Nothing Special   »   [go: up one dir, main page]

Cem 419 Exp 1

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Introduction to Ellipsometry

Laboratory 1
Objectives: During this laboratory, you will become familiar with the ellipsometer as
well as with the ellipsometry software. You will also learn how to calibrate the angle of
reflection and make measurements on a model Si/SiO2 system. Using this system, you
will investigate how refractive index varies with wavelength.

Experiments: In this lab, you will begin to make measurements and calculations using
the J. A. Woollam M-44 ellipsometer. The software uses the Fresnel equations along
with the theory of multiple reflections to calculate parameters such as refractive index,
thickness, and absorption coefficient. You can also use the ellipsometry software to
calculate delta (∆) and psi (Ψ) from a given model. The calculation of these quantities
has been covered to a large extent in the short course on ellipsometry.

Here is how to get started.

1. Power up. The computer and ellipsometer electronics should be off during this
procedure. Lighting the Xenon lamp requires a lot of power and can damage
sensitive equipment. Turn on the power supply to the lamp and then ignite the lamp.
Once the lamp is on, turn on the ellipsometer electronics and the computer.

Note that the ellipsometer works as described in the short course. Light from the Xenon
lamp is collimated with the optical fiber and then reflected from the sample at a fixed
angle. The detector contains a diffraction grating and a diode-array detector so that 44
wavelengths of light can be analyzed simultaneously. The ellipsometer also contains a
rotating polarizer. The intensity of the light is recorded as the polarizer rotates and then
the mathematics described in the short course are used to extract ∆ and Ψ.

2. Initialization and Calibration. Carefully place the silicon wafer standard on the
sample stage and start the WVASE program. Go to the hardware pull down menu
and select initialize. During initialization, software looks for the instrument
configuration, the rotating analyzer is brought up to speed and data acquisition is
synchronized with the rotating analyzer. Next, in the hardware menu, select align.
The ellipsometer contains a 4-quadrant detector. To insure that the sample is flat, the
signal to each of the quadrants should be identical. Adjust the tilt of the sample stage
so that the red cross is in the center of the cross-hairs as viewed through the eyepiece.
This alignment step can be completed only under the condition that the sample is flat.
Now go to the hardware menu and select calibrate. This calibration is necessary to
determine the exact positions of the input and analyzer polarizers and the relative
attenuation of the ac signal relative to the DC signal. (See the short course on
ellipsometry for details and an explanation of the specific parameters.)
3. Calibration of the incident angle. Although the ellipsometer is built to precise
standards and the angle of incidence should be 75.0°, to know this angle precisely
requires calibration. We do this with the standard silicon wafer included with the
instrument. The wafer consists of a well defined SiO2 layer on Si. As you might
guess, this is probably the best-characterized ellipsometric system because of its
relevance to electronics. First we need to set up Si/SiO2 as our model system.

a. Go to the model window and select adlayer. Select the file named si-jell.mat. You
just said that the bottom layer of your sample is silicon. Make sure that you are not
fitting any optical parameters in this file and that the thickness is on the mm scale.
We are assuming that the literature values of n and k for silicon are accurate.
b. Select adlayer again and select the file sio2.mat. Now you have told the software that
your system consists of a layer of SiO2 on Si. In the model window, select the SiO2
layer and put in 250 Å for the thickness. Check the fit box by thickness. Do not fit
the optical constants. This indicates that, after we make a measurement, we will fit
the data to this Si/SiO2 model using the thickness of SiO2 as the adjustable parameter.
c. Now go to the hardware/experimental_data/angles and click the 'Fit' box. This says
that along with thickness, you will use the angle of incidence to fit the data.
d. In the hardware window, go to experimental data and click “acquire single scan”.
This gathers your ellipsometric data.
e. In the fit window, select fit/normal fit. You should now see the thickness of the
Si/SiO2 film along with the angle of incidence. The ∆ and Ψ data were fitted by
varying the angle of incidence along with the film thickness. If the film thickness
isn't around 250 Å, something is not right.
f. Repeat the process to see how precisely you can determine the angle of incidence.
Try a few positions on the wafer.
g. Make this the angle of incidence for future experiments.

What you just did was to assume that the literature optical constants for Si and SiO2 are
correct. Then you could fit both the angle of incidence and the thickness of your sample.
You can make this determination at 44 wavelengths so it should be quite precise. Look at
the graph window to see how good the fit was.

Measurement of the Optical Constants of SiO2

1. Now that you know a little bit about the ellipsometer, you can start fitting optical
constants. Go ahead and make some measurements of the thickness and optical
constants of SiO2. Print out the optical constants and return them with your lab
report. You do this by going to the model and deciding to fit n for the SiO2. Just fit
refractive index, n, and thickness. The extinction coefficient, k, should be zero for
this transparent film. Make sure that you are not fitting the angle.

2. Refresh the optical constants of SiO2 by reloading the layer. Go to the model window
and delete the SiO2 layer. Then add it again. Use the thickness that you previously
measured, but don't fit it again. This time, fit the n and k values of the base Si layer.
How close are the values to the literature values in the original file.
Measurements on Gold Slides

A gold film is a simpler system to work with than a monolayer-coated slide. The reason
for this is that there is no surface film (at least in principle).

1. Rinse a gold slide with EtOH and dry it with N2. Measure its optical constants.
Write down the values at 633 nm. Next, clean the slide in piranha solution for 30
seconds. Rinse copiously with millipore water. Prepare 100 mL of the piranha
solution by mixing H2O2 with H2SO4 in a 1:3 ratio. Use extreme care
when working with piranha solution. You must wear a
lab coat or apron, gloves and a face shield. Mount the substrate
in a plastic hemostat being careful to grab only a corner. Next measure the optical
constants. Write down the values at 633 nm. Also note the ∆ and Ψ values.

SAVE THESE SUBSTRATES IN SCINTILLATION VIALS FOR LABORATORY 2

In your report, describe the reasons for the differences between the sample before
and after piranha cleaning.

You might also like