- Wafers with front side particles < 10’s of 0.25 um particles make satisfactory
low pressure nano-imprint substrates.
- The backside of wafers does not need to be as clean as the front, however
visible contamination must be avoided.
- Suitable process cleanliness can be achieved using a Class 100 environment
or even a Class 1000 environment with mini-environments around all process
equipment, and good process discipline.
- Substrates can be qualified by “eyeball” inspection under a off axis intense
light such as a projector or fiber light, that will probably detect sub 10 um
particles.
SPECIFICATION
In a high pressure imprint any particles will be crushed, allowing imprinters to be
used in office environments.
In low pressure imprint, we must assume that the particle is not crushed, but that it
will be pushed into the organic planarization layer. Given a planarization layer of 100
nm, and a residual layer of 50 nm, a 200-250 nm particle would be required to cause
a detectable thickness perturbation. A 500 – 600 nm particle would produce an easily
detected defect with a full thin film interference color cycle.
Based on air borne distributions shown below, it should be expected that a wafer
with < 10 particles > 200 nm, would have a 1 or 2 500 nm particles and no 1 um
particles. This is probably an acceptable level of wafer cleanliness for visually
acceptable imprints with thin residual layers.
CLASS Number of Particles per Cubic Meter by Micrometer Size
0.1 um 0.2 um 0.3 um 0.5 um 1 um 5 um
ISO 2 100 24 10 4
Single digit 0.25 um particles per wafer would have been required for a 0.5 um half
pitch process, a early 1990’s fab quality. Not state of the art, but does require care.
The industry standard tools are supplied by KLA-Tencor for both unpatterned and
patterned wafers.
Suitable process cleanliness can be achieved using a Class 100 environment or
even a Class 1000 environment with mini-environments around all process
equipment, and good process discipline.
INSPECTION PROCESS
Substrates can be qualified by “eyeball” inspection under a off axis intense light
suchas a projector or fiber light, that will probably detect sub 10 um particles.
A simple practical approach is to use a very bright off axis light and look at the wafer
against a black background. Particles will be seen as bright spots. Obviously this is a
qualitative technique, although we can estimate detection sensitivity.
The relative light scattered by a single particle has been estimated as 3*e-5 from a 2
um particle, and linear above 2 um. The eye’s relative detection threshold is 2 *e-4,
under dark adaptation of 1 minute from bright light ambient.
Comparing the eye’s dynamic range to the scattering from a 2 um particle, suggests
a detection limit for visual defects of 20 um. With a very intense light source, the
reflected intensity can be extended by up to 10 x, so that particles down to 2 um are
probably detectable.
Based on the size distributions shown above, if there are no 2 um particles, the 0.5
um particle count will be in the single digits and the wafer would be acceptable for
imprint.
Sources
http://www.particle.com/whitepapers_met/cleanroom%20standards.
htm#Class_Comparison
D Semleit, et al „Development of a particle deposition meter“ in Particles on surfaces
7 Detection, Adhesion and Removal , Ed K Mittal (2002) p27-40.
“Color Science Concepts and Methods, Quantitative Data and Formulae“. G.
Wyszecki, W.S. Stiles, p 519.
--------------------------------------------------------------------------------
Copyright © 2005 Impattern Solutions. All rights reserved.
