Chapter 4

The door beckoned, so he pushed through it. Even the street was better lit. He didn’t know
how, he just knew—the patron seated at the corner table was his man. Spectacles, unkept hair,
old sweater, one bottle, one glass. Elbows on the bar and foot on the rail, he strained for calmness
until the barkeep finally appeared. A meek “bourbon, rocks,” left his lips, then he shuffled a bill
he thought was a five onto the bar, pocketed the change, tugged on the drink, and let his feet move
him to the corner table. “I wanna talk to you....” The stranger slowly greeted him with his eyes,
as if he was expected, and replied “What about, son?” Resisting the urge to look away, the words
just fell “. . .those Rutherford rays. . .”

Scattering in One Dimension

The free state addressed in the last chapter is the simplest problem because the potential
is chosen to be zero. The next simplest problems are those where the potentials are piecewise
constant. A potential that is piecewise constant is discontinuous at one or more points. The
potential is chosen to be zero in one region and is of non-zero value in other regions without a
transition region. A discontinuity in a potential is not completely realistic though these problems
do model some realistic systems well. Transistors, semiconductors, and alpha decay (Rutherford
rays) are examples for which an abrupt change in potential is a key feature. Though they are
not perfect models, piecewise constant potentials do contain some realistic physics and serve to
illustrate features of potentials that are very steep.
The primary reason to address these problems is that a discontinuous change in potential
is much more mathematically tractable than a continuous change. These problems are usually
addressed using the position space, time-independent form of the Schrodinger equation. The
strategy is to divide space into regions at the locations where the potential changes and then
attain a solution for each region. The points at which the potential changes are boundaries.
The wavefunction and its first derivative must be the same on both sides of the boundaries.
Equating the wavefunctions and their first derivatives at the boundaries provides a system of
equations that yield descriptive information. Also, solving problems in one dimension is usually
mathematically attractive and can be realistic. One dimensional situations are often a precursor
to multi-dimensional situations.
There is some key quantum mechanical behavior in these problems. You should absorb the
idea that a particle has a non-zero probability to appear on the other side of a potential barrier
that it does not classically have the energy to surmount. This is known as barrier penetration or
tunneling. Also, a particle has a non-zero probability to be reflected at any boundary regardless
of energy. For instance, a particle possessing enough energy to surmount a potential barrier can
still be reflected. You should learn the terms reflection and transmission coefficients.
A particle is in either a free state (chapter 3), a bound state (chapter 5, 6, and parts of
others), or a scattering state. The bound state is described by a potential that holds a particle
for a non-zero time period. The scattering state can be described as an interaction of a free particle
with a potential that results in a free particle. If a free particle interacts with the potential and
does not become bound, the particle is in a scattering state. The hydrogen atom may be ionized.
The electron can escape and enter a free state, but it cannot escape without the influence of an
external photon. Scattering concerns the situation where there is no external influence.
Finally, you should also appreciate the presence of the postulates of quantum mechanics.

201
Though the position space, time-independent Schrodinger equation dominates the discussion, re-
member that it is simply a convenient form of the sixth postulate.

1. Find the reflection and transmission coefficients for a particle of energy E > V0 incident from
the left on the vertical step potential

½
0 if x < 0 region 1 ,
V (x) =
V0 if x > 0 region 2 .

Notice that region 2 extends to infinity. This is an approximation to a potential that is very steep
but not perfectly vertical, and of significant though not of infinite width. If the step is not vertical,
it is difficult to match boundary conditions, and if the step is not of infinite extent, “tunneling”
(see problem 3) is possible.
Particles that are incident from the left will either be reflected or transmitted. The reflection
coefficient, denoted R , is classically defined as the ratio of intensity reflected to intensity inci-
dent. The classical transmission coefficient, denoted T , is the ratio of intensity transmitted to
intensity incident. Quantum mechanically, intensity is analogous to probability density. Quantum
mechanical transmission and reflection coefficients are based on probability density flux (problem
2). The reflection and transmission coefficients must sum to 1 in either classical or quantum
mechanical regimes.
The two physical boundary conditions applicable to this and many other boundary value
problems are the wave function and its first derivative are continuous.

A general solution to the Schrodinger equation for a particle approaching from the left is

ψ1 (x) = A eik1 x + B e−ik1 x , x < 0 ,

ψ2 (x) = C eik2 x , x > 0 ,

where the subscripts on the wavefunctions and wave numbers indicate regions 1 or 2, and
√ p
2mE 2m (E − V0 )
k1 = and k2 = .
h̄ h̄
The ki are real because E > V0 . In region 1, the term A eik1 x is the incident wave and B e−ik1 x
describes the reflected wave. C eik2 x models the transmitted wave in region 2. A term like
D e−ik2 x , modeling a particle moving to the left in region 2, is not physically meaningful for a
particle given to be incident from the left. Equivalently, D = 0 for the same reason. Applying the
boundary condition that the wavefunction is continuous,

ψ1 (0) = ψ2 (0) ⇒ A e0 + B e0 = C e0 ⇒ A + B = C .

The derivative of the wavefunction is

ψ10 (x) = A ik1 eik1 x − B ik1 e−ik1 x , x < 0 ,

202
 ψ20 (x) = C ik2 eik2 x , x > 0 ,
so continuity of the first derivative means
¡ ¢
ψ10 (0) = ψ20 (0) ⇒ A ik1 eik1 (0) − B ik1 e−ik1 (0) = C ik2 eik2 (0) ⇒ k1 A − B = k2 C .

We now have two equations in the three unknowns A, B, and C. Eliminating C,

µ ¶
¡ ¢ ¡ ¢ ¡ ¢ ¡ ¢ k1 − k2
k1 A − B = k2 A + B ⇒ k1 − k2 A = k1 + k2 B ⇒ B= A.
k1 + k2

From this relation we can calculate the reflection coefficient. The incident probability density is
¯ ¯ ¡ ¢ ¡ ¢ ¡ ¢¡ ¢ ¯ ¯
¯ A eik1 x ¯2 = A eik1 x ∗ A eik1 x = A∗ e−ik1 x A eik1 x = A∗ A e0 = ¯ A ¯2 ,

or the intensity of the incident wave is the square of the norm of the coefficient of the incident
¯ ¯2
wave. Similarly, the intensity of the reflected wave is ¯ B ¯ , and the intensity of the transmitted
¯ ¯2
wave is ¯ C ¯ . In this specific problem the reflection coefficient is
¯ ¯2 µ ¶2 Á µ ¶2
¯B¯ k1 − k2 ¯ ¯2 ¯ ¯2 k1 − k2
R = ¯ ¯2 = ¯ A ¯ ¯ ¯
A = and
¯A¯ k1 + k2 k1 + k2

µ ¶2 µ ¶2 µ ¶2
k1 − k2 k1 + k2 k1 − k2 4k1 k2
R+T = 1 ⇒ T = 1−R = 1− = − =¡ ¢2 .
k1 + k2 k1 + k2 k1 + k2 k1 + k2

Postscript: Classically, R = 0 and T = 1 if E > V0 . Figure 4–2 illustrates the quantum

mechanical analogy. What does it mean that R > 0 when E > V0 ? For a single particle,
it means that a portion of the wavefunction is transmitted and a portion of the wavefunction is
reflected. Postulate 4 says that you can calculate the probabilities for finding the particle in region
1 or 2 from the probability amplitudes, but can do no better. If you look, you will find the particle
in either region 1 or region 2 per postulate 3, and the measurement will leave the state vector in
the eigenstate corresponding to the location that you measured per postulate 5. The probability
of finding the particle in the other region is then zero.
Notice that the particle will have longer wavelength and thus less momentum in region 2 since
the energy relative to the “floor” of the potential is less in region 2 than region 1.
The classical trajectory of a particle is continuous and a change in trajectory (the first deriva-
tive) will be continuous. Quantum mechanically, refined arguments are necessary. The validity of
the two boundary conditions can be demonstrated for a Gaussian wavefunction. The issue becomes
that a wave packet that is quantized in wave number will scatter from a potential with precisely
the same reflection and transmission coefficients as a Gaussian wavefunction. This can be done
for a wave packet quantized in wave number that is still smooth compared to the scale of the

203
potential variations. This is a non-trivial calculation that is beyond our scope. The point is that
the boundary conditions of continuity of the wavefunction and its first derivative are also justified
quantum mechanically.

2. (a) Calculate the amplitude C of the transmitted portion of the wavefunction in terms of the
coefficient A for the vertical step potential of problem 1.
(b) Show that the ratio | C | 2 / | A | 2 is 6= T from problem 1.
(c) Rectify the discrepancy between part (b) and the result of problem 1.

Parts (a) and (b) are intended to demonstrate a counter-intuitive fact that is explained in part
(c). Use intermediate results from problem 1 to solve for C in terms of A . Form the ratio
| C | 2 / | A | 2 which is not the same as T from problem 1. Part (c) requires that you know that
flux is velocity times intensity. Since the particle has a different height above the floor of the
potential in regions 1 and 2, it has not only a different wavenumber but a different velocity. These
velocities are related to the de Broglie relation. The ratio v2 | C | 2 /v1 | A | 2 is the transmission
coefficient found in problem 1.

(a) Using intermediate results from problem 1, the amplitude of the transmitted portion is
µ ¶ µ ¶ µ ¶ µ ¶
k1 − k2 k1 + k2 k1 − k2 2k1
C =A+B =A+ A= A+ A= A.
k1 + k2 k1 + k2 k1 + k2 k1 + k2

(b) The ratio of the squares of the magnitudes is not the transmission coefficient of problem 1,
¯ ¯2 µ ¶2
¯C ¯ 2k1 . 4k12 4k1 k2
2
¯ ¯2 = | A | |A|2 = ¡ ¢2 6= ¡ ¢2 .
¯A¯ k1 + k2 k1 + k2 k1 + k2

(c) These differ because the energy, and thus the velocity, of the wave changes as the particle
crosses the step. Flux is intensity times velocity. Probability density flux, probability density
pi h̄ki
times velocity, is appropriate for a quantum mechanical description. Since vi = = ,
m m
¯ ¯2 ¯ ¯2 ¯ ¯2 µ ¶2 ,
v2 ¯ C ¯ h̄k2 /m ¯ C ¯ k2 ¯ C ¯ 2k1 4k1 k2
T = ¯ ¯2 = ¯ ¯2 = ¯ ¯ 2 = k2 | A | 2 k1 | A | 2 = ¡ ¢2 ,
v1 ¯ A ¯ h̄k1 /m ¯ A ¯ k1 ¯ A ¯ k1 + k2 k1 + k2

consistent with the problem 1.

Postscript: A reflected wave is the same height above the floor of the potential as the incident
wave so the reflected wave has the same energy, velocity, and wave number as the incident wave.
The velocities cancel in region 1 so are not considered in calculating the reflection coefficient, i.e.,
¯ ¯2 ¯ ¯2
v1 ¯ B ¯ ¯B ¯
R= ¯ ¯2 = ¯ ¯2 .
v1 ¯ A ¯ ¯A¯

204
3. Determine the transmission coefficient for a particle with E < V0 incident from the left on the
rectangular barrier

½
V0 for −a < x < a ,
V (x) =
0 for | x | > a .

If the energy of the particle is less than the height of the step, and the “ potential plateau” is of
finite length, the particle incident from the left can appear on the right side of a barrier. This is
a non–classical phenomena known as barrier penetration or tunneling. Classically, if a ball
is rolled up a ramp of height h with kinetic energy K , the ball will roll back down the ramp if
K < mgh . A quantum mechanical ball has a non–zero probability that the ball would appear on
the other side of the ramp in the case K < mgh .
This is a boundary value problem with three regions because there are two boundaries. The
strategy is to require continuity of the wavefunction and its first derivative at all boundaries and
solve for the transmission coefficients in terms of the ratios of the squares of the appropriate
wavefunction coefficients.
(a) Divide “all space” into three regions at the boundaries of the potential ±a . The wavefunctions
consist of a linear combination of waves in both directions in each of the three regions:

ψ1 (x) = A eikx + B e−ikx for x < −a ,

κx −κx
ψ2 (x) = C e +De for −a < x < a,
ikx −ikx
ψ3 (x) = F e + G e for x > a,
√ p
where k = 2mE /h̄ and κ = 2m (V0 − E ) /h̄ . Notice that κ is defined so that it is real for
E < V0 . This is a technique that leads to minor simplifications. Conclude that G = 0 because it
is the coefficient of an oppositely directed incident wave, and therefore, is not physical.
(b) Apply the boundary condition that the wavefunction must be continuous at the boundaries.
This yields two equations. Differentiate the wavefunction and then apply the boundary condition
that the first derivative of the wavefunction must be continuous at boundaries. This yields two
more equations. Eliminating B from the two equations for the left boundary, you should find
µ ¶ µ ¶
−ika iκ −κa iκ
2A e = 1− Ce + 1+ D eκx .
k k
Using the two equations at the right boundary to solve for two of the unknown coefficients in terms
of the coefficient F , you should find
µ ¶ µ ¶
κa 1 ik ika −κa 1 ik
Ce = 1+ Fe and De = 1− F eika .
2 κ 2 κ

(c) Use the last two equations to eliminate the coefficients C and D from the first equation so
· ¸
−ika ika i(κ2 − k2 )
Ae =Fe cosh (2κa) + sinh (2κa) .
2kκ

205
 ex − e−x ex + e−x
Remember that sinh (x) = and cosh (x) = .
2 2

(d) A reflected and transmitted particle will be the same height above the floor of the defined
potential so the transmission coefficient is the ratio of probability density of that portion of the
wavefunction which “goes through” the barrier, represented by | F | 2 , to the incident probability
density, represented by | A | 2 . The reciprocal of this ratio, | A | 2 / | F | 2 , is
µ ¶
V02 2a p
T −1
= 1+ sinh2 2m (V0 − E ) ,
4E (V0 − E ) h̄

which is a more pleasing and compact expression than T that still indicates T 6= 0 when E < V0 .

(a) For the region x > a , ψ (x) = F eikx + G e−ikx , and G = 0 because it is the coefficient of an
oppositely directed incident wave that cannot be physical. So the wave function is

 A eikx + B e−ikx , for x < −a ,
ψ (x) = C eκx + D e−κx , for −a < x < a ,

F eikx , for x > a ,
√ p
where k = 2mE / h̄ and κ = 2m (V0 − E ) / h̄ .
(b) There are three regions so there are two boundaries. Continuity at the boundaries requires

A e−ika + B eika = C e−κa + D eκa , x = −a , and (1)

C eκa + D e−κa = F eika , x = a, (2)

so there are two equations in five unknowns. The derivative of ψ (x) is

 A ik eikx − B ik e−ikx , for x < −a ,
0
ψ (x) = C κ eκx − D κ e−κx , for −a < x < a ,

F ik eikx , for x > a.

Applying the boundary condition of continuity of the first derivative at x = −a ,

A ik e−ika − B ik eika = C κ e−κa − D κ eκa , and at x = a , (3)

C κ eκa − D κ e−κa = F ik eika . (4)

There are now four equations in five unknowns. Multiplying equation (1) by ik ,

B ik eika = C ik e−κa + D ik eκa − A ik e−ika .

Substituting this into equation (3) for B ik eika ,

A ik e−ika − C ik e−κa − D ik eκa + A ik e−ika = C κ e−κa − D κ eκa

⇒ 2A ik e−ika = C ik e−κa + C κ e−κa + D ik eκa − D κ eκa

= C (ik + κ) e−κa + D (ik − κ) eκa

206
 µ ¶ µ ¶
−ika iκ −κa iκ
⇒ 2A e = 1− Ce + 1+ D eκa . (5)
k k
Multiplying equation (2) by κ and solving for the term with the coefficient D,

D κ e−κa = F κ eika − C κ eκa

Substituting the right side into equation (4) for D κ e−κa ,

C κ eκa − F κ eika + C κ eκa = F ik eika

⇒ 2C κ eκa = F κ eika + F ik eika = F (κ + ik) eika

µ ¶
1 ik
⇒ C eκa = 1+ F eika . (6)
2 κ
Multiplying equation (2) by κ and solving for the term with the coefficient C,

C κ eκa = F κ eika − D κ e−κa

Substituting the right side into equation (4) for C κ eκa ,

F κ eika − D κ e−κa − D κ e−κa = F ik eika

−2D κ e−κa = −F κ eika + F ik eika = F (−κ + ik) eika

µ ¶
−κa 1 ik
⇒ De = 1− F eika . (7)
2 κ

(c) The signs on the exponentials in equations (6) and (7) are opposite those in equation (5), so
µ ¶ µ ¶
1 ik 1 ik
C eκa = 1+ F eika ⇒ C e−κa = 1+ F eika e−2κa ,
2 κ 2 κ
µ ¶ µ ¶
−κa 1 ik ika κa 1 ik
De = 1− Fe ⇒ De = 1− F eika e2κa ,
2 κ 2 κ
and now the signs of the exponentials are consistent. Substituting these into equation (5),
µ ¶ µ ¶ µ ¶ µ ¶
−ika iκ 1 ik ika −2κa iκ 1 ik
⇒ 2A e = 1− 1+ Fe e + 1+ 1− F eika e2κa
k 2 κ k 2 κ
µ ¶ µ ¶
F eika iκ ik κk −2κa F eika iκ ik κk
= 1− + + e + 1+ − + e2κa
2 k κ κk 2 k κ κk
µ ¶
F eika −2κa iκ −2κa ik −2κa 2κa iκ 2κa ik 2κa
= 2e − e + e + 2e + e − e
2 k κ k κ
· ¸
Fe ika ¡ −2κa 2κa
¢ iκ −2κa ik −2κa iκ 2κa ik2 2κa
2 2 2
= 2 e +e − e + e + e − e
2 kκ kκ kκ kκ
· ¸
F eika ¡ ¢ iκ2 ¡ 2κa ¢ ik2 ¡ 2κa ¢
= 2 e−2κa + e2κa + e − e−2κa − e − e−2κa
2 kκ kκ
· µ 2κa −2κa
¶ 2 2
µ 2κa ¶¸
ika e +e i (κ − k ) e − e−2κa
=Fe 2 +
2 kκ 2

207
 · ¸
−ika ika i (κ2 − k2 )
⇒ Ae =Fe cosh (2κa) + sinh (2κa) .
2kκ

(d) The transmission coefficient is the ratio of intensity transmitted, represented by | F | 2 , to

the intensity incident, represented by | A | 2 . It is conventional to calculate the reciprocal of this
ratio to arrive at a compact expression. Both sides are in the polar form of complex numbers, i.e.,
magnitude and phase. The last result can be arranged as the ratio
¯ ¯ ¯· ¸¯
¯ A e−ika ¯ ¯ i (κ2
− k 2
) ¯
¯ ¯ ¯ sinh (2κa) ¯¯ .
¯ F eika ¯ = ¯ cosh (2κa) + 2kκ
¯ ¯ · ¸1/2 · ¸ £ ¤
∗ 1/2
¯ ¯
¯ A e−ika ¯ A e −ika
A ∗ ika
e A A ∗ 1/2 A A ¯A¯
Now ¯¯ ¯=
¯ = = £ ¤ = ¯ ¯
F eika F eika F ∗ e−ika F F∗ 1/2 ¯F ¯
F F∗
¯· ¸ ¯2 µ 2 ¶2
|A|2 ¯ i(κ2 − k2 ) ¯ (κ − k2 )
⇒ ¯
= cosh (2κa) + ¯ 2
sinh (2κa) ¯ = cosh (2κa) + sinh2 (2κa),
|F |2 ¯ 2kκ 2kκ
where we have used the fact the product of complex conjugates is sum of the squares of the real
part and the coefficient of the imaginary part. Recalling that cosh2 (x) = 1 + sinh2 (x ) ,
µ 4 ¶
|A|2 2 κ − 2k2 κ2 + k4
= 1 + sinh (2κa) + sinh2 (2κa)
|F |2 4k2 κ2
µ ¶
κ4 − 2k2 κ2 + k4
= 1+ 1+ sinh2 (2κa)
4k2 κ2
µ 2 2 ¶
4k κ + κ4 − 2k2 κ2 + k4
= 1+ sinh2 (2κa)
4k2 κ2
µ 4 ¶
κ + 2k2 κ2 + k4
= 1+ 2 2
sinh2 (2κa)
4k κ
¡ 2 ¢2
κ + k2
= 1+ sinh2 (2κa) .
4k2 κ2
√ p
Substituting k = 2mE / h̄ and κ = 2m (V0 − E ) / h̄ ,
µ ¶2
2m 2m
2 (V0 − E ) + 2 E
µ ¶
|A|2 h̄ h̄ 2 2a p
=1+ µ ¶µ ¶ sinh 2m (V0 − E )
|F |2 2m 2m h̄
4 E (V 0 − E )
h̄2 h̄2
4m2 2 µ ¶
4 (V0 − E + E ) 2a p
=1+ h̄ sinh 2
2m (V − E )
0
4m2 h̄
4E (V 0 − E )
h̄4 µ ¶
|A|2 −1 V02 2 2a p
⇒ =T =1+ sinh 2m (V0 − E ) .
|F |2 4E (V0 − E ) h̄

4. Determine the transmission coefficient for a particle with E = V0 incident from the left on the
rectangular barrier defined in the last problem.

208
The intent of this problem is to further develop your skills applying boundary conditions. Require
continuity of the wavefunction and its first derivative at all boundaries.
The first step, which is non-trivial, is to write a proper wavefunction. Similar to problem 3,
the wavefunction in the three regions can be written as

ψ1 (x) = A eikx + B e−ikx for x < −a ,

ψ2 (x) = C + D x for − a < x < a,
ψ3 (x) = F eikx for x > a.

The wave number is the same in regions 1 and 3 because particle energy is the same height above
the “floor,” and is zero in region 2 for either possible definition because E = V0 . Follow parts
(b) through (d) of problem 3 to get the reciprocal of the transmission coefficient,

2mE 2
T −1 = 1 + a .
h̄2

p
When E = V0 , the wavenumber in the region −a < x < a is k = 2m (V0 − E ) / h̄ = 0 , so

 Aeikx + Be−ikx , for x < −a ,
the wavefunction can be written ψ (x) = C + Dx , for −a < x < a ,
 ikx
Fe , for x > a .

Continuity of the wavefunction at x = −a means

A e−ika + B eika = C − D a , (1)

and at x = a , C + D a = F eika . (2)

Continuity of the derivative of the wavefunction at x = −a means

A ik e−ika − B ik eika = D , (3)

ika
and at x = a , D = F ik e . (4)

Multiplying equation (1) by ik and solving for the term with the coefficient B yields

B ik eika = ik C − ik D a − A ik e−ika .

Substituting this into equation (3) eliminates the unknown coefficient B ,

A ik e−ika − ik C + ik D a + A ik e−ika = D

⇒ 2A ik e−ika = ik C + D (1 − ika) . (5)

Equation (2) can be written C = F eika − D a , and substituting this into equation (5),

2A ik e−ika = ik (F eika − D a) + D (1 − ika)

= ik F eika − ik D a + D (1 − ika)
= ik F eika + D (1 − 2ika ) .

209
Substituting equation (4) into the last equation eliminates the unknown coefficient D ,

2A ik e−ika = ik F eika + F ik eika (1 − 2ika)

⇒ 2A e−ika = F eika + F eika (1 − 2ika)
⇒ 2A e−ika = F eika (2 − 2ika)
⇒ A e−ika = F eika (1 − ika)
A e−ika
⇒ ika
= 1 − ika
¯F e −ika ¯ ¯ ¯
¯ Ae ¯ ¯ A ¯ |A|
and ¯ ¯ ¯ ¯
¯ F eika ¯ = ¯ F ¯ = | F | as previously shown,
2 ¯ ¯2
|A| ¯ ¯ = 1 + k2 a2 .
⇒ 2 = 1 − ika
|F |

√ 2
|A| 2mE 2
Substituting k = 2mE / h̄ , 2 = T −1 = 1 + a for E = V0 .
|F | h̄2

5. Show in the limit E → V0 from below, that the results of problems 3 and 4 are equivalent.

The intent of this problem is to demonstrate self-consistency by examining a limiting case. It is

a technique that is difficult to teach and important to learn. Start with the result of problem 3.
Expand sinh (x) and ignore the higher order terms so that sinh (x) ≈ x since E ≈ V0 . Of course,
in the limit E → V0 the results of problem 3 and 4 must be the same.

Start with the reciprocal of the transmission coefficient from problem 3. Ignoring higher order
terms as negligible in the series expansion of sinh (x) ⇒ sinh (x) ≈ x ,
µ ¶2
−1 V02 2a p
T =1+ 2m (V0 − E )
4E (V0 − E ) h̄
µ 2 ¶
V02 4a
=1+ 2m (V 0 − E )
4E (V0 − E ) h̄2
µ ¶
V02 a2
=1+ 2m
E h̄2
2mV02 2
=1+ a .
h̄2 E
2mE 2
Remembering that E ≈ V0 , T −1 ≈ 1 + a .
h̄2

6. (a) A particle of energy E > V0 is incident on a rectangular barrier of width 2a . Solve for
the value of the transmission coefficient and locate the positions of the maxima and minima in
terms of wave number. Plot transmission coefficient versus barrier width for a particle of constant
energy and variable barrier width.

210
(b) Locate the positions of the maxima in terms of particle energy. Plot transmission coefficient
versus particle energy where E > V0 and barrier width is fixed.

This problem introduces a phenomenon known as resonance scattering. At certain values of

particle wave number (energy), there is 100% transmission and no reflection. Part (a) should
illustrate resonance scattering for an incident particle of fixed energy and a barrier width that is
varied. Part (b) illustrates the same phenomenon where particle energy is varied and barrier width
is fixed. Start with an intermediate result from problem 10,
µ ¶2
1 1 k12 − k22
=1+ sin2 (2k2 a ) .
T 4 k1 k2

Solve for T instead of its reciprocal. The maxima occur where the sine term is zero at barrier
widths of 2a = nπ/k2 . The minima occur where the sine term is one, where the transmission
µ ¶2
2k1 k2
coefficient has the value T = . Part (b) is likely easiest if started with
k12 + k22
µ ¶
1 V02 2a p
= 1+ sin2 2m (E − V0 ) ,
T 4E (E − V0 ) h̄

the result of problem 10 in terms of energy. Again, solve for T instead of its reciprocal. This plot
is more difficult because the independent variable E appears in five places including the square
root of the sine squared term. The important part of this graph is the shape. If you do not have
access to a commercial graphing package, try to imagine what the graph should look like, and skip
to the solution. You should find the maxima at

π2 h̄2 2
E= n + V0 .
8ma2

This result is closely related to the infinite square well that is encountered in the next chapter.

µ ¶2
1 1 k12 − k22 4k2 k 2 k 4 − 2k12 k22 + k24
(a) = 1+ sin2 (2k2 a) = 12 22 + 1 sin2 (2k2 a)
T 4 k1 k2 4k1 k2 4k12 k22
¡ ¢
4k12 k22 + k14 − 2k12 k22 + k24 sin2 (2k2 a)
=
4k12 k22
4k12 k22
⇒ T = 2 2 ¡ 4 ¢ .
4k1 k2 + k1 − 2k12 k22 + k24 sin2 (2k2 a)

nπ
Maxima of T = 1 occur where sin2 (2k2 a) = 0 , i.e., where 2k2 a = nπ ⇒
are 2a =
k2
the barrier widths at which maxima occur, noting that the width of the barrier is given to be 2a .
The minima occur where the sine squared term is one, so minima occur at
µ ¶2
4k 2 k2 4k12 k22 2k1 k2
T = 2 2 ¡ 4 1 22 2 ¢ = ⇒ T = at minima.
4k1 k2 + k1 − 2k1 k2 + k24 · 1 k14 + 2k12 k22 + k24 k12 + k22

211
(b) In terms of particle energy,
µ ¶
1 V02 2a p
=1+ sin2 2m (E − V0 )
T 4E (E − V0 ) h̄
µ ¶
2 2 2a p
4E (E − V0 ) + V0 sin 2m (E − V0 )
h̄
=
4E (E − V0 )
4E (E − V0 )
⇒ T = µ ¶.
2 2 2a p
4E (E − V0 ) + V0 sin 2m (E − V0 )
h̄

The maxima occur where the sine squared term is zero, so

2a p p nπh̄
2m (E − V0 ) = nπ ⇒ 2m (E − V0 ) =
h̄ 2a
2 2 2 2
π h̄ 2 π h̄ 2
⇒ 2m (E − V0 ) = n ⇒ E − V0 = n
4a2 8ma2
π 2 h̄2 2
⇒ E= n + V0
8ma2

which are energies that are closely related to the

eigenenergies of a particle in an infinite square well.

Postscript: Resonance scattering is the circumstance where 100% transmission occurs.

7. (a) Show in position space that

³ ´ 2m Z ²
∆ ψ 0 (x) = 2 lim V (x) ψ (x) dx
h̄ ²→0 −²

is the general value of the discontinuity in the first derivative for an infinite potential.
(b) Calculate the value of the discontinuity in the first derivative for a particle interacting with
the potential well V (x) = −α δ (x) .

Use the conditions that ψ and ψ 0 are continuous at any boundary of a finite potential. There
must, however, be a discontinuity in the first derivative at all infinite boundaries. If the value of the
discontinuity can be calculated, the boundary condition for the first derivative is again useful. The
method of calculating the value of the discontinuity is to put the potential into the Schrodinger
equation, integrate around the infinite boundary, and let the limits approach zero from both sides.

For a general potential V (x) , the Schrodinger equation in position space is

h̄2 d2 ψ (x)
− + V (x) ψ (x) = E ψ (x) .
2m dx2
212
If we integrate around the point which is the discontinuity, this is
Z ² 2 Z ² Z ²
h̄2 d ψ (x)
− dx + V (x) ψ (x) dx = E ψ (x) dx,
2m −² dx2 −² −²
which assumes that the infinite boundary is at x = 0 . The quantity ² is infinitesimal. If we let
² → 0, the right side of the equation is zero. The integral of a finite quantity, ψ (x) , over an
arbitrarily small interval is zero. The term with the potential does not vanish because the integral
of an infinite quantity, the potential, can be non-zero even for an arbitrarily small interval. So,
Z ² 2 Z ²
d ψ (x) 2m
lim dx = 2 lim V (x) ψ (x) dx
²→0 −² dx2 h̄ ²→0 −²
Z ²
d ψ (x) ¯¯ d ψ (x) ¯¯ 2m
⇒ ¯ − ¯ = 2 lim V (x) ψ (x) dx
dx ²→0 dx −²→0 h̄ ²→0 −²
³ ´ 2m Z ²
0
⇒ ∆ ψ (x) = 2 lim V (x) ψ (x) dx (1)
h̄ ²→0 −²
is the general value of the discontinuity in the first derivative.
(b) A delta function potential is infinite, so employing equation (1),
Z ²
¡ 0 ¢¯¯ 2m ¡ ¢ 2m
∆ ψ (x) ¯ = 2 lim − α δ (x) ψ (x) dx = − 2 α ψ (0) ,
x=0 h̄ ²→0 −² h̄
because regardless of how closely ² approaches zero, −² < 0 < ² meaning that zero remains within
the limits of integration. Since the value of the delta function that makes its argument zero is
within the limits of integration, the integral is the wavefunction evaluated at the point that makes
the argument of the delta function zero.

Postscript: An infinite potential models a perfectly rigid and perfectly impenetrable wall. The
wavefunction must be zero at an infinite potential barrier and in the region of infinite potential
because the term V ψ in the Schrodinger equation would be infinite otherwise. There is a non-
smooth “corner” in the wavefunction as it goes to zero at any position other than ±∞ , thus there
is a discontinuity in the first derivative of the wavefunction. This physically unrealistic model is
useful to understanding concepts, is mathematically tractable, and is a first approximation to an
abrupt and large potential.

8. Calculate the reflection and transmission coefficients for a particle incident on the potential
V (x) = −α δ (x) .

This problem is intended to demonstrate how a potential that includes

a delta function is treated specifically, and how to treat a discontinu-
ity in general. First, write the wavefunction for two regions because
the delta function is of “zero” width. The wavenumber is the same in
regions 1 and 2 because the particle energy is the same height above
the potential “floor” in both regions. A ψ (0) is required. Use either
ψ1 (0) or ψ2 (0) , because the condition of continuity of the wavefunc-
tion at the boundary ensures that they are the same. We use ψ2 (0)
because it has one fewer terms than ψ1 (0) . Classically, the reflection
coefficient must be zero. The quantum mechanical result is non-zero.

213
The solution to the Schrodinger equation in position space is

ψ1 (x) = A eikx + B e−ikx , x < 0 , ψ2 (x) = C eikx , x > 0 ,

where the infinite potential well is at x = 0 . Continuity of the wave function means

A eik(0) + B e−ik(0) = C eik(0) ⇒ A + B = C .

The first derivatives in regions 1 and 2 are

ψ10 (x) = ik A eikx − ik B e−ikx , ψ20 (x) = ik C eikx .

We calculated the general value of the discontinuity of the first derivative in problem 7. This is
the difference of the first derivatives in regions 1 and 2, meaning
2m
ψ10 (0) − ψ20 (0) = − α ψ (0)
h̄2
2m
⇒ ik A eik(0) − ik B e−ik(0) − ik C eik(0) = − α ψ (0) .
h̄2
Using ψ2 (0) = C ,
2m
ik A − ik B − ik C = − αC
h̄2
µ ¶
2m
⇒ ik A − ik B = C ik − 2 α ,
h̄
and we use the continuity condition, A + B = C , to eliminate C , so
³ µ ¶
¢ 2m
ik A − ik B = A + B ik − 2 α
h̄
µ ¶ µ ¶
2m 2m
⇒ ik A − ik B = A ik − 2 α + B ik − 2 α
h̄ h̄
µ. . ¶ µ ¶
2m 2m
⇒ A ik − ik + 2 α = B ik + ik − 2 α
h̄ h̄
µ ¶ µ ¶
2m 2m
⇒ A α = B 2ik − α
h̄2 h̄2
·µ ¶Áµ ¶¸
2m 2m mαA
⇒ B=A 2 α 2ik − 2 α = .
h̄ h̄ ikh̄2 − mα
The reflection coefficient is
¯ ¯2 µ ¶µ ¶
¯B ¯ mαA∗ mαA 1 m2 α2 1
R = ¯ ¯2 = 2 2 = 4 = ± .
¯A¯ −ikh̄ − mα ikh̄ − mα | A | 2
k2 h̄ + m2 α2 1 + k2 h̄4 m2 α2

The transmission coefficient is

m2 α2 k2 h̄4 + m2 α2 − m2 α2 k2 h̄4 1
T =1−R=1− 4 = 4 = 4 = ± .
2
k h̄ + m α2 2 2 2
k h̄ + m α 2 2
k h̄ + m α 2 2 1 + m α2 k 2 h̄4
2

214
Postscript: Any curve that has a sharp corner has a discontinuity in it’s next order derivative.
Delta functions and theta functions can be useful in describing discontinuities. Techniques seen
in problems 7 and 8 are useful in a number of areas, the use of delta functions in particular is
becoming increasingly popular, and you will see these techniques in future chapters.

Practice Problems

9. For a particle that encounters a vertical step potential of height V0 , calculate the reflection and
transmission coefficients for the particle energies (a) E = 2V0 , (b) E = 3V0 /2 ,
(c) E = 1.1V0 , and (d) E = V0 . (e) Explain the result of part (d).

This problem is intended primarily to reinforce the quantum mechanical effect that reflection occurs
at a step even when E > V0 . It should further familiarize you with reflection and transmission
coefficients. The boundary value problem for a vertical step potential is solved in problem 1,
resulting in the general forms of the reflection and transmission coefficients
µ ¶2
k1 − k2 4 k1 k2
R= and T = 1 − R = .
k1 + k2 (k1 + k2 )2

Express the energies of parts (a) through (d) in terms of the wavenumbers to attain the reflection
and transmission coefficients using these equations. You should find an increasing amount of
reflection as the energy of the particle gets closer to the energy of the potential barrier, until at
E = V0 , the transmission coefficient is zero. The reason for this is there are artificialities built
into the model to make it more mathematically tractable. Reread the comments on tunneling to
identify the particular artificiality which founds the 100% reflection of part (d) to answer part (e).

10. Determine the transmission coefficient for a particle with E > V0 incident from the left on
the rectangular barrier defined in problem 3.

Reflection from a potential barrier of energy less than that of the incident particle is another solely
quantum mechanical phenomenon. Start with the wavefunction

 A eik1 x + B e−ik1 x , for x < −a ,
ψ (x) = C eik2 x + D e−ik2 x , for −a < x < a ,

F eik1 x , for x > a .

Require continuity of the wavefunction

√ and its first derivative
p at all boundaries. Follow the proce-
dures of problem 2. Define k1 = 2mE / h̄ and k2 = 2m (E − V0 ) / h̄ , to find
µ ¶
V02 2a p
T −1
=1+ sin2 2m (E − V0 ) .
4E (E − V0 ) h̄

215
11. Show in the limit E → V0 from above, that the result of problems 4 and 10 are equivalent.

This problem has the same intent as problem 5. Start with the result of problem 10. Expand
sin (x) and ignore the higher order terms so that sin (x) ≈ x since E ≈ V0 . You must find that
in the limit E → V0 the results of problems 10 and 4 are the same.

12. A particle of E > 0 approaches the negative vertical step potential

½
0, for x < 0,
V (x) =
−V0 for x > 0.

What are the reflection and transmission coefficients if E = V0 /3 , and E = V0 /8 ?

This problem parallels the discussion of the vertical step potential. It is intended to reinforce
the methods of addressing a boundary value problem. It also illustrates another completely non-
classical phenomenon. Classically, we expect the reflection coefficients to be zero for any E > 0 .
(a) Write the wavefunction. You have only two regions to consider. You should recognize that
the general form of the wavefunction is

ψ1 (x) = Aeik1 x + Be−ik1 x for x < 0,

ψ2 (x) = Ceik2 x + De−ik2 x for x > 0,

√ p p
where k1 = 2mE / h̄ and k2 = 2m (E − (−V0 )) / h̄ = 2m (E + V0 ) / h̄ . Can you conclude
that D = 0 because it is the coefficient of an oppositely directed incident wave?
(b) Apply the continuity conditions to the wavefunction and its first derivative at x = 0 . You
should have two equations in three unknowns. The square of the coefficient A is the intensity of
incidence, and the square of the coefficient B is the intensity of reflection. Since A and B are
the two coefficients you want to compare, you should combine your two equations to eliminate C .
You should find that
A (k1 − k2 ) = B (k1 + k2 ) .

(c) The reflection coefficient is the ratio | B | 2 / | A | 2 , which you can form from your result
of part (b). Use the definitions of the wavenumbers to establish R in terms of E and V0 , then
substitute E = V0 /3 and E = V0 /8 to get numerical answers. You should find R = 1/9 and 1/4 ,
respectively, and the transmission coefficients follow directly using R + T = 1 .

13. (a) Find the transmission coefficient for the rectangular well
½
−V0 , for −a < x < a ,
V (x) =
0 , for | x | > a .

(b) Show that T = 1 in the limit that V0 = 0 .

216
This problem is provided to give you some practice at solving a boundary value problem. The
procedures for this problem strongly parallel problems 3 and 4. You should find
µ ¶
V02 2a p
T −1 = 1 + sin2 2m (E + V0 )
4E (E + V0 ) h̄
√ p p
using k1 = 2mE / h̄ and k2 = 2m (E − (−V0 )) / h̄ = 2m (E + V0 ) / h̄ . Consider the
width of the well when V0 → 0 for part (b).

14. The transmission coefficient a particle of E < V0 approaching a rectangular barrier is

1
T = µ ¶.
V02 2 2a p
1+ sinh 2m (V0 − E )
4E (V0 − E ) h̄
µ ¶
V02 2 2a p
(a) Show that the term sinh 2m (V0 − E ) is non–negative and real, and
4E (V0 − E ) h̄
(b) show that this means R < 1 for the case E < V0 .

The transmission coefficient is non-zero for the case E < V0 , which is nonsense in a classical
regime. Since quantum mechanics “lives” in the world of complex numbers, this problem shows
that a non–zero transmission coefficient at a barrier is not a “trick” associated with complex or
imaginary numbers, and thus, there is actually a portion transmitted and only a portion reflected
when E < V0 . Start with the result of problem 3. There are energies, intrinsically positive and
real, two squares, and a product to consider. Can any of those make this term negative? Once
you know that the given term is non–negative and real, set it equal to a constant. Part (b) is the
algebra of showing that if the constant is positive and real, R is necessarily less than one.

15. Plot barrier width versus transmission coefficient for a rectangular barrier of width 2a for the
case E = V0 . What widths of the barrier result in 10%, 1%, 0.1%, and 0.01% transmission
when particle energy is equal to barrier height?

This problem illustrates that even a thin barrier prevents most transmission at E = V0 . A barrier
a few wavelengths in width results in miniscule transmission. Start with the reciprocal of the
transmission coefficient from problem 4 expressed in terms of energy and half barrier width. Use
the de Broglie relationship to express the energy in terms of wavelength, then solve for 2a . The
barrier width that results in 10% transmission is slightly less than one particle wavelength.

16. Calculate the reflection and transmission coefficients for a particle incident on the potential

V (x) = α δ (x) .

Problems 16 and 17 are intended to reinforced the ideas and the procedures in problems 7 and 8.
You should find the same reflection and transmission coefficients as found in problem 8.

217
17. Calculate the reflection and transmission coefficients for a particle incident on the potential

V (x) = −α δ (x − a) .

Again, follow problems 7 and 8. Again, you should find the same reflection and transmission
coefficients as found in problem 8 though the delta function located at other than zero requires
carrying exponentials. For instance,

mα A ei2ka
B= .
ikh̄2 − mα

Remember that a norm is the square root of the product of complex conjugates per problem 3 (d).

218
 Chapter 4 Homework Solutions

10. For the case E > V0 , the development is very similar to that of E < V0 . The primary
differences are all exponentials have imaginary factors, and we “add zero,” add and subtract the
same quantity, to arrive at a form of the transmission coefficient analogous to the case of E < V0 .
The wave function is

 A eik1 x + B e−ik1 x , for x < −a,
ψ(x) = C eik2 x + D e−ik2 x , for −a < x < a,

F eik1 x , for x > a,
√ p
where k1 = 2mE/h̄ and k2 = 2m(E − V0 )/h̄. Note since E > V0 , we define the wave
number in terms of E − V0 in the region −a < x < a versus V0 − E in the case E < V0 . The
derivative is 
 A ik1 eik1 x − B ik1 e−ik1 x , for x < −a,
0
ψ (x) = C ik2 eik2 x − D ik2 e−ik2 x , for −a < x < a,

F ik1 eik1 x , for x > a.
The continuity conditions are at x = −a,

A e−ik1 a + B eik1 a = C e−ik2 a + D eik2 a , (1)

and at x = a,
C eik2 a + D e−ik2 a = F eik1 a . (2)
Applying the boundary condition of continuity of the derivative at x = −a,

A ik1 e−ik1 a − B ik1 eik1 a = C ik2 e−ik2 a − D ik2 eik2 a . (3)

At x = a,
C ik2 eik2 a − D ik2 e−ik2 a = F ik1 eik1 a . (4)
Multiplying equation (1) by ik1 and solving for the term with the coefficient of B,

B ik1 eik1 a = C ik1 e−ik2 a + D ik1 eik2 a − A ik1 e−ik1 a .

Substituting the right side into equation (3),

A ik1 e−ik1 a − C ik1 e−ik2 a − D ik1 eik2 a + A ik1 e−ik1 a = C ik2 e−ik2 a − D ik2 eik2 a

⇒ 2A k1 e−ik1 a = C (k1 + k2 ) e−ik2 a + D (k1 − k2 ) eik2 a

µ ¶ µ ¶
−ik1 a k1 + k2 −ik2 a k1 − k2
⇒ 2A e =C e +D eik2 a . (5)
k1 k1
Multiplying equation (2) by ik2 and solving for the term with the coefficient of D,

D ik2 e−ik2 a = F ik2 eik1 a − C ik2 eik2 a .

Substituting this into equation (4),

C ik2 eik2 a − F ik2 eik2 a + C ik2 eik2 a = F ik1 eik1 a

1
 ⇒ 2C k2 eik2 a = F (k1 + k2 ) eik1 a
µ ¶
1 k1 + k2
⇒ C= F eik1 a e−ik2 a .
2 k2
Multiplying equation (2) by ik2 and solving for the term with the coefficient of C,
C ik2 eik2 a = F ik2 eik1 a − D ik2 e−ik2 a
and using this in equation (4),
F ik2 eik1 a − D ik2 e−ik2 a − D ik2 e−ik2 a = F ik1 eik1 a
−2D k2 e−ik2 a = F (k1 − k2 ) eik1 a
µ ¶
1 k2 − k1
D= F eik1 a eik2 a .
2 k2
Substituting the relationships for C and D into equation (5) yields
µ ¶µ ¶ µ ¶µ ¶
−ik1 a 1 k1 + k2 k1 + k2 ik1 a −2ik2 a 1 k2 − k1 k1 − k2
2A e = F e e + F eik1 a e2ik2 a
2 k2 k1 2 k2 k1
¡ ¢ ¡ ¢
F ik1 a k12 + 2k1 k2 + k22 −2ik2 a F ik1 a k12 − 2k1 k2 + k22 2ik2 a
= e e − e e
2 k1 k2 2 k1 k2
F h
= eik1 a k12 e−2ik2 a + 2k1 k2 e−2ik2 a + k22 e−2ik2 a
2k1 k2
i
− k12 e2ik2 a + 2k1 k2 e2ik2 a − k22 e2ik2 a
F h ¡ ¢ ¡ ¢ ¡ ¢i
= eik1 a − k12 e2ik2 a − e−2ik2 a + 2k1 k2 e2ik2 a + e−2ik2 a − k22 e2ik2 a − e−2ik2 a
2k1 k2
F £ ¤
= eik1 a −2ik12 sin(2k2 a) + 4k1 k2 cos(2k2 a) − 2ik22 sin(2k2 a)
2k1 k2
· ¸
ik1 a (k12 + k22 )
=Fe 2 cos(2k2 a) − i sin(2k2 a)
k1 k2
· ¸
A e−ik1 a i (k12 + k22 )
⇒ = cos(2k2 a) − sin(2k2 a) .
F eik1 a 2 k1 k2
¯ ¯ ¯ ¯
¯ Ae−ik1 a ¯ ¯ A ¯ | A |
Since ¯ ¯ ¯ ¯
¯ F eik1 a ¯ = ¯ F ¯ = | F | ,
2 ¯ ¯2
|A| ¯ i (k 2
+ k 2
) ¯
¯ 1 2
sin(2k2 a)¯¯
2 = ¯cos(2k2 a) −
|F | 2 k1 k2
µ ¶2
2 1 k12 + k22
= cos (2k2 a) + sin2 (2k2 a).
4 k1 k2
Adding zero in the form of sin2 (2k2 a) − sin2 (2k2 a),
2 µ ¶2
|A| 2 2 1 k12 + k22
2
2 = cos (2k2 a) + sin (2k2 a) − sin (2k2 a) + 4 sin2 (2k2 a)
|F | k1 k2
" µ ¶2 #
1 k12 + k22
=1+ − 1 sin2 (2k2 a). (6)
4 k1 k2

2
The expression in the square brackets reduces,
µ ¶2 µ ¶2
1 k12 + k22 k14 + 2k12 k22 + k24 4k12 k22 k14 − 2k12 k22 + k24 1 k12 − k22
−1= − 2 2 = = .
4 k1 k2 4k12 k22 4k1 k2 4k12 k22 4 k1 k2
√ p
In terms of energy where k1 = 2mE/h̄ and k2 = 2m(E − V0 )/h̄, this is
 2m 2m 2
à !2
E − (E − V 0 )
1  h̄2 h̄2  1 E − E + V0 V02
 2m √ p  = √ p = .
4 4 E (E − V 0 ) 4E(E − V0 )
E (E − V0 )
h̄2
Using this in equation (6) and expressing the wave number in the argument of the sine in terms
of energy, we arrive at the desired form,
2 µ ¶
|A| −1 V02 2 2a p
2 = T = 1 + sin 2m(E − V 0 ) .
|F | 4E(E − V0 ) h̄

12. (a) There are only two regions to consider, so the wave function is
½
Aeik1 x + Be−ik1 x , for x < 0,
ψ(x) =
Ceik2 x , for x > 0,
√ p p
where k1 = 2mE/h̄ and k2 = 2m(E − (−V0 ))/h̄ = 2m(E + V0 )/h̄, and the coefficient of
a term like De−ik2 x must be zero for x > 0 because it is the coefficient of an oppositely directed
incident wave so is non-physical.
(b) The derivatives of the wave function are
½
0 A ik1 eik1 x − B ik1 e−ik1 x , for x < 0,
ψ (x) =
C ik2 eik2 x , for x > 0.
Continuity of ψ(x) at x = 0 means A + B = C because the arguments of all exponentials are
zero. Continuity of ψ 0 (x) at x = 0 means
A ik1 − B ik1 = C ik2 .
Using C = A + B to eliminate C,
A ik1 − B ik1 = (A + B) ik2
= A ik2 + B ik2
⇒ Ai(k1 − k2 ) = Bi(k1 + k2 )
⇒ A(k1 − k2 ) = B(k1 + k2 ).

(c) The reflection coefficient is the ratio of the squares of the amplitudes of that portion reflected
to that portion incident. Using the result of part (b),
A (k1 − k2 ) = B (k1 + k2 )
B k1 − k2
⇒ =
A k1 + k2
¯ ¯2 µ ¶2
¯B ¯ k 1 − k 2
⇒ ¯¯ ¯¯ = .
A k +k
1 2

3
 √ p
Inserting values for the wave numbers, where k1 = 2mE/h̄ and k2 = 2m(E + V0 )/h̄,
¯ ¯2 Ã √ p !2
¯B ¯ 2mE/h̄ − 2m(E + V0 )/h̄
¯ ¯ = √ p
¯A¯ 2mE/h̄ + 2m(E + V0 )/h̄
Ã√ √ !2
E − E + V0
= √ √
E + E + V0
à p !2
1 − 1 + V0 /E
= p .
1 + 1 + V0 /E
If E = V0 /3, the ratio V0 /E = V0 /(V0 /3) = 3, and
¯ ¯2 µ √ ¶2 Ã √ !2 µ ¶2 µ ¶2
¯B ¯ 1 − 1 + 3 1 − 4 1−2 −1
¯ ¯ = √ = √ = =
¯A¯ 1+ 1+3 1+ 4 1+2 3
¯ ¯2
¯B ¯ 1 8
⇒ ¯¯ ¯¯ = R = ⇒ T = .
A 9 9
If E = V0 /8, the ratio V0 /E = V0 /(V0 /8) = 8, and
¯ ¯2 µ √ ¶2 Ã √ !2 µ ¶2 µ ¶2
¯B ¯ 1 − 1 + 8 1 − 9 1 − 3 −2
¯ ¯ = √ = √ = =
¯A¯ 1+ 1+8 1+ 9 1+3 4
¯ ¯2
¯B ¯ 1 3
⇒ ¯¯ ¯¯ = R = ⇒ T = .
A 4 4
√
13. (a)
p The wave numbers are
p comparable to those from problem 12 where k 1 = 2mE/h̄ and
k2 = 2m(E − (−V0 ))/h̄ = 2m(E + V0 )/h̄. The solution to this problem is procedurally the
same as problems 3 and 10 with the exception of the wave numbers. The differences are k in
problem 3 is replaced by k1 and κ in problem 3 is replaced with ik2 , and a sin emerges
instead of a sinh. The first substantial difference is when wave numbers are replaced by energies
in part (d). The comparable circumstance for the square well is
2
¡ 2 ¢2
|A| k1 − k22
2 = 1+ sin2 (2k2 a)
|F | 4k12 k22
so substituting the appropriate wave numbers,
µ ¶2
2m 2m
2 E− (E + V0 ) µ ¶
|A| h̄2 ¶ µh̄2 2 2a p
2 =1+
µ ¶ sin 2m(E + V0 )
|F | 2m 2m h̄
4 E (E + V 0 )
h̄2 h̄2
4m2 2 µ ¶
4 (E − E − V0 ) 2a p
=1+ h̄ sin 2
2m(E + V0 )
4m2 h̄
4E (E + V0 )
h̄4
2 µ ¶
|A| −1 V02 2 2a p
⇒ 2 = T =1+ sin 2m(E + V0 ) .
|F | 4E(E + V0 ) h̄

4
(b) V0 = 0 ⇒ a = 0. If a = 0, the sin2 term is zero, so T = 1 when V0 = 0.

15. The result of problem 4 for E = V0 is

1 2mE 2
= 1+ a .
T h̄2
The de Broglie wavelength is

h h √ h h2
λ= = √ ⇒ 2mE = ⇒ E= ,
p 2mE λ 2mλ2

so the reciprocal of the transmission coefficient is

µ ¶
h2
2m
1 2mλ2 4π 2 a2
=1+ µ ¶2 a2 = 1 +
T h λ2
2π
µ ¶ µ ¶1/2
4π 2 a2 1 2 2 2 1 λ 1
⇒ 2
= −1 ⇒ 4π a = λ −1 ⇒ a= −1 .
λ T T 2π T
Remember 2a is the width of the barrier, so
µ ¶1/2
λ 1
2a = −1 .
π T

This is barrier width in terms of wavelength

and transmission coefficient, and the graph is
at the right. It can be interpreted as the bar-
rier width in terms of wavelength for a given
value of the transmission coefficient. A numer-
ical value is

µ ¶1/2
λ 1 λ³ ´1/2 λ 1/2
2a = −1 = 10 − 1 = (9) = 0.955λ
π 0.1 π π
for 10% transmission. Notice that this is less than one particle wavelength. The others are
calculated similarly, using T = 0.01, 0.001, and 0.0001 respectively, so the four requested
numerical values are
% transmission T 2a
10% 0.1 0.955λ
1% 0.01 3.167λ
0.1% 0.001 10.061λ
0.01% 0.0001 31.829λ