Tackling tin whisker test
Richard A. Quinnell, Contributing Technical Editor -- Test & Measurement World, 9/1/2005
The European Union's Restriction of certain Hazardous Substances (RoHS) initiative to reduce the use of lead (Pb) is driving the electronics industry to consider alternatives to tin-lead alloys in component plating. The industry's preference, with respect to factors such as solderability, ease of manufacture, and compatibility with existing assembly methods, would be to use pure tin plating as a simple and cost-effective alternative.
 |
| Many different types of whiskers can grow from tin platings, but all of them are potentially hazardous to electronic systems. Courtesy of NASA Goddard Space Flight Center. |
Unfortunately, pure tin and high tin-content alloys can grow fine metallic extrusions called "whiskers." The cause of their growth is uncertain, but whiskers lengthen over time after a plating has been applied or reflowed, and they can become as long as several millimeters, sufficient to create short-circuits in high-density packaging. Unlike dendrites and electro-migration, tin whiskers grow in normal environmental conditions without needing the action of a solvent, and they grow with or without electric fields present.
Whiskers can pose a significant threat to product reliability, yet many electronics manufacturers have never heard of them. As a result, they may not consider the propensity for tin whisker growth during the validation of new plating systems.
Those who do know of the phenomenon have used different methods for evaluating their likelihood, with the result that there has been no consistent evaluation of the various plating alternatives. Recently released test standards can now help the industry prepare consistent measurements of whisker growth and, thus, enable manufacturers to properly evaluate plating technologies.
Lost by a whiskerWhisker growth can cause a number of failure conditions. In high-impedance, low-current circuits, a whisker can form a permanent short-circuit; it can take as much as 50 mA to burn out a whisker. In other circuits, a whisker would form a temporary or intermittent short before being vaporized. In some space applications, however, the whisker's vaporization can create a current-carrying plasma that can then grow to a capacity of hundreds of amperes.
Tin whiskers can also have an effect even when they do not form short-circuits. In high-frequency RF (above 6 GHz) or in fast digital circuits (with a rise time under 59 ps), the whiskers act as a stub, affecting the circuit impedance and causing reflections.
In electromechanical systems, such as disk drives, whiskers may break off and cause head crashes or contaminate bearing surfaces. Similarly, they may contaminate the optics in electro-optical systems, and they may reduce the effective isolation of high-voltage components in power-supply systems.
Standard measurements ariseTo help manufacturers evaluate the various plating processes, JEDEC (www.jedec.org) has introduced JESD22A121, "Measuring Whisker Growth on Tin and Tin Alloy Surface Finishes." In addition to describing a series of tests, the standard provides a consistent inspection protocol for measuring the results and offers a standard reporting format for documenting them. Embracing the standard would give the industry a consistent means of comparing plating alternatives.
JESD22A121 contains three sets of tests designed to accelerate whisker formation under a variety of conditions (Table 1). One test uses temperature cycling; the other two tests use static storage conditions.
After performing these stress tests, a manufacturer must evaluate whisker growth. The standard specifies which surfaces to examine, how to employ optical or scanning electron microscope (SEM) imaging, and how to select samples for further study. Using the selected samples, the manufacturer must measure the occurrence and length of whiskers.
The specification also provides an optional test sample pre-conditioning step (Table 2). These four pre-test conditions are intended to simulate the environments that the component might experience:
- Condition A imposes no preconditioning and is used in conjunction with the ambient temperature/humidity storage test.
- Condition B simulates component storage, such as in a stockroom, and is used in conjunction with the remaining tests as well as being used prior to preconditioning types C and D. It applies to samples that do not use whisker mitigation techniques such as under-plating or post-bake mitigation.
- The remaining two conditions mimic the temperature cycles used in the soldering of components onto a circuit board. Condition C simulates backward-compatible soldering processes that use tin-lead. Condition D simulates assembly using lead-free soldering processes.
If the standard is used as part of a qualification test, companies should be certain to agree on the preconditions to be used in the tests.
Another optionFor those without the interest or means of conducting their own plating evaluation tests using the JEDEC standard, there are other options. One comes from the International Electronics Manufacturing Initiative (iNEMI; www.inemi.org), which developed the original test methods that formed the basis of the JEDEC standard.
The organization's Tin Whisker User Group, made up of 11 manufacturers of electronic assemblies, has published a report that includes an evaluation of lead-free finishes for various applications. The report, "Recommendations on Lead-Free Finishes for Components Used in High-Reliability Products, Version 3," was updated in May 2005 and is available on the iNEMI Web site.
While the JEDEC standard and the iNEMI report can help developers and test engineers pick out and evaluate a finish that will minimize their risk of tin whiskers, they do not provide a guarantee. JEDEC clearly warns that the test conditions and preconditions have not been correlated with field results under similar conditions.
Nor is there a way of extracting a failure probability from the test data. As long as the mechanisms that produce tin whiskers are not understood, the tests can only serve as a guideline and a means of comparison.
The industry has known about tin whiskers for decades, but they were not the subject of prolonged study until recently. As new information and insights are gathered, the JEDEC specification and the iNEMI recommendations will continue to evolve. But they currently provide a starting point for consistent test and evaluation, which is the first step toward solving the problem of tin whiskers in electronic systems.
| Stress Type | Ref. Spec. | Test Conditions | Inspection Intervals | Minimum Duration |
| Temperature cycling | JESD22-A104 | Min Temperature: –55 to –40°C (+0/–10°C) | 500 cycles | 1000 cycles |
| Max Temperature: +85°C (+10/–0°C), air to air; 5–10-min soak; ~3 cycles/hr | ||||
| Ambient temperature/humidity storage | 30 ±2 °C and 60 ±3% RH | 1000 hrs | 3000 hrs | |
| High temperature/humidity storage | 60°C, ±5°C and 87% RH, +3/–2% RH | 1000 hrs | 3000 hrs |
| Condition | Preconditioning Temperature Exposure | Thermal Profile Exposure | Use Guidelines |
| A | None | Normal ambient exposure | Intended to test for whisker growth under ambient temperature/humidity storage. |
| B | Room temperature storage for a minimum of 4 weeks after the finish is applied | 15–30°C 30–80% RH | Intended for samples without underplating or post-bake mitigation before exposure to high temperature/humidity storage, temperature cycling or preconditioning per conditions C or D. |
| C | Sn-Pb temperature preconditioning | Sn-Pb profile per clause 5.1.2.1 | Intended to test for whisker growth after thermal exposure to Sn-Pb SMT assembly temperatures (backward compatibility). |
| D | Pb-free temperature preconditioning | Pb-free profile per clause 5.1.2.1 | Intended to test for whisker growth after thermal exposure to Pb-free SMT assembly temperatures (Pb-free compatibility). |
