ICT Evening Seminar - "Surface Finishing".
9th September 2004
The Institute of Circuit Technology held an evening symposium on the subject of Surface Finishing at Arundel on 9th September 2004.
ICT Chairman Andrew Hall introduced papers by Dr Martin Goosey of Rohm and Haas Electronic Materials Europe Ltd, and Thomas Michels of TMT Trading GmbH.
Martin Goosey gave an overview of PCB Solderable Finishes for Lead-Free Assembly, with reference to RoHS legislation, and described the characteristics, advantages and limitations of available metallic and organic finishes. Finishes would need both to be lead-free and to be compatible with lead-free solders, of which there were many variants. It appeared that two alloys, Sn,4Ag,0.5Cu (commonly termed SAC), melting in the range 215-220ºC and Sn,0.7Cu, with a eutectic melting point of 227ºC, would become the preferred lead-free solders in Europe.
Both alloys had melting points significantly higher than the 183ºC of eutectic tin-lead.
Attributes required of a suitable solderable finish included: providing good solderability for reflow, wave and hand soldering; surviving multiple soldering operations (up to 7 in extreme cases); performance at least equal to tin-lead HASL; surviving higher temperatures; compatibility with a wide range of solder alloys, as well as handling, surface insulation resistance and environmental and health and safety considerations. In addition, certain applications demanded such properties as wire-bondability.
Martin listed and compared their properties of lead-free HASL, OSP, immersion tin, electroless nickel immersion gold, electroless palladium and immersion. Most of these had been available for several years in response to the demand from surface-mount assemblers for planar finishes.
Comparing market share of the different finishes present and projected, he commented that although there were considerable differences in statistics from different sources, the trend was towards a substantial decrease in the use of HASL, even though lead-free HASL would maintain a small share, and slight growth in the use of electroless nickel-immersion gold and OSP. The major growth areas were predicted to be in immersion silver and, particularly, in immersion tin.
Although certain aspects of long-term reliability were still being investigated, and there were some questions about the influence of the conductivity of finishes on high frequency signal transmission, the choice of finish would be governed by considerations of functionality, ease-of-use and cost. Immersion tin, immersion silver and OSP were expected to be predominant, although Martin stressed that there was unlikely ever to be a single universal solution.
Thomas Michels took up the specific question of whether immersion tin would become a suitable replacement for HASL and focused on a proprietary white tin from Azzintec in Switzerland, which several multinational OEM and CEM assemblers had adopted as a preferred lead-free finish.
There had been much debate on the subject of tin whiskers, and Thomas presented the results of comprehensive investigations of the phenomenon. Many factors contributed to the kinetics of whisker growth, particularly those which caused internal stress within the deposit. It had also been observed that the condition of the underlying copper surface was a major contributor – tin deposited on a matt copper surface giving the least and on a bright surface the most whiskering.
Immersion tin formulations had been developed with deliberately-introduced metallic impurities such as copper, which virtually eliminated whiskering, but it seemed that the major users preferred the attributes of the pure tin finish, which offered significantly improved long-term solderability, and did not consider minor whiskering to constitute a significant area of concern. And in any case, if tin applied as a solderable finish showed evidence of whiskering before soldering, any whiskers present had been observed to disappear during the soldering process.
A 1micron immersion pure tin deposit had been demonstrated to exhibit 12 months shelf life and maintain solderability after three reflow cycles, whilst yielding ionic contamination figures well within military limits of acceptability. The mechanism of accelerated ageing had been shown to be the progressive diffusion of base copper into the tin coating, to form various copper-tin intermetallics.
Proprietary immersion tin now available gave much finer-grain deposit than traditional processes and, being based sulphuric acid rather than sulphonic acid, there were no solder mask compatibility issues. They had better throwing power and were more easily rinsed. Application involved four stages: clean, micro-etch, pre-dip and immersion tin plate, but control was very straightforward, the process was regenerable and plating time for 1 micron was only 9 minutes.
Thomas referred to an ITRI report which had concluded that metallic coatings were in general superior to organic coatings, and that of the metallic coatings, tin gave the best all-round performance. The only function for which tin was not appropriate was wire bonding.
Thomas concurred with Martin Goosey in accepting that no one finish would provide the universal solution, but believed that immersion tin would emerge as the solderable finish of choice.
In his closing remarks, Andrew Hall gratefully acknowledged Artetch Circuits for supporting the event.
Pete Starkey
ICT Council