Ingemar
Your improvement by sputtering Au onto your pads indicates your surface was
dirty.  Presumably, there was a plasma clean before your deposition but, if
not, you cleaned it somehow and its not important to discussion.  You then
had a clean, oxide-free (again, Au doesn't form an oxide) surface that
allowed full contact with the wire.  This supports wire bonding is not deep,
but along the interface.

Auger analysis (a process that shoots in electrons and looks at core
electrons in atoms) would NOT have helped you because you always expect
impurities on the surface, and Auger cannot distinguish chemistry (for the
most part).

It does sound like your group needs the regular input of a thin film
metallurgist/materials scientist.  I suggest you connect up with someone in
the ericsson company arena (I'm thinking on the semiconductor side since PWB
shops are not there yet) on a regular basis.  Then one can design a series
of experiments to address issues proactively (it never works in a panic,
reactive mode).  Unfortunately, board shops do not have the mentality to do
this, and the cost is high.  But soon, it will be cost-effective as board
prices climb higher as does volume/shop.

Good luck

Carey

-----Original Message-----
From: Ingemar Hernefjord (EMW) <[log in to unmask]>
To: 'Carey Pico' <[log in to unmask]>; TechNet E-Mail Forum. <[log in to unmask]>;
Nicholas LAI <[log in to unmask]>; Christian Skeppstedt (EMW)
<[log in to unmask]>; Ingemar Hernefjord (EMW)
<[log in to unmask]>
Date: Tuesday, September 28, 1999 12:12 AM
Subject: RE: [TN] COB,FP,TAB/Wire bond/plasma clean


>
>Try John DiLazzaro at <[log in to unmask]> or phone (503) 526-5263. He
must know a lot in that region. Advanced MCM maker. Thanks for wire bond
response. Superthin "lubricants" are nasty things, yes. I made following
experiment on a hopeless non-bondable gold : sputtered approx. 50nm Au on
it, and found that it had changed to 100% bondability! Now, exactly what
happened, I don't know. The molecular bombardment may just have removed such
a thin film that you talk about, or covered it with Au. Agree, Carey, I have
been starting up such investigations a couple of times, and each time I find
that it ends in such scientific a panorama that time runs out and costs
become unproportional. Best way is to talk with the supplier, or even change
supplier if you have the choice. I have FlipChip questions in this matter
too, but I will save it so you can get your breath back.
>Take care(y)/Ingemar
>
>
>3) Wire bonding: A rough surface (e.g., spiked) will just "moosh" (for lack
>of a better term) all around.  Surface cleanliness depends on how you clean
>your surface and the residue it leaves behind.  Water and oxides will
always
>be present (except Au oxide is unstable at room temperature).  Thin films
>and interfaces can be extremely strong.  The strength of thin films can be
>1000X higher than bulk equivalents and their other properties (e.g.,
>electrical) vary as well with thickness.  Auger (pronounced o-jay (a french
>"je") for you non-surface scientists) is useful because it can easily
>explore the composition of the top 500-3000 angstroms (it's accuracy is
>about 50 A).  But you pay $150/hr (one way or another in a big company).
>But that's only for gross contamination during film deposition (plating or
>sputtering).  Also, AES (another way to say Auger) is not useful on rough
>surfaces because you can't tell it's depth from the surface.  About solving
>this for you, I don't think I can from here.  Yes, I'm a thin film
>metallurgist (presently looking for some work in Portland in this field if
>any of you have connections), but the problem you have is very
>equipment/metal-system dependent.  It's like long distance surgury.  I am
>available as a consultant (but I get expensive), but first I recommend
>calling your suppliers and trying to network through them to find an
>industrial partner (nearby) that you can visit and get ideas.  Serious
>problems are not solved in a day or even a week (which many of my clients
>loath to hear).
>
>Carey
>
>
>

##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe:   SUBSCRIBE TECHNET <your full name>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Gayatri Sardeshpande at [log in to unmask] or
847-509-9700 ext.5365
##############################################################