Examination of nickel underlayer as a tin whisker mitigator

Tin (Sn) whiskers are electrically conductive crystal structures that may spontaneously erupt from Sn-finished surfaces. Tin whiskers present a reliability hazard in electronics, because of their potential to create unintended short circuits. In the past, researchers have suggested that the addition of a nickel (Ni) underlayer between the copper (Cu) base material and Sn plating may act as a mitigation strategy for whisker formation. To evaluate this claim, samples with Sn plated directly on Cu, and samples with a Ni underlayer between the Sn and Cu, were subjected to environmental exposure to induce whisker growth. Prior to the test, all samples were stored in an office environment for 2.5 years and little to no whisker growth was observed. Testing consisted of 1000 temperature cycles (-55degC to +85degC, 10 min dwells) followed by two months of elevated temperature humidity exposure (60degC and 85%RH). During the test, whisker length sand densities on the samples were periodically measured. All whisker lengths were measured in accordance with the JESD201 standard, taking the effective shorting distance of the whisker (between the root of the whisker and the point farthest away). We present a simple methodology for whisker length measurement as an alternative to the JEDEC-suggested technique of varying the observation angle of the whisker to see its maximum length. All of the samples were found to have whiskers after the first 500 temperature cycles. Further exposure to temperature cycling and elevated temperature/humidity did not significantly increase whisker density. Samples with the Ni underlayer had a greater average whisker density (around 2900 whiskers/mm²) compared with samples where the Sn was plated directly on the Cu (around 1800 whiskers/mm²). During temperature cycling, whisker lengths were similar for both sets of samples, with average lengths of around 12 mum. Elevated temperature humidity exposure induced whiskers with lengths - greater than 200 mum exclusively on samples with a Ni underlayer. Upon completion of the experiment, whisker length and diameter data was gathered from 877 whiskers. No correlation was found between whisker diameter and its length. In addition, whisker growth angles were calculated for 588 whiskers, and then binned in 10deg intervals to see whether any preferential growth orientation existed. The results demonstrated the absence of favored growth angles; however, very few whiskers grew at angles close to the surface. Measurements of plating thickness using X-ray Fluorescence (XRF) revealed that two specimens had a Sn plating thickness of 4.5 mum while the remainder had thicknesses ranging from 6.7 mum - 9.5 mum. A thickness of 1.2 mum was measured for Ni on specimens with a Ni underlayer. Distinctly fewer whiskers were found on the 4.5 mum Sn finish (less than 200 whiskers/mm² compared to the 2000-4000 whiskers/mm² seen on thicker-plated Sn). However, longer whiskers were found on the thinner plating. Observations 1 year after exposure to the environmental test conditions found no further changes in whisker lengths or densities. Thus, massive whisker growth appeared to be due solely to exposure to the environmental test conditions.