Hello Waclaw,
Bild 2: If it is correct that this is a cross-section of an unsoldered pad
on a flexible circuit substrate from your supplier, showing the Immersion
Au-over electroless Ni-Phosphorous-over Cu, then this is a terrible Ni-P
surface. Well executed Ni-P produces a relatively smooth surface when
viewed in cross-section. The actual Ni-P surface is typically looks like a
mixture of tennis and golf balls pressed together, resulting in what appear
to be Ni-P grains that have rounded perimeters and few straight edges or
sharp corners. Typical "good" electroless Ni-P contains 8-10 weight % P,
and this deposit is quite amorphous. Lower values of P make the Ni-P more
susceptible to corrosion by a typical immersion Au bath. The topography of
the NiP surface generally shows a greater peak-to-valley height difference
when the Ni-P is thicker, but at the 3 micrometers you mention, the surface
should show only small peak to valley differences. The deep valleys between
"grains" that is visible in this optical micrograph is quite remarkable. It
is not a good Ni-P surface. It exhibits a roughness that is much more
extreme than other photomicrographs that I have seen in Ni-P surfaces
exhibiting Black Pad corrosion and solder joint separation. Since the
electroless Ni appears to be continuous at the interface with the Cu, this
surface roughness does not appear to be attributable to a poorly cleaned and
poorly activated Cu surface. So I can only conclude that your supplier used
excessively aggressive rinse-clean-rinse of the Ni before Au plating, or the
Au plating operation caused extremely severe corrosion at the "grain"
boundaries of the electroless Ni. Again, this roughness is so severe it is
uncharacteristic of even severe Black Pad. This is why I suggest that
perhaps your supplier used an excessively aggressive hydrochloric acid
"cleaning" prior to immersion Au. In any case, this roughness is very bad.
Part of the problem may be that the solder could not even properly wet down
into the deep valleys in the Ni surface due to viscosity and surface tension
effects, and possible poor Au coating at the bottom of the valleys. Or,
this is really severe Black Pad, and these valley regions are so rich in P
that the solder would not wet and form intermetallics. Black Pad generally
originates at the "grain" boundaries of the Ni-P and as the attack at these
boundaries gets more severe, corroding deeper valleys, the black pad
corrosion begins to move up from the valleys to the convex surface of the
Ni-P grains. If your situation is black pad, the fact that the valleys
appears so deep, almost guarantees that the entire convex surface of the
Ni-P grains will also be corroded and will exhibit an enriched P layer on
top of the normal Ni-P ( ~8% P, typically). This is the most severe case of
black pad, and separated surfaces will show very little Ni3Sn4 remaining on
the surface of the Ni. Any Ni3Sn4 formed will generally be seen in the
separated solder surface.
Bild 6 and Bild 8: Photo Bild 6 is somewhat unclear, but both cross-section
do not appear to show the rough Ni-P surface shown in Bild 2. When viewing
the left-side half of the Bild 6, it appears that I can see a Ni-P layer on
the Cu, and a VERY thick Ni3Sn4 layer between the Ni-P layer and the bulk
solder. ( I am obviously making assumptions about the chemistry of what I
am viewing based upon your input). I can say that the interface between the
Ni and Cu looks good, and the surface of what appears to be the Ni remaining
after reflow, also look relatively smooth without ANY evidence of the deep
valleys seen in the Ni surface in Bild 2. I do not understand this. I can
also comment that what appears to be the Ni3Sn4 layer is VERY thick. It
appears to be much thicker than what appears to be the Ni-P layer. This is
unexpected. A very thick intermetallic (IMC) layer is generally not good
and can lead to brittle failure. The VERY thick apparent Ni3Sn4 layer makes
me wonder about the reflow conditions and the chemistry of the solder. This
IMC can grow rapidly if Cu is present in the solder, and it has also been
reported that Ag in PbSn eutectic solder can also accelerate IMC growth. Is
this a PbFree solder that contains Ag and/or Cu? If the solder is eutectic
PbSn or near eutectic PbSn with 2% Ag, what was the reflow profile? Was the
peak temperature too high or the time above 183C too long? If I am
interpreting the cross-section correctly, even these conditions make it hard
to explain what appears to be an EXTREMELY thick IMC layer, since as the Ni
diffuses into the solder to form the Ni3Sn4, the P-enriched/Ni depleted
surface in contact with the molten solder typically acts as a barrier to
diffusion.
So, at the moment I cannot positively identify the problem. Is the sure
that the Ni is electroless Ni-P? Is it possible that the Ni is electrolytic
Ni? If so, this could explain what appears to be extreme Ni3Sn4 thickness,
IF your reflow profile was too hot, to long above 183C, or if the solder was
62Sn-36Pb-2Ag, or if the solder is a PbFree solder containing Ag or Cu. If
this is is true, then excessive IMC growth and brittle failure may be the
cause of the problem.
Best regards,
Leo
Leo M. Higgins III, Ph.D.
ASAT Inc.
Central Region Mgr. / Director of Sales
[log in to unmask]
3755 Capital of Texas Hwy-So
Suite 100
Austin, TX USA 78726
tel: 512-383-4593
fax: 512-383-1590
mobile: 512-423-2002
-----Original Message-----
From: karpinski [mailto:[log in to unmask]]
Sent: Thursday, February 19, 2004 10:33 AM
To: [log in to unmask]
Subject: Re: [TN] Ni/Au solderability on kapton
Hello ,
Thank for your help.
I send you some pictures showing X-section of the solder joints.
I am not expert on this topic. Please comment on them.
Fig. 2
X-section through non soldered pad with Au layer showing very rough Ni
Layer
Fig. 6
X-section of a unseperated solder joint
Fig.8
X-section (etched) of a solder joint showing a strange Ni-solder interface
Fig. 10
Pictures of broken pads and peeled SMD resistor
regards
Waclaw
Leo Higgins schrieb:
>This sounds like either typical black pad problem, or a passivated Ni
>surface under the very thin immersion gold layer. A determination of the
>failure interface as suggested by another respondent is necessary. A
>cross-section of a failed interface and an unseparated interface from the
>same sample set would help in making a diagnosis (SEM / EDX map and line
>scans for O, Ni, Sn, P). A scanning Auger map of the surface can help in
>seeing if the Au deposit is low density or porous, and if it has allowed
Ni
>to oxidize and prevent good wetting after the Au is dissolved in the
solder.
>
>
>Regards,
>Leo
>
>------------------------------------
>Leo M. Higgins III, Ph.D.
>ASAT Inc.
>Central Region Mgr. / Director of Sales
>[log in to unmask]
>3755 Capital of Texas Hwy-So
>Suite 100
>Austin, TX USA 78726
>tel: 512-383-4593
>fax: 512-383-1590
>mobile: 512-423-2002
>------------------------------------
>
>
>-----Original Message-----
>From: karpinski [mailto:[log in to unmask]]
>Sent: Wednesday, February 18, 2004 3:38 PM
>To: [log in to unmask]
>Subject: Re: [TN] Ni/Au solderability on kapton
>
>
>Hi Vladimir,
>
>As I know it is Pd.
>regards
>
>Vladimir Igoshev schrieb:
>
>
>
>>Hi Waclaw,
>>
>>Do you know what material (Pd or Ru) your board supplier uses to deposit
>>
>>
>electroless Ni?
>
>
>>Vladimir Igoshev,
>>Research in Motion
>>
>>Voice: (+1) 519-888-7465, ext. 5283
>>Fax: (+1) 519-886-0863
>>E-mail: [log in to unmask] <mailto:[log in to unmask]>
>>
>>
>>-----Original Message-----
>>From: karpinski [mailto:[log in to unmask]]
>>Sent: Wednesday, February 18, 2004 4:15 PM
>>To: [log in to unmask]
>>Subject: [TN] Ni/Au solderability on kapton
>>
>>
>>Recently we have solderability problems with Ni/Au kapton flex boards
>>(0,1 µm Au, 3 µm Ni, 35m Cu).
>>We have produced and assembled large number of boards.
>>On 10 % of the boards we observe interconnection failures after final
>>assembly of the pcbs. If the suspected solder joints are lightly
>>stressed , the connections are easily broken, leaving an open circuits.
>>It is easy to peel these parts with solder material from the boards.
>>The pads below the removed parts are discolored dark gray and are sleek .
>>Some times the handsoldering of the broken joins was successful but
>>after small strain the joints broke again.
>>Additional vapor phase soldering did not improved the the strength of
>>the joints.
>>This is new problem for us. Does anyone have any experience with this?
>>How to rework the boards? How to determine when a board should be
>>reworked or
>>scrapped? What about reliability issues? Latent failures?
>>I would appreciate your help very much
>> regards
>>
>>--
>>Waclaw Karpinski
>>I. Physikalisches Institut
>>RWTH Aachen
>>D - 52074 Aachen
>>
>>---------------------------------------------------
>>Technet Mail List provided as a service by IPC using LISTSERV 1.8e
>>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
>847-509-9700 ext.5315
>
>
>>-----------------------------------------------------
>>
>>---------------------------------------------------
>>Technet Mail List provided as a service by IPC using LISTSERV 1.8e
>>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
>847-509-9700 ext.5315
>
>
>>-----------------------------------------------------
>>
>>
>>
>>
>>
>>
>
>--
>Waclaw Karpinski
>I. Physikalisches Institut
>RWTH Aachen
>D - 52074 Aachen
>Sommerfeldstr 14, Turm 28
>Tel.: +241 802 7181
>Fax.: +241 802 2623
>Mobil: 0049 173 70 22 548
>
>---------------------------------------------------
>Technet Mail List provided as a service by IPC using LISTSERV 1.8e
>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
>847-509-9700 ext.5315
>-----------------------------------------------------
>
>---------------------------------------------------
>Technet Mail List provided as a service by IPC using LISTSERV 1.8e
>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
847-509-9700 ext.5315
>-----------------------------------------------------
>
>
>
>
--
Waclaw Karpinski
I. Physikalisches Institut
RWTH Aachen
D - 52074 Aachen
Sommerfeldstr 14, Turm 28
Tel.: +241 802 7181
Fax.: +241 802 2623
Mobil: 0049 173 70 22 548
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
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
847-509-9700 ext.5315
-----------------------------------------------------
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
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 847-509-9700 ext.5315
-----------------------------------------------------
|
