Bio-based PEDOT: nanocellulose hybrids as efficient hole-transport layers for photoelectrochemical devices

Abstract

Developing sustainable hole-transport materials that can match the performance of conventional poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS) remains a key challenge for environmentally compatible optoelectronic devices. In this work, nanocrystalline cellulose (NCC) is demonstrated as a renewable dopant and stabilizer for PEDOT, forming bio-based hybrids with competitive photoelectrochemical performance. Two crystalline allomorphs, NCC Type I and NCC Type II, were compared as templates for aqueous oxidative polymerization of EDOT, producing stable dispersions of PEDOT nanoparticles (50–100 nm) electrostatically anchored to the NCC surface. Spectroscopic and thermal analyses revealed that the higher ester sulfate content and distinct morphology of NCC-II promoted enhanced polaron stabilization and 10–15% higher PEDOT incorporation compared to NCC-I. FTIR and UV–vis spectroscopy showed more pronounced polaronic bands for PEDOT:NCC-II hybrids, evidencing enhanced charge delocalization and doping. Optimal performance was achieved for the PEDOT:NCC-II (50 : 50) composition, which formed stable, conductive networks and served as an efficient hole-transport layer in poly(3-hexylthiophene)-based photoelectrochemical devices. These bio-based electrodes achieved photocurrent densities above 18 µA cm⁻², matching or exceeding PEDOT:PSS reference, and maintained stable operation over 300 s of cycling (>10 light/dark cycles at 0.3 Hz) with reproducible ON/OFF photoresponse. These findings establish nanocellulose-doped PEDOT as a sustainable alternative for next-generation optoelectronic interfaces.

---

Antonio Domínguez-Alfaro

Antonio Domínguez-Alfaro is a multidisciplinary scientist working at the interface of chemistry, engineering, advanced manufacturing, and biology. He obtained his BSc and MSc degrees in Chemical and Industrial Engineering from the University of Huelva and completed his PhD in Applied Chemistry at the University of the Basque Country in 2021 under the supervision of Prof. David Mecerreyes and Prof. Maurizio Prato. He subsequently conducted postdoctoral research at CIC biomaGUNE with Prof. Aitziber Cortajarena, focusing on the development of conducting protein-based hybrid materials. He later joined the University of Cambridge as a Margarita Salas Fellow, working with Prof. George G. Malliaras on bioelectronic systems. In 2024, he joined the Institute of Microelectronics of Sevilla (IMSE-CNM-CSIC) as a Momentum Fellow in the Neuromorphic Laboratory led by Prof. Bernabé Linares-Barranco. Since April 2026, he holds a Ramón y Cajal Fellowship at IMSE-CNM, CSIC leading a new research line on advanced printed electronics. His research aims to bridge chemistry and materials science with emerging manufacturing technologies, enabling the development of next-generation bioelectronic platforms to address current and future scientific, technological, and societal challenges.

1. Introduction

The rapid expansion of Internet of Things (IoT) devices, wearable electronics, and renewable energy systems has intensified the need for optoelectronic materials that combine high performance with environmental sustainability. Electronic waste already exceeds 45 million tons annually, a figure expected to rise as device lifetimes shorten and global production accelerates.^1,2 Addressing this challenge is key to international sustainability agendas, such as the European Green Deal and United Nations Sustainable Development Goals, which call for renewable, biodegradable materials in advanced technologies.^3–5

Poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS) remains the benchmark hole-transport material for optoelectronic and photoelectrochemical devices. Its tunable conductivity (10⁻²–10³ S cm⁻¹), optical transparency (∼90% at 100 nm film thickness), mechanical flexibility, and environmental stability support its widespread use.⁶ However, the petroleum-derived, non-biodegradable PSS counterion raises long-term environmental and recyclability concerns. The incompatibility of PSS with circular economy principles has motivated the search for bio-based dopants that can maintain the electrical performance of PEDOT while enabling sustainable processing and recyclability.^4,7

Nanocellulose is an attractive nanosized renewable alternative due to the abundance, mechanical strength, and tunable surface functionality of its source material, cellulose. In composites with PEDOT:PSS, cellulose nanofibers (CNFs) and nanocrystalline cellulose (NCC) have produced printable supercapacitors, wearable conductors, and flexible electrodes with enhanced processability and hierarchical organization.^8–13 Sulfated nanocellulose has even been used directly as a dispersant and dopant for PEDOT, enabling aqueous colloidal stability and controlled conductivity.^14,15 Nevertheless, most reported systems rely on physical blending with PEDOT:PSS or employ nanocellulose of undefined crystalline form, ignoring how distinct NCC polymorphs might influence polymer–dopant interactions. NCC, as its parent feedstock, exists in several well-defined crystalline allomorphs, being Type I (NCC-I) and Type II (NCC-II) the most relevant, which differ in chain orientation and ester sulfate surface density.¹⁶ The sulfate ester groups, introduced during the acid hydrolysis synthesis stage, provide negatively charged sites analogous to PSS functionality, with tuneable charge density through allomorph selection.¹⁷ Morphologically, NCC-I appears as needle-like rods with parallel chain alignment, while NCC-II exhibits shorter and more twisted ribbon-like nanostructures with antiparallel chains that generate different electrostatic and interchain hydrogen-bonding environments. These structural differences are expected to affect PEDOT nucleation, its doping efficiency, and charge delocalization, yet their impact on PEDOT polymerization and optoelectronic performance remains unexplored. Jeon et al. reported 57.8 S cm⁻¹ conductivity using sulfuric acid-treated cellulose nanofibers for flexible flame-retardant films,¹⁸ while Xia et al. reported 91 S cm⁻¹via electrostatic potential-enhanced polymerization, obtaining durable thermoelectric performance.¹⁹ Atifi et al. demonstrated 23.8 S cm⁻¹ using NCC, two orders of magnitude above conventional PEDOT:PSS, enabling shear-thinning printable inks.²⁰ However, these studies primarily focus on conductivity optimization for flexible or thermoelectric applications; none of them establishes how nanocellulose allomorphs govern the PEDOT doping or charge transport in energy-conversion devices. This knowledge is particularly relevant for photoelectrochemical (PEC) systems, where the hole-transport layer (HTL) must ensure efficient interfacial charge extraction, optical transparency, and operational stability under illumination.²¹ The central question remains whether rational selection of NCC allomorphs can produce PEDOT hybrids that match or surpass the performance of PEDOT:PSS in PEC applications.

In this work, we present the first systematic comparison of NCC-I and NCC-II as direct dopants for PEDOT in photoelectrochemical HTLs. By exploiting their different morphologies and surface charge densities, we control PEDOT polymerization and hybrid nanostructure to establish a clear structure–property–performance relationship. This study demonstrates that the nanocellulose allomorph choice is not only a sustainability factor, but a tuneable design parameter for optimizing electronic interfaces, advancing the rational development of sustainable, high-performance materials for next-generation optoelectronic devices.

2. Experimental part

2.1 Materials and reagents

The cellulose source employed in this work was microcrystalline cellulose (MCC) in powder form from cotton linters with 20 μm average particle size (Sigma-Aldrich, ref 310697). Sulfuric acid (98%) was acquired from Labkem (Barcelona, Spain). Any use of water reported here corresponds to ultrapure water, obtained from a Siemens Ultra Clear device, with a conductivity of 0.055 μS cm⁻¹. 3,4-Ethylenedioxythiophene (EDOT; 98%), ammonium persulfate ((NH₄)₂S₂O₈), and iron(III) chloride hexahydrate (FeCl₃) was ordered from Fisher Scientific. Regio-regular poly(3-hexylthiophene) with 95% RR and MW > 5000 and tetrahydrofuran were acquired from Sigma-Aldrich (ref 445703) and PANREAC, respectively.

2.2 Methods

Synthesis of NCC. All experiments were carried out starting from 10 g of microcrystalline cellulose (MCC) and followed a procedure that was optimized to end up with a tailored allomorph depending on the selected experimental conditions.²² In a typical experiment, the MCC was dispersed in 45 mL of ultrapure water by sonication in an ultrasonic bath (45 kHz) for 10 minutes. The dispersion was cooled to 0 °C with stirring until the medium's temperature stabilized. Subsequently, 45 mL of H₂SO₄ (98%) was added dropwise. Once the acid addition was complete, the next step determined the crystalline type to be generated, so two different treatment methods were used. To obtain NCC-I, the reaction medium was heated to 70 °C for 10 minutes, whereas for NCC-II synthesis, the conditions were set at 27 °C for at least 1 hour. After this process, the reaction crude was added to 1 L of fridge-cold ultrapure water and left to rest overnight at 4 °C. The sediment was separated by decantation and dialyzed using SpectraPor1 regenerated cellulose dialysis membranes from Spectrum Labs, with a molecular weight cut-off of 6–8 kDa. The dialysis water was constantly replaced until a neutral pH was reached. The dialyzed suspension was divided into 45 mL aliquots, which were centrifuged at 16 000 rcf for 1 minute, collecting the supernatant. The remaining pellet was redispersed in 45 mL of water and subjected to the same centrifugation conditions, repeating the process until the supernatants were colourless. The resulting dispersions had a concentration of approximately 2–3 mg mL⁻¹, and the conversion yield from MCC to NCC was in the range of 10–20%. Unless otherwise specified, the term NCCx refers generically to either allomorph.

Synthesis of PEDOT:NCCx. All the dispersions were prepared at a total solid content of 1.3 wt%. Firstly, 10 mL of a nanocellulose (NCC-I or NCC-II) dispersion, containing 10, 25, or 50 wt% of NCCx in MilliQ water, was sonicated until a homogeneous white colloid was obtained. Subsequently, ammonium persulfate (APS) (1.5 equivalents relative to the EDOT monomer) and a catalytic amount of FeCl₃ were added to the solution. The mixture was stirred for 15 min at room temperature. Following this, the EDOT monomer was added to reach final EDOT:NCCx weight ratios of 90 : 10, 75 : 25, and 50 : 50, and the reaction mixture was vigorously stirred (≈800 rpm). After 2 hours, the initially white dispersion gradually turned blueish, indicating the onset of oxidative polymerization. After 48 hours, the dispersion developed a deep-blue colour, characteristic of fully formed PEDOT. The resulting mixture was dialyzed using a 1000 kDa molecular weight cut-off membrane, with the water being changed three times daily over the course of two days. Finally, the blue aqueous solution was collected and freeze-dried, yielding a dark solid with a lightweight, foam-like texture.

P3HT nanoparticle synthesis. The P3HT nanoparticles were synthesized accordingly to a previous report method.^23,24 Firstly, P3HT was dissolved in THF at a concentration of 1 mg mL⁻¹ and magnetically stirred overnight at room temperature. The polymer solution (2 mL) was then injected into an aqueous solution under vigorous stirring (2 mg in 10 mL). Next, to increase the dispersibility, a tip sonicator was used for 3 min. The remaining THF was evaporated via a vacuum evaporator operating at 35 °C. After filtration, the solution was used for further experiments.

Thermogravimetric analysis (TGA). Thermogravimetric analyses were performed using a TGA Discovery (TA Instruments) under nitrogen, with a flow rate of 25 mL min⁻¹. The samples were equilibrated at 100 °C for 20 min and then heated at a rate of 10 °C min⁻¹ in the range from 100 °C to 800 °C.

Fourier transform infrared spectroscopy (FTIR). FTIR spectra were recorded using a Thermo Scientific model Nicolet 6700 FT-IR spectrometer, and KBr pellets in solid-state. The spectrum was recorded after applying ten to twenty scans in transmission mode. For the formation of the pellets, 1 mg of the PEDOT:NCCx composites was diluted with 1500 mg of KBr. The components were mixed in a mortar until homogenized. Finally, the resulting mixture was compressed, forming a slightly black pellet of ≈1 mm in thickness.

Transmission electron microscopy (TEM). The interaction between PEDOT and NCCx was evaluated using an JEOL JEM-2100F model EM-20014, which features a 200 kV Schottky field-emission gun (FEG). The copper grid was put in contact with a drop of 20 µl of 0.1 mg mL⁻¹ of PEDOT:NCCx or NCCx for 30 seconds, subsequently the grid was put in contact with another 20 µl single drop of the Milli-Q water for another 30 seconds. Finally, the grid was dried under air conditions at room temperature for several hours.

Scanning electron microscopy (SEM). The morphology of the was analysed by JEOL JSM-6490LV at 15 kV, running in a point-by-point scanning mode. The samples were first dispersed at 0.1 mg mL⁻¹ and then deposited by drop casting on an indium tin oxide/glass (ITO) substrate as a conductive flat surface. The sample was sputtered with Au for 60 seconds and then mounted on an aluminium holder with double-sided carbon tape.

Photoelectrode preparation. Photoelectrodes were prepared by sequential spray deposition of aqueous dispersions onto fluorine-doped tin oxide (FTO) coated glass substrates. Initially, a thin and uniform layer of PEDOT:NCC was deposited onto cleaned FTO substrates using a spray-coating system (Nadetech ND-SP Ultrasonic PRO) operated at a flow rate of 42 mL h⁻¹ to cover an area of 1 cm² (Fig. S14A). The spray nozzle was placed 60 mm above the substrate, and the substrate temperature was maintained at 80 °C using a hot plate. Subsequently, a layer of P3HT nanoparticles dispersed in water was deposited on top of the PEDOT:NCC layer using the same spray-coating conditions. The resulting FTO/PEDOT:NCC/P3HT-NP multilayer structures were used as photoactive electrodes for further optoelectronic characterization.

Photoelectrochemical measurements. The as-prepared photoelectrodes were tested in a three-electrode electrochemical cell (Fig. S14B), using 0.1 M NaClO₄ in dry acetonitrile as the electrolyte. The spray-coated photoelectrodes functioned as working electrodes (WE), with an Ag/AgCl (calibrated at 0.19 V vs. NHE) and a carbon rod used as the reference and counter electrodes, respectively. The WE were illuminated through a quartz window by a 150 W Xenon arc lamp with a power density of 300 mW cm⁻² (LOT-Oriel GmbH, Germany) under a nitrogen atmosphere. Continuous and on–off cyclic voltammetry (CV) was performed at a scan rate of 0.02 V s⁻¹ with the light beam periodically interrupted at a frequency of ≈0.3 Hz. Chronoamperometric experiments were carried out at −0.5 V vs. Ag/AgCl, with 15 s long light–dark cycles, for 300 s.

3. Results and discussion

3.1 Formation and characterization of PEDOT:NCC Hybrids

Understanding how nanocellulose allomorphs influence PEDOT polymerization and doping is essential to establishing the structure–property relationships governing the hybrid's performance. The PEDOT:NCCx hybrids were synthesized via aqueous oxidative polymerization, varying the NCC type (I or II) and PEDOT-to-NCC feed ratio (90 : 10, 75 : 25, and 50 : 50) (Fig. 1A). The polymerization followed the established oxidative mechanism in which Fe³⁺ oxidizes EDOT monomers to radical cations (EDOT˙⁺), initiating dimerization and chain propagation to form conjugated PEDOT backbones.^7,25–27 Concurrently, persulfate generates sulfate radical anions (SO₄˙⁻) that oxidize the polymer into its positively charged state (PEDOT⁺). The negatively charged sulfate ester groups on the NCC surface electrostatically stabilize PEDOT⁺, preventing aggregation and promoting colloidal stability. The dispersions remained readily re-dispersible after freeze-drying, confirming the stabilization role of the sulfated nanocellulose. The reaction achieved >95% monomer conversion within 48 hours, as evidenced by the characteristic colour transition from white to deep blue associated with fully oxidized PEDOT.^7,28 Both NCC-I and NCC-II supported efficient polymerization, although subsequent characterization revealed differences in their doping efficiency and polymer-substrate interactions.

Fig. 1 (A) Scheme of the chemical oxidative polymerization of EDOT in presence of NCCx. The reaction is carried out during 2 days, using ammonium persulfate as oxidant and a catalytic amount of iron chloride, (B) FTIR of all compositions studied of PEDOT:NCC-II using KBr pellets and diluted 1 mg of sample in 1000 mg of salt, (C) UV-vis of the different PEDOT:NCCII dispersions obtained after dialysis diluted at 25 v/v% in MilliQ water, with inset showing the deep-blue colour characteristic of fully polymerised PEDOT solution, TGA plots of (D) derivative and (E) relative weight obtained under nitrogen of PEDOT:NCC-II and NCC-II and PEDOT:Cl as control.

FTIR spectroscopy confirmed the coexistence of characteristic PEDOT and NCC features in all hybrid compositions (Fig. 1B and S1). The characteristic PEDOT bands at ≈1515 cm⁻¹ and ≈1340 cm⁻¹, corresponding to the asymmetric and symmetric C=C stretching vibrations of the thiophene ring, respectively, and the C–S–C deformation at ≈840 cm⁻¹ are present in all hybrids. The intensity of these signals increased with PEDOT content, confirming progressive polymer formation within the hybrids. Vibrations associated with nanocellulose were also evident: broad O–H stretching near 3400 cm⁻¹, C–H stretching around 2900 cm⁻¹, and C–O/C–O–C vibrations in the 1060–1160 cm⁻¹ region. The overlap between the ether C–O–C bands of NCC and PEDOT indicates a closer molecular association between the two components. A pronounced difference between the two allomorphs emerged in the region near ≈1200 cm⁻¹. NCC-II based hybrids displayed a broader and more intense band in this range that shifted towards lower wavenumbers as the NCC-II content increased. This behavior is consistent with stronger electrostatic or hydrogen-bonding interactions between the more sulfated NCC-II surface and the oxidized PEDOT⁺ chains. Concurrently, the sulfate-related peak at ≈850 cm⁻¹ decreased in intensity with higher NCC-II content, further supporting a greater degree of PEDOT doping. Conversely, the NCC-I hybrids showed no significant spectral evolution with composition, indicating a comparatively uniform doping level irrespective of the NCC-I content.

UV-vis spectroscopy further revealed how the nanocellulose allomorph and PEDOT loading influence the electronic structure of the hybrids. All hybrids exhibited the characteristic π–π* transition of PEDOT below 400 nm, confirming successful polymerization, together with broad near-infrared bands in the 700–1000 nm region associated with polaronic and bipolaronic transitions²⁹ (Fig. 1C). Distinct spectral differences were observed between NCC-I and NCC-II hybrids. In NCC-I systems, the dominant polaron band appeared near 800 nm and shifted only slightly to higher wavelengths as the NCC-I content increased, suggesting a relatively constant doping level across compositions (Fig. S2). In contrast, NCC-II hybrids showed a clear evolution of this band with the composition: the intensity of the 700–1000 nm polaron band decreased progressively as the NCC-II content increased from 10% to 50%. This is coupled with the emergence of a shoulder at ≈675 nm for the PEDOT:NCC-II (50 : 50) sample, which points towards the formation of bipolaronic states that are not observed for the other two compositions. These two features thus indicate a higher degree of doping in the NCC-II hybrids because of the highly oxidized polaronic states, which increase as more cellulose is introduced in the hybrid.^29,30

These spectral features indicate that NCC-II promotes a higher oxidation level and stronger dopant interaction within the hybrid. The differences in electronic structure observed here suggest that variations in nanocellulose surface chemistry and morphology play a decisive role in PEDOT doping behaviour.

Thermogravimetric analysis (TGA) under nitrogen was used to assess the composition and thermal behavior of the PEDOT:NCC hybrids (Fig. 1D, E and S3). Pristine NCC-II showed the expected two-step degradation pattern: (i) a mass loss below 250 °C attributed to physiosorbed and chemisorbed water, and (ii) a major decomposition between 270 and 500 °C corresponding to glycosidic bond cleavage and aromatization processes.¹⁶ The PEDOT:Cl reference displayed an onset of degradation at ≈326 °C and a maximum decomposition rate near ≈362 °C, leaving approximately 77% residual mass at 343 °C. This intermediate temperature (343 °C) was selected to estimate the PEDOT fraction in the hybrids, since it lies between the onset and maximum degradation of PEDOT, while NCC degradation remains minimal.

At all feed ratios, the NCC-II series incorporated a larger proportion of PEDOT than the corresponding NCC-I samples (Table 1). This higher incorporation efficiency reflects the denser distribution of sulfate ester groups on NCC-II, which provides additional nucleation sites and electrostatic stabilization for PEDOT⁺ during oxidative polymerization. A mass loss feature near 272 °C, common to all hybrids, corresponds to the disruption of the PEDOT-NCC ionic interactions and confirms that the polymer–dopant interface is governed by ionic rather than covalent bonding. These results demonstrate that NCC-II promotes more efficient polymer growth and yields hybrids with higher PEDOT content, consistent with the enhanced doping behavior identified by FTIR and UV–vis spectroscopy, and further supporting the role of the chemical surface of nanocellulose in stabilizing the hybrid interface. Transmission electron microscopy (TEM) revealed morphology differences related to NCC allomorph and PEDOT loading. Pristine NCC-I consisted of straight, needle-like rods approximately 200–300 nm in length and 5–10 nm in width, while NCC-II formed shorter, ribbon-like fibrils of 50–100 nm length and 15–20 nm width (Fig. S4). The PEDOT:Cl reference showed large, heterogeneous aggregates roughly 300 nm in diameter (Fig. S5). The PEDOT:NCC hybrids exhibited clear evidence of polymerization occurring on or near the cellulose surface, with PEDOT domains intimately associated with the nanocellulose fibrils. The resulting hybrids consisted of well-dispersed PEDOT nanoparticles of 50–100 nm anchored along the NCC surfaces (Fig. S6). Increasing the PEDOT fraction led to progressively greater surface coverage. In the 50 : 50 hybrids, PEDOT formed discrete clusters that partially decorated the nanocellulose fibrils, leaving regions of exposed NCC, and maintaining intimate polymer-NCC contact. At higher PEDOT ratios (90 : 10), continuous polymer enveloped the fibrillar network (Fig. 2). This more extensive coating suggests that portions of the PEDOT phase were no longer directly interfaced with the NCC surface. NCC-II-based hybrids consistently displayed denser and more homogeneous PEDOT coverage than NCC-I, consistent with the higher PEDOT incorporation and stronger ionic interactions between PEDOT⁺ and the NCC-II surface, evidenced by the TGA and UV–vis analyses.

Fig. 2 TEM images for 50 : 50 and 90 : 10 PEDOT:NCC composites using type-I and II nanocellulose (scale bar: 200 nm). Amorphous PEDOT nanoparticles highlighted in red; surface of NCC depicted in yellow.

Table 1 Chemical oxidative polymerization of EDOT in the presence of NCCx, in water at room temperature for 2 days

	EDOT (wt%)	NCC (wt%)	PEDOT^a (wt%)
a PEDOT (wt%) is calculated by TGA at 343 °C.
Type I	90	10	65
	75	25	50
	50	50	45
Type II	90	10	70
	75	25	62
	50	50	52

---

Scanning electron microscopy (SEM) corroborated these nanoscale observations (Fig. S7). All films exhibited interconnected PEDOT networks entangled with residual nanocellulose features, though surface morphology varied with composition. At lower PEDOT loading (50 : 50), cellulose fibrils remained partially visible, whereas at higher PEDOT content, smoother and more compact coatings with fewer exposed fibrils were observed. These morphological observations indicate that film continuity and surface texture evolve with hybrid composition, reflecting the balance between polymer coverage and accessible nanocellulose surface area.

Overall, the microscopy results show that the surface chemistry of nanocellulose governs PEDOT nucleation, distribution, and film continuity within the hybrids. The NCC-II surface provides a more effective template for uniform polymer growth, promoting intimate contact between PEDOT⁺ and the cellulose dopant sites. At intermediate PEDOT loading (50 : 50), the hybrids retain an optimal balance between polymer coverage and exposure nanocellulose surface area, favouring stronger ionic interactions. PEDOT content has a direct impact on the electrical resistivity, showing a clear dependence on both composition and NCC polymorph (Table S1). As a reference, PEDOT:PSS films exhibit sheet resistances that depend on the number of spray-coated layers, ranging from ≈1.5 to 5 kΩ for 30 and 5 spray-coated layers, respectively. In intermediate compositions (50 : 50 and 75 : 25 PEDOT:NCC), the hybrids generally display moderate sheet resistances, indicating that higher nanocellulose content partially disrupts the conductive PEDOT pathways. In contrast, the 90 : 10 PEDOT:NCC systems reach the lowest sheet resistances among the hybrids, comparable to PEDOT:PSS, with values of 2.5 kΩ for 90 : 10 PEDOT:NCCI and 2.3 kΩ for 90 : 10 PEDOT:NCCII. This suggests that moderate NCC incorporation enhances charge transport by promoting a more interconnected PEDOT network while preserving conductive pathways; overall, the electrical response is governed by PEDOT percolation and film homogeneity rather than NCC polymorph.

3.2 Photoelectrochemical performance

The hole-transport functionality of the PEDOT:NCC hybrids was evaluated in photoelectrochemical (PEC) cells using poly(3-hexylthiophene) (P3HT) nanoparticles as the photoactive layer (Fig. 3A). P3HT nanoparticles, 50–100 nm in diameter (Fig. S8), were synthesized via mini-emulsion polymerization. This approach is known to enhance molecular ordering and promote face-to-face π–π stacking and H-aggregate character that facilitates efficient exciton dissociation and hole mobility.³¹ These features make P3HT an excellent model system for probing interfacial charge extraction processes in hybrid photoelectrodes. The PEC devices were assembled on FTO substrates sequentially spray-coated with the PEDOT:NCC interlayer and the P3HT layer, while PEDOT:PSS and bare P3HT electrodes served as controls. In this configuration, photogenerated holes migrate from the P3HT absorber through the PEDOT:NCC hybrid to the FTO anode, while electrons are transferred to the electrolyte to drive cathodic reactions.^32,33 This configuration allows the interfacial properties of the bio-based PEDOT:NCC layers to be directly correlated with hole extraction efficiency in the photoelectrochemical device. To ensure a fair comparison between devices, the thickness of the P3HT active layer was kept constant across all samples by maintaining identical deposition parameters, including solution concentration, spray-coating conditions, and processing protocol. The PEDOT:NCC hybrid interfacial layers were deposited by spray coating 15 layers (Fig. 3B), which resulted in comparable film coverage and morphology across all conditions, as shown in Fig. S9. Gravimetric analysis revealed a similar amount of deposited material for all samples under these conditions (0.38 ± 0.1 mg), Fig. S10.

Fig. 3 (A) Automatic spray-coating of active layers on ITO substrates, (B) Photoelectrochemical set-up consisting of a three-electrode cell with the layer-by-layer assembled as working electrode (WE), graphite counter electrode (CE) and Ag/AgCl reference electrode. Measurements were performed in 0.1 M NaClO₄ at a scan rate was 0.02 V s⁻¹. During illumination experiments, films were periodically irradiated with a 150 W xenon lamp at 0.3 Hz. Inset: schematic representation of the multilayer architecture comprising the hole-transport layer (PEDOT:NCCx), the photoactive P3HT layer, and the ITO collector.

Cyclic voltammetry performed under dark conditions (Fig. S11) showed the characteristic oxidation peaks of P3HT near 0.51 V and 0.87 V, corresponding to electrochemical doping and polaron formation. The capacitance at low potentials (from 0.2 to −0.3 V), increased with higher NCC content in the PEDOT:NCCx hybrids, with NCC-II compositions yielding significantly higher capacitances than NCC-I. This behavior indicates a larger amount of electrochemically active PEDOT within the PEDOT:NCC-II hybrids, in good agreement with the higher doping level demonstrated from FTIR and UV-vis analyses (Fig. 1B and C). Incorporating a PEDOT-based interlayer substantially enhanced the cathodic photocurrent compared to bare P3HT electrodes, validating the improved hole extraction ability of the hybrids (Fig. 4A–B). The PEDOT:PSS reference displayed the expected enhancement, while both PEDOT:NCC series achieved comparable or superior photocurrent responses. Among these, the PEDOT:NCC-II (50 : 50) electrode delivered the highest photocurrent density, reaching 18.4 µA cm⁻² at −0.5 V, surpassing the 15.1 µA cm⁻² recorded for the PEDOT:PSS reference (Fig. S12). The 50 : 50 ratio consistently outperformed the 75 : 25 and 90 : 10 compositions for both NCC allotropes, underscoring the critical role of interfacial doping balance in governing PEC performance. Across all feed ratios, NCC-II hybrids exhibited higher photocurrent densities than NCC-I analogues, which is attributed to the higher sulfate density and ribbon-like morphology of NCC-II that promote stronger PEDOT doping and improved contact with the P3HT nanoparticles. SEM analysis (Fig. S13) supports this interpretation, showing that the shorter, more uniform NCC-II structures form intimate interfacial contact with the P3HT matrix, likely facilitating efficient hole extraction. The resulting performance hierarchy follows the order: P3HT/PEDOT:NCC-II (50 : 50) > P3HT/PEDOT:NCC-I (50 : 50) ≥ P3HT/PEDOT:PSS > P3HT/PEDOT:NCC (75 : 25) > bare P3HT.

Fig. 4 Cyclic voltammetry of P3HT, P3HT/PEDOT:PSS control and P3HT/PEDOT:NCCx (50 : 50) hybrid films with both NCC (A) type I or (B) type-II, measured at a scan rate of 20 mV s⁻¹. (C) Chronoamperometric measurement of the as-prepared films. These were intermittently illuminated by a 150 W Xenon arc lamp at a frequency of 0.3 Hz under an applied potential of −0.5 V.

Chronoamperometric stability tests under repeated 15 s light/dark cycles (0.3 Hz) demonstrated robust operational behavior or all hybrid electrodes (Fig. 4C). PEDOT:NCC-II (50 : 50) sample retained stable photocurrents above 9.5 µA cm⁻² for over 300 s, maintaining sharp and reproducible ON/OFF responses. The slight initial current decay followed by stabilization is attributed to capacitive charge redistribution within the NCC matrix, consistent with the enhanced capacitive features observed in the CV profiles. Both PEDOT:NCC series delivered higher and more stable photocurrents than bare P3HT and the PEDOT:PSS control, confirming their effectiveness as hole-transport layers. These results highlight that precise control of the PEDOT:NCC composition and nanocellulose allomorph enables bio-based hybrids with reliable photoelectrochemical performance comparable to conventional PEDOT:PSS systems. The combination of strong ionic interactions, effective doping, and stable charge transport across the hybrid interface reinforces the durability of these materials during continuous illumination.

The combined structural, spectroscopic, and electrochemical analyses reveal a coherent understanding of how nanocellulose allomorph and surface chemistry govern the evolution of PEDOT:NCC hybrid properties and their performance as hole-transport layers. At the molecular level, the negatively charged sulfate groups of NCC electrostatically stabilize oxidized PEDOT⁺ chains, leading to colloidally stable dispersions and uniform nanoparticle formation. The higher sulfate density and ribbon-like morphology of NCC-II promote more homogeneous PEDOT nucleation and improved charge delocalization as a result of a higher degree of doping, which is reflected in a more prominent C–S band at 850 cm⁻¹, red-shifted NIR polaron bands, and slightly higher PEDOT incorporation (≈10–15%) compared with NCC-I. TEM and SEM imaging demonstrate that these molecular-level interactions translate into denser and more continuous PEDOT coatings for NCC-II hybrids, yielding films with superior connectivity and uniformity. Nevertheless, both NCC allomorphs enable well-organized hybrid architectures in which the polymer remains effectively anchored to the NCC framework, maintaining stable electronic networks even at high PEDOT loadings. The resulting film morphology directly influences device-level behavior. The increased surface roughness of the PEDOT:NCC-I and PEDOT:NCC-II hybrids at the 50 : 50 composition enhances interfacial contact with the P3HT layer, producing a three-dimensional junction morphology reminiscent of bulk heterojunction architectures.³⁴ In photoelectrochemical configurations, hybrids with intermediate NCC content (50 : 50) consistently exhibit stronger and more stable photocurrents than P3HT controls, with operational stability maintained over repeated light/dark cycling. While NCC-II promotes greater polaron stabilization and morphological uniformity, both NCC-I and NCC-II hybrids demonstrate effective charge extraction and performance once incorporated into optimized device architectures. These results confirm that nanocellulose, irrespective of allomorph, can successfully replace PSS as a bio-based dopant for PEDOT. Altogether, these results establish clear structure–property–performance correlations: nanocellulose sulfate density controls polymer stabilization and dispersion, morphology dictates film architecture and roughness, and both collectively determine charge extraction efficiency and operational stability. This integrated understanding provides a rational framework for the design of sustainable, high-performance conductive polymers based on renewable nanocellulose templates for interfacial applications in photoelectrochemical and related optoelectronic devices.

4. Conclusions

This work demonstrates that nanocrystalline cellulose can serve as a green dopant for PEDOT, effectively replacing polystyrene sulfonate in hole-transport layers without compromising device performance. Systematic comparison between the two main nanocellulose allomorphs, NCC-I and NCC-II, establishes a clear structure–property–performance relationship that links ester sulfate density on surface and morphology to polymerization behavior, charge delocalization, and interfacial transport efficiency. NCC-II, characterized by its higher sulfate density and ribbon-like morphology, consistently promoted greater PEDOT incorporation (approximately 10–15% higher than NCC-I), stronger electrostatic stabilization of PEDOT⁺ species, and more continuous film formation. These structural and chemical advantages translated directly into improved photoelectrochemical performance: the PEDOT:NCC-II (50 : 50) hybrid achieved a photocurrent density of 18.4 µA cm⁻², surpassing that of PEDOT:PSS, while maintaining stable operation over extended light/dark cycling. Mechanistically, the enhanced interfacial charge transport in PEDOT:NCC-II-based electrodes arises from the combined effects of the chemical surface and short, ribbon-like architecture of NCC-II. These features facilitate homogeneous PEDOT nucleation, efficient polaron delocalization associated with a higher degree of doping, and intimate coupling with the P3HT photoactive layer. Overall, these results show that optimized PEDOT:NCC hybrids combine high photoelectrochemical performance and operational robustness with environmental friendliness and aqueous processability. By correlating nanocellulose crystalline type and surface chemistry with PEDOT doping efficiency and charge transport behavior, this study demonstrates how interface chemistry can be rationally tuned to control device performance without relying on synthetic additives. Nanocellulose-doped PEDOT hybrids thus emerge as viable, scalable, and environmentally sustainable alternatives to PEDOT:PSS for next-generation photoelectrochemical and organic electronic applications, expanding the horizons of the wooden age of electronics.³⁵

Conflicts of interest

There are no conflicts to declare.

Data availability

The authors declare that the main data supporting the findings of this study are available within the article and supplementary information (SI). Supplementary information: additional TEM and SEM images for morphology characterization, UV-vis, TGA, and FTIR for structural material characterization. See DOI: https://doi.org/10.1039/d5nr04688b.

Acknowledgements

Funding from Spanish MCIN/AEI/10.13039/501100011033 and “ERDF A way of making Europe” under project grants PID2022-139671OB-I00 and PID2023-147116OB-I00, as well as from the Gobierno de Aragón (DGA) under projects in priority lines and of a multidisciplinary nature (project grant PROY_T41_24) and T03_23R (Grupos de Investigación Reconocidos), is acknowledged. A. D.-A and B. L.-B. acknowledges the great momentum, hired under the Generation D initiative, promoted by Red.es, an organisation attached to the Ministry for Digital Transformation and the Civil Service, for the attraction and retention of talent through grants and training contracts, financed by the Recovery, Transformation and Resilience Plan through the European Union's Next Generation funds ID: MMT24-IMSE,CNM-02.

References

1. Y. Li, S. Chen, H. Yan, H. Jiang, J. Luo, C. Zhang, Y. Pang and Y. Tan, Chem. Eng. J., 2023, 468, 143572.
2. D. H. P. Kang, M. Chen and O. A. Ogunseitan, Environ. Sci. Technol., 2013, 47, 5495–5503.
3. V. R. Feig, H. Tran and Z. Bao, ACS Cent. Sci., 2018, 4, 337–348.
4. R. Rai and D. Mantione, J. Mater. Chem. C, 2023, 11, 11803–11813.
5. J. A. Chiong, H. Tran, Y. Lin, Y. Zheng and Z. Bao, Adv. Sci., 2021, 8, 2101233.
6. J. Rivnay, S. Inal, B. A. Collins, M. Sessolo, E. Stavrinidou, X. Strakosas, C. Tassone, D. M. Delongchamp and G. G. Malliaras, Nat. Commun., 2016, 7, 11287.
7. D. Mantione, I. del Agua, W. Schaafsma, J. Diez-Garcia, B. Castro, H. Sardon and D. Mecerreyes, Macromol. Biosci., 2016, 16, 1227–1238.
8. K. M. Alam, P. Kar, U. K. Thakur, R. Kisslinger, N. Mahdi, A. Mohammadpour, P. A. Baheti, P. Kumar and K. Shankar, J. Mater. Sci.: Mater. Electron., 2019, 30, 1390–1399.
9. F. Jiao, J. Edberg, D. Zhao, S. Puzinas, Z. U. Khan, P. Mäkie, A. Naderi, T. Lindström, M. Odén, I. Engquist, M. Berggren and X. Crispin, Adv. Sustainable Syst., 2018, 2, 1700121.
10. K. Jain, Z. Wang, L. D. Garma, E. Engel, G. C. Ciftci, C. Fager, P. A. Larsson and L. Wågberg, Appl. Mater. Today, 2023, 30, 101703.
11. G. Méhes, M. Vagin, M. Y. Mulla, H. Granberg, C. Che, V. Beni, X. Crispin, M. Berggren, E. Stavrinidou and D. T. Simon, Adv. Sustainable Syst., 2020, 4, 1900100.
12. R. Brooke, M. Lay, K. Jain, H. Francon, M. G. Say, D. Belaineh, X. Wang, K. M. O. Håkansson, L. Wågberg, I. Engquist, J. Edberg and M. Berggren, Polym. Rev., 2023, 63, 437–477.
13. S. Zhou, Z. Qiu, M. Strømme and Z. Wang, Adv. Funct. Mater., 2020, 30, 2005757.
14. X. Feng, X. Wang, M. Wang, S. Zhou, C. Dang, C. Zhang, Y. Chen and H. Qi, Chem. Eng. J., 2021, 418, 129533.
15. K. Matura, R. D'Orsi, L. Spagnuolo, F. Mayr, M. Cobet, C. Putz, A. Operamolla and S. Tekoglu, J. Mater. Chem. C, 2024, 12, 16701–16713.
16. J. M. González-Domínguez, A. Ansón-Casaos, L. Grasa, L. Abenia, A. Salvador, E. Colom, J. E. Mesonero, J. E. García-Bordejé, A. M. Benito and W. K. Maser, Biomacromolecules, 2019, 20, 3147–3160.
17. H. Kargarzadeh, M. Mariano, D. Gopakumar, I. Ahmad, S. Thomas, A. Dufresne, J. Huang and N. Lin, Cellulose, 2018, 25, 2151–2189.
18. S. H. Jeon, B. Ozlu and B. S. Shim, Biomacromolecules, 2024, 25, 644–654.
19. Y. Xia, J. Li, Z. Ji, K. Zhou, Y. Zhang, Y. Liu, S. W. Tsang, K. W. Wong, Q. Wang, W.-J. Wang, A. Cabot, X. Yang and K. H. Lim, Adv. Mater., 2025, 37, 2506338.
20. S. Atifi, M.-N. Mirvakili and W. Y. Hamad, ACS Appl. Polym. Mater., 2022, 4, 5626–5637.
21. J. Liu, M. Durstock and L. Dai, Energy Environ. Sci., 2014, 7, 1297–1306.
22. V. Calvo, M. Á. Álvarez Sánchez, L. Güemes, C. Martínez-Barón, S. Baúlde, A. Criado, J. M. González-Domínguez, W. K. Maser and A. M. Benito, ACS Macro Lett., 2023, 12, 152–158.
23. E. Colom, J. Hernández-Ferrer, A. Galán-González, A. Ansón-Casaos, M. Navarro-Rodríguez, E. Palacios-Lidón, J. Colchero, J. Padilla, A. Urbina, R. Arenal, A. M. Benito and W. K. Maser, Chem. Mater., 2023, 35, 3522–3531.
24. E. Istif, J. Hernández-Ferrer, E. P. Urriolabeitia, A. Stergiou, N. Tagmatarchis, G. Fratta, M. J. Large, A. B. Dalton, A. M. Benito and W. K. Maser, Adv. Funct. Mater., 2018, 28, 1707548.
25. A. Dominguez-Alfaro, N. Casado, M. Fernandez, A. Garcia-Esnaola, J. Calvo, D. Mantione, M. R. Calvo and A. L. Cortajarena, Small, 2024, 20, 2307536.
26. C.-Y. Lo, Y. Wu, E. Awuyah, D. Meli, D. M. Nguyen, R. Wu, B. Xu, J. Strzalka, J. Rivnay, D. C. Martin and L. V. Kayser, Polym. Chem., 2022, 13, 2764–2775.
27. E. Zheng, P. Jain, H. Dong, Z. Niu, S. Chen, S. Zhong and Q. Yu, ACS Appl. Polym. Mater., 2019, 1, 3103–3114.
28. I. del Agua, D. Mantione, N. Casado, A. Sanchez-Sanchez, G. G. Malliaras and D. Mecerreyes, ACS Macro Lett., 2017, 6, 473–478.
29. I. Zozoulenko, A. Singh, S. K. Singh, V. Gueskine, X. Crispin and M. Berggren, ACS Appl. Polym. Mater., 2019, 1, 83–94.
30. N. Massonnet, A. Carella, O. Jaudouin, P. Rannou, G. Laval, C. Celle and J.-P. Simonato, J. Mater. Chem. C, 2014, 2, 1278–1283.
31. E. Istif, A. Kagkoura, J. Hernandez-Ferrer, A. Stergiou, T. Skaltsas, R. Arenal, A. M. Benito, W. K. Maser and N. Tagmatarchis, ACS Appl. Mater. Interfaces, 2017, 9, 44695–44703.
32. M. A. Fusella, Y. L. Lin and B. P. Rand, in Handbook of Organic Materials for Electronic and Photonic Devices (Second Edition), ed. O. Ostroverkhova, Woodhead Publishing, 2019, pp. 665–693.
33. E. Palacios-Lidón, E. Istif, A. M. Benito, W. K. Maser and J. Colchero, Nanoscale, 2019, 11, 11202–11208.
34. G. Tullii, C. Bellacanzone, H. Comas Rojas, F. Fumagalli, C. Ronchi, A. Villano, F. Gobbo, M. Bogar, B. Sartori, P. Sassi, G. Zampini, G. Quaglia, L. Latterini, H. Amenitsch and M. R. Antognazza, ACS Appl. Mater. Interfaces, 2025, 17, 22434–22447.
35. J. Garemark and G. Panzarasa, Nat. Rev. Mater., 2026 DOI:10.1038/s41578-026-00906-6.

---