ZnO-CoO Nanoparticles Encapsulated in 3D Porous Carbon Microspheres for High-performance Lithium-Ion Battery Anodes

Accepted Manuscript Title: ZnO-CoO Nanoparticles Encapsulated in 3D Porous Carbon Microspheres for High-performance Lithium-Ion Battery Anodes Author: Lianjun Liu Cunyu Zhao Huilei Zhao Qianyi Zhang Ying Li PII: DOI: Reference:

S0013-4686(14)00968-2 http://dx.doi.org/doi:10.1016/j.electacta.2014.05.001 EA 22693

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

5-3-2014 1-5-2014 1-5-2014

Please cite this article as: L. Liu, C. Zhao, H. Zhao, Q. Zhang, Y. Li, ZnO-CoO Nanoparticles Encapsulated in 3D Porous Carbon Microspheres for High-performance Lithium-Ion Battery Anodes, Electrochimica Acta (2014), http://dx.doi.org/10.1016/j.electacta.2014.05.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ZnO-CoO Nanoparticles Encapsulated in 3D Porous Carbon

Lianjun Liu, Cunyu Zhao, Huilei Zhao, Qianyi Zhang, Ying Li *

ip t

Microspheres for High-performance Lithium-Ion Battery Anodes

cr

Mechanical Engineering Department, University of Wisconsin-Milwaukee, 3200 N. Cramer St.

us

Milwaukee, WI, 53211, USA Corresponding Author:

an

Ying Li, Ph.D. Assistant Professor

M

Email: [email protected]

Ac ce p

te

d

Tel.: +1 414-229-3716

1

Page 1 of 25

ABSTRACT: In this paper we report a novel architecture of hierarchical 3D porous carbon microspheres (PCM) to encapsulate ZnO-CoO nanoparticles that serves as an advanced anode for high-performance lithium-ion battery (LIB). The PCM is fabricated by a facile aerosol spray

ip t

pyrolysis method, and ZnO-CoO composite nanoparticles are infiltrated into the PCM by a simple one-pot hydrothermal procedure (i.e., ZnO-CoO@PCM). The developed hybrid material

cr

provides several advantages: (1) partial replacement of CoO with ZnO to offer a low-cost and

us

eco-friendly candidate anode, (2) a continuous and large surface area (1236 m2 g-1) carbon network for improved electrical conductivity and uniform dispersion of ZnO-CoO nanoparticles,

an

and (3) porous structure for good electrolyte diffusion and fast Li-ion transport and to buffer the large volume expansion of the metal oxides. As a result, this new ZnO-CoO@PCM

M

nanocomposite demonstrates a higher reversible capacity (1250 mAh g-1 after 150 cycles at a

d

current density of 100 mA g-1), more excellent cycling stability, and better rate capability than a

te

ZnO-CoO/PCM mixture and than a non-porous ZnO-CoO/carbon black mixture. The 3D porous nanocomposite architecture in this work could shed light on the design and synthesis of other

Ac ce p

metal oxides electrodes for energy storage.

KEYWORDS: ZnO-CoO nanoparticles, porous carbon microsphere, confinement effect, lithium-ion batteries, reversible capacity

2

Page 2 of 25

1. Introduction The increasing demand of rechargeable lithium-ions batteries (LIBs) for portable electronic devices has promoted the development of new electrode materials with high capacity and/or high

ip t

energy density. Nanostructured metal oxides (MxOy, M = Fe, Co, Ni, Mn, Zn, etc) have been

cr

considered as alternative anode materials for next-generation of LIBs because of their high reversible capacities (500 – 1000 mAh g-1), compared with conventional graphite of 372 mAh g[1-4]. Among the studied metal oxides, CoO is a promising candidate due to its relatively high

us

1

an

theoretical capacity (715 mAh g-1) and reversible electrochemical reaction (CoO + 2Li+ + 2e → Co +Li2O) [5-10]. However, the practical application of CoO for LIBs may be limited by its high

M

cost and toxicity. Moreover, CoO suffers from fast capacity decay due to its poor electrical conductivity, limited ions transport kinetics, and large volume expansion (> 200%) and

d

contraction during the charge/discharge process [9, 11-14].

te

Various strategies have been attempted to circumvent the above mentioned problems. First,

Ac ce p

CoO has a rocksalt structure, where Co2+ at the octahedral site could be partially replaced by eco-friendly and cheaper elements such as Mn2+ and Ni2+ without changing its structure and sacrificing the electrochemical performance [13, 15-17]. Second, incorporation of CoO with conductive materials (e.g., carbon) can greatly improve the conductivity [5, 6, 9, 11, 12, 18]. Third, downsizing CoO into nanoscale could reduce the diffusion lengths for Li+ ions and electrons, and making a hierarchically hollow CoO structure could alleviate the volume variation [1, 18, 19]. Following the three outlined strategies, many efforts have been made to either control the CoO morphology (e.g., nanocages, nanoflakes, nanodisks, nanospheres, and nanorods) [7, 8, 12, 20, 21], or directly grow CoO nanowire arrays on a metallic substrate (e.g., Ti foil, Ni foam) [22], or anchor CoO particles on the conductive matrices (e.g., one-dimensional (1D) carbon

3

Page 3 of 25

nanotube, nanofiber, and two-dimensional (2D) graphene) [6, 9, 14, 23-25]. Till now, only few studies have reported the partially substituted CoO electrodes by ZnO, NiO or MnO [13, 15-17]. Among them, ZnO is more attractive due to its high theoretical capacity (987 mAh g-1) and

ip t

electrochemical activity of Zn with respect to lithium [26, 27]. Recently, two studies have been conducted to directly grow ZnO-CoO nanotube or flower-like nanowall arrays on a metallic

cr

copper substrate [13, 16]. Although the morphology of ZnO-CoO could be tailored by adjusting

us

the reaction conditions, its rate capability and cycling performance are still unsatisfactory. The limited studies demand more research in this eco-friendly and promising ZnO-CoO electrode by

an

developing an advanced architecture, in which the electroactivities of each component are

M

manifested and a fast ion and electron diffusion is warranted.

It is noteworthy besides 1D carbon nanotube and 2D graphene, three dimensional (3D)

d

ordered porous carbon framework has been reported as an ideal scaffold for fabrication of

te

monolithic composite electrodes for supercapacitors and LIBs [28, 29]. However, no research

Ac ce p

has yet been done on ZnO-CoO encapsulated in a 3D porous carbon. On the other hand, the above mentioned CoO in various morphologies and CoO/carbon composites were usually tailored by using some organic complex precursors [6], substrates [13, 16, 22], and surfactants [10, 21], but the high cost and toxicity may limit a large scale application. Furthermore, the synthesis routes used for CoO/carbon composites were either conducted in a complicated process or often led to an unsatisfying dispersion and contact of CoO on the carbon support [10, 12, 18, 20]. It is hence highly desirable to develop a facile and general approach for the synthesis of CoO or substituted CoO anchored firmly on carbon. In this paper, we synthesized a new 3D porous carbon microsphere (PCM) from low cost raw materials (i.e., sucrose) via a spray pyrolysis method, and developed a novel architecture of

4

Page 4 of 25

ZnO-CoO nanoparticles decorated in the PCM (i.e., ZnO-CoO@PCM) by a simple one-pot hydrothermal method. The developed ZnO-CoO@PCM hybrid may offer the following merits: (1) the cost and toxicity of CoO could be minimized by partially replacing CoO with ZnO, (2)

ip t

embedding ZnO-CoO in the PCM framework could greatly improve the conductivity, (3) nanosized ZnO-CoO and porous structure of PCM could facilitate electrolyte and ions transport

cr

in the electrode matrix, and (4) large surface area and pore volume of PCM could allow the

us

uniform dispersion of ZnO-CoO nanoparticles, and provide sufficient buffer spaces to relax the volume change during the charge/discharge processes. Hence, a high capacity, good rate

an

performance, and excellent cycling stability of ZnO-CoO@PCM are expected.

M

2. Experimental Section Material preparation

d

Porous Carbon Microsphere (PCM): The PCM was synthesized through a spray pyrolysis

te

method with the aid of silica templates. In brief, 4.2 g sucrose and 20 ml 20 wt.% silica sol (40

Ac ce p

nm, Nissan Chemical) (weight ratio of carbon : silica = 1:2) was first prepared. 0.25 ml 95% sulfuric acid was then added into the evenly mixed solution. The solution was then placed in a round-bottom flask and heated up to 90 ˚C in an oil bath under vigorous stirring. After reflux for 30 h for sucrose carbonization, the solution was then transferred to a Collison nebulizer (BGI Inc.) for spray pyrolysis through a cylindrical quartz tube placed inside a tube furnace. The tube furnace temperature was set at 600 ˚C and compressed air was used as the carrier gas. Powder samples (C/SiO2) were collected on a glass fiber filter and then finally were calcined in N2 at 800 ˚C for 4 hours. The PCM was produced by removing the silica template in a 10% HF solution (Caution: the HF is harmful).

5

Page 5 of 25

Pure ZnO-CoO: ZnO-CoO nanostructure was prepared by a one-pot hydrothermal method. 1 mmol of Zn(NO3)2•6H2O and 2 mmol of Co(NO3)2•6H2O (molar ratio Zn:Co = 1:2) were dissolved into 100 ml H2O. 5 mmol of urea and 2 mmol of NH4F were added into the solution.

ip t

After stirring 30 min, the mixture was transferred into a 200 mL Teflon-line autoclave, sealed and maintained at 120 oC for 5 h. After it was cooled down to room temperature, the precipitate

us

and finally calcined at 400 oC for 2 h in a N2 atmosphere.

cr

was collected and washed with distilled water. Next, the precipitate was dried at 80 oC overnight

ZnO-CoO/carbon composites: ZnO-CoO nanoparticles encapsulated in the PCM, i.e., ZnO-

an

CoO@PCM, was synthesized by a one-pot hydrothermal method. In brief, 75 mg of PCM (the weight percentage of carbon to ZnO-CoO was 30%), Zn(NO3)2•6H2O (1 mmol), Co(NO3)2•6H2O

M

(2 mmol), urea (5 mmol), and NH4F (2 mmol) were dissolved into 100 ml water/ethanol (volume ratio of water/ethanol = 1:1). After stirring 10 min, the solution was transferred into Teflon-lined

te

d

autoclave, sealed and maintained at 120 oC for 5 h. After cooled down to room temperature, the precipitate was collected, washed with distilled water, and dried at 80 oC overnight.

Ac ce p

For comparison, ZnO-CoO simply mixed with PCM (i.e., ZnO-CoO/PCM) or mixed with carbon black (i.e., ZnO-CoO/CB) was prepared by a similar hydrothermal method but without adding ethanol in the aqueous solution described in the previous paragraph. Ethanol played an important role in promoting the infiltration of Zn-Co solution into the pores of PCM. Hence, for the sample of ZnO-CoO@PCM, the ZnO-CoO nanoparticles could be entirely confined within the pores of PCM; whereas, for the sample of ZnO-CoO/PCM, the majority of ZnO-CoO nanoparticles may be aggregated outside the PCM and only partially infiltrated in the pores of the PCM. The different materials morphologies have been verified by the materials

6

Page 6 of 25

characterization in a later section of the paper. All the samples prepared in this work were calcined at 400 oC for 2 h in a N2 atmosphere.

Material characterization

ip t

The carbon content of electrode materials was measured on a thermalgravimetric analyzer (TGA-

cr

DAT-2960 SDT) at a heating rate of 10 °C min−1 from 25 to 800 °C in air. The crystal structures of the electrode materials were identified by X-ray diffraction (XRD, Scintag XDS 2000) using

us

Cu Kα irradiation at 45 kV and a diffracted beam monochromator at 40 mA. The specific surface area and pore volume were analyzed by nitrogen adsorption-desorption at 77 K using the

an

Brunauer-Emmett-Teller (BET) method (Micromeritics, ASAP 2020). The valence state of Zn

M

and Co elements were identified by X-ray photoelectron spectroscopy (XPS), a PHI 5000 Versaprobe system using monochromatic Al KR radiation (1486.6 eV). All binding energies

d

were referenced to the C 1s peak at 284.6 eV. Scanning electron microscopy (SEM) (Hitachi

te

S4800) was used to obtain the surface morphology. The surface dispersion of Zn and Co elements was analyzed by X-ray elemental mapping. The real near-surface contents of Zn and

Ac ce p

Co were measured by Energy dispersive X-ray spectroscopy (EDS). The lattice structure of ZnOCoO@PCM was visualized by phase-contrast high resolution transmission electron microscopy (HRTEM) carried out with 300 keV electrons in a Hitachi H9000NAR instrument with 0.18 nm point and 0.11 nm lattice resolution. Amplitude contrast TEM images were used to obtain the information about the sizes and morphology.

Electrochemical measurement The active material powder was mixed with carbon black and Poly(vinylydene fluoride), PVDF, dissolved in n-methyl pyrolidinone, NMP, 8 wt.%, in a weight ratio 75:10:15. The slurry was mixed to obtain a homogeneous black paste which was then coated on an copper foil. The as-

7

Page 7 of 25

coated copper foils were dried under vacuum at 90 ˚C for 12 h. The working electrode and Li metal foil counter electrode were assembled into coin cells by using Celgard 2400 as the separator and a solution of 1M LiPF6 in ethylene carbonate (EC)/diethyl carbonate (DEC) as the

ip t

electrolyte. The cells were constructed in an Argon-filled glove box. The charge/discharge curves were measured using an Arbin potentiostat at a cutoff voltage of 0.01-3.0 V under various

cr

current densities of 100 mA g-1, 200 mA g-1, 400 mA g-1, 800 mA g-1 and 1000 mA g-1. The

us

specific capacity and current density were calculated based on ZnO-CoO mass only.

an

3. Results and discussion 3.1. Crystal structure and valence state

M

The crystal structures of ZnO-CoO and ZnO-CoO/carbon composites were identified by XRD, as shown in Figure 1. All the ZnO-CoO-based materials displayed similar diffraction peaks, which

d

were indexed to a mixture of ZnO (JCPDS: 89-1397) and CoO (JCPDS: 78-0431) crystals. No

te

new ZnCo2O4 phase or peak position shift was observed, but ZnO-CoO showed sharper peaks

Ac ce p

than ZnO-CoO@PCM, suggesting that ZnO-CoO had a larger crystallite size (ZnO: 26.3 nm, CoO: 20.6 nm) than ZnO-CoO@PCM (ZnO: 16.1 nm, CoO: 13.8 nm). This difference was probably caused by the confinement effect of PCM that restricted the growth of ZnO-CoO crystals, as evidenced by SEM and TEM presented in a later section in the paper.

8

Page 8 of 25

  

(d) ZnO-CoO/PCM    



(c) ZnO-CoO@PCM

cr

(b) ZnO-CoO/CB

50

2 (degree)

60

70

an

40

us

(a) ZnO-CoO

30

ip t



 ZnO  CoO

Figure 1. XRD patterns for pure ZnO-CoO, ZnO-CoO mixed with CB (i.e., ZnO-CoO/CB),

M

ZnO-CoO mixed with PCM (i.e., ZnO-CoO/PCM), and ZnO-CoO encapsulated in PCM (i.e.,

d

ZnO-CoO@PCM)

te

The XPS results further confirmed the existence of ZnO and CoO on the PCM surface. Figure 2 shows the XPS spectra of Co 2p and Zn 2p on ZnO-CoO@PCM. Co 2P spectrum (Figure 2a)

Ac ce p

displays the Co 2p3/2 and Co 2p1/2 peaks at 780.5 and 796.5 eV, respectively with two shake-up satellite peaks at 785.6 and 802.5 eV, corresponding to the typical Co2+ of CoO phase [13, 16]. The Zn 2p3/2 (1021.6 eV) and Zn 2p1/2 (1044.6 eV) peaks are assigned to ZnO [16] (Figure 2b). The final amount of carbon in ZnO-CoO/PCM (27 wt%), ZnO-CoO@PCM (28 wt%) and ZnOCoO/CB (31 wt%), obtained by thermo-gravimetric analysis (TGA), is comparable, very close to the nominal percentage (30 wt%) calculated from precursor concentrations. For the relative content of CoO to ZnO in the nanocomposites, the EDS analyses showed that the real nearsurface atomic ratio of Co/Zn was 1.6, 1.1, and 1.0 on ZnO-CoO/CB, ZnO-CoO/PCM, and ZnOCoO@PCM, respectively, lower than the nominal Co/Zn ratio of 2 calculated from the precursor

9

Page 9 of 25

concentration. The lower real ratio of Co/Zn was probably because Zn(OH)2 has a smaller solubility product constant (Ksp = 3.5 × 10-17) than Co(OH)2 (Ksp = 1.3 × 10-15) that led to an easier formation of Zn(OH)2 during the hydrothermal process [30, 31]. Upon dehydoroxylaion

ip t

and crystallization of the Zn(OH)2-Co(OH)2 mixture, ZnO may partially cover the surface of CoO, as indicated by XRD results in Figure 1 where ZnO indeed exhibited stronger diffraction

(a)

cr

peaks that overlap with those of CoO.

us

(b)

Zn 2p3/2

Co 2p3/2

780.5 eV

1021.6 eV

ZnO

CoO Intensity (a.u.)

Intensity (a.u.)

Co 2p1/2

an

796.5 eV

Zn 2p1/2

765

770 775 780 785

790 795 800

805

810

815

1015

1020

1025

1030

1035

1040

1045

1050

1055

Binding energy (eV)

te

Binding energy (eV)

d

M

1044.6 eV

Figure 2. XPS spectra of (a) Co 2p and (b) Zn 2p for ZnO-CoO encapsulated in PCM (i.e., ZnO-

Ac ce p

CoO@PCM)

3.2. Morphology and surface ZnO-CoO dispersion The encapsulated ZnO-CoO@PCM was characterized by SEM (Figure 3), X-ray elemental mapping (Figure 4), and TEM/HRTEM (Figure 5) to investigate the morphology and surface ZnO-CoO dispersion. SEM was also conducted on PCM, ZnO-CoO, ZnO-CoO/CB and ZnOCoO/PCM for comparison. As shown in Figure 3a, pure PCM showed a porous 3D hierarchical structure composed of many mesopores. The pores themselves were continuously connected by carbon walls to form one microsphere framework. ZnO-CoO, as shown in Figure 3b, displayed a mixed-shape of nanobelts/nanowires with a length longer than 5 μm. For ZnO-CoO@PCM 10

Page 10 of 25

(Figure 3c), no ZnO-CoO nanobelts/nanowires were observed; rather, ZnO-CoO likely remained in the porous structure of the microspheres. The different morphology of ZnO-CoO was probably due to the hydrophobicity and the confinement effect of the PCM. The contact angle

ip t

measurement showed that the PCM had a contact angle of 110o (θ), indicating its hydrophobic surface. In a mixture of water/ethanol (H2O/EtOH) medium, the PCM was homogeneously

cr

dispersed because of its affinity to EtOH, and the Zn2+/Co2+ solutions were easily infiltrated into

us

the pores. The pores would confine the growth of ZnCo nanobelts/nanowires during the hydrothermal crystallization process. Hence, ZnO-CoO nanoparticles were formed and

an

encapsulated in the mesopores. The effect of EtOH on the ZnO-CoO morphology in the hydrophobic PCM was further investigated by preparing ZnO-CoO/PCM in an aqueous solution

M

(without EtOH). This control experiment showed that wire/flake-like ZnO-CoO was formed and aggregated outside the PCM, as shown in Figure 3d, again demonstrating EtOH indeed promotes

te

d

the infiltration of Zn2+/Co2+-containing solution into the PCM. For ZnO-CoO/CB, as shown in Figure S1, it clearly shows that ZnO-CoO/CB is a mixture of carbon nanoparticles with a size of

Ac ce p

~70 nm and ZnO-CoO nanowires with a length greater than 2 µm.

11

Page 11 of 25

(b)

(c)

(d)

M

an

us

cr

ip t

(a)

d

Figure 3. SEM images of (a) PCM, (b) ZnO-CoO, (c) ZnO-CoO encapsulated in PCM (i.e.,

te

ZnO-CoO@PCM), and (d) ZnO-CoO mixed with PCM (i.e., ZnO-CoO/PCM)

Ac ce p

To further identify the location and distribution of ZnO-CoO particles over ZnO-CoO@PCM, x-ray elemental mapping was conducted for a selected area of the PCM at a high magnification. As shown in Figure 4, Zn (red color) and Co (green color) elements were evenly distributed over the selected area of the mesopores. Because the surface pore structure of the PCM did not change after ZnO-CoO incorporation, we can infer that ZnO-CoO species were indeed infiltrated inside the porous structure within the PCM.

12

Page 12 of 25

(a)

ip t

(b)

(d)

Co

Zn

M

an

us

(c)

cr

Co, Zn

Figure 4. (a) SEM image of ZnO-CoO encapsulated in PCM (i.e., ZnO-CoO@PCM), and X-ray

d

elemental mapping images showing the dispersion of (b) Zn and Co elements, (c) Co element,

te

(d) Zn element for the selected area (yellow rectangle) in (a).

Ac ce p

TEM and HRTEM (Figure 5) were conducted on ZnO-CoO@PCM to investigate the nanostructure of the PCM and the size and dispersion of ZnO-CoO particles within the PCM. The TEM images at a low magnification in Figure 5a-b showed that the PCM was composed of aggregated small hollow carbon rings with a diameter around 40 nm and a thickness around 2 nm. The TEM image at a high magnification in Figure 5c shows that a few dark spots (about 20 nm) were attached on the wall of the hollow carbon rings. Due to the larger atom number and heavier atom mass of Zn and Co than C, the dark spots should be associated with ZnO-CoO particles. The HRTEM image in Figure 5d further confirmed the accommodation of ZnO-CoO with a

13

Page 13 of 25

lattice spacing of 0.32 nm in the pore [16, 26, 27]. The carbon was amorphous, since no lattice fringes of graphite carbon were observed.

ip t

(b)

(a)

an

us

cr

(b)

(c)

(d)

0.32 nm

Ac ce p

te

d

M

ZnCo

Figure 5. (a) TEM image of a single microsphere of ZnO-CoO@PCM, (b) TEM image of the yellow rectangle area in (a), (c) TEM image showing ZnO-CoO particles encapsulated in the PCM, and (d) HRTEM image showing the lattice fringe of ZnO-CoO in the PCM. The porous structure of PCM before and after loading ZnO-CoO was also investigated by N2 adsorption/desorption isotherms, as shown in Figure 6. Bare PCM displayed a typical IV-type isotherm, corresponding to a mesoporous structure. The sharp incremental of N2 adsorption amount at the relative high pressure p/p0 = 0.93 – 1.0 indicated the existence of macroporosity [25, 32]. In addition, the T-plot fitting dada suggested the presence of micropores in the PCM (not shown). Clearly, the PCM has a multi-modal pore size distribution (e.g., macropore,

14

Page 14 of 25

mesopore and micropore) with mesopores being dominant. ZnO-CoO@PCM also displayed an IV-type isotherm for mesopores, but the quantity of adsorbed N2 remarkably decreased compared with that of the PCM. The incorporation of ZnO-CoO species with the PCM also

ip t

resulted in a decrease of the BET surface area from 1236 m2/g to 394 m2/g and pore volume from 3.1 cm3/g to 1.1 cm3/g, but its surface area was still more than 5 times larger than that of the

cr

ZnO-CoO/CB (44 m2/g, 0.16 cm3/g). Furthermore, the incorporation of ZnO-CoO particles

us

resulted in an expansion in the mesopore size of PCM from 13.9 nm to 16.1 nm (see the inset picture in Figure 6). These analyses again supported the SEM, X-ray element mapping and

an

TEM/HRTEM results that the pores of PCM were partially occupied by ZnO-CoO particles

M

without changing its mesoporous structure and spherical morphology. 0.035

PCM ZnO-CoO@PCM

0.030

0.025

dV/dD

d

1600

0.020

te

3

-1

Adsorbed quality (cm g STP)

2000

Ac ce p

1200

0.015

0.010

0.005

(a)

800

0

10

0.000 20

30

40

r (nm)

400

(a) PCM (b) ZnO-CoO@PCM (c) ZnO-CoO/CB

(b)

(c)

0

0.2

0.4

0.6

0.8

1.0

Relative pressure (P/PO)

Figure 6. N2 adsorption-desorption isotherms for PCM, ZnO-CoO particles encapsulated in PCM (i.e., ZnO-CoO@PCM), and ZnO-CoO mixed with CB (i.e., ZnO-CoO/CB). The inset is the pore size distribution of PCM and ZnO-CoO@PCM

15

Page 15 of 25

3.3. Electrochemical performance The electrochemical properties of the as-synthesized ZnCo/carbon electrodes were measured by configuring them as laboratory-scale coin cells. Figure 7a compares the voltage-capacity profiles

ip t

of PCM, ZnO-CoO and ZnO-CoO/carbon electrodes for the first charge/discharge cycle. Two

cr

discharge plateaus appeared on all four ZnCo-based samples. Among them, ZnO-CoO@PCM showed the longest discharge plateaus. The first plateau at 1.25 V may be ascribed to the reaction

us

of ZnO-CoO mixture with lithium [13, 26]. The second between 0.75 V and 0.5 V is associated with the conversion of oxides to Co and Zn, the insertion of lithium into PCM and the formation

an

of a solid electrolyte interphase (SEI) [6, 15, 16]. Surprisingly, an impressively high initial

M

discharge capacity of 2385 mAh g-1 was achieved on ZnO-CoO@PCM, more than twice as high as that of ZnO-CoO (922 mAh g-1) and ZnO-CoO/CB (975 mAh g-1), and even 50% higher than

d

the PCM (1641 mAh g-1) and ZnO-CoO/PCM (1604 mAh g-1). The initial capacity of ZnO-

te

CoO@PCM was also nearly twice as high as the sum of ZnO-CoO and PCM (72%×922 + 28%×1641 = 1123 mAh g-1), demonstrating a synergy between the PCM and the incorporated

Ac ce p

ZnO-CoO nanoparticles. To the best of our knowledge, this 2385 mAh g-1 initial discharge capacity is the highest reported in the literature for CoO-based LIB anodes. However, a relatively low charge capacity of 990 mAh g-1 is achieved, corresponding to a low Coulombic efficiency of 42%. According to the literature reports [6, 12, 15, 18], this large irreversible capacity loss during the first discharge/charge process is probably due to the fact that the irreversible formation of the SEI layer induced the lithium loss.

16

Page 16 of 25

(a) 3.0

(b) -1

1.0 (4) (5)

0.5

(3) ZnO-CoO/CB (4) ZnO-CoO/PCM

1500

(5) ZnO-CoO@PCM (5)

1000

(4) (3)

500

(2)

0

500

1000

1500

2000

0

2500

0

-1

Capacity (mAh g )

30

60

d

90

te

85

80 0

30

60

90

(2) PCM (3) ZnO-CoO/CB (4) ZnO-CoO/PCM

150

100 mA/g

(5) ZnO-CoO@PCM

(5)

M

ZnO-CoO/CB ZnO-CoO/PCM ZnO-CoO@PCM

120

(1) ZnO-CoO

100 mA/g 200 mA/g

-1

Capacity (mA h g )

1200

95

90

Cycle number

an

(d)1500

Ac ce p

Coulombic efficiency (%)

(c) 100

us

(1)

0.0

ip t

1.5

(1) (2) (3)

(1) ZnO-CoO (2) PCM

cr

2.0

Discharge Capacity (mAh g )

2.5

Potential (V)

2000

(1) ZnO-CoO (2) PCM (3) ZnO-CoO/CB (4) ZnO-CoO/PCM (5) ZnO-CoO@PCM

400 mA/g

900

(4) 800 mA/g 1000 mA/g

600

(3) (2)

300

(1) 0

120

Cycle number

150

0

20

40

60

80

100

Cycle number

Figure 7. Electrochemical characterization and battery performance results of ZnO-CoO, PCM, ZnO-CoO mixed with CB (i.e., ZnO-CoO/CB), ZnO-CoO mixed with PCM (i.e., ZnOCoO/PCM), and ZnO-CoO encapsulated in PCM (i.e., ZnO-CoO@PCM) samples: (a) the first cycle charge-discharge profiles between 0.01 V and 3.0 V at a current density of 100 mA g-1, (b) cyclic performance over 150 cycles at a current density of 100 mA g-1, (c) the Coulombic efficiency over 150 cycles at 100 mA g-1, and (d) comparison of the rate capability at a current density from 100 mA g-1 to 1000 mA g-1.

17

Page 17 of 25

The cycling performance and Coulombic efficiency of PCM, ZnO-CoO, ZnO-CoO/CB, ZnOCoO/PCM and ZnO-CoO@PCM samples over 150 cycles at 100 mA g-1 have been conducted for further comparison, as shown in Figure 7b. The discharge capacity of ZnO-CoO fast faded to

ip t

100 mAh g-1 after the 25th cycle, indicating the poor cycling stability. The PCM demonstrated a higher and more stable discharge capacity (492 mAh g-1 after 100 cycles) than ZnO-CoO, and

cr

even better than the commercial graphite (372 mAh g-1) and the literature reported carbon

us

nanofibers (CNFs, 177 mAh g-1 after 100 cycles at 140 mA g-1) [25], CMK-3 (400 mAh g-1 after 20 cycles at 100 mA g-1) [10], and graphene (420 mAh g-1 at 100 mA g-1 after 60 cycles) [23].

an

The ZnO-CoO/CB electrode showed a slightly improved discharge capacity and cycling stability. Compared to ZnO-CoO and ZnO-CoO/CB, the ZnO-CoO/PCM displayed a higher and more

M

stable capacity (850 mAh g-1) till 100 cycles, which still remained in 787 mAh g-1 over 150

d

cycles (91% retention). Two possible reasons are accounted for the higher capacity of ZnO-

te

CoO/PCM than ZnO-CoO/CB. First, compared with the 0D CB nanoparticles, the hierarchically 3D-structured PCM itself may have a higher capacity because of its continuous carbon walls,

Ac ce p

larger surface area and larger pore volume. Second, the ZnO-CoO may be partially encapsulated inside the PCM and the good electrical contact between ZnO-CoO and the carbon walls of the PCM has led to a higher capacity. The most important finding in this paper is that ZnOCoO@PCM not only has an over ten times higher and more excellent cycling performance than ZnO-CoO, but also shows a much higher long-term capacity (150 cycles) than PCM, ZnOCoO/CB, and ZnO-CoO/PCM. Again, this result demonstrated the existence of the synergistic effect between the confined ZnO-CoO nanoparticles and PCM. More interestingly, as shown in Figure 7b, the reversible capacity of ZnO-CoO@PCM shows an upward trend with the cycles, keeps a high capacity above 1000 mAh g-1 over 100 cycles, and

18

Page 18 of 25

finally reaches 1250 mAh g-1 after the 150th cycle. The increasing capacity was probably caused by three reasons. First, ZnO-CoO nanoparticles decorated in the pores of PCM might not be fully activated during the initial charge/discharge process. As the cycling proceeded, the residual ZnO-

ip t

CoO nanoparticles may further participate in an electrochemical reaction with Li+ ions. Second, the extra capacity is possibly due to the large surface area of the PCM that contains more active

cr

sites (e.g., disorders/defects) to enhance lithium storage and induce the reversible

us

formation/dissolution of polymeric gel-like film from the decomposition of electrolyte [6, 23, 33]. Third, the formation of metallic Co from the reduction of partial CoO could enhance the

an

conductivity and have a catalytic effect on the decomposition of electrolyte [34, 35]. Similar phenomenon in raising capacity was also observed on the other Fe3O4/CNTs, CoO/graphene,

M

CoO/CNFs, and MnO/graphene composites reported in the literature [6, 25, 33, 34, 36]. This electrochemical performance of ZnO-CoO@PCM (1250 mAh g-1 after 150 cycles) is also much

te

d

better than the ZnO-CoO nanorods arrays (900 mAh g-1 after 50 cycles) [16], MnO-CoO solid solutions (450 mAh g-1 after 20 cycles) [17], CoO-particles/CNFs (633 mAh g-1 after 50 cycles)

Ac ce p

[14], CoO-particles/graphene (740 mAh g-1 after 50 cycles) [37], mesoporous CoO/CMK-3 (1000 mAh g-1 after 15 cycles) [10], CoO/C polyhedra (510 mAh g-1 after 50 cycles) [20], CoONiO flowers/C-particles (650 mAh g-1 after 60 cycles) [15] and ZnO-CoO flowers/C-particles (500 mAh g-1 after 50 cycles) [13] at a same current density of 100 mA g-1 reported in the open literatures.

The superiority of ZnO-CoO@PCM was possibly due to the combined benefits of (1) the unique 3D hierarchical PCM with a large surface area (1236 m2 g-1), large pore volume (3.1 cm3 g-1) and continuous carbon walls, and (2) the uniformly dispersed ZnO-CoO nanoparticles that were confined in the mesopores. In comparison with the literature results, a 3D CoO/C polyhedra

19

Page 19 of 25

assembled by the aggregation of CoO particles have a small surface area (47.8 m2 g-1) and pore volume (0.21 cm3 g-1) [20]; the excess CoO particles at a high loading in a 2D CMK-3 (1168 m2 g-1, pore volume 1.1 cm3 g-1) could aggregate and destroy the mesopore structure of CMK-3 [10].

ip t

In Figure 7c, one can easily find that after the 15th cycle, the encapsulated ZnO-CoO@PCM show a high and stable Coulombic efficiency thereafter over 97%, higher than the literature

cr

reported hierarchical-like CoO@C microsheets (90%) [18], CoO-QDs/graphene (95%) [6] and

us

CoO nanowalls/graphene (93%) [38].

Figure 7d shows the rate capability of the ZnCo/carbon composites at various current

an

densities from 100 mA g-1 to 1000 mA g-1. Obviously, the order of rate capability was in the order of ZnO-CoO@PCM > ZnO-CoO/PCM > ZnO-CoO/CB > PCM > ZnO-CoO. ZnO-

M

CoO@PCM showed an excellent rate performance, with specific capacities of 1023, 1003, 840, 648, and 566 mAh g-1 for 20 cycles at 100 mA g-1, 200 mA g-1, 400 mA g-1, 800 mA g-1 and

te

d

1000 mA g-1, respectively. Moreover, when the rate was changed from 1000 mA g-1 back to 100 mA g-1, ZnO-CoO@PCM released a higher reversible capacity (1265 mAh g-1) after the 100th

Ac ce p

cycle than the initial one. This capacity of 1265 mAh g-1 is also comparable to that in the cycling performance at the same 100 mA g-1 (Figure 7b), even though various high current densities (i.e., 200-1000 mA g-1) have been conducted for 80 cycles prior to the last process (Figure 7d), again demonstrating the extraordinarily high long-term cycling ability of ZnO-CoO@PCM. There are four possible reasons that ZnO-CoO nanoparticles encapsulated in the 3D hierarchical PCM (i.e., ZnO-CoO@PCM) demonstrates a superior rate capability and cycling performance over the ZnO-CoO/PCM mixture, non-porous ZnO-CoO/CB and pure ZnO-CoO. First, ZnO-CoO nanoparticles that were uniformly anchored in the PCM, as evidenced by the SEM, X-ray mapping and HRTEM results, could avoid serious aggregation and reduce the

20

Page 20 of 25

diffusion length of Li+-ions, which may enhance the reversible capacity of the PCM. Second, the 3D hierarchical PCM itself may make additional contributions because it demonstrated a high capacity, good cycling performance and rate capability (see Figure 7). Third, the large surface

ip t

area and the continuous framework of PCM could enhance the electrochemical reaction area, warrant a high dispersion of ZnO-CoO nanoparticles and provide each ZnO-CoO particle with

cr

the required conductivity, therefore ensuring a high reversibility. Fourth, the large pore volume

us

and the large mesopores/macropores opened in the PCM provide valid space for electrolyte filtration/diffusion, easy access for Li+-ions and electrons transport, and in the meantime,

an

accommodate ZnO-CoO nanoparticles volume expansion during the charge/discharge process, thus leading to an excellent rate capability. The integrated four factors may induce a strong

M

synergy between the PCM and ZnO-CoO nanoparticles that accounts for the superior

te

4. Conclusion

d

electrochemical performance of ZnO-CoO@PCM.

Ac ce p

In this work, we have developed a facile strategy for assembling a novel architecture of binary ZnO-CoO nanoparticles uniformly encapsulated in a 3D porous carbon microsphere (i.e., ZnOCoO@PCM). The strategy involves the fabrication of PCM by a simple aerosol spray pyrolysis method and subsequent incorporation of ZnO-CoO into the PCM by a one-pot hydrothermal crystallization. This 3D hierarchical hybrid exhibits a high reversible capacity (1250 mAh g-1 in the 150th cycle), high Coulombic efficiency (99%), excellent cycling stability (150 cycles), and good rate capability for Li-ion battery anode. The superior electrochemical performance of ZnOCoO@PCM was due to the following integrated advantages: (1) ZnO-CoO nanoparticles are uniformly dispersed and confined in the pores that avoid the strong aggregation and large volume expansion during the repeated Li+ insertion-extraction process, and (2) large surface area and

21

Page 21 of 25

porous carbon framework not only greatly improves the conductivity of ZnCo oxides, enlarges the electrochemical reaction area, but also provides valid space for the electrolyte penetration and ions transport. Our low-cost, eco-friendly and multifunctional ZnO-CoO@PCM materials

ip t

may hold a significant promise for the construction of advanced electrodes for high performance

cr

energy storage devices.

us

ACKNOWLEDGMENT

an

The authors thank Dr. Fei Gao in Nanjing University, China, for the XPS analysis.

REFERENCES

J. Liu, D. Xue, Nanoscale Research Letters 5 (2010) 1525-1534.

[2]

J. Jiang, Y. Li, J. Liu, X. Huang, C. Yuan, X.W.D. Lou, Adv. Mater. 24 (2012) 5166–

d

5180.

M

[1]

M.V. Reddy, G.V.S. Rao, B.V.R. Chowdari, Chemical Reviews 113 (2013) 5364-5457.

[4]

X. Su, Q. Wu, X. Zhan, J. Wu, S. Wei, Z. Guo, Journal of Materials Science 47 (2012)

Ac ce p

2519-2534.

te

[3]

[5]

W. Yao, J. Chen, H. Cheng, Journal of Solid State Electrochemistry 15 (2011) 183-188.

[6]

C. Peng, B. Chen, Y. Qin, S. Yang, C. Li, Y. Zuo, S. Liu, J. Yang, ACS Nano 6 (2012) 1074-1081.

[7]

H. Guan, X. Wang, H. Li, C. Zhi, T. Zhai, Y. Bando, D. Golberg, Chemical Communications 48 (2012) 4878-4880.

[8]

Y. Sun, X. Hu, W. Luo, Y. Huang, Journal of Materials Chemistry 22 (2012) 1382613831.

[9]

Y. Sun, X. Hu, W. Luo, Y. Huang, Journal of Physical Chemistry C 116 (2012) 2079420799.

[10]

B. Sun, H. Liu, P. Munroe, H. Ahn, G. Wang, Nano Research 5 (2012) 460-469.

[11]

H. Qiao, L. Xiao, Z. Zheng, H. Liu, F. Jia, L. Zhang, Journal of Power Sources 185 (2008) 486-491.

22

Page 22 of 25

[12]

F.D. Wu, Y. Wang, Journal of Materials Chemistry 21 (2011) 6636-6641.

[13]

Z. Wu, L. Qin, Q. Pan, Journal of Alloys and Compounds 509 (2011) 9207-9213.

[14]

M. Zhang, E. Uchaker, S. Hu, Q. Zhang, T. Wang, G. Cao, J. Li, Nanoscale 5 (2013) 12342-12349. Y.F. Wang, L.J. Zhang, Journal of Power Sources 209 (2012) 20-29.

[16]

Y. Feng, R. Zou, D. Xia, L. Liu, X. Wang, Journal of Materials Chemistry A 1 (2013)

ip t

[15]

[17]

cr

9654-9658.

T. Kokubu, Y. Oaki, E. Hosono, H. Zhou, H. Imai, Advanced Functional Materials 21

us

(2011) 3673-3680. [18]

J. Liu, Y. Zhou, C. Liu, J. Wang, Y. Pan, D. Xue, CrystEngComm 14 (2012) 2669-2674.

[19]

S. Xiong, J.S. Chen, X.W. Lou, H.C. Zeng, Advanced Functional Materials 22 (2012)

[20]

an

861-871.

W. Yuan, J. Zhang, D. Xie, Z. Dong, Q. Su, G. Du, Electrochimica Acta 108 (2013) 506-

[21]

M

511.

Z. Wen, F. Zheng, Z. Jiang, M. Li, Y. Luo, Journal of Materials Science 48 (2013) 342-

[22]

d

347.

J. Jiang, J. Liu, R. Ding, X. Ji, Y. Hu, X. Li, A. Hu, F. Wu, Z. Zhu, X. Huang, Journal of

te

Physical Chemistry C 114 (2010) 929-932. Y. Qi, H. Zhang, N. Du, D. Yang, Journal of Materials Chemistry A 1 (2013) 2337-2342.

[24]

J.-C. Kim, I.-S. Hwang, S.-D. Seo, D.-W. Kim, Materials Letters 104 (2013) 13-16.

[25]

W.H. Ryu, J. Shin, J. Jung, I. Kim, Journal of Materials Chemistry A 1 (2013) 3239-

Ac ce p

[23]

3243. [26]

W. Luo, X. Hu, Y. Sun, Y. Huang, Journal of Materials Chemistry 22 (2012) 8916-8921.

[27]

L. Hu, B. Qu, C. Li, Y. Chen, L. Mei, D. Lei, L. Chen, Q. Li, T. Wang, Journal of Materials Chemistry A 1 (2013) 5596-5602.

[28]

H. Jiang, P.S. Lee, C. Li, Energy & Environmental Science 6 (2013) 41-53.

[29]

J. Lee, J. Kim, T. Hyeon, Advanced Materials 18 (2006) 2073-2094.

[30]

A. Moezzi, M. Cortie, A.M. McDonagh, European Journal of Inorganic Chemistry (2013) 1326–1335.

[31]

M. Iwunze, Equilibrium Concept in Analytic Chemistry, Authorhouse, Bloomington, 2009.

23

Page 23 of 25

[32]

C. Zhao, L. Liu, H. Zhao, A. Krall, Z. Wen, J.C.P. Hurley, J. Jiang, Y. Li, Nanoscale 6 (2014) 882-888.

[33]

P. Wu, N. Du, H. Zhang, J. Yu, D. Yang, Journal of Physical Chemistry C 115 (2011) 3612–3620. L. Zhang, P. Hu, X. Zhao, R. Tian, R. Zou, D. Xia, Journal of Materials Chemistry 21

ip t

[34]

(2011) 18279–18283.

Y. Sun, C. Du, X. Feng, Y. Yu, I. Lieberwirth, C. Chen, Applied Surface Science 259

cr

[35]

(2012) 769–773.

Y. Sun, X. Hu, W. Luo, F. Xia, Y. Huang, Advanced Functional Materials 23 (2013)

us

[36]

2436–2444.

M. Zhang, M. Jia, Y. Jin, X. Shi, Applied Surface Science 263 (2012) 573-578.

[38]

J. Zhu, Y.K. Sharma, Z. Zeng, X. Zhang, M. Srinivasan, S. Mhaisalkar, H. Zhang, H.H.

an

[37]

Ac ce p

te

d

M

Hng, Q. Yan, Journal of Physical Chemistry C 115 (2011) 8400-8406.

24

Page 24 of 25

Ac

ce

pt

ed

M

an

us

cr

i

Graphical Abstract (for review)

Page 25 of 25