FIELD OF THE INVENTION

This invention relates to printed electronics, and in particular, to a method of sintering a printed pattern comprising metallic-particle ink printed on a substrate and an apparatus for performing the method of sintering.

BACKGROUND OF THE INVENTION

Metallic-particle inks are widely used for fabricating conductive tracks in 3D printed electronics and have significantly better electrical conductivity compared to other types of conductive inks such as metallic-organic decomposition (MOD) and conductive polymer inks. Metallic-particle inks are dispersions of metallic particles (which may be nanoparticles or microparticles) in either organic or aqueous solvents. Typical metallic-particle inks include those comprising (but not limited to) silver, gold, copper, aluminium or nickel nanoparticles. The metallic particles in the inks are coated with organic additives and stabilising agents to prevent agglomeration of the metallic particles in the inks and help promote better printability and ink stability. However, these organic additives and stabilising agents act as an electrical insulator and prevent contact of neighbouring metallic particles for electron flow between the metallic particles that is needed to effect electrical conductivity. Therefore, metallic particles inks are initially electrically non-conductive when they are first deposited on a substrate as printed patterns.

A post-deposition process, also known as sintering, is essential to introduce electrical conductance within the printed patterns composed of the deposited metallic particles. Sintering is a critical process as it can determine the final electrical conductivity of the printed patterns. During sintering, energy is introduced to the printed patterns via various means to first decompose the encapsulating organic additives and stabilising agents around the metallic particles and then aggregate neighbouring metallic particles into a single continuous entity. The printed patterns should experience three different sequential stages in an ideal sintering process: initial contact with neighbouring metallic particles, formation of sintering necks, and grain growth and densification. The first stage of sintering is to first decompose the organic additives and stabilising agents that envelope the metallic particles [25] Organic shells around the metallic particles are removed during this stage, so that neighbouring metallic particles can form contacting points among themselves [14, 26] Electrons can now flow freely throughout the entire printed patterns as the neighbouring metallic particles make physical contact with each other [14, 27] However, electrical resistivity of the printed patterns is still very high [25] Note that shrinkage in the volume of the printed patterns can also be observed during the first stage of the sintering process [14] At the second stage of sintering, sinter necks are formed between neighbouring metallic particles. This stage is also known as the aggregation of the metallic particles [28] Triggered by the diffusive [25] Ostwald ripening phenomenon [29] and the need to reduce total interfacial energy [30], contacting metallic particles are able to form sinter necks between each other at elevated temperatures. These metallic particles, with the aid of the introduced energy, release high surface energy to achieve a state of thermodynamic equilibrium and aggregate into larger particles [25] through the formation of sinter necks [31] However, the Ostwald ripening phenomenon is not perpetual and this process will end when diameter of the particles is approximately 1.5 times larger than the original particle size [25] At this stage, the printed patterns have a highly porous structure and their electrical conductivities are still not optimised [25] In the third stage of sintering, inter-particle atomic diffusion and grain growth are observed [14], where the additional input of energy and sintering time promote grain growth and densification [32, 33] It is also interesting to note that grain boundaries within the sintered structures affect electrical conductance [28] Ingham et al. observed that grain growth took substantially longer than aggregation of the metallic particles [28] In addition, densification is also affected by grain growth kinetics and initial grain size [33] After sintering, electrons can flow throughout the entire printed patterns to conduct electricity, where all the individual metallic particles are integrated into a single continuous entity to form a continuous path, having electrical conductivity comparable to that of bulk material.

However, sintering is not favourable to substrates that are not heat resistant, as these substrates are likely to be damaged, deformed, warped or deteriorated if they are exposed to high temperatures for a long period. Hence, the sintering process has become a limiting factor for‘fully additive’ printed electronics, as it limits the choice of substrates that can be used. For example, low heat resistance substrates such as paper, textile, polyethylene terephthalate (PET), polycarbonate (PC), are not suitable for undergoing currently available sintering processes. More costly substrates such as polyimide, glass, ceramics that are heat resistant would be required to endure available sintering processes without damage. Furthermore, sintering processes that utilise electromagnetic radiation as a heating source (e.g. intense pulse light sintering, infrared (IR) sintering, laser sintering, ultraviolet sintering and microwave sintering) have been shown to have limited penetration depth for sintering metallic particles. For example, IR sintering has a negative feedback system, in which light absorption rates of silver nanoparticles in a printed pattern slow down as they aggregate into larger particles. Silver nanoparticles turn silvery in colour as aggregation occurs and reflect the IR electromagnetic waves away. Although this negative feedback system might be helpful for preventing the printed patterns from overheating, it can also prevent the bottommost layers from getting sintered especially if the printed patterns are very thick. Therefore, IR sintering may not be able to sinter all silver nanoparticles thoroughly in thicker printed patterns. This results in incomplete sintering at the bottom layers of the printed patterns, resulting in many areas of high resistivity near the bottom. Scattered areas of high resistivity in the conductive printed patterns are not desired as it may result in points of high heat concentration during usage, hence affecting their electrical performance. Furthermore, substrates may be damaged if the substrates are overexposed to high-intensity IR electromagnetic waves.

Induction heating is a heating technique especially used for heating electrically conductive materials directly through electromagnetic induction, typically through two main mechanisms: eddy currents losses and hysteresis losses [35-38] Hysteresis losses are losses that only occur in ferromagnetic materials, which is caused by alternating magnetic field. The alternating magnetic field forces the magnetic domains in ferromagnetic materials to continuously rotate back and forth and these repeated movements internally generate heat within the materials. However, hysteresis losses have insignificant contributions to the effects of induction heating especially in compact materials [36] and they are normally neglected in calculations and hence, hysteresis losses are not discussed in further detail here. The main contributor to the heating effects of induction heating is due to eddy current losses. In an induction heating system, an alternating voltage is applied to an induction coil to generate an alternating current within the coil, which in return generates an alternating magnetic field through the coil. The alternating magnetic field interacts with the conductive workpiece and induces eddy currents within the workpiece. Note that the alternating current, magnetic field and as well as the eddy currents all have the same frequency. The induced eddy currents hence generate a heating effect within the conductive workpiece through the Joule heating effects (also known as eddy current losses).

However, while the principles of induction heating may appear simple, it is crucial to understand the electromagnetic properties of the conductive workpiece (for instance electrical conductivity and relative magnetic permeability) to maximise the heating efficiency of induction heating. The key to efficient and effective heating of workpieces in induction heating is dependent on two main factors: materials properties of the workpiece and the frequency of the electromagnetic field that produces the eddy currents.

Skin Effect and Penetration Depth

As alternating current passes through the induction coil, the bulk of the current concentrates on the surfaces of the induction coil, also known as the skin effect [35-38] Skin effect is a phenomenon that demonstrates the non-uniformity of the alternating current distribution across the cross-section of the conductor. The skin effect phenomenon is also evident on the induction heated workpiece, whereas the induced eddy current concentrates on workpieces’ surfaces. 86% of the electrical power is concentrated on the surfaces of the workpiece and induces heating of the workpiece for a certain depth from the surface, also known as the penetration depth, d. The penetration depth is highly influenced by the electromagnetic properties of the workpiece and as well as the frequency of the eddy current (frequency of the eddy current is the same as the frequency of the alternating current that passes through the induction coil). The formula of penetration depth [37], d is defined as

where <5 is the penetration depth, p is the electrical resistivity of the workpiece material, p _r is the relative magnetic permeability of the workpiece material and f is the frequency of the induced eddy current in the workpiece. From equation (1) above, it can be observed that the penetration depth is a function of the electrical resistivity and relative magnetic permeability of the material and the frequency of the induced eddy current in the workpiece. However, during the process of induction heating, the electrical resistivity of the material may increase as its temperature increases and the material’s electrical resistivity can be expressed as a function of temperature shown below: p(T) = po [1 + (T _m - To)] (2) where T _m is the temperature of material, T _Q is the ambient temperature, p(T) is the electrical resistivity at temperature T, p _o is the electrical resistivity of material at temperature To, a is the temperature coefficient of electrical resistivity of the material. Note that during the induction heating process, increase in the temperature of the workpiece results in the increase of the electrical resistivity of its material and hence increases the penetration depth. Therefore, it can be deduced that the penetration depth is also a function of temperature and frequency of the induced eddy current in the workpiece and can be expressed as

In printed electronics, there are a few types of metallic particles inks that are of great interests to researchers and industries. These are silver, gold, copper, aluminium and nickel. Typical induction heating equipment in today’s market operates in the range from 10 kHz - 15 MHz. FIG. 1 shows penetration depths of five different metals at various operating frequencies ranging from 10 kHz - 15 MHz: silver, gold, copper, aluminium and nickel, and is plotted using the governing equation of the penetration depth (equation (1) above). These metals have different values of electrical resistivity and relative magnetic permeability as shown in Table 1 [39] below.

Metals Electrical Resistivity Relative Magnetic

(W-m) Permeability

Silver 1.586 x 10- ⁸ 0.99998

Gold 2.240 x 10 ⁸ 1

Copper 1.678 x 10 ⁸ 1.00002

Aluminium 2.655 x 10 ⁸ 0.999991

Nickel 6.840 x 10 ⁸ 600

Table 1 For effective induction heating, the diameter of the workpiece must not be smaller than the penetration depth as there will be poor power absorption and low heating efficiency due to the skin effect phenomenon. If the diameter of the workpiece is smaller than the penetration depth, the workpiece is prone to poor heating efficiency as there are prominent current cancelling effects in the workpiece and hence causing trivial energy absorption and Joule’s heating effects. From FIG. 1 , a governing relationship can be observed that the penetration depth decreases with increasing operating frequencies of the induction heater. From the above, it can be appreciated that induction heating is inadequate for sintering printed patterns comprising metallic-particle inks. This is because as-deposited printed patterns are not electrically conductive and eddy currents are not able to be induced in them to generate the necessary heating effect to sinter the metallic particles in the printed patterns. Although individual metallic particles within the printed patterns are electrically conductive, it is technically challenging to have an induction heater to individually heat these metallic particles due to the skin effect phenomenon. For instance, to effectively heat an individual silver nanoparticle of 50 nm average diameter, the penetration depth must be at least 50 nm. From equation (1) above, it can be calculated that to obtain a penetration depth of 50 nm, the induction heater is required to operate at a frequency of 1605 GHz. It is technically impossible to have an induction heater that is able to operate at such high frequencies. Current induction heating equipment normally operates at frequencies between 10 kHz - 15 MHz. Hence, heating individual nanoparticles at this operating frequency is extremely inefficient as there will be extremely poor power absorption and low heating efficiency.

Thus, although various different sintering techniques are currently available, none of them can achieve fast and selective sintering at a low sintering temperature to give the sintered patterns good electrical conductivity while minimizing damage to both the substrates and printed patterns.

SUMMARY OF INVENTION

A method of sintering in printed electronics is disclosed. In printed electronics, metallic particles in a metallic-particle ink are deposited or printed on a planar surface of a substrate. The printed metallic particles form a continuous path with an exposed surface and a bottom surface which is in contact with the planar surface of the substrate. The metallic particles are normally coated with a thermally decomposable, non-conductive outer layer and hence, the printed continuous path has limited conductivity. In the presently disclosed method of sintering, the printed metallic particles are first subjected to a non-contact heat source to thermally decompose at least part of the non-conductive outer layer to form an electrically conductive path at the exposed surface. The non-contact heat source introduces an initial electrical conductivity within the printed patterns. The electrically conductive path is then subjected to an induction field wherein heat is generated via induction heating of the electrically conductive path which further heats and sinters adjacent layers of un-sintered metallic particles. The induction step further sinters the printed patterns as they are electrically conductive after exposure to the non-contact heat source. Eddy currents are induced on the conductive printed patterns and generate Joule’s heating effect within the printed patterns to sinter at least most (if not all) of the printed metallic particles.

The disclosed method of sintering in printed electronics has numerous advantages, including: (i) selective sintering where only electrically conductive printed patterns will be sintered, (ii) minimum damage caused to substrates and printed patterns, (iii) short sintering time, (iv) good electrical conductivity, (v) non-contact sintering and (vi) good surface morphologies.

According to a first aspect, there is provided a method of sintering a printed pattern, the printed pattern comprising a metallic-particle ink deposited on a substrate, the method comprising the steps of:

(a) exposing the printed pattern to a non-contacting heat source to form an electrically conductive path of at least partially sintered metallic particles on an exposed surface of the printed pattern; and

(b) subjecting the electrically conductive path to an alternating magnetic field to generate heat via induction heating of the electrically conductive path to sinter metallic particles in the printed pattern in a direction from the electrically conductive path towards the substrate.

Exposing the printed pattern to the infrared irradiation may be for a duration ranging from 0.5 minutes to 4 minutes.

Forming the electrically conductive path may comprise infrared sintering of the metallic particles on the exposed surface at a temperature ranging from 120 °C to 150 °C.

Subjecting the electrically conductive path to an alternating magnetic field may comprise operating an induction heater at 5 kW at a frequency of 1.5 MHz.

According to a second aspect, there is provided an apparatus for sintering printed patterns, each printed pattern comprising a metallic-particle ink printed on a substrate, the apparatus comprising:

a conveyor belt to support thereon and to move the printed patterns;

a non-contacting heat source provided above the conveyor belt to form an electrically conductive path of at least partially sintered metallic particles on an exposed surface of each printed pattern when each printed pattern is positioned under the non contacting heat source by the conveyor belt; and

induction coils provided above the conveyor belt subsequent to the non-contacting heat source to provide an alternating magnetic field that generates heat via induction heating of the electrically conductive path to sinter metallic particles in each printed pattern in a direction from the electrically conductive path towards the substrate when each printed pattern is positioned under the induction coils by the conveyor belt.

The induction coil may be configured to provide the alternating magnetic field in pulses.

For both aspects, the non-contacting heat source may comprise infrared irradiation. The alternating magnetic field may be provided in pulses. The metallic-particle ink may comprise silver nanoparticles. The substrate may comprise polyimide.

BRIEF DESCRIPTION OF FIGURES

In order that the invention may be fully understood and readily put into practical effect there shall now be described by way of non-limitative example only exemplary embodiments of the present invention, the description being with reference to the accompanying illustrative drawings.

FIG. 1 (prior art) is a graph of induction heating penetration depths for different materials at different operating frequencies of the induction heater.

FIG. 2 is flow chart of a two-step method of sintering for printed electronics.

FIG. 3a is a schematic side view of a printed pattern.

FIG. 3b is a schematic top view of a printed pattern sample.

FIG. 4a is a schematic illustration of an exemplary embodiment of a first step of infrared sintering in the method of FIG. 2.

FIG. 4b to 4d is a schematic illustration of an exemplary embodiment of a second step of induction sintering in the method of FIG. 2.

FIG. 5 is a boxplot of twelve thermally sintered samples.

FIG. 6 is a boxplot of samples sintered by thermal sintering, infrared sintering and the

present two-step method of sintering.

FIG. 7 is FESEM images of an unsintered printed pattern at different magnifications.

FIG. 8a is FESEM images of printed pattern sample number 1 sintered by thermal sintering at 120 °C for 30 min at different magnifications. FIG. 8b is FESEM images of printed pattern sample number 2 sintered by thermal sintering at 120 °C for 60 min at different magnifications.

FIG. 8c is FESEM images of printed pattern sample number 3 sintered by thermal sintering at 120 °C for 120 min at different magnifications.

FIG. 8d is FESEM images of printed pattern sample number 4 sintered by thermal sintering at 120 °C for 240 min at different magnifications.

FIG. 9a is FESEM images of printed pattern sample number 5 sintered by thermal sintering at 150 °C for 30 min at different magnifications.

FIG. 9b is FESEM images of printed pattern sample number 6 sintered by thermal sintering at 150 °C for 60 min at different magnifications.

FIG. 9c is FESEM images of printed pattern sample number 7 sintered by thermal sintering at 150 °C for 120 min at different magnifications.

FIG. 9d is FESEM images of printed pattern sample number 8 sintered by thermal sintering at 150 °C for 240 min at different magnifications.

FIG. 10a is FESEM images of printed pattern sample number 9 sintered by thermal sintering at 200 °C for 30 min at different magnifications.

FIG. 10b is FESEM images of printed pattern sample number 10 sintered by thermal

sintering at 200 °C for 60 min at different magnifications.

FIG. 10c is FESEM images of printed pattern sample number 11 sintered by thermal

sintering at 200 °C for 120 min at different magnifications.

FIG. 10d is FESEM images of printed pattern sample number 12 sintered by thermal

sintering at 200 °C for 240 min at different magnifications.

FIG. 11 is FESEM images of printed pattern sample number 13 sintered by IR sintering at different magnifications.

FIG. 12 is FESEM images of printed pattern sample number 14 sintered by the present two- step method of sintering at different magnifications.

FIG. 13 is a graph of temperature profile of a printed pattern during induction sintering in pulses.

FIG. 14 is a schematic illustration of an apparatus for performing the method of sintering of FIG. 2.

FIG. 15 is a schematic illustration of a set-up for performing IR sintering.

FIG. 16 is a schematic cross-sectional view of the set-up of FIG. 15.

FIG. 17 is a schematic illustration of a set-up for performing induction sintering. DETAILED DESCRIPTION

Exemplary embodiments of a method of sintering 10 a printed pattern for printed electronics and an apparatus 100 for performing the method 10 will be described below with reference to FIGS. 2 to 17.

As shown in FIGS. 2 and 4, in general, there is disclosed a method of sintering 10 a printed pattern for printed electronics comprises a first step 20 of exposing a printed pattern 11 comprising a metallic-particle ink deposited on a substrate 18 to a non-contacting heat source 22 to form an electrically conductive path 24 of at least partially sintered metallic particles 12S on an exposed surface of the printed pattern 11 , and a second step 30 of subjecting the electrically conductive path 24 to an alternating magnetic field 32 to generate heat via induction heating of the electrically conductive path 24 to sinter metallic particles 12 in the printed pattern 11 in a direction from the electrically conductive path 24 towards the substrate 18.

Exposure 20 to the non-contact heat source 22 should comprise a short sintering time (e.g. 0.5 min to 4 min) and low sintering temperature (e.g. 120 °C to 150 °C) such that the substrate 18 is not affected. This step may comprise thermal sintering, intense pulse light sintering, infrared sintering, laser sintering, ultraviolet sintering, microwave sintering, low pressure plasma sintering, localised atmospheric plasma sintering and so on. The temperature range for the non-contact heat source 22 used in the first step 20 is highly dependent on the decomposition temperatures of the organic additives 14 in the metallic- particle ink 11 and the mechanical and optical properties of the substrate 18. The temperature range should ideally be set at a temperature higher than the decomposition temperatures of the organic additives 14 the in metallic-particle ink 11 but lower than the glass transition temperatures of the substrates 18. The actual determination of the temperature range and heating duration can be done empirically through design of experiments (DOE) or through computer simulations as ink suppliers typically do not disclose the type of organic additives used in metallic-particle inks.

The preferred non-contact heat source 22 is highly dependent on the metallic-particle inks and substrates used. The selected non-contact heat source 22 should provide the most effective sintering effect on the metallic-particle ink to form the electrically conductive path 24 but with the least damaging effects on the substrates. Bulk of the organic additives in the printed patterns should be vaporized in the first step 20 of the method 10 but without damaging the substrate 18, in order to prevent defects in the sintered printed patterns after the second step 30 of the two-step sintering method 10 has been completed.

The electrically conductive path 24 is a topmost layer of the printed pattern 11. Therefore, subsequent exposure of the printed pattern to induction heating 30 will generate a heat source that is confined to only the top layer of the printed pattern 11. This heat will sinter an underlying layer of metallic nanoparticles adjacent to the top conductive layer, and by limiting the duration of exposure to induction heating, the conductive path 24 and remainder 16 of the printed pattern 11 between the conductive path 24 and the substrate 18 can be sintered with minimal or no damage to the substrate 18.

Oscillation frequency of the alternating magnetic field 32 is preferably in the ultrahigh frequency range to prevent excessive power loss and allowing the sintering of small printed patterns. The required oscillation frequency is highly dependent on the application, type of metallic-particle ink and the size of the printed patterns.

Printed Pattern Preparation

As shown in FIG. 3a, before application of the present method 10 of sintering, a printed pattern 11 is deposited on a substrate 18. The printed pattern 11 comprises a metallic- particle ink which contains metallic particles 12 coated with a non-conductive substance 14 containing organic additives and stabilising agents to prevent agglomeration of the metallic particles 12 in the ink, so as to facilitate ink flow during printing.

In an exemplary embodiment of printed pattern preparation, silver nanoparticle ink (UTDAg40TE) from UT Dots Inc. was used as the metallic-particle ink. UTDAg40TE has a weight/volume percentage concentration of 25% to 40% of silver nanoparticles and is recommended to be used with the aerosol jet’s ultrasonic atomiser (UA) for ink depositions [10] This ink is also recommended to be thermally sintered between 120 °C - 140 °C to achieve good electrical conductivity. The ink is preferably shaken for at least 10 minutes before use to ensure homogeneous dispersions of the metallic particles within the liquid medium of the ink. Shaking of the ink bottle can be done either manually or through the use of a laboratory orbital shaker. If the metallic particles are found to be agglomerating in the liquid medium, ultrasonic treatment may be useful to disperse the agglomerates.

Polyimide was chosen as a substrate due to its high-temperature resistance and flexible properties. Polyimide films of about 75 p thickness were cut into squares having dimensions of 50 mm by 50 mm. The polyimide films were first wiped clean using isopropanol and then cleaned in an ultrasonic bath using ethanol for 15 minutes. Lastly, they were cleaned in an ultrasonic bath using distilled water for 15 minutes to rinse any remaining ethanol off their surfaces. These substrates were air dried before printing.

The silver nanoparticle ink was then deposited onto the substrates using Optomec’s Aerosol Jet 5x system. The aerosol jet printer is a fully-digital and non-contact‘fully additive’ printing technique, employing aerodynamic focusing technology to deposit concentrated aerosol ink droplets directly onto the substrates accurately and precisely. The aerosol jet printer is capable of handling inks with viscosities ranging between 1 - 1000 cps and printing fine lines up to 10 pm line width [1-4] Each printed pattern 11 comprised five resistive tracks 111 labelled A to E and two circular resistive pads 112 printed on a single piece of substrate 18, as can be seen in FIG. 3b, which also shows critical dimensions of the printed pattern 11. The five resistive tracks 111 were used to characterise the electrical properties with respect to different sintering conditions, while the two circular resistive pads 112 were used to study the effects of sintering methods on their surface morphologies.

IR Sintering

In an exemplary embodiment of the first step 20 of the method 10, as shown in FIG. 4a, the printed pattern 11 is exposed to a non-contacting heat source 22 by placing the printed pattern 11 about 5cm below a low-intensity IR quartz lamp 22 of 800 W and 50cm in length, for a few minutes (e.g. 1.5 min) for IR sintering 20 to occur. IR sintering 20 is selected for being easy to set up, fast and low cost. The intensity of the IR lamp 22 to the printed pattern 11 can be controlled by either controlling the power input into the lamp 22 through a potentiometer or varying the distance between the IR lamp 22 and the printed pattern 11. By exposing the printed pattern 11 to IR irradiation (indicated by the downward arrows in FIG. 4a under the IR lamp 22), a topmost layer 24 of metallic particles 12 (e.g. silver nanoparticles) in the exposed surface of the printed pattern 11 absorbs the IR electromagnetic waves and converts the light energy into thermal energy as the metallic particles 12 experience a surface plasmon resonance (SPR) phenomenon.

During IR sintering 20, the thermal energy first decomposes enveloping organic shells 14 of the metallic particles 12 and subsequently increases the temperature of the metallic particles 12 for sintering to occur. The thermal energy is also thermally conducted to the layers below 16 and sinters more metallic particles 12. Notably, the printed pattern 11 is exposed to the low-intensity infrared light 22 for only a short period of time to ensure minimum or no damage is sustained by the substrate 18. During the short IR sintering process 20, only a topmost layer 24 of the exposed surface of the printed pattern 11 is sintered by the IR light 22 and the bottom layers or remainder 16 of the printed pattern 11 under the sintered topmost layer 24 is still wet. After sintering, organic additives in the topmost layer 24 are decomposed away and the sintered metallic particles 12S are at least partially coalesced. In the bottom layers or the rest 16 of the printed pattern 11 between the sintered top layer 24 and the substrate 18, the metallic particles 12 are still enveloped in the non-conductive substance 14 and are still not in electrical contact with each other. In other words, only the metallic particles 12S in the topmost layer 24 are sintered to form an electrically conductive path 24 and the rest 16 of the printed pattern 11 is not electrically conductive.

Absorption of IR radiation is dependent on two main factors: wavelength and angle of incidence. Different materials have different absorption wavelengths and hence, it is critical to understand the absorption wavelengths of the metallic particles in the metallic-particle ink in the IR sintering process. For example, the optical properties of silver nanoparticles are tunable by simply modifying their shapes, sizes and structures. Hence, the optical properties of the spherical silver nanoparticles are directly dependent on their diameters [40] From the extinction spectra of spherical silver nanoparticles with diameters ranging from 10 nm to 100 nm, it can be observed that spherical silver nanoparticles with diameters less than 50 nm have peak absorptance of light wavelengths at around 400 nm. As the diameters of the spherical silver nanoparticles increase, the peaks broaden and shift towards longer wavelengths. These nanoparticles absorb the light at high efficiencies when they are exposed to the wavelengths of light that correspond to their peaks. During IR sintering, shapes and sizes of the silver nanoparticles change and in turn affect their optical properties [40]

As shown in FIGS. 15 and 16, an exemplary set-up 120 to perform the IR sintering in the first step 20 of the method 10 may comprise an IR quartz lamp 22 (e.g. line source quartz infrared heater with housing, using SSR 24-280VAC, voltage 240V, power 1600W, wavelength 0.8-1.4 mhi, length 500mm), a parabolic reflector 123 provided over the IR lamp 22 to direct the IR radiation downwards, a heater controller with potentiometer (e.g. 40A, not shown) to control the IR lamp 22, a thermal insulating base plate 124 (movable in the xy-axis) on which the printed pattern 11 (on the substrate 18) is placed and a gantry 125 (movable in the z-axis) on which the IR lamp 22 is mounted. The intensity of the IR lamp 22 to the base plate 124 can be controlled by either controlling the power input into the lamp 22 through the potentiometer or varying the height h of the lamp 22 from the base 124 using the gantry 125.

Induction Sintering

The IR sintering process 20 described above serves as a pre-induction sintering step for partially sintering as-deposited printed patterns 11 so that the printed patterns 11 have some initial electrical conductivity to allow the subsequent use of induction heating 30 for sintering the printed pattern 11. In the topmost layer 24 of the printed pattern 11 that has been at least partially sintered by IR sintering 20, the metallic particles 12S have coalesced with their neighbouring particles to form a larger conductive network. The partially sintered patterns 11 are now not significantly affected by the skin effect phenomenon discussed above as the conductive area 24 is now much larger than without IR sintering. Thus, there is better power absorption and heating efficiency in the IR partially sintered printed patterns 11 in the step of induction heating 30.

After IR sintering, the topmost layer 24 of the printed pattern 11 is now electrically conductive and comprises an electrically conductive path 24. The printed pattern 11 can now proceed to the second step 30 of subjecting the electrically conductive path 24 to an alternating magnetic field 30 for induction sintering 30, as shown in FIG. 4b. In the exemplary embodiment, induction sintering 30 is done by placing the electrically conductive printed pattern 11 above an induction coil 32 connected to an induction heater (not shown) and activating the induction heater to induce eddy currents in the topmost layer or electrically conductive path 24 of the printed pattern 11 for 1 min. The induction heater 32 may be operated at 5 kW at a frequency of 1.5MHz, for example. As the eddy currents flow through the electrically conductive topmost layer 24, Joule’s heating effect takes place. The heat generated from Joule’s heating in the topmost layer 24 further sinters the metallic particles 12S within that layer 24 and further improves its electrical conductivity.

Some of the heat is thermally conducted to the wet layers 16 below and this heat (indicated by the horizontal black line) decomposes the organic additives 14 and coalesces the metallic particles 12S together, as shown in FIG. 4c. Hence, the layer below the topmost layer 24 also becomes electrically conductive. Eddy currents can now flow through these layers and extend the induction sintering effects to these layers, in a direction from the topmost electrically conductive path 24 towards the substrate 18, until the entire printed pattern 11 is fully sintered, as shown in FIG. 4d.

As the heat generated by induction sintering 30 may potentially be thermally conducted to the underlying substrate 18 (as indicated by the horizontal black line in FIG. 4d) and damage the substrate 18, operating parameters of the induction sintering step 30 should be optimised to ensure that the underlying substrate 18 is not damaged during the process 30. For example, it is also possible to sinter the printed patterns with the induction sintering step 30 performed in pulses. These pulsating inductions can allow the substrates to cool down and prevent them to overheat during the induction sintering process while sintering the printed patterns. FIG. 13 shows the temperature profile of the top surface of a printed pattern sintered using induction sintering 30 in pulses.

Notably, induction sintering 30 is a selective process in which only electrically conductive materials are heated by induction heating. This is extremely advantageous as the induction sintering technique 30 is able to differentiate conductive materials from non-conductive materials. Therefore, electrically non-conductive substrates are not affected by induction heating and the present method 10 can thus minimise damage to the substrate.

As shown in FIG. 17, an exemplary set-up 130 to perform the second step 30 of the method 10 may comprise an induction coil 32 (e.g. solenoid type) connected to an induction heater 133 (e.g. model no. DW-UHP-6KW-I from HLQ Induction Equipment Co., Ltd, of input power 220V AC 50 or 60 kHz, oscillation frequency 900 - 2000 kHz, maximum oscillation power 6kW, duty cycle 80%, water cooled at 0.2MPa or 5L/min at water temperature under 30 °C), wherein the printed pattern 11 on substrate 18 is placed directly above the induction coil 32. A motorized platform 134 movable in the x and y axes is provided to support and position the printed pattern 11 and substrate 18. A water cooling system (not shown) was used to cool down the induction coil 32 due to the tremendous amount of heat generated in the coil 32 as a result of high currents passing through the coil 32. The water cooling system consisted of a water pump (e.g. PV55 from Pedrollo S.p.A. capable of delivering water pressure more than 3 bars), a water tank and a refrigerated circulator. The water pump was used to pump water from the water tank through the induction coil 32. Water in the water cooling system is preferably maintained under 30°C during operation using the refrigerated circulator.

Comparison and Analysis of Electrical Resistivity

Three different sintering methods: thermal sintering, IR sintering and the present two-step method 10 were tested for sintering a printed pattern comprising five resistive tracks prepared as described above. Fifteen unique samples of the printed pattern were prepared and each sintered using one of the three different sintering methods. After sintering, the electrical resistance, average line width and cross-sectional thickness of each resistive track 111 of the printed pattern 11 were measured to calculate its electrical resistivity.

The electrical resistance of each resistive track 111 was measured using a two-wire multi meter (Tektronix DMM4050 6-1/2 digit precision multi-meter). The measuring probes were placed in the centre of the circular electrodes during measurement. The cross-sectional thickness and width of the printed resistive tracks were measured with a 3D laser scanning confocal microscope (KEYENCE, VK-X200 series). This 3D laser scanning confocal microscope has a 408-nm wavelength violet laser and it is capable of giving a 0.5 nm display resolution. In addition, it is a non-contact, high accuracy surface measurement technique which will not damage the samples during the measurement process. With the collected data above, the electrical resistivity of each printed track was calculated and tabulated using the equation below:

R-W-tc

where p is the electrical resistivity of the material, R is the electrical resistance of the resistive track, / is the length of the resistive track 111 between the two measuring probes, w is the width of the resistive track and t _c is the thickness of the resistive track 111. The electrical resistivity of each printed resistive track of the fifteen samples is shown in Table 2 below.

Table 2

Some values of the electrical resistance of the resistive tracks could not be collected as there were accidental scratches to the tracks. These scratches had resulted in broken and open-circuited tracks. This is one of the major disadvantages of printed electronics as accidental scratches may easily cause serious damage to the conductive tracks. The electrical resistivity of the twelve thermally sintered samples numbers 1 to 12 are shown as box plots in FIG. 5. From FIG. 5, it can be observed that the resistive tracks can achieve much more consistent electrical resistivity at higher thermal sintering temperatures with longer sintering duration. As expected, higher thermal sintering temperatures and longer sintering also give lower electrical resistivity sintered patterns. However, the sintering time for the thermal sintering technique is too time-consuming for effective and efficient mass productions of printed electronics. From the results in Table 2, the samples sintered at 150 °C had a much lower electrical resistivity as compared to the samples sintered at 120 °C.

Therefore, it can be deduced that better coalescence of the silver nanoparticles in the metallic-particle ink occurred after the bulk of the organic solvents had decomposed or evaporated from the printed patterns. The best results for thermal sintering as shown in Table 2 can be achieved by sintering at 150 °C for 4 hours, which was performed for sample number 8. The sintering conditions in sample number 8 allow the sintered patterns to achieve both lowest electrical resistivity and high consistency.

The best result from thermal sintering was compared with IR sintering and the present method of sintering 10 for better evaluation. The electrical resistivity of the samples numbers 8, 13 and 14 are shown in a box plot in FIG. 6, which corresponds to the thermal sintering, infrared sintering and present two-step sintering method 10. Comparing sample number 14 with sample number 8, it can be seen in FIG. 6 that the present two-step sintering method 10 is able to achieve approximately 4 times lower electrical resistivity with 0.01 of the time taken by the thermal sintering technique. It is noted that thermal sintering and the present two-step sintering method are both consistent and reliable sintering techniques, which are able to achieve low variability in their electrical resistivity values. By comparison, although infrared sintering is able to obtain considerably low electrical resistivity in a mere 1.5 minutes, the variability of the electrical resistivity is too large for IR sintering to be used as a reliable and consistent sintering technique for printed electronics.

Comparison and Analysis of Surface Morphologies

The surface morphologies of the printed patterns sintered by the different methods as shown in Table 2 were taken with a Field Emission Scanning Electron Microscope (FESEM), JEOL JSM-7600F. FESEM is capable of capturing images at the nanoscale level. Hence, it is extremely helpful to have FESEM to capture and analyse the surface morphologies of the sintered patterns, as the inks are suspensions of metallic particles. FIG. 8, FIG. 9 and FIG. 10 show the FESEM images of the thermally sintered patterns under different sintering conditions, while FIG. 11 shows the FESEM images of the printed pattern sintered by IR sintering and FIG. 12 shows the FESEM images of the printed pattern sintered by the present two-step sintering method 10. For reference, FIG. 7 shows the FESEM images of unsintered printed patterns, in which individual discrete silver nanoparticles enveloped in organic additives can be observed. Several conclusions can be drawn from these FESEM images.

After thermal sintering of the printed patterns at 120 °C for 30 and 60 minutes respectively, the FESEM images (FIGS. 8a and b) show that the silver nanoparticles (approximately 10 nm in diameter) were still mostly discrete. Most of the organic shells of the silver nanoparticles had been decomposed and neighbouring particles were now able to form contacting points with each other. Thus, electrons could now flow freely throughout the entire printed pattern and the printed pattern some initial electrical conductivity. However, the above sintering time was not adequate for the printed patterns to have significant formation of sintering necks.

After thermal sintering of the printed patterns at 120 °C for 120 minutes (FIG. 8c), distinct neck formations were observed and these silver nanoparticles had coalesced into large silver clusters of continuous networks. This stage is also known as the aggregation of the nanoparticles [28] Triggered by the diffusive [25] Ostwald ripening phenomenon [29] and the need to reduce the total interfacial energy [30], contacting nanoparticles were able to form sinter necks between each other at elevated temperatures. These nanoparticles, with the aid of the introduced energy, released high surface energy to achieve a state of thermodynamic equilibrium and aggregated into larger particles [25] through the formation of sinter necks [31] At this stage, the structures of the printed patterns were highly porous and their electrical conductivity was still not optimised [25] As the sintering time was increased to 240 minutes, neck formations were more prominent and inter-particle atomic diffusion and grain growth was observed [14] (see FIG. 8d). The additional input of energy and sintering time promoted grain growth and densification [32, 33]

Similar sintering phenomenon was also observed in the printed patterns thermally sintered at 150°C and 200°C. In FIGS. 9a and 10a, formation of sintering necks between neighbouring particles were observable in the first 30 minutes of thermal sintering at 150 °C and 200 °C respectively. As thermal sintering proceeded, there was significant grain growth and densification of the silver nanoparticles (see FIGS. 9b - d and FIGS. 10b - d). A comparison can be made among thermal sintering at the three different temperatures of 120°C, 150°C and 200°C: at the end of 240 minutes, the average grain sizes of the silver nanoparticles inks thermally sintered at 120°C, 150°C and 200°C were approximately 150 nm, 200 nm and 250 nm respectively. From the observations above, it can be concluded that the grain sizes grew with increasing sintering temperatures and increasing sintering time. In FIGS. 8, 9 and 10, it can be observed that thermally sintered printed patterns tend to have high porosity. Porous structures are disadvantageous as they directly affect the specific resistivity of the entire printed pattern. High porosities limit electrons movement throughout the entire structures by narrowing their conductive paths and hence increasing the specific resistivity of the entire printed pattern [41] FIG. 10d suggests that if the sintering temperature is too high, cracks may be induced within the printed pattern, thereby increasing electrical resistivity of the printed pattern. The resistivity measurements in the previous section validated that resistivity of printed pattern sintered at 200°C for 240 minutes is indeed higher than resistivity of printed pattern sintered at 150°C for 240 minutes.

From the FESEM images of the printed pattern sintered by IR sintering as shown in FIG. 11 , it could be observed that the topmost surface of the printed pattern is well sintered with grain sizes of approximately 200 nm in diameter. It was also observed that the silver nanoparticles in the layers below started to have formations of necks between neighbouring particles. Thus, it can be concluded that IR sintering technique is a top-to-bottom sintering technique in which the topmost layer is sintered first. The light energy absorbed by the topmost layer is converted into heat energy, which is then thermally conducted to the layers below for sintering. The printed patterns are now partially sintered and have some initial electrical conductivity. Therefore, with the initial electrical conductivity within the printed patterns, the printed patterns can now be sintered using induction sintering subsequently. Notably, the topmost surface has better electrical conductivity compared to the layers below, as the bottommost layers may not be well-sintered due to the negative feedback mechanism of IR sintering as discussed above.

Pores within microstructures are defects as they may degrade mechanical and electrical properties of the printed patterns, and they have a higher tendency to cause failures. Therefore, it is desirable to have highly-dense surface morphologies with low porosities in sintered patterns, as better electrical and mechanical properties are commonly associated with low-porosity microstructures. The FESEM images of the printed patterns sintered using the present two-step method of sintering 10 show that the two-step sintering method 10 is able to produce highly-dense sintered patterns with low porosities, as can be seen in FIG. 12. The grain sizes are approximately in the range of 400 - 450 nm in diameter. Larger grain sizes can help to facilitate electrons movement by widening electrons flow paths and hence reducing the electrical resistivity of the sintered patterns. Therefore, highly-dense microstructures with larger grain sizes and low-porosities can significantly reduce the electrical resistivity of the sintered pattern and eventually improve its electrical performance.

The present two-step method of sintering 10 can be summarised as follows: IR sintering 20 is first used in which infrared electromagnetic waves are absorbed by the metallic particles in the ink. This light energy is converted into thermal energy by the SPR phenomenon. The thermal energy decomposes most organic shells of the metallic particles in the topmost layer so that these metallic particles can form contact points with neighbouring particles. At this stage, electrons are able to flow through these contacting metallic particles. With the further absorption of the infrared electromagnetic waves, more light energy is converted to thermal energy. This thermal energy further drives aggregation of the metallic particles, in which sinter necks are formed among neighbouring particles. During IR sintering, grain growth can only be observed in the topmost layer (see FIG. 11) as only the topmost layer absorbs the bulk of the infrared electromagnetic waves. At the end of the first sintering step 20, the printed pattern is partially sintered and has some initial electrical conductivity. Proceeding to the second step 30 of the present two-step sintering method, the printed patterns can now be further sintered by induction sintering 30 as the topmost layer of the printed pattern is now electrically conductive. Eddy currents are induced in the printed pattern and generate a Joule’s heating effect within the printed pattern. The generated heat is the main driving force that promotes further grain growth and densification of the silver nanoparticles as seen in FIG. 12.

Evaluation of Sintering Methods

To evaluate if the present two-step method of sintering 10 is better than conventional thermal sintering, a feasibility analysis was conducted based on six most important criteria in the sintering process: final electrical conductivity of the printed patterns, surface morphology, selective sintering, sintering speed, controllability and R2R compatibility, as further described below. In this feasibility study, each sintering technique is ranked for each of the criteria using a score system as follows:

Poor: +

Fair: ++

Good: +++

Very Good: ++++

Excellent: +++++

For“electrical conductivity”, scores were given according to the electrical resistivity of the sintered patterns. The sintering method which produced sintered patterns with the lowest electrical resistivity will have the best score. The printed patterns sintered by the present two-step method of sintering 10 had a much lower electrical resistivity compared to thermally sintering, but the electrical resistivity of the thermally sintered patterns was quite low too. Therefore, an excellent score was given to the present method 10 (+ + + + +) and a very good score was given to thermal sintering (+ + + +).

For“surface morphologies”, scores were given according to the density and porosity of the microstructures of the sintered patterns, in which highly-dense and low porosities microstructures had the highest score. The printed patterns sintered by the present method 10 had highly-dense microstructures with low porosities while thermal sintering produced sintered patterns with high porosities and cracks. Therefore, an excellent score was given to the present method 10 (+ + + + +) and a good score was given to thermal sintering (+ + +).

For“selective sintering”, scores were given according to the ability to selectively sinter the printed patterns while not affecting the substrates. The present method 10 was able to selectively heat the printed patterns by distinguishing the electrically nonconductive substrates from the conductive printed patterns was therefore given an excellent score (+ + + + +). For thermal sintering, the substrates were heated together with the printed patterns, hence, a poor score is given to thermal sintering (+).

For“sintering speed”, scores were given according to the time taken for sintering. The present method 10 can sinter printed patterns in 2.5 minutes, whereas thermal sintering needed 4 hours. However, the present method 10 is not the fastest sintering method compared to other sintering methods. Therefore, a“very good” score was given to the present method 10 (+ + + +) and a poor score was given to the thermal sintering technique (+)

For“controllability”, scores were given according to how well the sintering process can be controlled. The present method 10 can be controlled and optimised by varying the intensity of the infrared lamp to the printed patterns, operating frequency of the induction heater, operating power of the induction heater and sintering time. Thermal sintering can only be controlled by varying the temperature of the oven and the sintering time. Therefore, a“very good” score was given to the present method 10 (+ + + +) as there are many options to control and optimise it. A poor score was given to thermal sintering (+ +) as there are very limited options for controlling it.

For“R2R compatibility”, scores were given according to the ability to integrate the sintering technique into a roll-to-roll (R2R) printing process for large-scale manufacturing of printed electronics. The present method 10 is able to integrate well with the R2R printing process, whereas thermal sintering is incompatible with R2R. Therefore, an“excellent” score is given to the two-step sintering technique (+ + + + +) and a poor score is given to the thermal sintering technique (+). Table 3 below presents the qualitative feasibility comparison for electrical conductivity, surface morphologies, selective sintering, sintering speed, controllability and R2R compatibility for thermal sintering technique and the present method 10.

Table 3

The present two-step method of sintering 10 has been proven to be superior to thermal sintering, as the former can produce sintered patterns with better electrical properties and surface morphologies in a shorter time. At the same time, the present method 10 is able to selectively sinter printed patterns and have better controllability and good R2R compatibility. The present method 10 is a selective sintering technique, in which the induction sintering step can only heat up electrically conductive materials. Hence, the conductive temperature- sensitive substrates are not heated up by induction sintering, thereby minimising damage to the substrate. Thus, the present method 10 allows the use of a wider range of substrates for different applications, including flexible temperature-sensitive substrates such as low-cost polymer films, paper and textiles. Temperature-sensitive polymer films with low glass transitional temperatures (for examples, polyethylene terephthalate (PET), polycarbonate (PC) and polyethylene naphthalate (PEN)) can also be used. Hence, the present method allows for exploration of more potential applications in the field of wearable electronics and cheap disposable electronics (for examples, printed health monitoring sensors on textiles and disposable printed radio-frequency identification (RFID) tags on paper substrates).

The present method 10 can also be easily integrated into R2R printing processes. For instance, as-printed patterns from R2R printing processes (for instance, flexographic printing, offset lithography, gravure printing and screen printing) can directly be sintered after printing. FIG. 14 shows an exemplary apparatus 100 for performing the above-described method of sintering 10, as integrated into R2R printing. The apparatus 100 is configured for sintering printed patterns 11 , each printed pattern 11 comprising metallic-particle ink printed on a substrate. The apparatus comprises a conveyor belt 110 to support the printed patterns thereon, a non-contacting heat source 22 (such as an IR lamp 22) provided above the conveyor belt 110, and induction coils 32 provided above the conveyor belt 110 subsequent to the non-contacting heat source 22. The non-contacting heat source 22 is provided to form an electrically conductive path of at least partially sintered metallic particles on an exposed surface of each printed pattern when each printed pattern is positioned under the non contacting heat source 22 by the conveyor belt 110. The induction coils 32 are provided to generate heat via induction heating of the electrically conductive path to sinter metallic particles in each printed pattern in a direction from the electrically conductive path towards the substrate when each printed pattern is positioned under the induction coils 32 by the conveyor belt 110.

Using the apparatus 100, printed patterns 11 on the conveyor belt 110 thus pass through two sintering stations: an IR sintering station comprising the IR lamp 22 followed by an induction sintering station comprising the induction coils 32. As the printed patterns 11 pass through the IR sintering station, the printed patterns 11 will be exposed to the IR radiation 20 and absorb the IR radiation from the IR lamp 22, becoming heated up such that sintering of the metallic particles in the printed patterns 11 starts to occur, resulting in the printed patterns 11 having some initial electrical conductivity. The conductive printed patterns 11 then proceed to the induction sintering station on the conveyor belt 110 where the induction coils 32 are directly placed or located above the printed patterns 11. The induction coils 32 induce induction heating 30 on the conductive printed patterns 11 and allow further sintering to take place in a direction from the topmost layer of the printed pattern towards the substrate. The effects of sintering can be controlled according to these few parameters: intensity of the IR lamp 22, distance of the IR lamp 22 from the printed patterns 11 , power of an induction heater used to heat the induction coils 32, operating frequency of the induction heater, distance of the induction coils 32 from the printed patterns, design of the induction coils 32 and speed of the conveyor belt 110.

With the ability of the present method 10 to be intergrated into R2R printing processes and its ability to significantly reduce the time taken for sintering, the present method 10 and apparatus 100 are able to streamline the printed electronics fabrication processes, providing the potential of mass manufacturing cheap and flexible printed electronics.

Whilst there has been described in the foregoing description exemplary embodiments of the present invention, it will be understood by those skilled in the technology concerned that many variations in details of design, construction and/or operation may be made without departing from the present invention.

REFERENCES

[1] C. K. Chua and K. F. Leong, 3D Printing and Additive Manufacturing - Principles and

Applications, 5th ed.: World Scientific Publishing Co. Pte Ltd, 2017.

[2] H. W. Tan, C. K. Chua, and T. Tran, "A Study of the State-of-the-Art Printed Passive

Electronic Components Through Fully Additive Manufacturing Methods," in Proceedings of the 2nd International Conference on Progress in Additive Manufacturing, C. K. Chua, W. Y. Yeong, M. J. Tan, E. Liu, and S. B. Tor, Eds., ed Singapore: Research Publishing Services, 2016, pp. 27-32.

[3] H. W. Tan, T. Tran, and C. K. Chua, "A review of printed passive electronic components through fully additive manufacturing methods," Virtual and Physical Prototyping, vol. 11 , pp. 271-288, 2016/10/01 2016.

[4] H. W. Tan, T. Tran, and C. K. Chua, "Aerosol Jet Printed Strain Gauge," presented at the

IJIE 2016: The 18th International Conference on Industrial Engineering, Seoul, South Korea, 2016.

[5] N. Kaihovirta, T. Makela, X. He, C.-J. Wikman, C.-E. Wilen, and R. Osterbacka, "Printed all-polymer electrochemical transistors on patterned ion conducting membranes," Organic Electronics, vol. 11 , pp. 1207-1211 , 2010.

[6] E. Hrehorova, M. Rebros, A. Pekarovicova, B. Bazuin, A. Ranganathan, S. Garner, et al.,

"Gravure Printing of Conductive Inks on Glass Substrates for Applications in Printed Electronics," Journal of Display Technology, vol. 7, pp. 318-324, 2011.

[7] Y. Kawahara, S. Hodges, B. S. Cook, C. Zhang, and G. D. Abowd, "Instant inkjet circuits: lab-based inkjet printing to support rapid prototyping of UbiComp devices," in

Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing, Zurich, Switzerland, 2013, pp. 363-372.

[8] R. Bonadiman and M. M. P. Salazar, "Reliability of Ag ink jet printed traces on polyimide substrate," in Electronic System-Integration Technology Conference (ESTC), 2010 3rd, Berlin, Germany, 2010, pp. 1-5.

[9] L. Basirico, P. Cosseddu, B. Fraboni, and A. Bonfiglio, "Inkjet printing of transparent, flexible, organic transistors," Thin Solid Films, vol. 520, pp. 1291-1294, 12/1/ 2011.

[10] E. Halonen, T. Viiru, K. Ostman, A. L. Cabezas, and M. Mantysalo, "Oven Sintering Process Optimization for Inkjet-Printed Ag Nanoparticle Ink," Components, Packaging and Manufacturing Technology, IEEE Transactions on, vol. 3, pp. 350-356, 2013.

[11] K. Myong-Ki, H. Jun Young, K. Heuiseok, K. Kyungtae, L. Sang-Ho, and M. Seung-Jae, "Laser sintering of the printed silver ink," in Assembly and Manufacturing, 2009. ISAM 2009. IEEE International Symposium on, 2009, pp. 155-158.

[12] T. Niizeki, K. Maekawa, M. Mita, K. Yamasaki, Y. Matsuba, N. Terada, et al., "Laser sintering of Ag nanopaste film and its application to bond-pad formation," in Electronic Components and Technology Conference, 2008. ECTC 2008. 58th, 2008, pp. 1745- 1750.

[13] T. Kumpulainen, J. Pekkanen, J. Valkama, J. Laakso, R. Tuokko, and M. Mantysalo, "Low temperature nanoparticle sintering with continuous wave and pulse lasers," Optics & Laser Technology, vol. 43, pp. 570-576, All 2011.

[14] Z. Michael, E. Oleg, S. Amir, and K. Zvi, "Laser sintering of copper nanoparticles,"

Journal of Physics D: Applied Physics, vol. 47, p. 025501 , 2014.

[15] A. Denneulin, A. Blayo, C. Neuman, and J. Bras, "Infra-red assisted sintering of inkjet printed silver tracks on paper substrates," Journal of Nanoparticle Research, vol. 13, pp. 3815-3823, 2011.

[16] H. S. Kim, S. R. Dhage, D.-E. Shim, and H. T. Hahn, "Intense pulsed light sintering of copper nanoink for printed electronics," Applied Physics A, vol. 97, pp. 791-798, 2009.

[17] J. Perelaer, M. Klokkenburg, C. E. Hendriks, and U. S. Schubert, "Microwave Flash Sintering of Inkjet-Printed Silver Tracks on Polymer Substrates," Advanced Materials, vol. 21 , pp. 4830-4834, 2009.

[18] M. L. Allen, M. Aronniemi, T. Mattila, A. Alastalo, K. Ojanpera, M. Suhonen, et al.,

"Electrical sintering of nanoparticle structures," Nanotechnology, vol. 19, p. 175201 , 2008.

[19] S. Ma, V. Bromberg, L. Liu, F. D. Egitto, P. R. Chiarot, and T. J. Singler, "Low

temperature plasma sintering of silver nanoparticles," Applied Surface Science, vol. 293, pp. 207-215, 2/28/ 2014.

[20] J. Perelaer, R. Jani, M. Grouchko, A. Kamyshny, S. Magdassi, and U. S. Schubert, "Plasma and Microwave Flash Sintering of a Tailored Silver Nanoparticle Ink, Yielding 60% Bulk Conductivity on Cost-Effective Polymer Foils," Advanced Materials, vol. 24, pp. 3993- 3998, 2012.

[21] I. Reinhold, C. E. Hendriks, R. Eckardt, J. M. Kranenburg, J. Perelaer, R. R. Baumann, et al., "Argon plasma sintering of inkjet printed silver tracks on polymer substrates," Journal of Materials Chemistry, vol. 19, pp. 3384-3388, 2009.

[22] F. M. Wolf, J. Perelaer, S. Stumpf, D. Bollen, F. Kriebel, and U. S. Schubert, "Rapid

lowpressure plasma sintering of inkjet-printed silver nanoparticles for RFID antennas," Journal of Materials Research, vol. 28, pp. 1254-1261 , 2013/005/14 2013.

[23] S. Wunscher, S. Stumpf, J. Perelaer, and U. S. Schubert, "Towards single-pass plasma sintering: temperature influence of atmospheric pressure plasma sintering of silver nanoparticle ink," Journal of Materials Chemistry C, vol. 2, pp. 1642-1649, 2014.

[24] S. Wunscher, S. Stumpf, A. Teichler, O. Pabst, J. Perelaer, E. Beckert, et al., "Localized atmospheric plasma sintering of inkjet printed silver nanoparticles," Journal of Materials Chemistry, vol. 22, pp. 24569-24576, 2012.

[25] J. Perelaer and U. S. Schubert, "Novel approaches for low temperature sintering of

inkjetprinted inorganic nanoparticles for roll-to-roll (R2R) applications," Journal of Materials Research, vol. 28, pp. 564-573, 2013/002/28 2013.

[26] Y. Galagan, E. W. C. Coenen, R. Abbel, T. J. van Lammeren, S. Sabik, M. Barink, et ai, "Photonic sintering of inkjet printed current collecting grids for organic solar cell applications," Organic Electronics, vol. 14, pp. 38-46, 1// 2013.

[27] J. Reder, "Conductive Tracks and Passive Electronics," in Inkjet Technology for Digital Fabrication, ed: John Wiley & Sons, Ltd, 2012, pp. 159-181.

[28] B. Ingham, T. H. Lim, C. J. Dotzler, A. Henning, M. F. Toney, and R. D. Tilley, "How

Nanoparticles Coalesce: An in Situ Study of Au Nanoparticle Aggregation and Grain Growth," Chemistry of Materials, vol. 23, pp. 3312-3317, 2011/07/26 2011.

[29] P. W. Voorhees, "The theory of Ostwald ripening," Journal of Statistical Physics, vol. 38, pp. 231-252, 1985.

[30] S.-J. L. Kang, "1 - SINTERING PROCESSES," in Sintering, ed Oxford: Butterworth- Heinemann, 2005, pp. 3-8.

[31] B. J. Perelaer, A. W. M. de Laat, C. E. Hendriks, and U. S. Schubert, "Inkjet-printed

silver tracks: low temperature curing and thermal stability investigation," Journal of Materials Chemistry, vol. 18, pp. 3209-3215, 2008.

[32] H. J. Hwang, K.-H. Oh, and H.-S. Kim, "All-photonic drying and sintering process via flash white light combined with deep-UV and near-infrared irradiation for highly conductive copper nano-ink," Scientific Reports, vol. 6, p. 19696, 01/25 2016. [33] S.-J. L Kang, "11 - GRAIN GROWTH AND DENSIFICATION IN POROUS MATERIALS," in Sintering, ed Oxford: Butterworth-Heinemann, 2005, pp. 145-170.

[34] J. Perelaer, P. J. Smith, D. Mager, D. Soltman, S. K. Volkman, V. Subramanian, et al., "Printed electronics: the challenges involved in printing devices, interconnects, and contacts based on inorganic materials," Journal of Materials Chemistry, vol. 20, pp.

8446- 8453, 2010.

[35] R. E. Haimbaugh, "Practical Induction Heat Treating (2nd Edition)," ed: ASM

International, 2001.

[36] R. C. Goldstein and V. S.Nemkov, "Design Principles for Induction Heating and

Hardening," in Handbook of Metallu gical Process Design, ed: CRC Press, 2004, pp.

591- 640.

[37] V. Rudnev, D. Loveless, R. L. Cook, and M. Black, Handbook of induction heating : CRC Press, 2002.

[38] S. Semiatin, Elements of induction heating: design, control, and applications. ASM

International, 1988.

[39] Magnetic Materials. Available:

https://www.microwaves101.com/encyclopedias/magnetic-materia ls

[40] nanoComposix. Silver Nanoparticles: Optical Properties. Available:

https://nanocomposix.com/pages/silver-nanoparticles-optical- properties

[41] K. Ryu, Y.-J. Moon, K. Park, J.-Y. Hwang, and S.-J. Moon, "Electrical Property and

Surface Morphology of Silver Nanoparticles After Thermal Sintering," Journal of Electronic Materials, vol. 45, pp. 312-321 , January 01 2016.

[42] G. Weibing and C. Zheng, "Photonic sintering of nano-silver conductive ink for printed electronics," in 2016 6th Electronic System-Integration Technology Conference (ESTC), 2016, pp. 1-4.

[43] UT Dots Inc. (n.d.). Our Products. Available: http://www.utdots.com/products.html