Relative damage-index based on the leadfree interconnect transient strain history from digital image correlation, explicit finite-elements, cohesive-zone elements, and component's survivability envelope has been developed for life-prediction of two-leadfree electronic alloy systems. Life prediction of pristine and thermally-aged assemblies, have been investigated. Solder alloy system studied include Sn1Ag0.5Cu, and 96.5Sn3.5Ag. Transient strains during the shock-impact have been measured using digital image correlation in conjunction with high-speed cameras operating at 50,000 fps. Both the board strains and the package strains have been measured in a variety of drop orientations including JEDEC horizontal drop orientation, vertical drop orientation and intermediate drop orientations. In addition the effect of sequential stresses of thermal aging and shock-impact on the failure mechanisms has also been studied. The thermal aging condition used for the study includes 125°C for 100hrs. The presented methodology addresses the need for life prediction of new lead-free alloy-systems under shock and vibration, which is largely beyond the state of art. Three failure modes have been predicted including interfacial failure at the copper-solder interface, solder-PCB interface, and the solder joint failure. Explicit non-linear finite element models with cohesive-zone elements have been developed and correlated with experimental results. Velocity data from digital image correlation has been used to drive the attachment degrees of freedom of the submodel and extract transient interconnect strain histories. Explicit finite-element sub-modeling has been correlated with the full-field strain in various locations, orientations, on both the package and the board-side. The survivability of the leadfree interconnections under sequential loading (thermal aging and shock-impact) from simulation has been compared with pristine circuit assemblies subjected to shock-impact. Sequential loading changes the failure modes and decreases the drop reliability as compared to the room temperature experimental results. Damage index based survivability envelope is intended for component integration to ensure reliability in harsh environments.

Explicit submodeling and digital image correlation based life-prediction of leadfree electronics under shock-impact

In this paper, a design-envelope approach based on optical feature extraction techniques has been investigated for drop and shock survivability of electronic packaging has been presented for 6-leadfree solder alloy systems. Solder alloy systems investigated include, Sn1Ag0.5Cu, Sn3Ag0.5Cu, Sn0.3Ag0.7Cu, Sn0.3Ag0.7Cu0.1Bi, Sn0.2Ag0.7Cu0.1Bi- 0.1Ni, 96.5Sn3.5Ag. Previously, digital image correlation (DIC) has been used for measurement of thermally-induced deformation and material-characterization. In this paper, DIC has been used for transient dynamic measurements, and optical feature extraction. Board assemblies have been subjected to shock-impact in various orientations including the JEDEC zero-degree drop and the vertical free-drop. Transient deformation has been measured using both digital image correlation and the strain gages. Measurements have been taken on both the package and the board side of the assemblies. Accuracy of high-speed optical measurement has been compared with that from discrete strain gages. Package architectures examined include-flex ball-grid arrays, tape-array ball-grid arrays, and metal lead-frame packages. Explicit finite-element models have been developed and correlated with experimental data. Models developed include, smeared property models, Timoshenko-beam models, and explicit sub-models. The potential of damage identification and tracking for various solder alloys has been investigated. Data on identification of damage proxies for competing failure mechanisms at the copper-to-solder, solder-to-printed circuit board, and copper-to-package substrate has been presented. Design envelopes have been developed based on statistical pattern recognition. The design-envelope is intended for component integration to ensure survivability in shock and vibration environments at a user-specified confidence level.

Design Envelopes and Optical Feature Extraction Techniques for Survivability of SnAg Lead-free Packaging Architectures Under Shock and Vibration

http://ws2.binghamton.edu/park/publications/2010-1_Da_Yu_Free_Drop_Test.pdf

Dynamic responses of PCB under product level free drop impact

We investigate the drop reliability of different solder joint microstructures using a coupled board-level finite-element (FE) analysis and the microscale peridynamic (PD) simulations that can capture fracture in the heterogeneous solder joint. A new PD model for elastic behavior across a material interface is introduced and used in computing the damage and failure in the two-phase microstructure of a solder joint. This new model eliminates oscillations in strains at an interface. The microstructural geometry is digitized from the scanning electron microscopy images of experimentally tested samples. To simulate the drop test conditions, we employ the input acceleration method in the board-level FE model and calculate nodal velocities for the boundary conditions imposed on the microscale PD model. We test two different microstructures of intermetallic component (IMC) (AuSn4) and SAC305: flash and thick Au content. The results show significantly more fracture in the samples with larger Au content compared to the flash samples, which correlate well with experimental observations. The computed results show that the failure mechanism for solder joints with high Au content is a fracture through the AuSn4 IMC as well as along the interface between the inclusions and the matrix. A newly introduced measure, the quasi-damage index, estimates the locations in the microstructure of the solder joint with the highest risk for initiation of cracks and damage.

The Effect of Solder Joint Microstructure on the Drop Test Failure—A Peridynamic Analysis

Electronic packages subjected to drop and shock have been simulated using alternative theory known as peridynamic theory in the realm of FEM. In peridynamics, problems are formulated in terms of integral equation which is a substitute to conventional solid mechanics theory where problems are formulated in terms of partial differential equations. This alternative approach is more potent than conventional finite-element method because the integral equations remain valid even at discontinuities which enables to model crack initiation and crack propagation which can occur at multiple locations simultaneously. Previously, Agwai [2008] has demonstrated the use of peridynamic models using EMU code. In this paper, a peridynamic-theory based drop and shock model has been developed via finite element code using truss elements [Macek 2007]. The method enables the implementation of peridynamic theory in a commercial finite-element platform. Model predictions of peridynamic-models have been validated with experimentation for leadfree solder alloys system.

Peridynamic-models using finite elements for shock and vibration reliability of leadfree electronics

Product durability due to drop shock is a critical element for assessment of reliability for handheld devices. So far, no simulation model for a board-level drop has been extensively validated by experiments for its predictions of global (full-field) dynamic response. Accelerometers and strain gages which are traditionally utilized to measure the response at selected locations, fail to assess the global strain gradients and complex mode shapes. In this work, a novel non-contact optical technique has been proposed for measurement of full-field impact response. Pair of synchronized high-speed cameras capture the images of board assembly subjected to JEDEC standard impact, at rates up to 15,000 pictures per second. A digital image correlation (DIC) system has been integrated with the cameras to analyze the acquired images to give dynamic deformation, shape and strain over the entire surface of board during impact. A finite element model for the drop test has been developed using ANSYS/LS-DYNA. The numerical solution has been fully validated against experimental measurements of acceleration, strain and warpage at series of instants of time after impact. Effect of tightening torque at PCB mounts has been studied comprehensively with regards to the eigenvalues and mode shapes, and the necessity for accurate modeling of dynamic contact conditions existing at the supports has been demonstrated. Simulation model has been further used to assess the drop impact reliability of components on the board. Excellent correlation of all simulation results with the measured data validates the experimental and numerical propositions made in this work for analyzing a board-level drop impact.

Transient Dynamic Simulation and Full-Field Test Validation for A Slim-PCB Of Mobile Phone under Drop / Impact

Small form factor WLP (or WLCSP) type component has been increasingly applied to portable electronic devices. However, wafer level package type active components are more vulnerable to the failure than the passive or conventional BGA or QFP type active components. WLP constitute a neck point in terms of PCB assembly reliability. Therefore, it is very important to optimize the structure and material to secure high reliability of PCB assembly. Three variables are chosen for WLP reliability evaluation. They are solder ball size, PCB pad size, and underfill (use or non-use). DOE (design of experiment) technique is adopted to analyze variables effects on WLP solder joint lifetime. Board assembly level thermal cycling, high temperature and high humidity, and 4-point bending tests are employed as output parameters. The TC test results are compared with the finite element analysis. Solder joint cracks were mainly initiated from the triple point in IC side where solder, redistribution layer, and UBM meet together. WLP ball size was the major factor to determine solder joint lifetime whereas PCB pad size was not a critical parameter in all three reliability tests. Ironically, for the small ball sized WLPs, underfill decreased thermal fatigue lifetime. This is quite a different result compared with other previous reports. This discrepancy is presumably due to improper combination of underfill material property with high stress induced WLP passivation material, BCB (Benzocyclobuten).

Tae-Sang Park

Papers

Transient Dynamic Simulation and Full-Field Test Validation for A Slim-PCB Of Mobile Phone under Drop / Impact

Wafer level package solder joint reliability study for portable electronic devices