This item was submitted to Loughborough's Research Repository by the author. Items in Figshare are protected by copyright, with all rights reserved, unless otherwise indicated.

Tin whisker mitigation: research into mechanisms and strategies. Part 2,Tin whisker mitigation: research into mechanisms and strategies. Part 2,Post-plating mitigation methodsPost-plating mitigation methods

PLEASE CITE THE PUBLISHED VERSION

VERSION

AM (Accepted Manuscript)

PUBLISHER STATEMENT

This work is made available according to the conditions of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) licence. Full details of this licence are available at:https://creativecommons.org/licenses/by-nc-nd/4.0/

LICENCE

CC BY-NC-ND 4.0

REPOSITORY RECORD

Ashworth, Mark A., G.D. Wilcox, Helen E. Pearson, and D. Haspel. 2017. “Tin Whisker Mitigation: Researchinto Mechanisms and Strategies. Part 2, Post-plating Mitigation Methods”. figshare.https://hdl.handle.net/2134/23687.

Tin Whisker Mitigation – Research into Mechanisms & Strategies

M.A. Ashworth, G.D. Wilcox, H. Pearson and D. Haspel

Tin Whisker Mitigation Methodologies Loughborough University

24th November 2016

PART 1: EFFECT OF PLATING METHODOLOGIES PART 2: POST-PLATING MITIGATION METHODS

Tin Whisker Mitigation - Research into Mechanisms & Strategies

Part 2: Outline of presentation

• Introduction • Novel conformal coatings • Electrochemical oxidation • Atomic layer deposition • Summary

Introduction • Although whisker growth may be reduced by choice of

electroplating bath composition and optimisation of deposition parameters, such methods cannot be relied upon to fully mitigate whisker growth

• Given the unpredictable nature of whisker growth, even the ‘best’ electroplated tin coatings should be considered ‘whisker-resistant’ rather than ‘whisker-proof’

• Additional control measures, in the form of surface coatings or treatments, are required to further suppress the growth of whiskers

1. NOVEL CONFORMAL COATINGS

Post-plating mitigation methods

Tin whiskers and conformal coatings • Conformal coatings are applied to protect

printed circuit boards and associated electronic components

• Currently, tin whisker mitigation is attempted with conformal coatings that have not been designed to prevent whisker growth

• WHISKERMIT 2 research programme aimed to develop novel conformal coatings specifically designed to mitigate whisker growth

Research Aims and Objectives

• Engender polymers with physical barriers to whisker growth through the inclusion of nano-fillers in the conformal coating polymer formulation.

• Apply concept to commercial conformal coatings that are currently used for environmental protection in electronic components

• Evaluate the microstructure, mechanical properties and whisker resistance of the modified coating formulations

Evaluation of whisker mitigation

Only view/count whiskers that are growing out of the coating

• Whisker growth has been investigated using brass coupons electroplated with 2 µm of bright tin at 10 mA cm-2

• Modified coatings based on a HumiSeal acrylic formulation

• All conformal coatings applied by spraying • Three sets of trials currently under evaluation

(batches 1, 2 and 3) • Samples stored in an environmental chamber at

55°C/85% humidity • Whisker growth evaluated at periodic intervals

using a stereo microscope

Evaluation of whisker growth: Batch 1

N.b. 418 days analysis conducted using x64 magnification; all others used 50x

3% nanofiller modified acrylic coatings vs. unmodified 2x coating (29-7-15), Samples stored throughout at 55°C/85% humidity

0

10

20

30

40

50

60

0 50 100 150 200 250 300 350 400 450 500

Whi

sker

den

sity

(cm

-2)

Days after coating

Whisker penetration reduced by ~ 2/3 for

3% nanofiller modified samples

Unmodified

100µm

3% nanofiller 100µm

Evaluation of whisker growth: Batch 2

0

20

40

60

80

100

120

140

0 50 100 150 200 250 300 350 400

Pene

trat

ing

whi

sker

den

sity

(cm

-2)

Days after coating

3% nanofiller modified acrylic coatings vs. unmodified 2x coating (9-10-15), Samples stored at 55°C/85% humidity and RT

Whisker penetration

reduced by ~ 2/3 for 3% nanofiller modified samples

Unmodified

3% nanofiller

0

50

100

150

200

250

300

350

400

Unmodified 3% 5% 7%

Whi

sker

den

sity

(cm

-2)

Evaluation of whisker growth: Batch 3

0

100

200

300

400

#27 #45 #63 #43 #59 #75 #21 #37 #69 #31 #39 #47

Whi

sker

den

sity

(cm

-2)

Sample #

0

10

20

30

40

50

60

70

80

#1 #2 #72 #5 #6 #35 #9 #53 #61 #13 #23 #71

Aver

age

coat

ing

thic

knes

s (µm

)

Sample #

top of sample middle of sample base of sample

Unmodified

3%

5%

7%

5% modified samples are typically

thicker than unmodified 7% modified

samples are comparable or thinner than

the unmodified

Modified acrylic coatings vs. unmodified 2x coating (11-5-16), Samples stored at 55°C/85% humidity

Unmodified

3% nanofiller

5% nanofiller 7% nanofiller

Very significant reduction in whisker

penetration

3% modified samples are thinner than unmodified

Batch 2: mechanical property data

Yield stress (MPa) % elongation at failure Failure stress (MPa) Young’s modulus (MPa)

Samples tested approximately 1 month after fabrication

Modulus and yield strength are significantly improved for the modified formulations whilst elongation to failure is only slightly reduced

2

3

4

5

100

120

140

160

180

200

220

240

300

320

340

360

380

400

420

440

460

480

500

2

3

4

5

6

2. ELECTROCHEMICAL OXIDATION

Post-plating mitigation methods

The role of the surface oxide in whisker growth In 1994, Tu proposed his “cracked oxide theory” 1

whisker growth occurs at certain weak spots on the surface where the oxide has been broken

In the absence of an oxide no whisker growth would occur

Later adding 2 …

“if the surface oxide is very thick, it will physically block the growth of any hillocks and whiskers”

1 Physical Review B 49, 2030, 1994

2 Proceedings of the IEEE Electronic Components and Technology Conference, 2002 p1194–1200

Can we mitigate whisker growth by increasing the thickness of the surface oxide? Electrochemical oxidation in borate buffer and potassium

carbonate/bicarbonate solutions used to develop thicker oxides

Although the oxide film thickness is increased, other characteristics of the Sn coating (e.g. grain size, grain orientation and residual plating stresses) should be unaffected

XPS analysis to evaluate the thickness and composition of the oxide layers formed

Whisker growth evaluated using SEM and optical microscopy

Electrochemical oxidation • Electrochemical oxidation can be used to develop oxide

films that are much greater in thickness than a native Sn oxide

XPS sputter depth profiles comparing the thickness of the oxide films 1 day after Sn deposition and electrochemical oxidation (at 1.2 V vs. SCE)

Comparable thicknesses of oxide are

developed for borate buffer and

potassium bicarbonate-carbonate

baths Increased oxide

thickness

0

10

20

30

40

50

60

70

80

0 100 200 300 400 500 600

Oxy

gen

conc

entr

atio

n (a

t%)

Sputter time (s)

Native

Borate buffer (40 mC cm-2)

Potassium bicarb-carb. (40 mC cm-2)

Effect of electrochemical oxide on whisker growth

2 µm SnCu deposits on Cu electroplated at 10 mAcm-2

Whisker growth is considerably reduced for samples given a post electroplating oxidation treatment at 1.2 V vs SCE

Comparable reduction in whisker growth achieved for borate buffer and potassium bicarbonate-carbonate solutions

Still effective at mitigating whisker growth after ~ 3 years

0

500

1000

1500

2000

0 200 400 600 800 1000 1200

Whi

sker

den

sity

(cm

-2)

Time (days)

Native air formed oxide

Borate buffer solution

Potassium bicarbonate-carbonate solution

Whisker growth evaluated over ~ 3 years

Electrochemical oxidation at 1.2 V vs SCE

Order of magnitude

reduction in whisker growth

Whisker development with time

• Study underway to compare the growth of whiskers, and the development of the oxide film, on native and electrochemically oxidised deposits as a function of time

• 2 µm SnCu deposits on Cu

• Electrochemical oxidation at 2 V vs. SCE in potassium bicarbonate/carbonate solution

Native oxide

Electrochemical oxide

0

500

1000

1500

2000

2500

0 10 20 30 40 50

Whi

sker

Den

sity

(cm

-2)

Days after electroplating

0

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

Native 1.2V 0.12C 1.6V 0.12C 2V 0.12C 1.2V 0.24C 1.6V 0.24C 2V 0.24C

Whi

sker

Den

sity

(cm

-2)

0.12 C charge passed 0.24 C charge passed

Whisker growth evaluated after ~ 9 months

Electrochemical oxidation at higher potentials

05

101520253035404550

0 100 200 300 400 500 600

Sn (o

xide

) Con

cent

ratio

n (a

t%)

Sputter time (s)

1.2V

1.6V 2V

XPS depth profiles for Sn(oxide)

Electrochemical oxidation of SnCu deposits on Cu in potassium bicarbonate-carbonate electrolyte

Oxide film thickness is increased at higher potentials

However, there is no clear improvement in whisker mitigation at higher oxidation potentials compared with 1.2 V

Whisker mitigation: Sn deposits on brass • Electrochemical oxidation

of 2 µm pure Sn deposits on brass in borate buffer solution

• Whisker growth is greatly reduced for electrochemically oxidised Sn deposits on brass

• Whisker mitigation is more effective at the higher oxidation potential

• Still an effective barrier to whisker growth after almost 4 years 5915 83 61 39

0

50

100

150

200

250

300

350

400

450

500

Native 1.2V 1.6V 2.0V

Whi

sker

den

sity

(cm

-2)

Whisker growth evaluated after ~ 44 months

Whisker mitigation mechanisms For Sn deposits on Cu, IMC formation appears unaffected by electrochemical oxidation

Whisker mitigation must simply be derived from the increased thickness of the oxide layer

No reduction in the driving force for tin whisker growth (intermetallic formation)

Cu6Sn5 intermetallic

2 µm

Cu6Sn5 intermetallic

2 µm

Electrochemical oxidation at 1.2V borate buffer

2 µm

No ZnO formation

Reduced intermetallic

formation 2 µm

Pronounced ZnO

formation

Large intermetallics

For Sn deposits on brass, electrochemical oxidation reduces whisker growth by preventing the formation of Zn oxide at the deposit surface

Reduced IMC formation may also influence whisker formation

Reduced driving force for whisker growth

Native air formed oxide

SnCu

dep

osits

on

Cu

Sn d

epos

its o

n br

ass

Electrochemical oxidation at 1.2V borate buffer Native air formed oxide

3. ATOMIC LAYER DEPOSITION Post-plating mitigation methods

ALD as a whisker mitigation strategy • Atomic layer deposition (ALD) has been

investigated as a potential whisker mitigation strategy

• ESA funded collaborative research programme between Loughborough University, Picosun and Oy Poltronic Ab

• Initial studies to investigate the effect of coating process variables on whisker mitigation

Experimental approach

• Copper coupons electroplated with 2 µm Sn-Cu to promote whisker growth

• Electroplated coupons shipped to Picosun for ALD coating

• A range of pre-treatments and process conditions investigated

• Uncoated control samples included in each batch of ALD processed samples

• Whisker growth evaluated over ~12 month period using optical and scanning electron microscopy

ALD Coating Trials Batch Pre-treatment Coating

Material Coating

Thickness

1 Pre-treatment

trials ‘ALD 1’ ‘Standard’ 2

3

4 - 5x ‘standard’

5 - ‘ALD 2’

‘Standard’

6 - 5x ‘standard’

7 - ‘ALD 3’

‘Standard’

8 - 5x ‘standard’

Batch 5

Batch 6

Batch 8

Whisker growth test coupons

Analysis of whisker growth

For the majority of the process batches

investigated, Filament whisker growth is

considerably reduced on the ALD coated samples

compared with control samples

Control sample

Batch 6 (‘ALD 2’, 5x standard thickness)

10µm 10µm

10µm 10µm

ALD coated

Batch 4 (‘ALD 1’, 5x standard thickness)

Control sample ALD coated

18 46 0 353 1 357 2 345 1 370 103 346 329 336 0

50

100

150

200

250

300

350

400

450

B1 B2 B3 B3-C B4 B4-C B5 B5-C B6 B6-C B7 B7-C B8 B8-C

Aver

age

num

ber o

f whi

sker

s cou

nted

(20

fram

es p

er sa

mpl

e)

Batch No.

Effect of ALD coating on filament whisker growth

Batch 3

Batch 4

Batch 5

Batch 6

Batch 7 Batch 8

Batch 2

Batch 1

Suffix ‘C’ refers to control samples

For batches 3, 4, 5 and 6 the growth of filament whiskers is almost fully

prevented by ALD coating (‘ALD 1 and 2’)

Batches 1 and 2 showed increased whisker growth but had a longer delay between electroplating and ALD coating

Reduced whisker mitigation for batches 7 and 8 (‘ALD 3’)

Whisker growth prior to coating? • SEM analysis indicates that the whiskers on the ALD

processed samples are typically encased within a

shell of oxide, i.e. they were already present at the

time of coating

• No examples where whisker growth has definitely initiated post ALD coating

1 µm

1 µm

Batch 4 Example of whisker growth on

Batch 2 (‘ALD 1’)

1 µm

Examples of whisker growth on Batch 4 (‘ALD 1’, 5x thickness)

Batch 2

Exposed Sn

whisker

Whiskers are encased in the

ALD coating

1 µm

Exposed tin whisker

Cracked ALD coating

Example of whisker growth on Batch 7 (‘ALD 3’)

Batch 7

SEM analysis of batch 8 (‘ALD 3’) Control sample ALD coated

5 µm 10 µm

10 µm 10 µm

ALD coated

Coating failure may be attributed to the combined effect of coating material, lack of pre-treatment and the application of too thick a layer

2 µm

1 µm

However, SEM analysis suggests whiskers were present prior to deposition of the ALD coating

Exposed tin whisker

Oxide shell

Filament whisker density for batch 8 was comparable to

that of the uncoated samples

Summary • Conformal coating studies

Novel conformal coatings are under development. Whisker growth studies show a significant improvement in whisker mitigation compared with unmodified coatings

• Electrochemical oxidation treatment

Electrochemical oxidation has been shown to provide long term whisker mitigation for both SnCu deposits on Cu and Sn deposits on brass

• Atomic layer deposition

Significant reductions in whisker growth have been achieved as a result of ALD processing

Any questions or comments?