1. Axiomatic Design of CMP Machine [top]
Axiomatic design is
investigated as a design methodology for large or complex system design.
Particular considerations of system design are described and the
suitability of axiomatic design for such considerations is discussed.
Then, tools to enable successful application of axiomatic design to
systems are developed. A design of a machine tool system for polishing
silicon wafers using chemical mechanical polishing (CMP) is presented. The
CMP system architecture is decomposed from top level requirements using
the principles of axiomatic design, and the theorems developed. The CMP
system was designed and fabricated by this research group, and has
demonstrated excellent capability to remove material from the surface of a
wafer while maintained an increased control of the removal.
|
2. Process Analysis for Cu CMP [top]
This research establishes a
theoretical framework to relate the process parameters to the different
wafer/pad contact modes to study the behavior of wafer-scale
polishing. Several models of polishing - microcutting, brittle
fracture, surface melting and burnishing - are reviewed. Blanket
wafers coated with a wide range of materials are polished to verify the
models. Plastic deformation is identified as the dominant mechanism
of material removal in fine abrasive polishing. Additionally,
contact mechanics models, which relate the pressure distribution to the
pattern geometry and pad elastic properties, explain the die-scale
variation of material removal rate (MRR) on pattern geometry. Experiments
study the effects of pattern geometry on the rates of pattern
planarization, oxide overpolishing and Cu dishing.
|
3. Modeling Dielectric Erosion and Cu Dishing in Single- and
Multi-Step Cu CMP [top]
A formidable
challenge in the present multi-step Cu CMP process, employed in the
ultra-large-scale integration (ULSI) technology, is the control of wafer
surface non-uniformity, which primarily is due to dielectric erosion and
Cu dishing. In contrast with the earlier experimental and
semi-theoretical investigations, a systematic way of characterizing and
modeling dielectric erosion in both single- and multi-step Cu CMP
processes is presented. Wafer- and die-level erosion are defined, and the
plausible causes of erosion at each level are identified in terms of
several geometric and physical parameters. The local pressure distribution
is estimated at each polishing stage based on the evolving pattern
geometry and pad deformation. The single-step model is adapted for the
multi-step polishing process, with multiple sets of slurry selectivities,
applied pressure, and relative velocity in each step. Based on the
developed multi-step erosion model, the physical significance of each
model parameter on dielectric erosion is determined, and the optimal
polishing practices for minimizing erosion in both multi-step and
single-step polishing are suggested.
|
4. Slurry-Switching Point and
End-Point Detection in Cu CMP [top] A novel
in situ detection technique, based on the change of the reflectance of the
patterned surface at different polishing stages, is developed to detect
the process end-point and slurry-switching point in multi-step Cu
CMP. Models that employ light scattering theory and statistical
treatment correlate the sampled reflectance with the surface topography
and Cu area fraction for detecting the process regime and end-point.
Developed multi-step dielectric erosion and Cu dishing models are applied
to minimize resulting Cu thinning.
|
|