C composite as high-performance anode material for alkaline nickel–iron rechargeable batteries

C composite as high-performance anode material for alkaline nickel–iron rechargeable batteries

Journal of Power Sources 291 (2015) 29e39 Contents lists available at ScienceDirect Journal of Power Sources journal homepage: www.elsevier.com/loca...

4MB Sizes 1 Downloads 89 Views

Journal of Power Sources 291 (2015) 29e39

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

FeS/C composite as high-performance anode material for alkaline nickeleiron rechargeable batteries Enbo Shangguan a, b, *, Fei Li a, Jing Li a, Zhaorong Chang a, Quanmin Li a, **, Xiao-Zi Yuan c, Haijiang Wang c a

Collaborative Innovation Center of Henan Province for Green Manufacturing of Fine Chemicals, School of Chemistry and Chemical Engineering, Henan Normal University, Xinxiang, 453007, PR China Post-Doctoral Research Center of Henan Huanyu Group Co. Ltd., Xinxiang, 453002, PR China c National Research Council of Canada, Vancouver, BC, V6T 1W5, Canada b

h i g h l i g h t s  The FeS/C composite was successfully synthesized via a simple route.  FeS/C is first evaluated as host anode materials for nickeleiron batteries.  The FeS/C composite exhibits attractive rate capability even at a high rate of 5C.  The FeS/C composite exhibits superior electrochemical cycle stability.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 December 2014 Received in revised form 10 March 2015 Accepted 7 May 2015 Available online 15 May 2015

FeS and its composite, FeS/C, are synthesized via a simple calcination method followed by a coprecipitation process. The electrochemical properties of the bare FeS and FeS/C composite as anode materials for alkaline nickeleiron batteries are investigated. The results show that the FeS/C-3wt%Bi2O3mixed electrode delivers a high specific capacity of 325 mAh g1 at a current density of 300 mA g1 with a faradaic efficiency of 90.3% and retains 99.2% of the initial capacity after 200 cycles. For the first time, it is demonstrated that even at a discharge rate as high as 1500 mA g1 (5C) the FeS/C-3wt%Bi2O3-mixed electrode delivers a specific capacity of nearly 230 mAh g1. SEM results confirm that after 200 discharge echarge cycles, the size of FeS/C particles reduces from 5 to 15 mm to less than 300 nm in diameter and the particles are highly dispersed on the surface of carbon black, which is likely caused by the dissolution-deposition process of Fe(OH)2 and Fe via intermediate iron species. As a result, the FeS/C composite exhibits considerably high charge efficiency, high discharge capacities, excellent rate capability and superior cycling stability. We believe that this composite is a potential candidate of highperformance anode materials for alkaline iron-based rechargeable batteries. © 2015 Elsevier B.V. All rights reserved.

Keywords: Anode material Nickeleiron rechargeable batteries Iron electrode Electrochemical properties

1. Introduction Because of their high energy efficiency and scalability, rechargeable batteries are especially suitable for large-scale storage of electrical energy [1e5]. Many battery systems such as lithium-

* Corresponding author. Collaborative Innovation Center of Henan Province for Green Manufacturing of Fine Chemicals, School of Chemistry and Chemical Engineering, Henan Normal University, Xinxiang, 453007, PR China. ** Corresponding author. E-mail addresses: [email protected], [email protected] (E. Shangguan). http://dx.doi.org/10.1016/j.jpowsour.2015.05.019 0378-7753/© 2015 Elsevier B.V. All rights reserved.

ion, lead-acid, nickel-metal hydride (Ni-MH) batteries are commercially available and being tested for large-scale energy storage applications [5]. From a cost-efficiency perspective, the simultaneous demands on cost and durability present a tremendous challenge for the smart grid application of these battery systems. Unfortunately, almost none of them are sufficiently robust or cost-effective to respond to the growing market needs of load leveling, peak shaving and micro-grids [5]. Compared with other battery materials, iron-based material is cost effective, globally abundant, and non-toxic, and has a large theoretical specific capacity [5e10]. Thus, iron-based rechargeable battery systems using an alkaline electrolyte such as nickeleiron and iron-air batteries are

30

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

particularly attractive as sustainable and inexpensive solutions for large-scale electrical energy storage owing to their low cost, ecofriendliness, safety and long-term durability [4e10]. However, the iron electrode in such iron-based batteries always suffers from the intrinsic drawbacks of relatively low efficiency [4], considerable hydrogen evolution [5,6], high self-discharge [11,12] and low energy and power densities [4]. One of the principal limitations of the iron-based electrode is its passivation caused by iron hydroxide produced during the discharge process, preventing further anodic utilization and leading to a poor performance at high discharge rates [3,4]. As a consequence, a practical capacity of only about one-third of the theoretical value (962 mAh g1) can be achieved [3,5,13,14]. Also, the iron electrode has a low hydrogen overpotential, which makes overcharge necessary and gives rise to poor charging efficiency of the iron electrode, further limiting its applications in commercial batteries. Further, hydrogen evolution occurring on the iron electrode at open circuit can lead to the corrosion of the iron electrode and consequently a high self-discharge rate [15]. Therefore, mitigation of passivation and hydrogen evolution is a major challenge for a large-scale utilization of iron-based alkaline batteries. Considerable effort has been devoted to improving the electrochemical properties of iron electrodes for iron-based alkaline batteries, including new iron anode materials [16,17], anode additives [5,7,15,18], electrolyte additives [19,20], and nanosized materials [3,21e29]. For example, the performance of iron carbide (Fe3C) as a new anode active material for an alkaline battery has been reported with a low discharge capacity of below 200 mAh g1 [17]. Wang's group reported Fe2P nanoparticles as an anode material for a nickeleiron battery. The reversible discharge capacity of the Fe2P nanoparticles electrode reached 413 mAh g1 at a discharge current of 100 mA g1 with a low charge efficiency of 68.8% [16]. Recently, two types of FeOx-graphene hybrid with high-rate charge/ discharge ability or high discharge capacity have been successfully synthesized and applied as anode materials for nickeleiron batteries [29]. Nevertheless, these materials are still far from achieving the challenging cost goals for large-scale energy storages. As such, a cost effective iron material with high charge efficiency is required, which is meaningful for the iron electrode to exhibit its advantages for rechargeable alkaline batteries. Researchers have made considerable efforts on sulphur containing additives, such as bismuth sulphide [7,14], lead sulphide [30], iron sulphide [8,30e34], sodium sulphide and potassium sulphide [10,31,35e37]. The addition of these sulphur-containing substances greatly suppresses the hydrogen evolution and electrode passivation of the batteries and significantly increases the iron electrode capacity, especially at high discharge rates. Generally, the function of sulfide additives has been ascribed to the following factors:

These organo-sulfur compounds form strongly adsorbed layers on the iron electrode and block the electrochemical process of hydrogen evolution. Although various sulphides have been employed as functional additives to improve the discharge performance of the iron electrode, their roles are not yet adequately understood. Iron sulfide (FeS), having outstanding catalytic and electrochemical properties, has been intensively studied and has drawn particular research interests as a new type of anode material for lithium ion batteries [40e42]. Up to date, the reported FeS was mainly utilized as anode additives to increase hydrogen overpotential of the iron electrode and mitigate the electrode passivation in rechargeable alkaline batteries. For instance, Caldas et al. described the beneficial role of FeS in increasing the ionic conductivity of the passive film on the iron electrode [18]. Recently, Manohar et al. [5] reported a high-performance iron electrode based on carbonyl iron, bismuth oxide and iron sulfide additives. These electrodes exhibit a high charging efficiency of 92% and the electrodes are capable of being discharged at an impressive 3C rate. Reported literature on regarding the use of FeS as primary anode materials in iron electrodes is scarce. Recently, Mulder et al. [43] reported FeS electrode as an anode for a large-scale energy storage system. The FeS electrode prepared by ball milling method exhibit better cyclic stability and superior high-rate performance, but with a very low discharge capacity of only 120 mAh g1. Till date, the electrochemical characteristics of FeS in alkaline solution remain largely unknown. Hence, it is interesting to investigate the electrochemical properties of FeS in alkaline solutions and further enhance its performance for iron-based alkaline batteries. As is generally known, compositing with carbon is an effective modification technique to enhance the electrochemical performance of iron-based anode materials [3,22,23,25,28,44]. The coated carbon layer on the particle surface can not only facilitate electron transport, but also prevent electrochemical aggregation and maintain high capacity [45]. In this context, the main objective of this work is to prepare FeS/C anode materials using a low-cost approach and to explore the feasibility of using FeS/C as an anode material for alkaline batteries. In this work, FeS and its composite, FeS/C, as anode materials were prepared via a simple calcination method followed by a coprecipitation process. Characteristics of FeS/C as an anode material for alkaline batteries were investigated using electrochemical measurements, composition analysis and morphological observation. As expected, the as-prepared FeS/C composite exhibits a high

(1) modification of the iron electrode texture and morphology via the adsorption of sulphide ion at the electrode/electrolyte interface with subsequent incorporation into the oxide lattice [34,35]; (2) improvement of the Fe/Fe(OH)2 reaction rate and enhancement of the bulk conductivity and the anodic current density of the electrode, leading to improved high-rate ability [18,33,34,38]; (3) promotion of the iron compounds dissolution and prevention of the rapid electrode passivation, leading to improved cycleability [18,34,37,39]. Recently, S.R. Narayanan's group [19] has discovered that addition of organo-sulfur compounds to the aqueous alkaline electrolyte can significantly reduce the hydrogen evolution rate by 90%.

Fig. 1. XRD patterns of FeS and FeS/C composite.

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

31

Fig. 2. FE-SEM and EDS images of the FeS sample: (a) SEM image of FeS; (b) Bright-field FE-SEM image of FeS sample; FESEM-EDS analysis showing the distributions of Fe, S, and C of the FeS sample. Boxed area was scanned for Fe (c), S(d) and C (e) elemental mapping, respectively. (f) EDS spectra of the FeS sample.

Fig. 3. FESEM-EDS images of the FeS/C composite: (a) FESEM image of FeS/C composite; (b) Bright-field FE-SEM image of FeS/C composite; FESEM-EDS analysis showing the distributions of Fe, S, and C of the FeS/C composite. Boxed area was scanned for Fe (c), S(d) and C (e) elemental mapping, respectively. (f) EDS spectra of the FeS/C composite.

32

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

specific capacity with a high charge efficiency, excellent rate capability and superior cycling stability. 2. Experimental 2.1. Synthesis of FeS/C composite and bare FeS All the reagents used in this work are of analytical grade without further purification. In a typical synthesis of the FeS/C composite, 15 g of FeSO4$7H2O was dissolved in 150 mL distilled water under constant mechanical stirring. Then fresh Na2S$9H2O solution (15.5 g in 130 mL of distilled water) was added into the solutions under constant mechanical stirring. The mixture was under continuous stirring for 0.5 h at 30  C. A black precipitate was then formed. After the reaction completed, the precipitates in the solution were filtrated, washed several times with distilled water, and then dried in a vacuum environment at 80  C for 12 h. The above FeS powders was adopted as the raw material and glucose (C6H12O6) was used as the carbon source. The FeS powders and the carbon source were mixed in mass ratio of FeS: glucose ¼ 3:1. The mixtures were milled with a gyratory ball mill in ethanol for 6 h at a rotation speed of 300 r/min. Then, the obtained powders were pyrolyzed in a tube furnace at 700  C for 1 h in an inert atmosphere to obtain the final product of FeS/C composite. As comparison, bare FeS was also prepared by the identical procedures for FeS/C composite except for the addition of glucose. 2.2. Structure characterization

separate the cathode and anode. A solution of 6 M KOHþ15 g L1 LiOH was used as the electrolyte. Galvanostatic charge/discharge measurements were conducted using a Land CT2001A battery performance testing instrument (Wuhan LAND electronics Co. Ltd., China). For activation, thirty chargeedischarge cycles at a rate of 60 mA g1 were performed, and the electrodes were discharged to 0.8 V. To evaluate the rate performance, the cell was cycled at different current densities, varying from 60 to 1500 mA g1. The cut-off voltages were set at 0.8 V. The discharge capacity of the negative electrode was based on the amount of active material (FeS) without taking into account the conducting additives in the electrode. Cycle voltammogram (CV) and electrochemical impedance spectroscopy (EIS) were carried out on a Solartron SI 1260 impedance analyzer with a 1287 potentiostat interface at room temperature (25 ± 1  C) in 6 M KOHþ15 g L1 LiOH solution. A threeelectrode cell assembly was used in the test with a Hg/HgO electrode as reference electrode and a nickel ribbon as counter electrode. The CV test was scanned from 0.4 V to 1.25 V starting from 0.4 V to the negative direction, and the scanning rate was between 1 mVs1and 10 mVs1. 3. Results and discussion Fig. 1 shows the XRD patterns of the as-prepared FeS and FeS/C powders. All the diffraction peaks of the FeS in Fig. 1 can be exclusively attributed to FeS (JCPDS 37e0477), indicating the successful synthesis of FeS powders. The peaks at 2q ¼ 30.22 , 33.98 , 43.50 , and 53.40 correspond to the (110), (112), (114), and (300)

The phase structure of the samples was identified by a D8 X diffractometer (Germany, Bruker) with Cu Ka radiation. The particle morphology of the synthesized samples were characterized by field-emission scanning electron microscope (FE-SEM, S-3400NII) equipped with energy-dispersive X-ray spectrometry (EDS). Thermogravimetric (TG) and differential scanning calorimetry (DSC) analysis of the FeS and FeS/C composite was investigated on a NETZSCH STA 409 PC/PG thermal analyzer (NETZSCH, Germany) with a heating rate of 10  C per minute from room temperature to 800  C in an air atmosphere. 2.3. Electrochemical measurements The benefit of bismuth additives in reducing hydrogen evolution in nickeleiron batteries has been widely reported in the literature [3,5e7]. In accordance with the previous literature, we chose Bi2O3 as anode additives to further improve the performance of the FeS electrodes. The iron electrodes with Bi2O3 additives were prepared by incorporating slurries containing 82 wt.% active material, 10 wt.% carbon black, 3 wt.% Bi2O3, and 5 wt.% polytetrafluoroethylene (PTFE, 10 wt.%, in diluted emulsion, Tianjin city fine chemical research institute, China). The iron electrodes without Bi2O3 additive comprised 85 wt.% of active material, 10 wt.% carbon black, and 5 wt.% polytetrafluoroethylene. The admixtures were mixed homogeneously to form a slurry. Then, the slurry was poured into a foam nickel sheet (1.5 cm  1.5 cm) and dried in a vacuum coven at 80  C for 6 h with a total electroactive material loading of 110 mg per electrode. Subsequently, the pasted electrodes were pressed at 10 MPa to 0.5 mm thickness before measurement. The positive electrode was the commercially sintered Ni(OH)2 electrode (Henan Huanyu power sources Co. Ltd.) whose capacity was much higher than that of the iron electrodes in order to make full use of the active material in iron electrodes. Test cells were assembled using the prepared iron electrode as the anode, two commercial Ni(OH)2 electrode as the cathode, and polypropylene to

Fig. 4. TG and DSC plots for the as-prepared FeS(a) and FeS/C composite(b).

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

planes of the troilite FeS, respectively. No iron oxide impurites are detected. For the XRD spectrum of the FeS/C composite, the main peaks are similar to those of the bare FeS powders but with strong peak intensities, indicating a better crystallinity. The result may suggest that to some extent the carbon coating layer helps in enhancing the crystallinity of FeS in the calcination process. This result is similar to that of [email protected] reported by Fei et al. [40]. The morphology, structure, and composition of the synthesized bare FeS and FeS/C composite powders were investigated by FESEM and EDS, shown in Figs. 2 and 3, respectively. The images (Figs. 2 and 3a) show that both samples appear to be aggregates of similar irregular shapes, but the surface features differ greatly. For the bare FeS (Fig. 2a), its surfaces are glossy and smooth. In contrast, surfaces of the FeS/C composite particles (Fig. 3a) are rough, resulting from the carbon coating. To further confirm the surface composition of the bare FeS and FeS/C composite, two bright-field FE-SEM images and the corresponding EDS elemental maps are shown in Fig. 2 (bee) and Fig. 3 (bee), respectively, and the EDS spectra for the two samples are shown in Figs. 2 and 3f, respectively. Elemental mapping analysis suggests the presence of S, Fe and O components with uniform elemental distribution in the two samples. For both samples, the strong signals from Fe and S can be abundantly detected, exhibiting an S and Fe atomic ratio of 1:1. As shown in Fig. 2e, the C signal for the FeS sample is relatively low, resulting from the conducting resin. For the FeS/C composite (Fig. 3e), the C signal is much higher, mainly resulting from the carbon coating other than the conducting resin. The result clearly indicates that carbon is indeed coated onto the surface of FeS. In addition, it can be found that the Na signal in the FeS and FeS/C

33

samples is high, which comes from the raw reagent of Na2S.9H2O. Na elements exist as little impurities as the synthesis of high purity FeS is very difficult. Fig. 4 displays the TG and DSC curves of the bare FeS and FeS/C composite. As shown in Fig. 4a, the initial weight loss of ca. 4.72 wt.% from 25 to 200  C corresponds to the evaporation of the adsorbed water. With increasing temperature, an obvious and independent weight gain can be observed near 400  C extending to 500  C. This weight gain corresponds to two physicochemical processes: the oxidations of Fe and S. The oxidation of S to SO2 causes a weight loss, while the oxidation of Fe to Fe2O3 results in a weight gain. As illustrated in Fig. 4b, the initial weight loss of ca. 5.76 wt.% from 25 to 200  C corresponds to the evaporation of the adsorbed water. With increasing temperature, an obvious and independent weight gain can be observed near 400  C extending to 500  C. This weight gain also corresponds to two physicochemical processes: the oxidations of carbon and FeS. The oxidation of carbon to CO2/CO and S to SO2 causes weight losses, while the oxidation of Fe to Fe2O3 results in weight gains. The two processes take place in a mixed mode, but the weight gain plays a dominant role. TG profile of the FeS/C composite shows a weight loss of 15.3% by heating in the air up to 800  C, implying that the carbon content in the FeS/C composite is about 7 wt.%. Cyclic voltammograms of the FeS, FeS/C composite and FeS/ CeBi2O3-mixed electrodes during the initial ten cycles at a scan rate of 1 mVs1 are depicted in Fig. 5. From these profiles, it is clear that the carbon coating and the Bi2O3 additive strongly affect the redox behavior of the FeS electrode. In case of the FeS electrode (Fig. 5a), three oxidation peaks are observed around 0.93 V (Ox0), 0.88 V

Fig. 5. Cyclic voltammograms of the as-prepared FeS (a), FeS/C composite (b) and FeS/C-3wt%Bi2O3-mixed (c) electrodes for the initial ten cycles at a scan rate of 1 mV s1 in the voltage range of 1.2e0.4 V; (d) Change of redox peak current in cyclic voltammetry.

34

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

(Ox1), and 0.61 V (Ox2), and two reduction peaks are present around 1.04 V (Re1) and 0.98 V (Re2). For the FeS/C electrodes B (Fig. 5b) and C (Fig. 5c), three oxidation peaks appear around 0.92 V (Ox0), 0.87 V (Ox1), and 0.60 V (Ox2) and three reduction peaks occur around 1.05(Re1), 0.99 V(Re2) and 0.58 V(Re3). In these voltammograms, peak Ox0 observed in the oxidation direction may be attributed to the formation of the initial a-Fe(OH)2 layers or oxidation of adsorbed hydrogen atoms [15,37]. The previous investigation [15] indicates that both current peaks of Ox0 and Ox1 involve the electrooxidation of iron to Fe(II) species and peak (Ox0) appears only with the addition of S2 ion. In our case, peak (Ox0) appears in all the samples. According to the previous investigation [46], we believe that the appearance of peak (Ox0) is due to oxidation of iron to [Fe(OH)]ads. This step should be formulated as Fe / [Fe(OH)]ads. Peaks Ox1/Re1 and peaks Ox2/Re2 are assigned, respectively, to the following reactions (1) and (2) of iron oxidation/reduction [17]:

Re1 =Ox1 : FeðOHÞ2 þ 2e %Fe þ 2OH

E0 ¼ 0:877 V

Re2 =Ox2 : Fe3 O4 þ 4H2 O þ 2e %3FeðOHÞ2 þ 2OH E0 ¼ 0:560 V

(1)

(2)

In addition, the voltammogram of the two FeS/C electrodes exhibits one additional peak (Re3) at 0.58 V, and this additional reduction peak (Re3) gradually decays after the initial ten cycles. According to previous report [47], this peak may be due to the Fe(OH)2/FeOOH redox couple or the high electron-conductive carbon coating layer, which may be an intermediate activation process

of the FeS/C electrode. The Fe(OH)2/FeOOH redox couple may be represented as [48]:

FeðOHÞ2 þ OH %FeOOH þ H2 O þ e

E0 ¼ 0:560 V

(3)

Nevertheless, further investigation is needed to better understand this phenomenon. Changes of the Re2/Ox2 peak currents in the potential sweep cycles are shown in Fig. 5d. In case of the FeS electrode, it is clear that the Re2/Ox2 peak current apparently increases at the second cycle and then gradually decreases during the following cycles. This phenomenon is also observed during the study of pure iron electrode [25e28]. The decrease in redox currents on the pure iron electrode may be explained by the Fe(OH)2 layer formed during the cycles. But as for FeS electrode, further investigation is needed to better understand this phenomenon because that the real reaction mechanism on FeS electrode is complex and still unknown. For the FeS/C electrode, the Re2/Ox2 peak current shows a similar variation trend as the FeS electrode with the change at a smaller scale. It is worth noting that the redox current of the ten cycles is very close to each other for the FeS/C electrode with Bi2O3 additives. This result indicates that a good electrochemical reversibility of the FeS/ CeBi2O3-mixed electrode is established after the first cycle. Moreover, the peak current of Re2 (Fe3O4 / Fe(OH)2) increases rapidly to a maximum at cycle 2 or 3 and then decreases gradually with increasing the sweep cycle number, indicating an increase in the content of Fe3O4 in the reductive reaction shown in reaction (2) after the first cycle. This phenomenon was also observed during the study of Fe3C [17]. In the discharge process, FeS is oxidized

Fig. 6. Cyclic voltammograms of the as-prepared FeS (a), FeS/C composite (b) and FeS/C-3wt%Bi2O3-mixed (c) electrodes at various scan rates after 200 cycles at the scan rate of 1 mV s1. (d) Relationship between the main redox peak current and the square root of the scan rate for the three electrodes.

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

irreversibly to Fe3O4. During charging, the produced Fe3O4 is subsequently reduced to Fe through Fe(OH)2, contributing to the increased discharge capacity after the second discharge. Consequently, according to the previous report [17], it can be inferred that the reversible redox reactions of Fe3O4 % Fe(OH)2 % Fe take place after the oxidation of FeS to produce Fe3O4. The result will be proved by our further observation (see Fig. 10). Fig. 6 illustrates the cyclic voltammograms of the FeS, FeS/C composite and FeS/CeBi2O3-mixed electrodes at various scan rates after 200 cycles. It is clear that the CV curves of all the electrodes even after 200 cycles still remain a similar shape at a scan rate of 1 mVs1, confirming that all the FeS electrodes exhibit excellent cycling stability. As the scan rate increases from 1 to 10 mVs1, the anodic and cathodic peaks on the CV curves shift toward more positive and negative potentials, respectively, primarily due to the resistance and polarization of the electrode material. Also, the strong anodic and cathodic peaks of the cyclic voltammogram imply their high reversibility at a high current density. As shown in Fig. 6d, the redox current observed on the three electrodes scales linearly with the square root of the scan rate. This linearity might indicate a diffusion controlled process involving OH diffusion from the electrolyte to the electrode surface during the reduction step and from the electrode to the solution during the oxidation step. The electrochemical properties of the as-prepared bare FeS and composite FeS/C samples were evaluated by charge/discharge measurements. Generally, a newly-prepared iron electrode in a commercial secondary battery undergoes formation/activation during which the electrode is charged and discharged repeatedly

35

for 30 to 50 cycles. Then a stable discharge capacity is achieved. Since increased manufacturing time and capital cost of equipment for activation greatly contributes to the final cost of the battery, the activation process is a critical step in battery manufacturing. Unfortunately, till now, iron electrodes prepared by different methods suffer a slow activation rate. The enhancement of the activation rate of the iron electrode is of great significance for its practical application. Therefore, more attention has been paid to investigate the activation rate of the FeS electrode in the present work. The initial 30 chargeedischarge cycles of the FeS, FeS/C and FeS/ CeBi2O3-mixed electrodes during the activation process at a rate of 60 mA g1 are illustrated in Fig. 7. Obviously, after only ten chargeedischarge cycles, all the FeS electrodes reach about 90% of their maximum discharge capacities, exhibiting fast activation rates. Especially, preliminary discharge capacities of the two FeS/C electrodes are 198.6 mAh g1 and 199.2 mAh g1, respectively, in comparison with only 54.5 mAh g1 for the FeS electrode. After the activation process, the specific discharge capacities of the two FeS/C electrodes reach 325 mAh g1 and 308 mAh g1 with a faradaic efficiency of about 90.3% and 85.5%, respectively, while only 281 mAh g1 with a faradaic efficiency of about 78.0% is achieved for the FeS electrode. The result also reveals that the addition of Bi2O3 significantly influenced the formation process of the FeS electrodes. The charge/discharge curves of the initial 30 cycles of the three electrodes are shown in Fig. 7 (bed). For the FeS electrode, on the curve of the first and second discharges, only one plateau is observed because of the oxidation of FeS to produce Fe3O4. For the

Fig. 7. (a) The initial thirty chargeedischarge activation cycles of FeS, FeS/C composite and FeS/C-3wt%Bi2O3-mixed electrodes at the rate of 60 mA g1. Charge and discharge curves of FeS (b), FeS/C composite (c) and FeS/C-3wt%Bi2O3-mixed (d) electrodes during the activation cycles.

36

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

Fig. 8. (a) Rate performance of FeS, FeS/C composite and FeS/C-3wt%Bi2O3-mixed electrodes charged at 120 mA g1 and discharged at various rates. Discharge curves of the FeS(b), FeS/C composite(c) and FeS/C-3wt%Bi2O3-mixed(d) electrodes at different current rates.

third discharge, two plateaus are observed, which can be attributed to the oxidation reactions of Fe to Fe(OH)2 (Eq. (1)) and Fe(OH)2 to Fe3O4 (Eq. (2)). In contrast, for both the FeS/C electrodes, two plateaus at 1.30 V and 1.10 V, respectively are observed in the initial two discharge curves. This can be explained by the reduction of the electrode resistance originated from the carbon coating [44]. The reduced iron electrode resistance is beneficial to promoting the oxidation of FeS to produce Fe3O4, leading to a fast activation process. Moreover, it is noted that the lengths of the two discharge voltage plateaus of all the electrodes continuously increase during the activation process, indicating that more and more active materials participate in the electrochemical reactions. These charge/ discharge profiles of the FeS electrodes are very similar to the reported Fe3C [17], Fe2P [16] and carbonyl iron electrodes [5,13,14]. The rate capability of the prepared FeS, FeS/C and FeS/CeBi2O3mixed electrodes was evaluated by cycling the electrode in a step mode at different current densities. The results are presented in Fig. 8. As expected, both the FeS/C composite electrodes exhibit higher specific capacity and more stable cycling performance at each current density. As we can see that the specific discharge capacities at the same discharge rate increase in the following trend: A (FeS) < B (FeS/C) < C (FeS/C þ Bi2O3), implying that both the carbon coating and the addition of Bi2O3 play an important role in the improvement of the rate performance. Especially for the FeS/C composite with Bi2O3 additives a specific discharge capacity of about 325 mAh g1 is obtained at a rate of 60 mA g1, with a faradaic efficiency of about 90.3%. Even at rates of 600 (2C) and

1500 (5C) mA g1, specific discharge capacity values of 305 and 231 mAh g1 are achieved with a faradaic efficiency of 84.7% and 64.2%, respectively, showing distinctively high rate performance. In addition, it can be found that when the current density switches back from 1500 mA g1 to 60 mA g1, the specific capacity of all the electrodes can be fully recovered, demonstrating the excellent rate capability and good structural stability of the FeS

Fig. 9. Cyclic performance of the prepared FeS, FeS/C composite and FeS/C-3wt%Bi2O3mixed electrodes at charge/discharge rate of 300 mA g1.

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

electrodes. Also, it is worth noting that the FeS electrodes are capable of being discharged at an impressively high rate of 1500 mA g1, indicating that the FeS/C composite has an enormous potential as a high-rate anode material for nickel-based rechargeable batteries. Generally, the discharge rate capability of the iron electrodes was limited to 0.2C rate by the formation of a passive layer of iron (II) hydroxide [5]. Table 1 summarizes the latest literature on iron electrodes. Clearly, our result on high-rate performance is better than the previously reported data [3,5e7,14,16,43]. The cyclic performance of the bare FeS, FeS/C composite and FeS/CeBi2O3-mixed electrodes after the activation process is illustrated in Fig. 9 at a current density of 300 mA g1 for 200 cycles. Obviously, all the electrodes exhibit superior cycling stability. The specific discharge capacity at the same chargeedischarge cycles increase in the following trend: A (FeS) < B (FeS/C) < C (FeS/ C þ Bi2O3). In case of the FeS/C composite with Bi2O3 additives, about 99.2% of the initial capacity is maintained after 200 chargeedischarge cycles. In a previous report, it has been found that iron sulfide can be incorporated into the Fe(OH)2 lattice and interacts with Fe(II) in the oxide film to enhance the ionic conductivity of the iron electrode, retard passivation and promote the dissolution of iron, consequently, improving the cycle performance. Narayanan et al. [5] reported that the presence of an electronically conductive iron sulfide phase in iron electrode could mitigate the insulating nature of the passivation layer formed by iron (II) hydroxide. In our case, the excellent rate and cycle performance comes from an inherent characteristic of the FeS electrode. In terms of specific capacity, rate capability and cycling stability, the superior electrochemical performance of the iron electrode with FeS/C composite and Bi2O3 can be ascribed to the carbon

37

conductive layer and the addition of Bi2O3, which has a unique inherent characteristic of good rate and cycle performance. Carbon coating enhances the electron conductivity, which significantly facilitates the electrode reaction kinetic processes and therefore improves the rate performance. The existence of active carbon can also prevent the FeS particles from aggregation, and thereby leading to an excellent cycling stability. The addition of Bi2O3 in iron electrodes facilitates the in situ formation of elemental bismuth and a consequent increase in the overpotential for the hydrogen evolution reaction, resulting in to an increase in the charging efficiency and discharge capacity [5,7]. To further investigate the morphology and microstructure variation of the FeS electrodes after 200 chargeedischarge cycles, FE-SEM, EDS, and XRD measurements of the three electrodes were performed in the charged state. The results are illustrated in Fig. 10. FE-SEM images and XRD patterns of the FeS and FeS/CeBi2O3mixed electrodes, which are similar to those of the FeS/C sample, are not included in Fig. 10 for the sake of brevity. As shown in Fig. 10a, the bright spots in the images indicate the uniform dispersion of iron species. An average diameter of iron particles after 200 chargeedischarge cycles is less than 300 nm. Regarding the EDS and XRD analysis shown in Fig. 10c-d, we can clearly see that most particles in Fig. 10a are well-crystallized iron metal. Fig. 10d also clearly shows the existence of Fe3O4, which is produced during the discharge reaction. In comparison, an irregular morphology with diameters of 5e15 mm is observed for the FeS/C composite sample before the cycling measurement, as shown in Fig. 3a. As such, it is assumed that during the chargeedischarge process, the FeS/C electrode undergoes a “self-assembling” process via a dissolution-deposition mechanism. Most likely, the dispersion

Fig. 10. (a, b) FE-SEM images of FeS/C composite in the charged state after 200 cycles at different magnifications; (c) EDS spectra and (d) XRD pattern of the FeS/C composite in the charged state after 200 cycles.

38

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39

Table 1 Summary of properties of iron electrodes. Iron materials

Particle size [mm]

Electrode design

Additives

Discharge specific capacity [mAh g1]

Maximum output current [mA g1]

Ref.

FeOx-graphene

Carbonyl iron

0.02e0.1 5e10 0.5e3

5 wt% Bi2O3 1 wt% Bi2S3 10 wt% Bi2O3

380 322 150

1000 2C rate 0.2C rate

[3] [6] [5]

Carbonyl iron

0.5e3

3C rate

[5]

Carbonyl iron

0.5e3

1C rate

[5]

Carbonyl iron

0.5e3

Carbonyl iron

0.5e3

Commercial Fe3O4 electrode (Changhong) Fe2P FeS FeS

1e3

Powder rolled Powder spread with PTFE Pressed powder with PE binder Pressed powder with PE binder Pressed powder with PE binder Pressed powder with PP binder Pressed powder with PE binder Pressed pocket plate

0.02e0.3 0.1e0.4 5e15

Pressed powder no Pressed powder no Powder rolled with PTFE 3 wt% Bi2O3

a-Fe and Fe3O4

10 wt% Bi2O3 þ5 wt% 200 FeS 80 5 wt%Bi2O3 þNa2S 5

wt%Bi2S3

250

0.2C rate

[7]

5

wt%

290

0.2C rate

[14]

120

0.2C rate

[14]

343 110 231

500 1250 1500

[16] [43] Our study

Bi2S3

no

of iron nanoparticles is attributable to the dissolution and deposition processes of Fe and Fe(OH)2 via an intermediate species of iron: HFeO 2 [10,17,26]. Tsutsumi et al. [17] observed a similar phenomenon in their Fe3C electrodes. Further studies on the exact electrochemical reaction mechanisms are still needed. Additionally, it should be pointed out that after 200 chargeedischarge cycles, the reduced size of FeS/C particles during the dissolution-deposition process should take the credit for the excellent cycle performance of the electrodes. 4. Conclusions In summary, a new type of FeS/C composite was successfully synthesized via a simple calcination method followed by a coprecipitation process. The results indicate that the FeS/C particles are composed of FeS and carbon coating with a particle size around 5e15 mm. The reversible discharge capacity of FeS/C-3wt%Bi2O3mixed electrode is 331 mAh g1 at a discharge current of 60 mA g1 with a faradaic efficiency of 91.9%. At 120, 300 and 600 mA g1, discharge capacities of 329, 325 and 305 mAh g1 with a faradaic efficiency of 91.4%, 90.3%, and 84.7% are achieved, respectively, exhibiting attractive rate capability. For the first time, it is demonstrated that the FeS/C-3wt%Bi2O3-mixed electrode reaches a specific capacity of nearly 230 mAh g1 at a discharge rate as high as 1500 mA g1 (5C). Owing to the inherent electrochemical activity and high electronic conductivity resulted from the carbon layer, the FeS/C composite exhibits enhanced high discharge rate capability and cycling stability. These amazing results make the composite material very promising even without the optimization of the synthesis process and electrode structure. It is also found that after 200 dischargeecharge cycles, the size of the FeS/C micrometer particles reduces to less than 300 nm in diameter. This may be attributed to the dissolution-deposition process of Fe(OH)2 and Fe via intermediate iron species. To the best of our knowledge, the application of FeS/C composites as primary anode materials for nickel-based rechargeable batteries has not been reported so far. Due to the outstanding performance, the FeS/CeBi2O3-mixed electrode demonstrated in this work is extremely attractive for designing economically-viable large-scale energy storage systems based on alkaline nickeleiron rechargeable batteries. Acknowledgements The authors thank the financial supports from the Joint Funds of

the Natural Science Foundation of China (No. U1304211), Henan Education Bureau Foundation of China (Nos.13A150507 and 13A150512), Henan Provincial Department of Science and Technology Foundation of China (Nos.132300410294 and 132300410299) and Henan Key Science and Technology Program of China (No. 132102210256). References [1] M. Armand, J.M. Tarascon, Nature 451 (2008) 652e657. [2] J.B. Goodenough, Y. Kim, J. Power Sources 196 (2011) 6688e6694. [3] W. Jiang, F. Liang, J. Wang, L. Su, Y. Wu, L. Wang, RSC Adv. 4 (2014) 15394e15399. [4] S.R. Narayanan, G.K.S. Prakash, A. Manohar, B. Yang, S. Malkhandi, A. Kindler, Solid State Ionics 216 (2012) 105e109. [5] A.K. Manohar, C. Yang, S. Malkhandi, G.K.S. Prakash, S.R. Narayanan, J. Electrochem. Soc. 160 (2013) A2078eA2084. [6] A.S. Rajan, S. Sampath, A.K. Shukla, Energy Environ. Sci. 7 (2014) 1110e1116. [7] J.O. Gil Posada, P.J. Hall, J. Power Sources 268 (2014) 810e815. [8] N. Jayalakshmi, V.S. Muralidharan, J. Power Sources 32 (1990) 341e351. [9] C. Chakkaravarthy, P. Periasamy, S. Jegannathan, K.I. Vasu, J. Power Sources 35 (1991) 21e35. [10] P. Periasamy, B.R. Babu, S.V. Iyer, J. Power Sources 58 (1996) 35e40. [11] P. Periasamy, B. Ramesh Babu, S. Venkatakrishna Iyer, J. Power Sources 62 (1996) 9e14. [12] C.A.C. Souza, I.A. Carlos, M. Lopes, G.A. Finazzi, M.R.H. de Almeida, J. Power Sources 132 (2004) 288e290. [13] A.K. Manohar, C. Yang, S. Malkhandi, B. Yang, G.K. Surya Prakash, S.R. Narayanan, J. Electrochem. Soc. 159 (2012) A2148eA2155. [14] A.K. Manohar, S. Malkhandi, B. Yang, C. Yang, G.K. Surya Prakash, S.R. Narayanan, J. Electrochem. Soc. 159 (2012) A1209eA1214. [15] B.T. Hang, T. Watanabe, M. Egashira, I. Watanabe, S. Okada, J.-I. Yamaki, J. Power Sources 155 (2006) 461e469. [16] Y. Wang, L. Zhang, H. Li, Y. Wang, L. Jiao, H. Yuan, L. Chen, H. Tang, X. Yang, J. Power Sources 253 (2014) 360e365. [17] K. Ujimine, A. Tsutsumi, J. Power Sources 160 (2006) 1431e1435. [18] C.A. Caldas, M.C. Lopes, I.A. Carlos, J. Power Sources 74 (1998) 108e112. [19] B. Yang, S. Malkhandi, A.K. Manohar, G.K. Surya Prakash, S.R. Narayanan, Energy Environ. Sci. 7 (2014) 2753e2763. [20] S. Malkhandi, B. Yang, A.K. Manohar, G.K.S. Prakash, S.R. Narayanan, J. Am. Chem. Soc. 135 (2012) 347e353. [21] G. Huo, X. Lu, Y. Huang, W. Li, G. Liang, J. Electrochem. Soc. 161 (2014) A1144eA1148. [22] B.T. Hang, D.H. Thang, N.T. Nga, P.T.L. Minh, E. Kobayashi, J. Electrochem. Soc. 160 (2013) A1442eA1445. [23] B.T. Hang, D.H. Thang, E. Kobayashi, J. Electroanal. Chem. 704 (2013) 145e152. [24] C.-Y. Kao, Y.-R. Tsai, K.-S. Chou, J. Power Sources 196 (2011) 5746e5750. [25] C.-Y. Kao, K.-S. Chou, J. Power Sources 195 (2010) 2399e2404. [26] B.T. Hang, H. Hayashi, S.-H. Yoon, S. Okada, J.-I. Yamaki, J. Power Sources 178 (2008) 393e401. [27] B.T. Hang, T. Doi, S. Okada, J.-I. Yamaki, J. Power Sources 174 (2007) 493e500. [28] B.T. Hang, T. Watanabe, M. Eashira, S. Okada, J.-I. Yamaki, S. Hata, S.-H. Yoon, I. Mochida, J. Power Sources 150 (2005) 261e271. [29] H. Wang, Y. Liang, M. Gong, Y. Li, W. Chang, T. Mefford, J. Zhou, J. Wang, T. Regier, F. Wei, H. Dai, Nat. Commun. 3 (2012) 917.

E. Shangguan et al. / Journal of Power Sources 291 (2015) 29e39 [30] T.S. Balasubramanian, A.K. Shukla, J. Power Sources 41 (1993) 99e105. [31] M. Jayalakshmi, B. Nathira Begum, V.R. Chidambaram, R. Sabapathi, V.S. Muralidharan, J. Power Sources 39 (1992) 113e119. [32] P.R. Vassie, A.C.C. Tseung, Electrochim. Acta 21 (1976) 299e302. [33] K. Micka, Z. Z abranský, J. Power Sources 19 (1987) 315e323. [34] K. Vijayamohanan, A.K. Shukia, S. Sathyanarayana, J. Electroanal. Chem. Interface 289 (1990) 55e68. [35] N.A. Hampson, R.J. Latham, A. Marshall, R.D. Giles, Electrochim. Acta 19 (1974) 397e401. [36] R. Carta, S. Dernini, A.M. Polcaro, P.F. Ricci, G. Tola, J. Electroanal. Chem. Interface 257 (1988) 257e268. [37] G.P. Kalaignan, V.S. Muralidharan, K.I. Vasu, J. Appl. Electrochem. 17 (1987) 1083e1092. [38] K. Vijayamohanan, A.K. Shukla, S. Sathyanarayana, J. Power Sources 32 (1990) 329e339. [39] P. Periasamy, B. Ramesh Babu, S. Venkatakrishna Iyer, J. Power Sources 63

39

(1996) 79e85. [40] L. Fei, Q. Lin, B. Yuan, G. Chen, P. Xie, Y. Li, Y. Xu, S. Deng, S. Smirnov, H. Luo, ACS Appl. Mat. Interf. 5 (2013) 5330e5335. [41] C. Dong, X. Zheng, B. Huang, M. Lu, Appl. Surf. Sci. 265 (2013) 114e119. [42] B.-C. Kim, K. Takada, N. Ohta, Y. Seino, L. Zhang, H. Wada, T. Sasaki, Solid State Ionics 176 (2005) 2383e2387. [43] Mulder, F. Marten, WO patent no. 2014/142666 A1; 2014. [44] B.T. Hang, S.-H. Yoon, S. Okada, J.-I. Yamaki, J. Power Sources 168 (2007) 522e532. [45] Z. Zeng, H. Zhao, J. Wang, P. Lv, T. Zhang, Q. Xia, J. Power Sources 248 (2014) 15e21.  [46] J. Cerný, K. Micka, J. Power Sources 25 (1989) 111e122. [47] M.K. Ravikumar, T.S. Balasubramanian, A.K. Shukla, J. Appl. Electrochem. 26 (1996) 1111e1115. [48] A.K. Shukla, M.K. Ravikumar, T.S. Balasubramanian, J. Power Sources 51 (1994) 29e36.