AIMCAL, CHARLESTON, SC, 2013 The Effects of Defects on the Moisture-Barrier Performance of Clear Coatings on Polymer Substrates. The National Centre for Printed Electronics ALF part SMITH of Centre for Process Innovation Ltd. Copyright CPI 2013. All rights reserved
CPI: National Printable Electronics Centre UK national facility for printed and plastic electronics (established in 2008) Purpose - scale up from R&D to pilot processing 1500 m 2 of clean room facilities Substrates from 4 square to 370mm x 470mm Sedgefield? London?
CPI - Printable Electronics Overview CPI Printable Electronics is a design, development and prototyping facility for the emerging Printable Electronics (PE) industry. Funded by :- ~33% Catapult (TSB, UK Government) ~33% Collaborative Development (UK, EU) ~33% Commercial Development (Global) CPI Printable Electronics works with clients to bring new PE products and processes to market quickly and efficiently, by offering facilities and expertise that help reduce the level of R&D risk and capital investment.
De-risking investment Technology Push Universities don t have the experience Industry doesn t have the time Consequence: Research fails to reach Market Market Need Product Challenges Better products Innovative effects Higher yields Business Challenges Lower material, operating and fixed costs Faster production rates More sustainable products Research The Innovation Gap Production 1 2 3 4 5 6 7 8 9 Basic Idea Concept Developed Proof of Concept Process Validation In Lab Process Validation Production Scale Process Capability Validated Capability Validated Economic Run Capability Validated Range Of Parts Capability Validated Over Long Period
CPI Printable Electronics Facilities Fabrication Prototyping Access to Scaleable Toolsets Commercialisation Support Incubator Space Facilities 2 clean rooms (Class100 &1000) Formulation, optical, & electrical test labs 12 incubator offices Expert techno-commercial team
CPI s activities in materials formulation and processing Electronic Design Scale up capability Core competencies in formulation and processing of electro-active materials Application process Material design Formulation capability Barrier film technology OTFT Flexible display backplane arrays Organic Logic & Memory Low cost sensor arrays Biosensors OLED Lighting Photovoltaics Portable power Architectural Automotive/ Aerospace Low cost offgrid generation
Ultra-barrier
Ultra-barrier Market
Ultra-barrier Market The market size estimates are: 43% of the total printable and plastic electronic markets are predicted to be flexible devices requiring high grade barrier encapsulants in 2021 (IDTecEx 2012). Total market size $23B in 2021, of which $17B are flexible OLED displays, lighting and PV. Around $1.4bn for barrier products alone for flexible electronics applications in 2021 (IDTecEx 2012). BIPV flexible CIGS total market $635M by 2017 alone, with a market of $120-170M for clear flexible barrier films. (CIGS Photovoltaics markets-2012 report, Nanomarkets).
MARKET NEEDS FOR ULTRA-BARRIER
Barrier layer requirements on plastic substrates Water vapour barrier g.m 2.day -1 Polymer films 1s-100s Food packaging 0.1-10 Electro-Chromics 0.001-0.05 Flexible PV 0.00001-0.001 Flexible OLEDs 0.0000001-0.00001 Barrier retention over product lifetime!
What does this mean? ~100m ~50m Imagine a polymer sheet the size of a football pitch: How much water would pass through this over a MONTH at various barrier performance levels?
What does this mean? Raw Film Food Packaging OPV OLED 100 10 1 1 X 1O -2 1 X 1O -4 1 X 1O -6
Single layer v Multilayer approach
Single and multi-layer barriers A key enabling technology for flexible electronics on plastic Inter-layer Barrier Barrier layer Single perfect layer Plastic film Plastic film Single layer Multi-layer Can be the substrate and encapsulation (packaging) for devices Perfect layer can provide high barrier (high density, low defects) Multi-layer approach (e.g. Vitex)
H 2 O (g/m 2 ) Water vapour transmission through barriers of equivalent total thickness Low barrier Time Medium barrier High barrier Single barrier Polymer H 2 O (g/m 2 ) Lag times Double barrier Time assumes dry polymer films at outset follows R.Ash et al, Brit.J.Appl.Phys. 1965 see also G. Graff et al, J.Appl. Phys. 2004
Deposition Systems.
Optilab TM Sputter Coater 12 250 micron webs Web speed 0.1 20m/min Plasma Treatment Medium Frequency AC dual rotatable magnetron 720mm length, 152mm Ø 6-12mm target thickness Max. 400mm web width (356mm coating width) In-line optical and electrical resistance monitoring Planar 10kW pulsed power DC magnetron 650 x 135mm target area Medium Frequency AC or pulsed DC dual planar magnetron 720mm length, 152mm Ø 6-25mm target thickness
Atomic Layer Deposition (ALD) Oxford Instruments FlexALTM tool designed for processing semiconductor wafers Atomic Layer Deposition is a type of CVD process that builds up a thin film coating as a succession of layers, each ~1Ǻ thick Batch process: coating up to max 200mm wafer 140mm x 140mm Polymeric (PET, PEN) or rigid (Si, Glass) samples Coating is dense, with few defects relatively free of pinholes as required for barriers and encapsulation Different materials can be combined to form nano-scale laminates Much work taking place on continuous ALD projects
Barrier + Defect Measurement.
Barrier Measurement Mocon Water Vapour Transmission (Aquatran 1, 6 measurement stations installed) lowest sensitivity 5 x 10-4 g/m2/day calibrated standard Ca Test developing technique - issues with encapsulation - absolute measurement difficult spatial information
Barrier + Defect Measurements Mocon Aquatran 1Water Vapour Transmission (6 measurement stations installed) Best sensitivity 5 x 10-4 g/m2/day Calibrated standard: Quantitative Ca Test Developing technique Issues with encapsulation Absolute measurement difficult Spatial information: Qualitative Surfscan Particle Number Measurement Optical Microscopy Optical Interferometry (CCI)
PETeC Calium Test facility Class 100 Nitrogen Purged Glove Box Calcium Deposition Unit Calcium Buttons on 100mm glass slide
WVTR for ALD Barriers Using the Ca Test Method- Optical Measurements The sample is aged in an environmental oven and measured at intervals Is R H2O Plastic Barrier Glass A Epoxy B T Im Absorbing calcium button Camera or small-area OD meter OD meter samples ~10mm2 Camera images the 28mm dia Ca button in transmitted light Water vapour oxidises Ca metal to transparent Ca(OH) 2
Laser-test rig Film with colour and texture is used to show the laser beam. The transmitted beam is ~ 5 x 12 mm; the receiver slit is 1x10 mm.
Results on single layers, barrier and defects.
Sputtered Al 2 0 3 barrier coatings Al 2 0 3 coatings 30-60nm thick Interlayers, 2-3 micron polymer 175micron heat stabilised PET Water vapour barrier g.m 2.day -1 (Mocon) Single Al 2 0 3 layer 5.10-2 5.10-1 Double Al 2 0 3 layer 4-6.10-3 Triple Al 2 0 3 layer < 5.10-4 Single Al 2 0 3 layer (non-conventional) <5.10-4 1.10-3
Comparison of layers Fast-evap AlOx -MOCON 5x10-1 g/m 2 /day Sputtered AlOx (conventional) -MOCON 5x10-2 g/m 2 /day Sputtered AlOx (non-conventional) - MOCON ~5x10-4 g/m 2 /day ALD AlOx - MOCON < 5x10-4 g/m 2 /day (meaurement conditions 38 C, 90%RH)
Defect-mapping Ca-test 30x30 mm Ca buttons on a 100x100 mm glass panel
Sputter and evap AlOx Sputter (Conventional) Failed at 3 days (21 C / 50 %RH) Fast Evap (Packaging AlOx R2R) Failed at 3 hours (21 C / 50 %RH)
Sputter (Non-conventional) Day -3 (21 C / 50 %RH) Day 0 (60 C / 90 %RH)
Sputter (Non-conventional) Day 1 (60 C / 90 %RH) Day 3 (60 C / 90 %RH)
Sputter (Non-conventional) Day 7 (60 C / 90 %RH) D: 0-1-3-7 (60 C / 90 %RH)
Y\X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 0 PPP survey 1.5 3 4.5 6 1 2 30x25 3 15x15 4 25x30 25x20 7.5 5 35x20 Example of the defect count on 20x20 mm or as-received planarised PEN film 9 11 12 14 15 17 18 20 21 23 24 26 6 10x10 20x30 20x15 5x5 40x35 7 40x50 10x10 30x30 8 40x45 15x15 9 50x30 40x25 10 50x30 11 30x50 40x60 12 40x70 13 20x55 14 70x70 15 15x15 50x90 16 30x20 45x80 17 40x40 25x20 15x25 5x5 27 18 40x20 5x5 19 5x5 20x30 40x120 120x80 250x200 20x30
ALD-Samples to compare particle counts to bright spots. Days: 0-4-24-35-46-56-81-91 12k2911 ALD-Feb20-03
Comparison of particle-count defects. 12k2911 has high-barrier siblings, but low-ish particle count (~100) Particle count at 0.48-5.2 µm: 67 150 79 ALD-Feb20-03 has mediumbarrier siblings, but medium particle count (~150) Particle count at 0.48-5.2 µm: 64 146 150 The Surf-scan estimates that there are ~1 particle per square cm generated by the ALD-barrier deposition. The CCI estimates that there are 8 to 12 particles per square cm on the as-received film, which doubles on deposition. The number of bright-spots per square cm is in excess of 50-100. The number of bright spots is ~50-100x the number of particle adders on reference wafers; therefore (all of) the bright spots are not caused by particles in the process. The number of bright spots is ~10x the number of as-received film defects on reference wafers; therefore (all of) the bright spots are not caused by defects in the base film. It is not known if this version of the calcium test induces defects in the barrier film. There are some bright-spot defects (e.g. 3 in 9 cm 2 ) that release as much moisture as all the rest (e.g. 900 in 9 cm 2 ) combined. The low-count/high-flow defects are the ones that must be reduced/removed.
Looking through the glass (and epoxy)
Looking through barrier film
Looking through the glass (and epoxy) Looking through the barrier film
mm 0 1.5 3 4.5 6 7.5 9 10.5 12 13.5 15 16.5 18 19.5 21 22.5 24 25.5 27 28.5 Y\X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 0 P 30x20x10 1 X G 50x20 G 30x20 X X G 20x10 P 40x30 200nm G 50x50 1.5 2 X G 50x20 G 35x20 X X P 5x5 150nm 3 H 3 G 35x25 G 30x40 G 10x10 40x40x0.6 P 15x30 200nm 4.5 4 G 35x30 G 35x30 P 25x35 P 5x5 700nm G 50x30 6 G 40x40 P 10x10 G 15x15 5 G 30x20 1200nm 7.5 9 11 6 G 20x20 G 40x20 G 30x30 G 20x40 H 40x40 G 20x20 7 H 30x40 H 40x20 600nm G 40x70 G 10x10 G 40x30 (3micron resolution) 12 14 15 17 8 9 10 G 40x30 Fibre 350x550 P 10x10 150nm 11 G 50x60 G 40x50 G 20x30 G 50x50 P 50x50 1000nm P 5x5 180nm H 130x70 1200 nm CCI feature map 18 20 21 23 24 26 27 29 12 X G 30x10 G 15x20 X G 3x3 G 20x15 13 X G 80x60 G 60x40 X X X G 30x30 G 10x10 14 X G 15x15 X X X G 25x15 G 25x40 15 X X G 10x30 16 G 25x20 17 G 20x15 G 20x25 G 15x15 G 30x30 18 G 35x25 G 60x30 19 20 G 30x30 Fibre 250x30 G 160x200 G 50x30 G 30x20 G 20x30 G 25x15 G 50x40 Fibre 500x500 G 20x20 Fibre 500x500 G 20x20 P 10x10 1000nm P 10x10 1000nm G 20x15 P 5x5 1500nm G 15x20 G 20x10 G 30x30 G 30x50 G 15x10 H 10x10 100nm G 15x25 G 15x15 P 5x5 120nm G 25x10 G 35x40 G 30x40 G 10x20 G 30x30 G 30x20 G 15x20 G 25x20 P 10x10 1500nm
Y\X 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 0 1.5 1 3 2 30x25 4.5 6 7.5 9 11 3 15x15 4 25x30 25x20 5 35x20 6 10x10 20x30 20x15 5x5 40x35 7 40x50 10x10 30x30 8 40x45 15x15 12 9 50x30 40x25 Substrate surface map 14 3micron res. 10 50x30 15 PPP survey 17 18 20 21 23 24 26 11 30x50 40x60 12 40x70 13 20x55 14 70x70 15 15x15 50x90 16 30x20 45x80 17 40x40 25x20 15x25 5x5 27 18 40x20 5x5 19 5x5 20x30 40x120 120x80 250x200 20x30
0hr Top, transmission
24hr Top, transmission
4days Top, transmission
6days Top, transmission
13days Top, transmission
CCI image of defects Large blob lower-right of lower-left cross 11-14, dot upper-right of lower-left cross
Attempts to ID brightening spot 4-1 5-1
ALD samples produced: Extrinsic barrier (large area measurement) MOCON < 5x10-4 g/m 2 /day Intrinsic barrier (small area) Laser spot 1x10-6 g/m 2 /day Excellent reproducibility at AlOx thickness >20nm Barrier for 5nm AlOX extrinsic MOCON < 5x10-4 g/m 2 /day Already produced, with lower reproducability.
Higher throughput ALD, towards R2R ALD.
Platform type CPI Batch ALD Tool (100x100mm) Present ALD - batch type Coming ALD - R2R type Present Sputter - R2R + batch type Future Combinations R2R (Sputter, ALD, etc...) CPI R2R Sputter Tool (400mm web)
Proposed CPI R2R ALD development Key focus for R2R ALD development: production line speeds of >1 m/min small machine footprint web width 0.5m thin high density barrier layers methods to reduce defects + effect of defects robust mechanical barrier materials low production cost
Technology outlook Collaboration between hardware and process development companies. Collaboration with BENEQ (Finland) Second generation system(s) currently under construction for CPI. COMING... Facilitates rapid technology engagement with industry CPI will collaborate with institutes and companies Globally for wide range application for R2R ALD. Roadmap for R2R ALD Q2 2014 Scaling beyond 500 mm, line-integration Further web speed improvement by PEALD Pre-and post-processes to address particles Excellent prospects Operational to scaling to +1 meter widths, at due CPI to reasonable mechnical tolerances 0.5 mm gap across 500 mm web width
21.10.2013 Beneq 2013 58
Ultra-barrier Targets Line speed >5m/min Barrier 1 x 10-6 g/m 2 /day Barrier coating cost $2-4/m 2 Other coatings FUTURE as direct encapsulation (OLED, PV, PE, SENSORS) as Nano-laminates (new materials) as Contacts (work function control) as Gate dielectrics TFT devices PV structures Surface active coatings. 21.10.2013 Beneq 2013 60
Thanks for listening! Contact: alf.smith@uk-cpi.co.uk