Minimizing Tin Whiskers

Estimated reading time: 9 minutes

The iNEMI Tin Whisker User Group was formed to minimize the risk of failures from tin whiskers. This article reviews group recommendations, which reflect the latest data available regarding tin whisker formation and mitigation, including information on commonly used lead-free finishes and their effectiveness in reducing tin whiskers.

By Joe Smetana

Unalloyed tin electroplating has a long history of whisker formation and growth, resulting in reliability problems for various types of electronics equipment. The predominant whisker mitigation strategy for more than 50 years has been the addition of lead (Pb) to tin plating. Legislation that will eliminate the use of lead in electronics products sold in the European Union (EU) - due to be implemented July 1, 2006 - has led many electronics component suppliers to propose the removal of lead from tin/lead (SnPb) plating, leaving essentially pure tin. This approach is the most convenient and least costly lead-elimination strategy for the majority of component manufacturers. However, this approach presents reliability risks to the high-reliability user community due to the whisker-forming tendencies of pure tin and tin-alloy plating (Figure 1).

Figure 1. Whiskers grown on component leads. This test used a 64 LQFP with 10 µm (target) of matte Sn over CDA 194 alloy lead-frame.

No scientific consensus on whisker formation and growth fundamentals exist. Nor is there a standard set of accelerated tests that can relate whisker formation and growth during testing to field conditions with any reasonable degree of certainty. Therefore, any claims of a “whisker-free” tin-plating processes, or guaranteed lifetimes without whisker failure, must be regarded with skepticism.

The Tin Whisker User Group, comprised of 11 large, high-reliability electronics assembly manufacturers, was organized by members of the International Electronics Manufacturing Initiative (iNEMI) to develop recommendations for lead-free surface finishes for high-reliability electronic applications. The group’s intention is to minimize the risk of failures from tin whiskers, and recommend a combination of known-mitigation practices, process controls and some level of testing.

Figure 2. A tin whisker growing on a Cu-based lead-frame IC with an Sn finish. This whisker is in the dumbar region of the component.

It is the consensus of the iNEMI User Group that pure-tin electroplating presents a risk in high-reliability applications. However, there are cost-effective alternatives available to minimize this risk. This article examines commonly used lead-free finishes and their effectiveness in reducing tin whiskers.1 Whisker experimentation has been inconsistent relative to growth rates, incubation times and other parameters. The mitigation practices and techniques outlined here are not always effective in reducing tin whiskers, and should not be construed as a means of preventing whisker growth (Figure 2). They are methods for reducing whisker risks. Manufacturers should evaluate these alternatives in terms of reliability risk and cost benefits for their market applications.

Board Finishes

Silver - Silver finishes are not prone to whisker growth in most environments. However, rapid growth of silver whiskers or dendrites may form in the presence of H2S (found in some cases where environmental air pollution contains SO2). Many users also prefer to avoid silver finishes due to potential issues with electromigration and reduced solderability shelf life.

Immersion tin - Immersion tin is a chemical-displacement process that results in a relatively thin (<40 µin. or 1 µm) and stress-free tin film. Whiskers have been grown immersion tin; however, whisker lengths typically are less than half the threshold of concern. Throughout the years, or through several high-temperature thermal cycles, this finish converts to SnCu intermetallics entirely (due to the thin amount of tin). Solderability shelf-life considerations make this finish inappropriate for electronic components. It is, however, one of the lead-free-finish options for PCBs.

Component Lead Finishes

Non-tin plating - Nickel/palladium/gold (NiPdAu) or nickel/palladium should be considered for lead-frame applications. This plating has more than ten years of history in field applications, and early solderability issues have been resolved. NiPdAu is not prone to whisker growth in most environments, although gold can grow whiskers in certain environments. This finish is strongly recommended for most lead-frame applications to essentially eliminate whisker growth. Users should be aware that molding compounds do not adhere as well to noble metals, such as Pd and Au, as they do to copper (Cu). Therefore, it may be more difficult for NiPdAu packages to achieve MSL 1 and 2 performance at the higher temperatures associated with SnAgCu lead-free assembly. NiPdAu also has demonstrated corrosion in accelerated tests using high hydrocarbon and sulfur atmospheres. This corrosion has not been noted in actual field conditions.

Nickel underlay - Adding a nickel (Ni) underlay between tin plating and a copper base metal mitigates whisker formation by alleviating a major source of compressive stress in the tin film. This is thought by most to be the driving forces for whisker growth. The nickel prevents the formation of SnCu intermetallics that can preferentially grow in the grain boundaries of tin, and result in compressive stress in the tin layers. The thickness, porosity and ductility of the nickel plating are important to ensure an effective barrier layer for copper, even after lead forming. Similarly, control of the tin-bath impurities, particularly copper, is important if this underlay is to be effective.

Silver underlay - Adding a silver (Ag) underlay between tin plating and copper base metal has been proposed to mitigate whisker formation, similar to nickel underlay. As with nickel, silver underlay is expected to eliminate SnCu intermetallics and resulting stress build-up in the tin layers. There is limited test data supporting the effectiveness of silver underlay for whisker mitigation. Further investigation of this technique is recommended.

Fusing - Fusing tin plating shortly after plating (within 48 hours, for example) mitigates whisker formation. Fusing is a reflow operation, usually done by dipping the tin-plated surfaces into a hot oil bath. The fusing process affects both the intermetallic layer and the grain structure of the tin. The result is a low-stress finish that has an excellent field history without tin whisker problems. However, fusing done as part of the PCB assembly process does not appear to be an effective mitigation practice and may, in some cases, increase whisker growth.

Hot-dip tin - Hot-dip tin is a molten tin-bath process. Although not prevalent in lead-frame construction intended for electronics components, it has been used for structural steel parts, connectors and devices such as relays. Hot dipping is considered to be whisker-free. There is evidence that this mitigation process may not be effective with pure tin. Hot dipping with Sn4Ag or SnAgCu is a generally effective mitigation practice.

Annealing/heat treating - Heat treating at 150°C for one hour has shown promise as a tin whisker mitigation technique when used on matte tin-plated copper alloy lead-frames. This approach has been endorsed strongly by the IC packaging industry. However, conflicting test data for this mitigation practice continues to cause concern among users. The differences in test results may be related to the quality and repeatability of the Sn plating itself, irrespective of the mitigation practice. In some cases, it may be due to differences in test-and-inspection conditions or sample sizes.

Matte tin - Matte tin is a tin film with lower internal stresses and larger grain sizes than “bright tin.” It is also less prone to whisker formation and growth than bright tin film. Matte and bright tin finishes are defined in Table 1. Many suppliers claim a proprietary version of matte tin is “whisker-free.” These claims are most likely premature, and should be considered with skepticism.

Tin-bismuth - Tin-bismuth (SnBi) alloy finishes are controversial when used in conjunction with eutectic SnPb solder, but when added to tin in amounts of 2 to 4% by weight, they may aid in suppressing whisker growth. There is a ternary SnPbBi peritectic with a melting point of 135°C that is thermodynamically viable for small additions of Bi. There also is an SnPbBi eutectic with a melting point of 96°C. With lead-free solder, SnBi is a viable candidate for component finishes. With eutectic tin/lead solder, it will be necessary to control the bismuth content of the finish between 3 to 5%, so as to have enough bismuth to suppress whisker formation without getting into the compositional range of the ternary eutectic. Keeping the Bi content low also is necessary to retain solderability of formed leads. Some data suggest that even low amounts of Bi in SnPb solder joints affect fatigue performance of the solder joint adversely. This problem has been noted on Alloy 42 lead frames primarily.

Plated tin-silver - Plated SnAg (2 to 4% Ag) alloys have, in limited testing, shown promise for reducing tin whisker growth. Further investigation of this finish for tin whisker mitigation is encouraged.

Thick finishes - Industry data indicate that thicker tin finishes show a lower propensity for tin whiskers and/or a greater incubation time before tin whiskers occur. Tin thickness for components without a nickel or silver underlayer should be 7 µm minimum, with a preferred nominal thickness of 10 µm or greater. When nickel or silver underlayer plating is used, the minimum tin thickness should be 2 µm to ensure shelf life for good solderability.

Surface chemical etching - Surface chemical etching of copper-based alloys prior to plating has, in limited testing, shown promise in reducing tin whisker growth when the etching depth is in the range of 3 to 4 µm. Further investigation of this technique for tin whisker mitigation is encouraged.

Conformal coating - The use of conformal coating after assembly has shown some promise in reducing the rate of whisker growth. Its success appears to be specific to the material types used and the environment. Data does not support conformal coating as a cure for whisker growth; however, it does add an insulation barrier that may prevent shorting, should long whisker growth occur.

Conclusion

Until we can isolate and understand the causes of tin whisker growth, the industry must take measures to minimize their occurrence. iNEMI’s Tin Whisker User Group has outlined recommendations for surface finishes and mitigation strategies, along with acceptance criteria, to reduce whiskers, especially in high-reliability products. The recently released JEDEC standard, JESD22A121, “Measuring Whisker Growth on Tin and Tin Alloy Surface Finishes,” also provides an industry-standard method for measuring and comparing whisker propensity for different plating or finish chemistries and processes.2 By combining the practices outlined in these documents, manufacturers can reduce the risk of tin whiskers in their products.

Acknowledgement

The author recognizes the iNEMI Tin Whisker User Group for their development of the recommendations for lead-free finishes, upon which this article is based.

Click here to download a table summarizing results of a survey of the iNEMI Tin Whisker User Group regarding conditions under which they would accept or not accept individual mitigation practices.

REFERENCES

1. Recommendations on Lead-Free Finishes for Components Used in High-Reliability Products, version 3, iNEMI Tin Whisker User Group (updated May 2005); http://www.inemi.org/cms/projects/ese/tin_whisker_activities.html.
2. JESD22A121, “Measuring Whisker Growth on Tin and Tin Alloy Surface Finishes,” published May 2005; http://www.jedec.org/download/.

Joe Smetana, principal engineer, advanced technology, Alcatel, and chairman of the iNEMI Tin Whisker User Group, may be contacted at (972) 519-6135; e-mail: joseph.smetana@alcatel.com.