APS/123-QED

Heun equation, Teukolsky equation, and type-D metrics

D.Batic∗
Institute for Theoretical Physics
Swiss Federal Institute of Technology
CH-8093 Zürich, Switzerland

H.Schmid
arXiv:gr-qc/0701064v2 15 Jan 2007

UBH Software & Engineering GmbH

†
D-92263 Amberg, Germany
(Dated: February 7, 2008)

Abstract
Starting with the whole class of type-D vacuum backgrounds with cosmological constant we show
that the separated Teukolsky equation for zero rest-mass fields with spin s = ±2 (gravitational
waves), s = ±1 (electromagnetic waves) and s = ±1/2 (neutrinos) is an Heun equation in disguise.

PACS numbers: 02.30Gp, 02.30.Hq, 0420.Jb, 0.462.+v, 0.470.Bw

Keywords: Heun equation, Teukolsky equation, type-D metrics, QFT in curved spacetimes

∗
Electronic address: [email protected]
†
Electronic address: [email protected]

1
I. INTRODUCTION

According to Ronveaux (1995) the Heun equation (HE) is the most general second order
linear ODE of the form
 
d2 y 1 + 2α1 1 + 2α2 1 + 2α3 dy αβz − qA
2
+ + + + y=0 (1)
dz z z−1 z − zS dz z(z − 1)(z − zS )
where b
a ∈ C\{0, 1}, α1 , α2 , α3 , α, β, qA are complex arbitrary parameters, and 0, 1, b
a, and
∞ are regular singularities with exponents {0, −2α1 }, {0, −2α2 }, {0, −2α3 }, and {α, β},
respectively. Equation (1) has been originally constructed by the German mathematician
Karl Heun (1889) as a generalization of the hypergeometric equation (HYE). In order to see
how HE degenerates to the HYE we can first multiply (1) by z(z − 1)(z − b
a), then we set
b
a = 1, and qA = αβ, and finally we take out a factor (z − 1), leaving the HYE in its standard
form. Hence, we can always think to a HYE as a degenerated equation of Heun’s type. To
underline the importance of (1) we recall that it contains the generalized spheroidal equation
(GSWE), the Coulomb spheroidal equation, Lamé, Mathieau, and Ince equations as special
cases. The fields of applications of the HE in physics are so large that it is not possible
to describe them here in detail. However, a review of many general situations relevant to
physics, chemistry, and engineering where the HE occurs can be found in Ronveaux (1995)
(pp 341). Here, we will show which role plays the HE in quantum field theory in curved
spacetimes.
To understand the motivation underlying the present work we give a short review on studies
concerning exact solutions of the Teukolsky equation in some black hole geometries. In
the 70’s, and 80’s we find a large number of publications regarding the angular equation
obtained after separation of variables from the Teukolsky wave equation on Kerr manifolds.
See, for instance, Press, and Teukolsky (1973), Breuer et al. (1977), Fackerell, and Crossman
(1977), Leahy, and Unruh (1979), Chakrabarti (1984), and Seidel (1989). According to these
references we will also name the solution of the angular equation as spin-weighted spheroidal
function (SWSF). The radial equation has been investigated by Bardeen, and Press (1973),
Page (1973), Lee (1976), Arenstorf et al. (1978). The common picture emerging from
all previous studies is that the radial equation cannot be in general related to any known
differential equation of mathematical physics. This view changed with the work of Blandin
et al. (1983). They showed that the SWSF’s may be obtained by means of an elementary
transformation from Heun confluent functions. Three years later Leaver (1983) proved that

2
the radial, and angular parts of the Teukolsky master equation (TME) in the Kerr geometry
are generalized spheroidal wave equations. Finally, Suzuki et al. (1998) showed that the
radial, and angular part equations arising from the TME in the Kerr-Newman-deSitter
metric (KNdS) after separation of variables are Heun equations. It is interesting to observe
that if we let the cosmological constant go to zero (i.e. the KNdS geometry goes over to
the Kerr-Newman metric) their HEs become a confluent HE which, in turn, coincides with
the GSWE given by Leaver in 1983. Hence, the following question arises quite naturally,
namely: is it possible to reduce the TME in any physical relevant type D metric to a HE?
To conclude this short review we cite what Wu, and Cai wrote in 2003: ”it is not clear until
now whether the generalized Teukolsky equation in the general type D vacuum backgrounds
with cosmological constant can be transformed into a Heun equation.”
Our paper is organized as follows: in Sec. II we shortly present some results due to Kamran,
and McLenaghan (1987) concerning the separation of the TME in any type-D background.
In Sec. III-VII we show that the TME can be transformed in any physical relevant type D
metric into a HE.

II. BACKGROUND

Let D0 denote the class of algebraically special Petrov type D vacuum metrics with
cosmological constant. According to Thm. 2.1 in Kamran, and McLenaghan (1987) there
exists a system of local coordinates (u, v, w, x) in which such metrics can be written as


ds2 = 2 θ1 θ2 − θ3 θ4 (2)

with a symmetric null tetrad (θ1 , θ2 , θ3 , θ4 ) given by

p  
1 Z(w, x) f W (w) dw
θ = √ (ǫ1 du + m(x) dv) + 2 ,
2T (w, x) Z(w, x) g W (w)
p  
2 Z(w, x) W (w) f dw
θ = √ (ǫ1 du + m(x) dv) − 2 ,
2T (w, x) Z(w, x) g W (w)
p  
3 Z(w, x) X(x) dx
θ = √ (ǫ2 du + p(w) dv) + i = θ4 ,
2T (w, x) Z(w, x) X(x)
r
1 + f2
Z(w, x) := ǫ1 p(w) − ǫ2 m(x), g :=
2
where all functions are real-valued and ǫ1 , ǫ2 , and f are constants such that ǫ21 + ǫ22 6= 0.
Depending on whether f W 2 (w) is positive, negative or zero the metric (2) possesses a two-

3
parameter abelian group of isometries whose orbits are timelike, spacelike or null at a given
point, respectively. By integration of the Einstein-Maxwell field equation it results that the
general solution Af∗ in the class D0 can be specified as follows (Thm. 2.2 ibid.)

ǫ1 = b2 cos γ, ǫ2 = sin γ,
 
2 2 2 2 ℓ2 (1 − b2 cos2 γ)
m(x) = − c x + b k + ǫ2 − 2cℓx,
ǫ22
 
2 2 2 2 k 2 (b2 − ǫ22 )
p(w) = + b c w + ℓ + 2
cos γ + 2b2 ckw,
cos γ
T (w, x) = a(cw cos γ + k)(cx sin γ + ℓ) + 1,
f W 2 (w) = c2 b4 g4 w 4 cos2 γ + cf3 w 3 cos γ + f2 w 2 + f1 w + f0 ,
X 2 (x) = c2 g4 x4 sin2 γ + cκ1 x3 sin γ + κ2 x2 + g1 x + g0 ,
κ1 := acf1 cos γ − 2akf2 + 3ak 2 f3 + 4(ℓ − ab4 k 3 )g4 ,

κ2 := 3acℓf1 cos γ − (1 + 6akℓ)f2 + 3k(1 + 3akℓ)f3 + 6[ℓ2 − b4 k 2 (1 + 2akℓ)]g4

where f0 , f1 , f2 , f3 , g0 , g1 , g4 , a, b, c, k, ℓ, and γ are real parameters satisfying the relations

acg1 sin γ − 3a2 cℓ2 f1 cos γ + 2aℓ(1 + 3akℓ)f2 − [1 + 3akℓ(2 + 3akℓ)]f3

+ 4[b4 k − aℓ3 + 3ab4 k 2 ℓ(1 + akℓ)]g4 = 0, (3)

c2 (g0 sin2 γ − b4 f0 cos2 γ) + c[(2aℓ3 + b4 k)f1 cos γ − ℓg1 sin γ] − (b4 k 2 + ℓ2 + 4akℓ3 )f2
+ (b4 k 3 + 3kℓ2 + 6ak 2 ℓ3 )f3 + (3ℓ4 − b8 k 4 − 6b4 k 2 ℓ2 − 8ab4 k 3 ℓ3 )g4 = 0, (4)

and are restricted to a range such that (2) is non-singular with signature minus two. More-
over, the cosmological constant Λ is expressed in terms of the above parameters by

Λ = −3[a2 c2 f0 cos2 γ − a2 ckf1 cos γ + a2 k 2 f2 − a3 k 3 f3 + (1 + a2 b4 k 4 )g4 ]. (5)

√
Let A ∗ denote the subclass of solutions in Af∗ obtained by setting b = 1, c = 2, k = ℓ = 0
and γ = π/4. If in addition a = f = 1, such solutions recover the vacuum case with
cosmological constant of the seven-parameter family of Plebanski and Demianski (1976)
containing the Kerr-Newman-de Sitter metric as a special case.
Let B−0
be the subclass of solutions in Af∗ such that a = 0, b = c = 1, ℓ = 0 and γ = π/2.
0
If in addition f = 1, B− reduces to the vacuum case with cosmological constant of Carter’s
Be− (1968). By B 0 we will denote the subclass of solutions in Af∗ obtained by setting a = 0,
+

4
b = c = 1, and k = γ = 0. If we let f = 1, such a solution becomes the vacuum case with
e+ . The Carter’s B
cosmological constant of Carter’s B e± describe all non-accelerating type

D metrics in a coordinate system in which the components of the metric and the Maxwell
field are rational functions.
√
Let C ∗ denote the subclass of solutions in Af∗ obtained by setting a = 1, b = 0, c = 2,
k = ℓ = 0 and γ = π/4. If in addition f = 1, C ∗ reduces to the accelerating C -metric of
Levi-Civita (1918).
Finally, let C 00 denote the subclass of solutions in Af∗ obtained by setting a = 0, b = 1,
c = 0, k = 0, ℓ = 1 and γ = 0. If in addition f = 1, C 00 becomes the Robinson-Bertotti
solution (1959).
Following Kamran, and McLenaghan (1987) the Teukolsky equation can be written in a
compact form by introducing a spin parameter s which can assume the values ±2, ±1, and
±1/2. For s = 2, 1 and 1/2 we have

[(D − (2s − 1)ǫ + ǫ − 2sρ − ρ)(∆ − 2sγ + µ) − (δ + π − α − (2s − 1)β − 2sτ )(δ + π − 2sα)
1
− 2(s − )(s − 1)Ψ2 ]Φs = 0, (6)
2

and for s = −2, −1 and −1/2

[(∆ − (2s + 1)γ − γ − 2sµ + µ)(D − 2sǫ − ρ) − (δ − τ + β − (2s + 1)α − 2sπ)(δ − τ − 2sβ)
1
− 2(s + )(s + 1)Ψ2 ]Φs = 0 (7)
2

where Ψ2 is the non-zero Newman-Penrose component of the Weyl tensor and the Φs ’s are
defined in terms of the field components as given in Table 1 (pp 286 ibid.). According to
Thm. 3.1 (ibid.) for all solutions in the class D0 , and for all s = ±2, ±1, ±1/2 equations
(6), and (7) possess a separable solution of the form

T |s|+1 i|s|B/2
Φs = ei(ru+qv) e Θs (w, x)
Z |s|/2
′ ′
where r, q are arbitrary real constants, dB = Z −1 (ǫ1 m dw + ǫ2 p dx), and Θs (w, x) =
Gs (w)Hs (x). Moreover, (6), and (7) separate into the pair of decoupled ODE’s

Dws Lws Gs (w) + fs (w)Gs (w) = λs Gs (w), (8)

Dxs Lxs Hs (x) − gs (x)Hs (x) = λs Hs (x) (9)

5
where λs is a separation constant, the functions fs , and gs are given in Table 1 (pp 290-291
ibid.), and
 
d i 1 pr − ǫ2 q dW
Dws = W + 2
f (1 + ǫ(s)) − (1 − ǫ(s)) + (1 − |s|) ,
dw 1 + f f W dw
d ǫ1 q − mr dX
Dxs = −iX + iǫ(s) + i(|s| − 1) ,
dx X dx
d i  2
 pr − ǫ2 q dW
Lws = −f W + (1 + ǫ(s)) − f (1 − ǫ(s)) − |s|f ,
dw 1 + f 2 W dw
d ǫ1 q − mr dX
Lxs = iX + iǫ(s) + i|s| ,
dx X dx
where ǫ(s) is a sign function such that ǫ(s) = +1 for s > 0, and ǫ(s) = −1 for s < 0.

√
III. THE METRIC A ∗ : b = 1, c = 2, k = ℓ = 0 , γ = π/4

In this case we have

√
2 √ √
ǫ1 = ǫ2 = , m(x) = − 2 x2 , p(w) = 2 w 2 , T (w, x) = awx + 1,
2
X3 Y4
f W 2 (w) = g4 w 4 + fn w n = g4 (w − wi ),
n=0 i=1
4
Y
X 2 (x) = g4 x4 + af1 x3 − f2 x2 + g1 x + g0 = g4 (x − xi ).
i=1

Moreover, for i,j = 1, · · · , 4 let wi and xi denote the i-th root of the polynomial equations
f W 2 (w) = 0, and X 2 (x) = 0, respectively. Throughout this section we shall assume that
wi 6= wj , and xi 6= xj for every i 6= j. Furthermore, in the present case (5) reads
 
2 Λ
g4 = − a f0 + .
3
Making use of the expression for Ψ2 given by (2.7g) in Kamran, and McLenaghan (1984),
the functions fs (w), and gs (x) entering, respectively in (8), and (9) are computed to be
h  √  i
fs (w) = −(1 − 2|s|) 2g4 (1 − |s|)w 2 + 2i 2rǫ(s) + ag1 (1 − |s|) w ,
h  √  i
gs (x) = −(1 − 2|s|) 2g4 (1 − |s|)x2 − 2 2rǫ(s) − af1 (1 − |s|) x .

Hence, the equation for Gs (w) becomes

 
d 2 dGs dGs
fW + Γ(w) + Qs (w)Gs = 0 (10)
dw dw dw

6
with
√ f2 − 1
Γ(w) := i 2 2 (2rw 2 − q),
f +1

   
2
√ f2 − 1
Qs (w) := λs +2g4 (1−|s|)(1−2|s|)w + 2i 2r −2s + 2 + ag1 (1 − |s|)(1 − 2|s|) w+
f +1
 ′′ ′ 2
 √ W
′
2f (2rw 2 − q)2
2
f |s| W W + (1 − |s|)(W ) + i 2s(2rw − q) + .
W (1 + f 2 )2 W2

Equation (10) can be further simplified. To this aim let us make the substitution

Gs (w) = eh(w) ϕs (w).

′
If we require that h = −Γ/(2f W 2 ) we obtain the following equation for ϕs (w)
 
d 2 dϕs
fW + Rs (w) ϕs = 0 (11)
dw dw

where

Rs (w) := λs + 2g4 (1 − |s|)(1 − 2|s|)w 2 − Cs w+

 ′′ ′ 2
 √ W
′
(2rw 2 − q)2
f |s| W W + (1 − |s|)(W ) + i 2s(2rw 2 − q) +
W 2f W 2

with
√
Cs := 4i 2sr − ag1 (1 − |s|)(1 − 2|s|).

Let us introduce the following functions

σ±q (w) := 2rw ± q, fs (w) := 2g4 (1 − |s|)(1 − 2|s|)w 2 − Cs w + λs ,

and let us define constants

3
Y
c−1
i := g4 (wi − wj ), i = 1, 2, 3.
j=1
j6=i

By means of the homographic substitution

w − w1 w2 − w4
z= (12)
w − w4 w2 − w1

mapping the points w1 , w2 , w3 , w4 , ∞ to 0, 1, zS , ∞, z∞ with

w2 − w4 w3 − w1
z∞ := , zS := z∞ (13)
w2 − w1 w3 − w4

7
equation (11) becomes
d 2 ϕs dϕs es (z)ϕs = 0
2
+ P (z) +Q (14)
dz dz
with

1 1 1 2
P (z) = + + − ,
z z − 1 z − zS z − z∞
es (z) = B1 + B2 +
Q
B3
+
2
+
A1
+
A2
+
A3
+
A∞
z 2 (z − 1)2 (z − zS ) 2 (z − z∞ ) 2 z z − 1 z − zS z − z∞

where for i = 1, 2, 3
3
!
1 X 3ag 1
A∞ = (2|s|2 − 3|s| + 2) wi + (1 − |s|)(1 − 2|s|) − (2 − |s|)(1 + 2|s|)w4 ,
z∞ (w4 − w1 ) i=1
g 4
 2
2
s ci σ−q (wi )
Bi = − − i√ ,
2 2(wi − w4 )
c1 w2 − w1 w3 − w1
A1 = − gs (w1 ), A2 = c2 gs (w2 ), A3 = c3 gs (w3 ),
z∞ w4 − w1 zS (w4 − w1 )

and
√
g4 2s
gs (wi ) = |s|p(wi) − fs (wi ) + i σ+q (wi(w4 − 2wi )) +
2 wi − w4
  
X3 3
2
g 4 ci  Y 
σ−q (wi2) σ+q (wi2 ) wj − 2wiσ+q  wj  ,
wi − w4 j=1 j=1
j6=i j6=i

 
3
X 3 3
X  Y
p(wi ) := 2(2|s| − 3)wi2 + (4 − 3|s|)wi wj + (|s| − 2)w4  wj − 2wi  + 2(|s| − 1) wj .
j=1 j=1 j=1
j6=i j6=i j6=i

(15)
If we make the F-homotopic transformation

ϕs (z) = z α1 (z − 1)α2 (z − zS )α3 (z − z∞ )ϕ

es (z)

and require that αi2 = −Bi , then (14) becomes

d2 ϕ
es dϕes
2
+ Pbs (z) bs (z)ϕ
+Q es = 0 (16)
dz dz

with

1 + 2α1 1 + 2α2 1 + 2α3 b b b b

Pbs (z) = + + , bs (z) = A1 + A2 + A3 + A∞ .
Q
z z−1 z − zS z z − 1 z − zS z − z∞

8
Now, a direct computation gives
4
!
X
b∞ = (1 − |s|)(1 − 2|s|)
A wi +
ag1
, (17)
z∞ (w4 − w1 ) i=1
g4
3 4
!
X X
bi = − (1 − |s|)(1 − 2|s|)
A wi +
ag1
. (18)
i=1
z∞ (w4 − w1 ) i=1
g4

Taking into account that for the metric A ∗ equation (3) reduces to ag1 − f3 = 0, and
replacing f3 by ag1 in the equation f W 2 = 0 it can be checked that the following relation
holds
4
X ag1
wi = − .
i=1
g4
Hence, the coefficients (17), and (18) are zero, and (16) reduces to an Heun equation with
terms αβ and accessory parameter qA (according to the notation in (1)) given by

b1 − zS A
αβ = −(zS + 1)A b2 − A
b3 = zS A
b3 + A
b2 , b1
qA = −zS A (19)

where

b1 = A1 − α1 − α2 (1 + 2α1 ) − α1 + α3 (1 + 2α1 ) − 1 ,
A (20)
zS z∞
b2 = A2 + α1 + α2 (1 + 2α1 ) − α2 + α3 (1 + 2α2 ) −
A
1
, (21)
zS − 1 z∞ − 1
b3 = A3 + α2 + α3 (1 + 2α2 ) + α1 + α3 (1 + 2α1 ) −
A
1
. (22)
zS − 1 zS z∞ − zs

Concerning (9) we find the following equation for Hs (x), namely

  h
d 2 dHs
X + −λs + 2g4 (1 − |s|)(1 − 2|s|)x2 + Cbs x+
dx dx
  √ ′ 
′′ ′2 2 X (2rx2 + q)2
|s| XX − (|s| − 1)X − 2s(2rx + q) − Hs = 0 (23)
X 2X 2

with
√
Cbs := 4 2rs + (1 − |s|)(1 − 2|s|)af1 .

Let us introduce the following notation

bfs (x) = 2g4 (1 − |s|)(1 − 2|s|)x2 + Cbs x − λs ,

Y 3
−1
b
ci = g4 (xi − xj ), i = 1, 2, 3.
j=1
j6=i

9
By means of the homographic substitution

x − x1 x2 − x4
z= (24)
x − x4 x2 − x1

mapping the points x1 , x2 , x3 , x4 , ∞ to 0, 1, xS , ∞, x∞ with

x2 − x4 x3 − x1
x∞ = , xS = x∞ (25)
x2 − x1 x3 − x4

our equation (23) becomes

d 2 Hs dHs
2
+ Ps (z) + Qs (z)Hs = 0 (26)
dz dz

with

1 1 1 2
Ps (z) = + + − ,
z z − 1 z − xS z − x∞
Be1 e2
B Be3 2 e1
A Ae2 e3
A e∞
A
Qs (z) = 2 + + + + + + +
z (z − 1)2 (z − xS )2 (z − x∞ )2 z z − 1 z − xS z − x∞

where for i = 1, 2, 3
3
!
1 X 3ag1
e∞ =
A (2|s|2 − 3|s| + 2) xi + (1 − |s|)(1 − 2|s|) − (2 − |s|)(1 + 2|s|)x4 ,
x∞ (x4 − x1 ) i=1
g4
 2
2
ei = − s + √b
B
ci σ+q (xi )
,
2 2(xi − x4 )
e1 = − b
A
c1
b
gs (x1 ), A e2 = b
c2
x2 − x1
b
gs (x2 ), e3 = b
A c3
x3 − x1
b
gs (x3 ),
x∞ x4 − x1 xS (x4 − x1 )

and
√
g4 2s
b p(xi ) − bfs (xi ) −
gs (xi ) = |s|e σ−q (xi (x4 − 2xi )) −
2 xi − x4
  
3
X 3
c2i
g4 b  Y 
σ+q (x2i ) σ−q (x2i ) xj − 2xi σ−q  xj  ,
xi − x4 j=1 j=1
j6=i j6=i

 
3
X 3 3
X  Y
pe(xi ) := 2(2|s|−3)x2i +(4−3|s|)xi xj +(|s|−2)x4  xj − 2xi  +2(|s|−1) xj . (27)
j=1 j=1 j=1
j6=i j6=i j6=i

If we make the F-homotopic transformation

e s (z)
Hs (z) = z αb1 (z − 1)αb2 (z − xS )αb3 (z − x∞ )H

10
 ei , then (56) becomes
bi2 = −B
together with the requirement that α

es
d2 H e
+ b s (z) dHs + Q
P b s (z)H
es = 0 (28)
dz 2 dz

with

P
α1 1 + 2b
b s (z) = 1 + 2b +
α2 1 + 2b
+
α3
, b s (z) = A1 + A2 + A3 + A∞ .
Q
z z−1 z − xS z z − 1 z − xS z − x∞

Now, a direct computation gives

4
!
(1 − |s|)(1 − 2|s|) X af1
A∞ = xi + , (29)
x∞ (x4 − x1 ) i=1
g4
3 4
!
X (1 − |s|)(1 − 2|s|) X af1
Ai = − xi + . (30)
i=1
x∞ (x4 − x1 ) i=1
g4

Going back to the polynomial equation X 2 (x) = 0 it can be verified that

4
X af1
xi = − .
i=1
g4

This implies that (29), and (30) are zero, and (28) becomes an Heun equation with

αβ = xS A3 + A2 , qA = −xS A1 (31)

bi
where the Ai ’s can be directly obtained from the formulae (20)-(22) for the coefficients A
by means of the formal substitutions A bi −→ Ai , Ai −→ A ei , αi −→ αbi , zS −→ xS and
z∞ −→ x∞ .

IV. 0 : b = c = 1, a = ℓ = 0 , γ = π/2
THE METRIC B−

In the present case we have

ǫ1 = 0, ǫ2 = 1, m(x) = −(x2 + k 2 ), p(w) = 2kw, T (w, x) = 1,

f W 2(w) = f2 w 2 + f1 w + f0 = f2 (w − w1 )(w − w2 ),
Y4
2 4 2
X (x) = g4 x + κ2 x + g1 x + g0 = g4 (x − xi ), κ2 = −f2 + 3kf3 − 6k 2 g4 .
i=1

Moreover, for i,j = 1, · · · , 4 let w1 , w2 , and xi denote the roots of the polynomial equations
f W 2 (w) = 0, and X 2 (x) = 0, respectively. Throughout this section we shall assume that

11
w1 6= w2 , and xi 6= xj for every i 6= j. Finally, let us recall that in the present situation (3),
(4) and (5) read, respectively

4kg4 − f3 = 0, g0 − k 2 (k 2 g4 − kf3 + f2 ) = 0, Λ = −3g4 . (32)

Since the term x3 is not present in the expression for X 2 it can be easily checked that the
roots of the polynomial equation X 2 (x) = 0 satisfy the condition x1 + x2 + x3 + x4 = 0.
Taking into account the expression for Ψ2 given by (2.8g) in Kamran, and McLenaghan
(1984), the functions fs (w) and gs (x) entering, respectively in (8), and (9) are computed to
be

fs (w) = 0,
 ′ 
(X 2 (x)) 6X 2 (x) + CΛ (k + ix)
gs (x) = −2irǫ(s)(1 − 2|s|)(k + ix) + (1 − |s|)(1 − 2|s|) i −
k − ix 3(k − ix)2
with CΛ := −2k(3f2 + 4Λk 2 ) + 3ig1 . The equation for Gs (w) becomes
 
d 2 dGs b dGs
fW + Γ(w) + Qs (w)Gs = 0 (33)
dw dw dw
with
2
b f −1
Γ(w) := 2i 2 (2rkw − q),
f +1

2irk
 ′′ ′2

2
Qs (w) := λs − 1 + ǫ(s) − f (1 − ǫ(s)) + f |s| W W + (1 − |s|)W +
1 + f2
′
W 4f (2rkw − q)2
2is(2rkw − q) + .
W (1 + f 2 )2 W2
In order to simplify (33) we make the transformation
b
Gs (w) = eh(w) ϕs (w)

requiring that b b
′
h = −Γ/(2f W 2 ). Hence, we end up with the following equation for ϕs (w)
 
d 2 dϕs
fW + Rs (w)ϕs = 0 (34)
dw dw
where
 ′′ ′2
 W
′
(2rkw − q)2
Rs (w) := λs − 2irkǫ(s) + f |s| W W + (1 − |s|)W + 2is(2rkw − q) + .
W fW2
By means of the transformation
w − w1
z= (35)
w2 − w1

12
mapping the points w1 , w2 , ∞ to 0, 1, ∞ our equation (34) becomes

d 2 ϕs dϕs es (z)ϕs = 0
2
+ Ps (z) +Q (36)
dz dz

with
1 1 es (z) = B1 + B2 + A1 + A2
Ps (z) = + , Q
z z−1 z2 (z − 1)2 z z−1
where
 2  2
s 2rkw1 − q s 2rkw2 − q
B1 = − +i , B2 = − −i ,
2 f2 (w2 − w1 ) 2 f2 (w2 − w1 )
Y2
λs + 2irk(ǫ(s) − s) |s|(|s| − 2) 2
A1 = − + + 2 (2rkwi − q), A2 = −A1 .
f2 2 f2 (w2 − w1 )2 i=1

If we make the F-homotopic transformation

ϕs (z) = z α1 (z − 1)α2 ϕ
es (z)

with the requirement that αi 2 = −Bi then (36) reads

′′ ′
es (z) + Pbs (z)ϕ
ϕ bs (z)ϕ
es (z) + Q es (z) = 0 (37)

with

1 + 2α1 1 + 2α2 bs (z) = (A1 + A2 )z + α2 + α1 (1 + 2α2 ) − A1 .

Pbs (z) = + , Q
z z−1 z(z − 1)

Since A2 = −A1 it follows that (37) is an HYE. Notice that an HYE can be always thought
as a degenerate Heun equation since it can be recovered from (1) by setting for instance
zS = 1 and qA = αβ or zS = q = 0 or 1 + 2α3 = 0 and q = zS αβ. Concerning the ODE (9)
for Hs (x) we find
  h  
d 2 dHs ′′ ′2
X + − (λs − 2irkǫ(s)(1 − 2|s|)) + 4rsx + |s| XX + (1 − |s|)X −
dx dx
′
X (x2 + k 2 )2
2rs(x2 + k 2 ) − r2 +
X X2 
′
(X 2 ) 6X 2 + iCΛ (x − ik)
(1 − |s|)(1 − 2|s|) − Hs = 0. (38)
x + ik 3(x + ik)2

Let us introduce the following function

σ± (x) := x ± k 2 .

13
Moreover, let b
ci be defined as in the previous section. By means of the homographic substi-
tution (24) equation (38) becomes
d 2 Hs dHs
2
+ Ps (z) + Qs (z)Hs = 0 (39)
dz dz
with
1 1 1 2
Ps (z) = + + − ,
z z − 1 z − xS z − x∞

e1
B e2
B e3
B 2 Bek
Qs (z) = + + + + +
z2 (z − 1)2 (z − xS )2 (z − x∞ )2 (z − xk )2
e1
A Ae2 e3
A e∞
A Aek x1 + ik
+ + + + , xk := x∞
z z − 1 z − xS z − x∞ z − xk x4 + ik
where for i = 1, 2, 3
3
!
1 X
e∞ =
A xi − 3x4 ,
x∞ (x4 − x1 ) i=1
 !
+ 2 2 4
Y
ei = − s rb
c σ
i k (xi ) ek = −2(1 − |s|)(1 − 2|s|) 1 + kC Λ
B + , B (xj + ik)−1 ,
2 xi − x4 3g4 j=1

e1 = − b
A
c1
b
gs (x1 ), e2 = b
A c2
x2 − x1
b
gs (x2 ), Ae3 = b c3 (x3 − x1 )
b
gs (x3 ),
x∞ x4 − x1 xS (x4 − x1 )
4
! 3
Y Y
ek = (1 − |s|)(1 − 2|s|)(x4 + ik)
A τ (k) + i
CΛ
p(k) (xj + ik) −1
(xi + ik)−1 ,
x∞ (x1 − x4 ) 3g4 j=1 i=1

with

g4 2rs c2i r 2
2g4b
b
gs (xi ) = λs − 2irkǫ(s)(1 − 2|s|) + |s|ep(xi ) + σ+ (xi x4 ) − σ+ (x2i )·
2 xi − x4 xi − x4
  
X3 3  
 2 Y  (1 − |s|)(1 − 2|s|) xi − x4 i xi − ik
· σ− (xi ) xj − 2xi σ−  xj  − − CΛ ,
j=1 j=1
xi + ik b
ci 3 xi + ik
j6=i j6=i

3
!
X
τ (k) := 3x4 − xi k 2 + 2i(x1 x2 + x1 x3 − x2 x4 − x3 x4 − x1 x4 + x2 x3 )k+
i=1

3x1 x2 x3 − x2 x3 x4 − x1 x2 x4 − x1 x3 x4 ,

3
! 3
!
X X
p(k) := 3k 4 − xi + 5x4 k3 + x1 x2 + x2 x3 + x1 x3 − 3x4 xi k2 +
i=1 i=1

i(x1 x2 x4 + x1 x3 x4 + x2 x3 x4 − 3x1 x2 x3 )k − x1 x2 x3 x4 ,

14
and pe(xi ) given by (27). Inserting the roots of the polynomial equation X 2 (x) = 0 into the
expressions for B ek , and A
ek we checked with the help of Maple 9.5 that Bek = A ek = 0. Let

us work with the F-homotopic transformation

e s (z).
Hs (z) = z αb1 (z − 1)αb2 (z − xS )αb3 (z − x∞ )H

ei . Then (39) becomes

bi2 = −B
Moreover, we require that α

e ′′ (z) + P
H b s (z)H
e ′ (z) + Q
b s (z)H
e s (z) = 0 (40)
s s

with
P
α1 1 + 2b
b s (z) = 1 + 2b +
α2 1 + 2b
+
α3
, Qb s (z) = A1 + A2 + A3 .
z z−1 z − xs z z − 1 z − xs
Taking into account that the Ai ’s can be directly obtained from the formulae (20)-(22) for
the coefficients A bi by means of the formal substitutions A bi −→ Ai , Ai −→ A
ei , αi −→ α
bi ,
zS −→ xS and z∞ −→ x∞ , and that A ek = 0 implies that

Y4
CΛ
τ (k) + i p(k) (xj + ik)−1 = 0,
3g4 j=1

we obtain
3
X 3
X  
ei − 1 1 1
Ai = A + +
i=1 i=1
x∞ x∞ − xS x∞ − 1
" 3
(1 − |s|)(1 − 2|s|)(x1 − x2 ) X
= 3x4 − xi +
(x4 − x2 )(x1 − x4 ) i=1
3 4
!#
Y CΛ Y
(xi + ik)−1 T (k) − i p(k)(x4 + ik) (xj + ik)−1
i=1
3g4 j=1
(1 − |s|)(1 − 2|s|)(x1 − x2 )
= M(k)
(x4 − x2 )(x1 − x4 )
with
3
X 3
Y
M(k) := 3x4 − xi + (T (k) + (x4 + ik)τ (k)) (xi + ik)−1 , (41)
i=1 i=1

  
3
! 3 3
X  X  X 
T (k) := xi (2x4 − xi ) − 3x24 k2 + i  xi 2x24 + x2j  −
i=1 i=1 j=1
j6=i
   
3 3 3 3 3 3
!
 X Y   X Y  Y X
4x4 x1 xi + xi  k + x24 x1 xi + xi  + xi xi − 6x4 .
i=1 i=1 i=1 i=1 i=1 i=1
i6=1 i6=1 i6=1 i6=1

15
Inserting the roots of X 2 = 0 into (41) it can be verified that M(k) = 0 for all real k. Hence,
we have reduced (40) to an Heun equation with terms αβ, and qA given by (31).

V. 0 : b = c = 1, a = k = γ = 0
THE METRIC B+

In the present case we have

ǫ1 = 1, ǫ2 = 0, p(w) = w 2 + ℓ2 ,
m(x) = −2ℓx, T (w, x) = 1,
4
Y
2 4 2
f W (w) = g4 w + f2 w + f1 w + f0 = g4 (w − wi ),
i=1
2 2
X (x) = κ2 x + g1 x + g0 = κ2 (x − x1 )(x − x2 ), κ2 = −(f2 + 2ℓ2 Λ)

where for i = 1, · · · , 4 wi , and x1 , x2 denote the roots of the polynomial equations f W 2 (w) =
0, and X 2 (x) = 0, respectively. Throughout this section we shall assume that wi 6= wj for
every i 6= j, and x1 6= x2 . Finally, since the term w 3 is not present in the expression for f W 2
it can be easily verified that the roots of the equation f W 2 (w) = 0 satisfy the condition
w1 + w2 + w3 + w4 = 0. Taking into account the expression for Ψ2 given by (2.9g) in Kamran,
and McLenaghan (1984), the functions fs (w) and gs (x) entering, respectively in (8), and (9)
are computed to be
" ′
#
fW2 bΛ (w + iℓ)
6f W 2 − C
fs (w) = −2irǫ(s)(1 − 2|s|)(w + iℓ) − (1 − |s|)(1 − 2|s|) −
w − iℓ 3(w − iℓ)2
gs (x) = 0

bΛ := −2iℓ(3g2 + 4Λℓ2 ) + 3f1 . Concerning the ODE (8) for Gs (w) we find
with C
 
d 2 dGs dGs
fW + Γℓ (w) + Qs (w)Gs = 0 (42)
dw dw dw
with
f2 − 1 2
Γℓ (w) = 2ir 2 (w + ℓ2 ),
f +1

2ir  
Qs (w) = λs − 2ǫ(s)(1 − 2|s|)rℓ + 2irǫ(s)(1 − 2|s|)w − 2
1 + ǫ(s) − f 2 (1 − ǫ(s)) w+
1+f
  ′
4f r 2 (w 2 + ℓ2 )2
′′ ′2 2 2 W
f |s| W W + (1 − |s|)W + 2irs(w + ℓ ) + +
W (1 + f 2 )2 W2
" ′
#
fW2 bΛ (w + iℓ)
6f W 2 − C
(1 − |s|)(1 − 2|s|) − .
w − iℓ 3(w − iℓ)2

16
 ′
By means of the transformation Gs (w) = eh(w) ϕs (w) with the requirement that h =
−Γℓ /(2f W 2 ) equation (42) simplifies to
 
d 2 dϕs
fW + Rs (w)ϕs = 0 (43)
dw dw

where
  ′
r 2 (w 2 + ℓ2 )2
e ′′ ′2 2 2 W
Rs (w) = λs − 4irsw + f |s| W W + (1 − |s|)W + 2irs(w + ℓ ) + +
W fW2
" ′
#
fW 2 2 b
6f W − CΛ (w + iℓ)
(1 − |s|)(1 − 2|s|) −
w − iℓ 3(w − iℓ)2

with
es := λs − 2ǫ(s)(1 − 2|s|)rℓ.
λ

For notational purposes let us introduce the function σ± (w) := w ± ℓ2 . By means of the
homographic substitution (12) equation (43) becomes

d 2 ϕs dϕs es (z)ϕs = 0
2
+ Ps (z) +Q (44)
dz dz

with
1 1 1 2
Ps (z) = + + − ,
z z − 1 z − zS z − z∞

es (z) = B1 + B2 +
Q
B3
+
2
+
Bℓ
+
z 2 (z − 1)2 (z − zS ) 2 (z − z∞ ) 2 (z − zℓ )2
A1 A2 A3 A∞ Aℓ w1 − iℓ
+ + + + , zℓ := z∞
z z − 1 z − zS z − z∞ z − zℓ w4 − iℓ

where for i = 1, 2, 3
3
!
1 X
A∞ = wi − 3w4 ,
z∞ (w4 − w1 ) i=1
 2 4
!
s rci σ+ (wi2 ) bΛ Y
ℓC
Bi = − −i , Bℓ = −2(1 − |s|)(1 − 2|s|) 1 − i (wj − iℓ)−1 ,
2 wi − w4 3g4 j=1
c1 w2 − w1 c3 (w3 − w1 )
A1 = − gs (w1 ), A2 = c2 gs (w2 ), A3 = gs (w3 ),
z∞ w4 − w1 zS (w4 − w1 )
4
! 3
(1 − |s|)(1 − 2|s|)(w4 − iℓ) CbΛ Y Y
−1
Aℓ = τb(ℓ) − i pb(ℓ) (wj − iℓ) (wi − iℓ)−1 ,
z∞ (w1 − w4 ) 3g4 j=1 i=1

17
with

e g4 2irs 2g4 c2i r 2

gs (wi ) = −λs + |s|p(wi) − σ+ (wi w4 ) − σ+ (wi2 )·
2 w − w4 wi − w4
  i 
X3 3  
 2 Y  (1 − |s|)(1 − 2|s|) wi − w4 i b wi + iℓ
· σ− (wi ) wj − 2wi σ−  wj  − − CΛ ,
j=1 j=1
wi − iℓ ci 3 wi − iℓ
j6=i j6=i

3
!
X
τb(ℓ) := 3w4 − wi ℓ2 − 2i(w1 w2 + w1 w3 − w2 w4 − w3 w4 − w1 w4 + w2 w3 )ℓ+
i=1

3w1 w2 w3 − w2 w3 w4 − w1 w2 w4 − w1 w3 w4 ,

3
! 3
!
X X
pb(ℓ) := 3ℓ4 + i wi + 5w4 ℓ3 + w1 w2 + w2 w3 + w1 w3 − 3w4 w i ℓ2 −
i=1 i=1

i(w1 w2 w4 + w1 w3 w4 + w2 w3 w4 − 3w1 w2 w3 )ℓ − w1 w2 w3 w4 ,

constants ci defined as in Section III, and p(wi ) given by (15). Inserting the roots of the
polynomial equation f W 2 (w) = 0 into the expressions for Bℓ , and Aℓ we checked with the
help of Maple 9.5 that Bℓ = Aℓ = 0. In order to reduce (44) to a Heun equation we just
need to introduce the F-homotopic transformation

ϕs (z) = z α1 (z − 1)α2 (z − zS )α3 (z − z∞ )ϕ

es (z).

with exponents αi such that αi 2 = −Bi , and to proceed as we did for the equation (40) in
Section IV. Concerning the equation for Hs (x) we find
   ′
d 2 dHs X
X + −λs + 2ǫ(s)ℓr − 2s(2ℓrx + q) +
dx dx X
 ′′ ′2
 (2ℓrx + q)2 
|s| XX + (1 − |s|)X − Hs = 0. (45)
X2
By means of the transformation z = (x − x1 )/(x2 − x1 ) mapping the points x1 , x2 , ∞ to
0, 1, ∞ our equation (45) becomes
d 2 Hs dHs
+ Ps (z) + Qs (z)Hs = 0 (46)
dz 2 dz
with
1 1 e1
B e2
B e1
A e2
A
Ps (z) = + , Qs (z) = + + +
z z−1 z2 (z − 1)2 z z−1

18
where
 2  2
e1 = − s 2rℓx1 + q s 2rℓx2 + q
e2 = −
B − , B
+ ,
2 κ2 (x2 − x1 ) 2 κ2 (x2 − x1 )
Y2
e1 = − λs − 2ℓr(ǫ(s) − s) + |s|(|s| − 2) −
A
2
(2rℓxi + q), e2 = −A
A e1 .
κ2 2 κ22 (x2 − x1 )2 i=1

If we make the F-homotopic transformation

e s (z)
Hs (z) = z αb1 (z − 1)αb2 H

ei then (46) reads

bi2 = −B
with the requirement that α

H b s (z)H
e s′′ (z) + P b s (z)H
es′ (z) + Q e s (z) = 0 (47)

with
e e e1
P
α1 1 + 2b
b s (z) = 1 + 2b +
α2
, b s (z) = (A1 + A2 )z + α
Q
b2 + α
b1 (1 + 2b
α2 ) − A
.
z z−1 z(z − 1)
e2 = −A
Since A e1 it follows that (47) is an HYE.

√
VI. THE METRIC C ∗ : a = 1, c = 2, b = k = ℓ = 0 , γ = π/4

In this case we have

√
2 √
ǫ1 = 0, ǫ2 = , m(x) = − 2 x2 , p(w) = 0, T (w, x) = wx + 1,
2
3
X 3
Y
f W 2 (w) = fn w n = f3 (w − wi ), X 2 (x) = g4 x4 + f1 x3 − f2 x2 + g1 x + g0 .
n=0 i=1

Taking into account that (3), and (4) imply in the present framework that

g1 = f3 , g0 = 0,

the function X 2 (x) can be written as follows

3
Y
2 3 2


X (x) = x g4 x + f1 x − f2 x + f3 = g4 x (x − xi ).
i=1

Moreover, for i,j = 1, · · · , 3 let wi and xi denote the i-th root of the polynomial equations
f W 2 (w) = 0, and X 2 (x) = 0, respectively. Throughout this section we shall assume that
wi 6= wj , and xi 6= xj for every i 6= j. Furthermore, in the present case (5) reads
 
Λ
g4 = − + f0 .
3

19
Finally, the roots of X 2 (x) = 0 satisfy the following useful relations
3
X 3
Y
f1 f3
xi = − , xi = − .
i=1
g4 i=1
g4

Making use of the expression for Ψ2 given by (2.10g) in Kamran, and McLenaghan (1984),
the functions fs (w), and gs (x) entering, respectively in (8), and (9) are computed to be

fs (w) = −f3 (1 − |s|)(1 − 2|s|)w,

 
√ (1 − |s|)(1 − 2|s|) dX 2 2 2
gs (x) = 2 2rǫ(s)(1 − 2|s|)x − − X + f3 .
x dx x
Hence, the equation for Gs (w) becomes
 
d 2 dGs dGs
fW + Γq + Qs (w)Gs = 0 (48)
dw dw dw
with
√ 1 − f2
Γq := i 2q ,
1 + f2

 ′′ ′2

Qs (w) := λs + f3 (1 − |s|)(1 − 2|s|)w + f |s| W W + (1 − |s|)W −
′
√ W 2f q 2 1
i 2sq + .
W (1 + f ) W 2
2 2

Equation (48) can be further simplified. To this aim let us transform Gs (w) according to

Gs (w) = eh(w) ϕs (w).

′
If we require that h = −Γq /(2f W 2) we obtain the following equation for ϕs (w)
 
d 2 dϕs
fW + Rs (w) ϕs = 0 (49)
dw dw
where
 ′′ ′2

Rs (w) := λs + f3 (1 − |s|)(1 − 2|s|)w + f |s| W W + (1 − |s|)W −
′
√ W q2
i 2sq + .
W 2f W 2
For notational purposes let us define constants
3
Y
c−1
i := f3 (wi − wj ), i = 1, 2, 3.
j=1
j6=i

20
By means of the homographic substitution

w − w1
z= (50)
w2 − w1

mapping the points w1 , w2 , w3, ∞ to 0, 1, zS , ∞ with zS := (w3 − w1 )/(w2 − w1 ), equation

(49) becomes
d 2 ϕs dϕs es (z)ϕs = 0
2
+ P (z) +Q (51)
dz dz
with

1 1 1
P (z) = + + ,
z z − 1 z − zS
es (z) = B1 + B2 +
Q
B3
+
A1
+
A2
+
A3
z 2 (z − 1)2 (z − zS ) 2 z z − 1 z − zS

where for i = 1, 2, 3
 2
s qci
Bi = − + i√ , Ai = −(w1 − w2 )ci gs (wi ),
2 2
  
  3
 |s| X 
gs (wi ) = λs + f3 (1 − |s|)(1 − 2|s|)wi − (2 − |s|) − q 2 c2i  wj − 2wi  .
2 j=1
j6=i

If we make the F-homotopic transformation

ϕs (z) = z α1 (z − 1)α2 (z − zS )α3 ϕ

es (z)

and require that αi2 = −Bi , then (51) becomes

d2 ϕ
es dϕes
2
+ Pbs (z) bs (z)ϕ
+Q es = 0 (52)
dz dz

with
1 + 2α1 1 + 2α2 1 + 2α3 b b b
Pbs (z) = + + , Qbs (z) = A1 + A2 + A3 .
z z−1 z − zS z z − 1 z − zS
b1 + A
Finally, it can be checked that A b2 + A
b3 = 0 with

b1 = A1 − α1 − α2 (1 + 2α1 ) − α1 + α3 (1 + 2α1 ) ,
A
zS
α
b2 = A2 + α1 + α2 (1 + 2α1 ) − 2 + α3 (1 + 2α2 )
A ,
zS − 1
b3 = A3 + α2 + α3 (1 + 2α2 ) + α1 + α3 (1 + 2α1 ) .
A
zS − 1 zS

21
Hence, (52) reduces to an Heun equation with terms αβ and accessory parameter qA given
by (19). Concerning (9) we find the following equation for Hs (x), namely
    
d 2 dHs
√ ′′ ′2 √ 2X
′
x4
X + −λs + 4 2rsx + |s| XX − (|s| − 1)X − 2 2rsx − 2r 2 2 +
dx dx X X
′
!#
X2 X 2 f3
(1 − |s|)(1 − 2|s|) −2 2 + Hs = 0. (53)
x x x

Let us introduce the following notation

−1 −1 −1
b
c1 = g4 x1 x2 , b
c2 = g4 x1 (x1 − x2 ), b
c3 = g4 x2 (x2 − x1 ).

By means of the homographic transformation

x x1 − x3
z= (54)
x − x3 x1
mapping the points 0, x1 , x2 , x3 , ∞ to 0, 1, xS , ∞, x∞ with
x1 − x3 x2 x∞
x∞ = , xS = (55)
x1 x2 − x3
our equation (53) becomes
d 2 Hs dHs
+ Ps (z) + Qs (z)Hs = 0 (56)
dz 2 dz
with
1 1 1 2
Ps (z) = + + − ,
z z − 1 z − xS z − x∞
Be1 e2
B Be3 2 e1
A Ae2 e3
A e∞
A
Qs (z) = 2 + + + + + + +
z (z − 1)2 (z − xS )2 (z − x∞ )2 z z − 1 z − xS z − x∞
where

e∞ = x1 + x2 − 3x3 ,
A
x3 x∞
 2  2
e s2 e s √ c2 x21
b e s √ c3 x22
b
B1 = − , B2 = − + 2r , B3 = − + 2r
4 2 x1 − x3 2 x2 − x3
e1 = − b
A
c1 I
b
gs , Ae2 = b c2 x1 II
b
gs , A e3 = bc3 x2 III
b
g4
gIs = λs + |s|A,
gs , b
x∞ x3 x3 xS 2

g4 √ x1 x3
gII
bs = λs + |s|B + 2 2rs − (1 − |s|)(1 − 2|s|) [g4 (x1 − x3 )(x1 − x2 ) + f3 x1 ] −
2 x1 − x3
x4 x2
c22 g4 1
4r 2 b ,
x1 − x3

22
  
g4 √ x2 x3 f3
gIII
bs = λs + |s|C + 2 2rs − (1 − |s|)(1 − 2|s|) g4 (x2 − x3 )(x2 − x1 ) − −
2 x2 − x3 x2
x1 x42
4r 2 b
c23 g4 ,
x2 − x3

A := (|s| − 2)x3 (x1 + x2 ) + 2(|s| − 1)x1 x2 ,

B := 2(2|s| − 3)x21 + (4 − 3|s|)x1 x2 + (|s| − 2)x3 (x2 − 2x1 ),
C := 2(2|s| − 3)x22 + (4 − 3|s|)x1 x2 + (|s| − 2)x3 (x1 − 2x2 ).

If we make the F-homotopic transformation

e s (z)
Hs (z) = z αb1 (z − 1)αb2 (z − xS )αb3 (z − x∞ )H

ei , then (56) becomes

bi2 = −B
together with the requirement that α
es
d2 H e
+ b s (z) dHs + Q
P b s (z)H
es = 0 (57)
dz 2 dz
with
b s (z) = 1 + 2b
P
α1 1 + 2b
+
α2 1 + 2b
+
α3
, b s (z) = A1 + A2 + A3 .
Q
z z−1 z − xS z z − 1 z − xS
ei with i = 1, 2, 3 it can be checked that
Now, employing the expressions for the coefficients A
X3 X 3  
ei − 1 1 1
Ai = A + + = 0.
i=1 i=1
x∞ x∞ − xS x∞ − 1

Hence, (57) becomes an Heun equation with

αβ = xS A3 + A2 , qA = −xS A1

where again the Ai ’s can be directly obtained from the formulae (20)-(22) for the coefficients
bi by means of the formal substitutions A
A bi −→ Ai , Ai −→ A ei , αi −→ α
bi , zS −→ xS and
z∞ −→ x∞ .

VII. THE METRIC C 00 : b = ℓ = 1, a = c = k = γ = 0

In this case we have

ǫ1 = 1 = p(w) = T (w, x), ǫ2 = 0 = m(x),

X2
2
f W (w) = fn w n = f2 (w − w1 )(w − w2 ),
n=0
X (x) = −Λx2 + g1 x + g0 = −Λ(x − x1 )(x − x2 ).
2

23
Throughout this section we shall assume that w1 6= w2 , and x1 6= x2 . Making use of the
expression for Ψ2 given by (2.11g) in Kamran, and McLenaghan (1984), the functions fs (w),
and gs (x) entering, respectively in (8), and (9) are computed to be

2
fs (w) = Λ(1 − |s|)(1 − 2|s|), gs (x) = 0.
3

The equation for Gs (w) becomes

 
d 2 dGs b dGs + Qs (w)Gs = 0
fW +Γ (58)
dw dw dw

b := 2ir(f 2 − 1)/(f 2 + 1), and

with Γ
′
2 W
Qs (w) := λs − Λ(1 − |s|)(1 − 2|s|) + 2irs +
3 W
 ′′ ′ 2
 4r 2 f 1
f |s| W W + (1 − |s|)(W ) + .
(1 + f ) W 2
2 2

′
By means of the transformation Gs (w) = eh(w) ϕs (w) with the requirement that h =
−Γ/(2f W 2 ) we obtain the following equation for ϕs (w)
 
d 2 dϕs bs (w) ϕs = 0
fW +Q (59)
dw dw

where
′   2
bs (w) := λs − 2 Λ(1 − |s|)(1 − 2|s|) + 2irs W + f |s| W W ′′ + (1 − |s|)(W ′ )2 + r .
Q
3 W fW2

By means of the variable transformation z = (w − w1 )/(w2 − w1 ) mapping the points

w1 , w2 , ∞ to 0, 1, ∞ equation (59) becomes

d 2 ϕs dϕs es (z)ϕs = 0
2
+ P (z) +Q (60)
dz dz

with
1 1 es (z) = B1 + B2 + A1 + A2
P (z) = + , Q
z z−1 z2 (z − 1)2 z z−1
where
 2  2
s r s r
B1 = − +i , B2 = − −i ,
2 f2 (w2 − w1 ) 2 f2 (w2 − w1 )
λs − (2/3)Λ(1 − |s|)(1 − 2|s|) |s|(|s| − 2) 2r 2
A1 = − + + 2 , A2 = −A1 .
f2 2 f2 (w2 − w1 )2

24
If we make the F-homotopic transformation

ϕs (z) = z α1 (z − 1)α2 ϕ
es (z)

and require that αi2 = −Bi for i = 1, 2, then (60) becomes

d2 ϕ
es dϕes
2
+ Pbs (z) bs (z)ϕs = 0
+Q (61)
dz dz
with
1 + 2α1 1 + 2α2 bs (z) = (A1 + A2 )z + α2 + α1 (1 + 2α2 ) − A1 .
Pbs (z) = + , Q
z z−1 z(z − 1)
Since A2 = −A1 it follows that (61) is an HYE. Concerning (9) we find the following equation
for Hs (x), namely
     ′ 
d 2 dHs ′′ ′2 X q2
X + −λs + |s| XX + (1 − |s|)X − 2sq − 2 Hs = 0. (62)
dx dx X X
By means of the transformation z = (x−x1 )(x2 −x1 ) mapping the points x1 , x2 , ∞ to 0, 1, ∞
our equation (62) becomes
d 2 Hs dHs
+ Ps (z) + Qs (z)Hs = 0 (63)
dz 2 dz
with
1 1 e1
B e2
B e1
A e2
A
Ps (z) = + , Qs (z) = + + +
z z−1 z2 (z − 1)2 z z−1
where
 2  2
e1 s q e2 = − s q
B =− + , B − ,
2 Λ(x2 − x1 ) 2 Λ(x2 − x1 )
e1 λs |s|(|s| − 2) 2q 2 e2 = −A
e1 .
A =− + − 2 , A
Λ 2 Λ (x2 − x1 )2
Finally, with the F-homotopic transformation

e s (z),
Hs (z) = z αb1 (z − 1)αb2 H

ei for i = 1, 2 (63) becomes

bi2 = −B
and the requirement that α
es
d2 H e
+ b s (z) dHs + Q
P b s (z)H
es = 0 (64)
dz 2 dz
with
e e e1
P
α1 1 + 2b
b s (z) = 1 + 2b +
α2
, b s (z) = (A1 + A2 )z + α
Q
b2 + α
b1 (1 + 2b
α2 ) − A
z z−1 z(z − 1)
e2 = −A
Taking into account that A e1 it results that (64) is an HYE.

25
 Acknowledgments

The research of D.B. was supported by the EU grant HPRN-CT-2002-00277. One of

us (D.B.) is indebted to N. Kamran, McGill University, Montreal, Canada and S.Q. Wu,
Hua-Zhong Normal University, Wuhan, China in finding some useful references.

[1] Arenstorf, R. F., Cohen, J. M., and Kegeles, L. S. ”Electromagnetic radiation near black holes
and neutron stars” ’, J. Math. Phys. 19, 833 (1978)
[2] Bardeen, J. M., and Press, W. H., ”Radiation fields in the Schwarzschild background” ’, J.
Math. Phys. 14, 7 (1973)
[3] Bertotti, B., ”Uniform Electromagnetic Field in the Theory of General Relativity” ’, Phys.
Rev. 116, 1331 (1959)
[4] Blandin, J., Pons, R., and Marcilhacy, G., ”General Solution of Teukolsky’s equation” ’, Lett.
Nuovo Cimento 38, 561 (1983)
[5] Breuer, R. A., Ryan, M. P., and Waller, S., ”Some properties of spin-weighted spheroidal
harmonics” ’, Proc. R. Soc. London A 358, 71 (1977)
[6] Carter, B., ”Hamilton-Jacobi and Schrödinger separable solutions of Einstein’s equations” ’,
Comm. Math. Phys. 10, 280 (1968)
[7] Chakrabarti, S. K., ”On mass-dependent spheroidal harmonics of spin one-half” ’, Proc. R.
Soc. London A 391, 27 (1984)
[8] Fackerell, E. D., and Crossman, R. G., ”Spin-weighted angular spheroidal functions” ’, J. Math.
Phys. 18, 1849 (1977)
[9] Heun, K., ”Zur Theorie der Riemmanschen Functionen Zweiter Ordnung mit vier Verzwei-
gungspunkten” ’, Mathematische Annalen 33, 161 (1889)
[10] Kamran, N., and McLenaghan, R. G., ”Exhaustive integration and a single expression for the
general solution of the type D vacuum and electrovac field equations with cosmological constant
for a nonsingular aligned Maxwell field” ’, J. Math. Phys. 25, 1955 (1984)
[11] Kamran, N., and McLenaghan, R. G., ”Separation of variables for higher spin, zero rest-mass
field equations on type D vacuum backgrounds with cosmological constant” ’in Gravitation and
geometry, (Monogr. Textbooks Phys. Sci., 4, Bibliopolis, Naples), 279 (1987)

26
[12] Levi-Civita, T., ”Il sottocaso B2 : soluzioni oblique” ’, Rend. Acc. Lincei 27, 343 (1918)
[13] Leahy, D. A., and Unruh, W. G., ”Angular dependence of neutrino emission from rotating
black hole” ’, Phys. Rev. D 19, 3509 (1979)
[14] Leaver, E. W., ”Solutions to a generalized spheroidal wave equation: Teukolsky’s equations
in general relativity, and the two-center problem in molecular quantum mechanics” ’, J. Math.
Phys. 27, 1238 (1986)
[15] Lee, C. H., ”Coupled gravitational and electromagnetic perturbations around a charged black
hole” ’, J. Math. Phys. 17, 1226 (1976)
[16] Page, D. N., ”Particle emission rates from a black hole: Massless particles from an uncharged,
non rotating hole” ’, Phys. Rev. D 13, 198 (1973)
[17] Plebanski, J. F., and Demianski, M., ”Rotating, Charged, and Uniformly Accelerating Mass
in General Relativity” ’, Ann. Phys. 98, 98 (1976)
[18] Press, W. H., and Teukolsky, S. A., ”Perturbations of a Rotating Black Hole. 2. Dynamical
Stability of the Kerr metric” ’, Astrophys. J. 185, 649 (1973)
[19] Robinson, I., ”A solution of the Maxwell-Einstein Equations” ’, Bull. Acad. Polon. Sci. 7, 351
(1959)
[20] Ronveaux, A., Heun’s Differential Equations, (Oxford University Press, Oxford, New York,
Tokyo, 1995)
[21] Seidel, E., ”A comment on the eigenvalues of spin-weighted spheroidal function” ’, Class.
Quant. Grav. 6, 1057 (1989)
[22] Suzuki, H., Takasugi, E. and Umetsu, H. ”Perturbations of Kerr-de Sitter Black Holes and
Heun’s Equations” ’, Prog. Theor. Phys. 100, 491 (1998)
[23] Wald, R.M., General Relativity, (The University of Chicago Press, Chicago, 1984)
[24] Wu, S. Q.,and Cai, X., ”Massive Complex Scalar Field in a Kerr-Sen Black Hole Background:
Exact Solution of Wave Equation and Hawking radiation” ’, J. Math. Phys. 44, 1084 (2003)

27