Tin whiskers on plated electrical leads and printed circuit boards (PCBs) pose a threat to component reliability and safety. They can cause circuit failures, intermittent upsets and electrical arcing. Tin whiskers are electrically conductive protrusions of crystalline tin from the surface of a plating finish.
Tin whiskers already have caused the complete loss of more than one space satellite. For aviation, the whiskers can pose a threat to aircraft as they age in service. The incubation period for whisker growth can range from a few days to eight, 10 or 20 years without causing a noticeable failure, and then suddenly unexpected havoc occurs. Some particulars:
It has been demonstrated in the laboratory that whiskers can generate extremely high voltage gradients in low voltage (120-volt and even lower) circuits. Voltage gradients, which can lead to arcing and fire, can occur in the range of 1,000 volts per millimeter (25 kilovolts per inch) on 120-volt circuits. Worse, the tin can vaporize into a highly conductive plasma of tin ions capable of carrying hundreds of amperes. Indeed, the arcing can be sustained by tin evaporated from the area immediately surrounding the arc.
The whiskers can break away, creating debris contamination that can cause short circuits in areas remote from the whiskers' origins (a troubleshooting nightmare).
Like sandpaper, whiskers can abrade polymer insulation on wiring.
In mixed metal conductors and plating the whiskers potentially can lead to metal oxidation and electrolytic effects that theoretically can lead to conductor corrosion over a long period.
The whisker problem may be especially acute in aerospace applications because of two developments: electrical component density on aircraft is increasing, and more closely spaced circuit elements mean the whiskers have less space to bridge and cause electrical arcing. Moreover, the use of lead in the solder, a traditional mitigation against whisker growth, has been banned.
Cathode ray tube (CRT) instrument displays and flat plasma screens fail with disconcerting frequency in airline service--a problem manifested with the benefit of lead soldering and plating. With the lead out, a whole screen of instrumentation may disappear at some critical stage in the not too distant future.
A March update by the National Electronics Manufacturing Initiative Inc. (NEMI) Tin Whisker User Group provides a succinct statement of the problem: "Pure tin electroplating has a long history of whisker formation and growth that has resulted in reliability problems for various types of electronic equipment. The predominant whisker mitigation strategy for more than 50 years has been the addition of lead to the tin plating. Legislation that will eliminate the use of lead in electronic products sold in the European Union (due to be implemented on July 1, 2006) has led many electronic component suppliers to propose the removal of lead from tin-lead plating, leaving essentially pure tin. (See Editor's Note, page 6.) This approach is the most convenient and least costly [short term] lead-elimination strategy... [however] the pure tin strategy presents reliability risks due to the whisker formation tendencies of pure tin and alloy plating."
A May 12 paper by the same group contains two sobering observations: "At this writing, the exact mechanism for whisker growth is unknown." And "at this time there is no method known to guarantee the elimination of tin whiskers over the life of a product."
Despite these uncertainties, some countermeasures against whisker growth are available, according to a 2002 paper in the Semiconductor Equipment and Materials International Journal:
Use thicker layers in the connector plating, which absorb more stress on the metal and retard whisker growth. Whiskers form because of compressive stress generated by irregular intermetallic formation. Stress can build in the tin layer, and whiskers can extrude through ruptures along grain boundaries in the metal.
Apply a post-bake after plating. Heating to 150 degrees C (300 degrees F) reduces stress through more uniform intermetallic diffusion. However, PCBs cannot be baked without causing damage.
Use an underlayer of nickel or gold to provide a more uniform intermetallic bond and hence less stress.
However, these countermeasures are only partially effective. For instance, thicker layers may retard but not eliminate whisker growth. Eventually, whiskers can still grow.
Dave Douthit, manager of LoCan, an electronics reliability consulting firm in Mesa, Ariz., says baking, too, has limitations: "Baking components has shown promise in lowering but not removing or stopping whiskers on certain types and styles of finishes and certain types of components."
Confounding the potential hazard is the fact that there is no accelerated test available for whisker growth. Accelerated testing of wiring, with considerable capability to predict the onset of insulation crack growth, has been developed. But for the connections, the technology has not advanced to the same level.
Douthit fears the whisker problem "is becoming a runaway freight train." Others are not as alarmed, saying whisker growth is not "a barn burner," but it definitely is an issue of electrical system degradation. Proliferating whiskers must be taken into consideration as an inspection and maintenance issue, as an aircraft ages in service.
Ordinary tin is a silver-white metal. It is malleable, somewhat ductile, and has a highly crystalline structure. During warming, gray, or alpha tin, with a cubic structure, changes into white, or beta tin--the ordinary form of the metal--which has a tetragonal structure. When tin is cooled below 13.2 degrees C (55 degrees F), it changes from white to gray, "morphing" back to the cubic structure. This change is affected by impurities and is inhibited by lead.
Let me hazard an hypothesis: whiskering may reflect a "quicksilvering effect" of magnetism (i.e., an electromagnetic field) on tin. After all, tin sits on the periodic table not far from elements like mercury, which have this property of fluidity. Tin's fluidity manifests as dimorphous crystal growth. Dimorphous, as in the classical definition of crystallizing in two forms fundamentally different while having the same chemical composition.
Whiskers may grow along the force lines of the magnetic field, induced by current flow when a printed circuit board is powered up--the growth occurring as the tin's temperature changes and its crystals "morph." Perhaps tin crystals without their soothing lead balm are as schizophrenic as Jekyll without his tempering Hyde. Remove the current (and the field), and the tin is impassive. Store it in the Earth's magnetic field with the same orientation for a long enough period of time, and researchers have found a slower rate of whisker growth than occurs when a PCB is generating its own strong local electromagnetic field.
Those more familiar with magnetism may say, "But tin isn't ferromagnetic." However, there are other types of magnetic properties, such as used in magnetic resonance imaging (MRI), which exploit the diamagnetic (i.e., magnetically repelled) properties of the body's carbon content. Many ceramics have magnetic properties. Could an electromagnetic field be the catalyst for the crystal growth of tin whiskers?
A few techniques that might confirm this notion:
Slowly rotate a sampling of PCBs -- changing their aspect in relation to the Earth's magnetic field -- and compare the results to non-rotated samples.
Utilize the electromagnetic shielding of a Faraday cage. Store half of one batch and eventually compare their whiskering with the unshielded batch lot.
Use twin components in PCBs such that switching power between them periodically will either reverse or cut across (i.e., transverse) the electromagnetic field associated with that PCB. Compare the effects with a similar "control" PCB without this field-switching feature.
Experiment with inducing and reinforcing whisker growths by strongly augmenting directional electromagnetic fields and seeing whether they cause accelerated whisker growth of a revealing orientation on a test item.
Monitor growth rates via time-elapsed scanning electron microscope (SEM) photography to look for differences in growth rates when the electromagnetic field is collapsed and when it is reversed.
If an object lesson exists here, it is yet another by-product of the law of unintended consequences. With the prohibition of lead in the plating, whisker woes increase. Has the tin-plating of the lead prohibitionists created a new failure mode? An Eliot Ness of avionics -- recalling the famed G-man of the Prohibition era -- may be needed to sling some lead back into the plating.
(NASA maintains a Web site devoted exclusively to the tin whisker problem. See http://nepp.nasa.gov/whisker/index.html. For a sobering discussion of the arcing potential of whiskers, see http://nepp.nasa.gov/whisker/reference/tech_papers/brusse2002-slides-tin-whiskers-attributes-mitigation-CARTS-US.pdf.)
David Evans can be reached by e-mail at [email protected].