EDUCATION Revista Mexicana de Fı́sica E 17 (1) 1–5 JANUARY–JUNE 2020

An alternative solution to the radial quantum condition for the hydrogen atom
S. Álvareza , R. Acostab , and R. de Cossa
a
Departamento de Fı́sica Aplicada, Centro de Investigación y de Estudios Avanzados del IPN,
Apartado Postal 73 Cordemex 97310 Mérida, Yucatán, México.
e-mail: santiago.alvarez@cinvestav.mx; romeo.decoss@cinvestav.mx
b
Facultad de Ciencias Quı́micas,
Universidad Central de Ecuador, Quito, Ecuador.
e-mail: reacosta@uce.edu.ec
Received 29 January 2019; accepted 1 July 2019
H
The quantization method applied to the hydrogen atom involves the solution of the phase integral pr dr = nr h, which was named by
Sommerfeld as the radial quantum condition and it was solved using complex integration. In this work, we present an alternative solution to
the radial quantum condition using real variable integration methods as an accessible way for students of introductory quantum mechanics
courses. In addition, we show that in the Sommerfeld model the degeneracy of the energy levels is related to the geometric properties of the
ellipse describing the electron motion around the nucleus.

Keywords: Hydrogen atom; elliptical orbits; phase integral.

H
La aplicación del método de cuantización al átomo de hidrógeno implica resolver la integral de fase pr dr = nr h, la cual fue denominada
por Sommerfeld como la condición de cuantización radial y fue resuelta empleando técnicas de integración de variable compleja. En
este trabajo, presentamos una solución alternativa para la condición de cuantización radial, en la cual se utilizan únicamente métodos
de integración de variable real como una forma accesible para los estudiantes de cursos introductorios de mecánica cuántica. Además,
mostramos que dentro del modelo de Sommerfeld, la degeneración que presentan los niveles de energı́a está relacionada con las propiedades
geométricas de la trayectoria elı́ptica del electrón alrededor del núcleo.

Descriptores: Átomo de hidrógeno; órbitas elı́pticas; integral de fase.

PACS: 03.65.Sq; 03.65.Vf DOI: https://doi.org/10.31349/RevMexFisE.17.1

1. Introduction where r represents the distance between the particles. Thus,

the total energy E = T + V is given by
One of the pillars of the atomic structure was established by
Bohr, by introducing the concept of quantization of circular m³ 2 ´ Ze2
E= ṙ + r2 φ̇2 − , (4)
orbits for the electron around a fixed nucleus and developing 2 r
a theory, which was in quantitative agreement with the hydro-
which in terms of pr = mṙ and pφ = mr2 φ̇ can be rewritten
gen spectrum data [1]. Subsequently, Sommerfeld extended
as à !
Bohr’s theory by assuming the existence of elliptical orbits in 2
1 pφ Ze2
an attempt to explain the splitting of the spectral lines in the E= p2r + 2 − , (5)
2m r r
hydrogen spectrum [2–4]. The hydrogen atom was studied
by Sommerfeld as a planetary system according to the for- where pr and pφ are the radial and angular momentum, re-
mulation of Kepler’s law in the form: The electron moves in spectively.
an ellipse, in one focus of which the nucleus is situated. Con- The quantization for the phase-space was conducted over
sequently, the motion in the elliptic path represents a prob- many years before the Sommerfeld work [2–4]. First, Planck
lem of two degrees of freedom. For the present analysis, will suggested the rule defined by
consider the polar coordinates measured from the nucleus, Z Z
namely azimuth φ and the radius r (see Fig. 1). Thus, the dp dq = h, (6)
element of orbit ds of the electron is given by
where q is a generalized coordinate and p is the correspond-
ds2 = dr2 + r2 dφ2 , (1) ing conjugated momentum. At the same time Sommerfeld
proposed that for every molecular process, the exchange
with kinetic energy, T : quantity of action is
m³ 2 ´
Zτ
T = ṙ + r2 φ̇2 , (2) h
2 L dt = , (7)
2π
and potential energy, V (r), in CGS is: 0

Ze2 where τ is the duration of the process and L is the La-

V =− , (3) grangian. It is interesting to mention that Sommerfeld and
r
2 S. ÁLVAREZ, R. ACOSTA, AND R. DE COSS

other involved in the development of the old quantum the-

ory were all experienced in the Hamiltonian formulation of
classical dynamics. Therefore, it was natural that they ex-
press the generalization of the quantum conditions in that lan-
guage [2–4]. Hence, Sommerfeld attempted to formulate the
idea of quantization in a universal way in order to be applied
to any classically finite motion. Thus, the Sommerfeld pro-
posal states that among all possible motions are allowed, only
those that satisfies the condition,
Z
pk dqk = nk h, (8) F IGURE 1. Geometrical representation of the hydrogen atom.

where q and p refer to a pair of canonical variables and the The Eq. (10) has to be restated in terms of the orbital
integration is over a full period of classical motion. Conse- equation of the ellipse. From Fig. 1 and according to elemen-
quently, by setting nk = 1, 2, . . . in (8) we fix the first, sec- tary analytical geometry [12], we have that:
ond, etc. quantized phase-orbit of the k-th freedom degree.
OP = a
Interestingly, this quantization rule was proposed indepen-
dently by Wilson [5], Ishiwara [6], and Sommerfeld [7]. OQ = b
The Wilson-Sommerfeld quantization conditions applied
to elliptical orbits wich takes the form [8–10]: NP = a(1 + ²)
I NM = a(1 − ²)
pφ dφ = nφ h, (9)
NO = a², (14)
I
pr dr = nr h, (10) where a and b are the major and minor axis respectively, and
ε is the ellipse eccentricity (0 < ε < 1), which are related by
where h is the Planck constant. The number nφ is the az- p
imuthal quantum number and it can take the values: 1, 2, 3, b = a 1 − ²2 . (15)
· · · ; while nr is the radial quantum number and it can take the To obtain the equation of the electron orbit around the nu-
values: 0, 1, 2, · · · . These integrals have to be computed for cleus, we will start by analyzing Fig. 1. We denote as d the
one complete period. The Eqs. (9) and (10) were called the distance between the point N and the point where the electron
azimuthal quantum condition and the radial quantum condi- is located (see Fig. 1), so d can be expressed geometrically
tion [13], respectively. as
From the subject of central forces in classical mechan- 2
d2 = r2 sin2 φ + (2aε + r cos φ) , (16)
ics, we know that for a particle of mass m moving in a po-
tential V (r), the angular momentum L is a constant of mo- and using the trigonometric identity sin2 φ + cos2 φ = 1, the
tion [11]. This fact can be easily demonstrated by taking the last equation is reduced to:
time derivative of the angular momentum vector, L = r × p,
d2 = r2 + 4aε (aε + r cos φ) . (17)
resulting:
dL But, by construccion, the left and right focal radii obey the
= r × F. (11)
dt relationship d + r = 2a. Therefore,
But for the case of a central potential, the general form of the 2
force is given by (2a − r) = r2 + 4aε (aε + r cos φ) . (18)
r
F = f (r) , (12) Thus, the parametric formula of the ellipse in polar coor-
r
dinates is written as:
where f (r) = −dV /dr. Thus, substituting this expression µ ¶
for F in (11) we find that dL/dt = 0, which means that for 1 1 1 + ε cos φ
= . (19)
the case of a particle moving in a force field derived from a r a 1 − ε2
central potential V (r), the angular momentum is a constant
Then, before solving the integral in the radial quantum
of motion. Hence, according to the law of conservation of
condition (10), we notes that
angular momentum pφ = mr2 φ̇ = L is constant [11]. Thus,
the integral (9) becomes dr dr dφ
pr = m =m
dt dφ dt
pφ = nφ ~, (13) µ ¶
1 dφ dr L dr
= 2 mr2 = 2 , (20)
where ~ = h/2π. r dt dφ r dφ

Rev. Mex. Fı́s. E 17 (1) 1–5

 AN ALTERNATIVE SOLUTION TO THE RADIAL QUANTUM CONDITION FOR THE HYDROGEN ATOM 3

so that
I Z2π µ ¶2
L dr
pr dr = dφ = nr h.. (21)
r2 dφ
0

From the parametric formula of the ellipse (19), we have that

µ ¶2
1 dr ε2 sin2 φ
= 2. (22)
r2 dφ (1 + ε cos φ)
Thus, we find that the integral (10) becomes
Z2π p
2 sin2 φ F IGURE 2. Plot of g(u) for ε = 1/2.
Lε 2 dφ = nr h. (23)
(1 + ε cos φ)
0 Then
1 − u2
The constant L, is also so-called areal constant and deter- cos φ = , (28)
1 + u2
mines the size of the ellipse. Thus, the problem of quanti-
zation for the elliptical orbit has been reduced to solve the and
integral, I, in Eq. (23): 2 du
dφ = . (29)
1 + u2
Z2π
sin2 φ Because the function (27) presents a discontinuity at φ = π,
I≡ 2 dφ. (24)
(1 + ε cos φ) the limits corresponding to the change of variable in the Eq.
0 (26) are as follows:
This integral was evaluated by Sommerfeld using complex
Z
+∞
integration on a closed contour [13]. We would like to men- 2(1 − u2 )du
tion that details of this solution and other alternative methods, −εI =
(1 + u2 )[(1 − ε)u2 + (1 + ε)]
based also on complex variable techniques, can be found in 0
reference [14]. In the next section, we show an alternative Z0
solution which does not require the use of the complex vari- 2(1 − u2 )du
+ . (30)
able. (1 + u2 )[(1 − ε)u2 + (1 + ε)]
−∞

2. The radial quantum condition solved using The integrand in the equation is a function of u that we will
real variable integration methods denote by g(u). As we can see g(u) = g(−u), that means
that g is an even function in u. In order to observe thepparity
In this section we present a solution of the integral (24) using of g(u), in Fig. 2 it is shown the plot of g(u) for ε = 1/2.
real variable integration methods. We start integrating (24) Thus, as a result of the parity of g(u) we have that:
by parts [15], obtaining
Z
+∞
Z2π 2(1 − u2 )du
2
sin φ −εI = 2 . (31)
I= (1 + u2 )[(1 − ε)u2 + (1 + ε)]
2 dφ
(1 + ε cos φ) 0
0

Z2π Now, separating in partial fractions we obtain:

1 cos φ
=− dφ. (25) Z
+∞ Z
+∞
ε 1 + ε cos φ 2 2
ε du ε du
0 −εI = 2 −2 . (32)
1 + u2 (1 − ε)u2 + (1 + ε)
Now, we take the right hand integral and separate the interval 0 0
[0, 2π] in two parts [0, π] and [π, 2π]. Then we have:
Manipulating algebraically we arrive to:
Zπ Z2π
cos φ cos φ " Z
+∞
−εI = dφ + dφ. (26) 2 du 2
1 + ε cos φ 1 + ε cos φ −εI = 2 − √
0 π ε 1+u 2
ε 1 − ε2
0
These integrals are rational functions of cos φ, then we can q
Z
+∞ 1−ε #
1+ε du
performe the following change of variable [16]:
× ³q ´2 . (33)
φ 1−ε
u = tan . (27) 0 1+ε u + 1
2

Rev. Mex. Fı́s. E 17 (1) 1–5

4 S. ÁLVAREZ, R. ACOSTA, AND R. DE COSS

Integrating the two terms of the right side, we obtain: but L = pφ = nφ ~, then
"
2 m Z 2 e4
−εI = 2 lim arctan u E=− . (43)
u→+∞ ε 2~ (nr + nφ )2
2

Ãr !# This result shows that the energy levels of the elliptical orbits
2 1−ε are degenerate because there are two quantum numbers (nr
− √ arctan u . (34)
ε 1 − ε2 1+ε and nφ ) in the denominator. In this way, more than one set
of values for nr and nφ are producing the same value for the
Finally, we find that:
energy. From here, the total or principal quantum number n
µ ¶
2π 1 rises and corresponds to the sum of the radial and azimuthal
I= √ −1 . (35) quantum numbers:
ε2 1 − ε2
Thus, with (35) in (23) we obtain: n ≡ nr + nφ . (44)
µ ¶
1 Then the expression of the total energy in terms of the prin-
2πL √ − 1 = nr h. (36)
1 − ε2 cipal quantum number is:
This result is the same expression that was obtained by Som- m Z 2 e4
E=− . (45)
merfeld solving (24) using complex variable methods [13]. 2~2 n2
To conclude, a relationship between the quantum numbers
3. Quantization of energy orbits and the geometric properties of ellipse can be obtained com-
bining (13), (15) and (36). From that we find
After solving the phase-integral of the radial quantum con-
a nr + nφ
dition, in this section, we are obtaining the expression for = , (46)
the quantization of energy. We start rewriting (4) using b nφ
φ̇ = L2 /mr2 , to obtain which can be rewritten as:
a n
p2r L2 Ze2 = . (47)
E(r, pr ) = + 2
− . (37) b nφ
2m 2mr r
From Fig. 1 we can see that when the electron is at the points This result shows that the ratio between semi-major (a) and
P and M, the radial component of the momentum vanishes semi-minor (b) axes of the elliptic orbit is equal to the ratio
(pr = 0), because there are turning points [11]. Notice that between the principal and azimuthal quantum numbers.
the distance r at P and M corresponds to rP = a(1 − ε) and Finally, we want to mention that the problem of treating
rM = a(1 + ε), respectively. Thus, evaluating (37) at the a particle in an elliptical orbit as a de Broglie wave conducts
points P and M, we obtain: to the original quantum conditions proposed by Wilson [5],
Ishiwara [6], and Sommerfeld [7]. A didactic exposition of
L2 Ze2 this topic can be found in reference [17].
EP = − , (38)
2ma2 (1 − ε)2 a(1 − ε)
L2 Ze2 4. Final remarks
EM = − . (39)
2ma2 (1 + ε)2 a(1 + ε)
In summary, we presented an alternative solution to the ra-
It is important to remember that the total energy is a constant dial quantum condition, based on the Sommerfeld model of
of motion, consequently EP = EM = E. Hence, in order to the hydrogen atom, using real variable integration methods,
find an expression for E in terms of L, we take the difference which does not require the use of complex variable tech-
(1 + ε)EM − (1 − ε)EP and we obtain: niques. Adittionally, we used a simple geometric approach
L2 to show that the degeneracy of energy levels in the Sommer-
E=− . (40) feld model is related to the properties of the ellipse describing
2ma2 (1 − ε2 )
the electron motion around the nucleus. Both subjects can be
On the other hand, to find an expression for E in terms of Z, useful for the discussion of the Sommerfeld model in an in-
we take the difference (1 + ε)2 EM − (1 − ε)2 EN to obtain troductory quantum mechanics courses.
Ze2
E=− . (41)
2a Acknowledgments
Now, combaining (36), (40) and (41) the following relation
One of the authors, S. Álvarez acknowledges a student fel-
for the energy E can be get
lowship from the Consejo Nacional de Ciencia y Tecnologı́a
m Z 2 e4 (CONACyT, México). This work was partially supported by
E=− , (42) CONACyT-México under Grant No. 288344.
2 (nr ~ + L)2

Rev. Mex. Fı́s. E 17 (1) 1–5

 AN ALTERNATIVE SOLUTION TO THE RADIAL QUANTUM CONDITION FOR THE HYDROGEN ATOM
1. N. Bohr, Philos. Mag. 27 (1914) 506.
2. M. Jammer, The Conceptual Development of Quantum Me-
chanics (Mc Graw-Hill, New York, 1966), pp. 89-93.
3. J. Mehra and H. Rechenberg, The Historical Develpment of
Quantum Theory (Springer-Verlag, 1982), pp. 200-221.
4. M. Taketani, The Formation and Logic of Quantum Mechan-
ics: Vol. II. The Way to Quantum Mechanics (World Scientific,
2001), p.103-119.
5. W. Wilson, Philos. Mag. 29 (1915) 795.
6. J. Ishiwara, Proc. Tokyo Math. Phys. Soc. 8 (1915) 173.
7. A. Sommmerfeld, Ann. Phys. 51 (1916) 1.
8. R.M. Eisberg, Fundamentals of Modern Physics (John Wiley
and Sons, Inc., 1961), pp. 128-132.
9. L. Pauling and E. Bright Wilson Jr., Introduction to Quantum
Mechanics with Applications to Chemistry (Dover, New York,
1985), pag. 41.
Rev. Mex. Fı́s. E 17 (1) 1–5
5
10. L. De la Peña, Introducción a la Mecánica Cuántica (Fondo de
Cultura Económica, 2012), pp. 22-24.
11. J.B. Marion and S.T.Thorton, Classical Dynamics of Particles
and Systems (Harcount Collegue Publishers, 1995), pp. 291-
309
12. C.H. Lehman, Geometrı́a Analı́tica (Limusa, México, 1989),
pp. 173-190.
13. A. Sommerfeld, Atomic Structure and Spectral Lines (Methuen
& Co. Ltd., London, 1923).
14. L.A. de Castro, C.A. Brasil, and R.J. Napolitano, Rev. Bras.
Ensino Fis. 38 (2016) 3318.
15. J.T. Cushing, Am. J. Phys. 38 (1970) 1149.
16. G.H. Hardy, A Course of Pure Mathematics (Cambridge Uni-
versity Press, 1976).
17. T. Deeney and C. O’Sullivan, Am. J. Phys. 82 (2014) 883.