|
Determination of ultra thin film uniformity utilizing x-ray photoelectron spectroscopy
Introduction
Thin films less than 5 nm thick are used in current microchip production and various other nanotechnology applications. Desirable electrical and mechanical properties depend upon thin films having a uniform thickness across a deposition surface. This article outlines a novel, non-destructive method for estimating thin film (less than 5 nm thick) uniformity across a sample surface using X-ray photoelectron spectroscopy (XPS).
Experimental Methods
The deposition surface was a 100 nm thick thermal SiO2 grown on a 200 mm diameter Si(100) substrate. The initial wafer was reduced in size to a 5×5 cm2 portion. After a brief acetone/isopropyl alcohol/deionized water rinse followed with a compressed N2 dry, the sample was loaded into a Baltec MED 020 magnetron sputterer deposition tool. Cr was then deposited at 8×10-3 mBar Ar pressure and 100 mA power over a 10 s time duration. The sample was then quickly loaded into the XPS system, a Physical Electronics Quantum 2000 Scanning ESCA Mircroprobe.Al Ka radiation is used as the x-ray source.
Results and Discussion
The Cr 2p orbital serves as a confirmation of Cr deposition with the Cr 2p orbital doublet having peak centers at 576.7 and 586.4 eV. Metallic Cr0 has doublet peaks located at 574 eV and 583 eV for the 2p3/2 and 2p1/2 orbitals, respectively. The ~3 eV binding energy peak shift away from the metallic Cr0 values represents the bonding between the Cr and O from atmospheric oxidation during transfer between the Cr deposition tool and XPS. The Si 2p peak is centered at 103.8 eV, the expected thermally grown SiO2 value.
Figure 1
Figure 1 represents the Si 2p intensity at several lateral locations moving across the sample using 0.5 cm increments. The Si 2p intensity (I) is related to the film thickness through the Beer-Lambert Law relationship d = -λ LN (I/Io) where d is overlayer film thickness, Io is Si 2p intensity without an overlayer film and λ is the effective electron attenuation length. It requires an in situ experimental setup using XPS analysis before and after the overlayer deposition to determine an accurate quantitative film thickness. Even though we can not get the quantitative thickness, we can estimate relative thickness variation. Estimating λ using the value for elemental Cr (~1.9 nm), Beer-Lambert calculates a variation between the maximum and minimum thickness points in the film as ~2.1 nm. This analysis method allows the simple identification of the film’s physical locations that are either thicker or thinner in comparison to the film average.
|