0% found this document useful (0 votes)
34 views9 pages

Material Model

Uploaded by

78ryng6gwy
Copyright
© © All Rights Reserved
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
0% found this document useful (0 votes)
34 views9 pages

Material Model

Uploaded by

78ryng6gwy
Copyright
© © All Rights Reserved
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

FITTING THE DISPERSIONS

z The Reflectivity and Ellipsometry equations are used in the


analysis of the DBS and SE measurements.

z Among the parameters that need to be varied in the equations


are the n’s and k’s – which actually are n(λ)’s and k(λ)’s – for
for all the materials composing the film (layers and substrate).
substrate).

z To solve the equations, one should know or calculate the


values of n, k for each material and each wavelength, which is
is impractical (too many variables).
DISPERSION MODELS

z Instead, we use simplified mathematical models to represent the n


n and k dispersions, each model using a small number of
parameters.
z Four Dispersion Models are available on our systems.
y Cauchy Polynomial

y Cubic Polynomial

y Harmonic Oscillator (HO)

y Bruggeman Effective Medium Approximation (BEMA)

And, for a few well known and stable materials, the dispersion is given
by:
Tabular model: fixed table of n and k values for the whole set of
wavelengths.
CAUCHY POLYNOMIAL

The Cauchy dispersions for n and k are of the form:

N1 N2 N3 K K2 K3
n(λ )= N0 + + + + ... k(λ )= K0 + 1
+ 4+ + ...
λ
2
λ
4
λ
6
λ
2
λ λ
6

z The Ni and Ki are constants. They can be treated as unknowns and


varied during the calculations to match the spectra.

z The Cauchy model has no physical constraints. This means that the
solutions may not be physically possible.

z The Cauchy model works well for the visible wavelength range, and
and for materials with simple monotonic dispersions.
CUBIC POLYNOMIAL

The Cubic Polynomial dispersions for n and k are of the form:

n(λ)=N0+N1λ +N2λ 2+N3λ 3 k(λ)=K0+K1λ +K2λ 2 +K3λ 3


z The Ni and Ki are constants. They can be treated as unknowns and
varied during the calculations to match the spectra.
z This model has no physical constraints. This means that the solutions
solutions may not be physically possible.
z This model is good for metals and silicides, in the visible
wavelength range. However, it is not often used, and the tabular
model is usually preferred, if available.
HARMONIC OSCILLATOR

z The Harmonic Oscillator (HO) model is the most commonly


used one. It describes the movement of electrons in a damping
medium when they are subjected to an electromagnetic (EM)
field.
No EM field EM field (light)

nucleus oscillation

e- cloud
e- cloud
HO MODEL PARAMETERS

A material HO model is usually composed of 1 to 5 oscillators.


Many parameters enter in the complicated HO equation which
represents a material dielectric constant. Here are some of them:
ε : dielectric constant = (n - i k)2
NM : number density of oscillators for oscillator M (in nm-3)
EnM : resonant energy of oscillator M (in eV)
EgM : damping energy of oscillator M (in eV)
φM : phase shift for oscillator M (in radians)
The Harmonic Oscillator model respects the physical constraints.
Peaks in the dispersion curve correspond to oscillator resonance energies.
energies. Often, oscillators have peaks outside the wavelength range, but
but dispersion tails in the spectrum range.
BRUGGEMAN EFFECTIVE MEDIUM
APPROXIMATION (BEMA)

z This model describes a material as a composite mixture of


materials.

z The n and k dispersion of each component in the mixture is


assumed to be known (or it has been calculated separately).

z The volume fractions of the different components are treated as


as a variables in the BEMA dispersion model.

z As an example, polysilicon can be treated as a mixture of


crystalline silicon, amorphous silicon, and voids.
Model Summary

Table 1: Dispersion models, icon representation, and applications.

Icon Model Description Typical


applications

Tabular Table of n, k vs λ values All films

3 Cauchy n = N0 + N1/λ2 + N2/λ4 Dielectrics


k = K0 + K1/λ2 + K2/λ4 Photoresist
Polyimide
(visible range)
4

Bruggeman Combination of up to 8 All films


different material models Also use for
modeling roughness
or scattering
Harmonic Host material containing All films
Oscillator damped oscillators especially,
Polysilicon, TiN
OLSA

An off-line spectral analysis software (OLSA) is available as an


option for the calculation and optimization of dispersion models
using spectral measurement data downloaded from the system to a
a floppy.

This program also allows the simulation of films and dispersions.


dispersions.

The use an off-line analysis program frees the tool.

Optimized recipes can be copied back to the tool for routine


production monitoring applications.

You might also like