Reliability investigation of printed wiring boards processed with water soluble flux constituents

Abstract

The purpose of this research was to investigate the factors that enhance conductive anodic filament (CAF) formation in printed wiring boards. The variables studied were (1) flux formulation, (2) conductor spacing, (3) operating voltage, and (4) temperature.

A Weibull distribution of failure times due to CAF was observed. A novel test circuit was designed and implemented that allowed the mean time to failure to be determined for boards processed with three different fluxes, at 0.5 mm and 0.75 mm conductor spacings and at 150V and 200V. The boards were aged at 85%RH and a temperature of 75°C, 85°C or 95°C.

It was found that the flux formulation affected the rate of CAF formation. A modified linear aliphatic polyether flux with a chloride activator had a significantly different activation energy than control printed wiring boards or those boards processed with a poly(ethylene/propylene) glycol flux or a poly(ethylene/propylene) glycol flux with a bromide activator.

The addition of bromine to a poly(ethylene/propylene) glycol flux decreased the rate of CAF formation as compared to poly(ethylene/propylene) glycol without a halide activator. The inter-relation between voltage and conductor spacing was quantified as a L4/V2 relationship for the plated through hole test pattern used in this study. 325V/mm was found to be a critical voltage gradient for the formation of CAF. The maximum temperature of the reflow profile also greatly enhances CAF formation and decreases the mean time to failure.

Microscopic analysis showed distinct differences in CAF morphology between the various processed boards. Control boards had small halo-like CAF formations around a separated fiber / epoxy interface. CAF that formed on boards processed with poly(ethylene/propylene) glycol or poly(ethylene/propylene) glycol with a bromine activator had a stratified appearance that penetrated well into the epoxy. Boards that were processed with the modified linear aliphatic polyether with chlorine activator had a striated morphology that also penetrated into the epoxy. All CAFs were consistently copper and chlorine containing despite the use of a bromine containing flux. Electron diffraction revealed that a CAF observed in this study was synthetic atacamite. Stainless steel (i.e., iron, nickel, and chromium) residues were also observed as a result of drill bit breakage during PTH formation.