Contents

Cover

Title page

Copyright page

Foreword by Peter L. Bocko

Preface

Part I: Flexible Glass & Flexible Glass Reliability

Chapter 1: Introduction to Flexible Glass Substrates
1. 1.1 Overview of Flexible Glass
2. 1.2 Flexible Glass Properties
3. 1.3 Flexible Glass Web for R2R Processing
4. 1.4 Flexible Glass Laser Cutting
5. 1.5 Summary
6. References
Chapter 2: The Mechanical Reliability of Thin, Flexible Glass
1. 2.1 Introduction
2. 2.2 The Mechanical Reliability of Glass
3. 2.3 Applied Stress
4. 2.4 The Strength of Thin Glass Sheets
5. 2.5 Summary
6. References
Chapter 3: Low Modulus, Damage Resistant Glass for Ultra-Thin Applications
1. 3.1 Introduction
2. 3.2 Young’s Modulus and Basic Fracture Mechanics
3. 3.3 Vickers Indentation Cracking Resistance of Calcium Aluminoborosilicate Glasses
4. 3.4 Summary
5. References

Part II: Flexible Glass Device Fabrication

Chapter 4: Roll-to-Roll Processing of Flexible Glass
1. 4.1 Introduction
2. 4.2 Roll-to-Roll Manufacturing Process Equipment
3. 4.3 R2R Deposition and Patterning of ITO on Thin Flexible Glass and Plastic Films
4. 4.4 Conclusions
5. 4.5 Future
6. Acknowledgements
7. References
Chapter 5: Thin-Film Deposition on Flexible Glass by Plasma Processes
1. 5.1 Introduction
2. 5.2 Substrate Requirements for Vacuum Processes
3. 5.3 Types of Vacuum Processes
4. 5.4 Large Area Coatings onto Flexible Glass
5. 5.5 Thermal Pre- and Post-Treatment for Flexible Glass
6. 5.6 Future Trends in Vacuum Processing on Flexible Glass
7. References
Chapter 6: Printed Electronics Solutions-Based Processes with Flexible Glass
1. 6.1 Introduction
2. 6.2 Printing Processes
3. 6.3 Summary of Different Printing Processes
4. 6.4 Example – Printed OPV Cell on Ultra-Thin Flexible Glass
5. 6.5 Future
6. References

Part III: Flexible Glass Device Applications

Chapter 7: Flexible Glass in Thin Film Photovoltaics
1. 7.1 Introduction
2. 7.2 General Substrate Requirements for Photovoltaic Applications
3. 7.3 Requirements for CdTe Superstrates
4. 7.4 Standard CdTe Device Stack and Processing
5. 7.5 Flexible CdTe Device Performance
6. 7.6 Flex and Bend Testing of CdTe
7. 7.7 Future Trends/Directions
8. References
Chapter 8: Ultra-Thin Glass for Displays, Lighting and Touch Sensors
1. 8.1 Introduction and Overview
2. 8.2 Ultra Thin Glass Substrates for Flexible Displays
3. 8.3 Thin film Device Processing on Ultra Thin Glass
4. 8.4 Thin Glass Displays
5. References
Chapter 9: Guided-Wave Photonics in Flexible Glass
1. 9.1 Flexible Guided-Wave Photonics
2. 9.2 Flexible Polymer Passive Waveguide Photonics
3. 9.3 Flexible Polymer Active Waveguide Photonics
4. 9.4 Flexible Polymer Waveguides for Electro-Optic Applications
5. 9.5 Flexible Glass Optical Substrates
6. 9.6 Ultrafast-Laser Fabrication of Embedded Waveguides
7. 9.7 Embedded Waveguides in Flexible Glass
8. 9.8 Prospective of Thermal Poling in Flexible Glass Waveguides
9. 9.9 Summary and Future
10. References
Chapter 10: Flexible Glass for Microelectronics Integration
1. 10.1 Introduction
2. 10.2 Integration Technology Description: Why Flexible Glass for Electronics/Sensor Integration (3 Dimensional Integrated Circuits – 3DIC)
3. 10.3 Example of Microelectronics/Sensor Integration
4. 10.4 Fabrication Techniques
5. 10.5 Future Direction
6. References

Index

End User License Agreement

List of Illustrations

Chapter 1

Figure 1.1 Optical transmission of glass substrates of differing thicknesses in the (a) UV to near-IR and (b) UV spectrum. Note that the data was smoothed to reduce significant optical interference fringes in the thinner glass substrates.
Figure 1.2 Optical transmission of glass substrates of differing thicknesses in the IR spectrum.
Figure 1.3 (a) Refractive index of Willow Glass and polymer film substrates in the UV to near-IR, and (b) Willow Glass in the IR spectrum.
Figure 1.4 Optical haze of flexible glass and polymer film substrates. (Error bars are standard deviation.)
Figure 1.5 Color of flexible glass and polymer film substrates.
Figure 1.6 Optical measurements of Willow Glass and polymer film substrates before and after extended UV exposure. (a) Optical transmission at 550 nm and (b) Color (L*).
Figure 1.7 Surface roughness of flexible glass and representative substrates used in flexible electronics.
Figure 1.8 Surface energy of flexible glass and reference flexible substrates.
Figure 1.9 Young’s modulus and hardness of flexible glass and representative flexible substrates.
Figure 1.10 Relative flexural rigidity of glass and representative materials.
Figure 1.11 Failure strength distributions of Willow Glass samples cut with mechanical scribe and break (squares) and CO₂ laser crack propagation process (circles) measured by 2-point bend.

Chapter 2

Figure 2.1 Reliability “bathtub” diagram showing stages to product life [1].
Figure 2.2 Subcritical crack growth in window glass in air and under a constant load [2].
Figure 2.3 Michalske and Freiman’s model for the interaction between water and strained silica bonds [3].
Figure 2.4 Representation of water-induced bond rupture in silica glass [4].
Figure 2.5 Crack velocity – stress intensity factor diagram illustrating the regions of subcritical crack growth in glass. The stress intensity factor K_l = Yσ_a, where Y is the crack shape parameter, σ_a is the applied stress and a is the crack depth.
Figure 2.6 Strength degradation while under a static load.
Figure 2.7 Static fatigue test results of abraded soda-lime silicate microscope slides [13]. The applied stress is normalized by the strength in liquid nitrogen, σ_n, and time to failure is represented by the measured value divided by t_0.5 = t(σ/σ_n = 0.5), where t_0.5 = is the time-to-failure for a given abrasion condition where the applied stress is 50% of the liquid nitrogen strength.
Figure 2.8 Strength degradation during dynamic loading conditions and the effect of stressing rate on measured strength in a fatigue environment.
Figure 2.9 Effect of loading rate on the strength of an abraded boro-aluminosilicate glass at room temperature.
Figure 2.10 Ingredients for establishing the mechanical reliability of glass.
Figure 2.11 Required initial (inert) strength for range of applied stresses and duration under stress. An n value of 20 is assumed. “Inert” refers to the experimental condition of strength testing in an environment where no fatigue takes place during the test. Testing in liquid nitrogen has been used extensively [14].
Figure 2.12 A design diagram for thin glass sheets in bending. The required strength for a given bend condition is given for three stress durations.
Figure 2.13 Schematic of mechanical proof testing of glass. The proof testing of glass is a process intended to eliminate flaws weak enough to fail while the glass is in-service.
Figure 2.14 Allowable stress as a function of proof stress for several common stress events.
Figure 2.15 Failure probability design for initial (inert) strength of S₀ = 200 MPa and m = 5.
Figure 2.16 Design diagram for various fiber lengths subjected to bending. 125 μm silica-clad fiber proof tested to 700 MPa.
Figure 2.17 Stresses during roller conveyance of flexible glass web.
Figure 2.18 Bend-induced stress in flexible glass. Young’s modulus was taken to be 73 GPa and Poisson’s ratio equal to 0.23.
Figure 2.19 Tensile testing flawless silica fibers in air, vacuum, and at low temperature, from Proctor et al., [19].
Figure 2.20 Crack systems resulting from mechanical contact: (a) shows crack systems resulting from sharp contact [29] and (b) shows Hertzian cracks generated by contact with a blunt object [4]. Note that Hertzian cracks can also form from plastic zones created by sharp contact.
Figure 2.21 The ball-on-clamped-ring test configuration. The thin glass sheet in clamped about its perimeter and a ball is pressed onto the glass in the center of the test area [32].
Figure 2.22 The test area is determined by the contact region of the ball.
Figure 2.23 The stress beneath the ball in the ball-on-clamped-ring test as a function of applied load. The glass initially wraps around the ball under minimal membrane stress and then is loaded to failure with increasing membrane stress.
Figure 2.24 The strength of thin glass sheets as measured by the ball-on-clamped-ring method. Similar strengths were found for the three thicknesses studied. The median strengths of near 1 GPa are indicative of submicron flaw depths.
Figure 2.25 Two point bend test method. The glass specimen is long enough to be parallel to both the top and bottom platens.
Figure 2.26 The edge strength of 100 μm thick glass sheets tested in two-point bending in ambient conditions. The edges were created by the mechanical score and break method. Failure from surface damage is treated as suspended data.

Chapter 3

Figure 3.1 ²⁷Al MAS NMR of Glasses I-V. The dotted lines are reference to the location of Al (IV) and Al(V) peaks. The NMR data demonstrates that the coordination state is primarily tetrahedral for aluminum cations.
Figure 3.2 ¹¹B MAS NMR of Glasses I-V. The NMR data shows that the boron is primarily in the trigonal coordination state, but does contain a small fraction of tetrahedral coordinated boron that increases as the boron content in the glasses increases.
Figure 3.3 Measured modulus vs. calculated modulus for glasses I-V.
Figure 3.4 Plot of modulus vs. dissociation energy, packing density. Dissociation energy and packing density were calculated using boron speciation from NMR. Across the glass composition series studied, the increase Young’s modulus is shown to be highly dependent on the increase in dissociation energy.
Figure 3.5 Bend induced tensile stress vs. the bend radius for 100 micron thick glasses with different Young’s moduli.
Figure 3.6 Stress intensity vs. bend radius for glasses with different moduli and flaw size at 100 micron thickness.
Figure 3.7 Measured Young’s modulus vs. fictive temperature for glasses I, III and V. Glass I with highest B₂O₃, lowest SiO₂ has largest slope of Young’s modulus with fictive temperature.
Figure 3.8 The bend induced tensile stresses vs. bend radius for Glass I at 100 micron thickness at various fictive temperatures. The highest fictive temperature corresponds to the lowest Young’s modulus and lowest bend induced tensile stress for a given bend radius.
Figure 3.9 Stress intensity vs. bend radius for Glass I at different fictive temperatures for 1 and 10 micron flaw sizes. The shape factor used corresponds to a surface scratch type crack, Y = 1.12.
Figure 3.10 Characteristic indents made at 500, 1000, and 2000 gf in the endpoint glasses, Glasses I and V, with T_f = anneal point.
Figure 3.11 NMR comparison of Glass I heat treated so T_f = 664 °C (anneal point temperature) vs. T_f = 622 °C (strain point temperature).
Figure 3.12 NMR comparison of Glass V heat treated so T_f = 793 °C (anneal point temperature) vs. T_f = 741 °C (strain point temperature).
Figure 3.13 Schematic of indentation cracking threshold for tips of increasing sharpness. In the extremes of sharpness, the cube corner (C.C.) indenter eliminates any compositional advantage to crack resistance.
Figure 3.14 10 gf cube corner indents in (a) Glass I and (b) Glass V show that in the extremes of sharp contact, the advantage of inherent damage resistance is negated.

Chapter 4

Figure 4.1 CHA chamber design and web path. The web can be run in forward and reverse directions.
Figure 4.2 Cross web uniformity of 22 inch wide Si deposition in the CHA vacuum deposition tool.
Figure 4.3 CHA tool with the vacuum chamber shown open, the image is of the web handling side.
Figure 4.4 Example roll map and process recipe for the CHA roll-to-roll sputter tool.
Figure 4.5 GVE full contact web path with process zones labeled.
Figure 4.6 The inside of the GVE chamber prior to installation at CAMM site.
Figure 4.7 GVE Si rotatable targets. Target composition is 2%Al/98%Si. The targets are 770 mm long with an outer diameter of 151 mm.
Figure 4.8 O₂% vs target voltage hysteresis curve for CHA intrinsic silicon target.
Figure 4.9 CAMM Hollmuller Siegmund wet processing tool.
Figure 4.10 Oscillating sprays in the Hollmuller Siegmund wet processing tool. Sprays are located on the top and bottom of the web to allow for two sided processing.
Figure 4.11 Azores 600 roll-to-roll step and repeat photolithography tool outfitted with Northfield roll handlers.
Figure 4.12 Simplified image of the Azores stage and handlers system. Stage and handlers move in concert in the X and Y directions while the optics remain fixed.
Figure 4.13 Image of the Azores projection system through the mask. Masks need to be fabricated right reading chrome side up to ensure proper orientation of the final image onto the substrate [25].
Figure 4.14 Bulk resistivity and transmission measurements as a function of O₂%. Transmission values are of the PEN substrate and ITO film stack at ambient temperature measured at 550 nm.
Figure 4.15 Bulk resistivity and transmission measurements as a function of deposition pressure (at 2% O₂) on PEN at ambient temperature. Transmission measurements are of substrate and ITO film stack at 550 nm.
Figure 4.16 Bulk resistivity and transmission measurements as a function of O₂ partial pressure for annealed samples. Transmission measurements of the coated films were taken at 550 nm.
Figure 4.17 Bulk resistivity and transmission measurements as a function of deposition pressure for annealed samples. Transmission measurements of the coated substrates were taken at 550 nm.
Figure 4.18 X-ray diffraction measurement of PEN substrate, 0.80 Pa, 2.0% O₂ sample and 90 minute annealed 0.80 Pa, 2.0% O₂ sample.
Figure 4.19 Patterned ITO thin films on flexible glass web (a) 50 nm thick, 120 Ω/sq ITO film with 30 μm lines with 1 mm spacing; 150 nm thick 30 Ω/sq ITO film with (b) 10 μm lines and spaces and (c) 5–30 μm lines and spaces.
Figure 4.20 Transmission measurements of Willow Glass, PEN and PET films uncoated and with ITO coatings.
Figure 4.21 X-ray diffraction measurements of ITO depositions on PEN, PET and Willow Glass.
Figure 4.22 SEM images of (a) ITO deposition on Willow Glass. (b) ITO deposition on Teonex Q65FA PEN. (c) ITO deposition on Melinex ST505 PET.
Figure 4.23 AFM images of (a) Uncoated Willow Glass, (b) ITO deposition on Willow Glass.
Figure 4.24 Comparison of sheet resistance of ITO coatings on Willow Glass and Teonex Q65FA PEN in an 85% relative humidity and 85 °C environment.

Chapter 5

Figure 5.1 Cross-section scheme of an inverse sputter etching tool with gutter shaped extraction grid for generation of a divergent ion-(plasma) beam.
Figure 5.2 Cross-section scheme of a linear ion source (LIS) for generation of a divergent ion-(plasma) beam.
Figure 5.3 Overview of commonly used vacuum coating technologies for thin film coatings in a dynamic sheet-to-sheet and roll-to-roll manufacturing processes.
Figure 5.4 Material feed for different evaporation processes. (a) Set of ceramic boats for thermal evaporation (b) Crucible filled with material for electron beam evaporation.
Figure 5.5 Hollow-cathode plasma source for plasma-assisted evaporation: left image: operation principle of a hollow-cathode plasma source, right image: extract of a set of 16 hollow-cathode mounted on a flange for large-area plasma assisted evaporation of Al₂O₃ in a roll-to-roll machine.
Figure 5.6 (a) Setup and components of a rectangular magnetron and (b) setup of a dual magnetron system for bipolar pulsed reactive sputtering.
Figure 5.7 Overview about the various sputtering modes [19].
Figure 5.8 Various pulse magnetron sputtering modes applied for the deposition of dielectric layers. The pulse packet technology is a combination of unipolar and bipolar pulse sputtering[19].
Figure 5.9 (a) Plasma discharge of a rotatable dual magnetron system and (b) cylindrical ZnO:Al targets with a length of 820 mm.
Figure 5.10 Overview of typically used sputtering modes and target types for various layer materials.
Figure 5.11 Examples of different PECVD plasma sources and deposition techniques. (a) HF capacitive PECVD source; (b) Linear microwave PECVD; (c) Dual-magnetron PECVD; (d) Hollow-cathode PECVD.
Figure 5.12 Steps of one ALD cycle.
Figure 5.13 Overview about spatial ALD concepts. (a) Showerhead spatial ALD according to Maas et al. [60] Scheme taken from Poodt et al. [61]. (b) Roll-to-Roll ALD setup according to Poodt et al. [62]; (c) Roll-to-Roll ALD setup from Beneq Oy (Maydannik et al.) [63].
Figure 5.14 Scheme of in-line coating tool for sheet-to-sheet processing.
Figure 5.15 Pilot scale in-line coating machine ILA 900 with vertical substrate handling for sheet-to sheet thin-film and process development. The maximum target length is 900 mm. This coating tool can also be used for flexible glass.
Figure 5.16 Principle of roll-to-roll vacuum coating machines.
Figure 5.17 Example of a versatile pilot-scale roll-to-roll vacuum coater with two process drums (novoFlex^® 600 by Fraunhofer FEP, Dresden, Germany).
Figure 5.18 Principle of an optical inline spectrometer for vacuum coating machines with substrate movement and more than one measurement spot for transmission and reflection over the substrate width.
Figure 5.19 Important components of an optical in-line monitor. (a) Example for an optical vacuum feed-through, (b) Reflection measurement probes.
Figure 5.20 Requirements on the sheet resistance for non-annealed ITO coatings for different applications on flexible substrates.
Figure 5.21 Transmittance spectra of ITO thin films deposited by DC sputtering on 3 mm thick low iron glass and 100 μm thick flexible glass prepared at room temperature (RT) and 400 °C.
Figure 5.22 DC Sputtered ITO thin films with a thickness of 100 nm (left side) and 1000 nm (right side) deposited on flexible glass.
Figure 5.23 Influence of the used sputtering process pressure for 1000 nm thick ITO films on the curvature of 100 μm thick flexible glasses and the mechanical film stress.
Figure 5.24 Mechanical film stress for 400 nm thick ITO films coated at different sputtering process pressures and power impact on the target.
Figure 5.25 Scheme of typical broadband AR layer stack based on four layers.
Figure 5.26 Multi-layer stack based on Silica and Titania on flexible glass (left side, substrate size 250 × 300 mm²) and PET film (right side, substrate width 200 mm). Both substrates have the same thickness of 100 μm. Note; the maximum temperature during the deposition was lower than 50 °C. The Young’s modulus for flexible glass is 70 GPa, for PET 3700 MPa.
Figure 5.27 Mechanical film stress for titania thin films in dependence on an additional oxygen gas flow during sputtering for adjustment of the stoichiometry for different used modes of powering.
Figure 5.28 Left side – single side AR coatings on flexible glass. The mode of powering for the titania films in this layer stack were varied. Right side – measurement of the bending for 130 μm thick glass stripes coated with AR layer stack.
Figure 5.29 Spectral distribution of reflectance for four-layer antireflection coatings on flexible glass deposited at different powering modes for the sputtering process of Titania films. Only one side is coated. The backside of the substrate was noncoated.
Figure 5.30 Scheme of temperature profile for heat impact using ultra-fast annealing on surfaces.
Figure 5.31 Sheet resistances (left side) for 560 nm thick ITO films on 3 mm thick low iron float glass and 100 μm thick flexible glass in as-deposited state and after annealing in vacuum and by FLA. The right side shows the transmittance spectra and calculated absorbance spectra in the visible range for the coatings on flexible glass. The films were annealed in vacuum at 400 °C for 15 min and by FLA with a used pulse time for the flash of 2 ms.
Figure 5.32 Influence of energy density and pulse duration parameters for a flash lamp annealing process on the temperature depth profile and the thermal penetration depth of thin films or substrates [86].
Figure 5.33 Overview of needed energy densities and pulse duration times for ultra-fast flash lamp treatment of materials/ thin films usually used in optoelectronic devices [86].
Figure 5.34 Influence of different used energy densities for FLA on the sheet resistance of ITO films coated on flexible glass. The reference sample is the ITO film in as-deposited state. The pulse duration time for the flash was fixed on 2 ms.

Chapter 6

Figure 6.1 Magnified image of typical graphics print.
Figure 6.2 ZnO formulations for gravure printing process, (a) original ZnO dispersion and (b) formulated with better edge formation and homogeneity [9].
Figure 6.3 Flexographic printing process.
Figure 6.4 An anilox cylinder and engraved surface of the roll.
Figure 6.5 Flexography printing plate.
Figure 6.6 Gravure printing process.
Figure 6.7 A gravure cylinder and engraved cells.
Figure 6.8 (a) principle of the reverse offset printing process, (b) An example of micro structured metal grid formed by revere offset printing.
Figure 6.9 Principle of gravure offset printing process [42].
Figure 6.10 Example of gravure offset printed 23-inch touch sensor pattern with silver metal on flexible glass [43].
Figure 6.11 Screen printing process.
Figure 6.12 Mesh screen with stencil.
Figure 6.13 Principle of the rotary screen printing.
Figure 6.14 A rotary screen cylinder.
Figure 6.15 Operation principle of continuous-stream inkjet.
Figure 6.16 Operation principle of thermal inkjet.
Figure 6.17 Roll-to-roll pilot line at VTT used for OPV cell printing on the ultra slim glass.
Figure 6.18 (a) Printed OPV cells on the ultra slim flexible glass. (b) JV-curves of printed OPV cells on flexible glass.
Figure 6.19 Hybrid integration process (a) printing, (b) assembly (c) foil over-moulding.
Figure 6.20 Examples of hybrid integration. (a) large area flexible LED luminaire and (b) injection moulded flexible glass with patterned ITO coating in an elastomer material.

Chapter 7

Figure 7.1 Example configurations (superstrate/substrate) for different PV technologies. (Top) CdTe, OPV, and perovskites are grown in superstrate geometries. The growth is on a transparent electrode on a transparent growth superstrate. (Bottom) CIGS, CZTS, and a:Si are grown in a substrate geometry, which does not require a transparent growth substrate.
Figure 7.2 (a) Optical transmission curves for various materials – soda lime glass is the most common frontsheet in use today. (b) Spectral response curves for different PV technologies.
Figure 7.3 Water vapor transmission rate relative to accumulated volume/layer thickness over the course of various times. Comparing an accumulated layer thickness/volume of water to an expected operational lifetime and a device’s constituent layers helps explain WVTR requirements.
Figure 7.4 (Top) Specific power relative to efficiency for products in 2015. This highlights that the critically important role of package weight on specific power. (Bottom) Hypothetical packaging examples to achieve different module areal densities.
Figure 7.5 Two different layer stacks explored by the CdTe community. Commercial modules are 17% efficient now and may use different layers.
Figure 7.6 Examples of a 2-point (left) and 4-point (right) bend test apparatus. In the case of the 2-point bend test, there is a linearly increasing moment between the bend points. In the case of the 4-point bend test, there is a constant moment between the central two bend points.

Chapter 8

Figure 8.1 Various P-Cap touch display realizations.
Figure 8.2 Surface topology of thin glass, PES and metal foil [28] on an area of 20 μm × 20 μm.
Figure 8.3 Definition of parameters that define the planarity of display substrates.
Figure 8.4 Barrier Properties of flexible substrates and requirements of different display techologies.
Figure 8.5 Layer stack on the active matrix backplane substrate [50].
Figure 8.6 Layer stack on the color filter frontplane substrate [51].
Figure 8.7 Reduction of layer stress in sputtered molybdenum alloy. Left: standard process, Right: optimized parameters [51].
Figure 8.8 Layer stress respectively layer curvature as a function of molybdenum alloy sputter power (100% corresponds to 5.5 KW on a 488 mm × 88 mm target) upon 10 × 12 cm² substrate size.
Figure 8.9 Compressive stress in a PECVD stack on thin glass.
Figure 8.10 Microscopic images of the ultra-thin glass substrates with fully processed active [50] matrix backplane (left) and color filter (right).
Figure 8.11 TN LCD test cells with a size of 10 cm × 12 cm before (a,b) and after (c,d) cell assembly process optimizations (top crossed polarizers, bottom: parallel polarizers) [51].
Figure 8.12 Typical switching properties of a 10 cm × 12 cm TN test cell on flexible ultra thin glass (V_p = 0 … 5 V; f = 1 kHz, rectangular pulse).
Figure 8.13 Comparision of the viewing angle characteristics of ultra thin glass (left) and standard glass (right) liquid crytal cells.
Figure 8.14 Alignment tolerance of frontplane and active matrix backplane substrates of a 4 inch AMLCD on ultra-thin glass substrates.
Figure 8.15 AMLCD on ultra-thin glass with bonded TAB driver chips [52].
Figure 8.16 Transfer characteristics of nine test TFTs manufactured on 75 μm flexible glass and on a standard 0.7 mm glass [51].
Figure 8.17 Typical output characteristic of test TFTs on 75 μm thin glass and 0.7 mm standard glass.
Figure 8.18 Thin glass 4 inch AMLCD Demonstrator [50, 51].

Chapter 9

Figure 9.1 (a) Schematic of a polymer waveguide fabrication process, (b) Microscopic cross-section view of a 50 × 50 mm² waveguide, and (c) a 4 × 12 waveguide array [8].
Figure 9.2 Schematic of a 10-channel board-to-board optical link developed by IBM, and photograph [8].
Figure 9.3 (a) Schematic view of flexible waveguide fabrication process, (b) photograph of waveguide after the handler substrate removal, (c) example of ChG glass with different composition, and (d) flexible waveguide bending test (after [12]).
Figure 9.4 (a) chemical structure of QB-Er complex ligand, (b) schematic sketch of PMMA active waveguide structures. The SEM image and guided mode profile at 1540 nm is shown in (c).
Figure 9.5 (a) SEM photograph of polymer waveguides (SU8+ nanocrystals) on the top of oxidized silicon wafers, (b) the gain of polymer waveguide with different Er³+ and Ce³+ concentration(after [17]).
Figure 9.6 Chemical structures of EO polymer mixtures including (a) PMMA host, (b) AJC146 chromophore, and (c) BMI cross-linker. (d) schematic illustration of blends of three chemicals before and after thermal poling and lattice hardening [11].
Figure 9.7 (a) Optical transmission of flexible glass and other polymeric materials [20] and (b) Surface roughness of flexible glass, PEN, and Polyimide.
Figure 9.8 Schematic sketch and photograph of ultrafast laser direct writing technique.
Figure 9.9 Microscope cross-section and near-field guided mode images of optical waveguides fabricated by the ultrafast laser in (a) sapphire, (b) LiTaO₃, and (c) Chalcogenide glass (Gallium Lanthanum Sulphide).
Figure 9.10 Optical Microscope photo of (a) Waveguides written in GLS glasses by 240-fs pulses at 400-nJ on the left and 200-nJ on the right, and (b) waveguide fabricated in flexible glasses using 300-nJ pulse.
Figure 9.11 Schematic of ultrafast laser writing setup using a telescope laser beam shaping tool.
Figure 9.12 (a) Schematic diagram of waveguide writing using an amplitude mask. Computer simulations of energy distributions around the focal point produced by (a) a symmetric beam and (b) an aperture with aspect ratio of 6. The black scale bar is the laser wavelength (800 nm), f was set as 0.4 (20x objective), the aperture was set as 0.5 mm × 3 mm.
Figure 9.13 Schematic and photograph of the experiment setup for sample characterization.
Figure 9.14 (a) Waveguide formed using an 80X aberration-corrected objective under ‘perfect focusing’ (Top: image of the waveguide cross section; Bottom: Guided mode profile at 1550 nm); (b) Waveguide formed by spatially shaped laser beam to introduce strong astigmatism to control the focal volume (Top: image of the waveguide cross section; Bottom: Guided mode profile at 1550 nm); (c) From top to bottom, waveguides formed in 100 μm (top), 50 μm, 35 μm, and 25 μm (bottom) flexible glass substrates [36].
Figure 9.15 MFD as functions of pulse energy and laser writing speed in 100 mm glass.
Figure 9.16 (a) Propagation loss measurement by OFDR. Propagation loss of a waveguide in an 11.4 cm-long flexible glass substrate is indicated by the slope of the fitted data between the two facets. (b) Waveguide propagation losses measured on 10 waveguides (WG) written using the same processing conditions as in (a) [36].
Figure 9.17 Flexible glass waveguides written in (a) 100 μm thick substrate with 13.5 cm bending radius; (b) 50 μm thick substrate with 2.1 cm bending radius. (c) 35 μm thick substrate with 1.0 cm bending radius; (d) The same 35 μm thick substrate with 1.0 cm bending radius [36].
Figure 9.18 Insertion loss of the ultrafast-lase-written waveguide in 25 μm, 35 μm, 50 μm, and 100 μm thick flexible glass substrates with bending radius down to 0.67 cm [36].
Figure 9.19 Mode profiles (top) and microscope images (bottom) of waveguides written with (a) 1000-nJ and (b) 1160-nJ laser pulses, after 1 hour baking for each temperature. (c) Waveguide attenuation for each temperature, written with 1000-nJ pulses (d) Waveguide attenuation for each temperature, written with 1160-nJ pulses [36].
Figure 9.20 (a) Optical microscope photographs (a) topview of a 3rd-order Bragg grating waveguide and (b) end view of Bragg grating waveguide in 50 μm thick glass.
Figure 9.21 (a) Experimental setup for Bragg grating waveguide characterization, and (b) a typical reflection and transmission spectra of a 3rd order Bragg grating waveguide with 25% duty cycle.
Figure 9.22 (a) Top view of ultrafast laser written Bragg grating waveguides in flexible glass using modulation with 25% duty cycle. (b) Bragg grating reflection spectra after two heating cycles [36].
Figure 9.23 (a) Schematic of experimental setup for BGW birefringence measurement. (b) Bragg grating reflection spectra for two perpendicular polarized light.
Figure 9.24 Sketch of thermal poling process.
Figure 9.25 Waveguide formed ~15-mm under the surface. The dark zone between the surface & waveguide is laser-induced voids.
Figure 9.26 Distribution of SH light in fs-written glass. (a) and (d) are ordinary transmission images, (b) and (e) are SH images, and (c) and (f) are overlay images of ordinary transmission and SH images.
Figure 9.27 Illustration of thermal poling from backside in 50 mm flexible glass.

Chapter 10

Figure 10.1 An example process flow that was used to generate the solar cells using a combination of MEMS and IC processing steps on a (111) oriented Si wafer using KOH etching. Junctions and metallization layers are formed using standard IC processing steps. Sidewalls of the micro-scale solar cells are passivated by a nitride layer which also protects them from KOH etching. Cells are released by the KOH etchant which selectively etches the desired crystal plane.
Figure 10.2 Silicon-on-insulator (SOI) wafer based process flow for fabrication of micro-scale solar cells. The cells are tethered to the substrate at the end of the process and transferred to a flexible receiving substrate.
Figure 10.3 Photomicrograph of MEPV silicon cells with electroplated pillars, after the SOI based process flow.
Figure 10.4 SEM of MEPV silicon cells made with the SOI process flow, with electroplated Cu and solder pillars. Portion of the suspended frame is visible around the cell.
Figure 10.5 150 mm diameter, wafer format Corning^® Willow^® Glass receiver substrate processed in standard microfabrication tools for creating the series-parallel circuits for the flexible demo.
Figure 10.6 Transfer of cells from wafer to flexible substrate with solder reflow and mechanical detachment.
Figure 10.7 Singulated Willow Glass substrate and MEPV silicon solar cells. This circuit has 476 cells (34 in parallel and 14 in series).
Figure 10.8 Flexible demo system fabricated on polyimide and copper receiver circuit.
Figure 10.9 Flexible demo system (polyimide) showing 1 mm bend radius.
Figure 10.10 Single layer of the sensor array – red squares are 1 mm × 1 mm individual sensors.
Figure 10.11 Ceramic receiver for 4-plate assembly. Pads along the ledges are connected to larger vias at the periphery of the ceramic receiver, enabling stacking of elements.
Figure 10.12 Ten stacked receivers, providing a total volume of 35 mm × 35 mm × 12.5 mm for 10 × 10 × 40 sensor array. Each 1 mm sensor would have multiple pixels, further increasing sensor resolution and functionality.
Figure 10.13 Example of a cluster tool used in CMOS fabrication, with automated wafer handling and transport capabilities. (image courtesy of Applied Materials).
Figure 10.14 PECVD deposition system used in flat panel manufacturing, capable of handling very large glass substrates. (image courtesy of Applied Materials).
Figure 10.15 Section of a wafered silicon solar cell fabrication line, capable of very high throughput levels. (image courtesy of Meyer Burger).
Figure 10.16 A roll-to-roll manufacturing setup, with customizable sections. (image courtesy of Coatema).

List of Tables

Chapter 1

Table 1.1 Reference flexible substrate materials used for comparison purposes to 100 µm-thick Willow Glass.
Table 1.2 Flexible glass thermal properties. Attributes from both Willow Glass and 0211 Microsheet are shown to highlight variations due to composition.
Table 1.3 Dielectric constant (D_k) and loss tangent (D_f) for Willow Glass and representative polymer films.
Table 1.4 Density of flexible glass and representative flexible substrates.

Chapter 2

Table 2.1 The allowable stress as a function of initial, S, and measured strength in a fatigue environment, a. An n value of 20 is assumed.

Chapter 3

Table 3.1 Cationic radii, packing factors, and dissociation energies of oxides [1–2, 4].
Table 3.2 Calcium aluminoborosilicate glass compositions and properties.
Table 3.3 Vickers indentation cracking threshold results.

Chapter 4

Table 4.1 Bend radius possible for 100 MPa bend stress for different glass thicknesses [10].
Table 4.2 Performance and operating specifications for Azores 6600 R2R photolithography system [25].
Table 4.3 As deposited properties of the ITO thin films at ambient temperature on PEN.
Table 4.4 Comparison of bulk resistivity and figure of merit values for as deposited and 90-minute annealed samples of ITO deposited on PEN. Figure of merit was calculated using the method proposed by Haacke in [43].
Table 4.5 ITO properties deposited on PEN following a 90 minute anneal at 110 °C.
Table 4.6 Summary of XRD data collected for samples as deposited and following 90 minute anneal times on PEN substrates.
Table 4.7 Summary of etch rates for ITO deposited at ambient temperatures. All etching was performed at room temperature, except where stated otherwise.
Table 4.8 Summary of sheet resistance, bulk resistivity, transmission and figure of merit for ITO films deposited on Willow Glass, PEN and PET samples

Chapter 5

Table 5.1 Maximum temperature (in °C) of the substrates during 500 nm TiO₂ deposition [25].
Table 5.2 Properties of 500 nm thick TiO₂ layers deposited onto unheated glass [25].
Table 5.3 Selection of layer materials, precursors and application for Atomic Layer Deposition.
Table 5.4 Properties of ITO films deposited by DC sputtering on 3 mm thick low iron glasses and on 100 µm thick flexible glasses. The transmittance T_vis. is defined as the integrated brightness in the visible range of the spectrum (380 nm to 780 nm) related to a light source similar to daylight (D 65) corresponding to an angle of 2 degrees by a standard observer. The extinction coefficient k and the refractive index are given for 550 nm wavelength. T_L*, T_a* and T_b* are the color values for the transmittance.
Table 5.5 Mechanical film stresses for the single layers of the AR layer stack and the whole AR layer stack prepared at different modes of powering for the sputtering process of titania:
Table 5.6 Technologies for ultra-fast thermal post annealing:

Chapter 6

Table 6.1 Properties of different printing methods.
Table 6.2 Advantages and disadvantages of different printing methods.

Chapter 7

Table 7.1 Examples of flexible from various thin-film PV technologies and their temperature, moisture, and chemical requirements. The growth architecture typically dictates whether or not a transparent substrate is required. The list of examples for each technology is meant to be illustrative rather than exhaustive.
Table 7.2 Water vapor transmission rates for various materials.
Table 7.3 T_s and CTE for superstrate materials.

Chapter 8

Table 8.1 Qualitative Comparision of Lighting Technologies [6].
Table 8.2 Thermal Parameters of various substrate materials.
Table 8.3 Required substrate characteristics.
Table 8.4 Comparison of switching times for thin glass and reference liquid crystal cells.
Table 8.5 Comparison of TFT parameters on thin and standard glass [52].
Table 8.6 Characteristics of thin glass AMLCD Demonstrator.

Chapter 9

Table 9.1 Key Properties of Optical Polymers for Waveguide Fabrications [6].
Table 9.2 Comparison between polymer flexible photonics and glass flexible photonics technology

Scrivener Publishing
100 Cummings Center, Suite 541J
Beverly, MA 01915-6106

Advances in Roll-to-Roll Vacuum Coatings Technology

Often new technologies, processes or materials suddenly appear that attract some publicity. It is not always easy to find reviews of these advances that allow the reader to compare and contrast the different technologies. This series of books aims at providing a source of information that will enable the reader to obtain an overview of groups of recent advances in technologies, processes or materials.

Series Editor: Charles A. Bishop
E-mail: cabuk8@btinternet.com

Publishers at Scrivener
Martin Scrivener (martin@scrivenerpublishing.com)
Phillip Carmical (pcarmical@scrivenerpublishing.com)

This edition first published 2017 by John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA and Scrivener Publishing LLC, 100 Cummings Center, Suite 541J, Beverly, MA 01915, USA
© 2017 Scrivener Publishing LLC
For more information about Scrivener publications please visit www.scrivenerpublishing.com.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.

Wiley Global Headquarters
111 River Street, Hoboken, NJ 07030, USA

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.

Limit of Liability/Disclaimer of Warranty
While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials, or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read.

Library of Congress Cataloging-in-Publication Data
ISBN 978-1-118-94636-7

Foreword by Peter L. Bocko

Technological revolutions are often built upon a foundation of self-delusion and naiveté. That bleak statement requires some explanation. While a scientific revolution can be nucleated by an individual’s insight, the delivery of a breakthrough technology requires a shared vision of the innovation’s benefit followed by broad and protracted collaboration among materials, process, systems, device and application specialists. And if these collaborators realized at the outset the level of resolve and resources ultimately required to deliver a revolutionary technological platform, few would get off the ground.

Fortunately, a revolution in electronics based upon flexible glass has progressed well beyond initial naiveté and subsequent (and periodic) stages of disillusionment. This book is a major milepost in this platform’s development, documenting over a decade of hard won advances in the flexible glass platform through collaboration across relevant component technologies and applications. As an early promoter, champion and sponsor for the applications of flexible glass, I am excited that the building blocks for broad innovation have achieved critical mass, and for the first time are accessible in one place.

Glass has a capability of being drawn under heat and tension into a film of arbitrary thickness while retaining its desirable surface, mechanical and optical properties. This is simple and intuitive. After explaining to a customer engineer the process of drawing molten glass into precise sheet for LCDs, he asked “How do you make it thinner?”. “Pull harder.” I answered. Since glass-sheet manufacture has been automated, processes have been pushed to draw glass to the limits of sufficient thinness to achieve flexibility, motivated by the desire to minimize weight, enhance conformability or to enable in-line processing.

While the forming of precise ultra-thin glass has been established across multiple glass manufacturing platforms over the last 20 years, what has been missing were the constellation of enabling component technologies: packaging, handling, deposition, patterning and device design that can be used to transform flexible glass from the glass maker’s forming tool and adapt it to a functional system. This has resulted in skepticism and resistance of the electronics industry for commitment to large scale development of flexible glass platforms.

Things have changed since then, but I expect that it will still take time and much hard work to drive flexible glass to the high-volume applications that fully leverages its potential. The editor of this work as well as chapter author, my erstwhile colleague from Corning, Dr. Sean Garner, is in large part responsible for promoting flexible glass in the technology community and structuring the collaborations that have brought us to the verge of breakthrough of flexible glass into enabling advanced electronic applications. This book represents a major contribution to the field. The long-incubated flexible glass revolution is upon us.

Peter L. Bocko
Adjunct Professor of Materials Science & Engineering,
Cornell University
Former Chief Technology Officer, Corning Glass
Technologies, Corning Incorporated

Preface

Flexible glass continues to emerge as a significant material component for electronic and opto-electronic applications. Its use goes well beyond earlier capacitor applications. For example, new opportunities in fields of displays, sensors, lighting, backplanes, circuit boards, photonic substrates, and photovoltaics continue to be identified. This is much more than just transitioning the devices that exist currently on thicker rigid glass onto a thinner, flexible substrate. Flexible glass substrates in these applications enable new device designs, manufacturing processes, and performance levels not possible or practical with alternative substrate materials and may include electronic applications such as fully-integrated, large-area, smart surfaces. In addition, these new applications require specifically optimized fabrication processes, manufacturing equipment, and device designs that take advantage of the unique properties of flexible glass.

Although there have been previous discussions of flexible glass substrates and devices at conferences and in published journals, they have focused on very specific aspects or applications. This book, however, provides a much broader overview as well as detailed descriptions that cover flexible glass properties, device fabrication methods, and emerging applications. This book is not meant to provide a comprehensive, detailed description of all attributes and possibilities but rather, it provides the basis for identifying new device designs, applications, and manufacturing processes for which flexible glass substrates are uniquely suited. Information in this book encourages and enables the reader to identify and pursue advanced flexible glass applications that do not exist today and provides a launching point for exciting future directions.

Information in this book is based on over 10 years of valuable discussions and collaborations focused on truly defining what flexible glass means in the context of these emerging electronic and opto-electronic applications. This learning is also built upon decades of previous activities in earlier applications. What started personally for me as an “exploratory investigation” has occupied most of my career as I collaborated on various aspects of flexible glass’ definition, processing, and applications. The chapters included here are from some of my more significant collaborations meant to provide an overall, well-rounded perspective.

The chapters are grouped into three sections. The first focuses on flexible glass and flexible glass reliability and has three chapters with authors from Corning. The second section focuses on flexible glass device fabrication which includes chapters on roll-to-roll processing, vacuum deposition, and printed electronics. These chapters are authored by established experts in their respective fields that have extensive experience in processing flexible glass substrates in toolsets that range from research to pilot scale. The third section focuses on flexible glass device applications and includes chapters on photovoltaics, displays, integrated photonics, and microelectronics integration. These are authored by experts with direct experience in fabricating and characterizing flexible glass devices. The diverse list of authors and their depth of experience in working with a variety of material systems, processes, and device technologies significantly adds valuable context to the overall flexible glass discussion.

Sean Garner
June 2017