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-to-
roll 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-rulle-
processer 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 PROBLEM FORMULATION ........................................................................................2

1.2 MAIN CONTRIBUTION OF THE THESIS.......................................................................3
2 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 PAPER ...................................................................................................................14

4.2 POLYMER FILMS ....................................................................................................17
5 CONDUCTOR PERFORMANCE EVALUATION ............................................19

5.1 PATTERN DEFINITION ............................................................................................19

5.1.1 Line width .....................................................................................................21
5.1.2 Raggedness ...................................................................................................21

xi
 5.2 ELECTRICAL CONDUCTIVITY .................................................................................21
5.2.1 4-point electrical resistance measurements .................................................22
6 SINTERING ..............................................................................................................23

6.1 OVEN SINTERING ...................................................................................................24

6.2 ELECTRICAL SINTERING ........................................................................................25
6.3 MICROWAVE SINTERING .......................................................................................25
6.4 CHEMICAL SINTERING ...........................................................................................26
6.5 INTENSE PULSED LIGHT SINTERING........................................................................27
7 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 Atomic Force Microscopy

AgNP Silver nanoparticle
CIJ Continuous Inkjet
CuNP Copper nanoparticle
CuO Copper(II) Oxide
DLS Dynamic Light Scattering
DOD Drop On Demand Inkjet
IPL Intense Pulsed Light
LWC Lightweight Coated
MEMS Micro-electromechanical systems
NP Nanoparticle
OLED Organic Light-Emitting Diode
PCA Principal Component Analysis
PDMS Polydimethylsiloxane
PE Polyethylene
PI Polyimide
PLS Projection to Latent Structures (Partial Least Squares regression)
P-OLED Polymer Organic Light-Emitting Diode
PVP Polyvinylpyrrolidone, water-soluble polymer
PPS Parker Print Surf, measurement method for surface roughness
RFID Radio Frequency Identification
SEM Scanning Electron Microscopy
SEM-EDS SEM - Energy Dispersive X-ray Spectroscopy
TEM Transmission Electron Microscopy
ZnO 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 inkjet-
printed 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 inkjet-
printed 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-on-
demand 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 heat-
sensitive 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 Viscosity range (mPa·s)

Thermal inkjet 1-5
Piezo inkjet 5-30
Gravure 50-200
Flexography 50-500
Screen 1000-10000
Offset 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 particle-
particle 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 5-
50 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 temperature-
sensitive 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 latex-
suspension, 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. a-
b) 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, non-
existent 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 de-
stabilisation 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 ion-
implanted 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

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 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 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 Electric IPL sintering Microwave

sintering sintering sintering

Conductivity 15/18 50/40 40/30 40/35

% of bulk Ag
(Photo/LC)

Limiting Substrate Contact Delamination Local

factor deformation method overheating

Time scale 1 – 10 minutes 0.1 - 1 s 0.1 – 10 ms 1 – 10 s

Feedback Neutral Positive Negative Positive

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 roll-
to-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 solution-
processed 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 inkjet-
printable 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 dispersion-
stabilizing 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 IV-
VI 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, custom-
designed 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 ink-
substrate interaction and process development for IPL processing, could possibly
solve these issues.

45
46
13 REFERENCES

1. Meggs, P.B., A history of graphic design. 1998: John Wiley & Sons.
2. 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.
3. Hsien-Hsueh, L., C. Kan-Sen, and H. Kuo-Cheng, Inkjet printing of nanosized
silver colloids. Nanotechnology, 2005. 16(10): p. 2436.
4. 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.
5. Kordás, K., et al., Inkjet Printing of Electrically Conductive Patterns of Carbon
Nanotubes. Small, 2006. 2(8-9): p. 1021-1025.
6. Jang, J., Conducting Polymer Nanomaterials and Their Applications
Emissive Materials Nanomaterials. 2006, Springer Berlin / Heidelberg. p. 189-
260.
7. Li, D., et al., Conductive copper and nickel lines via reactive inkjet printing.
Journal of Materials Chemistry, 2009. 19(22): p. 3719-3724.
8. Park, B.K., et al., Direct writing of copper conductive patterns by ink-jet
printing. Thin Solid Films, 2007. 515(19): p. 7706-7711.
9. Wang, J.Z., et al., Low-cost fabrication of submicron all polymer field effect
transistors. Applied Physics Letters, 2006. 88(13): p. 133502-3.
10. Garnier, F., et al., All-Polymer Field-Effect Transistor Realized by Printing
Techniques. Science, 1994. 265(5179): p. 1684-1686.
11. Li, J., et al., A stable solution-processed polymer semiconductor with record high-
mobility for printed transistors. Sci. Rep., 2012. 2.
12. 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.
13. Calvert, P., Inkjet Printing for Materials and Devices. Chemistry of Materials,
2001. 13(10): p. 3299-3305.
14. Sirringhaus, H., et al., High-Resolution Inkjet Printing of All-Polymer
Transistor Circuits. Science, 2000. 290(5499): p. 2123-2126.
15. 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.
16. Sambucetti, C.J., Magnetic ink for jet printing. Magnetics, IEEE Transactions
on, 1980. 16(2): p. 364-367.
17. 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.

47
18. Patel, N.G., et al., Screen-printed biosensors using different alcohol oxidases.
Sensors and Actuators B: Chemical, 2001. 75(1–2): p. 101-110.
19. 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.
20. Xu, T., et al., Viability and electrophysiology of neural cell structures generated
by the inkjet printing method. Biomaterials, 2006. 27(19): p. 3580-3588.
21. Xu, T., et al., Inkjet printing of viable mammalian cells. Biomaterials, 2005.
26(1): p. 93-99.
22. 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.
23. 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.
24. Mabrook, M.F., C. Pearson, and M.C. Petty, An inkjet-printed chemical fuse.
Applied Physics Letters, 2005. 86(1): p. 013507-013507-3.
25. 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.
26. 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.
27. MacFarlane, D.L., et al., Microjet fabrication of microlens arrays. Photonics
Technology Letters, IEEE, 1994. 6(9): p. 1112-1114.
28. Jahn, S.F., et al., Polymer Microsieves Manufactured by Inkjet Technology.
Langmuir, 2008. 25(1): p. 606-610.
29. 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.
30. Li, Y., et al., RFID Tag and RF Structures on a Paper Substrate Using Inkjet-
Printing Technology. Microwave Theory and Techniques, IEEE Transactions
on, 2007. 55(12): p. 2894-2901.
31. 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.
32. 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.
33. Hyora, S., et al., An electrothermochromic paper display based on
colorimetrically reversible polydiacetylenes. Nanotechnology, 2014. 25(9): p.
094011.
34. Shimoda, T., et al., Inkjet Printing of Light-Emitting Polymer Displays. MRS
Bulletin, 2003. 28(11): p. 821-827.

48
35. Comiskey, B., et al., An electrophoretic ink for all-printed reflective electronic
displays. Nature, 1998. 394(6690): p. 253-255.
36. 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.
37. 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.
38. Sarfraz, J., et al., A printed H2S sensor with electro-optical response. Sensors
and Actuators B: Chemical, 2014. 191(0): p. 821-827.
39. Liana, D.D., et al., Recent Advances in Paper-Based Sensors. Sensors, 2012.
12(9): p. 11505-11526.
40. 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.
41. Sekitani, T., et al., Printed Nonvolatile Memory for a Sheet-Type
Communication System. Electron Devices, IEEE Transactions on, 2009. 56(5):
p. 1027-1035.
42. Lian, K., et al., Printed flexible memory devices using copper phthalocyanine.
Materials Science and Engineering: B, 2010. 167(1): p. 12-16.
43. Lien, D.-H., et al., All-Printed Paper Memory. ACS Nano, 2014. 8(8): p. 7613-
7619.
44. 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.
45. 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.
46. Gaikwad, A.M., et al., A flexible high potential printed battery for powering
printed electronics. Applied Physics Letters, 2013. 102(23): p. -.
47. 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.
48. 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.
49. Nakano, S., et al., Low operation voltage of inkjet-printed plastic sheet-type
micromechanical switches. Applied Physics Letters, 2008. 92(5): p. -.
50. Kopola, P., et al., Gravure printed organic light emitting diodes for lighting
applications. Thin Solid Films, 2009. 517(19): p. 5757-5762.
51. Villani, F., et al., Inkjet Printed Polymer Layer on Flexible Substrate for OLED
Applications. The Journal of Physical Chemistry C, 2009. 113(30): p. 13398-
13402.

49
52. 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.
53. 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.
54. Tobjörk, D. and R. Österbacka, Paper Electronics. Advanced Materials.
23(17): p. 1935-1961.
55. Kamyshny, A., Metal-based Inkjet Inks for Printed Electronics. The Open
Applied Physics Journal, 2011. 4: p. 19-36.
56. C, C.R. and S.R. G, Fluid droplet recorder with a plurality of jets. 1968, Google
Patents.
57. 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.
58. Ross R. Allen, J.D.M., and William R. Knight Thermodynamics and
Hydrodynamics of Thermal Ink Jets. Hewlett-Packard Journal, 1985(5): p. 21-
27.
59. Setti, L., et al., An amperometric glucose biosensor prototype fabricated by
thermal inkjet printing. Biosensors and Bioelectronics, 2005. 20(10): p. 2019-
2026.
60. 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.
61. Roda, A., et al., Protein microdeposition using a conventional ink-jet printer.
BioTechniques, 2000. 28(3): p. 492-496.
62. Chen, P.-H., et al., Pressure response and droplet ejection of a piezoelectric inkjet
printhead. International Journal of Mechanical Sciences, 1999. 41(2): p. 235-
248.
63. 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.
64. 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.
65. Wijshoff, H., The dynamics of the piezo inkjet printhead operation. Physics
Reports, 2010. 491(4–5): p. 77-177.
66. Kipphan, H., Handbook of Print Media: Technologies and Production Methods.
2001: Springer.
67. Holmberg, K. and B. Jönsson, Surfactants and Polymers in Aqueous Solution.
2003: John Wiley & Sons.
68. Magdassi, S., The Chemistry of Inkjet Inks. 2010: World Scientific.

50
69. 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.
70. 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.
71. Rahaman, M.N., Ceramic Processing and Sintering. 2003: Taylor & Francis.
72. Studart, A.R., E. Amstad, and L.J. Gauckler, Colloidal Stabilization of
Nanoparticles in Concentrated Suspensions. Langmuir, 2006. 23(3): p. 1081-
1090.
73. Fritz, G., et al., Electrosteric Stabilization of Colloidal Dispersions. Langmuir,
2002. 18(16): p. 6381-6390.
74. 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.
75. 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. 453-
459.
76. 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.
77. 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.
78. 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.
79. 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.
80. 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.
81. 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.
82. Tang, X.-F., Z.-G. Yang, and W.-J. Wang, A simple way of preparing high-
concentration 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.
83. 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. Paquet, C., et al., Photosintering and electrical performance of CuO nanoparticle
inks. Organic Electronics, 2014. 15(8): p. 1836-1842.
85. Deegan, R.D., et al., Capillary flow as the cause of ring stains from dried liquid
drops. Nature, 1997. 389(6653): p. 827-829.
86. Smith, P., et al., Direct ink-jet printing and low temperature conversion of
conductive silver patterns. Journal of Materials Science, 2006. 41(13): p. 4153-
4158.
87. Ihalainen, P., et al., Influence of Surface Properties of Coated Papers on Printed
Electronics. Industrial & Engineering Chemistry Research, 2012. 51(17): p.
6025-6036.
88. Bollström, R., et al., A multilayer coated fiber-based substrate suitable for printed
functionality. Organic Electronics, 2009. 10(5): p. 1020-1023.
89. 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.
90. 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.
91. International Organization for, S., Information technology - office equipment -
measurement of image quality attributes for hardcopy output - binary
monochrome text and graphic images. 1996: ISO.
92. Pawlow, P., Z. Phys. Chem, 1909. 65(1).
93. 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.
94. Li, S., P. Liu, and Q. Wang, Study on the effect of surface modifier on self-
aggregation behavior of Ag nano-particle. Applied Surface Science, 2012.
263(0): p. 613-618.
95. Magdassi, S., et al., Triggering the Sintering of Silver Nanoparticles at Room
Temperature. ACS Nano. 4(4): p. 1943-1948.
96. Mark, L.A., et al., Electrical sintering of nanoparticle structures.
Nanotechnology, 2008. 19(17): p. 175201.
97. Sueyoshi, H. and S. Kakiuchi, Microwave Heating of Thin Au Film.
MATERIALS TRANSACTIONS, 2007. 48(3): p. 531-537.
98. 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.
99. 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.
100. 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. 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.
102. 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.
103. Bomke, H.A., et al., Annealing of ion‐implanted silicon by an incoherent light
pulse. Applied Physics Letters, 1978. 33(11): p. 955-957.
104. Loisel, B., et al., Flash lamp crystallization of amorphous silicon films on glass
substrates. Thin Solid Films, 1984. 117(2): p. 117-123.
105. 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.
106. Pécz, B., et al., Crystallization of amorphous-Si films by flash lamp annealing.
Applied Surface Science, 2005. 242(1–2): p. 185-191.
107. 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. 249-
260.
108. Ohdaira, K., et al., Explosive crystallization of amorphous silicon films by flash
lamp annealing. Journal of Applied Physics, 2009. 106(4): p. -.
109. McMahon, R.A., et al., Flash-lamp annealing of semiconductor materials—
Applications and process models. Vacuum, 2007. 81(10): p. 1301-1305.
110. 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.
111. 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.
112. 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.
113. 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.
114. Tobjörk, D., et al., IR-sintering of ink-jet printed metal-nanoparticles on paper.
Thin Solid Films, 2012. 520(7): p. 2949-2955.
115. 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.
116. 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. 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.
118. 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