We are only a couple of months away from the approaching dreaded deadline of Restriction on Hazardous Substances (RoHS) implementation (July 1, 2006). However, the industry is far from even coming close to becoming RoHS compliant.
There are three main reasons for this — lack of uniformity in legislation in different European Union (EU) member countries, supply chain issues, and numerous technical issues in RoHS compliance and implementation. For example, the EU Parliament passed the RoHS and Waste Electrical and Electronic Equipment (WEEE) directives back in May 2001, but there are major countries that have yet to pass legislation for companies to follow. The companies are not obligated to follow the directive of the EU parliament but rather the legislation of the countries in which they operate.
The deadline for the WEEE directive has come and gone (August 2005 and then January 2006) but still there is no legislation in the UK. Additionally, different countries interpret WEEE regulation slightly differently, resulting in different rules in different countries. For example, some countries require a visible fee to be charged when a product is sold, others do not. If this is not enough, fines for non-compliance are different in different countries. All this creates confusion for companies that operate in multiple countries.
The problem is further compounded by ongoing attempts for exemption by major companies. More and more exemptions are being sought, even though final decisions are still pending for many exemptions. Be that as it may, very few companies will be ready for RoHS compliance by the due date. At least as far as I can tell, no one is talking about a delay for the date of implementation even though so few companies will be ready.
So what should you do?
Well, you can hide your head in the sand and hope for the best, or you can do your level best to get compliant to the extent technically possible. This will give you an edge over the companies who are hoping for a deadline extension, which seems unlikely at this point.
With tin/lead eutectic compositions, there was a big difference between the melting point (183°C) and peak temperature (220°C). One could maintain peak temperatures between 190° to 225°C, a variation of almost 35°C, and still achieve good reflow soldering results. Such a wide process window is not available for lead-free, especially since some components, such as aluminum electrolytic capacitors, restrict the maximum temperatures as well as the duration of time above 230°C to which they can be subjected. Additional restraints are dictated by low-cost laminates, plastic connectors and moisture-sensitive components, if used. To accommodate such constraints, the peak temperature in lead-free assemblies should be maintained between 230° and 245°C, a variation of only 15°C — a tight process window, indeed. This is approximately a 60 percent drop from the 35°C variation in tin/lead assemblies. The process window is further reduced if large components with high thermal mass are used with smaller, temperature-sensitive components. Large components with high thermal mass require higher peak temperatures for longer durations, and smaller, temperature-sensitive components require lower temperatures for shorter durations. This process window reduction will require tight process control and narrow temperature bandwidth across the board. Many assembly houses may have difficulty meeting such requirements, especially on complex boards without concerted time and efforts in developing reflow profiles.
This is not the whole story about complications in developing reflow profiles for lead-free. So far, we have been under the illusion that we are close to agreeing on a lead-free composition of SAC 305 (3 percent silver, 0.5 copper and the rest tin) with a melting point of about 220°C. However, there is talk of using some components with SAC 205 and SAC 105 compositions. These have 2 percent and 1 percent silver compositions respectively but with same copper (0.5 percent) composition. These alloys have higher melting points closer to 225°C. SAC 105 and SAC 205 are being suggested for applications such as PDAs and cell phones for better mechanical shock and slightly lower cost. Because a given device may not be available in the commonly used SAC 305 composition, you may end up with two or three different compositions of balls with different melting points.
Now add this to the complication of meeting different requirements for different types of components, such as the electrolytic aluminum capacitors discussed above, and you have a real challenge on your hands to develop reflow profiles with tight bandwidths. Backward compatibility issues can further compound this problem.
Let us first take up the subject of backward compatibility — a scenario in which some components are only available with lead-free surface finishes and must be used, even if a company is exempt from RoHS compliance. The reason is simple: It is not economical for many component suppliers to supply both tin/lead and lead-free versions of the same component. It is not an issue when using leaded components, such as SOIC, PLCC or fine-pitch with lead-free surface finishes. Most tin/lead components have 85 percent tin surface finish with about 15 percent lead. With the complete removal of lead in lead-free components, solderability is essentially the same, especially with the addition of a small amount of bismuth. The real problem arises when using lead-free BGAs on a primarily tin/lead board. If a tin/lead profile with a maximum peak temperature of 220°C is used, the BGA balls will not reflow or will partially reflow, creating serious solder joint reliability issues. So what type of reflow profile should be used? There are two options:
Option 1. Use the standard tin/lead reflow profile with peak reflow of 210° to 220°C for all the components except the lead-free BGAs. Do not place lead-free BGAs with other tin/lead components. After the tin/lead components have been reflow soldered, use a selective soldering process with a laser to selectively place and solder all the lead-free BGAs. The laser selective soldering system will selectively place and solder only the lead-free BGAs without impacting the neighboring tin/lead components already soldered in the convection oven. If you do not have access to a laser selective soldering system, see Option 2.
Option 2. If you want to solder all the tin/lead components along with some lead-free BGAs in the same oven, you have to use a peak temperature that will not damage the tin/lead components but also will be sufficient to reflow the lead-free BGAs. Keep in mind that you are using tin/lead solder paste because most of the components on the board are tin/lead. Therefore, a peak temperature of 210° to 220°C will be fine for tin/lead but inadequate for lead-free BGA balls with melting points of 217° to 221°C. However, a peak temperature of 226° to 228°C with 45 to 60 seconds time above liquidus (TAL) will be sufficient to reflow lead-free BGAs without seriously damaging the tin/lead components on the same board.
If the tight reflow temperature band of 226° to 228°C is difficult to achieve for soldering both tin/lead and lead-free BGAs in a backward compatibility scenario, consider selective laser soldering or find an alternative source for BGAs with tin/lead balls.
The industry has not quite settled on a lead-free board surface finish either. OSP surface finishes were once considered unacceptable for lead-free, but now high-temperature OSPs are making inroads. This is because of the problem associated with immersion silver finishes that has shown tendencies toward micro voids (also known as champagne voids) even though the problem is not found on PCBs from all suppliers and all immersion silver finish chemistries. But before we think that OSP is the panacea, keep in mind that BGA ball drops have been seen with OSPs where, depending on reflow profile, copper can migrate from the PCB pad to the top of the ball to form a ternary copper/tin/nickel intermetallic that causes the ball to drop and not connect. The solution to this problem has been to move away from OSPs to an Electroless Nickel Immersion Gold (ENIG) surface finish only to face potential issues with a black pad. What is the best surface finish to use? Well, like every thing else in life, nothing is risk-free. Look at your application and perform due diligence to select materials, processes and suppliers to minimize the problem.
So where do we go from here, considering the confusion in government legislation in different European countries, practically no legal direction in the USA except in California and possibly a serious legislation in China? And if this is not enough, the supply chain does not yet have complete availability of RoHS compliant parts, and technical challenges in processing lead-free parts are far from being solved. I believe we should see all these challenges as opportunities to build environmentally friendly RoHS-compliant products. If you do not, you will make your competitor happy because he is hoping that you will hide your head in the sand so that he can take your market share. Remember, legislation or no legislation, after driving gas guzzling cars, we feel a little less guilty when we buy that lead-free or even reduced lead cell phone to tell the boss that we are stuck in rush hour traffic and will be late for the meeting.
Ray Prasad held key technology positions at Boeing and Intel for 15 years before starting his consulting practice in 1994, which focuses on SMT, BGA, fine-pitch and lead-free implementation. In 2000, he joined BeamWorks, a laser selective assembly equipment manufacturer for lead-free and tin/lead products.
Author of the textbook Surface Mount Technology: Principles and Practice published by Walter Kluwer Academic Publishers, now also translated into Chinese, and numerous papers, Ray is a popular workshop leader at national and international conferences. A long time member of IPC, he is currently the chairman of BGA committee IPC-7095 “Design and Assembly Process Implementation for BGA.” He is the past chairman of the Surface Mount Land Pattern (IPC-SM-782) and Package Cracking (J-STD-020) committees.
Ray is the recipient of SMTA’s Member of Distinction Award and IPC President’s award for his contribution to SMTA and the advancement of the electronics industry. He is a columnist for SMT magazine and also serves on its advisory board. Ray received his BS in Metallurgical Engineering from the Regional Institute of Technology, Jamshedpur in India, and his MS in Materials Science and Engineering as well as his MBA from the University of California at Berkeley. He is a registered Professional Metallurgical Engineer.
|
