Ann. Phys. (Berlin) 528, No. 3–4, 264–271 (2016) / DOI 10.1002/andp.
201500314
Rapid Research Letter
Four five-parametric and five four-parametric independent
confluent Heun potentials for the stationary Klein-Gordon
equation
A.S. Tarloyan1 , T.A. Ishkhanyan1,2 , and A.M. Ishkhanyan1,3,∗
Received 9 October 2015, revised 12 January 2016, accepted 12 January 2016
Published online 22 February 2016
We present in total fifteen potentials for which the station-
ary Klein-Gordon equation is solvable in terms of the con-
fluent Heun functions. Because of the symmetry of the con-
fluent Heun equation with respect to the transposition of
its regular singularities, only nine of the potentials are in-
dependent. Four of these independent potentials are five-
parametric. One of them possesses a four-parametric ordi-
nary hypergeometric sub-potential, another one possesses
a four-parametric confluent hypergeometric sub-potential,
and one potential possesses four-parametric sub-potentials
of both hypergeometric types. The fourth five-parametric
potential has a three-parametric confluent hypergeomet-
ric sub-potential, which is, however, only conditionally in-
tegrable. The remaining five independent Heun potentials
are four-parametric and have solutions only in terms of irre-
ducible confluent Heun functions.
1 Introduction
The Klein-Gordon equation [1, 2] is a relativistic version
of the Schrödinger equation that describes the behavior
of spinless particles. The equation has a large range of
applications in contemporary physics, including particle
physics, astrophysics, cosmology, classical mechanics,
etc. (see [1–4] and references therein). For the station-
ary problems, when the Hamiltonian does not depend
on time, particular solutions can be obtained by apply-
ing the separation of variables that reduces the problem
to the solution of the stationary Klein-Gordon equation.
This approach is widely used to treat particles in various
external fields or curved space-time using functions of
the hypergeometric [5–14] or the Heun [15–20] classes.
In the present paper we consider the reduction of the
one-dimensional stationary Klein-Gordon equation to
264
the single-confluent Heun equation [21, 22]. This equa-
tion possesses two regular singular points located at fi-
nite points of the complex z-plane and an irregular sin-
gularity of rank 1 [21]. Owing to such a structure of the
singularities, the confluent Heun equation directly in-
corporates, by simple choices of the involved exponent
parameters, the Gauss ordinary hypergeometric and the
Kummer confluent hypergeometric equations as well as
the algebraic form of the Mathieu equation and several
other familiar equations. Because of the richer structure
of the singularities, it is clear that the confluent Heun
equation is potent to suggest a set of potentials that can-
not be treated by the hypergeometric equations in rea-
sonable limits. In the meantime, since the parameters
standing for different singularities are clearly separated
so that the influence of the each feature originating from
a particular singularity is well identified, it is expected
that the confluent Heun generalizations will suggest a
clear route to follow the details relevant to a particular
prototype hypergeometric or Mathieu potential.
We show that, to derive energy-independent poten-
tials that are in addition proportional to an energy-
independent continuous parameter and for which the
potential shape is independent of the latter parameter,
there exist only 15 permissible choices for the coordinate
transformation. Each of these transformations leads to
a four- or five-parametric potential solvable in terms of
the confluent Heun functions. However, because of the
symmetry of the confluent Heun equation with respect
to the transposition of its regular singularities, only nine
∗ Corresponding author E-mail: aishkhanyan@gmail.com
1 Institute for Physical Research, NAS of Armenia, 0203, Ashtarak,
Armenia
2 Moscow Institute of Physics and Technology, Dolgoprudny,
Moscow Region, 141700, Russia
3 Institute of Physics and Technology, National Research Tomsk
Polytechnic University, Tomsk, 634050, Russia

C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Ann. Phys. (Berlin) 528, No. 3–4 (2016)

Rapid Research Letter

of these potentials are independent. Four of the indepen- V (x) is written as [1]
dent potentials are five-parametric and five others are
four-parametric. d2 ψ 1  
+ 2 2 (E − V (x))2 − m2 c4 ψ = 0. (1)
The five-parametric Heun potentials all possess hy- dx2  c
pergeometric sub-potentials while the four-parametric
ones do not. One of the five-parametric potentials Applying the independent variable transformation
has a four-parametric sub-potential solvable in terms z = z(x), this equation is rewritten for the new argument
of the Gauss hypergeometric function, another po- z as
tential has a four-parametric sub-potential solvable in
terms of the Kummer confluent hypergeometric func- ρz 1 (E − V (z))2 − m2 c4
ψzz + ψz + 2 2 ψ = 0, (2)
tion and there is a potential that possesses four- ρ  c ρ2
parametric sub-potentials of both hypergeometric types.
where ρ = dz/dx and the lowercase Latin index denotes
Finally, the fourth five-parametric Heun potential pos-
differentiation. Further transformation of the dependent
sesses a three-parametric confluent hypergeometric
variable ψ = ϕ (z) u(z) reduces this equation to the fol-
sub-potential which is, however, only conditionally inte-
lowing one for the new dependent variable u(z):
grable in the sense that in this case the potential cannot
 
be presented as being proportional to a parameter and ϕz ρz
uzz + 2 + uz
having a shape that is independent of that parameter. ϕ ρ
Methodologically, we follow the route for reduction  
of the problem to a target second-order ordinary dif- ϕzz ρz ϕz 1 (E − V (z))2 − m2 c4
+ + + 2 2 u = 0. (3)
ferential equation having rational coefficients applied ϕ ρ ϕ  c ρ2
for the case of the Schrödinger equation in [23–26] and
to the quantum two-state problem in [27–29]. The ba- This equation becomes the single-confluent Heun
sic approach suggested in [23] for construction of ex- equation [21, 22]:
actly integrable energy-independent potentials by trans-  
forming the dependent and independent variables rests γ δ αz − q
uzz + + + ε uz + u = 0, (4)
on the observation that, if a potential is proportional z z−1 z(z − 1)
to an energy-independent parameter and the potential
if
shape is independent of both energy and that parame-
ter, then the logarithmic derivative ρ  (z)/ρ(z) of the func- γ δ ϕz ρz
+ +ε =2 + (5)
tion ρ = z (x), where z = z(x) is the coordinate transfor- z z−1 ϕ ρ
mation, cannot have poles other than the finite singu-
and
larities of the target equation. It then follows that the
function ρ should necessarily be of the Manning form αz − q ϕzz ρz ϕz 1 (E − V (z))2 − m2 c4
ρ(z) = i (z − zi ) Ai [30] with zi being the finite singulari- = + + 2 2 . (6)
z(z − 1) ϕ ρ ϕ  c ρ2
ties of the target equation and the exponents Ai all being
integers or half-integers.
Resolving equation (5) for ϕ:
The paper is organized as follows. In section 2 we de-
rive the nine independent potentials and present the so- eεz/2 γ δ
lution of the problem for these potentials in terms of the ϕ= √ z 2 (z − 1) 2 , (7)
ρ(z)
single-confluent Heun functions. In section 3 we discuss
the exactly and conditionally exactly solvable hyperge- and substituting this into equation (6), we get
ometric sub-potentials and present a new conditionally  
exactly solvable potential written in terms of the Lambert C 0 + C 1 z + C 2 z2 + C 3 z3 + C 4 z4 ρz2 ρzz
+ −
W-function. The paper is concluded by a brief discussion z2 (z − 1)2 4ρ 2 2ρ
of the derived results. 1 (E − V (z))2 − m2 c4
+ = 0, (8)
2 c 2 ρ2

2 Confluent Heun potentials where the constants C 0,1,2,3,4 are defined by the parame-
ters of the confluent Heun equation.
The one-dimensional Klein-Gordon equation for a parti- We suppose that the potential is energy-independent
cle of rest mass m and energy E in a scalar potential field and is proportional to an independent parameter μ:


C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.ann-phys.org 265
 A.S. Tarloyan et al.: Four five-parametric and five four-parametric independent . . .
Rapid Research Letter

V (x) = μf (x), with a potential shape f (x) that is inde- m2

pendent of that parameter: f = f (μ). A key observation 1
of [23] for this case is that, for an E -independent coor-
dinate transformation z = z(x), equation (8) can be satis-
fied only if 1/ 2

z (x) = ρ = zm1 (z − 1)m2 /σ (9)

−1 − 1/ 2 0 1/ 2 1 m1
with integer or half-integer m1,2 .
The lines leading to this conclusion are as follows.
− 1/ 2
Taking the second derivative of equation (8) with respect
to E , we see that
−1
1 r0 + r1 z + r2 z2 + r3 z3 + r4 z4 r(z)
= ≡ , (10)
ρ2 z (z − 1)
2 2
z (z − 1)2
2
Figure 1 Fifteen possible pairs m1 , m2 . The nine independent
cases are marked by filled shapes. The cases possessing ordinary or
where ri = (c2 2 /2) d2 Ci /dE 2 . Rewriting now the polyno-
 confluent hypergeometric sub-potentials are marked by squares or
mial r(z) as r(z) = i (r − si ), taking the limit E , μ → 0 triangles, respectively. The rhombs indicate the cases that possess
and applying the identity hypergeometric sub-potentials of both types. The asterisks mark
   2 the conditionally integrable cases.
ρzz ρz ρz
= + , (11)
ρ ρ z ρ
is a polynomial of at most fourth degree. By direct in-
we see that the roots si are 0 or 1 (in other words, spection it is then shown that the last requirement is
one can say that the logarithmic derivative ρz /ρ can- fulfilled only for certain permissible sub-sets of the pa-
not have poles other than the finite singularities z = rameters v0,1,2,3,4 . For the nine independent cases of m1,2
0, 1 of the confluent Heun equation). With this and the resultant potentials can be conveniently written in
equation (10), we arrive at equation (9) with inte- the form presented in Table 1. Four of the independent
ger or half-integer m1,2 and arbitrary σ . Besides, since potentials are five-parametric, while the remaining five
z2 (z − 1)2 /ρ 2 = z2−2m1 (z − 1)2−2m2 σ 2 is a polynomial of at potentials are four-parametric (V0,1,2 , x0 , σ are arbitrary
most fourth degree, we have the inequalities −1 ≤ m1,2 ≤ complex constants). The types of the hypergeometric
1, 0 ≤ m1 + m2 ≤ 2. This leads to 15 possible sets of m1,2 sub-potentials for the four five-parametric cases pos-
shown in Fig. 1. We note that, because of symmetry of the sessing such sub-potentials are indicated in the last col-
confluent Heun equation with respect to the transposi- umn of the table.
tion z ↔ 1 − z, only nine of these cases are independent. The solution of the problem is readily written taking
The independent cases are marked in the figure by filled the pre-factor ϕ (z) as
shapes.
The next step is matching the cross-term −2E V (z)/ρ 2
ϕ = eα0 z zα1 (z − 1)α2 (14)
and the term V 2 (z)/ρ 2 with the rest in equation (8). Tak-
ing the first derivative with respect to E , we first get
Then, collecting the coefficients at powers of z in the
V (z) numerators of equations (5) and (6), we get eight equa-
z2 (z − 1)2 2 = v0 + v1 z + v2 z2 + v3 z3 + v4 z4 ≡ v(z),
ρ tions which are linear for the five parameters γ , δ, ε, α, q
(12) of the confluent Heun function u (z) and are quadratic for
the three parameters α0,1,2 of the pre-factor.
and further taking the limit E → 0 find that the parame- Resolving these equations, we finally get that the so-
ters v0,1,2,3,4 should be so chosen that lution of the stationary Klein-Gordon equation (1) is ex-
plicitly written in terms of the confluent Heun function
V 2 (z) as
z2 (z − 1)2 = z2m1 −2 (z − 1)2m2 −2 v2 (z)/σ 2 ≡ w(z)
ρ2
(13) ψ = eα0 z zα1 (z − 1)α2 HC (γ , δ, ε; α, q; z) (15)

266 www.ann-phys.org 
C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
 Ann. Phys. (Berlin) 528, No. 3–4 (2016)

Rapid Research Letter

Table 1 Nine independent potentials. V0,1,2 and x0 , σ are arbitrary (complex) constants.

N m1 , m2 Potential V (z) Coordinate transformation z(x) or x(z) Hypergeom. sub-potential

V1 V2 x − x0
1 0, 0 V0 + + z(x) = 1 F1 [31]
z z−1 σ
V1 √ √
2 1/2, −1/2 V0 + x(z) = x0 + σ ( z(z − 1) − sinh−1 ( z − 1)) —
z−1
V1 (x − x0 )2
3 1/2, 0 V0 + z(x) = —
z−1 4σ 2
 
x − x0
4 1/2, 1/2 V0 + V1 z z(x) = cosh2 —
2σ
V1 V2
5 1, −1 V0 + + x(z) = x0 + σ (z − log (z)) Conditionally solvable: 1 F1
z−1 (z − 1)2 Lambert W: z(x) = −W(−e−(x−x0 )/σ )
V1 √ √
6 1, −1/2 V0 + x(z) = x0 + 2σ ( z − 1 − tan−1 ( z − 1)) —
z−1
V2 x−x0
7 1, 0 V0 + V1 z + z(x) = e σ 1 F1 [32]
z−1
2 F1 [33]
 
x − x0
8 1, 1/2 V0 + V1 z z(x) = sec2 —
2σ
1
9 1, 1 V0 + V1 z + V2 z2 z(x) = 2 F1 [35]
e(x−x0 )/σ + 1

with the involved parameters being given by the equa- and the auxiliary parameters r0,1,2,3,4 , v0,1,2,3,4 and
tions w0,1,2,3,4 for each row of Table 1 are readily calculated
through the definitions (10), (12) and (13):
γ = 2α1 + m1 , δ = 2α2 + m2 , ε = 2α0 , (16)
r(z) = r0 + r1 z + r2 z2 + r3 z3 + r4 z4 = z2−2m1 (z − 1)2−2m2 ,
α = α0 (m1 + m2 + 2(α1 + α2 − α0 )) (22)
1   
+ 2 2 E 2 − m2 c4 r3 − 2E v3 + w3 , (17)
 c v0 + v1 z + v2 z2 + v3 z3 + v4 z4 = r(z)V (z), (23)

q = α1 (2 − m1 − m2 ) + (2α1 + m1 )(α0 − α1 − α2 ) w0 + w1 z + w2 z2 + w3 z3 + w4 z4 = r(z)V 2 (z) (24)

1   
+ 2 2 E 2 − m2 c4 r1 − 2E v1 + w1 . (18)
 c
The coordinate transformation x(z) or z(x) calculated
using equation (9) is presented in the third column of
The equations for the exponents α0,1,2 read
Table 1.
1  2   The derived solution applies to any set of the involved
α02 + E − m2 c4 r4 − 2E v4 + w4 = 0, (19) parameters. It should be stressed that the parameters in
2 c 2
general may be chosen complex. For example, putting
x0 → x0 + i π σ , one may change the sign of the expo-
α12 − α1 (1 − m1 ) nents involved in the coordinate transformations in the
1  2   7th and 9th rows of Table 1.
+ E − m2 c4 r(0) − 2E v (0) + w (0) = 0, (20)
2 c 2
3 Hypergeometric sub-potentials
α22 − α2 (1 − m2 )
1    Consider the hypergeometric reductions of the above
+ E 2 − m2 c4 r(1) − 2E v (1) + w (1) = 0 (21) confluent Heun potentials. We first demand for the
2 c 2


C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.ann-phys.org 267
 A.S. Tarloyan et al.: Four five-parametric and five four-parametric independent . . .
Rapid Research Letter

Table 2 Confluent and ordinary hypergeometric potentials. The two ordinary hypergeometric potentials are transformed into each
other by the change x0 → x0 + i π σ

N m1 , m2 Potential V (z) Coordinate transformationz(x) Reference

V1 x − x0
1 0, 0 V0 + z(x) = 1 F1 Coulomb [31]
z σ
x−x0
7 1, 0 V0 + V1 z z(x) = e σ 1 F1 Exponential (Morse [32])

V1 x−x0
7 1, 0 V0 + z(x) = e σ 2 F1Hulthén [33]
z−1 (Eckart [34])
1
9 1, 1 V0 + V1 z z(x) = 2 F1Woods-Saxon [35]
e(x−x0 )/σ + 1 (Eckart [34])

parameters of the hypergeometric sub-potentials to be
independent of each other. Then, the results are as
follows.
The confluent Heun equation is reduced to the Kum-
mer confluent hypergeometric equation if δ = 0 ∪ q =
α or γ = 0 ∪ q = 0. Examining these two possibilities
through equations (16)–(24), one readily reveals that this
is possible only in three cases when m1,2 are integers
(half-integer m1,2 lead to constant potentials) obeying
the inequality 0 ≤ m1 + m2 ≤ 1. Because of the symme-
try of the potentials with respect to the transposition
m1 ↔ m2 ∪ z ↔ 1 − z, the number of the independent
cases is reduced to two. These are the Coulomb poten-
tial [31] and the exponential potential shown in Table 2,
where the numbers in the first column indicate the num-
ber of the confluent Heun potential to which the particu-
lar hypergeometric sub-potential belongs. The first of the
two confluent hypergeometric potentials, the Coulomb
potential, has been applied in the past by many authors
starting from the early days of quantum mechanics [1, 2].
The second potential can be viewed as a truncated one-
term version of the Morse potential [32].
The confluent Heun equation is reduced to the Gauss
ordinary hypergeometric equation if ε = α = 0. It then
follows from Eqs. (16), (17), (19) that in this case r3 =
v3 = w3 = 0 and r4 = v4 = w4 = 0, so that the polynomi-
als r(z), v(z) and w (z) are of the second degree. Accord-
presented in Table 2. The first one is identified as a ver-
sion of the Hulthén potential [33], which presents a one-
term four-parametric specification of the two-term five-
parametric Eckart potential [34]. The second potential is
the Woods-Saxon potential [35], which is again a four-
parametric one-term specification of the Eckart poten-
tial. We note that the two hypergeometric sub-potentials
are transformed into each other by simple change x0 →
x0 + i π σ . Thus, there exists only one independent ordi-
nary hypergeometric potential. This potential has been
explored in the past by many authors on several occa-
sions (see, e.g., [6–9]).
We would like to conclude this section by noting that
if a weaker requirement of conditional solvability (that is
if a parameter of the potential is fixed to a specific value
or if the parameters standing for characteristics of dif-
ferent physical origin are dependent) is examined, there
may exist other hypergeometric sub-potentials. An ex-
ample of this kind of sub-potentials is as follows. Con-
sider the case m1,2 = (1, −1). It is then readily verified
that for
√
c 3c
V1 = − √ , V2 = − , (25)
3σ 2σ
that is for the potential
 
c 1 3/2

ingly, m1,2 obey the inequality 1 ≤ m1 + m2 ≤ 2 (see Eq. V = V0 − √ +

x−x0
, z = −W −e− σ
(22)). Hence, hypergeometric sub-potentials may exist 3σ z − 1 (z − 1)2
only for the six sets m1,2 close to the upper right-hand (26)
corner in Fig. 1. Because of the symmetry of the hyperge-
ometric equation with respect to the transposition z ↔ it holds δ = 0 and α = q so that the confluent Heun
1 − z, the number of the independent cases is reduced to equation (4) is reduced to the scaled Kummer conflu-
four. A closer inspection further reveals that the sets with ent hypergeometric equation. This is a conditionally in-
m1 = 1/2 or m2 = 1/2 do not lead to non-constant poten- tegrable potential since the interaction strengths V1,2 de-
tials. Thus, we arrive at two four-parametric potentials pend on the space scale σ . More strictly, this potential

268 www.ann-phys.org 
C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Ann. Phys. (Berlin) 528, No. 3–4 (2016)

Rapid Research Letter

and

 
1−ε 2 m2 c4 − E 2 E
a = α1 + − + + . (33)
2 3ε 3εV02 εV0

Here any combination of the signs + or − is applica-

ble for α1 and ε. We note that by choosing different signs
we get different independent fundamental solutions.

4 Discussion
Figure 2 Potential (28) for σ = 1, 3, 10 (c =  = 1). The inset
presents the coordinate transformation z(x). There have been many studies employing reductions of
the stationary Klein-Gordon equation to the Heun class
of differential equations (for the reductions to the single-
cannot be presented as V = μf (z) with f = f (μ). This
confluent Heun equation see, for instance [15–19], and
three-parametric potential (V0 , x0 , σ are arbitrary) and
for a list of sub-cases belonging to the hypergeometric
its counterpart for m1,2 = (−1, 1) are marked in Fig. 1 by
class see [14]). However, to the best of our knowledge,
asterisks.
there has not been a general discussion of the solvability
Choosing x0 = −σ and
of the Klein-Gordon equation in terms of the hypergeo-
c metric or more advanced special functions.
V0 = √ , (27) In the present paper we have presented a systematic
2 3σ
treatment of this question using the general approach
we get a single-parametric potential defined for a posi- developed in [23–26] for the stationary Schrödinger
tive σ on the positive half-axis x > 0 that has a singularity equation and in [27–29] for the time-dependent quan-
at the origin and vanishes at infinity (Fig. 2): tum two-state problem (see also [36–38]). This is an
approach for searching for exactly solvable energy-
z(z − 4)   independent potentials via energy-independent coordi-
V = V0 , z = −W −e−1−x/σ . (28)
(z − 1)2 nate transformation that is designed for identification of
solvable potentials that are proportional to an energy-
independent parameter and have a shape that does not
In the vicinity of the origin the behavior of the poten-
depend on that parameter [23]. It should be noted, how-
tial is Coulomb-like:
ever, that the technique is also potent to generate con-
√
3c/4 ditionally integrable potentials - we have presented an
V |x→0 = − + O(1), (29) example of this kind of potentials (see another example
x
in [25], several other examples are presented in [29, 36,
while at infinity the potential vanishes exponentially: 37]).
Discussing the representative example of the single-
2c
V |x→+∞ = − √ e−x/σ + O(e−2x/σ ). (30) confluent Heun equation which directly incorporates
3σ the two hypergeometric equations, we have shown that
there exist in total fifteen permissible choices for the
The solution of the Klein-Gordon equation for this coordinate transformation each leading to a four- or
potential is explicitly written as five- parametric potential. Because of the symmetry of
the single-confluent Heun equation with respect to the
ψ = zα1 (1 − z)1/2 e1
εz/2
F1 (a; 1 + 2α1 ; −εz) (31) transposition z → 1 − z, only nine of these potentials
are independent. Five of the independent potentials are
with four-parametric. A peculiarity of these potentials is that
 they do not posses hypergeometric sub-potentials. Four
σ  2 4 2σ other independent confluent Heun potentials are five-
α1 = ± m c − E 2, ε=± m2 c4 − (E − V0 )2
c c parametric. These potentials present distinct generaliza-
(32) tions of certain hypergeometric potentials.


C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.ann-phys.org 269
 A.S. Tarloyan et al.: Four five-parametric and five four-parametric independent . . .
Rapid Research Letter

Among the independent five-parametric confluent
Heun potentials, one potential extends the Coulomb
confluent hypergeometric potential [31], another one [4]
extends the Hulthén ordinary hypergeometric po-
[5]
tential [33], and there is a potential that possesses
four-parametric sub-potentials of both hypergeomet- [6]
ric types (exponential potential, which is a truncated
version of the Morse confluent hypergeometric poten- [7]
tial [32], and the Woods-Saxon ordinary hypergeometric [8]
potential [35]). The fourth five-parametric Heun poten-
tial possesses a confluent hypergeometric sub-potential [9]
which is a conditionally integrable in the sense that it
[10]
cannot be presented as being proportional to an energy-
independent parameter and having a shape that is [11]
independent of that parameter. This is a three-
parametric potential explicitly given through a co- [12]
ordinate transformation written in terms of the Lambert
W-function [39, 40]. The Schrödinger counterpart of this [13]
potential has been presented in [41].
[14]
We would like to conclude by noting that the ap-
plied approach and the lines of the presented analysis [15]
are rather general and can be extended to other target
equations or to treat other structurally similar problems, [16]
for instance, to derive the Heun solutions of the Klein-
Gordon equation on manifolds with variable geometry [17]
[42] or to identify the potentials for which the Klein-
[18]
Gordon and other relativistic quantum mechanical wave
equations [1, 3] are solvable in terms of the general or [19]
multiply-confluent Heun functions.
[20]
[21]
Acknowledgments. This research has been conducted within the
[22]
scope of the International Associated Laboratory (CNRS-France &
SCS-Armenia) IRMAS. The work has been supported by the Ar- [23]
menian State Committee of Science (SCS Grants no. 13YR-1C0055,
no. 15T-1C323 and no. 13RB-052) and the project “Leading Rus- [24]
sian Research Universities” (Grant No. FTI 120 2014 of the Tomsk [25]
Polytechnic University). [26]
[27]
PACS numbers: 03.65.Ge Solutions of wave equations: bound
states, 02.30.Ik Integrable systems, 02.30.Gp Special functions [28]
Key words. Klein-Gordon equation, integrable potentials, conflu- [29]
ent Heun equation.
[30]
[31]
References
[32]
[1] W. Greiner, Relativistic Quantum Mechanics. Wave [33]
equations (Springer, Berlin, 2000).
[2] H. Kragh, Am. J. Phys. 52, 1024–1033 (1984). [34]
[3] T. Ohlsson, Relativistic Quantum Physics - From Ad- [35]
vanced Quantum Mechanics to Introductory Quan-
tum Field Theory (Cambridge University Press, Cam-
bridge 2011).
P. Gravel and C. Gauthier, Am. J. Phys. 79, 447–453
(2011).
D. J. Rowan and G. Stephenson, J. Phys. A 10, 15–23
(1977).
F. Dominguez-Adame, Phys. Lett. A 136, 175–177
(1989).
M. C. Zhang, Chin. Phys. B 17, 3214–3216 (2007).
W. C. Qiang, R. S. Zhou, and Y. Gao, Phys. Lett. A 371,
201–204 (2007).
C. Rojas and V. M. Villalba, Phys. Rev. A 71, 052101
(2005).
N. A. Rao and B. A. Kagali, Phys. Lett. A 296, 192–196
(2002).
M. C. Zhang and Z. B. Wang, Chin. Phys. Lett. 22,
2994–2996 (2005).
C. Y. Chen, D. S. Sun, and F. L. Lu, Phys. Scr. 74, 405–
409 (2006).
M. R. Setare and O. Hatami, Commun. Theor. Phys. 51,
1000–1002 (2009).
L. K. Sharma, P. V. Luhanga, and S. Chimidza, Chiang
Mai J. Sci. 38, 514–526 (2011).
D. V. Gal’tsov and A. A. Ershov, Sov. Phys. 32, 764–769
(1989).
S. Q. Wu and X. Cai, J. Math. Phys. 44, 1084–1088
(2003).
V. B. Bezerra, H. S. Vieira, and A. A. Costa, Class. Quan-
tum Grav. 31, 045003 (2014).
H. S. Vieira, V. B. Bezerra, and C. R. Muniz, Ann. Phys.
350, 14–28 (2014).
A. Schulze-Halberg and J. Wang, Few-Body Syst. 55,
1223–1232 (2014).
K. Bakke and C. Furtado, Ann. Phys. 355, 48–54 (2015).
A. Ronveaux, Heun’s Differential Equations (Oxford
University Press, London, 1995).
S.Yu. Slavyanov and W. Lay, Special functions (Oxford
University Press, Oxford, 2000).
A. Ishkhanyan and V. Krainov, arXiv:1508.06989
(2015).
A. M. Ishkhanyan, Eur. Phys. Lett. 112, 10006 (2015).
A. M. Ishkhanyan, Phys. Lett. A 380, 640–644 (2016).
A. M. Ishkhanyan, Theor. Math. Phys. 186 (2016).
A. M. Ishkhanyan and A. E. Grigoryan, J. Phys. A 47,
465205 (2014).
A. M. Ishkhanyan, T. A. Shahverdyan, and T. A.
Ishkhanyan, Eur. Phys. J. D69, 10 (2015).
T. A. Shahverdyan, T. A. Ishkhanyan, A. E. Grigoryan,
and A. M. Ishkhanyan, J. Contemp. Physics (Armenian
Ac. Sci.) 50, 211–226 (2015).
M. F. Manning, Phys. Rev. 48, 161 (1935).
E. Schrödinger, Annalen der Physik 76, 361–376
(1926).
P. M. Morse, Phys. Rev. 34, 57–64 (1929).
L. Hulthén, Ark. Mat. Astron. Fys. 28A, 5 (1942); L.
Hulthén, Ark. Mat. Astron. Fys. 29B, 1–12 (1942).
C. Eckart, Phys. Rev. 35, 1303–1309 (1930).
R. D. Woods and D. S. Saxon, Phys. Rev. 95, 577–578
(1954).

270 www.ann-phys.org 
C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Ann. Phys. (Berlin) 528, No. 3–4 (2016)

Rapid Research Letter

[36] A. M. Ishkhanyan, J. Phys. A 30, 1203–1208 (1997).
[37] A. M. Ishkhanyan, J. Phys. A 33, 5539–5546 (2000).
[38] A. M. Ishkhanyan and K.-A. Suominen, J. Phys. A 34,
6301 (2001).
[39] J. H. Lambert, Acta Helvitica 3, 128 (1758).
[40] L. Euler, Acta Acad. Scient. Petropol. 2, 29–51
(1783).
[41] B. W. Williams, Phys. Lett. A 334, 117 (2005).
[42] P. P. Fiziev and D. V. Shirkov, Theor. Math. Phys. 167,
680–691 (2011).


C 2016 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.ann-phys.org 271