Turkish Journal of Physics

Volume 42 Number 5 Article 4

1-1-2018

General solution of the Schrödinger equation with potential field

quantization
HASAN HÜSEYİN ERBİL

Follow this and additional works at: https://journals.tubitak.gov.tr/physics

Part of the Physics Commons

Recommended Citation
ERBİL, HASAN HÜSEYİN (2018) "General solution of the Schrödinger equation with potential field
quantization," Turkish Journal of Physics: Vol. 42: No. 5, Article 4. https://doi.org/10.3906/fiz-1801-10
Available at: https://journals.tubitak.gov.tr/physics/vol42/iss5/4

This Article is brought to you for free and open access by TÜBİTAK Academic Journals. It has been accepted for
inclusion in Turkish Journal of Physics by an authorized editor of TÜBİTAK Academic Journals. For more
information, please contact [email protected].
 Turkish Journal of Physics Turk J Phys
(2018) 42: 527 – 572
http://journals.tubitak.gov.tr/physics/
© TÜBİTAK
Research Article doi:10.3906/fiz-1801-10

General solution of the Schrödinger equation with potential field quantization

Hasan Hüseyin ERBİL∗,

Faculty of Science, Ege University, Izmir, Turkey

Received: 10.01.2018 • Accepted/Published Online: 12.07.2018 • Final Version: 12.10.2018

Abstract: A simple procedure has been found for the general solution of the time-independent Schrödinger equation
(SE) with the help of quantization of potential area in one dimension without making any approximation. Energy
values are not dependent on wave functions. So, to find the energy values, it is enough to find the classic turning
points of the potential function. Two different solutions were obtained, namely, symmetric and antisymmetric in bound
states. These normalized wave functions are always periodic. It is enough to take the integral of the square root of the
potential energy function to find the normalized wave functions. If these calculations cannot be made analytically, they
should then be performed by numerical methods. The SE has been solved for a particle in many one-dimension and the
spherical symmetric central potential wells as examples. Their energies and normalized wave functions were found as
examples. These solutions were also applied to the theories of scattering and alpha decay. The results obtained with the
experimental values were compared with the calculated values. One was seen to be very fit.

Key words: Potential quantization, Schrödinger equation, infinitely high potential well, potential wells of any form,
α -decay theory, scattering, energy values of states

1. Introduction
Mechanical total energy (E) is the sum of the kinetic energy (T ) and potential energy (U ) . That is, E = T + U .
If E = U then T = 0. If E > U then T > 0; If E < U then T < 0 (it is not possible, classically). In quantum
mechanics, the total energy is equal to the eigenvalue of the total energy operator Ĥ (Hamiltonian). One-
ℏ d 2 2
dimension Hamiltonian is: Ĥ= − 2m dx2 +U (x) . This operator is hermitic and its eigenvalues are real numbers.
The equation of eigenvalues of this Hamiltonian is given as follows (time-independent Schrödinger Equation,
SE):
[ ]
ℏ2 d2
Ĥψ (x) = − +U (x) ψ (x) = Eψ (x) (1)
2m dx2

The eigenvalue E can have two values: −E and +E, (E > 0). So, from Eq. (1), the following differential
equation is obtained:
d2 ψ (x) [ 2 ]
2
+ k −m1 2 U (x) ψ (x) = 0, (2)
dx
where,
/
k 2 = −m1 2 E for − E and k 2 = m1 2 E for + E; m1 2 = 2m ℏ2 (3)
∗ Correspondence: [email protected]

527
This work is licensed under a Creative Commons Attribution 4.0 International License.
 ERBİL/Turk J Phys

Although we succeed in solving the time-independent SE for some quantum mechanical problems in one
dimension, an exact solution is not possible in complicated situations. We must then resort to approximation
methods. For the calculation of the stationary states and energy eigenvalues, these include the variation method,
the method of Nikiforov–Uvarov [1–4], the supersymmetric quantum mechanics, the supersymmetric Wentzel–
Kramers–Brillouin (WKB) and the WKB approximations [5, 6], and perturbation theory. Perturbation theory
is applicable if the Hamiltonian differs from an exactly solvable part by a small amount. The variation method
[18,37] is appropriate for the calculation of the ground state energy if one has a qualitative idea of the form
of the wave function. The WKB method [13,15] is applicable in the nearly classical limit. The exact general
solution of the differential equation given in Eq. (2) has not been achieved yet. This is a very challenging
problem for theoretical physicists.
In this study, we followed a simple procedure for the exact general solution of the time-independent SE
in one dimension without making any approximation. We applied this simple procedure to various quantum
mechanical problems: one-dimension potentials, spherically symmetric potentials, tunneling effect, scattering
theory and presented some examples.

2. Solution of the time-independent SE in one dimension

Let us rewrite the time-independent SE in one dimension given in Eq. (2):

d2 ψ (x) [ 2 ]
+ k −m1 2 U (x) ψ (x) = 0. (4)
dx2

Here, E and U (x) are respectively the total and effective potential energies of a particle of mass m .
Two kinds of solutions to the SE correspond precisely to bound and scattering (unbound) states. When
E < U (−∞) and U (+∞) the solutions correspond to bound states; when E > U (−∞) or U (+∞) the so-
lutions correspond to scattering states [6]. In real life, many potential functions go to zero at infinity, in which
case the criterion is simplified even further: when E < 0 , bound states occur; when E > 0 , scattering states
occur. In this section, we will explore potentials that give rise to both kinds of states. We will solve Eq. (4) in
two steps.

2.1. The first step

Inspired by the theorem given in [8,9], we consider the following integral function:
∫
S (x) = U (x) dx. (5)

If f (x) = S(x) in this theorem, we get:

∫ +∞
F (ε) = S (x) y (x−x0 , ε) dx. (6)
−∞

From this function for ε = 0 , the following is obtained:

∫ +∞
F (ε)= F (0) = S (x) δ (x−x0 ) dx = S(x0 )= S. (7)
−∞

528
 ERBİL/Turk J Phys

Now, let us take the potential and wave functions respectively as follows:
∫ x2
S= U (x) dx; U (x) = S (x0 ) δ (x−x0 ) = Sδ (x−x0 ) and ψ (x) = F (x). (8)
x1

δ(x−x0 ) is Dirac function. x1 and x2 are the roots of the equation E = U (x) that is the apsis of the classical
turning point of the potential function. If we take x0 = (x1 +x2 )/2 and d =x2 −x1 , (x2 >x1 ) , then we find
x1 = x0 −d/2, x2 = x0 +d/2 (see Figure 1). If we take these functions, the SE given in Eq. (4) becomes as
follows:

Figure 1. Regions relevant to a particle of energy E moving in a one dimensional potential field U (x) . (a), in domains
I and III, E < U (x), (unbound state); in domain II, E > U (x) , (bound state). (b), in domains I and III, E > U (x),
in domain II, E < U (x), (unbound state). The roots of the equation E = U (x) are turning points of the corresponding
classical motion. (c), regions for U (x) < 0 and E < 0 .

d2 F (x) 2
+k F (x) =m1 2 Sδ (x−x0 ) F (x). (9)
dx2
To evaluate the behavior of F (x) at x =x0 , let us integrate Eq. (9) over the interval [x1 ,x2 ] = [x0 −d/2,x0 +d/2]
and also consider the limit d → 0, then we obtain the following:
′ ′
F (x0 +d/2) −F (x0 −d/2) =m21 SF (x0 ). (10)

529
 ERBİL/Turk J Phys

Eq. (10) shows that the derivation of F(x) is not continuous at the x =x0 point [6,9], whereas the wave functions
F (x) should be continuous at the same point.
To solve the differential equation given in Eq. (9), we can perform the transformation of Fourier of the
equation, and then we obtain the following function:
[ √ ]
−k|(x−x0 )| a2 π 2 2m
F (x) = Ae , A= , a = − √ SF (x0 ) . (11)
k 2 ℏ2 2π

From Eq. (11), we get the following function:

F (x) = Aek(x−x0 ) for x <x0 and F (x) = Ae−k(x−x0 ) for x >x0 (12)

Substituting the functions given in Eq. (12) into Eq. (10) and taking the limit d → 0 , the following values are
obtained:
m m
k = 2 S or E = ± 2 S 2 . (13)
ℏ 2ℏ
The same values given in Eq. (13) can also be obtained from Eq. (11). To find the constant A, the function
F (x) can be normalized to 1:
∫ x0 ∫ +∞
AA∗ e2k(x−x0 ) dx+ AA∗ e−2k(x−x0 ) dx = 1.
−∞ x0

From this equation, we obtain the coefficient of normalization as:

√ √ /
|A| = k = mS ℏ. (14)

From Eq. (12), by the linear combinations of these functions, we also have the following functions:

F (x) = Aek(x−x0 ) +Be−k(x−x0 ) , (15a)

1 [ k(x−x0 ) −k(x−x0 ) ]
F (x) = A e +e = Acosh [k(x−x0 )] , (15b)
2

1 [ k(x−x0 ) −k(x−x0 ) ]
F (x) = A e −e = Asinh [k(x−x0 )] . (15c)
2

2.2. The second step

Let us assume the wave function ψ (x) to be ψ (x) = F (x) eiG(x) , (we assume that G(x) is a real function). If
we substitute this function into Eq. (4), we get:

′′ ′
[ ′ ′ ′′
]
F (x) −F (x) G 2 (x) −k 2 F (x) −m1 2 U (x) F (x) +i 2F (x) G (x) +F (x) G (x) = 0. (16)

From the real and imaginary parts of Eq. (16), we can have the following two equations:

′′ ′
F (x) −F (x) G 2 (x) −k 2 F (x) −m1 2 U (x) F (x) = 0, (17)

530
 ERBİL/Turk J Phys

′ ′ ′′
2F (x) G (x) +F (x) G (x) = 0. (18)

From Eq. (17), the following equations are obtained:

[ ′
]
For F (x) = Aek(x−x0 ) ; F (x) m1 2 U (x) +G 2 (x) = 0, (19a)

[ ′
]
For F (x) = Ae−k(x−x0 ) ; F (x) m1 2 U (x) +G 2 (x) = 0, (19b)

[ ′
]
For F (x) = Aek(x−x0 ) +Be−k(x−x0 ) ; F (x) m1 2 U (x) +G 2 (x) = 0, (19c)

[ ′
]
For F (x) = A cosh [k (x−x0 )] ; F (x) m1 2 U (x) +G 2 (x) = 0, (19d)

[ ′
]
For F (x) = A sinh [k (x−x0 )] ; F (x) m1 2 U (x) +G 2 (x) = 0. (19e)

From Eqs. (19a)–(19e), the following two equations are obtained:

′
m1 2 U (x) +G 2 (x) = 0, (20a)

F (x) = 0. (20b)

From Eq. (20a), the function G (x) is obtained as follows:

∫ √ ∫ √
G (x) = ±m1 −U (x)dx = ±im1 U (x)dx = ±iQ (x) (21a)

∫ √
Q (x) = m1 U (x)dx. (21b)

Thus, the wave function ψ (x) is written as follows:

ψ (x) = F (x) e±iG(x) or ψ (x) = F (x−x0 ) e±iG(x−x0 ) . (22)

The functions given in Eq. (22) can also be written as follows:

[ ]
ψ (x) = F (x) AeiG(x) +Be−iG(x) (23a)

[ ]
ψ (x) = F (x−x0 ) AeiG(x−x0 ) +Be−iG(x−x0 ) (23b)

In the functions given in Eqs. (23a) and (23b):

√ ∫√
(a) For E > U (x) ; k = m1 −E, G (x) = m1 −U (x)dx,
√ ∫√
(b) For E < U (x) ; k = m1 E, G (x) = m1 U (x)dx .

531
 ERBİL/Turk J Phys

As shown in Figure 1a, the total energy is equal to the potential energy at the points x1 = x0 −d/2 and
x2 = x0 +d/2 . So, the kinetic energy is zero at these points. Now, let us examine the behavior of the function
in the interval [x0 −d/2,x0 +d/2]
Firstly, let us consider Eq. (20b):
F (x) = 0.

Let us assume that the roots of this equation are x1 = x0 −d/2 and x2 = x0 +d/2 . So, we can write the
following:
F (x1 ) = F (x0 −d/2) = 0; F (x2 ) = F (x0 +d/2) = 0. (24)

From Eq. (24), the following expressions can be obtained:

√ √ √
(a) For F (x) = Aek(x−x0 ) +Be−k(x−x0 ) ; k = m1 −E= im1 E= iK, [K =m1 E> 0]

a11 A+a12 B = 0;a21 A+a22 B = 0;

a11 = e−idK/2 ;a12 = eidK/2 ;a21 = eidK/2 ;a22 = e−idK/2 .

Since A and B have to nonzero values, the determinant of coefficients in this system of equations must
be zero. Thus, the following equation can be written as follows:

a11 a12
det = −2i sin (dK) = 0. (25)
a21 a22

Here, let us take dK = q. From Eq. (25), we have the following values:

sin (dK) = sin (q) = 0→ q = nπ, [n = 1, 2, 3, 4, . . . integer].

So, we obtain the quantization condition of energy as follows:

√
2m
Kd = |E|d = q, [q = nπ, (n = 1, 2, 3, . . . integer numbers) ] (26)
ℏ2

Here, the coefficient B is obtained as B = −A . Thus the function F (x) is written as follows:
[ ]
F (x) = A eiK(x−x0 ) −e−iK(x−x0 ) = 2iA sin [K (x−x0 )] = B sin[K (x−x0 ) ]. The coefficient B in this
√
function is found by normalizing the function in the interval [x1 ,x2 ] . One is found as |B| = 2/d .
Thus, the functions F (x) and ψ (x) are written as follows:
√ √
F (x) = 2/dsin[K(x−x0 )]; ψ (x) = 2/dsin[K(x−x0 )]eiG(x−x0 ) (27)

(b) For F (x) = A cosh [k (x−x0 )]; k = iK

F (x1 ) = F (x0 −d/2) = 0; F (x2 ) = F (x0 +d/2) = 0;

F (x1 ) = F (x2 ) = A cosh(dk/2) = A cos(dK/2) = 0

dK/2 = q/2 =(2n − 1)π/2 → q = (2n − 1) π, [n = 1, 2, 3, . . . integer].

532
 ERBİL/Turk J Phys

So, we obtain the quantization condition of energy as follows (symmetric case):

√
2m
Kd = |E|d = q, q = (2n − 1)π, (n = 1, 2, 3, . . . integer numbers) ]. (28)
ℏ2

The coefficient A in this function is found by normalizing the function in the interval [x1 ,x2 ] . That is,
√
|A| = 2/d . Thus, the functions F (x) and ψ (x) are written as follows:
√ √
F (x) = 2/d cos[K(x−x0 )]; ψ (x) = 2/d cos[K(x−x0 )]eiG(x−x0 ) . (29)

(c) For F (x) = A sinh [k (x−x0 )] ; k = iK ;

F (x1 ) = F (x0 −d/2) = 0; F (x2 ) = F (x0 +d/2) = 0;

F (x1 ) = −A sinh(dk/2) = −iAsin(dK/2) = 0

F (x2 ) = A sinh(dk/2) = iAsin(dK/2) = 0

dK/2 = q/2 =nπ → q = 2nπ, [n = 1, 2, 3, . . . integer numbers].

So, we obtain the quantization condition of energy as follows (antisymmetric case):

√
2m
Kd = |E|d = q, [q = 2nπ, (n = 1, 2, 3, . . . integer numbers) ]. (30)
ℏ2

The coefficient A in this function is found by normalizing the function in the range [x1 ,x2 ] . That is,
√
|A| = 2/d . Thus, the functions F (x) and ψ (x) are written as follows:
√ √
F (x) = 2/d sin [K (x−x0 )] ; ψ (x) = 2/d sin [K (x−x0 )] eiG(x−x0 ) (31)

It is possible to combine Eqs. (28) and (30) in one equation as follows (general case):
√
2m
Kd = |E|d = q, [q = nπ, (n = 1, 2, 3, . . . integer numbers)] . (26)
ℏ2

2
p
Now let us write the kinetic energy of the particle as follows: T = 2m = E − U (x) . By integrating this
equation from x1 to x2 and using Eq. (8), the following equation is obtained:
∫ x2 ∫ x p2 ∫x ∫x ∫x
x1
T dx = x12 2m dx = x12 [E − U (x)] dx = x12 Edx− x12 U (x)dx = Sk , and the following equation is writ-
ten:
Sk = E (x2 −x1 ) −S = Ed − S. (32)

From Figure 1, we can observe that:

(a) For the case where E > U (x), the kinetic energy is positive and [E (x2 −x1 ) −S] > 0 (bound state).
(b) For the case where E < U (x), the kinetic energy is imaginary and [E (x2 −x1 ) −S] < 0 (unbound
state).
(c) For the case where E = U (x) , the kinetic energy is zero and [E (x2 −x1 ) −S] = 0 (ground state).

533
 ERBİL/Turk J Phys

In addition, for the bound states, in the interval [x1 ,x2 ] , the kinetic energy is positive; outside this
interval, the kinetic energy is imaginary. The minimum point of the potential corresponds to ground state. In
ground state, the kinetic energy is zero, namely, T = 0 or Sk = 0 . Thus, at the minimum point of the potential,
we can write that:
E0 (x2 −x1 ) −S = 0 or S =E0 (x2 −x1 ) = E0 d.

By substituting this value of S into Eq. (13), we get the ground state energy expression as follows:

m 2 m 2ℏ2
E0 = − 2S = − 2 E0 d →E0 = − md2 .
2 2
(33)
2ℏ 2ℏ

Here, E0 represents the ground state energy. The negative sign indicates that the state is bound and it can be
omitted for the positive energies in the calculations.

3. Boundary conditions
Let us divide the potential field into three domains as shown in Figure 1 and represent the functions ψ1 (x) ,ψ2 (x),
and ψ3 (x) in each domain. The wave functions and their derivatives should be continuous. Because of these
conditions, the above functions must satisfy the following conditions:
′ ′
ψ1 (x1 ) = ψ2 (x1 ) ; ψ1 (x1 ) = ψ2 (x1 ); ψ2 (x2 ) = ψ3 (x2 ) ,
′ ′
ψ2 (x2 ) = ψ3 (x2 ); lim ψ1 (x) → 0 and lim ψ3 (x) → 0. (34)
x→+∞ x→+∞

The normalization of the bound state function requires that the functions vanish at infinity. With these boundary
and normalization conditions of the wave functions, we can find the integral constants, A, B and the energy E.
As it is also seen above, in the bound states, we do not need the solutions of the SE. It is sufficient to know
only the classical turning points, x1 and x2 , of the potential function.
Quantization of the energy values can also be found by means of boundary conditions. Let’s find them
now. In bound states, let us apply the conditions given in Eq. (34) to the following functions:

Aek(x−x0 ) +Be−k(x−x0 ) ; A cosh [k (x−x0 )]; B sinh [k (x−x0 )] .

(a) For the function Aek(x−x0 ) +Be−k(x−x0 )

According to Figure 1a, in domain I, E < U (x) ; in domain II, E > U (x); in domain III, E < U (x) .
According to Eqs. (23a) and (23b), the corresponding wave functions are written as follows:

ψ1 (x) = A1 eik(x−x0 ) e−Q(x−x0 ) (35a)

ψ2 (x) = [A2 ek(x−x0 ) +B2 e−k(x−x0 ) ]e

iQ(x−x0 )
(35b)

ψ3 (x) = A3 e−ik(x−x0 ) e−Q(x−x0 ) (35c)

∫√ √ √
Here, Q (x−x0 ) = m1 −U (x−x0 )dx, and k = m1 −E= im1 E

534
 ERBİL/Turk J Phys

′ ′ ′
Boundary conditions: ψ1 (x1 ) = ψ2 (x1 ) ,ψ2 (x2 ) = ψ3 (x2 ) ,ψ1 (x1 ) = ψ2 (x1 ) ,ψ2 (x2 ) =
′
limx→+∞ ψ3 (x2 ) , ψ1 (x) → 0, and ψ3 (x) → 0
From these conditions, four linear equations are obtained as follows:
  
a11 a12 a13 a14 A1
 a21 a22 a23 a2   A2 
   (36a)
 a31 a32 a33 a34   B2  = 0.
a41 a42 a43 a44 A3

For this system of equations to have a solution different from zero, the determinant of coefficients should
vanish, namely:
a11 a12 a13 a14
a21 a22 a23 a2
det = 0. (36b)
a31 a32 a33 a34
a41 a42 a43 a44

If it is taken as follows

x1 = x0 −d/2,x2 = x0 +d/2, Q′ (−d/2) = Q′ (d/2) = k, k = iK, dK = q

from Eqs. (36a) and (36b), following equation is obtained:

[ ]
e[−iq−(1−i)Q(−d/2)+(1+i)Q(d/2) −1+e2iq = 0. (37)

For the equality in Eq. (37) to be realized, it has to be q = nπ, (n = 1, 2, 3, . . .) Thus, we obtain the
quantization condition of energy as follows:
√
2m
Kd = |E|d = q, [q = nπ, (n = 1, 2, 3, . . . integers)] (38)
ℏ2

(b) For the function cosh [k (x−x0 )]

According to Figure 1a, in domain I, E < U (x) ; in domain II, E > U (x); in domain III, E < U (x) .
According to the equations given in Eqs. (23a) and (23b), the corresponding wave functions are written
as follows:

ψ1 (x) = A1 cosh [ik (x−x0 )] eiiG(x−x0 ) , ψ2 (x) = A2 cosh [k (x−x0 )] eiG(x−x0 ) ;

ψ3 (x) = A3 cosh [−ik (x−x0 )] eiiG(x−x0 ) , x1 = x0 −d/2 and x2 = x0 +d/2

According to the conditions given in Eq. (34), it can be written that:

ψ1 (x1 ) = ψ2 (x1 ) or ψ1 (x1 ) −ψ2 (x1 ) = 0, (39a)

ψ2 (x2 ) = ψ3 (x2 ) or ψ2 (x2 ) −ψ3 (x2 ) = 0. (39b)

From Eqs. (39a) and (39b), two linear equations are obtained as follows:

a11 A1 +a12 A3 = 0;a21 A1 +a22 A3 = 0 (39c)

a11 = e−G(−d/2) cos(dk/2), a12 = 0; a21 = 0, a22 = −eG(d/2) cos(dk/2).

535
 ERBİL/Turk J Phys

For this system of equations to have a solution different from zero, the determinant of coefficients should
vanish, namely:
a11 a12 1
det = 0 → − e−G(−d/2)+G(d/2) [1+ cos (dk)]= 0. (40)
a21 a22 2
√ √ ( √ )
In Eq. (40), − 12 e−G(−d/2)+G(d/2) ̸= 0, k =m1 −E= im1 E= iK, E > 0, K =m1 E . So,
1+ cos (dk) = 1+ cos (idK) = 1+ cos (dK) = 0. From the last equation, we have that:
dK = q = nπ, (n = 1, 3, 5, 7, . . . odd integer). This is also written as follows:

dK = q = (2n − 1) π, (n = 1, 2, 3, . . . integer) (41)

The solution of the system of equations in (39c) gives: A1 = 0 and A3 = 0 . The coefficient A2 = Ais found
by the normalization of the function, ψ2 (x) = A2 cosh [k (x−x0 )] eiG(x−x0 ) namely:
∫ x0 +d/2
A cosh [k (x−x0 )] eiG(x−x0 ) A∗ cosh [k (x−x0 )] e−iG(x−x0 ) dx =
x0 −d/2

AA∗ [dK+ sin (dK) ] AA∗ d d|A|

2 √ √
= = = 1→ |A| = 2/d = 2K/q.
2K 2 2
So, the normalized wave functions in the bound state are as follows:

ψ (x) = Acos [K (x−x0 )] eiG(x−x0 ) or ψ (x) = Acos [Kx] eiG(x) . (42)

(c) For the function sinh [k (x−x0 )]

According to Figure 1a, in domain I, E < U (x) ; in domain II, E > U (x); in domain III, E < U (x)
According to Eqs. (23a) and (23b), the corresponding wave functions are written as follows:

ψ1 (x) = A1 sinh [ik (x−x0 )] eiiG(x−x0 ) ; ψ2 (x) = A2 sinh [k (x−x0 )] eiG(x−x0 )

ψ3 (x) = A3 sinh [−ik (x−x0 )] eiiG(x−x0 ) ; x1 = x0 −d/2 and x2 = x0 +d/2.

According to the conditions given in Eq. (34) we have the following:

ψ1 (x1 ) = ψ2 (x1 ) or ψ1 (x1 ) −ψ2 (x1 ) = 0, (43a)

ψ2 (x2 ) = ψ3 (x2 ) or ψ2 (x2 ) −ψ3 (x2 ) = 0. (43b)

From Eqs. (43a) and (43b), the following two linear equations are obtained:

a11 A1 +a12 A3 = 0; a21 A1 +a22 A3 = 0 (44)

a11 = −ie−G(−d/2) sin(dk/2); a12 = 0; a21 = 0; a22 = ieG(d/2) sin(dk/2).

For this system of equations to have a solution different from zero, the determinant of coefficients should
vanish, namely:
a11 a12
det = 0 → e−G(−d/2)+G(d/2) sin2 (dk/2)= 0. (45)
a21 a22

536
 ERBİL/Turk J Phys

√ √ ( √ )
In Eq. (45), e−G(−d/2)+G(d/2) ̸= 0, k =m1 −E= im1 E= iK, E > 0, K =m1 E . So,

sin2 (dk/2) = sin2 (idK/2) = sin2 (dK/2)= 0. From the last equation, we have the following value:

dK = q = 2nπ, (n = 1, 2, 3, . . . integer) (46)

The solution of the system of equations in Eq. (44) gives A1 = 0 and A3 = 0 . The coefficient A2 = A is
found by the normalization of the function ψ2 (x) = A2 sinh [k (x−x0 )] eiG(x−x0 ) namely:
∫ x0 +d/2
A sinh [k (x−x0 )] eiG(x−x0 ) A∗ sinh [k (x−x0 )] e−iG(x−x0 ) dx =
x0 −d/2

AA∗ [−dK+ sin (dK) ] AA∗ d √ √

=− = 1→ |A| = 2/d = 2K/q.
2K 2

So, the normalized wave functions in the bound state are as follows:

ψ (x) = A sin [K (x−x0 )] eiG(x−x0 ) or ψ (x) = A sin [Kx] eiG(x) . (47)

Eqs. (41) and (46) can be combined as follows:

dK = q = nπ, (n = 1, 2, 3, 4, . . . integer). (48)

So, in bound states, the normalized wave functions are:

√ √
ψ (x) = 2/d cos [K(x−x0 )] eiG(x−x0 ) and ψ (x) = 2/d sin [K(x−x0 )] eiG(x−x0 ) , (49a)

√ √
Or, ψ (x) = 2/d cos [Kx)] eiG(x) and ψ (x) = 2/d sin [Kx)] eiG(x) , (49b)

The energy:
ℏ2 q 2 q2 ℏ2
Eq = = M h , (M h = ). (49c)
2m d2 d2 2m
Now, in bound states, let us see the relations between the potential areas (see Figure 2):
∫ ∫
(a) According to the partial integration, udv= uv− vdu can be written as follows:
∫x ∫x ∫x ′
S =Sp = x12 U (x) dx = [xU (x)]x21 − x12 xU ′ (x) dx =x2 U (x2 ) −x1 U (x1 ) −St ; St = x12 xU (x) dx;
x

U (x1 ) = U (x2 ) = Eq ; d =x2 −x1 .

Thus, the following can be written:

Sp = Ed−St or Ed =Sp +St . (50a)

(b) Total energy=kinetic energy+potential energy; E = T + U (x). From here, with the following integration:

∫ x2 ∫ x2 ∫ x2
Edx = T dx+ U (x)dx → Ed =Sk +Sp . (50b)
x1 x1 x1
∫ x2 ′
If Eqs. (50a) and (50b) are compared, it is seen that Sk = St = x1
xU (x) dx .

537
 ERBİL/Turk J Phys

Figure 2. Potential area in the bound state: (a), U (x) > 0 and E > 0; (b) , U (x) < 0 and E < 0 .

√
(c) According to Eq. (8): E = − 2ℏ
m
2 S = − 2ℏ2 Sp →Sp =
2 m 2
m |E|
2ℏ2

√
2m ℏ2 q 2 q2 q
Kd = q → |E|d = q → |E| = = M h → Sp = 2Mh . (50c)
ℏ2 2m d 2 d 2 d

(d) From (50a) and (50b), we obtain the following:

q
Sk =Mh (q − 2). (50d)
d

Thus from Eqs. (50a)–(50d), let us rewrite these potential areas as follows:
∫ x2
q
Sp = U (x) dx =2Mh (P otential energy f ield), (51a)
x1 d

538
 ERBİL/Turk J Phys

∫ x2
′ q
Sk = xU (x) dx =Mh (q − 2) (Kinetic energy f ield), (51b)
x1 d

q2
SE = Sp +Sk = Mh (T otal energy f ield), (51c)
d

For q = 2 , Sk = 0, the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .), the excited states occur. Eqs. (51a)–
(51c) indicate that potential areas are quantized.

4. Summary
It is possible to summarize the results above as follows. We can write the general solutions of the SE in one
dimension as follows:
ψ (x) = Aekx e±iG(x) or ψ (x) = Aek(x−x0 ) e±iG((x−x0 ) , (52a)

ψ (x) = Ae−kx e±iG(x) or ψ (x) = Ae−k(x−x0 ) e±iG((x−x0 ) , (52b)

[ ] [ ]
ψ (x) = Aekx +Be−kx e±iG(x) or ψ (x) = Aek(x−x0 ) +Be−k(x−x0 ) e±iG(x−x0 ) , (52c)

ψ (x) = Acosh (kx) e±iG(x) or ψ (x) = Acosh[k (x−x0 ) ]e±iG(x−x0 ) , (52d)

ψ (x) = A sinh (kx) e±iG(x) or ψ (x) = A sinh [k (x−x0 ) ]e±iG(x−x0 ) , (52e)

ψ (x) = Aekx±iG(x) +Be−kx∓iG(x) or ψ (x) = Aek(x−x0 )±iG(x−x0 ) +Be−k(x−x0 )∓iG(x−x0 ) , (52f)

In these functions, we have the following values:

∫ √
√
(a) For E > U (x) ; k =m1 −E, G (x) = m1 −U (x)dx (53a)

√ ∫ √
(b) For E < U (x) ; k =m1 E, G (x) =m1 U (x)dx. (53b)

√ / √ /
m1 = 2m ℏ = 2m ℏ2 , [m mass or reduced mass of particle, ℏ =h/(2π), h Planck constant]; x1 and x2 (x2 >x1 )
are, depending on E, the roots of the equation E = U (x); and x0 = (x1 +x2 )/2; d =x2 −x1 .
This solution is similar to that of the WKB approach but not exactly the same. There is approximation
in the WKB method, but there is no approach in the method we have given here. Our procedure gives exact
results. Those who are familiar with the WKB approach can easily see the differences.
In bound states, the normalized wave functions are as follows [G is taken as real function):

ψ (x) = A cos [Kx] eiG(x) or ψ (x) = A cos [K (x−x0 )] eiG(x−x0 ) , (54a)

ψ (x) = B sin [Kx] eiG(x) or ψ (x) = B sin [K (x−x0 )] eiG(x−x0 ) , (54b)

539
 ERBİL/Turk J Phys

√ ∫ √
√ √ √ 2m √
A=B= 2/d = 2K/q; k = m1 |E| = |E|; G (x) = m1 |U (x) |dx.
ℏ2

x1 and x2 are the roots of the equation: |E| = |U (x) | or |E| = |U (x−x0 ) |. From this, we have the following:

x0 = (x1 +x2 )/2; d =x2 −x1 ; x1 = x0 −d/2; x2 = x0 +d/2; E = U (x) ; |E| = |U (x1 )| = |U (x2 )|

or |E| = |U (−d/2)| = |U (d/2)|; 2|E| = |U (−d/2)| + |U (d/2)|. (54c)

√
2m ℏ2 q 2 q2 ℏ2
Kd = q → |E|d = q → |E| = = Mh 2 ; [Mh = ] (55)
ℏ2 2m d 2 d 2m

For q = 2 , ground state occurs; for q = nπ, (n = 1, 2, 3, . . .) , excited states occur. The energy values are also
given by the following formulas:
√
m m 2ℏ2 q q
E = − 2 S 2 = − 2 Sp 2 ; Sp = |E|; S = S p (x0 , d) = 2Mh ; Sk (x0 , d) =Mh (q − 2) . (56)
2ℏ 2ℏ m d d

4.1. Practical procedure to find the energy values

As seen in Eq. (55), the total energy values depend on d; so the d value should be calculated. It can be
calculated by one of the equations given in Eq. (54c), but to find the values of d and energy E , the practical
procedure can be given as follows:
/
First, solving the equation |U(x) |= Mh q 2 y, (y =d2 ), the following values are found: x1 ,x2 , d =x2 −x1 ,
d2 (y) = d ∗ d . Then, the equation y =d2 (y) is solved, and the y value is found. So, the energy value is obtained
as follows:
q2 q2 ℏ2 q 2
|E| = Mh = Mh 2 = . (57)
y d 2m d2

For q = 2, the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .) the excited states occur.

5. Examples of one dimensional potentials

p
5.1. Potential: U (x) = a |x| , (a > 0, p > 0)
5.1.1. Energy
According to the practical procedure above:

[ ]1/p [ ]1/p
q2
p Mh q 2 Mh q 2
E = U (x) = a|x| = Mh → x1 = − ; x2 =
y a y a y

[ ]1/p [ ]2/p
Mh q 2 Mh q 2
d =x2 −x1 = 2 ; d2 (y) = d ∗ d = 4 ; y =d2 (y)
a y a y
[ ]−2/(p+2) [ −p ]2/(p+2)
a2−p q2 2 a2
y= ; E q = Mh = Mh q (5.1.1)
Mh q 2 y Mh q 2

540
 ERBİL/Turk J Phys

For q = 2 , the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .), the excited states occur. In the formula
4 ℏω .
q
given in (5.1.1), if p = 2 and a = 12 mω 2 , it is found that Eq = This is the energy of the simple harmonic

2 ℏω 2 ℏω, (n
1 nπ
oscillator. We have E0 = ground state energy for q = 2; Eq =
= 1, 2, 3, ..) excited state energy
( 1)
for q = nπ. The well-known energy of the simple harmonic oscillator is En = n+ 2 ℏω, (n = 0, 1, 2, . . .) . The

2 ℏω 2 ℏω
1 nπ
ground state energy E0 = is the same but the excited energy Eq = En = is different [8]. This
difference is observed in the experimentally measured energy spectra of nuclear nuclei, but this phenomenon
cannot be explained by harmonic oscillatory models. This event can be explained here. We think that the latter
is more accurate because there is no approximation in our solutions.
For the same potential, the ground state energy obtained from the supersymmetric quantum mechanics
(SQ) is as follows [10]:
 ( ) p/(p+2)
0.8862Γ 32 + p1 ℏ2 2
E0 =  ( ) ap  . (5.1.2)
Γ 1+ p1 2m

From Eq. (5.1.2), for p = 2 and a = 12 mω 2 , we find that E0 = 0.999985 12 ℏω ≈ 12 ℏω. In this poten-
[ 2 2 ]1/3
tial, for p = 1 (V-Form potential), E0 ≈ 0.794 ℏ ma is obtained from the formula given in Eq. (5.1.1);
[ ]1/3 [ ]1/3 [ ]1/3
ℏ2 a2 ℏ2 a2 ℏ2 a2
E0 ≈ 0.763 m from the SQ [10]; E0 ≈ 0.813 m from the variation method [11]; and E0 ≈ 0.885 m

from the WKB method [11]. These three values are approximately the same.

5.1.2. Wave functions

∫ √ √ ∫ √
2m √ 2m √ 2|x|
(p+2)/2
G (x) = m1 |U (x) |dx = U (x)dx= Q (x) = a ,
ℏ2 ℏ2 p+2
√ √
ψ (x) = A cos [Kx)] eiG(x) = 2K/q cos [Kx] eiQ(x) = 2/d cos [Kx] eiQ(x) (5.1.3a)

√ √
ψ (x) = B sin [Kx)] eiG(x) = 2K/q sin [Kx] eiQ(x) = 2/d sin [Kx] eiQ(x) . (5.1.3b)
√
2m √ 2|x|
(p+2)/2
mω 2
In the case of simple harmonic oscillator, Q (x) = ℏ2 a p+2 = 2ℏ x . The well-known wave
function of the harmonic oscillator is as follows:
√
−ρ2 /2 mω
ψn (ρ) = An e Hn (ρ) ; [ρ = x] (5.1.4)
ℏ

Here, Hn (ρ) is Hermit polynomials. If Eqs. (5.1.3a) and (5.1.3b) are expanded in series, polynomials are
obtained. So, the well-known function given in Eq. (5.1.4) of harmonic oscillator is an approximate function.

5.2. Infinitely high square potential well or finite square potential well
We consider a particle of mass m captured in a box limited by 0 ≤ x ≤ a . The corresponding potential is given:
U (x) = 0 for 0 < x < a ; U (x) = ∞ for x < 0 and x > a .

541
 ERBİL/Turk J Phys

5.2.1. Energy
The turning points of this potential are given by the following equation: U (x) = E. From this equation, we can
find the classical turning points of the potential function as follows: x1 = 0 and x2 = a ; x0 = (x1 +x2 )/2 =
a/2, d =x2 −x1 = a. By substituting this value of d into the equation dK = q, and then solving for |E|, we can
2 2
ℏ2 q
have the energy value as Eq = 2m a2 = Mh aq 2 ; for q = 2 , the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .) ,
2
ℏ2 (nπ)
the excited states occur. The well-known energy is En = 2m a2 [8]. It is seen that the two energy values
are the same.

5.2.2. Wave functions

∫√
Here, U (x) = 0 . So, G (x) = m1 |U (x) |dx = 0 ,
√ √
ψ (x) = A cos [Kx] eiG(x) = 2/a cos [Kx] = 2K/q cos [Kx],
√ √
ψ (x) = B sin [Kx] eiG(x) = 2/a sin [Kx] = 2K/q sin [Kx],
[ a ] √ [ a ] √ [ a ]
ψ (x) = A cos K(x− ) = 2/a cos K(x− ) = 2K/q cos K(x− ) ,
2 2 2
[ a ] √ [ a ] √ [ a ]
ψ (x) = B sin K(x− ) = 2/a sin K(x− ) = 2K/q sin K(x− ) .
2 2 2

5.3. Trigonometric potential well

We consider the potential energy of the particle U (x) so that U (x) = U0 cotg 2 (πx/a), [U 0 > 0, a > 0 and
0 < x < a].

5.3.1. Energy
The roots of the equation
q2 q2
Eq = U (x) = Mh →U0 cotg 2 [πx/a] = Mh ;
y y
[ √ ] [ √ ]
1 Mh q 2 1 Mh q 2
x1 = − arccotg ;x2 = arccotg ; d =x2 −x1 ; y = d ∗ d
π U0 y π U0 y
[√ ]
4 Mh q 2 q2 q2
y = 2 arccotg 2 ; E q = Mh 2
= Mh . (5.3.1)
π U0 y d y

Eq. (5.3.1) is not solved analytically. So, it may be solved by the numerical method for the exact values.
We have for q = 2 , the ground state; for q = nπ, (n = 1, 2, 3, . . .), the excited states.

5.3.2. Wave functions

∫ √ ∫ √ ∫ √ √
2ma2 U0 [ πx ]
G (x) =m1 |U (x) |dx = m1 U (x)dx=m1 2
U0 cotg (πx/a)dx= Q(x), Q (x) = ln sin( ) ,
ℏ π
2 2 a
√ √
ψ (x) = A cos [Kx)] eiG(x) = 2K/q cos [Kx] eiQ(x) = 2/d cos [Kx] eiQ(x) ,
√ √
ψ (x) = B sin [Kx)] eiG(x) = 2K/q sin [Kx] eiQ(x) = 2/d sin [Kx] eiQ(x) .

542
 ERBİL/Turk J Phys

5.4. Infinitely high parabolic potential well

(a )
x 2
We consider the potential energy of the particle, U (x), so that U (x) = U0 x−a , [ U0 > 0, a > 0 , x > 0].

5.4.1. Energy
(a )
x 2
2
The positive roots of the equation E = U (x) = U0 x−a = Mh qy are as follows:

v [ ] v [ ]
u √ u √
u a2 Mh q 2 +2yU0 −q √Mh Mh q 2 +4yU0 u a2 Mh q 2 +2yU0 +q √Mh Mh q 2 +4yU0
t t
x1 = ; x2 =
2yU0 2yU0

v [ ] v ] 2
u √ √ u[ √ √
u u
a t Mh q +2yU0 +q Mh Mh q +4yU0 t Mh q +2yU0 −q Mh Mh q +4yU0 
2 2 2 2
d2 = d2 = (x2 −x1 ) =   .
2
−
2 yU0 yU0 

√ 2 2
√ √
2 ℏ2 U0
The root of the equation d2 = y isy = a2 MUh0q and Eq = Mh dq 2 = Mh qy = Mh U0
a2 q → Eq = 2ma2 q .

We have ground state for q = 2 ; the excited states for q = nπ, (n = 1, 2, 3, . . .) [8]. For m = 1andℏ = 1,
√
this energy value is En = π2 n 2Ua2 . For this potential, the energy values obtained from the supersymmetric
0

WKB [12] and standard WKB [14], respectively, are as follows:

√ √
2U0 [ √ ] 2U0 [ √ ]
En = 2n+ 1 + 8a2 U0 −2U 0 and En = 2n + 1+ 2a2U
0 −2U0 .
a2 a2

5.4.2. Wave functions

∫ √ ∫ √ ∫ √ ( ∫ √ (
a x )2 a x)
G (x) = m1 |U (x)|dx = m1 U (x)dx = m1 U0 − dx = m1 U0 − dx= Q(x);
x a x a
√ [ ]
2mU0 x2
Q (x) = a ln (x) −
ℏ2 2a

√ √ √
ψ (x) = A cos [Kx] e±iQ(x) ; ψ (x) = B sin [Kx] e±iQ(x) ; |A| = |B| = 2/d = 2K/q, [δ = Eq 2 −4ab].

5.5. Potential U(x) = ax2 +b/x2

We consider the potential energy of the particle as U (x) = ax2 +b/x2 , where a and b are positive constants.

5.5.1. Energy
/ 2
The roots of the equation U (x) = ax2 +b x2 = Mh qy = Eq are as follows:

√ √ √ √ √
Eq +δ Eq −δ Eq −δ Eq +δ
x1 = − ,x2 = ,x3 = − ,x4 = ; [δ = Eq 2 −4ab].
2a 2a 2a 2a

543
 ERBİL/Turk J Phys

From these grandeurs, we get

√ √
Eq +δ Eq −δ √
d31 = x3 −x1 = − + ; d31 2 = Eq /a−2 b/a;
2a 2a

√ √
Eq +δ Eq −δ √
d42 = x4 −x2 = − ; d42 2 = Eq /a−2 b/a.
2a 2a
√
From these equations, we can easily obtain d31 = −d42 = d,d2 = Eq /a−2 b/a .
2
From the solution of the equation Eq = Mh dq 2 , we get the positive energy value as follows:

√ √
√ 2
√ ℏ2 2
Eq = ab+ ab + aMq q = ab+ ab + a q (5.5.1)
2m

For q = 2 , the ground state occurs; for q = nπ, n = 1, 2, 3, . . .) , the excited states occur.
For m = 1 and ℏ = 1 , this energy becomes:

√
√ a
Eq = ab+ ab+ π 2 n2 , (n = 1, 2, 3, . . .). (5.5.2a)
2

The energies obtained from the supersymmetric WKB and standard WKB [8,12–14] respectively are as
follows:
[ √ ]
√ 1
En = 2a 2n + 1+ +2b , (n = 0, 1, 2, 3, . . .), (5.5.2b)
4

√ [ √ ]
En = 2a 2n + 1+ 2b , (n = 0, 1, 2, 3, . . .). (5.5.2c)

It can be assumed that the values given in Eqs. (5.5.1) and (5.5.2a) are more accurate because there is
no approximation.

5.5.2. Wave functions

∫ √ ∫ √ ∫ √
G (x) = m1 |U (x)|dx = m1 U (x)dx=m1 ax2 +b/x2 dx

1 {√ √ [ ( ) √ √ ]}
=m1 b + ax4 + b ln x2 − ln (b+ b b + ax4 ) = Q(x),
2

√
m {√ √ [ ( ) √ √
4 + b ln x2 − ln (b+ b b + ax4 )
]}
Q (x) = b + ax ,
2ℏ2
√ √
ψ (x) = A cos [Kx] eiQ(x) and ψ (x) = B sin [Kx] eiQ(x) ; |A| = |B| = 2/d = 2K/q .

544
 ERBİL/Turk J Phys

6. The radial Schrödinger equation for spherical symmetric potentials

The Schrödinger equation (SE) in three dimensions is given as follows:

2m
∆Ψ (⃗r) + [E − V (⃗r)] Ψ (⃗r) = 0. (58)
ℏ2
Here, E and V are the total and potential energies, respectively, m is the mass or reduced mass of particle.
Spherical polar coordinates x = r sin (θ) cos (ϕ) , y = r sin (θ) sin (ϕ) , and z = r cos (θ) are appropriate for the
symmetry of the problem. The SE given in Eq. (58), expressed in these coordinates, is as follows:
[ ]
∂2 2 ∂ 1 2m
+ Ψ (r, θ, ϕ) + 2 L̂2 (θ, ϕ) Ψ (r, θ, ϕ) + 2 [E − V (r, θ, ϕ)] Ψ (r, θ, ϕ) = 0. (59a)
∂r2 r ∂r r ℏ

Here,
∂2 ∂ 1 ∂2
L̂2 (θ, ϕ) = 2
+cotg (θ) + 2 . (59b)
∂θ ∂θ sin (θ) ∂ϕ2
The potential energy of a particle which moves in a central, spherically symmetric field of force depends only upon
the distance r between the particle and the force center. Thus, the potential energy should be V (r, θ, ϕ) = V (r).
Solution of Eqs. (59a) and (59b) can be found by the method of separation of variables. To apply this method,
the solution is assumed to be in the following form:

Ψ (r, θ, ϕ) = R (r) Y (θ, ϕ) or Ψ (r, θ, ϕ) = R (r) |jm >. (60)

In Eq. (60), R(r) is independent of the angles and Y (θ, ϕ) or |jm > is independent of r . Substituting Eq.
(60) into Eq. (59a) and rearranging it, the following two equations are obtained:
{ }
∂ 2 R(r) 2 ∂R(r) 2m C
+ + [E − V (r)] − 2 R (r) = 0, (61)
∂r2 r ∂r ℏ2 r

L̂2 (θ, ϕ) Y (θ, ϕ) +CY (θ, ϕ) = 0, (62)

where C is constant. Eq. (62) is independent of the total energy E and of the potential energy V (r)
and, therefore, the angular dependence of the wave functions is determined by the property of spherical
symmetry, and admissible solutions of Eq. (62) are valid for every spherically symmetric system regardless
of the special form of the potential function. The solutions of Eq. (62) can be found in any quantum
mechanics and mathematical physics textbook [15–18] and the solutions are known as spherical harmonic
functions, Ylµ (θ, ϕ) , whereC = l (l + 1) , (l = 0, 1, 2, 3, . . .) are positive integer numbers and µ = −l, −l+1, −l+
2, . . . 0, 1, 2, . . . , l. Eq. (61) is the radial SE. Substituting C = l (l + 1) and F (r) = rR(r) values into Eq. (61),
the radial wave equation is obtained as follows:

∂ 2 F (r) 2m
+ 2 [E − U (r)] F (r) = 0. (63)
∂r2 ℏ
ℏ l(l+1)
2
Here, U (r) = V (r) + 2m r2 is the effective potential energy. Eq. (63) is a one dimensional differential
equation and is the same as Eq. (2). In Eq. (2), the free variable is x , while in Eq. (63) the free variable is
r . So, the solution procedure of one dimensional differential equation has been given above in Eqs. (52a)–(57).
Now let us give some examples.

545
 ERBİL/Turk J Phys

6.1. Coulomb type central potential well

/
The potential energy of hydrogen-like atom is V (r) = −Ze2 r . If the centrifugal potential function is added to
[ ]
ℏ2 l(l+1)
this potential V (r), we have the following effective potential function: U (r) = − ar + rb2 , a = Ze2 and b = 2m r2

6.1.1. Energy
2
The classical turning points of this effective potential are given by the following equation: − ar + rb2 = − |E| = −Mh qy .
From this equation, we can find the classical turning points of this effective potential function and d2 as follows
[21]:
√ √ [ ]
ay− a2 y 2 −4bMh yq 2 ay+ a2 y 2 −4bMh yq 2 2 y ya2 −4bMh q 2
r1 = ;r2 = ; d2 = (r2 −r1 ) = .
2Mh q 2 2Mh q 2 Mh 2 q 4

Mh q 2 [Mh q 2 +4b] 2 2
The root of the equation d2 = y is y = a2 , and |Eq | =Mh dq 2 = Mh qy ;

2me4 Z 2 me4 Z2 Z2 me4

Eq = − =− 2 [ / ] = −E0 [ / ] , E0 = (6.1.1)
ℏ2 2 2
[4l + 4l +q ] 2
2ℏ l (l + 1) +q 4 2
l (l + 1) +q 4 2ℏ2

We have the ground state for q = 2 ; the excited states for q = nπ, (n = 1, 2, 3, . . .) . For the hydrogen
atom; Z = 1 atom number, m electron mass, e electron charge and in the ground state, l = 0 . From Eq.
(6.1.1), E0 = −13.6eV is obtained. This is the well-known ground state energy of the hydrogen atom. The well-
2
known excited state energy of hydrogen-like atoms is given as follows: En = −E0 Zn2 , (n = 1, 2, 3, . . .) . There is
no obvious l = 0 quantum number in this formula, but in Eq. (6.1.1) this number of quantum is clearly visible
[19].

6.1.2. Wave functions

∫ √ √ ∫ √ ∫ √
2m √ 2m a b
G (r) =m1 |U (r)|dr = −U (r)dr = − dr = Q(r);
ℏ2 ℏ2 r r2
√ [ √ ]
2m √ √ ar − b
Q (r) = 2 ar − b− barctan( ) ,
ℏ2 b
√ √ [ ]
F (r) = A cos [Kr] eiQ(r) and F (r) = 2
d cos [Kr] e iQ(r)
= 2K
q cos [Kr] e
iQ(r)
, F (r) = B sin dq r eiQ(r) , and
√ [ q ] iQ(r) √ [ q ] iQ(r)
2 2K
F (r) = d sin d r e = q sin d r e ,

F (r) F (r)
Ψ (r, θ, ϕ) = R (r) Y (θ, ϕ) = Y (θ, ϕ) = |jm >. (6.1.2)
r r

In Eq. (6.1.2), m is not mass, it is magnetic quantum number. The known wave function of the hydrogen
atom is an exponential function including Laguerre polynomials. Our results are more accurate because there
is no approach.

546
 ERBİL/Turk J Phys

6.2. Spherical symmetric square well (infinitely high or finite)

Consider a particle of mass m captured in a box limited by 0 ≤ r ≤ a . The corresponding central potential can
be given by V (r) = 0 for 0 ≤ r ≤ a; V (r) = ∞ for r < 0 and r > a. With this potential, the effective potential
is as follows:
b ℏ2
U (r) = ; [b = l (l + 1)].
r2 2m

6.2.1. Energy
b Mh q 2
With this effective potential, the equation U (r) = r2 can be written. From this equation, the
= y
√
2
classical turning points and some grandeur are found as follows [19]: r1 = 1q M by
h
;r2 = a; d2 = (r2 −r1 ) =
[ √ ]2
by
a− 1q M h
.
M h a2 q 2
From the solution of the equation d2 = y ; we can obtain y = √ √ 2 and the energy value as follows:
[ b± Mh q ]

[√ √ ]2
b± Mh q ℏ2 [ √ ]2
Eq = = l (l + 1)±q . (6.2.1)
a2 2ma2

For q = 2 , the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .) the excited states occur.
The known allowed energies are given as follows [9]:

ℏ2
Enl = βnl 2 . (6.2.2)
2ma2

Here, βnl is the n th zero of the lth spherical Bessel functions. Eq. (6.2.1) can be written as follows:

ℏ2 [ √ ]2
Eq = Enl = l(l + 1)±nπ . (6.2.3)
2ma2

Some values of βnl and energies calculated according to Eqs. (6.2.2) and (6.2.3) [with sign + in (6.2.3)] are
given in the Table 1.
/
Table 1. Some energy values of the infinitely high spherical symmetric square well (Unit ℏ2 (2ma2 ) .

nl βnl Enl From (6.2.2) Enl From (6.2.3)

1s 3.142 9.872 9.870
1p 4.493 20.187 20.755
1d 5.763 33.212 31.260
2s 6.283 39.476 39.478
2p 7.725 59.676 59.250
2d 9.095 82.719 76.260

547
 ERBİL/Turk J Phys

6.2.2. Wave functions

∫ √ √ ∫ √ ∫ √ √ ∫ √
2m b 2m b
G (r) = m1 |U (r) |dr = |U (r) |dr = m1 dr = dr,
ℏ2 r2 ℏ2 r2

∫ √ √
√
b 2m √
G (r) = m1 dr = b ln (r) = l(l + 1) ln (r) = Q(r),
r2 ℏ 2

F (r) = A cos [Kr] eiG(r) = A cos [Kr] eiQ(r) ;

[q ] [q ]
F (r) = B sin [Kr] eiG(r) = sin [Kr)] eiQ(r) ; F (r) = A cos r eiG(r) = A cos r eiQ(r) ;
d d
[q ] [q ] √ √
F (r) = B sin r eiG(r) = sin r eiQ(r) ; |A| = |B| = 2/d = 2K/q
d d

F (r)
Ψ (r, θ, ϕ) = R (r) Y (θ, ϕ) = Y (θ, ϕ) .
r

6.3. Three dimensional isotropic harmonic oscillator potential

1 2 2
The potential energy of three dimensional isotropic harmonic oscillators is given by V (r) = 2 mω r . With
this potential, the effective potential U (r) is as follows:

1 ℏ2 l(l + 1) 1 b 1 ℏ2 l (l + 1)
U (r) = mω 2 r2 + 2
= ar2 + 2 ; [a = mω 2 and b = ].
2 2m r r 2 2m

6.3.1. Energy
√
2 Eq −δ
From this effective potential, the positive roots of the equation Eq = Mh qy = U (r) are r1 = 2a ;r2 =
√ √
Eq +δ q2
2a ; [Eq = M h y ; δ = Eq 2 −4ab].
√ √
Eq +δ Eq −δ
With these roots, d =r2 −r1 = 2a − 2a ;d2 = y = d ∗ d =d .
2

2
By substituting this value of d2 into the equation y =Mh qy , and then solving for Eq , we can obtain the
appropriate energy as follows [19]:

1 √ √
Eq = ℏω[ l (l + 1)+ l (l + 1) +q 2 ]. (6.3.1)
2

For q = 2 , the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .) , the excited states occur.
The known energy values are given as follows [20]:

Enl = (2n + l+3/2)ℏω, (n = 0, 1, 2, 3, . . . .). (6.3.2)

Some values of the energies calculated according to Eqs. (6.3.1) and (6.3.2) are given in Table 2.

548
 ERBİL/Turk J Phys

Table 2. Some energy values of the isotropic harmonic oscillator (unit ℏω) .

nl Enl [according to (6.3.2)] Enl [according to (6.3.1)]

1s 3.500 1.571
1p 4.500 2.430
1d 5.500 3.217
2s 5.500 3.142
2p 6.500 3.927
2d 7.500 4.597

6.3.2. Wave functions

∫ √ √ ∫ √ ∫ √
2m √ 2m b
G (r) =m1 |U (r) |dr = U (r)dr = ar2 + 2 dr = Q(r),
ℏ2 ℏ2 r
{ [ √ √ ]}
1 √ 4 √ b+ ar4 +b
Q (r) = m1 ar +b− b ln 2 ,
2 r2
[q ]
F (r) = A cos [Kr] eiG(r) = A cos [Kr] eiQ(r) = A cos r eiQ(r) ,
d
[q ]
F (r) = B sin [Kr] eiG(r) = B sin [Kr] eiQ(r) = B sin r eiQ(r) ,
d
√ √ F (r)
|A| = |B| = 2/d = 2K/q; Ψ (r, θ, ϕ) = R (r) Y (θ, ϕ) =
Y (θ, ϕ) .
r
( ) √
The known radial wave functions are as follows [20]: Rnl (ρ) = Aρl+1 e− 2 ρ Ln l+1/2 ρ2 . Here, ρ = mω/ℏr
1 2

and L are the Laguerre polynomials.

6.4. Three dimensional isotropic harmonic oscillator with spin-orbit coupling (shell model in
nuclear physics)
1 2 2
The simple potential energy of a three dimensional isotropic harmonic oscillator is given by V0 (r) = 2 mω r .
The spin −orbit interaction potential must be added to this simple potential. The spin −orbit interaction
potential is given as follows:
{ }
ℏ2 1 dV0 (r) ⃗ ⃗ 1
Vlsj (r) = − l⃗
s ; l.⃗
s = [j (j + 1) −l (l + 1) −s (s + 1)]
2m2 c2 r dr 2

Here l, s , and j are the orbital, spin, and total angular momentum quantum numbers of a particle, respectively. It
is possible to find the derivation of this expression in any quantum mechanics textbook. If Vlsj (r) is calculated,
the following value is found:
2
ℏ ω2
Vlsj (r) = − 4mc 2 [j (j + 1) −l (l + 1) −s (s + 1)] = −Clsj. If s =1/2 (for fermions) is accepted,

2 2 [ ]
ℏ ω
Vlsj (r) = − 4mc 2 j (j + 1) −l (l + 1) − 34 = −Clsj is obtained. Thus, the harmonic oscillator potential and
effective potential with spin-orbit coupling becomes as follows:
[ ]
/ 1 ℏ2
V (r) = V0 (r) +Vlsj (r) ; U (r) = ar −Clsj +b r2 ;
2 2
a = mω , b = l (l + 1) .
2 2m

549
 ERBİL/Turk J Phys

6.4.1. Energy
2
From this effective potential, the positive roots of the equation Eq = Mh qy = U (r) are as follows:

√ √ √
(Eq +Clsj ) −δ (Eq +Clsj ) +δ 2
r1 = ; r2 = ; [δ = (Eq +Clsj ) −4ab].
2a 2a
√ √
Eq +δ Eq −δ
With these roots, d =r2 −r1 = 2a − 2a ;d2 = y = d ∗ d =d2 . By substituting this value of d2 into the
2
equation y =Mh qy , and then solving for Eq , we can obtain the appropriate energy as follows:

[ √ ]
1 √ √ 2
Eq = ℏω 2
l(l + 1)−Cj + ( l(l + 1)−Cj ) +q , [Cj = Clsj /(ℏω)] (6.4.1)
2

For q = 2 , the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .) , the excited states occur.
The known energy values (with first order perturbation) are as follows [7,9,20]:

C0
Enl = (2n + l+3/2) ℏω− [j (j + 1) −l (l + 1) −s(s + 1)] ℏω. (6.4.2)
4ℏω

Here, n = 0, 1, 2, . . , integer numbers and C0 is a positive parameter. In Table 3, some energy values
calculated according to Eqs. (6.4.1) and (6.4.2) are given, with the following values:

C0
C0 = 0.015ℏω, s =1/2 and Cj = [j (j + 1) −l (l + 1) −3/4] ℏω.
4ℏω

Table 3. Some energy values of the isotropic harmonic oscillator with spin-orbit coupling.

States According to (6.4.2) (unit ℏω) According to (6.4.1) (unit ℏω)

1d5/2 3.493 3.211
1f5/2 4.515 4.083
1f7/2 4.489 4.061
1g7/2 5.519 4.986
1g9/2 5.485 4.955

6.4.2. Wave functions

∫ √ √ ∫ √
2m
G (r) = m1 |U (r) |dr = U (r)dr = Q (r)
ℏ2
√ { √ [ √ ]
2m 1 Clsj 2ar2 −Clsj
Q (r) = ar −Clsj r +b− √ ln 2 ar −Clsj r +b+
4 2 4 2 √ −
ℏ2 2 4 a a
√ [ √ ] }
b √
ln 2 b(ar4 −Clsj r2 +b)−Clsj r +2b + b ln (r)
2
2
[q ]
F (r) = A cos [Kr] eiG(r) = A cos [Kr] eiQ(r) = A cos r eiQ(r)
d

550
 ERBİL/Turk J Phys

[q ]
F (r) = B sin [Kr] eiG(r) = B sin [Kr] eiQ(r) = sin r eiQ(r)
d
√ √ √ F (r)
|A| = |B| = 2/d = 2K/q = q/d; Ψ (r, θ, ϕ) = R (r) Y (θ, ϕ) = Y (θ, ϕ)
r

The known radial wave functions are [20]: Rnl (ρ) = Aρl+1 e− 2 ρ Ln l+1/2 (ρ2 )
1 2

√
Here ρ = mω/ℏr and L is the Laguerre polynomial.

6.5. Three axial deformed harmonic oscillator potential (Nilsson model in the nuclear physics)
It was examined in detail in [36].

6.6. Periodic potential of arbitrary form

It was examined in detail in [35] with examples.

6.7. Trigonometric and Hyperbolic Pöschl–Teller potential wells

It was examined in detail in [34].

6.8. Saxon–Woods type central potential

Saxon–Woods type potential function is given as V1 (r) = − 1+e[(r−R
V0
0 )/a]
; [a, V0 ,R0 parameters]. The spin-orbit
interaction potential energy term can be added to this potential. The spin-orbit interaction potential energy
term is given as follows: Vlsj (r) = − 4µℏ2 c2 1r dVdr
2
1 (r)
[j (j + 1) −l (l + 1) −s(s + 1)] , [ µis the reduced mass]. If this
function is calculated, the following potential is found:

1 er/a ℏ2 c2 V0 e−R0 /a
Vlsj = −Clsj [ ]2 ; {C lsj = [j (j + 1) −l (l + 1) −s (s + 1)] .
r [ r−R0
] 4µ2 c4 a
1+e a

ℏ2 l(l+1) 1
The centrifugal energy is given by Vc (r) = 2µ r2 . The Coulomb potential Vcp (r) for charged par-
ticles must also be gotten if they exist. So, the effective potential can be obviously written as U (r) =
er/a ℏ2 l(l+1) 1
V1 (r) +Vlsj +Vc (r) +Vcp (r) or U (r) = − V0
r−R0 −Clsj 1r [ r−R0 ]2 + 2µ r2 + Vcp (r) .
1+e[ a ] 1+e[ a ]

The classical turning points of this effective potential cannot be found analytically and, therefore, it must
/
be found numerically. The classical turning points are the roots of the equation U (r) = − |Eq | = −Mh q 2 y .
2
Let these roots be r1 (y) and r2 (y). With these roots, first, d2 = [r2 (y) −r1 (y)] = d2 (y) is found, and after
/
solving the equation d2 (y) = y, the following is found: y and Eq = Mh q 2 y .

6.8.1. Numeric calculations

Let us calculate the energy values of Cu (29, 68). We have taken the potentials as follows:

3Rco 2 −r2
Vcp (r) = (Z − 1) e2 , [Coulomb potential in the sphere]
2Rco 3

551
 ERBİL/Turk J Phys

V0
V (r) = − , (Saxon − W oods potentials)
1+e(r−R0 )/a0
α (L, S, J) = 0.5 [J (J + 1) −L (L + 1) −S(S + 1)]

ℏ2 1 Vs0 e(r−Rs0 )/as0

VLSJ (r) = − α (L, S, J) [ ] , (spin − orbit potential)
2µ r as0 1+e(r−Rs0 )/as0 2

For protons, U (r) = V (r) +VLSJ (r) +Vc (r) +Vcp (r) ; for neutrons, U (r) = V (r) +VLSJ (r) + Vc (r). The
parameters in these potentials have been calculated by the method of Volya [21,22] and their values are
as0 = a0 = 0.662; Rco = R0 = 5.142885; V0 = 47.655271; Vso = 28.422020; Rso = 4.726557. The values of
d are obtained by solving the following equation:

ℏ2 q 2
U (−d/2) + U (d/2) = −2 ,
2µ d2

and the energy values have been calculated with the following formula:

ℏ2 q 2
Eq = − .
2µ d2

The results are seen in the Table 4.

Table 4. A few energy values of the Cu (29,68) nucleus with Saxon–Woods potential (unit MeV).

States Eq (M eV )(neutron) Eq (M eV )(proton)

1 s 1/2 −45.6256 −33.9536
1 p 3/2 −15.1995 −11.3653
1 p 1/2 −15.1987 −11.3644
1 d 5/2 −6.5021 −4.9028
1 d 3/2 −6.5011 −4.9015
2 s 1/2 −45.6061 −33.9536
1 f 7/2 −3.4830 −2.6484
2 p 3/2 −25.1702 −18.8825

6.9. Relativistic Dirac equation in a central potential

Consider a Dirac particle (spin is 1/2) of mass m captured in a central potential well, V (r). With this
potential, Dirac Hamiltonian can be written as follows [8,9]: HD = α
⃗ p⃗+βm + V (r). Here, we have the
following relativistic units: ℏ = c = 1, p⃗= −i∇;
⃗ α⃗ = (αx ,αy ,αz ) , and β are hermitical 4-operators acting on the
spin variables alone. Including the position vector ⃗r; α
⃗ = d⃗r/dt(velocity). The Hamiltonian HD and radial
equation of Schrödinger can be brought to the form below [23]:

d2 F (r)
+ [α − U (r)] F (r) = 0; α =E 2 −m2 ; U (r) = Ure (r) +Uj (r) +UL (r)±Urez (r)
dr2
/ /
2 2
Ure (r) = 2EV (r) −V (r) ; Uj (r) = (j+1/2) r2 ; UL (r) = L(L + 1) r2 ;
√ /
2 2
Urez (r) = (j+1/2) r4 −V ′ (r) .

552
 ERBİL/Turk J Phys

√ ∫√
For α > U (r) boundstate : k = i α; G (r) = U (r)dr. r1 and r2 are the roots of the following equation:
( )
[α − U (r) ] = 0 or E 2 −m2 −U (r) = 0 and d =r2 −r1 ; (r1 <r2 ).

The bound state energies are given by the solution of the following equation:

√ √ √
Kd = |α|d = d |E 2 −m2 |= d m2 −E 2 = q, [m > |E| ].

For q = 2 , the ground state occurs; for q = nπ, (n = 1, 2, 3, . . .), the excited states occur.
Now let us find the function F (r) in bound states. In bound states, always E > U (r) thatis, α > U (r)
Therefore:
√ √ √
K= |α|= |E 2 −m2 |= m2 −E 2 = K > 0,
∫ √ ∫ √
G (r) = i U (r)dr = i −|U (r) |dr = Q(r),

[q ]
F (r) = A cos [Kr] eiG(r) = A cos [Kr] eiQ(r) = A cos r eiQ(r) ,
d
[q ]
F (r) = B sin [Kr] eiG(r) = B sin [Kr] eiQ(r) = sin r eiQ(r) .
d

The application to the atoms of hydrogen-like was given in detail in [23].

7. Transmission coefficient for an arbitrary form potential barrier

7.1. Determination of the wave functions
Let us consider the solution as follows:

F (r) = Aekr±iG(r) +Be−kr∓iG(r) (64)

√ ∫√
Here, (a) For the case where E > U (r) , k = im1 E, G (r) = im1 U (r)dr ,
√ ∫√
(b) For the case where E < U (r) , k =m1 E, G (r) =m1 U (r)dr .
Let us divide the potential into three domains, as seen in Figure 3. In region I, E >U1 ; in region II,
E <U2 ; and in region III, E >U3 . Now, consider that a particle with total energy E comes from the left as in
Figure 3 and hits the barrier at the point r1 .

Figure 3. The unbounded state (potential barrier) and tunneling.

553
 ERBİL/Turk J Phys

According to the function given in Eq. (64), the wave functions can be obtained for these three regions
as follows:
F1 (r) = A1 eiKr±Q1 (r) +B1 e−iKr∓Q1 (r) ,

F2 (r) = A2 eKr∓iQ2 (r) +B2 e−Kr±iQ2 (r) , (65)

F3 (r) = A3 eiKr±Q3 (r) .

√ ∫√
Here, k = m1 E,Qp (r) = m1 Up (r)dr, (p = 1, 2, 3).
In the calculations of the above functions, the fact that the waves travel both from left to right and right
to left in region I and only from left to right in region III was taken into account. Since there is no wave coming
from right to left in region III, the coefficient B3 must be zero.

7.2. Calculation of the transmission coefficient T

To calculate the transmission coefficient T, the coefficients Ai and Bi in the functions given in Eq. (65) must
be found. In order to find these coefficients, the following boundary conditions are used:
′ ′ ′ ′
F1 (r1 ) = F2 (r1 ) ;F1 (r1 ) = F2 (r1 ) ; F2 (r2 ) = F3 (r2 ) ;F2 (r2 ) = F3 (r2 )
√ √ √
E =U1 (r1 ) = U2 (r1 ) ; E =U2 (r2 ) = U3 (r2 ) ; m1 E = m1 Up (r1 ) = m1 Up (r2 )= K;
√ √ √
m1 E = m1 Up (r1 ) = m1 Up (r2 )= K
∫ √ √
′ ′ ′
Qp (r) = m1 Up (r)dr →Qp (r) = m1 Up (r); Qp (r1 ) = Qp (r2 ) = K.

In the functions given in Eq. (65), there are five unknown coefficients. Four of them can be found in term of
A1 . All of the coefficients have been calculated in this study, but here, only the coefficient A3 is sufficient.
According to the signs of the exponential terms in Eq. (65), two expressions for A3 can be found.
For the lower part of signs:

2exp [(1 + i) Kr1 + (1 − i) Kr2 −Q1 (r1 ) +iQ2 (r1 ) +iQ2 (r2 ) +Q3 (r2 )]
A3 = A1 .
exp [2Kr1 +2iQ2 (r1 )] +exp [2Kr2 +2iQ2 (r2 )]

For the upper part of signs:

2exp [(1 + i) Kr1 + (1 − i) Kr2 +Q1 (r1 ) +iQ2 (r1 ) +iQ2 (r2 ) −Q3 (r2 )]
A3 = A1 .
exp [2Kr1 +2iQ2 (r2 )] +exp [2Kr2 +2iQ2 (r1 )]

The transmission coefficient is defined as follows:

A3 A3 ∗ 2
T = = (66)
A1 A1 ∗ cosh [2Kd] + cos[2P ]
√ ∫ r2 √
[ P =Q2 (r2 ) −Q2 (r1 ) = 2m
ℏ2 r1
U2 (r)dr ]

∫ √ √ ∫ √
2m
Q2 (r) = m1 U2 (r)dr = U2 (r)dr.
ℏ2

554
 ERBİL/Turk J Phys

In the literature [6], the transmission coefficient T (or the barrier penetration probability) which is calculated
by the WKB method is known as follows:
√ ∫
−2g 2m r2 √
T =e , [g = U2 (r) −Edr]. (67)
ℏ2 r1

In Eqs. (66) and (67), r1 and r2 are abscises of the points that the particle hits and leaves the potential barrier,
respectively.

7.3. Application to cold emission

7.3.1. Calculation of transmission coefficient
Cold emission of electrons from a metal surface is the basis of an important device known as scanning tunneling
microscope (STM). An STM consists of a very sharp conducting probe which is scanned over the surface of
a metal (or any other solid conducting medium). A large voltage difference is applied between the probe and
the surface. The surface electric field strength immediately below the probe tip is proportional to the applied
potential difference, and inversely proportional to the spacing between the tip and the surface. Electrons
tunneling between the surface and the probe tip cause a weak electric current. The magnitude of this current
is proportional to the tunneling probability T. It follows that the current is an extremely sensitive function of
the surface electric field strength and, hence, of the spacing between the tip and the surface (assuming that the
potential difference is held constant). An STM can thus be used to construct a very accurate contour map of
the surface under investigation. In fact, STMs are capable of achieving sufficient resolution to image individual
atoms.
Suppose that a cold metal surface is subject to a large uniform external electric field of strength ∈ ,
which is directed such that it accelerates electrons away from the surface. The electrons just below the surface
of a metal can be regarded as being in a potential well of depth W , where W is called the work function
of the surface. Adopting a simple one dimensional treatment of the problem let the metal lie at x < 0, and
the surface at x = 0 . The applied electric field is shielded from the interior of the metal. Hence, the energy
E , for example, of an electron just below the surface is unaffected by the field. In the absence of the electric
field, the potential barrier just above the surface is simply U (x) −E = W . The electric field modifies this to
U (x) −E = W − e ∈ x . The potential barrier is sketched in Figure 4. It can be seen in Figure 4 that an electron
just below the surface of the metal is confined by a triangular potential barrier which extends from x = x1 tox2 ,
where x1 = 0 and x2 = W /(e ∈) . In Eq. (66), if it is put that:

√ √
2m 2m √ √ √
d =x2 −x1 = W / (e ∈) ;m1 = ; k= E = m1 E = m1 − |E|= im1 W
ℏ2 ℏ 2

∫ √ √
√ 2 2me ∈ 3/2 8me ∈ 3/2
Q2 (x) = m1 −e ∈ x= i x =i x .
3ℏ 9ℏ2

With these values, the following transmission coefficient is obtained:

2 2
T = T new = = [ √ ] [ √ ]. (68)
cosh [2Kd] + cos [2Q2 (x2 )] cos ℏe∈ W 3/2 +cosh 43ℏe∈
2 2m 2m
W 3/2

555
 ERBİL/Turk J Phys

Figure 4. The potential barrier for an electron in a metal surface subject to an external electric field.

Here, in calculation of Eq. (68), the following is used:

cosh (iy) = cos (y) and cos (iy) = cosh(y).

Using the WKB approximation [6], the probability of such an electron tunneling through the barrier and
consequently being emitted from the surface is calculated as follows:
[ ∫ ] [ ∫ ]
x2 √ x2 √
Twkb = exp −2m1 U (x) −E dx = exp −2m1 W − e ∈ x dx
x1 x1

[ √ ]
4 2m 3/2
Twkb = exp − W . (69)
3ℏe ∈

The above result given in Eq. (69) is known as the Fowler–Northeim formula. This formula is the result of the
WKB approximation. The formula given in Eq. (68) is an exact formula because there is no approximation.

7.3.2. Numerical calculations and comparison of Eqs. (68) and (69)

The barrier penetration probabilities or the transmission coefficients T have been calculated from Eqs. (68) and
1/2
(69). In the calculations, the electron mass, mc2 = 0.511003M eV ; the electron charge, e = 1.19999 (M eV.f m)
have been taken. The obtained values for the different metals are seen in Table 5. In Table 5, it can be seen that
the new method is more appropriate than the classical WKB method. In the calculations of the current-voltage
characteristics of a diode in semiconductor physics, it is expected to have better results [ℏc = 197.329M eV.f m] .

7.4. Application to alpha decay in atom nuclei and calculation of half-life

7.4.1. Calculation of half-life formula
An α -particle is the nucleus of a helium atom. It consists of two protons and two neutrons. In the process
of α -decay of nuclei, an α -particle is assumed to move in a spherical region determined by the daughter nucleus.
The central feature of this one-body model is that the α -particle is preformed inside the parent nucleus. The
theory does not prove that the α -particle is preformed but it proves that it behaves as if it is [24]. Figure
5 shows a plot, suitable for the purposes of the theory, of the potential energy between the α -particle and

556
 ERBİL/Turk J Phys

Table 5. Comparison of the transmission coefficients values calculated with the classical and the new method.
Work Transmission coefficient Transmission coefficient
function Tnew from equation (68) Twkb from equation (69)
Metals
W (eV) Electric field ∈ (V /cm) Electric field ∈ (V /cm)
[4] 5 × 106 5 × 107 1 × 107 5 × 106 5 × 107 1 × 107
Na 2.46 5.12448 ×10−23 0.0206 1.43171 ×10−11 1.28112 ×10−23 0.0051 3.57928 ×10−12
Al 4.08 5.07471 ×10−49 0.000052 1.42474 ×10−24 1.26868 ×10−49 0.000013 3.56185 ×10−25
Cu 4.70 1.40147 ×10−60 3.60174 ×10−6 2.36768 ×10−30 3.50368 ×10−61 9.00435 ×10−7 5.91919 ×10−31
Zn 4.31 3.25862 ×10−53 0.000020 1.14169 ×10−26 8.14656 ×10−54 4.91018 ×10−6 2.85422 ×10−27
Ag 4.73 3.68836 ×10−61 3.15165 ×10−6 1.21464 ×10−30 9.22090 ×10−62 7.87911 ×10−7 3.03659 ×10−31
Pt 6.35 4.59212 ×10−95 1.28249 ×10−9 1.35530 ×10−47 1.14803 ×10−95 3.20624 ×10−10 3.38826 ×10−48
Pb 4.14 4.19617 ×10−50 0.000040 4.09691 ×10−25 1.04904 ×10−50 0.000010 1.02423 ×10−25
Fe 4.50 9.20656 ×10−57 8.67486 ×10−6 1.91902 ×10−28 2.30164 ×10−57 2.16872 ×10−6 4.79754 ×10−29

the residual nucleus for various distances between their centers. The horizontal line Eα is the disintegration
energy. There are three regions of interest. In the spherical region r <r1 we are inside the nucleus and speak
of a potential well with a depth of −U0 , where U0 is taken as a positive number. Classically, the α -particle
can move in this region with a kinetic energy Eα +U0 but cannot escape from it. The region r1 < r <r2 forms
a potential barrier because here the potential energy is more than the total available energy Eα . The region
r >r2 is a classically permitted region outside the barrier. From the classical point of view, an α -particle in the
spherical potential well would reverse its motion every time it tried to pass beyond r =r1 of tunneling through
such a barrier. A consistent model for this process assumes that the α -particle is bounded to the nucleus by a
spherical potential well V1 (r) or a spherical effective potential well U1 (r) and that the α -particle is repelled
from the residual nucleus by the central Coulomb potential barrier V2 (r) or the effective central Coulomb
potential barrier U2 (r). The original radioactive nucleus has the charge Ze and the α -particle has the charge
2e. So the Coulomb potential barrier is as follows:

Figure 5. Effective potential function for the α -decay process of the nuclei.

2 (Z − 2) e2 c
V2 (r) = = , [c = 2 (Z − 2) e2 ].
r r
c b ℏ2
Thus, the corresponding effective potential function is obtained as U2 (r) = r + r 2 , [b = 2m l (l + 1) This effective
potential is depicted in Figure 5. There are three domains in this effective potential.
According to the one-body theory, the disintegration constant λ of an alpha emitter is given by λ = f T ,
[T transmission coefficient] . Here, f is the frequency with which the α -particle presents itself at the barrier and
v
T is the probability of transmission through the barrier. The quantity f is roughly of the order of 2(r1 −r3 ) ,

557
 ERBİL/Turk J Phys

where v is the relative speed of the alpha particle inside the nucleus. t1/2 = 0.693/λ is used for the calculations
of nuclear half-life. The relative speed of the alpha particle can be found from its kinetic energy that is equal
to the difference between the disintegration energy of the alpha particle, Eα , and the ground state energy of
the nucleus, E0 , namely:
1
K.E. = mv 2 = [Eα −(− |E0 | )] = Eα +|E0 |. (70)
2
From Eq. (70), the following is obtained:
√ √
2(Eα +|E0 |) 1 m
v= and = .
m v 2(Eα +|E0 |)

If the values of 1/v and λ = f T are substituted into t1/2 = 0.693/λ, then the following formula is obtained:
√
0.693 0.693 2(r1 −r3 ) 1 2(r1 −r3 ) m
t1/2 = = = 0.693 = 0.693 .
λ fT T v T 2(Eα +|E0 |)

To simplify the numerical calculations, this formula can be rewritten in following form:
√
2(r1 −r3 ) mc2 1
t1/2 = 0.693 . (71)
c 2(Eα +|E0 |) T

Here, m = mmnn+m
mα
α
is the reduced mass. ( mn and mα are the mass of the nucleus and α -particle).

7.4.2. Determination of the potential functions

In our numerical calculations, we have taken the harmonic oscillator type central potential as seen in Figure 6.
The central potential parts V1 (r) ,V2 (r),V3 (r) and the central effective potential parts U1 (r),U2 (r),U3 (r) are as
follows:

Figure 6. Central harmonic oscillator type effective potential.

b
V1 (r) = −U0 +ar2 ; U1 (r) = −U0 +ar2 + ; [r3 ≤ r ≤r1 ] [I region]
r2
b
; [r1 ≤ r ≤rm ]
V21 (r) = ar2 ; U21 (r) = ar2 + (II 1 region)
r2
c c b
V22 (r) = ; U22 (r) = + 2 ; [rm ≤ r ≤r2 ] (II 2 region)
r r r
c c b
V3 (r) = ; U3 (r) = + 2 ; [r2 ≤ r ≤ ∞] (III region)
r r r

558
 ERBİL/Turk J Phys

[ ]
1/3
Here, R =R0 (A − 4) +41/3 is the total rayon of the nucleus and alpha particle. From the solution of the
√ √ √ √
(Eα +U0 )+ (Eα +U0 )2 −4ab (Eα +U0 )− (Eα +U0 )2 −4ab
equation Eα = U1 (r)= −U0 +ar2 + rb2 , we can obtain r1 = 2a ; r3 = 2a .
√
2
From the solution of the equation Eα = U22 (r) = , we can obtain r2 = c+ c2E+4bE
c b
r + r2 α
α
. From the
√ √
(E0 +U0 )− (E0 +U0 )2 −4ab
solution of the equation E0 = U1 (r) = −U0 +ar2 + rb2 , we can obtain r11 = 2a ;
√ √
(E0 +U0 )+ (E0 +U0 )2 −4ab 2ℏ2
r12 = 2a ; d0 = r12 −r11 ; d =r2 −r1 . From the solution of the equation E0 = − md 0
2 ,

we obtain the ground state energy as follows:

[ √ ]
√ 1 8aℏ2 √
E0 = ab− U0 + 4ab− −4 abU0 +U0 .
2
2 m

With these grandeurs the coefficients of transmission are written as follow:

√ (∫ rm ∫ r2 √ )
2m √
Twkb = e−2Pwkb ;Pwkb = U 21 (r) −Eα dr+ U22 (r) −Eα dr (72)
ℏ2 r1 rm

2 2
Tnew = = , (73)
cosh [2Kd] + cos [2 {Q2 (r2 ) −Q2 (r1 )}] cosh [2Kd] + cos [2Pnew ]
√ (∫ ∫ )
2m rm √ r2 √
Pnew = U21 (r)dr+ U22 (r)dr , (74)
ℏ2 r1 rm

√
2(r1 −r3 ) mc2 1
twkb
1/2 = 0.693 , (75)
c 2(Eα −E0 ) Twkb

√
2(r1 −r3 ) mc2 1
tnew
1/2 = 0.693 . (76)
c 2(Eα −E0 ) Tnew

The parameters a and rm in potentials can be calculated. To calculate the parameter a , the following equation
is used: U1 (r) = −U0 +ar2 + rb2 = 0 . The roots of this equation are found as follows:
√ √ √ √
U0 + U0 −4ab ′
2
U0 − U0 2 −4ab
r1′ = ;r3 = .
2a 2a

As it can be seen in Figure 6, the r1′ can be taken as sum of the radii of the nucleus and the alpha particle. That
1/3
is, since the radius of the alpha particle is Rα = R0 41/3 and the radius of the nucleus RN = R0 (A − 4) ,
[ ]
1/3
the total radius is Rc = Rα +RN = R0 41/3 +(A − 4) .
′ 2
√ /
Thus it can be taken as follows: (r1 ) = Rc 2 = [U0 + U0 2 −4ab] (2a) . From this equation, the value
U0 Rc 2 −b
of a is obtained as a = RUc02 − Rbc 4 = . To calculate the parameter rm the equation U1 (r) = U22 (r)
Rc 4
[ √ ]
√
3 √
3 ( ) 2/3
is used and the resolution of this equation gives: rm = 2 3aU 0+ 2X
; X = 9a2
c+ 3a3 27ac2 −4U 3 .
62/3 aX 0

Therefore, the potential depends only on R0 and U0 parameters, which simplifies the calculations.

559
 ERBİL/Turk J Phys

7.4.3. Numerical calculations

To calculate the transmission coefficient T (or barrier penetration probability), Eqs. (72) and (73) have been
used and to calculate the half-life values, Eqs. (75) and (76) have been used. Calculations were made for five
nuclei and the results are visible in Tables 6 and 7. In these tables, the half-life values calculated from the new
method and WKB method are compared with the experimental results. For the R0 and U0 parameters, first,
R0 = 1.25f m and U0 = 40M eV values have been taken (Table 6) and it was seen that tnew 1/2 values had better

1/2 values. Later, R0 values in the interval 1.10 − 1.60f m by step 0.01 and U0 values in the
results than twkb
interval 30-50 MeV by step 1 have been changed and the most consistent values with the experimental results
have been taken (Table 7). In all the tables, Ii πi and If πf are the initial and final state spins (parity) of the
nucleus, respectively. lα is the angular momentum of the alpha particle. From these tables, it can be seen that
the new method is more appropriate than the classical method (WKB). Table 7 shows that the new method
gives very good results with the experiment [25–32]. In these tables “exp” is the experimental value of the
half-life. y : year, d : day .

Table 6. Comparison of the half-life values of the classical method with the new method for R0 = 1.25f m and
U0 = 40M eV in the harmonic oscilator potential model.
/ /
Z
AX Eα (MeV) I i πi I f πf lα texp
1/2
twkb
1/2
(75) tnew
1/2
(76) twkb
1/2
texp
1/2
tnew
1/2
texp
1/2
208 P o 5.215 0+ 0+ 0 2.898 y 0.0463 y 5.6950 y 0.16 1.97
84
210 P o 5.407 0+ 0+ 0 138.28 d 1.2767 d 267.865 d 0.009 1.94
84
247 Bk 3− 5−
97 5.889 2 2
2 1380 y 16.6389 y 2521.68 y 0.012 1.83
209 P o 1− 5−
84 4.979 2 2
2 102 y 3.6377 y 96.3285 y 0.036 0.94
5+ 3−
83
237 N p 4.959 2 2
1 2.14 ×106 y 76826.3 y 1.44 ×106 y 0.036 0.67

Table 7. Comparison of the experimental half-life values and the results calculated using the new methods with the
harmonic oscillator type well potential, calculated values a and rm parameters for the parameters r0 and U0 which are
very well matched with the experimental results.
/
Z
A
X Eα (MeV) Ii πi If πf lα r0 (fm) U0 (MeV) texp
1/2 tnew
1/2 tnew
1/2 texp
1/2
208
84 Po 5.215 0+ 0+ 0 1.30 45 2.898 y 2.899 y 1.0004
210
84 Po 5.407 0+ 0+ 0 1.30 47 138.40 d 138.41 d 1.0001
247 1− 5−
97 Bk 4.979 2 2 2 1.24 35 102 y 102 y 1.0010
5+ 3−
84
209
Po 4.959 2 2 1 1.23 49 2.14 × 10 y6
2.15 × 10 y 6
1.0028
237 3− 5−
83 Np 5.889 2 2 2 1.29 43 1380 y 1375 y 0.9964

Here, the general transmission coefficient formula for a potential barrier with an arbitrary form has been
easily calculated without making any approximation. In this calculation, a new method that we developed for
the solution of the radial SE has been used. The transmission coefficient obtained from the new method is given
∫√
by the formula in Eq. (73). In this formula, it could be difficult to calculate the integral U2 (r)dr . If these
calculations cannot be made analytically, they should then be performed by numerical methods.
In the application of the general transmission coefficient formula to the α -decay, three dimensional
harmonic oscillator potential well has been used. The results are given in the tables along with the experimental
values. The tables also contain the “ratio” column for comparison. It can be seen in Table 7 that the ratios for
the most of the nuclei are close to 1 (one). The deviations from 1 are within the experimental error. Hence, it is
said that the results obtained using the new method are more realistic. In the W KB method, the wave function

560
 ERBİL/Turk J Phys

is sinusoidal inside and outside the potential barrier. However, it is not a sinusoidal but an exponential function
while entering into the potential barrier. So, the entering wave into the potential barrier is not sinusoidal and
after the potential barrier, it becomes sinusoidal again. But in the new method, the wave function includes a
sinusoidal multiplier inside the potential barrier, but the sinusoidal multipliers are different inside and outside
the potential barrier. Thus, the wave function has different phases inside and outside the potential barrier,
but it advances everywhere as sinusoidal functions. It can also be said that it is more accurate and realistic.
Besides, the W KB method gives an approximate wave function. In the new method, the wave function is exact
because there is no approximation. That is why the theoretical calculated half-life values match better with
the experimental values. From these, we conclude that the transmission coefficient given in Eq. (73) is more
correct and realistic. By using the new transmission coefficient and half-life formulas, the half-life values of
nuclei can easily be calculated. The general transmission coefficient formula can be used for the other tunneling
phenomenon such as the cold emission from the metals.

8. Scattering theory
8.1. Calculation of the scattering amplitudes
Let us consider a spherical wave progressing at the direction of 0z axis from left to right, and arriving to a
central potential field, sitting at the origin of the 0xyz coordinate system. When we consider scattering, we
shall assume that the interaction between the scattering particle and the scatter can be represented by an
effective central potential energy function U (r) , where r is the relative radial variable. The effective potential
U (r) can include attractive and repulsive parts. Such a central potential is schematically represented in Figure
7. The total energy of the incoming particle beam is E and the incoming particle beam can be represented by
the spherical wave. This progressive spherical wave progress from right to left and arrives to the point r =r1 in
Figure 7. We divide the potential region into four zones and examine the motion of the particle beam in these
four zones.

Figure 7. General schematic representation of scattering by central potential.

Zone I is the region before the effective potential from where free particle comes; zone II, III, and IV
are the effective potential regions where the particle beam is affected. These regions may include attractive and
repulsive potential segments. Zone IV is the region where the particle beam is not able to penetrate because
the potential is infinite. So, the wave function is zero in this region. These are presented with four regions
along with the central potentials in Figure 7. We assume that r =r1 and r = r2 at the interface between zone
I and II, and zone II and III, respectively. The effective potential segments in the zones are represented
as U1 (r) , U2 (r) , and U3 (r) according to the zone numbers. The effective potential U4 (r) can be assumed
infinite since the particle does not enter this region. The central potential can be taken as zero at much far from
zone I so that the particle is free in that region and the effective potential is composed of only the centrifugal

561
 ERBİL/Turk J Phys

term due to the incoming particle angular momentum or spin. The Coulomb interaction potential should also
be added to U1 (r) if it is available. The total energy of the incoming particle and the centrifugal term are
always positive and the latter is less than the former. According to the functions given in Eqs. (52a)–(52f), the
following functions are determined for the zones that are taken into account:
√ √
In zone I :E > 0,U1 (r) > 0 and E >U1 (r) ; k = im1 E= iK; (K =m1 E) ;
∫ √ [ √ ∫ √ ]
2m
G1 (r) = im1 U1 (r)dr = iQ1 (r); m1 = , Q1 (r) = m1 U1 (r)dr .
ℏ2
√
In zone II :E > 0,U2 (r) > 0 and E <U2 (r) , k =m1 E= K
∫ √ [ √ ]
√ 2m
G2 (r) = m1 U2 (r)dr =Q2 (r) ; k = m1 E,m1 = .
ℏ2

In zone III, the conditions E < 0 (This zone corresponds to the bound state. So, the negative energy should
be taken.), U3 (r) < 0 and E >U3 (r) leads to the expressions below:
√ √ [ √ ]
k = im1 E= im1 −|E|= ±K, K =m1 |E| ,
∫ √ ∫ √
G3 (r) = im1 U3 (r)dr = im1 −|U3 (r) |dr =±Q3 (r).

The wave function in zone I should vanish at large distance. Under these circumstances, the radial wave
functions in the three zones can be put in the forms below with regard to the general functions given in Eqs.
(52a)–(52f):
F1 (r) = A1 eiKr−Q1 (r) +e−iKr−Q0 (r) ; [B1 = 1,Q1 (r) > 0,Q0 (r) > 0] , (77a)

F2 (r) = A2 eKr±iQ2 (r) +B2 e−Kr∓iQ2 (r) , (77b)

F3 (r) = A3 e−Kr±iQ3 (r) +B3 eKr∓iQ3 (r) . (77c)

The wave function in zone IV vanishes since the effective potential in this region is infinite. Here, Q0 (r) is
the function resulting from the angular momentum of the incoming particle. This function Q0 (r) can also be
taken as zero because it does not have any contribution to the calculation of the scattering cross-section as it
will be seen soon.
The potential in zone III can also be complex in some cases (usually called the optical potential). If
U3 (r) is the optical potential, it can be written as follows:
√
U3 (r) = |U3 (r)| ei∅ = U31 (r) +iU32 (r) = U31 2 (r)+U32 2 (r)ei∅ . (78a)
[ ] √ √ [ √ ]
U32 (r) U32 (r)
Here, tan (∅) = U31 (r)
, ∅ = arctan U31 (r) ; k = im 1 E= im 1 −|E|= K, k = m 1 |E| ;
∫ √ ∫ √
G3 (r) = im1 U3 (r)dr = im1 − |U3 (r)|dr =±Q3 (r)
∫ √ ∫ √√ ∫ √
|U3 (r) |dr =m1 2 2
U31 2 (r)+U32 2 (r)dr.
4
Q3 (r) = m1 U31 (r)+U32 (r)dr =m1 (78b)

562
 ERBİL/Turk J Phys

In Eqs. (77a)–(78b), the functions Qp (r) can also be written briefly as follows:
∫ √
Qp (r) = m1 |Up (r) |dr, [p = 0, 1, 2, 3].

The terms containing A1 and B1 in the functions in Eqs. (77a)–(77c) give the outgoing and incoming waves,
respectively. We assume that the amplitude of the incoming wave at the boundary of zone I and II is constant.
The second (77b) and the third (77c) functions represent the states of the wave in the effective region of the
′
potential. Applying the continuity conditions on Fp (rj ) and Fp (rj ) , [p, j = 1, 2, 3] functions, the coefficients
Ai and Bi in Eqs. (77a)–(77c) can be determined. These conditions at the boundary points of the three zones
can be written in the following form:

′ ′ ′ ′
F1 (r1 ) = F2 (r1 ) ; F1 (r1 ) = F2 (r1 ) ; F2 (r2 ) = F3 (r2 ) ; F2 (r2 ) = F3 (r2 ) (79)

′ ′
F3 (r3 ) +F3 (r3 ) /K = 0, Qp (rj ) = K, (p, j = 1, 2, 3).

The coefficients A1 ,A2 ,B2 ,A3 ,B3 in the functions given in Eqs. (77a)–(77c) can be found by solving five linear
equations, which can be obtained by using the conditions given in Eq. (79) for each of the functions given in
Eqs. (77a)–(77c).
The essential coefficient for the scattering cross-section is A1, as described below. Therefore, there is no
need to give other coefficients here. The A1 coefficient, which is obtained from four equations, is computed by
taking into account the lower and upper signs in the exponential expressions in Eq. (79) as follows:

8.1.1. The A1 coefficient found using the lower signs:

1 { }
A1 = (1 + i) e−2iKr1 −Q0 (r1 )+Q1 (r1 ) −1+ (2 − i) e[2K(r3 −r1 )+2i[Q2 (r1 )−Q2 (r2 )+Q3 (r2 )−Q3 (r3 )] (80a)
2

8.1.2. The A1 coefficient found using the upper signs:

(1 + i) e2(1−i)Kr1 −Q0 (r1 )+Q1 (r1 )+2i[Q2 (r1 )+Q3 (r2 )]

A1 = . (80b)
(2 + i) e[2Kr3 +2i[Q2 (r2 )+Q3 (r3 )] −e[2Kr1 +2i[Q2 (r1 )+Q3 (r2 )]

The terms with A1 in the functions given in Eqs. (77a)–(77c) representing the outgoing wave from the center
of potential, includes both the scattered by the potential and the incoming wave. Therefore, we must subtract
away the latter to find the amplitude of only the scattered wave. Thus, we obtain the scattering amplitude and
radial wave function representing the scattered wave as:

Cs (r1 ) = A1 e−Q1 (r1 ) −e−Q0 (r1 ) , (scatteringamplitude) (81a)

Fs (r) eiKr [ ] eiKr

Rs (r) = = Cs (r1 ) = A1 e−Q1 (r1 ) −e−Q0 (r1 ) . (81b)
r r r

Eqs. (81a) and (81b) represent the elastic scattering wave by the potential.

563
 ERBİL/Turk J Phys

8.2. Calculation of particle currents

8.2.1. Calculation of the scattered particle current
[ ∗
]
Using Eq. (81b) and the equation Js (r) = 2mi ℏ
Rs ∗ (r) dRdr
s (r)
−Rs (r) dRdr
s (r)
, the current of scattered
particles per unit area at r =r1 point can be found as:

1 ℏK 2
Js (r1 ) = |Cs (r1 ) | . (82)
r1 2 m

8.2.2. Calculation of the incoming particle current

The radial wave function of the incoming beam, passing through zone I and toward zone II, can be written as
−iKr [ ]
Rg (r) = gr = 1r e−iKr−Q0 (r1 ) = Cg (r1 ) e r , Cg (r1 ) = e−Q0 (r1 ) . Here, Cg (r1 ) = e−Q0 (r1 ) represents
F (r)

the amplitude of the incoming wave. The incident current per unite area at r =r1 point can be obtained in the
way that is applied to the current equation:

1 ℏK 2 1 ℏK −2Q0 (r1 )
Jg (r1 ) = |Cg (r1 ) | = 2 e . (83)
r1 2 m r1 m

8.3. Calculations of scattering cross-sections

8.3.1. Calculation of elastic scattering differential cross-section
The probability per unit differential surface of a sphere of radius r1 that an incident particle is scattered into
the differential surface area on the sphere of radius r1 , dS =r1 2 dΩ, [dΩ = sin (θ) dθdϕ] is expressed as the ratio
of the scattered current to the incident current, that is:

dσs dσs Js (r1 ) dσs Js (r1 ) 2

= 2 = → = r1 . (84)
dS r1 dΩ Jg (r1 ) dΩ Jg (r1 )

The differential elastic cross-section can be expressed in a simple form by putting Eqs. (82) and (83) into Eq.
(84):
Cs (r1 ) Cs ∗ (r1 ) 2
2
dσs |Cs (r1 ) | 2
= r1 = 2 r1 . (85)
dΩ Cg (r1 ) Cg ∗ (r1 ) |Cg (r1 ) |

Since the scattering is azimuthally symmetrical, the angle ϕ can be integrated out so that the expression given
in Eqs. (84) and (85) can be written as follows:

2
dσs Js (r1 ) 2 |Cs (r1 ) | 2
= 2π r1 sin (θ) = 2π 2 r1 sin(θ). (86)
dθ Jg (r1 ) |Cg (r1 ) |

The expressions given in Eq. (86) show the elastic scattering differential cross-sections in the angle dθ which is
usually measured experimentally.

8.3.2. Calculation of differential inelastic or reaction (no-elastic) cross-section

Differential reaction (capture of particle, emission of particle, inelastic collision…) cross-section per the solid
angle can be found through the difference between the incoming current and the outgoing current divided by

564
 ERBİL/Turk J Phys

the former. By analogy with Eq. (86), the differential reaction cross-section can be expressed as follows:
[ 2
]
2
dσr [Jg (r1 ) −Js (r1 )] 2 |Cg (r1 ) | −|Cs (r1 ) |
= 2π r1 sin (θ) = 2π 2 r1 2 sin(θ). (87)
dθ Jg (r1 ) |Cg (r1 ) |

8.3.3. Calculation of total cross-sections

The total elastic scattering cross-section is the total probability to be elastic scattered in any direction and it
can be determined through the integral of differential cross-section given in Eq. (85):
∫ ∫ ∫∫
dσs Js (r1 ) 2
σs = dσs = dΩ = r1 sin (θ) dθdϕ,
dΩ Jg (r1 )

2
Js (r1 ) |Cs (r1 ) |
σs = 4πr1 2 = 4πr1 2 2. (88)
Jg (r1 ) |Cg (r1 ) |

By analogy with Eq. (88), the total reaction cross-section can be expressed as follows:

[Jg (r1 ) −Js (r1 )]

σr = 4πr1 2 = 4πr1 2 −σs . (89)
Jg (r1 )

In Eq. (89), it is seen that if Js (r1 ) = Jg (r1 ), then σr = 0, full elastic scattering; if Js (r1 ) >Jg (r1 ) , then
σr < 0, it is taken out of the particle from the target (emission of particle from target) and if Js (r1 ) <Jg (r1 ) ,
then σr > 0, it is captured (absorbed) the particle by the target.
The total scattering cross-section, including all process [elastic plus reaction (all of no-elastic events)]:

Js (r1 ) [Jg (r1 ) −Js (r1 )]

σt = σs +σr = 4πr1 2 +4πr1 2 = 4πr1 2 . (90)
Jg (r1 ) Jg (r1 )

Then the cross-sections σs ,σr ,σt can be expressed through the A1 coefficients given in Eqs. (80a) and (80b).
σs elastic scattering cross-section found using the coefficient (80a) is as follows:

X1 = 3e4Kr1 +5e4Kr3 +2e4Kr1 [cos (2Kr1 ) + sin (2Kr1 )] ,

X2 = −2e2K(r1 +r3 ) [3 cos (2 (Kr1 −Y )) +2 cos (2Y ) + sin (2 (Kr1 −Y )) + sin(2Y ),

σs 1
2
= e−4Kr1 {X1 +X2 } . (91a)
4πr1 2
σs elastic scattering cross-section found using the coefficient (80b) is as follows:

P1 = 3e4Kr1 +5e4Kr3 +2e4Kr1 [cos (2Kr1 ) + sin(2Kr1 )] ,

P2 = −2e2K(r1 +r3 ) [3cos (2Kr1 −2Y ) +2 cos (2Y ) + sin (2Kr1 −2Y ) + sin(2Y )] ,

P3 = e4Kr1 +5e4Kr3 −2e2K(r1 +r3 ) [2 cos (2Y ) + sin(2Y )]

σs P1 +P2
= . (91b)
4πr1 2 P3

565
 ERBİL/Turk J Phys

In both cases, the reaction and total scattering cross-section σr and σt are as follows:

σr σs σt σs σr
=1− and = + = 1 or σt = 4πr1 2 . (91c)
4πr1 2 4πr1 2 4πr1 2 4πr1 2 4πr1 2

In these expressions, Y is given by the following equation:

∫ r1 √ ∫ r2 √
Y =Q2 (r1 ) −Q2 (r2 ) +Q3 (r2 ) −Q3 (r3 ) = m1 |U2 (r)|dr+m1 |U3 (r)|dr. (91d)
r2 r3

The integrals in Eq. (91d) can be solved numerically if the functions Q2 (r) and Q3 (r) cannot be calculated
analytically. It can be seen from these formulas given in Eq. (91a)–(91d) that the scattering cross-sections
(σ s and σr ) depend on the total energy E [withK(E)] , Y integral and the effective radius r1 ,r2 ,r3 of the
scatter potential, separately, but total scattering cross-section σt = σs +σr only depends on the parameter r1
so the energy E . Here, r1 can be considered as impact or collision parameter, classically.

8.4. Examples of the calculation of scattering cross-section

8.4.1. Model potentials, wave functions, and their ingredients
To calculate a scattering cross-section, a model potential should be considered. Here as an example, we consider
the Wood–Saxon shape potential plus the spin-orbit, centrifugal, and Coulomb potentials. The potential zones
are defined in Figure 7 and shown in Figure 8 for these model potentials. Wood–Saxon potential depends on
three parameters (V 0 ,ac ,Rc ) . The calculations have been thus performed with these potentials. The scattering
event affects only the relative motion. In the center of the mass reference frame (CM), the scattering cross-
section of the incoming (incident or projectile) particle σ(Ω) depends on the energy Er = Mt EL / (Mp +Mt ) ,
where Mp and Mt are respectively mass of incident (projectile) and target particles; EL , Laboratory; and
Er , relative energies. The boundary values of the potential zones taken in the calculations: the maximum
potential energy occurs at the distance r2 = Rm = R0 (Ap 1/3 +At 1/3 ) , the addition of the projectile (incident
particle) radius to the target radius, provided that they are spherical, where Ap and At are the mass number
of the projectile and the target, respectively. The zones and rk , (k = 1, 2, 3) values are shown in Figure 8. The
effective potential energy is as follows:

U (r) = Vws (r) +VLSJ (r) +VS (r) + Vc (r) , (92)

, Wood–Saxon potential, VLSJ (r) = − 2Mℏi 2 c2 1r dVdr(r) <L.

2
Vws (r) = − 1+Exp[(r−R
V0
c )/ac ]
⃗ S>,
⃗ spin-orbit potential,

Figure 8. Effective Wood–Saxon potential and Coulomb potential zones used in the calculations.

566
 ERBİL/Turk J Phys

1
⃗ S>
<L. ⃗ = [J (J + 1) −L (L + 1) −S(S + 1)]
2

=ℏ
2
b L(L+1)
VS (r) = r 2 ; [b centrifugal potential,
2Mi ],
[ ]
Vc (r) = Crc , Cc = (Zp e) (Zt e) = Zp Zt e2 potential energy of Coulomb.
Here, [L, S, J] are the relative orbital, spin, and total angular momentum quantum numbers between
the target and projectile, respectively. Mi is the reduced mass of the target and projectile. Rc = R0 At 1/3 ,
( R0 is the parameter).
√ /
If rsc = [Cc + Cc 2 +4bEr ] [2Er ] is the positive root of the equation Er = rb2 + Crc = Vs (r) +Vc (r)
[ √ ]/
and r2 = Rm then we get r1 = r2 +rsc = r2 + Cc + Cc 2 +4bEr [2Er ].Zp and Zt are the charge numbers of
the projectile and the target, respectively. The r3 value is determined by equalizing Vs (r) toEr . Consequently,
r3 ,r2 ,r1 values are obtained as follows:
√
√ ( 1 1
) Cc + Cc 2 +4bEr
r3 = b/Er ; r2 = Rm = R0 Ap 3 +At 3 ; r1 = r2 + .
2Er

The wave functions for zones 1–3 used in the calculation can be expressed in the following ways, respectively:

∫ √ √
ℏ2 L0 (L0 +1)
Q0 (r) = mℏ 2
dr = L0 (L0 +1) ln (r) ,
2Mi r

∫ √ { [√ ]}
Cc b √ √ b+Cc r
Q1 (r) = mℏ + dr = 2mℏ b+Cc r− barctanh ,
r r2 b

∫ √ { [√ ]}
Cc b √ √ b+Cc r
Q2 (r) = mℏ + dr = 2mℏ b+Cc r− barctanh ,
r r2 b

∫ √ ∫ √
Q3 (r) = mℏ |U3 (r) |dr =mℏ |Vws (r) +VLSJ (r) +VS (r) |dr. (93)

√ /
Here, mℏ = 2Mi ℏ and L0 is the angular momentum of the incident particle.
It is seen from Eq. (91c) that even though σs and σr have changed with the parameters V0 and ac ;σt =
σs +σr has not changed for a certain value of R0 . In other words, the total cross-section does not depend on the
parameters V0 and ac and, therefore, does not also depend on the potential. So, R0 parameter can be obtained
from the solution of the equation 4πr1 2 = σt exp as follows (σt exp is experimental total cross-section):

√ [ √ ] √
− π10 Cc + Cc 2 +4bEr +Er 10σt exp
R0 = √ [√ √ ] (94)
20Er π 3 Ap + 3 At

To calculate σs and σr separately, the parameter V0 in the range 20–60 by step 0.0001 and ac in the range
0.40–0.60 by step 0.01 have been changed in the potential until σr cal > 0 and round [ σr exp ] = round [ σrcal ] and
floor [ σr exp ] = floor [ σr cal ]}, during the calculations. So, rough V0 and ac values have been found. Then, with

567
 ERBİL/Turk J Phys

{ }
these approximate values, σs exp = σs cal and σr exp = σr cal system of equations has been solved and the
exact values of parameters V0 and ac have been found. With these V0 and ac parameters, the exact σs and σr
have been recalculated. We have taken in the calculations: e2 = 1.439976M eV f m;Mu = 931.502M eV /c2 ;
ℏc = 197.329M eV f m.

8.4.2. Thermal neutron cross-sections

For example, thermal neutron (E L = 0.025eV ) scattering cross-sections for some targets have been calculated
and compared with the measured values. In calculation, because their angular momentums are zero even
nuclei have been taken as targets. So, the relative angular momentums are those of the projectile, they are
L = 0, S =1/2, J =1/2 for neutron. For the calculations of σs ,σr , and σt , the formulas given in Eq. (84) which
has been obtained by the lower sign functions. The results are compared with the measured total cross-sections
in Table 8. The experimental values have been taken from [25–32]. It is seen that agreements are fairly good
as seen in Table 8. In this table, the first column: target; the second column: (R0 ,V0 , andac ) parameters; the
third column: (Rc ,Rm = r2 ,r1 ); the fourth column: cross-sections; the fifth column: values of calculation; and
the sixth column: values of the experiment.

Table 8. [n(0, 1) + Xn(Z, N )] Thermal neutron cross-section comparisons with those measured.

R0 Rc
Target Cross-sections Calculations Experiment
V0 Rm
Xn(Z, N ) (mb) (mb) (mb)
ac r1
2.29845 2.89586 σs → 3390 3390 ± 12
H (1,2) 21.3275 5.19431 σr → 0.519 0.519 ± 0.007
0.40 5.19431 σt → 3390.52 3390.52
1.86896 4.27885 σs → 4746 4746 ± 2
C (6,12) 35.2934 6.14781 σr → 3.53 3.53 ± 0.07
0.40 6.14781 σt → 4749.53 4749.53
1.5543 3.91659 σs → 3761 3761 ± 6
O (8,16) 41.2421 5.47089 σr → 0.190 0.190 ± 0.019
0.40 5.47089 σt → 3761.19 3761.19
1.02922 3.12533 σs → 1992 1992 ± 6
Si (14,28) 21.3192 4.15456 σr → 177 177 ± 5
0.40 4.15456 σt → 2169 2169
1.1803 4.03655 σs → 3010 3010 ± 8
Ca (20,40) 39.0960 5.21685 σr → 410 410 ± 20
0.40 5.21685 σt → 3420 3420

3
8.4.3. 2 He Cross-sections on some targets in the intermediate energy
The cross-sections σt at three different He-3 energies have been calculated for 5 different targets
40
[ 94 Be, 12 16 28
6 C, 8 O, 14 Si, and 20 Ca]. The calculated cross-sections have been compared with those measured
taken from [32,38–40]. The comparisons are made in the way that is described above and given in Table 9.
There is no need for relative angular momentums here.

568
 ERBİL/Turk J Phys

Table 9. [ He[2, 3] + Xn[Z, N ] Total cross-section comparison with those measured.

Energy r1 (f m)
Xn[Z, N ] (projectile) R0 (f m) Rc (f m) Rm = r2 (f m) (impact Calculation Experiment
(target) EL (MeV) parameter) σt (mb) σt (mb)
96.4 0.673279 1.40048 2.37051 2.53101 805 805 ± 30
Be[4,9] 137.8 0.623868 1.29770 2.19747 2.30905 670 670 ± 30
167.3 0.607056 1.26273 2.13825 2.23016 625 625 ± 30
96.4 0.620244 1.42000 2.31455 2.53885 810 810 ± 40
C[6,12] 137.8 0.594922 1.36203 2.22006 2.37697 710 710 ± 30
167.3 0.572480 1.31065 2.13631 2.26556 645 645 ± 35
96.4 0.631332 1.59086 2.50140 2.78546 975 975 ± 35
O[8,16] 137.8 0.606261 1.52768 2.40206 2.60079 850 850 ± 50
167.3 0.595506 1.50058 2.35945 2.52313 800 800 ± 25
96.4 0.600731 1.82417 2.69058 3.15392 1250 1250 ± 65
Si[14,28] 137.8 0.603057 1.83124 2.70099 3.02513 1150 1150 ± 70
167.3 0.590377 1.79273 2.64420 2.91119 1065 1065 ± 40
96.4 0.544438 1.86195 2.64717 3.28976 1360 1360 ± 90
Ca[20,40] 137.8 0.563942 1.92866 2.74200 3.19154 1280 1280 ± 85
167.3 0.565988 1.93565 2.75195 3.12222 1225 1225 ± 75

The calculation of cross-sections through solution of the radial SE by the partial wave expansion is
very difficult. In many cases, some approximations are needed for these kinds of solutions. In the present
study, firstly, differential elastic scattering, inelastic (or reaction) scattering, and total cross-sections have been
calculated without using any approximation. These calculations have been performed using a simple method,
improved for the solution of the radial SE, for an incident particle being in a central field of any form. We
have obtained the general formulas of the scattering amplitudes and elastic, inelastic (no-elastic) and total
scattering cross-sections. Secondly, we have made some applications. In these applications, the potentials have
been assumed to have Saxon–Woods shape plus spin-orbit interaction, centrifugal and Coulomb potentials.
With these potentials, first, for the thermal neutrons; the elastic, inelastic (neutron radiative capture), and
total scattering cross-sections of different targets have been calculated. Then, total scattering cross-sections
for 2 3 He particles of three different energies on 5 targets have been calculated. The calculated results have
been compared with the experimental results. The results calculated have given satisfactory agreement with
the available experimental results. More cross-section calculations can be found in [33].
The calculations have also shown that the total cross-sections depend on the mean potential range. Thus,
it is also proved that the total cross-sections can be calculated easily using even very complex potentials. The
same calculations have also been performed using optical potentials, but the results have not been included here
due to getting the same scattering distance and cross-sections. The use of two parameters is seen to be enough
in the agreement of the calculated results with the measured results, whereas this agreement is ensured using
more parameters in the partial wave expansion method.

9. Conclusion and some explanations

We have found a simple procedure for the general solution of the time-independent SE in one dimension without
using any infinite series or other approximations. The wave functions, which are always periodic in the bound

569
 ERBİL/Turk J Phys

states, are given in Eqs. (52a)–(57). In our procedure of solution, there are two difficulties: one is to solve the
√ ∫
equation E = U (x) to find the energy; and the other is to integrate U (x) namely, to calculate U (x) dx to
find the exact normalized wave function. If these calculations cannot be done analytically, then it should be
done by numerical methods. To find the energy values, there is no need to calculate this integral, it is sufficient
to find the classical turning points by solving the equation E = U (x). Thus, there is no need to know the wave
functions to find the energy values of the states; it is enough to know only the potential energy function. The
SE has been solved for a particle in many potential wells and their energies and normalized wave functions were
calculated as examples.
Using a simple procedure, the solution of the radial SE for spherically symmetric potentials without
using any infinite series has been achieved in this study. The wave functions which are always periodicals
are given in Eqs. (52a)–(57). By using this procedure, the radial SE has been solved for a particle found in
many spherically symmetric central potential wells and two different solutions have been found. One of them is
symmetric function and the other is antisymmetric function. From these expressions, it is observed that these
functions are periodic and they are similar to each other in form for all potential wells. This simple solution
was applied to scattering and tunneling theories and it yielded good results.
The solution that we propose is a general solution. The points of view supporting the method we presented
here is more realistic which are as follows:
As it is known, the SE is a second order differential equation with variable coefficients. The solutions of
such equations are based generally on series method and special functions (Hermit, Bessel…) in quantum physics.
In the expanding of power series, the consecutive relations between the coefficients of the series are found. Some
approaches are taken to make the series convergent and by using them, the energy values and the wave functions
are determined. In all of these solutions, one or more approximations are used. Some other methods such as
perturbation, WKB, and variation are also applied for solving the SE and approximate solutions are obtained.
However, the solution that is proposed here is neither based on series methods nor special functions, and no
approximation is used. We think that these solutions, which do not have any approach, are more realistic.
It is seen that the functions found with the known methods and with the present method are different in
form. However, when we do numerical calculations, the results are not very different from each other and are
consistent with their well-known values. For example, for the one dimensional harmonic oscillator, our result
and the result obtained by the known methods are very close to each other.
The SE, the fundamental equation of the quantum mechanics, is also known as the Schrödinger wave
equation. In physics, the harmonic waves are represented with periodic functions e.g., sin and cos. However,
most of the known solutions of the SE are in the form of polynomials. Especially the solution of the harmonic
oscillator should be periodic function. However, the known solution is polynomial. As it is seen in the functions
given in Eqs. (5.1.3) and (5.1.4), our solution gives sin and cos functions, which makes it more realistic.
It is said that the quantum mechanics includes classical mechanics. Hence, the results of the classical
mechanics should be obtained from the results found in quantum mechanics. As it is seen from the functions
given in Eq. (5.1.4), it is very difficult to obtain sin and cos functions from polynomials. However, the classical
solutions could be easily obtained if Ψ (x) is taken very small in the functions given in (5.1.3).
In quantum mechanics textbooks, it is said that all infinitely high potential wells are similar to each
other. But when we look at their solutions, we see that all of the solutions are different in form; whereas our
solutions are similar to each other in form for all such potential wells.
In quantum mechanics, if there is no exact solution, sometimes, the variation principle is used to find the

570
 ERBİL/Turk J Phys

ground state energy. The calculations of variation are made with a trial function, and it is seen that the results
are not very dependent on the trial function. However, the wave functions are not necessary to find the ground
or excited state energy in our solutions. It is sufficient to know the classical turning points of the potential
function.
Special conditions are not required in order to find the energy values and wave functions. The continuity
of the wave functions and their derivatives at the classical turning points are enough.
The application of our procedure is very easy. Most of the problems that could not be solved analytically
with the known methods can easily be solved using our procedure. A complete solution of the SE used in all
branches of physics has been made. A problem that has been worked on by many theoretical physicists has been
solved. I think that this solution is very useful and helpful for those physicists interested in quantum mechanics
and applications.

Acknowledgements
I would like to express my sincere gratitude to my wife Özel and my daughters Işıl and Beril for their support
and patience during my study and for their help in editing.

References

[1] Büyükkılıç, F.; Eğrifes, H.; Demirhan, D. Theor. Chem. Acc. 1997, 98, 192-196.
[2] Eğrifes, H.; Demirhan, D.; Büyükkılıç, F. Phys. Scripta 1999, 59, 90-94.
[3] Eğrifes, H. ; Demirhan, D.; Büyükkılıç, F. Phys. Scripta 1999, 60, 195-198.
[4] Eğrifes, H. ; Demirhan, D.; Büyükkılıç, F. Phys. Lett. A 2000, 275, 229-237.
[5] Powell, J. L.; Crasemann, B. Quantum Mechanics; Addison-Wesley Publising Company Inc.: USA, 1965.
[6] Schwabl, F. Quantum Mechanics, 2nd ed.; Springer-Verlag Berlin Heidelberg: New York, NY, USA, 1995.
[7] Griffiths, D. J. Introduction to Quantum Mechanics; Pearson Prentice Hall: NJ, USA, 1995.
[8] Erbil, H. H. The International Review of Physics Vol. 1 2007, 4, 197-213.
[9] Greiner, W. Quantum Mechanics, 2nd ed.; Springer-Verlag Berlin Heidelberg: New York, NY, USA , 1993.
[10] Cooper, F.; Khare, A.; Sukhatme, U. Supersymmetry in Quantum Mechanics; World Scientific, 2001.
[11] Gönül, B.; Zorba, İ. Phys. Lett. A 2000, 269, 83-88.
[12] Gonul, B.; Ozer, O.; Koçak, M.; Tutcu, D.; Cançelik, Y. J. Phys. A: Math. Gen. 2001, 34, 8271-8279.
[13] Griffiths, J. Introduction to Quantum Mechanics; Prentice Hall, 1994.
[14] Kasap, E.; Gönül, B.; Şimşek, M. Chem. Phys. Lett. 1990, 172, 499-502.
[15] Landau, L. D.; Lifstits, E. M. Quantum Mechanics; Pergamon Press, 1958.
[16] Hassani, S. Foundations of Mathematical Physics; Prentice-Hall International Inc.: USA, 1991.
[17] Ayant, Y.; Borg, M. Fonctions Speciales; Dunod: Paris, France, 1971.
[18] Messiah, A. Quantum Mechanics; North Holland, 1958.
[19] Erbil, H. H. The International Review of Physics Vol. 2 2008, 1, 1-10.
[20] Pal, M. K. The Theory of Nuclear Structure; Scientific and Academic Editions, 1983.
[21] Erbil, H. H,; Ertik. H.; Şirin, H.; Kunduracı, T. The International Review of Physics 2012, 6, 359-366.
[22] Volya, A. http//www.volya.net.

571
 ERBİL/Turk J Phys

[23] Erbil, H. H. The International Review of Physics 2013, 7, 376-384.

[24] Krane, K. S. Introductory Nuclear Physics; John-Wiley & Sons: New York, NY, USA, 1988.
[25] Firestone, R. B. Table of Isotopes; Wiley Interscience, 1996.
[26] Harmatz, B. Nucl. Data Sheets 1981, 34, 1-807.
[27] Schmorak, M. R. Nucl. Data Sheets 1990, 45, 1-207.
[28] Rytz, A. At. Data Nucl. Data Tables 1991, 47, 205-239.
[29] Rytz, A. At. Data Nucl. Data Tables 1979, 23, 507-598.
[30] Rytz, A. At. Data Nucl. Data Tables 1973, 12, 407-498.
[31] Schmorak, M. R. Nucl. Data Sheets, 1981, 32, 1-209.
[32] Avrigeanu, M.; Von Oertzen W.; Fischer, U.; Avrigeanu, V. Nucl. Phys. A 2005, 759, 327-341.
[33] Erbil, H. H.; Selvi, A. S.; Kunduracı, T. The International Review of Physics 2013, 7, 230-238.
[34] Erbil, H. H. The International Review of Physics 2012, 6, 144-152.
[35] Erbil, H. H. The International Review of Physics 2012, 6, 439-447.
[36] Erbil, H. H,; Kunduracı, T. The International Review of Physics, 2012, 6, 269-278.
[37] Erbil, H. H. Kuantum Fiziği; Ege University: İzmir, Turkey, 2001.
[38] Mughabghab, S. F.; Divadeenam, M.; Holden, N. E. Neutron Cross Sections; Academic Press Inc.: New York, NY,
USA, 1981.
[39] Mughabghab, S. F. Thermal Neutron Capture Cross Section, Resonance Integrals and G-Factors, International
Nuclear Data Committee, Vienna, 2003.
[40] Mughabghab, S. F.; Divadeenam, M.; Holden, N. E. Neutron Cross Sections 1/A; Academic Press Inc.: London,
UK, 1981.

572