Tutorial: Possible Wave Functions for Systems with a Single Particle

Confined in One Spatial Dimension

 A “system” refers to an electron in a potential energy well, e.g., an electron in a

one-dimensional infinite square well. The system is specified by a given
Hamiltonian.
 Assume all systems are isolated.
 TISE and TDSE are abbreviations for the Time Independent Schrödinger Equation
and Time Dependent Schrödinger Equation, respectively.
 ˆ, p
x ˆ, H ˆ refer to position, momentum and Hamiltonian operators, respectively.

That is, the symbol ^ denotes an operator.

 The symbol in all questions denotes a sum over a complete set of states.

 Ψ(x)
refers to a general state,  (x) refers to a stationary state wave function
n

for the system at time t=0, where n  1,2,3...

 In Ψ E  (x) n
no summation is implied, unless explicitly written as
n

.
E n n ( x)

 For a one-dimensional infinite square well of width a, the stationary states are
, and the allowed energies are E  n 2 2 2 /(2ma 2 ) ,
 n ( x)  2 / a sin( nx / a ) n

where n  1,2,3...
 When answering the questions, refer to the figures below for infinite and finite
square wells.

Learning Goals:

1 edited 10/17/2020
 You will learn about the properties of possible wavefunctions for a quantum
system with a given potential energy V(x) (single particle confined in one spatial
dimension)
 Wavefunction must be continuous.
 Wavefunction must be smooth (continuous in first derivative) when V(x) is
finite.
 Wavefunction must go to zero at +/- infinity.
 Wavefunction must be normalizable.
 Wavefunction must satisfy the boundary condition for the system.

 You will learn to differentiate between all possible wavefunctions and stationary
state wavefunctions.
 A possible wavefunction need NOT be symmetric or anti-symmetric even if
the potential energy is symmetric (the symmetry of the potential energy may
be reflected in the energy eigenfunctions but NOT in other possible wave
functions of the system constructed by taking linear superposition of energy
eigenfunctions).
 A possible wavefunction can be written as a superposition of stationary state
wave functions.
 ANY linear superposition of stationary state wave functions must be a
possible wavefunction if properly normalized.
 The probability density for a stationary state wave function does not change
with time. However, the probability density for other possible wavefunctions
which are not stationary state wave functions (but can be obtained from a
linear superposition of stationary states) changes with time.

 Challenging ideas.
 A possible wave function for a finite square well need not have a non-zero tail
in the classically forbidden region. One can take a linear superposition of
stationary state wave functions for a finite square well to make possible wave
functions that go to zero everywhere in the classically forbidden region.
 Scattering state wave function is not a possible wave function for a particle
since it is not normalizable.
 Although scattering state wave function with a given energy is not
normalizable, we can construct a normalizable wave function by taking a
linear superposition of these scattering state wave functions with energies
localized in a narrow range (so that the wave function close to the finite
square well looks “effectively” like a scattering state wave function with a
given energy).

2 edited 10/17/2020
QuILT for possible wavefunctions

1. What does the Hamiltonian acting on a stationary state wave function give, i.e.,
ˆ (x) =?
H n

A. It cannot be determined without knowing the Hamiltonian explicitly.

B. It cannot be determined without knowing the energy eigenfunctions explicitly.
C. where En is the energy eigenvalue of the nth state.
En n (x)

D. E (x) where E is the average energy of the system.

E. None of the above.

2. An electron is confined in a one dimensional infinite square well. Is wavefunction

1
 ( x)   1 ( x)   2 ( x) a possible wavefunction at time t=0? Choose the
2
correct answer from the statements below.
A. It is a possible wave function because it satisfies the TISE Hˆ  ( x)  E ( x)
where Ĥ is the Hamiltonian operator.
B. It is possible even if it does not satisfy Hˆ  ( x)  E ( x) .
C. It is not possible because it is not an energy eigenfunction but a linear
superposition of energy eigenfunctions.
D. It is a possible wave function for two electrons but not for one electron.
E. None of the above.

3. Is the graph below a possible wavefunction for an electron in a 1-D infinite square
well at time t=0?

3 edited 10/17/2020
A. It is a possible wave function.
B. It is not possible because it does not satisfy Hˆ  ( x)  E ( x) where Ĥ is the
Hamiltonian operator.
C. It is not possible because it is neither symmetric nor anti-symmetric about the
center of the well.
D. It is not a possible wave function because it does not satisfy the boundary
conditions for the system.
E. None of the above

4. Consider the following conversation between Sally and Harry.

Sally: I don’t understand the answer to question 3. The wavefunction is neither
symmetric nor anti-symmetric. Why is it a possible wavefunction for a symmetric
potential energy?
Harry: When the potential is symmetric, an energy eigenfunction must be symmetric
or anti-symmetric. But a superposition of energy eigenfunctions is not necessarily
symmetric. For example, the sum of an even function and an odd function is neither
even nor odd.
Sally: But the possible wave function must be an eigenstate of a particular operator,
e.g., the Hamiltonian or position operator.
Harry: That’s not true. The possible wavefunction need not be an eigenstate of a
particular operator. It can be a superposition of the eigenstates.

Do you agree with Sally or Harry? Explain.

Simulation 1: 1-D Infinite Square Well

4 edited 10/17/2020
Answer questions (a) and (b) before running the simulation.
(a) Draw the energy eigenfunctions for the ground state and the first excited
 1 ( x)

state for a 1D infinite square potential energy well.

 2 ( x)

1
(b) Predict and sketch the shape of the wavefunction  ( x)   1 ( x)   2 ( x) .
2

Double click “ejs_qm_superposition.jar” to open the simulation. You can find two
windows as shown below. The first window shows the absolute value of the
wavefunction. You can change the width of the infinite square well and start/stop the
time evolution to observe how the wavefunction changes with time. The option
“phase as color” should be selected in our tutorial. Unselecting this option would
separate the wavefunction to real and imaginary parts. The second window shows the
coefficients of the superposition wavefunction. You can input the coefficients of
different energy eigenstates (marked by “quantum #”) to build a superposition
wavefunction. “Re” is the real part and “Im” is the imaginary part of the coefficients.
You can use the button “Normalize” at the bottom of second window to normalize the
coefficients.

(c) Input 1 as the coefficient for the ground state and -1 for the first excited state in
the simulation coefficient box. Click “normalize” and observe the absolute value

5 edited 10/17/2020
 of the superposition wavefunction. Is the shape the same as your prediction in step
(b)? Explain. (Note that the simulation shows the absolute value of the
wavefunction.)

(d) What would be the approximate shape of the wavefunction

1
 ( x)   1 ( x)  2 2 ( x) at time t=0? Sketch your prediction below and
5
check your prediction with the simulation by inputting the correct coefficients.

(e) Build a superposition wavefunction with three different energy eigenstates.

Record your coefficients for each eigenstate and sketch the absolute value of the
superposition wavefunction below.

5. Choose all of the following statements that are correct about a possible
1
wavefunction  ( x)   1 ( x)   2 ( x) for an infinite square well at time t=0.
2
(I) It satisfies the boundary conditions of the infinite square well, i.e.,
 (0)  0,  (a )  0 .
(II) It is a continuous function.
(III) It has a continuous first derivative (slope) at all points except at the
well boundaries.
(IV) It is normalizable because it is zero outside the well everywhere
beyond x=0 and x=a.
A. (I) and (II) only
B. (II) and (III) only
C. (I), (II) and (III) only
D. (I), (III) and (IV) only
E. All of the above.

6. Choose all of the following statements that are correct about the wave function
below for an electron interacting with an infinite square well of width a between
d
x=0 and x=a.  (x ) and are continuous everywhere.
dx

6 edited 10/17/2020
 (I) It is a possible wave function because it is a continuous, smooth and
normalizable function that satisfies the boundary conditions.
(II) It is a possible wave function, and can be obtained by a superposition
of energy eigenfunctions according to Fourier series analysis.
(III) It is not a possible wave function because it is not symmetric about
the center of the well.
A. (I) only
B. (II) only
C. (III) only
D. (I) and (II) only
E. None of the above.

7. Choose all of the following statements that are correct about the wave function
below for an electron at given time t=0 interacting with an infinite square well of
d
width a between x=0 and x=a.  (x ) and are continuous everywhere.
dx

(I) It is a possible wave function because it is a continuous function

whose first derivative is also continuous.
(II) It is not a possible wave function because it does not go to zero at the
boundaries of the well; there is a finite probability of finding the
electron outside the well.
(III) It is not a possible wave function because it is not an energy
eigenfunction.
A. (I) only
B. (II) only
C. (III) only
D. (II) and (III) only
E. None of the above.

8. Choose all of the following statements that are correct about the wave function
1
 ( x,0)   1 ( x)   2 ( x) at time t=0 which is a superposition of the ground
2

7 edited 10/17/2020
 state and first excited wave functions for an electron interacting with a finite
square well.
(I) It is a possible wavefunction.
(II) It satisfies the TISE Ĥ  E .
1
(III) The wavefunction becomes  ( x, t )   1 ( x)   2 ( x)e iEt /  at
2
time t where E is the average energy of the system.

A. (I) only
B. (II) only
C. (III) only
D. (I) and (II) only
E. All of the above

9. Consider the following conversation between Kate and Carol:

1
Kate: I don’t understand. How come  ( x,0)   1 ( x)   2 ( x) at t=0 is a
2
possible wave function but doesn’t satisfy the TDSE or the TISE?
Carol: It does not satisfy the TISE because it is not a stationary state wave
function; it is a superposition of stationary state wave functions. It does not satisfy
the TDSE because it is only specified at time t=0. If you write  ( x.t ) you will
see that it does satisfy the TDSE.
Kate: What is  ( x, t ) ?
Carol: It is the wave function given at time t.
 ( x, t ) 
1
 1 
( x)e iE1t /    2 ( x)e iE2t /  .
2
Do you agree with Carol? Explain.

10. Substitute  ( x, t ) 
1
 1 
( x)e iE1t /    2 ( x)e iE2t /  into the left hand side and
2
 ( x, t )
the right hand side of the TDSE i  Hˆ  ( x, t ) and show that the
t
TDSE is satisfied.

8 edited 10/17/2020
Simulation 2: Time evolution of states in 1-D infinite square well
Use the same simulation program as in “Simulation 1” to answer the following
questions.
(a) Predict and sketch the shape of the absolute value of the wavefunction
1
 1 ( x)   2 ( x) at time t=0. Check your prediction with the simulation.
2

(b) When the initial wave function of an electron in a 1-D infinite square well is
1
 ( x,0)   1 ( x)   2 ( x) , would the shape of the wavefunction and
2
probability density change with time? Explain by writing down the wavefunction
 ( x, t ) at time t.

(c) Check your prediction in (b) with the simulation by starting the time evolution.
Observe the simulation result and sketch the absolute value of the wavefunction at
time t=2 units. Is the shape of the wavefunction same as at t=0?

(d) If the initial state at t=0 is , will the shape of the wavefunction change with
 1 ( x)

time?

(e) Set the initial state in the simulation to be purely . Check your prediction
 1 ( x)

in (d) by observing the time evolution result and sketching the absolute value of
wavefunction at time t=0 and t=2 unit.

11. An electron is in a one dimensional finite square well. The normalization of the
wave function implies that

9 edited 10/17/2020
 (I) The wave function should go to zero both as x   and as
x   .
(II) The probability of finding the electron anywhere in space should add
up to 1.
(III) The wave function must be zero in regions where the energy of the
system E is less than the potential energy V(x).
Choose all of the above statements that are correct.
A. (I) only
B. (II) only
C. (III) only
D. (I) and (II) only
E. All of the above

12. After normalizing the wave function, which one of the following equations is
satisfied?
A.  ( x, t ) *  ( x, t )  1

B.   ( x, t )dx  1



C.   ( x, t ) *  ( x, t )dx  1



D.  x ( x, t ) *  ( x, t )dx  1


E. None of the above.

13. An electron is interacting with a one dimensional finite square well with a wave
function  ( x,0) at t = 0. Choose all of the following statements that are correct:
(I) 2 must be symmetric about the center of the well.
 ( x , 0)

(II)  ( x,0) must reflect the symmetry of the potential energy well.
(III) Any single-valued, smooth normalizable function is a possible
wavefunction  ( x,0) .
A. (I) only
B. (II) only
C. (III) only
D. (I) and (III) only
E. (II) and (III) only

14. Choose all of the following statements that are correct about the wave function
below for an electron interacting with a finite square well of width a (between x=0
and x=a) at time t=0:

10 edited 10/17/2020
 (I) It is a possible wave function because it is the first excited state.
(II) It is a possible wave function because it is anti-symmetric about the
center of the well.
(III) It is not a possible wave function because it goes to zero at the
boundaries of the well.
(IV) It is not possible because its derivative is not continuous at the
boundaries of the well.
A. (I) only
B. (III) only
C. (IV) only
D. (I) and (II) only
E. (III) and (IV) only

Simulation 3: Finite square well

(a) Draw an approximate sketch of the energy eigenfunctions for the ground state
and the first excited state for an electron in a finite square well at
 1 ( x)  2 ( x)

time t=0.

1
(b) Predict and sketch the shape of the wave function  ( x)   1 ( x)   2 ( x) at
2
time t=0.

11 edited 10/17/2020
Double click on “bound-states.jar” to open the simulation program. You can change
the depth and width of the 1-D finite square well and select the energy level to
observe the absolute square of energy eigenfunction. You can click the button
“Superposition State” to setup the coefficients of different energy eigenstates to build
a superposition state. After you have input the coefficients in the window of
“Superposition State”, you can choose “Normalize” and then click “Apply” to observe
the new superposition state. You can also observe the time evolution of the absolute
square of the wavefunction (probability density) by clicking the button “Play/Pause”.

1
(c) Build the superposition state  ( x)   1 ( x)   2 ( x) in the simulation. Is the
2
shape of wavefunction the same as your prediction in step (b)? Explain.

12 edited 10/17/2020
(d) Build a superposition wavefunction with three different energy eigenstates.
Record your coefficient for each eigenstate and sketch the absolute value of the
superposition wavefunction below. Is your superposition wavefunction
symmetric? Explain.

15. Choose all of the following statements that are correct about the wave function
below for an electron interacting with a finite square well of width a (
d
V ( x )  V0 when 0  x  a and V ( x )  0 anywhere else).  (x ) and are
dx
continuous and single valued everywhere.

Smooth function that goes to zero within the region x=0 and x=a.

(I) It is a possible wave function because it is a continuous, smooth

normalizable function that satisfies the boundary conditions.
(II) It is not a possible wave function because it doesn’t satisfy the
boundary conditions; it goes to zero inside the well.

13 edited 10/17/2020
 (III) It is not a possible wave function because the probability of finding
the particle outside the finite square well is zero but quantum
mechanically it must be nonzero.
A. (I) only
B. (II) only
C. (III) only
D. (II) and (III) only
E. None of the above.

16. Choose all of the following statements that are correct about the wave function
below for an electron interacting with a finite square well of width a (
d
V ( x )  V0 when 0  x  a and V ( x )  0 anywhere else).  (x ) and are
dx
continuous everywhere.

(I) It is a possible wave function because it is a continuous, smooth and

normalizable function that satisfies the boundary conditions.
(II) It is not a possible wave function because it goes to zero at the
boundaries of the well.
(III) It is not a possible wave function because it is not an energy
eigenfunction.
A. (I) only
B. (II) only
C. (III) only
D. (II) and (III) only
E. None of the above.

17. Choose all of the following statements that are correct about the wave function
below for an electron interacting with a finite square well of width a (
d
V ( x )  V0 when 0  x  a and V ( x )  0 anywhere else).  (x ) and are
dx
continuous everywhere.

14 edited 10/17/2020
 (I) It is a possible wave function because it is a continuous, smooth and
normalizable function that satisfies the boundary conditions.
(II) It is not a possible wave function because it goes to zero at x=a.
(III) It is not a possible wave function because it is not an energy
eigenfunction.
(IV) It is not a possible wave function because it is not symmetric with
respect to the well.
A. (I) only
B. (II) only
C. (II) and (III) only
D. (II) and (IV) only
E. (III) and (IV) only

Simulation 4: Time evolution of a state in a finite square well

1
(a) When the initial wave function at time t=0 is  1 ( x)   2 ( x) , predict whether
2
the shape of the wavefunction will change with time. Explain by writing down the
wavefunction  ( x, t ) at time t.

Then use the same simulation program as in “Simulation 3” to answer the following
questions.

1
(b) Build the superposition state  1 ( x)   2 ( x) in the simulation and start the
2
time evolution to check your prediction in (a).

(c) When the initial state is , will the shape of the absolute square of the
 2 ( x)

wavefunction (probability density) change with time? Use the simulation to check
your prediction.

15 edited 10/17/2020
18. Sketch the ground state for a finite square well and the ground state for an infinite
square well separately. Compare and explain the similarities and differences of
these two ground states.

19. Which one of the following graphs correctly represents an energy eigenfunction of
a free particle? The function in Figure (a) below oscillates over all space (not
shown).

(a) (b)

(c) (d)

(e) None of the above

20. Is the energy eigenfunction of a free particle Aeikx normalizable? Explain. A is

a constant.

16 edited 10/17/2020
21. Can Aeikx be a possible wavefunction for a free particle? Explain. A is a
constant.

22. Consider the following conversation between Sally and Harry.

Sally: How can the energy eigenfunction not be a possible wave function for the free
particle?

Harry: Because the absolute square of the wavefunction must be normalizable.

Otherwise the total probability of finding the particle would be infinite.

Sally: I disagree. If the energy eigenfunction is not a possible wave function, that
means we cannot have a free particle with definite value of energy or momentum. But
classically we can always have a free particle moving with a constant momentum.

Harry: Well, the free particles in reality exist as wave packets. The magnitude of the
2k 2 2
momentum of a free particle is p  k and the energy is E  , where k 
2m 
is the wave vector. A wave packet could consist of plane waves e with different
ikx

wave vectors k in a very narrow range. Thus, we can consider the wave packet as
“effectively” having a definite energy and momentum, if the distribution of
energy/momentum is highly localized about a given wave vector k .

Do you agree with Sally or Harry? Explain.

23. Consider the following conversation between Harry and Sally:

Sally : My friend in the condense matter research group told me that they often use
the plane wave eikx as the wavefunctions to analyze solid state system.

Harry: Yes. Although the plane waves are not normalizable when the space expands
from   to   , it is often useful to confine particles in a box with length L and a
periodical boundary condition ( ( x  L)   ( x) ). For a particle confined in a box
with length L and a periodical boundary condition, the stationary state wavefunction

17 edited 10/17/2020
 1 ikx n
could be written as e , where k  is discrete. However, for a macroscopic
L L
system, L is very large ( mm , cm as opposed to 10 8 m which is the size of the
atom) so that k could be considered as “quasi” continuous. We often use a large but
finite box with periodical boundary condition to study the bulk properties of a
macroscopic physical system. In this case, for particle confined in one spatial
1 ikx
dimension, using plane wave e as a possible wavefunction would be
L
acceptable.

1 ikx 1
Sally: For the wavefunction e , is the normalization constant, right?
L L
L
Harry: Yes. You can calculate the integral result of 1 ikx 1  ikx
 e e dx to prove
0 L L

that this wave function is normalizable.

Do you agree with Harry? Explain. Write down the integral result that Harry
mentioned above.

Simulation 5. Free particle

Double click the “qm_FreeParticleWavepacket.jar” to open the simulation program.
You can see the “Position Space” windows as below. (Another window “Momentum
Space” may also pop up. Just close it since we only need to observe the position space
wave packet in this tutorial.) The “position space” window shows the absolute value
of the wave packet in position space. And you can start/stop the time evolution of the
absolute value of wavefunction or reset the initial wave packet.

18 edited 10/17/2020
(a) Sketch the wave packet in the simulation at time t=0. Will the shape of the
absolute value of wave packet change with time? Explain.

(b) Start the time evolution and observe the wave packet at time t=1 unit. Is this result
consistent with your prediction in (a)? Explain.

24. Consider the following conversation between Sally and Harry.

Sally: Why does the wave packet in the simulation spread out with time?

Harry: Because the wave packet is not a stationary state wavefunction. As shown in
the graph below1, the wave packet consists of different energy eigenstates with
different wave vectors k . The different energy eigenstates would evolve with time at
different phase velocities. In other words, the magnitude of phase velocity
 E
vp   is different for the different components of the wave packet. Therefore,
k p
the wave packet cannot remain localized.

Do you agree with Harry? Explain.

1
The image is edited based on the graphs at http://hyperphysics.phy-astr.gsu.edu/Hbase/Waves/wpack.html#c2
and http://www.physics.csbsju.edu/QM/fall.12.html

19 edited 10/17/2020
  2k 2
25. Use the expression for the free particle energy eigenvalue Ek  to find the
2m
magnitude of phase velocity in terms of wave vector k .

26. Construct a general wave packet for a free particle at time t. Remember that for
discrete energies and stationary states (n=1, 2, … ) , we can write a
En n 

general wavefunction at time t as  ( x, t )   cn e  iE nt
 n ( x ) , where cn are
/

n 1

the time-independent coefficients for each stationary state. You can use this
analogy to find  ( x, t ) for a free particle except that the scattering states are
continuous instead of discrete. So the summation should be replaced with an
integral when constructive the wave packet. The energy eigenfunctions for a free
 2k 2
particle are of the form eikx with energy eigenvalues Ek  .
2m

20 edited 10/17/2020