As-Received Analysis

SAMPLE: Silicon <111> Wafer in “As-Received” Condition from Supplier

SAMPLE STORAGE, HANDLING & PREPARATION
Sample was received inside an air-tight plastic bag that was cut open.
The cassette case was opened and one of the 50 wafers was removed using wafer tweezers.
Sample wafer was turned upside-down on UHV clean aluminum foil.
Scribed at very edge, and cleaved to form a ~20 x 20 mm square from the very center.
Note: All samples to be analyzed must be small enough to fit on and be trapped on the 80 mm diameter sample mount.
The upside of the sample was immediately clipped onto the sample mount and loaded into the load-lock.
This sample piece, is in its’ “as-received” condition, will then be analyzed by XPS.
Sample was not cleaned or treated by any method by scientists at XPS Metrology.
The pressure in the load-lock was reduced to 10^-5 torr within 5 minutes and sample transferred into the analysis chamber.
By analyzing any sample in its’ “As-Received” condition you might find un-expected contaminants that might cause your problem.
To analyze the bulk of a wafer, the freshly exposed edge can be held so the X-ray beam strikes the exposed bulk edge.

Therefore: Analyzed As-Received = “As-Received Analysis” is recommended.

Silicon wafers in Cassette in Clean Room	Silicon wafer with naturally formed Native Oxides	Silicon wafer with 50 nm of SiO2

p-Si <111> Sample As-Received
Raw Survey Spectrum with no labels

p-Si <111> Sample As-Received
Survey Spectrum with peak labels, peak integration, peak assignment, and atom% composition

p-Si <111> Sample – After Ar+ Ion Etching for 20 sec using 3 kV acceleration
Raw Survey Spectrum with no labels

p-Si <111> Sample – After Ar+ Ion Etching for 20 sec using 3 kV acceleration
Survey Spectrum with peak labels, peak integration, peak assignment, and atom% composition

Summary Atom % Table for Elements
Shows Change in Atom %s after “As-Received” Sample was “Ar+ Ion Etched”

Information about the sample received from the customer.

The sample is a polished Silicon wafer having a <111> crystalline structure that has been in a multi-wafer sample rack inside a sealed plastic bag for 1 year after being produced.
The final treatment of the polished wafer was a rinse with clean DI water, followed by rapid drying with warm Nitrogen gas, and then placed in the rack.
After the rack was filled with 50 identical wafers, the rack was placed inside a plastic bag, that was shrink-wrapped to exclude air while pulling a slight vacuum.
The surface is expected to be pure Silicon and has a trace level dopant to make it behave as a p-type semi-conductor.

Based on the atom%s of the elements detected by XPS, we make the following observations and interpretations.

By analyzing in the “As-Received” condition before any ion etching, we know that there was roughly 16 angstroms of Carbon on the as-received surface.
We also know that the amount of Oxygen decreased to almost zero (0) atom% because Ar+ ions removed roughly 31 angstroms of a layer that has Oxygen.
The amount of Silicon greatly increased after the ion etching by roughly 2X.
The ratio of Si : O in the as-received condition is roughly 1.7 : 1.0.
The amount of Nitrogen is very low, but it does increase by ~3X after the ion etching.
The Argon signal appears because the sample was ion etched with Ar+ ions which become implanted and trapped within the pure silicon lattice.
If the top 150 angstroms (15 nm) of this surface was pure SiO₂, then the ratio of Si : O would be roughly 1 : 2. Because the atomic ratio is far from 1 : 2, we know from experience, that top surface (~10 nm) the Silicon wafer, that should be pure Silicon, is actually a natice oxide of Silicon on top of the pure silicon wafer.
From experience we also know that pure silicon wafers often easily form a 20-25 angtrom thick layer of Silicon native oxide.
Because XPS measures the top 10 nm we can know that the atom% results are due to the presence of pure silicon and silicon native oxides.
To learn more, we run high energy resolution spectra that show multiple chemical state peaks from all of the main elements.
The Silicon native oxides usually include 4 different silicon oxides: Si₂O, SiO, and Si₂O₃, and SiO₂ which are due to Si ¹⁺, Si ²⁺, Si ³⁺ and Si ⁴⁺.

To track which XPS signal of each element is used to generate atom%s, this table includes the Spin-Orbit value of the XPS peak used to generate atom% values.

“As-Received Naturally Formed Native Oxides on Metal Surfaces”

Effects of Argon Ion Etching

Effects of Argon Ion Etching are readily visible from SEM images (below) which are images of the top 10-50 angstrons. The Ar+ ion gun at 2 kV was turned on for 0 seconds, 10 seconds, 60 seconds, and 120 seconds.

When EDX (EDS) is used, then you do not measure the element information from the top 10-50 angstroms. EDX data is from the bulk depths which ranges from 0.5 to 3.0 microns (5,000 angstroms – 30,000 angstroms). Chemistry in the bulk is always different from the top surface. EDX never gives element information that matches XPS.