Report

Renewable Cathode Materials from Biopolymer/Conjugated Polymer Interpenetrating Networks

See allHide authors and affiliations

Science  23 Mar 2012:
Vol. 335, Issue 6075, pp. 1468-1471
DOI: 10.1126/science.1215159

Abstract

Renewable and cheap materials in electrodes could meet the need for low-cost, intermittent electrical energy storage in a renewable energy system if sufficient charge density is obtained. Brown liquor, the waste product from paper processing, contains lignin derivatives. Polymer cathodes can be prepared by electrochemical oxidation of pyrrole to polypyrrole in solutions of lignin derivatives. The quinone group in lignin is used for electron and proton storage and exchange during redox cycling, thus combining charge storage in lignin and polypyrrole in an interpenetrating polypyrrole/lignin composite.

Renewable energy systems based on intermittent sources require methods for power balancing over time, and thus some means of storage. Charge storage in organic polymers rarely gives energy and power densities, gravimetric or volumetric, that match the needs for secondary batteries and supercapacitors. This was one reason for the abandonment of efforts to make polymer batteries from conjugated polymers two decades ago (1), because inorganic insertion electrodes are superior. Widespread application of electrical power storage may require more abundant materials than those available in inorganics (which often require rare metals), and at a lower cost. Materials for charge storage are desired from easily accessible and renewable sources (2). Combining cellulose materials and conjugated polymers for charge storage (3, 4) has again attracted attention (5).

Biopolymers with redox functions are used in energy conversion processes in plants. The highly sophisticated structures designed to split water to oxygen and protons, in photosystem II in green plants, are made from protein structures combined with a manganese complex and use temporary proton storage on amino acid residues to accomplish this four-electron oxidation step (6). Electron and proton storage is found in the metabolism of plants and bacteria, where quinones are used as soluble electron/proton transport agents. With hydroquinone (Q/QH2), two electrons and protons are stored in a structure of 6 carbon and 2 oxygen atoms, an electronic charge density of 2 Faraday per 108 g, 1787 C/g, or 496 mAh/g. This is a favorable number compared with standard electrochemical systems; in lithiated carbon materials, a maximum doping level is 6 carbons per lithium, equivalent to 344 mAh/g and, in the olivine FePO4 system, 170 mAh/g (7). It is desirable to use the quinone redox function in electroactive materials to enhance charge-storage capacity.

Conjugated polymers with added quinone groups within the conjugation path on the main chain give improved charge storage (810). The inclusion of anionic redox species as the counter ion, also incorporating quinones, in doped conjugated polymers is well demonstrated (1114). The combined redox processes of polymer and redox anion contribute to charge capacity in these materials.

Phenol and quinones are found in plants and wood. A by-product of paper processing is brown liquor, also called spent sulfite liquor, an abundant and cheap residue mainly used as a surfactant. Brown liquor incorporates derivatized lignins rich in phenol groups, which can be further converted to quinones through oxidation processes.

We report synthesis and characterization of a class of materials based on the combination of polypyrrole and lignin derivatives with redox functions. Among biopolymers, lignin is second only to cellulose in biosynthesis and makes up some 20 to 28% of wood. The lignin polymer (15) is chemically variable and electronically insulating. The electrochemistry of lignins has been studied in thin layers (16) and evaluated for electrocatalysis (17, 18). If lignin is incorporated into an electrode material with sufficient electronic and ionic conductivity to allow charge transport to and from the quinone site, it is possible to use this redox function for charge storage. We show that polypyrrole is suitable and that quinone electrochemistry and polypyrrole conductivity combine to create an electroactive conjugated polymer/biopolymer composite. Previous reports of electrical properties of composites of polypyrrole with lignin derivatives (19, 20) did not report the use of the redox capacity in lignin.

Electrochemical polymerization of pyrrole at Au electrodes in aqueous lignosulfonate solutions (21), using a three-electrode system, generates a solid, black conducting product [Ppy(Lig)] that adheres to the substrate. Growth of this material (fig. S1) can continue unabated for more than 2 hours, giving a film of thickness 3 μm. The conductivity of these materials is ~1 S/cm. From elemental analysis (21), we deduce the characteristic N:S ratio, which indicates a mass fraction of ~50% doped polypyrrole and 50% lignosulphonate.

Cyclic voltammetry (CV) of Ppy(Lig) electrodes in 0.1 M aqueous HClO4 reveals two redox waves, one very well defined and narrow at ~0.5 V versus Ag/AgCl and one less well pronounced in the potential range where oxidation/reduction of polypyrrole is typically found (Fig. 1 and fig. S2). The redox process of the Ppy(Lig) electrode in aqueous electrolytes with pH buffers shows a systematic dependence of the redox potential with pH, with 58 mV/pH unit, close to the 59 mV/pH unit expected for a 1-electron/1-proton process (fig. S3).

Fig. 1

CV of the Ppy(Lig) composite electrode. (A) Voltammograms recorded between 0.1 and 0.4 V. (B) Voltammograms recorded between 0.1 and 0.75 V versus Ag/AgCl, scan rates 5 to 25 mV s−1 (inner to outer). (C) Dependence of the redox peak currents on scan rate. Film thickness, 0.5 μm.

Discharge under galvanostatic conditions in 0.1 M HClO4 probe the available charge of these materials (Fig. 2) and the rate at which this charge can be extracted. We note that for the thinner film, almost no dependence of the discharge rate is observed, but the thicker film shows clear limitation of capacity and rate within the interval of 1 to 17 A/g. We note from the discharge curves two different slopes of voltage versus charge capacity, with a transition at 0.35 to 0.55 V (depending on the film thickness and the discharge current), which should be a reflection of the two contributing materials with different redox potentials. The transition is where the quinone system should be reduced from its Q form to its QH2 form, with remaining capacity due to the polypyrrole. For the thinner film, with little dependence on the discharge rate and therefore reflecting a situation of minor diffusion limitations, we find the quinone charge capacity to be ~40 mAh/g and the polypyrrole charge to be 30 to 35 mAh/g.

Fig. 2

Galvanostatic discharge curves for (A) thinner (0.5 μm) and (B) thicker (1.9 μm) Ppy(Lig) composite film. Two regions are visible, assigned to electrochemical activity of Ppy and lignin-derived quinones, along with linear regression lines used for capacitance analysis. For clarity, in (A) the regression lines are shown for the highest discharge current only; the other ones nearly overlap with each other.

Capacitance analysis was done with the experiments reported in Fig. 2. The two slopes found in these discharge curves are interpreted as due to the two electrode mechanisms of quinone and polypyrrole electrochemistry. The capacitance values calculated from the inverse of the slopes are plotted in Fig. 3 and span from 1000 F/g for the thin-film electrode quinone system at low discharge rates to 350 F/g for the thick electrode at high discharge rates, with diffusion limitations. The overall capacitance in the composite material, a combination of both processes, is lower.

Fig. 3

Capacitance versus discharge current of Ppy(Lig) electrodes, as evaluated in terms of two contributing capacitances due to the quinone (Clig) and the polypyrrole (Cppy) redox processes. The data are extracted from the slopes of the linear parts of the discharge curves shown in Fig. 2. The overall capacitance (Coverall) was calculated from the charge stored and the voltage change. The thinner electrode (A) (0.5 μm) shows only a minor dependence of the three variables (Clig, Cppy, and Coverall) on the discharge current, whereas the thicker electrode (B) (1.9 μm) is strongly influenced by diffusion limitations.

We have studied the mass change of the electrode, both during the synthesis process and during oxidation and reduction, using a quartz crystal microbalance with dissipation monitoring (21). The synthesis of the polymer material under galvanostatic conditions leads to a mass that grows linearly with charge (fig. S1).

During redox cycling of the electrode material in 0.1 M aqueous HClO4, mass changes are small and indicative of a mixed anion and cation exchange with the electrolyte (figs. S4 and S5). By comparison with standard polypyrrole (ClO4), we note smaller mass change and suppressed variation of the mechanical modulus of the material during redox.

The presence of two distinct redox waves in the CV of composite electrodes further illustrates the difference between the Ppy(Lig) material and standard forms of polypyrrole. The location and shape of the second redox wave shows that the quinone groups of the lignosulfonate are responsible for this charge. The wave can be deconvoluted into two redox peaks 0.3 V apart when measured by differential pulse polarography (fig. S2C). The different monolignols in the lignosulfonate material show different chemical structures adjacent to the quinone groups, and the sinapyl element—with a methoxy group near the quinone group—is plausibly the one found at lower potential, as the methoxy group injects electrons into the phenyl ring (18).

Integration of the charge in the CV in Fig. 1B indicates that the ratio of charge in the redox of polypyrrole and the quinone groups is ~1:1.4. Because the fraction of polypyrrole to lignosulfonate is almost 1:1 in the composite material, as deduced from elemental analysis, we conclude that the major charge storage is within the quinone system. This is true also for thick films and indicates that the quinone redox site is readily accessible. The redox process requires both ion and electron transport, and the redox peak amplitude is linear in sweep rate for thin electrodes (0.5 μm), indicating a surface-bound electroactive species. As we go to thicker films (1.9 μm), we transit to a diffusion-limited redox peak for the quinone group (fig. S2B).

We propose that the redox activity in the composite electrode is due to anion insertion in the first wave and a proton release in the second wave (Fig. 4). Ppy0(LigQH2)Ppy+(ClO4)(LigQH2)+ePpy+(ClO4)(LigQH2)Ppy+(ClO4)(LigQ)+2e+2H+ The net reaction (if completed) is

Ppy0(LigQH2)Ppy+(ClO4)(LigQ)+3e+2H+

Water can be the proton donor/acceptor in quinone electrochemistry and is a most probable site where the proton is stored. This is why the use of nonaqueous solutions almost completely suppresses electroactivity (21). However, polypyrrole has acid-base properties leading to marked pH sensitivity of conductivity (22, 23). It is a conceivable proton acceptor, but within this pH range conductivity is retained.

Fig. 4

A simplified reaction of oxidative electrochemical redox of quinone functions in a lignosulfonate biopolymer within a polypyrrole matrix.

Polypyrrole may thus be a site for proton storage after oxidation and for retrieval upon reduction of the quinone group.

We know the mass fraction of –OH groups in the lignosulfonate material and can calculate the fraction of phenol groups in the composite electrode (21). In the lignosulfonate used, two of the monomers (sinapyl and coniferyl alcohol) dominate, and we neglect the contribution from the p-coumaryl alcohol group. These monomers could each contribute one quinone group, giving a maximum of 7% by weight of quinone in the composite electrode. This gives us a value of 69 mAh/g. For the polypyrrole fraction, elemental analysis indicates a low amount of anionic dopant species due to ClO4, and we can estimate a upper limit to this charge capacity of 40 mAh/g, based on anion exchange only. Assuming that polypyrrole can be charged to one charge per four monomers (with some fixed counterions carried on the lignosulfonate) and that cation exchange accounts for the ion flow, we obtain values of 90 mAh/g. We did not observe mass loss during oxidation of the electrode, so the cation insertion mechanism must be a minor one. The 50/50% composite material could therefore store ~80 mAh/g, based on the 1:1 stoichiometry. In measurements, we find charge densities at lower but comparable numbers, 70 to 75 mAh/g, for the thin-film electrode.

The theoretical ratio between charge capacity in the lignosulfonate and in polypyrrole is between 1.75 and 0.8, depending on the method of estimating polypyrrole capacity. Our experimental value is 1.4. However, as we must assume that the charge capacity in polypyrrole is higher than that of the oxidized lignosulfonate in order to explain our experimental results, we must also conclude that we used a large fraction of all quinone groups incorporated in the polymer film. The observation that a large fraction of the redox capacity of the lignosulfonate is accessible for electrochemistry is consistent with the almost molecular miscibility of polypyrrole and lignosulfonate, as also proven by the absence of nanostructure in electron microscopy (21). The density of the material is 1.4 g/cm3, which leads to a volumetric charge density of 100 mAh/cm3.

Self discharge (21) is a problem with these electrodes and will need further study. However, we observe considerable differences of performance between different lignosulfonate compounds. The fraction of phenolic groups varies widely in lignosulfonates, depending on origin and processing of lignin. This means that there is room for optimization of the Ppy(Lig) materials using different sources of processed lignins with varying loading, with varying charge densities, and with a possibility to improve upon present results.

We compare the charge capacity and the capacitance per mass of these materials to those for polypyrrole/carbon composites [table 1 in (24)]—as analyzed for the three-electrode situation, which does not take counter electrode and electrolyte into account—and find that the charge density and capacitance of Ppy(Lig) are higher than reported for most of these materials. Only polypyrrole combined with nanostructured carbons is close to the values reported here.

We have demonstrated interpenetrating networks of lignosulfonate and polypyrrole that can be used for charge and energy storage. The use of the renewable biopolymer should lead to low-cost electrodes with improved safety and nontoxicity, operating in water. There is ample room for further developments to improve charge density and capacitance by searching through the universe of lignins.

Supporting Online Material

www.sciencemag.org/cgi/content/full/335/6075/1468/DC1

Materials and Methods

Figs. S1 to S9

References (2529)

References and Notes

  1. Materials and methods are available as supporting material on Science Online.
  2. Acknowledgments: This work was supported by the Knut and Alice Wallenberg Foundation, and O.I. is a Wallenberg Scholar. We thank R. Gabrielsson, N. Solin, A. Elfving, and V. Andersson for experimental support and discussion. The kind donation of lignosulfonate samples from Borregaard LignoTech AS is gratefully acknowledged. The authors have applied for a Swedish patent on the class of materials reported here.

Stay Connected to Science

Navigate This Article