0
RESEARCH PAPERS

# Experimental In Situ Characterization and Creep Modeling of Tin-Based Solder Joints on Commercial Area Array Packages at $−40°C$, $23°C$, and $125°C$

[+] Author and Article Information

Center for Composite Materials,  University of Delaware, Newark, DE 19716

Jay G. Sloan, Mark A. Lamontia

DuPont Engineering Technology, Beech Street, Engineering Center, Wilmington, DE 19803-0840

Subhotosh Khan

DuPont Thermount® Business Team, Richmond, VA 23234

John J. Gillespie1

Center for Composite Materials, Department of Materials Science and Engineering, and Department of Civil and Environmental Engineering,  University of Delaware, Newark, DE 19716gillespie@ccm.udel.edu

Hereafter, the area array package assemblies connected through solder balls to the printed wiring board shall be called AAP∕PWB assemblies.

1

To whom correspondence should be addressed.

J. Electron. Packag 127(4), 430-439 (Dec 21, 2004) (10 pages) doi:10.1115/1.2070049 History: Received June 08, 2004; Revised December 21, 2004

## Abstract

The objective of this work was to experimentally determine the in situ creep behavior and constitutive model equations for a commercial area array package and printed wiring board assembly at $−40$, 23, and 125 °C through shear loading. The chip is connected to the printed circuit board by means of solder joints made of 62%Sn–36%Pb–2%Ag alloy. It was shown that the creep rate of solder ball arrays could be investigated using a stress relaxation method. Under the shear relaxation mode, the creep strain increases with temperature and can be described by a power law model with coefficients determined by finite element modeling (FEM). An analytical model was developed to describe the stress relaxation of an array with an arbitrary number of solder balls by defining an equivalent solder ball shear area as a fitting parameter. The resulting constitutive model is in excellent agreement with both FEM and experimental results at all test temperatures. A parametric study is conducted to investigate the creep response as a function of temperature for arrays consisting of a wide range of solder balls.

<>

## Figures

Figure 1

(a) Three-dimensional schematic of AAP∕PWB specimen, showing the solder ball arrays, chip, and PWB; and (b) schematic of center cross section of AAP∕PWB specimen, showing the central area free of solder balls

Figure 2

3×3 solder ball grid of half-symmetry model with specifications: ABAQUS FEA code, 20-noded elements mesh, symmetry bounded conditions on xz plane, the solder balls are constrained in the z direction on top and bottom of steel block, and move top block in the x direction

Figure 3

Shear stress distribution for 3×3 solder ball array

Figure 4

Mechanical testing fixtures: (a) fixture used for shear measurements and (b) fixture with AAP∕PWB specimen

Figure 5

Experimental results of shear relaxation test at 125 °C

Figure 6

Curve fitting of the average experimental results of shear relaxation test at −40, 23, and 125 °C

Figure 7

FE modeling of shear load test on half solder ball. The load is introduced by shearing (moving) the top surface of the solder ball in the x direction, while the bottom surface was fixed.

Figure 8

Shear stress distribution for half solder ball loaded under shear for 10 s at 23 °C. Maximum stress levels are reached at the interface with the printed wiring board and chip.

Figure 9

Creep strain distribution for half solder ball model after shear loading for 10 s at 23 °C. Maximum creep strain levels are reached at the edges.

Figure 10

Experimental results and FEM simulation for solder ball specimens tested in shear relaxation mode at −40, 23, and 125 °C

Figure 11

Fitting the analytical relaxation Eq. 16 with FEM data at 23 °C for determination of the equivalent shear area parameter (S). A good fit was obtained when S=0.045mm2.

Figure 12

Analytical prediction of P‐t curves at −40 and 125 °C using Eq. 16

Figure 13

Analytical prediction of P‐t curves as a function of number of solder balls in AAP∕PWB specimen at −40°C

Figure 14

Analytical prediction of P‐t curves as a function of number of solder balls in AAP∕PWB specimen at 23 °C

Figure 15

Analytical prediction of P‐t curves as a function of number of solder balls in AAP∕PWB specimen at 125 °C

Figure 16

Predicted degree of the relaxation (DX), after 5000 s at different temperatures, as a function of the number of solder balls. The linear fit equations are: y=−0.32x+99.8,R2=1 for 125 °C, y=−0.65x+94.7,R2=0.99 for 23 °C and −0.674x+90.9,R2=0.97 for −40°C.

Figure 17

Experimental and simulated relaxation behavior of solder balls at 23 °C using different models

Figure 18

Experimental and simulated relaxation behavior of solder balls at 125 °C using different models

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections