Data Loading...
Whisker-Inpenetrable Metal Cap Process for Electronic ... Flipbook PDF
Title: Microsoft PowerPoint - TinWhiskerInpenetrable-Web Author: Irwin Created Date: 11/26/2010 6:09:08 PM
110 Views
83 Downloads
FLIP PDF 385.21KB
Whisker-Inpenetrable Metal Cap Process for Electronic Assemblies Bob Landman, Gordon Davy, Dennis Fritz LDF Coatings, LLC
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Abstract There are many negatives due to the Pb-free RoHS legislation. There are, still today, no good options available to high-rel OEMs for what are now Pbfree COTS components and assemblies. Component-by-component “mitigation” of tin whisker growth risk is very labor-intensive and, at best, it worked only when the majority of components were not Pb-free. The ideal universal process should be much less expensive and should be applicable to an entire assembly, not just “mitigating” the risk of whisker growth but preventing whiskers from growing, with a permanently impenetrable coating. A process is described that deposits a nickel (or other whisker-impenetrable metal) cap selectively over all Pb-free tin and high-tin alloys, but will not deposit nickel on any insulating surfaces (component bodies, solder mask, etc..) of an assembly.
Copyright 2010 LDF Coatings, LLC All Rights Reserved
What Is The Problem? •
Pb-free tin on electronic assemblies can grow whiskers that can cause circuit failure.
•
Because high-reliability assemblies have long design life, whisker problem is of particular concern to designers.
•
Regulations and economics have resulted in many otherwise desirable COTS electronic components and assemblies being available only with
– Pb-free tin as the termination finish OR – Pb-free SAC solder balls under BGAs •
High-rel customers lack clout – don’t buy enough.
– Often impractical to procure desired SnPb termination finish or BGA balls.
Copyright 2010 LDF Coatings, LLC All Rights Reserved
What is the Solution? A paradigm shift is needed Is mitigation prevention ? • Component-by-Component mitigation is unreliable and difficult to manage • LDF Coatings solution protects the entire assembly at once Copyright 2010 LDF Coatings, LLC All Rights Reserved
Tin Whisker Mitigation • COTS Components – 85% have a tin termination finish. – If OEM uses SnPb solder, whisker risk is primarily for unsoldered termination areas
• COTS Assemblies – BOM cannot be “scrubbed,” hence no mitigation. – If “dual-use,” has Pb-free (high-Sn) solder.
• “Mitigate” means “reduce” not “prevent.” • If prevention is possible, why mitigate? Copyright 2010 LDF Coatings, LLC All Rights Reserved
Mitigating Whisker Risk Component-By-Component If Pb-free Sn is not allowed
OEMS must use:
• No COTS assemblies • Expert “scrubbing” of OEM-BOM to catch non-conforming components • 100% receiving inspection to confirm that components match spec
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Mitigating Whisker Risk Component-By-Component Pb-free components must be subjected to a process that modifies termination finish to include Pb.
– Chip components: AEM process • Single source • High setup costs • Logisitics and delays
– Larger components: solder dipping • • • •
High setup costs Many components can probably withstand manual dipping Risk of undetected damage from heat or handling Logistics and delays
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Mitigating Whisker Risk Component-By-Component If contract allows Sn, OEM must do more: • Have experts “scrub” the OEM-BOM looking for “highrisk” components. • Pinsky model identifies numerous risk areas – Examples: Ni underlayer, reflow, spacing • Verify that received components match product descriptions. • Cannot mitigate risk for COTS assemblies. All agree that risk with Pb-free Sn on component terminations can at best only be mitigated. Copyright 2010 LDF Coatings, LLC All Rights Reserved
Mitigating Whisker Risk Component-By-Component Kim et al, recommend electroplating instead of solder dip.
•
Where Pb-free cap layer over Sn termination finish must: 1. Remain solderable during storage 2. Remain impenetrable, even as IMC grows
• Investigated Ni, Au, Pd over 6 μm (240 μin.) Sn 1. Solderability During soldering, will thicker Sn under-layer (which expands as it melts), break the cap? Example: Oxide on Sn not attacked by Type R flux, but does break up. Ni – NO If cap is continuous, but not for islands 2. Impenetrability Ni, Au, Pd: YES! Pb: NO (Li, AEM, anecdote, no photos)
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Mitigating Tin Whisker Risk (Kim, et al) Cap Type & Attributes Ni
50 nm
200 nm
Electrodeposit continuity? WB solderable (8 hr steam)? Whisker penetration?
No – Islands Significant decrease Promotes whiskers
Yes Unsolderable None in 5 years
Au
Mostly AuSn2 as deposited (1 day) Yes Yes None in 4 years
Largely AuSn2 after 2000 hours Yes Yes None in 4 years
Very little PdSn4 IMC as deposited Yes Yes None in 4 years
PD reaction rate with Sn < Au > Ni Yes Yes None in 4 years
Electrodeposit continuity? WB solderable (8 hr steam)? Whisker penetration?
PD Electrodeposit continuity? WB solderable (8 hr steam)? Whisker penetration?
Copyright 2010 LDF Coatings, LLC All Rights Reserved
200 nm Ni Cap Impenetrable For Five Years Kim, et al
Discontinuous Ni (50 nm) over Sn Stop-start whisker growth pattern Uncoated Sn (6 μm)
All other coatings over Sn (50 and 200 nm)
Copyright 2010 LDF Coatings, LLC All Rights Reserved
600 nm Copper Cap Penetrated In 3 Days (Reinbold et al) Virtually no growth of White spots are whiskers far from ledge. whiskers seen at 49 Cu diffused away from days; first whisker ledge; whiskers were on Cu-top seen at 3 found growing ahead of days where Cu could be detected IMC precipitation plastically deforms nearby Sn; 600 nm thick layer of whisker nucleates when Cu vapor deposited with stress exceeds a critical shadow mask over 1200 value. nm Sn (w/10 nm Cr Cu and Ni diffusion rate adhesion layer) difference in Sn cannot For Ni, explain huge whisker 1) Sn whisker is not an irresistible force, or penetration difference. 2) IMC precipitation does not deform Sn above critical nucleation value Copyright 2010 LDF Coatings, LLC All Rights Reserved
The Prevention Challenge Develop a practical process that can be applied to an assembly to protect it from whiskerinduced shorts. – Coat all tin with a material that permanently prevents whiskers from penetrating from below. – Prospects for a non-material added process are gloomy. – No more mitigation! Copyright 2010 LDF Coatings, LLC All Rights Reserved
The Ideal Prevention Process For Assemblies • Permanently suppresses whisker growth (with supporting data) • Eliminates BOM scrubbing • Eliminates testing for Pb-free Sn at receiving • Eliminates all mitigation efforts • Solves whisker problem with COTS assemblies • Applies to all electronic assemblies • Covers all tin – no shadowing • Negligible risk of damage to assembly function, reliability, and legibility • Negligible fixed and maintenance costs for equipment
Copyright 2010 LDF Coatings, LLC All Rights Reserved
The Ideal Prevention Process For Assemblies • Small equipment footprint • Inherently safe – doesn’t use exotic chemicals (i.e., not already in use in electronics industry) • Simple training of work force • Quick: handling time req’t ≯5 minutes/assembly, so no bottleneck • Can be performed immediately after cleaning and before conformal coating; no additional cost for masking What would it be worth per square foot?
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Whisker Prevention Coating Processes For Assemblies (Not Mitigation!) Three Game Changing Options: 1. “ALD-Cap” conformal ceramic - Sundew Technologies 2. “Whisker-Tough” conformal polymer - Smith & Co. 3. Selective metal cap - LDF Coatings, LLC Copyright 2010 LDF Coatings, LLC All Rights Reserved
Sundew ALD-Cap • Thin layer (~ 200 nm) of ceramic applied after proper surface prep. • Takes only minutes to apply • Batch process • Coating properties unlikely to change with temp, humidity, or time • Substitutes for polymer conformal coating • Rework not a problem • Process inherently coats under BGAs Copyright 2010 LDF Coatings, LLC All Rights Reserved
Smith & Company “Whisker Tough” •
Penetration and adhesion of flexible films was studied extensively to determine formulation requirements – Unique concept: tent and redirect a whisker’s growth rather than prevent its growth – no penetration. – Coating material formulated per findings to give at least 0.1 mil coverage along sharp edges of terminations (thixotropy).
•
Replaces conventional conformal coating
•
FOD virtually eliminated with specially designed dip tank
– marginal costs involve only extra cost of material, if any – – – –
A reasonable amount of operator training required Pot life ≮ 3 hrs (extendable indefinitely by replacing dragout) Overnight bake w/ temperature ramp, incl. 2 hrs at 85°C for full cure Compliance verification • • •
Completely transparent Any area of non-coverage findable by non-fluorescence under UV illumination Proper composition and cure easily verified by measuring coating modulus
– Rework comparable to conventional urethane CC’s Copyright 2010 LDF Coatings, LLC All Rights Reserved
LDF Coatings Selective Metal Cap Available whisker-preventing cap processes: – Ceramic – “ALD-Cap” – Polymer – “Whisker-Tough” – Metal Apply to a functioning assembly
1. Must apply to surfaces selectively: – Must coat all exposed tin – OK to coat other metals (mask contact areas) – Must not coat any insulating surface 2. Must be proven whisker-impenetrable Solderability, ductility not requirements! 1. Reduced to practice; patent applied for 2. Whisker penetration data: • Kim: No penetration of 200 nm Ni in 5 years • Chason: Penetration of 600 nm Cu in 3 days • Li: Penetration of Pb Copyright 2010 LDF Coatings, LLC All Rights Reserved
Selective Metal Cap Layer Process For Coating Tin • Must be a chemical process • Theoretically, three options – Electrochemical
• All tin surfaces must be electrically connected
– Immersion (replacement)
• Few candidates; too thin
– Electroless (autocatalytic)
• Except with special procedures, does not deposit on insulators • Numerous metals known to electrolessly deposit on tin • Ni and Pd are known to be impenetrable
What metal would you try first? What experiment would you try first? Copyright 2010 LDF Coatings, LLC All Rights Reserved
LDF Coatings, LLC is Formed • •
Landman, Davy, Fritz Provisional patent applied for
• •
Actual estimated deposit thickness needed ~ 1-2 μm < 5 min. Kim et al. found no penetration of 200 nm (0.2 μm) Ni in 5 yrs.
June 26, 2010 reduced to practice on an H&L Instruments fiberoptic pcb – Masked with tape and removable mask coating – Uses conventional cleaning – Total immersion 1 hr. ~25 μm (1 mil) Ni deposited – Ni reacts with air to form a protective oxide layer – Ni reacts with tin very slowly to form IMC; reaction rate drops as t1/2 Plenty of metallic Ni cap layer will remain in 30 years
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Process Applied To This Commercial Product
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Concept: Testing For Permanent Impenetrability Whisker boundary Greatest whisker distance Bare brass coupon
Sn coating
Observe impenetrable coating appearance along this edge Tapered coating over coupon
Tapered coating over Sn
L – Coupon with coating of tapered thickness of Sn R – Coupon with coatings of perpendicular tapered thicknesses of Sn and impenetrable coating. Greatest whisker distance is distance from edge to farthest whisker in a single observation. Whisker boundary is greatest whisker distance remaining unchanged for three successive monthly observations. Copyright 2010 LDF Coatings, LLC All Rights Reserved
How LDF Coatings Process Compares to Ideal • • • • • • • • •
Eliminates BOM scrubbing and testing components for Pb-free Sn at receiving Solves termination finish whisker problem with COTS assemblies Permanently suppresses whisker growth (Kim et al. > 5 yrs for 200 nm Ni) Applicable to all electronic assemblies (mask, and add components later if necessary) All tin is covered – no shadowing – hot aqueous immersion Risk of damage to assembly function, reliability, and legibility is negligible Fixed and maintenance costs negligible Small equipment footprint Process is inherently safe – does not use exotic chemicals (i.e., not already in use in electronics industry) – Long history, wide use of electroless Ni Copyright 2010 LDF Coatings, LLC All Rights Reserved
How LDF Coatings Process Compares to Ideal • Training of work force is simple – expect a broad process window • Quick, so no bottlenecks: time requirement ≯ 5 minutes per assembly • Process can be performed immediately after cleaning and before conf. coating; no additional cost for masking • Easy cleanup of assembly for next process step – water and steam then dry • Built-in practical compliance verification – bubbles, coated Cu looks different (+XRF) • No interference w/ rework - masking allows components to be added later Copyright 2010 LDF Coatings, LLC All Rights Reserved
LDF Coatings Next Steps • • •
Provisional patent applied for Continue research and publish, while seeking select companies for participation in trials Seek a corporate licensing partner – For more R&D and to commercialize • • • • •
–
Verify min. thickness for permanent suppression Does IMC prevent penetration? Explore process window for robustness Effects of porosity, P content, etc. Pd over Ni, Ni-Pd alloy for extra corrosion resistance?
To assume marketing, production, installation, distribution, customer service (including recycling)
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Problem and Solution Review • High-rel assemblies risk tin whiskers from COTS components and sub-assemblies. • Preventing whisker risk with an ideal process on an assembly beats “mitigating” risk componentby-component. • Prepared and presented requirements list for an ideal process for assemblies. • Analyzed three processes for assemblies: ceramic, polymer, and selective metal coating.
Copyright 2010 LDF Coatings, LLC All Rights Reserved
Problem and Solution Review • Process for selective deposition of a thin cap layer of whisker-impenetrable metal, coating all tin but not insulating surfaces – – – –
Process described Reduced to practice Compares favorably to the ideal Assemblies are likely to remain permanently impenetrable
• LDF Coatings, LLC was formed to – Publicize the process – Seek assemblies from OEMs for further testing and to ensure that the process does not impair function – Seek a corporate licensing partner – Promote the implementation. Copyright 2010 LDF Coatings, LLC All Rights Reserved
Footnote Who says we won’t be able to buy SnPb solder?
Source: IPC Global Quarterly Solder Statistical Program, Electronics Industries Market Data Update, Spring 2010. Free to members at http://www.ipc.org/Update. Copyright 2010 LDF Coatings, LLC All Rights Reserved
Copyright 2010 LDF Coatings, LLC All Rights Reserved