SMART Group – ICT Joint Symposium, 30 November 2000

Citation

(2001), "SMART Group – ICT Joint Symposium, 30 November 2000", Circuit World, Vol. 27 No. 3. https://doi.org/10.1108/cw.2001.21727cac.001

SMART Group – ICT Joint Symposium, 30 November 2000

SMART Group – ICT Joint Symposium, 30 November 2000

Solderable finishes – past, present and future

Keywords: ICT, SMART Group, Solderable finishes

ICT Chairman Steve Payne introduced an evening symposium on the theme of solderable finishes for PCBs and components, to an audience of local circuit fabricators and assemblers, at the Mill House Hotel in Arundel on 30 November 2000. The SMART Group Technical Committee was represented by Pete Starkey, who reviewed historical trends and factors that had influenced choice of PCB finishes to date. His paper served as an introduction to the main presentation, from Dr Martin Goosey representing the Institute of Circuit Technology, which examined environmental issues and their effect on final finish options for the future.

Pete Starkey described the solderable finish as the primary interface between the fabrication technology and the assembly technology and, asking the question "Who decides what solderable finish to use?", made the point that the board manufacturer is generally only a service provider, not making own product but simply offering a manufacturing service to the OEM's design and specification. Although the OEM might also be the assembler, the assembly operation was increasingly carried out by yet another service provider; the contract assembler. Contract assemblers were again organisations not making their own products but providing an assembly service to the OEM's instructions and specifications. If the board manufacturer wanted to take the initiative to introduce a new finish, it was its responsibility to sell it to the customer and to get it qualified and included in the customer's specification. In many instances it was the proprietary process supplier, using skilful marketing techniques, who was most influential in determining the choice of finish. In the ideal scenario, the OEM, the assembler, the board manufacturer and the process supplier would work together to define the right finish for the job. This situation did not often happen.

After referring briefly to ancient Roman and Egyptian history for the origins of soldering, and having taken a short excursion into the chemistry and metallurgy of solder joint formation and factors which affect the solderability of metal surfaces, Starkey showed statistics illustrating the relative popularity of various finishes over the last few years.

He described how the demise of fused electroplated tin-lead had been accelerated by the demand for "selective plating", particularly for the elimination of fusible metal under solder resist. This had led to the emergence and almost universal predominance of hot air solder levelling (HASL) until the trend to fine-pitch SMT initiated a demand for flat finishes. HASL remained the mainstream finish, providing a "pre-soldered" surface with copper-tin intermetallic already formed in-situ. It also provided the "self-inspecting" benefits of the thermal shock involved in the process. However, even though the practicability of lead-free HASL had been demonstrated, demand for HASL appeared to have peaked and there was increasing competition from various flat finishes.

As a flat and cosmetically attractive finish, electroless nickel-immersion gold had become fashionable in the early-mid 1990s, and had been useful when additional functions such as wire-bonding or key-pad contacts were required. Commercial opportunism on the part of supply-houses and vague specification on the part of OEMs had resulted in poorly-characterised, difficult-to-control processes getting the market off to a shaky start, although the situation had progressively improved. Remarking on the fact that the finish introduces a hard nickel layer between ductile conductor and ductile solder and that the integrity of the solder joint depended on the formation of a thin tin-nickel intermetallic, Starkey referred to reliability studies carried out by Hewlett Packard, Celestica, Hadco and others. These had concluded that failure mechanisms were complex and were believed to involve some form of corrosive attack on nickel by gold. However, there remained cautious confidence for the future.

Organic solderability protectives (OSPs) were comparable in popularity with electroless nickel-immersion gold. The processes were straightforward and cost-effective, but the cosmetic unattractiveness of OSPs had been a commercial barrier to their acceptance in "high-technology" applications, even though contemporary benzimidazole and alkyl imidazole products had been demonstrated to be very effective in multiple reflow sequences.

The catch-all classification "other finishes" included various whisker-resistant modified immersion tins and tin-organics, immersion silver-organics and palladium variants, which were taking a small but steadily growing market share.

Starkey concluded his presentation by listing available sources of information, then posed the question "Where do we go from here?" as a lead-in to the second paper.

Martin Goosey began by putting lead in a realistic perspective from the environmental point of view. Referring again to ancient history, he explained how certain schools of thought blamed lead poisoning for the fall of the Roman Empire, and that more recently lead from petrol additives had been associated with various undesirable effects on human health. Current concerns related to the possibility of lead leaching into water from landfill, and had focused attention on potential contamination from scrap electronic assemblies. Although lead in electronics represented less than 2 per cent of the total use of lead and the electronics industry had previously been considered too important to hinder by legal restriction, various legislation was appearing. For example, the draft "WEEE" directive and the Integrated Pollution Prevention and Control directive, had initiated detailed and thorough re-appraisals by the electronics industry.

Whatever legislative obligations might be imposed upon the industry, general public concern about perceived environmental issues had created circumstances in which companies also saw significant commercial potential in becoming lead-free, and the Japanese in particular had taken the initiative to promote "the added value of environmental technologies" in establishing lead-free electronic production.

Having established the inevitability that lead would eventually be replaced in electronics, whether for reasons of legal compliance or commercial opportunism, Dr Goosey went on to review the alternatives to lead alloys in soldering. It appeared that the tin-silver-copper systems, with silver in the range 3-4 per cent, copper in the range 0.5-0.7 per cent and melting points around 217°C were becoming the generally preferred choice for many mainstream applications.

Dr Goosey defined the principal requirements of a PCB finish to suit lead-free processing. It should perform at least as well as tin-lead HASL, be able to survive higher soldering temperatures, be compatible with a wide range of solders, and obviously contain no lead. For surface mount applications, it should have good planarity and be able to withstand multiple solder cycles. Other properties, such as wire-bondability, might also be required.

Many of the non-HASL finishes earlier described by Starkey were suitable for lead-free applications and were already well characterised. The trend to lead-free had added impetus to their adoption. Dr Goosey listed and compared the various merits of lead-free HASL, OSPs, immersion tins, nickel-golds, palladium and immersion silvers. He commented that immersion silver was fast gaining industry acceptance as a relatively low-cost high-reliability finish. It had been qualified by major multinationals such as Intel, Lucent, Motorola and Raytheon and possibly represented the closest drop-in replacement for conventional HASL.

The issue of lead-free finishes for components was of more serious concern; there was still a distinct shortage of "lead-free" components although the situation was gradually improving. Lead-free component finishes included tin (electroplated or dipped), tin-copper, tin-bismuth, tin-silver, nickel-palladium, nickel-palladium-gold and silver-palladium.

Electroplated tin offered excellent solderability and corrosion resistance and was widely used on passive components, although applications were limited on fine pitch devices because of the tendency of tin to form whiskers. Work was continuing to develop whisker-free processes and tin-copper was a potential option. Tin-bismuth required a totally lead-free system, and tin-silver was relatively expensive and involved a slow plating process. Nickel-palladium had been used for some years on fine pitch devices but needed increased soldering temperature to enable good wetting. Nickel-palladium-gold gave improved wettability at the expense of higher material cost, and silver-palladium was not really suitable for lead-free applications because of potential crack formation on thermal cycling.

Summarising his conclusions, Dr Goosey made it clear that lead-free electronics is already happening and that lead would eventually be prohibited in electronics assembly. Planar finishes had been in use for some time and the trend to fine pitch and higher density would force more use of planar finishes. Suitable PCB and component finishes were available for lead-free assembly although more lead-free components were required. The need for wire bondability would provide a market for thicker finishes such as gold and palladium, although the principal PCB finishes were likely to be based on tin and silver. Functionality, ease of use and cost would be key factors, assuming that reliability requirements were met. There was confidence that, with care, lead-free assembly could be successfully implemented, although there was unlikely to be a universal solution.