Advanced Ti‐Doped Fe₂O₃@PEDOT Core/Shell Anode for High‐Energy Asymmetric Supercapacitors

Abstract

An effective strategy to significantly boost the capacitive properties of Fe2O3-based anodes by Ti doping and poly(3,4-ethylenedioxythiophene) (PEDOT) coating is successfully demonstrated. The Ti-Fe2O3@PEDOT electrode exhibits a significant capacitance improvement and exceptionally cyclic stability. A remarkable energy density of 0.89 mWh cm−3 can be obtained for a high-performance asymmetric supercapacitor device consisting of Ti-Fe2O3@PEDOT and a MnO2 anode. The development of advanced energy-storage and delivery systems is highly pursued with the ever-increasing demand for renewable energy sources and growing concerns for the environment.1-5 Asymmetric supercapacitors (ASCs), also known as electrochemical hybrid supercapacitors, have attracted increasing attention due to their potential applications in hybrid electric vehicles, hand-held electronics, microelectro­mechanical systems and sensors.6-10 ASCs typically are made up of a battery-type Faradaic cathode as the energy source and a double-layer-type anode as the power source, and thus can be operated in a wider working voltage range and deliver a substantially higher energy density.11-13 It is well known that the functionality of the electrode materials is essential to the overall properties of the ASCs. With this in mind, considerable interest has been sparked in exploring high-performance cathode and anode materials for ASCs.14-18 Over the past few years, great achievements have been made on the performance of cathode materials, whereas the progress on the anode materials has been relatively slow, which therefore has become the main barrier for practical applications of ASCs.19 Carbon nano­materials, such as activated carbon (AC),20 graphene,21-23 and carbon nanotubes (CNTs)24, 25 with excellent electrical conductivity and high surface area, have been extensively studied as anodes. Recent reports have shown that ASCs based on these carbon anodes have good power density, but usually suffer from a low energy density as a result of the low capacitance of the carbon materials.26-28 Therefore, the exploration of new, state-of-the-art anode materials with high capacitance is still highly valuable and significant. Hematite (α-Fe2O3), an earth-abundant, low cost and environmentally benign material with a high theoretical specific capacitance and suitable negative working window, has received growing attention as the high-performance anode for ASCs in recent years.29-32 However, the reported capacitive performance of α-Fe2O3 electrodes, especially the rate capability and energy density, is still unsatisfactory because of its poor electrical conductivity (ca. 10−14 S cm−1).33 Some strategies have been employed to improve the capacitive performance of hematite electrodes, including the development of nanostructures to increase the effective surface area and to shorten the diffusion pathway for ions and electrons, the incorporation of carbon materials for improving the electrical conductivity as well as the introduction of oxygen vacancies into hematite.17, 32, 34-36 For instance, an areal capacitance of 0.68 F cm−2 at 3 mA cm−2 (equal to 908 F g−1) observed for a Fe2O3/graphene composite hydrogel was much higher than that of pristine Fe2O3 (0.23 F cm−2) in 1 m KOH electrolyte, but its cyclic stability needed to be further improved (Only 69% of its initial capacitance was retained after 200 cycles).37 Recently, oxygen-deficient α-Fe2O3 nanorods (NRs) have been developed and yielded a remarkable areal capacitance of 382.7 mF cm−2 at 0.5 mA cm−2 with enhanced cycling stability (ca. 95% of its capacitance could be retained after 10 000 cycles).17 Nevertheless, it remains a challenge to develop high-performance Fe2O3 electrodes with satisfactory capacitive properties and cycling durability. In this work, we report the rational design and fabrication of Ti-doped Fe2O3@poly(3,4-ethylenedioxythiophene) (denoted as Ti-Fe2O3@PEDOT) core/shell NR arrays grown on flexible carbon cloth as the high-performance anode for ASCs. Ti4+ has been reported to be an electron donor by substitutionally replacing Fe3+ and reducing Fe3+ to Fe2+. It is anticipated that the donor density of Fe2O3 could be significantly enhanced after Ti doping, and thus could boost its capacitive performance. In addition, poly(3,4-ethylenedioxythiophene) (PEDOT) is an ultrahigh stable and conductive polymer. It can not only effectively improve the conductivity of nanomaterials, but also act as a protective layer to prevent the architectures from destruction/degradation. Herein, we have developed a new kind of Fe2O3-based anode using Ti-doped Fe2O3 nanorods as the core and a highly stable, conductive PEDOT layer as the shell. Such unique core/shell architectures can offer a high electrical conductivity of the overall electrode for charge transport, a large interfacial area for reaction, and numerous channels for rapid diffusion of electrolyte ions within the electrode, which endows the designed Ti-Fe2O3@PEDOT electrode with an excellent capacitive performance. The as-prepared Ti-Fe2O3@PEDOT core/shell electrode showed a remarkably large areal capacitance of 1.15 F cm−2 (311.6 F g−1 and 28.8 F cm−3 at 1 mA cm−2) with outstanding rate capability. The Ti-Fe2O3@PEDOT electrode also exhibited ultrahigh cycling durability with more than 96% capacitance retention after 30 000 cycles. To the best of our knowledge, these are the best areal capacitance and capacitance retention values ever achieved for α-Fe2O3 electrodes. Based on this advancement, a flexible high-performance ASC device with a maximum energy density of 0.89 mWh cm−3 and a maximum power density of 0.44 W cm−3 was achieved. This work constitutes a promising strategy to rationally design and fabricate novel Fe2O3-based nanostructured anodes with largely enhanced capacitive behavior, which hold great promise in energy storage/conversion devices. Ti-Fe2O3@PEDOT NRs were synthesized on a conductive carbon cloth substrate via a two-step process, as illustrated in Figure 1a. Firstly, Ti-Fe2O3 NRs were grown directly via a hydrothermal method according to a previously published report with slight modification.17 Pristine Fe2O3 NRs were also prepared via a similar synthetic procedure for comparison (see Experimental Section). Figure S1 in the Supporting Information displays typical scanning electron microscopy (SEM) images of the as-prepared Ti-Fe2O3 NRs, which clearly reveal that the Ti-Fe2O3 NRs with a diameter of approximately 50 nm were grown uniformly on the entire surface of the carbon fiber. Compared to pristine Fe2O3 NRs (Figure S1a), the diameter of Ti-Fe2O3 NRs is much smaller, suggesting they have a larger specific surface area. Brunauer–Emmett–Teller (BET) results showed that the specific surface area of Ti-Fe2O3 NRs/carbon cloth is 21.3 m2 g−1, which is almost double that of the Fe2O3 NRs/carbon cloth (13.1 m2 g−1), confirming that the specific surface area of the Fe2O3 NRs could be enhanced after Ti doping. After deposition of the NRs a uniform and conductive PEDOT polymer shell was electrochemically deposited on the Ti-Fe2O3 NRs surface (details in the Experimental Section). SEM images revealed that there is no obvious morphological change after PEDOT coating (Figure 1b). Figure S2a in the Supporting Information presents typical X-ray diffraction (XRD) patterns of Fe2O3, Ti-Fe2O3, and Ti-Fe2O3@PEDOT samples. Excluding the peaks of the substrate, all the diffraction peaks of Fe2O3 can be indexed to hematite (Joint Committee on Powder Diffraction Standards (JCPDS) #33–0664). Compared to Fe2O3, a slight shift in peak position toward lower angles is observed for the Ti-Fe2O3 (Figure S2b), suggesting that Ti4+ has been successfully doped into the Fe2O3 lattice as the ionic radius of Ti4+ is larger than that of Fe3+.38-40 To further study the detailed microstructure of the NRs, transmission electron microscopy (TEM) was performed. It is clear that an amorphous PEDOT shell of about 5 nm in thickness is enveloping the Ti-Fe2O3 (Figure 1c). The inset in Figure 1c displays a high-resolution TEM (HRTEM) image of Ti-Fe2O3@PEDOT. A lattice fringe of 0.27 nm that corresponds to the plane of hematite (JCPDS #33–0664) can clearly be observed, implying that the core is made of highly crystalline α-Fe2O3 NRs (Figure S3a,b in the Supporting Information). Energy-dispersive spectroscopy (EDS) mapping of Fe, Ti and S, as shown in Figure 1d–f clearly demonstrates that the Ti atoms are homogeneously distributed throughout the Fe2O3 NRs and that the PEDOT layer is coated uniformly on the surface of the Fe2O3 NRs. To gain insight into the chemical composition and valence state of the Ti-Fe2O3@PEDOT NRs, we performed Raman and X-ray photoelectron spectroscopy (XPS) studies. Figure 2a shows the Raman spectra collected for the Fe2O3, Ti-Fe2O3, and Ti-Fe2O3@PEDOT NRs. Compared to the Raman spectrum of pristine Fe2O3, the characteristic Raman peaks of the Ti-Fe2O3 sample were shifted towards more negative values and became more broadened (inset in Figure 2a), indicating that the sample possessed more oxygen vacancies after Ti doping.41 After coating with PEDOT the peak intensity of Fe2O3 reduced substantially and the peaks of PEDOT emerged, suggesting the formation of PEDOT on the surface of the Ti-Fe2O3 core. XPS survey spectra of the Fe2O3, Ti-Fe2O3, and Ti-Fe2O3@PEDOT samples are presented in Figure 2b, evidencing the existence of Ti (Ti 2p) and S (S 2p), which again confirms that Ti was included in the Fe2O3 and that PEDOT was successfully coated on the surface of the Ti-Fe2O3. The two broad peaks located at 464.7 and 458.9 eV in Figure S4 (Supporting Information) correspond to the characteristic Ti 2p1/2 and Ti 2p3/2 peaks of Ti4+, evidencing that the doped Ti is Ti4+.41 Figure 2c compares the Fe 2p core-level XPS spectra of the Fe2O3 and Ti-Fe2O3 samples. The Fe 2p3/2 and Fe 2p1/2 peaks of the two samples are in line with the typical peaks of Fe3+, revealing the presence of an Fe3+ species in the samples, which is further confirmed by the satellite peak of the Fe 2p3/2 line centered at 719.3 eV.17 Additionally, the peaks of the Fe 2p XPS spectrum for the Ti-Fe2O3 sample shifted towards more negative values and was broadened compared to those of the pristine Fe2O3 sample, which indicates the existence of Fe2+ (oxygen vacancies).17 Figure 2d shows the O 1s spectra of the three samples. Compared to Fe2O3, Ti-Fe2O3 exhibits an oxygen-defect peak (located at 531.8 eV) with higher intensity, which again suggests that the Ti-Fe2O3@PEDOT sample has more oxygen defects. To evaluate the electrochemical performance of the Ti-Fe2O3@PEDOT NRs as a supercapacitor electrode, electrochemical tests were carried out in a conventional three-electrode electrochemical cell with a 5 m LiCl solution as the electrolyte (Experimental Section). Figure 3a compares the cyclic voltammetry (CV) curves of the pristine Fe2O3, Ti-Fe2O3, and Ti-Fe2O3@PEDOT electrodes collected at 100 mV s−1. All these CV curves show a quasi-rectangular shape, suggesting that the capacitance of these electrodes consists mainly of double-layer capacitance and some surface Faradaic reaction. This is also consistent with other literature.42-44 As expected, the Ti-Fe2O3 electrode exhibited a considerably larger current density than the pristine Fe2O3 electrode, indicating an improvement of the electrochemical capacitance after Ti doping. Considering that the amount of Ti doping has a great influence on the donor density of Fe2O3, we further investigated the interplay between the capacitive performance of the Ti-Fe2O3 electrodes and the Ti content. As displayed in Figure 3b the areal capacitance of the Ti-Fe2O3 electrode first increased dramatically from 91.5 mF cm−2 to 163.6 mF cm−2 as the Ti content continuously increased, and then decreased when the Ti content was higher than 23.1%. Thus, the Ti-Fe2O3 electrode containing 23.1% Ti had the most optimal capacitive performance and was selected for coating with PEDOT. A substantial improvement of the capacitive current density and CV shape were observed for the Ti-Fe2O3@PEDOT electrode, demonstrating that the PEDOT shell can further enhance the electrochemical activity of Ti-Fe2O3 electrodes. In addition, CV curves of the Ti-Fe2O3@PEDOT electrode obtained at different scan rates ranging from 10 to 400 mV s−1 exhibited quasi-rectangular shapes with no significant change (Figure S5a, Supporting Information), revealing the ideal capacitive behavior of the Ti-Fe2O3@PEDOT electrode. The highest areal capacitance of 395.6 mF cm−2 was achieved for the Ti-Fe2O3@PEDOT electrode at 10 mV s−1, whereas only 368.7 mF cm−2 and 173.6 mF cm−2 were achieved for Ti-Fe2O3 and Fe2O3 electrodes, respectively, at the same current density. The galvanostatic charge–discharge curves of these three samples at a current density of 2 mA cm−2 are collected in Figure 3c. In comparison to the Fe2O3 and Ti-Fe2O3 electrode, more symmetrical charge–discharge curves and longer discharge times can clearly be identified for the Ti-Fe2O3@PEDOT electrode. This indicates that the Ti-Fe2O3@PEDOT electrode possesses a superior Coulombic efficiency and enhanced capacitance, and again confirms our hypothesis that Ti doping and PEDOT coating dramatically improve the capacitive performance of Fe2O3 electrodes. Figure 3d shows the relationship between the calculated areal capacitance of these electrodes and their discharge current density (detailed calculation see Supporting Information). Remarkably, the areal capacitance of the Ti-Fe2O3@PEDOT electrode is relatively larger than the values obtained for Fe2O3 and Ti-Fe2O3 electrodes at the same discharge current density. For instance, the calculated areal capacitance at 1 mA cm−2 for the Ti-Fe2O3@PEDOT electrode reached around 1.15 F cm−2, which is much higher than that of Fe2O3 (0.46 F cm−2) and Ti-Fe2O3 (0.81 F cm−2) electrodes. Additionally, it is an impressive value when compared to those of many previously reported Fe2O3-based electrodes, for instance, Fe2O3 nanotubes (0.18 F cm−2);29 V2O5-doped Fe2O3 (0.183 F cm−2);45 oxygen-deficient Fe2O3 (0.31 F cm−2);17 Fe2O3/C nanocomposites (0.59 F cm−2);46 and Fe2O3/N-rGO composites (1.24 F cm−2).47 Furthermore, the Ti-Fe2O3@PEDOT electrode exhibited a substantially enhanced rate capability, which led to an excellent capacitance retention of 66.8% when the current density increased from 1 to 8 mA cm−2, which is higher than that of Fe2O3 electrode and Ti-Fe2O3 electrodes. It is worth noting that the Ti-Fe2O3@PEDOT electrode has ultrahigh electrochemical durability with 96.1% areal capacitance retention after 30 000 cycles (Figure 3e), which is markedly higher than that of Ti-Fe2O3 (80.7%) and Fe2O3 (81.8%) electrodes. As far as we know, this is the best cyclic stability ever reported for Fe2O3-based electrodes.17, 45-47 There was almost no change in the CV curve of the Ti-Fe2O3@PEDOT electrode after 30 000 cycles, which further demonstrates the excellent cycling stability of Ti-Fe2O3@PEDOT. These results convincingly confirm that the PEDOT layer can remarkably enhance the conductivity and durability of Fe2O3 electrodes. To better clarify the boosted electrochemical properties of the Ti-Fe2O3@PEDOT electrode, electrochemical impedance spectroscopy (EIS) was conducted. The Nyquist plots for the three electrodes are presented in Figure 3f. All these plots are made up of two parts, a semicircle and a straight slope in the high- and low-frequency regions, respectively. The semicircle is often considered to be related to an electron-transfer-limited process, whereas the straight slope corresponds to a diffusion-limited electron-transfer process. From the plots, the charge-transfer resistance (fitted according to the equivalent circuit as shown in Figure S6, Supporting Information) of the Ti-Fe2O3@PEDOT electrode was the smallest, and that of Ti-Fe2O3 was the second smallest. This clearly reveals that the conductivity of Fe2O3 can be greatly improved by Ti doping and coating with a PEDOT layer, which was also confirmed by internal resistance analysis. Figure S5b shows the iR drops of these electrodes, which originate from the discharge curves plotted as a function of current density. In comparison to the Fe2O3 and Ti-Fe2O3 electrodes, the slope of the iR drop plots of the Ti-Fe2O3@PEDOT electrode is substantially less steep. As the slope is proportional to the equivalent series resistance (ESR) of the electrodes this again confirms the enhanced conductivity of the electrode. Moreover, the Fe2O3, Ti-Fe2O3, and Ti-Fe2O3@PEDOT electrodes should possess similar ion-diffusion rates as they have similar morphologies. Therefore, the superior capacitive performance and cycling durability of the Ti-Fe2O3@PEDOT electrode can be ascribed to the following benefits: 1) Ti-doping of Fe2O3 results in a higher specific surface area, leading to a better use of the pseudocapacitance of the electrode; 2) Ti doping introduces oxygen vacancies, which serve as donors to boost the electric conductivity; 3) the highly conductive and stabilized PEDOT layer not only relaxes the transfer of electrons, but also acts as a protective layer to protect Fe2O3 from degradation during long-term cycling, contributing to excellent cycling performance. To demonstrate the feasibility of the as-prepared Ti-Fe2O3@PEDOT electrode as high-performance ASC anode, a flexible solid-state ASC device based on Ti-Fe2O3@PEDOT as the anode and a MnO2 cathode was fabricated (details see Experimental Section). For this, MnO2 was coated on a carbon cloth as described in the literature.9 The SEM images and 2p XPS spectrum are displayed in Figure (Supporting Information). Figure shows a uniform MnO2 was coated on the carbon The 2p1/2 peak and 2p3/2 located at eV and eV are consistent with MnO2 reported CV and galvanostatic curves of MnO2 electrode are presented in Figure in the Supporting the MnO2 electrode achieved a large areal capacitance of mF cm−2 at a scan rate of 10 mV s−1 and mF cm−2 at a discharge current density of 1 mA These results convincingly reveal that the as-prepared MnO2 electrode has an excellent capacitive performance. to the ASC the charge between the MnO2 cathode and the Ti-Fe2O3@PEDOT anode needed to be and the areal of these two electrodes was calculated to be (details see Supporting Information and Figure Figure shows the CV curves of the ASC device collected at 100 mV s−1 with an voltage ranging from to indicating that the ASC device is stable up to an voltage of Figure displays the CV curves of the ASC device at scan All of the CV curves exhibited and quasi-rectangular shapes that are of a good and typical capacitive Additionally, all the charge–discharge curves at different current density show further confirming the superior capacitive performance of the ASC device (Figure As shown in Figure the ASC device achieved a maximum capacitance of F cm−3 at a discharge current density of 1 mA cm−2, which is substantially larger than that of previously reported such as ASC F ASC F and ASC F Furthermore, the ASC device retained a rate capability of at a high discharge current density of 8 mA Moreover, as shown in Figure the CV curves of our ASC device almost show no change different which reveals that our device has excellent This result indicates its properties and potential for flexible energy-storage devices. The cyclic durability of the ASC device was further at 100 mV s−1 for cycles and it was that the device of the initial capacitance, revealing its good cycling performance (Figure Supporting Information). Figure 5 displays the plots of our ASC and some values of previously reported are also for our ASC device was to an energy density of 0.89 mWh cm−3 at a current density of 1 mA cm−2, and can hold at an energy density of mWh cm−3 at 8 mA cm−2, further demonstrating that the ASC device an outstanding rate capability. Such energy density value is substantially higher than that of reported ASC such as mWh ASCs mWh ASC mWh and ASC mWh and to that of ASC mWh the energy density of the ASC device is a lower than the the maximum power density of 0.44 W cm−3 that is by the ASC device is much larger than the reported ASCs In an strategy has been reported to remarkably boost the capacitive performance and cycling durability of Fe2O3 Ti doping and PEDOT Compared to a pristine Fe2O3 electrode and a Ti-Fe2O3 electrode, the Ti-Fe2O3@PEDOT electrode an enhanced areal capacitance of 1.15 F cm−2 at 1 mA Moreover, the Ti-Fe2O3@PEDOT electrode exhibits exceptionally cyclic stability that can more than 96% after 30 000 cycles, which is the best cycling performance achieved far by a Fe2O3-based electrode. Additionally, a flexible high-performance ASC device based on the Ti-Fe2O3@PEDOT electrode as the anode and a MnO2 electrode as the cathode a maximum energy density of 0.89 mWh cm−3 and an impressive rate capability. The of stable and Fe2O3-based electrodes new for Fe2O3 materials in high-performance energy-storage devices. of Fe2O3 and Fe2O3 Ti-doped Fe2O3 samples were synthesized by a previously reported method with slight m and 1 m to a solution to which 8 was as Ti source, by the solution was to about After the was into a with a of A of carbon cloth 3 was into the solution in the The was and at for 1 and then to After the sample was it was with and at in After the sample was further at in for 1 to the Ti-doped Fe2O3 For comparison Fe2O3 was synthesized in the same The of Ti-doped Fe2O3 and Fe2O3 was obtained by of PEDOT was coated Ti-doped Fe2O3 by of using a electrochemical The was carried out in a solution consisting of and at for 5 at The of PEDOT was by of MnO2 on Carbon MnO2 was synthesized by using a electrochemical The electrolyte for MnO2 was obtained by and in and the MnO2 was deposited by a voltage of for at of ASC The solid-state were using MnO2 as the cathode and Ti-Fe2O3@PEDOT as the anode with a and the polymer electrolyte was To the charge between the two electrodes, the area of MnO2 electrode to Ti-Fe2O3@PEDOT electrode was to be The electrolyte was fabricated by LiCl and into and at for 1 the electrodes and the were into the electrolyte and then the electrolyte was to at for was and at for to in the The area and thickness of the fabricated were about 1 and and The and of the were by SEM Raman spectroscopy XPS and X-ray The surface area of the sample was obtained from at that were on an to the samples were at 100 for 5 voltammetry galvanostatic and electrochemical impedance spectroscopy were carried out using an electrochemical The electrochemical of the electrodes were performed in a three-electrode cell in 5 m LiCl solution within the potential of approximately to The electrode and electrode were a and a electrode respectively. This work was by the of and the of and for and the of As a to our and this by the Such materials are and be for but are not from than should be to the The is not for the content functionality of by the than should be to the for the