PEDOT:PSS: From conductive polymers to sensors

Poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) is one of the most promising conductive polymers.^1–4 During the last two decades, research on PEDOT:PSS has increased dramatically.^5–7 PEDOT:PSS polymers with various compositions, doping, and chemical modifications have been developed to meet different application requirements. PEDOT:PSS has attracted much attention commercially, and its market continues to grow, with an estimated annual value of 50 billion dollars.⁸ The key advantages of PEDOT:PSS include adjustable conductivity,^9–13 good transparency to visible light,^14–17 excellent thermal stability,¹⁸ and a high level of biocompatibility,¹⁹ which makes it successfully applied in electrostatic coatings, flexible electronics,^20–22 bioelectronics,^23,24 energy storage,¹⁴ and tissue engineering.^25–27 Since commercial PEDOT:PSS is usually in the form of an aqueous dispersion,²⁸ it is compatible with many solution-based manufacturing processes, such as coating techniques (e.g., dip-coating, drop-coating, spin-coating, and spray-coating), printing techniques (e.g., inkjet printing and screen printing), and lithography (e.g., soft lithography and nanoimprint lithography).^29–32 

In recent years, there has been a dramatic increase in the number of studies of PEDOT:PSS-based devices for sensing applications, including gas,^33,34 biological,^35,36 electrochemical,^37–39 and physical^40–42 sensing. PEDOT:PSS biological and chemical sensing have been reviewed in Refs. 1 and 43. Liao et al.⁴⁴ reviewed the development of biological and chemical sensors for OECTs using PEDOT:PSS as the channel. Gao et al.⁴⁵ reviewed the latest developments in the use of PEDOT:PSS and composites based upon it in chemical sensors. However, there has been no review summarizing the use of PEDOT:PSS in physical sensing applications, such as in humidity,^46–48 temperature,^29,49,50 pressure,^51–53 and strain sensors.^54–56 Therefore, this review focuses on physical sensors using PEDOT:PSS and on methods to improve the sensing performance, especially with regard to the enhancement of the sensing performance of PEDOT:PSS at the nanometer scale.^29,57,58

In this review, we discuss the latest developments concerning PEDOT:PSS, from materials to sensors, covering basic material properties, sensor fabrication processes, and sensing principles, focusing especially on the influence of micro/nanostructures on sensor performance. First, the polymer structure of PEDOT:PSS is discussed, including aqueous dispersions, solid films, and the emerging PEDOT:PSS hydrogels. Second, we introduce the manufacturing process of PEDOT:PSS-based devices, such as coating, printing, conventional lithography, and alternative lithography. The advantages and disadvantages of these fabrication techniques in terms of resolution, cost, and efficiency are compared. Finally, we summarize the latest developments in PEDOT:PSS-based physical sensors, including those for humidity, temperature, pressure, and strain.

The chemical structures of PEDOT and PSS are shown in Fig. 1(a). PEDOT:PSS aqueous dispersions with different ratios of PEDOT and PSS can be obtained through oxidative polymerization of 3,4-ethylenedioxythiophene (EDOT) monomers in the presence of a PSS matrix. Completely oxidized PEDOT chains under ideal conditions possess one charge carrier per three monomers.⁵⁹ Negatively charged PSS serves as a counterion to balance the positive charges of PEDOT, and, together with the PEDOT matrix, forms a homogenous and stable aqueous dispersion.

PEDOT:PSS aqueous dispersion presents a typical tertiary structure.⁶⁰ As shown in Fig. 1(b), EDOT and styrene sulfonate units represent the primary structure of the polymer. Relatively short positively charged PEDOT chains attach electrostatically to the longer negatively charged PSS chains, forming the secondary structure. The polyionic complex is eventually dispersed in water in the form of colloidal particles, with a hydrophilic PSS-rich shell and a hydrophobic PEDOT-rich core (tertiary structure). As a reference, the particle size of PEDOT:PSS in aqueous solution ranges from 20 to 200 nm for the commercially available Clevios P.⁶¹ The hydrodynamic radius of colloidal particles in PEDOT:PSS aqueous solution for Orgacon ICP 1050 is about 250 nm.⁶² The colloidal gel particles repel each other owing to the electronegative PSS shell, forming a uniform and stable water dispersion. The interactions between the colloidal gel particles are dominated by electrostatic forces.⁶² 

Pristine PEDOT:PSS solid films in general have a multilayered structure. Randomly distributed colloidal gel particles precipitate out during the process of water evaporation, forming a pancake-like structure [Fig. 2(a)].^61,63 Each “pancake” possesses a hydrophobic, conductive PEDOT-rich core and a hydrophilic, insulating PSS-rich shell. As a reference, the average diameter of PEDOT:PSS “pancakes” is 30–50 nm, with a 5–10 nm-thick PSS-rich shell (a 25 nm-thick solid film made from unfiltered Clevios P). The cohesive strength between “pancakes” is based on hydrogen bonding between sulfonic acid groups⁶¹ and π-orbital stacking between PEDOT chains. This pancake model is in accordance with the granular structure observed under bright-field transmission electron microscopy (TEM), as shown in Fig. 2(b). Secondary doping^64,65 and post-treatment^66,67 alter the microstructures of PEDOT:PSS solid films, enhancing their conductivity. For example, counterion exchange between an ionic liquid and PEDOT:PSS leads to rearrangement of PEDOT and PSS chains. Interconnected PEDOT crystalline nanofibrils are observed in ionic-liquid-doped PEDOT:PSS solid films under TEM, as shown in Fig. 2(c). In addition, it has been shown that PEDOT:PSS-produced film has excellent thermal stability, chemical stability, and durability.^68–70 For instance, it retains a stable high electrical conductivity after being tested for 60 days in an environment of high humidity at 65 °C.^71–73 

Conductive hydrogels can be used to compensate for the mechanical mismatch between biomaterials and inorganic materials. Their superior mechanical properties, together with their stability in water, make conductive hydrogels one of the best candidates for applications in neuronal interfaces and tissue engineering.^25–27 Compared with other conductive hydrogels, PEDOT:PSS-based hydrogels are endowed with high conductivity even at a low solid content (46 S/m at a solid content of 0.78%).⁷⁵ PEDOT:PSS-based hydrogels with high water content normally exhibit a porous 3D structure, as shown in Fig. 3. This porosity of PEDOT:PSS hydrogels makes them compatible with a variety of post-processing methods to tune their mechanical performance or to introduce new capabilities. For example, polymer monomers can be injected into the porous structure to form a double-network hydrogel. In addition, PEDOT:PSS is biocompatible and nontoxic, allowing it to act as a scaffold for cell proliferation and differentiation.^25,26 Besides, PEDOT:PSS has superb environmental stability in acidic, alkaline, high-temperature, and other harsh environments. PEDOT:PSS-based hydrogels are expected to have a wide range of applications in the fields of tissue engineering, drug delivery, sealing, wound dressing, interface engineering, environmental pollution control, sensors, actuators, bioelectronics, energy storage, and catalysis.^76–78 

Currently, available PEDOT:PSS-based hydrogels described in the literature include PEDOT:PSS hydrogels themselves,^75,79,80 PEDOT:PSS polymer blend hydrogels made from combinations of PEDOT:PSS and one or more other polymer components,^81,82 and PEDOT:PSS composite hydrogels^83–85 consisting of PEDOT:PSS and inorganic materials. PEDOT:PSS polymer blend hydrogels have been applied successfully in tissue engineering and other domains.⁸⁶ However, PEDOT:PSS hydrogels are still in their early stage, with new fabrication techniques being developed and new applications emerging.^87,88

As PEODT: PSS can be dispersed in aqueous solution, it is compatible with many solution processes. Coating is a simple process with low cost and high material utilization rate. It has been widely used in research laboratories and industrial production.^89–91 At present four coating processes are available, namely, dip coating,^29,92 drop coating,^93,94 spray coating,^95,96 and spin coating,^97,98 as shown in Fig. 4. Dip coating can be used for large-area film formation on a variety of substrates, but is limited by the viscosity of the material solution.^99,100 Ding et al.¹⁰¹ used dip coating to make conductive sponges by immersing melamine sponge in an aqueous dispersion of PEDOT:PSS so that the latter adhered to the micrometer-scale framework of the sponge. As a much simpler process, drop coating needs only a one-step operation. For instance, a transparent PEDOT:PSS electrode was fabricated by dropping PEDOT:PSS solution directly on the surface of a polyethylene terephthalate substrate without special equipment.¹⁰² However, owing to the influence of solvent evaporation, the thickness and size of the films formed by drop coating are difficult to control.^103,104 Spray coating has also been used.^105,106 Kumar et al.¹⁰⁷ sprayed a suspension of PEDOT:PSS/reduced graphene oxide (rGO) onto a carbon cloth using a single-step spray deposition technique and thereby fabricated large-area flexible electrodes. However, spraying has higher instrumentation requirements and relatively high costs.^108–110 Compared with the above coating methods, spin coating can produce uniform PEDOT:PSS films with good repeatability.^111,112 A simple strategy of cryo-controlled quasi-congealing spin coating was developed by Chen et al.¹¹³ to prepare flat and smooth PEDOT:PSS films with high conductivity and moisture resistance. However, this approach requires a flat substrate and leads to some wastage of materials.^114,115

Printing is a straightforward way to deposit materials on a wide variety of substrates. It has the advantages of high efficiency, low cost, and less material waste.^27,30,116 Current printing processes include screen printing,^117,118 inkjet printing,^119,120 and 3D printing,^121,122 as shown in Fig. 5. Screen printing is a mature process and suitable for various substrates.^123,124 A mixed solution of PEDOT:PSS and poly(dimethylsiloxane-b-ethylene oxide) was screen-printed on knitted cotton to prepare conductive textiles.¹²⁵ However, the thickness of the formed film is large, and this method is not suitable for solutions with low viscosity and high volatility.^126–128 Lo et al.¹²⁹ used inkjet printing to prepare a PEDOT:PSS gas-sensitive film with FeCl₃ additive on a polyimide film. It has been shown that inkjet printing is a simple and low-cost method to prepare thin films with controllable thickness.^130,131 However, the charged ink droplets will affect the resolution and performance of the device.^132–134 Recently, 3D printing has emerged as a technique to rapidly print arbitrary PEDOT structures, benefitting from its freedom with regard to design, its ability to print complex structures, and its compatibility with many materials.^135–137 One typical application is that by Lei et al.,¹³⁸ who used 3D printing technology with hybrid electrohydrodynamics to construct multiscale conductive scaffolds for cardiac tissue engineering. The scaffolds were composed of micrometer-scale fibers of polycaprolactone and sub-micrometer-scale conductive fibers of PEDOT:PSS/polyethylene oxide. However, with 3D printing, it is difficult to print high-precision complex curved surfaces,¹³⁹ and the anisotropy of the product will affect its mechanical properties.^140,141

Conventional top-down lithography is a common technique for making micro- or nanoscale structures on flat substrates.^142,143 For example, photolithography is a key technology in the manufacture of integrated circuits and microchips, and it has made a revolutionary contribution to the semiconductor industry.^31,144 Figure 6 shows three commonly used conventional lithography techniques, including ordinary lithography,^145,146 ion beam lithography,^147,148 and electron beam lithography.^149,150 Zhu et al.⁴⁶ proposed a one-step photopatterning method based on photolithography, using ultraviolet rays to transfer microstructures from a polydimethylsiloxane (PDMS) photomask to a PEDOT:PSS/polyethylene glycol hybrid film, thereby producing a microstructured thin film with a resolution of 2 μm. In ion beam lithography, the collision of high-momentum ions [e.g., gallium ions or hydrogen ions (protons)] with the resist layer leads to the removal of the resist particles.^151,152 A PEDOT:PSS-coated silica single-mode fiber was etched into a patterned phase photonic sieve by focused ion beam lithography to improve the optical coupling between the fiber and a silicon photonic waveguide.¹⁵³ Their high momentum means that the ions propagate in a straight line, resulting in high-resolution etching.^154,155 However, it also damages the substrate and reduces the device yield.^156,157 Electron beam lithography has extremely high resolution (below 10 nm) owing to the relatively small wavelength of the electron beam.¹⁵⁸ Santoro et al.¹⁵⁹ used electron beam lithography to fabricate quartz nanostructures, and then spin-coated a composite film containing PEDOT:PSS to make a cell–material interface with nanometer resolution. However, electron beam lithography is time-consuming^160,161 and the process is complicated, which increases its cost, and thus it is only suitable for a small-scale production involving high-precision processing.^162,163

Alternative lithographic techniques have been developed that have advantages of low cost, convenience, and high throughput, and that can replicate micro-to nanostructures on a large scale.¹⁶⁶ After decades of development, alternative lithography has become a viable commercial process and is expected to overcome the limitations of conventional lithography.^167,168

Soft lithography is a typical alternative lithographic technique that uses elastomer impressions for printing, molding, transferring, and embossing, relying on conformal contact between the elastomer and the target substrate [Fig. 7(a)].¹⁶⁹ Soft lithography has been widely applied to the fabrication of microfluidics, microelectromechanical systems, and flexible electronics/photonics. However, owing to the mechanical instability of the elastomers, such techniques are usually limited to patterning micrometer-scale structures.^170,171

Nanoimprint lithography (NIL) is another commonly used alternative lithographic technique and is able to achieve high-resolution nanoscale patterning.^172,173 Typically, it can be classified into thermal and ultraviolet (UV) imprinting, as shown in Fig. 7(b).^174,175 Hot embossing was the earliest NIL technique, and requires just a simple one-step heating and pressing process to transfer a pattern from a hard mold to a thermoplastic.^32,176 During the process, the structure on the mold directly contacts the substrate in a molten state, and the “opposite” structure is replicated on the substrate.^177,178

Although NIL can pattern very small structures (down to 5 nm), the high-temperature or UV processes cannot fulfill the requirements of PEDOT patterning. To solve this problem, we developed a hybrid nanoscale printing approach by combining the advantages of soft lithography and nanoimprinting. Thermal NIL was used to pattern nanochannels on PDMS and then transfer the nanostructures to the substrate through capillary action.^179–181 As shown in Fig. 7(c), the soft flat PDMS was coated with Mr-I T85 imprint resist, and was then heated and pressurized to replicate the nanostructures from a rigid silicon mold. Next, the PDMS was combined with a flat substrate to form nanochannels. The material solution was sucked into the nanochannel under the action of capillary force. Finally, the imprint resist was removed after the solvent had evaporated.^182,183 This approach has been successfully applied to patterning PEDOT:PSS nanowires on both hard and flexible substrates.^29,57 This hybrid approach can be performed at room temperature without a cleanroom requirement, greatly improves process efficiency and safety, and can produce nanopatterns on a larger scale with high speed and high yield.

PEDOT:PSS in the form of an aqueous dispersion is compatible with a variety of manufacturing processes, and this has promoted its application in electrical devices. Figure 8 shows some electrical devices based on PEDOT:PSS, namely, a polymer light-emitting device (PLED), a supercapacitor, a perovskite solar cell (PSC), a thermoelectric device, and an electrochromic window. As a common hole injection layer in PLEDs, PEDOT:PSS with a high work function is spin-coated between the emitter layer and the anode to reduce the energy barrier and improve the surface roughness of the anode [e.g., indium tin oxide (ITO)], thereby producing a more efficient PLED.^186,187 PEDOT:PSS is also used as an electrode in supercapacitors, since it can increase the specific capacitance compared with carbon materials.¹⁸⁸ Khasim et al.¹⁸⁹ doped rGO into PEDOT:PSS to increase its conductivity fourfold, thereby showing that secondary doping can help improve the performance of PEDOT:PSS-based supercapacitors. PEDOT:PSS is also used as the electrode in PSCs to solve the problems of high cost and poor stability associated with traditional electrodes (Au and Ag) and thus promote the commercial development of PSCs.¹⁹⁰ In addition, PEDOT:PSS has proved to be a promising organic thermoelectric material because of its enhanced high electrical conductivity and low thermal conductivity. Through doping and dedoping treatment, the thermoelectric quality factor and response speed of PEDOT:PSS-based thermoelectric devices can be enhanced.¹⁹¹ In electrochromic windows, PEDOT:PSS can be used as a solid proton source for the electrochromic material WO₃. When the protons released by PEDOT:PSS under the action of an electric field are inserted into the WO₃ lattice, transition between the small polarization states associated with W ions of different valences results in optical absorption that causes coloration of the WO₃.¹⁹² 

As one of its most important applications, PEDOT:PSS has had a profound impact as a sensor material, owing to its excellent and adjustable electrical conductivity, hygroscopicity, and flexibility.¹ There are many reviews of the use of PEDOT:PSS-based devices to detect biological and chemical substances, but few have summarized research on physical sensing.^44,45 Therefore, this section focuses on PEDOT:PSS-based physical sensors, including those for humidity, temperature, pressure and strain.

PEDOT:PSS has been shown to be a hygroscopic material,¹⁹³ making it promising for sensing humidity.^194,195 The sensing principle is dependent on water adsorption and desorption by the PSS.^196,197 In high-humidity condition, the insulating and hydrophilic PSS shell absorbs water and swells, resulting in an increase in the distance between adjacent conductive and hydrophobic PEDOT cores, which leads to an increase in the resistivity of PEDOT:PSS for the resistive type of sensing device, as shown in Fig. 9(a). By contrast, low humidity causes water to desorb out of the PSS, reducing the volume of the PSS and the distance between adjacent PEDOTs and thereby reducing the resistivity.⁵⁷ So far, PEDOT:PSS has been used to develop various types of humidity sensors, including resistance,^{47,57,198–200} capacitance,^46,201 and resonator^48,195,196 types.

A resistive-type humidity sensor converts changes in humidity into changes in the resistivity of PEDOT:PSS.^41,57 Zhou et al.⁵⁷ developed a PEDOT:PSS humidity sensor with high sensitivity (5.46%) and ultrafast response (0.63 s) on a flexible substrate using PEDOT:PSS nanowires as shown in Fig. 9(b). Owing to their ultrahigh surface-to-volume ratio, the PEDOT:PSS nanowires were readily exposed to the environment and further accelerated the water absorption/release rate. Such a device has been successfully applied for human breath testing. In addition, PEDOT:PSS has been used in combination with other moisture-sensitive materials to overcome the shortcomings of a relatively small dynamic response range from a single material. Siddiqui et al.²⁰² proposed a humidity sensor with dual sensing elements composed of PEDOT:PSS and MoS₂, as shown in Fig. 9(c). MoS₂ and PEDOT:PSS respond to humidity changes in the ranges 0%RH–40%RH and 40%RH–80%RH, respectively. By combining these two sensing materials in series in the detection circuit, a very high sensitivity (50 kΩ/%RH) and a wide response range (0%RH–80%RH) were obtained.²⁰² This strategy has been applied to make a humidity sensor that can detect the full humidity range (0%RH–100%RH) through a combination of three humidity-sensitive materials: graphene oxide (GO), PEDOT:PSS, and C₁₅H₁₅N₃O₂ (methyl red).²⁰⁰ 

Capacitive humidity sensors have also been reported. Figure 9(d) shows a typical humidity sensor combining PEDOT:PSS film and interdigital capacitors.²⁰¹ The sensor used PEDOT:PSS film as the dielectric layer, which demonstrated good sensitivity (3.4 pF) to humidity at 52.0%RH–93.4%RH. The sensing principle is that humidity causes a change in the effective dielectric constant of the PEDOT:PSS-based dielectric layer.²⁰¹ Zhu et al.⁴⁶ proposed a one-step photopatterning method to manufacture a PEDOT:PSS/polyethylene glycol capacitive humidity sensor, with PEDOT:PSS being used for the humidity electrodes to enhance sensitivity.

A piezoelectric resonator coated with a PEDOT:PSS thin film will respond to humidity changes owing to adsorption and desorption of water molecules by the film through a typical mass loading effect.¹⁹⁶ As shown in Fig. 9(e), a microwave resonator coupled with PEDOT:PSS film was developed for humidity sensing.⁴⁸ The transmission coefficient and resonance frequency of the sensor changed with the humidity and showed high repeatability in the humidity range of 40%RH–60%RH.⁴⁸ Julian et al.¹⁹⁵ developed a quartz crystal microbalance (QCM)-based humidity sensor in which the sensing chips were coated with PEDOT:PSS mixed with polyvinyl alcohol (PVA) nanofibers. The sensor demonstrated excellent sensitivity (up to 33.56 Hz/%RH), fast response/recovery times (5.6 s/3.5 s), long-term stability, and low hysteresis (<1.3%).

PEDOT:PSS has been shown to be one of the most promising candidates for wearable temperature sensors,^203,204 owing its good flexibility and high thermal response.^63,205 PEDOT:PSS itself is very sensitive to temperature changes,²⁰⁶ since these result in microstructural changes.²⁰⁷ Owing to the hygroscopicity of PSS, temperature changes affect the water content and volume of the insulating and hydrophilic PSS-rich shell.^29,208 At room temperature, the hydrophilic PSS absorbs moisture from the air by forming hydrogen bonds between the sulfonic acid groups of the PSS chains and water molecules. The insulating PSS chains, together with nonconductive water, form boundaries to prevent electrons hopping between PEDOT:PSS nanoparticles. At high temperature, water loss from PSS not only reduces the total number of particle boundaries but also decreases the effective “size” of these boundaries. When the temperature decreases, the electrons may not have enough energy to overcome these barriers caused by the particle boundaries, and thus the resistance increases,²⁰⁸ as shown in Fig. 10(a).

In recent years, attempts have been made to improve the thermal sensitivity of PEDOT:PSS-based devices.^29,49,50 The most commonly used techniques are secondary doping^198,209 and thermal expansion.^49,50 It has been found that secondary doping of PEDOT:PSS using materials with higher thermal conductivity [e.g., GO,²⁹ rGO,²⁰⁹ multiwalled carbon nanotubes (MWCNTs),¹⁹⁸ and silver nanoparticles (AgNPs)²¹⁰] is a relatively easy strategy. Zhou et al.²⁹ optimized the temperature-sensing performance of a PEDOT:PSS nanodevice by doping the PEDOT:PSS with GO. Nanoscale GO flakes filled the gaps between adjacent PEDOT:PSS nanoparticles, serving as temperature-dependent conductive paths for electron hopping. The temperature coefficient of resistance (TCR) of the PEDOT:PSS/GO nanowire reached a maximum of −1.2%/°C when the mixing ratio of PEDOT:PSS and GO was 13:1. There is a threshold at which the GO flakes completely fill the gaps between adjacent PEDOT:PSS nanoparticles. Excess GO will then lead to a deterioration in the cohesiveness of adjacent PEDOT:PSS nanoparticles, while insufficient GO will reduce the path of electronic transition, thereby reducing electron mobility and the sensitivity of the sensor,²⁹ as shown in Fig. 10(b). A second method is thermal expansion of the substrate, which can create microcracks in the sensing layer to improve the temperature sensitivity of the sensor. Yu et al.⁴⁹ fabricated a PEDOT:PSS sensing layer with microcracks on a PDMS substrate through pre-stretching and sulfuric acid treatment, as shown in Fig. 10(c). Increasing temperature caused the PDMS substrate to swell and expanded the cracks in the PEDOT:PSS sensing layer, resulting in an increase in resistance. The positive TCR caused by PDMS expansion is much larger than the inherent TCR of PEDOT:PSS (about −0.45%/°C), resulting in a high temperature sensitivity of 4.2%/°C of the entire sensor. By contrast, the crack-free PEDOT:PSS-PDMS sensor has a rather low sensitivity of only 0.1%/°C.⁴⁹ In addition, a method that combines secondary doping and material swelling has been reported to improve the sensitivity of temperature sensors.⁵⁰ Oh et al.⁵⁰ fabricated an Au-poly(N-isopropylacrylamide) (pNIPAM)/PEDOT:PSS/carbon nanotube (CNT) multilayer structured temperature sensor on a PDMS substrate. The enhanced electron hopping between PEDOT:PSS and CNTs in the PEDOT:PSS/CNT composites due to increased temperature improved the thermal sensitivity. Besides, at high temperature, the microstructure of pNIPAM was coiled and hydrophobic, leading to expulsion of absorbed water. The reduced volume of the pNIPAM made the closely attached PEDOT:PSS/CNT network more densely packed, which further reduced the resistance of the sensor. By contrast, at low temperature, the microstructure of the pNIPAM became hydrophilic, absorbed water, and swelled. The closely attached PEDOT:PSS/CNT network was loosened, increasing the resistance of the sensor. This approach helped to improve the sensitivity of the sensor to −2.6%/°C. The structure of the sensor is shown in Fig. 10(d).

The compressibility of pure PEDOT:PSS is limited by its rigid conjugated backbone,²¹¹ limiting the potential of pristine PEDOT:PSS as a sensing element in pressure sensors.^212,213 However, Wang et al.⁵¹ did develop a pure PEDOT:PSS pressure sensor. In this sensor, 0.87–1.88 μm thick pure PEDOT:PSS film was directly compressed to test its pressure response. Figure 11(a) shows the rapid decrease in sensor sensitivity with increasing pressure caused by the fairly small compressibility in the thickness direction. Therefore, a prerequisite for PEDOT:PSS-based pressure sensor design is to endow the PEDOT:PSS with greater compressibility.²¹⁴ When a compressible elastic structure with PEDOT:PSS as the main conductive path is deformed under pressure, its electrical conductivity is changed, which makes it able to sense pressure. We classify PEDOT:PSS pressure sensors into three categories according to the method used to realize “compressible” PEDOT:PSS, namely, doping with soft polymers or plasticizers,⁵² making compressible skeletons or dielectrics,^215–217 and designing microstructures.^218,219

The first approach involves changing the compressibility of PEDOT:PSS directly through doping soft polymers or plasticizers into it.⁴² Tseng et al.⁵² introduced poly (acrylic acid) (PAA) into the PEDOT:PSS and obtained an increased stretchability (from 20% to 40%) and a much-decreased Young’s modulus, as shown in Fig. 11(b). A pyramidal-microstructure-based pressure sensor was then developed using this compressible and conductive material. Here, a noteworthy problem is that soft polymers or plasticizers, nonconductive in most cases, will lead to a decrease in the overall conductivity. There is always a trade-off between compressibility and conductivity. Methanol treatment was adopted to solve this problem.⁵² The sensitivity of the pressure sensor doubled after such treatment.

Increased compressibility of PEDOT:PSS can also be achieved indirectly by attaching it to a compressible skeleton or dielectrics.^215,216 The simplest and easiest method is to soak a sponge in PEDOT:PSS solution, as shown in Fig. 11(c).^53,101 PEDOT:PSS is then anchored to the sponge as the water evaporates. An advantage of a sponge-based sensor is that its sensitivity can be adjusted by changing the size and density of the pores in the sponge over a large range.⁵³ In another study, Lee et al.²²⁰ coated a thin layer of PEDOT:PSS onto a pyramidal PDMS substrate to achieve high-sensitivity pressure detection, since the point contact condition at the beginning of compression enabled a large response under a small pressure, as shown in Fig. 11(d). The ultrasensitivity of this pressure sensor was demonstrated by tracking the blood pulse at the wrist.²²⁰ 

Structural design is another strategy to improve the compressibility of PEDOT:PSS.²¹⁹ Out-of-plane buckling is frequently used in PEDOT:PSS stretchability enhancement. In-plane tensile strain is relieved when the buckling PEDOT:PSS film becomes flat.²¹⁸ In fact, pleats on PEDOT:PSS films not only improved the in-plane stretchability but also increased the compressibility along the thickness direction.²¹⁸ A pressure sensor based on pleated PEDOT:PSS film was developed by Tan et al.²¹⁸ and exhibited a high sensitivity of 642.5 kPa⁻¹ and a rapid response time (200 μs), as shown in Fig. 11(e).

PEDOT:PSS-based strain sensors have attracted widespread attention in various applications, such as artificial electronic skin and health monitoring/diagnosis,^54,56 owing to the excellent electrical conductivity and mechanical properties of PEDOT:PSS.²⁰ Intrinsic PEDOT:PSS is not suitable for stretchable strain sensors because of its brittleness and susceptibility to breakage.⁵⁵ However, the stretchability of PEDOT:PSS can be improved by mixing with elastomers^55,221,222 or by forming textile fibers^56,223,224 or aerogel.^54,209,211 PEDOT:PSS can be incorporated into a stretchable elastic structure to make the latter conductive. Strain changes the shape of the PEDOT:PSS-based stretchable structure, thereby changing its electrical conductivity. According to this principle, the PEDOT:PSS-based structure can be used for strain sensing. For example, latex⁵⁵ has been mixed with PEDOT:PSS to form an elastomer matrix at an appropriate mixing ratio to produce a conductive film with excellent stretchability.²²¹ Panwar and Anoop²²² mixed PEDOT:PSS with a biocompatible polymer poly (vinylidene fluoride–trifluoroethylene–chlorotrifluoroethylene) P(VDF-TrFE-CTFE) and heat-cured the mixture to create a conductive elastomer, as shown in Fig. 12(a). Owing to the molecular bonding between the biocompatible polymer and the PEDOT:PSS, the strain sensor can achieve a maximum strain of 230%.

A dip-coating method has been used to facilitate attachment of PEDOT:PSS from solution to realize textile-fiber-based strain sensors.^225,226 Excellent flexibility and skin adhesion makes these conductive textiles good candidates for constructing wearable flexible sensors.²²⁷ For example, PEDOT:PSS was dipped into a polyurethane (PU) fiber core coated with CNTs. Here, the PU fiber was pre-stretched to produce cracks on the PEDOT:PSS, which reduced the initial resistance of the strain sensor and improved its strain sensitivity (by about six times), as shown in Fig. 12(b). The CNTs bridging the PEDOT:PSS prevent complete rupture of the conductive path when the sensor is subjected to a large strain, thereby increasing the sensing range (0.1%–150% strain).⁵⁶ 

Zhang et al. prepared rGO/PEDOT:PSS aerogel by doping GO into PEDOT:PSS and then reducing GO, as shown in Fig. 12(c). A small amount of PEDOT:PSS nanoparticles greatly improved the strain sensitivity of aerogels (from 580% to 5760%) due to the specific cell size of the aerogel and the bridging effect of PEDOT:PSS nanoparticles.²⁰⁹ Zhou and Hsieh⁵⁴ developed another strong and highly conductive aerogel based on cellulose nanofibrils (CNFs) and PEDOT:PSS in which surface carboxylates of the CNFs were protonated into carboxyls and hydrogen-bonded with the PSS. The strength of the aerogel was increased by an order of magnitude, which is attributed to the change in the PEDOT benzenoid structure. Annealing in ethylene glycol vapor increased the conductivity of the aerogel by two orders of magnitude, and a strain sensor formed by injecting PDMS into the aerogel had a good stretchable range (a maximum strain of 90%) and sensitivity (a gauge factor of 14.8).

Ultrasmall sensors (<1 μm) have emerged that reduce the minimum detection limit of strain sensors.²²⁸ Tetsu et al.²²⁹ developed an ultrathin (thickness ∼1 μm) strain sensor that conforms to the epidermal structure. It can detect small deformations of the human skin (∼2%) and reduce the influence of the natural deformation of the skin during measurement. Tang et al.⁵⁸ prepared highly aligned PEDOT:PSS nanowires that can cover a large area through a mechanism involving capillary action. A nanomorphology-based strain sensor using these nanowires had greatly improved sensitivity (a gauge factor of 35.8). It could detect slight bending changes up to 200 μm with a rapid response (230 ms) and had a negative or positive response to inward or outward bending, respectively. This is due to the fact that the inward bending of the sensor reduces the distance between adjacent PEDOT:PSS nanocrystals, thereby reducing the charge transfer distance and the sensor resistance. The opposite is true for outward bending. In addition, the nanowire sensor exhibits anisotropy, i.e., it is only sensitive to the bending strain parallel to the nanowire, but not to the bending strain perpendicular to the nanowire. However, such PEDOT:PSS-based strain sensors exposed to the air will be affected by the environment owing to the sensitivity of PEDOT:PSS to humidity. A processing method in which PEDOT:PSS is injected into microchannel can prevent interference by air and humidity, as well as solving the problem of connection damage caused by frequent bending.²³⁰ Bhattacharjee et al.²³¹ injected a conductive solution of PEDOT:PSS into a microchannel with a diameter of 275 μm formed by PDMS to make a strain sensor. When the sensor was stretched, the axial deformation of the microchannel reduced the effective volume fraction of the internal PEDOT:PSS. The creation of an electrical discontinuity led to a decrease in the conductive path inside the channel, thereby increasing the resistance of the sensor. This structure enabled the sensor to achieve an ultrahigh gauge factor of about 12 000, which is 400 times higher than that of most strain sensors.

Over the years, PEDOT:PSS, as the most commercially successful conductive polymer, has been extensively developed and studied. PEDOT:PSS in the form of aqueous dispersions is compatible with a variety of fabrication techniques. It can be fabricated into high-performance sensors through simple and low-cost methods. In this review, the morphology of PEDOT:PSS in the forms of aqueous dispersions and solid films has been described, and the potential of emerging hydrogels has also been discussed. Traditional processes (coating, printing, and lithography) and emerging processes (soft lithography and nanoimprint lithography) used to prepare PEDOT:PSS-based equipment have been presented, and the latest developments of PEDOT:PSS in physical sensors (for humidity, temperature, pressure and strain sensors) have been described. We have specifically pointed out that PEDOT:PSS-based nanowires made by nanoscale patterning improves the sensitivity and response speeds of humidity and temperature sensors.

Although much progress has been made in these areas, there are still issues that need to be resolved to allow further development. First, the uniformity of the films is important for stability and consistency of performance in each area. However, to date, there have been no studies of the control of uniformity during the preparation and conductivity enhancement of PEDOT:PSS solid films. Particular attention needs to be paid to the heterogeneity caused by phase separation during the drying process, which is also a problem that needs to be solved when PEDOT:PSS films are applied to precision instruments and metal-free devices in the future. Second, the high-precision micro/nanoprocessing currently used for PEDOT:PSS-based devices has the disadvantages of high cost and low yield, which limits their application. In the future, it will be necessary to develop low-cost and high-precision processes that can be used in industrial mass production. Third, sensors based on PEDOT:PSS are susceptible to environmental interference, since PEDOT:PSS is sensitive to temperature, humidity, and strain. Reasonable packaging and appropriate compensation are the major directions for the development of PEDOT:PSS-based sensors to take in the future. Finally, the microstructure of PEDOT:PSS-based hydrogels is not yet completely clear. More efforts are required to study the polymer chain system of PEDOT:PSS-based hydrogels on the microscopic scale. In general, therefore, PEDOT:PSS research, from material to manufacturing process to sensors, still has a long way to go.

This work was supported by the National Natural Science Foundation of China (Grant Nos. 62001325, 91743110, 52075384, and 21861132001), the National Key R&D Program of China (Grant No. 2018YFE0118700), Tianjin Applied Basic Research and Advanced Technology (Grant No. 17JCJQJC43600), the Foundation for Talent Scientists of Nanchang Institute for Microtechnology of Tianjin University, and the “111” Project (Grant No. B07014).

The authors have no conflicts to disclose.

X.Z. and W.Y. contributed equally to this work.

Data sharing is not applicable to this article as no new data were created or analyzed in this study.