Fully soluble self-doped poly(3,4-ethylenedioxythiophene) with an electrical conductivity greater than 1000 S cm−1

See allHide authors and affiliations

Science Advances  12 Apr 2019:
Vol. 5, no. 4, eaav9492
DOI: 10.1126/sciadv.aav9492


Wet-processable and highly conductive polymers are promising candidates for key materials in organic electronics. Poly(3,4-ethylenedioxythiophene):poly(4-styrenesulfonate) (PEDOT:PSS) is commercially available as a water dispersion of colloidal particles but has some technical issues with PSS. Here, we developed a novel fully soluble self-doped PEDOT (S-PEDOT) with an electrical conductivity as high as 1089 S cm−1 without additives (solvent effect). Our results indicate that the molecular weight of S-PEDOT is the critical parameter for increasing the number of nanocrystals, corresponding to the S-PEDOT crystallites evaluated by x-ray diffraction and conductive atomic force microscopic analyses as having high electrical conductivity, which reduced both the average distance between adjacent nanocrystals and the activation energy for the hopping of charge carriers, leading to the highest bulk conductivity.


Although investigations of electrically conductive conjugated polymers date back to the early 1960s, particularly on polypyrrole and polyaniline (1), the development of conductive polymers in the 1970s by A. G. MacDiarmid, A. J. Heeger, and H. Shirakawa (2000 Nobel Prize in Chemistry), where electrical conductivity of polyacetylene films increased remarkably upon iodine doping (2, 3), accelerated explosively the use of organic electronics in low-cost, lightweight, and flexible electronic devices such as organic light-emitting diodes (OLEDs), organic field effect transistors, and organic photovoltaics (OPVs) (4, 5) since the early 1980s. Furthermore, in the era of the “Internet of things”, the field of organic electronics is developing via increased interest in printed electronics, stretchable electronics, and, recently, wearable electronics for use in flexible displays and touch panels, soft sensors, and actuators (6, 7). In such devices, a flexible, wet-processable, and highly conductive polymer would be a promising candidate as a key material.

Poly(3,4-ethylenedioxythiophene):poly(4-styrenesulfonate) (PEDOT:PSS), developed by Bayer AG in 1990 (8, 9) and commercially available as a water dispersion of colloidal particles, is one of the most successful conductive polymers of the last two decades. Because of its high electrical conductivity (up to 1000 S cm−1), transparency, and thermal stability, PEDOT:PSS has been used in a wide field of applications, including antistatic coatings, hole injection layers in OLEDs and OPVs, cathode materials in solid aluminum electrolytic capacitors (10), and transparent electrodes for various electronic devices, as an alternative to indium tin oxide (11, 12). However, PEDOT:PSS has some technical issues that impede its further practical application; these issues stem from the use of PSS as external ions to compensate the positive charges on PEDOT. Since the hydrophobic PEDOT core is surrounded by a shell of excess hydrophilic PSS (13) for water dispersion as a colloidal particle with a diameter of several tens of nanometers (11), (i) forming ultrathin layers thinner than the colloidal particle size is difficult (14), (ii) the PEDOT:PSS colloids aggregate and precipitate when stored for an extended period, and (iii) the pristine PEDOT:PSS exhibits poor electrical conductivity (<1 S cm−1) (15); additives, such as ethylene glycol (EG), dimethyl sulfoxide (DMSO), and sorbitol, are therefore indispensable for improving electrical conductivity, namely, by the “solvent effect” (1520), which can be interpreted in terms of the crystallization of the PEDOT core and the removal of the insulating PSS shell of the colloidal particles to enhance the transport of charge carriers between the PEDOT nanocrystals by a hopping mechanism (2124). That is, the insoluble and intractable nature of PEDOT is responsible for these drawbacks of PEDOT:PSS having a 2.5-fold excess of PSS over PEDOT by weight (24). Therefore, prescient and intensive studies on fully soluble self-doped PEDOT (S-PEDOT) have been reported in the literature (2529). Zotti and co-workers (26) synthesized an EDOT derivative bearing a sodium alkylsulfonate side chain, sodium 4-[(2,3-dihydrothieno[3,4-b][1,4]dioxin-2-yl)methoxy]butane-1-sulfonate (fig. S1A), and polymerized it to obtain S-PEDOT with a mass average molecular weight (Mw) of 5000 g mol−1 and a maximum electrical conductivity of 1 to 5 S cm−1. On the other hand, Konradsson, Berggren, and co-workers reported chemical polymerizations of the same S-PEDOT and found that the electrical conductivities were 12 S cm−1 (27) and 30 S cm−1 (28), respectively. Since these conductivity values were two or three orders of magnitude lower than that of the highly conductive PEDOT:PSS (up to 1000 S cm−1), improving the electrical conductivity of S-PEDOT has thus far been considered impossible.

In this study, we first synthesized a novel EDOT derivative (S-EDOT) bearing a sodium alkylsulfonate side chain, sodium 4-[(2,3-dihydrothieno[3,4-b][1,4]dioxin-2-yl)methoxy]butane-2-sulfonate (fig. S1B). From the viewpoints of safety, low environmental load, and mass production for commercialization, the S-EDOT monomer was designed because of the lower toxicity of 2,4-buthanesultone (table S1) (30) used in the synthetic process. Furthermore, the fully soluble S-PEDOT was synthesized by the oxidative polymerization of different concentrations (CS-EDOT) of the S-EDOT monomer in a range of 1 to 10 weight % (wt %) (Fig. 1) and characterized by means of gel permeation chromatography (GPC), UV-vis-NIR (ultraviolet–visible–near infrared) spectroscopy, x-ray diffraction analysis (XRD), conductive atomic force microscopy (cAFM), and the four-point technique. It was found that the electrical conductivity achieved as high as 1089 S cm−1, which is two orders of magnitude higher than those of previously reported S-PEDOTs (2529), and exceeded the conductivity of PEDOT:PSS (up to 1000 S cm−1). The results led us to conclude that the molecular weight of S-PEDOT is the critical parameter for increasing the number of nanocrystals, corresponding to the S-PEDOT crystallites evaluated by XRD and cAFM, which reduced both the average distance between adjacent nanocrystals and the activation energy for the hopping of charge carriers, leading to the highest bulk conductivity.

Fig. 1 Synthesis of S-PEDOT by oxidative polymerization of the S-EDOT monomer.


Molecular weight and solution properties

S-PEDOT was synthesized by the oxidative polymerization of different concentrations (CS-EDOT) of the S-EDOT monomer in a range of 1 to 10 wt % using (NH4)2S2O8 and FeSO4 as oxidant and catalyst, respectively (Fig. 1). The GPC elution curves of various S-PEDOTs are shown in Fig. 2A. Since the GPC measurement is not applicable to the PEDOT:PSS colloidal dispersions, these results provide clear evidence that the S-PEDOT is fully soluble, similar to other self-doped conductive polymers (2529, 3136). A flow-through test was carried out using a membrane filter with a pore diameter of 0.1 μm [Optimizer V-47, 0.1 μm hydrophobic ultrahigh molecular weight polyethylene (UPE), Entegris]. An S-PEDOT water solution (0.5 wt %) permeated without resistance, whereas a PEDOT:PSS water dispersion (0.5 wt %; Clevios PH1000, Heraeus) hardly passed through the membrane filter; only transparent water was permeated under pressurization (fig. S2). The GPC curves were unimodal with polydispersity indices (PDIs) of 3.15 to 4.15, indicative of a broad molecular weight distribution. This is probably because the termination reaction is due to recombination rather than disproportionation (9). The weight average molecular weight (Mw) and the number average molecular weight (Mn) of the S-PEDOT reached maximum values of 22,300 and 5380 g mol−1, respectively, at CS-EDOT = 5 wt %. Here, the number average degree of polymerization (DP) was ca. 18, which was more than twice higher than the value of the previously reported S-PEDOT (8 units average) measured by matrix-assisted laser desorption ionization time-of-flight mass spectrometry (26). The decrease of both Mw and Mn at CS-EDOT > 5 wt % was likely due to a steep increase in the viscosity of the reaction solution; for example, at CS-EDOT = 10 wt %, the stirring of the reaction solution became difficult in 30 min after adding the (NH4)2S2O8 solution, which may decrease the diffusion of reactants such as monomer, oxidant, and catalyst, leading to the inhomogeneous polymerization and/or low monomer conversion. Figure 2B shows that the viscosity of the S-PEDOT solutions (1 wt %) reached a maximum (24.3 mPa·s) at CS-EDOT = 5 wt %, reflecting the molecular weight, whereas the pH values were nearly constant (1.86 to 2.04). These values are comparable to those of the PEDOT:PSS water dispersion (1 wt %) with a viscosity and pH of 16 mPa·s and 1.89, respectively. As shown in Fig. 2C, the UV-vis-NIR spectra of the S-PEDOT water solutions (0.01 wt %) were less dependent on CS-EDOT, demonstrating the same electronic structure: A broad shoulder at approximately 800 nm is consistent with the low energy absorbance of neutral PEDOT, and a free carrier of bipolarons in the NIR region corresponds to PEDOT in the oxidized state (3739). To compare the optical properties in more detail, absorption and transmittance spectra of S-PEDOT and PEDOT:PSS (Clevios PH1000, Heraeus) thin films were measured (fig. S3). The absorption of the S-PEDOT thin film (125 nm thick) was slightly higher than that of the PEDOT:PSS thin film (128 nm thick) in the whole experimental range of wavelength, which is associated with the higher optical density of PEDOT chromophore in S-PEDOT than in PEDOT:PSS. On the other hand, the transmittance of the S-PEDOT thin film was lower than that of the PEDOT:PSS, where the value at 550 nm, near the center of visible light, was 83%. Furthermore, while PEDOT:PSS can be dispersed only in water and in several mixed solvents of water and alcohol, S-PEDOT is soluble not only in water but also in various organic solvents (table S2), which is a great advantage for use in organic electronic devices easily degraded by water and for the fabrication of composites with water-insoluble polymers.

Fig. 2 Solution properties of S-PEDOT.

(A) GPC elution curves, weight average molecular weight (Mw), number average molecular weight (Mn), and PDI of S-PEDOT synthesized with different CS-EDOT. (B) Viscosity and pH dependencies of the S-PEDOT water solutions (1 wt %) on CS-EDOT. (C) UV-vis-NIR spectra of the S-PEDOT water solutions (0.01 wt %) synthesized with different CS-EDOT.

Crystalline structure

XRD analysis was performed to clarify the structure of S-PEDOT; the results are shown in Fig. 3A. Although S-PEDOT is a new polymer, the diffraction patterns are simpler than the PEDOT:PSS (13, 40) and similar to those of PEDOTs doped with perchlorate (PEDOT:ClO4), p-toluenesulfonate (PEDOT:TsO), and hexafluorophosphate (PEDOT:PF6), which indicates that the S-PEDOT unit cell belongs to an orthorhombic system (4043). Therefore, the diffraction peaks at 2θ = 5.3° (d = 16.6 Å), 10.5° (d = 8.4 Å), and 24.7° (d = 3.6 Å) can be assigned to the diffractions from the (100), (200), and (020) planes of the orthorhombic unit cell of the S-PEDOT crystal with lattice parameters of a = 16.6 Å, b = 7.2 Å, and c = 7.8 Å. Since the main chain of the S-PEDOT matches the c axis, the c value corresponds to the repetition period of two monomer units (7.8 Å) similar to that of PEDOT (4043). However, the π-stacking distance between two S-PEDOT chains, representing one-half of the value of b (3.6 Å), is slightly longer than that of PEDOT (3.4 Å) (4043) due to the steric hindrance and/or electrostatic repulsion between the S-PEDOT side chains. Unlike polyalkylthiophenes, in which the dopant ions are present in the vicinity of the polymer chains (44), PEDOT exhibits dopant ions between the π-stacked columns along the a axis direction (4043). Thus, the a value (16.6 Å) of the S-PEDOT crystal is determined by the length of the alkylsulfonic acid side chains. Since the unit lattice contains four monomer units with the side chains alternatingly arranged in opposite directions with respect to the polymer chain, the theoretical density of the S-PEDOT crystal is calculated to be 1.59 g cm−3, which is consistent with the actual density measured by the method of Archimedes (1.50 to 1.53 g cm−3). Assuming that a broad halo in the 2θ range of 15° to 35° corresponds to the amorphous S-PEDOT, the degree of crystallinity (Xc) can be expressed as the ratio of total area of crystalline peaks to total area of all diffraction peaks. As shown in Fig. 3B, the Xc value of the S-PEDOT film reached 72% at CS-EDOT = 5 wt %, the tendency of which agrees with the molecular weight and viscosity results (Fig. 2B). However, the result is inconsistent with the trend observed in common polymers (such as polyethylene); because of the entanglement of the molecular chains, a higher molecular weight results in a lower Xc (45). This trend suggests that the S-PEDOT exhibits less entanglement in water since the polymer chains have an extended coil structure due to the rigid π-conjugated backbone and electrostatic repulsion between the alkylsulfonic acid side chains, favoring crystallization by π-stacking in the solid state. Notably, the Xc values are substantially higher than those of PEDOT:PSS (24) and are associated with the higher weight fractions of PEDOT main chains in S-PEDOT (45 wt %), the absence of amorphous PSS chains (the weight fraction of PEDOT in PEDOT:PSS is only 29%), and/or the side chains of S-PEDOT being highly ordered and incorporated in the crystalline phase. Assuming that the doping level of S-PEDOT is the same as that of PEDOT (0.22 to 0.35) (46), corresponding to one charge per three to four monomer units, 22 to 35% of the sulfonic acid groups of the S-PEDOT side chains dissociate and compensate for the positive charges on S-PEDOT main chains as counter ions. However, 65 to 78% of the undissociated sulfonic acid groups in the S-PEDOT side chains can form intermolecular hydrogen bonds and stabilize the crystalline structure, as shown in the inset of Fig. 3A. Furthermore, the crystallite sizes normal to the (100) and (020) planes calculated using the Scherrer’s equation were D100 ≈ 6 nm and D020 ≈ 3 nm, respectively, regardless of the CS-EDOT (Fig. 3B). Therefore, the high Xc value can be explained in terms of the large number of crystallites, in which approximately four columns of eight S-PEDOT molecules π-stacked in the b axis direction are accumulated in the a axis direction (fig. S4).

Fig. 3 Crystalline structure of S-PEDOT.

(A) X-ray diffraction (XRD) patterns of S-PEDOT films with different CS-EDOT and a possible crystalline structure of S-PEDOT (inset). (B) Crystallinity (Xc) and crystallite size (D100 and D020) dependencies of S-PEDOT films on CS-EDOT. a.u., arbitrary units.

Surface morphology and current images

As shown in Fig. 4A, the surface morphologies of S-PEDOT thin films were similar irrespective of CS-EDOT, where numerous particles with an average particle size (Dp) of ca. 11 nm were densely packed to form a coherent solid film with a particle number (Np), defined as the number of particles in 1 μm2, of 3000 to 3600 (Fig. 4B). In addition, S-PEDOT thin films were smooth, with a surface roughness (Ra) of ca. 0.4 nm, compared to PEDOT:PSS films (1 ≤ Ra ≤ 2 nm) (21), because S-PEDOT is not dispersed as a colloid but is fully dissolved in water. Figure 5A shows current images representing the distribution of local conductivity measured by cAFM; these images reveal that the highly conductive S-PEDOT nanocrystals (shown in the bright region) are sparsely distributed in less conducive amorphous matrices (shown in the blue region), similar to PEDOT:PSS (24). The average size of S-PEDOT nanocrystals (Dnc) was found to be an almost constant value of ca. 5 nm (Fig. 5B), coincident with the crystallite sizes (D100 and D020) in Fig. 3B, suggesting that the S-PEDOT nanocrystal corresponds to a single crystal. However, the number of nanocrystals in 1 μm2 (Nnc) shows a maximum of 5800 μm−2 at CS-EDOT = 5 wt %. The tendency of Nnc is similar to those of the molecular weight (Fig. 2A) and the Xc (Fig. 3B), demonstrating that one particle in Fig. 4A consists of ca. two nanocrystals. Therefore, the average distance between adjacent nanocrystals (Lnc) reached a minimum (3.9 nm) at CS-EDOT = 5 wt %. This interparticle distance is suitable for the hopping of charge carriers between adjacent nanocrystals, which is critical to achieving high bulk conductivity (24).

Fig. 4 AFM height images of S-PEDOT.

(A) Surface morphologies of S-PEDOT thin films with different CS-EDOT. (B) Average particle size (Dp), number of particles (Np), and surface roughness (Ra) dependencies of S-PEDOT thin films on CS-EDOT.

Fig. 5 cAFM images of S-PEDOT.

(A) Current mapping images of S-PEDOT thin films with different CS-EDOT. (B) Average nanocrystal size (Dnc), number of nanocrystals (Nnc), and average length between adjacent nanocrystals (Lnc) dependencies of S-PEDOT thin films on CS-EDOT.

Electrical conductivity and transport of charge carriers

To clarify the correlation between the current images at the nanoscale and the transport of charge carriers in the bulk state, electrical conductivity was measured by the four-point technique. We should emphasize that an electrical conductivity as high as 1089 S cm−1 at CS-EDOT = 5 wt % (Fig. 6A) was attained without additives (solvent effect). This value is more than two orders of magnitude higher than that of the S-PEDOT (2529) and exceeds the conductivity of PEDOT:PSS with additives such as EG and DMSO (up to 1000 S cm−1). In addition, the S-PEDOT was stable in nitrogen and air, where the electrical conductivity of the film held in air at room temperature for 1 year was almost constant to be 1058 S cm−1. On the other hand, the electrical conductivity was reduced to 836 S cm−1 after the film was soaked in water for 3 days. The value is still higher than those of the oxidative chemical vapor–deposited zwitterionic PEDOT thin films before (668 S cm−1) and after water soaking (172 S cm−1) (47), demonstrating that the S-PEDOT is stable even in water.

Fig. 6 Electrical conductivity and carrier transport properties of S-PEDOT.

(A) Electrical conductivity dependency of S-PEDOT films measured by the four-point method on CS-EDOT. (B) Electrical conductivity and activation energy (Ea) dependencies of S-PEDOT films on Lnc. (C) Lnc, electrical conductivities at 77 K (σ77K) and 293 K (σ293K), and Ea of S-PEDOT films with different CS-EDOT. The vertical and horizontal axes of (B) (top) correspond to the vertical axis of (A) and the horizontal axis of (B) (bottom), respectively.

A strong correlation is observed between Lnc and electrical conductivity (Fig. 6B), demonstrating that the distribution of nanocrystals plays an important role in electrical conduction. According to the nearest-neighbor hopping model (48), the activation energy (Ea) required for hopping of charge carriers can be calculated by measuring the electrical conductivity (σ) at room temperature (T = 298 K) and that at liquid-nitrogen temperature (T = 77 K) and using the equation σ = σ0 exp.(−Ea/kBT), where σ0 and kB are the conductivity factor and the Boltzmann constant (8.617 × 10−5 eV K−1), respectively (21). A linear relationship between Lnc and Ea demonstrates that the longer the distance between adjacent nanocrystals, the lower the probability of charge carrier hopping (Fig. 6B). Since the values of Ea (Fig. 6C) are lower than the thermal energy at room temperature, ca. 26 mV, the charge carriers in S-PEDOT have sufficient energy for hopping, which is responsible for the extremely high electrical conductivities at room temperature.


Figure 7A shows the hierarchical structure of S-PEDOT. The S-EDOT monomer (primary structure), sodium 4-[(2,3-dihydrothieno[3,4-b][1,4]dioxin-2-yl)methoxy]butane-2-sulfonate, was newly synthesized and chemically polymerized into S-PEDOT, where Mw (secondary structure) reached a maximum (22,300 g mol−1) at CS-EDOT = 5 wt % (Fig. 7B). Furthermore, the values of Xc and Nnc (tertiary structure) increased in proportion to Mw, reaching 72% and 5800 μm−2, respectively (Fig. 7C). Here, the larger the Nnc, the lower the Lnc (3.9 nm) (quaternary structure) and Ea (4.4 meV) (Fig. 7D), which resulted in a remarkable increase in bulk conductivity up to as high as 1089 S cm−1 (Fig. 7E). Thus, the results led us to conclude that molecular weight was the key parameter responsible for improving electrical conductivity, which arose from the fact that the Mw of the S-PEDOT with low conductivity (1 to 5 S cm−1) was only 5000 g mol−1 (26). In addition, synthesis of S-PEDOT with a narrow molecular weight distribution by optimization of polymerization conditions will be important for improving the electrical conductivity. Notably, the relation between Lnc and electrical conductivity (Fig. 7E) has not yet saturated, suggesting that a further improvement in conductivity can be achieved by decreasing the Lnc value (facilitating hopping of charge carriers); further investigations are currently underway.

Fig. 7 Correlation between the hierarchical structure and properties of S-PEDOT.

(A) Chemical structures of the S-EDOT monomer (primary structure) and S-PEDOT (secondary structure), the S-PEDOT nanocrystal (tertiary structure), and the distribution of S-PEDOT nanocrystals (quaternary structure). (B) Relation between CS-EDOT and Mw. (C) Dependencies of Xc and Nnc on Mw. (D) Dependencies of Lnc and Ea on Nnc. (E) Relation between Lnc and electrical conductivity.


Synthesis of S-EDOT

Hydroxymethyl EDOT (HMEDOT) was synthesized by the reaction of 3,4-dihydroxythiophene-2,5-dicarboxylic acid diethyl ester (Sigma-Aldrich) and 2,3-dibromo-1-propanol (Tosoh Finechem), with subsequent hydrolysis and decarboxylation. The S-EDOT monomer, sodium 4-[(2,3-dihydrothieno[3,4-b][1,4]dioxin-2-yl)methoxy]butane-2-sulfonate (fig. S1B), was synthesized by the sulfonation of HMEDOT with 2,4-buthanesultone (Fujifilm Wako Pure Chemical), where the overall yield was 42%, and the purity evaluated by liquid chromatographic analysis was 99.2% (fig. S5) (49). The molecular structure of the as-synthesized S-EDOT monomer was fully characterized by 1H nuclear magnetic resonance (NMR), 13C NMR, and Fourier transform infrared spectroscopies and by elemental analysis (fig. S6).

Polymerization of S-PEDOT

S-PEDOT was synthesized by the oxidative polymerization of the S-EDOT monomer at different concentrations (CS-EDOT) in the range of 1 to 10 wt % (Fig. 1). A typical polymerization procedure at CS-EDOT = 5 wt % was as follows. The S-EDOT monomer (50 g, 151 mmol) and FeSO4·7H2O (25.2 g, 90.6 mmol; Fujifilm Wako Pure Chemical) were dissolved in a 1 M sulfuric acid water solution. Meanwhile, (NH4)2S2O8 (68.9 g, 302 mmol; Fujifilm Wako Pure Chemical) was dissolved in pure water (41.2 wt %) and added dropwise to the S-EDOT solution. The total weight of the reaction solution was typically 1 kg, and the oxidative polymerization was carried out under vigorous stirring at 20°C for 20 hours in a nitrogen atmosphere. After impurity ions were removed using cation and anion exchange resins (Lewatit Monoplus S108H and MP62WS, Lanxess), the solution was homogenized at 70 MPa using a high-pressure homogenizer (R5 10.38, SMT). At different CS-EDOT, the polymerization procedure was exactly the same except that the FeSO4 and (NH4)2S2O8 were 0.6- and 2-fold, respectively, to the S-EDOT monomer. 1H NMR measurement of the S-PEDOT was carried out, but clear spectrum could not be obtained similarly to the previous work (26). This can be explained in terms of the Knight shift, where NMR line shift and line broadening are brought about probably by the increase of Pauli paramagnetic spin susceptibility (50) in the doped highly conductive S-PEDOT.

GPC and solution characteristics

The Mw, Mn, and PDI of S-PEDOT were measured by GPC on an instrument (Prominence, Shimadzu) equipped with a TSKgel α-M column and guard column α (Tosoh); polystyrene sulfonate standards (PSS, American polymer standard) were used. The S-PEDOT aqueous solution (1 wt %) was neutralized with N,N-diisopropylamine (Junsei Chemical), dried on a hot plate at 75°C, and then dried in a vacuum at 65°C for 2 hours. The GPC measurements were performed on DMSO solutions of the neutralized S-PEDOT (5 × 10−3 wt %) using a DMSO solution with 10 mM tetrabutylammonium bromide (Tokyo Chemical Industry) and 10 mM N,N-diisopropylamine as an eluent at flow rate, column temperature, and injection volume of 0.3 ml min−1, 50°C, and 20 μl, respectively. The UV-vis-NIR spectra of the S-PEDOT water solutions and spin-coated thin films were measured with a spectrophotometer (V-670, Jasco). The viscosity and pH of the S-PEDOT water solutions (1 wt %) were measured at 25°C with a torsional oscillation-type viscometer (Viscomate VM-10A-L, Sekonic) and a pH meter (Laqua F-74S, Horiba), respectively.

Electrical conductivity

The solid films were fabricated by casting the S-PEDOT water solution (1 wt %) onto a glass substrate (35 mm long, 25 mm wide, and 1 mm thick) at 120°C with a moisture analyzer (MOC-120H, Shimadzu) and subsequent heating at 200°C for 30 min in a vacuum. The electrical conductivity of the S-PEDOT films (10 mm2, 7.2 to 10.2 μm thick) was evaluated by the normal four-point method with a Loresta-GP (MCP-T610, Mitsubishi Chemical Analytech), where the thickness of the films was measured with a stylus profilometer (D-100, KLA-Tencor) in dry air.

XRD analysis

The XRD patterns were measured by an x-ray diffractometer (MiniFlex600, Rigaku) equipped with a one-dimensional high-speed detector (D/tex Ultra, Rigaku) at 40 kV and 15 mA. The values of Xc and Dhkl were estimated by the profile fitting method and Scherrer’s equation, respectively, using the integrated x-ray powder diffraction software (PDXL 2, Rigaku).

Scanning probe microscopy

The S-PEDOT water solutions (1 wt %) were spin coated onto n-doped Si wafers (0.5 mm thick, resistivity was 0.02 ohm cm) at 2000 rpm with a spin coater (MS-A150, Mikasa); the coated wafers were then heat treated at 200°C for 10 min in air to stabilize the structure of the S-PEDOT. The thermal stability of the S-PEDOT film was measured in nitrogen at a constant heating rate of 10°C min−1 with thermogravimetry and a differential thermal analyzer (TG-DTA2000S, NETZSCH Japan), where the decomposition temperature (Td) of the S-PEDOT film was ca. 240°C (fig. S7). The cAFM measurements were carried out with a scanning probe microscope (SPM-9600, Shimadzu) equipped with a conductive probe (CONTPt, NanoWorld), where height and current images were measured in contact mode under a bias voltage of −0.1 V.


Supplementary material for this article is available at

Table S1. Oral median lethal dose (LD50) of various sultones in rats.

Table S2. Dispersibility of PEDOT:PSS and solubility of S-PEDOT in various solvents.

Fig. S1. Chemical structure of S-EDOT isomers.

Fig. S2. Flow-through test.

Fig. S3. Optical properties of thin films.

Fig. S4. Possible structure of the S-PEDOT nanocrystal.

Fig. S5. Synthetic route of the S-EDOT monomer.

Fig. S6. Characterization of the S-EDOT monomer.

Fig. S7. TG-DTA curves of the S-PEDOT film.

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.


Acknowledgments: We greatly appreciate H. Shirakawa at the University of Tsukuba for the advice and useful comments on the synthesis of S-PEDOT. Funding: This work was financially supported, in part, by Grant-in-Aid for Scientific Research (B), JSPS KAKENHI grant number 16H04203. Author contributions: H.O. proposed and supervised this research. H.Y. designed and synthesized the S-EDOT monomer and wrote the manuscript. The polymerization and characterization of S-PEDOT were carried out by K.K. K.M. mainly contributed to the analysis of molecular weight by GPC measurements. Competing interests: H.O. and H.Y. are inventors on a patent application related to this work filed by the University of Yamanashi and Tosoh Corporation (no. P2018-123213A, filed on 31 January 2017). The other authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.

Stay Connected to Science Advances

Navigate This Article