Additively Manufactured Polymers for Electronic Components

Abstract

Over the last decade, polymers have attracted increasing attention for the fabrication of electronic devices due to the innovative results that can be achieved using additive manufacturing (AM) processes. Intrinsically conductive polymers are commonly used to obtain flexible and stretchable devices. They also enable the customisation of electronic devices when processed through AM. However, their main limitation is the reduction in electrical conductivity under mechanical deformation, such as bending. Extrinsically conductive nanocomposites, incorporating conductive fillers into polymer matrices, demonstrate the ability to retain electrical conductivity even following repeated bending, presenting a promising solution to the limitations of intrinsically conductive polymers. However, a gap remains in optimising their processing conditions for diverse 3D printing technologies. Moreover, fillers should be carefully selected according to the application’s specific needs. Dielectric polymers are also very promising for various electronic applications, but they are less investigated and have lower visibility than their conductive counterparts. This review presents three classes of polymer materials, i.e., intrinsically and extrinsically conductive polymers and insulators, discussing their advantages, drawbacks, and applications for 3D printing in electronics. This overview concludes with assessing future investigation areas needed to unlock the possibilities of 3D-printed polymers in electronics.

1. Introduction

There is an increasingly prominent trend of using polymers in the electronics industry, which is currently valued at $5.9 billion. The market is expected to grow to $7.7 billion by 2028, with a projected compound annual growth rate (CAGR) of 3.8%. Polymers are used in the electronics field for various applications, such as flexible electronic circuits, devices, sensors, actuators, light-emitting devices, energy harvesting, and packaging.

Conventional processes suited to wafer-scale fabrication of electronics, such as high-temperature doping, vacuum-based deposition, photolithography, and dry or wet etching, are complex and costly, requiring cleanrooms and specialised equipment. By contrast, printing techniques that use liquid inks and simpler, non-vacuum processes offer a promising alternative for scalable and low-cost electronics manufacturing [1].

In the last few years, 3D printing has been utilised to fabricate embedded electronics, 3D structural electronics, conformal electronics, stretchable electronics, and organic light-emitting diodes (OLEDs). Three-dimensional printing applications have significantly expanded, and 3D-printed electronics are considered the emerging frontier in additive manufacturing (AM) [2]. New markets and applications have been fostered by the possibility of integrating functional components, such as sensors and circuits, by pausing the printing process at any specific layer [3]. Furthermore, AM technologies can create electronic devices on stretchable, flexible, biocompatible substrates to enhance conformability to irregularly shaped surfaces [4]. Therefore, the application of 3D printing in electronics presents significant potential for the fabrication of complex objects with multifunctional properties.

In the context of AM for electronics, polymers are emerging as the material of choice to fabricate flexible electronic devices [3,5], and they can be either intrinsically or extrinsically conductive [6]. Intrinsically conductive polymers (ICPs), also known as conjugated polymers, have a conjugated π-electron, enabling electron transfer and conductivity. Extrinsically conductive polymers (ECPs) are insulating polymers made electrically conductive by adding fillers, such as metal particles, carbon nanotubes (CNTs), and graphene, which allow electronic transfer [7,8]. In addition to this, special effort has been channelled into developing thermally conductive polymer nanocomposites that can be additively manufactured [9,10]. More importantly, insulating polymers are also employed in electronics to avoid short circuits, as well as to protect and isolate devices. These three categories of polymers will be considered in the following section.

Moreover, an additional challenge for the electronics industry is how polymeric materials are deposited on the substrate or patterned surfaces. Nowadays, the electronics industry relies on the planar fabrication method mainly through lithographic techniques. Therefore, the fabrication of a third dimension is unexploited, limiting design choices. AM technologies are being studied as alternative fabrication methods. Furthermore, most of the studies in the literature present the use of metallic materials [11,12], such as silver nanoparticles (NPs) [13], copper [14], CNTs [15,16,17], and graphene-based [18,19] and graphite-based materials [20,21], for AM owing to their exceptional electrical properties. Of all the materials manufactured via AM for electronics, polymers are the least explored because most electronic components require conductive materials to operate. Therefore, the majority of research and development efforts have been directed toward this specific need.

AM has several advantages, such as: (i) the ability to process a large variety of materials, i.e., metals, polymers, and ceramics; (ii) the reduction of waste; (iii) design flexibility; and (iv) the possibility of manufacturing structures and morphologies with peaks, valleys, pillars, or cavities characterised by a considerable height difference. However, certain issues are raised at times, such as a lower level of resolution and lower fabrication speeds, which are relevant for industrial purposes, compared to more conventional lithographic technologies. There are several types of AM technologies. Among these, vat photopolymerisation is an AM process in which the curing of a liquid photopolymer selectively occurs by light-based irradiation. Thus, a light source, e.g., a laser, selectively targets the liquid resin, solidifying it layer by layer [22]. There are two main types of vat photopolymerisation technologies, precisely, stereolithography (SL) and digital light processing (DLP) (Figure 1a) [23]. The former uses a laser that selectively irradiates the polymeric resin and cures the material to fabricate the desired object layer by layer. This process works by triggering a chemical reaction, which can be either free radical or cationic polymerisation, within a light-sensitising resin. When exposed to light, small reactive molecules in the resin link together to form a solid, cross-linked structure, allowing the object to take shape one layer at a time. The main advantage of this technology is a relatively higher resolution, defined by the laser spot size, compared to other AM technologies. Limitations include the limited fabrication speed and the rheological characteristics of the material, i.e., the viscosity, which cannot be too high to allow reflow under the printing platform. The latter instead employs a digital mask to irradiate the polymeric resin. The digital mask has the shape of every single layer to be printed. The entire surface of a layer is exposed at once thanks to a dynamic projection system (often based on micromirror arrays). It is thus possible to drastically increase the fabrication speed since the curing of a single layer occurs with a single irradiation. The resin cures at once across the illuminated pattern, following the same basic chemical reaction principles. However, the resolution is much more limited because it depends on the resolution of the digital mask.

The powder bed fusion process consists of local melting or sintering of a powder of thermoplastic polymeric materials (Figure 1b) [23]. A roller uniformly distributes the polymer powder across the building platform. Subsequently, a laser selectively heats the powder, facilitating bonding of the particles through melting or sintering processes [24,25]. Here, the laser acts like a precise heat source, melting or partially melting the surfaces of the powder particles. As the particles cool, they fuse to form a solid layer. The exact mechanism, which implies melting versus sintering, depends on the material and temperature, but, in both cases, the key is achieving enough heat to enable inter-particle bonding without degrading the polymer. Inkjet printing (IJP) is an AM technology based on the ejection of ink droplets through a nozzle. It is broadly categorised into two groups, precisely, continuous inkjet (CIJ) and drop-on-demand (DoD) inkjet. In CIJ systems, droplets are continuously ejected using an electric field and a charging electrode. While DoD printing systems eject droplets using a voltage waveform with piezoelectric or thermal inkjet methods, each is suitable for various printing resolutions, materials, and applications [26]. Piezoelectric inkjet methods use a mechanical structure known as a piezoelectric inkjet printhead (PIP) (Figure 1c) [27]. This device has an ink chamber connected to an ink cartridge through a restrictor. A piezoelectric membrane at the chamber’s top pressurises the ink with an electrical pulse, increasing fluid velocity at the nozzle to form droplets. Instead, in a thermal inkjet printing (TIJ) printhead, an electrical pulse heats the fluid in the resistor, creating a vapour bubble that pushes the fluid through the nozzle to form droplets [22,28,29]. In both cases, the goal is to generate small pressure changes inside the printhead that force tiny ink droplets with a volume of a few picolitres out of the nozzle. Once on the surface, the droplets solidify either by evaporation, cooling, or chemical curing, depending on the ink’s composition. Aerosol jet printing (AJP) is a novel fabrication technology that belongs to this group. It relies on focusing an aerosol stream of the material under consideration on a substrate [25,30,31]. AJP uses a directed aerosol stream to ensure consistent deposition (Figure 1e) [32]. It is a non-contact technique, and the nozzle is placed from 1 mm to 5 mm away from the substrate [33]. This approach enables precise patterning of more intricate surfaces [34]. AJP is a technique for the exact digital placement of droplets in precisely defined positions. There are two main types of AJP systems based on the atomisation method, namely, the ultrasonic and the pneumatic AJP [35]. In principle, the AJP technique is compatible with any material that can be suspended within an aerosol. Commercial systems employ ultrasonic or pneumatic atomisation to create aerosols from inks with viscosities spanning from 1 mPas to 1000 mPas. The working principle involves generating a fine mist of droplets, and then focusing this mist into a narrow stream by using a surrounding sheath gas. This focused aerosol jet can be precisely guided onto the substrate, making it ideal for printing fine features, even on irregular surfaces, with minimal overspray.

Material extrusion processes encompass AM technologies based on the continuous flow of inks or melted polymeric filaments through a nozzle (Figure 1d) [23]. The deposition of inks is also referred to as direct ink writing (DIW) [36,37,38]. Ink filaments are deposited layer by layer until the entire structure is created. This technology offers several advantages, including flexibility and low processing costs. It operates at room temperature. However, the materials used in this process typically have poor mechanical properties, with the subsequent need for the addition of reinforcement fillers [23,24]. Specific rheological properties are required to print polymeric inks using DIW [39,40]. The ink should be shear-thinning, i.e., the viscosity decreases with increasing applied shear rate or deformation, thus allowing proper extrusion of the material. Furthermore, the ink should also have a sufficiently high zero-shear rate viscosity, i.e., the viscosity of the material when no deformation is applied, to keep its shape after the extrusion process. The ink is pushed through a nozzle using air pressure or a mechanical plunger and, once extruded, its rheological behaviour ensures it retains its shape. Depending on the formulation, the material solidifies by light-induced curing, solvent evaporation, or heat, allowing the printed shape to set as it builds up layer by layer. After the deposition of the material on the substrate, the ink needs to solidify. Solidification can happen upon UV irradiation to cross-link a photopolymer by solvent evaporation or by thermal curing. The process of material extrusion for melted polymers is known as fused deposition modelling (FDM), also referred to as fused filament fabrication (FFF). This method involves melting and extruding a thermoplastic polymer. A solid filament is fed into a heated nozzle where it melts. The molten polymer is then deposited in thin strands that cool and solidify upon contact with the built surface. As each new layer is laid down, thermal bonding occurs between layers, helping to form a cohesive, three-dimensional object. Other AM technologies have recently begun using polymeric materials, such as laser-induced forward transfer (LIFT) technology. LIFT is a printing process that allows the deposition of solid and liquid materials with high resolution. Bohandy et al. presented it for the first time almost 30 years ago (Figure 1f) [41]. Since then, various LIFT-based methods have emerged to customise the process according to the printing material properties and application requirements. The transfer of pixels in a solid phase involves irradiating a thin layer of an absorbing material (the donor) with a pulsed laser through a transparent substrate. This interaction generates high local pressure, ejecting small pixels from the donor onto a nearby target substrate (the receiver). Transparent layers can also be transferred using a dynamic release layer (DRL) between the substrate and the transferable film [42]. The principle behind LIFT is based on using a laser pulse to generate a rapid pressure or thermal effect in a thin donor layer. This causes a small amount of material to be ejected and deposited onto a nearby surface, without any contact between nozzle and substrate. This mechanism allows for high-resolution patterning and is particularly useful for complex or viscous materials that are otherwise hard to print. The shape and size of the ejected material are affected by the dimensions and the features of the incident laser spot [43].

Figure 1. Schematic representing the AM technologies mentioned in this review: (a) vat polymerisation [23], (b) powder bed fusion [23], (c) inkjet printing [27], (d) direct ink writing [23], (e) aerosol jet printing [32], and (f) laser-induced forward transfer technology [41].

Figure 1. Schematic representing the AM technologies mentioned in this review: (a) vat polymerisation [23], (b) powder bed fusion [23], (c) inkjet printing [27], (d) direct ink writing [23], (e) aerosol jet printing [32], and (f) laser-induced forward transfer technology [41].

Considering all of these AM processes, this review provides a comprehensive overview of polymers in AM for electronics, assessing the advantages, limitations, and applications of various polymer classes, including ICPs, ECPs, and insulators. It also highlights the potential of 3D-printed polymers in electronics and suggests future research directions in the field.

2. Methodology

Web of Science was used as the research database for this review. The following search string was used: (additive manufacture* OR 3D print* or 3DP) AND (polymer* and nanocomposite*) AND (electronic*). The search was refined for the years 2025–2013. The newest and most relevant papers were then considered and cited. Papers outside this database and older ones were referred to and cited when they were particularly interesting for the specific topic.

3. Polymers Used for AM in Electronic Applications

3.1. Intrinsically Conductive Polymers (ICPs)

Although ICPs exhibit lower mobility and conductivity when compared to their inorganic counterparts, like silicon and copper, they have emerged as significant contenders in the electronics and optoelectronics industries. This is primarily due to the possibility of tuning their properties according to specific application needs and to the ease of processing them as a solution. These characteristics, along with their cost effectiveness and suitability for near-room temperature solution processing, make ICPs attractive for consumer electronics [44]. Furthermore, CPs possess unique chemical, physical, and processing properties that position them as competitive candidates for the next generation of wearable or implantable electronics. These applications demand stretchable electronic devices, accommodating the three-dimensional movements of the human body while preserving their electrical performance. Many of the solution processing techniques developed for CPs in applications such as organic photovoltaics (OPV), OLEDs, or organic field-effect transistors (OFETs) can be adapted for the fabrication of wearable and epidermal organic electronics [28]. ICPs are composed of unsaturated building blocks, including arenes, olefins, or acetylenes, which are linked by single bonds that possess some π-bond character. This arrangement results in the formation of extensive domains containing delocalised and polarisable π-electrons. Some classical examples of ICPs include polyacetylene, poly(p-phenylene), poly(p-phenylene vinylene), poly(p-phenylene ethynylene), and polyfluorene. The potential for synthesising a wide range of diverse conjugated polymers is evident. This can be achieved by altering the aromatic repeat unit, such as substituting larger polycyclic aromatic hydrocarbons for the benzene ring, or by modifying the position of the substitutes on the aromatic rings, as exemplified by poly(m-phenylene) and poly(o-phenylene). Moreover, the incorporation of heteroatoms into the aromatic ring opens avenues for even more complex conducting polymers, including polythiophene, polypyrrole, polycarbazole, or the addition of a functional group to the backbone of the polymer, as seen in polyaniline, as shown in Figure 2 [44,45]. Among all of the conjugated polymers, the most used and investigated is poly(3,4-ethylenedioxythiophene):poly(styrenesulfonic acid) (PEDOT:PSS), represented in Figure 3. It is widely recognised as the most remarkable conducting polymer due to its adjustable conductivity, high transparency, excellent film-forming properties, and good thermal stability [46].

Recently, there have been notable advancements in DIW of PEDOT:PSS [46,47]. This innovative and versatile PEDOT:PSS processing method enables precise control over the diameter and arrangement of printed strands in a localised and highly efficient manner. As a result, it offers significant potential for developing microelectronic array devices. PEDOT:PSS pillars with varying aspect ratios have been successfully created by adjusting various printing parameters, such as pulling speed, pulling time, polymer solution concentration, and tip diameter. This includes the production of high-aspect-ratio pillars measuring 7 μm in diameter and 5000 μm in height [48]. Wearable electronic devices have also been developed by combining PEDOT with another semi-conductive polymer, poly(3-hexyl thiophene) (P3HT). In the study conducted by Sarojini et al., the PEDOT:PSS layer, approximately 300 nm thick, exhibited a low sheet resistance of around 70 Ω sq^–1 [49].

A PEDOT:PSS aqueous dispersion was employed for in situ printing using SL into a hydrogel matrix already formed by photopolymerising poly(ethylene glycol) diacrylate (PEGDA), as depicted in Figure 4. Computer-aided design-based architectural models featuring square pores and different line spacing options of 500, 600, and 800 μm were chosen to guide the printing process of 3D strctures via UV laser exposure, resulting in well-integrated scaffolds with predefined geometries [50]. Recent research has focused on the development of three-dimensional PEGDA:PEDOT structures produced using SL for the long-term monitoring of volatile organic compounds. The researchers demonstrated the feasibility of monitoring hazardous compounds by controlling variations in structure and conductivity [51].

Furthermore, wearable electronic devices have been produced using electrohydrodynamic (EHD) printing of inks based on PEDOT. One example is the utilisation of PEDOT:PSS ink mixed with poly(ethylene oxide) (PEO) to create a uniform EHD printing solution, where an electric field is employed to generate printed patterns. The inclusion of PEO resulted in an increase in ink viscosity. By printing PEO/PEDOT:PSS walls with varying numbers of layers, three-dimensional structures were achieved, featuring wall widths ranging from 49.5 to 62.5 μm and wall heights spanning from 0.77 μm to 57.25 μm [52].

DIW is another technology that is suitable for AM of PEDOT:PSS. For this technology, PEDOT:PSS inks are prepared by dissolving the polymer in a suitable solvent. As stated in Section 1, the ink needs to have shear thinning behaviour to allow the extrusion process. After extrusion, the ink should have sufficiently high viscosity to maintain its shape until the hardening process. One simple and efficient way to address this problem is to use conducting polymers, producing highly stable inks by simply adjusting the concentration [53,54]. Wu et al. reported the DIW of PEDOT:PSS to fabricate supercapacitors. Different PEDOT:PSS inks were fabricated, changing the polymer concentration from 1 wt% to 9 wt%. With increasing PEDOT:PSS concentration, a transition from a low-viscosity liquid-like state to a more dough-like state occurred due to the increased entanglement of the PEDOT:PSS chains (Figure 5a). Figure 5b illustrates the shear rate-dependent rheological viscosity of inks with different PEDOT:PSS contents. Higher viscosities were observed in inks with a higher quantity of PEDOT:PSS. Additionally, all inks exhibited shear thinning behaviour, with apparent viscosity decreasing as the shear rate increased. To assess ink printability, interdigitated electrode patterns were printed using all of the inks, and the relationship between viscosity and printing behaviour was summarised (Figure 5c). Inks with low PEDOT:PSS concentrations (1~3 wt%) were unable to maintain the pattern shape, as they exhibited lateral spreading after printing. By contrast, inks with high PEDOT:PSS concentrations (7~9 wt%) frequently led to nozzle clogging due to PEDOT:PSS aggregates in the ink, resulting in broken patterns. The ink with an optimal concentration of 5 wt% PEDOT:PSS allowed for stable printing, maintaining a well-defined pattern without spreading after printing [47].

Recent efforts have extended additive manufacturing (AM) technologies, especially DIW, to other ICPs. For example, fully additive fabrication of polyaniline (PANI)-based electrodes has been demonstrated via DIW, enabling 3D printing of electrical circuits [55]. In another study, PANI nanofibres were dispersed in photocurable acrylic resins to enable DLP printing of conductive microstructures with enhanced print quality and surface conductivity, overcoming the limitations of pure PANI systems [56]. Moreover, pressure sensors have been fabricated via FDM-DIW using PANI and CNTs within acrylic matrices, achieving a conductivity of around 15 × 10⁻³ S cm⁻¹ with a filler concentration of 2 wt% [57].

Furthermore, polypyrrole (PPy) has also been integrated into direct-write processes. DIW printing has been used to fabricate PPy-containing 3D electrodes and conductive hydrogel scaffolds, such as PPy nanotube inks printed for flexible supercapacitors [58] or composite hydrogels (PPy/alginate-gelatin), delivering conductivity values of up to 1.5–2 S m⁻¹ [59].

3.2. Extrinsically Conductive Polymers (ECPs)

The electrical behaviour of ECPs can be described by what is called the percolation threshold, which defines the particular relationship between electrical conductivity and filler content (Figure 6) [52].

An evident increase in the material’s electrical conductivity occurs at a specific volume fraction of filler, known as the percolation concentration. This concentration is directly related to the type of filler used and its dispersion within the matrix. When the filler concentration is below this threshold, the material acts like an insulator, exhibiting an electrical conductivity similar to that of the material without any filler added. By increasing the filler content, a rapid increase in the electrical conductivity of the material is observed. This is due to the formation of a three-dimensional network between the filler particles, which ensures electron transfer. It depends not only on the particles’ dimensions but also on the shearing rate used in the dispersion procedure. The main fillers used for polymer nanocomposites are metallic particles [62,63], ceramic particles [64,65,66,67], carbon allotropes [68,69,70], carbon fibres [71], and their hybrid combinations [72,73,74,75,76,77,78,79,80]. The interaction between fillers and matrices also has a direct effect on other characteristics of the material, like mechanical ones. Carbon black is one of the main fillers used to impart electrical conductivity to polymeric matrices, but it is responsible for lowering the elongation at break. Hence, the filler loading must be low enough to maintain the good mechanical properties of the material. Percolation concentrations of carbon black and CNT composites can be respectively below 0.5 wt% and 0.1 wt% [81]. CNTs and graphene are both advanced carbon-based materials that possess unique properties, making them preferable in electronic applications compared to other fillers, such as carbon black. CNTs and graphene are ideal for mechanical reinforcement in nanocomposites thanks to their exceptional mechanical strength and stiffness, with a tensile strength several times greater than steel. Moreover, they are excellent electrical conductors, suitable for electronic applications like transistors [82], sensors [83,84,85,86], and other devices [60,87,88,89,90].

3.2.1. Carbon Nanotubes

Theoretically, carbon nanotubes have great mechanical properties due to the strong s bonds that characterise them. To make a comparison, multiwall CNTs (MWCNTs) have a tensile strength of 11–63 GPa [91], while Kevlar is characterised by a tensile strength of 3 GPa [92]. MWCNTs may present some percentage of defects due to their growing mechanism, so their mechanical properties could be experimentally lower than their theoretical ones. Their Young’s modulus is independent of chirality but not of diameter, and the highest value of 1 TPa can be obtained for diameters of 1 nm and 2 nm.

CNTs have been widely investigated in the literature and processed both in various forms and with different technologies [93,94,95]. In analogy with ICPs, the fabrication of CNT-based materials with AM has also gained wide interest in industrial applications [96]. Electrical conductivity was enhanced by coupling PEDOT with CNTs. In fact, a π–π interaction is established between the thiophene rings of the PEDOT backbone and MWCNTs, allowing the charge to be more delocalised [48]. PEDOT is a highly conductive polymer that is suitable for use in biosensors and energy storage applications. Its electrical conductivity arises from the delocalisation of π-electrons within its chemical structure, as well as the presence of sulfonated polystyrene (PSS) [97]. Three-dimensional samples were obtained and showed good mechanical properties and considerable electrical conductivity (0.05 S/cm¹). Epoxy and acrylic CNT composites have been largely investigated due to the possibility of being directly printed through vat photopolymerisation processes. SL can effectively print functional polymeric materials with tailored chemical-physical properties by optimising the formulation [98]. Nevertheless, a lot of critical aspects should be taken into account when UV-curing these systems. Gonzalez et al. [99] developed a photocurable acrylic resin with CNTs for SL printing using PEGDA and poly(ethylene glycol) methyl ether methacrylate (PEGMEMA) as monomers, with CNTs dispersed via sonication. PEGMEMA was used to decrease the viscosity of the resin, enhanced by the addition of the filler, since low viscosities are required for the material to flow inside the vat for this 3D printing technology. Good electrical properties were obtained, reaching a value of 2 × 10⁻⁵ S/cm for a filler content of up to 0.5 wt%. The problem related to UV-curing of these types of systems is the shielding effect of CNTs, which makes polymerisation much more difficult. In fact, CNTs contribute to the absorption of UV light, competing with the photoinitiator. This results in a decrease in the cross-linking density, which reduces both resolution and printability. This drawback can be overcome by increasing the intensity of the SL light source or by increasing the photoinitiator content.

Hence, 3D printing can be implemented to directly fabricate electrically conductive systems, such as microelectromechanical systems (MEMS), as illustrated in Figure 7. They are devices that combine electric and mechanical functions, such as pressure sensors, accelerometers, and gyroscopes. They are commonly produced via integrated circuit (IC) batch processing techniques and can have dimensions ranging from a few micrometres to millimetres. These devices can sense, control, and actuate on a microscale and have an impact on the macroscale. IC processing involves photolithography and chemical etching to add or subtract two-dimensional layers on a substrate (typically silicon). Among all AM techniques, SL, DLP, and two-photon lithography (2PL) are suitable for printing these microstructures, as they offer high resolution and improved surface finish. However, they can work with a limited number of materials, and there is a need for a metallic finish on the printed object to provide electrical conductivity to the device since conventional MEMS are made of silicon and glass [100]. A way in which a metallic finish is applied is a wet process through electroless metal deposition. This process allows the coating of an insulating substrate with a metal, which provides it with the high level of electrical conductivity that is required for its applications. Therefore, the need for wet metallisation in electroless metal deposition can be avoided by developing a printable conductive material with an electrical conductivity comparable to that of metals.

3.2.2. Graphene

Graphene is the strongest nanomaterial reinforcement [101], showing a two-dimensional structure of sp² hybridised carbon atoms arranged in a hexagonal lattice. Graphene acts as the fundamental structural component in various carbon allotropes, including carbon nanotubes, graphite, fullerenes, and others. As a nanomaterial, graphene exhibits a wide range of properties that extend beyond specific domains. It is renowned for displaying phenomena such as the room-temperature quantum Hall effect, carrier mobility [102], and the degeneration of valence and conduction bands at the K-point (Brillouin zone). These properties can be modified by altering the deposition of layer-by-layer graphene sheets, akin to the structure of graphite. In terms of electronic properties, bi-layer graphene demonstrates parabolic electronic bands with an open energy gap [103], while tri-layer graphene exhibits a combined band structure. The number of layers is crucial in changing the electrical properties of graphene, as an increase in the number of layers leads to valence band–conduction band overlapping [104]. Reinforcement in polymer composites can be obtained using graphene and exploiting its remarkable properties [105]. Wei and colleagues demonstrated that composite structures can be created through 3D printing using a graphene loading of up to 5.6 wt% [106]. Zhang et al. developed a polymer-graphene composite for direct 3D printing using a solution-based approach, achieving a high electrical conductivity of 4.76 S cm⁻¹ with 6 wt% graphene. The printing process aligned the graphene, enhancing conductivity due to the preferential orientation of the graphene particles during the process [107]. These graphene composites enabled the production of flexible printed circuits with strong interlayer bonding strength and mechanical durability in both axial and transverse directions [108]. Three-dimensional-printed samples of poly(lactic acid) and polypropylene filled with 20% graphene nanoplatelets exhibited an electrical conductivity of up to 5 S cm⁻¹ [109]. Acrylic and epoxy polymers are the most commonly used resins for SL technology and, when combined with graphene-based filler, the resolution and the quality of SL-printed parts are affected by the curing strategy, laser intensity, exposure duration, and scan speed [110]. Whereas the addition of photoinitiators and UV absorbers controls the depth of polymerisation [111]. Although SL offers several advantages compared to other AM technologies, including high-resolution printing and nozzle-free processing, SL has some drawbacks if graphene-based inks are used, including issues with graphene dispersion and stability, curing compatibility, limited layer thickness, and material uniformity [110]. SL encompasses various printing methods, such as single vector array, DLP, continuous liquid interface production, and two-photon greyscale lithography. Graphene-based inks are known for their ink formation characteristics, mechanical durability, and excellent conductivity [107]. In this context, graphene oxide (GO) offers advantages over pristine graphene due to its easier production, hydrophilic functional groups, and the possible large-scale incorporation of active materials [112]. GO dispersion in water is a key factor in enhancing ink printability. When GO is uniformly dispersed in water, it creates an isotropic dispersion with characteristics like a nematic liquid crystal [110]. This uniform dispersion also leads to increases in the storage modulus (G′) and loss modulus (G″), as shown in Figure 8 [113]. The concentration of GO can be tuned to control the rheological properties and be adjusted for ink printability according to specific requirements.

3.2.3. MXenes

In recent years, MXenes, which are a family of two-dimensional transition metal carbides, nitrides, and carbonitrides, have emerged as a highly promising class of conductive fillers for additively manufactured polymer composites. Unlike CNTs and graphene, which are well established in enhancing electrical and mechanical properties, MXenes offer several unique advantages, such as surface chemistry and enhanced dispersibility, high electrical conductivity, and good performance in 3D-printed architectures. MXenes feature surface terminations (–OH and –O) that confer hydrophilicity and enable facile dispersion in aqueous or polar solvents [114]. This is particularly advantageous for formulating inks compatible with PEDOT:PSS [115], which are DIW-printed, showing an electrical conductivity of up to 2 S m⁻¹ and excellent mechanical resilience. Similarly, 3D-printed MXene/RGO hybrid aerogels achieved conductivities above 1000 S m⁻¹ and impressive electromagnetic interference (EMI) shielding (max. 86.9 dB) [116].

Moreover, with metallic-level conductivity and tunable surface chemistries, MXenes create robust interfacial coupling with polymer matrices, facilitating efficient electron transport that is similar to graphene but often superior to unfunctionalised CNTs. As a result, ink formulations with MXenes can simultaneously deliver high conductivity and superior mechanical properties [117].

Furthermore, MXenes exhibit excellent biocompatibility along with high electrical conductivity. Hence, they are regarded as more promising materials in the field of biomedical implantable devices, compared to CNTs and graphene, whose cytotoxicity is well known [118]. Nevertheless, MXenes may undergo oxidation over time, which necessitates thoughtful stabilisation strategies [119].

3.2.4. Filler–Polymer Interactions

Understanding and engineering the polymer–filler interface is essential to enhance polymer–filler adhesion and improve conductivity in ECPs. Surface functionalisation of fillers can introduce strong covalent or intermolecular bonding, which can enhance phase affinity, reduce interfacial tension, improve dispersion, and lower percolation thresholds, while reinforcing mechanical cohesion. For example, CNTs and graphene can be functionalised with –COOH, –OH, amine groups, or polymer wrapping to improve interaction with the polymer host, enhance dispersion, and facilitate π–π or hydrogen-bonding interactions [120]. MXenes also present in-built tunable surface terminations (–OH, –F, and –O), enabling strong hydrogen bonding and ion–dipole coupling with polar polymers. This improves dispersion, antioxidant surface passivation, and resistance to humidity-induced degradation [121].

In situ polymerisation is also a strategy successfully applied to polyaniline/CNT composites wherein the polymer matrix forms around the filler, leading to covalent or supramolecular anchoring, and improving electrical pathways and mechanical fatigue resistance [122].

3.3. Insulating Polymers

Insulator materials with high thermal conductivity are becoming promising for electronic applications because standard metallic materials have limitations such as high cost, weight, and electrical conductivity. This has led to the additive manufacturing of different polymer-based nanocomposites containing boron nitride particles for effective thermal management in electronic devices [9,10,65,67,123,124]. Moreover, the dielectric properties of insulating layers in high-density interconnects can significantly influence the performance of electronic circuits and signal propagation speed. Indeed, the polarity of dielectric materials can limit signal propagation speed and the maximum interconnect density on a chip. Furthermore, lower dielectric permittivity enables closer spacing of interconnect lines, reducing crosstalk and increasing device layout density. Therefore, the use of insulating layers with lower dielectric constants (low-k), such as low-k polymer materials, can boost circuit density and speed, making them suitable for cost-effective, high-performance circuits [125].

Benzocyclobutene (BCB)-based materials have gained significant commercial interest due to their low dielectric constants, making them suitable as dielectrics in on-chip connections, such as resin-coated copper foil. Moreover, BCB-based polymers are thermally activated and are obtained as clean reaction products of highly reactive units in precisely controlled curing processes. This is particularly notable in lithography materials, where the control over linear polymers as precursors and the accuracy and reactivity of benzocyclobutene units have resulted in low-error materials [126]. Dow Chemical Co. has developed a range of low-k polymers that are based on various benzocyclobutene monomers, also known as BCB (biscyclo[4.2.0]octa-1,3,5-triene or 1,2-dihydrobenzocyclobutene). One of their commercial products is CYCLOTENE, which is a family of thermosetting polymers derived from the monomer 1,3-divinyl-1,1,3,3-teramethyldisiloxane-bis-benzocyclobutene (DVS-bis-BCB), as depicted in Figure 9. In many instances, CYCLOTENE is simply referred to as BCB in the literature. BCB has been specifically designed for electronic coating applications due to its low-k characteristics, making it ideal as an interlayer dielectric in radio frequency components. Since the early 1990s, it has been widely used in applications such as bumping and redistributing chips, planarisation in flat panel displays, MEMS hermitisation, wafer bonding, passivation, and gap filling. Its versatility comes from its properties, including effective planarisation, minimal outgassing, and low copper migration [127]. Regarding BCB’s fundamental electrical parameters, such as dielectric constant (k), loss tangent (tan δ), and breakdown strength, which are critical for the performance of advanced electronic devices, BCB-based materials exhibit an ultra-low k that is between 2.65 and 2.43 in the range of 1 MHz and 1THz, combined with a very low loss tangent in the range of 0.0008–0.002 [128], which minimises electrical energy losses due to parasitic capacitive effects [129]. The moisture-induced dielectric constant increases by just ~1.2% in 85 °C/85% RH, and the breakdown strength ranges from ~2.25 MV/cm to 3.2 MV/cm, with varying electrode spacing. Compared to other dielectric materials, BCB’s properties enable increased signal propagation velocity with lower crosstalk, closer interconnection spacing, and negligible dependency on moisture absorption, which are essential in electronic circuits and high-density on-chip interconnects [125]. AM of BCB-based material with IJP and AJP was recently presented.

BCB was used with AM technologies for the first time in two studies by our research group, which investigated the processing of this polymer using IJP and AJP. Regarding IJP, an in-depth study was conducted on the ink properties and processing parameters [130]. The study demonstrated that low ejection speeds lead to incomplete droplet ejection, while high speeds produce satellite droplets. Furthermore, it examined the interaction between the substrate and two types of printable inks, which created an array of isolated drops. The prints showed the occurrence of the coffee ring effect (CRE) for inks with enhanced wettability on the substrate. This phenomenon arises from a lower contact angle with the SiO₂ substrate, which leads to CRE and the formation of drops with larger diameters. Furthermore, when printing a square array pattern, CRE results in an accumulation of material, severely compromising the morphological quality. This underscores the need to optimise ink properties to effectively prevent the occurrence of CRE. Material agglomerations and depletions create strong inhomogeneities that negatively impact film quality and should be avoided. Finally, a bonding pattern already obtained with other conventional technologies was successfully printed with BCB ink, demonstrating the ability to print complex shapes down to 130 μm. Metrology confirmed that the printed features matched the design specification (Figure 10).

The same approach was followed using AJP technology. The effects of various printing parameters on printing times and pattern morphology were studied to achieve high printing speeds of 10 mm/s and to ensure uniform line morphology [131]. Although some overspray was consistently present in all prints, optimal printing parameters were identified, reducing printing time while maintaining quality, and allowing large depositions of 3 cm × 3 cm. The study also demonstrated the printing of square patterns with thicknesses ranging from 1 µm to 43 µm and intricate geometries, including interdigitated lines and features as small as ~60 µm, using AJP technology.

Another dielectric polymer widely used in the electronics industry is the epoxy-based resin SU-8. Compared to BCB, SU-8 has a relatively higher dielectric constant (k ≈ 3.0–3.2) and a breakdown strength of about 4–6 MV/cm, which ensures electrical insulation in MEMS and biosensor platforms but may present some limitations in high-density devices [132]. It is preferred for protecting electrodes against ionic solutions, which can lead to corrosion, and for ensuring stability for low-cost electronic components, such as electronic biosensors used in direct contact with ionic solutions. However, SU-8 is typically processed using technologies available in cleanrooms, thus increasing device costs and complicating the fabrication process. To reduce fabrication expenses, the AJP of SU-8 employing ultrasonic aerosolisation for electronic components was demonstrated, as shown in Figure 11. This study showed that SU-8 can be printed without reformulation and effectively acts as a passivation layer for conductive silver lines when exposed to ionic solutions. Enlarging the printed SU-8 film by 100 μm beyond the electrodes, the leakage current was reduced by six orders of magnitude while maintaining stability over 20 voltage sweeps. The optimal post-printing cure time was 15 min at 160 °C, further minimising leakage current [133]. This example and all of the previous works are summarised in

Table 1. Summary of the most used additive manufacturing technologies for processing polymers in electronics, reporting their corresponding resolution, the viscosities of processable materials, and their advantages and disadvantages.

Technology	SL	DIW	LIFT	IJP	AJP
Materials	PEDOT:PSS	PEDOT:PSS	PEDOT:PSS	PEDOT:PSS	BCB
	CNT nanocomposite	CNT nanocomposite		BCB	SU-8
	Graphene nanocomposites	Graphene nanocomposites		Graphene nanocomposites	CNT nanocomposites
Resolution	100 µm	200 µm	20 µm	50 µm	75 µm
Viscosity	Up to 5 Pa·s	Up to thousands of Pa·s	Wide range of viscosities	Maximum 0.005–0.01 mPa·s	For ultrasonic atomisation, a maximum of 0.01 Pa·s
		Shear thinning behaviour	Both Newtonian and non-Newtonian behaviours		For pneumatic atomisation, a maximum 1 Pa·s
Advantages	Relatively high resolution in z-direction	Suitable for all types of inks	Suitable for all types of inks	Suitable for all types of inks	Suitable for all types of inks
	Mechanically stable material	Suitable for high-viscosity inks	High printing speed	High fabrication speed	Can process inks with different viscosities
			Can process inks with different viscosities		Versatility in terms of resolution
Limitations	Fabrication speed is low	Low resolution	Material needs to be coated on a donor substrate	Suitable only for low-viscosity inks	Ink formulation is critical
	Limited to UV-curable materials	Limited to shear thinning materials	Expensive equipment	Not suitable for 3D structures	Overspray
Applications	Volatile organic compound sensors; 3D high-frequency devices	Piezoresistive sensors; Microcapacitors; Tissue engineering	OLED fabrication; Micro-battery electrodes; Biomolecule printing	Sensors; OFETs; OSCs; Wafer bonding	Sensors, Transistors, Wafer bonding
References	[48,51,110,134]	[23,47,54,105]	[41]	[49,102,130,135,136]	[103,131,133,137]

, listing the main AM technologies used for processing polymers in electronics, the corresponding resolution obtainable with AM technologies, the viscosities of processable materials, and the advantages and disadvantages of each technology.

4. Applications in the Electronic Field of Additively Manufactured Polymers

4.1. Sensors

The transition to Industry 4.0 and the growing demand for large-area Internet of Things (IoT) technologies have emphasised the need for certain electronic components, such as cost-effective, low-power miniaturised sensors. These sensors play a crucial role in enabling a more connected world than today. Additionally, addressing challenges like integration densification, affordable flexible substrates, environmental sustainability, cost-efficient fabrication processes, and extending device lifetime has become paramount in this context [135].

To meet these demands, DIW has emerged as a viable manufacturing technique for large-area printed sensors and electronic systems. Although there are still challenges to overcome, direct-write methods show great promise in enabling smart sensors and electronics [39,138]. They continue to be a focus of development in both academic and industrial sectors. Flexible printed sensors, in particular, are seen as fundamental building blocks for IoT, and for sensors and systems for smart city and smart community applications due to their unique advantages [139,140].

Despite extensive research and development, the market for printed electronics is still unexplored, and many potential applications are still waiting to be developed and commercialised. The major issues preventing the translation of lab-scale innovations into marketable products are scalability and the compatibility of materials with AM technologies. In this context, AJP can be a promising technology to address these challenges. AJP offers compatibility with various materials and the ability to print at very fine resolutions down to the micrometre level. Since the early 2000s, AJP has found applications in printed microelectronics, electrochemical systems, and various types of sensors, including those for strain, gas, biological, and optical sensing, among others. It represents a valuable tool to produce diverse sensors and electronic devices [141].

4.1.1. Strain Sensors

Strain gauge devices serve to quantitatively translate applied strain or deformations into variations in resistance, capacitance, or piezoelectricity [142]. Traditional strain gauges typically consist of a sensitive grid (referred to as the active layer), a substrate, a covering layer, and lead wires. When strain is applied to the sensitive grid, the original geometry of the foil pattern is changed, leading to a subsequent change in its electrical resistance. The sensitive grid, which transforms surface strain into a change of resistance, is often made of thin alloy wires or foil traces, such as copper-nickel. These traces are patterned in tortuous ways and placed on polymeric substrates, e.g., polyimide, epoxy, or fibre glass-reinforced epoxy.

However, traditional foil-based strain gauges are gradually becoming outdated due to their rigidity and susceptibility to failure under high strain rates [143]. Ensuring strong adhesion between conventional strain gauges and the structural materials of interest presents its own set of challenges [144]. Direct-write methods, instead, offer an excellent alternative for replacing conventional sensors [139,145]. They enable direct printing of functional material components, including strain gauge patterns and insulating backing layers, onto the structural platform. Furthermore, AJP demonstrates its versatility in creating next-generation strain sensor designs on unconventional substrate platforms, including intricate 3D-printed parts [146], soft structural textiles [143], and temperature-sensitive materials like PVC conduits [147] and store-bought bandages [148].

4.1.2. Sensors for Human Health Monitoring

AJP has garnered increasing attention as a promising method for developing healthcare-related hybrid and heterogeneous material (HHM)-based devices. Researchers have effectively combined AJP with photonic sintering techniques to create wearable strain sensors tailored for healthcare applications, even on materials sensitive to temperature changes, such as off-the-shelf bandages (Figure 12a). For instance, Agarwala and colleagues showcased the direct patterning of strain sensors using silver nanoparticle inks on bandage materials [148]. Moreover, AJP has played a crucial role in the healthcare sector by producing smart personal protective equipment (PPE). In a recent study, a silver strain sensor fabricated using AJP was attached to a polyimide substrate and integrated into the flexible portion of a mask near the nose bridge. This enabled the wireless monitoring of healthcare professionals’ temperature, including detecting irregular respiratory breathing and fever, as well as tracking strain-related issues like face irritation and bruising [149].

4.2. Actuators

Shape memory polymer (SMP) actuators can mimic human muscular response to multiple stimuli—i.e., light, electricity, magnetism, temperature, pH, or moisture levels. Although they have limitations like fatigue and slow responses, 3D printing of SMP materials has shown the possibility of fabricating SMP actuators cheaply and with a fast responsive ability to various stimuli, including light, electricity, magnetism, heat, pH changes, and moisture [150]. SMPs also offer several advantages, including low density, large strain recovery, biocompatibility, and biodegradability [151]. For example, 3D-printed SMPs were used to produce a simple gripper. The research studied the effect of multiple parameters, such as nozzle temperature control and filament speed during the extrusion process, to optimise the quality of the actuator in terms of part density, dimensional accuracy, and surface roughness [152]. Additionally, different deformations, referred to as predetermined self-evolving behaviours, were achieved by 3D-printed polymer materials after water immersion (Figure 13c) [153].

Bi-stability is another intriguing property of SMPs, allowing structures with zero degrees of freedom to have two or more stable positions under specific conditions. These structures can pass from a stable configuration to another with a slight deformation. For example, a 3D-printed bi-stable actuator made of nylon was successfully fabricated (Figure 13d) [154].

Soft actuators can be manufactured very effectively using 3D printing technology [155,156,157]. These soft actuators can achieve substantial strain, up to nearly 200%, due to their low Young’s modulus and high elastic energy density. For instance, a two-layer dielectric elastomer (DE) actuator was 3D printed using an acrylic-based photopolymer, leading to the creation of a facial robotic system that mimicked human facial muscles [150]. Moreover, 3D printing has enabled the creation of generators using dielectric elastomers, which can be applied for energy harvesting, as in orthotic shoe inserts that capture energy during gait. In another application, a hexapod robot with multiple degrees of freedom was created using 3D-printed DE soft actuators, with all parts manufactured through 3D printing [158].

Soft piezoelectric materials, or piezoelectric polymers, are also used as actuators due to their flexibility, softness, transparency, and lightweight properties. They can convert electrical voltage into mechanical strain, making them suitable for various applications. While many piezoelectric actuators are produced via electrospinning, an exception is a polyvinylidene fluoride actuator created using inkjet printing. This actuator was used as a reciprocating membrane pump in a smart lab-on-a-chip system, showing its practical utility [159]. Furthermore, a cost-effective approach for producing active piezoelectric nanoparticles was developed using microscale dynamic optical projection stereolithography. This innovative method enhanced printing resolution and throughput compared to traditional SL techniques, particularly for 3D printing of piezoelectric polymers [160].

4.3. Thin-Film Transistors

Concerning conducting polymers, recent advancements have initiated a new era of entirely polymer-based electronic components and circuits [137]. What sets polymers apart from conventional silicon-based technology is the simplicity of their processing steps, which translates into cost efficiency. While polymers may currently lag behind silicon in terms of switching speed, their true value shines when cost takes precedence over speed. In the fabrication of a typical organic thin-film transistor (OTFT), four essential layers are required: precisely two conductive electrodes, a dielectric layer, and a semi-conductive layer. A critical aspect of OTFT fabrication is achieving precise alignment of these layers at the micro scale, which poses significant challenges for certain printing techniques due to difficulties in maintaining micro-scale tolerance in printed patterns [161]. For this reason, there have been many reports on printed high-performance transistors utilising various printing techniques. Numerous studies have demonstrated the use of metallic nanoparticle inks, including silver, copper, gold, and aluminium, as well as conductive PEDOT:PSS, for printing conductive electrodes in OFETs. In an effort to enhance electrical conductivity, CNT-based inks have been employed as alternatives to metallic nanoparticle inks. For instance, thin lines of composites based on CNTs and poly(9,9-dioctylfluorene-co-dipenthoxybithiophene) were precisely aerosol jet-printed to fabricate OTFT with the aid of solvents like chloroform and xylene [92]. Furthermore, the synergistic combination of inkjet printing and spin-coating has paved the way for the creation of all-polymer thin-film transistors [162,163]. One illustrative example involved the use of Baytron-P, an aqueous solution comprising poly(styrene sulfonic acid) (PSS) and conducting oligomeric PEDOT counter-ions, for printing electrodes [164]. This was achieved by dissolving the upper polymer layer and subsequently re-depositing it within the crater-like intrusion formed due to the pinning of the contact line, enabling the creation of transistor-based inverter circuits. Moreover, researchers have achieved successful inkjet printing of all-polymer transistors, incorporating materials such as poly(4-vinyl phenol) and polythiophene [164].

4.4. OLED

The 3D e-jet printing method was used to directly print high-resolution OLED pixels onto 3D mechanical frames created using DLP. This technology can print a variety of functional materials for optoelectronic devices at high resolution, achieving a minimum line width of 2.6 µm, while adapting to different 3D geometries through synchronised five-axis movement [165].

This hybrid 3D printing system can sequentially print all components of optoelectronic devices, from mechanical frames to OLED layers, resulting in a transparent, freeform 3D display. Importantly, it eliminates the need for additional thermal drying or annealing, allowing for the direct fabrication of high-resolution 3D devices under ambient conditions.

Unlike previous methods for flexible or stretchable devices, which relied on substrates and were limited to surface attachment on non-planar objects, this hybrid 3D printing system can embed optoelectronic devices within arbitrarily shaped 3D structures at specified locations. This is achieved because all of the device components and mechanical frames are printed together as freeform optoelectronics.

This innovation was demonstrated through various 3D architectural examples, including structures with high-resolution OLED pixels integrated inside, as well as a transparent, eyeglass-type display designed for a wireless augmented reality system. These applications showcase the potential of the hybrid 3D printing system for creating freeform 3D optoelectronic devices.

The OLED pixels, each measuring 20 µm × 80 µm, are patterned through hybrid 3D printing (Figure 14), maintaining a gap of approximately 50 µm between the nozzle tip and the surface of the printed structure to ensure accurate deposition. This is possible using a five-axis system that enables z-axis translation and tilting of the xy-plane. It keeps consistent spacing even on non-planar surfaces. The printed OLEDs are integrated into a transparent, eyeglass-style display developed for a wireless augmented reality AR system.

4.5. Organic Solar Cells

Organic solar cells (OSCs) are a promising class of photovoltaic devices that utilise conjugated polymers and small molecules as photoactive materials and electrodes. Their intrinsic advantages include mechanical flexibility, lightweight architecture, and the possibility of low-temperature, solution-based processing, enabling scalable fabrication on plastic substrates through printing and roll-to-roll techniques [166]. Typical OSC architectures employ polymer donors such as P3HT blended with fullerene-based materials (PCBM), forming bulk heterojunction (BHJ) active layers, while transparent conducting electrodes can be produced with printable materials like PEDOT:PSS or printed metallic grids [167].

AM methods have recently been explored to leverage the solution-processable nature of OSC materials for cost-effective and patternable device fabrication. Inkjet printing and EHD printing are among the techniques capable of depositing functional layers with controlled thickness and morphology. For instance, Esa et al. demonstrated the use of EHD printing for a controlled deposition of P3HT:PCBM active layers, enabling microstructured photoactive films via high-resolution spraying [168]. Similarly, fully inkjet-printed OSC stacks have been reported, where all layers, including the ITO-free transparent electrode, hole transport layer (PEDOT:PSS), BHJ layer, and top electrode, were patterned in air at room temperature [169]. Moreover, AM technologies can enable the fabrication of unconventional device geometries and 3D-structured electrodes. Despite progress, printed OSCs still lag in efficiency and durability. Bench-top OSCs have reached ~19–20% of power conversion efficiency (PCE), but fully printed devices usually show much lower PCE (a few percent) [136,170].

4.6. Wafer Bonding

Wafer bonding is an important process in semiconductor manufacturing in which two or more semiconductor wafers are bonded together to create complex integrated circuits and MEMS. This critical technique plays a fundamental role in achieving enhanced device performance, the integration of heterogeneous materials, and the development of new electronic and optical devices. In the last few years, the combination of 3D printing and wafer bonding technologies has been an exciting direction for transforming semiconductor fabrication. The use of 3D printing in wafer bonding enables the development of intricate three-dimensional structures and customised device architectures. It also reduces manufacturing flows. One of the AM technologies investigated for this purpose is AJP, as presented in a recent study, which used BCB-based ink and assessed the material bonding quality after a thorough characterisation of the AJP process. A bonding test was performed using two silicon (Si) wafers, with a thermally grown Si oxide layer, bonded by a BCB layer deposited using AJP. The process, illustrated in Figure 15a, involved an initial heat of 150 °C for 30 min, above the 23 °C glass transition temperature of the BCB ink. This allowed the BCB to flow and evenly coat the surfaces of the wafers under a compressive force of 2.4 kN. After curing and bonding, gel content analysis revealed over 99% cross-linked material, indicating successful cross-linking. It was, nevertheless, noted that BCB is prone to oxidation upon curing. To assess potential oxidation, Fourier transform infrared spectroscopy (FTIR) analysis was performed, comparing BCB cured under nitrogen and air atmospheres (Figure 15b). In the FTIR spectra, BCB cured in the air showed a higher peak at 3000–3200 cm⁻¹, attributed to OH group stretching, along with oxidised benzylic CH₂ groups detected at 1700–1850 cm⁻¹. This indicated a higher concentration of oxidised species in BCB cured under an air atmosphere.

Die shear tests were also conducted to measure the shear strength of wafers bonded with a BCB layer. These tests were performed on a pad with a constant cross-section to calculate shear strength accurately and to compare coatings from AJP and spin-coating. Testing complex patterns with a high aspect ratio would not yield precise results and would not be feasible with spin-coating. Figure 15e presents the die shear test setup, where a tip applies a displacement to the top silicon wafer while the bottom wafer is secured by a physical barrier. The test is concluded when the bonding layer breaks, and the force at which this happens is recorded [131]. Figure 15f shows the die shear test results for wafers bonded with a BCB layer deposited using AJP and spin-coating. The shear strengths were 38.74 ± 1.88 MPa for AJP and 36.25 ± 2.33 MPa for spin-coating. These comparable values indicated that AJP did not negatively impact bonding quality compared to spin-coating, and these results were consistent with the literature results for spin-coated BCB [171,172,173].

Table 2. Summary of the most used applications for 3D printing of polymers in electronics.

Application	Material	AM Technology	Highlights	Reference
Strain sensor	CNT nanocomposite	AJP	High mechanical durability	[54,104,139]
	Graphene nanocomposites	IJP	Relatively high electrical conductivity
	PEDOT:PSS	DIW
Sensor for human health monitoring	Silver nanocomposites	SL	Relatively high conductivity	[53,142]
	PEDOT:PSS	AJP	Poor mechanical properties
Actuators	CNT	SL	High electrical conductivity	[150,157]
	Graphene nanocomposites	DIW
Thin-film transistor	PEDOT:PSS	SL	High performance	[92,174]
		LIFT	High resolution
OLED	PEDOT:PSS	SL	Low viscosity	[165,175,176]
		IJP	High resolution
			High performance
OSCs	PEDOT:PSS	IJP	Freedom of design	[168,169,170]
	P3HT:PCBM	EHD	Low efficiency
Wafer bonding	BCB	IJP	Good resolution	[130,131]
		AJP	Ink viscosity is critical
			Overspray for AJP
Packaging	SU-8	IJP	Good printing quality	[133,177]
		AJP	Presence of overspray

displays a summary of all of the abovementioned applications with the materials and the technologies employed.

5. Current Limitations and Overcoming Strategies

Three-dimensional-printed polymers have been used in a vast variety of electronic applications. ICPs are mostly utilised to make thin-film transistors and OLEDs. The main advantage of using these kinds of materials over more traditional ones is their flexibility, which enables, for instance, their use in electronic skins and wearable devices. The main challenge is the durability and electrical conductivity under bending. Indeed, many works show that electrical properties are reduced during bending or after bending cycles. Therefore, there should be a focus on how to improve the durability of the electrical performance of intrinsically conductive polymers. A solution is to add fillers to improve the mechanical properties of the material. For example, Soni et al. developed a PEDOT:PSS/GO nanocomposite, which maintained its properties after 1000 cycles with a bending strain of 30% [178]. Theoretically, many fillers can be employed to achieve this goal, such as graphene, silver nanoparticles, and MXenes. The choice of the most appropriate filler depends on the desired properties and the mechanical stress to which the devices are subjected. The possibility of enhancing PEDOT:PSS properties through the addition of fillers is widely investigated in the literature. However, the same level of knowledge is not present for other ICPs, such as polyacetylene, polyaniline, and polypyrrole. Moreover, the number of studies regarding AM of ICPs other than PEDOT:PSS is limited and should be improved, taking advantage of all of the studies regarding PEDOT:PSS.

In addition to mechanical stresses, the long-term reliability of additively manufactured conductive polymers requires careful evaluation under realistic operational conditions, such as thermal cycling, prolonged humidity exposure, and electrical ageing. These factors critically influence the performance of polymer-based devices, particularly in wearable electronics, embedded sensors, and soft robotics, where environmental fluctuations are unavoidable. Thermal cycling, for example, induces differential expansion and contraction in multilayer systems, potentially leading to delamination as envisaged by Kaltenbrunner et al. [179]. Similarly, humidity exposure is especially critical for materials like PEDOT:PSS, resulting in a progressive decrease in conductivity above 70% relative humidity [180]. Despite extensive research on bending-induced conductivity loss, few studies have systematically quantified these other degradation pathways in printed systems. A recent investigation, for instance, revealed that PEDOT:PSS films can still show a decrease in conductivity under high-temperature and high-humidity conditions [181]. To strengthen the design of durable printable electronics, it is essential to integrate accelerated ageing protocols in formulation workflows. Addressing these gaps is essential not only to ensure device functionality in real-world applications but also to guide formulation strategies that enhance material robustness beyond laboratory conditions.

Regarding extrinsically conductive nanocomposites, the literature provides plenty of research papers [182]. Knowledge regarding the formulation of electrically conductive polymer-based nanocomposites is well established. What is missing is a full understanding of the most appropriate processing conditions for the different 3D printing technologies. For example, considering some works on SL printing of CNT- and graphene-based polymer nanocomposites, the loading of conductive fillers is relatively limited, being 5 wt% at most. This is due to two main factors. First, the addition of fillers drastically increases the viscosity of the nanocomposite by several orders of magnitude, making processing very challenging. This is true not only for SL technology but for many others, e.g., IJP, AJP, and DIW. Second, graphene and CNTs interact with the UV-curing process by absorbing a large amount of UV radiation that should be used for cross-linking. Consequently, too high loading of fillers hinders cross-linking, leading to poorly cured or uncured nanocomposites that, therefore, have detrimental mechanical properties. There are different approaches to improve the situation. Adding some rheological modifiers could help to reduce the viscosity while increasing the filler concentration in the nanocomposite. Among all of the rheological modifiers, reactive diluents are a very interesting category for their two-fold effect. They help to reduce the viscosity while also helping with the cross-linking reaction at the same time. However, reactive diluents have some drawbacks as well. Cured nanocomposites exhibit a lower glass transition temperature as a result of the chemical nature of the reactive diluents. These diluents promote mobility in polymer chains, thereby increasing the free volume and lowering the T_g. It would be challenging when the T_g of the nanocomposites is lowered to a level close to the service temperature because it would yield a material without proper thermal and mechanical characteristics. Another issue is that the overall mechanical properties of nanocomposites decrease. This occurs in the opposite direction of the effect of the addition of filler, which provides an increase in both electrical and mechanical properties. Therefore, the concentration of the type of reactive diluent should be carefully selected.

Recent advances in artificial intelligence and machine learning are now facilitating the optimisation of data-driven formulation. These tools have been used not only to identify correlations between formulation parameters and rheological properties but also to predict the electrical performance and mechanical integrity of printed components under operational stress [183]. Generative models and neural networks have also shown potential in optimising and controlling AM process parameters, thereby reducing the number of development cycles [184]. Integrating machine learning into the design of conductive inks and composites could likewise speed up the discovery of printable, durable inks tailored for electronic applications [185]. Machine learning methods and AI-driven formulation design have already been applied for this purpose to optimise the viscosity and mechanical properties of additively manufactured ECPs, as demonstrated in the creation of 3D lattice structures [186] and smart responsive materials [187]. Another challenge is the interaction between polymers and fillers, which is critical but scarcely understood to date. Poor polymer–filler interaction leads to inadequate electrical and mechanical performance. Surface modification of fillers can improve this interaction, but further research is needed to understand how surface modifications influence the overall properties of nanocomposites.

Furthermore, long-term electrical ageing due to current stress has been shown to induce material degradation in conductive 3D-printed polymer composites as a result of local Joule heating, thus impairing mechanical integrity [188]. Encapsulation strategies have been proposed to mitigate humidity-induced degradation in transient electronic devices [189]. However, their integration into AM-compatible printable formulations remains underexplored.

Meanwhile, MXenes, which are a rapidly developing class of 2D transition metal carbides/nitrides, are proving to be strong contenders as conductive fillers. Their high conductivity, tunable surface chemistry, and good dispersion in aqueous PEDOT:PSS systems make them ideal for printable composites [119]. Indeed, a PEDOT:PSS/MXene composite ink has demonstrated high conductivity and robust mechanical properties in 3D-printed hydrogels [115]. Compared to CNTs and graphene, MXenes are often easier to process in water-based formulations and can support stronger polymer–filler interfaces, although stability under long-term exposure remains an open challenge. Comparative studies of MXenes, CNTs, and graphene in AM are still few, making this a valuable direction for future research.

Dielectric polymers are a category of polymers that have been the least explored in the area of additive manufacturing within electronics, leading to difficulties when developing or utilising a new ink. There is a lack of information available in the literature, leading to trial-and-error tests, which are both time consuming and inefficient. The increase in research works focused on 3D printing of dielectric polymers could help to develop new materials in a faster and easier fashion. Moreover, dielectric polymers are used for many applications in electronics, but very few papers propose an approach to use AM with them. We believe that this is due to difficulties in either modifying or improving the electrical properties of dielectric polymers. Hence, they attract less interest from the scientific community, which is more focused on investigating and improving conductive polymers and composites. Furthermore, the fabrication methods for dielectric polymers are well established in the electronics industry, which uses spin-coating coupled with lithographic process or screen printing. Currently, more novel fabrication technologies, such as IJP or AJP, cannot replace conventional fabrication technology due to resolution and throughput limitations. However, IJP and AJP have sufficient resolution for some applications or specific patterns, and the material throughput could be improved by working on the fabrication technology itself. The ability to additively manufacture dielectric polymers would be extremely important for the electronics industry to reduce material waste, operator mistakes, and the overall fabrication time of the complete device. Indeed, the development of a system capable of fully 3D printing a device is important. Hence, AM of dielectric polymers should be investigated to achieve this goal.

Although the potential and feasibility of AM with polymeric materials in the electronics field are established, numerous challenges and issues still need to be addressed. For instance, AM methods such as DIW typically produce features ranging from 10 µm to 100 µm, depending on the ink rheology, nozzle diameter, extrusion parameters, and system configuration, which constrain miniaturisation in printed electronics [190]. For applications like printed thin-film transistors, where channel length and interconnect width directly influence carrier mobility, this limited resolution often results in reduced device performance compared to traditional photolithographic fabrication [191].

Surface roughness induced by layer-by-layer deposition can also degrade device performance, as it was demonstrated that AJP-printed Ag transmission lines can perform competitively with PCB copper traces, if roughness does not exceed a few micrometres [192]. Hence, optimisation of printing parameters becomes crucial to minimise surface roughness and improve the performance of electronic devices.

Moreover, cracking and void formation during drying and sintering of metallic inks are common defects that emerge from shrinkage-induced stress and poor wetting on substrates, leading to discontinuities that break electrical pathways and increase the series resistance of printed traces. Sridhar et al. reported cracks in silver inkjet-printed tracks on plasma-treated glass-reinforced epoxy substrates, where microvoids resulting from substrate roughness reduced conductivity and adhesion [193]. Achieving low defect, reliable electrical contacts in printed electronics requires careful management of ink formulation, substrate preparation, and sintering conditions. Addressing these issues, robust material–process integration is essential to translate additively manufactured devices from lab prototypes to dependable real-world systems.

A higher material throughput was demonstrated only on a lab scale. Hardware with multiple printheads needs to be developed to compete with the industrial machines available nowadays for industrial applications. Furthermore, certain technologies impose stringent limitations on the types of materials that can be utilised, consequently restricting the available options. During vat photopolymerisation (SL/DLP), conductive fillers like CNTs, graphene, or MXenes can significantly interfere with UV-curing by either absorbing or scattering incident light, leading to incomplete cross-linking of the polymer matrix. Eng et al. demonstrated that even at low loading (0.25 wt%), CNTs reduced cure depth, necessitating post-UV or thermal treatments to achieve mechanical integrity and reliable conductivity [194]. Similarly, Shah et al. reviewed vat photopolymerisation of nanocomposite resins and highlighted the need for a trade-off between filler content and photo-curing due to UV attenuation [195]. As a result, poor curing leads to defects, such as weak interlayer adhesion, dimensional inaccuracy, and zones of mechanical brittleness, which undermine electrical continuity and device robustness. For these reasons, post-processing treatments, such as thermal post-curing, become crucial to improve interlayer bonding.

Moreover, it is essential to expand the versatility of many AM platforms, which still restrict usable materials. In light of this, technologies like LIFT and AJP offer the opportunity to handle diverse inks with different rheological behaviours. However, novel concepts applied to existing technologies, as well as emerging additive manufacturing techniques, so far primarily used for metals and their oxides [196,197], can be explored for polymers in future research, thereby broadening the possibilities. In addition, future developments should not only address current processing limitations but also focus on creating a fully digital, feedback-driven AM workflow. This includes in situ monitoring of print quality, real-time property mapping, and adaptive printing strategies to enable quality assurance and complex device fabrication.

6. Conclusions

AM of polymers is extremely promising and can find applications in a variety of electronic devices. ICPs have been primarily exploited in thin-film transistors and OLEDs for their flexibility, which enables their integration into wearable and stretchable devices and electronic skins.

However, a weakness of ICPs is the loss of mechanical and electrical performance under cyclic bending. Dispersing conductive fillers, such as CNTs, in PEDOT:PSS matrices can transform this drawback into a benefit, if appropriate fillers are selected for achieving the expected mechanical and electrical properties. Moreover, many studies have been dedicated to investigating extrinsically conductive nanocomposites, but further work is needed to understand the optimal parameters for processing these materials through different 3D printing technologies. For instance, a maximum threshold can be reached for the filler content in liquid polymer resins using SL technology due to the viscosity increase from increasing the filler percentage. Adding reactive diluents to the resin can help to reduce viscosity, supporting cross-linking. However, a trade-off between filler loading and reactive diluent concentration should be found to mitigate the possible reductions in glass transition temperature and mechanical performance induced by the added reactive diluent. Furthermore, a crucial and still underexplored aspect is the improvement of interactions between polymers and fillers to enhance electrical and mechanical performance. Regarding dielectric polymers in the field of additive manufacturing for electronics, they have gained less interest than their conductive counterparts over the last decade. Further investigation into the formulation of these polymers for AM will certainly boost progress in electronics. In addition, the scale-up of additively manufactured polymers in industrial environments requires developing hardware, such as multiple printheads, to reduce the production time. Therefore, numerous challenges remain, and additional innovations are still necessary, although the feasibility of AM of polymer materials in electronics is evident.