Wave Mechanics Lecture
Wave Mechanics Lecture
Wave Mechanics Lecture
Wave Mechanics
Semester 1 2009
J D Cresser
[email protected]
Rm C5C357
Ext 8913
Semester 1 2009 PHYS201 Wave Mechanics 1 / 86
Quantum Mechanics and Wave Mechanics I
Quantum mechanics arose out of the need to provide explanations for a range of
physical phenomena that could not be accounted for by classical physics:
Black body spectrum
Spectra of the elements
photoelectric effect
Specic heat of solids . . .
Through the work of Planck, Einstein and others, the idea arose that electromagnetic
and other forms of energy could be exchanged only in denite quantities (quanta)
And through the work of De Broglie the idea arose that matter could exhibit wave like
properties.
It was Einstein who proposed that waves (light) could behave like particles (photons).
Heisenberg proposed the rst successful quantum theory, but in the terms of the
mathematics of matrices matrix mexhanics.
Semester 1 2009 PHYS201 Wave Mechanics 2 / 86
Quantum Mechanics and Wave Mechanics II
Schr odinger came up with an equation for the waves predicted by de Broglie, and that
was the start of wave mechanics.
Schr odinger also showed that his work and that of Heisenbergs were mathematically
equivalent.
But it was Heisenberg, and Born, who rst realized that quantum mechanics was a
theory of probabilities.
Something that Einstein would never accept, even though he helped discover it!
Through the work of Dirac, von Neumann, Jordan and others, it was eventually shown
that matrix mechanics and wave mechanics were but two forms of a more fundamental
theory quantum mechanics.
Quantum mechanics is a theory of information
It is a set of laws about the information that can be gained about the physical world.
We will be concerned with wave mechanics here, the oldest form of quantum
mechanics.
Semester 1 2009 PHYS201 Wave Mechanics 3 / 86
The Black Body Spectrum I
Environment at temperature T
Black body at temperature T
A
A
A
A
A
A
A
AU
1
f
S
Observed spectrum
Rayleigh-Jeans spectrum
1
The rst inkling of a new physics lay in
understanding the origins of the black
body spectrum:
A black body is any object that absorbs
all radiation, whatever frequency, that falls
on it.
The blackbody spectrum is the spectrum
of the radiation emitted by the object when
it is in thermal equilibrium with its
surroundings.
Classical physics predicted the
Rayleigh-Jeans formula which diverged at
high frequencies the ultra-violet
catastrophe.
Semester 1 2009 PHYS201 Wave Mechanics 4 / 86
The Black Body Spectrum II
Planck assumed that matter could absorb or emit black body radiation energy of
frequency f in multiples of hf where h was a parameter to be taken to zero at the end
of the calculation.
But he got the correct answer for a small but non-zero value of h:
S(f , T) =
8hf
3
c
3
1
exp(hf /kT) 1
.
where
h = 6.6218 10
34
Joules-sec
now known as Plancks constant.
Semester 1 2009 PHYS201 Wave Mechanics 5 / 86
Einsteins hypothesis
Einstein preferred to believe that the formula hf was a property of the electromagnetic
eld
Planck thought it was a property of the atoms in the blackbody.
Einstein proposed that light of frequency f is absorbed or emitted in packets (i.e.
quanta) of energy E where
E = hf Einstein/Planck formula
Sometime later he extended this idea to say that light also had momentum, and that it was
in fact made up of particles now called photons.
Einstein used this idea to explain the photo-electric effect (in 1905).
A rst example of wave-particle duality: light, a form of wave motion, having
particle-like properties.
Semester 1 2009 PHYS201 Wave Mechanics 6 / 86
The Bohr model of the hydrogen atom I
@
@
@
@
@
@
@I
n=1
n=2
n=3
n=4
emitted
photons
Rutherford proposed that an atom consisted of a
small positively charged nucleus with electrons in
orbit about this nucleus.
According to classical EM theory, the orbiting
electrons are accelerating and therefore ought to
emit EM radiation.
The electron in a hydrogen atom should spiral into the
nucleus in about 10
12
sec.
Bohr proposed existence of stable orbits (stationary states) of radius r such that
angular momentum = pr = n
h
2
, n = 1, 2, 3, . . .
Leads to orbits of radii r
n
and energies E
n
:
r
n
= n
2
0
h
2
e
2
m
e
E
n
=
1
n
2
m
e
e
4
8
2
0
h
2
n = 1, 2, . . .
Semester 1 2009 PHYS201 Wave Mechanics 7 / 86
The Bohr model of the hydrogen atom II
Bohr radius radius of lowest energy orbit a
0
= r
1
=
0
h
2
e
2
m
e
= 0.05 nm.
The atom emits EM radiation by making a transition between stationary states,
emitting a photon of energy hf where
hf = E
m
E
n
=
m
e
e
4
8
2
0
h
2
_
1
n
2
1
m
2
_
Einstein later showed that the the photon had to carry both energy and momentum hence
it behaves very much like a particle.
Einstein also showed that the time at which a transition will occur and the direction of
emission of the photon was totally unpredictable.
The rst hint of uncaused randomness in atomic physics.
Bohrs model worked only for hydrogen-like atoms
It failed miserably for helium
It could not predict the dynamics of physical systems
But it broke the ice, and a search for classical physics answers was abandoned.
Semester 1 2009 PHYS201 Wave Mechanics 8 / 86
De Broglies Hypothesis
In 1924 Prince Louis de Broglie asked a philosophical question:
If waves can sometimes behave like particles, can particles sometimes exhibit
wave-like behaviour?
To specify such a wave, need to know its
frequency f use Einsteins formula E = hf
wavelength make use of another idea of Einsteins
Special relativity tells us that the energy E of a zero rest mass particle (such as a
photon) is given in terms of its momentum p by
E = pc = hf .
Using the fact that f = c gives pc = hc/
which can be solved for = h/p.
As there is no mention in this equation of anything to do with light, de Broglie assumed
it would hold for particles of matter as well.
Semester 1 2009 PHYS201 Wave Mechanics 9 / 86
Evidence for wave properties of matter
50
interference maximum
of reected electrons
nickel crystal
incident electron
beam of energy
54 eV
=
2r
3
3
r
3
In 1926 Davisson and Germer accidently showed
that electrons can be diffracted (Bragg diffraction)
Fired beam of electrons at nickel crystal and
observed the scattered electrons.
They observed a diffraction pattern identical to that
observed when waves (X-rays) were red at the
crystal.
The wavelengths of the waves producing the
diffraction pattern was identical to that predicted by
the de Broglie relation.
G P Thompson independently carried out the
same experiment. His father (J J Thompson) had
shown that electrons were particles, the son
showed that they were waves.
Fitting de Broglie waves around a circle gives Bohrs quantization condition
n = 2r
n
= pr
n
= n
h
2
Semester 1 2009 PHYS201 Wave Mechanics 10 / 86
The Wave Function
Accepting that these waves exist, we can try to learn what they might mean.
We know the frequency and the wavelength of the wave associated with a particle of
energy E and p.
Can write down various formulae for waves of given f and .
Usually done in terms of angular frequency and wave number k:
= 2f k = 2/
In terms of = h/2
E = hf = (h/2)(2f ) and p = h/ = (h/2)(2/)
i.e. E = and p = k
Some possibilities for these waves:
(x, t) =Acos(kx t)
or =Asin(kx t)
or =Ae
i(kxt)
or =. . .
But what is doing the waving?? What are these waves made of?
In absence of any better knowledge give a generic name:
the wave function.
Semester 1 2009 PHYS201 Wave Mechanics 11 / 86
What can we learn from the wave function? I
To help us understand what might be, let us reverse the argument
I.e. if we are given (x, t) = Acos(kx t), what can we learn about the particle?
This wave will be moving with a phase velocity given by v
phase
=
k
.
Phase velocity is the speed of the crests of the wave.
Using E = =
1
2
mv
2
and p = k = mv where v is the velocity of the particle we get:
v
phase
=
k
=
E
p
=
1
2
mv
2
mv
=
1
2
v
That is, the wave is moving with half the speed of its associated particle.
Semester 1 2009 PHYS201 Wave Mechanics 12 / 86
What can we learn from the wave function? II
Thus, from the phase velocity of the wave, we can learn the particles velocity via
v = 2v
phase
.
What else can we learn E and p, of course.
But what about the position of the particle?
(x, t)
x
v
phase
-
1
This is a wave function of constant amplitude and wavelength. The wave is the same
everywhere and so there is no distinguishing feature that could indicate one possible
position of the particle from any other.
Thus, we cannot learn where the particle is from this wave function.
Semester 1 2009 PHYS201 Wave Mechanics 13 / 86
Wave Packets I
We can get away from a constant amplitude wave by adding together many waves with
a range of frequencies and wave numbers.
cos(5x) + cos(5.25x)
cos(4.75x) + cos(4.875x) + cos(5x) + cos(5.125x) + cos(5.25x)
cos(4.8125x) + cos(4.875x) + cos(4.9375x) + cos(5x) + cos(5.0625x) + cos(5.125x) + cos(5.1875x)
An integral over a continuous range of wave numbers produces a single wave packet.
1
Semester 1 2009 PHYS201 Wave Mechanics 14 / 86
Wave Packets II
To form a single wave packet we need to carry out the sum (actually, an integral):
(x, t) =A(k
1
) cos(k
1
x
1
t) + A(k
2
) cos(k
2
x
2
t) + . . .
=
n
A(k
n
) cos(k
n
x
n
t)
1'-1"
8"
(a) (b)
2k
2x
x k
k
(x, t)
A(k)
1
(a) The distribution of wave numbers k of harmonic waves contributing to the wave
function (x, t). This distribution is peaked about k with a width of 2k.
(b) The wave packet (x, t) of width 2x resulting from the addition of the waves with
distribution A(k). The oscillatory part of the wave packet (the carrier wave) has wave
number k.
Semester 1 2009 PHYS201 Wave Mechanics 15 / 86
Wave packets III
2x
-v
group
x
(x, t)
9
1
Particle in here
somewhere??
The wave function of a wave
packet is effectively zero
everywhere except in a region of
size 2x.
Reasonable to expect particle to
be found in region where wave
function is largest in magnitude.
The wave packet ought to behave in some way like its associated particle e.g. does it
move like a free particle.
We can check the speed with which the packet moves by calculating its group velocity
v
group
=
_
d
dk
_
k=k
.
From E =
p
2
2m
we get =
2
k
2
2m
and hence =
k
2
2m
a dispersion relation which gives v
group
=
k
m
=
p
m
The wave packet moves as a particle of mass m and momentum p = k.
Semester 1 2009 PHYS201 Wave Mechanics 16 / 86
Heisenberg Uncertainty Relation I
Thus the wave packet is particle-like:
It is conned to a localized region is space . . .
It moves like a free particle.
But at the cost of combining together waves with wave numbers over the range
k k k k + k
This corresponds to particle momenta over the range
p p p p + p
and which produces a wave packet that is conned to a region of size 2x.
I.e. the wave packet represents a particle to a certain extent, but it does not x its
momentum to better than p nor x its position to better than x.
The bandwidth theorem (or, Fourier theory) tells us that xk 1.
Multiplying by gives the Heisenberg Uncertainty Relation.
xp
Semester 1 2009 PHYS201 Wave Mechanics 17 / 86
Heisenberg Uncertainty Relation II
Heisenbergs uncertainty relation is a direct result of the basic principles of quantum
mechanics
It is saying something very general about the properties of particles
No mention is made of the kind of particle, what it is doing, how it is created . . .
It is tells us that we can have knowledge of the position of a particle to within an uncertainty
of x, and its momentum to within an uncertainty p, with xp
If we know the position exactly x = 0, then the momentum uncertainty is innite and vice versa
It does not say that the particle denitely has a precise position and momentum, but we
cannot measure these precise values.
It can be used to estimate size of an atom, the lowest energy of simple harmonic
oscillator, temperature of a black hole, the pressure in the centre of a collapsing star . .
.
Semester 1 2009 PHYS201 Wave Mechanics 18 / 86
The size of an atom
p
p
a
electron
proton
Can estimate the size of an atom by use of the
uncertainty relation.
The position of the electron with respect to the
nucleus will have an uncertainty x = a
The momentum can swing between p and p so will
have an uncertainty of p = p
Using the uncertainty relation in the form xp we get ap .
The total energy of the atom is then
E =
p
2
2m
e
2
4
0
a
2
2ma
2
e
2
4
0
a
The radius a that minimizes this is found from
dE
da
= 0, which gives
a
4
0
2
me
2
0.05 nm (Bohr radius) and E
min
1
2
me
4
(4
0
)
2
2
13.6 eV.
Semester 1 2009 PHYS201 Wave Mechanics 19 / 86
THE TWO SLIT EXPERIMENT
Shall be considered in three forms:
With macroscopic particles (golf balls);
With classical waves (light waves);
With electrons.
The rst two merely show us what we would expect in our classical world.
The third gives counterintuitive results with both wave and particle characteristics that
have no classical explanation.
Semester 1 2009 PHYS201 Wave Mechanics 20 / 86
An Experiment with Golf balls: Both Slits Open.
We notice three things about this experiment:
Erratic
golfer
Slit 1
Slit 2
Fence
The golf balls arrive in lumps: for each golf
ball hit that gets through the slits, there is a
single impact on the fence.
The golf balls arrive at random.
If we wait long enough, we nd that the golf ball arrivals tend to form a pattern:
Erratic
golfer
Slit 1
Slit 2
Fence
This is more or less as we might expect:
The golf balls passing through hole 1 pile
up opposite hole 1 . . .
The golf balls passing through hole 2 pile
up opposite hole 2.
Semester 1 2009 PHYS201 Wave Mechanics 21 / 86
An Experiment with golf balls: one slit open
.
If we were to perform the experiment with one or the other of the slits closed, would
expect something like:
Slit 1
blocked
Slit 2
blocked
i.e. golf ball arrivals accumulate opposite
the open slits.
expect that the result observed with both
slits open is the sum of the results
obtained with one and then the other slit
open.
Express this sum in terms of the probability
of a ball striking the fence at position x.
Semester 1 2009 PHYS201 Wave Mechanics 22 / 86
Probability for golf ball strikes
Can sketch in curves to represent the probability of a golf ball striking the back fence at
position x.
Thus, for instance,
P
12
(x)x = probability of a golf ball striking screen between x and x + x when both slits
are open.
P
1
(x) P
2
(x) P
12
(x)
Now make the claim that, for golf balls: P
12
(x) = P
1
(x) + P
2
(x)
Semester 1 2009 PHYS201 Wave Mechanics 23 / 86
Probability distributions of golf ball strikes . . .
Golf balls behave like classical particles
The fact that we add the probabilities P
1
(x) and P
2
(x) to get P
12
(x) is simply stating:
The probability of a golf ball that goes through slit 1 landing in (x, x + x) is completely
independent of whether or not slit 2 is open.
The probability of a golf ball that goes through slit 2 landing in (x, x + x) is completely
independent of whether or not slit 1 is open.
The above conclusion is perfectly consistent with our classical intuition.
Golf balls are particles: they arrive in lumps;
They independently pass through either slit 1 or slit 2 before striking the screen;
Their random arrivals are due to erratic behaviour of the source (and maybe random
bouncing around as they pass through the slit), all of which, in principle can be measured
and allowed for and/or controlled.
Semester 1 2009 PHYS201 Wave Mechanics 24 / 86
An Experiment with Waves
Now repeat the above series of experiments with waves shall assume light waves of
wavelength .
Shall measure the time averaged intensity of the electric eld E(x, t) = E(x) cos t of
the light waves arriving at the point x on the observation screen.
The time averaged intensity will then be
I(x) = E(x, t)
2
=
1
2
E(x)
2
Can also work with a complex eld
c(x, t) =
1
2
E(x)e
it
so that
E(x, t) =
2Rec(x, t)
and
I(x) = c
_
=I
1
(x) + I
2
(x) + 2
_
I
1
(x)I
2
(x) cos
Last term is the interference term which explicitly depends on slits separation d the
waves probe the presence of both slits.
We notice three things about these experiments:
The waves arrive everywhere at once, i.e. they do not arrive in lumps.
The single slit result for waves similar to the single slit result for golf balls.
We see interference effects if both slits open for waves, but not for golf balls.
Semester 1 2009 PHYS201 Wave Mechanics 27 / 86
An experiment with electrons
Now repeat experiment once again, this time with electrons.
Shall assume a beam of electrons, all with same energy E and momentum p incident
on a screen with two slits.
Shall also assume weak source: electrons pass through the apparatus one at a time.
Electrons strike a uorescent screen (causing a ash of light) whose position can be
monitored.
Semester 1 2009 PHYS201 Wave Mechanics 28 / 86
An experiment with electrons: one slit open
Electron gun
Slit 1
Slit 2
uorescent ashes
P
2
(x)
P
1
(x)
A
A
A
A
A
AK
*
With one slit open observe electrons striking uorescent screen in a random fashion,
but mostly directly opposite the open slit exactly as observed with golf balls.
Can construct probability distributions P
1
(x) and P
2
(x) for where electrons strike, as for
golf balls.
Apparent randomness in arrival of electrons at the screen could be put down to
variations in the electron gun (cf. erratic golfer for golf balls).
Semester 1 2009 PHYS201 Wave Mechanics 29 / 86
An experiment with electrons: both slits open
P
12
(x)
Two slit interfer-
ence pattern!
Electron gun
The following things can be noted:
Electrons strike the screen causing individual ashes, i.e. they arrive as particles, just as
golf balls do;
They strike the screen at random same as for golf balls.
Can construct probability histograms P
1
(x), P
2
(x) and P
12
(x) exactly as for golf balls.
Find that the electron arrivals, and hence the probabilities, form an interference pattern
as observed with waves:
P
12
(x) = P
1
(x) + P
2
(x) + 2
_
P
1
(x)P
2
(x) cos
_
2d sin
_
P
1
(x) + P
2
(x) The expected result for particles.
Semester 1 2009 PHYS201 Wave Mechanics 30 / 86
An experiment with electrons: both slits open
Interference of de Broglie waves
Problem: we have particles arriving in lumps, just like golf balls, i.e. one at a time at
localised points in space . . .
but the pattern formed is that of waves . . .
and waves must pass through both slits simultaneously, and arrive everywhere on the
observation screen at once to form an interference pattern.
Moreover, from the pattern can determine that = h/p the de Broglie relation.
Interference of de Broglie waves seems to have occurred here!
Semester 1 2009 PHYS201 Wave Mechanics 31 / 86
What is going on here?
If electrons are particles (like golf balls) then each one must go through either slit 1 or
slit 2.
A particle has no extension in space, so if it passes through slit 1 say, it cannot possibly be
affected by whether or not slit 2, a distance d away, is open.
Thats why we claim we ought to nd that
P
12
(x) = P
1
(x) + P
2
(x)
but we dont.
In fact, the detailed structure of the interference pattern depends on d, the separation
of the slits.
So, if an electron passes through slit 1, it must somehow become aware of the presence of
slit 2 a distance d away, in order to know where to land on the observation screen so as to
produce a pattern that depends on d.
i.e. the electrons would have to do some strange things in order to know about the
presence of both slits such as travel from slit 1 to slit 2 then to the observation screen.
Maybe we have to conclude that it is not true that the electrons go through either slit 1
or slit 2.
Semester 1 2009 PHYS201 Wave Mechanics 32 / 86
Watching the electrons
We can check which slit the electrons go through by watching next to each slit and
taking note of when an electron goes through each slit.
If we do that, we get the alarming result that the interference pattern disappears we
regain the result for bullets.
It is possible to provide an explanation of this result in terms of the observation
process unavoidably disturbing the state of the electron.
Such explanations typically rely on invoking the Heisenberg Uncertainty Principle.
Semester 1 2009 PHYS201 Wave Mechanics 33 / 86
The Heisenberg Microscope
electron momentum p
de Broglie wavelength
= h/p
Weak incident light eld: minimum of one
photon scatters off electron.
In order to distinguish the images of the
slits in the photographic plate of the
microscope, require wavelength of photon
at least
l
d. Momentum of photon is
then h/d
Photon-electron collision gives electron a
sideways momentum kick of p = h/d.
So electron deected by angle of
p/p (h/. d/h) = d/
is approximately the angular separation between an interference maximum and a
neighbouring minimum!
Semester 1 2009 PHYS201 Wave Mechanics 34 / 86
An uncertainty relation explanation
Recall the Heisenberg uncertainty relation for a particle:
xp
1
2
_
=I
1
(x) + I
2
(x) + 2
_
I
1
(x)I
2
(x) cos
Note that the interference term occurs because waves from two slits are added, and then
the square is taken of the sum.
The time average is not an important part of the argument it is there simply as we are
dealing with rapidly oscillating light waves.
So, it looks like the interference pattern for electrons comes about because waves from
each slit are being added, and then the square taken of the sum!!
Semester 1 2009 PHYS201 Wave Mechanics 37 / 86
Interference of de Broglie waves II
Moreover, the wavelength of these waves in case of electrons is the de Broglie
wavelength = h/p
So can be understood electron interference pattern as arising from the interference of
de Broglie waves, each emanating from slit 1 or slit 2, i.e.
12
(x, t) =
1
(x, t) +
2
(x, t)
and hence
P
12
(x) =
1
(x, t) +
2
(x, t)
2
=
1
(x, t) +
2
(x, t)
2
=
1
(x, t)
2
+
2
(x, t)
2
+ 2Re
_
1
(x, t)
2
(x, t)
_
=P
1
(x) + P
2
(x) + 2
_
P
1
(x)P
2
(x) cos .
Probabilities are time independent as we have averaged over a huge number of
electron arrivals.
Note that we have taken the modulus squared since (with the benet of hindsight) the
wave function often turns out to be complex.
Semester 1 2009 PHYS201 Wave Mechanics 38 / 86
Born interpretation of the wave function
But
12
(x, t) is the wave function at the point x at time t.
And P
12
(x) =
12
(x, t)
2
is the probability of the particle arriving at x (at time t)
So squaring the wave function a wave amplitude gives the probability of the particle
being observed at a particular point in space.
This, and other experimental evidence, leads to the interpretation:
P(x, t) = (x, t)
2
dx = probability of nding the particle in region (x, x + dx) at time t.
This is the famous Born probability interpretation of the wave function.
As P(x, t) is a probability, then (x, t) is usually called a probability amplitude (as well as the
wave function).
From this interpretation follows the result that energy is quantized!
Semester 1 2009 PHYS201 Wave Mechanics 39 / 86
Probability interpretation of the wave function I
The Born interpretation of the wave function (x, t) is
P(x, t) = (x, t)
2
dx = probability of nding the particle in region (x, x + dx) at time t.
So what does probability mean?
To illustrate the idea, consider the following
setup:
()ve charge
(+)ve mirror charge
x
liquid
helium
A negatively charged electron oats above a
sea of liquid helium.
It is attracted by its mirror image charge
But repelled by the potential barrier at the surface
of the liquid
Classical physics says that it should come to rest
on the surface.
Quantum mechanics says otherwise.
Suppose we have N = 300 identical copies of the
same arrangement, and we measure the height
of the electron above the surface.
Semester 1 2009 PHYS201 Wave Mechanics 40 / 86
Probability interpretation of the wave function II
Suppose we divide the distance above the surface of the liquid into small segments of
size x
Measure the height of the electron in each of the
N identical copies of the experimental
arrangement.
Let N(x) be the number of electrons found to be
in the range x to x + x, and form the ratio
N(x)
Nx
x
a
0
no. of detections N(x) in in-
terval (x, x + x)
0 28
0.5 69
1.0 76
1.5 58
2.0 32
2.5 17
3.0 11
3.5 6
4.0 2
4.5 1
Distance x from surface in units of a
0
P(x, t) = |(x, t)|
2
= 4(x
2
/a
3
0
)e
2x/a0
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0
0.1
0.2
0.3
0.4
0.5
0.6
N(x)
Nx
Data on the left plotted as a histogram.
The ratio
N(x)
N
is the fraction of all the electrons
found in the range x to x + x.
In other words
N(x)
N
probability of nding an electron in
the range x to x + x
Semester 1 2009 PHYS201 Wave Mechanics 41 / 86
Probability interpretation of the wave function III
So the probability of nding the electron at a certain distance from the liquid helium
surface
is just the fraction of the times that it is found to be at that distance if we repeat the
measurement many many times over under identical conditions.
The results of the measurements will vary randomly from one measurement to the next.
According to the Born interpretation:
N(x)
N
probability of nding an electron in the range x to x + x
P(x, t)x
(x, t)
2
x
where (x, t) is the wave function for the electron.
Since the electron must be found somewhere between and , then
all x
N(x)
N
=
N
N
= 1
Semester 1 2009 PHYS201 Wave Mechanics 42 / 86
Probability interpretation of the wave function IV
Or, in terms of the wave function, for any particle conned to the x axis, we must have:
_
+
(x, t)
2
dx = 1.
This is known as the normalization condition.
All acceptable wave functions must be normalized to unity.
This means that the integral must converge
Which tells us that the wave function must vanish as x .
This is an important conclusion with enormous consequences.
There is one exception: the harmonic wave functions like Acos(kx t) which go on forever.
See later for their interpretation.
The requirement that (x, t) 0 as x is known as a boundary condition.
Used when nding the solution to the Schr odinger equation.
Semester 1 2009 PHYS201 Wave Mechanics 43 / 86
Expectation value
The results of measuring the position of a particle varies in a random way from one
measurement to the next.
We can use the standard tools of statistics to work out such things as the average
value of all the results, or their standard deviation and so on.
Thus, suppose we have data for the position of a particle in the form
N(x)
N
= fraction of all measurements that lie in the range (x, x + x).
The average of all the results will then be
(x)
all x
x
N(x)
N
all x
xP(x, t)x
all x
x(x, t)
2
x.
So, in the usual way of taking a limit to form an integral:
(x) =
_
+
x(x, t)
2
dx
known as the expectation value of the position of the particle.
Semester 1 2009 PHYS201 Wave Mechanics 44 / 86
Uncertainty
We can also calculate the expectation value of any function of x:
(f (x)) =
_
+
f (x)(x, t)
2
dx.
A most important example is (x
2
):
(x
2
) =
_
+
x
2
(x, t)
2
dx
because then we can dene the uncertainty in the position of the particle:
(x)
2
= ((x (x))
2
) = (x
2
) (x)
2
.
This is a measure of how widely spread the results are around the average, or
expectation, value (x).
The uncertainty x is just the standard deviation of a set of randomly scattered results.
Semester 1 2009 PHYS201 Wave Mechanics 45 / 86
Example of expectation value and uncertainty I
Can use the data from before (where a
0
7.6 nm)
x
a
0
no. of detections N(x) in
interval (x, x + x)
0 28
0.5 69
1.0 76
1.5 58
2.0 32
2.5 17
3.0 11
3.5 6
4.0 2
4.5 1
The average value of the distance of the electron
from the surface of the liquid helium will be
(x) 0
28
300
+ 0.5a
0
69
300
+ a
0
76
300
+ 1.5a
0
58
300
+ 2a
0
32
300
+ 2.5a
0
17
300
+ 3a
0
11
300
+ 3.5a
0
6
300
+ 4a
0
2
300
+ 4.5a
0
1
300
= 1.235a
0
.
This can be compared with the result that follows for
the expectation value calculated from the wave
function for the particle:
(x) =
_
+
x (x, t)
2
dx =
4
a
3
0
_
0
x
3
e
2x/a
0
dx = 1.5a
0
.
Semester 1 2009 PHYS201 Wave Mechanics 46 / 86
Example of expectation value and uncertainty II
Similarly for the uncertainty: (x)
2
all x
(x (x))
2
N(x)
N
Using the data from earlier:
(x)
2
(0 1.235)
2
a
2
0
28
300
+ (0.5 1.235)
2
a
2
0
69
300
+ (1 1.235)
2
a
2
0
76
300
+ (1.5 1.235)
2
a
2
0
58
300
+ (2 1.235)
2
a
2
0
32
300
+ (2.5 1.235)
2
a
2
0
17
300
+ (3 1.235)
2
a
2
0
11
300
+ (3.5 1.235)
2
a
2
0
6
300
+ (4 1.235)
2
a
2
0
2
300
+ (4.5 1.235)
2
a
2
0
1
300
= 0.751a
2
0
which gives x = 0.866a
0
.
The exact result using the wave function turns out to be the same.
Semester 1 2009 PHYS201 Wave Mechanics 47 / 86
Particle in innite potential well I
Consider a single particle of mass m conned to within a region 0 < x < L with
potential energy V = 0 bounded by innitely high potential barriers, i.e. V = for x < 0
and x > L.
x = 0 x = L
V = V =
forbidden
region
forbidden
region
V = 0
This simple model is sufcient to describe
(in one dimension):
1. the properties of the conduction electrons
in a metal (in the so-called free electron
model).
2. properties of gas particles in an ideal gas
where the particles do not interact with
each other.
More realistic models are three dimensional, and the potential barriers are not innitely
high (i.e. the particle can escape).
Semester 1 2009 PHYS201 Wave Mechanics 48 / 86
Deriving the boundary conditions
The rst point to note is that, because of the innitely high barriers, the particle cannot
be found in the regions x > L and x < 0.
The wave function has to be zero in these regions to guarantee that the probability of nding
the particle there is zero.
If we make the not unreasonable assumption that the wave function has to be
continuous, then we must conclude that
(0, t) = (L, t) = 0.
These conditions on (x, t) are known as boundary conditions.
These boundary conditions guarantee that the particle wave function vanishes as
x .
The wave function is zero even before we get to innity!
Semester 1 2009 PHYS201 Wave Mechanics 49 / 86
Constructing the wave function I
Between the barriers, the energy E of the particle is purely kinetic, i.e. E =
p
2
2m
.
Using the de Broglie relation p = k we then have E =
2
k
2
2m
k =
2mE
2mE
.
The negative value will be written k.
From the Einstein-Planck relation E = we also have = E/.
So we have the wave number(s) and the frequency of the wave function, but we have
to nd a wave function that
Satises the boundary conditions (0, t) = (L, t) = 0
And is normalized to unity.
Semester 1 2009 PHYS201 Wave Mechanics 50 / 86
Constructing the wave function II
In the region 0 < x < L the particle is free, so the wave function must be of the form
(x, t) = Acos(kx t)
Or perhaps a combination of such wave functions as we did to construct a wave packet.
In fact , we have to try a combination as a single term like (x, t) = Acos(kx t) is not zero
at x = 0 or x = L for all time.
Further, we have the picture of the particle bouncing back and forth between the walls, so
waves travelling in both directions are probably needed.
The easiest way to do this is to guess
(x, t) = Ae
i(kxt)
+ Be
i(kxt)
.,,.
+ Ce
i(kx+t)
+ De
i(kx+t)
.,,.
wave travelling to the right wave travelling to the left
A, B, C, and D are coefcients that we wish to determine from the boundary conditions
and from the requirement that the wave function be normalized to unity for all time.
Semester 1 2009 PHYS201 Wave Mechanics 51 / 86
Applying the boundary conditions I
Quantization of energy
First, consider the boundary condition at x = 0. Here, we must have
(0, t) = Ae
it
+ Be
it
+ Ce
it
+ De
it
= (A + D)e
it
+ (B + C)e
it
= 0.
This must hold true for all time, which can only be the case if A + D = 0 and B + C = 0.
Thus we conclude that we must have
(x, t) = Ae
i(kxt)
+ Be
i(kxt)
Be
i(kx+t)
Ae
i(kx+t)
= A(e
ikx
e
ikx
)e
it
B(e
ikx
e
ikx
)e
it
= 2i sin(kx)(Ae
it
Be
it
).
Semester 1 2009 PHYS201 Wave Mechanics 52 / 86
Applying the boundary conditions II
Quantization of energy
Now check on the other boundary condition, i.e. that (L, t) = 0, which leads to:
sin(kL) = 0
and hence
kL = n n an integer
This implies that k can have only a restricted set of values given by
k
n
=
n
L
n = 1, 2, 3, . . .
The negative n do not give a new solution, so we exclude them.
An immediate consequence of this is that the energy of the particle is limited to the
values
E
n
=
2
k
2
n
2m
=
2
n
2
2
2mL
2
=
n
n = 1, 2, 3, . . .
i.e. the energy is quantized.
Semester 1 2009 PHYS201 Wave Mechanics 53 / 86
Normalizing the wave function I
Now check for normalization:
_
+
(x, t)
2
dx = 4
Ae
it
Be
it
2
_
L
0
sin
2
(k
n
x) dx = 1
We note that the limits on the integral are (0, L) since the wave function is zero outside that
range.
This integral must be equal to unity for all time, i.e. the integral cannot be a function of
time. But
Ae
it
Be
it
2
=
_
Ae
it
Be
it
__
A
e
it
B
e
it
_
= AA
+ BB
AB
e
2it
A
Be
2it
which is time dependent!!
Can make sure the result is time independent by supposing either A = 0 or B = 0.
It turns out that either choice can be made we will make the conventional choice and put
B = 0.
Semester 1 2009 PHYS201 Wave Mechanics 54 / 86
Normalizing the wave function II
The wave function is then
(x, t) = 2iAsin(k
n
x)e
it
but we still havent nished normalizing the wave function!
I.e. we still have to satisfy
_
+
(x, t)
2
dx = 4A
2
_
L
0
sin
2
(k
n
x) = 2A
2
L = 1
This tells us that
A =
_
1
2L
e
i
where is an unknown phase factor.
Nothing we have seen above can give us a value for , but it turns out not to matter (it
always found to cancel out).
Semester 1 2009 PHYS201 Wave Mechanics 55 / 86
Normalizing the wave function III
We can choose to suit ourselves so we will choose = /2 and hence
A = i
_
1
2L
.
The wave function therefore becomes
n
(x, t) =
_
2
L
sin(nx/L)e
i
n
t
0 < x < L
= 0 x < 0, x > L.
with associated energies E
n
=
2
n
2
2
2mL
2
n = 1, 2, 3, . . . .
The wave function and the energies have been labelled by the quantity n, known as a
quantum number.
The quantum number n can have the values n = 1, 2, 3, . . . (negative numbers do not
give anything new)
n = 0 is excluded, for then the wave function vanishes everywhere there is no particle!
Semester 1 2009 PHYS201 Wave Mechanics 56 / 86
Comparison with classical energies
The particle can only have discrete energies E
n
The lowest energy, E
1
is greater than zero, as is required by the uncertainty principle.
Recall uncertainty in position is x
1
2
L and uncertainty in momentum is p p, so, by the
uncertainty relation
xp p
2
L
E
2
2
mL
2
E
1
.
Classically the particle can have any energy 0.
This phenomenon of energy quantization is to be found in all systems in which a
particle is conned by an attractive potential.
E.g. the Coulomb potential binding an electron to a proton in the hydrogen atom, or the
attractive potential of a simple harmonic oscillator.
In all cases, the boundary condition that the wave function vanish at innity guarantees
that only a discrete set of wave functions are possible.
Each allowed wave function is associated with a certain energy hence the energy
levels of the hydrogen atom, for instance.
Semester 1 2009 PHYS201 Wave Mechanics 57 / 86
Some Properties of Innite Well Wave Functions
The above wave functions can be written in the form
n
(x, t) =
_
2
L
sin(nx/L)e
iE
n
t/
=
n
(x)e
iE
n
t/
the time dependence is a complex exponential of the form exp[iE
n
t/]. The time
dependence of the wave function for any system in a state of given energy is always of this
form.
The factor
n
(x) contains all the spatial dependence of the wave function.
n
(x) =
_
2
L
sin(nx/L)
Note a pairing of the wave function
n
(x) with the allowed energy E
n
. The wave function
n
(x) is known as an energy eigenfunction and the associated energy is known as the
energy eigenvalue.
Semester 1 2009 PHYS201 Wave Mechanics 58 / 86
Probability Distributions I
The probability distributions corresponding to the wave functions obtained above are
P
n
(x) =
n
(x, t)
2
=
_
2
L
sin(nx/L)e
iE
n
t/
2
=
2
L
sin
2
(nx/L) 0 < x < L
= 0 x < 0, x > L
These are all independent of time, i.e. these are analogous to the stationary states of
the hydrogen atom introduced by Bohr states whose properties do not change in
time.
The nomenclature stationary state is retained in modern quantum mechanics for such
states.
Semester 1 2009 PHYS201 Wave Mechanics 59 / 86
Probability Distributions II
1
(x) =
_
2
L
sin(x/L) P
1
(x) =
2
L
sin
2
(x/L)
-
0 L
x
-
0 L
x
Note that the probability is not uniform across the well. In this case there is a maximum
probability of nding the particle in the middle of the well.
Semester 1 2009 PHYS201 Wave Mechanics 60 / 86
Probability Distributions III
2
(x) =
_
2
L
sin(2x/L) P
2
(x) =
2
L
sin
2
(2x/L)
-
0 L
x
-
0 L
x
Once again, the probability distribution is not uniform.
However, here, there is are two maxima, at x = L/3 and x = 2L/3 with a zero in
between.
Seems to suggest that the particle, as it bounces from side to side does not pass
through the middle!
Semester 1 2009 PHYS201 Wave Mechanics 61 / 86
Probability Distributions IV
3
(x) =
_
2
L
sin(3x/L) P
3
(x) =
2
L
sin
2
(3x/L)
-
0 L
x
-
0 L
x
Yet more maxima.
Semester 1 2009 PHYS201 Wave Mechanics 62 / 86
Probability Distributions V
20
(x) =
_
2
L
sin(20x/L) P
20
(x) =
2
L
sin
2
(20x/L)
-
0 L
x
-
0 L
x
If n becomes very large the probability oscillates very rapidly, averaging out to be 1/L,
so that the particle is equally likely to be found anywhere in the well.
This is what would be found classically if the particle were simply bouncing back and
forth between the walls of the well and observations were made at random times, i.e.
the chances of nding the particle in a region of size x will be x/L.
Semester 1 2009 PHYS201 Wave Mechanics 63 / 86
Expectation values and uncertainties I
Recall that if we were to measure the position of the particle in the well many many
times
Making sure before each measurement that the particle has the same wave function (x, t)
each time it is in the same state each time.
We get a random scatter of results whose average value will be
(x) =
_
+
x(x, t)
2
dx
And whose standard deviation x is
(x)
2
= ((x (x))
2
) = (x
2
) (x)
2
For a particle in an innite potential well in the state given by the wave function
n
(x, t),
we have
n
(x, t)
2
=
2
L
sin
2
(nx/L) 0 < x < L
so the expectation value is
(x) =
2
L
_
L
0
x sin
2
(nx/L) dx =
1
2
L
Semester 1 2009 PHYS201 Wave Mechanics 64 / 86
Expectation values and uncertainties II
I.e. the expectation value is in the middle of the well.
Note that this does not necessarily correspond to where the probability is a maximum. In
fact, for, say n = 2, the particle is most likely to be found in the vicinity of x = L/4 and
x = 3L/4.
To calculate the uncertainty in the position, we need
(x
2
) =
2
L
_
L
0
x
2
sin
2
(nx/L) dx = L
2
2n
2
2
3
6n
2
2
.
Consequently, the uncertainty in position is
(x)
2
= (x
2
) (x)
2
= L
2
n
2
2
3
n
2
2
L
2
4
= L
2
n
2
2
6
12n
2
2
.
Semester 1 2009 PHYS201 Wave Mechanics 65 / 86
Further developments
The basic ideas presented till now can be extended in a number of ways:
Look at the properties of combinations of innite square well wave functions, e.g.
(x, t) =
1
2
(
1
(x, t) +
2
(x, t))
This we can do already: nd that this is not a stationary state, i.e. the probability (x, t)
2
oscillates in time.
Work out expectation values of other quantities e.g. energy or momentum.
Work out the wave function for particles in other potentials, or for more complex
particles (e.g. atoms, molecules, solid state, gases, liquids, . . . )
The last set of aims can only be realised once we have the equation that the wave function
satises: the Schr odinger wave equation.
Semester 1 2009 PHYS201 Wave Mechanics 66 / 86
Derivation of the Schr odinger wave equation I
There is no true derivation of this equation.
But its form can be motivated by physical and mathematical arguments at a wide variety of
levels of sophistication.
Here we will derive the Schr odinger equation based on what we have learned so far
about the wave function.
The starting point is to work out, once and for all, what the wave function is for a
particle of energy E and momentum p.
In the absence of any further information, we have used cos(kx t) and other harmonic
functions.
But the correct result we get from the solution for the wave function of a particle in an innite
potential well.
Semester 1 2009 PHYS201 Wave Mechanics 67 / 86
Derivation of the Schr odinger wave equation II
For the particle of energy E = and momentum p = k (we wont worry about the n
subscript here), this wave function is:
(x, t) =
_
2
L
sin kxe
it
= i
_
1
2L
_
e
ikx
e
ikx
_
e
it
= i
_
1
2L
_
e
i(kxt)
e
i(kx+t)
_
Classically, we would expect the particle to bounce back and forth within the well.
Here we have (complex) waves travelling to the left: e
i(kx+t)
, and right: e
i(kxt)
.
This suggests that the correct wave function for a particle of momentum p and energy E
travelling in the positive x direction is
(x, t) = Ae
i(kxt)
.
We want to nd out what equation this particular function is the solution of.
All dynamical laws in nature involve rates of change e.g. velocity and acceleration in
Newtons laws.
The rates of change here will be with respect to both time t and position x, i.e. we will end up
with a partial differential equation.
Semester 1 2009 PHYS201 Wave Mechanics 68 / 86
Derivation of the Schr odinger wave equation III
Taking derivatives with respect to position rst gives:
x
2
= A
2
x
2
e
i(kxt)
= k
2
Ae
i(kxt)
= k
2
2
2m
x
2
=
p
2
2m
.
Similarly, taking the derivatives with respect to time:
t
= i
This can be written, using E = as
i
t
= = E.
Semester 1 2009 PHYS201 Wave Mechanics 69 / 86
Derivation of the Schr odinger wave equation IV
We now generalize this to the situation in which there is both a kinetic energy and a
potential energy present, then E = p
2
/2m + V(x) so that
E =
p
2
2m
+ V(x)
where is now the wave function of a particle in the presence of a potential V(x).
Assuming that
2
2m
x
2
=
p
2
2m
and i
t
= = E still hold true then we get
p
2
2m
+ V(x) = E
2
2m
x
2
+ V(x) = i
t
Thus we arrive at
2
2m
x
2
+ V(x) = i
t
which is the famous time dependent
Schr odinger wave equation.
Semester 1 2009 PHYS201 Wave Mechanics 70 / 86
Some comments on the Schr odinger equation
The solution to the Schr odinger equation describes how the wave function changes as
a function of space and time.
But it does not tell us how a particle moves through space.
Newtons equations tell us the position and velocity of a particle as it moves through space
Schr odingers equation tells us how the information about the particle changes with time.
Though called a wave equation, it does not always have wave-like solutions.
If the potential is attractive, i.e. tends to pull particles together, such as a Coulomb force
pulling an electron towards an atomic nucleus, then the wave function is not a wave at all:
these are called bound states.
In other cases, the particle is free to move anywhere in space: these are called unbound or
scattering states.
There is a whole industry built around solving the Schr odinger equation in principle
all of chemistry ows from solving the Schr odinger equation.
We will concentrate on one part of solving the Schr odinger equation: obtaining the
wave function for a particle of a given energy E.
Semester 1 2009 PHYS201 Wave Mechanics 71 / 86
The Time Independent Schr odinger Equation I
We saw that for the particle in an innitely deep potential well that for a particle of
energy E, the solution looked like
(x, t) =
_
2
L
sin(kx)e
iEt/
= (x)e
iEt/
where the space and time parts of the wave function occur as separate factors.
This suggests trying a solution like this for the Schr odinger wave equation, i.e. putting
(x, t) = (x)e
iEt/
If we substitute this trial solution into the Schr odinger wave equation, and make use of
the meaning of partial derivatives, we get:
2
2m
d
2
(x)
dx
2
e
iEt/
+ V(x)(x)e
iEt/
= i. iE/e
iEt/
(x) = E(x)e
iEt/
.
Semester 1 2009 PHYS201 Wave Mechanics 72 / 86
The Time Independent Schr odinger Equation II
The factor exp[iEt/] cancels from both sides of the equation, giving
2
2m
d
2
(x)
dx
2
+ V(x)(x) = E(x)
After rearranging terms:
2
2m
d
2
(x)
dx
2
+
_
E V(x)
_
(x) = 0
which is the time independent Schr odinger equation.
Note that the quantity E the energy of the particle is a free parameter in this
equation.
I.e. we can freely choose a value for E and solve the equation.
Can emphasize this by writing the solution as
E
(x).
It appears that we would get a solution for any energy we choose, but it is not as
simple as that . . .
Semester 1 2009 PHYS201 Wave Mechanics 73 / 86
The nature of the solutions
The solutions to the Schr odinger equation assume two different forms:
Those for bound states, i.e. where the particle is trapped by an attractive potential, e.g. the
electron in a hydrogen atom is bound to the proton by an attractive Coulomb potential.
The wave function must vanish as x .
This leads to the quantization of energy.
Those for scattering states where the particle is free to move through space, though its
behaviour is inuenced by the presence of some potential, e.g. an alpha particle scattering
off a gold nucleus as in the original experiments by Rutherford.
The wave function does not vanish as x : in fact it looks something like e
i(kxt)
.
Have to be careful (creative?) with the probability interpretation of .
In addition, there is a technical requirement: (x) and
k/m.
To determine the properties of a quantum simple harmonic oscillator, we need the
potential energy of such an oscillator. This is just V(x) =
1
2
kx
2
=
1
2
m
2
x
2
.
The quantum mechanical simple harmonic oscillator of energy E is then described by
the Schr odinger equation
2
2m
d
2
dx
2
+
1
2
m
2
x
2
= E
Semester 1 2009 PHYS201 Wave Mechanics 75 / 86
Allowed energies of the quantum harmonic oscillator
Recall that E is a free parameter that we can choose as we like.
Unfortunately, for almost all choices of E we run into an awkward problem: the solution
looks like
(x) e
1
2
(x/a)
2
for x a =
_
m
,
i.e. it diverges.
We shall see an example shortly.
So, whats wrong with that?
Recall that the solution must be normalized to unity, i.e. we must have
_
+
(x)
2
dx = 1
and that can only happen if (x) 0 as x .
Only if the energy is chosen to be
E
n
= (n +
1
2
) n = 0, 1, 2, . . .
do we get solutions that vanish as x .
Semester 1 2009 PHYS201 Wave Mechanics 76 / 86
Unacceptable solutions of SHO Schr odinger equation
We can illustrate this by plotting the solutions as a function of E, starting at E less than
5
2
and slowly increasing until E reaches
5
2
(i.e. n = 2).
The solutions initially diverge to for x but as E approaches
5
2
, the solution
tends to extend along the x axis, ultimately asymptoting to zero along the x axis.
Semester 1 2009 PHYS201 Wave Mechanics 77 / 86
Comparison of quantum and classical energies.
Once again, we see that the energy is quantized
E
n
= (n +
1
2
),
and the lowest energy, E
0
=
1
2
is > 0.
This is the same situation as found with the particle in the innite potential well.
Classically, the lowest energy would be zero, but once again, because of the uncertainty
relation xp
1
2
, the lowest energy is not zero.
Suppose we had a classical oscillator of total energy
1
2
.
The amplitude a of oscillation of such an oscillator will be given by
1
2
m
2
a
2
=
1
2
n
(x) =
_
1
2
n
n!a
H
n
(x/a)e
1
2
(x/a)
2
where H
n
(z) are known as the Hermite polynomials:
H
0
(z) = 1 H
1
(z) = z H
2
(z) = z
2
1 etc.
For n = 0, the solution is:
E
0
=
1
2
0
(x) =
_
1
a
1
2
(x/a)
2
P
0
(x) =
0
(x)
2
=
1
a
e
(x/a)
2
.
We can use this solution to show non-classical behaviour of a quantum harmonic
oscillator.
Semester 1 2009 PHYS201 Wave Mechanics 79 / 86
Non-classical properties of quantum SHO
The ground state probability distribution P
0
(x) and the potential V(x) are plotted on the
same gure:
P
0
(x) =
1
a
1
2
(x/a)
2
V(x) =
1
2
m
2
x
2
a
a
classically
allowed region
classically
forbidden region
classically
forbidden region
A classical oscillator of energy
1
2
will oscillate between a
and +a.
But the wave function is
non-zero for x > a and x < a.
There is a non-zero probability
(0.1573) of nding the
oscillating mass in a region
outside the allowed range of a
classical oscillator.
Semester 1 2009 PHYS201 Wave Mechanics 80 / 86
Quantum scattering
The other class of problems that can arise are so-called scattering problems
Scattering problems involve particles that are not bound by an attractive potential.
The wave function does not vanish at innity so the usual probability interpretation is no
longer valid, but we can still make sense of the solutions.
Find that quantum mechanics allows particles to access regions in space that are
classically forbidden.
I.e. according to classical physics, the particle does not have enough energy to get to some
regions in space, but quantum mechanically, this has a non-zero probability of occurring.
Have already seen this occurring with the simple harmonic oscillator.
The possibility of a particle being found in forbidden regions in space gives rise to
barrier penetration and quantum tunnelling.
Shall look at the rst in detail, the latter in outline.
Semester 1 2009 PHYS201 Wave Mechanics 81 / 86
Scattering by a potential barrier
The classical version
Shall consider a particle of energy E incident on a potential barrier V
0
with E < V
0
.
Consider the classical version of the problem:
V(x)
x
V = 0
V
0
incoming
particle
particle reected
o barrier
classically
forbidden region
for E < V
0
The diagram shows a barrier dened by
V(x) = 0 x < 0
= V
0
x > 0
For x < 0, as V = 0, the particle will have
entirely kinetic energy E =
p
2
2m
.
For x > 0 its energy will be E =
p
2
2m
+ V
0
.
But E < V
0
p
2
2m
< 0 i.e. it has negative kinetic energy which is impossible.
Thus the region x > 0 is classically forbidden, i.e. a particle cannot reach the region x > 0 as
it does not have enough energy.
The particle must simply bounce off the barrier.
Semester 1 2009 PHYS201 Wave Mechanics 82 / 86
Schr odinger equation for potential barrier
Recall that the time independent Schr odinger equation is given by
2
2m
d
2
(x)
dx
2
+ V(x)(x) = E(x)
for a particle of energy E moving in the presence of a potential V(x).
Here, the potential has two parts: V = 0 for x < 0 and V = V
0
for x > 0.
The Schr odinger equation will also have two parts (with
d
2
dx
2
=
):
2
2m
+ 0. = E x < 0
2
2m
+ V
0
= E x > 0
We have to solve both these equations, and then join the solutions together so that
(x) and
+
2mE
2
= 0 x < 0
2m
2
(V
0
E) = 0 x > 0
For x < 0, the energy is entirely kinetic as V = 0, so
E =
p
2
2m
=
2
k
2
2m
using p = k k
2
=
2mE
2
We will put k = +
2mE
+ k
2
= 0 x < 0
which has the general solution
(x) = Ae
ikx
+ Be
ikx
x < 0 A and B are unknown constants.
This solution can be conrmed by substituting it back into the equation on the left hand
side, and showing that the result equals the right hand side, i.e. = 0.
Semester 1 2009 PHYS201 Wave Mechanics 84 / 86
Solution of Schr odinger equation for potential barrier II
The physical meaning of this solution is best found by putting back the time
dependence.
Recall that the solution to the time dependent wave equation for a particle of energy E is
(x, t) = (x)e
iEt/
= (x)e
it
E =
In this case we have
(x, t) = Ae
i(kxt)
+ Be
i(kx+t)
= wave travelling to the right + reected wave travelling to the left
V(x)
x
V = 0
V
0
classically
forbidden region
for E < V
0
incoming
wave
wave reected
o barrier
Semester 1 2009 PHYS201 Wave Mechanics 85 / 86
Solution of Schr odinger equation for potential barrier III
For the region x > 0, we have to solve
2m
2
(V
0
E) = 0
We are interested in the case in which E < V
0
, i.e. V
0
E > 0
This is the case in which the particle has insufcient energy to cross over into x > 0.
So we put =
_
2m
2
(V
0
E) where > 0
to give
2
= 0
which has the solution
(x) = Ce
x
+ De
x
C and D unknown constants.
We can check that this is indeed a solution by substituting the expression into the
equation for (x).
Semester 1 2009 PHYS201 Wave Mechanics 86 / 86
Solution of Schr odinger equation for potential barrier IV
There is a problem with this solution:
As > 0, the contribution e
x
will grow to innity as x .
This would mean that the probability of nding the particle at x = would be innite, which
is not physically acceptable.
This is a boundary condition once again.
So we must remove the e
x
contribution by putting D = 0 leaving:
(x) = Ce
x
x > 0.
If we put the time dependence back in, we get
(x, t) = Ce
x
e
it
which is NOT a travelling wave.
The particle is not moving in the region x > 0.
Semester 1 2009 PHYS201 Wave Mechanics 87 / 86
Matching the two solutions
Our overall solution so far is
(x) = Ae
ikx
+ Be
ikx
x < 0
(x) = Ce
x
x > 0
These solutions must join together smoothly at x = 0.
This requires (x) and
(x) to be continuous at x = 0.
The rst matching condition gives
_
Ae
ikx
+ Be
ikx
_
x=0
=
_
Ce
x
_
x=0
i.e. A + B = C
The second matching condition gives
_
Aike
ikx
Bike
ikx
_
x=0
=
_
Ce
x
_
x=0
i.e. ik(A B) = C
These two equations can be easily solved for B and C in terms of A:
B =
ik +
ik
A C =
2ik
ik
A
Semester 1 2009 PHYS201 Wave Mechanics 88 / 86
Interpreting the solutions I
We can write out the full solution as
(x) = Ae
ikx
+
ik +
ik
Ae
ikx
x < 0
= A
2ik
ik
e
x
x > 0.
We have a bit of difculty using the usual probability interpretation as we cannot
normalize this solution.
I.e. there is no way for
_
+
(x)
2
dx = 1
as the waves for x < 0 do not go to zero as x .
But lets work out (x)
2
anyway for these waves for x < 0 to see what they could mean:
(x)
2
for incident waves = A
2
(x)
2
for reected waves = A
2
ik +
ik
2
= A
2
Semester 1 2009 PHYS201 Wave Mechanics 89 / 86
Interpreting the solutions II
If these were water waves, or sound waves, or light waves, this would mean that the
waves are perfectly reected at the barrier.
For our probability amplitude waves, this seems to mean that the particle has a 100%
chance of bouncing off the barrier and heading back the other way.
Recall: the particle has an energy which is less than V
0
, so we would expect that the
particle would bounce off the barrier.
So this result is no surprise.
What we are actually doing here is calculating the particle ux associated with the incident
and reected waves.
But the wave function is not zero inside the barrier!
For x > 0 we have
(x)
2
= A
2
2ik
ik
e
x
2
= A
2
4k
2
2
+ k
2
e
2x
.
This means that there is a non-zero probability of the particle being observed inside
the barrier. This is the region which is classically forbidden.
Semester 1 2009 PHYS201 Wave Mechanics 90 / 86
Barrier penetration
V(x)
x
V = 0
V
0
|(x)|
2
=
4k
2
k
2
+
2
e
2x
(2)
1
The particle will penetrate a
distance 1/2 into the
wall.
Suppose we had a particle detector inside a solid wall and we red a particle at the wall
But its energy is so low that it cannot penetrate through the wall.
Nevertheless, in the region x > 0 inside the wall, there is a chance that the particle
detector will register the arrival of the particle!!
If the particle is detected, then, of course, there is no reected wave.
If the detector doesnt go off, then there is a reected wave, i.e. the particle has bounced off
the wall.
Semester 1 2009 PHYS201 Wave Mechanics 91 / 86
An example I
Suppose a particle of energy E approaches a potential barrier of height V
0
= 2E.
The energy of the particle when it approaches the barrier is purely kinetic:
E =
p
2
2m
p =
2mE
Using the de Broglie relation p = k, we get as before k =
2mE
.
For the x > 0 part of the wave function we have (using V
0
= 2E):
=
2m(V
0
E)
2mE
= k
Hence
1
= k
1
=
2
. The penetration depth is therefore
1
2a
=
4
.
Thus, in this case of E =
1
2
V
0
, the penetration depth is roughly the de Broglie
wavelength of the incident waves. This gives a convenient estimate of the penetration
depth in different cases.
Semester 1 2009 PHYS201 Wave Mechanics 92 / 86
An example II
Consider two extreme cases:
The particle is an electron moving at 1 ms
1
, then the penetration depth turns out to be
1
2
= 0.06 mm for an electron
The particle is a 1 kgm mass also moving at 1ms
1
. The penetration depth is then
1
2
= 5.4 10
32
m for a 1 kg mass.
Clearly, the macroscopic object effectively shows no penetration into the barrier
Hence barrier penetration is unlikely to ever be observed in the macroscopic world.
The microscopic particle penetrates a macroscopically measurable distance, and so
there is likely to be observable consequences of this.
Perhaps the most important is quantum tunnelling.
Semester 1 2009 PHYS201 Wave Mechanics 93 / 86
Quantum tunnelling
If the barrier has only a nite width, then it is possible for the particle to emerge from
the other side of the barrier:
V(x)
x
V = 0
V
0
incoming
wave
wave reected
o barrier
transmitted
wave
Semester 1 2009 PHYS201 Wave Mechanics 94 / 86