Several points are: * There�s quite a bit of background on IMC in the fine SMTnet Archives. I�ll try not to duplicate it. * If you have a smooth and well-formed solder connection using fairly common materials, you assuredly will have IMC. Any evidence of an IMC layer is proof of a metalurgical bond to the base metal. * You can look at and measure your IMC using common microsection analysis and an optical microscope. You probably don�t need to etch, unless you�re into detailed analysis. * IMC sometimes get a bad rap from well meaning people. Every proper solder connection will have IMC. Your job is to minimize the development of the IMC during your watch. You minimize IMC formation by: - Buying boards with thin IMC when reflow soldering. - Keeping the temperature as low as possible, while still making good product. - Keeping the time at temperature as short as possible, while still making good product. * Thick IMC are not necessarily bad, but thinner IMC are better than thicker IMC, all other things the same. * Based on a CALCE study: IMC grow over time as a function of temperature above absolute temperature, diffusion coefficients, and what not. So, at 80�C (353�K), Time||IMC Thickness 10 sec||0.02 microns 1000 sec (~15 min)||0.2 microns 100000 sec (~1 day)||2 microns 10000000 sec (~100 days)||20 microns As a bonus, I will give you their formula, providing you don�t ask me to explain it, any of the factors, or coefficients. * From that same CALCE study: For a standard solder joint, there is drop off in strength after the intermetallic thickness exceeds approximately 8-10 microns. * Solder gurus do not agree with each other about IMC, probably because times change and peoples' intrepetation of data differ. * I like Howard Manko, but it�s difficult to understand him for: - First, stating in �Solders & Soldering" that IMC are not necessary for a solder connection. - Then about a page later, almost as an after-thought, stating that the �IMC are not necessary� thing he referred to earlier worked �because liquid sodium is capable of wetting solid tantalum and niobium with out these reactions.� [the formation of the IMC]. - Then later in the same book, going on to discuss Sn/Cu IMC.
reply »