2.2.3.1 Substrate Cleaning

The first and definitely one of the most important steps in device fabrication is substrate cleaning. Several different types of substrates can be used for electronic circuit development. Silicon wafers are very common, especially in silicon-based semiconductor industry, but other kinds of substrates may also be employed. Glass slides, conformable substrates (like Parylene-C), or even less conventional substrates such as textiles and silk are among them. Despite the fact that some of those substrates may have special requirements or even incompatibilities with the standard cleaning protocols, the cleaning step cannot be easily omitted. Keeping the substrate free of contaminants is a matter of great importance as the degree of its cleanness affects the quality of the deposited film. Moreover, particles on the substrate could potentially lead to damage of the photomask during contact photolithography exposure or even cause, in some cases, undesirable masking effects due to light diffraction. Among the contaminants that should be removed before coating the substrate with photoresist are atmospheric dust from operators and equipment, organic particles, moisture, H₂O residue films, solvent stains, smoke particles, residual resist, particulates, and chunks of granular matter [7, 11].

The cleanliness of the fabrication environment is of critical importance; hence, all fabrication steps take place in a clean room environment (typically class 100) that allows the presence of up to 100 particles (sized 0.5 µm or larger) per cubic foot of atmosphere. An environment such as this minimizes the amount of unwanted particles in the milieu and as a result it minimizes the number of unwanted contaminants on the device as well. Taking this environment as granted, typical cleaning procedures may include both wet and dry methods. Sonication in water soap baths or solvent baths is normally employed for the removal of particles (both inorganic and organic). In some cases piranha solution (a 3 : 1 mixture of sulfuric acid and hydrogen peroxide) can also be used to detach organic particles from the substrate. Additional approaches include thermal treatment at high temperature (dehydration bake), plasma cleaning, vapor cleaning, and supercritical cleaning with CO₂ during which supercritical fluid of carbon dioxide is used for removal of inorganic and organic contaminants from cracks and clefts owing to its ability to penetrate into crevices [7, 11].

2.2.3.2 Deposition of the Photoresist

Once substrate cleanness is ensured, deposition of the photoresist on the substrate follows. Among the ways of depositing polymers on a substrate spin-coating is the one that can guarantee uniformity, reproducibility, and precision during deposition. It is a well-known, traditional technique that is easy to use and offers control of the film thickness. The main drawback is the fact that the majority of the processed material is wasted.

Consequently, spin-coating deposition has been the method of choice for photoresist thin-film formation during fabrication for several decades now. During this approach, a small droplet of the photoresist is placed in the middle of the substrate, which is secured on a chuck via vacuum. Centripetal acceleration spreads the photoresist on the substrate. Although almost 98% of the initial material is spun off, eventually a thin film of photoresist is deposited on the substrate, as shown in Figure 2.2.

Thickness h of the photoresist is controlled through specific parameters of the process. The angular spinning speed ω and time t as well as the liquid density ρ, material viscosity η, and evaporation rate e_e are the most important factors affecting the film’s formation. During this process, complex nonequilibrium phenomena take place, and it is believed that two parts contribute to the rate by which the thickness of the film changes over time: a part that refers to the effect of the angular spinning speed and a part connected to the evaporation rate of the photoresist e_e. Generally, the spinning cycle can be separated into two stages: a very fast coating stage (when the photoresist is spread on the substrate) and a longer drying stage (during which the solvent evaporates). In any case, the rheology behind the film formation is complex, especially if the evaporation of the photoresist is taken into account. Therefore, the film thickness is usually given by the empirical expression (2.1) [7, 11, 12]

2.1

where K is an overall calibration constant, C is the polymer concentration in g/100 ml, and η is the solution’s viscosity. The exponential parameters α, β, and γ are determined experimentally. Once these parameters are set, a calibration curve is obtained that can provide the film thickness for a given polymer and solvent. Usually, film thickness is inversely proportional to the square root of the angular spinning speed ω and proportional to the solutions viscosity η to the 0.4–0.6 power [12].

2.2

There are two common ways to realize the photoresist’s dispersion on the substrate: the static dispense and the dynamic dispense. During static dispense a drop of photoresist is deposited on the substrate near its center while it is stationary. The amount of material deposited is in direct correlation with the viscosity of the photoresist (more viscous photoresists require more material to be placed) and the size of the substrate (bigger substrates require more material to ensure the total coverage of the substrate). On the other hand, dynamic dispersion dictates an initial step of spinning at a low speed (typically 500 rpm) while the dispensing takes place. After that, the substrate is accelerated to its final speed. Theoretically, this approach facilitates the wetting of the substrate and consequently the spreading of the material, leading to less waste.

For both approaches, the angular spinning speed and the time of the spinning are the two parameters that affect the final thickness of the film. In general, high speed and longer spinning times lead to thinner film formation.

2.2.3.3 Post-apply Bake

What follows is a thermal treatment step called post-apply bake (PAB) (or soft bake). It usually lasts for a minute or two on a hot plate at 110 °C. The purpose of this step is to evaporate the remaining solvent from the photoresist and to densify it just before exposure. This renders the coated film more stable and reduces the probability of the covered substrate to stick to the mask during exposure.

2.2.3.4 Use of the Mask/Alignment/Exposure

Exposure stands at the very heart of the fabrication process. The basic principle behind photolithography, after all, is altering a photoresist’s solubility by delivering energy to it via radiation. The stencil that is used to transfer the desired pattern on the photosensitive film is called the mask. Generally, a mask is made of glass (transparent to UV radiation) with a metal pattern on it (usually a 800 Å chromium film is used). The glass windows allow the radiation to pass through it with very little absorption while the metal pattern protects the underlying photoresist from any interaction with light. Masks are constructed with electron beam lithography, which can result in higher resolution than photolithography [7]. Special care is also taken in the proper alignment of different device layers to each other during the exposure. As previously stated this is one of photolithography’s figures of merit (Registration) and is handled with the use of special marks (alignment marks) strategically placed on the different layers [9]. After all, registration is one of the main advantages of photolithography compared to the rest of the techniques, along with high throughput due to its parallel nature.

There are three different ways to perform the exposure: contact, proximity, and projection mode, as shown in Figure 2.3.

2.2.3.4.2 Proximity Mode

Here, the mask and the substrate are not in contact as there is a small gap (10–50 µm) maintained between the two. That protects the mask from damage but at the same time lowers resolution due to diffraction effects.

Contact and proximity mode printing are known together as shadow printing. The resolution r for them is given by the formula [9, 11, 12]

2.3

where λ is the radiation wavelength, s is the distance between the mask and the substrate, and d is the photoresist thickness.

2.2.3.4.3 Projection Mode

This is the mode of choice used in semiconductor industry from the mid-1970s to date. In projection printing, there is no direct contact between the mask and the substrate as the mask is projected onto the substrate through a lens system. This approach protects the mask from damage since there is no physical contact involved, as shown in Figure 2.3. In addition, the demagnification of the mask pattern achieved with the optics results in high resolution and makes the mask fabrication a little easier [7].

The resolution r for projection printing is given by [9, 11, 12]

2.4

where k is a coefficient that depends on process-related factors and NA is the numerical aperture.

2.2.3.5 Development

After exposure, a development step will allow the latent resist pattern formed to be revealed. Selective dissolution of the resist creates a relief that will serve as a mold for the next fabrication steps. Development is of extreme importance as it controls the quality of the transferred motif.

There are two main approaches to performing this step: wet development and dry development.

2.2.3.6 Descumming and Post-baking

Descumming is a mild oxygen plasma treatment to remove any residual resist after development. It removes tiny amounts of unwanted material without harming the desired features. Patterned resist is also affected, but as long as only a few hundred angstroms are removed this does not cause any fabrication concerns.

Just before the pattern printed in the photoresist is transferred onto the substrate, a post-baking step (also known as hard baking) takes place. Hard baking promotes interfacial adhesion of the film and removes the residual solvent. It usually occurs at the temperature of 120 °C (slightly higher than the temperature used for soft baking), which additionally cross-links the photoresist, making it harder and more resistant to the etching steps that follow. Special care should be taken in order for the temperature not to cause flow or melting of the photoresist as this will degrade the profile of the resist.

2.2.3.7 Pattern Transfer

The previous steps create the desired pattern on the photoresist. The next goal is to transfer this pattern (or its negative) from the photoresist onto the substrate. There are two different methods to achieve this goal: a subtractive process and an additive one [4, 12].

In the subtractive method, first a material film is deposited on the substrate. Photolithography creates a positive image of the pattern and then etching removes the excess material, leaving behind the desired structure. The additive method, on the other hand, uses photolithography to create first a negative image of the pattern and then to realize its positive version via selective deposition of material. Both these methods will be further developed with case studies later in this chapter.

Etching is one of the most crucial parts in fabrication. Selective etching creates the required polymer microstructure during the subtractive approach, while it controls the material deposition in the additive one. In general, etching is a method of removing material that is not protected under the photoresist. It can be done chemically, mechanically, or with a combination of the two mechanisms.

Wet etching was initially the method of choice in the microelectronic industry. An acidic solution was used to erode the thin film not covered by the photoresist, creating a selective 3D structure. However, the method’s isotropic nature usually resulted in an undercut profile damaging the overall resolution. Consequently, new dry etching approaches quickly became popular as they could provide etching in an anisotropic way. Plasma etching, in particular, uses plasma (an ionized gas) to anisotropically and selectively etch only the patterned material and not the photoresist above it, allowing fabrication with sub-micrometer resolution. As dry techniques are easily automated and remove the need for toxic developers, it is not a surprise they quickly rose to dominance [11].

Many different dry etching techniques have been developed, but among them reactive ion etching (RIE) offers the benefits of both the chemical and the physical etching worlds. RIE uses plasma to create ionized atoms that can be accelerated by an electrical field and cause a directional sputtering of the substrate. This is extremely important in giving anisotropy to the technique. The charged molecules gain kinetic energy, which they transfer to the film in the collision, etching it vertically. At the same time they provide the energy for an etching reaction to take place, which is selective due to its chemical nature.

The deposition of the material of interest mentioned above is done with a number of different techniques. Chemical vapor deposition (CVD), physical vapor deposition (PVD), sputtering, and electroplating are among them. Here, we will focus on thermal evaporation as it is widely used for metal film deposition and it will prove to be extremely useful for the fabrication of organic devices that follows.

2.2.3.8 Stripping

The last step of the fabrication process is the removal of the remaining photoresist. That will create patterns by selectively discarding the evaporated material that was deposited on the photoresist while leaving the rest intact. The photoresist acts now as a sacrificial layer and is removed along with the metal layer on top of it, creating the desired metal profile. Photoresist stripping is usually performed with the help of organic solvents. Acetone is very commonly used for this task, along with other phenol-based commercial strippers. Nevertheless, environmental issues favor the use of dry stripping methods such as oxygen plasma. In any case, the ultimate criterion in the stripping approach is not to destroy the target material film. Especially for organic materials, this criterion poses a number of extra difficulties due to incompatibilities with the majority of solvents.

2.2.4.1 Sacrificial Layer Method

An alternative way of patterning organic materials, developed by DeFranco and coworkers, is based on the use of a poly(monochloro-p-xylylane (Parylene-C) sacrificial buffering layer [4]. Parylene-C (a polymer widely used as a barrier layer) is employed to protect the organic film during each step of the photolithography fabrication (deposition, development, and strip of the photoresist). After Parylene-C deposition, the formed film is inert and resistant enough to withstand a photolithography step on it.

From this point, two different fabrication methods (an additive and a subtractive one) lead, eventually, to the organic material patterning [4, 13, 14]. For the subtractive method the developed photoresist serves as a mask to selectively etch and remove both the Parylene-C layer and the organic film under it. In the additive method, on the other hand, the photoresist acts as contact mask and an etching step leaves behind voids in the Parylene-C to be filled with the polymer. Both these approaches can give high-quality patterned polymer films, and are shown schematically in Figure 2.4.

The next two case studies are paradigms of the aforesaid additive and subtractive methods implemented in organic device fabrication. The active area of those devices is covered with a thin poly(3,4-ethylenedioxythiophene) doped with poly(styrene sulfonate) (PEDOT:PSS) polymer film while the electrodes and their wiring are gold patterned with the use of conventional photolithography techniques. Both methods are versatile, generic, and can be used for direct patterning of polymer films in a variety of organic devices (polymer-covered electrodes, organic transistors, etc.).

2.2.4.1.1 Subtractive Patterning

A subtractive method that can result in high performing devices was presented in 2011 by Khodagholy and coworkers [15]. In this approach, PEDOT:PSS-covered gold electrodes were fabricated on a 2 µm Parylene-C film that served as a flexible substrate. Initially, gold electrodes, interconnections, and pads were patterned on a Parylene-C film via standard photolithography. A second 2 µm thick film of Parylene-C was used to insulate the device while a second photolithography step followed by oxygen etching (RIE) opened windows over the recording sites and pads. The polymer (PEDOT:PSS) was deposited through spin casting and the devices were coated with a third (sacrificial) layer of Parylene-C. The final photolithography and etching step defined the PEDOT:PSS-coated electrodes. Immersion of the device in deionized water promoted the removal of the Parylene-C sacrificial layer exposing the electrodes (Figure 2.5).

The success of the process relies on the fact that the polymer film adheres better to gold than to the Parylene-C film above it. In addition, due to its hydrophobic character, deionized water facilitates the sacrificial layer’s peeling off without affecting the quality and conductivity of the organic film. The conducting film’s integrity is also maintained during the Parylene-C deposition process. The above points render the method generic, versatile, and usable for different types of conducting polymers as they also become hydrophilic when doped.

Parylene-C plays a key role in the fabrication approach studied. It not only protects the organic film, which is sensitive to solvents, but also offers electrical insulation for the device, which is imperative for its functionality. Parylene-C is a member of the greater poly p-xylylene family and is produced with the substitution of Cl on the aromatic ring (Figure 2.6).

It has been extensively used in the past for coating purposes. It is a green chemistry polymer as it is chemically inert [16] and needs no initiator for solvent-free deposition as a coating film [17]. Hence it can be easily deposited on and removed from the polymer films without causing their chemical deterioration [15].

The material is deposited from its vapor phase via a CVD method proposed by Gorham. Its dimer is heated at 150 °C (P = 1 Torr) creating the vapor phase of the material. A pyrolysis stage follows at 680 °C (P = 0.5 Torr) that cracks the dimers to give birth to monomer units. The final polymerization step takes place on the device substrate at 25 °C (P = 0.5 Torr) resulting in the formation of a thin polymer film [17]. Most importantly, coating thickness can be controlled accurately and reproducibly through the amount of dimer used.

2.2.4.1.2 Additive Patterning

An additive method of polymer patterning was presented by Sessolo and coworkers in 2013 [3]. Once more, gold electrodes, contact pads, and their interconnections were patterned lithographically on a glass substrate. The device was coated with a 2 µm Parylene-C film that adhered on the substrate with the use of 3-(trimethoxysilyl)propyl methacrylate (−174 Silane). A soap solution was spin cast on Parylene-C to act as an antiadhesive layer between the first and a second 2 µm Parylene-C (sacrificial) film. Photolithography and an etching step were then used to open windows above the electrodes and the pads. After that, the polymer (PEDOT:PSS) was deposit by spin coating on the device, and a final peel-off step defined the final polymer device structure, shown in Figure 2.7.

Once again, the method is generic and versatile, and it can be adapted according to the desired device geometry. Moreover, different conducting polymers can be used as active layers as long as they can be deposited from solution.

2.2.4.2 Orthogonal Photoresist Method

A different way of dealing with the polymer patterning challenge comes with the utilization of orthogonal solvents. The term orthogonal refers to solvents in which the organic compounds are insoluble, a feature that allows not only the patterning of organic electronic materials but also their multilayer deposition.

Hydrofluoroethers (HFEs) in particular belong to a class of solvents, which, besides being nontoxic and environment friendly, are also orthogonal to many organic materials [18]. Consequently, they are ideal candidates for polymer patterning as long as a photoresist compatible with them is synthesized.

A photoresist such as this was presented in 2009 by Taylor et al. [5]. The HFE compatible material is a copolymer composed of a highly fluorinated monomer 1 (3.3.4.4.5.5.6.6.7.7.8.8.9.9.10.10.10-hepta-decafluorodecyl methacrylate) and a photosensitive monomer 2 (2-nitro-benzyl methacrylate) (Figure 2.8). Its solubility in HFE solvents can be modified after UV exposure due to structural changes to the photosensitive 2 part of the molecule, resulting in a negative tone photoresist. In addition, it is acid stable, a feature extremely useful when it is used to pattern acidic polymers.

As a proof of concept, a bottom-contact organic thin-field transistor was fabricated with a pentacene channel and PEDOT:PSS drain and source electrode by the same group. On a Si wafer, a 360 nm oxide was grown thermally just before PEDOT:PSS was spin cast and baked at 180 °C for 10 min. Photoresist 3 was then spun on the PEDOT:PSS layer and patterned lithographically with HFE-7200 (an isomeric mixture of methyl nonafluorobutyl ether and methyl nonafluoroisobutyl ether) acting as its developer. The image was transferred on the PEDOT:PSS with oxygen etching and the remaining photoresist was lifted off in a propan-2-ol (10% by volume)/HFE-7100 mixture. Photoresist 3 was spun again on the patterned PEDOT:PSS film, this time followed by UV light exposure and a development step. Pentacene was thermally evaporated, and removal of the photoresist by the previously used solvent mixture defined a pentacene channel connecting the PEDOT:PSS source and drain electrodes (Figure 2.9).

In a similar approach, Hwang and coworkers were able to pattern polymer materials using supercritical carbon dioxide (scCO₂) as the solvent [19] to fabricate an organic light-emitting diode (OLED). ScCO₂ is an environment-friendly fluid used in the dry photolithography process (DPP) for resist stripping owing to its physical and chemical advantages. Most importantly, it is a poor solvent for most ionic, high molecular weight, and low pressure organic materials.

In their work, a light-emitting polymer (LEP) was patterned on top of a PEDOT:PSS active layer. PEDOT:PSS was first spin cast on glass coated with indium tin oxide (ITO). A negative tone copolymer was synthesized from 1H,1H,2H,2H-perfluorodecyl methacrylate (FDMA) and tert-butyl methacrylate (TBMA) and deposited on the PEDOT:PSS layer. After UV exposure, scCO₂ was used for the development followed by oxygen plasma cleaning treatment and an LEP spin-casting step. A thermally deposited CsF(1 nm)/Al(40 nm) film completed the ITO/PEDOT:PSS/LEP/CsF/Al structure (Figure 2.10).

2.3.1.1 Gravure

Gravure printing is a process coming from intaglio developed in the nineteenth century. It uses a rotary printing press and operates at high speed processing (up to 100 m min⁻¹). It is adapted for large volume runs such as books or magazines. Gravure is based on the direct transfer of the ink to the substrate from an engraved cylinder to the substrate (Figure 2.11). This cylinder is electroplated, for instance, with copper, and engraved electromechanically or by laser to create microcells. A chromium zinc layer is generally added to protect the cylinder. The volume of the ink that is deposited is determined by the geometry of the microcells. Then, the ink is transferred by capillary force from the microcells to the substrate when pressure is applied between the gravure cylinder and the substrate. Moreover, a doctor blade station is used in order to remove any excess ink and to ensure good film uniformity. Gravure allows designs with high resolution up to 20 µm and presents good printing quality and process reproducibility. It is compatible with a wide range of solvents or water-based inks. The required viscosity of the inks is comprised between 10 and 500 cP. Nevertheless, this process exhibits some limitations in terms of maintenance and cost of manufacturing mainly because of the price and the weight of cylinders.

This technique has been studied for electronic applications such as for the realization of field-effect transistors (FETs) or photovoltaic cells [21]. A few groups tested the printability of conducting polymers to print functional multilayers to fabricate FETs [22] or to realize solar cells with laboratory gravure modules [23–25]. Gravure-printed PEDOT:PSS electrodes demonstrated high mechanical flexibility and stretchability in comparison with conventional indium tin oxide (ITO) on flexible substrates. For instance, Hübler et al. fabricated photovoltaic cells by printing PEDOT:PSS as the anode and a poly(3-hexylthiophene-2,5-diyl):[6,6]-phenyl-C 61 butyric acid methyl ester blend (P3HT:PCBM) as the photoactive and hole transport layers on paper [4]. Yang et al. used an industrial gravure printing proofer to process the photovoltaic module, where they print conducting polymers on flexible substrates already coated with ITO [26]. Recently, gravure has been investigated to realize impedance-based electrochemical biosensors [27], which paves the way to fabricate the new generation of flexible devices for biological applications.

2.3.1.2 Flexography

Conventionally, flexography is used for printing food packages and magazines. Flexography is compatible with flexible substrates such as plastics or cardboards. It is a process combining a system of cylinders and a flexible relief plate (Figure 2.12). An engraved anilox roll is used to transfer the ink from the pan. The amount of the ink is fixed by the volume of the microcells’ cylinder. A doctor blade ensures that the thickness of the ink is uniform. Then, the printing of the ink is assured by the contact between the substrate and the plate cylinder. These plates are usually made from rubber or photopolymer materials and produced by photolithography.

This technology is compatible with a wide variety of inks: UV-cured inks and solvent- or water-based inks with a viscosity ranging from 10 to 100 cP. Resolution up to 50 µm can be reached. The use of the plates lowers the cost of fabrication and allows the printing of large volume runs with high processing speed (up to 180 m min⁻¹). The costs are reduced in comparison with gravure but plates can be degraded with the use of solvents. One of the major undesirable effects of this process is the halo that appears on the edge of the printed pattern. The default comes from the deformation of the plate when the pressure is applied between the plate and the substrate squeezing the ink.

Only few studies chose this approach for printed electronic applications [28, 29]. Hübler et al. [30] use a laboratory flexographic test printing device to print the gate and contact electrodes using conducting polymers to fabricate transistors. Krebs et al. [31] reported using this process to improve the wettability of surface to slot die conducting polymer while Yu et al. [32] printed the silver layer to realize grids to replace ITO in polymer cells.

2.3.1.3 Screen Printing

Screen printing is a simple and versatile technique used in many applications such as textiles or advertisements. It has been originally invented in China, and imported into Europe in the eighteenth century and developed in the second half of the twentieth century. The first patent was published in 1907 by Samuel Simon, and the technique became more popular in the 1960s with artists such as Andy Warhol [33].

Screen printing is a stencil method that uses a mesh stretched over a frame to transfer the ink to the substrate. The ink is pushed into the opening of the mesh through a squeegee (Figure 2.13). The mesh is usually made from nylon, polyester, or a stainless steel. Two different categories of screen printing have been developed and can be integrated for roll-to-roll manufacturing: the flatbed screen or the rotary screen. The flatbed screen (Figure 2.13a) is used in laboratories or in production and the rotary is adapted for large-scale process with a production speed up to 100 m min⁻¹. In this case, the squeegee and the ink are placed in the rotary screen (Figure 2.13b), making the maintenance procedure difficult. The printing resolution can be up to 100 µm and depends on parameters such as the surface energy of the substrate, the ink viscosity, the mesh size, and counts.

Screen printing is one of the most robust technologies that have been developed in the last few years as it is compatible with a wide range of substrates such as textiles, clothes, papers, and glass. This process requires high-viscosity inks (up to 10 000 cP) and the deposited thickness can go up to few hundreds of microns. Screen printing is already used to print interconnections for printed circuit boards and antennas for RFIDs (Radio Frequency IDentifications), active layers [34] and electrodes [35, 36] for solar cells or organic thin-film transistors [37], and emitting layers in OLEDS [38–40] (Figure 2.14). Many studies [41–43] and patents [44–46] have reported the use of screen printing for biosensing applications. They demonstrated the feasibility of such devices to detect bioelements or to fabricate amperometric biosensors.

2.3.2.1 Aerosol Jet

Aerosol jet printing has been developed since the 2000s and shows a promising future for industrial applications. This technology allows printing a large range of ink viscosities (from 1 to 2500 cP). The principle is based on the ejection of fluids even with nanoparticles from a chamber to the substrate (Figure 2.15), a new technology that uses the atomization of suspension by an ultrasonic or a pneumatic system. The formed droplets are guided by a jet stream up to the printing nozzles. The aerosol printing can produce conformal coating and resolution down to 10 µm with processing speed up to 1.2 m min⁻¹.

Few studies mention aerosol printing in literature [47, 48]. Recently, it has been evaluated for diverse applications such as solar cells and thin-film transistors. For instance, aerosol jet has been used to print the conducting polymer as an active layer in solar cells [49] and to fabricate electrodes for TFTs [50]. Aerosol has been also studied for the deposition of DNA or enzymatic layers and for the functionalization of electrodes in biological sensors [51].

2.3.2.2 Inkjet

Inkjet is a noncontact technology commonly used for daily use printing applications (home or office). This process can be integrated in industry in roll-to-roll processes in the production line. Advances in these technologies have been seen recently, for instance, the standardization and commercialization of 3D printing and the creation of new functional inks for electronics.

2.3.2.2.2 Basic Principles

The inkjet mechanism is based on the formation and the ejection of fluid droplets, released from a chamber under the variation of internal pressure in the printhead cavity and through nozzles. This fluid ejection depends on rheological parameters such as the ink’s properties (surface tension, viscosity, density) and the chamber’s pressure. The resolution can reach 20 µm depending on the nozzle diameter, drop volume, surface energy of the substrate, and drop-to-drop spacing. Originally, two technologies have been developed for inkjet: CIJ and DOD technology (Figure 2.16).

CIJ (Continuous Inkjet) Printing

The CIJ technology consists of the generation and the ejection of continuous drops. Those droplets pass through an electrostatic charging electrode and then a high-voltage deflector plate leads trajectory deviation toward a recycling recipient. Two different categories of CIJ have been developed: the binary and the multiple deflections. In a binary deflection system (Figure 2.17) there are two states regulated by the electrostatic deflector. The drops that are charged during the printing step are going directly to the substrate and the non-charged ones are deviated and collected in a gutter and then recycled. In a multiple deflection system, the non-charged drops go to the gutter to be reprocessed while the charged drops are deviated to be deposited onto the substrate.

CIJ technology is used in industry for marking and encoding the products before commercialization. It presents advantages such as good printing resolution down to 50 µm, high ejection frequencies of droplets up to 60 kHz at high printing speed, and good process stability without nozzles clogging. However, this technique shows limitations regarding the type of inks to be printed, which have to match the requirements of the system especially for the electrostatic charging step. The ink recycling step can induce materials contamination. Finally, printing of complex designs and customization is not easy because of the continuous ejection of the drops. This is why this technology has never been adopted for printed electronics.

(DOD) Drop-On-Demand Inkjet Printing

The drop-on-demand technique can be divided into four categories: piezoelectric, thermal, electrostatic, and acoustic inkjet. Thermal and piezoelectric inkjets are the most successful technologies adopted in industry and laboratories.

In thermal inkjet, the heating element is integrated within the printhead chamber. A local increase in the temperature of the ink, generally up to 300 °C, creates an air bubble within the ink inside the ejection chamber. This change in local pressure provokes the formation and the ejection of individual droplets (Figure 2.18a). Today, thermal inkjet is widely used for graphic printing and especially for home printers. However, requirements for heating before ejection make this technology very limited in terms of inks. A few studies mentioned the possibility of fabricating biosensors [59, 60]. For instance, Setti et al. [61] reported fabricating an amperometric sensor for the detection of glucose.

The piezoelectric inkjet technology is based on the use of a piezo membrane (Figure 2.18b). This membrane is mechanically deformed when an external voltage is applied thus causing a change in the pressure inside the printhead chamber. This causes the formation of the droplet in the aperture in the printhead. The jetting of the ink is given by the pressure release and a large type of inks can be used. Here, the nozzle generates a droplet with a volume of a few picoliters. The ink viscosity ranges from 5 to 20 cP and the resolution down to 20 µm.

With the rise in printed electronics, several companies such as FujiFilm, Ceradrop, and Meyer Burger developed new inkjet machines that can print different functional materials using piezo-DOD technology (Figure 2.19). The printing parameters such as the voltage applied to the piezo element, the drop spacing, the waveform, and the temperature of the ink can be tuned. Some printers offer the possibility to realize the alignment to print complete devices with several layers and different inks. The machines show flexibility and are easy to use, making the process adaptable to different materials.

Inkjet offers many possibilities for new developments. In the literature, inkjet is widely studied for electronic applications such as for organic thin-film transistors (OTFTs) [62, 63], OLEDs and solar cells or organic photovoltaics (OPVs) [64–68], and sensors [69]. Studies demonstrated the feasibility of using commercial PEDOT:PSS ink to print OTFTs with dimensions in the range of micrometer [70, 71] and sub-micrometer [72] scales by controlling the hydrophobicity of the substrate. Bharathan et al. [73] combined inkjet with spin-coating to pattern electroluminescent devices. Inkjet has also been used to print active layers of solar cells [74, 75] with device efficiency above 3%. Printed conducting polymers have also been studied for chemical sensing such as ammonia detection [76, 77].

2.3.2.2.3 Printability Requirements

In the DOD technology, the formation and ejection of the droplet depend on different rheological parameters of the ink such as viscosity, density, and surface tension (Figure 2.20).

Printability of the ink depends on these rheological parameters combined with the characteristics of the printhead (nozzle diameter, frequency of the drops’ ejection, applied voltage, channel length, and internal pressure). The rheological behavior of the ink can be described with dimensionless parameters such as the Reynolds number and the Weber number or the Ohnesorge number, defined as

2.5

where μ is the viscosity of the ink, is the density, is the surface tension, R is the radius of the nozzle, and is the velocity.

These numbers are directly dependent on the fluid’s rheological properties and they are used to predict the ejection of the ink from the nozzle. In the 1980s, Fromm [78] defined the criteria for ink printability such as . Later, Reis and Derby [79, 80] discussed Fromm’s assumption and propose their own rheological requirements for ink jetting. Using a numerical simulation of drop formation, Reis suggested that the condition is more adapted in order to match the ink with the printing requirements. They propose a chart [80] where one can define which area is the most adequate combing the above-mentioned dimensionless number with DOD printing. The lower limit corresponds to a high viscosity domain that prevents the jetting while the upper limit causes the formation of unwanted satellite drops. This theory is often used in the literature [81–83] for the formulation of new functional inks dedicated to inkjet printing technology.

2.3.3.1 Metallic Inks

The inks are usually composed of metallic nanoparticles, for instance, gold, silver, or copper in suspension in solvents. Surfactants can be added to the formulation for the tuning of surface tension to fit the requirements of the printing process. Many manufacturers offer a large choice of metallic inks for printing processes and the market is in expansion. Different methods exist to produce nanoparticles, such as chemical reactions or mechanical attrition. The first method is based on the use of precursors [84] where an organic complex containing metal salt is dissolved in a solvent. For the mechanical attrition, ball milling [85] is usually used to downscale materials to nanoscale. The size of nanoparticles can vary from a few to hundreds of nanometers and they can be produced in different shapes such as nanoplates or nanospheres depending on the preparation conditions. These nanoparticles are subsequently embedded in organic shell polymers for protection against the oxidation and for solubility in water or in vehicle solvents. Poly(vinyl pyrrolidone) (PVP), for example, is one of the most commonly used polymers for the encapsulation of nanoparticles.

Thermal curing is a key step in the fabrication of metal structures. After printing, an annealing step allows the evaporation of organic compounds and the coalescence of nanoparticles. New types of selective curing techniques such as microwave curing [86], laser sintering [87], or photonic sintering [88] permit the coalescence of nanoparticles without thermomechanical degradation or damage to underlying materials such as plastic substrates. Electrical conductivity of between 10% and 30% of that of bulk metal is generally obtained, depending on the thermal cycle and the size and the shape of the nanoparticles [89].

2.3.3.2 Dielectric Inks

Dielectric inks are usually used in printed electronics to fabricate dielectric isolation layers in circuits. They can be UV-based inks also composed of nanoparticles (TiO₂, BaTiO₃). UV-curable inks comprise monomers, oligomers, photoinitiators, and additives [90]. Photoinitiators absorb energy from UV leading to the polymerization of the monomers, for instance, acrylic monomers after exposure to UV light (Figure 2.21).

These UV inks offer advantages such as facile processing without clogging the nozzle. They avoid the evaporation of volatile organic compounds from the ink since polymerization occurs without the need for a drying step. The polymerization step takes a few seconds and can be easily integrated in a roll-to-roll process.

2.3.3.3 Conducting Polymer Inks

Since their discovery in the 1970s by Shirakawa et al. [91], conducting polymers have been broadly studied because of their electrical conductivity, high mechanical flexibility, and long-term stability. With the emergence of printed electronics, conducting polymers are formulated in new conducting inks. Polyaniline, polypyrroles, polythiophenes, and PEDOT:PSS are especially considered for electronic applications. They are dispersed in organic solvents, and secondary dopant solvents such as ethylene glycol or dimethyl sulfoxide (DMSO) are added to the formulation to boost the conductivity and to fix the viscosity while surfactants such as Triton X-100 adapt the surface tension to match the substrate surface energy.

Polypyrrole is a dielectric polymer that can be doped by oxidation process to reach conductivity up to 100 S cm⁻¹ [92]. This polymer is usually electrochemically deposited [93] for applications such as capacitors [94] or chemical sensors [95]. Only few studies mention polypyrrole as an ink [96, 97] for printed electronics applications.

Polyaniline is a conducting polymer that combines electronic and optical properties. Polyaniline exists in three stable oxidation states that exhibit different conductivities and colors. Among them, the emeraldine state presents the highest conductivity in its doped state. Polyaniline has advantages such as stability at room temperature, ease of synthesis, and transparency. Companies such as Panipol are developing polyaniline inks compatible with inkjet and others process. This polymer has been studied for applications such as gas sensing [98], especially ammonia [77, 99], combined with graphene for capacitor electrodes [100] and recently for biomedical applications such as biosensors [101, 102]. However, this polymer is not compatible [103] for implantable biodevices.

PEDOT in its neutral state is a conducting polymer but remains difficult to process mostly because it is insoluble and rapidly oxidizes in air. In combination with PSS, a water-soluble polymer that serves as a charge-balancing counter ion [104], the complex is stable in solution. In its oxidized state, PEDOT:PSS is a stable polymer and it is one of the most studied polymers currently used for printing transparent electrodes [105] or hole transport layers [35] in applications such as solar cells and OLEDs. PEDOT:PSS inks are available commercially in water-based formulations from a few suppliers such as Heraeus and Agfa (Orgacon) that are compatible with printing technologies such as inkjet or screen printing. PEDOT:PSS thin films present high transparency, good thermal stability, and high conductivity, which can be even tuned by doping the polymer with solvents [106]. Recent research [107, 108] highlights PEDOT:PSS as a good candidate for biomedical applications owing to its higher electronic and ionic properties combined with its biocompatibility[109].