Issue link: https://iconnect007.uberflip.com/i/304995
14 SMT Magazine • May 2014 Indeed, testing such intricate phenom- ena of tin whisker formation and growth is not straightforward, not to mention its labo- rious and costly nature. Nonetheless, a well thought-out test plan including the properly selected parameters is the prerequisite in order to draw a viable conclusion, positive or nega- tive, from the test results. As selecting testing parameters that are in sync with the intrinsic properties of the system is a critical step, it is plausible to choose the test parameters based on the anticipated underlying process and/or a postulated theory so that the tests can capture the action. In contrast to testing the mechanical behav- ior of solder joints (e.g., thermal fatigue, me- chanical shock) the test parameters should set to monitor the nucleation and growth pattern of tin whiskers or lack thereof. More impor- tantly, the tests for the intended purpose are to gauge the relative susceptibility to whiskering. Testing the absence of whiskers is as meaning- ful as the presence of whiskers. The end-game is to secure a tin-whisker-resistant system (not necessarily tin-whisker-proof) or to discern be- tween the tin-whisker-resistant and tin-whis- ker-prone systems. To this end, one has to de- fine what is deemed to be tin-whisker-resistant in a practical sense. Tests should monitor: • First appearance of whisker, if feasible • Max length of whisker at high T • Max length of whisker at low T • Density of whiskers • Overall pattern and appearance Desirably: • Rate of formation over a temperature range • Activation energy Ideally: • Accelerated test vs. real-life phenomena Among the various sources of causes and factors as presented in the literature and in my last column, Tin Whiskers, Part 4: Causes and Contributing Factors, they can be refined into primary factors (e.g., tin and alloy composition, coating thickness, coating chemistry, crystal structure, surface morphology) and aggravating factors (mechanical force, intermetallics at the interface, CTE mismatch, substrate base metal). The use of this refinement that affects the way the tests are run improves the odds in achieving the sound conclusions. Real-life stresses either introduced at or subsequent to the tin plating or during its ser- vice life may lead a different tin whisker be- havior as in accelerated tests (e.g., tempera- ture cycling, elevated temperature storage). Alloy-making process to achieve homogeneity needs to be taken into consideration. For an "impurity" system, how the process that adds elements into tin could also affect the whisker propensity. Tin whiskers occur for certain reasons. Thermodynamics shows that nature sponta- neously directs to a state of lower energy in the absence of external energy input, and the stress/strain has a tendency to be released by making changes. The fact that tin whiskers spontaneously grow out of the surface of the coating with time should be associated with the change in energy state (stress/strain) in the coating to the direction that lowers its energy state. Additionally, in order to grow whiskers, there must be a supply of the mate - rial (tin atoms) through a passable path at a rate that is fast enough in a finite time frame. There is also fundamental distinction between the crystal growth within the lattice and the growth out of surface (protruding from the surface, like whiskers). Relatively speaking, the energy to drive grain growth is very low and so it tends to oc- cur at much slower rates and is easily changed by the presence of second phase particles or sol- ute atoms in the structure. The external tem- perature (test temperature) drives the kinetics of defect dynamics in the tin layer by affecting stress relaxation and atomic mobility-related mechanisms. For instance, a high temperature (relative to tin's recrystallization temperature) is expected to impede the continued growth along the protruding direction, resulting in short whiskers. It is also worth noting that tin's recrystallization temperature changes with the level of its purity. In other words, when add- ing elements into tin, tin's behavior in relation SMT proSpeCTS & perSpeCTiveS TIN WHISKERS, PaRT 5: ImPaCT oF TESTING CoNDITIoNS continues