An Expert Looks at the Issues
Dr. Ning-Cheng Lee on Solder and Lead-Free Substitutes
Dr. Lee is Vice President of Technology for Indium Corp. of America, Clinton, N.Y. and a widely respected expert on solder. His background includes nearly 15 years in the development of fluxes and solder pastes. He received a bachelor's degree in chemistry from the National Taiwan University, Taipei, and a doctorate in polymer science from the University of Akron (Ohio). He is a prolific author and has contributed several articles to Chip Scale Review, including
"Lead-Free Soldering of Chip-Scale Packages," in the March-April issue.
Dr. Ning-Cheng Lee
Q What are the most promising lead-free solder substitutes?
A The answer depends on two key factors: the end product and the soldering process used, i.e., reflow, wave soldering, etc. With these known, the most promising lead-free choices will be determined by the following:
Reliability (physical properties such as mechanical property, electrical property, fatigue, corrosion resistance)
Manufacturing cost (solder materials, components, boards, operating costs such as soldering equipment, soldering atmosphere and, particularly, process yield for the intended applications)
Availability of flux, solder materials, components, boards, etc., for the volume intended
For reflow applications where reliability is emphasized, a eutectic Sn-Ag-Cu system, such as 95.5Sn-4.0Ag-0.5Cu, 95.5Sn-3.9Ag-0.6Cu, and 95.5Sn-3.8Ag-0.7Cu, will be the best choice. A Sn-Ag-Bi system, with low Bi content, such as 91.8Sn-3.4Ag-4.8Bi, is also a good candidate. Its outstanding wetting performance likely would offer a better process yield. However, its sensitivity toward the presence of Pb will delay the acceptance of this system until the Pb is completely phased out of solder interconnects. If low cost is being emphasized, Sn-Zn-Bi systems would be good choices, due to their relatively low melting point; hence virtually no upgrade in thermal stability of the components and boards would be required.
Q Can lead-free substitutes be as reliable as current solders?
A The answer is "yes," if the infrastructure supporting lead-free soldering is developed properly. Currently, there are some issues that need to be clarified, including the impact of fillet lifting and the increasing extent of voiding on reliability. Both phenomena appear to be more profound in lead-free systems.
Overall, however, the limited reliability data available on lead-free solder joints strongly suggests that many of those alloy systems can outperform the eutectic Sn-Pb system. This is particularly true in terms of mechanical strength, creep properties and fatigue performance. On the other hand, being intrinsically poorer in wetting than eutectic Sn-Pb, the reliability performance is highly dependent on the progress of supporting systems, such as surface finishes of components and boards, flux technology upgrades, thermal stability of parts and reflow technology.
Q What changes will have to be made in solder reflow systems to handle solder substitutes?
A For eutectic Sn-Pb systems, the commonly used reflow peak temperature is about 40±15°C above the melting temperature of solder. For the currently prevailing lead-free systems, such as Sn-Ag-Cu, Sn-Ag, or Sn-Ag-Bi alloys, the melting temperature is around 215°C. If the same practice is applied, the peak temperature will have to be 255±15°C, or 240° to 270°C.
This, obviously, is too harsh for the parts to sustain the reflow process; hence the high end has to be lowered. Data indicates that a range of 230° to 260°C would be a good compromise between soldering and thermal stability considerations. However, lead-free solders tend to form a grainier surface texture due to a greater tendency to crystallize into larger grains. To offset this intrinsic attribute of lead-free alloys, a more rapid cooling rate is desirable. Before flux technology can be upgraded to deliver proper wetting, an inert reflow atmosphere may be needed for satisfactory soldering.
Q Lead-free solders aside, what major changes in composition or attributes are solders likely to offer in the near future?
A Sn will remain the primary solder constituent, mainly because of its high reactivity with a variety of base metals (which makes it good for wetting), and its low cost. The alloys may shift from binary to ternary or quaternary systems, such as Sn-Ag-Cu, with major emphasis on better fatigue resistance through a grain-refining approach.
Compositions with more than four elements will be rare, due to the difficulty of achieving consistent manufacturability. Some additives, such as bismuth, may be employed to enhance the wetting-either through lowering the surface tension or through promoting a reaction with base metals.
Other additives, such as indium, may be used to increase the compliance as well as the tolerance against contamination. Other changes in attributes may include better tombstoning resistance and better oxidation resistance.
Alloys for low-temperature soldering may garner interest, while Pb-containing high-temperature solders may still have room for more improvement. Still, the drive toward lead-free compounds may discourage investment in this direction.
Q What is the prospect of having more new alloys developed for lead-free applications?
A The continuous demand for better and cheaper materials will always drive the industry to search for more, new lead-free alloys. However, industry also tends to standardize on the materials used, to cut the cost of handling and processing.
At this stage, several alternatives appear to be very promising. Once the industry has settled on a few adequate alloy choices, the motivation for changing to an incrementally better material will be fairly slim, and the incentive to develop more new alloys will diminish.
Q Will the toxic landfill issue be better addressed with the recycling approach instead of the lead-free approach?
A Recycling is a more logical choice for any society with limited resources, regardless of the lead-poisoning issue. In Japan, projected landfill capacity will drop to zero by 2008. In other countries, although not as urgent as in Japan, the full depletion of landfill capacity is merely a matter of time.
Recycling not only theoretically avoids the lead-poisoning issue, but also reduces the waste and slows down the consumption of landfill capacity. However, the cost of 100% recycling will be prohibitively high and unrealistic. In addition, the waste generated from reclaiming lead will very likely also be harmful. Obviously, the best solution will be lead-free recycling.