Tin whisker mitigation. research into mechanisms & strategies: part 1: effect of plating methodologies.

Transcription

1 Loughborough University Institutional Repository Tin whisker mitigation research into mechanisms & strategies: part 1: effect of plating methodologies This item was submitted to Loughborough University's Institutional Repository by the/an author. Citation: ASHWORTH, M.A. and WILCOX, G.D., Tin whisker mitigation research into mechanisms & strategies: part 1: effect of plating methodologies. Tin Whisker Mitigation Methodologies: SMART Group Seminar, Holywell Park, Loughborough University, 24th November Additional Information: These are conference slides. Metadata Record: Version: Accepted for publication Rights: 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: Please cite the published version.

2 Tin Whisker Mitigation Research into Mechanisms & Strategies M.A. Ashworth and G.D. Wilcox Tin Whisker Mitigation Methodologies Loughborough University 24 th November 2016

3 Whiskers Research at Loughborough WHISKERMIT Deposition current density Process optimisation (commercial bright tin) Deposit thickness Pulse plating In-service mitigation strategies Incorporation of particulates Conformal coating WHISKERMIT present Development of novel conformal coatings for whisker mitigation Electrochemical oxidation to mitigate whisker growth

4 Tin Whisker Mitigation - Research into Mechanisms & Strategies PART 1: EFFECT OF PLATING METHODOLOGIES PART 2: POST-PLATING MITIGATION METHODS

5 Part 1: Outline of presentation Introduction Experimental approach Deposition onto copper Deposit characterisation Deposition onto brass Effect of electroplating parameters Pulse plating Summary

6 Tin whiskers Crystalline growths from a metal surface (e.g. Sn, Zn and Cd) Uncertain incubation period before growth Numerous growth morphologies possible A few micrometres in diameter and up to several millimetres in length Although investigated for ~70 years, whisker related problems are increasing due to environmental legislation and device miniaturisation 1 mm Spiral Kinked Odd-shaped eruption Hillock/nodule Filament 10µm 1µm 10µm 1µm 20µm

7 Factors that influence whisker growth Electroplating bath chemistry Pure tin or tin alloy (Sn-Pb, Sn-Bi, Sn-Cu etc.) Bright or matte tin Electroplating parameters Current density, temperature, agitation Deposit characteristics Grain size and morphology, orientation, deposit thickness, Sn oxide Substrate Cu, brass, alloy 42 Intermetallic formation, elemental diffusion from substrate Environmental conditions Temperature, humidity, thermal cycling, applied external stress

8 Experimental approach All samples were electroplated using a commercial bright tin electroplating bath (Tinmac) Effect of deposition current density, deposit thickness and substrate composition investigated 5, 10, 20, 30, 40 and 50 ma cm -2 2, 5 and 10 µm Copper, brass (63Cu/37Zn) and Alloy 42 (58Fe/42Ni) substrates 3 samples per condition electroplated late 2011 Selected samples stored at 55 C/85% humidity

9 Sn on Cu: whisker growth after ~2 years 10mAcm -2 2µm #1 10mAcm -2, 2µm, #3 20mAcm -2 2µm #1 10µm 2µm 1µm 30mAcm -2 2µm #1 2µm Deposit thickness 2 µm 5 µm 10 µm 10 ma cm ma cm ma cm ma cm -2 #1 #2 #3 HT #1 #2 #3 HT #1 #2 #3 HT #1 #2 #3 HT X X X X X X X X X 30mAcm -2 5µm #1 2µm 10mAcm -2 10µm #1 10mAcm -2 5µm #3 20mAcm -2 10µm #3 30mAcm -2 10µm #1 1µm 2µm 2µm 2µm Samples from batch #3 stored in oven for 5000h at 55 C/85% RH

10 Whisker growth on Sn deposits on Cu Examples of whisker growth present on Sn deposits on copper (batch 2) after storage at room temperature for 5 years.no long filament whiskers observed 2µm, 40 ma cm -2 5µm, 30 ma cm -2 5µm, 40 ma cm µm 2 µm 10 µm 5µm, 40 ma cm -2 10µm, 20 ma cm -2 10µm, 30 ma cm µm 2 µm 10 µm

11 Whisker growth on oven stored samples 2µm, 10 ma cm -2 2µm, 10 ma cm -2 Overall, whisker growth on the heat treated samples remains comparable to that on the samples stored entirely at room temperature 10µm, 30 ma cm -2 2 µm 10 µm 10µm, 30 ma cm µm 2 µm

12 Sn deposits on Cu - summary Sn deposits on copper produced in this study demonstrate low levels of whisker growth, even after 5 years storage, despite being produced from a bright tin electroplating bath Variation in whisker density between samples from different batches Only significant whisker growth observed for thinnest deposits (2 µm) electroplated at lowest current density (10 ma cm -2 ) No increase in whisker growth with storage at 55 C/85% RH for 5000 hours Why such low whisker growth? Characterisation of deposit microstructures

13 Effect of deposition current density 5 ma cm ma cm ma cm -2 1µm 1µm 30 ma cm ma cm ma cm -2 1µm Facetted cuboidal grains formed at 5 macm -2 Reduction in grain size at higher current densities 1µm 1µm 1µm

14 Effect of deposition current density on microstructure 5 macm macm macm -2 As deposition FIB crosssection (after 1 month) current density increases: Grain size reduces and become less 2 µm 2 µm 2 µm facetted Grain orientation map (deposit surface) µm 5 µm 5 µm 010 Deposit becomes increasingly columnar All deposits show a strong 001 orientation EBSD Deposit orientation 110

15 Effect of storage at elevated temp/humidity FIB analysis ~ 2 years after tin deposition Cu 6 Sn 5 1µm Cu 3 Sn Cu 6 Sn 5 1µm 5 µm tin 20 macm -2 2 years room temperature storage Schematic of Cu 6 Sn 5 formation 1 10 µm tin 20 ma cm -2 2 years storage includes 5000 hours at 55 C/85% RH Cu 3 Sn layer develops during storage at 55 C IMC layer is more planar compared with that developed at room temperature No acceleration in whisker growth observed compared with samples stored at room temperature 1 B. Z. Lee and D. N. Lee, Acta Mater. 46, 3701 (1998)

16 From the literature A novel electrolyte for the high speed electrodeposition of bright pure tin at elevated temperatures, Zhang, W., Guebey, J., and Toben, M., Metal Finishing, 2011, 109, 1-2, p The very low whiskering of the deposit is attributed to a combination of the mixed equiaxed and columnar grain structure and the preferential <112>->101>-<103> multi-peak texture of the deposit Intensity Angle 2θ XRD pattern for a 10 µm deposit electroplated at 20 ma cm -2 in the current study WO 2010/ A1: Preventing or mitigating growth formations on metallic films, J.W. Osenbach, 2010 Patent suggests that tin deposits having a <001> preferred texture are less susceptible to whisker growth

17 Acceleration of whisker growth To more fully investigate the effect of process variables (and study mitigation techniques) more rapid whisker growth was required Deposition onto brass is known to promote rapid and profuse whisker growth For tin deposits on brass, whisker growth is primarily driven by zinc oxide formation at the deposit surface rather than intermetallic growth at the Snbrass interface Zn oxide 5 µm tin deposit on brass, electroplated at 10 ma cm -2 ~29 months after tin deposition Blocky Cu 6 Sn 5 intermetallic 1µm

18 Zn incorporation into whiskers? Auger analysis carried out to investigate the composition of the whiskers formed on Sn deposits on brass Prior to sputter Zn present on whisker surface After sputter Zn absent after oxide removed C Sn O Zn Sn O Zn Kinetic energy (ev) Kinetic energy (ev) Zn incorporation into whisker? Whiskers are pure Sn

19 Effect of deposition current density 10µm 10µm 10µm 10µm (a) 2 µm, 10 ma cm -2 (b) 2 µm, 20 ma cm -2 (c) 2 µm, 30 ma cm -2 (d) 2 µm, 40 ma cm -2 10µm 10µm 10µm 10µm (e) 10 µm, 10 ma cm -2 (f) 10 µm, 20 ma cm -2 (g) 10 µm, 30 ma cm -2 (h) 10 µm, 40 ma cm -2 Evaluation of whisker density for 2 µm tin deposits on brass Analysis approx. 33 months after tin deposition Storage at room temperature

20 Effect of deposition current density µm deposits µm deposits mAcm mAcm-2 Number of features /mm mAcm-2 30mAcm-2 40mAcm-2 Number of features /mm mAcm-2 30mAcm-2 40mAcm Filaments Eruptions Eruptions/nodules < Whisker feature type 10 µm Filaments Eruptions Eruptions/nodules < Whisker feature type 10 µm Analysis approx. 33 months after tin deposition Storage at room temperature

21 Extension of current density range 5 ma cm ma cm -2 Further reductions in deposition current density to 5 ma cm -2 generate increasing densities of long 100µm filament whiskers Further increases in deposition current density to 50 ma cm -2 result in low densities of 100µm large eruptions Number of features /mm mAcm mAcm-2 40mAcm mAcm Filaments Eruptions Eruptions/nodules < Whisker feature type 10 µm 5 µm thick deposits Whisker size and distribution ~18 months after deposition

22 Effect of deposit thickness µm deposits 10µm (a) 2 µm, 10 ma cm -2 10µm 10µm (e) 10 µm, 10 ma cm -2 10µm Whisker density (whiskers/mm 2 ) µm deposits 10µm deposits (d) 2 µm, 40 ma cm -2 (h) 10 µm, 40 ma cm Analysis approx. 33 months after tin deposition Storage at room temperature Deposition current density

23 Whisker growth after ~ 5 years Density of Tin Whiskers (Total)/mm ma cm-2 20 ma cm-2 30 ma cm-2 40 ma cm-2 2 μm 5 μm 10 μm Deposit Thickness Density of Tin Whiskers /mm Filaments Eruptions Whisker Feature Type Whisker growth for the batch 3 samples has been evaluated after ~5 years storage at room temperature 2 µm deposit 5 µm deposit Density of Tin Whiskers /mm 2 Density of Tin Whiskers /mm Filaments 10 µm deposit Eruptions Whisker Feature Type Filaments Eruptions Whisker Feature Type

24 Effect of deposition current density Results demonstrate that the growth of filaments, eruptions and nodules, whisker growth is reduced as the deposition current density is increased At higher deposition current densities the growth of nodules and eruptions is favoured Whisker growth reduces as the deposit thickness increases Reduced whisker growth could be achieved by deposition of thicker coatings at higher deposition current density

25 Effect of Pulse Plating Pulse plating may be used to manipulate the grain structure and orientation of the tin deposit µm tin on copper 10% duty cycle 20 Hz frequency 2µm 2µm 110 Results, for tin deposits on brass, showed that whisker growth could be influenced by control of pulse plating parameters (duty cycle and pulse frequency)

26 Effect of pulse plating on whisker growth Long filaments Deposits produced using a 10% duty cycle and 2.5 Hz showed reduced Large eruptions whisker growth compared with direct current deposits 100µm Direct current, 20 ma cm µm Pulse plated, 10 % duty 2.5 Hz Pulse plated, 10 % duty 20 Hz 3 weeks after deposition BUT, pulse plating could also result in greatly accelerated whisker growth compared with direct current deposits low pulse frequencies preferable 1mm

27 But.. The observed influence of current density on whisker growth may only be applied with regards to the electroplating bath used for the current study Other electroplating bath formulations may demonstrate a different relationship between current density, deposit microstructure and whisker growth E.g. it is often thought that fine grained columnar structures and higher deposition current densities are more prone to whisker growth Elevated temperature and humidity may also affect other tin deposits in a different way

28 Whisker growth on consumer products 500 µm 500 µm 500 µm Freeview Cu Kα digital recorder, circa 2009 Zn Kα Examples of tin whiskers found on the connector pins of an electroplated SCART connector XRF analysis showed that the connector was electroplated with ~ 5 µm of bright tin. Sn Lα.. analysis also showed that the substrate was composed of ~ 63% Cu and 37% Zn i.e. ideal conditions to generate whisker growth

29 Summary Bright tin electrodeposited onto Cu does not always result in significant whisker growth In the present study, whisker growth was reduced by increasing deposit thickness and by deposition at higher current densities Large eruptions, rather than filament whiskers were preferentially formed at high deposition current densities Exposure of Sn deposits on Cu to 55 C and 85% humidity did not accelerate whisker growth

30 Any questions? We would like to thank the UK EPSRC Innovative Electronics Manufacturing Research Centre (IeMRC) for funding this research