Electrochemical Performance of Iron Oxide Nanoﬂakes on Carbon Cloth under an External Magnetic Field

: In this work, the iron oxide (Fe 2 O 3 ) nanoﬂakes on carbon cloth (Fe 2 O 3 @CC) were triumphantly prepared and served as the electrode of supercapacitors. By applying an external magnetic ﬁeld, we ﬁrst ﬁnd that the magnetic ﬁeld could suppress the polarization phenomenon of electrochemical performance. Then, the inﬂuences of the mono-/bi-valent cations on the electrochemical properties of the Fe 2 O 3 @CC were investigated under a large external magnetic ﬁeld (1 T) in this work. The chemical valences of the cations in the aqueous electrolytes (LiNO 3 and Ca(NO 3 ) 2 ) have almost no inﬂuences on the speciﬁc capacitance at different scan rates. As one of important parameters to describe the electrochemical properties, the working potential window of the Fe 2 O 3 @CC electrode was also investigated in this work. The broad potential window in room-temperature molten salt (LiTFSI + LiBETI (LiN(SO 2 CF 3 ) 2 + LiN(SO 2 C 2 F 5 ) 2 )) has been obtained and reached 1.2 V, which is higher than that of the traditional aqueous electrolyte (~0.9 V).


Introduction
Electrochemical capacitors have been considered as a reliable supplement in power sources aspect, which is due to their fast charging/discharging processing, good cycling properties and high power density [1][2][3]. Nevertheless, their low energy density becomes a stumbling block to the rapid development of supercapacitors. In order to get higher energy density several strategies have been explored to design and fabricate electrode materials with various porous and hierarchical nanostructures, which are conducive to enhance specific capacitances and modify the rate performances [4][5][6][7][8]. Iron oxides (such as Fe 2 O 3 and Fe 3 O 4 ) have been brought into focus as electrode candidates for supercapacitors in a variety of materials [9][10][11][12]. Specifically, α-Fe 2 O 3 materials have been synthesized and reported elsewhere because of their high theoretical capacitance, low cost, and abundant chemical valences. Currently, various α-Fe 2 O 3 nanostructures, for instance nanoparticles, nanotubes, nanosheets, and nanospheres with high super-paramagnetic behaviors, porous structure, and high surface area have been researched and exhibited excellent electrochemical properties [13][14][15][16][17]. However, the weak electrical conductivity of α-Fe 2 O 3 materials (10 −4 Ω −1 ·m −1 for bulk and 2.5 × 10 −3 Ω −1 ·m −1 for nanostructures) still makes it difficult to completely use the active material, giving rise to finite enhancement of the specific capacitance [18][19][20]. There is literature reporting that very small external magnetic fields (mT) have induced capacitance enhancement of the Fe 2 O 3 -graphene nanocomposites [21]. This could explain why the external magnetic field is deemed to facilitate the electrons/ions transmission between the electrolyte and the electrode during charging/discharging procedure. However, one wonders if the very large magnetic field has an effect on the capacitance. More recently, a finer morphology of nanoporous magnetic material has been prepared by applying an external magnetic field (~0.5 T) which modulates different dealloying stages of the nanoporous structure formation. We have reported the effect of an external magnetic field on the dealloying process of the Ni-Al alloy in alkaline solution [22].
Munichandraiah et al. found that the aqueous electrolytes with bivalent cations (Ca 2+ or Mg 2+ ) contributed to modify the MnO 2 electrochemical utilization and, thus, improved the specific capacitance [23,24]. However, the common aqueous electrolytes cannot overcome the thermodynamic limitation of water splitting,~1.23 V [25,26], and the non-aqueous electrolytes (e.g., organic electrolytes) are generally combustible and expensive. Over the years, the inventive "water-in-salt" hybrids and room-temperature molten salts have been proven to broad the working voltage window (much larger than 1.23 V) [27][28][29]. Herein, this work investigates the influences of the mono-/bi-valent cations on the electrochemical properties of the Fe 2 O 3 @carbon cloth (Fe 2 O 3 @CC) under an external magnetic field (1 T). The electrochemical performances of the Fe 2 O 3 @CC electrode in a room-temperature molten salt electrolyte have been also studied for the broad working potential window.

Materials and Methods
The carbon cloth (WOS 1002 PHYCHEMi Co., Ltd., Wuhan, China) in the work was conducted the hydrophilic treatment. The detailed process is following: the carbon cloth was washed with concentrated H 2 SO 4 and H 2 O 2 mixed solution, and deionized water under sonication for 2 h. The treated carbon cloth (CC) served as the substrate for the growth of Fe 2 O 3 nanoflakes. Electrochemical deposition was performed at 1.5 V for 5 min in a conventional three-electrode system: the treated CC, the Pt, and an Ag/AgCl as the working, counter and reference electrodes, respectively. The working electrolyte was 0.05 M FeCl 2 aqueous solution for iron deposition. Then the working electrode was washed with deionized water and ethanol, and dried for a night. Lastly, the dried precursors were calcinated at 650 • C for 2 h in an inert atmosphere.
The structural information, constitute and morphology features of the Fe 2 O 3 @CC composites were determined by X-ray diffraction (XRD, Rigaku D/Max-2500, Japan) with native Standard Measurement software, and field emission scanning electron microscopy (FESEM, FEI Verios 460 L, Eindhoven, The Netherlands). The electrochemical measurements were carried out in a three-electrode system: the Fe 2 O 3 @CC (5 × 5 mm), the Pt (50 × 50 mm) and Ag/AgCl were used as working electrode, counter electrode, and reference electrode, respectively. All electrochemical measurements were carried out under an external magnetic field (EMP-5, East Changing, Beijing, China). The magnetic field was controlled through adjusting the magnitude of the current and the magnetic flux density was then measured by a Digital Gauss Meter (PF-045B, Mianyang, China). The direction of magnetic field was perpendicular to the surface of working electrode. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) of the Fe 2 O 3 @CC electrode were performed using electrochemical workstation (CHI 760E, Shanghai, China) in different aqueous electrolytes of 1 M LiNO 3 , 1 M Li 2 SO 4 , 1 M KNO 3 , 1 M Ca(NO 3 ) 2 , and 1 M LiTFSI + LiBETI.

Results and Discussion
The phase and constitute of the composites were characterized by XRD analysis and the corresponding result is shown in Figure 1. The distinct diffraction peak located at around 25 • can be ascribed to the carbon cloth substrate. Although the intensity of the remaining peaks is very weak, the peak positions can be matched well with the   Figure 2 shows the microstructure of the Fe2O3@CC composite electrode. After electrochemical deposition, the whole braided structure of the carbon cloth is successfully maintained and the diameter of the carbon fiber is approximately unchanged. Comparing to the smooth surface of the carbon cloth, the Fe2O3@CC composite exhibits a rough surface, which indicates the successful deposition of Fe2O3. In Figure 2b, the flimsy Fe2O3nanoflakes with around 20 nm thickness are interconnecting and uniformly decorate the carbon fiber. To verify the external magnetic field effect on the electrochemical performances of the Fe2O3@CC electrode, the CVs were carried out in different electrolytes with or without applying an external magnetic field and the results are shown in Figure 3. In Figure 3a, it is seen that the contribution from the CC substrate can be negligible and the specific capacitance of the Fe2O3@CC electrode is mainly from the contribution of Fe2O3. In terms of CV shape, the specific capacitance of the Fe2O3@CC electrode is the combination of the double layer and pseudo-capacitive characteristic [30][31][32]. The specific capacitances of the Fe2O3@CC electrode are almost the same by integrating the enclosed area with or without the external magnetic field. But their shapes of the CV curves have some differences to a certain extent. Concretely, the polarization phenomenon of the CV plot is obtaining efficient control when adding the external magnetic field. That is to say, the splitting of water in the electrolyte is suppressed by the external magnetic field. Therefore, the following electrochemical tests in different aqueous electrolytes are conducted under the external magnetic field. Figure 3 depicts CV plots of the Fe2O3@CC electrode in various aqueous electrolytes of 1 M Li2SO4, 1 MLiNO3, 1 MKNO3, and 1 MCa(NO3)2, for exploring the better specific capacitance.The measurements were performed at the magnetic flux density (1 T). Among the CV in monovalent cationic electrolytes, the one in LiNO3 solution presents the largestenclosed area. The CV plot shapes are nearly the same for the bivalent   Figure 2 shows the microstructure of the Fe2O3@CC composite electrode. After electrochemical deposition, the whole braided structure of the carbon cloth is successfully maintained and the diameter of the carbon fiber is approximately unchanged. Comparing to the smooth surface of the carbon cloth, the Fe2O3@CC composite exhibits a rough surface, which indicates the successful deposition of Fe2O3. In Figure 2b, the flimsy Fe2O3nanoflakes with around 20 nm thickness are interconnecting and uniformly decorate the carbon fiber. To verify the external magnetic field effect on the electrochemical performances of the Fe2O3@CC electrode, the CVs were carried out in different electrolytes with or without applying an external magnetic field and the results are shown in Figure 3. In Figure 3a, it is seen that the contribution from the CC substrate can be negligible and the specific capacitance of the Fe2O3@CC electrode is mainly from the contribution of Fe2O3. In terms of CV shape, the specific capacitance of the Fe2O3@CC electrode is the combination of the double layer and pseudo-capacitive characteristic [30][31][32]. The specific capacitances of the Fe2O3@CC electrode are almost the same by integrating the enclosed area with or without the external magnetic field. But their shapes of the CV curves have some differences to a certain extent. Concretely, the polarization phenomenon of the CV plot is obtaining efficient control when adding the external magnetic field. That is to say, the splitting of water in the electrolyte is suppressed by the external magnetic field. Therefore, the following electrochemical tests in different aqueous electrolytes are conducted under the external magnetic field. Figure 3 depicts CV plots of the Fe2O3@CC electrode in various aqueous electrolytes of 1 M Li2SO4, 1 MLiNO3, 1 MKNO3, and 1 MCa(NO3)2, for exploring the better specific capacitance.The measurements were performed at the magnetic flux density (1 T). Among the CV in monovalent cationic electrolytes, the one in LiNO3 solution presents the largestenclosed area. The CV plot shapes are nearly the same for the bivalent To verify the external magnetic field effect on the electrochemical performances of the Fe 2 O 3 @CC electrode, the CVs were carried out in different electrolytes with or without applying an external magnetic field and the results are shown in Figure 3. In Figure 3a, it is seen that the contribution from the CC substrate can be negligible and the specific capacitance of the Fe 2 O 3 @CC electrode is mainly from the contribution of Fe 2 O 3 . In terms of CV shape, the specific capacitance of the Fe 2 O 3 @CC electrode is the combination of the double layer and pseudo-capacitive characteristic [30][31][32]. The specific capacitances of the Fe 2 O 3 @CC electrode are almost the same by integrating the enclosed area with or without the external magnetic field. But their shapes of the CV curves have some differences to a certain extent. Concretely, the polarization phenomenon of the CV plot is obtaining efficient control when adding the external magnetic field. That is to say, the splitting of water in the electrolyte is suppressed by the external magnetic field. Therefore, the following electrochemical tests in different aqueous electrolytes are conducted under the external magnetic field. Figure 3 depicts CV plots of the Fe 2 O 3 @CC electrode in various aqueous electrolytes of 1 M Li 2 SO 4 , 1 MLiNO 3 , 1 MKNO 3 , and 1 MCa(NO 3 ) 2 , for exploring the better specific capacitance.The measurements were performed at the magnetic flux density (1 T). Among the CV in monovalent cationic electrolytes, the one in LiNO 3 solution presents the largestenclosed area. The CV plot shapes are nearly the same for the bivalent case, suggesting the same electrochemical procedure. Moreover, the enclosed areas in CV for the bivalent casepossess larger enclosed area than that in LiNO 3 solution. case, suggesting the same electrochemical procedure. Moreover, the enclosed areas in CV for the bivalent casepossess larger enclosed area than that in LiNO3 solution. The variation of the corresponding specific capacitance with the scan rates of the Fe2O3@CC electrode in different electrolytes has exhibited in Figure 4. The calculated specific capacitance of the Fe2O3@CC electrode measured in 1 M Li2SO4 electrolyte are 52, 40, 29, 26, 23, and 19 mF cm −2 at 1, 2, 5, 8, 10, and 20 mV s −1 , respectively. As reported in the literature, the specific capacitances reduce with increase in the scan rate in any electrolytes [33][34][35][36]. The situation could be seen in other electrolytes. The values of specific capacitance obtained in 1 M LiNO3 solution are 53, 44, 28, 24, 24, and 18 mF·cm −2 , respectively. However, these values obtained in 1 M KNO3 solution decrease to 30,24,20,17,15, and 9 mF·cm −2 , respectively. The slightly higherspecific capacitance of Fe2O3@CC electrodeobtained in lithium ions electrolyte could be ascribed to more effortless migration for lithium ions. It possesses the applicable crystal size (0.69 Å for Li + , 1.38 Å for K + , and 0.99 Å for Ca 2+ ) [37]. Simultaneously, the lower saturated concentration of the KNO3 solution (only 3.1 M, about 10 M for LiNO3) results in its lowest specific capacitances at all scan rates. Interestingly, the specific capacitance, variation trend and retention are practically consistent in both LiNO3 and Ca(NO3)2 solutions. Thus, the different valence of the cations in the electrolytes has no effects on the electrochemical performances of the Fe2O3@CC electrode. The variation of the corresponding specific capacitance with the scan rates of the Fe 2 O 3 @CC electrode in different electrolytes has exhibited in Figure 4. The calculated specific capacitance of the Fe 2 O 3 @CC electrode measured in 1 M Li 2 SO 4 electrolyte are 52, 40, 29, 26, 23, and 19 mF cm −2 at 1, 2, 5, 8, 10, and 20 mV s −1 , respectively. As reported in the literature, the specific capacitances reduce with increase in the scan rate in any electrolytes [33][34][35][36]. The situation could be seen in other electrolytes. The values of specific capacitance obtained in 1 M LiNO 3 solution are 53, 44, 28, 24, 24, and 18 mF·cm −2 , respectively. However, these values obtained in 1 M KNO 3 solution decrease to 30, 24, 20, 17, 15, and 9 mF·cm −2 , respectively. The slightly higherspecific capacitance of Fe 2 O 3 @CC electrodeobtained in lithium ions electrolyte could be ascribed to more effortless migration for lithium ions. It possesses the applicable crystal size (0.69 Å for Li + , 1.38 Å for K + , and 0.99 Å for Ca 2+ ) [37]. Simultaneously, the lower saturated concentration of the KNO 3 solution (only 3.1 M, about 10 M for LiNO 3 ) results in its lowest specific capacitances at all scan rates. Interestingly, the specific capacitance, variation trend and retention are practically consistent in both LiNO 3   respectively. Thereis not strictly parallel to each other in the low-frequencies, indicating similar Warburg impedance. According to the literatures, the Rct value could be usually linked to the diffusive resistance and the distributed resistance of the electrode material, which hints the negligible influences by the nature of electrolyte [25]. Therefore, there is no distinct differences on the EIS between LiNO3 and Ca(NO3)2 solution. Lately, Yamada et al. explored a room-temperature hydrate melt of Li salts (LiTFSI + LiBETI) served as aqueous electrolyte, which expanded the stable working voltage window [28]. In this work, 1 M LiTFSI + LiBETI aqueous solution was configured, and used as the electrolyte to investigate the electrochemical performances compared with 1 M LiNO3 aqueous solution. Firstly, Figure 6a depicts CV curves of the Fe2O3@CC electrode with different working voltage window in 1 M LiTFSI + LiBETI 35 Ω, respectively. Thereis not strictly parallel to each other in the low-frequencies, indicating similar Warburg impedance. According to the literatures, the R ct value could be usually linked to the diffusive resistance and the distributed resistance of the electrode material, which hints the negligible influences by the nature of electrolyte [25]. Therefore, there is no distinct differences on the EIS between LiNO 3 and Ca(NO 3 ) 2 solution.  respectively. Thereis not strictly parallel to each other in the low-frequencies, indicating similar Warburg impedance. According to the literatures, the Rct value could be usually linked to the diffusive resistance and the distributed resistance of the electrode material, which hints the negligible influences by the nature of electrolyte [25]. Therefore, there is no distinct differences on the EIS between LiNO3 and Ca(NO3)2 solution. Lately, Yamada et al. explored a room-temperature hydrate melt of Li salts (LiTFSI + LiBETI) served as aqueous electrolyte, which expanded the stable working voltage window [28]. In this work, 1 M LiTFSI + LiBETI aqueous solution was configured, and used as the electrolyte to investigate the electrochemical performances compared with 1 M LiNO3 aqueous solution. Firstly, Figure 6a depicts CV curves of the Fe2O3@CC electrode with different working voltage window in 1 M LiTFSI + LiBETI Lately, Yamada et al. explored a room-temperature hydrate melt of Li salts (LiTFSI + LiBETI) served as aqueous electrolyte, which expanded the stable working voltage window [28]. In this work, 1 M LiTFSI + LiBETI aqueous solution was configured, and used as the electrolyte to investigate the electrochemical performances compared with 1 M LiNO 3 aqueous solution. Firstly, Figure 6a depicts CV curves of the Fe 2 O 3 @CC electrode with different working voltage window in 1 M LiTFSI + LiBETI electrolyte. Through expanding the negative voltage from −0.9 to −1.4 V, we investigate the stable working voltage window of the LiTFSI + LiBETI electrolyte. The CV shape always maintains rectangular-like when expanding the negative voltage from −0.9 to −1.2 V. With further expanding the negative voltage, the response current increases rapidly, hinting the splitting of water in electrolyte. Thus, the stable working voltage window is −1.2 to 0 V, while it is −0.9 to 0 V in the above aqueous electrolyte. Then the stability of the Fe 2 O 3 @CC electrode in 1 M LiTFSI + LiBETI is in comparison with that in 1 M LiNO 3 solution (Figure 6b). The primary rectangular-like shapes are similar, while a severe polarizationoccurred in 1 M LiNO 3 solution due to the easier ions migration than 1 M LiTFSI + LiBETI with relatively high viscosity. This finding is in accordance with the expanded working voltage window obtained for the room-temperature hydrate melt of Li salts electrolyte. electrolyte. Through expanding the negative voltage from −0.9 to −1.4 V, we investigate the stable working voltage window of the LiTFSI + LiBETI electrolyte. The CV shape always maintains rectangular-like when expanding the negative voltage from −0.9 to −1.2 V. With further expanding the negative voltage, the response current increases rapidly, hinting the splitting of water in electrolyte. Thus, the stable working voltage window is −1.2 to 0 V, while it is −0.9 to 0 V in the above aqueous electrolyte. Then the stability of the Fe2O3@CC electrode in 1 M LiTFSI + LiBETI is in comparison with that in 1 M LiNO3 solution (Figure 6b). The primary rectangular-like shapes are similar, while a severe polarizationoccurred in 1 M LiNO3 solution due to the easier ions migration than 1 M LiTFSI + LiBETI with relatively high viscosity. This finding is in accordance with the expanded working voltage window obtained for the room-temperature hydrate melt of Li salts electrolyte. The variations of the specific capacitances in these two different electrolytes with the scan rates are exhibited in Figure 6c to illustrate the electrolytes effects. Not surprisingly, the variation trends of the specific capacitances are similar, and the specific capacitances decrease with increasing the scan rates. Moreover, the specific capacitances in 1 M LiNO3 at all scan rates are higher than that in 1 M LiTFSI + LiBETI, which is due to the lower ionic conductivity,slower ions transportation and diffusion limitation between the electrode and electrolyte. However, the capacitance retention in 1 M LiTFSI + LiBETI is higher than that in 1 M LiNO3, that is to say, the capacitance attenuation in traditional Lisalt aqueous electrolytes is more significant.
Likewise, the EIS (after fitting) of the Fe2O3@CC electrode in 1 M LiTFSI + LiBETI electrolyte has been measured to contrast with that in 1 M LiNO3. The whole shapes are similar, which are constituted by a semi-circle and a straight line. By best fitting with the some equivalent circuit in The variations of the specific capacitances in these two different electrolytes with the scan rates are exhibited in Figure 6c to illustrate the electrolytes effects. Not surprisingly, the variation trends of the specific capacitances are similar, and the specific capacitances decrease with increasing the scan rates. Moreover, the specific capacitances in 1 M LiNO 3 at all scan rates are higher than that in 1 M LiTFSI + LiBETI, which is due to the lower ionic conductivity, slower ions transportation and diffusion limitation between the electrode and electrolyte. However, the capacitance retention in 1 M LiTFSI + LiBETI is higher than that in 1 M LiNO 3 , that is to say, the capacitance attenuation in traditional Li-salt aqueous electrolytes is more significant.
Likewise, the EIS (after fitting) of the Fe 2 O 3 @CC electrode in 1 M LiTFSI + LiBETI electrolyte has been measured to contrast with that in 1 M LiNO 3 . The whole shapes are similar, which are constituted by a semi-circle and a straight line. By best fitting with the some equivalent circuit in Figure 5, the R s values for the Fe 2 O 3 @CC electrode in LiNO 3 and LiTFSI + LiBETI solution are 2.32 and 3.72 Ω, respectively, which are approximatelyequal. The R ct values in LiNO 3 and LiTFSI + LiBETI solution are 287.12 and 379.28 Ω, respectively. The higher charge transfer resistance may be attributed to the high viscosity and the low ionic conductivity. Thus the straight line in 1 M LiTFSI + LiBETI is more inclined, suggesting larger Warburg impedance. [38] And above all, the EIS plot in 1 M LiTFSI + LiBETI indicates a higher resistance and more difficult ions diffusion than that in 1 M LiNO 3 .

Conclusions
In summary, the electrochemical properties of the prepared Fe 2 O 3 @CC electrode have been investigated in various aqueous electrolytes with or without external magnetic field. There is no distinct influence of the larger external magnetic field (1 T) on the specific capacitance, while the external magnetic field suppresses the water splitting. In addition, the specific capacitances at all scan rates of the Fe 2 O 3 @CC electrode in usual aqueous electrolytes have been virtually unaffected by the chemical valences of the cations. Finally, the Fe 2 O 3 @CC electrode in 1 M LiTFSI + LiBETI electrolyte exhibits a broad working voltage window (1.2 V), but shows slightly low specific capacitances at all scan rates. This may be due to the high viscosity and low conductivity of the room-temperature molten salts.