Novel highly stable conductive polymer composite PEDOT:DBSA for bioelectronic applications

In this work, a novel conductive polymer composite consisting of poly(3,4-ethylenedioxythiophene) doped with dodecylbenzenesulfonic acid (PEDOT:DBSA) for bioelectronic applications was prepared and optimized. The novel PEDOT:DBSA composite possesses superior biocompatibility toward cell culture and electrical characteristics comparable to the widely used PEDOT:PSS. The cross-linking processes induced by the cross-linker glycidoxypropyltrimethoxysilane (GOPS), which was investigated in detail using Fourier transform Raman spectroscopy and XPS analysis, lead to the excellent long-term stability of PEDOT:DBSA thin films in aqueous solutions, even without treatment at high temperature. The electrical characteristics of PEDOT:DBSA thin films with respect to the level of cross-linking were studied in detail. The conductivity of thin films was significantly improved using sulfuric acid posttreatment. A model transistor device based on PEDOT:DBSA shows typical transistor behavior and suitable electrical properties comparable or superior to those of available conductive polymers in bioelectronics, such as PEDOT:PSS. Based on these properties, the newly developed material is well suited for bioelectronic applications that require long-term contact with living organisms, such as wearable or implantable bioelectronics. A novel polymer material PEDOT:DBSA was prepared using oxidative polymerization and was modified by cross-linker GOPS and H2SO4-posttreatment. This material exhibits biocompatibility toward cell culture comparable to the glass substrate. The cross-linking process resulted in complete stabilization of PEDOT:DBSA thin film in an aqueous environment, whereas such stabilization was achieved even without high-temperature treatment. The model OECT device proved that the proposed material possesses electrical properties comparable to or even better than other organic mixed conductors used for transistors. This all shows a great potential of PEDOT:DBSA for bioelectronics applications.

However, achieving long-term biocompatibility together with suitable electronic and ionic conductivity of such optimized PEDOT:PSS remains challenging. Therefore, much effort has been dedicated to designing novel materials with improved biocompatibility compared to PEDOT:PSS, which would meet all requirements necessary for bioelectronics. Great attention has been devoted to PEDOT-based materials containing biologically active molecules as the counterion, replacing acidic PSS, such as anionic polysaccharides or poly(ethylene glycol) [26][27][28][29][30][31][32][33][34][35][36]. However, preparing stable PEDOT-based thin films of high quality still involves several challenges. Some difficulties arise from the synthesis, as the method most commonly used for such purposes is electropolymerization [26,[37][38][39][40], which still involves problems regarding controllability or film thickness and homogeneity. Thus, attention is focused on finding an optimal method to improve the processability of prepared composites [41].
One reason why PEDOT:PSS is so widely utilized is its water solubility, as this property allows simple and environmentally friendly processing. However, the water solubility is also disadvantageous, because it makes the material easily dispersible in an aqueous environment and thus unsuitable for many applications [42,43]. Therefore, PED-OT:PSS is often stabilized by cross-linking utilizing (3-glycidyloxypropyl) trimethoxysilane (GOPS) [44][45][46]. This film modification technique is well known to prevent redispersion and delamination of polymer thin films [24,47], which is important for the effectiveness and long life of bioelectronic devices. However, it is at the cost of compromised electrical properties of the polymer [44,47]; upon addition of 0.1 v/v% GOPS, the conductivity of PEDOT:PSS drops by one order of magnitude [47]. Moreover, crosslinking requires additional heat treatment (<100°C) [47], restricting its application in the presence of living cells or temperature-sensitive materials. Thus, preparing materials that possess good stability and conductivity at the same time, together with excellent biocompatibility, remains a major challenge. These materials would further extend bioelectronics applications and help researchers prepare more effective and better performing bioelectronic devices.
In this study, we addressed this challenge and developed a novel polymer material consisting of PEDOT doped with DBSA (PEDOT:DBSA). DBSA has already been used in the field of bioelectronics as a surfactant and stabilizer during the synthesis of organic materials [48][49][50][51] or as an additive of PEDOT:PSS due to its positive impact on film processing [52][53][54] and conductivity [55,56], Furthermore, we were interested in whether the cross-linking mechanism due to GOPS is retained for PEDOT:DBSA. Here, we show that its utilization as a PEDOT counterion results in a material that possesses excellent biocompatibility and longterm stability as well as improved electrical characteristics compared to the widely used PEDOT:PSS. A novel approach of PEDOT-based material fabrication was used, as a PEDOT:DBSA dispersion, easily processible into thin films, was prepared by oxidative polymerization. To customize its properties for bioelectronics applications, crosslinking by GOPS and sulfuric acid posttreatment were studied. The applicability of this prepared material in the bioelectronic field was demonstrated using a model transistor device. The results indicated that the PEDOT:DBSA conductive polymer is well suited for bioelectronic applications that require long-term contact with living organisms, such as wearable or implantable bioelectronics, and shows improved substitution for widely used PEDOT:PSS.

Synthesis of PEDOT:DBSA
Dodecylbenzesulfonic acid (DBSA, Abeson K, 96%, Spolchemie Czechia) (50 mmol) was dissolved in 400 ml of water and poured into a 1500 ml flask equipped with a stirrer. Sodium persulfate (Aldrich, 98%) 100 mmol and 0.1 g ferric trichloride hexahydrate dissolved in 400 ml of water were added. A mixture of 1 g of Triton X 100 (nonionic surfactant) and 100 mmol of EDOT (99.8% by HPLC, a product of VUOS Rybitvi, Czechia) was poured into the flask under vigorous stirring. A white milky EDOT emulsion was created. The reaction mixture was diluted to 1000 ml with water and stirred for 2 days under an inert atmosphere at laboratory temperature. A dark blue product was isolated by filtration. The filter cake was washed with 3 l of water to remove unreacted substances, salts and sulfuric acid. The wet filter cake was stirred in 1000 ml of water. The dispersion was treated in a high-pressure homogenizer Panda Plus (GEA Niro Soavi, Italy). The final dispersion had a particle size under the detection limit of the particle analyzer Horiba LA-950 (Horiba Ltd., Japan, detection limit 0.01 µm), pH = 6.5 and dry solid content of 2.0%.

Substrate preparation
Prior to utilization, all used substrates were pretreated by sonication in a NaOH and isopropanol bath for 10 min each. The substrates were then rinsed with deionized water and isopropanol and dried with dry air. Afterward, they were activated by plasma treatment (RPS50+, Ceplant, 15.5 V, 1 min) [57].

Sample preparation
First, colloidal solutions were prepared by mixing the PEDOT:DBSA dispersion with 0 and 5 v/v% cross-linking agent (3-glycidyloxypropyl)trimethoxysilane (GOPS, Sigma-Aldrich, cat. No. 440167). The prepared solutions were sonicated for 10 min to homogenize and were then spin-coated on prepared substrates (substrates used for individual analyses are described below) to form 100 nm thin films. Each solution was used to prepare 2 sets of thin films. The first set of samples was subsequently treated at 140°C for 1 h, and the second was treated at 25°C for 72 h. To perform the sulfuric acid posttreatment, deposited samples were heated for 15 min at 200°C (pristine PEDOT:DBSA) or 100°C (PEDOT:DBSA/GOPS), which was followed by immersion in concentrated H 2 SO 4 for 5 min at 160°C, washing with deionized water, drying with dry air and heating for another 5 min at 160°C. For the needs of the delamination test, 2 sets of samples were prepared by drop-casting 50 µl of the solutions on prepared substrates. The first set was subsequently treated at 140°C for 1 h, and the second set was treated at 25°C for 72 h. The prepared samples were then immersed in deionized water, each in an isolated container. For the 4-point probe method, solutions containing 1, 2.5, 7.5, and 10 v/v% GOPS were prepared utilizing the same procedure as mentioned above and were spin-coated on substrates to form 100 nm thin films [57].

Characterization methods
The film thickness was measured with a DektakXT stylus profilometer (Bruker).
Raman spectra were obtained utilizing silicon wafer substrates by means of Fourier transform Raman spectroscopy (FT-Raman) using a Nicolet TM iS50 Spectrometer (Thermo Fisher Scientific). All measurements were taken at room temperature on the exchangeable Raman module, which is equipped with a 1064 nm diode laser giving a maximum laser power of 500 mW [57].
XPS analyses utilizing indium tin-oxide substrates were carried out with an Axis Ultra DLD spectrometer using a monochromatic Al Kα (hυ = 1486.7 eV) X-ray source operating at 75 W (5 mA, 15 kV). The spectra were obtained using an analysis area of~300 × 700 µm. The Kratos charge neutralizer system was used for all analyses. High-resolution spectra were measured with a step size of 0.1 eV and a 20 eV pass energy. The instrument base pressure during the measurement was consistently at 2·10 -8 Pa. Spectra were analyzed using CasaXPS software (version 2.3.15) and were charge-corrected to the main line of the carbon C 1 s spectral component (C-C, C-H) set to 285.0 eV. A standard Shirley background was used for all sample spectra [57]. X-ray diffraction (XRD) experiments were carried out using Empyrean (Malvern Panalytical) equipped with HighScore Plus software. The measurement was conducted using the Cu anode (λ = 0.154 nm), generator voltage 40 kV and tube current 30 mA in the range of 2θ from 5°to 90°with a step size of 0.013°.
During the delamination test, the absorbance value of the bulk of deionized water in each container containing immersed sample on glass or PEN substrate was measured immediately after sample immersion and then every 24 h for 21 days on a Varian Cary 50 UV-Vis spectrophotometer [57].
To measure the resistance of the prepared samples on glass substrates, a 4-point probe method was performed using a Keithley 2100 6 ½ digit multimeter (Tektronix) [57]. The conductivity value was then calculated based on the sheet resistance and film thickness.
For OECT characterization, glass substrates (20 × 15 × 1 mm) patterned by Au interconnects with a 50 µm channel length and 1 mm channel width were used. The channel was put into direct contact with an electrolyte solution (NaCl; 200 mS) with an immersed Au gate electrode. The drain and gate voltage and current were measured and applied by a Keithley 2602 A instrument [57].

PEDOT:DBSA and PEDOT:DBSA/GOPS characterization
To study the novel polymer composite PEDOT:DBSA and observe changes in its structure caused by cross-linking, Raman spectroscopy was conducted. Special attention was focused on the effect of temperature on the cross-linking activity of dopant GOPS. Figure 1 shows Raman spectra obtained for pristine polymer (black line) and GOPS-doped PEDOT:DBSA treated at 25°C (red line) or 140°C (blue line). The spectrum of the pristine polymer shows the highest band at 1420 cm −1 . This band is related to the conjugation length and, together with the band at approximately 1530 cm −1 , indicates the symmetrical and asymmetric stretching vibrations of C α = C β in PEDOT rings. The band at 1360 cm −1 corresponds to the deformation vibrations of aromatic C β = C β , while symmetrical C α -C α' interring stretching vibrations are demonstrated by the peak near 1260 cm −1 . The same peaks can be observed for both GOPS-doped samples treated at different temperatures. The spectra of these samples are approximately equal, showing redshifted bands located at 1420 and 1260 cm −1 compared to the spectrum of the pristine polymer. The redshift in the spectra of PEDOT:DBSA/GOPS indicates an increase in the effective conjugation length, which is related to the turn of the PEDOT structure into the quinoid form, losing its aromaticity and decreasing the bond order. This observation is in accordance with previously published results, as the same behavior was observed even for cross-linked PEDOT:PSS [42]. Since a redshift was observed for both doped samples, the cross-linking activity of GOPS was induced even spontaneously, and PEDOT:DBSA structure modification was achieved without a high curing temperature.
XPS analysis was performed to characterize the novel polymer composite PEDOT:DBSA and observe the crosslinking mechanism in GOPS-doped samples. Figure 2 shows the S 2p core-level signal of the pristine polymer composite (a) and PEDOT:DBSA/GOPS treated at 25°C (b) or 140°C (c). The obtained spectra show typical 2p 1/2 and 2p 3/2 doublets due to spin-orbit splitting. The signal of the pristine polymer ( Fig. 2a) consists of two major contributions. The lower binding energy peaks (163-166 eV) correspond to the spin-split components of the sulfur atoms of PEDOT [58], and the higher energy peaks (167-171 eV) represent the contribution of the sulfonate groups of DBSA due to their electron density deficit caused by the presence of three electronegative oxygens [41,47]. The DBSA signal may be further fitted with the following components: the higher binding energy doublets (S 2p 3/2 168.8 eV) of sulfonate groups interacting with small ions and the lower binding energy doublets (S 2p 3/2 167.9 eV) related to DBSA molecules interacting with PEDOT [47,59]. The PEDOT signal may also be deconvoluted into two doublets, namely, neutral (S 2p 3/2 163.9 eV) and positively charged sulfur atoms in thiophene units (S 2p 3/2 165.4 eV) [60]. PEDOT peaks exhibit asymmetric tail stretching toward high binding energy, which originates from the presence of positive charges delocalized over several rings [59].
The signals obtained for GOPS-doped samples are approximately equal and show the spin-split doublets at the same positions as in the case of pristine PEDOT:DBSA, see Fig. 2b, c. When these spectra were compared with the one of pristine material, increased intensity of sulfonate group contribution with respect to PEDOT signal was observed. The signals of doped samples showed a slight drop in the binding energy of neutral PEDOT (−0.25 eV), while the doublets of positively charged PEDOT + are located at the same position as in the case of pristine PEDOT:DBSA, unaffected by GOPS molecules. The presence of the cross-linker also did not influence the tail  stretching of PEDOT peaks. Regarding the contribution of the DBSA moiety, a slight redshift of +0.17 eV was observed for both doublets, denoting the decreased electron density of these groups. This can be explained by the reaction of the sulfonate groups of DBSA with GOPS, using the negative delocalized charges of DBSA molecules to form new bonds. Because both sulfonate group doublets show the same shift in binding energy, the new bonds are created not only between GOPS and the surplus DBSA but also DBSA molecules interacting with PEDOT are involved, indicating cross-linking through the whole polymer film. The increased DBSA signal intensity with respect to the PEDOT contribution indicates that such GOPS-DBSA interactions affected the electrostatic attraction between PEDOT and DBSA, promoting the hydrophobic interactions of PEDOT chains (see Fig. 3). This resulted in phase segregation within the film, resulting in the accumulation of DBSA molecules in the surface area. These changes within the composition of the polymer film might explain the decreased binding energy of the neutral PEDOT molecules as it affected its surroundings. The tail stretching of the PEDOT peaks detected in cross-linked samples shows that the material modification did not cause the neutralization of PEDOT segments. Since the spectra detected for both GOPS-doped samples are identical, it shows that the posttreatment temperature does not affect the structure of the resulting films and confirms that crosslinking occurs even at room temperature. A very similar cross-linking mechanism was observed for PEDOT:PSS (see Fig. S1, Supplementary Information). However, in this case, the GOPS bonded only with PSS interacting with small counterions, as the peak corresponding to PSS interacting with PEDOT was not affected by the cross-link. This suggests that, as opposed to PEDOT:DBSA, GOPS is located only in the PSS-rich regions and does not affect the electrostatic interactions between PEDOT and PSS, nor does it cause phase segregation within this material [47].
To further clarify the interaction mechanism between PEDOT:DBSA and GOPS and to analyze the structure of the studied materials, XRD measurements of pristine PED-OT:DBSA and PEDOT:DBSA doped with 5 or 10 v/v % treated at 25 or 140°C were conducted. As shown in Fig. 4, the spectrum of pristine PEDOT:DBSA consists of two major diffraction peaks located at 19°and 24.8°that are ascribed to the π-π stacking of DBSA benzene rings and d 010 spacing of ππ stacking of PEDOT chains, respectively [61][62][63][64]. This indicates that PEDOT-rich and DBSA-rich segments are formed within the PEDOT:DBSA thin film. Upon the addition of GOPS, a single broad peak was detected for all GOPSdoped samples that was further fitted with the two components observed for the pristine material. In contrast to pristine PEDOT:DBSA, cross-linked materials show the prevailing intensity of π-π stacking of DBSA molecules over the intensity of (010) diffraction. The intensity of both fitted peaks became stronger with the increasing amount of GOPS and higher posttreatment temperature, indicating a higher density of ordered crystalline zones with much more distinctive DBSA segregation and improvement in ordering within the PEDOTrich phase. The changes in the spectra detected for PED-OT:DBSA/GOPS treated at room temperature and the doped material treated at high temperature were identical, only to a lesser extent; thus, we can conclude that the used temperature does not affect the character of structural changes induced within the material by cross-linking but facilitates these changes and therefore only influences the degree to which they manifest. The obtained results are in accordance with the results of XPS analysis and confirm that upon the addition of GOPS, the electrostatic interactions between PEDOT and DBSA are weakened; as a result, enhanced phase segregation within the material is attained with highly ordered PEDOT grains that are separated from each other by DBSA-rich phases.

Stability and biocompatibility enhancement
The ability of PEDOT:DBSA films to resist delamination and redispersion in an aqueous environment and the effect of cross-linking on this ability were studied using a delamination test. Again, the effect of the temperature applied during GOPS-doped samples preparation was the focus. PEDOT:PSS served as a reference material.
The delamination test consisted of observation of samples condition when immersed in deionized water. Photos of individual samples deposited on a glass substrate after 21 days of such a test are displayed in Fig. 5a). The film of pristine PEDOT:PSS delaminated as a whole within the first 24 h of the experiment (as has already been reported [65]) and formed a stable free-standing film. Pristine PED-OT:DBSA exhibited continuous delamination of small pieces during the whole experiment. When cross-linked spontaneously, only its middle part delaminated in the larger pieces, see Fig. 5a, while the edge stayed attached during the whole experiment. Delamination was fully prevented for the film of heated PEDOT:DBSA/GOPS, which remained unchanged during the whole experiment. This shows better attachment to the substrate demonstrated by the pristine PEDOT:DBSA film compared with PED-OT:PSS and a significantly positive effect of cross-linking on its ability to resist delamination.
To gain insight into the extent of redispersion of individual materials into the solution, the UV-Vis spectrum was measured regularly. The absorbance values at 224 nm, representing the PEDOT main absorption peak, were collected and examined since these values directly indicate the amount of dispersed material. The evolution of absorbance in individual samples deposited on glass substrates is displayed in Fig. 5b. Values detected for the solution above PEDOT:PSS exhibited a steady increase from approximately 0.5 to 0.8, while both pristine and cross-linked PEDOT:DBSA, heated or not, exhibited stable values during the majority of the experiment. These samples showed a slight increase in absorbance during the first 24 h, but it then remained almost constant for the rest of the experiment. The absorbance values determined for the solution above pristine PEDOT:DBSA were~0.6. Solutions above both samples of doped PEDOT:DBSA exhibited lower absorbance values, i.e.,~0.4 for material cross-linked spontaneously and~0.17 for heated sample. These results showed fast delamination of PEDOT:PSS from the glass substrate and an increasing amount of material released into the solution. This is in accordance with the literature, as it is known that washing this material in water significantly contributes to the reduced film thickness [66][67][68]. Compared with PEDOT:PSS, PED-OT:DBSA showed an improved ability to attach to the substrate and excellent long-term resistance against redispersion. Moreover, film modification by cross-linking significantly contributed to its further stabilization, as it completely prevented film delamination and limited the amount of released material. Since a significant increase in the material concentration in the solution above both GOPS-doped samples was detected only on the first day of the test and then remained unchanged, thoroughly washing PEDOT:DBSA/GOPS should be sufficient to obtain completely stabilized thin films. Notably, the excellent stability in wet conditions was achieved even without the need for a high-temperature treatment. In contrast, heating is necessary to stabilize the PEDOT:PSS thin film [19]. In this case, complete stabilization of the material was observed only when cross-linking was induced by high temperature, while PEDOT:PSS/GOPS treated at room temperature exhibited gradual redispersion (see Fig. S2).
Very similar trends were observed even for samples deposited on the PEN substrate (see Fig. S3). Again, pristine PEDOT:PSS exhibited a steady increase in absorbance values, while PEDOT:DBSA showed stable values during the majority of the experiment. Almost constant values were observed for all cross-linked samples, indicating that the material modification led to the complete stabilization of both studied materials. All studied materials restricted delamination during the whole test except for cross-linked PEDOT:PSS, which delaminated from the substrate during the first hour of the test and formed a free-standing film.
The biocompatibility of pristine and cross-linked PED-OT:DBSA thin layers deposited on glass was evaluated in terms of the cell culture viability of 3T3 fibroblasts. The MTT assay was used for this purpose [11,69]. PED-OT:PSS, plain glass and standard cell culture plastics served as reference materials. The relative viabilities of cell cultures growing on the studied materials are shown in Fig. 6. The value obtained for reference material PEDOT:PSS reached only approximately 40% of the value determined for culture plastics, whereas the viability of cells growing on pristine polymer composite PEDOT:DBSA is comparable with the glass control (approximately 70%). There was a trend toward improved biocompatibility due to GOPSmediated cross-linking, which was promoted by heat treatment since the viability of cells growing on spontaneously cross-linked and heat-treated PEDOT:DBSA/GOPS was approximately 75 and 90%, respectively. The lattermentioned treatment led to a significant improvement in biocompatibility toward PEDOT:PSS. These results confirm that PEDOT:PSS in its pristine state does not provide a suitable environment for living cells, which corresponds with the literature [11][12][13][14][15]. Furthermore, the result indicates that the biocompatibility of pristine PEDOT:DBSA was higher than that of PEDOT:PSS, as was suggested by earlier studies [26,38], especially if GOPS cross-linking is applied. A positive effect of material cross-linking was reported for PEDOT:PSS [42] but not reported for PEDOT:DBSA. The mechanism behind this is attributed mainly to the stabilization of the thin film [42]. Since the biocompatibility of all PEDOT:DBSA samples was higher than that of PED-OT:PSS and similar to that of other organic semiconductors used for bioelectronics, e.g., diketopyrrolopyrrole or (P3HT) [11], PEDOT:DBSA presents great potential for application in this field.

Application in model bioelectronic device
To reveal the applicability of the novel polymer composite PEDOT:DBSA for bioelectronic transistor devices, its electrical properties were investigated using the 4-point probe (4PP) method. Moreover, to study the impact of cross-linking on the conductivity of the material, thin films doped with various concentrations of GOPS were analyzed. Standard culture plastic was used as a control material, and its response was established as 100%. Error bars represent the standard deviation for four individual measurements. Asterisks indicate statistically significant differences (p < 0.05) from PEDOT:PSS. All other differences were not significant It should be emphasized that no other treatment was applied to the studied samples to identify the effect of cross-linking alone and that the conductivity of the studied materials can be increased by using any of the conductivity-enhancing methods. Figure 7 shows the dependence of the material conductivity on the amount of cross-linker in the thin film. The value determined for pristine PEDOT:DBSA is~0.06 S•sq −1 . The presence of 1 v/v% cross-linker in the polymer films resulted in a conductivity drop, reaching approximately onethird of the value obtained for the pristine material. Increasing the concentration of GOPS in the studied films resulted in a further drop in electrical conductivity, which was more than 95% when the material was doped with 10 v/v% cross-linker. Since pristine PEDOT:PSS exhibited under the same conditions an electrical conductivity of~0.1 S•sq −1 , it shows that pristine PEDOT:DBSA possesses suitable electrical properties for utilization in bioelectronic devices. The conductivity of this novel material is improved compared to that of other PEDOT-based materials, as shown in Table 1. However, a distinctly negative impact of cross-linking on the material electrical properties was confirmed. This deteriorating effect of GOPS has already been reported [42,47] and is attributed mainly to the morphology changes in the thin film induced by its cross-linking and to the insulating nature of the siloxane network that is formed upon the reactions of GOPS molecules. As the XPS and XRD analysis showed, the cross-link mechanism is based on reactions of GOPS with sulfonate groups of DBSA, which affects the electrostatic interactions between PEDOT and its counterion, resulting in phase segregation. The PEDOT-rich conducting grains are formed that are electrically insulated from each other by the surrounding siloxane network, which results in decreased conductivity of the cross-linked material. Notably, the negative impact of cross-linking is much more pronounced for PEDOT:PSS, as 1 v/v% of GOPS added into this material resulted in a 99,8% drop in conductivity, from 1 to 0.002 S•cm −1 [47]. We assume that such a difference in the extent of the negative impact of GOPS on these two materials is caused by the different cross-linking mechanisms. In contrast with PED-OT:PSS, GOPS leads to weakened electrostatic interactions between PEDOT and DBSA, which leads to enhanced phase segregation and improved ordering within PEDOT grains, which, opposite to cross-linking, has a positive impact on the conductivity of the material.
To further test the suitability of the polymer composite PEDOT:DBSA for bioelectronic devices, organic electrochemical transistors (OECTs) based on this material, pristine or cross-linked, were prepared as model devices, and their performance was characterized. Sulfuric acid posttreatment was applied to overcome the negative effect of material cross-linking. It is well known that this treatment enhances material conductivity, as it supports PEDOT chain extension and causes the exchange of some of its counterions for HSO 4 ions [70,71]. The transfer curves of OECTs based on the studied materials and their first deviationtransconductance g mare illustrated in Fig. 8. All these curves show typical pinch-off saturation behavior as the devices work in depletion mode, similar to devices based on PEDOT:PSS (see Fig. S4, Supplementary Information). The shape of these curves is not considerably affected by either polymer cross-linking or sulfuric acid posttreatment. To analyze the impact of these modifications on the resultant electrical properties of the device, the product of the electronic mobility and volumetric charge storage capacity (µC*), an important parameter for OECTs [72], was calculated using the following equation: where W, d and L capture the channel geometry, specifically its width, thickness and length, and V th and V G represent the threshold or gate voltage, respectively. The values obtained for individual OECTs are shown in Table 2. For the device based on pristine PEDOT:DBSA, the µC* value was determined to be 14 F•cm −1 •V −1 •s −1 . The cross-linking caused a significant drop in this value (to approximately 20%), regardless of whether the process was induced spontaneously or by heat. Sulfuric acid posttreatment also exhibits a very large impact on the electrical properties of the resulting devices. The µC* value was increased by almost 400% for posttreated pristine PEDOT:DBSA compared with the same material without the posttreatment. An even more pronounced shift of electrical properties was observed for cross-linked PEDOT:DBSA. In this case, the µC* of the posttreated GOPS-doped material was more than 15 times higher than that of the nontreated PEDOT:DBSA/ GOPS and 3 times higher than that of the pristine polymer  Table S1, Supplementary Information). It is worth mentioning that the sulfuric acid treatment of conductive thin films did not affect the biocompatibility toward cell culture (see Fig. S5, Supplementary  Information).
The model transistor device based on the proposed materials showed a typical transistor behavior and characteristics comparable to the device that was previously used in bioelectronic applications [73]; therefore, we assume that PED-OT:DBSA, pristine or cross-linked, is suitable for bioelectronic transistor applications. The H 2 SO 4 posttreated PEDOT:DBSA, pristine or cross-linked, exhibited sufficient electrical properties with regard to these applications, as they are comparable or even better than other organic mixed conductors used for transistors. For example, the µC* of PEDOT:PSS determined at the same experimental setup is Bold entries denote the materials that are analyzed in this study  Table S1, Supplementary Information). Other materials often used in bioelectronics were characterized outside of this study [24,72,74].

Conclusion
This study addresses the challenge of preparing material that simultaneously possesses good stability and conductivity, as well as excellent biocompatibility toward cell culture. Therefore, a novel PEDOT-based material, PEDOT:DBSA, was synthesized using oxidative polymerization. Using this approach, a polymer composite easily processed into thin films was prepared. To improve the long-term stability of the proposed material, cross-linking using the dopant GOPS was applied, and sulfuric acid posttreatment was used to enhance the resultant electrical properties. In this manner, we obtained material exhibiting biocompatibility comparable to the glass substrate; and thus significantly improved over PEDOT:PSS and other materials commonly used in bioelectronics. After cross-linking, PEDOT:DBSA completely resists delamination and redispersion under aqueous conditions. In contrast to PEDOT:PSS, temperature treatment is not necessary to reach such stabilization. The cross-linking process has a negative impact on the electrical conductivity of the material; however, when compared to PEDOT:PSS, this impact is less pronounced and can be easily overcome by the application of sulfuric acid posttreatment, which is not possible with PEDOT:PSS. This phenomenon is attributed to a difference between the cross-linking mechanism in these two materials.
The model OECT device based on this material showed typical transistor behavior comparable with the OECT containing PEDOT:PSS as an active layer. Moreover, the µC* value determined for the proposed material is comparable or even better compared with other organic mixed conductors used for transistors, showing suitable electrical properties of PEDOT:DBSA. This indicates that the proposed polymer composite PEDOT:DBSA is a promising material for bioelectronic transistor applications, especially for those that require long-term contact with living organisms, and a potential substitution for PEDOT:PSS.
Acknowledgements The authors gratefully acknowledge that the work was supported by the Czech Science Funding Open access publishing supported by the National Technical Library in Prague.

Compliance with ethical standards
Conflict of interest The authors declare no competing interests.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.