Gunnar Ingelman

Nicusor Timneanu

Quantum mechanics, advanced course

Scattering theory
Scattering processes :
• main tool for investigating structure of matter
→ understanding of matter,
incl. bio-molecules (life !)
• major part of experimental work
• mustt have
h proper theory
th framework
f k
energy is most important variable
Comment to lecture
on Nobel prize:
2 ferromagnetic
half-metallic CrO2 layers,
layers
N(E) N(E)
separated by
non-magnetic
g layer
y

”… depends on density
of available states …”

2π
wi →[ n ] = Vni ρ (En )En ≈ Ei
2
Fermi’s golden rule
h
2π
or wi →n = Vni δ (En − Ei )
2
• exp. resolution ΔE
h • fundamental
to be integrated with ∫ dEn ρ (En ) ΔE~h/Δt
 Wh t iis scattering?
What tt i ?

• Français: dispersion
• Auf Deutsch: Streuen
• In Italiano: dispersione
• På svenska: spridning
• E português:
Em t ê dispersar
di
• Românã: împrãştiere
• 简体中文版: 驱散 散射
• English: cause to separate and go in different directions,
deflection, collision, dispersion,
p diffusion, sprinkling
p g
 Constituents & binding force

ncertaint relation ΔxΔp ~ h → Δp ~ h/Δx

Heisenberg uncertainty

system size Δx fluctuation binding binding

[m] Δp [eV/c] energy force
atom 10-10
10 103 1 kkeV
V el-magn.
l
nucleus 10-14 107 10 MeV strong (~ van der Waals)
proton 10-15 108 100 MeV strong (f
(fundamental
ndamental QCD)
quark ? < 10-18 > 1011 100 GeV super-strong ?
 To see = scattering experiment
pole

wave
length

Analysis of scattering pattern gives “picture” of object

Smaller objects than wave length of probe cannot be observed
→ looks point-like

High resolution ⇔ short wave length ⇔ high energy

λ = hc / E
Quantum mechanical duality: particle ↔ wave
High energy particles (photon, electron, proton…) → high resolution
Scattering of On Energy Result Field
(projectile) (target) Ti
Time

Electrons Atoms eV, KeV Electron Atomic

1920
structure physics

Protons Nuclei 20 MeV Mass distrib., Nuclear

nuclear struct. physics 1940
Electrons Nuclei 200 MeV Charge Nuclear
distribution physics 1950

Electrons Protons, 200 MeV Charge Nuclear 1960

neutrons distribution physics

Electrons Protons, 10 GeV Quarks, Particle

neutrons inner structure physics 1970
 Quarks observed 1968
Unexpectedly many
electrons scattered
to large angles θ
⇒ point charges
in proton !!!
Nobelpris 1990: Friedman, Kendall, Taylor

Cf. Rutherford → atomic nucleus

Cf
Surprisingly many α-particles Q2 ~ θ2 ~ 1/λ2
scattered byy g
gold foil to
large angles
⇒ atomen’s massa
i smallll nucleus
in l
 Energy distribution of quarks in proton
Known el.-magn. interaction via photon γ e- E,θ
→ number of scattered e as function of E,θ
,
e-
γ x p
dσ 2πα
[ ] (
2

dx dQ 2
=
x Q2
1 + (1 − y )2
F2 x , Q 2
) q

x,y,Q2 depends on E,θ

andd proton ““structure ffunction
i ” ( ) [ ( ) (
F2 x, Q 2 = ∑ e 2f xq f x, Q 2 + xq f x, Q 2 )]
f
quark charge
with probability qf (x,Q2) to find (with resolution Q~θ ~1/λ)
quark of “flavour” f =u,d,s… and fraction x of proton energy
→ proton = uud
{+ q{
q Quark-antiquark pairs created/annihilated
valens quantum due to Heisenberg uncertainty relation
quarks fl
fluctuatio
i ns
proton quantum numbers ΔEΔt ~ h/4π
 The Scattering Cross Section

Cross section = effective area for collision

Beam off particles

B ti l shot
h t towards
t d a target
t t ⇒ Statistical
St ti ti l nature
t
Size of beam >> target size (probability of interaction)

The cross section is a measure of interaction between particles

Units = units of area = m2

Standard unit is barn=10-24 cm2 = 10-28 m2

 Classical scattering (Newton)

Particles
P ti l h have wellll d
defined
fi d ttrajectories
j t i (orbits)
( bit )
Trajectories are completely and uniqely determined by
Newton’s
Newton s equations of motion
 Cross sections

Elementary cross section

d b*db*d
dσ=b*db*dϕ
Total cross section
b max 2π
σ= ∫ b ⋅ db ∫ dϕ = π b 2
max
0 0

Correlation between b ↔ θ
(definite trajectory, Newton’s law)

db db(θ ) dΩ
dσ = b ⋅ ⋅ dθ ⋅ dϕ = b ⋅ ⋅
dθ dθ ssin θ
b = impact parameter
dΩ = sinθ dθ dϕ = solid angle Differential cross section
dσ db((θ ) 1
= b⋅ ⋅
dΩ dθ sin θ
 Famous example: Rutherford scattering

1911 – discovery of the nucleus

Experiment:
Au foil bombarded with α particles

Observed angles of 140o deflection

Inconsistent with Thompson model

Ernest Rutherford (1871-1937)

father of nuclear physics
 Rutherford scattering: assumptions

• Atom contains a nucleus with positive charge and almost

the entire mass of the atom
• Atom
At contains
t i Z electrons
l t moving
i around d the
th nucleus
l
• Target nucleus is much more heavy than incident particles
(no recoil for nucleus) ⇒ force is central
• Classical mechanics can be used to describe collision
(conservation of energy, momentum and angular momentum)
• Interaction is pure Coulomb interaction V( r)~1/r (both target
and projectile are point-like charges)
• There
Th is
i no other
th force
f acting,
ti no interaction
i t ti with
ith electrons
l t
• Scattering is elastic
Consequence: central force + conservation of angular momentum
⇒ the trajectory is planar
 Rutherford scattering: derivation
Z1e ⋅ Z 2e 1
Interaction is pure Coulomb V (r) = ⋅
4 πε0 r
dφ
Angular momentum L = m ⋅ r2 ⋅ = const.
dt
1 r r
Energy of particle E = m ⋅ v 2 + V (r ) = const.
2
⎛
m dr ⎞ 2
mr dφ
2
⎛ ⎞ 2
r
Use polar coordinates: rr, φ E= ⎜ ⎟ + ⎜ ⎟ + V (r )
2 ⎝ dt ⎠ 2 ⎝ dt ⎠
Equation
q of motion => dr 2 r L2
= [E − V (r
( r))] − 2 2
dt m m r
If E is kinetic only, then … L = m ⋅ v ⋅ b = 2mE ⋅ b

dr 2 r 2Eb 2
= [E − V (r
( r))] −
dt m mr 2
 Rutherford scattering: derivation
Using angular momentum dφ mr 2
dt = m ⋅ r ⋅ 2
= dφ
L b 2mE
Replace dt and obtain … b ⋅ dr
dφ = r
( r)) b 2
V (r
… connection between r and φ r2 1− − 2
E r
Integrating over the half trajectory π −θ ∞
b ⋅ dr
∫
d
⎛ π −θ⎞ =
r ∈ (rmin ,∞) φ ∈ ⎜0, ⎟ 2 r
V (r ) b2
⎝ 2 ⎠ rmin
r2 1− − 2
E r
⇒relation between b and θ
⎛ Z1e ⋅ Z 2e ⎞ 2
1 2θ
P f
Performing
i theth integral
i t l… b 2
= ⎜ ⎟ cot
⎝ 4 πε 0 ⎠ 16E 2
2
2
dσ db(θ ) 1 ⎛ Z e ⋅ Z 2e ⎞ 1 1
Differential cross section = b⋅ ⋅ = ⎜⎜ 1 ⎟⎟
dΩ dθ sin θ ⎝ 4πε 0 ⎠ 16 E 2 sin 4 θ
2
 Rutherford cross section
Differential cross section
2
dσ ⎛ Z1e ⋅ Z 2 e ⎞ 1 1
⎜
=⎜ ⎟
dΩ ⎝ 4πε 0 ⎟⎠ 16 E 2 sin 4 θ
2

Probability for scattering in a certain direction:

-does not depend on angle ϕ, only on θ (trajectory planar)

-proportional to the strength of the interaction squared
-inverse proportional to energy squared
decreases drastically for large scattering angles θ
-decreases
Scattering as
a tool
t l for
f
exploring the
structure
t t off
matter

The search for

f d
fundamental
t l
particles
 Quantum Scattering
There are no localized (defined) trajectories

Incoming particles – described rigorously by wavepackets

(superposition of plane waves)

r r rr i
− Ek t r
φ ( x , t ) = ∫ d k ⋅ e ⋅ e h ⋅ c(k )
3 ik ⋅ x

Wave Wave Plane Stationary state Coefficient

function Vector Wave (time dependent) (amplitude)
r r
ik ⋅ x
Simplify – use single momentum plane waves
r er
OK approx.
pp particle of well defined momentum p = hk
for p
since then Δxi ~ h / Δpi is large, i.e. extended wave packet
 Incident wave Transmitted wave Scattered wave
Plane wave Plane wave Spherical wave
Elastic scattering r
r r̂ˆ k′
k′ = k ⋅ r

r r
k r
target

r r rr ik⋅r
ik ⋅ x e
φ =e ik ⋅ x
φ +ψ scatt = e + f k (θ , ϕ )
r
 Scattering cross section
Defined using flux (current density)
j = #particles / unit area / unit time dA
r r
jin r jscatt
dσ r
t
target
t
Unit area
r r
Conservation of #particles: jini ⋅ dσ = jscatttt ⋅ dA dA=r2*dΩ

− ih *
( )
r
Flux
u in qua
quantum
tu mechanics:
ec a cs jin = φ ⋅ ∇φ − φ ⋅ ∇φ *
Probability distribution
2m
− ih *
( )
(density of states) r
jscatt = ψ scatt ⋅ ∇ψ scatt −ψ scatt ⋅ ∇ψ scatt
*

ρ = φ =φ φ
2 * 2m
 Scattering cross section
r
dσ jscatt 2
= r r = f k (θ , ϕ )
2
Differential cross section:
Ω
ddΩ jin
fk((θ,ϕ)=scattering
,ϕ) g amplitude
p
2π 1
σ = ∫ dϕ ∫ d (cos θ ) f k (θ , ϕ )
2
Total cross section:
0 −1

transition probability Φ → Ψscatt

wave :ψ scatt
prob. amplitude
dA
r f k (θ , ϕ ) r
jin r jscatt
wave : φ dσ r
target
 Schrödinger equation
∂Ψ (r , t )
Time dependent phenomena ih = H Ψ (r , t )
∂t 2
h r
With hamiltonian H =− ∇ + V (r )
2

2m
Since V(r) is independent of time, separate the time dependence
r r
Time independent S
S.E.
E H Ψ( x) = E Ψ( x)
h 2k 2
Scattering happens for positive energies notation: E = >0
2m
(bound states - negative energy levels)
r 2m r
U (x) = 2 V (x)
⎡ h2 2 h2 r ⎤ r h 2k 2 r h
⎢− ∇ + U ( x )⎥ Ψ ( x ) = Ψ( x)
⎣ 2m 2m ⎦ 2m

Inhomogeneous
Helmholtz equation
( r
) r r
∇ + k Ψ( x) = U ( x) Ψ( x)
2 2
 Solution?

Homogeneous equation ( r
∇ + k Ψ( x) = 0
2 2
)
rr
ik ⋅ x
r r r r
general solution is a plane wave e = cos(k ⋅ x ) + i sin( k ⋅ x )
For inhomogeneous equation find a particular solution:
Introduce the Green function ( r r r r
)
∇ + k G( x, x ' ) = δ ( x, x ' )
2 2

r r
ik x − x '
r r 1 e
...with
with solution G( x, x ' ) = − ⋅ r r
(wave propagation x→x’) 4π x − x '
Final solution is sum of general solution+particular solution
r rr
3r r r r r
Integral equation:Ψ ( x ) = e
ik ⋅ x
+ ∫ d x ' G( x, x ' )U ( x ' ) Ψ( x ' )
 Solution?
r r
ik x − x '
r rr
1 3r e r r
∫
ik ⋅ x
Ψ( x) = e − d x ' r r U ( x ' )Ψ ( x ' )
4π x − x'
Identify the scattering amplitude far from origin:

1 r r r
r r
f k (θ , ϕ ) = − ∫
− i k '⋅ x '
3
d x' e U ( x ' )Ψ ( x ' )
4π
Symbolically Ψ = Ψ + GUΨ0

Solve integral equation by iteration

Ψ = Ψ 0 + GUΨ = Ψ 0 + GU (Ψ 0 + GUΨ ) = ...

= Ψ + GUΨ + GUGUΨ + ...
0 0 0
 Born approximation

If the potential is weak (compare to E) – smaller terms

⇒ expressioni converges

Born approximation: Ψ = Ψ 0 + GUΨ 0

Onlyy keep
p terms of order one in p
potential U
Scattering amplitude becomes:
1 r r r
r rr
f k (θ , ϕ ) = − ∫
3 −ik '⋅ x ' ik ⋅ x '
d x' e U ( x ' )e
r 4π
k' r r r
q = k −k' r r i (k − k ' )⋅ x '
r r r
1
=− ∫
3
θ d x ' U ( x ' ) e
r 4π
k r r r θ
Momentum transfer (scattering vector) q = k − k ' = 2k sin
2
 Born approximation

2m 3r r iqr⋅ xr ' r
2 ∫
f k (θ , ϕ ) = − d x 'V ( x ' )e = f ( q )
4πh
Scattering amplitude = Fourier transform of the potential
2π 1
σ = ∫ dϕ ∫ d (cos θ ) ⋅ f k (θ , ϕ )
2
Scattering cross section
0 −1

Interpretation of Born approx. Higher terms

– one step scattering – two step scattering
r
r k' r r r
G( x, x ' ) k'
k r
x' r
k
 Born approximation in X-ray scattering
r r
Wave function Ψ (x ) Electric field u (x )
r r
Potential U (x ) Electron density ρ (x )
Schrödinger equation Wave equation
( r
) r r
∇ + k Ψ ( x ) = U ( x )Ψ ( x )
2 2
( 2 r2
) r r
∇ + k u ( x ) = ρ ( x )u ( x )
Scattering amplitude = Scattered electric field =
Fourier transform of potential Fourier transform of density
r r r 2πixr⋅( krs − krin ) r
f (q ) = F (U ( x )) E (k s , kin ) = ∫ ρ ( x )e dx
Cross section in direction θ,ϕ Intensity of scattered field

dσ I scatt
= F (h, k , l )
2
r 2
= f (q ) IT
dΩ
 Born approximation: Coulomb potential
Screened Coulomb potential Pure Coulomb in limit
e − μr V0 Z1e ⋅ Z 2 e
r
V (r ) = −V0 μ → 0 and →
μr μ 4πε 0
F i transform
Fourier t f can be
b solved
l d analytically
l ti ll
r 2m 3r r iqr⋅ xr ' 2m 2 sin (qr ')
2 ∫
f (q ) = − d x 'V ( x ' )e = − 2 ∫ dr 'V (r ' )r '
4πh h qr '
r 2mV0 1
f (q ) =
μh 2 μ 2 + 4k 2 sin
i 2θ /2

For pure Coulomb, the cross section becomes classic Rutherford

2 2
dσ r 2 ⎛ 2m Z1e ⋅ Z 2 e 1 ⎞ ⎛ Z1e ⋅ Z 2 e ⎞ 1
= f (q ) → ⎜⎜ 2 ⎟⎟ = ⎜⎜ ⎟⎟
dΩ ⎝ h 4πε 0 4 k 2
sin 2
θ / 2 ⎠ ⎝ 4πε 0 ⎠ 16 E 2
sin 4
θ /2
 Born approximation: Bragg condition
r r r
If potential has translational symmetry (crystal) V (r ) = V (r + R)

Scattering amplitude r 2m 3r r iqr⋅ xr '

2 ∫
f (q ) = − d x 'V ( x ' )e
4πh
r r r
Change of variables x ' = x + R
r 2m r r r rr r r r r r

4πh 2 ∫
iq ⋅ x iq ⋅ R iq ⋅ R
⇒ f (q ) = − d 3
x V ( x + R ) e e = f ( q ) e

Implies f (
r
q ) 1 −( e
r r
iq ⋅R
=) 0 Condition for scattering: f (q ) ≠ 0
r
r r
Final condition is: q ⋅ R = 2πn Bragg law:
( )
r r r
k − k ' ⋅ R = 2πn
⎛ krˆ − krˆ ' ⎞ ⋅ Rr = 2π n = λn θ
⎜ ⎟ 2 R sin = λn
⎝ ⎠ k 2
 Scattering angle vs. Bragg angle

r
k'
r θB
k θS

θ S = 2θ B
 Resolution
Momentum transfer ⇔ scattering angle ⇔ resolution
r r r θ
Momentum transfer (scattering vector) q = k − k ' = 2k sin
2
S ll angle
Small r l scattering
tt i ⇒ low
l resolution
l ti Bragg law:
k' r
θ →0 q→0 1 2 sin θ / 2
q = →0
r d λ
k
Small momentum transfers
r 2m 3r r iqr⋅ xr ' 2m 3r r
2 ∫ 2 ∫
f (q ) = − d x 'V ( x ' ) e → − d x 'V ( x ' )
4πh 4πh
Scattering only probes the whole potential (electron density)

Low reslution gives the overall features of the target (molecule)

 Resolution
Momentum transfer ⇔ scattering angle ⇔ resolution

Large angle scattering ⇒ high resolution Bragg law:

θ → θ max q → qmax 1 2 sin θ max / 2

r →
r q d λ
k' High reslution gives finer details of the target (molecule)
θ
r
k Bragg law:
Extremum
E t rcase: backwards
b k d scattering
tt i 1
=
2
qmax λ
θ =π q = 2k d min
r r smallest distance
k' k between resolvable
Higest reslution is limited by ½ wavelength lattice planes

r 2m 3r r i 2 kr⋅ xr ' Scattering probes (samples)

2 ∫
f (q ) = − d x 'V ( x ' ) e
4πh the potential (density) at λ/2
 Method of Partial Waves

Method for solving the Schrödinger equation for scattering

(also used in wave physics, optics, nuclear physics)
Used for spherically symmetric potentials V(r)
Can be easily
y used for inelastic scattering
g as well,,
for any energies and potentials strengths

Very powerful computational method

method, both analytic and numeric

Uses: calculations of atomic form factors (X-ray crystallography)

calculations
l l ti off electrons
l t scattering
tt i iin crystals,
t l used d ffor
determination of radiation damage (project in Uppsala)
...
 Spherical harmonics Yl ,m (θ , ϕ )
l=0, S-wave
l=1, P-wave
l=2, D-wave
...
 Plane wave expansion
rr ∞ r
ˆ rˆ
e ik ⋅ r
= ∑ (2l + 1) ⋅ i ⋅ jl (kr
l
k ) ⋅ Yl (k ⋅ r )
l =0

∞
=∑
l =0
×
 Partial waves ψ E ,l ,m (r ,θ , ϕ ) = RE ,l (r ) ⋅ Yl ,m (θ , ϕ )

The target induces a phase shift in the incoming wave

δl

r
V (r )

1 lπ e iδ l lπ
sin( kr − ) sin(kr − + δ l )
kr 2 k
kr 2

Each partial wave (l=0

(l 0,1,2,...)
1 2 ) scatters independently
If δl=0, then there is no scattering for l wave
 Scattering amplitude & cross section in partial waves:
∞
sum of partial
f k (θ ) = ∑ (2l + 1) f l (k ) Pl (cos θ ) • amplitudes
l =0
• cross sections
dσ ∞
σ tot = ∫ dΩ = ∫ dΩ f l (k ) = 4π ∑ (2l + 1) f l (k )
2 2

dΩ l =0

1 iδ l
For elastic scattering: f l (k ) = e sin δ l
k
4π ∞
⇒ σ tot = 2
k
∑ (
l =0
2l + 1) sin 2
δl in terms of phase shifts

Exact results!
Approximation when only some partial waves (l -values) included

Low energy scattering → l = 0 dominates (only l = 0 for k→0 )

Optics in a water droplet

The distribution of light intensity

inside the water droplet. The
spherical shape of the droplet
makes it act as a microlens and
focuses the laser light into a ‘hot
spot’. The droplet is 22.6 µm
across The color scale shows
across.
the increase in light intensity in
the droplet, with the red areas
b
boosted db
by a ffactor off 100
100.
 Colliders – advanced machines for
investigating the structure of matter

Electrons Electrons 200 GeV New particles: W & Z

Electrons Protons 300 GeV Proton inner structure,

quarks and g
q gluons
Protons Antiprotons 2 TeV Bottom & top quarks
Higgs?
Protons Protons 14 TeV Higgs? Supersymmetry?
Extra dimensions?

1 GeV = 109 eV, 1 TeV = 1012 eV

CERN Conseil Europeen pour la Recherche Nucleaire Geneve
LEP, Large Electron Positron collider

27 km circumference

Di
Discovery W
W,Z
Z iin 1983
 Dagens största “elektronmikroskop”:
HERA (Hadron-Elektron-Ring-Anlage)
(Hadron Elektron Ring Anlage) @ DESY i Hamburg

900 GeV
p

30 GeV
e-

Teori → datormodell Observerat

 Test av QCD-teorin
Accelererad laddning emitterar kvanta,
t.ex. oscillerande elektrisk laddning → antenn −
α=e2/4π
QED’s kopplings-styrka −

p
αs =g2/4π QCD
QCD’ss kopplings-styrka
kopplings styrka
beror av energin

Observerade 2-jet-händelser:
• påvisar gluonen
• mäter αs
Emission av flera gluoner
i teori och observationer
QCD-kraftens styrka
12π
α s (Q 2 ) =
⎛ Q2 ⎞
(33 − 2n f )log⎜⎜ Λ2 ⎟⎟
⎝ ⎠
Q ≈ energi E
αs(E)
• stor energi → liten koppling αs
Q2 → ∞ : αs → 0
asymptotic freedom
kvarkarna “fria”
⇒ teorin OK, beräkningsbar

• liten energi → stor koppling αs

Λ 0 2 GeV: αs(Q2) → ∞
Q2 → Λ≈0.2 E
confinement Gross, Wilczek, Politzer
kvarkarna fängslade
g pp
“för upptäckten av asymptotisk
y p frihet
⇒ teorin ej löst, problem i teorin för den starka kraften”
 Forskargruppen i teoretisk högenergifysik
sommaren 2004
Johan Rathsman Johan Alwall Nic Timneanu Leszek Motyka
forskarassistent doktorand doktorand postdoc
→ lektor → postdoc SLAC → forskare biofysik → Hamburg

Rikardd
Rik
Enberg
doktorand
→ postdoc
Berkeley

2005

Magdalena Larfors GI Korinna Zapp David Eriksson

X-jobb-student diplom-student doktorand
→ doktorand teor. fysik → doktorand Heidelberg
 Fermilab Chicago
Discover bottom quark in 1977
and top quark in 1995

6 km circumference

TEVATRON
ATLAS at Large Hadron Collider at CERN

Tord Ekelöf et al.

proton-proton collisions at 14 000 GeV energy

• Higher energi → new,
new massive particles can be created !?
• Higher resolution → new, smaller structures can be observed !?
 SLAC Stanford Linear Acceleration Center
Discover quarks
Discover 1969,1969,
quarks and charm 19741974
and charm
3 km long

To become photonics center, home of

LCLS Linac Coherent Light SourceUses 3 km from the
(2009)
linear accelerator
Cost ~ 300 million $
to determine
Photons 1.5 – 15 Å
structure of Pulse 10
10-250
250 fs
proteins etc
FLASH (2004) & European X-Ray Laser Project (2012)

FLASH (known as VUV-FEL) @Hamburg

- 300 m long EXFEL facts
- pulses 25 fs short - 3.4
34k km llong
- intensity 1012 photons/pulse - Cost ~ 1 billion Euro
- wavelength: 30nm-6nm (Sweden is part of it)
- underground 6-38 m
- pulse 100 fs
- wavelength
g 0.85 – 60 Å