A simpler graphical solution and an approximate formula for energy eigenvalues in

finite square quantum wells

F. M. S. Lima

Citation: American Journal of Physics 88, 1019 (2020); doi: 10.1119/10.0001694

View online: https://doi.org/10.1119/10.0001694
View Table of Contents: https://aapt.scitation.org/toc/ajp/88/11
Published by the American Association of Physics Teachers

ARTICLES YOU MAY BE INTERESTED IN

TM
A LEGO dynamic force “macroscope”
American Journal of Physics 88, 906 (2020); https://doi.org/10.1119/10.0001700

A completely algebraic solution of the simple harmonic oscillator

American Journal of Physics 88, 976 (2020); https://doi.org/10.1119/10.0001702

Comment on “The negative flow of probability” [Am. J. Phys. 88, 325–333 (2020)]
American Journal of Physics 88, 1023 (2020); https://doi.org/10.1119/10.0001773

Schrödinger's original struggles with a complex wave function

American Journal of Physics 88, 433 (2020); https://doi.org/10.1119/10.0000852

Physical pendulum model: Fractional differential equation and memory effects

American Journal of Physics 88, 962 (2020); https://doi.org/10.1119/10.0001660

Exploring delay dynamics with a programmable electronic delay circuit

American Journal of Physics 88, 1006 (2020); https://doi.org/10.1119/10.0001695
NOTES AND DISCUSSIONS

A simpler graphical solution and an approximate formula for energy

eigenvalues in finite square quantum wells
F. M. S. Limaa)
Instituto de Fısica, Universidade de Brasılia, P.O. Box 04455, 70919-970 Brasılia, DF, Brazil
(Received 11 January 2020; accepted 17 July 2020)
https://doi.org/10.1119/10.0001694
The computation of allowed energy levels for a particle bounded to a finite square well potential is
ubiquitous in modern physics and introductory quantum mechanics textbooks. For stationary
bound states, the matching conditions for the wave functions lead to a pair of transcendental
equations whose roots correspond to the energy eigenvalues. However, the graphical solutions
available do not make clear the dependence of the energies on the well potential parameters. In this
note, I present a simpler graphical solution involving only one dimensionless parameter that
determines a straight-line crossing identical sinusoidal curves. I then reduce this solution to a
single cosine curve, and from a three-point interpolation, I derive an approximate formula for all
energy levels valid for any square quantum well and that demands only a pocket calculator. VC 2020
American Association of Physics Teachers.
https://doi.org/10.1119/10.0001694

In modern physics or introductory quantum mechanics solid-state and nanophysics (e.g., in semiconductor hetero-
textbooks, we find only a few quantum systems whose bound junctions, ultra-thin metallic films, and atomic force micros-
state energy levels are easy to be determined. One of the first copy), it is pedagogically valuable to search for a simple
examples of exactly solvable quantum states is that of a par- approximation scheme that yields correct results for both
ticle of mass m confined in an infinitely deep square quantum infinitely deep and very shallow SQWs.
well (SQW). On choosing x ¼ 0 at the center of the well For a better comparison to infinite SQWs, let us choose
potential, whose width is L, the potential energy is VðxÞ ¼ 0 the potential of the finite SQW as VðxÞ ¼ 0 for ‘ < x
inside it (i.e., for L=2 < x < þL=2) and 1 otherwise. For < þ‘, where ‘  L=2, and VðxÞ ¼ V0 > 0 otherwise. For
this potential, the wave functions for stationary states are this potential, the one-dimensional, time-independent
wn ðxÞ ¼ A cos ðkn xÞ; n odd, and wn ðxÞ ¼ B sinðkn xÞ;n even.1 Schr€odinger equation can be written as
From the boundary condition wn ð6L=2Þ ¼ 0— i.e., the van-
ishing of the de Broglie wave function at the impenetrable w00 þ k2 w ¼ 0 ; ‘ < x < þ‘; (2a)
walls, forming standing waves—it follows that kn ¼ n p=L,
so the allowed energy levels are1 w00  q2 w ¼ 0 ; jxj > ‘; (2b)
pffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 kn2
h p2 h2 where k  2mE =h and q  2mðV0  EÞ =h. Since 0 < E
n ¼ ¼ n2 ¼ n2 1 : (1) < V0 for bound states, both k and q are positive real num-
2m 2mL2
bers, which makes these second-order differential equations
However, this naive model is an oversimplification, which is with constant coefficients solvable in terms of trigonometric
never realized in practice. A more realistic model is the finite and exponential functions. The boundary condition wð61Þ
SQW, in which a particle moves between walls with a finite ¼ 0 reduces the general solution to
height V0, penetrating into the walls (i.e., the classically 8
forbidden regions), which leads to longer de Broglie wave- >
>
> 6C eqx ; x < ‘
>(
>
lengths and thus lower energy eigenvalues. This problem is < A cos ðk xÞ
treated in all modern physics and quantum mechanics textbooks, wðxÞ ¼ (3)
> B sin ðk xÞ ; ‘ < x < þ‘
>
but, in spite of its simplicity, the transcendental equations arising >
>
>
: C eq x ;
from the matching conditions for the wave functions at the x > þ‘ :
potential walls, whose roots correspond to the energy eigenval-
ues, do not admit exact closed-form solutions.2,3 Though we can Here, the upper (lower) sign and trigonometric function are
appeal to numerical root-finding (iterative) routines, some alter- for symmetric (antisymmetric) states. The matching condi-
natives are found in the literature (approximate analytical or tions for wðxÞ and w0 ðxÞ at x ¼ 6‘ then yield
graphical solutions or series expansions), but they are of limited
practical value because they are either of poor accuracy or fail A cos ðk ‘Þ ¼ C eq ‘ ; (4a)
in the limit of infinitely deep or very shallow well potentials
(see, e.g., Refs. 5–10). A k sin ðk ‘Þ ¼ C q eq ‘ (4b)
Since the finite SQW model has important applications in
many branches of physics, from atomic and nuclear to for symmetric wave functions and

1019 Am. J. Phys. 88 (11), November 2020 http://aapt.org/ajp C 2020 American Association of Physics Teachers
V 1019
 B sin ðk ‘Þ ¼ C eq ‘ ; (5a) which is just the number of allowed energy levels. Here, bxc
is the integer part of x. Note that the line y ¼ h=hmax crosses
B k cos ðk ‘Þ ¼ C q eq ‘ (5b) the trigonometric curves at points with higher and higher
ordinates, as indicated by vertical dashed lines in Fig. 1,
for antisymmetric ones. Dividing Eq. (4b) by Eq. (4a) and which makes the abscissae hn of these points move toward
Eq. (5b) by Eq. (5a), these pairs of equations are reduced to the lower end of the n-th interval with the increase in n since
each trigonometric curve is a decreasing function of h.
h tan h ¼ q ‘; (6a) (Therefore, ðn  1Þp=2 < hn < hn1 þ p=2 and ðN  1Þp=2
h coth ¼ q ‘; (6b) < hN < minfhmax ;hN1 þ p=2g. These upper bounds should
be taken into account in the optimization of numerical root-
pffiffiffiffiffiffiffiffiffiffi  finding routines applied to this problem.) This simple
respectively, where h  k ‘ ¼ 2mE =h ‘ is a suitable
dimensionless variable. Though these equations are enough graphical solution allows for a better understanding of the
for a graphical solution, as done, e.g., in Refs. 2–4, on seek- dependence of the energy eigenvalues on the SQW physical
ing for a simpler solution, let us rewrite them as parameters. The increase in V0 for a given L, or that of L for
a given V0, increases hmax, thus reducing the slope of the line
h sin h ¼ q ‘ cos h; (7a) y ¼ h=hmax , which increases the number N of intersections
with the trigonometric curves. In the limit of infinitely deep
h cos h ¼ q ‘ sin h : (7b) SQWs, hmax ! 1, and then, the crossing-points will tend to
the points on the axis of abscissae with hn ¼ np=2, which
On taking the square on both sides of each equation, we makes En ¼ h2 hn2 =ð2m‘ 2 Þ tend to n2 1 , which is just the
apply the trigonometric identity sin2 h þ cos2 h ¼ 1 to fur- well-known result for infinite SQWs [see Eq. (1)]. Inversely,
ther reduce them to the decrease in V0 for a given L, or that of L for a given V0,
reduces hmax, thus increasing the slope of the line y ¼ h=hmax
h until the number N of intersections to be reduced to only
j cos hj ¼ ; (8a)
hmax one, which occurs when hmax < p=2, or, equivalently,
V0 < p2 h2 =ð2mL2 Þ ¼ 1 , the ground-state energy for the cor-
h
j sin hj ¼ ; (8b) responding infinitely deep SQW. This shows that there is at
hmax least one stationary bound state in a finite SQW, no matter
pffiffiffiffiffiffiffiffiffiffiffiffi  how shallow it is. For very shallow SQWs, hmax  p=2, so
where hmax  2mV0 =h ‘ corresponds to the maximum the Taylor approximation cosh  1  h2 =2! becomes accu-
value that h can assume. Since h > 0 by virtue of Eq. (6), one rate, and we are left with 1  h2 =2 ¼ h=hmax whose only posi-
must add the conditions tan h > 0 to Eq. (8a) and tive solution is
coth ¼ 1= tan h < 0 to Eq. (8b), which is why only the right-
hand “half” of the trigonometric curves in Eq. (8) were taken qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 
1 2 h3
into account in Fig. 1. In this figure, hn, n ¼ 1; 2; …; N, are the h1 ¼ 1 þ 2 hmax  1  hmax  max ; (10)
abscissae of the points where the straight-line y ¼ h=hmax inter- hmax 2
sects those curves. The total number N of such points is given 2
by the ratio hmax =ðp=2Þ rounded up to the nearest positive inte- where hmax  1 justifies the binomial approximation. On
ger, i.e., writing this result in terms of E1 and V0 and neglecting high-
order terms, one promptly finds
 
hmax
N¼ þ 1; (9) 2m ‘2 V02
ðp=2Þ E1  V0  ; (11)
h2

which is the correct result for very shallow SQWs, as estab-

lished, e.g., in Eq. (4.19) of Ref. 11.
Since the trigonometric curves in Fig. 1 are identical, I
realized that all the roots hn, n ¼ 1; …; N, can be computed
by reducing the trigonometric functions in Eq. (8) to a single
cosine curve over ½0; p=2 containing all “roots” ~h n ¼ hn
ðn  1Þ p=2, as indicated in Fig. 2. Our approximation
scheme consists in interpolating the function y ¼ cos h by its
three points (0, 1), ðp=3; 1=2Þ, and ðp=2; 0Þ, marked with
black bullets in Fig. 2. This yields the parabola (see the red,
dashed curve)
3 2 1
y¼ h  h þ 1: (12)
p2 2p
The exact h~ n values will be approximated by an, the abscis-
Fig. 1. Graphical solution of Eqs. (8a) and (8b). The straight-line sae of the points where the parallel straight-lines
y ¼ h=hmax crosses the trigonometric functions given in their LHS. Each
trigonometric branch has a length p=2 and contains a root hn, n ¼ 1; 2; h p=2
…; N. In this example, the parameter hmax has been chosen as 7p=4, for y¼ þ ðn  1Þ (13)
which Eq. (9) yields N ¼ 4. hmax hmax

1020 Am. J. Phys., Vol. 88, No. 11, November 2020 F. M. S. Lima 1020
 numerically.) Our approximate values are in good agreement
with the corresponding exact ones.
In conclusion, in this note, I have presented a simpler
graphical solution and an approximate analytical formula for
the energy levels of stationary bound states of a particle in a
finite SQW. Our graphical solution demands only one
dimensionless parameter, namely, hmax, which determines
the straight-line y ¼ h=hmax plotted in Fig. 1, which crosses
some identical trigonometric curves, allowing students to see
that finite SQWs always have a finite number N  1 of sta-
tionary bound states and, since this figure also shows that the
decrease in V0 increases the slope of the straight-line, they
can understand why the energy eigenvalues En deviate more
and more from n , the corresponding eigenvalues for infinite
SQWs. In Fig. 2, we succeeded in reducing those curves to a
single cosine curve over the interval ½0; p=2. As this curve
Fig. 2. Our approximation scheme. After reducing the trigonometric func-
tions in Eq. (8) to a single cosine branch over ½0; p=2 containing all roots
resembles a parabola, I have taken the endpoints, together
~h n ; n ¼ 1; …; N, we interpolate cos h with the parabola through the points the “midpoint” ðp=3; 1=2Þ, for a quadratic interpolation.
(0, 1), ðp=3; 1=2Þ, and ðp=2; 0Þ, as indicated by a red dashed curve. These This has led to an accurate formula for the “roots” ~h n , from
three points are marked with black bullets. The dotted lines are geometric which one promptly finds the corresponding energy eigen-
continuations of the original straight-line y ¼ h=hmax . The red bullets are values and all this without artificial numerical factors or
the points where the parallel lines cut the interpolation curve.
adjustable parameters. We also give a numerical example to
illustrate how our approximation scheme works, and we
intersect the interpolation curve, within the interval ð0; p=2Þ. think that it is so simple that the reader can easily repeat the
These points are marked with red bullets in Fig. 2. For each computations for arbitrary values of L and V0 in a pocket
n ¼ 1; …; N, with N as given in Eq. (9), the only positive calculator. As our method is not iterative, it is free of diver-
solution of gence issues, or slow convergence, as well as spurious roots,
with the advantage of becoming more and more accurate in
 
1 1 n1 both limits of infinitely deep and very shallow SQWs, as
3 z2 þ þ zþ  1 ¼ 0; (14) given in Eqs. (1) and (11), respectively.
2 Z 2Z
Of course, some authors have already developed graphical
where z  h=p and Z  hmax =p, is solutions for the finite SQW problem similar to the one pro-
posed here. In Sec. 4.2 of Ref. 11, a more advanced textbook,
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Davies arrives at a pair of simple equations in his Eq. (4.14),
2 ðn  1Þ 4
2 zn ¼ b þ b2  þ ; (15) which is the same as our Eq. (8), but, astonishingly, he turns
3Z 3 back to the more complicated original result, our Eq. (6), to
where b  1=6 þ 1=ð3 ZÞ. Since ~h n  an ¼ p zn , develop a graphical solution! In Ref. 12, the authors also
arrive at that pair of equations, but their graphical analysis is
2
h
2 somewhat obscured by some equations involving complex
En  2
pzn þ ðn  1Þp=2 ; numbers, which follow from their choice of the reference
2m‘ level for the well potential (null in the barriers and negative
h2
p2  into the well). Moreover, they adopt k for the axis of abscis-
¼ ð2zn þ n  1Þ2 ;
2mL2 sae, which is not a dimensionless quantity. In Ref. 13, Mitin
¼ ð2zn þ n  1Þ2 1 : (16) et al. employed a method very similar to ours, but they prefer
to use the inverse trigonometric function arcsinx, which is
Let us show how our approximation scheme works using a not a single-valued function. In comparison to those methods
SQW whose physical parameters, in atomic units, (In this found in textbooks, our method certainly is both simpler and
unit system, lengths are expressed in units of aB  h2 = more elegant.
ðm0 ke e2 Þ ¼ 0:529 Å, the Bohr radius, and energies are in On regarding previous attempts to approximate the
units of Ht  m ke2 e4 =h2 ¼ 27:2114 eV, the Hartree energy. allowed energy levels in finite SQWs, distinct methods are
The main advantage of this system is that m0 ¼ h found in the literature, but they either present a poor accu-
¼ ke ¼ e ¼ 1.) are L ¼ 2 and V0 ¼ 49 p2 =32 ¼ 15:11, racy or they do not remain accurate for both very shallow
which were chosen in a manner that hmax ¼ 74 p, the value and very deep SQWs.5–7,9 Reed’s method,7 e.g., which is
used in Figs. 1 and 2. For this value, Eq. (9) returns often mentioned in recent papers (see, e.g., Ref. 10), involves
N ¼ b3:5c þ 1 ¼ 4. Then, since Z ¼ 7=4, one promptly a rather complicated function, which certainly demands the
finds b ¼ 1=6 þ 4=21 ¼ 5=14, with which our approximate use of a computer, which in turn could be used for solving
formula in Eq. (15) returns 2 z1 ¼ 0:8515; 2 z2 ¼ 0:6821; Eq. (6) numerically or even for implementing a full numeri-
2 z3 ¼ 0:4789, and 2 z4 ¼ 0:2068. With these values in hands, cal solution of the Schr€odinger equation! As our interpola-
being 1 ¼ p2 h2 =ð2mL2 Þ ¼ p2 =8 a.u., Eq. (16) returns the fol- tion of the cosine function includes the endpoints (0, 1) and
lowing approximate values (in a.u.): E1 ¼ 0:8946 ð0:8805Þ; ðp=2; 0Þ, our approximate formula in Eq. (15) will also
E2 ¼ 3:4905 ð3:4864Þ; E3 ¼ 7:5811 ð7:6788Þ, and E4 furnish the exact result for infinite SQWs and the correct
¼ 12:6868 ð12:9883Þ. (The values in parenthesis are the exact result for very shallow SQWs. Finally, for those who insist
energy eigenvalues, found by solving Eqs. (8a) and (8b) in solving Eq. (6) numerically, our approximate energies

1021 Am. J. Phys., Vol. 88, No. 11, November 2020 F. M. S. Lima 1021
 7
could be taken as good starting points to improve the conver- B. C. Reed, “A single equation for finite square well energy eigenvalues,”
gence of numerical routines applied to this problem. Am. J. Phys. 58, 503–504 (1990).
8
B. I. Barker, G. H. Rayborn, J. W. Ioup, and G. E. Ioup, “Approximating
a) the finite square well with an infinite well: Energies and eigenfunctions,”
Electronic mail: [email protected]
1 Am. J. Phys. 59, 1038–1042 (1991).
R. A. Serway, C. J. Moses, and C. A. Moyer, Modern Physics, 3rd ed.
9
(Thomson Brooks/Cole, Belmont, CA, 2005), Secs. 6.4 and 6.5. D. W. L. Sprung, Hua Wu, and J. Martorell, “A new look at the square
2 well potential,” Eur. J. Phys. 13, 21–25 (1992).
D. J. Griffiths and D. F. Schroeter, Introduction to Quantum Mechanics,
10
3rd ed. (Cambridge U. P., Cambridge, UK, 2018), Sec. 2.6. V. Barsan, “Understanding quantum phenomena without solving the
3 Schr€odinger equation: The case of the finite square well,” Eur. J. Phys. 36,
R. W. Robinett, Quantum Mechanics, 2nd ed. (Oxford U. P., Oxford, UK,
2006), Secs. 6.2 and 8.2. 065009 (2015).
4 11
L. I. Schiff, Quantum Mechanics, 3rd ed. (McGraw-Hill, New York, J. H. Davies, The Physics of Low-Dimensional Semiconductors: An
1968), Sec. 2.9. Introduction (Cambridge U. P., New York, 1998), Sec. 4.2.
5 12
S. Garrett, “Bound state energies of a particle in a finite square well,” Am. C. Cohen-Tannoudji, B. Diu, and F. Lalo€e, Quantum Mechanics (John Wiley
J. Phys. 47, 195–196 (1979). & Sons, New Jersey, 1977), Vol. 1, Sec. I.D.b and its Complement HI.
6 13
H. A. Gersch and C. H. Braden, “Approximate energy levels and sizes of V. V. Mitin, D. I. Sementsov, and N. Z. Vagidov, Quantum Mechanics for
bound quantum system,” Am. J. Phys. 50, 53–59 (1982). Nanostructures (Cambridge U. P., New York, 2010), Sec. 3.3.

1022 Am. J. Phys., Vol. 88, No. 11, November 2020 F. M. S. Lima 1022