The third factor, the solder used to join the component termination to the
PCB would also have to be taken into account.

Some empirical studies of the various combinations would have to be
undertaken but from the theoretical point of view you can make some predictions based
on analysis of the likely mixtures of the three parts.

The situation is complicated by the question of whether there is perfect
mixing within the joint, i.e. do the component finish and the PCB finish dissolve
completely and mix uniformly in the solder in the fillet during the soldering
process or do they remain partly undissolved at the interfaces.

And the situation could be different for reflow and wave soldering.   In the
former everything is confined to the little droplet of solder that forms the
fillet and you can calculate what the final composition will be.   In the
latter the finishes could be largely flushed off the surfaces as they pass through
the wave but elements dissolved from the component and PCB finish can build up
in the solder bath effectively changing its composition.  The estimation of
the net composition of the solder in the fillet is complicated.

The final factor is how the fillet solidifies since that determines whether
segregation of dissolved elements can occur.   The way the solder in the fillet
solidifies depends on cooling rate and on whether the alloy is a eutectic.
If the cooling rate is fast enough then there is not much time for segregation
so the resulting fillet should be fairly homogeneous.   If the cooling rate
is slow, perhaps because the component or a connected ground plane has a high
thermal mass then there could be segregation of dissolved elements to the last
point to solidify, which might be the solder/substrate interface.   If the
alloy is non-eutectic (e.g. the SAC305) then segregation is more likely to occur
even at moderate cooling rates.

Having said all that the final result will then depend on what ingredients
there in the mix.    Any mainly tin alloy finishes are unlikely to cause
problems since they will just slightly alter the average tin content of the final
solder fillet.    If the component has a lead-containing finish experience
indicates that there can be problems with segregation of lead to the interface.
The gold on an ENIG finish is thin enough that it dissolves off quickly and is
unlikely to build up to levels where it could form intermetallic in quantities
that might affect reliability.  Similarly for the silver from an immersion
silver finish.     The use of the SnCuNi solder rather than one of the SAC alloys
should not introduce any further complications.   The small amount of Ni is
selectively incorporated in the Cu6Sn5 intermetallic. And Ni dissolved from an
ENIG finish would not alter the situation significantly.   For the reasons
discussed recently in this forum the use of  bismuth-containing solder or
component finish would  be a problem only if there is also lead in the system.

For wave soldering it is important to check the composition regularly to
ensure that elements dissolved from PCB and component termination finishes have
not built up to levels where they could cause problems.

Keith Sweatman
Nihon Superior Co., Ltd

-------------------------------------------------------------------------------Leadfee Mail List provided as a service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Leadfree
To temporarily stop delivery of Leadree for vacation breaks send: SET Leadfree NOMAIL
Search previous postings at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
-------------------------------------------------------------------------------