Scribd
Upload
31 views8 pages

Lecture 5

- The document discusses quantum physics concepts including the Fourier transform relationship between position space wavefunctions ψ(x) and momentum space wavefunctions φ(k), and how this relates to position and momentum distributions and uncertainties. - It defines expectation values, uncertainties, and proves the Heisenberg uncertainty principle that ∆x∆p ≥ h/2. - It discusses Compton scattering experiments where x-rays are scattered by electrons, resulting in a shift in the wavelength of scattered x-rays according to the Compton shift formula. This provided evidence that light has quantized particle properties.

Uploaded by

Oliver58
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
31 views8 pages

Lecture 5

- The document discusses quantum physics concepts including the Fourier transform relationship between position space wavefunctions ψ(x) and momentum space wavefunctions φ(k), and how this relates to position and momentum distributions and uncertainties. - It defines expectation values, uncertainties, and proves the Heisenberg uncertainty principle that ∆x∆p ≥ h/2. - It discusses Compton scattering experiments where x-rays are scattered by electrons, resulting in a shift in the wavelength of scattered x-rays according to the Compton shift formula. This provided evidence that light has quantized particle properties.

Uploaded by

Oliver58
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

8.

04 Quantum Physics Lecture V


Last time
|(x)|
2
and |(k)|
2
as probability densities in position and momentum space,
respectively. . .
Parsevals theorem:

dx|(x)|
2
=

dk|(k)|
2

Normalization: 1=

dx (x)
2
=

dk (k)
2

| |

| |
Measurement:
Measurementofxwithresolutionx
app
<xyieldsrandomresultwithinx
(with probability given by |(x)|
2
x
app
) and changes the momentum distri-
bution: back action of the measurement.
You can know the probability distributions |(x)|
2
and |

(p)|
2
exactly and
simultaneously without violating Heisenberg uncertainty. In apartic-
ular realization of the experiment, you cannot predict the outcomes of
a position measurement followed by a momentum measurement to better than
xp
h
.
2
Since (x) uniquely determines (k), the wavefunction (x) in position space
encodesboth the spatial and the momentum distribution of the parti-
cle. Dierent
1
(x),
2
(x)withthesameprobabilitydensity|
1
(x)|
2
=|
2
(x)|
2
describe particles with the same position distribution but dierent momentum
distributions.
(k)=
1

dx(x)e
ikx
(x)= dk(k)e
ikx
1

= Fourier transform (5-1)


= inverse Fourier transform (5-2)
Dirac delta function:
0
x
Area1
w
Figure I: Dirac delta function.
Massachusetts Institute of Technology V-1

8.04 Quantum Physics Lecture V


x 1
2
(x)= lim e

(5-3)
w 0

2w
2w
2

0 for x=x
0
,
(xx
0
) =

(5-4)

for x=x
0
.

dx(xx
0
)=1 (5-5)

dx(xx
0
)f(x) =f(x
0
) (5-6)

dxe
ikx
= 2(k) (5-7)

dke
ikx
= 2(x) (5-8)

Today
Finish proof: inverse Fourier transform
Exact denition of expectation values, uncertainties
Proof (see notes): xk
1
, xp
h
, t
1
, Et
h
2 2 2 2
Compton scattering
Photoelectric eect
Back to our integral I(y) =

dke
iky
. We have already determined the area under

I(y) as being equal to 2. Consequently, we have another useful identity:

dke
iky
= 2(y) (5-9)

dxe
ikx
= 2(k) (5-10)

We can now nish our proof of the inverse Fourier transformation.


Massachusetts Institute of Technology V-2

8.04 Quantum Physics Lecture V


Inverse Fourier Transform.

dx(x)e
ikx
(k) =
2

2
dk

(k)e
ikx
=
2
dx

(x

) dke
ik(xx

= dx

(x

)2(xx

)
2

=(x)
(k)=
1

dx(x)e
ikx
(x)=
1

dk(k)e
ikx
= Fourier transform (5-11)
inverse Fourier transform
= (5-12)
(Fourier decomposition)
Notethatsince(k),andhencethemomentumdistribution|

(p)|
2
iscompletelyde-
terminedby(x),thewavefunctioncarriesmomentuminformationaswellasspatial
information about the particle.
Expectation values and a precise denition of un-
certainty
Giventhat|(x)|
2
dxistheprobabilitytondtheparticlewithin[x, x+dx],wecan
calculate the expectation value of the particles position as

2
x=

dxx|(x)| = average position. (5-13)


Similarly, we can dene

n 2
x = dxx
n
|(x)| (5-14)

and for any function of positionf(x),

2
f(x)= dxf(x)|(x)| . (5-15)

We dene the uncertainty of the particles position x via the following relation:
(x)
2
=(xx)
2
0 = precise denition ofx (5-16)
Massachusetts Institute of Technology V-3




8.04 Quantum Physics Lecture V
We can expand the RHS:
(xx)
2
=x
2
2xx+x
2

=x
2
2xx+x
2
(x)
2
=x
2
x
2
(5-17)
Similarly, since |

(p)|
2
is the probability density for momentum,

(p)
2 2
p= dpp

dk( hk) (k) (5-18) | | | | =

(p)
2 2
f(p)= dpf(p)

dkf( hk) (k) (5-19) | | | |

(p)
2
=(pp)
2
=p
2
p
2
Using these denitions we can prove the following theorem:
(5-20)
Theorem 5.1. For any function (x), we have the inequality
xk
1
2
. (5-21)
Multiplication by h then yields the Heisenberg uncertainty relation
xp
h
2
. (5-22)
The equality occurs only for Gaussian functions.
Proof. Consider the positive quantity
x(x) +
d
dx
2
0, (5-23)

I() = dx

where is a real number. We can assume without loss of generality that the origin
ofthecoordinatesystemischosenattheexpectationvalueoftheparticlesposition,
such that x=0. When we factor out the integrand in I(), there are three terms.
The rst is

dxx
2
|(x)|
2
=< x
2
>=(x)
2
(5-24)

Massachusetts Institute of Technology V-4



8.04 Quantum Physics Lecture V
since x=0. The second term can be integrated by parts.

d
dxx +

= dxx (

)
dx dx dx

x= 2
= x

(x)(x)|
x=
dx|(x)|


=0 if (x) van- = since (x) is
ishes suciently normalized
fast as x
= (5-25)
The last term can be represented in terms of the Fourier Transform.

dx

2
=
2

dx

d 1

dk

(k)e
ikx

dx

dx


d 1

dx

2
dk

(k

)e
ik

2
dx

dkik

(k)e
ikx

dk

ik

(k

)e
ik

2
dk

dk

kk

(k)

(k

dxe
i(kk

)x
2



2(kk

)
=
2

dk

dk

kk

(k)

(k

)(kk

=
2
dkk
2
|(k)|
2

=
2
k
2

=
2
(k)
2
. (5-26)
Whereinthelaststepwehaveagainassumedwithoutlossofgeneralityacoordinate
system where p=0. Consequently, we have
I()=(x)
2
+
2
(k)
2
, and I()0 for all
1 1
=(k)
2

(k)
2
+
4(k)
2

4(k)
2
+(x)
2

2
1 1
=(k)
2

2(k)
2
+(x)
2

4(k)
2
. (5-27)
I()0 thus requires (x)
2

1
or xk
1
.
4(k)
2
2
Massachusetts Institute of Technology V-5
8.04 Quantum Physics Lecture V
You will show in a homework problem that the equality holds for Gaussian
wavepackets. We note without proof that the equality holds only for Gaussian
wavepackets. The same calculation in the time-frequency domain yields
1
t (5-28)
2
or, multiplying by h, E = h,
h
Et (5-29)
2
energy-timeuncertainty. TheHeisenberguncertaintyarisesbecauseofourinability
to measure a frequency (wavelength) accurately in a nite time interval (position
interval).
Compton Scattering: Electrons scattering x-rays
Whenvisiblelightisscatteredbymatter(electrons),thescatteredlighthasapproxi-
matelythesamefrequencyastheincidentlight. Forx-rays,Compton(1922)observed
A

G
M
M

Figure II: Compton Scattering


thatthewavelengthofthescatteredlightincreases,andthattheincreaseislongerfor
largerscatteringangle. Forbackscattering(=)thewavelengthshift=

isequalto= 4.8510
12
m,independentofthewavelengthofthex-rays. Comp-
ton interpreted the eect as a (relativistic) scattering process between two particles,
anelectronandanx-rayphoton,obeyingenergyandmomentumconservation. Elec-
troninitiallyatrest,treatitasfreeparticle(initialenergyofelectronboundinatom
10 eV, x-ray photon energy 10 keV).
Massachusetts Institute of Technology V-6
8.04 Quantum Physics Lecture V

in
=0

in
=/4
(a) =0 (b) =/4

in
=/2

in
=0.71A
=
0.0485A
(c) =/2 (d) =
Figure III: ...
A

G
M
M
- F
=c|k| =
2c

p = 0

=c|k

| =
2

Figure IV: Compton Scattering


Massachusetts Institute of Technology V-7
8.04 Quantum Physics Lecture V
Electron: E =m
0
c
2
E
2
=p
2
c
2
+m
2
0
c
4
Momentum conservation: hk=p

+ hk

Energy conservation: h h=E

+
Inyourhomework,youwillshowthattheseassumptionsresultinawavelengthshift
=

=
c
(1cos) = Compton shift (5-30)
where
c
=
h
= 0.024

a A is constant called the Compton wavelength of the
mec
electron. According to
dB
=
h
p
, it is the deBroglie wavelength associated with the
(ctional)momentump=m
e
c. Thisformulaexplainstheshiftedinelasticscattering
peak,theunshiftedelasticpeakobservedat

=canbeexplainedbythescattering
ofstronglyboundinner-shellelectrons. Inthiscase,thewholeatom(with10
4
times
larger mass) absorbs the momentum, resulting in a much smaller recoil energy shift
that was unobservable to Compton. The energy loss of the x-ray photon is due to
the KE imparted tothe electron as a consequence of momentum conservation in the
scattering process of an individual photon. If h 0 (i.e., E = h and p = hk
h
associated with individual photons become very small,
c
= 0, and there is
mec

no shift): the Compton shift is a QM eect associated with the grainy quantized
natureoflight. Comptonscatteringshowsthatthereareindividualphotonscarrying

energy E = h and momentum p =


c
and that the interaction between = hk
h
,
electrons and light can be viewed as a scattering process obeying momentum and
energy conservation.
Massachusetts Institute of Technology V-8

You might also like