METAL FILMS FOR PRINTED ELECTRONICS

Thesis for the degree of PhD METAL FILMS FOR PRINTED ELECTRONICS Ink-substrate Interactions and Sintering Thomas Öhlund Supervisor: Prof. Håkan Olin...
7 downloads 0 Views 10MB Size
Thesis for the degree of PhD

METAL FILMS FOR PRINTED ELECTRONICS Ink-substrate Interactions and Sintering Thomas Öhlund

Supervisor: Prof. Håkan Olin Assistant supervisors: Prof. Hans-Erik Nilsson Dr. Mattias Andersson

Department of Natural Sciences Digital Printing Center Mid Sweden University Doctoral Thesis 210 Örnsköldsvik, Sweden 2014

i

Department of Natural Sciences Digital Printing Center Mid Sweden University, SE-891 18 Örnsköldsvik Sweden ISSN 1652-893X ISBN 978-91-87557-98-9 Akademisk avhandling som med tillstånd av Mittuniversitetet framläggs till offentlig granskning för avläggande av teknologie doktorsexamen, torsdag den 18/12 2014, klockan 10:00 i Mediacenter DPC, Mittuniversitetet, Järnvägsgatan 3 Örnsköldsvik. Seminariet kommer att hållas på engelska. © Thomas Öhlund, 2014 Printed by Kopieringen Mid Sweden University, Sundsvall, Sweden, 2014

ii

Dedicated to the memory of my admirable father, Erik Öhlund

iii

iv

ABSTRACT A new manufacturing paradigm may lower the cost and environmental impact of existing products, as well as enable completely new products. Large scale, roll-toroll manufacturing of flexible electronics and other functionality has great potential. However, a commercial breakthrough depends on a lower consumption of materials and energy compared with competing alternatives, and that sufficiently high performance and reliability of the products can be maintained. The substrate constitutes a large part of the product, and therefore its cost and environmental sustainability are important. Electrically conducting thin films are required in many functional devices and applications. In demanding applications, metal films offer the highest conductivity. In this thesis, paper substrates of various type and construction were characterized, and the characteristics were related to the performance of inkjet-printed metal patterns. Fast absorption of the ink carrier was beneficial for well-defined pattern geometry, as well as high conductivity. Surface roughness with topography variations of sufficiently large amplitude and frequency, was detrimental to the pattern definition and conductivity. Porosity was another important factor, where the characteristic pore size was much more important than the total pore volume. Apparent surface energy was important for non-absorbing substrates, but of limited importance for coatings with a high absorption rate. Applying thin polymer–based coatings on flexible non-porous films to provide a mechanism for ink solvent removal, improved the pattern definition significantly. Inkjet-printing of a ZnO-dispersion on uncoated paper provided a thin spot-coating, allowing conductivity of silver nanoparticle films. Conductive nanoparticle films could not form directly on the uncoated paper. The resulting performance of printed metal patterns was highly dependent on a well adapted sintering methodology. Several sintering methods were examined in this thesis, including conventional oven sintering, electrical sintering, microwave sintering, chemical sintering and intense pulsed light sintering. Specially designed coated papers with modified chemical and physical properties, were utilized for chemical low-temperature sintering of silver nanoparticle inks. For intense pulsed light sintering and material conversion of patterns, custom equipment was designed and built. Using the equipment, inkjet-printed copper oxide patterns were processed into highly conducting copper patterns. Custom-designed papers with mesoporous coatings and porous precoatings improved the reliablility and performance of the reduction and sintering process.

v

The thesis aims to clarify how ink-substrate interactions and sintering methodology affect the performance and reliability of inkjet-printed nanoparticle patterns on flexible substrates. This improves the selection, adaptation, design and manufacturing of suitable substrates for inkjet-printed high conductivity patterns, such as circuit boards or RFID antennas.

Keywords: inkjet printing, silver nanoparticles, paper, flexible substrates, sintering, printed electronics, IPL sintering, flash sintering, copper films, coatings, thin films

vi

SAMMANFATTNING Ett nytt tillverkningsparadigm kan minska kostnad och miljöpåverkan för existerande produkter, såväl som att möjliggöra helt nya produkter. Storskalig tillverkning av flexibel elektronik och annan funktionalitet i rulle-till-rulleprocesser har stor potential. Ett kommersiellt genombrott förutsätter dock att material- och energiåtgång är lägre än för konkurrerande tillverkningstekniker, samt att en tillräckligt hög prestanda och tillförlitlighet på produkterna kan bibehållas. Substratet utgör en stor del av produkten vilket innebär att substratets kostnad och miljöpåverkan är viktiga. Elektriskt ledande filmer är nödvändiga i många produkter och tillämpningar. Den högsta ledningsförmågan för krävande tillämpningar erhålls med metallfilmer. I avhandlingen har papper av olika typ och konstruktion karaktäriserats, och egenskaperna relaterats till inkjet-tryckta metallfilmers prestanda. Snabb absorption av bläckets lösningsmedel var gynnsam för ledningsförmåga och väldefinierad mönstergeometri. Ytråhet med tillräckligt stor amplitud och spatiell frekvens, korrelerade negativt med ledningsförmåga och mönsterdefinition. Porositet var en viktig faktor, där karaktäristisk porstorlek var avsevärt viktigare än total porvolym. Ytenergi var mycket viktig för icke-absorberande substrat, men av begränsad betydelse för bestrykningar med snabb absorption. Genom att bestryka plastfilmer med polymerbaserade suspensioner, och därmed införa en mekanism för separering av lösningsmedel och nanopartiklar, så förbättrades mönsterdefinitionen avsevärt. Inkjet-tryckning av en ZnO-dispersion på obestruket papper fungerade som en lokal tunn bestrykning, vilken medgav elektriskt ledande silvernanopartikel-filmer ovanpå. Likadana silverfilmer kunde inte fås ledande när de trycktes direkt på det obestrukna papperet. Den slutliga prestandan på tryckta filmer var starkt beroende av en väl anpassad sintringsmetodik. Ett flertal sintringmetoder utvärderades, däribland konventionell ugnssintring, elektrisk sintring, mikrovågssintring, kemisk sintring samt blixtsintring. Specialkonstruerade bestrukna papper, med anpassade kemiska och fysikaliska egenskaper, användes för kemisk lågtemperatur-sintring av silvernanopartikel-bläck. Egen utrustning för materialprocessning och blixtsintring konstruerades och byggdes inom ramen för avhandlingsarbetet. Med denna utrustning kunde inkjet-tryckta mönster av kopparoxid omvandlas till kopparfilmer med hög ledningsförmåga. Specialkonstruerade papper med mesoporösa bestrykningar och porösa förbestrykningar, förbättrade tillförlitligheten i blixtsintringsprocessen och den resulterande kopparfilmens prestanda.

vii

Avhandlingen syftar till att klargöra hur växelverkan mellan bläck och yta, samt sintringsmetodiken, påverkar tillförlitligheten och prestandan på inkjet-tryckta, elektriskt ledande mönster på flexibla substrat. Detta förbättrar urval, anpassning, konstruktion och tillverkning av lämpliga substrat för inkjet-tryckta mönster med hög ledningsförmåga, såsom kretskort och antenner för RFID.

viii

ACKNOWLEDGEMENTS

To begin with, this work has been funded by the Gunnar Sundblad Research Foundation, the European Regional Development Fund, the KK Foundation and the Kempe Foundation, all of whose support I gratefully acknowledge. I would like to thank my Supervisors Prof. Håkan Olin, Dr. Mattias Andersson and Prof. Hans-Erik Nilsson for giving me great freedom, trust and support during my time as a Ph.D. student. I would also like to thank Dr. Henrik Andersson, Dr. Jonas Örtegren, Dr. Renyun Zhang, Dr. Magnus Hummelgård, Britta Andres, Niklas Johansson and Sven Forsberg, who have been co-authors or made other contributions regarding this work. I am grateful to Dr. Dan Bylund and Dr. Joakim Bäckström for allocating time within their busy schedules to share their expertise in characterisation methods. I would like to thank Prof. Magnus Norgren and Prof. Håkan Edlund for teaching interesting courses that have been important for this work. I would like to thank Dr. Anna Schuppert and Dr. Wolfgang Schmidt for a rewarding co-operation. I am grateful to Prof. Jouko Peltonen and all the other nice and skilled people that I know at Åbo Akademi University, for giving me such an inspiring and educational time during my visit. I am grateful to Christina Westerlind and Boel Nilsson at SCA R&D, who have always helped with characterisation and instrument training whenever needed. I would like to thank Marie Tjärnström and Leif Kassman at MoRe Research for a good cooperation. Last, but most importantly, I would like to thank my friends and family.

Örnsköldsvik, November 2014 Thomas Öhlund

ix

x

TABLE OF CONTENTS

ABSTRACT ....................................................................................................................... V SAMMANFATTNING ................................................................................................. VII ACKNOWLEDGEMENTS ............................................................................................ IX ABBREVIATIONS AND ACRONYMS ................................................................... XIII LIST OF PAPERS ........................................................................................................... XV RELATED PAPERS NOT INCLUDED IN THE THESIS...................................... XVI 1

INTRODUCTION ......................................................................................................1 1.1 1.2

2

PROBLEM FORMULATION ........................................................................................2 MAIN CONTRIBUTION OF THE THESIS.......................................................................3

INKJET .........................................................................................................................4 2.1 THERMAL INKJET ....................................................................................................5 2.2 PIEZOELECTRIC INKJET............................................................................................5 2.2.1 Printing conditions .........................................................................................6

3

FUNCTIONAL INKS FOR PIEZOELECTRIC INKJET ......................................7 3.1 IMPORTANT INK PROPERTIES ...................................................................................8 3.1.1 Viscosity .........................................................................................................8 3.1.2 Surface tension ...............................................................................................8 3.1.3 Volatility .........................................................................................................9 3.1.4 Particle size ....................................................................................................9 3.1.5 Particle concentration, density and colloidal stability ...................................9 3.2 METAL NANOPARTICLE INKJET INKS .....................................................................11 3.3 COFFEE STAINING..................................................................................................13

4

FLEXIBLE SUBSTRATES FOR PRINTED ELECTRONICS.............................14 4.1 4.2

5

PAPER ...................................................................................................................14 POLYMER FILMS ....................................................................................................17

CONDUCTOR PERFORMANCE EVALUATION ............................................19 5.1 PATTERN DEFINITION ............................................................................................19 5.1.1 Line width .....................................................................................................21 5.1.2 Raggedness ...................................................................................................21

xi

ELECTRICAL CONDUCTIVITY .................................................................................21 5.2 5.2.1 4-point electrical resistance measurements .................................................22 6

SINTERING ..............................................................................................................23 6.1 6.2 6.3 6.4 6.5

7

OVEN SINTERING ...................................................................................................24 ELECTRICAL SINTERING ........................................................................................25 MICROWAVE SINTERING .......................................................................................25 CHEMICAL SINTERING ...........................................................................................26 INTENSE PULSED LIGHT SINTERING........................................................................27

CUSTOM INKJET INK DEVELOPMENT ..........................................................30 7.1 CUO NANOPARTICLE INK ......................................................................................31 7.1.1 Synthesis and stabilisation ...........................................................................31 7.1.2 Tuning of ink properties ...............................................................................32 7.1.3 Application example : copper films ..............................................................33 7.2 ZNO NANOPARTICLE INK.......................................................................................33 7.2.1 Tuning of ink properties ...............................................................................33 7.2.2 Application example : spot-coating ..............................................................33

8

COMMENTS ON THE INKJET SYSTEM INTERACTION ............................34

9

SUMMARY OF PUBLICATIONS .........................................................................35

10

DISCUSSION .......................................................................................................42

11

CONCLUSIONS ...................................................................................................44

12

OUTLOOK.............................................................................................................45

13

REFERENCES........................................................................................................47

xii

ABBREVIATIONS AND ACRONYMS

AFM AgNP CIJ CuNP CuO DLS DOD IPL LWC MEMS NP OLED PCA PDMS PE PI PLS P-OLED PVP PPS RFID SEM SEM-EDS TEM ZnO

Atomic Force Microscopy Silver nanoparticle Continuous Inkjet Copper nanoparticle Copper(II) Oxide Dynamic Light Scattering Drop On Demand Inkjet Intense Pulsed Light Lightweight Coated Micro-electromechanical systems Nanoparticle Organic Light-Emitting Diode Principal Component Analysis Polydimethylsiloxane Polyethylene Polyimide Projection to Latent Structures (Partial Least Squares regression) Polymer Organic Light-Emitting Diode Polyvinylpyrrolidone, water-soluble polymer Parker Print Surf, measurement method for surface roughness Radio Frequency Identification Scanning Electron Microscopy SEM - Energy Dispersive X-ray Spectroscopy Transmission Electron Microscopy Zinc Oxide

xiii

xiv

LIST OF PAPERS This thesis is mainly based on the following six papers, herein referred to by their Roman numerals:

I

T. Öhlund, J. Örtegren, H. Andersson and H-E. Nilsson, “The Importance of Surface Characteristics for Structure Definition of Silver Nanoparticle Ink Patterns on Paper Surfaces”, Proc. NIP26: 26th Int. Conf. on Digital Printing Tech., 309-313. (2010)

II

T. Öhlund and M. Andersson, “Effect of Paper Properties on Electrical Conductivity and Pattern Definition for Silver Nanoparticle Inkjet Ink", Proc. LOPE-C, 115-119. (2012)

III

T. Öhlund, J. Örtegren, S. Forsberg and H-E. Nilsson, “Paper Surfaces for Metal Nanoparticle Inkjet Printing”, Applied Surface Science, 259, 731-739. (2012)

IV

T. Öhlund, H. Andersson. J. Örtegren and H-E. Nilsson, ”Sintering Methods for Metal Nanoparticle Inks on Flexible Substrates”, Proc. NIP25: 25th Int. Conf. on Digital Printing Tech., 614-617. (2009)

V

T. Öhlund, A.K. Schuppert, B. Andres, H. Andersson, S. Forsberg, W. Schmidt, H-E. Nilsson, M. Andersson and H. Olin, “Assisted sintering of silver nanoparticle inkjet inks on paper with active coatings”, Submitted for publication. (2014)

VI

T. Öhlund, M. Andersson, H-E. Nilsson and H. Olin, “Flash sintering of inkjet-printed copper patterns on flexible substrates: Achieving 69% of bulk conductivity using active pre-coatings", In manuscript.

xv

RELATED PAPERS NOT INCLUDED IN THE THESIS

S1

H. Andersson, A. Manuilskiy, C. Lidenmark, J. Gao, T. Öhlund, S. Forsberg, J. Örtegren, W. Schmidt, H-E. Nilsson, “The influence of paper coating content on room temperature sintering of silver nanoparticle ink”, Nanotechnology 24 (45), 455203. (2013)

S2

H. Andersson, C. Lidenmark, T. Öhlund, J. Örtegren, A. Manuilskiy, S. Forsberg, H-E. Nilsson, “Evaluation of coatings applied to flexible substrates to enhance quality of inkjet printed silver nano-particle structures”, IEEE Transactions on Components, Packaging and Manufacturing Technology, 2(2), 342-348. (2012)

S3

H. Andersson, K. Hammarling, J. Siden, A. Manuilskiy, T. Öhlund, H-E. Nilsson, “Modified EAS Tag Used as a Resistive Sensor Platform”, Electronics 1 (2), 32-46. (2012)

S4

H. Andersson, T. Öhlund, A. Manuilskiy, J. Örtegren, S. Forsberg, H-E. Nilsson, “Evaluation of InkAid surface treatment to enhance print quality of ANP silver nano-particle ink on plastic substrates”, Proc. LOPE-C 241-245. (2010)

S5

R. Zhang, H. Andersson, M. Andersson, B. Andres, H. Edlund, P. Edström, S. Edvardsson, S. Forsberg, M. Hummelgård, N. Johansson, K. Karlsson, H-E. Nilsson, M. Norgren, M. Olsen, T. Uesaka, T. Öhlund, H. Olin, “Soap-film coating: High-speed deposition of multilayer nanofilms”, Scientific Reports 3. (2013)

xvi

Chance favours the prepared mind. Louis Pasteur

xvii

xviii

1

INTRODUCTION

The invention of printing technology as we know it today is usually attributed to Johannes Gutenberg back in the 15th century [1], although it is known that early printing techniques were used more than a thousand years before that in Asia [2]. Since that time, graphic printing has become an important part of society with extensive applications such as packaging, books and newspapers. The latter two are declining in importance due to the competition with increasingly popular digital media. However, a new important concept has emerged over the last few decades, in which printing is used for the deposition of functional materials. This extends the scope of printing beyond the conveying of text and graphics. New exciting possibilities exist, such as the fabrication of electronic devices. Printable functional materials include electrically conducting [3-8], semiconducting [9-12] or insulating materials [13, 14], as well as materials with magnetic [15, 16], biological [17-22], chemical [23, 24], optical [25-27] or mechanical [28, 29] functions. Target applications for functional printing include radio frequency identification (RFID)[30, 31], flexible displays [32-36], sensors [37-40], printed memories [41-43], printed batteries [44-46], micro-electromechanical systems (MEMS) [47-49] and light-emitting devices [50-53]. In recent years, the global interest in functional printing research has expanded greatly and new commercial applications continually arise. Most methods known in traditional graphical printing can be used and have been used for the printing of functional fluids. The main advantage of digital printing technologies is a quick and simple workflow due to the elimination of processing steps between digital pattern information and final print. Among the digital technologies, inkjet is arguably the most flexible one, in that it is a non-contact method and can be used for a wide range of fluid compositions and substrates. One of the most important advantages of digital fabrication is that the substrates, materials and processing enable environmentally friendly, low-cost production. Extended knowledge in how to combine various materials and processing with flexible substrates, will permit cost-savings and novel products. This will arguably enable new markets with a large value. Existing products might be reduced in cost due to the new manufacturing methods and materials. The society will benefit from substituting part of the traditional electronics market with flexible electronics, manufactured using environmentally friendly substrates, materials and processes. A lower cost and an extended range of products add to the benefit. Manufacture of large-area functional devices further extends the advantage of digital fabrication.

1

1.1 Problem formulation There is considerable interest world-wide in printed functionality on low-cost, flexible substrates. Sensors, RFID and other functionality for logistics, communication or entertainment, are potentially important additions to future packaging and other paper media [54]. However, the combination of paper-based substrates and inks based on functional nanoparticle materials is challenging, since paper is inherently complex both in physical and chemical composition. Also, a fairly high temperature heating process is often called for to allow sintering which changes the properties of a printed inorganic functional material to enable the required performance [55]. This is again challenging since paper, or other low-cost flexible substrates, typically deform at the temperatures normally needed. Therefore more knowledge is required to understand how substrates interact with inorganic nanoparticle inks, and to identify the aspects of the substrate that are most important. Exploring alternative sintering processes compatible with papers or other low-temperature substrates is an additional necessity. This thesis attempts to share relevant knowledge in these matters to contribute to the understanding of the possibilities, limitations and requirements for successful application of functional nanoparticle inks on paper and other flexible substrates. The following questions can be formulated: •

Which are the most important characteristics of paper substrates regarding inkjet-printed nanoparticle layers?



Can common paper characterisation methods be used to estimate the performance of inkjet-printed conductors?



Do low cost paper alternatives exist that allow high performance inkjetprinted conductive features?



How significant are the benefits of coatings on papers and smooth plastic films?



Which methods and process conditions for sintering are appropriate for temperature-sensitive substrates?



Which properties and mechanisms are involved in low-temperature chemical sintering on coated substrates?

2

1.2 Main contribution of the thesis This thesis focuses mainly on material, interaction and processing aspects of inkjetprinted metal patterns on flexible substrates. The thesis highlights how the substrate properties, as well as sintering methodology, affect the performance of inkjet-printed conductors using metal nanoparticle inkjet inks. The thesis discusses a number of important properties, their relative importance, as well as suitable methods for the characterisation. Such knowledge is an important prerequisite for successful fabrication of highly conductive features, such as RFID antennas and electrodes for sensors and light-emitting devices, as well as devices for energy harvesting and storage. The gained knowledge should be useful also for other coated substrates and for inkjet dispersions based on other nanoparticles, such as magnetic nanoparticles for memory applications or metal-oxide nanoparticles for semiconducting layers. The thesis shows that certain low cost, high volume production, coated papers can be used as relatively high performance substrates for inkjet-printed conductors. The thesis presents custom-designed papers that are effective in assisting low-temperature sintering of silver nanoparticle films. Within the work of the thesis, equipment for intense pulsed light (IPL) processing of thin films has been designed and built by the author. Using this equipment in combination with custom-designed coated papers, inkjet-printed copper patterns with very high conductivity have been realized. The included work for publication can be summarised as: •

Three studies of conductor performance on a large set of paper substrates which were extensively experimentally characterised. The studies are focused on: -Pattern definition of conductors using different inks. A multivariate PCA model is used to extract correlation structure in the experimental data. -Relating pattern definition to conductivity. A multivariate PLS model is used to extract correlation structure in the experimental data. -Extended characterisation of the substrates and the printed nanoparticle layers. Analysis of the various characterisation methods. Analysis of important interaction mechanisms and properties.



A study of several sintering methods on two paper substrates with different properties. The study provides insight into the comparative performance of the methods for each paper, and how the paper properties may change the optimal sintering operational procedure.

3



A study of chemically assisted low-temperature sintering of silver nanoparticle films on specially designed coated papers. The papers are custom-manufactured with controlled variations. The study is an evaluation of how sintering behaviour is affected by variations in coating pore size, as well as surface chemistry.



A study of IPL processing of CuO patterns on specially designed coated papers. The papers employ different precoatings. It is demonstrated that a suitable precoating design significantly improves the performance and reliability of IPL processing.

2 INKJET The earliest inkjet technology is known as continuous inkjet (CIJ). CIJ uses a continuous ink stream that is broken into droplets by acoustic pressure waves [56]. Droplets not to be included in the printed pattern are charged during drop formation and deflected when passing an electric field. The deflected ink is collected and re-used in the process. A later development of inkjet is drop-ondemand inkjet (DOD). In DOD, the printing nozzles eject droplets only when they are called for in the print pattern. Below, the two most common drop-on-demand technologies will be briefly described: piezoelectric inkjet and thermal inkjet.

Figure 1. The piezoelectric functional materials inkjet printer used during the work of this thesis.

4

2.1 Thermal inkjet In thermal inkjet, each ink chamber contains a heating element. A current pulse through the heater causes rapid vaporization of the ink to form a bubble, causing a pressure increase. The pressure pushes a droplet of ink through the nozzle. Immediately after, condensation and contraction of the vapour bubble pulls new ink into the chamber (figure 2a) [57]. As a consequence, the ink must have a volatile component for the vapour bubble to form. Thermal inkjet is frequently used for aqueous dispersions of living cells, proteins, enzymes, DNA or other biological material. It can be argued that the pulse heating of the element in the chamber, reaching temperatures of 300-400 °C [58], would limit the compatibility with heat sensitive inks, such as those containing biological material. However, the bubble expands quickly and thermally isolates the main part of the ink from the heater. Therefore, only a small fraction of the ink is thermally affected, and heatsensitive materials can be printed with retained functionality [59-61].

a

b

Figure 2. Inkjet principle schematics. a) Thermal inkjet. b) Piezoelectric inkjet. (Courtesy of http://bucarotechelp.com/)

2.2 Piezoelectric inkjet Piezoelectric inkjet technology uses piezoelectric material in the ink-filled chamber behind the print head nozzles. A voltage is applied over the material to make it flex, generating a pressure pulse in the fluid that forces a droplet out of the nozzle (figure 2b) [62]. The waveform (the change of voltage over time) is tailored to the printhead characteristics and fluid properties, to achieve the most well-behaved and reliable droplet ejection possible [63, 64]. Further, the jetting reliability depends on a minimum of air bubbles and wetting of the nozzle plate [65]. Typically the waveform contains a positive voltage section to push ink out of the nozzle, and a negative section to pull the piezo element the opposite direction and

5

refill the ink chamber (figure 3). Piezoelectric printheads are usually preferred for functional inkjet inks, since the piezoelectric technology allows a larger flexibility in the actuating waveform, compared with thermal inkjet. Also, ink formulation is generally more flexible, since a volatile ink component is not required.

Figure 3. Waveform editor for the Dimatix piezoelectric inkjet printer.

2.2.1

Printing conditions

The end result of a printed feature depends on many factors, such as the properties of ink and substrate, the geometry of the feature, and the extent of preprocessing and postprocessing. Moreover, the characteristics of the inkjet system, and the environmental conditions, have influence as well. A few important properties are listed below. Nozzle size The diameter of the nozzle determines the drop volume and therefore affects the maximum resolution and layer thickness. A smaller nozzle diameter allows higher possible resolution but is more prone to nozzle blockage and instability. Printheads with a nozzle diameter of 9 µm are available, corresponding to drop volumes as low as 1 pL. Such printheads were tested during this thesis work, although printheads with 10 pL (22 µm) nozzles were chosen due to a significantly better jetting reliability.

6

Temperature and humidity A higher temperature and lower humidity, increases the evaporation rate of a drop in position for jetting at the nozzle opening. A high evaporation rate can be problematic for the jetting reliability, particularly for small nozzle sizes. This is because evaporation increases the particle concentration in the drop which may lead to nozzle blockage. Voltage amplitude Increasing the maximum amplitude of the electric pulses controlling the piezoelectric element increases the flow velocity and the magnitude of forces acting on the drop at ejection. Generally, inks with high surface tension or high viscosity require a higher voltage for drop ejection. If the voltage is too high, the droplet may split into several parts ('satellites') that change direction and cause print quality issues. The choice of voltage is dependent on the printhead, as well as the properties of the ink. With the 10 pL printheads, a voltage of 16-24 V is suitable for the dispersions used throughout the thesis work. Voltage waveform The waveform defines the voltage variation, during the time interval that is necessary for ejection of one droplet. The waveform usually contains a positive voltage section to eject the ink droplet and a negative section to refill the ink chamber. The pulse duration can be increased to better adapt the waveform to an ink with higher viscosity. The amplitude and waveform is tailored to the ink rheology and printhead design to achieve reliable drop formation and ejection. A total pulse duration of 10 µs is common with the Dimatix system (figure 3).

3 FUNCTIONAL INKS FOR PIEZOELECTRIC INKJET To be able to print electrically conductive, or any other functional material, the material needs to be dispersed in a carrier solution to form an ink. The rheological requirements differ depending on the intended printing technology. In this thesis, piezoelectric inkjet has been the chosen method. Ink viscosity guidelines for different printing methods are compared in table 1.

7

Table 1. Comparison of recommended viscosity ranges for different printing methods [66].

Printing method Thermal inkjet Piezo inkjet Gravure Flexography Screen Offset

Viscosity range (mPa·s) 1-5 5-30 50-200 50-500 1000-10000 40000-100000

3.1 Important ink properties 3.1.1

Viscosity

A viscosity in the 8-12 mPa·s range is recommended by the manufacturer of the inkjet system used in this thesis. Extending the viable range to approximately 5-40 mPa·s or slightly more, is possible with proper adaptation of the driving voltage and waveform. The low viscosity and narrow viscosity range associated with inkjet is necessary for reliable droplet formation and ejection. The viscosity of a dispersion of spherical rigid particles is given by Einstein’s equation of viscosity of dispersions [67]: η/η0 = 1+2.5φ

(1)

Where η is the viscosity of the dispersion, η0 is the viscosity of the medium (in absence of the particles) and φ is the volume fraction of particles. The viscosity increases with higher volume concentration of particles due to particle-particle interactions. Therefore, for nanoparticle inkjet inks, the volume concentration of material cannot be too high, in order to retain the viscosity within the viable range. Inks with modest concentration of particles may need a high viscosity additive such as glycerol. The inkjet system that was used throughout this thesis has the option of heating the ink up to a maximum temperature of 60 ˚C to reduce the viscosity. 3.1.2

Surface tension

The surface tension of the ink is a very important rheological property for inkjet. With the system used in this thesis, the recommended range is 28-32 mN/m according to the manufacturer. The surface tension of water is 72 mN/m which is ill-suited for reliable drop formation and ejection. To lower the surface tension of water-based dispersions, alcohols such as isopropanol may be added. However, a

8

quite large proportion is required to reduce the surface tension into the ideal range. Surfactants are much more effective at low concentrations. However, surfactants may pose some problems, since associated foaming can easily block ink channels and nozzles. To counteract foaming, de-foaming agents can be used. However, the surface activity of surfactants and de-foamers make them prone to stick to walls in storage containers or within the inkjet printhead, increasing the risk of fluctuations in surface tension, and system clogging [68]. 3.1.3

Volatility

For piezoelectric inkjet, the ink volatility should be sufficiently low. Volatile solvents compromise the reliability of drop formation, since fast evaporation of drops at the nozzle-air interface increases the material concentration, which may block the nozzle [68]. In contrast, when the ink droplets reach the substrate, quick solvent evaporation is advantageous. That is because quick solvent removal facilitates a quick setting and immobilization of the nanoparticle film. A quick setting is beneficial for the geometric pattern definition and functionality, and reduces the effects of coffee staining. This is discussed in papers I-III. A quick solvent removal at the substrate can be achieved by substrate heating or by absorption. The porous nature of paper substrates, with the associated absorption capability, can be used to advantage in this regard. If the main solvent of the ink is water or a more volatile solvent, it is usually neccesary to decrease the overall volatility by incorporating a suitable low-volatility additive. An example of 'humectant' that can be used for this purpose, is ethylene glycol. 3.1.4

Particle size

Maintaining a particle size smaller than 1/100 of the nozzle diameter is considered as a rule of thumb to reduce the risk of blocked nozzles. With the 10 pL printheads that were used throughout this thesis, the nozzle diameter is 22.5 µm, implying that the largest particles should preferably be less than 200 nm in the longest dimension. The particle size and particle shape depend on the synthesis method and the synthesis conditions. Using wet chemical synthesis methods, the size and shape of the nanoparticles are usually affected by reaction temperatures and the concentrations of the reagents [69, 70]. 3.1.5

Particle concentration, density and colloidal stability

Dispersions of solid particles in a liquid become unstable when the homogeneous distribution of particles in the medium no longer is maintained. The probability of aggregation increases with larger volume concentration of particles, since the probability of collisions increases. Sedimentation adds to the problem, since when the solid particles settle to the bottom of the ink over a period of time, the volume

9

concentration of particles is increased at the bottom. The speed of sedimentation will be given by: Vsed = 2(ρp - ρs)*g*rp / ηs

(2)

Where ρp and ρs are the densities of particle and solvent, respectively; g is the acceleration constant of gravity, rp is the particle radius and ηs is the viscosity of the solvent. Particles of a metal nanoparticle inkjet ink have high sedimentation velocities since the particle density is high and the solvent viscosity is low. It follows from equation (2) that the only practical means to decrease the sedimentation rate of such inks is to reduce the particle size. However, fabrication of nanoparticles that are small enough to avoid sedimentation, may imply complex processes and high cost. The silver nanoparticle inks used within the work of this thesis commonly have particle diameters of more than 20 nanometers. For high density metals such as silver, sedimentation will occur during long-term storage, and effective stabilisation is required to maintain good stability. The stabilisation serves to prevent particle collisions that cause irreversible coagulation by introducing particle-particle energy barriers. The main types of stabilisation are referred to as electrostatic stabilisation, steric stabilisation and electrosteric stabilisation [71]. A higher concentration of solid material in the ink implies an increased dry film thickness. Therefore, the desired particle concentration will depend on the application requirements of film thickness. A higher particle concentration decreases the sensitivity to surface roughness of the substrate, and reduces coffee staining effects. The maximum volume concentration of solids is limited, since if the concentration is too large, particle-particle interactions will imply a viscosity above the feasible range. Furthermore, the higher the concentration, the more challenging the colloidal stability will be. This is since a smaller average particleparticle distance increases the probability of particle collisions and agglomeration. Commonly, the dispersed nanoparticles are stabilised with polymers that adhere to the particle surface [72]. In solution, ligands will extend from the particles and prevent contact with other particles. In close proximity, overlapping ligands will increase system entropy thereby creating an effective energy barrier [67] (steric stabilization). If the polymers are charged in the solvent, the charges may improve the stabilisation by electrostatic repulsion [73] (electrosteric stabilisation). The volume concentration of solids in nanoparticle inkjet dispersions is rarely more than 10 vol%, but more commonly 1-5 vol%. Among commercially available silver nanoparticle dispersions, the highest concentrations available correspond to approximately 10 vol% (50-60 wt%).

10

Because of the limited material concentration and the small drop volumes (in the 550 pL range), the thickness of the dried layers is low, typically below 1 µm. This is unfavourable if very low sheet resistance is desired in conductive structures such as coils or antennas. However, in other applications, very thin layers are mandatory for the desired performance, such as for active semiconductor layers in thin film transistors [74] or emissive layers in P-OLED displays [75].

3.2

Metal nanoparticle inkjet inks

Several types of conductive materials can be formulated into inkjet inks, such as conductive polymers [6, 76, 77], carbon nanotubes [5, 78] and metal nanoparticles [55]. When high conductivity is required, metal nanoparticle inks offer the best performance. Metal nanoparticle inks of gold [79, 80], silver [3,4] and copper [81, 82] can be used for inkjet-printing of conductive structures on various substrates. Frequently, silver nanoparticles have been used due to low reactivity in air. Silver has the highest conductivity of all metals, and many inkjet-printable dispersions are commercially available. Gold is, due to its high cost, hardly justified unless the material consumption is small, and its specific properties are needed in the application. For example, gold electrodes can be combined with silver electrodes in electrochemical sensors [29]. The relatively high cost of silver remains a limitation for cost-sensitive applications. Copper has approximately 95% of the electrical conductivity of silver, but only 1% of the price. Therefore, copper-based ink formulations are highly interesting. However, the reactivity of copper nanoparticles in ambient conditions is challenging, compared with silver nanoparticles. The reactivity issues can be by-passed by using ambient-stable dispersion of copper oxide (CuO) nanoparticles which can be converted into pure copper using intense pulsed light processing (IPL) [83, 84]. This possibility has been utilized in paper VI. IPL processing is further described in the sintering section of this thesis. In table 2, conductivity/price ratios for various metals are specified. Aluminium has the highest conductivity/price ratio of all metals, and copper the second highest. However, aluminium has the highest reactivity with oxygen, and lower conductivity than copper. Therefore, copper is arguably the most promising metal for low-cost, highly conductive layers in printed electronics applications.

11

Table 2. Comparison of conductivity/price ratios for different metals. Stock market prices in SEK (swedish krona), 30/7-2014. Metal Conductivity (MS/m) Price (SEK/kg) Conductivity/Price (kS·kg/(m·SEK)) Al 35 13.6 2579 Cu 59.6 49.1 1215 Zn 16.9 16.5 1026 Ni 14.3 127 112 Ag 63 5070 12.4 Au 41 320000 0.128

Patterned metal films manufactured by inkjet printing of nanoparticle dispersions have been the focus of interest throughout this thesis. Very high conductivity can be achieved at high processing temperatures. Under certain conditions, high conductivity can also result at low processing temperatures. Those conditions are of special interest when highly conductive features are required on temperaturesensitive substrates.

Figure 4. Dried layer of silver nanoparticles deposited on LWC paper with inkjet printing. The particles are 20-100 nm in diameter. (TEM, Magnus Hummelgård)

12

3.3 Coffee staining One, occasionally problematic, effect observed with nanoparticle inks is known as 'coffee staining'. This means that the particle concentration of a deposited liquid will increase at the edges of the line/droplet. That is because liquid is evaporating at a higher rate at the contact line (the edge), leading to a convective flow of solution from the interior. Therefore, there is a net transport of particles toward the edges of the structure [85]. It has been shown [86] that a higher concentration of particles reduces the impact of coffee staining. This has been confirmed by unpublished experiments by the thesis author (figure 6).

Figure 5. Coffee staining.

a

b

Figure 6. Coffee staining effect on mesoporous coated paper using different concentrations of dispersed silver nanoparticles. a) 40 wt% silver, slight coffee staining visible. b) 20 wt% silver, strong coffee staining is evident. Note that the spreading of nanoparticles is larger for the 20 wt% Ag ink, and therefore not allowing the complete conductor width to fit within the AFM scan range. (AFM scans showing cross-sections of minimum width conductors, thesis author).

13

4 FLEXIBLE SUBSTRATES FOR PRINTED ELECTRONICS 4.1 Paper Paper-based substrates for printed electronics are interesting for several reasons. Environmental friendliness, flexibility and low cost are key benefits. The mechanical, physical and chemical properties may be altered by additives, or by changes in the manufacturing process. Therefore, it is possible to tailor papers or coatings, adapting them to a specific functional ink or application. Such tailoring has been described in paper V and paper VI. Uncoated papers are generally too porous and rough to be compatible with most functional inks which means that coatings need to be applied. Important coating properties, and their relation to the performance of inkjet-printed metal films, have been examined and discussed in the included papers of this thesis and in other papers [87]. Examples of important properties are pore size, surface roughness, absorption rate, mechanical stability and surface chemistry, as has been shown in papers I-III. These properties must be controlled in order to prevent problems occurring within the printed layer. To optimize the properties, the entire system must be understood and accounted for. This system includes substrate, ink rheology and composition, printing- and sintering method, as well as the application. The main challenge of using paper as a substrate for nanomaterial deposition is that paper is fibrous and non-uniform to its nature. High porosity and surface roughness of uncoated paper surfaces will in most cases render difficulties with reaching the intended functionality. This has been shown and discussed in paper III. Furthermore, paper is readily affected by environmental factors such as temperature and humidity that might change mechanical properties and cause dimensional- and roughness changes. However, paper as a substrate for printed electronics, or other printed functionality, has many merits. The attractiveness of paper substrates for such applications derives from a number of factors: •

As a renewable resource, paper has environmental advantages compared to most other substrates.



Industrial processes for paper production are mature and cost-efficient.



Certain applications demand, or benefit from, flexibility. The flexibility of paper substrates can vary from very flexible to fairly stiff, depending on composition and construction.

14



Due to the porous nature of paper, the weight is low, leading to reduced transport costs. The porosity also has the effect of increasing liquid absorption, which is beneficial especially for low viscosity inks, as is the case with inkjet.



Roll-to-roll production is possible for truly low-cost, large scale manufacturing of functionality on flexible substrates. In such applications, the absorption capability of paper is an asset. The far-reaching possibility to adapt the chemical and mechanical properties to suit the intended inks and application is a large advantage.



Widespread use. Paper is found everywhere: in packaging, magazines, tickets etc. Printing functional materials directly on paper surfaces is therefore of general interest.

With the inherent difficulties, actions that might be taken to increase the probability of success are 1) Adapting the printing technology and printing fluid to suit the paper substrates. 2) Utilizing suitable coatings of the paper surface.

Bollström et al. developed a multilayer-coated paper, suitable for printed electronics [88]. The paper utilized a thin and smooth top coating made by mineral pigments and latex binder, and an effective barrier layer made by a latexsuspension, containing optional mineral pigments.

15

a

c

e

b

d

f

Figure 7. Cross sections (top row) and surfaces (bottom row) of AgNP layers on papers. ab) Mesoporous coated paper. c-d) LWC paper. e-f) Uncoated office paper. The Uncoated paper does not allow formation of continuous films, due to large pore structures and large surface roughness. Note that the scale varies between images. Paper III.

Figure 8. Custom-designed inkjet-papers (Cross sections, SEM). The top white layers are AgNP films. The type of precoating differs. (a) Reference paper with polyethylene (PE) barrier. (b) 'Active' paper with CaCO3 precoating. This triggers low-temperature sintering of certain AgNP films (paper V) and improves IPL processing of CuO NP films. Paper VI.

16

Various paper designs, as well as important paper- and coating properties, have been discussed in paper III and V. Coatings with nanosized pores designed for quickly receiving the ink carrier fluid are of particular interest. A problem with paper substrates in general and coated papers in particular, is permanent deformation when exposed to typical sintering temperatures. Suitable ink formulations and sintering techniques are usually called for. Sintering methods compatible with papers and other temperature-sensitive substrates, are examined in papers IV-VI.

Figure 9. Inkjet-printed narrow conductor on mesoporous coated inkjet-paper. The conductor width is approximately 45 µm and the height 0.5 µm. (AFM, thesis author)

4.2 Polymer films Flexible polymer films are extensively used for printed electronics. Compared to paper substrates, uncoated polymer films usually have very smooth surfaces, homogeneous properties and no porosity. While this is beneficial in several aspects, the smooth, non-absorbing surface is non-ideal in other aspects. Since the typical inkjet dispersion contains a large proportion of carrier fluid, it is difficult to achieve fast drying which may be crucial in a roll-to-roll process. Further, nonexistent absorption and a very low surface roughness, poses challenges regarding pattern definition and coffee staining, as discussed in paper I. Paper S2 evaluates

17

the effect of applying thin polyvinylpyrrolidone (PVP) coatings on polyethylene and polyimide films; it is concluded that coatings of as little as 5 µm wet film thickness, improve the pattern definition considerably (figure 11). Some of the polymer materials have high temperature stability, albeit those tend to be relatively expensive.

Figure 10. Surface topography of different substrates (AFM, thesis author). a) Mesoporous coated paper. b) Lightweight coated (LWC) paper. c) PET film. Surface roughness values are given as RMS values over 20x20 µm. Paper V.

Figure 11. Pattern definition of AgNP ink on untreated polymer films (a-b) and after application of a thin (5 µm wet thickness) PVP-based coating (c-d). The pattern definition is significanly improved by the application of coatings. a) Untreated PI. b) Untreated PE. c) Coated PI. d) Coated PE. e) Nominal printing pattern. Paper S2.

Polyimide (PI, Kapton) films can withstand prolonged exposure to 300 °C and still keep their flexibility. The high surface energy normally causes a large ink wetting. Due to the large wetting, the smallest possible structures can generally not be achieved without reducing the surface energy [89]. The cost of polyimide film is relatively high.

18

Polyethylene (PE), Polycarbonate (PC) and Polyethylene terephtalate (PET) are commonly used as low-cost flexible substrates for printed functionalities. Their softening points are below 150 °C and therefore they are not compatible with oven processing if high sintering temperatures are needed. Polytetraflourethylene (PTFE, Teflon) has excellent temperature stability and resistance to aging and chemicals. It is, however, expensive and its low surface energy can pose problems with “line bulging”, which can be described as unwanted, local broadening of a printed structure [90].

5 CONDUCTOR PERFORMANCE EVALUATION When evaluating the performance of conducting films and patterns, two of the most important aspects are conductivity and geometric pattern definition. High conductivity with small variations are generally desired, and sometimes crucial. For example, paper S3 demonstrates a simple RFID tag using a coil antenna. The coil resistance must be sufficiently low in relation to the inductance. Otherwise, the resonance circuit Q-value will be too low and the tag cannot be detected by the reading electronics. Well-controlled ink spreading implies well-defined geometry of the printed features. This allows closer packing of the features, and smaller conductor widths. Not only does this reduce material consumption, but it also improves device reliability and performance, decreasing its power consumption and increasing its high-frequency limit of operation.

5.1 Pattern definition Determination of pattern definition has been made by using parts of the ISO 13660 standard for print quality evaluation. Straight minimum-width lines, using a single nozzle, have been printed and scanned into digital images. The print quality measures are determined from the image by pixel-level calculations defined in the standard. In-house software has been used to calculate line width and raggedness according to the standard. The ISO standard requires that the analysed pattern image has been acquired digitally with a minimum resolution of 600 dpi. During this work, the acquirement has been made with a flatbed scanner, using an optical scanning resolution of 2400 dpi. Detailed information is given in the ISO 13660 document [91]. A brief description of the calculation procedure, is given below.

19

Figure 12. Pattern definition of polar and non-polar AgNP ink on various papers and PI film. Line width and raggedness are given in µm. Mesoporous coated papers (1-4) with a large absorption rate and low surface roughness show the best definition. Slow-absorbing (6-7) or non-absorbing (10) substrates compromise pattern definition. Excessive surface roughness compromises definition as well. Paper I.

20

5.1.1

Line width

A horisontal line is divided into a large number of vertical segments, each segment being one pixel wide. The line width is defined as the average height of the segments. The height of each segment is determined by using a threshold value T given by T = Rmax - 0.6(Rmax – Rmin)

(3)

Where Rmax and Rmin are the maximum and minimum reflectance in the image, respectively. 5.1.2

Raggedness

Using the same line segmentation as when calculating the line width, the raggedness is defined as the standard deviation of all distances d between the edge threshold contour and a fitted straight edge. The threshold value T is the same as calculated in (3).

Figure 13. The raggedness calculation.

5.2 Electrical conductivity Electrical conductivity is defined as the reciprocal of resistivity, where the resistivity ρ is defined as ρ=R·A/L

(4)

Where R is the electrical resistance, A is the cross-section area, and L is the conductor length. For thin films, the sheet resistance Rs is sometimes used, defined as Rs = ρ / t

(5)

Where t is the thickness of the film.

21

The electrical conductivity or resistivity of the printed conductors has been determined by 4-point resistance measurements and measurements of conductor geometry. The geometry has been determined using AFM (film thickness) and image analysis (length and width). 5.2.1

4-point electrical resistance measurements

To improve the accuracy of the resistance measurements, in particular for low resistance structures, a 4-point method should preferably be used. With this method, separate probe/wire pairs are used for the current-sourcing and the voltage sensing functions. As illustrated in figure 14, current is sourced through the points at 1 and 4, via a pair of probes and current wires. This current generates a voltage drop across the structure to be measured according to Ohm's law V=R·I. The current also generates a voltage drop across the current wires themselves. To avoid a measurement error from the voltage drop, a pair of sense probes are placed immediately adjacent to the target resistance R, at points 2 and 3. Because almost no current flows in the sense wires, the voltage drop over the sense contact points and wires is extremely low and the resistance R is therefore effectively measured without contribution from wiring and contact-point resistances. The sense probes are normally arranged as the inside pair, and the source probes as the outside pair. If the probe tips are equidistant, and the distance between the tips is small compared to the distance from the tips to the nearest edge of the conducting film, it can be derived that Rs = (V / I) · (π / ln2)

(6)

where Rs is the sheet resistance. The resistivity can then be determined if the film thickness is known, according to equation (5).

Figure 14. Schematic of a 4-point resistance measurement setup. The influence of the wire resistances Rw and the contact point resistances are effectively eliminated with the setup.

22

6

SINTERING

The process of thermal sintering can be thought of as a two-phase process. The first phase starts at the temperature where the protective organic shell of the particles degrades to the extent that direct particle-particle contact starts to form. The second, actual sintering phase, starts at a higher temperature and involves coalescence of particles and neck formation, so that the particle matrix transforms into larger structures. Because of the small size of a nanoparticle, there will be a high relative amount of partially unbounded surface atoms. These atoms represent a higher energy state which therefore lowers the melting point of the NP, compared to the corresponding bulk material [92]. Generally, the smaller the particle size, the lower the temperature that is required for efficient sintering. However, the amount and composition of organic stabilizers and other additives strongly affect the sintering temperature [93]. Stabilizing molecules with weaker bonding to the metal NPs, require less energy for de-stabilisation and therefore reduce the sintering temperature [94]. Introducing a suitable chemical compound in the substrate, or altering another effective substrate property, may allow substrate-induced destabilisation and low-temperature sintering of the NP matrix [95]. Such an approach is demonstrated in paper V of this thesis. Using an oven for heat treatment is the traditional and most common method due to its simplicity. However, faster and more effective methods are continually being researched, including methods compatible with temperature-sensitive substrates. Different alternative methods have been evaluated and compared in paper IV. Separate studies of chemically assisted sintering (paper V) and intense pulsed light sintering (paper VI) are also included. Some of the most promising methods are briefly described below.

23

a

b

c

d

e

f

Figure 15. AgNP ink layer on a mesoporous inkjet photo paper (top row) and a laser printer paper (bottom row). Oven sintering at the highest substrate-tolerable temperature is not sufficient for neck formation and particle growth. Electrical sintering implies substantial particle growth and high conductivity. The scale bars represent 0.5 µm. a) Unsintered. b) Oven 110 °C, 3 min. c) Electrically sintered. d) Unsintered. e) Oven 150 °C, 20 min. f) Electrically sintered. Paper IV.

6.1 Oven sintering Simplicity, as well as a large degree of control giving predictable and repeatable results, are the main advantages of oven sintering. Oven sintering relies mainly on heat convection in air, and heat conduction in the material that is being sintered. However, the method requires a rather long exposure time, minutes or more. Neither is it energy-efficient, nor is it suitable for roll-to-roll-manufacturing. Oven sintering is non-selective in that it will heat the substrate to the same extent as the deposited material. Therefore, the sintering temperature needs to be constrained when using heat sensitive substrates, such as many papers and plastic films (figure 16b). Using sintering temperatures below the substrate softening or deformation limits, typically limits the conductivity of metal nanoparticle films (figure 15 b, e). If higher conductivity is required, more sophisticated sintering methods should be utilised. Some of them are briefly described below.

24

a

b

Figure 16. a) Resistivity of oven-sintered AgNP films on laser-printer paper. Resistivity reaches a minimum within 10 minutes at temperatures above 110 °C. b) Cracks in a mesoporous absorption coating due to thermal expansion of the PE precoating (130 °C, 1 min). Paper IV.

6.2 Electrical sintering Electrical sintering was first proposed by Allen et al. [96]. It is based on running electrical current through the conducting structure. The resistive loss induces heat that sinters the structure. The sintering process can be quick and very effective (figure 15 c, f). The structure to be sintered needs to have an initial conductivity to be able to start the process, thus pre-sintering might be necessary with another method. The required physical contact of current electrodes to the pattern and the fact that only limited pattern geometries are suitable, limits the versatility of electrical sintering. Electrical sintering can be successfully applied with simple pattern geometries such as coil antennas for RFID.

6.3 Microwave sintering Exposure to microwaves is an effective way of heating thin films provided that the film thickness is below the penetration depth of the film material at the microwave frequency [97]. Microwave heating has been performed for thin films of gold and silver [97-99]. Experiments that have been conducted in paper IV show that microwave sintering can be fast and very effective, but the interaction between the pattern geometry and the electromagnetic field is difficult to predict and control (figure 17). Microwave sintering is highly dependent on the pattern geometry, size, location and direction. Therefore, it is challenging to use in actual practice.

25

b

a

c

Figure 17. a) Short-time exposure is preferred during microwave sintering of AgNP films. b) Extensive particle growth and high conductivity. c) Failure due to local overheating. Paper IV.

6.4 Chemical sintering With chemical sintering, the requirement of exposing the film to elevated temperatures is removed or reduced. Instead, the conductivity is increased by chemically induced coalescence of nanoparticles. The coalescence may be induced by post-processing at room temperature, exposing the film to vapour [100] or liquid [101, 102]. It is also possible to tailor the ink-substrate interaction, exploiting a complementary design of ink and substrate. Such an approach may allow a spontaneous coalescence of nanoparticles upon contact with the substrate. Paper V presents a study using specially designed coated papers, allowing chemically assisted sintering of a silver nanoparticle ink (figure 8, 18). a

b

Figure 18. Room-temperature sintering of AgNP film on tailored coated paper. a) No sintering on a reference paper with mesoporous coating and PE precoating. b) Spontaneous sintering on the same coating but with a CaCO3 precoating. Equal scale. Paper V.

26

6.5 Intense pulsed light sintering Intense pulsed light (IPL) processing means that a thin film is effectively heated by high-power pulsed irradiation from a flash lamp. The pulse duration is typically no longer than a few milliseconds, and therefore the heat transfer to the underlying substrate is limited. This enables fast high-temperature processing of the deposited film, with small impact on the substrate. Since the IPL may cover a large area, the method is feasible for roll-to-roll processing. IPL processing of thin films was first researched in semiconductor materials processing, such as annealing of ionimplanted silicon [103], crystallization of amorphous silicon films [104] and metallization of silicon [105]. The interest in semiconductor materials processing with IPL has been renewed in recent years [106-109]. The utilisation of IPL for processing of printed metal nanoparticle films was first proposed by Schroder et al. [110]. IPL sintering of silver films has been conducted in several studies [111-113], including paper IV in this thesis. Many advantages of IPL processing derive from the short pulse durations and high peak temperatures. One highly interesting advantage is that it enables solution processing of copper films in ambient conditions. Partly oxidized Cu NPs, or even solid CuO NPs, can be effectively reduced to pure copper films. Ink stabilising polymers and/or alcohol solvents can serve as reduction agents. By providing the necessary activation energy using IPL, chemical reduction and sintering can be achieved simultaneously, effectively transforming the deposited CuO layer into pure copper [83]. If the heating and cooling process is completed within a few milliseconds, re-oxidation can be avoided in ambient conditions. Paper VI presents a study demonstrating IPL processing of inkjet-printed CuO films on specially designed coated papers. a

b

c

D

Figure 19. IPL-processed Cu films on paper. a) Inkjet-printed antenna (unprocessed CuO). b) After IPL processing. c) Inkjet-printed circuit board. d) Rod-coated large-area film. Paper VI.

27

a b

c d

e

Figure 20. a) Black dashed: Oven sintered Ag, Black solid: Chemically assisted oven sintered Ag, Red: IPL-processed Cu. b) Insufficient exposure. c) Suitable exposure. d) Overexposure. e) PET film. CuO pattern could not be successfully processed. Paper VI.

Figure 21. CuO film on custom mesoporous papers, before and after IPL processing using equal IPL exposure (SEM). A CaCO3 precoating increases the effectiveness of the IPL processing. a) Unprocessed CuO. b) IPL processed Cu (PE precoating). c) IPL processed Cu (CaCO3 precoating). Equal scale. Paper VI.

A large part of the work in this thesis concerns ink-substrate interactions. Effective sintering reduces the importance of ink-substrate interactions, which is discussed in paper III and paper V. However, the sintering process in itself may depend on favourable ink-substrate interactions for its effectiveness. This is obviously true for substrate-induced chemical sintering (paper V) where the sintering is an integral part of the ink-substrate interaction. Successful IPL processing is also dependent on suitable ink-substrate interactions, which is discussed in paper VI. In addition to mechanical interactions (discussed throughout this thesis), and chemical interactions (predominantly discussed in paper III and V), we must include additional thermal and optical interactions to be able to better describe and understand the IPL process. Figure 22 shows a schematic with the proposed

28

interactions of different types. Each interaction type is associated with a number of variables given in the schematic. Arguably, the proposed variables are among the important variables governing the IPL processing.

Figure 22. IPL ink-substrate interaction schematic.

Variables involved in thermal interaction include the thicknesses, geometry and thermal properties of the film and substrate, and the thermal resistance of the interface. Arguably, one of the most important thermal properties is the thermal expansion. If there is a large thermal expansion mismatch between the film and the substrate, cracking and de-lamination may result. Optical interaction may be exemplified by the total light absorption of the film and its dependence on substrate properties such as roughness and internal light scattering. Tobjörk et al. used continuous IR-radiation to sinter inkjet-printed gold and silver nanoparticle films on paper [114]. The absorbance of the films was larger on paper than on a plastic or glass substrate. This was explained by the scattering properties of the paper and possible surface plasmon effects. The commercial CuO dispersion employed in paper VI uses nanoparticles with an average size of around 120 nm, according to the manufacturer. One way to possibly lower the required energy and further improve performance could be to mix the dispersion with smaller nanoparticles. Joo et al. found that mixing 20-50 nm CuNPs with 2 µm Cu microparticles resulted in lower resistivity [115]. Park et al. successfully applied flash sintering for spin-coated nickel nanoparticle dispersion on a polyimide substrate [116]. They observed that films with a 5-500

29

nm particle size distribution were easier to sinter, compared to films with a narrow particle size distribution with an average of 50 nm. This was explained by a more effective absorption in a wider wavelength range, and that the smallest nanoparticles possibly initiate the sintering process with a low melting temperature. During this thesis work, equipment for IPL processing (figure 23) has been designed and built by the author. The equipment is capable of sintering and material conversion of thin films. To allow process optimization, the pulse duration and the pulse intensity are variable. The available range of pulse duration is approximately 300 µs – 5 ms (t10), with a maximum electrical discharge power of approximately 1 MW. Successful IPL processing of inkjet-printed films of gold, silver and copper oxide has been performed during the thesis work. a b

Figure 23. a) IPL processing machine. b) Flash pulse (voltage waveform from photo diode).

7

CUSTOM INKJET INK DEVELOPMENT

Numerous silver nanoparticle dispersions for inkjet printing are commercially available; a few of them have been used in this thesis. Inkjet-printable dispersions of other metals, metal oxides or other semiconducting/insulating materials are much less commonly available. Therefore, custom ink formulations are necessary if less common materials are to be used, or if special control of the formulation is required. Particular care should be taken to ensure that the strict requirements on ink rheology and dispersion stability are fulfilled. Water-based inks are desirable for environmental- and cost reasons, but since water has a high surface tension and too low viscosity, water-based dispersions require rheological modification to be compatible with piezo-electric inkjet. While copper nanoparticles in water

30

dispersion are very prone to oxidize [117], CuO NPs do obviously not suffer from oxidation issues. Conveniently, additives working well for inkjet rheology tuning, have beneficial side-effects as reduction agents in IPL reduction of CuO [117]. Examples are ethanol and propanol, reducing surface tension in aqueous dispersions, and ethylene glycol and glycerol, increasing viscosity and reducing volatility [68]. Additionally, common dispersion-stabilizing polymers such as PVP, also function as reduction agents [118]. During the thesis work, custom inkjet ink formulation was performed by the author. Below, the development processes for two water-based functional inkjet dispersions will be briefly described, together with application examples. The first ink is a CuO dispersion which can be used for IPL conversion to copper. The second ink is a ZnO dispersion which can be used as a spotcoating of uncoated paper. Both of these inks should be usable also for semiconducting layers in devices. Some details in the recipes, processes and applications have been left out, since these activities have not yet been published.

7.1 CuO nanoparticle ink 7.1.1

Synthesis and stabilisation

Suitable amounts of distilled water, copper acetate and acetic acid are mixed and heated while stirring. When the desired reaction temperature is reached, NaOH is quickly added. The solution mix undergoes a quick colour transition from blue to black as the CuO nanoparticles are precipated from solution. The size and shape of the synthesised CuO nanoparticles will be affected by the reaction temperature and the relative concentrations of NaOH and acetic acid. Here, experimental conditions are used that give rise to spherical nanoparticles of less than 10 nm size. The precipated nanoparticles are collected and the reaction bi-products are removed by centrifugation. The particles are washed with analytical grade ethanol followed by centrifugation, drying and re-dispersion in distilled water at low particle concentration. This is because it was assumed that a lower concentration of particles facilitates the adhesion of stabiliser molecules at the surface of the nanoparticles. For functionalisation/stabilisation, a suitable amount and molecular weight of PVP was added, followed by overnight stirring at room temperature. This allows adhesion of PVP to the nanoparticles, creating a steric barrier that prevents agglomeration. The optimal PVP amount is determined by particle size measurements using DLS, using the PVP concentration that minimises the particle size. The synthesis and stabilisation process is schematically described in figure 24a. A TEM image of the particles after 2 months of storage is seen in figure 24b. It is seen that the stabilisation is effective since the particles are still very small and well dispersed.

31

a

b

Figure 24. a) CuO nanoparticle synthesis and stabilisation. b) The stabilisation is still effective 2 months after the ink formulation (TEM, Renyun Zhang).

7.1.2

Tuning of ink properties

The CuO concentration of the dispersion is increased to 10 wt% CuO by vacuum evaporation in a rotary evaporator. Glycerol is added to increase viscosity. Ethylene glycol is added to decrease the evaporation rate. A surfactant is added to decrease the surface tension to approximately 30 mN/m. A small amount of PDMS emulsion is added as de-foaming agent. Finally, a small amount of ascorbic acid is added as a reduction agent for IPL processing.

Figure 25. CuO ink formulation.

32

7.1.3

Application example : copper films

The CuO dispersion was confirmed to be inkjet-printable using a Dimatix materials printer. Printed CuO films on mesoporous inkjet photo paper could be transformed into highly conductive copper films by IPL processing (figure 26).

Figure 26. Conductive Cu film realised with IPL processing of the custom-developed CuO ink.

7.2 ZnO nanoparticle ink A quicker route for manufacturing a ZnO inkjet ink was possible, since water dispersion of ZnO nanoparticles was commercially available (50 nm average particle size, 50 wt%). 7.2.1

Tuning of ink properties

Rheology and evaporation rate were modified by the addition of glycerol, ethylene glycol, surfactant and de-foamer. Similar concentrations of additives were used, as with the CuO dispersion. a

b

Figure 27. a) ZnO ink formulation. b) Droplet ejection from the nozzles of the printhead.

7.2.2

Application example : spot-coating

The ZnO dispersion was confirmed to be inkjet-printable, using a Dimatix materials printer. The ZnO dispersion was inkjet-printed on an uncoated office paper as a spot-coating. Because of the high volume concentration of solids in the

33

dispersion, a single layer was effective in smoothing the high surface roughness of the uncoated paper. When inkjet-printing an AgNP dispersion on top of it, the ZnO spot-coated layer allowed a continuous, conductive AgNP film to form. In contrast, the AgNP dispersion could not form conductive layers directly on the uncoated paper surface (figure 28).

Figure 28. Inkjet-printed ZnO-dispersion used as spot-coating on uncoated copy paper (UCP). The upper half of the area has been covered with the ZnO layer. An inkjet-printed AgNP layer is conducting when printed on the spot-coated area but non-conducting when printed on the uncoated area.

8

COMMENTS ON THE INKJET SYSTEM INTERACTION

The end performance of printed metal films is not only a matter of ink-substrate interaction, but also environmental conditions, and sintering, are parts of the system as well. The selection of printing method is dependent on the requirements of speed and resolution, as well as the properties of the intended ink and substrate. For example, if the intended substrate is uncoated paper and the requirements of resolution and process speed are sufficiently low, screen printing would be a logical choice. This is since screen printing uses high viscosity inks, with a high concentration of solids and large particle size. Such ink properties give rise to comparably thick films which are therefore compatible with the large surface roughness and pore sizes of the typical uncoated paper. On the other hand, if the application requires very thin films and higher resolution, or if the material consumption needs to be minimized, inkjet is probably the best choice. Inkjet technology requires low viscosity and small particle size and therefore the requirements on the substrate are much more critical with respect to the surface roughness. However, the thin films of nanoparticles allow sintering at lower temperatures. With a suitable tailoring of the ink, or the substrate, sintering

34

can even occur at room temperature as an integral part of the ink-substrate interaction. This is shown in paper V. The intended application must be factored in as well. This concerns production cost, production speed and the intended lifetime of the product, as well as the flexibility and reliability requirements. An overview of different system components is suggested in figure 29. There are a number of factors for each component that have an impact on other parts of the system, as well as the performance and reliability of the complete system. All of the proposed factors have been discussed in this thesis.

Figure 29. Inkjet system performance.

9

SUMMARY OF PUBLICATIONS

Paper I is focused on studying pattern definition on various paper substrates; i.e., minimum width and raggedness of conductors printed with polar and non-polar ink. The paper substrates are characterized with regard to surface material content and physiochemical properties such as apparent surface energy, absorption rate, porosity and surface roughness. A Principal Component Analysis (PCA) model is constructed to extract the multivariate correlations in the data. For both inks, the PCA highlights a high absorption rate and large total pore volume as the most important prerequisites for high print definition. This is consistent with the observation that all the paper substrates in the study compared favourably to a homogeneous non-absorbing polymer film. It is indicated that fast absorption is of more concern than chemical ink-substrate interactions, since conductor width differences between different inks are small for fast absorbing substrates but large for slow absorbing substrates. The following is observed and concluded:

35



For substrates with low surface roughness and a low absorption rate, the print definition depends on a suitable matching with respect to the surface energy of the substrate and the surface tension of the ink. For substrates with high surface roughness or a high absorption rate, the matching is only of minor importance.



The PCA multivariate analysis suggests that the most important substrate properties for high pattern definition are a high absorption rate and large total pore volume.



Substrates with high surface roughness show significantly compromised pattern definition due to ink feathering. On the other hand, the smoothest substrate, the non-porous polymer film, shows the worst pattern definition of all substrates. It is argued that sufficiently low levels of surface roughness are not harmful to pattern definition, and may to some extent be beneficial, in particular if the absorption is slow or nonexistent.

Paper I was written in cooperation with Jonas Örtegren, Henrik Andersson and Hans-Erik Nilsson. The thesis author performed the major part of the design and implementation of the experiments and characterisation, as well as the major part of the reasoning and conclusions. The thesis author wrote the manuscript.

Paper II has the main part of the substrate selection and characterisation in common with paper I. In addition to the variables describing the pattern definition, the electrical conductivity is included as a measured variable. A multivariate PLS (projection to latent structures) model is constructed to extract correlations among conductivity, pattern definition and paper characteristics. The model indicates that the strongest correlations in the data are: • • •

Conductivity and pattern definition are correlated. Conductivity and pattern definition are negatively correlated to surface roughness. Conductivity and pattern definition are correlated with absorption rate.

Two of the papers did not allow conductivity. A microscopic analysis revealed that the reasons for this were a too large surface roughness or pore structure, that disrupted the film continuity. The pore structure is obviously an important parameter. However, the model correlation between the conductivity and the total pore volume is weak. This suggests that the total pore volume is not a good

36

descriptor for the pore characteristics of the paper. The observations and conclusions can be summarised as: •

Conductor width and raggedness are related, and both are strongly negatively correlated with conductivity. Therefore, the conductor should be geometrically well defined, in order for the conductivity to be high.



The PLS model suggests that low surface roughness and a high absorption rate are important paper characteristics for inkjet printing of high performance conductors.



The model shows an appropriate fit and predictive ability. It is therefore indicated that multivariate modelling based on the suggested paper characterisation methods can be used for reasonable prediction of the conductivity and pattern definition of inkjet-printed conductors on paper. Improvement of the multivariate model is suggested by describing the porosity with a characteristic pore size, instead of the total pore volume per unit area.

Paper II was written in cooperation with Mattias Andersson. The thesis author performed the major part of the design and implementation of the experiments and characterisation, as well as the major part of the reasoning and conclusions. The thesis author wrote the manuscript.

Paper III has the main part of the substrate selection and characterisation in common with paper I and II. The characterisation is extended to include the investigation of surface- and interface details with atomic force microscopy and scanning electron microscopy. This study is focused on a thorough examination and analysis of the electrical conductivity resulting from polar silver nanoparticle ink at different sintering temperatures. Key factors and mechanisms explaining the conductivity for each paper type are suggested, and these factors are further discussed in detail. It is suggested that the importance of chemical ink-substrate interactions decreases greatly as sintering temperature is raised. Beyond a certain temperature, physical-mechanical properties of the surface dominate, among which the most crucial ones are porosity and surface roughness. It is claimed that when the characteristic pore size and short-scale surface roughness amplitudes are much smaller than the typical nanoparticle layer thickness, a continuous film will form. It is observed that a large absorption rate reduces the relative importance of other parameters, in particular surface energy. Furthermore, it is concluded that

37

dimensional stability of the surface during printing and sintering is important to avoid stress and cracks in the film which, if present, will seriously impair the conductivity. Paper III was written in cooperation with Jonas Örtegren, Sven Forsberg and HansErik Nilsson. The thesis author performed the major part of the design and implementation of the experiments and characterisation, as well as the major part of the reasoning and conclusions. The thesis author wrote the manuscript.

Paper IV examines different methods for selective sintering of printed conductors on two types of paper substrates with different heat tolerance and ink-substrate adhesion. The selective methods are microwave sintering, electrical sintering and IPL sintering. The selective methods are compared with conventional oven sintering. For each of the methods examined, the maximum achievable conductivity for each paper is considered as well as the sintering process conditions affecting it, such as the power level and sintering time. Observations of how the sintering affects the reliability of the paper/conductor combination are made, as well as observations and explanations of the overheating conditions. Some observations and conclusions are: •

It is possible to reach similar levels of conductivity for inkjet photo paper and laser copy paper even though the surface properties of the photo paper are better adapted to inkjet printing. The reasons for this are that the laser copy paper has a higher heat tolerance as well as higher adhesion to the silver nanoparticle layer, allowing a larger sintering energy.



Sintering methods that heat selectively provide faster sintering and higher conductivity, but a suitable choice of method and fine tuning of the sintering procedure is needed to maximize conductivity but still prevent damage to the ink layer or substrate.



All selective methods resulted in better conductive performance than the conventional oven sintering due to the limited heat tolerance of the paper substrates. The oven sintering was found more effective at shorter sintering times compared to longer times, since the shorter times allowed the temperature to be raised somewhat, resulting in higher performance overall.

38



The intense pulsed light sintering method has the attractive attribute of negative feedback behaviour, distributing the energy more evenly over the area of the conductive film and protecting it from local overexposure. This is because the film reflectivity increases during the sintering. The opposite, positive feedback behaviour, is inherent for the electrical and the microwave sintering methods. Positive feedback means that as the sintering progresses, the current will increase. Consequently, if certain pattern areas start to sinter before other areas, local current differences tend to increase until there is an excessive current condition, damaging or destroying an area. In particular with the microwave sintering, issues with local overheating were evident. However, when successful, the electrical and microwave sintering methods gave rise to very effective sintering, resulting in high conductivity.

The sintering method comparison is summarised in table 3. Table 3. Comparision of sintering methods. AgNP tracks sintered on mesoporous photo paper (Photo) and laser copy paper (LC).

Oven sintering

Electric sintering

IPL sintering

Microwave sintering

15/18

50/40

40/30

40/35

Limiting factor

Substrate deformation

Contact method

Delamination

Local overheating

Time scale

1 – 10 minutes

0.1 - 1 s

0.1 – 10 ms

1 – 10 s

Feedback

Neutral

Positive

Negative

Positive

Conductivity % of bulk Ag (Photo/LC)

Paper IV was written in cooperation with Jonas Örtegren, Henrik Andersson and Hans-Erik Nilsson. The thesis author performed the major part of the design and implementation of the experiments and characterisation, as well as the major part of the reasoning and conclusions. The thesis author wrote the manuscript.

39

Paper V provides insight into low-temperature sintering mechanisms using coated papers of special design and custom manufacturing. The mesoporous absorption coatings use boehmite pigments, and contain a small amount of chloride as an active sintering agent. The coating pigment size is varied in order to create a series of papers with varying characteristic pore size. One paper series uses a conventional polyethylene precoating/barrier, while another series exploits a novel, porous CaCO3 precoating. The variations in coating morphology and precoating type significantly affect low-temperature sintering of inkjet-printed silver nanoparticle films. Observations and conclusions are: •

The presence of chloride effectively induces low-temperature sintering for one AgNP dispersion, while for another dispersion, the sintering is inhibited.



The surface morphology variations have large impact on resistivity when sintering is inhibited.



The precoating type has a large influence on low-temperature sintering. It is believed that this is due to the altered surface chemistry, but also that the altered absorption rate plays a role.



Effective sintering greatly diminishes the influence of both surface chemistry and morphology.



Compared with traditional PE precoatings, porous CaCO3 precoatings can be used to increase absorption speed, improve temperature stability and to alter surface chemistry.

Two concepts for assisting, or inhibiting, low-temperature sintering of a particular AgNP ink are proposed. 1) Incorporating an active sintering agent in the coating recipe. 2) Adapting the precoating. Each of those concepts, or a combination, may be utilised to tailor the paper properties for a specific AgNP dispersion. Such tailored coated papers may reduce or remove the need for explicit sintering in rollto-roll fabrication of printed electronics. Paper V was written in cooperation with Anna Schuppert, Britta Andres, Henrik Andersson, Sven Forsberg, Wolfgang Schmidt, Hans-Erik Nilsson, Mattias Andersson and Håkan Olin. The thesis author performed the major part of the design and implementation of the experiments and characterisation, as well as the major part of the reasoning and conclusions. The thesis author wrote the

40

manuscript. Anna Schuppert and Wolfgang Schmidt designed and manufactured the papers.

Paper VI focuses on intense pulsed light processing (IPL) of inkjet-printed CuO patterns. The custom-designed coated papers, manufactured for paper V, are used and compared with commercial flexible substrates. The porous CaCO3 precoating beneath the absorption layer improved the effectiveness and reliability of IPL processing compared with the PE precoating. The performance advantage remained over a range of coating pore sizes, as well as when compared with commercial substrates. Further, the resulting low-cost Cu films compared very favourably with oven-sintered Ag films. Observations and conclusions: •

With the novel precoating, the processing is realisable within 5 ms, using a single pulse. A resistivity of 2.43±0.15 µΩcm was achieved at best, corresponding to 69% of the conductivity of bulk copper. This performance is higher than any results reported before for solutionprocessed copper, arguably due to a favourable film-substrate interaction regarding chemical, mechanical, optical and thermal properties.



The beneficial effect of the CaCO3 precoating is mainly because of its porosity, allowing quick removal of the water, as well as its impact on surface chemistry, improving the film-substrate adhesion. The disadvantage of the CaCO3 precoating is a faster re-oxidation of the Cu films in ambient conditions.



Presence of acetic acid in coatings prevents re-oxidation during IPL processing and protects the Cu films from de-generation during ambient storage.



IPL processing of CuO patterns with varying feature size is challenging, since the optimal exposure energy tends to be dependent on feature size. The CaCO3 precoating significantly increased the exposure latitude which resulted in a pronounced reliability improvement in these challenging situations.

Paper VI was written in cooperation with Mattias Andersson, Hans-Erik Nilsson and Håkan Olin. The thesis author performed the major part of the design and implementation of the experiments and characterisation, as well as the major part of the reasoning and conclusions. The thesis author wrote the manuscript.

41

10

DISCUSSION

With uncoated flexible polymer films, the lack of absorption is a disadvantage for inkjet printing. Since the volume concentration of particles is low for inkjetprintable metal nanoparticle dispersions, the excess solvent needs to be quickly removed, if not to compromise the geometric definition and conductivity of the pattern. If absorption is non-existent, only evaporation remains as a mechanism for solvent removal. For quick evaporation on the substrate, either the ink solvent needs to be volatile, or the substrate temperature needs to be elevated. Volatile solvents are frequently problematic with inkjet, since fast evaporation will cause instability at the nozzle-air interface and clogging of the inkjet nozzles. A high substrate temperature is problematic, not only because of the low temperature tolerance of low-cost polymer substrates, but also because convection will transport heat from the substrate to the nozzles above it, causing evaporation in the nozzles and instability of the jetting process. If neither absorption, nor fast evaporation is controlling the ink spreading on the non-absorbing polymer films, it follows that the contact angle of the ink will become an important factor. In paper s2 and s4, non-absorbing polymer films have been coated with thin polymer coatings. Although the coatings applied are not porous, they still provide a mechanism for solvent absorption. As a consequence, geometric definition improved significantly. As far as paper substrates are concerned, absorption can be thought of as a side effect of the porous nature of their construction. In that sense, paper is better adapted for inkjet printing of functional materials, than are non-porous substrates such as glass sheets or polymer films. Papers I-III show that absorption rate has a positive correlation with pattern definition and conductivity for silver nanoparticle films. However, the surface porosity needs to be well controlled, which is also true for the related property, surface roughness. Metal nanoparticle films are easily disrupted by surface perturbations or pores of the same order of magnitude as the film thickness. The film thickness resulting from a nanoparticle inkjet dispersion is commonly a few hundred nanometers or less. For this reason, uncoated papers are rarely usable for nanoparticle functional inks, since surface roughness and pore sizes typically are large, in comparison with the film thickness. The paper characterisation methods utilized in papers I-III, are apparently relevant to estimate the suitability for inkjet-printed conductors, and therefore probably for other inkjet-printed functionalities as well. For some of the methods, deviations from, and extensions to, the standard measurement practices were made to better suit the involved coating types, as described in paper III. A few of the standard characterisation methods are not well suited for predicting adaptability with

42

nanoparticle conductors. This is the case with the Bentsen and PPS methods for estimating surface roughness, since they determine only a single value, but reveal no information about the underlying character of the roughness. As has been concluded in paper III, the character of the roughness has great importance; the most important range is arguably in the micrometer-scale region. Therefore, profilometry and atomic force microscopy are the preferable methods for surface roughness characterisation. Besides surface roughness, pore size is another critical property. Either of these two properties, if sufficiently large, will make conductive function impossible with the inks used within this thesis. However, if the characteristic pore size is small, a large total pore volume poses no problems; the corresponding large absorption rate will instead be beneficial for geometric definition and conductivity. Therefore, when examining coating porosity using mercury porosimetry, the pore size distribution should be the main interest. An interesting and important aspect is the cost/performance ratio of paper coatings for printed electronics. Although the highest performance was demonstrated on relatively expensive mesoporous coatings, it has been shown in papers I-IV that low cost, high volume production coated papers of LWC type, can be used as relatively high performance substrates for inkjet-printed conducting patterns. Since inkjet printing of nanoparticle dispersions is highly demanding regarding absorption capability and surface smoothness, it is expected that in most cases the LWC paper grades will work well for other printing methods as well. A potential problem with metal nanoparticle ink films on temperature-sensitive substrates is the general need for sintering to reach the desired level of conductivity. Sintering can in some cases be induced at room temperature by tailoring the substrate surface chemistry to de-stabilize the ink dispersant polymers upon contact (paper V, [95]). More effective sintering and higher conductivity is usually achievable only by heating the nanoparticle film to fairly high temperatures. The required temperature depends on the material and size of the nanoparticles, the binding energy and evaporation temperature of the dispersionstabilizing molecules, as well as the evaporation temperatures of the solvent components. Commonly, the required temperature exceeds the deformation temperature of the substrate. Selective sintering methods can be used to increase the achievable conductivity on heat-sensitive substrates. The selective methods rely on a difference in material properties between the nanoparticle film and the substrate, in order to selectively heat the film with the least possible heat transfer to the substrate. A number of sintering methods have been examined in papers IVVI of this thesis. The sintering process conditions, such as the sintering time and

43

the applied energy, should be optimized with respect to the entire system of method, substrate, ink, application and required performance.

11

CONCLUSIONS

The resulting performance of inkjet-printed electrically conductive features depends on a complex interplay of many factors. These factors include printing technology; rheological, chemical and material properties of the ink; chemical and mechanical properties of the substrate; as well as the sintering method and process conditions. Important physical surface properties for inkjet-printing of nanoparticle films are the surface roughness, the characteristic pore size and the absorption rate. The chemical properties of the substrate are also important, particularly when sintering temperatures must be restricted. This is because a suitable chemical interaction with the dispersion stabilising molecules, can induce the coalescence of nanoparticles. With metal nanoparticle dispersions, chemical low-temperature sintering can be achieved this way. Increased sintering temperature reduces the importance of physical and chemical surface-ink interactions to a large extent. Applying coatings on non-porous flexible polymer films was evaluated as enhancing pattern definition for silver nanoparticle ink. Thin PVP-based coatings provided significant improvements in pattern definition. A ZnO dispersion was inkjet-printed on uncoated paper as a very thin but functional local porous coating. An AgNP film was conducting when printed on the ZnO layer, but not when printed on the uncoated areas. Selective sintering methods should be used if the highest possible electrical conductivity is a priority, and the sintering process conditions should be tailored to the specific combination of method, ink and substrate, to optimize performance. Sometimes it is desired to obtain high conductivity on a wide range of substrates for the same ink system. This can be achieved by choosing and optimising a suitable sintering method, or by modifying the surface properties with a suitable coating. Swellable, or other mechanically unstable coatings, should be avoided. Papers and paper coatings offer great possibilities for substrate customisation with respect to chemical and physical properties. Consequently, the substrate may be tailored for a specific ink, sintering method, or application. In this thesis, customdesigned papers have been utilised in papers V-VI. It was concluded that the

44

presence of chloride in the absorption coating, and a porous CaCO3 precoating, had a large impact on low-temperature sintering of silver nanoparticle films, and IPL processing of CuO films.

12

OUTLOOK

Partially, this thesis has been focussed on finding the important aspects of paper substrates that improve, or restrict, the performance of inkjet-printed, electrically conducting nanoparticle patterns. The relative importance of these factors and how they relate to each other, has been estimated. Partially, the thesis has been focussed on alternative sintering methods, and understanding the mechanisms that these methods rely on. To allow greater insight into matters of ink-substrate interactions and sintering, the researcher must have control and knowledge of all relevant components in the ink and the substrate. With such control, modifications and tailoring can be performed with respect to both ink and substrate, to observe the resulting effect on the system. Papers V-VI use controlled substrates. Nanomaterial synthesis and inkjet ink formulation will be a useful tool for further research within this topic. In the future research, one possible approach is to tailor the ink-substrate system in further detail, with the intention of achieving substrate-induced room-temperature sintering that approaches the effectiveness of the best external sintering methods. IPL processing enabling lost-cost fabrication of copper patterns is highly interesting for the future. At this point, it is a challenge to combine the high exposure energy required for the processing with a high reliability of the process, and a high durability of the resulting Cu film. Further research, focussed on inksubstrate interaction and process development for IPL processing, could possibly solve these issues.

45

46

13

REFERENCES

1. 2.

Meggs, P.B., A history of graphic design. 1998: John Wiley & Sons. Whitfield, R., et al., Caves of the thousand Buddhas: Chinese art from the silk route. 1990: Published for the Trustees of the British Museum by British Museum Publications Ltd. Hsien-Hsueh, L., C. Kan-Sen, and H. Kuo-Cheng, Inkjet printing of nanosized silver colloids. Nanotechnology, 2005. 16(10): p. 2436. Kim, D. and J. Moon, Highly Conductive Ink Jet Printed Films of Nanosilver Particles for Printable Electronics. Electrochemical and Solid-State Letters, 2005. 8(11): p. J30-J33. Kordás, K., et al., Inkjet Printing of Electrically Conductive Patterns of Carbon Nanotubes. Small, 2006. 2(8-9): p. 1021-1025. Jang, J., Conducting Polymer Nanomaterials and Their Applications Emissive Materials Nanomaterials. 2006, Springer Berlin / Heidelberg. p. 189260. Li, D., et al., Conductive copper and nickel lines via reactive inkjet printing. Journal of Materials Chemistry, 2009. 19(22): p. 3719-3724. Park, B.K., et al., Direct writing of copper conductive patterns by ink-jet printing. Thin Solid Films, 2007. 515(19): p. 7706-7711. Wang, J.Z., et al., Low-cost fabrication of submicron all polymer field effect transistors. Applied Physics Letters, 2006. 88(13): p. 133502-3. Garnier, F., et al., All-Polymer Field-Effect Transistor Realized by Printing Techniques. Science, 1994. 265(5179): p. 1684-1686. Li, J., et al., A stable solution-processed polymer semiconductor with record highmobility for printed transistors. Sci. Rep., 2012. 2. Subramanian, V., et al., Printed organic transistors for ultra-low-cost RFID applications. Components and Packaging Technologies, IEEE Transactions on, 2005. 28(4): p. 742-747. Calvert, P., Inkjet Printing for Materials and Devices. Chemistry of Materials, 2001. 13(10): p. 3299-3305. Sirringhaus, H., et al., High-Resolution Inkjet Printing of All-Polymer Transistor Circuits. Science, 2000. 290(5499): p. 2123-2126. Voit, W., et al., Application of inkjet technology for the deposition of magnetic nanoparticles to form micron-scale structures. Science, Measurement and Technology, IEE Proceedings -, 2003. 150(5): p. 252-256. Sambucetti, C.J., Magnetic ink for jet printing. Magnetics, IEEE Transactions on, 1980. 16(2): p. 364-367. Newman, J.D., A.P.F. Turner, and G. Marrazza, Ink-jet printing for the fabrication of amperometric glucose biosensors. Analytica Chimica Acta, 1992. 262(1): p. 13-17.

3. 4.

5. 6.

7. 8. 9. 10. 11. 12.

13. 14. 15.

16. 17.

47

18. 19.

20. 21. 22.

23. 24. 25.

26.

27. 28. 29.

30.

31.

32. 33.

34.

Patel, N.G., et al., Screen-printed biosensors using different alcohol oxidases. Sensors and Actuators B: Chemical, 2001. 75(1–2): p. 101-110. Setti, L., et al., Thermal Inkjet Technology for the Microdeposition of Biological Molecules as a Viable Route for the Realization of Biosensors. Analytical Letters, 2004. 37(8): p. 1559-1570. Xu, T., et al., Viability and electrophysiology of neural cell structures generated by the inkjet printing method. Biomaterials, 2006. 27(19): p. 3580-3588. Xu, T., et al., Inkjet printing of viable mammalian cells. Biomaterials, 2005. 26(1): p. 93-99. Allain, L.R., D.N. Stratis-Cullum, and T. Vo-Dinh, Investigation of microfabrication of biological sample arrays using piezoelectric and bubble-jet printing technologies. Analytica Chimica Acta, 2004. 518(1–2): p. 77-85. O'Toole, M., et al., Inkjet printed LED based pH chemical sensor for gas sensing. Analytica Chimica Acta, 2009. 652(1-2): p. 308-314. Mabrook, M.F., C. Pearson, and M.C. Petty, An inkjet-printed chemical fuse. Applied Physics Letters, 2005. 86(1): p. 013507-013507-3. Bharathan, J. and Y. Yang, Polymer electroluminescent devices processed by inkjet printing: I. Polymer light-emitting logo. Applied Physics Letters, 1998. 72(21): p. 2660-2662. Fakhfouri, V., et al. Inkjet printing of SU-8 for polymer-based MEMS a case study for microlenses. in Micro Electro Mechanical Systems, 2008. MEMS 2008. IEEE 21st International Conference on. 2008. MacFarlane, D.L., et al., Microjet fabrication of microlens arrays. Photonics Technology Letters, IEEE, 1994. 6(9): p. 1112-1114. Jahn, S.F., et al., Polymer Microsieves Manufactured by Inkjet Technology. Langmuir, 2008. 25(1): p. 606-610. Määttänen, A., et al., A low-cost paper-based inkjet-printed platform for electrochemical analyses. Sensors and Actuators B: Chemical, 2013. 177(0): p. 153-162. Li, Y., et al., RFID Tag and RF Structures on a Paper Substrate Using InkjetPrinting Technology. Microwave Theory and Techniques, IEEE Transactions on, 2007. 55(12): p. 2894-2901. Subramanian, V., et al., Progress Toward Development of All-Printed RFID Tags: Materials, Processes, and Devices. Proceedings of the IEEE, 2005. 93(7): p. 1330-1338. Gelinck, G.H., et al., Flexible active-matrix displays and shift registers based on solution-processed organic transistors. Nat Mater, 2004. 3(2): p. 106-110. Hyora, S., et al., An electrothermochromic paper display based on colorimetrically reversible polydiacetylenes. Nanotechnology, 2014. 25(9): p. 094011. Shimoda, T., et al., Inkjet Printing of Light-Emitting Polymer Displays. MRS Bulletin, 2003. 28(11): p. 821-827.

48

35. 36. 37.

38. 39. 40.

41.

42. 43. 44. 45.

46. 47.

48.

49. 50. 51.

Comiskey, B., et al., An electrophoretic ink for all-printed reflective electronic displays. Nature, 1998. 394(6690): p. 253-255. Drzaic, P., et al., 44.3L: A Printed and Rollable Bistable Electronic Display. SID Symposium Digest of Technical Papers, 1998. 29(1): p. 1131-1134. Unander, T., H.E. Nilsson, and B. Oelmann. Printed touch sensor for interactive packaging and display. in Polymers and Adhesives in Microelectronics and Photonics, 2007. Polytronic 2007. 6th International Conference on. 2007. Sarfraz, J., et al., A printed H2S sensor with electro-optical response. Sensors and Actuators B: Chemical, 2014. 191(0): p. 821-827. Liana, D.D., et al., Recent Advances in Paper-Based Sensors. Sensors, 2012. 12(9): p. 11505-11526. Molina-Lopez, F., D. Briand, and N.F. de Rooij, All additive inkjet printed humidity sensors on plastic substrate. Sensors and Actuators B: Chemical, 2012. 166–167(0): p. 212-222. Sekitani, T., et al., Printed Nonvolatile Memory for a Sheet-Type Communication System. Electron Devices, IEEE Transactions on, 2009. 56(5): p. 1027-1035. Lian, K., et al., Printed flexible memory devices using copper phthalocyanine. Materials Science and Engineering: B, 2010. 167(1): p. 12-16. Lien, D.-H., et al., All-Printed Paper Memory. ACS Nano, 2014. 8(8): p. 76137619. Hilder, M., B. Winther-Jensen, and N.B. Clark, Paper-based, printed zinc-air battery. Journal of Power Sources, 2009. 194(2): p. 1135-1141. Braam, K.T., S.K. Volkman, and V. Subramanian, Characterization and optimization of a printed, primary silver–zinc battery. Journal of Power Sources, 2012. 199(0): p. 367-372. Gaikwad, A.M., et al., A flexible high potential printed battery for powering printed electronics. Applied Physics Letters, 2013. 102(23): p. -. Fuller, S.B., E.J. Wilhelm, and J.M. Jacobson, Ink-jet printed nanoparticle microelectromechanical systems. Microelectromechanical Systems, Journal of, 2002. 11(1): p. 54-60. Xu, R., et al., Screen printed PZT/PZT thick film bimorph MEMS cantilever device for vibration energy harvesting. Sensors and Actuators A: Physical, 2012. 188(0): p. 383-388. Nakano, S., et al., Low operation voltage of inkjet-printed plastic sheet-type micromechanical switches. Applied Physics Letters, 2008. 92(5): p. -. Kopola, P., et al., Gravure printed organic light emitting diodes for lighting applications. Thin Solid Films, 2009. 517(19): p. 5757-5762. Villani, F., et al., Inkjet Printed Polymer Layer on Flexible Substrate for OLED Applications. The Journal of Physical Chemistry C, 2009. 113(30): p. 1339813402.

49

52. 53. 54. 55. 56. 57.

58.

59.

60.

61. 62.

63.

64.

65. 66. 67. 68.

Chang, S.-C., et al., Dual-color polymer light-emitting pixels processed by hybrid inkjet printing. Applied Physics Letters, 1998. 73(18): p. 2561-2563. Sandström, A., et al., Ambient fabrication of flexible and large-area organic light-emitting devices using slot-die coating. Nat Commun, 2012. 3: p. 1002. Tobjörk, D. and R. Österbacka, Paper Electronics. Advanced Materials. 23(17): p. 1935-1961. Kamyshny, A., Metal-based Inkjet Inks for Printed Electronics. The Open Applied Physics Journal, 2011. 4: p. 19-36. C, C.R. and S.R. G, Fluid droplet recorder with a plurality of jets. 1968, Google Patents. Chen, P.-H., W.-C. Chen, and S.H. Chang, Bubble growth and ink ejection process of a thermal ink jet printhead. International Journal of Mechanical Sciences, 1997. 39(6): p. 683-695. Ross R. Allen, J.D.M., and William R. Knight Thermodynamics and Hydrodynamics of Thermal Ink Jets. Hewlett-Packard Journal, 1985(5): p. 2127. Setti, L., et al., An amperometric glucose biosensor prototype fabricated by thermal inkjet printing. Biosensors and Bioelectronics, 2005. 20(10): p. 20192026. Okamoto, T., T. Suzuki, and N. Yamamoto, Microarray fabrication with covalent attachment of DNA using Bubble Jet technology. Nature Biotechnology, 2000. 18(4): p. 438-441. Roda, A., et al., Protein microdeposition using a conventional ink-jet printer. BioTechniques, 2000. 28(3): p. 492-496. Chen, P.-H., et al., Pressure response and droplet ejection of a piezoelectric inkjet printhead. International Journal of Mechanical Sciences, 1999. 41(2): p. 235248. Kye-Si, K., Experimental analysis of waveform effects on satellite and ligament behavior via in situ measurement of the drop-on-demand drop formation curve and the instantaneous jetting speed curve. Journal of Micromechanics and Microengineering, 2010. 20(11): p. 115005. Kye-Si, K., Waveform Design Methods for Piezo Inkjet Dispensers Based on Measured Meniscus Motion. Microelectromechanical Systems, Journal of, 2009. 18(5): p. 1118-1125. Wijshoff, H., The dynamics of the piezo inkjet printhead operation. Physics Reports, 2010. 491(4–5): p. 77-177. Kipphan, H., Handbook of Print Media: Technologies and Production Methods. 2001: Springer. Holmberg, K. and B. Jönsson, Surfactants and Polymers in Aqueous Solution. 2003: John Wiley & Sons. Magdassi, S., The Chemistry of Inkjet Inks. 2010: World Scientific.

50

69.

70.

71. 72.

73. 74. 75.

76.

77. 78.

79.

80.

81.

82.

83.

Van den Rul, H., et al., Water-based wet chemical synthesis of (doped) ZnO nanostructures. Journal of Sol-Gel Science and Technology, 2006. 39(1): p. 41-47. Jana, N.R., L. Gearheart, and C.J. Murphy, Wet chemical synthesis of silver nanorods and nanowires of controllable aspect ratio. Chemical Communications, 2001(7): p. 617-618. Rahaman, M.N., Ceramic Processing and Sintering. 2003: Taylor & Francis. Studart, A.R., E. Amstad, and L.J. Gauckler, Colloidal Stabilization of Nanoparticles in Concentrated Suspensions. Langmuir, 2006. 23(3): p. 10811090. Fritz, G., et al., Electrosteric Stabilization of Colloidal Dispersions. Langmuir, 2002. 18(16): p. 6381-6390. Chung, J.H., et al., Effect of thickness of ZnO active layer on ZnO-TFT's characteristics. Thin Solid Films, 2008. 516(16): p. 5597-5601. Bale, M., et al., Ink-jet printing: The route to production of full-color P-OLED displays. Journal of the Society for Information Display, 2006. 14(5): p. 453459. Yoon, H. and J. Jang, Conducting-Polymer Nanomaterials for High-Performance Sensor Applications: Issues and Challenges. Advanced Functional Materials, 2009. 19(10): p. 1567-1576. Yoshioka, Y. and G.E. Jabbour, Desktop inkjet printer as a tool to print conducting polymers. Synthetic Metals, 2006. 156(11–13): p. 779-783. Small, W.R. and M. in het Panhuis, Inkjet Printing of Transparent, Electrically Conducting Single-Walled Carbon-Nanotube Composites. Small, 2007. 3(9): p. 1500-1503. Huang, D., et al., Plastic-Compatible Low Resistance Printable Gold Nanoparticle Conductors for Flexible Electronics. Journal of The Electrochemical Society, 2003. 150(7): p. G412-G417. Ni, Z., et al., Self-aligned inkjet printing of highly conducting gold electrodes with submicron resolution. Journal of Applied Physics, 2007. 101(6): p. 064513. Youngil, L., et al., Large-scale synthesis of copper nanoparticles by chemically controlled reduction for applications of inkjet-printed electronics. Nanotechnology, 2008. 19(41): p. 415604. Tang, X.-F., Z.-G. Yang, and W.-J. Wang, A simple way of preparing highconcentration and high-purity nano copper colloid for conductive ink in inkjet printing technology. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2010. 360(1–3): p. 99-104. Kang, H., E. Sowade, and R.R. Baumann, Direct Intense Pulsed Light Sintering of Inkjet-Printed Copper Oxide Layers within Six Milliseconds. ACS Applied Materials & Interfaces, 2014. 6(3): p. 1682-1687.

51

84. 85. 86.

87.

88. 89. 90. 91.

92. 93.

94.

95. 96. 97. 98.

99.

100.

Paquet, C., et al., Photosintering and electrical performance of CuO nanoparticle inks. Organic Electronics, 2014. 15(8): p. 1836-1842. Deegan, R.D., et al., Capillary flow as the cause of ring stains from dried liquid drops. Nature, 1997. 389(6653): p. 827-829. Smith, P., et al., Direct ink-jet printing and low temperature conversion of conductive silver patterns. Journal of Materials Science, 2006. 41(13): p. 41534158. Ihalainen, P., et al., Influence of Surface Properties of Coated Papers on Printed Electronics. Industrial & Engineering Chemistry Research, 2012. 51(17): p. 6025-6036. Bollström, R., et al., A multilayer coated fiber-based substrate suitable for printed functionality. Organic Electronics, 2009. 10(5): p. 1020-1023. Chan, C.M., T.M. Ko, and H. Hiraoka, Polymer surface modification by plasmas and photons. Surface Science Reports, 1996. 24(1–2): p. 1-54. van Osch, T.H.J., et al., Inkjet Printing of Narrow Conductive Tracks on Untreated Polymeric Substrates. Advanced Materials, 2008. 20(2): p. 343-345. International Organization for, S., Information technology - office equipment measurement of image quality attributes for hardcopy output - binary monochrome text and graphic images. 1996: ISO. Pawlow, P., Z. Phys. Chem, 1909. 65(1). Perelaer, B.J., et al., Inkjet-printed silver tracks: low temperature curing and thermal stability investigation. Journal of Materials Chemistry, 2008. 18(27): p. 3209-3215. Li, S., P. Liu, and Q. Wang, Study on the effect of surface modifier on selfaggregation behavior of Ag nano-particle. Applied Surface Science, 2012. 263(0): p. 613-618. Magdassi, S., et al., Triggering the Sintering of Silver Nanoparticles at Room Temperature. ACS Nano. 4(4): p. 1943-1948. Mark, L.A., et al., Electrical sintering of nanoparticle structures. Nanotechnology, 2008. 19(17): p. 175201. Sueyoshi, H. and S. Kakiuchi, Microwave Heating of Thin Au Film. MATERIALS TRANSACTIONS, 2007. 48(3): p. 531-537. Cao, Z., N. Yoshikawa, and S. Taniguchi, Microwave heating behavior of nanocrystalline Au thin films in single-mode cavity. Journal of Materials Research, 2009. 24: p. 268-273. Perelaer, J., B.J. de Gans, and U.S. Schubert, Ink-jet Printing and Microwave Sintering of Conductive Silver Tracks. Advanced Materials, 2006. 18(16): p. 2101-2104. Grouchko, M., et al., Conductive Inks with a “Built-In” Mechanism That Enables Sintering at Room Temperature. ACS Nano, 2011. 5(4): p. 3354-3359.

52

101.

102.

103. 104. 105.

106. 107.

108. 109. 110. 111.

112.

113.

114. 115.

116.

Zapka, W., et al., Low Temperature Chemical Post-Treatment of Inkjet Printed Nano-Particle Silver Inks. NIP & Digital Fabrication Conference, 2008. 2008(2): p. 906-911. Long, Y., et al., Rapid sintering of silver nanoparticles in an electrolyte solution at room temperature and its application to fabricate conductive silver films using polydopamine as adhesive layers. Journal of Materials Chemistry, 2011. 21(13): p. 4875-4881. Bomke, H.A., et al., Annealing of ion‐implanted silicon by an incoherent light pulse. Applied Physics Letters, 1978. 33(11): p. 955-957. Loisel, B., et al., Flash lamp crystallization of amorphous silicon films on glass substrates. Thin Solid Films, 1984. 117(2): p. 117-123. Lue, J.T., Formation of nickel and palladium silicides by a short-pulse light-flash and its application in the metallization of solar cells. Solid-State Electronics, 1983. 26(8): p. 787-793. Pécz, B., et al., Crystallization of amorphous-Si films by flash lamp annealing. Applied Surface Science, 2005. 242(1–2): p. 185-191. Smith, M., et al., Modelling of flash-lamp-induced crystallization of amorphous silicon thin films on glass. Journal of Crystal Growth, 2005. 285(1–2): p. 249260. Ohdaira, K., et al., Explosive crystallization of amorphous silicon films by flash lamp annealing. Journal of Applied Physics, 2009. 106(4): p. -. McMahon, R.A., et al., Flash-lamp annealing of semiconductor materials— Applications and process models. Vacuum, 2007. 81(10): p. 1301-1305. Schroder, K.A., S.C. McCool, and W.F. Furlan, Broadcast Photonic Curing of Metallic Nanoparticle Films in NSTI-Nanotech 2006. 2006. p. 198-201. Abbel, R., et al., Photonic flash sintering of silver nanoparticle inks: a fast and convenient method for the preparation of highly conductive structures on foil. MRS Communications, 2012. 2(04): p. 145-150. Perelaer, J., et al., Roll-to-Roll Compatible Sintering of Inkjet Printed Features by Photonic and Microwave Exposure: From Non-Conductive Ink to 40% Bulk Silver Conductivity in Less Than 15 Seconds. Advanced Materials, 2012. 24(19): p. 2620-2625. Yung, K.C., et al., Ink-jet printing and camera flash sintering of silver tracks on different substrates. Journal of Materials Processing Technology, 2010. 210(15): p. 2268-2272. Tobjörk, D., et al., IR-sintering of ink-jet printed metal-nanoparticles on paper. Thin Solid Films, 2012. 520(7): p. 2949-2955. Sung-Jun, J., H. Hyun-Jun, and K. Hak-Sung, Highly conductive copper nano/microparticles ink via flash light sintering for printed electronics. Nanotechnology, 2014. 25(26): p. 265601. Park, S.-H. and H.-S. Kim, Flash light sintering of nickel nanoparticles for printed electronics. Thin Solid Films, 2014. 550(0): p. 575-581.

53

117.

118.

Dharmadasa, R., et al., Room Temperature Synthesis of a Copper Ink for the Intense Pulsed Light Sintering of Conductive Copper Films. ACS Applied Materials & Interfaces, 2013. 5(24): p. 13227-13234. Ryu, J., H.-S. Kim, and H.T. Hahn, Reactive Sintering of Copper Nanoparticles Using Intense Pulsed Light for Printed Electronics. Journal of Electronic Materials, 2011. 40(1): p. 42-50.

54

Suggest Documents