Hi Gunter,
I think that the problem looks like brittle fracture through the interface which forms between the IMC layer and the electroless Ni surface. I notice that the cracks are partial and located within the BGA joint (don't extend to the outer surfaces). We've seen several similar failures, and a good example which comes to mind was found to originate during cool-down from reflow.
The failure was characterised by formation of fractures between bottom of IMC layer and top of electroless Ni. There is always some discussion as to whether the fault was caused by excessive application of force or poor joint metallurgy thanks to (for example) non-ideal PCB plated condition. In this example, we went on to find subtle cracking in the laminate around some of the pads. Note that these joints were examined after soldering but before test or subsequent product build. As a result, we were able to conclude that cracking was induced during the cool-down phase following soldering. Perhaps surprising, but this generated enough force after joint formation (IMC layer growth and joint solidification) to partially crack the joints and produce some minor cracking in the laminate. One could make a case for this being described as subtle pad cratering. I have sent Steve a couple of SEM images with a request to post on his site. I have also posted these on our site (http://www.themcsgroup.co.uk/Gallery/ and look in "Electronics and Telecoms" section for or use these links...
http://www.themcsgroup.co.uk/images/gallery/01639_pb-free_bga_internal_crack_m10k.jpg
http://www.themcsgroup.co.uk/images/gallery/01639_pb-free_bga_internal_crack_main.jpg
) The similarities with you're situation are strong and I hope you'll find the comparison interesting.
We've looked at lot of broken BGA joints on ENIG and often find that the fracture occurs through a thin, discontinuous layer which, when analysed, suggests a compound composition of Sn-Ni-P.
As to whether the problem is "black-pad" or not, I think this comes down to a debate about definitions. We tend to refer to "black-pad" when we find evidence of the Ni-depletion layer in the non-soldered PCB condition (either unused PCB or unpopulated pad) or when we see corrosion "spikes" in and around the failed location.
Where we find a thin, continuous Ni-depletion zone after soldering (but no corrosion spikes) and joint failure between IMC layer and Ni-depletion layer we refer to this as brittle interface fracture.
When soldering to an ENIG pad, even in the absence of any corrosion of the electroless Ni, after soldering a Ni-depleted (or locally P-rich) layer forms immediately underneath the IMC layer. Does this necessarily mean that the PCB has the "black-pad" condition after production - no. Can this Ni-depleted structure suffer from brittle fracture under load whether black-pad is present or not - yes.
In every case we have analysed there tend to be subtle variations in the extent to which the unused PCB has the Ni-depletion (black-pad) condition and there are usually differences in the handling (and thus loading) of the BGA joints also. Thus, we see brittle interface fractures with no black-pad condition but "excessive" force, and black-pad fractures when see PCB plating corrosion and failed joints which have not experienced inappropriate loading.
For what its worth, I think that the situation which you describe is most likely a result of brittle fracture as a result of load applied to the BGA. Suggest that you take a slightly closer look at the interface with BSE and EDS analyses. I would also look out for any subtle evidence of damage in the laminate under the pads. Finally, consider 100% micro-section analysis through the affected devices (every joint) so that you can thoroughly "map" the location and extent of the damage across the device.
Kind regards,
Stewart
Stewart McCracken
MCS Ltd.
Centre House
Midlothian Innovation Centre
Roslin
Midlothian
EH25 9RE
m. +44(0)7711 541735
t. +44(0)131 440 9090
f. +44(0)131 440 9085
e. [log in to unmask]
w. www.themcsgroup.co.uk
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Grossmann, Guenter
Sent: 27 October 2010 08:37
To: [log in to unmask]
Subject: [TN] Cracks in BGA solderjoints
Hi Technetters
I have an effect in BGA solder joints that I have never seen before.
In microsections and in FIB preparations I see cracks between the IMC and a slightly darker region towards the Ni which I assume to be a P enriched zone of the Ni. However, the cracks are somewhere in the width of the solder joint and they are quite short running out to a row of pores that look somehow like Kirkendall voids.
Is there still a possibility to place one or two pictures? Kim's IPC site doesn't work.
Best regards
Günter
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts 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 847-615-7100 ext.2815
-----------------------------------------------------
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please contact helpdesk at x2960 or [log in to unmask]
______________________________________________________________________
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts 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 847-615-7100 ext.2815
-----------------------------------------------------
|
