Issue link: https://iconnect007.uberflip.com/i/339176
July 2014 • SMT Magazine 57 TIN WHISKER RISK MANAGEMENT By CoNFoRMAl CoATING continues fEATURE both regions of tension and compression. The effect of the bending was noticeably more sig- nificant for the Alloy 42 base metal test cou- pons and the regions of compression had high- er whisker density than the regions of tension. The negative effect of Alloy 42 base metal on the propensity of electrodeposited bright tin coatings to whisker has been shown in previous research. In addition, the effect of stressing tin plating resulting in increased tin whisker den- sity has also been previously reported. The conformal coatings used in this experi- ment mitigated tin whisker protrusions for the test coupons that were not stressed. Parylene coating at a thickness of 0.5 mils and both acrylic and polyurethane coatings with a mini- mum thickness of 1.0 mils did not exhibit any tenting following the 9.5 years of environmen- tal exposure to 50°C and 50% RH. Tenting was observable on the 1.0 and 2.0 mil thick acrylic coating in regions of compres- sion for the bent Copper C110 base metal sam- ples; however there were no indications of tin whisker protrusions. There was no disruption of the polyurethane coating of any thickness for the bent Copper C110 base metal samples. For the Alloy 42 base metal samples, in addi- tion to the tin whisker protrusion in the tension and compression regions for the 1.0 mil thick acrylic coating reported after 5.5 years, there was tin whisker protrusion of the 2.0 mil thick acrylic coating in the compression regions ob- served initially after 9.5 years. The 2.0 mil thick acrylic coating also exhibited tenting due to tin whisker growth in the tension regions. While there was observable tin whisker protrusions through the 1.0 mil thick polyurethane coating in regions of tension and compression for the bent Alloy 42 base metal samples, there was no observable tin whisker protrusions through the 2.0 mil thick polyurethane coating. The 2.0 mil thick polyurethane coating did exhibit tenting only in the compression regions, initially ob- served after 9.5 years. The improved tin whis- ker mitigation with thicker conformal coating is in agreement with the CALCE study stating that coatings of 1.0 mils and thickness and low modulus are at risk for tin whisker penetration. The conformal coating materials used in this testing mitigated the growth of tin whis- kers through the coating for this specific elec- trodeposited tin plating and this specific envi- ronmental exposure when there were no addi- tional stresses applied to the coupons. It should be noted that the tin plating selected and ap- plied during this experiment were intentionally designed to promote the growth of tin whiskers and would not normally be considered as an acceptable plating for component leads of real hardware. Parylene, which has a significantly higher modulus demonstrated in this experi- ment the ability to mitigate tin whiskers at a thickness of 0.5 mils; however there were no stressed (bent) samples for parylene. The bent samples indicate that stressed regions of tin plating will have a greater tendency to whisker. Additional testing on real world component leads mounted to circuit cards is warranted to determine minimum requirements for each coating type. For the tin coated braiding and stranded wire subjected to temperature and humidity exposure, there was no observable tin whisker growth on the surface following five years of ex- posure. The lack of observed tin whiskers may be due to the minimal thickness of the tin coat- ing and the lack of stressed regions within the strands. In addition, the thin tin coating may have been consumed by the tin-copper inter- metallic layer relatively quickly following man- ufacturing, resulting in the reduced risk of tin whisker formation. The braiding and stranded wire products should be considered acceptable for use in high-reliability assemblies without having to add mitigation steps to reduce the risk of tin whisker growth. The tin coated single strand wire did exhib- it a high concentration of tin whisker growth with whiskers measured up to 40µ in length. This does comply with the JESD201 maximum tin whisker length for Class 2 hardware; how- ever the presence of tin whiskers is not allow- able for JESD201 Class 3 hardware. While the JESD201 specification is not accepted industry- wide, the observed presence of tin whiskers on the tin coated single strand wire indicates that further analysis may be required to insure that the potential growth of tin whiskers on the product is accounted for in design and building of hardware. SMT