Mechanical Behaviour of Two Sorts of MCM Structures
- PDF / 969,036 Bytes
- 7 Pages / 414.72 x 648 pts Page_size
- 91 Downloads / 159 Views
Mat. Res. Soc. Symp. Proc. Vol. 390 01995 Materials Research Society
Schematically, the sort of samples we used is shown in Figure 1(a). Our samples had two different distributions of the solder bumps: one was called external distribution (Figure 1, b), the other is called external/internal distribution (Figure 1,c). The alloys constituting the bumps were: a Pb/Sn classical eutectic alloy, Pb37:Sn63; and a Pb rich alloy, Pb95/Sn5 (the numbers corresponding to the weight percent of the elements).
......... , *
Substrate (a)
0
* 00000909 (b)
0S S
0
(c)
Figure1: (a) Schematic representationof the chip on substrate investigatedsystems. (b) and (c) representthe two types of bump distribution.
Mechanical tests The mechanical characterization of these chip on substrate samples was performed by tensile and shear experiments. We set up two sort of mechanical devices for performing these experiments. Our chip on substrate systems were previously mounted and glued on silicon supports before the tests. To glue the samples to the supports was performed with a controlled force of 0.05 N, which corresponded to a 0.064 MPa compressive stress in a bump and a 10-6 order strain. For the shear experiments, the samples were located in the device and aligned with micrometric screws, and under an optical magnification. For to follow the microstructural evolution towards failure, we performed the same type of experiments but in a scanning electron microscope. In this case we designed the sample supports presented in Figure 2. With those supports not only tensile but also shear deformation, could be induced with a tensile microdevice. Experimental results Typical Force vs. Displacement curves were obtained from the experiments. Accounting of the rigidity of the devices, and considering a mean effective surface for the sample (which correponds to the number of bumps, times their contact surface with the chip or substrate), we deduced the flow stress, the maximum strength and the deformation to complete failure of the tested samples. The elastic parameters (Young Modulus, Shear modulus), were also deduced. These parameters are commented in our discussion.
154
Figure 2: Sample supports for
the in-situ tests. (a)tensile test, (b) shear test. The little squares represent the samples, the arrows indicate their location for the test.
! (a)
(b)
Microstructural observations After tensile experiments, typical elongated structures showing that the bumps were plastically deformed from spheres to cylinders were observed. For the samples deformed by shear, the rupture morphology showed the effect of the relative displacement. The in-situ tests confirmed these observations, which showed the ductile morphology of the rupture: a substantial plastic deformation ocurred in both tests (Figure 3). However, from tension as from shear experiments; performed macroscopically or in the scanning electron microscope: we observed that failure ocurred always at the same interface; the one which separates the chip from the bump. These microstructural o
Data Loading...
