Tim Jensen, Product Manager, PCB Assembly Materials, would surely be considered one of the most knowledgeable people in lead-free soldering and solder materials. I was able to catch up with him a few days ago for a few minutes and ask him some questions. Here are his responses:
Dr. Ron (DR): Tim, RoHS has been in effect now for almost 18 months. It appears that SAC 305 (96.5% Sn, 3.0% Ag, 0.5% Cu) appears to be the alloy of choice for solder paste. First, is this true from your perspective and if so, why is SAC 305 more dominant than the near eutectic SAC 387? Second, what about lead-free alloys that contain little or no silver (to save alloy cost). Do any of these alloys make sense for a solder paste, or just for solder for wave soldering? Or is the old adage still true that the major cost of solder paste is the processing, not so much the material?
Tim: SAC305 has become the alloy of choice for much of the industry. The reason was based primarily on the fact that SAC305 is a lower cost alloy than SAC387. Studies that have been conducted to this point don't show a negative impact on reliability. Therefore, a cheaper alloy with no downsides is an easy choice. A side benefit that has been realized with SAC305 is that because of the slight plastic range, it shows a lower occurrence of tombstoning defects than SAC387. However, the cost of Sn and Ag have steadily increased over the past couple of years. Because of this cost trend, some solder users are considered even lower Ag content alloys such as SAC105 (98.5Sn/1.0Ag/0.5Cu) and SAC0307 (99.0Sn/0.3Ag/0.7Cu). These alloys are lower cost, but have some significant processing disadvantages. They don't become fully liquid until 226-227 C which creates additional constraints on the already tight Pb-Free reflow process window. The low Ag should improve the alloy's drop test performance (for portable electronics) but it probably won't be as robust as SAC305 in typical thermal cycling environments.
Because of the tight margins on most consumer electronics, the assemblers have been working hard to reduce their costs through process improvements and through price pressures on their material suppliers. As the price of solder paste drops, the impact of metals costs becomes much more important. In addition, Ag is a high cost metal, so that also causes the SAC metal costs to have a greater impact that with the Sn/Pb alloy of the past. High volume producers of electronics will likely see a cost savings by using a low Ag alloy. However, the overall cost impact to process window and end product reliability is still not well known.
DR: From your perspective, what is the approximate percentages of lead-free vs tin-lead paste used in the US and around the world?
Tim: Depending on their customer base, every solder paste supplier's spread of Sn/Pb vs. Pb-Free will be different. Using general terms, the United States and Western Europe still use a significantly higher percentage of Sn/Pb than Pb-Free. On the surface, that may be surprising. Especially considering the drive for Pb-Free has come through the EU yet they are using a lot of Sn/Pb. If you dig a little deeper, it makes more sense. A significant portion of the manufacturing remaining in these locations is often higher reliability products that are currently exempt from the RoHS legislation. Some examples would include military, aerospace, medical, and automotive. In areas such as China, a much higher percentage of the solder paste used is Pb-Free. This also makes sense as much of the consumer electronics are manufactured in these regions. Worldwide, 30-60% of the paste made by most solder suppliers is probably Sn/Pb (depending on their customer base).
DR: Many people are concerned with voiding in solder joints, yet some reports from iNEMI and others suggest that voiding has not been shown to be a reliability issue. What are your thoughts on voiding and have there been recent advancements in low voiding solder pastes?
Tim: It is pretty well recognized that there isn't a direct correlation between percent voiding and solder joint reliability. Voids can impact the solder joint reliability, but typically only if it is in the location that sees the greatest stress (typically the ball to component interface or the ball to PCB pad interface). Voids in the bulk solder rarely have any impact on the solder joint reliability. The challenge is that most x-ray analysis provides only a 2 dimensional view of the solder joint and voiding. Therefore, you can see how much voiding is present, but not where (in the vertical direction) the void is located. Since this challenge exists, the most logical approach is to say that the less total voiding there is, the less likely that there will be a void at one of the interfaces. Many of the latest generation Pb-Free solder pastes, exhibit pretty low void performance on standard BGA's. The technology advancements that Indium Corporation has been developing are focused on providing low voiding in CSP's with via-in-pad technology and under QFN's on the large thermal pad.
DR: Reliability of lead-free solder joints continues to be a concern. What are your thoughts on this issue?
Tim: There is a significant level of concern surrounding the reliability of Pb-Free solder joints in high reliability electronics. Pb-Free has been used in consumer electronics for the past several years. While it is difficult to say whether or not Pb-Free solder joints have contributed to more field failures, this limited history does at least tell us that it doesn't cause dramatic fall out on these products (with expected life span ~5 years or less). For high reliability electronics, it is difficult to know for certain the impact of Pb-Free. A significant amount of work has been done on modeling to predict the life of the Pb-Free solder joints. However, until we have real data, it is impossible to know for certain how well the models correlate to real world applications.
In addition to the solder joint reliability, there are challenges with the other materials used which could also impact the end product reliability. For example, the circuit board used in a Pb-Free process may degrade more than through a Sn/Pb process. This will likely result in an end product that is less reliable. In cases like this, going Pb-Free isn't what hurts reliability, but rather it is the processing conditions that Pb-Free requires.
DR: Well that's all we have time for today. Stay tuned for further discussions with Tim on halide-free soldering.