- Email: [email protected]
Mesoporous spinel ferrite composite derived from a ternary MgZnFe-layered double hydroxide precursor for lithium storage
Journal of Alloys and Compounds 726 (2017) 306e314
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Mesoporous spinel ferrite composite derived from a ternary MgZnFe-layered double hydroxide precursor for lithium storage Duan Wang a, *, Jingli Wu a, Daxun Bai b, Rongrong Wang c, Feng Yao b, Sailong Xu b, * a
School of Chemical and Environmental Engineering, North University of China, Taiyuan 030051, China State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, China c Xi'an Institute of Electromechanical Information Technology, Xi'an 710065, China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 April 2017 Received in revised form 29 June 2017 Accepted 1 August 2017 Available online 4 August 2017
Layered double hydroxides (LDHs), also known as hydrotalcite-like anionic clays, are very convenient precursors with a tunable flexibility toward multifunctional nanomaterials, especially in energy storage. Typical methods to improve lithium storage are to introduce additional or self-generating carbonaceous supports to LDH-derived transition metal oxides as anode nanomaterials which can host lithium mainly though a conversion mechanism. Here, we describe a preparation of mesoporous spinel ferrite composite (MgFe2O4/ZnFe2O4) for lithium storage, which is assisted by a combined conversion and alloying mechanism. The composite is derived by a thermal decomposition of a scalablely produced singleresource precursor of ternary Mg2þZn2þFe3þ-layered double hydroxide (Mg2þZn2þFe3þ-LDH), and subsequent selective etching. Electrochemical test shows that the electrode delivers an exceptional electrochemical performance, i.e., a reversible capacity of 1190 mA h g1 after 100 cycles at 100 mA g1, and, in particular, a reversible capacity of 981 mA h g1 at 500 mA g1 after 330 cycles, as well as a reversible capacity of 541 mA h g1 at 2000 mA g1 after 1000 cycles. The high electrochemical performance could be attributed to the following features: the combined alloying and conversion mechanisms of ZnFe2O4, synergistic MgFe2O4, and slight-content MgO as a non-active matrix, as well as an appropriate specific area and mesoporous size distribution. Our results show that the cation-tunable LDH precursor-derived synthesis route might be an alternative to prepare multiple-component composites of spinel ferrites and transition metal oxides. © 2017 Elsevier B.V. All rights reserved.
Keywords: Layered double hydroxide precursor Spinel ferrite Sintering Domain structure Energy storage materials
1. Introduction Layered double hydroxides (LDHs), also known as hydrotalcitelike anionic clays, have attracted increasing interest in various potential applications, especially including electrochemical energy storage and conversion [1e7]. LDHs with structure based on Mg(OH)2-like layers, possess the general formula of [M2þ13þ xþ n 2þ (such as Mg2þ, xM x(OH)2] (A )x/n·yH2O, in which cationic M Zn2þ, Ni2þ, and Co2þ) and M3þ (such as Al3þ, Fe3þ, and Co3þ) are well-orderedly occupied within a LDH layer, Ane anions (such as CO2 3 , NO3 , and surfactant anion) are well-intercalated between the LDH interlayer galleries, and the value (x) of M3þ/(M2þ þ M3þ) molar ratio ranges typically between 0.2 and 0.33. Despite the huge
* Corresponding authors. E-mail addresses: [email protected] (D. Wang), [email protected]. cn (S. Xu). http://dx.doi.org/10.1016/j.jallcom.2017.08.005 0925-8388/© 2017 Elsevier B.V. All rights reserved.
family, the LDHs could be basically recognized as convenient precursors with a unique feature to easily tune the types of metal cations and intercalated anions, as well as M2þ/M3þ molar ratios over a wide range, which have enabled the LDHs as tunable precursors to prepare electrode nanomaterials for lithium ion batteries (LIBs). Two main approaches have hitherto been applied to prepare LDH precursor-derived products as anode nanomaterials for LIBs. On one hand, transition metal oxide/spinel/carbon composites were prepared typically by calcining tunable LDH precursors supported on the additionally introduced carbon (such as carbon nanotube (CNT) and graphene (G)). The TMOs/carbon composites included CoFe2O4/CNT [8], CoO/CoFe2O4/G [9], and NiO/NiFe2O4/G [10] derived from Co2þFe2þFe3þ-LDH/CNT, Co2þFe3þ-LDH/G, and Ni2þFe3þ-LDH/G precursors, respectively, which were able to exhibit highly cycling stability and rate capability compared with the spinels without carbonaceous support. On the other hand, diverse spinel/carbon composites were also synthesized via the
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
utilization of the self-generating carbon during the carbothermal process of different LDH/organic molecule precursors. Such spinel/ carbon composites included Ni@NiO/Al2O3/C from a NiAl-LDH/ glucose mixture [11], and C-NiO/NiFe2O4 derived by a rapid catalyzing C2H2 flow on NiFe-LDH surface at 300 C via chemical vapor deposition (CVD) [12]. The self-generating carbon is able to greatly promote the cycling stability and rate capability of the resulting spinel/carbon composites, with the aid of the high conductivity of the self-generating carbonaceous nanomaterials. Indeed, both types of spinel-containing composites delivered the high electrochemical performances, which were based mainly on the conversion mechanism of the nanosized active transition metal oxide/ spinel and also benefit from the additionally introduced or selfgenerating conductive carbonaceous supports. In this present study, we describe a preparation of MgFe2O4/ ZnFe2O4 composite as a conversion/alloying nanomaterials for lithium storage. The carbon-free composite was prepared via a thermal decomposition of a scalablely prepared Mg2þZn2þFe3þLDH single-source precursor and subsequent selective etching (Scheme 1), and thus endowed with the features of boosting the electrochemical performances: conversion/alloying ZnFe2O4, synergistic active MgFe2O4, and low-content non-active MgO, as well as an appropriate specific area and mesoporous size distribution. Electrochemical test shows that the composite is indeed able to deliver the expected electrochemical performances. 2. Experimental 2.1. Preparation of Mg2þZn2þFe3þ-LDH precursor The Mg2þZn2þFe3þ-LDH precursor was prepared via a scalable procedure of separate nucleation and aging steps (SNAS) [13]. In brief, a salt solution, which contained Mg(NO3)2·6H2O, Zn(NO3)2·6H2O and Fe(NO3)3·9H2O with a (MgþZn)/Fe molar ratio of 2: 1, was obtained via dissolution of the salts into a freshly deionized water to yield a resulting solution with a total cationic concentration of 1.2 M. An aqueous base solution was also acquired, with a fixed concentration of NaOH ([OH] ¼ 2 2þ 2þ 3þ 1.6([Mg ] þ [Zn ] þ [Fe ]) and Na2CO3 ([CO3 ] ¼ 2[Fe3þ]). The as-prepared solutions were introduced automatically and simultaneously into a commercial chemical reactor operated at a rotor speed of 3000 rpm. The resulting slurry was mixed continuously for 2 min, then collected, and aged at 70 C for 48 h. The Mg2þZn2þFe3þ-LDH precursor was obtained by centrifugation, then rinsing thoroughly the slurry several times with alternate deionized water and ethanol, and finally by drying in vacuum at 60 C overnight. 2.2. Preparation of MgFe2O4/ZnFe2O4/MgO composite The composite was prepared by calcining the as-prepared Mg2þZn2þFe3þ-LDH precursor. The calcination was carried out in
307
a muffle furnace at 900 C for 3 h with a ramping rate of 5 C/min. The calcined product was then rinsed by carefully stirring in an aqueous solution of (NH4)2SO4 (10 wt %) at 70 C for 48 h to remove the partial MgO, and then in aqueous solution of NaOH (1 M) at 40 C for 5 h to remove ZnO completely. For comparison of XRD patterns, the mixture of MgFe2O4 and ZnFe2O4 was prepared by physically mixing the individual MgFe2O4 and ZnFe2O4. The individual MgFe2O4 and ZnFe2O4 were derived from MgFe- and ZnFe-LDH precursors obtained under the same experimental conditions, respectively. Effective removals of MgO and ZnO were performed by using HCl aqueous solution (1 M) at 80 C for 3 h and NaOH aqueous solution (1 M) at 40 C for 5 h, respectively. 2.3. Characterization Powder X-ray diffraction (XRD) was used on a powder diffractometer (Rigaku XRD-6000, Cu-Ka radiation, l ¼ 1.542 Å) operated at a scanning speed of 10 /min. To acquire sample morphology, scanning electron microscopy (SEM) images were recorded on a Zeiss Supra 55 instrument. Transmission electron microscopy (TEM) observation was carried out on an electron microscope (JEOL JEM-2100) with an equipped EDX, which was operated at an accelerating voltage of 200 kV. The samples for SEM and TEM visualizations were dispersed in ethanol via sonication and then deposited manually onto silicon wafer or copper microgrid. Elemental analysis was conducted for metal ions by using an inductively coupled plasma emission spectrometer (ICP-ES, Shimadzu). Specific surface area and pore size distribution were determined quantitatively from nitrogen adsorption/desorption isotherms collected at 77 K on a Nova 1200, Quantachrome apparatus. For the pre-treatments before the measurement, the sample was kept for drying in vacuum at 393 K for 8 h. The surface area was computed via the BrunauereEmmetteTeller (BET) method, and the pore size distribution was calculated through the BarretteJoynereHalenda (BJH) method. X-ray photoelectron spectroscopy (XPS) measurement was applied on an X-ray Photoelectron Spectrometer (Kratos Axis ULTRA) assembled with a 165 mm hemispherical electron energy analyser. 2.4. Electrochemical measurement The assembly of all Swagelok-type cells was conducted in a commercial argon-filled glove box. To prepare the working electrodes, a mixer was prepared by using active material/poly(vinyl difluoride) (PVDF)/super-P acetylene black at 70/10/20 (wt %), and then carefully pasted on a clean surface of Cu foil. The loading mass of the composite was ca. 10 mg cm2. Lithium foil and glass fiber (GF/D, both from Whatman) were utilized as the counter electrode and the separator, respectively. The electrolyte consisted of a solution of 1 M LiPF6 salt, which was dissolved into a solvent mixture of ethylene carbonate (EC): dimethyl carbonate (DMC): diethyl
Scheme 1. Schematic illustration of preparing mesoporous MgFe2O4/ZnFe2O4/MgO composite derived from a ternary MgZnFe-LDH precursor: initial calcination at 900 C in air, subsequent partial removal of MgO by (NH4)2SO4 solution and removal of ZnO by NaOH solution.
308
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
carbonate (DEC) at 1: 1: 1 (wt %) plus vinylene carbonate (VC, 2 wt %). All were purchased commercially (from Tianjing Jinniu Power Sources Material Co. Ltd., China). Galvanostatic cycling was recorded for all the assembled cells by using a commercial cell-testing system (LAND CT2100A) between the voltage range from 0.01 to 3.0 V (vs. Liþ/Li). Cyclic voltammogram (CV) were measured on an electrochemistry work station (CHI 760E) at a scan rate of 0.1 mV s1 in the range of 0e3.0 V (vs. Liþ/Li). Electrochemical impedance spectrum (EIS) was recorded on an advanced electrochemical system (Parstat 2273) operated at an open circuit potential over the frequency range between 100 kHz and 0.1 Hz. 3. Results and discussion The Mg2þZn2þFe3þ-LDH precursor was prepared by using our SNAS method characteristic of a scalable production [13]. The obtained LDH precursor shows a XRD pattern with the typical layered feature of a LDH precursor (Fig. S1). The precursor was subjected to the calcination at 900 C in air. The calcination leads to the formation of cubic MgO, wurzite-structured ZnO, as well as the spinels of MgFe2O4 (JCPDS No. 88-1943) and ZnFe2O4 (JCPDS no. 82-1042), as shown by the XRD result in Fig. 1a. Further selective etching was conducted to remove the partial MgO phase and ZnO phases by using the acidic (NH4)2SO4 solution (10 wt %) and the basic NaOH solution (1 M), respectively. Considering that the amount of non-active MgO has an influence on electrochemical performance of the composite electrode, the residual amount of MgO were tuned by varying type (such as strongly acidic HNO3 and weakly acidic (NH4)2SO4) and concentration of etching solution, as well as etching temperature and duration [14]. A proper etching condition was optimized, for example, using (NH4)2SO4 solution (10 wt %, pH z 6.0) at a temperature of 70 C for 48 h. The post-etched product shows the very reflection peaks of MgFe2O4/ZnFe2O4 and MgO (Fig. 1a), the latter of which was not removed completely due to the use of weakly acidic (NH4)2SO4 solution. From previous studies [14e16], it is well-known that thermal decomposition of a typical MIIMIII-LDH precursor in air or under an inert atmosphere yields mixed metal oxides, which consist of metal oxide (MIIO) and spinel (MIIMIII2O4). Such studies include ZnO/ ZnAl2O4 composite calcined from ZnAl-LDH from 500 C to 800 C in air [16], CoO/CoFe2O4 from a CoFe-LDH precursor in a He
atmosphere at 700 C for 2 h [17], NiO/NiFe2O4 from a NiFe-LDH precursor 500 C in air [18], ZnO/ZnFe2O4 nanoparticle from ZnFe-LDH [19]. Furthermore, the formation of MIIO/MIIMIII2O4 has been corroborated by a transformation mechanism from a ZnAlLDH precursor monitored by using in situ XRD [16], i.e., the preferential nucleation and growth of ZnO between 200 and 500 C, as well as the subsequent doping of Al cation into ZnO and the formation of ZnAl2O4 between 500 and 800 C. Note that MII/MIII molar ratio of a MIIMIII-LDH precursor is typically in the range from 5:1 to 2:1, the doping of MIII cation into MIIO results in the good dispersion MIIMIII2O4 phase embedded in a matrix of excess MIIO continuous phase [14e16]. In addition, solid-solution metal oxide could be generated by the thermal decomposition of MIIMIII-LDH [20]. In distinct contrast to the above cases, graphene-supported Mn0.25Co0.75O solid solution was indeed formed by thermal decomposition of CoMn-LDH/graphene oxide at 600 C for 2 h in Ar [20]. The formation of Mn0.25Co0.75O could be ascribed mainly to the Mn doping into the CoO phase, owing to the carbothermal process in Ar atmosphere, as well as the comparable size and chemical nature of the Mn2þ and Co2þ cations, and the ease of miscibility of the MnO/CoO solid solution. Therefore, in the case of a typical precursor of MgZnFe-LDH, the thermal decomposition in air leads to the very formation of MgO/MgFe2O4/ZnO/ZnFe2O4 composite, instead of Mg0.64Fe2.36O4 and the like. To experimentally exclude a possibility that the formed spaniels may also contain a MgxZn1-xFe2O4 solid solution, a comparison of XRD pattern was carried out between the above spinel composite and the mixture of individual MgFe2O4 and ZnFe2O4. Perusal of the comparison clearly reveals that no shifts of XRD reflection peaks of the above spinel composite are observed with respect to the mixture (Fig. 1b). No shift of XRD pattern indeed rules out the possibility of MgxZn1xFe2O4 solid solution formed owing to the cationic doping of Mg or Zn. We thus believe that the composite is obtained, which consists of MgFe2O4, ZnFe2O4, and MgO, thus denoted as MZO/MgO. To quantitatively determine the exact content of each component, we further carried out element analysis to the MZO/MgO composite by using the ICP-ES measurement. The ICP result gives each concentration of Fe, Mg, and Zn cations, and the molarities of Fe, Mg, and Zn cations were determined, thus yielding a (MgþZn)/ Fe molar ratio of 0.54 and a Mg/Zn molar ratio of 0.63. With respect to the theoretic molar ratio (0.5) for both MgFe2O4 and ZnFe2O4 of the post-etched composite, the excess amount of Mg cation is determined to be MgO. Therefore, a MgFe2O4/ZnFe2O4 molar ratio
Fig. 1. XRD patterns of the products calcined from the MgZnFe-LDH precursor: (a) before and after the selective etching to remove the partial MgO, and ZnO. (b) No shift of XRD observed patterns between the MZO/MgO composite and the mMZO mixture.
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
is calculated to be 1: 2, and a (MgFe2O4 þ ZnFe2O4)/MgO molar ratio to be 11: 1. The excess of MgO is indeed consistent with the abovementioned XRD result of MgO observed, which is partially removed by using weakly acidic (NH4)2SO4 aqueous solution. As expected, we obtain the MZO/MgO composite that is composed of highcontent active MgFe2O4 and ZnFe2O4, as well as slight-content non-active MgO. From a previous study [21], it is expected that the slight amount of non-active MgO is expected to act as a matrix to effectively accommodate the large volume change during the lithiation/delithiation process, thus facilitating a high cycle stability. The morphology of the MZO/MgO composite was visualized by using SEM and TEM. Fig. 2a shows that the composite exhibits the irregular nanoplatelets of the MZO/MgO composite, which are much smaller than those of the Mg2þZn2þFe3þ-LDH precursor typically with the hydrotalcite-like nanoplatelets (as shown in Fig. 2b). Such a visible change in dimensional size can be ascribed to the dehydroxylation of the pristine LDH precursor during the thermal decomposition [22], as reported previously for the hollow CoFe2O4 sphere converted from Co2þFe2þFe3þ-LDH shell sphere [13], and ZnO/ZnAl2O4 nanoparticle from ZnAl-LDH [16]. Note that SEM and TEM images show the severe aggregation of the composite nanoparticles. The aggregation is typically observed for the MIIO/ MIIMIII2O4 composites calcined from different MIIMIII-LDH precursors. The aggregation can be ascribed to the initial dehydroxylation of the layers and loss of volatile species arising from decomposition of the interlayer carbonate anions below ca. 200 C [14,16], and the subsequent topotactic transformation to MIIO/MIIMIII3O4 phases during the sintering process at moderate and high temperatures [14,16,18]. TEM observation further manifests that the nanoplatelet-like morphology of the MgFe2O4/ZnFe2O4 composite is well preserved, and also the MgFe2O4/ZnFe2O4 nanoplatelet contains both domains in gray contrast and pores in bright contrast (Fig. 2c). The features could be ascribed to the dehydroxylation of LDH precursor, the topotactic transformation to the resultant transition metal/spinel composite, and the selective etching. A HRTEM image reveals that the MgFe2O4/ZnFe2O4 composite is highly crystalline, with the well-defined lattice fringes in
309
two regions determined to be 0.48 nm that corresponds to the (111) crystalline planes of the MgFe2O4 and ZnFe2O4 phases (Fig. 2d). Especially, a visible heterojunction nanostructure is observed with a slight mutual dislocation. Similar HRTEM visualization of interfacial nanodomains was also reported early, such as CoO/CoFe2O4 composite calcined from a CoFe-LDH precursor [17] and Co2SnO4/ Co3O4/Al2O3/C composite derived from a laurate anion-intercalated CoAlSn-LDH precursor [23]. The chemical states were examined to the species of the MZO/ MgO composite by using XPS technique. Fig. 3a shows that for Fe 2p, the bands are centered at 710.9 and 712.6 eV, as well as 725.3 and 718.9 eV, which can be attributed to Fe 2p3/2 and Fe 2p1/2, respectively [24]. In the case of Mg 1s, the strong peak is clearly visible at 1304.2 eV (Fig. 3b), corresponding to the existence of Mg2þ of the composite. In the case of Zn 2p, the peaks centered at 1022.1 and 1045.6 eV can be ascribed to Zn 2p3/2 and Zn 2p1/ 2þ specie 2,respectively (Fig. 3c), reflecting the oxidation state of Zn of the composite. For O 1s, the spectrum, with an additional shoulder, can be de-convoluted by three distinct peaks centered at 529.6, 530.7, and 532.2 eV (Fig. 3d). The strong one at 529.6 eV can be ascribed to the surface lattice oxygen (O2), viz., O-Fe(Mg, Zn) in the composite, while those other two at high energy can be assigned to either the chemisorbed/dissociated oxygen, or the OH species on the surface of the composite [25]. N2 adsorption-desorption isotherm was measured to examine the texture structure of the MZO/MgO composite. Fig. 4a shows that the isotherms of the sample possess a distinct H3-type hysteresis loop at a relative pressure ranges 0.05e0.94 P/P0, which can be ascribed to type IV (IUPAC classification). The surface area is calculated to be 25.6 m2 g1, and the pore size distribution is determined to rang between 2 and 10 nm (Fig. 4b). This pore size distribution clearly manifests the typical mesoporous characteristic of the MZO/MgO composite, well consistent with the pores in bright contrast by TEM observation. The formation of porous stacking for the resulting MIIMIII2O4 nanoparticles can be ascribed to the selective removal of MIIO from the MIIO/MIIMIII2O4 composite. Similar porous nanostructures were reported previously, such as ZnAl2O4 mesopore networks obtained by calcining ZnAl-
Fig. 2. SEM images of (a) the MZO/MgO composite and (b) MgZnFe-LDH precursor. (c) TEM and (d) HRTEM images of the MZO/MgO composite.
310
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
Fig. 3. XPS spectra of (a) Fe 2p, (b) Mg 1s, (c) Zn 2p, and (d) O 1s for the MZO/MgO composite.
Fig. 4. N2 adsorption-desorption isotherms and pore size distribution of MZO/MgO composite.
LDH precursor and subsequent etching of ZnO [22], MgMIII2O4 (M ¼ Al, Fe) macroporous frameworks available by decomposition of MgAl- and MgFe-LDH precursors and removal of MgO [14], as well as porous ZnFe2O4 prepared by calcining ZnFe-LDH precursor and subsequent etching of ZnO [19]. From a pervious study [26], it is well-known that the approprixate surface area is able to provide enough contact area and lithium storage sites for the electrodes,
and the mesoporous structure can effectively facilitate the transportation of Liþ and electrolyte molecules and also relieve the volume changes of the active materials during the repeated charge/ discharge processes. Therefore, we expect that the approprixate surface area and the narrow mesopore size distribution of the MZO/ MgO composite could greatly facilitate enhancing the lithium storage.
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
We initially carried out CV measurement to the MZO/MgO composite electrode. Fig. 5a shows that in the first cathodic sweep, two cathodic peaks at 0.71 V and 0.52 V are clearly visible. The weak peak at 0.71 V could be ascribed to the formation of the intermediate LinZnFe2O4, owing to the Li-intercalation into ZnFe2O4 (Eqn. (1)) [27e30]. The other strong peak at 0.52 V corresponds to the reaction of Liþ with the intermediate LinZnFe2O4 to yield Zn0 and Fe0 (Eqn. (2)), and the further lithiation of Zn0 to give rise to a LixZn alloy (Eqn. (3)) [27e30], as well as the reduction of MgFe2O4 to Fe0 and MgO (Eqn. (4)) [31], along with the formation of a solid electrolyte interface (SEI) layer and Li2O [31]. During the first oxidation sweep, a broad anodic peak is visible at 1.61 V, corresponding to the oxidation of the metallic Zn0 into Zn2þ and Fe0 into Fe2O3 (Eqns. (5) and (6)) [31e33], as inactive MgO can not participate in the reversible reaction of Eqn. (4). From the second cycle, both cathodic and anodic peaks appear to shift positively [27e30,34,35], only one strong cathodic peak appears at 0.89 V, and one remarkable anodic peak is observed to shift to 1.75 V. Apparently, a good overlap is visible between the further sweeps, indicating a highly reversible uptake and release of lithium upon the completion of the initial cycling. The electrochemical behavior of the MZO/MgO composite electrode could be summarized by the following equations [27e35]: ZnFe2O4 þ xLiþ þ xee / LixZnFe2O4
(1)
LixZnFe2O4 þ (8 e x)Liþ þ (8 e x)ee / Zn0 þ 2Fe0 þ 4Li2O
(2)
Zn0 þ xLiþ þ xee 4 LixZn
(3)
311
MgFe2O4 þ 6ee þ 6Liþ / MgO þ 2Fe0 þ 3Li2O
(4)
Zn0 þ Li2O 4 ZnO þ 2Liþ þ 2ee
(5)
2Fe0 þ 3Li2O 4 Fe2O3 þ 6ee þ 6Liþ
(6)
We then conducted the galvanostatic discharge/charge test to the MZO/MgO composite electrode between 0.01 and 3 V at 100 mA g1. Fig. 5b exhibits that during the first discharge process, the initial potential decreased rapidly, ascribed to the small amount of Liþ ion insertion into the electrodes. Two voltage plateaus appear, corresponding to the two reduction peaks in the first cathodic sweep of the above CV curve [28,29,34]. The short voltage plateau was shown up at ca. 0.95 V, manifesting the formation of metallic Fe0 and Zn0 nanoparticles into an amorphous Li2O matrix owing to the conversion reaction [28,29,35]. The obvious long plateau was visible at ca. 0.67 V, which could be associated with the complete decomposition of LixZnFe2O4 and MgFe2O4 to Zn0, Fe0, which, together with amorphous MgO, are embedded in a Li2O matrix [36]. At the end of the discharge profile, the potential smoothly decreases from 0.67 V to the lower cut-off voltage, which could be related to the formation of LixZn alloys [28,29,35]. Additional capacity in the tail of the curve at low potential is usually observed just after the conversion reaction. Capacitive-like behavior is characterized by a linear potential decay/increase in the potentialedischarge/charge curve [37]. The extra contribution was attributed either to a capacitive-like behavior for the conversion-reaction transition metal oxides (Fe, Co, and Nicontaining oxides) [37,38], or to SEI formation for conversionalloying transition metal oxides (such as ZnFe2O4) [39,40]. The
Fig. 5. (a) The first three CV curves for MZO/MgO composite electrode between 0.01 and 3.0 V at 0.1 mV s1, and (b) the first three charge/discharge profiles at 100 mA g1, (c) cycling behavior at a rate of 100 mA g1, and (d) rate capability.
312
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
initial discharge and charge capacities of the MZO/MgO composite electrode are determined to be 1911 and 1004 mA h g1, thus giving rise to a Coulombic efficiency of 53%. The loss of capacity is ascribed typically to the irreversible decomposition of electrolyte to form the solid electrolyte interface (SEI) layer. In addition, a good overlap is found between the second and third cycles, which strongly suggests a possibility of the good reversibility of the MZO/MgO electrode. We further examined the cycling stability of the MZO/MgO composite electrode at a current density of 100 mA g1. Fig. 5c shows that the charge capacity of the composite electrode exhibits an ascending trend after the 60th cycles, and almost hovers at 1190 mA h g1 over the rang from the 60th and the 100th cycles, indeed indicating a good cycling stability of the composite electrode. Such a type of “ascending-then-stable” trend was also reported previously, which is closely associated with both the activation of the oxides resulting from the nanosize effect [41,42]. Furthermore, the value of Coulombic efficiency remains stable at ca. 98% beginning with the second cycles. The rate capability was also evaluated to the electrode at different current densities. Fig. 5d depicts that the MZO/MgO composite electrode delivers the charge capacities of 993.4, 912.1, 806, 717.8, 620.6, and 557 mA h g1 with the current densities varying initially from 100, then to 200, 400, 800, 1000, eventually to 2000 mA g1, respectively. When the current density is reset back to the current density of 100 mA g1, the capacity is able to return to 1150 mA h g1, thus reflecting a good rate capability. We further estimated the long-term cycling stability of the MgZO/MgO composite electrode at large current densities. Fig. 6a shows that at a 500 mA g1, a similar “decrease-increase-stable” trend of charge capacity is observed. The reversible charge capacity is 989 mA h g1 after the 100th cycle, then 857 mA h g1 after the 130th cycle, and eventually maintains at 981 mA h g1 after the 330th cycle. In particular, at a high current density of 2000 mA g1, such a “decrease-increase-stable” trend is also clearly resolved in Fig. 6b, with a charge capacity of 541 mA h g1 after the 1000th cycle.
A further comparison of reversible specific capacity was conducted between the electrodes of our MZO/MgO composite and MgFe2O4 or ZnFe2O4 containing electrodes reported in previous studies [43e53]. Fig. 7 displays that the reversible charge capacity of our composite electrode could be comparable to those of the other electrodes when considered the approximate amount of electrochemical active nanomaterials used. Considering the significant improvements, we suggest that the following compositional and nanostructural characteristics can be responsible for the contribution. (i) The combination of alloying and conversion mechanisms of ZnFe2O4, together with the active MgFe2O4, could contribute greatly to the reversible capacity [54]. The alloying mechanism is considered to provide the additional capacity [28,30,55], which has been recognized to be a capacitive-like behavior [38]. (ii) The appropriate specific area could provide enough contact area between MgFe2O4/ZnFe2O4 and electrolyte, and the mesopore size distribution is capable of offering void space to accommodate the strain resulting from the volume expansion/ contraction during cycles [26]. (iii) Low-content MgO is wellknown to act as an inert matrix to accommodate strain change and prevent aggregation of active components [15]. This contribution is similar to that of nonactive Al2O3 reported in a previous study [19]. (iv) The mesoporous MZO/MgO composite may possess a relatively stable interfacial resistance, which could facilitate improving the stability of SEI structure and interface charge transfer properties of the composite electrode. To support this hypothesis, EIS measurement was carried out to the composite electrode. Fig. S2 shows that the Nyquist plot consists of two characteristic parts, a depressed semicircle over the high-to-middle frequency and an inclined line tail in the low frequency range. The MZO/MgO composite electrode exhibits a small diameter of the semicircle, suggesting low SEI resistance and contact resistance, as well as the charge-transfer impedance to facilitate the rapid electrode reaction at the electrode/electrolyte interface, thus leading to the good cycling and rate performances of the composite electrode. From various previous studies [17,19,23,56], it is well-known that LDHs have a unique feature of readily tuning types and molar ratios of metal cations, which has enabled LDHs to sever tunable precursors with the great flexibility of preparing various composites used as anode nanomaterials LIBs. Previous studies have demonstrated that CoO/CoFe2O4 nanocomposites [17], derived from scalably prepared CoFe-LDH single-resource precursors, exhibited tunable cycle performances and rate capabilities by varying Co/Fe molar ratio of LDH precursors. Co2SnO4/Co3O4/
Fig. 6. Super-long cycling performance of MgFe2O4/ZnFe2O4 electrodes at (a) 500 mA g1 and (b) 2000 mA g1.
Fig. 7. Comparison of reversible capacity between the electrodes of our composite and MgFe2O4 or ZnFe2O4-containing electrodes reported previously.
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
Al2O3/C composite [23], obtained from a laurate anion-intercalated CoAlSn-LDH single-source precursor, was able to deliver a greatly boosted reversible capacity of 1170 mA h g1 at 100 mA g1 after 100 cycles, in comparison with the bi-active composites without Al2O3 or carbon (Co2SnO4/Co3O4/C, Co2SnO4/Co3O4/Al2O3, and Co2SnO4/Co3O4) that were derived by controlling LDH precursors without Al cation or surfactant intercalation. Therefore, on the basis of chemically balancing types and molar ratios of active/non-active cations, our LDH-precursor synthesis route may be an alternative to design and prepare diverse metal oxide/spinel composites for energy storage.
[8]
[9]
[10]
[11]
4. Conclusions [12]
We have demonstrated a preparation of MgFe2O4/ZnFe2O4 composite with the high electrochemical performances for lithium storage. The composite is obtained by simple thermal decomposition of the ternary MgZnFe-LDH single-resource precursor produced scalablely and subsequent selective removing ZnO and partial MgO formed. The composite indeed delivers a high reverse charge capacity of 1190 mA h g1 after 100 cycles at 100 mA g1, a long cycling stability of a high specific capacity of 981 mA h g1 after 330 cycles at 500 mA g1, and, in particular, a super-long cycling stability of 541 mA h g1 after 1000 cycles at 2000 mA g1. The high enhancements could be attributed to the following distinct advantages: (i) the alloying/conversion ZnFe2O4, and co-existing active MgFe2O4, (ii) low-content inert MgO matrix; (iii) the appropriate specific area and the mesopore size distribution, and (iv) the low resistance. Considering the unique flexibility of chemically balancing types and molar ratios of active/non-active cations, our LDH-precursor synthesis route may be an alternative to design and prepare diverse metal oxide/spinel composites for energy storage.
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
Acknowledgements This work was financially supported by the National Basic Research Program of China (973 Program, 2014CB932102) and the National Natural Science Foundation of China.
[21]
[22]
Appendix A. Supplementary data [23]
Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jallcom.2017.08.005.
[24]
References
[25]
[1] H. Yin, Z. Tang, Ultrathin two-dimensional layered metal hydroxides: an emerging platform for advanced catalysis, energy conversion and storage, Chem. Soc. Rev. 45 (2016) 4873e4891. [2] Y. Jia, L. Zhang, G. Gao, H. Chen, B. Wang, J. Zhou, M.T. Soo, M. Hong, X. Yan, G. Qian, J. Zou, A. Du, X. Yao, A heterostructure coupling of exfoliated NieFe hydroxide nanosheet and defective graphene as a bifunctional electrocatalyst for overall water splitting, Adv. Mater. 29 (2017) 1700017. [3] J. Zhang, H. Hu, Z. Li, X.W. Lou, Double-shelled nanocages with cobalt hydroxide inner shell and layered double hydroxides outer shell as highefficiency polysulfide mediator for lithium-sulfur batteries, Angew. Chem. Int. Ed. 55 (2016) 3982e3986. [4] C. Tang, H.-F. Wang, X.-L. Zhu, B.-Q. Li, Q. Zhang, Advances in hybrid electrocatalysts for oxygen evolution reactions: rational integration of NiFe layered double hydroxides and nanocarbon, Part. Part. Syst. Char. 33 (2016) 473e486. [5] L. Zhang, H. Yao, Z. Li, P. Sun, F. Liu, C. Dong, J. Wang, Z. Li, M. Wu, C. Zhang, B. Zhao, Synthesis of delaminated layered double hydroxides and their assembly with graphene oxide for supercapacitor application, J. Alloys Compd. 711 (2017) 31e41. [6] L. Luo, B. He, W. Kong, Z. Wang, NiCo2S4/NieCo layered double hydroxide nanocomposite prepared by a vapor-phase hydrothermal method for electrochemical capacitor application, J. Alloys Compd. 705 (2017) 349e355. [7] Y. Lu, B. Jiang, L. Fang, F. Ling, F. Wu, B. Hu, F. Meng, K. Niu, F. Lin, H. Zheng, An
[26]
[27]
[28]
[29]
[30]
[31]
[32]
313
investigation of ultrathin nickel-iron layered double hydroxide nanosheets grown on nickel foam for high-performance supercapacitor electrodes, J. Alloys Compd. 714 (2017) 63e70. H. Fu, Z.-j. Du, W. Zou, H.-Q. Li, C. Zhang, Simple fabrication of strongly coupled cobalt ferrite/carbon nanotube composite based on deoxygenation for improving lithium storage, Carbon 65 (2013) 112e123. D. Du, W. Yue, Y. Ren, X. Yang, Fabrication of graphene-encapsulated CoO/ CoFe2O4 composites derived from layered double hydroxides and their application as anode materials for lithium-ion batteries, J. Mater. Sci. 49 (2014) 8031e8039. D. Du, W. Yue, X. Fan, K. Tang, X. Yang, Ultrathin NiO/NiFe2O4 nanoplates decorated graphene nanosheets with enhanced lithium storage properties, Electrochim. Acta 194 (2016) 17e25. Z. Liang, R. Huo, Y.-X. Yin, F. Zhang, S. Xu, Y.-G. Guo, Carbon-supported Ni@ NiO/Al2O3 integrated nanocomposite derived from layered double hydroxide precursor as cycling-stable anode materials for lithium-ion batteries, Electrochim. Acta 108 (2013) 429e434. J.M. Lee, T.H. Gu, N.H. Kwon, S.M. Oh, S.-J. Hwang, Rapid synthetic route to nanocrystalline carbon-mixed metal oxide nanocomposites with enhanced electrode functionality, J. Phys. Chem. C 120 (2016) 8451e8460. S. Xu, Y. Yang, T. Xu, Y. Kuang, M. Dong, F. Zhang, F. Besenbacher, D.G. Evans, Engineered morphologies of layered double hydroxide nanoarchitectured shell microspheres and their calcined products, Chem. Eng. Sci. 66 (2011) 2157e2163. L. Zou, X. Xiang, J. Fan, F. Li, Single-source precursor to complex metal oxide monoliths with tunable microstructures and properties: the case of Mgcontaining materials, Chem. Mater. 19 (2007) 6518e6527. G. Fan, F. Li, D.G. Evans, X. Duan, Catalytic applications of layered double hydroxides: recent advances and perspectives, Chem. Soc. Rev. 43 (2014) 7040e7066. X. Zhao, F. Zhang, S. Xu, D.G. Evans, X. Duan, From layered double hydroxides to ZnO-based mixed metal oxides by thermal decomposition: transformation mechanism and UV-blocking properties of the product, Chem. Mater. 22 (2010) 3933e3942. M. Li, Y.-X. Yin, C. Li, F. Zhang, L.-J. Wan, S. Xu, D.G. Evans, Well-dispersed bicomponent-active CoO-CoFe2O4 nanocomposites with tunable performances as anode materials for lithium-ion batteries, Chem. Commun. 48 (2012) 410e412. X.-F. Zhao, S.-L. Xu, L.-Y. Wang, X. Duan, F.-Z. Zhang, Exchange-biased NiFe2O4/NiO nanocomposites derived from NiFe-layered double hydroxides as a single precursor, Nano Res. 3 (2010) 200e210. M.A. Woo, T.W. Kim, I.Y. Kim, S.-J. Hwang, Synthesis and lithium electrode application of ZnO-ZnFe2O4 nanocomposites and porously assembled ZnFe2O4 nanoparticles, Solid State Ionics 182 (2011) 91e97. F. Wang, S. Zhang, D. Bai, F. Zhang, S. Xu, Graphene-supported binary active Mn0.25Co0.75O solid solution derived from a CoMn-layered double hydroxide precursor for highly improved lithium storage, RSC Adv. 6 (2016) 19716e19722. R.C. Jin, H. Liu, Y. Guan, J. Zhou, G. Chen, ZnFe2O4/C nanodiscs as high performance anode material for lithium-ion batteries, Mater. Lett. 158 (2015) 218e221. L. Zou, F. Li, X. Xiang, D.G. Evans, X. Duan, Self-generated template pathway to high-surface-area zinc aluminate spinel with mesopore network from a single-source inorganic precursor, Chem. Mater. 18 (2006) 5852e5859. B. Wu, S. Zhang, F. Yao, F. Zhang, S. Xu, Synergistic lithium storage of a multicomponent Co2SnO4/Co3O4/Al2O3/C composite from a singlesource precursor, RSC Adv. 5 (2015) 69932e69938. J.G. Kim, Y. Kim, Y. Noh, W.B. Kim, Formation of carbon-coated ZnFe2O4 nanowires and their highly reversible lithium storage properties, RSC Adv. 4 (2014) 27714e27721. C. Hu, S. Guo, G. Lu, Y. Fu, J. Liu, H. Wei, X. Yan, Y. Wang, Z. Guo, Carbon coating and Zn2þ doping of magnetite nanorods for enhanced electrochemical energy storage, Electrochim. Acta 148 (2014) 118e126. R. Jin, G. Liu, C. Liu, L. Sun, High electrochemical performances of hierarchical hydrangea macrophylla like NiCo2O4 and NiCo2S4 as anode materials for Liion batteries, Mater. Res. Bull. 80 (2016) 309e315. M. Wang, Y. Huang, X. Chen, K. Wang, H. Wu, N. Zhang, H. Fu, Synthesis of nitrogen and sulfur co-doped graphene supported hollow ZnFe2O4 nanosphere composites for application in lithium-ion batteries, J. Alloys Compd. 691 (2017) 407e415. D. Bresser, E. Paillard, R. Kloepsch, S. Krueger, M. Fiedler, R. Schmitz, D. Baither, M. Winter, S. Passerini, Carbon coated ZnFe2O4 nanoparticles for advanced lithium-ion anodes, Adv. Energy Mater. 3 (2013) 513e523. L. Cabo-Fernandez, F. Mueller, S. Passerini, L.J. Hardwick, In situ Raman spectroscopy of carbon-coated ZnFe2O4 anode material in Li-ion batteriesdinvestigation of SEI growth, Chem. Commun. 52 (2016) 3970e3973. F. Martinez-Julian, A. Guerrero, M. Haro, J. Bisquert, D. Bresser, E. Paillard, S. Passerini, G. Garcia-Belmonte, Probing lithiation kinetics of carbon-coated ZnFe2O4nanoparticle battery anodes, J. Phys. Chem. C 118 (2014) 6069e6076. A.K. Rai, T.V. Thi, J. Gim, J. Kim, Combustion synthesis of MgFe2O4/graphene nanocomposite as a high-performance negative electrode for lithium ion batteries, Mater. Char. 95 (2014) 259e265. Y. Pan, Y. Zhang, X. Wei, C. Yuan, J. Yin, D. Cao, G. Wang, MgFe2O4 nanoparticles as anode materials for lithium-ion batteries, Electrochim. Acta 109 (2013) 89e94.
314
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
[33] H. Qiao, L. Luo, K. Chen, Y. Fei, R. Cui, Q. Wei, Electrospun synthesis and lithium storage properties of magnesium ferrite nanofibers, Electrochim. Acta 160 (2015) 43e49. [34] C. Wang, Y. Li, Y. Ruan, J. Jiang, Q.-H. Wu, ZnFe2O4-nanocrystal-assembled microcages as an anode material for high performance lithium-ion batteries, Mater. Today Energy 3 (2017) 1e8. [35] Y. Chen, J. Wu, W. Yang, Y. Fu, R. Zhou, S. Chen, L. Zhang, Y. Song, L. Wang, Zn/ Fe-MOFs-derived hierarchical ball-in-ball ZnO/ZnFe2O4@carbon nanospheres with exceptional lithium storage performance, J. Alloys Compd. 688 (2016) 211e218. [36] S. Permien, S. Indris, M. Scheuermann, U. Schuermann, V. Mereacre, A.K. Powell, L. Kienle, W. Bensch, Is there a universal reaction mechanism of Li insertion into oxidic spinels: a case study using MgFe2O4, J. Mater. Chem. A 3 (2015) 1549e1561. pez, G.F. Ortiz, P. Lavela, J.L. Tirado, Towards an all-solid-state battery: [37] M.C. Lo preparation of conversion anodes by electrodepositioneoxidation processes, J. Power Sources 244 (2013) 403e409. pez, G.F. Ortiz, P. Lavela, R. Alca ntara, J.L. Tirado, Improved energy [38] M.C. Lo storage solution based on hybrid oxide materials, ACS Sustain. Chem. Eng. 1 (2013) 46e56. [39] A. Ponrouch, P.-L. Taberna, P. Simon, M.R. Palacín, On the origin of the extra capacity at low potential in materials for Li batteries reacting through conversion reaction, Electrochim. Acta 61 (2012) 13e18. [40] S.J. Rezvani, R. Gunnella, A. Witkowska, F. Mueller, M. Pasqualini, F. Nobili, S. Passerini, A.D. Cicco, Is the solid electrolyte interphase an extra-charge reservoir in Li-ion batteries? ACS Appl. Mater. Interfaces 9 (2017) 4570e4576. [41] J. Lu, Z. Chen, Z. Ma, F. Pan, L.A. Curtiss, K. Amine, The role of nanotechnology in the development of battery materials for electric vehicles, Nat. Nanotechnol. 11 (2016) 1031e1038. [42] P. Poizot, S. Laruelle, S. Grugeon, L. Dupont, J.M. Tarascon, Nano-sized transition-metal oxides as negative-electrode materials for lithium-ion batteries, Nature 407 (2000) 496e499. [43] J. Cai, C. Wu, Y. Zhu, P.K. Shen, K. Zhang, Hierarchical porous acetylene black/ ZnFe2O4@carbon hybrid materials with high capacity and robust cycling performance for Li-ion batteries, Electrochim. Acta 187 (2016) 584e592. [44] D. Narsimulu, B.N. Rao, M. Venkateswarlu, E.S. Srinadhu, N. Satyanarayana, Electrical and electrochemical studies of nanocrystalline mesoporous MgFe2O4 as anode material for lithium battery applications, Ceram. Int. 42 (2016) 16789e16797.
[45] L. Yao, H. Deng, Q.-A. Huang, Q. Su, G. Du, Three-dimensional carbon-coated ZnFe2O4 nanospheres/nitrogen-doped graphene aerogels as anode for lithium-ion batteries, Ceram. Int. 43 (2017) 1022e1028. [46] Y. Yin, N. Huo, W. Liu, Z. Shi, Q. Wang, Y. Ding, J. Zhang, S. Yang, Hollow spheres of MgFe2O4 as anode material for lithium-ion batteries, Scr. Mater. 110 (2016) 92e95. [47] B. Jiang, C. Han, B. Li, Y. He, Z. Lin, In-situ crafting of ZnFe2O4 nanoparticles impregnated within continuous carbon network as advanced anode materials, ACS Nano 10 (2016) 2728e2735. [48] L. Luo, D. Li, J. Zang, C. Chen, J. Zhu, H. Qiao, Y. Cai, K. Lu, X. Zhang, Q. Wei, Carbon-coated Magnesium ferrite nanofibers for lithium-ion battery anodes with enhanced cycling performance, Energy Technol. (2017), http:// dx.doi.org/10.1002/ente.201600686. [49] X. Yang, H. Xue, Q. Yang, R. Yuan, W. Kang, C.-S. Lee, Preparation of porous ZnO/ZnFe2O4 composite from metal organic frameworks and its applications for lithium ion batteries, Chem. Eng. J. 308 (2017) 340e346. [50] L. Liu, R. Gao, L. Sun, S. Han, D. Chen, Z. Hu, X. Liu, Ultrahigh cycling stability and rate capability of ZnFe2O4@graphene hybrid anode prepared through a facile syn-graphenization strategy, New J. Chem. 40 (2016) 3139e3146. [51] Y. Yin, W. Liu, N. Huo, S. Yang, Synthesis of vesicle-like MgFe2O4/graphene 3D network anode material with enhanced lithium storage performance, ACS Sustain. Chem. Eng. 5 (2017) 563e570. [52] J. Liu, J. Xiao, X. Zeng, P. Dong, J. Zhao, Y. Zhang, X. Li, Combustion synthesized macroporous structure MFe2O4 (M ¼Zn, Co) as anode materials with excellent electrochemical performance for lithium ion batteries, J. Alloys Compd. 699 (2017) 401e407. [53] D. Cai, H. Zhan, T. Wang, MOF-derived porous ZnO/ZnFe2O4 hybrid nanostructures as advanced anode materials for lithium ion batteries, Mater. Lett. 197 (2017) 241e244. [54] D. Bresser, S. Passerini, B. Scrosati, Leveraging valuable synergies by combining alloying and conversion for lithium-ion anodes, Energy Environ. Sci. 9 (2016) 3348e3367. [55] Y. Sharma, N. Sharma, G.V.S. Rao, B.V.R. Chowdari, Li-storage and cyclability of urea combustion derived ZnFe2O4 as anode for Li-ion batteries, Electrochim. Acta 53 (2008) 2380e2385. [56] J. Liu, Y. Li, X. Huang, G. Li, Z. Li, Layered double hydroxide nano- and microstructures grown directly on metal substrates and their calcined products for application as Li-ion battery electrodes, Adv. Funct. Mater. 18 (2008) 1448e1458.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Mesoporous spinel ferrite composite derived from a ternary MgZnFe-layered double hydroxide precursor for lithium storage Duan Wang a, *, Jingli Wu a, Daxun Bai b, Rongrong Wang c, Feng Yao b, Sailong Xu b, * a
School of Chemical and Environmental Engineering, North University of China, Taiyuan 030051, China State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, China c Xi'an Institute of Electromechanical Information Technology, Xi'an 710065, China b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 April 2017 Received in revised form 29 June 2017 Accepted 1 August 2017 Available online 4 August 2017
Layered double hydroxides (LDHs), also known as hydrotalcite-like anionic clays, are very convenient precursors with a tunable flexibility toward multifunctional nanomaterials, especially in energy storage. Typical methods to improve lithium storage are to introduce additional or self-generating carbonaceous supports to LDH-derived transition metal oxides as anode nanomaterials which can host lithium mainly though a conversion mechanism. Here, we describe a preparation of mesoporous spinel ferrite composite (MgFe2O4/ZnFe2O4) for lithium storage, which is assisted by a combined conversion and alloying mechanism. The composite is derived by a thermal decomposition of a scalablely produced singleresource precursor of ternary Mg2þZn2þFe3þ-layered double hydroxide (Mg2þZn2þFe3þ-LDH), and subsequent selective etching. Electrochemical test shows that the electrode delivers an exceptional electrochemical performance, i.e., a reversible capacity of 1190 mA h g1 after 100 cycles at 100 mA g1, and, in particular, a reversible capacity of 981 mA h g1 at 500 mA g1 after 330 cycles, as well as a reversible capacity of 541 mA h g1 at 2000 mA g1 after 1000 cycles. The high electrochemical performance could be attributed to the following features: the combined alloying and conversion mechanisms of ZnFe2O4, synergistic MgFe2O4, and slight-content MgO as a non-active matrix, as well as an appropriate specific area and mesoporous size distribution. Our results show that the cation-tunable LDH precursor-derived synthesis route might be an alternative to prepare multiple-component composites of spinel ferrites and transition metal oxides. © 2017 Elsevier B.V. All rights reserved.
Keywords: Layered double hydroxide precursor Spinel ferrite Sintering Domain structure Energy storage materials
1. Introduction Layered double hydroxides (LDHs), also known as hydrotalcitelike anionic clays, have attracted increasing interest in various potential applications, especially including electrochemical energy storage and conversion [1e7]. LDHs with structure based on Mg(OH)2-like layers, possess the general formula of [M2þ13þ xþ n 2þ (such as Mg2þ, xM x(OH)2] (A )x/n·yH2O, in which cationic M Zn2þ, Ni2þ, and Co2þ) and M3þ (such as Al3þ, Fe3þ, and Co3þ) are well-orderedly occupied within a LDH layer, Ane anions (such as CO2 3 , NO3 , and surfactant anion) are well-intercalated between the LDH interlayer galleries, and the value (x) of M3þ/(M2þ þ M3þ) molar ratio ranges typically between 0.2 and 0.33. Despite the huge
* Corresponding authors. E-mail addresses: [email protected] (D. Wang), [email protected]. cn (S. Xu). http://dx.doi.org/10.1016/j.jallcom.2017.08.005 0925-8388/© 2017 Elsevier B.V. All rights reserved.
family, the LDHs could be basically recognized as convenient precursors with a unique feature to easily tune the types of metal cations and intercalated anions, as well as M2þ/M3þ molar ratios over a wide range, which have enabled the LDHs as tunable precursors to prepare electrode nanomaterials for lithium ion batteries (LIBs). Two main approaches have hitherto been applied to prepare LDH precursor-derived products as anode nanomaterials for LIBs. On one hand, transition metal oxide/spinel/carbon composites were prepared typically by calcining tunable LDH precursors supported on the additionally introduced carbon (such as carbon nanotube (CNT) and graphene (G)). The TMOs/carbon composites included CoFe2O4/CNT [8], CoO/CoFe2O4/G [9], and NiO/NiFe2O4/G [10] derived from Co2þFe2þFe3þ-LDH/CNT, Co2þFe3þ-LDH/G, and Ni2þFe3þ-LDH/G precursors, respectively, which were able to exhibit highly cycling stability and rate capability compared with the spinels without carbonaceous support. On the other hand, diverse spinel/carbon composites were also synthesized via the
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
utilization of the self-generating carbon during the carbothermal process of different LDH/organic molecule precursors. Such spinel/ carbon composites included Ni@NiO/Al2O3/C from a NiAl-LDH/ glucose mixture [11], and C-NiO/NiFe2O4 derived by a rapid catalyzing C2H2 flow on NiFe-LDH surface at 300 C via chemical vapor deposition (CVD) [12]. The self-generating carbon is able to greatly promote the cycling stability and rate capability of the resulting spinel/carbon composites, with the aid of the high conductivity of the self-generating carbonaceous nanomaterials. Indeed, both types of spinel-containing composites delivered the high electrochemical performances, which were based mainly on the conversion mechanism of the nanosized active transition metal oxide/ spinel and also benefit from the additionally introduced or selfgenerating conductive carbonaceous supports. In this present study, we describe a preparation of MgFe2O4/ ZnFe2O4 composite as a conversion/alloying nanomaterials for lithium storage. The carbon-free composite was prepared via a thermal decomposition of a scalablely prepared Mg2þZn2þFe3þLDH single-source precursor and subsequent selective etching (Scheme 1), and thus endowed with the features of boosting the electrochemical performances: conversion/alloying ZnFe2O4, synergistic active MgFe2O4, and low-content non-active MgO, as well as an appropriate specific area and mesoporous size distribution. Electrochemical test shows that the composite is indeed able to deliver the expected electrochemical performances. 2. Experimental 2.1. Preparation of Mg2þZn2þFe3þ-LDH precursor The Mg2þZn2þFe3þ-LDH precursor was prepared via a scalable procedure of separate nucleation and aging steps (SNAS) [13]. In brief, a salt solution, which contained Mg(NO3)2·6H2O, Zn(NO3)2·6H2O and Fe(NO3)3·9H2O with a (MgþZn)/Fe molar ratio of 2: 1, was obtained via dissolution of the salts into a freshly deionized water to yield a resulting solution with a total cationic concentration of 1.2 M. An aqueous base solution was also acquired, with a fixed concentration of NaOH ([OH] ¼ 2 2þ 2þ 3þ 1.6([Mg ] þ [Zn ] þ [Fe ]) and Na2CO3 ([CO3 ] ¼ 2[Fe3þ]). The as-prepared solutions were introduced automatically and simultaneously into a commercial chemical reactor operated at a rotor speed of 3000 rpm. The resulting slurry was mixed continuously for 2 min, then collected, and aged at 70 C for 48 h. The Mg2þZn2þFe3þ-LDH precursor was obtained by centrifugation, then rinsing thoroughly the slurry several times with alternate deionized water and ethanol, and finally by drying in vacuum at 60 C overnight. 2.2. Preparation of MgFe2O4/ZnFe2O4/MgO composite The composite was prepared by calcining the as-prepared Mg2þZn2þFe3þ-LDH precursor. The calcination was carried out in
307
a muffle furnace at 900 C for 3 h with a ramping rate of 5 C/min. The calcined product was then rinsed by carefully stirring in an aqueous solution of (NH4)2SO4 (10 wt %) at 70 C for 48 h to remove the partial MgO, and then in aqueous solution of NaOH (1 M) at 40 C for 5 h to remove ZnO completely. For comparison of XRD patterns, the mixture of MgFe2O4 and ZnFe2O4 was prepared by physically mixing the individual MgFe2O4 and ZnFe2O4. The individual MgFe2O4 and ZnFe2O4 were derived from MgFe- and ZnFe-LDH precursors obtained under the same experimental conditions, respectively. Effective removals of MgO and ZnO were performed by using HCl aqueous solution (1 M) at 80 C for 3 h and NaOH aqueous solution (1 M) at 40 C for 5 h, respectively. 2.3. Characterization Powder X-ray diffraction (XRD) was used on a powder diffractometer (Rigaku XRD-6000, Cu-Ka radiation, l ¼ 1.542 Å) operated at a scanning speed of 10 /min. To acquire sample morphology, scanning electron microscopy (SEM) images were recorded on a Zeiss Supra 55 instrument. Transmission electron microscopy (TEM) observation was carried out on an electron microscope (JEOL JEM-2100) with an equipped EDX, which was operated at an accelerating voltage of 200 kV. The samples for SEM and TEM visualizations were dispersed in ethanol via sonication and then deposited manually onto silicon wafer or copper microgrid. Elemental analysis was conducted for metal ions by using an inductively coupled plasma emission spectrometer (ICP-ES, Shimadzu). Specific surface area and pore size distribution were determined quantitatively from nitrogen adsorption/desorption isotherms collected at 77 K on a Nova 1200, Quantachrome apparatus. For the pre-treatments before the measurement, the sample was kept for drying in vacuum at 393 K for 8 h. The surface area was computed via the BrunauereEmmetteTeller (BET) method, and the pore size distribution was calculated through the BarretteJoynereHalenda (BJH) method. X-ray photoelectron spectroscopy (XPS) measurement was applied on an X-ray Photoelectron Spectrometer (Kratos Axis ULTRA) assembled with a 165 mm hemispherical electron energy analyser. 2.4. Electrochemical measurement The assembly of all Swagelok-type cells was conducted in a commercial argon-filled glove box. To prepare the working electrodes, a mixer was prepared by using active material/poly(vinyl difluoride) (PVDF)/super-P acetylene black at 70/10/20 (wt %), and then carefully pasted on a clean surface of Cu foil. The loading mass of the composite was ca. 10 mg cm2. Lithium foil and glass fiber (GF/D, both from Whatman) were utilized as the counter electrode and the separator, respectively. The electrolyte consisted of a solution of 1 M LiPF6 salt, which was dissolved into a solvent mixture of ethylene carbonate (EC): dimethyl carbonate (DMC): diethyl
Scheme 1. Schematic illustration of preparing mesoporous MgFe2O4/ZnFe2O4/MgO composite derived from a ternary MgZnFe-LDH precursor: initial calcination at 900 C in air, subsequent partial removal of MgO by (NH4)2SO4 solution and removal of ZnO by NaOH solution.
308
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
carbonate (DEC) at 1: 1: 1 (wt %) plus vinylene carbonate (VC, 2 wt %). All were purchased commercially (from Tianjing Jinniu Power Sources Material Co. Ltd., China). Galvanostatic cycling was recorded for all the assembled cells by using a commercial cell-testing system (LAND CT2100A) between the voltage range from 0.01 to 3.0 V (vs. Liþ/Li). Cyclic voltammogram (CV) were measured on an electrochemistry work station (CHI 760E) at a scan rate of 0.1 mV s1 in the range of 0e3.0 V (vs. Liþ/Li). Electrochemical impedance spectrum (EIS) was recorded on an advanced electrochemical system (Parstat 2273) operated at an open circuit potential over the frequency range between 100 kHz and 0.1 Hz. 3. Results and discussion The Mg2þZn2þFe3þ-LDH precursor was prepared by using our SNAS method characteristic of a scalable production [13]. The obtained LDH precursor shows a XRD pattern with the typical layered feature of a LDH precursor (Fig. S1). The precursor was subjected to the calcination at 900 C in air. The calcination leads to the formation of cubic MgO, wurzite-structured ZnO, as well as the spinels of MgFe2O4 (JCPDS No. 88-1943) and ZnFe2O4 (JCPDS no. 82-1042), as shown by the XRD result in Fig. 1a. Further selective etching was conducted to remove the partial MgO phase and ZnO phases by using the acidic (NH4)2SO4 solution (10 wt %) and the basic NaOH solution (1 M), respectively. Considering that the amount of non-active MgO has an influence on electrochemical performance of the composite electrode, the residual amount of MgO were tuned by varying type (such as strongly acidic HNO3 and weakly acidic (NH4)2SO4) and concentration of etching solution, as well as etching temperature and duration [14]. A proper etching condition was optimized, for example, using (NH4)2SO4 solution (10 wt %, pH z 6.0) at a temperature of 70 C for 48 h. The post-etched product shows the very reflection peaks of MgFe2O4/ZnFe2O4 and MgO (Fig. 1a), the latter of which was not removed completely due to the use of weakly acidic (NH4)2SO4 solution. From previous studies [14e16], it is well-known that thermal decomposition of a typical MIIMIII-LDH precursor in air or under an inert atmosphere yields mixed metal oxides, which consist of metal oxide (MIIO) and spinel (MIIMIII2O4). Such studies include ZnO/ ZnAl2O4 composite calcined from ZnAl-LDH from 500 C to 800 C in air [16], CoO/CoFe2O4 from a CoFe-LDH precursor in a He
atmosphere at 700 C for 2 h [17], NiO/NiFe2O4 from a NiFe-LDH precursor 500 C in air [18], ZnO/ZnFe2O4 nanoparticle from ZnFe-LDH [19]. Furthermore, the formation of MIIO/MIIMIII2O4 has been corroborated by a transformation mechanism from a ZnAlLDH precursor monitored by using in situ XRD [16], i.e., the preferential nucleation and growth of ZnO between 200 and 500 C, as well as the subsequent doping of Al cation into ZnO and the formation of ZnAl2O4 between 500 and 800 C. Note that MII/MIII molar ratio of a MIIMIII-LDH precursor is typically in the range from 5:1 to 2:1, the doping of MIII cation into MIIO results in the good dispersion MIIMIII2O4 phase embedded in a matrix of excess MIIO continuous phase [14e16]. In addition, solid-solution metal oxide could be generated by the thermal decomposition of MIIMIII-LDH [20]. In distinct contrast to the above cases, graphene-supported Mn0.25Co0.75O solid solution was indeed formed by thermal decomposition of CoMn-LDH/graphene oxide at 600 C for 2 h in Ar [20]. The formation of Mn0.25Co0.75O could be ascribed mainly to the Mn doping into the CoO phase, owing to the carbothermal process in Ar atmosphere, as well as the comparable size and chemical nature of the Mn2þ and Co2þ cations, and the ease of miscibility of the MnO/CoO solid solution. Therefore, in the case of a typical precursor of MgZnFe-LDH, the thermal decomposition in air leads to the very formation of MgO/MgFe2O4/ZnO/ZnFe2O4 composite, instead of Mg0.64Fe2.36O4 and the like. To experimentally exclude a possibility that the formed spaniels may also contain a MgxZn1-xFe2O4 solid solution, a comparison of XRD pattern was carried out between the above spinel composite and the mixture of individual MgFe2O4 and ZnFe2O4. Perusal of the comparison clearly reveals that no shifts of XRD reflection peaks of the above spinel composite are observed with respect to the mixture (Fig. 1b). No shift of XRD pattern indeed rules out the possibility of MgxZn1xFe2O4 solid solution formed owing to the cationic doping of Mg or Zn. We thus believe that the composite is obtained, which consists of MgFe2O4, ZnFe2O4, and MgO, thus denoted as MZO/MgO. To quantitatively determine the exact content of each component, we further carried out element analysis to the MZO/MgO composite by using the ICP-ES measurement. The ICP result gives each concentration of Fe, Mg, and Zn cations, and the molarities of Fe, Mg, and Zn cations were determined, thus yielding a (MgþZn)/ Fe molar ratio of 0.54 and a Mg/Zn molar ratio of 0.63. With respect to the theoretic molar ratio (0.5) for both MgFe2O4 and ZnFe2O4 of the post-etched composite, the excess amount of Mg cation is determined to be MgO. Therefore, a MgFe2O4/ZnFe2O4 molar ratio
Fig. 1. XRD patterns of the products calcined from the MgZnFe-LDH precursor: (a) before and after the selective etching to remove the partial MgO, and ZnO. (b) No shift of XRD observed patterns between the MZO/MgO composite and the mMZO mixture.
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
is calculated to be 1: 2, and a (MgFe2O4 þ ZnFe2O4)/MgO molar ratio to be 11: 1. The excess of MgO is indeed consistent with the abovementioned XRD result of MgO observed, which is partially removed by using weakly acidic (NH4)2SO4 aqueous solution. As expected, we obtain the MZO/MgO composite that is composed of highcontent active MgFe2O4 and ZnFe2O4, as well as slight-content non-active MgO. From a previous study [21], it is expected that the slight amount of non-active MgO is expected to act as a matrix to effectively accommodate the large volume change during the lithiation/delithiation process, thus facilitating a high cycle stability. The morphology of the MZO/MgO composite was visualized by using SEM and TEM. Fig. 2a shows that the composite exhibits the irregular nanoplatelets of the MZO/MgO composite, which are much smaller than those of the Mg2þZn2þFe3þ-LDH precursor typically with the hydrotalcite-like nanoplatelets (as shown in Fig. 2b). Such a visible change in dimensional size can be ascribed to the dehydroxylation of the pristine LDH precursor during the thermal decomposition [22], as reported previously for the hollow CoFe2O4 sphere converted from Co2þFe2þFe3þ-LDH shell sphere [13], and ZnO/ZnAl2O4 nanoparticle from ZnAl-LDH [16]. Note that SEM and TEM images show the severe aggregation of the composite nanoparticles. The aggregation is typically observed for the MIIO/ MIIMIII2O4 composites calcined from different MIIMIII-LDH precursors. The aggregation can be ascribed to the initial dehydroxylation of the layers and loss of volatile species arising from decomposition of the interlayer carbonate anions below ca. 200 C [14,16], and the subsequent topotactic transformation to MIIO/MIIMIII3O4 phases during the sintering process at moderate and high temperatures [14,16,18]. TEM observation further manifests that the nanoplatelet-like morphology of the MgFe2O4/ZnFe2O4 composite is well preserved, and also the MgFe2O4/ZnFe2O4 nanoplatelet contains both domains in gray contrast and pores in bright contrast (Fig. 2c). The features could be ascribed to the dehydroxylation of LDH precursor, the topotactic transformation to the resultant transition metal/spinel composite, and the selective etching. A HRTEM image reveals that the MgFe2O4/ZnFe2O4 composite is highly crystalline, with the well-defined lattice fringes in
309
two regions determined to be 0.48 nm that corresponds to the (111) crystalline planes of the MgFe2O4 and ZnFe2O4 phases (Fig. 2d). Especially, a visible heterojunction nanostructure is observed with a slight mutual dislocation. Similar HRTEM visualization of interfacial nanodomains was also reported early, such as CoO/CoFe2O4 composite calcined from a CoFe-LDH precursor [17] and Co2SnO4/ Co3O4/Al2O3/C composite derived from a laurate anion-intercalated CoAlSn-LDH precursor [23]. The chemical states were examined to the species of the MZO/ MgO composite by using XPS technique. Fig. 3a shows that for Fe 2p, the bands are centered at 710.9 and 712.6 eV, as well as 725.3 and 718.9 eV, which can be attributed to Fe 2p3/2 and Fe 2p1/2, respectively [24]. In the case of Mg 1s, the strong peak is clearly visible at 1304.2 eV (Fig. 3b), corresponding to the existence of Mg2þ of the composite. In the case of Zn 2p, the peaks centered at 1022.1 and 1045.6 eV can be ascribed to Zn 2p3/2 and Zn 2p1/ 2þ specie 2,respectively (Fig. 3c), reflecting the oxidation state of Zn of the composite. For O 1s, the spectrum, with an additional shoulder, can be de-convoluted by three distinct peaks centered at 529.6, 530.7, and 532.2 eV (Fig. 3d). The strong one at 529.6 eV can be ascribed to the surface lattice oxygen (O2), viz., O-Fe(Mg, Zn) in the composite, while those other two at high energy can be assigned to either the chemisorbed/dissociated oxygen, or the OH species on the surface of the composite [25]. N2 adsorption-desorption isotherm was measured to examine the texture structure of the MZO/MgO composite. Fig. 4a shows that the isotherms of the sample possess a distinct H3-type hysteresis loop at a relative pressure ranges 0.05e0.94 P/P0, which can be ascribed to type IV (IUPAC classification). The surface area is calculated to be 25.6 m2 g1, and the pore size distribution is determined to rang between 2 and 10 nm (Fig. 4b). This pore size distribution clearly manifests the typical mesoporous characteristic of the MZO/MgO composite, well consistent with the pores in bright contrast by TEM observation. The formation of porous stacking for the resulting MIIMIII2O4 nanoparticles can be ascribed to the selective removal of MIIO from the MIIO/MIIMIII2O4 composite. Similar porous nanostructures were reported previously, such as ZnAl2O4 mesopore networks obtained by calcining ZnAl-
Fig. 2. SEM images of (a) the MZO/MgO composite and (b) MgZnFe-LDH precursor. (c) TEM and (d) HRTEM images of the MZO/MgO composite.
310
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
Fig. 3. XPS spectra of (a) Fe 2p, (b) Mg 1s, (c) Zn 2p, and (d) O 1s for the MZO/MgO composite.
Fig. 4. N2 adsorption-desorption isotherms and pore size distribution of MZO/MgO composite.
LDH precursor and subsequent etching of ZnO [22], MgMIII2O4 (M ¼ Al, Fe) macroporous frameworks available by decomposition of MgAl- and MgFe-LDH precursors and removal of MgO [14], as well as porous ZnFe2O4 prepared by calcining ZnFe-LDH precursor and subsequent etching of ZnO [19]. From a pervious study [26], it is well-known that the approprixate surface area is able to provide enough contact area and lithium storage sites for the electrodes,
and the mesoporous structure can effectively facilitate the transportation of Liþ and electrolyte molecules and also relieve the volume changes of the active materials during the repeated charge/ discharge processes. Therefore, we expect that the approprixate surface area and the narrow mesopore size distribution of the MZO/ MgO composite could greatly facilitate enhancing the lithium storage.
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
We initially carried out CV measurement to the MZO/MgO composite electrode. Fig. 5a shows that in the first cathodic sweep, two cathodic peaks at 0.71 V and 0.52 V are clearly visible. The weak peak at 0.71 V could be ascribed to the formation of the intermediate LinZnFe2O4, owing to the Li-intercalation into ZnFe2O4 (Eqn. (1)) [27e30]. The other strong peak at 0.52 V corresponds to the reaction of Liþ with the intermediate LinZnFe2O4 to yield Zn0 and Fe0 (Eqn. (2)), and the further lithiation of Zn0 to give rise to a LixZn alloy (Eqn. (3)) [27e30], as well as the reduction of MgFe2O4 to Fe0 and MgO (Eqn. (4)) [31], along with the formation of a solid electrolyte interface (SEI) layer and Li2O [31]. During the first oxidation sweep, a broad anodic peak is visible at 1.61 V, corresponding to the oxidation of the metallic Zn0 into Zn2þ and Fe0 into Fe2O3 (Eqns. (5) and (6)) [31e33], as inactive MgO can not participate in the reversible reaction of Eqn. (4). From the second cycle, both cathodic and anodic peaks appear to shift positively [27e30,34,35], only one strong cathodic peak appears at 0.89 V, and one remarkable anodic peak is observed to shift to 1.75 V. Apparently, a good overlap is visible between the further sweeps, indicating a highly reversible uptake and release of lithium upon the completion of the initial cycling. The electrochemical behavior of the MZO/MgO composite electrode could be summarized by the following equations [27e35]: ZnFe2O4 þ xLiþ þ xee / LixZnFe2O4
(1)
LixZnFe2O4 þ (8 e x)Liþ þ (8 e x)ee / Zn0 þ 2Fe0 þ 4Li2O
(2)
Zn0 þ xLiþ þ xee 4 LixZn
(3)
311
MgFe2O4 þ 6ee þ 6Liþ / MgO þ 2Fe0 þ 3Li2O
(4)
Zn0 þ Li2O 4 ZnO þ 2Liþ þ 2ee
(5)
2Fe0 þ 3Li2O 4 Fe2O3 þ 6ee þ 6Liþ
(6)
We then conducted the galvanostatic discharge/charge test to the MZO/MgO composite electrode between 0.01 and 3 V at 100 mA g1. Fig. 5b exhibits that during the first discharge process, the initial potential decreased rapidly, ascribed to the small amount of Liþ ion insertion into the electrodes. Two voltage plateaus appear, corresponding to the two reduction peaks in the first cathodic sweep of the above CV curve [28,29,34]. The short voltage plateau was shown up at ca. 0.95 V, manifesting the formation of metallic Fe0 and Zn0 nanoparticles into an amorphous Li2O matrix owing to the conversion reaction [28,29,35]. The obvious long plateau was visible at ca. 0.67 V, which could be associated with the complete decomposition of LixZnFe2O4 and MgFe2O4 to Zn0, Fe0, which, together with amorphous MgO, are embedded in a Li2O matrix [36]. At the end of the discharge profile, the potential smoothly decreases from 0.67 V to the lower cut-off voltage, which could be related to the formation of LixZn alloys [28,29,35]. Additional capacity in the tail of the curve at low potential is usually observed just after the conversion reaction. Capacitive-like behavior is characterized by a linear potential decay/increase in the potentialedischarge/charge curve [37]. The extra contribution was attributed either to a capacitive-like behavior for the conversion-reaction transition metal oxides (Fe, Co, and Nicontaining oxides) [37,38], or to SEI formation for conversionalloying transition metal oxides (such as ZnFe2O4) [39,40]. The
Fig. 5. (a) The first three CV curves for MZO/MgO composite electrode between 0.01 and 3.0 V at 0.1 mV s1, and (b) the first three charge/discharge profiles at 100 mA g1, (c) cycling behavior at a rate of 100 mA g1, and (d) rate capability.
312
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
initial discharge and charge capacities of the MZO/MgO composite electrode are determined to be 1911 and 1004 mA h g1, thus giving rise to a Coulombic efficiency of 53%. The loss of capacity is ascribed typically to the irreversible decomposition of electrolyte to form the solid electrolyte interface (SEI) layer. In addition, a good overlap is found between the second and third cycles, which strongly suggests a possibility of the good reversibility of the MZO/MgO electrode. We further examined the cycling stability of the MZO/MgO composite electrode at a current density of 100 mA g1. Fig. 5c shows that the charge capacity of the composite electrode exhibits an ascending trend after the 60th cycles, and almost hovers at 1190 mA h g1 over the rang from the 60th and the 100th cycles, indeed indicating a good cycling stability of the composite electrode. Such a type of “ascending-then-stable” trend was also reported previously, which is closely associated with both the activation of the oxides resulting from the nanosize effect [41,42]. Furthermore, the value of Coulombic efficiency remains stable at ca. 98% beginning with the second cycles. The rate capability was also evaluated to the electrode at different current densities. Fig. 5d depicts that the MZO/MgO composite electrode delivers the charge capacities of 993.4, 912.1, 806, 717.8, 620.6, and 557 mA h g1 with the current densities varying initially from 100, then to 200, 400, 800, 1000, eventually to 2000 mA g1, respectively. When the current density is reset back to the current density of 100 mA g1, the capacity is able to return to 1150 mA h g1, thus reflecting a good rate capability. We further estimated the long-term cycling stability of the MgZO/MgO composite electrode at large current densities. Fig. 6a shows that at a 500 mA g1, a similar “decrease-increase-stable” trend of charge capacity is observed. The reversible charge capacity is 989 mA h g1 after the 100th cycle, then 857 mA h g1 after the 130th cycle, and eventually maintains at 981 mA h g1 after the 330th cycle. In particular, at a high current density of 2000 mA g1, such a “decrease-increase-stable” trend is also clearly resolved in Fig. 6b, with a charge capacity of 541 mA h g1 after the 1000th cycle.
A further comparison of reversible specific capacity was conducted between the electrodes of our MZO/MgO composite and MgFe2O4 or ZnFe2O4 containing electrodes reported in previous studies [43e53]. Fig. 7 displays that the reversible charge capacity of our composite electrode could be comparable to those of the other electrodes when considered the approximate amount of electrochemical active nanomaterials used. Considering the significant improvements, we suggest that the following compositional and nanostructural characteristics can be responsible for the contribution. (i) The combination of alloying and conversion mechanisms of ZnFe2O4, together with the active MgFe2O4, could contribute greatly to the reversible capacity [54]. The alloying mechanism is considered to provide the additional capacity [28,30,55], which has been recognized to be a capacitive-like behavior [38]. (ii) The appropriate specific area could provide enough contact area between MgFe2O4/ZnFe2O4 and electrolyte, and the mesopore size distribution is capable of offering void space to accommodate the strain resulting from the volume expansion/ contraction during cycles [26]. (iii) Low-content MgO is wellknown to act as an inert matrix to accommodate strain change and prevent aggregation of active components [15]. This contribution is similar to that of nonactive Al2O3 reported in a previous study [19]. (iv) The mesoporous MZO/MgO composite may possess a relatively stable interfacial resistance, which could facilitate improving the stability of SEI structure and interface charge transfer properties of the composite electrode. To support this hypothesis, EIS measurement was carried out to the composite electrode. Fig. S2 shows that the Nyquist plot consists of two characteristic parts, a depressed semicircle over the high-to-middle frequency and an inclined line tail in the low frequency range. The MZO/MgO composite electrode exhibits a small diameter of the semicircle, suggesting low SEI resistance and contact resistance, as well as the charge-transfer impedance to facilitate the rapid electrode reaction at the electrode/electrolyte interface, thus leading to the good cycling and rate performances of the composite electrode. From various previous studies [17,19,23,56], it is well-known that LDHs have a unique feature of readily tuning types and molar ratios of metal cations, which has enabled LDHs to sever tunable precursors with the great flexibility of preparing various composites used as anode nanomaterials LIBs. Previous studies have demonstrated that CoO/CoFe2O4 nanocomposites [17], derived from scalably prepared CoFe-LDH single-resource precursors, exhibited tunable cycle performances and rate capabilities by varying Co/Fe molar ratio of LDH precursors. Co2SnO4/Co3O4/
Fig. 6. Super-long cycling performance of MgFe2O4/ZnFe2O4 electrodes at (a) 500 mA g1 and (b) 2000 mA g1.
Fig. 7. Comparison of reversible capacity between the electrodes of our composite and MgFe2O4 or ZnFe2O4-containing electrodes reported previously.
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
Al2O3/C composite [23], obtained from a laurate anion-intercalated CoAlSn-LDH single-source precursor, was able to deliver a greatly boosted reversible capacity of 1170 mA h g1 at 100 mA g1 after 100 cycles, in comparison with the bi-active composites without Al2O3 or carbon (Co2SnO4/Co3O4/C, Co2SnO4/Co3O4/Al2O3, and Co2SnO4/Co3O4) that were derived by controlling LDH precursors without Al cation or surfactant intercalation. Therefore, on the basis of chemically balancing types and molar ratios of active/non-active cations, our LDH-precursor synthesis route may be an alternative to design and prepare diverse metal oxide/spinel composites for energy storage.
[8]
[9]
[10]
[11]
4. Conclusions [12]
We have demonstrated a preparation of MgFe2O4/ZnFe2O4 composite with the high electrochemical performances for lithium storage. The composite is obtained by simple thermal decomposition of the ternary MgZnFe-LDH single-resource precursor produced scalablely and subsequent selective removing ZnO and partial MgO formed. The composite indeed delivers a high reverse charge capacity of 1190 mA h g1 after 100 cycles at 100 mA g1, a long cycling stability of a high specific capacity of 981 mA h g1 after 330 cycles at 500 mA g1, and, in particular, a super-long cycling stability of 541 mA h g1 after 1000 cycles at 2000 mA g1. The high enhancements could be attributed to the following distinct advantages: (i) the alloying/conversion ZnFe2O4, and co-existing active MgFe2O4, (ii) low-content inert MgO matrix; (iii) the appropriate specific area and the mesopore size distribution, and (iv) the low resistance. Considering the unique flexibility of chemically balancing types and molar ratios of active/non-active cations, our LDH-precursor synthesis route may be an alternative to design and prepare diverse metal oxide/spinel composites for energy storage.
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
Acknowledgements This work was financially supported by the National Basic Research Program of China (973 Program, 2014CB932102) and the National Natural Science Foundation of China.
[21]
[22]
Appendix A. Supplementary data [23]
Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jallcom.2017.08.005.
[24]
References
[25]
[1] H. Yin, Z. Tang, Ultrathin two-dimensional layered metal hydroxides: an emerging platform for advanced catalysis, energy conversion and storage, Chem. Soc. Rev. 45 (2016) 4873e4891. [2] Y. Jia, L. Zhang, G. Gao, H. Chen, B. Wang, J. Zhou, M.T. Soo, M. Hong, X. Yan, G. Qian, J. Zou, A. Du, X. Yao, A heterostructure coupling of exfoliated NieFe hydroxide nanosheet and defective graphene as a bifunctional electrocatalyst for overall water splitting, Adv. Mater. 29 (2017) 1700017. [3] J. Zhang, H. Hu, Z. Li, X.W. Lou, Double-shelled nanocages with cobalt hydroxide inner shell and layered double hydroxides outer shell as highefficiency polysulfide mediator for lithium-sulfur batteries, Angew. Chem. Int. Ed. 55 (2016) 3982e3986. [4] C. Tang, H.-F. Wang, X.-L. Zhu, B.-Q. Li, Q. Zhang, Advances in hybrid electrocatalysts for oxygen evolution reactions: rational integration of NiFe layered double hydroxides and nanocarbon, Part. Part. Syst. Char. 33 (2016) 473e486. [5] L. Zhang, H. Yao, Z. Li, P. Sun, F. Liu, C. Dong, J. Wang, Z. Li, M. Wu, C. Zhang, B. Zhao, Synthesis of delaminated layered double hydroxides and their assembly with graphene oxide for supercapacitor application, J. Alloys Compd. 711 (2017) 31e41. [6] L. Luo, B. He, W. Kong, Z. Wang, NiCo2S4/NieCo layered double hydroxide nanocomposite prepared by a vapor-phase hydrothermal method for electrochemical capacitor application, J. Alloys Compd. 705 (2017) 349e355. [7] Y. Lu, B. Jiang, L. Fang, F. Ling, F. Wu, B. Hu, F. Meng, K. Niu, F. Lin, H. Zheng, An
[26]
[27]
[28]
[29]
[30]
[31]
[32]
313
investigation of ultrathin nickel-iron layered double hydroxide nanosheets grown on nickel foam for high-performance supercapacitor electrodes, J. Alloys Compd. 714 (2017) 63e70. H. Fu, Z.-j. Du, W. Zou, H.-Q. Li, C. Zhang, Simple fabrication of strongly coupled cobalt ferrite/carbon nanotube composite based on deoxygenation for improving lithium storage, Carbon 65 (2013) 112e123. D. Du, W. Yue, Y. Ren, X. Yang, Fabrication of graphene-encapsulated CoO/ CoFe2O4 composites derived from layered double hydroxides and their application as anode materials for lithium-ion batteries, J. Mater. Sci. 49 (2014) 8031e8039. D. Du, W. Yue, X. Fan, K. Tang, X. Yang, Ultrathin NiO/NiFe2O4 nanoplates decorated graphene nanosheets with enhanced lithium storage properties, Electrochim. Acta 194 (2016) 17e25. Z. Liang, R. Huo, Y.-X. Yin, F. Zhang, S. Xu, Y.-G. Guo, Carbon-supported Ni@ NiO/Al2O3 integrated nanocomposite derived from layered double hydroxide precursor as cycling-stable anode materials for lithium-ion batteries, Electrochim. Acta 108 (2013) 429e434. J.M. Lee, T.H. Gu, N.H. Kwon, S.M. Oh, S.-J. Hwang, Rapid synthetic route to nanocrystalline carbon-mixed metal oxide nanocomposites with enhanced electrode functionality, J. Phys. Chem. C 120 (2016) 8451e8460. S. Xu, Y. Yang, T. Xu, Y. Kuang, M. Dong, F. Zhang, F. Besenbacher, D.G. Evans, Engineered morphologies of layered double hydroxide nanoarchitectured shell microspheres and their calcined products, Chem. Eng. Sci. 66 (2011) 2157e2163. L. Zou, X. Xiang, J. Fan, F. Li, Single-source precursor to complex metal oxide monoliths with tunable microstructures and properties: the case of Mgcontaining materials, Chem. Mater. 19 (2007) 6518e6527. G. Fan, F. Li, D.G. Evans, X. Duan, Catalytic applications of layered double hydroxides: recent advances and perspectives, Chem. Soc. Rev. 43 (2014) 7040e7066. X. Zhao, F. Zhang, S. Xu, D.G. Evans, X. Duan, From layered double hydroxides to ZnO-based mixed metal oxides by thermal decomposition: transformation mechanism and UV-blocking properties of the product, Chem. Mater. 22 (2010) 3933e3942. M. Li, Y.-X. Yin, C. Li, F. Zhang, L.-J. Wan, S. Xu, D.G. Evans, Well-dispersed bicomponent-active CoO-CoFe2O4 nanocomposites with tunable performances as anode materials for lithium-ion batteries, Chem. Commun. 48 (2012) 410e412. X.-F. Zhao, S.-L. Xu, L.-Y. Wang, X. Duan, F.-Z. Zhang, Exchange-biased NiFe2O4/NiO nanocomposites derived from NiFe-layered double hydroxides as a single precursor, Nano Res. 3 (2010) 200e210. M.A. Woo, T.W. Kim, I.Y. Kim, S.-J. Hwang, Synthesis and lithium electrode application of ZnO-ZnFe2O4 nanocomposites and porously assembled ZnFe2O4 nanoparticles, Solid State Ionics 182 (2011) 91e97. F. Wang, S. Zhang, D. Bai, F. Zhang, S. Xu, Graphene-supported binary active Mn0.25Co0.75O solid solution derived from a CoMn-layered double hydroxide precursor for highly improved lithium storage, RSC Adv. 6 (2016) 19716e19722. R.C. Jin, H. Liu, Y. Guan, J. Zhou, G. Chen, ZnFe2O4/C nanodiscs as high performance anode material for lithium-ion batteries, Mater. Lett. 158 (2015) 218e221. L. Zou, F. Li, X. Xiang, D.G. Evans, X. Duan, Self-generated template pathway to high-surface-area zinc aluminate spinel with mesopore network from a single-source inorganic precursor, Chem. Mater. 18 (2006) 5852e5859. B. Wu, S. Zhang, F. Yao, F. Zhang, S. Xu, Synergistic lithium storage of a multicomponent Co2SnO4/Co3O4/Al2O3/C composite from a singlesource precursor, RSC Adv. 5 (2015) 69932e69938. J.G. Kim, Y. Kim, Y. Noh, W.B. Kim, Formation of carbon-coated ZnFe2O4 nanowires and their highly reversible lithium storage properties, RSC Adv. 4 (2014) 27714e27721. C. Hu, S. Guo, G. Lu, Y. Fu, J. Liu, H. Wei, X. Yan, Y. Wang, Z. Guo, Carbon coating and Zn2þ doping of magnetite nanorods for enhanced electrochemical energy storage, Electrochim. Acta 148 (2014) 118e126. R. Jin, G. Liu, C. Liu, L. Sun, High electrochemical performances of hierarchical hydrangea macrophylla like NiCo2O4 and NiCo2S4 as anode materials for Liion batteries, Mater. Res. Bull. 80 (2016) 309e315. M. Wang, Y. Huang, X. Chen, K. Wang, H. Wu, N. Zhang, H. Fu, Synthesis of nitrogen and sulfur co-doped graphene supported hollow ZnFe2O4 nanosphere composites for application in lithium-ion batteries, J. Alloys Compd. 691 (2017) 407e415. D. Bresser, E. Paillard, R. Kloepsch, S. Krueger, M. Fiedler, R. Schmitz, D. Baither, M. Winter, S. Passerini, Carbon coated ZnFe2O4 nanoparticles for advanced lithium-ion anodes, Adv. Energy Mater. 3 (2013) 513e523. L. Cabo-Fernandez, F. Mueller, S. Passerini, L.J. Hardwick, In situ Raman spectroscopy of carbon-coated ZnFe2O4 anode material in Li-ion batteriesdinvestigation of SEI growth, Chem. Commun. 52 (2016) 3970e3973. F. Martinez-Julian, A. Guerrero, M. Haro, J. Bisquert, D. Bresser, E. Paillard, S. Passerini, G. Garcia-Belmonte, Probing lithiation kinetics of carbon-coated ZnFe2O4nanoparticle battery anodes, J. Phys. Chem. C 118 (2014) 6069e6076. A.K. Rai, T.V. Thi, J. Gim, J. Kim, Combustion synthesis of MgFe2O4/graphene nanocomposite as a high-performance negative electrode for lithium ion batteries, Mater. Char. 95 (2014) 259e265. Y. Pan, Y. Zhang, X. Wei, C. Yuan, J. Yin, D. Cao, G. Wang, MgFe2O4 nanoparticles as anode materials for lithium-ion batteries, Electrochim. Acta 109 (2013) 89e94.
314
D. Wang et al. / Journal of Alloys and Compounds 726 (2017) 306e314
[33] H. Qiao, L. Luo, K. Chen, Y. Fei, R. Cui, Q. Wei, Electrospun synthesis and lithium storage properties of magnesium ferrite nanofibers, Electrochim. Acta 160 (2015) 43e49. [34] C. Wang, Y. Li, Y. Ruan, J. Jiang, Q.-H. Wu, ZnFe2O4-nanocrystal-assembled microcages as an anode material for high performance lithium-ion batteries, Mater. Today Energy 3 (2017) 1e8. [35] Y. Chen, J. Wu, W. Yang, Y. Fu, R. Zhou, S. Chen, L. Zhang, Y. Song, L. Wang, Zn/ Fe-MOFs-derived hierarchical ball-in-ball ZnO/ZnFe2O4@carbon nanospheres with exceptional lithium storage performance, J. Alloys Compd. 688 (2016) 211e218. [36] S. Permien, S. Indris, M. Scheuermann, U. Schuermann, V. Mereacre, A.K. Powell, L. Kienle, W. Bensch, Is there a universal reaction mechanism of Li insertion into oxidic spinels: a case study using MgFe2O4, J. Mater. Chem. A 3 (2015) 1549e1561. pez, G.F. Ortiz, P. Lavela, J.L. Tirado, Towards an all-solid-state battery: [37] M.C. Lo preparation of conversion anodes by electrodepositioneoxidation processes, J. Power Sources 244 (2013) 403e409. pez, G.F. Ortiz, P. Lavela, R. Alca ntara, J.L. Tirado, Improved energy [38] M.C. Lo storage solution based on hybrid oxide materials, ACS Sustain. Chem. Eng. 1 (2013) 46e56. [39] A. Ponrouch, P.-L. Taberna, P. Simon, M.R. Palacín, On the origin of the extra capacity at low potential in materials for Li batteries reacting through conversion reaction, Electrochim. Acta 61 (2012) 13e18. [40] S.J. Rezvani, R. Gunnella, A. Witkowska, F. Mueller, M. Pasqualini, F. Nobili, S. Passerini, A.D. Cicco, Is the solid electrolyte interphase an extra-charge reservoir in Li-ion batteries? ACS Appl. Mater. Interfaces 9 (2017) 4570e4576. [41] J. Lu, Z. Chen, Z. Ma, F. Pan, L.A. Curtiss, K. Amine, The role of nanotechnology in the development of battery materials for electric vehicles, Nat. Nanotechnol. 11 (2016) 1031e1038. [42] P. Poizot, S. Laruelle, S. Grugeon, L. Dupont, J.M. Tarascon, Nano-sized transition-metal oxides as negative-electrode materials for lithium-ion batteries, Nature 407 (2000) 496e499. [43] J. Cai, C. Wu, Y. Zhu, P.K. Shen, K. Zhang, Hierarchical porous acetylene black/ ZnFe2O4@carbon hybrid materials with high capacity and robust cycling performance for Li-ion batteries, Electrochim. Acta 187 (2016) 584e592. [44] D. Narsimulu, B.N. Rao, M. Venkateswarlu, E.S. Srinadhu, N. Satyanarayana, Electrical and electrochemical studies of nanocrystalline mesoporous MgFe2O4 as anode material for lithium battery applications, Ceram. Int. 42 (2016) 16789e16797.
[45] L. Yao, H. Deng, Q.-A. Huang, Q. Su, G. Du, Three-dimensional carbon-coated ZnFe2O4 nanospheres/nitrogen-doped graphene aerogels as anode for lithium-ion batteries, Ceram. Int. 43 (2017) 1022e1028. [46] Y. Yin, N. Huo, W. Liu, Z. Shi, Q. Wang, Y. Ding, J. Zhang, S. Yang, Hollow spheres of MgFe2O4 as anode material for lithium-ion batteries, Scr. Mater. 110 (2016) 92e95. [47] B. Jiang, C. Han, B. Li, Y. He, Z. Lin, In-situ crafting of ZnFe2O4 nanoparticles impregnated within continuous carbon network as advanced anode materials, ACS Nano 10 (2016) 2728e2735. [48] L. Luo, D. Li, J. Zang, C. Chen, J. Zhu, H. Qiao, Y. Cai, K. Lu, X. Zhang, Q. Wei, Carbon-coated Magnesium ferrite nanofibers for lithium-ion battery anodes with enhanced cycling performance, Energy Technol. (2017), http:// dx.doi.org/10.1002/ente.201600686. [49] X. Yang, H. Xue, Q. Yang, R. Yuan, W. Kang, C.-S. Lee, Preparation of porous ZnO/ZnFe2O4 composite from metal organic frameworks and its applications for lithium ion batteries, Chem. Eng. J. 308 (2017) 340e346. [50] L. Liu, R. Gao, L. Sun, S. Han, D. Chen, Z. Hu, X. Liu, Ultrahigh cycling stability and rate capability of ZnFe2O4@graphene hybrid anode prepared through a facile syn-graphenization strategy, New J. Chem. 40 (2016) 3139e3146. [51] Y. Yin, W. Liu, N. Huo, S. Yang, Synthesis of vesicle-like MgFe2O4/graphene 3D network anode material with enhanced lithium storage performance, ACS Sustain. Chem. Eng. 5 (2017) 563e570. [52] J. Liu, J. Xiao, X. Zeng, P. Dong, J. Zhao, Y. Zhang, X. Li, Combustion synthesized macroporous structure MFe2O4 (M ¼Zn, Co) as anode materials with excellent electrochemical performance for lithium ion batteries, J. Alloys Compd. 699 (2017) 401e407. [53] D. Cai, H. Zhan, T. Wang, MOF-derived porous ZnO/ZnFe2O4 hybrid nanostructures as advanced anode materials for lithium ion batteries, Mater. Lett. 197 (2017) 241e244. [54] D. Bresser, S. Passerini, B. Scrosati, Leveraging valuable synergies by combining alloying and conversion for lithium-ion anodes, Energy Environ. Sci. 9 (2016) 3348e3367. [55] Y. Sharma, N. Sharma, G.V.S. Rao, B.V.R. Chowdari, Li-storage and cyclability of urea combustion derived ZnFe2O4 as anode for Li-ion batteries, Electrochim. Acta 53 (2008) 2380e2385. [56] J. Liu, Y. Li, X. Huang, G. Li, Z. Li, Layered double hydroxide nano- and microstructures grown directly on metal substrates and their calcined products for application as Li-ion battery electrodes, Adv. Funct. Mater. 18 (2008) 1448e1458.