Page 135 - Wire Bonding in Microelectronics

W ir e Bond Testing 113

28 Bond shear force and pull force
vs. bond deformation for aluminum
wedge bonds
24

20
Force (gf) 16

12

8
Bond pull test
4 Shear test, unpulled bonds
Shear test on wedge after pulling
0
Low Medium High
deformation deformation deformation
1.4 1.8 2.8
Deformation (wire diameters)

FIGURE 4-22 Data from bond-pull tests and bond-shear tests on ultrasonic
aluminum wedge bonds of 25 µm (1 mil) diameter wire on aluminum
metallization. The error bars represent one standard deviation of the mean
above and below the mean breaking strength ~20 bonds [4-9].

These results may be understood if the metallurgical nature of the
bond is considered. For Al ultrasonic wedge bonds, the bond heel
becomes metallurgically overworked and weakens as the bond defor-
mation increases, but the amount of the welded area increases simulta-
neously. The pull test is particularly sensitive to the weakening of the
bond heel. Therefore, the pull force decreases as the deformation
increases. The shear test, on the other hand, is completely independent
of the condition of the heel; it is sensitive only to the actual amount of
welded area. A high-shear value could be obtained from a wedge bond
with a cracked or completely broken neck (heel). From the above, it is
apparent that the shear test is not very useful for evaluating Al wedge
bonds made from small-diameter wire, particularly as the bond defor-
mation increases above two wire diameters. However, since current
shear testers can be easily positioned within several micrometers, they
are occasionally used to evaluate the welded interface on small Al wedge
bonds when the problems described above are understood. The mini-
mum wedge bond deformation recommendation for high frequency
bonds (made with ≥ 100 kHz) is approximately 1.25 × wire diameters.
Thus, shearing such with modern equipment is generally feasible