Tenké Vrstvy

TECHNICAL UNIVERSITY OF LIBEREC
Tenké vrstvy
(v optice)
Jan Václavík
jan.vaclavik@tul.cz
Electromagnetic Waves
• Transverse wave
• Oscillating quantities: electric and
magnetic fields, in phase,
perpendicular to each other and
direction of propagation
• Can travel through vacuum, speed is c = 3.0 x 108 m.s-1

• In transparent medium, speed is 𝒗 = 𝒄Τ𝒏 where n is a property of the
medium called refractive index
• When electromagnetic waves are emitted or absorbed by an atom, done
so in quanta of energy: E = h f
CP 551
Electromagnetic Spectrum
Electromagnetic Spectrum
VISIBLE SPECTRUM
700 nm 400 nm
c=f E=hf
Gamma rays
Increasing frequency
Increasing wavelength
Red orange yellow green blue indigo violet
X-rays
Ultraviolet
Visible
Infrared
Microwaves
Radio
CP 588
Optics
n1 < n2
1
normal
n1
1
2
2
n2
𝑛1 𝑠𝑖𝑛Θ1 = 𝑛2 𝑠𝑖𝑛Θ2
Interference of waves - water waves
Interference of waves - light
fringes
Interference
𝐸1 𝑡 = 𝐸 cos 𝜔𝑡 + 𝜙
𝐸2 𝑡 = 𝐸 cos 𝜔𝑡
𝐸𝑃 cos 𝜔𝑡 = 𝐸 cos 𝜔𝑡 + 𝜙 + 𝐸 cos 𝜔𝑡

𝐸𝑃2 = 𝐸 2 + 𝐸 2 − 2𝐸 2 cos 𝜋 − 𝜙
= 𝐸 2 + 𝐸 2 + 2𝐸 2 cos 𝜙
𝐸𝑃2 = 2𝐸 2 1 + cos 𝜙
𝜙  
𝐸𝑃2 = 4𝐸 2 cos2 1  cos   2 cos 2  
2 2
𝜖0 𝑐𝐸𝑃2 𝜙
𝐼= = 2𝜖0 𝑐𝐸 2 cos 2
2 2
𝜙
𝐼 = 𝐼0 cos2 𝐼0 = 2𝜖0 𝑐𝐸 2
2
Interference examples: Newton Rings

Definitions: TECHNICAL UNIVERSITY OF LIBEREC
Planes of Incidence and the Interface and the polarizations
Perpendicular (“S”) Incident medium

polarization sticks out
of or into the plane of
incidence.
ki kr
Ei Er ni
Plane of incidence
(here the xy plane) is i r
the plane that Interface
Plane of the interface (here the
contains the incident yz plane) (perpendicular to page)
y
and reflected k-
vectors.
t
x nt
z Et
Parallel (“P”) kt
polarization lies
parallel to the plane of
incidence. Transmitting medium
Definitions: TECHNICAL UNIVERSITY OF LIBEREC
Planes of Incidence and the Interface and the polarizations
Perpendicular (S)
polarization sticks up out
of the plane of incidence.
Parallel (P) polarization lies

parallel to the plane of
incidence.
Fresnel equations
r  E0r / E0i for the r  E0r / E0i for the
perpendicular parallel
t   E0 t / E0 i polarization t  E0t / E0i polarization
The Tangential Electric Field is

Continuous ki kr
Ei(x, y = 0, z, t) + Er(x, y = 0, z, t) = Ei Er ni
i
Et(x, y = 0, z, t)
Bi i i r Br
y
Interface
The Tangential Magnetic Field t z x
is Continuous Et nt
–Bi(x, y=0, z, t) cos(qi) + Br(x, y=0, Bt kt
z, t) cos(qr) =
–Bt(x, y=0, z, t) cos(qt)
Reflection & Transmission Coefficients TECHNICAL UNIVERSITY OF LIBEREC
for Perpendicularly and Parallel Polarized Light
r  E0 r / E0i   ni cos( i )  nt cos( t ) /  ni cos( i )  nt cos( t ) 
t  E0t / E0i  2ni cos( i ) /  ni cos( i )  nt cos( t )
r||  E0 r / E0i   ni cos(t )  nt cos(i ) /  ni cos(t )  nt cos(i )
t||  E0t / E0i  2ni cos(i ) /  ni cos(t )  nt cos(i ) 
i 1 m 2
wi
ni r  t 
nt 1 m 1 m
wt
t
m 2
wt cos(t ) 𝑛𝑡 r  t 
m  𝜌= m  m 
wi cos(i ) 𝑛𝑖
Reflection on Air-to-Glass and Glass-to-Air Interfaces

nglass  1.5, nair  1
crit  arcsin(nt /ni)
1.0 1.0
Critical
angle
Reflection coefficient, r
Reflection coefficient, r
Brewster’s angle
.5 .5 r
r||=0! r|| ┴
Total internal
reflection
0 0
Brewster’s
angle
-.5 r -.5
┴
Critical r||
angle
-1.0 -1.0
0° 30° 60° 90° 0° 30° 60° 90°

Incidence angle, i Incidence angle, i
Transmittance (T) and Reflectance (R)

2
E0t
 t2
  0 c0  2 E0i
2
I t At  t 2  0t  wt  nt E0t wt nt 2 cos(t )
n E 2
T      t
I i Ai   0 c0  2
 wi  ni E0i wi ni cos(i )
2
 i 2  0i
n E
 
i
wi
ni   nt cos t    2
T   t   mt
2
  ni cos i   
nt
wt
t
R  r2
wi i r
ni wi
nt
Reflectance of Air-to-Glass and Glass-to-Air Interfaces

TIR Applications
• Q: Why it is used?
Example (Diamond to Air)
• What is the critical angle c for light passing from diamond

(n1 = 2.41) into air (n2 = 1)?
• Solution: At the critical angle, 1  c
and 2  90
n1 sin 2
Rearranging the equation 
 n2 sin 1
 n n2
sin 1  sin c  sin(90) 
2
n1 n1

Substitution gives
1
 sin c  c  24.5
2.41
Reflectance at normal incidence
2
 nt  ni  4 nt ni
R    T 
 nt  ni 
2
 t
n  ni 
Air-glass interface (ni = 1, nt = 1.5)
R = 4 % and T = 96 %
Antireflection Coating
Thin-film Interference
CP 563
180° phase change
 
Ray  undergoes a
phase change on No phase change
reflection. Air
Film
doesanot
Ray  has phase t
undergodue
change a phase
to the path
nAir < nFilm
change on reflection..
difference.
Air
How the reflected rays  and  interfere?

Caution! The wavelength in the film is different than in air.
Dark lines in drawings are there to help you see the boundaries, and are not a separate medium.
180° phase change
The path length
difference is
approximately 2t.  No phase change
Air
There is a 180 phase
difference (½ of a Film
wavelength) due to the  t nAir < nFilm
 film =
first reflection. nfilm
Air
We will get destructive interference when the path difference is an integral

number of wavelengths:
𝜆
2𝑡 = 𝑚𝜆𝑓𝑖𝑙𝑚 = 𝑚 ⇒ 2𝑛𝑓𝑖𝑙𝑚 𝑡 = 𝑚𝜆, 𝑚 = 0,1,2 …
𝑛𝑓𝑖𝑙𝑚
180° phase change
Only for normal

incidence!!  No phase change
Air
Film
 t nAir < nFilm
 film =
nfilm
Air
2nfilmt =m,m=0,1, 2...

 1
2nfilm t =  m+  ,m = 0,1, 2...
 2
Effective layer thickness due to incidence angle

Why does the film look

dark at the top (where it
is thinnest)?
 2d nf 
  2    2  1
 o 
1   2  0
air | water | air

Antireflective Coating  2d nf 
  2    2  1
 o 
2 - 1 = 0
1= 
Destructive interference:
air n1 = 1 2=  - thinnest film
 = 2 (2d nf / o) + 0
nf d
film
d = o / (4 nf) = f / 4
glass n2
n 1 < nf < n2
Transfer Matrix Method
one wave with positive direction(E+)
Superposition of electric field -
one wave with negative direction(E )
Propagating through an interface:
𝐸𝑖 + 1 1 𝑟𝑖,𝑗 𝐸𝑗 +
− = −
𝐸𝑖 𝑡𝑖 𝑗
,
𝑟𝑖,𝑗 1 𝐸𝑗
𝑁 −𝑁 2𝑁 2𝑁𝑗
𝑟𝑖,𝑗 =-𝑟𝑗,𝑖 =𝑁𝑖+𝑁𝑗, 𝑡𝑖,𝑗 = 𝑁 +𝑁𝑖 , 𝑡𝑗,𝑖 =𝑁 +𝑁 ;
𝑖 𝑗 𝑖 𝑗 𝑖 𝑗
𝑟, 𝑡 are, respectively, the complex amplitude reflection and

transmission Fresnel coefficients; 𝑁 is the complex
refractive index of the layer
Propagating within a layer:
𝐸𝑖 + Φ−1 o E,i+ 2𝜋
− = Φ=𝑒 −𝑖 𝜆 𝑁𝑖𝑑
𝐸𝑖 o Φ E,i−
Propagation through
mediums at Where d is the thickness of medium, ω is the frequency of
the propagating light and c is the speed of light
normal incidence
𝞱𝑖 𝞱𝑖 At interface Within the layer

Medium i
P polarization:
Medium j
𝞱𝑗 𝑁𝑖 𝑐𝑜𝑠𝞱𝑗 − 𝑁𝑗 𝑐𝑜𝑠𝞱𝑖
𝑟𝑖,𝑗 =
𝑁𝑗 𝑐𝑜𝑠𝞱𝑖 + 𝑁𝑖 𝑐𝑜𝑠𝞱𝑗
2𝑁𝑖 𝑐𝑜𝑠𝞱𝑖
𝑡𝑖,𝑗 =
𝑁𝑗 𝑐𝑜𝑠𝞱𝑖 + 𝑁𝑖 𝑐𝑜𝑠𝞱𝑗
2𝜋
−𝑖 𝜆 𝑁𝑚 𝑑𝑐𝑜𝑠𝞱𝑗
Φ=𝑒
S polarization:
𝑁𝑖 𝑐𝑜𝑠𝞱𝑖 − 𝑁𝑗 𝑐𝑜𝑠𝞱𝑗
𝑟𝑖,𝑗 =
𝑁𝑖 𝑐𝑜𝑠𝞱𝑖 + 𝑁𝑗 𝑐𝑜𝑠𝞱𝑗
2𝑁𝑖 𝑐𝑜𝑠𝞱𝑖
𝑡𝑖,𝑗 =
𝑁𝑖 𝑐𝑜𝑠𝞱𝑖 + 𝑁𝑗 𝑐𝑜𝑠𝞱𝑗
Fresnel coefficiencts for

oblique incidence 29

𝐸1− 𝑇21
𝑟= +=
𝐸1+ 𝑇11 𝑇12 𝐸𝑁+ 𝐸1 𝑇11
− = 𝑇 𝑇22 0
𝐸1 21
𝐸𝑁+ 1
𝑡= +=
𝐸1 𝑇11
  nt cos t    2
T   t   mt R  r2
2
  ni cos i   
wt cos(t ) 𝑛𝑡
m  𝜌=
wi cos(i ) 𝑛𝑖
Coherence condition
Validity condition:
𝑑
n << 1
∆𝜆𝐶𝑜ℎ𝑒𝑟𝑒𝑛𝑐𝑒
θ
Phase difference between

transmission orders:
2𝜋
𝛿= 2𝑛𝑑 cos 𝜃
𝜆
When δ = 2𝑚𝜋 , 𝑚 ∈ ℕ there will

be constructive interference in T
Incoherent condition
෤ ෤
𝐼 = 𝐸 ∗ 𝐸 = |𝐸0 |2 𝑒 𝑖𝑘𝑛𝑑 𝑒 −𝑖𝑘𝑛𝑑 = |𝐸0 |2
phase information
𝑡
>> 1
∆𝜆𝐶𝑜ℎ𝑒𝑟𝑒𝑛𝑐𝑒
Phase relationships between

interior reflections is destroyed –
therefore there is no interference
Interaction of optical (electromagnetic) wave with TECHNICAL UNIVERSITY OF LIBEREC
material
𝑐
𝑛=
𝑣
• Change of speed
• Refraction - refractive index
• Izotropic/anisotropic medium
• Frequency dependent
• Dielectric materials
• Absorption
• Loss of energy
• Phase effects
• Metals
• Phenomenological model
Material and Electric Field

Electronic polarization
• The index of refraction is a consequence of
electrical polarization, especially electronic
polarization
1 1
v and c 
 0 0 𝑛 = 𝜀𝑟 𝜇𝑟
ℎ
𝐸= 𝑣
𝜆
– +
𝐸>0
Absorption
Absorption
absorption
• Ionic bond - high absorption

• ions of opposite charge move in
opposite directions in the same
electric field f0
f
• hence we get effectively twice the

interaction between the light and
the atoms
• Generally, we would expect
absorption mainly in the infrared
• because these frequencies match
the thermal vibrations of the atoms
• Atom-on-spring model
• Single resonance peak
• Phonons
• Lattice interaction
• Complicated structure
Atomic absorption
• Absorption or emission
due to excitation or
relaxation of the electrons
in the atoms
http://www.nhn.ou.edu/~kieran/reuhome/vizqm/figs/hydrogen.gif
http://www.sbu.ac.uk/water/images/molecul5.jpg
Molecular absorption
http://www.sbu.ac.uk/water/images/watopt.jpg
Band structure
Optical properties of metals
• Almost any frequency of light can be absorbed

• High concentration of electrons → practically all the
light is absorbed within about 0.1µm of the surface
• Metal films thinner than this will transmit light
• gold coatings space suit helmets, displays, heatshields
• Penetration depth
• water: 32 cm
• glass: 29 cm
• graphite: 0.6 µm 𝐼
= 𝑒 −1
• gold: 0.15µm 𝐼0
Optical properties of metals

• Excited atoms in the surface layers?
• Relaxation → photon emission
• The energy lost by the descending
electron is the same as the one
originally incident
• Metals reflects the light very well
• Opaque, reflective
• 90% for most metals
• The remaining energy is usually lost as
heat
• In terms of electrostatics, the field of
the radiation causes the free
electrons to move and a moving
charge emits electromagnetic
radiation
• the wave is re-emitted ≈ reflected
Metal properties – complex band structures
aluminum
spectrum is
relatively flat
gold reflects lots of

red wavelengths
blue red
http://www.thermo.com/eThermo/CMA/Images/Various/109Image_12275.gif
Non metals
• Dielectrics and semiconductors behave essentially

the same way, the only difference being in the size
of the bandgap
• Photons with energies greater than Eg are absorbed
• Production of electron-hole pairs
• May or may not re-emit photon on recombination
EC
EG
EV
hole
Non metals
• Impurity levels divide up the bandgap to allow transitions

with energies less than Eg
• Recombination can be either radiative (photon) or non-
radiative (phonon) depending on the transition probabilities
EC
hf1 phonon
hf2
EV
Absorption coefficients of various semiconductors

Lorentz model
Resonance
Lorentz model – Electrical field

Polarization and Dielectric Constant

Lorentz oscilator
Dielectric constant of water

Refractive index
Absorption coefficient
Optical constants of TiO2
7
n k
4
N, K
0
0.1 1 10 100
Wavelength (um)
Free carriers - Drude model
𝑑2 𝑥 1 𝑑𝑥 𝑑𝒗 1
𝑚0 2 + 𝑚0 = 𝑚0 + 𝑚 𝒗 = −𝑒𝑬 𝑬 𝑡 = 𝑅𝑒{𝑬 𝜔 𝑒 −𝑖𝜔t }
𝑑𝑡 𝜏 𝑑𝑡 𝑑𝑡 𝜏 0
−𝑒𝜏 1
𝒗(𝜔) = 𝑬(𝜔)
𝑚0 1 − 𝑖𝜔𝜏
−𝑒 1
𝑥(𝜔) = 𝑬(𝜔)
𝑚0 −𝜔 2 − 𝑖𝜔/𝜏
𝑛𝑒 2 1
𝑫 𝜔 = 𝜖0 𝑬 𝜔 + 𝑷 𝜔 = 𝜖0 𝑬 𝜔 + 𝜖 0 𝑬 𝜔 = 𝜖0 𝜖𝑟 𝑬 𝜔
𝑚0 𝜖0 −𝜔 2 − 𝑖𝜔/𝜏
𝜔𝑝2 𝑛𝑒 2
𝜖𝑟 𝜔 = 1 − 2 𝜔𝑝2 =
𝜔 + 𝑖𝜔/𝜏 𝑚0 𝜖0
Drude model
𝜔𝑝2 𝑛𝑒 2
𝜖𝑟 𝜔 = 1 − 2 𝜔𝑝2 =
𝜔 + 𝑖𝜔/𝜏 𝑚0 𝜖 0
𝑖𝜎ො
𝜖Ƹ 𝜔 = + 1 = 𝜖𝑟 + 𝑖𝜖𝑖
𝜖0 𝜔
𝜎2
𝜖𝑟 𝜔 = 1 − 𝜎2 = 𝜔𝜖0 (1 − 𝜖1 )
𝜔𝜖0
𝜎1
𝜖𝑖 𝜔 = 𝜎1 = 𝜖2 𝜖0 𝜔
𝜔𝜖0
𝑛ො = 𝑛 + 𝑖𝜅 = 𝜖Ƹ
Drude-Lorentz
Drude-Lorentz
Drude-Lorentz
Empiric models
Lorentz oscillator Drude oscillator
𝜔𝑝2 𝜔𝑝2
𝜖𝑟 𝜔 = 1 − 𝜖𝑟 𝜔 = 1 −
𝑖𝜔 𝑖𝜔
𝜔02 − 𝜔2 − 𝜔2 +
𝜏 𝜏
𝐵 𝐶 𝐸 𝐹
Cauchy formula 𝑛 𝜆 =𝐴+ +
𝜆2 𝜆4
𝑘 𝜆 =𝐷+ +
𝜆2 𝜆4
2
𝐵 𝑗 𝜆
Sellmeier formula 𝑛2 𝜆 = 𝐴 + ෍ 2
𝜆 − 𝜆𝑗2
𝑗
• Sellmeier
• real part of dielectric constant of dielectric oscillator (no damping)
• parameter A should be 1
• singularities at λj
• Cauchy
• Taylor series of Sellmeier formula – useful for regression
• Extinction coefficient arbitrary
Sellmeier formula
Lambert’s (Beer-Lambert’s) Law

• Wave - exponential loss of energy in material
• The fall of intensity
Δ𝐼 = −𝛼 ⋅ Δ𝑥 ⋅ 𝐼
𝛼 … absorption coefficient (m−1)
𝑑𝐼
= −𝛼𝐼
𝑑𝑥
𝐼
• More definitions of α! 𝛼𝑥 = − ln 𝐼
0
Ellipsometry
Phase measurement
2000 64
Ellipsometry
• ELLIPSOMETRY is a method based on measurement of the change of the
polarisation state of light after reflection at non normal incidence on the
surface to study
• The measurement gives two independent angles: Ψ and Δ
• It is an absolute measurement: do not need any reference
• It is a non-direct technique: does not give directly the physical
• Parameters of the sample - thickness and index
• It is necessary to always use a model to describe the sample
• SPECTROSCOPIC ELLIPSOMETRY (SE) gives more comprehensive results
since it studies material on a wide spectral range
Ellipsometry
EP rp
Ei
linear polarisation elliptical polarisation
Er rs
Ep
ES 0 Er Ep.rp
Ei Es Es.rs
Ambient (n0, k0)
Thin Film 1 (n1, k1, T1)

Thin Film 2 (n2, k2, T2)
Thin Film i (ni, ki, Ti)
Substrate (ns, ks)
Ellipsometry measures the complex reflectance ratio :
rp
= = tan Ψ.e j = f( ni, ki, Ti )
rs
Ellipsometry
• Change after reflection on the sample

p

rp 
Y
0 rs s
• ellipticity: tan() - the ratio of the large axis to sin 2𝜖 = sin 2𝜓 sin Δ
the small axis tan 2Θ = 𝑡𝑎𝑛 2𝜓 𝑐𝑜𝑠 Δ
• angle of rotation  - angle between the main axis
and the P axis
Physical meaning of Ψ and Δ
• Those parameters give all relevant information about the polarization

state of the light at a given wavelength.
• tan(Ψ) gives exactly the angle of the first diagonal of the rectangle in
which the ellipse is enclosed
• cos(Δ) gives roughly how fat is going to be the ellipse (shape)
• Its mathematically linked to the ratio of short axis to long axis of the
ellipse in its fundamental frame. 40.
1
tan 2Θ = 𝑡𝑎𝑛 2𝜓 𝑐𝑜𝑠 Δ

0.5
1 0.5 0.5 1
0.5
1
Physical meaning of Ψ and Δ

0 20. 40.
1 1 1
0.5 0.5 0.5
1 0.5 0.5 1 1 0.5 0.5 1 1 0.5 0.5 1
0.5 0.5 0.5
1 1 1
60. 80. 90.

1 1 1
0.5 0.5 0.5
1 0.5 0.5 1 1 0.5 0.5 1 1 0.5 0.5 1
0.5 0.5 0.5
1 1 1
• Example of some different phase shift (Δ) for a given Ψ value

• The long axis is the ellipse is represented and it has to be compared with the diagonal of
the rectangle.
• Because Ψ is fixed, the ellipse is enclosed in the same rectangle for each graphic.
• Modification of Δ change both angle of inclination and ellipticity
Ellipsometry
Goniometer
Analyser Arm
Polariser Arm
Sample
Mapping
rho/theta
Ellipsometry
Determination of ellipsometric parameters
𝐼 = 𝐸𝑑𝑃 𝐸෠𝑑𝑃 + 𝐸𝑑𝑆 𝐸෠𝑑𝑆

𝐼 𝑡 = 𝐼0 (1 + 𝛼cos(2𝜔 𝑡 + 𝛽sin(2𝜔 𝑡 )
A : angle between Analyser and plane of Incidence.

(t) : angle between Polariseur and plane of Incidence.
𝐸𝑑𝑃 1 0 cos 𝐴 sin 𝐴 𝑟𝑃 0 cos 𝑃 − sin 𝑃 1 0 𝐸𝑃

=
𝐸𝑑𝑆 0 0 − sin 𝐴 cos 𝐴 0 𝑟𝑆 sin 𝑃 cos 𝑃 0 0 𝐸𝑑𝑆
Detector Analyser Rotation Sample Rotation Polariser Lamp
𝐼 = 𝐸𝑑𝑃 𝐸෠𝑑𝑃 + 𝐸𝑑𝑆 𝐸෠𝑑𝑆

𝐼 𝑡 = 𝐼0 (1 + 𝛼cos(2𝑃 𝑡 + 𝛽sin(2𝑃 𝑡 )
A : Angle between Analyser axis and Plane of Incidence.

P(t) : Angle between Polarizer axis and Plane of Incidence.
Determination of ellipsometric TECHNICAL UNIVERSITY OF LIBEREC
parameters
Intensity HADAMART TRANSFORM
𝜋
4
𝑆1 = න 𝐼 𝑡 𝑑𝑡
0
𝜋
2
𝜋
4
3
𝜋
4
𝜋
2
𝜋
3
S1 S2 S3 S4 S1 4𝜋
Polarizer angle
𝑆1 − 𝑆2 − 𝑆3 + 𝑆4 𝑆1 + 𝑆2 − 𝑆3 − 𝑆4
𝛼= 𝛽=
2𝐼0 2𝐼0
𝑆1 − 𝑆2 − 𝑆3 + 𝑆4 𝑆1 + 𝑆2 − 𝑆3 − 𝑆4
𝛼= 𝛽=
2𝐼0 2𝐼0
cos 2 𝐴
𝐼0 =
tan2 𝜓 + tan2 𝐴
1+𝛼
tan 𝜓 = tan 𝐴
tan2 (𝜓) − tan2 (𝐴) 1−𝛼
𝛼=
tan2 (𝜓) + tan2 (𝐴)
𝛽
cos Δ =
2 cos Δ tan 𝜓 tan 𝐴 1 − 𝛼2
𝛽=
tan2 𝜓 + tan2 𝐴
Sample analysis
tan(Ψ) Physical Model

Experimental Estimated sample structure
cos(Δ) Measurement - Film Stack and structure

- Material n, k, dispersion
- Composition Fraction of Mixture
No
 Non-direct technique. Measurement = Model?
 Need the use of models to interpret Yes

the measurements and to get
physical parameters of the layers
Ti , ni , ki
REAL SAMPLE STRUCTURE

Spectroscopic ellipsometry sensitivity
Phase variation :  is extremelly

sensitive to ultra thin layers
Angle of incidence: 75°

Spectroscopic ellipsometry sensitivity
Sensitivity could reach 0.01 A !!
0nm
10 A
Angle of incidence: 75°

Thin Film Deposition Methods

Thin Film
Deposition
Physical Chemical
Deposition Deposition
Vacuum
Sputtering Chemical Vapour Atomic Layer
Evaporation
Resistive Diode CVD
E-beam Magnetron PECVD
PLD IBS LPCVD
Cathodic Arc LP-PECVD

Chemical Vapour Deposition

• Materials are deposited from the gaseous state during CVD.
• Precursors for CVD processes must be volatile
• Precursors must be stable enough to be able to be delivered
to the reactor.
• Typical Precursor Materials:

• Halides - TiCl4, TaCl5, WF6, etc.
• Hydrides - SiH4, GeH4, AlH3(NMe3)2, NH3, etc.
• Metal Alkyls - AlMe3, Ti(CH2tBu)4, etc.
• Metal Alkoxides - Ti(OiPr)4, etc.
• Metal Dialylamides - Ti(NMe2)4, etc.
• Metal Diketonates - Cu(acac)2, etc.
• Metal Carbonyls - Ni(CO)4, etc.
• Others – include a range of other metal organic compounds,
complexes and ligands
CVD
Low
atmospheric
pressure
Hot wall Cold wall Hot wall

PECVD Vertical flow
tube Tube tube
Continuous Hot wall Cold wall

Batch
motion batch single wafer
Hot wall
Cold wall
Types of CVD processes

1. APCVD (atmospheric pressure CVD)
2. LPCVD (low Pressure CVD)
3. PECVD (plasma enhanced CVD)
• Hot wall
• Parallel type
• Single wafer
4. MOCVD (metal organic CVD)
5. LCVD (laser CVD)
6. PCVD (photochemical CVD)
7. CVI (chemical vapor infiltration)
8. CBE (chemical beam epitaxy)
CVD – AP CVD
• APCVD reactors operate in mass transport limited region
• Substrates placed horizontaly, moving them under gas
stream
• They are used for depositing low temperature oxide films
• Advantages:
• Simple
• High deposition rate
• Low temp
• Disadvantages:
• Poor step coverage
• Particle contamination
• Require excess substrate handling
• Application:
• Doped & undoped low temp oxides
CVD – LP CVD
• The reactor consists of a quartz tube heated by a three zone furnace
• Substrates stand vertically, perpendicular to the gas flow
• It operates in a surface reaction rate limited mode
• Supply of equal flux of reactants is not required
• Therefore geometry can be such that it can accommodate a large no. of
substrates at a time
• Advantages:
• Excellent purity
• Comfortable step coverage
• Large wafer capacity
• Disadvantage
• High temp
• Low deposition rate
• Application
• Doped & undoped high temp oxides
• Silicon nitride
• polysilicon
CVD – PE CVD
• ‘RF induced’ glow discharge to

transfer energy into reactant gases
• Lower substrate temperature
(compared to APCVD & LPCVD)
• Parallel plate type
• Wafer must be loaded & unloaded
individually
• Chance of contamination
• Silicon dioxide
• Deposition on silicon nitride
• Hot wall type
• Uniformity
• Large no. of wafer deposition
• Contamination while loading and
unloading
CVD – PE CVD
• Single wafer type

• The reactor is load locked
• It offers cassette to
cassette operation
• It offers rapid radiant
heating of each wafer
• Wafer larger than 200mm
can be loaded
CVD – MO CVD
• MOCVD stands for Metal-Organic Chemical Vapour Deposition. This is a technique
for depositing thin layers of atoms onto a semiconductor wafer. Using MOCVD
you can build up many layers, each of a precisely controlled thickness, to create a
material which has specific optical and electrical properties
• Principle: Atoms that you would like to be in your crystal are combined with
complex organic gas molecules and passed over a hot semiconductor wafer.
• The heat breaks up the molecules and deposits the desired atoms on the surface,
layer by layer.
• By varying the composition of the

gas, you can change the properties of
the crystal at an almost atomic scale.
• It can grow high quality
semiconductor layers
Evaporation
• Thermal energy supplied to the

crucible or boat evaporates atoms
• Evaporated atoms travel through
the evacuated space between the
source and the sample and stick to
the sample
• Few, if any, chemical reactions occur
due to low pressure
• Can force a reaction by flowing a gas
near the crucible
• Surface reactions usually occur very
rapidly and there is very little
rearrangement of the surface
atoms after sticking
• Thickness uniformity and shadowing
by surface topography, and step
coverage are issues
Vacuum – Mean Free Path
• ≈63% of molecules undergo a collision

in a distance less than  and ≈0.6% Gas Symbol
ҧ
𝜆𝑝 ҧ
𝜆𝑝
travel more than 5  [m hPa] [m Pa]
Hydrogen H2 11.5·10-5 11.5·10-3
Nitrogen N2 5.9·10-5 5.9·10-3
Oxygen O2 6.5·10-5 6.5·10-3
Helium He 17.5·10-5 17.5·10-3
Neon Ne 12.7·10-5 12.7·10-3
Argon Ar 6.4·10-5 6.4·10-3
Air 6.7·10-5 6.7·10-3
𝑘𝑇 Krypton Kr 4.9·10-5 4.9·10-3

ҧ =
𝜆𝑝 Xenon Xe 3.6·10-5 3.6·10-3
2
2𝜋𝑑𝑚 Mercury
Water vapor
Hg
H2O
3.1·10-5
6.8·10-5
3.1·10-3
6.8·10-3
Carbon
CO 6.0·10-5 6.0·10-3
monoxide
𝜆ҧ Mean free path [m] Carbon
CO2 4.0·10-5 4.0·10-3
𝑑𝑚 Molecular diameter [m] dioxide
Hydrogen
𝑝 Pressure [Pa] HCl 3.3·10-5 3.3·10-3
chloride
Ammonia NH3 3.2·10-5 3.2·10-3
Chlorine Cl2 2.1·10-5 2.1·10-3
Vacuum flow
• Knudsen number
𝜆ҧ
𝐾𝑛 =
𝑑
Viscous flow Knudsen flow Molecular flow

Medium High / Ultra-
Low vacuum
vacuum high vacuum
Pressure < 10-3 bzw. <
103…1 1…10-3
range [hPa] 10-7
Pressure < 10-1 bzw. <
105…10 2 102…10 -1
range [Pa] 10-5
Knudsen
Kn < 0.01 0.01 < Kn < 0.5 Kn > 0.5
number
Re < 2,300:
Reynolds laminar
number Re > 4,000:
turbulent
0.6 > p · d >

p · d [hPa cm] p · d > 0.6 p · d < 0.01
0.01
Evaporation
• The vacuum is usually < 10-5 hPa

• 10-5 hPa = 1 mPa → 𝜆ҧ ≈ 6 m (N2)
• The source heater
• Resistance
• W, Mo, Ta filament or boat
• Contaminants Na, K - they are used in the
production of W
• E-beam (W, Mo, Ta, C, Al2O3 crucible)
• E-beam is often cleaner although S is a common
contaminant in graphite
• Top surface of metal is melted during evaporation
so there is little contamination from the crucible
• More materials can be evaporated (high melting-
point materials)
• A downside of e-beam is that X-rays are produced
when the electron beam hits the melt
• X-rays can create trapped charges, damage to
the substrate
• This damage must be removed by annealing
e-beam Evaporation
• Electron emitted by filament (≈102 mA)
• Acceleration by electric potential (6-12 kV)
• Beam formed by electrostatic optics
• Beam bended by magnetic field
• Hidden filament!
• Low efficiency: ≈10 %
• Multiple crucibles
Sputter deposition - machines

Sputter deposition
• The impact of an atom or ion on a surface

produces sputtering from the.
• No melting of the material
• History of Sputtering
• The verb to SPUTTER originates from Latin
SPUTARE(To emit saliva with noise).
• Phenomenon first described 150 years ago.
Grove (1852) and Plücker (1858) first reported
vaporization and film formation of metal films by
sputtering.
• 1897-- key for understanding discovery of electrons
and positive ions in low pressure gas discharges
and atom structure (J.J. Thomson, Rutherford)
Sputter deposition
• DIODE type
• Conductive plasma
• Pressure: 100-1000 Pa
• Power: 100-1000W
• For DC sputtering, voltage -2 to -5kV
• Incoming ion ≈ 500-1000eV.
• Sputtered atoms ≈ 3-10eV → ineffective!
• 95% of incoming energy goes to target heating & secondary electrons
• Need efficient cooling techniques to avoid target damage
• Sputtering yields and the ion currents limited → low deposition rate
• Energies of the atoms or molecules sputtered are about
10x higher than those thermally evaporated
→ better film quality
Gas Discharge Modes

Striking Voltage – Paschen’s law

• Dependent on distance and pressure!
• Minimal at specific p∙d
• Atmosphere (100 kPa): 327V @ 7.5µm
𝐵𝑝𝑑
𝑉𝐵 =
1
ln(𝐴𝑝𝑑) − ln ln 1 + 𝛾
𝑆𝐸
𝑝𝑑
𝑉𝐵 = 𝑎
ln 𝑝𝑑 + 𝑏
Sputter yield
• Threshold energy
• The yield increases with the energy
• Yield approaches saturation
• Saturation at higher energies for heavier
bombarding particles (Xe+ 100keV and Ar+
20KeV)
Sputtering yield vs. chamber TECHNICAL UNIVERSITY OF LIBEREC
pressure
Too few collisions limit yield
Too many collisions

prevent ionization
Plasma not Influence of working pressure and

Arcing in current on deposition rate for non-
sustained at
plasma (?) magnetron sputtering.
low pressure
For same power P=IV=constant, high current (ion number) comes with low voltage (ion energy)
100
Reactive sputtering
• Sputtering a compound target may not give what one wants.
• This doesn’t mean reactive sputtering will give what one wants – it is
just one more thing to try with.
• Certainly reactive sputtering can be done using DC sputtering, whereas
compound target (insulating) can only be used for RF sputtering.
• Chemical reaction takes place on substrate and target.
• Can “poison” target if chemical reactions are faster than sputter rate.
• Need to adjust reactive gas flow to get good composition (e.g. SiO2
rather than SiO2-X) without incorporating excess gas into film.
• A mixture of inert + reactive gases used for sputtering:
• Oxides – Al2O3, SiO2, Ta2O5 (O2 mixed with Ar)
• Nitrides – TaN, TiN, Si3 N4 (N2, NH3, mixed with Ar)
• Carbides – TiC, WC, SiC (CH4, C2H4, C3H8, mixed with Ar)
Magnetron sputtering
• In DC & RF sputtering, the efficiency of ionization from energetic collisions between the
electrons and gas atoms is low.
• Most electrons lose energy in non-ionizing collisions or are collected by the electrodes.
• Oscillating RF fields increasing ionization efficiency marginally.
• Deposition rates are low.
• To increase deposition rates, magnets are used to increase the percentage of electrons that take
part in ionization events, increasing the ionization efficiency.
• A magnetic field is applied at right angle to the electric field by placing large magnets behind the
target.
• This traps electrons near the target surface and causes them to move in a spiral motion until
they collide with an Ar atom.
• The ionization and sputtering efficiencies are increased significantly - deposition rates increase
by 10-100x, to 1 µm per minute.
• Unintentional substrate heating is reduced since the dense plasma is confined near the target
and ion loss to the substrates is less.
• A lower Ar pressure (0.1 Pa can still sustain plasma) can be utilized since ionization efficiency is
larger → improvement of film quality as less argon will be incorporated into the film.
• Magnetron sputtering can be done in DC or RF, but more often it is done in DC as cooling of
insulating targets is difficult in RF systems.
As a result:
Magnetron Sputtering
• Current density (proportional to
ionization rate) increases by 100 times.
• Required discharge pressure drops 100
times.
• Deposition rate increases 100 times.
Hoping radius r:
Orbital motion of electrons increases probability that For electron: r0.3cm
they will collide with neutral species and create ions. For Ar+ ion: r81cm
Magnetron Sputtering
Magnetron Sputtering– target TECHNICAL UNIVERSITY OF LIBEREC
erosion
Magnetron Sputtering Chamber Configuration

Ion beam sputter deposition (IBS)
• High-end thin film deposition process.
• Lower pressure sputter deposition (0.01 Pa), sputtered atoms retain kinetic energy (minimal collision in
low pressure environments).
• High quality, smooth, pin hole free films.
• Enhanced adhesion and micro-structure control.
• Yields excellent coverage at small thicknesses and on high aspect ratio features.
• Independent control of ion beam

parameters.
• Low energy ion source (left) assists

with end-hall ion source (right).
• Film properties exceed those

deposited by evaporation or
magnetron sputtering.
Další témata
• Glancing angle deposition GLAD
• Deposition time – pump down, substrate heating,
material drying/melting, deposition, layer
switching, cool down
• DIY resistive evaporator – easy
• Deposition rate
• Optical quality
• Process time
• Aluminium - fast
Comparison of typical thin film deposition technology
Comparison of evaporation and sputtering
Evaporation Sputtering
Vacuum Pumps
Pumping Methods
VACUUM PUMPS
(METHODS)
Gas Transfer Entrapment

Vacuum Pump Vacuum Pump
Positive Displacement Kinetic Adsorption

Vacuum Pump Vacuum Pump Pump
Reciprocating Rotary Drag Fluid Entrainment Ion Transfer Cold Trap

Displacement Pump Pump Pump Pump Pump
Diaphragm Liquid Ring Gaseous Ejector Bulk Getter Getter

Pump Pump Ring Pump Pump Pump Pump
Piston Rotary Turbine Liquid Jet Diffusion Getter Ion Sublimation

Pump Piston Pump Pump Pump Pump Pump Pump
Multiple Vane Sliding Vane Axial Flow Gas Jet Diffusion Self Purifying Evaporation
Rotary Pump Rotary Pump Pump Pump Ejector Pump Diffusion Pump Ion Pump
Rotary Radial Flow Vapor Jet Fractionating Sputter Ion

Plunger Pump Pump Pump Diffusion Pump Pump
Dry Roots Molecular

Cryopump
Pump Pump Drag Pump
Turbomolecular
Condenser
Pump
Pressure Ranges
RANGE PRESSURE
ROUGH (LOW) VACUUM 759 TO 1 x 10 -3 (mbar)
HIGH VACUUM 1 x 10 -3 TO 1 x 10 -8 (mbar)
ULTRA HIGH VACUUM LESS THAN 1 x 10 -8 (mbar)

Gas Load
Permeation
Outgassing
Real
Leaks
Diffusion Virtual
Backstreaming
GAS LOAD (Q) IS EXPRESSED IN:

mbar liters per second
Pump-down Curve
10+1
10-1
Volume
10-3
Pressure (mbar)
10-5 Surface Desorption
10-7
Diffusion
10-9
Permeation
10-11
10 1 10 3 10 5 10 7 10 9 10 11 10 13 10 15 10 17
Time (sec)
Pump Operating Ranges
Ultra High High Vacuum Rough Vacuum

Vacuum
Rotary Vane Mechanical Pump

Rotary Piston Mechanical Pump
Dry Mechanical Pump
Sorption Pump
Blower/Booster Pump
Venturi Pump
High Vac. Pumps
Ultra-High Vac. Pumps
10-12 10-10 10-8 10-6 10-4 10-2 1 10+2

P (mbar)
Typical Vacuum System
9 1 Chamber
8
2 High Vac. Pump
3 Roughing Pump
1
7 3a Foreline Pump
8 4 Hi-Vac. Valve
5 Roughing Valve
5 6 Foreline Valve
4
7 Vent Valve
7 8 Roughing Gauge
2 9 High Vac. Gauge
3a 3
Mechanical Vacuum Pump (gas transfer pump)

• Rough pumps system from 1 atmosphere
• Q = 1-1000 m3/h
• Backs high-vacuum pump

• can use one pump for both purposes, or use two
pumps
• Three general types
• Indirect drive (Welch)
• Direct drive (Alcatel, Edwards, etc.)
• Dry pumps (Edwards, Varian, Pfeiffer, Leybold,
Anest Iwata) Indirect drive (belt drive)
Dry (scroll pump) Direct drive (rotary-vane)

Rotary vane pumps
Oil Sealed Rotary Pumps Understanding Gas Ballast www2.avs.org/chapters/nccavs/pdf/Gas_Ballast_OilSealedPumps.pdf

Rotary vane pumps

• One stage vs. two stage – ultimate pressure
• One stage 103 − 102 Pa
• Two stage 101 − 100 Pa
• Belt Drive
• Slower 400-600 RPM
• Bigger, Cheaper
• Direct drive
• Faster 1500 to 1725 RPM
• Smaller, lighter
Dry scroll pump

• Use one fixed and one
orbiting scroll to create
crescent-shaped gas pockets
• Gas pockets are compressed
and air is forced through
central exit port
• No oil, completely dry and
contamination-free
High vacuum pumps
• Oil diffusion pumps (gas transfer)

• Operation range: 10−3 − 10−10 mbar
• Turbomolecular pumps (gas transfer)
• Operation range: 10−2 − 10−10 mbar
• Gas capture pumps
• Ion pumps, getter, titanium sublimation pump
• Must operate in conjunction with other high vacuum pumps
• Operation range: to 10−11 mbar
High Vacuum Pumps
Ultra High
High Vacuum Rough Vacuum
Vacuum
Roughing Pumps
Liquid Nitrogen Trap

Diffusion Pump
Turbo Pump
Cryo Pump
Ion Pump
Tit. Subl. Pump
10-12 10-10 10-8 10-6 10-4 10-2 1 10+2

Oil diffusion pump

• Oil heated – boils Water cooling coils
• Vapor streams up and is deflected out

and down through baffles
• The large oil vapor molecules transfer
momentum to air molecules that
randomly enter pump (3-stage stack at Foreline
right)
• Oil re-condenses on side of pump that
is actively cooled by circulating chilled
water Pump oil
• Air molecules build up at base of

pump
• Mechanical backing pump removes air
from base (4th stage)
Kurt J. Lesker Co. Heater
• 100 - 10000 l/s pump rate

• Can’t operate above ~ 10-2 mbar
(can “crack” the oil)
Turbo (molecular) pumps

• Gas molecules interact with
spinning blades and are
preferentially forced downward
• High vacuum (10-6 Pa) requires
rotation of 20,000 to 90,000 RPM
• Generally work between 10-3 and
10-7 mbar
• Ineffective before gas is in
“molecular flow”
• Purely mechanical
• Very clean, fast
• Back with mechanical pump
• Options
• Bearings: Ceramic (oil lubricated)
Magnetic (supported w/out physical
contact), also hybrid
• Rotor options (Blade configuration)
• Cooling (air or water)
Ion pump (gas capture)

• Principal: gases are taken up by
reaction with fine particles of metal,
or by ion implantation
• Use parallel array of short stainless
steel tubes (anode)
• Plates of Ti (or Ta) near ends of
tubes (cathode)
• Generate strong magnetic field

parallel to tubes
• Gas is ionized in tubes by electrons
released from cathode
• Ions strike cathode and sputter Ti
• Results in chemical reactions and
ion burial
• Generally used around electron gun Kurt J. Lesker Co.

Pump speed curves
Rotary Vane:
Pump speed lower at low vacuum
Pump Speed
Turbo pumps: opposite
Pressure
Vacuum Gauge
Ultra High
High Vacuum Rough Vacuum
Vacuum
Bourdon Gauge
Capacitance Manometer
Thermocouple Gauge
Pirani Gauge
Hot Fil. Ion Gauge
Cold Cathode Gauge
Residual Gas Analyzer
McLeod Gauge
Spinning Rotor Gauge
10-12 10-10 10-8 10-6 10-4 10-2 1 10+2

P (mbar)
Transport Phenomena Gauges

• Pirani: Filament = Sensor + Bridge
• Constant-voltage Pirani: Voltage exerted to the filament is constant, temperature is variable
Range: 10 to 10-3 mbar
• Constant-temperature Pirani: Range: 1000 to 10-3 mbar
Problem: Contamination ( by oil, etc. )
• Thermocouple: Change of temperature of the filament is monitored by a thermocouple
Range: 5 to 10-3 mbar ( Above 5, the filament temperature changes very little )
Advantage: Using very low power
Ionization Gauges
Penning
• Quick, free of X-ray problems, no filament
• Problems
• The ion-induced current is not linearly related to the pressure, rather, the
relationship is exponential
• Less accurate than HCGs
• Cleaning is necessary ( Oil vapors )
• Starting of the CCGs can be delayed ( LP ), can be turned on at higher pressure during
a pumpdown
Deposition Rate
Mass Evaporation Rate

Knudsen Cell
Cosine Law
Uniformity – flat surface

Equilibrium Vapor Pressure

Tenké Vrstvy

Uploaded by

Copyright:

Available Formats

Tenké Vrstvy

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tenké Vrstvy

Uploaded by

Copyright:

Available Formats

TECHNICAL UNIVERSITY OF LIBEREC

• Can travel through vacuum, speed is c = 3.0 x 108 m.s-1

Interference of waves - water waves

Interference of waves - light

𝐸𝑃 cos 𝜔𝑡 = 𝐸 cos 𝜔𝑡 + 𝜙 + 𝐸 cos 𝜔𝑡

Interference examples: Newton Rings

Planes of Incidence and the Interface and the polarizations

Perpendicular (“S”) Incident medium

Planes of Incidence and the Interface and the polarizations

Parallel (P) polarization lies

The Tangential Electric Field is

for Perpendicularly and Parallel Polarized Light

r  E0 r / E0i   ni cos( i )  nt cos( t ) /  ni cos( i )  nt cos( t ) 

t  E0t / E0i  2ni cos( i ) /  ni cos( i )  nt cos( t )

r||  E0 r / E0i   ni cos(t )  nt cos(i ) /  ni cos(t )  nt cos(i )

t||  E0t / E0i  2ni cos(i ) /  ni cos(t )  nt cos(i ) 

Reflection on Air-to-Glass and Glass-to-Air Interfaces

0° 30° 60° 90° 0° 30° 60° 90°

Transmittance (T) and Reflectance (R)

Reflectance of Air-to-Glass and Glass-to-Air Interfaces

Example (Diamond to Air)

• What is the critical angle c for light passing from diamond

Reflectance at normal incidence

Air-glass interface (ni = 1, nt = 1.5)

How the reflected rays  and  interfere?

We will get destructive interference when the path difference is an integral

Only for normal

2nfilmt =m,m=0,1, 2...

Effective layer thickness due to incidence angle

Why does the film look

air | water | air

Propagating through an interface:

𝑟, 𝑡 are, respectively, the complex amplitude reflection and

Propagating within a layer:

Transfer Matrix Method

𝞱𝑖 𝞱𝑖 At interface Within the layer

Fresnel coefficiencts for

Transfer Matrix Method

Phase difference between

When δ = 2𝑚𝜋 , 𝑚 ∈ ℕ there will

Phase relationships between

Material and Electric Field

• Ionic bond - high absorption

• hence we get effectively twice the

Optical properties of metals

• Almost any frequency of light can be absorbed

Optical properties of metals

Metal properties – complex band structures

gold reflects lots of

• Dielectrics and semiconductors behave essentially

• Impurity levels divide up the bandgap to allow transitions

Absorption coefficients of various semiconductors

Lorentz model – Electrical field

Polarization and Dielectric Constant

Dielectric constant of water

Optical constants of TiO2

Free carriers - Drude model

Lambert’s (Beer-Lambert’s) Law

Thin Film 1 (n1, k1, T1)