Electrochemical Characteristics of Lignin in CTMP for Paper Battery Electrodes

Abstract

Lignin has been extensively researched as a cathode active material in secondary batteries. In the present work, the energy storage potential of lignin naturally present in papers made of softwood chemi‐thermomechanical pulp (CTMP) is explored. More specifically, effects from softwood CTMP fines on the electrochemical characteristics have been studied. Compared to pulp fibers, fines are higher in lignin content and have higher specific surface area. It was expected that this would be positive for the electrode performance; however, the result points to the opposite. The fines do not significantly contribute to a higher lignin specific capacity, and they deteriorate the cycling stability. Higher fines content was found to result in a higher oxidative activity as well as more abundant competing reactions. These competing reactions are believed to be linked to the cycle stability. Therefore, we hypothesize that the electrochemical stability of lignin can be better understood by studying differences between fines and fiber lignin. As the theoretical specific capacity of this material is about 20 times larger than obtained here, identification of the reasons for this capacity discrepancy is needed to realize the full potential of lignin‐based paper batteries.

Lignin naturally present in papers made of unbleached mechanical pulps is electrochemically active. Here, softwood chemi‐thermomechanical pulp (CTMP) is studied as an active cathode material for application in paper batteries. Proof‐of‐concept measurements are shown, and it is found that softwood CTMP fines exhibit competing oxidative electrochemical reactions, which probably can be linked to the electrode's limited cycling stability.

Introduction

A rising demand for sustainable and low‐cost energy storage devices has rallied research on alternative material concepts. Here, a range of plant‐derived carbohydrates and polyphenolics can be processed and successfully employed. Depending on their nature, they can provide ionic or electrical conductivity,[ ¹ , ² ] charge storage, ^[3] and mechanical stability, ^[4] and have been found relevant for key components such as electrolytes, ^[5] membranes, ^[6] and electrodes. ^[7] Notably, lignocellulosic materials have been extensively researched given their abundance and long use in papermaking. ^[8] Devices coined “paper batteries” constitute a category of researched energy storage devices where cellulose is a bearing constituent, not only as a separator, but also in the electrodes.[ ⁹ , ¹⁰ ] In contrast to traditional electrode fabrication methods, paper electrodes are also believed to be a low‐cost alternative thanks to the high production rates on paper machines – over 1 km² or 50 ton per hour is common ^[11] – as well as to the low‐cost raw material. Although paper batteries would not be able to compete with e. g. lithium‐ion batteries (LIB) in terms of energy density, paper batteries should be relevant for less demanding applications where price and sustainability are more important. ^[8]

To store and conduct electric charges, paper batteries must include active electrode materials as the cellulose itself is insulative and electrochemically inert. As previously reported in this journal and elsewhere, lignin has attracted much attention as an active cathode material in secondary batteries and pseudo‐supercapacitors thanks to its electroactive structure.[ ¹² , ¹³ , ¹⁴ ] This ability is enabled after forming reversible catechol/orthoquinone (QH₂/Q) redox pairs (Figure 1A). Although rare in native lignin, the QH₂/Q structure can be formed by oxidizing phenolic terminal units of guiacyl or syringyl type. This oxidation occurs in the processing of lignin ^[15] but can also be done electrochemically. ^[16]

Figure 1.

(A) The experimental matrix of the four prepared pulps for evaluating the factors fines content and acetone‐soluble extractives and (B) the fines and lignin contents in the resulting papers made from the four different pulps.

Each QH₂/Q group can store the charge equivalent of two electrons, which translates to 53.6 mAh per 1 mmol of QH₂/Q. A guaiacyl‐type monolignol with a molecular mass of about 180 g/mol should thus exhibit a specific capacity of around 296 mAh/g after demethylation. This can be compared to polymeric lignins, which rather would range from 39 mAh/g (spruce wood) ^[17] to 240 mAh/g (softwood kraft lignin). ^[18]

Lignin cathodes have originally been researched with protons as mobile ions, which limits the cell operating window. However, protons in catechols can be substituted to Li⁺ which considerably increases the theoretical potential window. ^[19] Along with research on post‐LIB technologies with more abundant monovalent metal ions, investigations on lignin with e. g. Na^+[20] or K⁺, ^[21] but also divalent Zn²⁺, ^[22] have successfully been demonstrated.

Paper batteries with added lignin has been explored in several studies.[ ²³ , ²⁴ ] However, achieving a system where the lignin remains stable in the electrode is challenging. On the other hand, native lignin in wood is stabilized by α‐ether bonds to hemicellulose, which together forms what is known as the lignin‐carbohydrate complexes. ^[25] Exploiting native lignin in wood for energy storage has been researched ^[26] and it has been proved that such lignin exhibits a charge storage ability. The charge storage capacity can be improved by a light sulfonation of the wood, although heavier sulfonation of the wood leads to poorer cyclic stability. However, wood does lack the key feature for paper batteries of being producible on high‐speed paper machines. For paper batteries, we should rather turn our attention to pulps for papermaking, which normally contains different amounts of residual lignin depending on the pulping process.

The pulp class of interest is the mechanical pulps. These pulps are, in contrast to chemical pulps, made using processes which largely preserve the lignin. In this work, we have investigated the charge storage ability of unbleached chemi‐thermomechanical pulp (CTMP), a subclass of mechanical pulps. The CTMP used in this study was made from Norway spruce (Picea abies), in which lignin is mainly of guaiacyl type. ^[27] In the CTMP process, wood chips are treated with a light sulfonation prior to mechanical defibration in hot water. In this mechanical defibration, a substantial amount of fine particles of disintegrated wood, called fines, are formed. They differ from the fibers mainly by size, but also by chemical composition and shape. The fines are a heterogenous fraction of the pulp, where the three major subcategories fibril, flake, and ray cell fines all have distinctly different morphology and chemical composition. ^[28]

Mechanical pulps contain extractives such as resins and fatty acids, which are known to reduce accessible fiber surface area for fiber‐fiber bonding sites. ^[29] These extractives could potentially constitute an insulating barrier to electronic contacting of the lignin. In this work, the relevance of fines and extractives for the CTMP lignin electrochemistry has been investigated. The results from these experiments, in combination with a better understanding of the charge storage ability in CTMP lignin, constitute valuable insights for the development of biobased paper batteries.

Results and Discussion

Pulp and Paper Composition

The CTMP whole pulp contained 54 % fines which decreased to 19 % after fines removal by Celleco filtering (Figure S1). Compared to the whole pulp, pulp with fines removed contained slightly lower amounts of acetone‐soluble extracts (9.8 vs 10.9 mg/g pulp). This difference is consistent with previously reports on fines and fibers in mechanical pulps, where fines in general are richer in both extractives and lignin compared to fibers. ^[13]

Handsheets were prepared from the four different pulps which had been treated or not treated with fines removal and acetone extraction according to the experimental matrix (Figure 1A). The fines content in papers made from the whole pulps were lower compared to the fines content of the whole pulp itself (Figure S1); however, the papers can still be divided into two distinct categories of higher and lower fines content (Figure 1B). Acetone extraction did not affect the fines content but reduced the lignin content, which is in line with previous reports on acetone‐soluble fractions in CTMP lignin. ^[30]

Electrode Electrochemical Signature

The formed papers were made conductive by impregnation with aqueous poly(3,4‐ethylenedioxythiophene) polystyrene sulfonate (PEDOT : PSS). The general electrochemical responses were characterized by cyclic voltammetry (CV), as shown in Figure 2B. In a mildly acidic electrolyte (0.01 M HCl with NaCl 1 M), the CV clearly displayed lignin redox characteristics previously reported elsewhere. ^[16] The observed initial resting potential was close to 0.35 V, showing that the QH₂/Q redox pair (E=0.49 V vs Ag/AgCl) was in its reduced catechol form at rest.

Figure 2.

The measured electrochemical characteristics of CTMP lignin. Three electrochemical reactions known to occur in lignin were targeted (A); the oxidation of catechols (i), the oxidation of guaiacyls (ii), and the reduction of orthoquinones (iii). In a performed CV (B), reaction (i) was almost absent and reaction (ii) distinct in the first cycle, with the inverse case for the subsequent cycles. By integrating current over time in mixed‐mode galvanostatic charge/discharge measurements (C), the capacities of the reactions i‐iii in the first charge/discharge cycle were assumed to be approximately quantified by the capacities (a), (b), and (c).

The peak related to catechol oxidation (Figure 2A, reaction i) was of considerably lower magnitude in the first cycle compared to subsequent cycles (Figure 2B). This was expected since QH₂/Q moieties in general are not present in native lignin. In this system, they are instead formed in the electrochemical oxidation of guaiacyls (Figure 2A, reaction ii),[ ¹⁶ , ³¹ ] visible as an oxidative peak at 0.73 V. The guaiacyl peak was highest in the first cycle, after which it rapidly decreased for each cycle. As more and more of the guaiacyl groups are demethylated and turned into catechols, the peaks assigned to the QH2/Q redox pair are growing.

The capacities (a), (b), and (c) measured by the mixed‐mode galvanostatic charge/discharge technique (Figure 2C) were assumed to reflect reactions i‐iii in Figure 2A. However, due to a probable presence of other unidentified reactions, the measured capacities a, b, and c are referred to as the oxidation capacities at 0.5 and 0.65 V, and the discharge capacity, respectively. The reported capacities were normalized to the mass of lignin.

Influence of Capacitance on Capacities

The measured capacitance in the potential window of 0.0–0.2 V exhibited a clear non‐faradaic behavior (Figure S2). As the papers were equally impregnated with PEDOT : PSS (15 wt %), no difference between the different pulp batches was expected when normalized to the paper mass. This was confirmed by employing two‐way ANOVA on samples according to the experimental matrix in Figure 1A. Fines removal and/or acetone extraction all displayed p‐values ≫0.05, which states that the treatments have no effect on the capacitance (Figure S3). However, the variation among the samples was high, with a standard deviation of 0.56 compared to the average 3.27 F/g paper. This probably reflects variations in PEDOT : PSS loadings in the samples ^[32] following uneven impregnations. While this variation was unintended, it opened for investigating effects from the PEDOT : PSS content.

Variations in PEDOT : PSS loadings should affect the measured capacities in at least two ways. First, there exist both faradaic charge contributions from the lignin redox reactions, and non‐faradaic contributions from the capacitive PEDOT : PSS. Second, as lignin is electrically insulative, variations in PEDOT : PSS loadings should affect the amount of electronically connected lignin. Correlations of the specific capacitance with the respective capacities a, b, and c measured according to Figure 2C were therefore investigated with multiple linear regression. Such correlations were indeed found for all capacities (p <0.05) and are also observable graphically (Figure 3A–C). For the oxidative capacities a and b, which should comprise only faradaic charges since ΔV=0, the correlation with capacitance means that more PEDOT : PSS leads to additional lignin reactions. This points out that around 15 wt % impregnation with PEDOT : PSS was not enough for saturating the lignin with electric contacts.

Figure 3.

Results from multiple linear regression analysis (effects of acetone extraction excluded) on oxidation and reduction capacities; oxidation capacities at 0.5 V (A) and 0.65 V (B), and the discharge capacity over 0.65–0.35 V (C). Capacity measurements A, B, and C correspond to capacities a, b, and c in Figure 2C. Observations are displayed as dots, with their multiple linear regression functions as lines with their 95 % confidence interval as a shaded area. The difference in capacity retention between papers with higher respective lower fines content over 100 cycles is presented in (D), with the average as solid line and a 95 % confidence interval as shaded area. Colours represent higher (grey) and lower (green) fines content.

Effect from Acetone Extraction

Removing acetone‐soluble extractives from CTMP was initially hypothesized to enhance electrical contacts between lignin and PEDOT : PSS by removing insulating films of extractives. However, any such effects were absent in the results; neither the discharge capacity, nor the guaiacyl oxidation capacity were affected. The only significant effect from acetone extraction was a reduction in the oxidation capacity at 0.5 V (p=0.03). This can be understood as acetone‐soluble lignin having a higher content of catechols.

Competing Oxidative Reactions

An important result from this study is the discrepancy between the oxidation capacities (a) and (b), and the reduction capacity (c). In the ideal case, the amount of reducible orthoquinones should equal the sum of oxidized guaiacyls and catechols (Figure 2). However, the sums of the oxidation capacities were on average significantly higher than the discharge (reduction) capacity (p <0.01, Figure S4). This means that there are oxidative reactions competing with the formation of reducible orthoquinones.

Nevertheless, the oxidation capacity at 0.65 V was a significant explaining variable for the discharge capacity (p <0.01, Table S1), showing that guaiacyls do form orthoquinones in the oxidative process at 0.65 V. The question is how large portion of these oxidative reactions have orthoquinones as the product, in other words, the faradaic efficiency of the reaction. A precise quantification of the faradaic efficiency was not possible due to different approaches to measure the respective oxidation and reduction capacities but, clearly, the faradaic efficiency was less than the ideal case of 100 %.

Interestingly, the aspect of faradaic efficiency was also related to the fines content. Papers with higher fines content had significantly higher oxidation capacities at both 0.5 V and 0.65 V compared to papers with lower fines content (p <0.01 for both, Figure 3A, B). It was expected that these effects on the oxidative capacities would be mirrored in the discharge (reduction) capacity; however, this capacity was not related to the fines content (p ≫0.05, Figure 3C). This can be understood as the fines lignin, compared to fiber lignin, being more prone to competing oxidative reactions which do not result in reducible orthoquinones.

We also note that competing reactions, taking place more abundantly in fines, can be linked to the cycling stability of the paper electrodes. Repeated galvanostatic charge/discharge over 100 cycles revealed a significant difference in cycling stability between papers with lower respectively higher fines content, where a lower fines content provided higher capacity retention (Figure 3D). The competing reactions are thus potentially not only competing, but also destructive for the electrode.

Reactions in lignin‐carbohydrate chemistry are complex and extensive studies will be required to understand the nature of these competing reactions. However, as the results point out, differences between fines and fiber lignin could provide clues.

Fines and fibers differ both in chemical composition and morphology. Fines, especially CTMP fines, consist largely of fragments from the compound middle lamella (CML). ^[33] Secondary fiber wall lignin is organized wrapped around the microfibril bundles,[ ³⁴ , ³⁵ ] which is different from the less ordered structure found in the CML.

There are also reports on differences in chemical structure between fines and fiber lignin: from studies on softwood thermomechanical pulp, it has been suggested that fiber lignin, especially bulk fiber lignin inside the fiber wall, is richer in methoxyl groups compared to fines lignin.[ ³⁶ , ³⁷ ] Further, spruce CML lignin is reported to contain fewer β‐O‐4 bonds and more condensed structures compared to lignin in the secondary wall. ^[38]

However, it remains unclear how the differences between fines and fibers could be linked to the electrochemical results in this work. We propose that to gain a better understanding of this in future investigations, separate extraction of lignin from fines and fibers should be performed. Here, a mild protocol such as cellulolytic enzyme lignin (CEL) extraction would be recommended,[ ³⁹ , ⁴⁰ ] after which the extracted lignin should be characterized by its chemical structure (e. g. by nuclear magnetic resonance (NMR) such as ³¹P‐NMR, Carbon‐13 or 2D‐Carbon‐Proton NMR) and measures of electrochemical characteristics (e. g. by CV or step potential electrochemical spectroscopy).

Discharge Capacity

The lignin specific capacity in this work reached at its best 2.63 mAh/g over 0.65–0.35 V vs Ag/AgCl. This can be compared to the theoretical specific capacity of lignin calculated from the amount of phenolic hydroxyl groups (PhOH) reported elsewhere. Calculations based on results from ³¹P‐NMR applied on lignin extracted from thermomechanical pulp indicate a theoretical specific capacity of 60.5–73.4 mAh/g depending on the lignin extraction protocol. ^[39] The ³¹P‐NMR results are in line with PhOH quantification in high‐temperature CTMP using the periodate oxidation technique, ^[41] which translates to a theoretical specific capacity of 60.2 mAh/g lignin. With these values as benchmark, it should be possible for the lignin specific capacity to be considerably higher than what was achieved in this work.

Nevertheless, the 2.63 mAh/g lignin achieved herein is close to the 2.92 mAh/g lignin achieved in a previous work on heavily sulfonated wood veneers impregnated with a similar amount of PEDOT : PSS. ^[12] It is worth to note that mild sulfonation and defibration can reach similar results as heavy sulfonation in terms of accessing lignin electrochemically. However, choosing PEDOT : PSS as the electronic additive might be questionable since both this work on CTMP and the mentioned work on sulfonated wood underperforms in discharge capacity.

It was believed that a thorough impregnation of PEDOT : PSS would give electronic access to the bulk of the paper, and that PEDOT should exhibit good affinity to the cellulosic fibers as imaged from previous work on systems with PEDOT : PSS and nanocellulose. ^[42] Scanning electron microscopy with energy dispersive X‐ray spectroscopy (SEM/EDX) imaging revealed that PEDOT : PSS did thoroughly impregnate the bulk of the paper but, however, it also revealed the creation of PEDOT : PSS agglomerates (Figure 4). Thus, it appears as if the internal cohesion of PEDOT : PSS overarches the affinity to cellulose. Here, electrostatic repulsive forces between sulfonated lignin and PSS might at least partly explain this behavior.

Figure 4.

Representative SEM/EDX cross‐section of a CTMP paper impregnated with PEDOT : PSS, where sulphur (purple) denotes the distribution of PEDOT : PSS against the background signal from carbon (green).

Further, it appears as if the fiber walls have not been penetrated by the PEDOT : PSS. Poor affinity to the fiber surfaces could again play a role, but the hydrodynamic diameter of the PEDOT : PSS might also constitute a problem if it is too large in relation to any available pores in the wall. To overcome these challenges, polymerization in‐situ of a conducting polymer might be an attractive alternative to impregnation. For example, pyrrole has been demonstrated to penetrate and polymerize in‐situ inside the fiber walls of sulfonated wood. ^[43] The same study also points out the importance of loosening the fiber wall structure. This must be done carefully as blunt lignin‐dissolving chemical modifications, such as sulfonation, have got a proven negative effect on the electrode cycling stability. ^[26]

Conclusions

PEDOT : PSS‐impregnated papers made from unbleached spruce CTMP display the familiar electrochemical characteristics of lignin reported elsewhere. Acetone‐soluble extractives were present in the pulp, but acetone extraction did not lead to increased electrochemical activity. Rather, acetone extraction reduced the amount of lignin in the pulps and appears to particularly dissolve the small fraction of the CTMP lignin that carry catechol moieties.

For paper electrodes made from unbleached spruce CTMP, reactions competing with the formation of reducible ortho‐quinones were concluded to occur, as the sum of the oxidation capacities exceeded the reduction capacity. This happened to a larger extent in fines than in fibers. The fines content was also inversely related to the cycling stability. It is possible that this lower stability of fines lignin is related to the more abundant competing reactions occurring in fines. Thus, studying differences between fines and fiber lignin might reveal clues about the electrochemical stability of lignin.

The performance of an electrode has many different measurable aspects, of which the cycling stability is just one. The specific capacity of the electrode is another one, and this study concludes that there still is much left to be done to realize the full potential of CTMP for paper battery electrodes. The achieved specific capacity is only 4–5 % of the theoretical capacity. This can be due to limitations in using PEDOT : PSS as the electronic additive, why alternative conductors should be tested. Treatments that open the closed structure of the fiber wall to further facilitate access for the electronic conductor as well as the electrolyte is another suggested route to increase the capacity, although this must be done carefully to not affect the electrode stability. One could also consider chemistries which can provide fire retardancy in such paper electrodes e. g. through phosphorylation, ^[44] which is relevant in cases where non‐aqueous electrolytes are used.

While some aspects still need to be resolved before CTMP paper batteries can be turned into reality, the results from this study provides a better understanding of these remaining issues and provides directions for the continued research on low‐cost, biobased and sustainable energy storage.

Experimental Section

Pulp and Paper Preparation and Characterization

Wood is by nature a heterogeneous material, but the properties of wood‐derived materials can be better controlled the more processed they are. Wood of different species and origin, as well as different pulping techniques, give pulps of different properties. The present study employed a commercial‐grade unbleached CTMP pulp made of Norway spruce (Picea abies), kindly supplied in bale by Stora Enso Skoghall mill, Sweden. Since the pulp was delivered according to specification, the results of this study can be assumed to be reproducible with this very material.

The unbleached CTMP pulp was hot disintegrated according to ISO 5263‐3 prior to further processing. This pulp is henceforth referred to as whole pulp. Papers made from four different batches of pulp (Figure 1A) with or without acetone extraction and/or fines removal were prepared according to the following procedure.

For the removal of fines, 500 g of whole pulp were diluted to 0.5 % consistency in a 100‐litre chest with air agitation. This pulp suspension was pumped with a speed of 10 L/min and sprayed onto a 75 μm mesh in a Celleco filter, at which the fines were able to be accepted and the fibers rejected by the filter. The filtering procedure was continued for 3 h. Both whole pulp and fines‐removed pulp were collected separately and dewatered over 15 μm mesh in a vacuum funnel prior to further processing.

Soxhlet extraction with acetone was performed according to standard ISO 14453 on samples of both whole pulp and on fines‐removed pulp. Prior to extraction, the pulps were each separately solvent exchanged from water to ethanol (96 %, Sigma Aldrich). After extraction, the pulps were washed with water 10 times in the Soxhlet equipment, after which they were stored in water before the papermaking step.

Prior to the preparation of laboratory hand sheets, the four different CTMP pulps were hot disintegrated in accordance with ISO 5263–3, diluted to 2 g/L, and cooled to room temperature. Sheets were formed in a circular TAPPI manual sheet former (PTE GmbH), followed by couching with blotting papers and a stainless‐steel roll and finally dried in a speed dryer (PTE GmBH). The resulting papers had a grammage of about 50 g/m².

On the acquired pulps, as well as on hot disintegrated lab sheets made from the different pulps, the fines content was measured using a L&W FiberTester+, and the lignin contents were determined according to the acid hydrolysis method (ISO 21436, procedure B). Fines and lignin content measurements were performed in duplicates.

Preparation of Sample Paper Electrodes

Aqueous PEDOT : PSS solution with a concentration of 1.25 wt % (Orgacon ICP 1050) was kindly supplied by Agfa. Ethylene glycol (EG) bought from Sigma Aldrich was adopted as a secondary dopant and added to the PEDOT : PSS solution with a mass ratio of 5 : 1 and magnetically stirred overnight in room temperature. The liquid was used for impregnating samples of the hand sheets. Impregnation was made by wetting circular patches ($\varnothing$ =50 mm) of the paper samples in the PEDOT : PSS and EG solution in petri dishes. The amount of PEDOT : PSS solution was adjusted to enable an even wetting of the paper, which translated to 15 % of the paper dry weight. The samples were left for evaporation overnight, after which they were dried in 120 °C for 1 h to evaporate EG from the samples.

1 cm² patches of the papers were prepared with a laser cutter (Trotec Speedy 300). Wet‐proofed (polytetrafluoroethylene‐impregnated) polyacrylonitrile carbon fiber sheets (Toray carbon paper) were used as a conducting support for the patches, to which polyimide (Kapton) tape was taped on one side for mechanical support. Sample patches were glued on the tape‐free side of individual supports using a carbon ink (7102 Carbon Conductor, DuPont), and then cured 10 min on a hot plate at 150 °C.

Electrochemical Characterization of Paper Electrodes

An Ivium Octostat5000 potentiostat along with Ivium software was used for the electrochemical characterizations in a 3‐electrode setup. The 20 ml cell was filled with aqueous 0.01 M HCl to which 1 M NaCl was added as supporting electrolyte (Figure S5) and was equipped with a platinum mesh counter electrode and an Ag/AgCl reference electrode. All reported potentials refer to the Ag/AgCl reference. After each characterized sample, the electrolyte was removed, and the cell and electrodes were cleaned to exclude effects of any leaked lignin on the subsequent sample.

Cyclic voltammetry at 10 mV/s in the potential window of 0.0–0.75 V was used for an initial investigation of the electrochemical characteristics for the CTMP paper. CV at 5 mV/s in the non‐faradaic window of 0.0–0.2 V was used to determine the capacitance of each individual sample, by integrating the average charge in the forward and backward scan during the 3rd CV cycle according to Mathis et al. ^[45]

Mixed mode galvanostatic charge/discharge was performed to evaluate three main characteristics as described in Figure 2C according to the following. A constant current of 0.1 mA was applied until a potential of 0.5 V was reached. At 0.5 V, the measurement changed to constant potential until the current had dropped below 0.01 mA. Then, charging was continued at 0.1 mA until a potential of 0.65 V was reached, where the potential again was kept constant until the current had dropped below 0.01 mA. Lastly, a discharge current of −0.1 mA was employed until −0.1 V.

The discharge capacity retention was investigated by running galvanostatic charge/discharge for 100 cycles at 0.1 mA over the potential window of 0.35–0.65 V.

SEM/EDX Imaging

Paper electrode samples from all four batches were imaged all along their profile in cross‐section with SEM (Zeiss, EVO 50) and EDX (Bruker, Quantax 200). On the obtained X‐ray maps, sulfur was used as a marker for the PEDOT : PSS in contrast to the background carbon signals.

Statistical Evaluation

Given the heterogenous nature of papers, statistical evaluation of the resulting data is necessary in order to ensure repeatability. In this study, the electrochemical results in terms of specific capacitance, initial catechol oxidation, guaiacyl oxidation, and discharge capacity have been evaluated by two‐way analysis of variance (ANOVA) and with multiple linear regression. 6 samples of each pulp type were analyzed. The significance level α was set to 0.05, meaning that any result was regarded as significant when it could be stated with 95 % confidence.

Supporting Information

Supplementary figures and a table of the summarized statistical results are available in the supporting information.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Isacsson P., Björk E., Capanema E., Granberg H., Engquist I., ChemSusChem 2024, 17, e202400222. 10.1002/cssc.202400222

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.