Reducing and Preventing Tin Corrosion Whiskers

Estimated reading time: 7 minutes

By Olaf Kurtz, Ph.D.; Peter Kühlkamp, Jürgen Barthelmes; and Robert Rüther, Atotech Deutschland GmbH

For the connector and IC/leadframe industries, corrosion can occur on metalized components and products, resulting in malfunctions. The JEDEC Solid State Technology Association recommends the use of elevated temperature and high relative humidity (RH) test parameters for qualifying all tin electrodeposits. Extensive studies have therefore been carried out to characterize the chemical and physical parameters involved for whisker formation from matte and bright tin electrolytes, quantifying their influence and deriving various mechanisms for whisker formation.1-4 The industry needs systematic investigations to understand the phenomena of corrosion whisker formation and possible protection means to suppress or even prevent their onset under different storage conditions. Unfortunately, JEDEC/IPC publications on tin whisker formation provide no recommendations on prevention or suppression of corrosion whiskers.5

This article presents the work done on an effective anti-corrosion treatment* that can mitigate corrosion whisker formation effectively. The process benefits to tin solderability after storage at elevated temperature and humidity are highlighted, with emphasis on discoloration and electrical properties (e.g. contact resistance) of the finished coatings.

Investigations were made into the mechanism of whisker formation under high temperature and humidity conditions. The three storage conditions recommended by the JEDEC for the investigation of whisker growth on tin deposits include thermal cycling tests using a temperature range of -55° to +85°C, at a defined heating and cooling rate involving 1000 cycles. Furthermore, JEDEC standards describe isothermal storage and test conditions with an elevated relative humidity of 60 and 85% and temperatures of 30° and 55°C.6-8

Since tin has a relatively low melting point of 232°C, the inner crystallographic stresses occurring at elevated temperatures of 55°C can be relieved by thermally induced relaxation processes.9-11

If the relative humidity is then increased as described by JEDEC corrosion generally is observed on tin surfaces, leading to the growth of corrosion or tin whiskers. One explanation for this corrosion-related growth is due to the formation of a thin oxide film with a lower density than that of the original pure tin deposit. The lower density and resulting increase in volume of this oxide layer cause an increase in the exerted pressure at the deposit grain boundaries, increasing internal stress. This corrosion effect is accelerated significantly if exposed copper is present immediately adjacent to the tin, as is the case in ICs produced using the trim-and-form procedure or for connectors manufactured from tin-coated strip material.12, 13Under the influence of moisture and tramp ions, such as chloride or sulphate, the corrosion is increased further. Resulting whiskers can grow to several hundreds of microns long, with preference for the lead toe and dam bar interfaces.

ExperimentsA detailed study of influencing factors that promote or inhibit corrosion whisker growth was undertaken (see Acknowledgements) to identify the most important influencing factors for the promotion and suppression of corrosion whiskers. During the first study phase, post-dip treatments were selected for neutralization or active protection of the tin surface. A total of 30 combinations of the post-dip treatment were investigated at different process steps during this preliminary study to qualify and quantify their influence on whisker growth. To investigate the influence of ionic contamination, individual test samples were contaminated by immersion into identical concentrations (2 mmol) of chloride and sulphate solutions.

The study was carried out on ICs in a PSO 20 housing of C194 copper base material. The outer leads on the ICs had previously been coated with approximately 10 µm of a matte tin plating bath. After the corresponding treatment with the selected post-dip, the separated ICs (after the trim-and-form procedure) were placed in the so-called dead bug position, i.e. with the leads facing upwards, on a custom test holder. The samples were stored in this position in a climate chamber at 60°C and 85% RH for 4,000 hours.

Results to DateFrom the design of experiment (DOE) analysis, we saw a significant reduction in the corrosion behavior of tin electrodeposits using post-dip neutralization and final protective treatment. Surface chloride contamination exhibits a significant influence on both corrosion whisker growth and maximum whisker length (MWL).

Based on these successful preliminary studies, anti-corrosion agents are developed that can effectively retard or prevent the onset of corrosion whisker formation, resulting in significant improvements of technical properties such as solderability and contact resistance.

Corrosion Whisker TestsAfter proving the corrosion-inhibiting benefits of an anti-corrosion agent, corrosion whisker inhibition tests were undertaken using JEDEC storage conditions of 55°C/85% RH for 4,000 hours.

During this study, various IC packages were tin plated (10 µm), then treated with anti-corrosion agent and tested prior to trim-and-form processing. IC type FW 25 with FPG base material showed 0.1% Fe, 0.03% P, < 0.05% Imp., and the remaining Cu. PDIP 32 on C194 showed 2.4% Fe, 0.03% P, 0.012% Zn, and the remaining Cu. Finally, an LQFP on C7025 showed 3% Ni, 0.65% Si, 0.15% Mg, and the remaining Cu. All three package types were observed to be whisker-free after 4,000 hours in the climate chamber at 55°C/85% RH (Figure 1).

Overall, the anti-corrosion agent exhibits a strong growth rate inhibition of corrosion whiskers, even in the presence of chloride contamination.

SolderabilityAs a result of this effective corrosion protection, the packages' solderability after steam aging and pressure cooker test improved. This is particularly important at low reflow temperatures of 220°C or below (Figure 2). Solderability tests after steam aging were made to estimate the anti-corrosion effect. A non-activating flux was used for all the soldering results described.

Each wetting diagram curve in Figure 2 represents the mean of ten individual measurements. The measured zero crossing for the wetting is shown as zero crossing time (ZCT).

Anti-corrosion agents also can be used for protection of BGAs. Cleaning with the new post-treatment was performed after application and reflow of tin/silver (Sn/Ag) solder beads, prior to electric contact resistance measurement. Testing showed that yield and contact resistance properties of the BGAs could be improved significantly.

ConclusionOn the basis of these preliminary studies, a post-dip solution was developed to reduce or prevent corrosion attack and the growth of corrosion whiskers under JEDEC test conditions. Solderability improved significantly. It was discovered that the degree of improvement after treatment of the samples with anti-corrosion agent also improves according to the storage time and the increase in the corrosive test parameters in direct comparison with un-treated samples. Reproducible results were achieved. These results are of significance not only for the IC/leadframe industry, but also for the plug connector industry.

After treatment, the tin coatings showed no discoloration at the surface, after 24 hours in the pressure cooker test.

It was also proven that use of an anti-corrosion solution for protection of solder beads in BGA housings also allowed significant improvements in yield and contact resistance to be achieved.

*The anti-corrosion agent developed from these findings is Protectostan LF.

ACKNOWLEDGEMENTS We extend our warmest thanks to Paolo Crema of ST Microelectronics Italy and to Joseph Gauci, Adrian-Michael Borg and Robert Caruana of ST Microelectronics Malta for the cooperation, discussions, and the provisions of the experimental results and their evaluation, as well as their permission to publish the majority of these results.

REFERENCES1. J. Barthelmes, P. Kuehlkamp, F. Lagorce-Broc, O. Kurtz, D.-G. Neoh, G. Barlowe, "Quantification of the Influence of All Important Physical and Chemical Tin Plating Bath Parameters on the Propensity for Whisker Formation," Sur-Fin USA 2007.2. J. Barthelmes, P. Kuehlkamp, F. Lagorce-Broc, S. W. Kok, D.-G. Neoh, "Different Storage Conditions have Different Whisker Growth Mechanisms Can Bright Tin be an Alternative to Matte Tin?," IEMT Malaysia 2006.3. J. Barthelmes, P. Kuehlkamp, D.-G. Neoh, "Quantification of the Influence of All Important Physical and Chemical Tin Plating Bath Parameters on the Propensity for Whisker Formation," SEMICON Singapore 2006.4. J. Barthelmes, P. Kuehlkamp, F. Lagorce-Broc, P. Lam, "Pure Bright Tin Deposits as the Whisker--Minimized Lead-Free Technology," Productronica China 2006.5. STMicroelectronics, Nature of whiskers and whisker mitigation techniques, www.st.com/stonline/products/literature/an/10792.pdf, 2006.6. JEDEC/IPC Solid State Technology Association, "Current Tin Whisker Theory and Mitigation Practices Guidelines," JP002, March 2006.7. JEDEC Standard Solid State Technology Association "Measuring Whisker Growth on Tin and Tin Alloy Surface Finishes," JESD22A121, May 2005.8. JEDEC Standard Solid State Technology Association "Environmental Acceptance Requirements for Tin Whisker Susceptibility of Tin and Tin Alloy Surface Finishes," JESD201, March 2006.9. S. Chopin, P. Su, "Mitigating Whisker Growth on Electroplated Tin Finishes," SMT, June 2006.10. J. Barthelmes, P. Crema, P. Kuehl-kamp, "Whisker Growth during Heat-Humidity Storage: Mechanism and Factors of Influence," SEMICON Singapore 2007.11. J. Barthelmes, P. Crema, P. Kuehl kamp, J. Gauci, A-M. Borg, R. Caru-ana, "Whisker Growth from Plated Tin Surfaces during Heat-Humidity Storage: Mechanism, Influences, Mitigation," IPC/JEDEC USA 2007.12. Oberndorff P., Dittes M., Crema P., Chopin S., ECTC 2005;429-433.13. Osenbach JW, DeLucca JM, Potteiger BD, Shook RL, Baiocchi FA, ECTC NEMI Workshop, June; 2005.

Olaf Kurtz, Ph.D.; Peter Kühlkamp; Jürgen Barthelmes; and Robert Rüther may be contacted at Atotech Deutschland GmbH, Erasmusstr. 20, 10553 Berlin, Germany; olaf.kurtz@atotech.com; www.atotech.com.