xK0 (ax)K0 (bx)dx = , Re(a + b) > 0;
0 a 2 − b2
Z ∞
πm
xKm (ax)2 dx = ,
0 2 sin mπa2
m 6= 0, |Rem| < 1, Rea > 0;
Z ∞
1
xK0 (ax)2 dx = 2
, Rea > 0.
0 2a
Proof. Assume that 0 < Rem < 1. Then for small z
π π  z −m
Km (z) ≈ I−m (z) ≈ ,
2 sin πm 2 sin πmΓ(1 − m) 2
π π  z m−1
Km−1 (z) ≈ − Im−1 (z) ≈ .
2 sin π(m − 1) 2 sin πmΓ(m) 2
Therefore, for small y,
y  
aK m−1 (ay)K m (by) − bKm (ay)K m−1 (by)
a2 − b2
πy   ay m−1  by −m  ay −m  by m−1 
≈ 2 2 2
a −b
(a − b )(2 sin πm) Γ(m)Γ(1 − m) 2 2 2 2
m −m −m m
π(a b −a b )
= .
2 sin mπ(a2 − b2 )

8.2 Barnes integrals

Theorem 8.3 For Rez > 0, c > 0, c + Rem > 0,
Z ∞   x −2c−is−m
1 is   is
Km (z) = Γ c+ Γ c+ +m ds. (8.365)
8π −∞ 2 2 2

Proof.
1
Z 
is   is  z −2c−is−m ids
Γ c+ Γ c+ +m
2πi γ 2 2 2 2
∞ n ∞ n
X (−1)  z 2n−m X (−1)
  z 2n+m
= Γ(m − n) + Γ(−m − n)
n=0
n! 2 n=0
n! 2
∞  z 2n−m X ∞  z 2n+m
X π π
= +
n=0
n!Γ(1 − m + n) sin πm 2 n=0
n!Γ(1 + m + n) sin πm 2
π  
= I−m (z) − Im (z) = 2Km (z).
sin πm

57
The integral is convergent because of the estimates
  
Γ c + is Γ c + is + m  ≤ chsi2c+Rem e− π2 s .
  
(8.366)
 2 2 
  
 z −2c−is−m 
  ≤ |z|−2c−Rem es arg z . (8.367)
 2 
Of course, we have a version for the Hankel functions.
The following representation holds only for real x

Theorem 8.4 For 0 < c < 12 Rem,

∞
Γ(c + it2 )  x m+ 12 −2c−it
Z
1
Jm (x) = √ dt. (8.368)
4 π −∞ Γ(m + 1 − c − it2 ) 2

8.3 Integral of Sonine and Schafheitlin

([1] Exercise 4.14, p. 236):
Z ∞
Jm (xξ)Jk (yξ)dξ
0 ξλ
1+m+k−λ


yk Γ y2
 
2 1−m+k−λ 1+m+k−λ
=
F , ; k + 1; 2 , y < x;
2λ x1+k−λ Γ(k + 1)Γ 1+m−k−λ
2
2 2 x
1+m+k−λ
m


x2
 
x Γ 2 1+m−k−λ 1+m+k−λ
=
F , ; m + 1; 2 , x < y.
2λ y 1+m−λ Γ(m + 1)Γ 1−m+k−λ
2
2 2 y

8.4 Another integral

The following integrals essentially Watson, 13.31 (1):
Z ∞
exp(−qy 2 )Jm (ay)Jm (by)ydy (8.369)
0
1  a2 + b2   ab  π
= exp − Im , Rem > −1, | arg q| < . (8.370)
2q 4q 2q 2
Here is its another version:
Z ∞
exp(−qy 2 )Im (αy)Im (βy)ydy (8.371)
0
1  α2 + β 2   αβ  π
= exp Im , Rem > −1, | arg q| < . (8.372)
2q 4q 2q 2

58
9 Klein-Gordon equation in 1 + 1 dimensions
9.1 Hyperbolic coordinates
We have the coordinates
1 1
x+ := (t + y), x− := (t − y). (9.373)
2 2
The space R1,1 is divided into 4 sectors:
J++ = {(t, y) : t > |y|} = {x+ > 0, x− > 0},
J−− = {(t, y) : t < −|y|} = {x+ < 0, x− < 0},
J+− = {(t, y) : y > |t|} = {x+ > 0, x− < 0},
J−+ = {(t, y) : y < −|t|} = {x+ < 0, x− > 0}.
There are 4 hyperbolic coordinate systems:
J++ : t = r cosh φ, y = r sinh φ,
p 1 1
r = t2 − y 2 , φ = log(t + y) − log(t − y),
2 2
1 φ 1 −φ
x+ = re , x− = re ;
2 2
J−− : t = −r cosh φ, y = −r sinh φ,
p 1 1
r = t2 − y 2 , φ = log(−t − y) − log(−t + y),
2 2
1 φ 1 −φ
x+ = − re , x− = − re ;
2 2
J+− : t = r sinh φ, y = r cosh φ,
p 1 1
r = y 2 − t2 , φ = log(y − t) − log(y + t),
2 2
1 φ 1 −φ
x+ = re , x− = − re ;
2 2
J−+ : t = −r sinh φ, y = −r cosh φ,
p 1 1
r = y 2 − t2 , φ = log(−y + t) − log(−y − t),
2 2
1 1
x+ = − reφ , x− = re−φ ;
2 2

The d’Alembertian in all sectors is

2 = −∂t2 + ∂y2 = −∂x+ ∂x− = ∂r2 + r−1 ∂r − r−2 ∂φ2 . (9.374)

9.2 Plane waves

We look for solutions of the Klein-Gordon equation
(−2 + 1)F = 0

59
of the form Z Z
1
F (t, y) = F (ξ, η)ei(−tτ +yη) dξdη.
2π
We obtain
(−τ 2 + η 2 + 1)F (ξ, η) = 0.
Positive frequency plane waves are parametrized by ψ ∈ R and given by
fψ (x, y) := ei(−x cosh ψ+y sinh ψ) .
They are solutions of the Klein-Gordon equation. Here is the plane wave in
hyperbolic coordinates:
J++ : fψ (r, φ) = e−ir cosh(φ−ψ) ,
J−− : fψ (r, φ) = eir cosh(φ−ψ) ,
J+− : fψ (r, φ) = e−ir sinh(φ−ψ) ,
J−+ : fψ (r, φ) = eir sinh(φ−ψ) .
Positive frequency solutions are given by
Z
g(x, y) = fψ (x, y)g(ψ)dψ, (9.375)

where g is a distribution on R. We will denote by H the Hilbert space of

functions on R2 of the form (9.375) with g ∈ L2 (R). We will treat them as
“nice” solutions of the Klein-Gordon equation.

9.3 Hyperbolic waves

Let us introduce for µ ∈ R
Z
1
fµ := fψ eiµψ dψ. (9.376)
2π
Proposition 9.1 Here are the expressions of hyperbolic waves in hyperbolic
coordinates:
1
J++ : fµ (r, φ) = Kiµ (ir)eiµφ , (9.377)
π
1
J−− : fµ (r, φ) = Kiµ (−ir)eiµφ , (9.378)
π
π
e± 2 m
J+− : fµ (r, φ) = Kiµ (±r)eiµφ , (9.379)
π
π
e∓ 2 m
J−+ : fµ (r, φ) = Kiµ (±r)eiµφ . (9.380)
π
Proposition 9.2 We can expand plane waves in hyperbolic waves:
Z
fψ = fµ e−iµψ dµ. (9.381)

60
9.4 Wave equation in 1 + 1 dimension
We will use polar coordinates. Using the expressions (4.141) we obtain

2 := ∂x2 − ∂t2 = cos(2φ)∂r2 − 2 sin(2φ)r−1 ∂r ∂φ − cos(2φ)r−2 ∂φ2

+2r−2 sin(2φ)∂φ − r−1 cos(2φ)∂r . (9.382)

Thus

r2 2 = cos(2φ) r2 ∂r2 − r∂r − 2 sin(2φ) r∂r − 1 ∂φ − cos(2φ)∂φ2

On functions of the form rλ f (φ) we obtain an operator

Λλ = cos(2φ)(λ2 − 2λ) − 2 sin(2φ)(λ − 1)∂φ − cos(2φ)∂φ2 .

Let us substitute w = sin(2φ). One can distinguish two regions:

p √
∂φ = 2 1 − w2 ∂w , cos(2φ) = 1 − w2 , cos(2φ) > 0,
p √
∂φ = −2 1 − w2 ∂w , cos(2φ) = − 1 − w2 , cos(2φ) < 0.

We obtain
     
 λ 1 − λ + 2 λ − 1 w∂w + (1 − w2 )∂w
2
,
2 2 2
p
Λλ = −4 1 − w2    
 λ λ − 1 − λw∂w + (1 − w2 )∂w
2
.
2 2

This corresponds to the Jacobi equation

(1 − w2 )∂w
2
− 2(m + 1)w∂w + n(n + 2m + 1) (9.383)

with
λ
n = −m = , (9.384)
2
λ
−n = m= . (9.385)
2

10 Elements of partial differential equations

10.1 General formalism
Let X
P (k) = Pα k α (10.386)
α

be a polynomial in d variables k = (k1 , . . . , kd ). We set

1
Di := ∂i . (10.387)
i

61
We consider the differential operator
X
P (D) = Pα D α . (10.388)
α

One can consider two problems: find solutions of the homogeneous problem

P (D)ζ = 0, (10.389)

and, given f , find solutions of the inhomogeneous problem.

P (D)ζ = f. (10.390)

To solve the inhomogeneous problem, it is useful to introduce a Green’s function

or a fundamental solution of P (D), which is a distribution G satisfying

P G(x) = δ(x). (10.391)

Note that if we know Green’s function, then

Z
Gf (x) = G(x − y)f (y)dy (10.392)

solves the inhomogeneous equation.

Green’s function is not uniquely defined. In fact, if G is Green’s function
and ζ solves the homogeneous problem, then G + ζ is also Green’s function.
We will see, however, that often we will have distinguished Green’s functions.
Sometimes we will also have distinguished solutions.
We can look for Green’s functions using the Fourier transformation. In fact,
suppose that G ∈ S(Rd ). We can write
Z
dk
G(x) = G(k)eikx , (10.393)
(2π)d
Z
dk
δ(x) = eikx . (10.394)
(2π)d

Equation (10.391) becomes

Z Z
ikx dk dk
P (k)G(k)e d
= eikx . (10.395)
(2π) (2π)d

Thus formally, Z
1 ikx dk
G(x) = e . (10.396)
P (k) (2π)d
If (10.396) is well defined, then it provides a distinguished Green’s function for
1
P (D). Unfortunately, often, especially if P has zeros, P (k) is not a well defined
distribution and (10.396) is problematic.

62
10.2 Laplace equation in d = 1
Consider P (D) = −∂x2 , so that P (k) = k 2 .
Space of solutions is
a + bx. (10.397)
Examples of Green’s functions:

G+ (x) = −θ(x)x, (10.398)

G− (x) = −θ(−x)|x|, (10.399)
1
G0 (x) = − |x|. (10.400)
2
1
k2 is not a distribution, but it can be regularized.

10.3 Helmholtz equation in d = 1

Consider P (D) = −∂x2 + m2 , so that P (k) = k 2 + m2 .
Space of solutions is
a+ emx + a− e−mx (10.401)
Examples of Green’s functions:

sinh(mx)
G+ (x) = −θ(x) , (10.402)
m
| sinh(mx)|
G− (x) = −θ(−x) , (10.403)
m
e−m|x|
G(x) = . (10.404)
2m
1
k2 +m2 is a distribution. By the method of residues we can compute:

eikx e−m|x|
Z
1
dk = , (10.405)
2π k 2 + m2 2m

which reproduces (10.404).

10.4 Laplace equation in d = 2

Consider P (D) = −∂x2 − ∂y2 , so that P (k) = kx2 + ky2 . Introduce complex coor-
dinates
1 1
z = (x + iy), z = (x − iy). (10.406)
2 2
Then
∂z = ∂x − i∂y , ∂z = ∂x + i∂y . (10.407)
We have
−∂x2 − ∂y2 = ∂z ∂z . (10.408)

63
Therefore, solutions are sums of a holomorphic and antiholomorphic function:

ζ1 (z) + ζ2 (z). (10.409)

In polar coordinates
x = r cos φ, y = r sin φ,
the Laplacian is
1 1
−∂x2 − ∂y2 = − ∂r r∂r − 2 ∂φ2 . (10.410)
r r
We claim that rotationally symmetric Green’s functions have the form
1 1 1
a− log r = a − log(z) − log(z) + log 4. (10.411)
2π 4π 4π
We easily check that a + b log(r) solves the Laplace equation outside of the
origin, either in the polar coordinates, or noticing the decomposition into a
holomorphic and an antiholomorphic function. It is more difficult to determine
the coefficient in front of log r.
Note that P G = δ means that for any test function φ ∈ Cc∞ (R2 )
Z Z
P (D)φ(x, y)G(x, y)dxdy = φ(x, y)P (D)G(x, y)dxdy = φ(0, 0). (10.412)

Assume that φ is rotationally symmetric and G(r) = a + b log(r). We have

Z
P (D)φ(x, y)G(x, y)dxdy (10.413)
x2 +y 2 >2
Z ∞
= 2π (−∂r r∂r )φ(r)G(r)dr (10.414)

Z ∞
= 2π φ(r)(−∂r r∂r )G(r)dr (10.415)

+2π(φ0 ()G() − φ()G0 ()) (10.416)
→ −2πbφ(0). (10.417)
1
Hence, b = − 2π .
1
k θ(k) is not a distribution. We can regularize it as in (??). Then we obtain
a Green’s function with a rather strange looking constant:


eixkx +iyky − 1 dkx dky
Z Z
G(x, y) = (10.418)
(kx2 + ky2 ) (2π)2
2 +k 2 <1
kx y

eixkx +iyky dkx dky

Z Z
+ (10.419)
(kx2 + ky2 ) (2π)2
2 +k 2 >1
kx y

1 r γ
= − log − . (10.420)
2π 2 2π

64
10.5 Helmholtz equation in d = 2
Consider P (D) = −∂x2 − ∂y2 + m2 , so that P (k) = kx2 + ky2 + m2 . The method of
Fourier transformation gives a distinguished Green’s function

eixkx +iyky
Z Z
dkx dky
G(x, y) = (10.421)
(m + kx + ky ) (2π)2
2 2 2

ei|k|r cos φ
Z Z
1
= 2
|k|d|k|dφ (10.422)
(2π) (m2 + |k|2 )
1
= K0 (mr), (10.423)
2π
where K0 (z) is the 0th MacDonald function. Note the asymptotics around zero:
z
K0 (z) ' − log − γ. (10.424)
2
Thus, in order to obtain a zero-mass Green’s function, we need to renormalize.
Writing Gm (r) for the Green’s function with mass m, as defined in (10.420) and
(10.423), we obtain the massless Green’s functions by the following limit:
1

lim Gm + log m = G0 . (10.425)
m→0 2π

10.6 Wave equation in d = 1 + 1

Consider 2 = ∂t2 − ∂y2 . Introducing coordinates

1 1
u+ := (t + y), u− := (t − y), (10.426)
2 2
we have
2 = ∂u+ ∂u− . (10.427)
Therefore, the general solution is

χ+ (t + y) + χ− (t − y). (10.428)

We have the retarded Green’s function, the advanced Green’s function, and the
Pauli-Jordan solution:

D+ (t, y) = θ(t − |x|) = θ(t + x)θ(t − x), (10.429)

−
D (t, y) = θ(−t − |x|) = θ(−t − x)θ(−t + x), (10.430)
PJ
D (t, y) = θ(t − |x|) − θ(−t − |x|) (10.431)
= θ(t − x) − θ(−t − x) = θ(t + x) − θ(−t + x). (10.432)

Let us compute the retaqrded Green’s function by the Fourier transform method.
We introduce E, p, the dual variables to t, y. Besides, p+ := E + p and

65
p− := E − p. We have dEdp = 21 dp+ dp− .

ei(−Et+px) dEdp
Z
+
D (t, y) = (10.433)
(−E 2 + p2 − i0sgnE)(2π)2
e−i(u+ p− +u− p+ ) dp− dp+
Z
1
=
(10.434)
2 − p+ p− − i0sgn(p+ + p− ) (2π)2
Z −iu− p+ Z −iu+ p−
1 e dp+ e dp−
= − (10.435)
2 (p+ + i0)2π (p− + i0)2π
= θ(u+ )θ(u− ). (10.436)

The Feynman propagator is obtained by the Wick rotation from the Eu-
clidean propagator (Green’s function of the Laplacian). More precisely, we set
E = ikx , t = ix:
Z Z −iEt+ipy
1 e dEdp
DF (t, y) = 2 2 2
(10.437)
(2π) −E + p − i0
Z Z −ikx x+ipy
i e dEdp
= 2 2 2
= iDE (i−1 t, y). (10.438)
(2π) −E + p − i0

Setting
1  x2 + y 2  γ
DE (x, y) = − log − , (10.439)
4π 4 2π
we obtain
i  −t2 + y 2 + i0  iγ
DF (t, y) = −log −
4π 4 2π
i  | − t2 + y 2 |  1
iγ
= − log + θ |t| − |y| − ,
4π 4 2 2π
i  | − t2 + y 2 |  1
1
iγ
D(±) (t, x) = ∓ log + θ t − |y| − θ − t − |y| ∓ .
4π 4 2 2 2π

11 Miscellanea
11.1 ...
Identity

(1 − w2 ) m + κ 
+ ∂r mw + (1 − w2 )∂w


(2m + κ) ∂w + (2m + κ)w∂r =
r r
m 
− ∂r − (m + κ)w + (1 − w2 )∂w .


+
r

66
11.2
We set z = cos φ:
1 1
∂x2 + ∂y2 + ( − m2 ) 2
4 y
1 1 2 1 1 
= ∂r + ∂r + 2 ∂φ + ( − m2 ) 2
2
r r 4 sin φ
1 1  1 1 
= ∂r2 + ∂r + 2 (1 − z 2 )∂z2 − z∂z + ( − m2 )
r r 4 1 − z2
We use
1 1 1 1 1 z
(1 − z 2 )− 2 (m+ 2 ) ∂z (1 − z 2 ) 2 (m+ 2 ) = ∂z − (m + ) ,
2 1 − z2
obtaining
1 1
 1 1  1 1
(1 − z 2 )− 2 (m+ 2 ) (1 − z 2 )∂z2 − z∂z + ( − m2 ) (1 − z 2 ) 2 (m+ 2 )
4 1 − z2
1
2
= (1 − z 2 )∂z2 − (2m + 2)z∂z − m + .
2

11.3
√
We set t = cos 2φ, so that sin2 φ = 1−t
2 , cos2 φ = 1+t
2 , ∂φ = −2 1 − t2 ∂t .
1 1 1 1
∂x2 + ( − α2 ) 2 + ∂y2 + ( − β 2 ) 2 + ∂z2
4 x 4 y
2 1
= ∂r2 + ∂r + 2 (1 − w2 )∂w
2
− 2w∂w +
r r
!
1  2 2 1 2 2 1 2 2 
4(1 − t )∂t − 4t∂t + ( − α ) +( −β ) .
1 − w2 4 1+t 4 1−t

We use
1 1 1 1 1 1 1 1
(1 + t)− 2 (α+ 2 ) (1 − t)− 2 (β+ 2 ) ∂t (1 + t) 2 (α+ 2 ) (1 − t) 2 (β+ 2 )
1 1 1 1 1 1
= ∂t + (α + ) − (β + ) .
2 2 1+t 2 2 1−t
We obtain
1 1 1 1
(1 + t)− 2 (α+ 2 ) (1 − t)− 2 (β+ 2 )
 1 2 1 2 
× 4(1 − t2 )∂t2 − 4t∂t + ( − α2 ) + ( − β2)
4 1+t 4 1−t
1
(α+ 21 ) 1
(β+ 12 )
×(1 + t) 2 (1 − t) 2
!
α − β α+β+2  (α + β + 2)(α + β)
2 2
= 4 (1 − t )∂t + − t ∂t + .
2 2 4

67
.1 The digamma function
In our paper we use the digamma function:
∂z Γ(z)
ψ(z) := . (.440)
Γ(z)
Here are its properties:

1
ψ(1 + z) = ψ(z) + , (.441)
z
ψ(z) − ψ(1 − z) = −π cot(πz), (.442)
1  1 
ψ +z −ψ − z = π tan(πz), (.443)
2 2
 1
2 log 2 + ψ(z) + ψ z + = 2ψ(2z), (.444)
2
ψ(1) = −γ, (.445)
1
ψ = −γ − 2 log 2. (.446)
2
The inverse of the Gamma function is an analytic function with the derivative
1 ψ(z)
∂z =− , (.447)
Γ(z) Γ(z)
1 
∂z = (−1)n n!, n = 0, 1, 2, ... (.448)
Γ(z) z=−n


It is also useful to introduce

1 1
the shifted kth harmonic number Hk (z) := + ··· + , (.449)
z z+k−1
1 1
the kth harmonic number Hk := + · · · + = Hk (1), (.450)
1 k
Γ(z + k)
the Pochhammer symbol (z)k := (.451)
Γ(z)
(
(z)(z + 1) · · · (z + k − 1), k ≥ 0,
= 1
(z+k)(z+k+1)···(z−1) , k ≤ 0.

Some of their properties are collected below:

Hk+n (z) = Hn (z) + Hk (z + n), (.452)

Hk (z) = −Hk (1 − z − k), (.453)
ψ(z + k) = ψ(z) + Hk (z), (.454)
ψ(1 + k) = −γ + Hk , (.455)
k
(z)k = (−1) (1 − k − z)k , (.456)
∂z (z)n = Hn (z)(z)n . (.457)

68
References
[1] Andrews, G.E., Askey, R., Roy, R.: Special functions, Cambridge Univer-
sity Press, 1999
[2] Everitt, W.N. and Kalf, H.: The Bessel differential equation and the Han-
kel transform, to appear in J. Comput. Appl. Math.
[3] Watson, G.N. A treatise on the theory of Bessel functions, Cambridge
University Press, 2nd ed. 1948
[4] Titchmarsh, E.C.: Eigenfunction expansions I, Oxford University Press,
2nd ed. 1962

69