Sulfur Corrosion Failure Mechanisms and the Test Methods Developed - - PowerPoint PPT Presentation

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe Sulfur Corrosion Test September 16, 2016 1

Sulfur Corrosion

Failure Mechanisms and the Test Methods Developed to Mitigate Catastrophic Events

ACKNOWLEDGEMENT: Randy Schlueller, Dell Inc. provided much field data and photographs. Randy presented his paper “Creep Corrosion On Lead-Free Printed Circuit Boards In High Sulfur Environments” at the 2007 SMTA conference.

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Outline

 Overview – Sulfur Corrosion  Product Risk and Factors  PCB Finish Methods  Extent of the Impact  Mitigation:

• Design Recommendations
• Sulfur Corrosion Testing

 Test Results  Conclusion  Supporting Information

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Overview: Sulfur Corrosion

Case Study: Separately, HP and Dell recently experienced dramatic product failures within first 2 weeks of installation at multiple automotive design labs. Replacement systems failed similarly. Conclusion: Airborne sulfur from modeling clay corroded immersion silver PCB trace finish. “Sulfur Gas at the Parts per Million Level Can Cause Computer Systems to Fail Within 2 Months of Use.”Hewlett-Packard Failure Mechanism: Galvanic reaction with electrode

potential between Copper and Silver when copper is exposed to high levels Sulfur and >50% Humidity.

‘Creep Corrosion’

Copper Sulfide (Cu2S)

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Products at Risk

Vulnerable Product: Immersion silver (ImAg) PCB finish in corrosive environment applications

Presentation Focus: IT Hardware Initiatives Have Resulted in Sudden and Drastic Increases in Sulfur Corrosion Pb

Absorbs active sulfur

Electrolyte transfers CU, reacts with S & deposits CU2S

Surface moisture

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Corrosion Factors

• Aggressive dendritic formation growth in all directions
• Causes low resistance trace paths
• Voltage not required (not electrochemical migration)
• Thermal cycling induced: Cu2S created/dissolved in moisture

layer; forms dendrites when moisture dries

• Increased airflow (fans) accelerates sulfur exposure, failures

Initiatives Have Resulted in Sudden and Drastic Increases in Sulfur Corrosion Pb

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Sulfur Environments

• Vehicle exhaust fumes present
• Adhesives
• Desiccant
• Sewage/waste-water treatment
• Farms/cattle yards
• Swamps/bogs/wetlands

8.8 Mil Tons of Sulfur Produced in the US Annually

Majority of Failures Occur within 2 to 4 Months Rapid Failure Environments

• Rubber manufacturing
• Water treatment
• Paper mills
• Fertilizer production
• Petrochemical
• Automotive design (clay modeling lab)
• Packaging areas

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

PCB Finish Materials

PB Free Methods

• Immersion Silver (ImAg)
• Hot Air Solder Leveling (HASL)
• Organic Solder Preservative (OSP)
• Electroless-Nickel Immersion Gold (ENIG)
• White Tin

 Industry-Accepted ImAg Benefits

• Only cosmetic tarnishing expected
• Studies indicated electrochemical creep corrosion not a problem

 Recent Reports of Creep Corrosion in OSP Material

(Paper mill industry)

“While ENIG process has largest market penetration, last year more ImAg process lines were installed in PCB facilities than any other finish”. (Multicircuits, Inc.) ‘Finish’ Protects PCB Cu from Oxidation and Deterioration

Paper Mill

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Extent of the Impact

Supplier contamination ruled out: Creep corrosion found

• n PCBs supplied by 5 different vendors

HDDs with OSP in same sulfur environments did not show signs of corrosion  Dell study compared ImAg, OSP, ImSn, ENIG. Concluded ImAg, ENIG could not survive ISA G2 environment. Alcatel-Lucent ImAg testing (ISA Class G2 equivalent MFG at 69% RH) confirms creep corrosion in surface finish Failures Most Frequently Occur in Desktop/Storage Computers & HDDs, Rather Than Laptops

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

• Defined as: Hrs/yr where

temp >0°C, and RH >80%

• 6 Catagories 1 to 6
• Critical Zone: ≥10%/yr

Extent of the Impact

Impact of RH Factor Determined by Duration of Exposure ‘Time of Wetness’

Time of Wetness and Pollution Degree combined with ANSI/ISAS17-07 standard

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Extent of the Impact

2009 EU Directive for the Energy Performance of Buildings: Energy reduction via circulation of

• utside air. (EU Data Centre Code of Conduct, Section 3)

Impact: Significantly greater exposure to corrosive gasses with varying humidity environments. Result: Greater events of creep corrosion failures

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Mitigation: Sulfur Corrosion Testing

Many corrosion failures undetected by electrical testing Corrosion causes electrical shorting; various symptoms result

(depending on which conductors are the first to bridge)

Resistance decrease measured during re-exposure of corroded product to high humidity: >10 Mohm to < 1 Mohm

Standard Test Methods Don’t Capture Creep Corrosion Failures

ISA S71.04-1985 OnGuard Correlation Effects Class GX >2000 Å/30 Days Class GX >66 Å/24 Hours Electronic/electrical equipment not expected to survive due to corrosive attack. Class G3 <2000 Å/30 Days Class G3 <66 Å/24 Hours High probability that corrosive attack will

• ccur. Probable effect on equipment reliability

in less than 5 years. Class G2 <1000 Å/30 Days Class G2 <33 Å/24 Hours Effects of corrosion are measurable and are a factor in equipment reliability. Possible effects in less than 5 years. Class G1 <300 Å/30 Days Class G1 <10 Å/24 Hours Corrosion is not a factor in determining equipment reliability.

ISA Standard

"Environmental Conditions for Process Measurement and Control Systems: Airborne Contaminants” ISA-S71.04-1985

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Mitigation: Sulfur Corrosion Testing

Historical test methods for MFG (Mixed Flow Gas)

• IEC
• EIA
• Telecordia
• Battelle

 Does not properly produce creep corrosion  Does not provide a usable acceleration factor to determine proper test duration to simulate 1 year, 2 years or product life

“The industry requires a test method that will consistently

produce creep corrosion on ImAg control samples in a way as is observed in high sulfur environments.” Dell, inc.

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Mitigation: Design Changes

Design Recommendations:

• Plugging all non-test vias with soldermask
• Use of non-soldermask defined test vias and pads
• Spacing vias and pads sufficiently to prevent

bridging

• Solder paste to cover all remaining metal features on

the PCB

• Discontinue use of ImAg surface finish
• Test product samples

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Mitigation: Sulfur Corrosion Testing

Clay Test Method

• Early development by Randy Schueller, Dell, Inc.
• Utilizes high content (30-50% elemental sulfur) sulfur modeling clay
• Wetted/heated clay placed in sealed container
• Pre-cooled PWA samples placed in container with no direct clay contact
• Exposed to the sulfurous environment for approximately 12 hours
• Repeated process four times, measure corrosion

Creep corrosion typically visible after 2 days, pronounced after 5 days

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Mitigation: Sulfur Corrosion Testing

3 day exposure

Clay Test Results

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Conclusion

Failures Have Occurred, Products are at Risk Design/Mfg Process Must Consider Corrosive Customer Environments ImAg Poses Highest Known Risk Testing Methods Now Available

Future Work

• Refinements to the clay test method

Improve control of humidity, temperature, sulfur levels and include airflow

• Research other test methods
• Standard test method adoption

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Thank ank You

• u

David Rahe

512-775-4001 drahe@dlinnovations.com

3201 Industrial Terrace, Suite 120, Austin, TX 78758 www.dlinnovations.com

ASTR 2008 Oct 1 – Oct 3, Portland, Oregon

David Rahe, DLi Labs September 16, 2016

Sources

 Article: Accelerated Corrosion of Printed Circuit Boards due to High Levels of Reduced Sulfur Gasses in Industrial Environments, Paul Mazurkiewicz, Ph.D. Hewlett-Packard Company, Fort Collins, Co, USA  (http://www.multicircuits.com/pcb/tech/surface_finishes.html)