Reading time ( words)
By Tom Baggio and Dr. Kenichiro Suetsugu
Figure 1. Portable minidisk and PCB manufactured with a Sn/Ag/Bi solder paste.
With legislative directives banning lead in electronics processes afloat in Japan and Europe, it is important to understand how converting to lead-free solders will affect the SMT process as a whole.Is the world going to convert to a lead-free process for electronics manufacturing? The answer is yes. It is now just a question of when. Five years ago, the industry sentiment was maybe, but with the legislation directives in Japan and the European Union (EU), the target dates are now set. Japan's target is 2001 and the EU's proposed target is 2004. Even if these dates are pushed back, the reality is electronic manufacturing will be using a lead-free process before the end of the first decade of the 21st Century.
In October 1998, Matsushita (parent of Panasonic) introduced a minidisk player1,2 made using a tin/silver/bismuth (Sn/Ag/Bi) alloy at volumes of more than 40,000 units per month. With more than 500,000 of the lead-free minidisk players produced as of June 1999, the company is on its way to reaching this goal (Figure 1).
Figure 2. Matsushita's lead-free organization chart.
Historically, mankind has had significant experience with tin/lead (Sn/Pb) alloys. Only recently has the electronics industry begun to seriously consider an alternative. Finding a suitable replacement in three short years is not a trivial task. Matsushita has devoted a tremendous amount of resources to this effort, evaluating approximately 50 lead-free alloys over the past four years (Figure 2).
System ApproachUnderstanding lead-free technology and its impact on the entire system of processes is paramount. A total solution approach should be adopted for an effective and well-considered implementation. As a starting point, an alloy's properties must be identified, measured and understood, and a basic understanding of how these alloys interact with the finishes used on the printed circuit boards (PCB) and electrodes should be achieved. Switching to a lead-free process will inherently require more complex consideration and implementation than a simple drop-in replacement. Because of its influence on overall processes, it will have an impact on design, layout, mechanical and electrical properties, as well as product reliability. There will be tradeoffs in benefits, costs and applications among the alloys. To fully employ this technology, a manufacturer will need a thorough understanding of how being lead-free will impact the entire system of processes (Figure 3).
Figure 3. Various mechanical strength testing with several electrode finishes. Note that in some cases, the lead-free alloy is superior to the standard Sn/Pb.
Higher Melting Point of Lead-free Alloys"Escaping goblins to be caught by wolves,"3 or, as the proverb is commonly known, "out of the frying pan and into the fire," best describes the problems with the higher melting points of most lead-free alloys in the running. As the industry moves away from Sn/Pb alloys, an environmental issue may be solved, but a host of other considerations crop up. Most lead-free alloys being considered have a melting temperature between 200° to 230°C. This elevated temperature is the main concern for any lead-free process because it affects the components, reflow profile, PCB materials and operational costs (Figure 4).
Figure 4. Possible lead replacement alloys.4-6
The elevated liquidus temperatures narrow the process window. With lead-free alloys, there are concerns regarding component heat resilience, PCB laminate, temperature profile and flux medium makeup that require consideration.
ReliabilityThe finished product's reliability must be greater than or equal to the conventional Sn/Pb products. The properties of whatever alloy is used must be suitable for the application. With the many choices of alloy materials, electrode finishes and PCB surface finishes, the level of testing is now multiplied from conventional Sn/Pb solder. For each alloy selected, tests need to be performed for various component finishes, not just in reference to mechanical strength, but vibration analysis, thermal shock and variances in elemental analysis (Figure 5).
Figure 5. Failure mode analysis of 1608 resistor.
Elemental analysis needs to be revisited. Because most research has been focused on the Cu/Sn intermetallic compounds, the depth of research on lead-free alloys is sparse. Research focused on how the base metals form the alloy matrix, interact and inter-react with wetted elements, and the diffusion rates of these elements, warrants additional consideration. New techniques need to be developed for this analysis (Figure 6).
Figure 6. Bismuth separation. Bismuth tends to concentrate and drop out of the alloy matrix as its concentration is increased. Why bismuth tends to concentrate near the PCB land and electrode leads is not yet fully understood.
Component Temperature ResistanceA major concern with higher reflow temperatures is component reliability. Will current SMT components handle these higher processing temperatures and will higher temperatures during assembly affect their long-term reliability? As an example, Al caps typically are rated to 230°C; exposing them to 240°C or higher would cause an immediate catastrophic failure. Other components such as chips are capable of withstanding temperatures in excess of 250°C. Development efforts have been focused on reducing the melting point of the alloys and higher temperature resistance for components.
Temperature ProfileAs reflow temperatures rise, the control of the maximum ΔT across the PCB is critical. A typical 20° to 30°C ΔT is not suitable for some lead-free reflow processes because of the elevated melting point. It is recommended that the maximum deviation be limited to 20°C and, if possible, to a 10°C ΔT. PCB size is also a governing factor in alloy selection and lead-free process implementation. It is fairly reasonable to achieve low-temperature variations (ΔT less than 15°C) with relatively small and fairly dense PCBs, but when trying to process larger PCB assemblies having various thermal masses unevenly distributed across the card, it is a more difficult task (Figure 7).
Figure 7. A comparison of a Sn/Ag/Bi alloy (approximately 210°C melting point) with a standard Sn/Pb eutectic alloy (183°C melting point). With this alloy, the process window is cut in half, leaving only a 20°C window for the lead-free alloy as compared to a 40°C or more processing window.
With larger PCBs, the reflow ΔTs can exceed 40°C. Oven modifications are being developed to control these larger ΔTs. Mask technology and combination infrared (IR) and convection heating sources are being explored. Modified profiles, where raising the preheat temperature is coupled with lowering reflow temperatures (liquidus), are useful in limiting overall ΔT. These changes force the alloys and fluxes to change. Faster temperature rise times (exceeding 4°C per second), along with elevated temperatures, can lead to excessive demands on the flux medium makeup.
InfrastructureIf the current assembly infrastructure requires replacement when the world goes lead-free, it will be costly for everyone. Therefore, a strong goal for the lead-free business is to make the switch as "drop-in" as possible, i.e., streamlining as many causal relationships in the process as possible.
The area of main concern for existing equipment is the reflow process. As product demands are different from smaller to larger PCBs, it is best to consider the system of lead-free soldering as a whole.
Separate Wave from ReflowThe lead-free alloy used in the reflow process will most likely not be the same alloy used in the wave bath. Issues such as an excessive dross rate, which can be three to five times greater than standard Sn/Pb baths, are wave exclusive. With lead-free wave baths, removing impurities with the current methods of sampling, analyzing, skimming/drossing and reanalyzing may be insufficient because impurities can build quickly.
As an example, 0.2 percent of lead impurity in a Sn/Ag/Bi wave bath will sig-nificantly reduce the pull strength of the soldered joint. If electrodes with a Sn/Pb finish are soldered, this impurity must be monitored closely in some lead-free baths. Therefore, a continuous method of analyzing the wave bath and purification methods may be used in future lead-free wavesoldering baths.
Alloy Benefits and TradeoffsSn/Ag/Cu alloys offer excellent mechanical reliability for the finished product but require a high melting temperature of approximately 217°C. A Sn/Bi electrode finish shows excellent mechanical properties with a Sn/Ag/Bi alloy but has slightly lower properties with other finishes. Lead softening, the process where lead separates from the Sn/Pb matrix and concentrates in lead-rich zones, is a common problem with conventional Sn/Pb alloys. This phenomenon is not observed in many lead-free alloys.
In the manufacturing process, most lead-free alloys have lower wetting properties that place more emphasis on the placement equipment's skew control. While the lower wetting condition causes a skew issue (reduction of self-centering), tombstoning is reduced.
CostHigher melting points will demand more energy, leading to higher wave and reflow energy costs. Energy conservation techniques employed in the newer generation of soldering equipment will help reduce this, but the energy conservation techniques could have easily been employed with the Sn/Pb based alloys, as well. Therefore, energy consumption will increase.
Depending on wave or reflow, material costs are different. With wavesolder bars, the material costs are typically 90 percent of the finished goods. Any increases in raw material costs will directly increase the cost of the bars. With solder paste, the actual alloying material costs are much lower (approximately 20 to 40 percent), so the impact of higher raw material costs will not be as significant.
Incidental costs, such as startup, retraining, equipment modifications, testing service or equipment, will also be present in the conversion. Testing costs associated with harder alloys that cause more wear on test probes is another consideration.
ConclusionLead-free has quickly become an available technology. It is apparent that the momentum has shifted from a wait-and-see attitude to a global proactive approach.
The North American electronics assembly market must move forward with this technology even though no direct legislation has been proposed in the United States.
Considering the limitations on imports that the EU and the Pacific Rim's directives will enforce, U.S. companies will have to deal with lead-free issues sooner rather than later.
REFERENCES1 M. Judd and Briendly, Soldering in Electronics Assembly Second Edition, Butterworth-Heinemann, Oxford, 1999.
2 "Electric Times," April 1999.
3 J. R. R. Tolkien, The Hobbit, Ballantine, 1966, p. 103.
4 B. P. Richards, et. al., "Lead-free Soldering: An Analysis of the Current Status of Lead-free Soldering," Department of Trade and Industry, 1999, p. 20.
5 H. H. Manko, Solders and Soldering: Materials, Design, Production and Analysis for Reliable Bonding, McGraw Hill, 1992.
6 K. Suetsugu, et. al., "Thermoset Pb-free Solder Using Heat Resistant Sn/Ag Paste," National Technical Report, Vol. 43, No. 1, 1997
TOM BAGGIO may be contacted at Panasert Engineering, Panasonic Factory Automation, 9377 W. Grand Ave., Franklin Park, IL 60131; (847) 288-4400. DR. KENICHIRO SUETSUGU, manager of process and material development, circuits manufacturing technology laboratory, may be contacted at Matsushita Electric Industrial Co. Ltd., 2-7, Matsuba-cho, Kadoma, Osaka, 571-8502, Japan; (06) 6901-1171; Fax: (06) 6905-1537.
Top 10 Reasons Not To Convert to Lead-free
- Can't impress people with knowledge of Pb symbol.
- No more heavy metal jokes.
- Concerned over losing "Lead" Zeppelin fan club membership.
- Three-eyed fish taste better.
- Still enjoy great-grandmother's pewter tea set.
- Lead, dioxin and asbestos are new favorite food groups.
- Have sunk family fortune in lead mining.
- Still waiting for the United States to convert to the Lead Standard.
- Still working on the arcane alchemy experiments in basement.
- Paint chips ... the other white meat.