TECHNET Archives

November 2016

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"George M. Wenger" <[log in to unmask]>
Reply To:
TechNet E-Mail Forum <[log in to unmask]>, [log in to unmask]
Date:
Thu, 24 Nov 2016 13:06:53 +0000
Content-Type:
text/plain
Parts/Attachments:
text/plain (183 lines)
Ioan, 

I know TN strips off attachments but I included Dave's, Richard's and your email address in the email so the three of you should have gotten the attachment. If you didn't let me know and I will resend it. 

George 

----- Original Message -----

From: "Ioan Tempea" <[log in to unmask]> 
To: [log in to unmask] 
Sent: Thursday, November 24, 2016 7:40:01 AM 
Subject: Re: [TN] Gold embrittlement 

Thanks everybody! 

I'm having a better idea now. 

To answer Joyce's questions about the screening that has been done, well, I cannot disclose all these details, sorry! 

George, 
Thanks for the rigorous answers! Can you, please, send me directly the paper, Technet strips all attachments. 

All the best and Happy Thanksgiving to all south of the border! 

Ioan Tempea, P. Eng. 
Manufacturing Engineer, Satellite Systems 

MacDonald, Dettwiler and Associates Corporation, 21025 Trans-Canada Highway, Ste-Anne-de-Bellevue, QC, Canada H9X 3R2 
Tel: +1-514-457-2150 x3556 
www.mdacorporation.com 

This e-mail, and any attachments, are intended solely for the use of the intended recipient(s) and may contain legally privileged, proprietary and/or confidential information. Any use, disclosure, dissemination, distribution or copying of this e-mail and any attachments for any purposes that have not been specifically authorized by the sender is strictly prohibited. If you are not the intended recipient, please immediately notify the sender by reply e-mail and permanently delete all copies and attachments. This email is for informational purposes only and shall not be interpreted to authorize or conclude a binding agreement between MDA and any other party unless this email contains or is accompanied by an express written confirmation of MDA's intention to enter into a binding agreement, such confirmation shall only be provided by an authorized representative of MDA. 


-----Message d'origine----- 
De : TechNet [mailto:[log in to unmask]] De la part de George M. Wenger 
Envoyé : Wednesday, November 23, 2016 8:08 PM 
À : [log in to unmask] 
Objet : Re: [TN] Gold embrittlement 

Hi Dave and Richard - well I usually agree with both of you but because I'm not really a chemist or material scientist I'm not really sure about gold migration, Gibb's free energy issue, or gold nucleation time-variable event. I guess I'm just too simplistic and while I may not know the correct terms I think I can say what I've seen during my 45 years of doing solder assembly failure analysis. I'm not sure if the TN responses have answered Ioan's questions. 

Hi Ioan - let me re-list you questions and give you my thoughts in RED 

I have SnPb37 joints with 4-8% Au that did not fail thermal cycling, no cracks, pass electrical tests like a charm. This kinda contradicts J-STD-001 and common knowledge, so I would appreciate some insider info regarding Au embrittlement, to help me figure this one out. 

I know one could write volumes on this, but I'm sure the questions below can be answered in a brief manner. 


* How did Au embrittlement end up being a reliability issue? Real catastrophic failures? There have been some gold embrittlement problems in the industry and because of the problems mechanical strength evaluations were done which showed that when then the gold concentration was 3-5% the mechanical strength of the solder joint degreased dramatically. The strength didn't go to zero but the strength was reduced by at least 50%. So specifications were written to avoid solder joints with 3-5% gold content. 

* What is the effect on the long term reliability of solder joints? The effect is that the solder joint strength may be reduced with thermal aging but as long as the strength of the solder joint is greater than the stress on the solder joint there isn't a long term reliability issue. 

* What screening method creates catastrophic failure of embrittled joints (mechanical shock, thermal cycling, vibration, etc.)? Mechanical stress after isothermal aging should let you know if you could have a catastropic failure due to gold embrittlement. 

* It looks like the embrittlement process goes on as long as there's Au to combine with Sn. What is the speed of this reaction? Otherwise said, how long does it take for a joint with high Au content to become brittle? Here is where I go back to my comment to Dave and Richard, I can't say it is migration, nucleation, or what the speed might be. 

* Has anybody screen for embrittlement? Or, anyone tried to maximise embrittlement (e.g. 100 hours @ 150 C) and then screened the hardware? Look at Figure 8a and 8c in the attached paper (Case Studies of Brittle Interface Failures in Area Array Solder Interconnections) that we presented at ISTFA sixteen years ago. We had a gold embrittlement failure during a push test on a solder interconnection that was isothermal aged at 150C for 451 hours. 

I think you also made the following statement: 

I must make the statement that we're not talking about any product here. We've simply tried to see if components with thick Au plating could be qualifiable and to my surprise nothing seems to tell me we would waste money starting a full blown qualification and I would like to know why. 
Sections have been done, that's how I know there's so much Au in the joints. Here is what I know. Solder joints made to nickel are not as strong as solder joints made to copper. Solder joints with 3-5% gold are not as strong as solder joints with 0-2% gold. Gold does embrittle solder joints but that doesn't mean you will have a gold embrittlement failure. If your use (cycling, thermal aging, and mechanical stress) exceeds the strength of the solder joint you probably will have a gold embrittlement failure. So you need to determine what your use conditions are and test for them. 

I hope the attached paper and above information helps. 

Regards and Happy Thanksgiving, 

George 

George M. Wenger 
Failure Signature & Characterization Lab LLC 
609 Cokesbury Road, High Bridge, NJ 08829 
(908) 638-8771 (Home) (732)-309-8964 (Cell) [log in to unmask] 

----- Original Message ----- 

From: "David Hillman" <[log in to unmask]> 
To: [log in to unmask] 
Sent: Wednesday, November 23, 2016 1:20:52 PM 
Subject: Re: [TN] Gold embrittlement 

Hi Richard - well, we usually agree but not this time. The formation of the gold/tin IMC happens as the solder joint is molten and being formed. Gold and tin can form the IMC phase in a solid state condition but it is very very very slow and only in very rare occasions has been found to be a failure root cause. Gold will migrate and form gold/tin IMC in the presence of nickel plating but that is a Gibb's free energy issue and again, rarely is a failure root cause (think of all the ENIG boards soldered in the industry). If you don't have a gold embrittlement problem right after soldering, you have a very high probability of not having a gold embrittlement problem. The reason for all of the JSTD-001 criteria is an attempt to get folks to manage this issue proactively and not reactively. 

Dave 

On Wed, Nov 23, 2016 at 12:03 PM, Stadem, Richard D. < [log in to unmask]> wrote: 

> Just keep in mind that gold nucleation is a time-variable event. It 
> begins immediately after a solder joint is formed, but continues for a 
> long time, typically for two or three years or until equilibrium is 
> reached. The embrittlement due to gold nucleation will seldom be 
> present immediately after soldering, but solder joints will typically 
> fail perhaps between 6 months to two years later. 
> That is why the J-STD-001 reads that you must show objective proof, 
> and that proof needs to consider the time variable as a factor to test. 
> 
> -----Original Message----- 
> From: TechNet [mailto:[log in to unmask]] On Behalf Of Vladimir 
> Sent: Wednesday, November 23, 2016 11:27 AM 
> To: [log in to unmask] 
> Subject: Re: [TN] Gold embrittlement 
> 
> Hi Ioan, 
> 
> Au and Sn form intermetallic platelets and cracking goes at their 
> interface with the matrix. 
> 
> ‎The fact that your product did pass the test doesn't mean there were 
> no cracking in solder and the product will last a as long as it 
> supposed to in field (it will depend on how good a reliability model was). 
> 
> I'd definitely suggest you to section ‎joints after the test. 
> 
> Regards, 
> 
> Vladimir 
> 
> SENTEC 
> 
> Sent from my BlackBerry 10 smartphone on the Rogers network. 
> Original Message 
> From: Ioan Tempea 
> Sent: Wednesday, November 23, 2016 12:02 
> To: [log in to unmask] 
> Reply To: TechNet E-Mail Forum 
> Subject: [TN] Gold embrittlement 
> 
> Dear Technos, 
> 
> I have SnPb37 joints with 4-8% Au that did not fail thermal cycling, 
> no cracks, pass electrical tests like a charm. This kinda contradicts 
> J-STD-001 and common knowledge, so I would appreciate some insider 
> info regarding Au embrittlement, to help me figure this one out. 
> 
> I know one could write volumes on this, but I'm sure the questions 
> below can be answered in a brief manner. 
> 
> 
> * How did Au embrittlement end up being a reliability issue? Real 
> catastrophic failures? 
> 
> * What is the effect on the long term reliability of solder joints? 
> 
> * What screening method creates catastrophic failure of embrittled 
> joints (mechanical shock, thermal cycling, vibration, etc.)? 
> 
> * It looks like the embrittlement process goes on as long as there's 
> Au to combine with Sn. What is the speed of this reaction? Otherwise 
> said, how long does it take for a joint with high Au content to become brittle? 
> 
> * Has anybody screen for embrittlement? Or, anyone tried to maximise 
> embrittlement (e.g. 100 hours @ 150 C) and then screened the hardware? 
> 
> * Anything else I should know? 
> 
> Thanks, 
> 
> Ioan Tempea, P. Eng. 
> Manufacturing Engineer, Satellite Systems 
> 
> [cid:[log in to unmask]] 
> 
> MacDonald, Dettwiler and Associates Corporation, 21025 Trans-Canada 
> Highway, Ste-Anne-de-Bellevue, QC, Canada H9X 3R2 
> Tel: +1-514-457-2150 x3556 
> www.mdacorporation.com<http://www.mdacorporation.com/> 
> 
> This e-mail, and any attachments, are intended solely for the use of 
> the intended recipient(s) and may contain legally privileged, 
> proprietary and/or confidential information. Any use, disclosure, 
> dissemination, distribution or copying of this e-mail and any 
> attachments for any purposes that have not been specifically 
> authorized by the sender is strictly prohibited. If you are not the 
> intended recipient, please immediately notify the sender by reply 
> e-mail and permanently delete all copies and attachments. This email 
> is for informational purposes only and shall not be interpreted to 
> authorize or conclude a binding agreement between MDA and any other 
> party unless this email contains or is accompanied by an express 
> written confirmation of MDA's intention to enter into a binding 
> agreement, such confirmation shall only be provided by an authorized representative of MDA. 
> 

ATOM RSS1 RSS2