|
|
Indentations
in Microelectronic Bond Pads
|
|
Introduction
From the viewpoint of cost and
reliability, the bonding of leads from the contact pads on a microelectronic
device to the terminals in its package is one of the most critical steps. A
common method of attachment utilises an ultrasonic transducer. In this
method, the wire is brought into contact with the bond pad and then vibrated
under pressure; this breaks through any surface oxide or contamination and
effects a cold weld between the freshly exposed metallic surfaces. The
ultrasonic vibration in the plane of the bond imposes a varying horizontal
stress level with the maximum stress near the outer part of the contact. |
 |
|
As device sizes continue to decrease and the
number of connections (i.e., the pin count) to increase, in future devices
it will be necessary to reduce both the sizes and spacing of bond pads. By
the same token, wire bonding will require improvements in process speed. It
is clear, therefore, that the bond pad structure should exhibit consistent
and optimised mechanical properties. |
|
 |
|
Results
Two specimens were investigated, each
with multiple-layer bond pads of area 150 µm x 150 µm. Two indentations were
produced in each specimen using the measurement parameters given below. The
indentations were positioned using the NanoTest high resolution optical
microscope, and were at least 30 µm away from the edge of the bond pad.
It is clear from the figures that
whereas a well-behaved indentation response was obtained from the first
specimen, the loading curves for the second specimen contained anomalous and
reproducible discontinuities due to either adhesion failure or fracturing
phenomena. |
|
 |
|
Experimental Parameters
|
|
Depth or Load Controlled |
Load |
|
Maximum Load |
300 mN |
|
Initial Load |
100 µN |
|
Loading Rate |
3.99 mN.s -1 |
|
Dwell Time at Maximum Load |
0 s |
|
Diamond |
Berkovitch |
|
Z Distance Between Indentations |
50 µm |
|
Y Distance Between Indentations |
0 µm |
|
Retraction Between Indentations |
20 µm |
|
Analysis Method |
Doerner and Nix |
|
Depth Resolution |
0.457 nm |
|
Load Resolution |
16.55 µN |
|
No. of Measurements |
2 |
|
Amplifier Gain |
15% |
|
Filter |
None |
|
|