Porous carbon nanotubes decorated with nanosized cobalt ferrite as anode materials for high-performance lithium-ion batteries

Porous carbon nanotubes decorated with nanosized cobalt ferrite as anode materials for high-performance lithium-ion batteries

Accepted Manuscript Porous Carbon Nanotubes Decorated with Nanosized Cobalt Ferrite as Anode Materials for High-Performance Lithium-Ion Batteries Ling...

2MB Sizes 0 Downloads 10 Views

Accepted Manuscript Porous Carbon Nanotubes Decorated with Nanosized Cobalt Ferrite as Anode Materials for High-Performance Lithium-Ion Batteries Lingyan Wang, Linhai Zhuo, Haiyang Cheng, Chao Zhang, Fengyu Zhao PII:

S0378-7753(15)00384-5

DOI:

10.1016/j.jpowsour.2015.02.138

Reference:

POWER 20767

To appear in:

Journal of Power Sources

Received Date: 28 January 2015 Accepted Date: 25 February 2015

Please cite this article as: L. Wang, L. Zhuo, H. Cheng, C. Zhang, F. Zhao, Porous Carbon Nanotubes Decorated with Nanosized Cobalt Ferrite as Anode Materials for High-Performance Lithium-Ion Batteries, Journal of Power Sources (2015), doi: 10.1016/j.jpowsour.2015.02.138. 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.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Graphical Abstract

ACCEPTED MANUSCRIPT

Porous Carbon Nanotubes Decorated with Nanosized Cobalt

2

Ferrite as Anode Materials for High-Performance

3

Lithium-Ion Batteries

4

Lingyan Wanga,b,c, Linhai Zhuod, Haiyang Chenga, b, Chao Zhanga,b, Fengyu Zhaoa,b,*

5

a

6

Chemistry and Process, Changchun Institute of Applied Chemistry, Chinese Academy

7

of Sciences, Changchun 130022, China

8

c

9

d

RI PT

1

M AN U

SC

State Key Laboratory of Electroanalytical Chemistry, and bLaboratory of Green

University of the Chinese Academy of Sciences, Beijing 100049, China College of Chemistry and Chemical Engineering, Taishan University, Taian 271021,

China

11

* Corresponding author. Tel./fax: +86 431 85262410.

12

E-mail address: [email protected] (F. Zhao).

13

ABSTRACT: Generally, the fast ion/electron transport and structural stability

14

dominate the superiority in lithium-storage applications. In this work, porous carbon

15

nanotubes decorated with nanosized CoFe2O4 particles ([email protected]) have been

17 18

EP

AC C

16

TE D

10

rationally designed and synthesized by the assistance of supercritical carbon dioxide (scCO2). When tested as anode materials for lithium-ion batteries, the [email protected] composite exhibits outstanding electrochemical behavior with high lithium-storage

19

capacity (1077 mAh g-1 after 100 cycles) and rate capability (694 mAh g-1 at 3 A g-1).

20

These outstanding electrochemical performances are attributed to the synergistic

1

ACCEPTED MANUSCRIPT

effect of porous p-CNTs and nanosized CFO. Compared to pristine CNTs, the p-CNTs

2

with substantial pores in the tubes possess largely increased specific surface area and

3

rich oxygen-containing functional groups. The porous structure can not only

4

accommodate the volume change during lithiation/delithiation processes, but also

5

provide bicontinuous electron/ion pathways and large electrode/electrolyte interface,

6

which facilitate the ion diffusion kinetics, improving the rate performance. Moreover,

7

the CFO particles are bonded strongly to the p-CNTs through metal-oxygen bridges,

8

which facilitate the electron fast capture from p-CNTs to CFO, and thus resulting in a

9

high reversible capacity and excellent rate performance. Overall, the porous p-CNTs

10

provide an efficient way for ion diffusion and continuous electron transport as anode

11

materials.

12

Keywords: porous carbon nanotubes; oxygen bridges; ternary metal oxides;

13

synergistic effect; lithium-ion battery

14

1. Introduction

16 17 18

SC

M AN U

TE D

EP

High demand for electric vehicles (EV), plug-in hybrid electric vehicles (PHEV)

AC C

15

RI PT

1

and large energy-storage devices has enforced extensive research efforts in advanced rechargeable lithium-ion batteries (LIBs) with high energy and power density as well as good cycle stability. It is well known that the rate performance of an electrode is

19

determined by the rate of electron and ion transport, while the cycling stability will be

20

determined by the durability of such transport networks [1-3]. However, the

21

conventional graphitized carbons generally have slow Li+ ion diffusion and large 2

ACCEPTED MANUSCRIPT

resistance at the electrode/electrolyte interface at high rates due to the low porosity,

2

which leads to the kinetic problem of LIBs [4]. By contrast, non-graphitized carbon

3

materials, such as hard carbon spherules, three-dimensional (3D) ordered

4

macroporous and mesoporous carbons with higher porosity showed faster Li-ion

5

transfer than that in graphite due to higher porosity when studied as anode materials

6

for LIBs [5-9]. Very recently, extensive reports have also focused on designing and

7

preparing of novel hierarchical porous composites in order to realize fast and efficient

8

transport of ions/electrons and the stable structure during the charge/discharge process

9

[10-17]. As a result, they showed outstanding electrochemical performances including

10

extraordinary rate capabilities and long-term cycling stability when used as anode

11

materials in LIBs. For the excellent electrochemical performances, a common

12

explanation is that the porous 3D framework and conductive network can not only

13

increase the contact area between electrode and electrolyte, effectively facilitating the

14

ionic diffusion, but also provide more space for accommodating the stress caused by

15

volume changes, improving the cycle performance [13, 14]. In addition, the lithium

16

storage process involves both ion diffusion and electron transport in the electrode

17

materials, thus it is equally important to enhance the electrical conductivity in order to

19

SC

M AN U

TE D

EP

AC C

18

RI PT

1

further achieve high-rate capability. Therefore, it is an effective strategy to design porous nano-structured electrode materials with high conductivity and short path

20

length, since the diffusion time is proportional to the square of diffusion length and

21

inversely proportional to diffusivity [1].

22

Carbon nanotubes (CNTs) with 1D tubular structure have been considered as an 3

ACCEPTED MANUSCRIPT

ideal scaffold for loading active materials in energy conversion and storage

2

applications in view of extraordinary mechanical, electrical, and thermal properties

3

[18-25]. Besides, it is also a fine lithium-storage host at a low voltage, which makes it

4

an attractive anode material for LIBs. Compared with the above-mentioned 3D porous

5

carbonaceous materials, however, CNTs are still suffered from the kinetic problems

6

associated with ineffective Li+ ion transfer since Li+ ions cannot directly penetrate the

7

walls of CNTs. Previously, our group successfully synthesized a new kind of porous

8

carbon material (p-CNTs) by steaming multi-walled CNTs via acid vapor [26]. The

9

resultant p-CNTs showed high density of oxygen-containing functional groups and

10

numerous interconnected mesopores (6-10 nm) in the tubes with largely increased

11

specific surface area. As a derivative of CNTs, it is reasonable to believe that the

12

porous p-CNTs can provide multi-dimensional electronic network and promote the

13

electrolyte infiltration throughout the electrode, thus favoring the diffusion of both

14

electrons and Li+ ions.

TE D

M AN U

SC

RI PT

1

In this work, we take advantage of p-CNTs with unique structural superiority as

16

support to anchor electrochemical active material (CoFe2O4) on the surface of p-CNTs

17

by a method of supercritical carbon dioxide (scCO2)-expanded ethanol-assisted

19

AC C

18

EP

15

deposition as described in our previous work [5, 6]. As a straightforward, highly efficient and green technology, scCO2 deposition method can efficiently overcome the

20

common difficulties of traditional deposition methods and realize a uniform, accurate

21

and controllable coating of nanoparticles on support. As a result, the as-prepared

22

[email protected] nanocomposite exhibits excellent electrochemical performances with 4

ACCEPTED MANUSCRIPT

high reversible capacity, excellent cycling performance and enhanced rate capability

2

compared to the pristine [email protected] In this work, we focus our attention on

3

discussion of the relation between the structure and electrochemical performance of

4

the [email protected] as an anode material for LIBs. It is confirmed that the porous

5

carbon nanotubes could provide an efficient way for ion diffusion and continuous

6

electron transport. At the same time, the CFO particles are bonded strongly to the

7

p-CNTs through oxygen bridges, which facilitate the electron fast capture from

8

p-CNTs to CFO and thus resulting in a high reversible capacity and excellent rate

9

performance.

2. Experimental

11

2.1. Materials

TE D

10

M AN U

SC

RI PT

1

All of the reactants were of analytical grade and used without further purification.

13

The multi-walled CNTs with diameter of 40-60 nm were purchased from Shenzhen

14

Nanotech Port. Co., Ltd., and purified with 6 M HCl to remove the metal impurities or

15

catalysts before used.

17 18

AC C

16

EP

12

2.2. Synthesis of p-CNTs The p-CNTs were prepared according to our previous method [26]. Briefly, pristine

CNTs (0.3 g) were loaded on a porous SiO2 griddle of glass steamer and then placed

19

into a 50 mL Teflon-vessel, at the bottom of which 1 mL HNO3 was added previously.

20

Subsequently, the vessel was moved to the oven for steaming at 200 °C for 5 h.

21

Finally, the products were washed with distilled water and ethanol, and dried. 5

ACCEPTED MANUSCRIPT

1

2.3. Synthesis of [email protected] and [email protected] Composites The [email protected] composite was prepared by scCO2-expanded ethanol-assisted

3

deposition method. Typically, the as-prepared p-CNTs (20 mg) were dispersed into 10

4

mL of ethanol under ultrasonic treatment for 0.5 h. Subsequently, 0.15 mol of

5

Co(NO3)2·6H2O and 0.30 mol of Fe(NO3)3·9H2O were added. After stirring for 10 min,

6

the resultant solution was transferred into a high-pressure autoclave (50 mL), and

7

placed into an oil bath at 200 °C. Then, the autoclave was pumped CO2 up to 12.0

8

MPa by using a CO2 delivery pump to form a homogeneous expanded fluid under

9

rapid stirring. After 2 h, the vessel was cooled to room temperature and depressurized

10

slowly. The product was harvested by centrifugation and washed with deionized water

11

and ethanol. Finally, the as-prepared product was calcined at 450 °C for 2 h with a

12

heating rate of 3 °C min-1 under a nitrogen atmosphere, which was denoted as

13

[email protected] For comparison, the [email protected] composite was also prepared under

14

the similar procedure, except that the pristine multi-walled CNTs were used instead of

15

the p-CNTs.

16

2.4. Characterization

18

SC

M AN U

TE D

EP

AC C

17

RI PT

2

Power X-ray diffraction (XRD) patterns of the samples were measured from 10 to

80° by a Bruker D8 Advance XRD instrument with Cu Kα (λ = 1.5418 Å) radiation.

19

Thermogravimetry

analysis

(TGA)

was

20

thermogravimetric analyser. X-ray photoelectron spectroscopy (XPS) was recorded on

21

a PHI quantera SXM spectrometer with an Al Kα = 280.00 eV excitation source. 6

performanced

on

a

TGA

2050

ACCEPTED MANUSCRIPT

Scanning electron microscopy (SEM) images were performed on a Hitachi S-4800

2

field emission scanning electron microscope at an accelerating voltage of 10 kV. The

3

SEM sample on carbon tape was subject to the observation without any

4

sputter-coating. Transmission electron microscopy (TEM), high-angle annular

5

dark-field scanning transmission electron microscopy (HAADF-STEM) and element

6

mapping were performed on a Tecnai G20 instrumet with a fied emission gun

7

operating at 200 kV. Samples were dispersed in ethanol and the dispersions were

8

dropped on a copper grid and dried. Nitrogen absorption/desorption isotherms were

9

conducted on an ASAP 2020 accelerated surface area and porosimetry instrument

10

(Micromeritics) at 77 K with a pretreatment of the samples at 150 °C for 12 h under

11

vacuum

12

Brunauer-Emmett-Teller (BET) method. The pore size distribution and average pore

13

diameter were calculated by the Barret-Joymer-Halenda (BJH) equation.

14

2.5. Electrochemical Measurement

The

specific

surface

area

was

calculated

using

the

EP

TE D

condition.

M AN U

SC

RI PT

1

Electrochemical measurements were evaluated with CR2025-type cells

16

assembled in an argon-filled glove box with water and oxygen contents < 0.1 ppm.

17 18 19

AC C

15

For preparing the working electrodes, active materials, acetylene black and polyvinylidene fluoride (PVDF) were mixed at a weight ratio of 80:10:10 in N-methyl-2-pyrrolidinone (NMP) to form a uniform slurry, and then pasted on a Cu

20

foil. The foil was dried at 100 °C under vacuum and then cut into electrodes with a

21

diameter of 1.2 cm and an active material load of about 1.5 mg cm-2. In assembling

22

the cells, a lithium foil was used as the counter electrode, and Celgard 2400 7

ACCEPTED MANUSCRIPT

membrane was used as a separator, and 1 M LiPF6 (in ethylene carbonate/dimethyl

2

carbonate, EC/DMC, 1:1, v/v) was used as the electrolyte. Galvanostatic

3

charge/discharge tests were carried out on a multi-channel NEWARE Battery

4

Measurement System in the voltage range of 0.01-3.0 V (vs. Li+/Li). The capacity is

5

calculated based on the total mass of the active material. After cycles, the cell was

6

disassembled in the glove-box, and the working electrode was taken out and washed

7

with pure DMC solution for the TEM characterization. Cyclic voltammetry (CV) and

8

electrochemical impedance spectroscopy (EIS) measurements were performed on a

9

VMP3 electrochemical workstation (Bio-logic Inc.).

M AN U

SC

RI PT

1

3. Results and discussion

11

3.1. Synthesis and Structural Analysis

TE D

10

Scheme 1 illustrates the synthetic procedure of [email protected] composite. First,

13

porous p-CNTs are synthesized through steaming multi-walled CNTs via acid vapor

14

according to our previous work [26]. Next, the p-CNTs are immersed into an ethanol

15

solution of Fe(NO3)3⋅9H2O and Co(NO3)2⋅6H2O, and then the mixture is expanded by

16

introducing high pressure CO2 to obtain a homogeneous fluid at a temperature of

18

AC C

17

EP

12

200 °C. During this process, the metal complexes are precipitated out in situ and evenly coated on the surface of p-CNTs. Finally, the precipitates are annealed at

19

450 °C under N2 to form the [email protected] composite. Fig. 1a shows the XRD

20

patterns of the as-prepared [email protected] and [email protected] composites. Both

21

patterns display sharp diffraction peaks at 2θ values of 30.3, 35.7, 43.2, 57.3 and 8

ACCEPTED MANUSCRIPT

62.9°, corresponding to (220), (311), (400), (511) and (440) planes of a face-centered

2

cubic spinel CoFe2O4 (JCPDS No. 22-1086) with the Fd3m space group. Besides, the

3

peak appeared at 26° can be indexed into the (002) diffraction of the multi-walled

4

CNTs [24, 27, 28]. To quantify the content of CFO in the composite,

5

thermogravimetric analysis (TGA) was carried out in air from room temperature to

6

800 °C at a rate of 10 °C/min (Fig. 1b). Both samples present a rapid mass loss

7

between 350 and 600 °C, which is ascribeld to the oxidation of CNTs. The amounts of

8

CFO in the [email protected] and [email protected] composites are about 61% and 62%,

9

respectively, as estimated according to the residual weight ratio.

M AN U

SC

RI PT

1

In order to analyze the oxidation level of the p-CNTs, the C 1s XPS spectra of the

11

pristine CNTs and p-CNTs were tested (Fig. 2a). It is clear that an extra peak presents

12

at 288.9 eV (carboxylate, O-C=O) for p-CNTs, besides the same peaks with CNTs at

13

284.5 eV (carbon sp2, C=C), 285.1 eV (carbon sp3, C-C), and 286.2 eV

14

(epoxy/hydroxyl, C-O-C/C-OH) [29]. It suggests that the surfaces of p-CNTs possess

15

rich acidic groups after oxidation treatment. Since the carboxyl groups have the

16

highest affinity for metal ions in a wide range of pH, the p-CNTs are more promising

17

to immobilize foreign materials compared to the pristine CNTs. Fig. 2b-d show the

19

EP

AC C

18

TE D

10

XPS spectra of the [email protected] and [email protected] composites. By using a Gaussian fitting method, the Co 2p spectra (Fig. 2b) are well fitted into two spin-orbit doublets,

20

characteristic of Co2+ and Co3+, and two shakeup satellites (identified as “sat.”). The

21

Fe 2p spectra are best fitted into two spin-orbit doublets, characteristic of Fe2+ and

22

Fe3+ (Fig. 2c). Moreover, there are two satellite peaks at the high binding energy sides 9

ACCEPTED MANUSCRIPT

of Fe 2p3/2 and 2p1/2 edge. Thus, both of the electron couples of Co3+/Co2+ and

2

Fe3+/Fe2+ exist in the spinel CoFe2O4 structure. In addition, the O 1s XPS signals (Fig.

3

2d) can be divided into three peaks at 529.9, 531.9 and 533.4 eV. Specifically, the

4

peak at 529.9 eV is typical of metal-oxygen bonds. The peak sitting at 531.9 eV is

5

usually associated with defects, contaminants, and a number of surface species

6

including hydroxyls, chemisorbed oxygen, under-coordinated lattice oxygen, or

7

species intrinsic to the surface of the spinel. The peaks at 533.4 eV can be attributed

8

to multiplicity of physic- and chemisorbed water at the surface [30].

M AN U

SC

RI PT

1

Fig. 3 displays the morphologies of the pristine CNTs, p-CNTs and their

10

composites anchored with CFO. Fig. 3a shows the SEM image of the randomly

11

entangled CNTs with an outer diameter of approximately 40-60 nm with relatively

12

smooth surfaces. By contrast, the surfaces of p-CNTs treated with acid vapor steaming

13

become quite rough (Fig. 3d). Importantly, large quantities of interconnected pores

14

around 8 nm and structural defects present along the sidewalls on the p-CNTs (Fig.

15

3d). It is worth to expect that such novel porous carbon nanotubes should have

16

wide-spread applications, for the interconnected mesopores can largely facilitate the

17

penetration and circulation of the molecules inside the tubes, and large surface area

19

EP

AC C

18

TE D

9

together with high density of functional groups can offer more active sites and space for loading and anchoring the molecules. For the [email protected] (Fig. 3e, f), a layer of

20

CFO particles are uniformly and completely coated on the surfaces of p-CNTs,

21

forming a core/shell nanostructure. By contrast, some naked CNTs are found for the

22

[email protected] as indicated by arrows in Fig. 3b and 3c, which may be ascribed to the 10

ACCEPTED MANUSCRIPT

1

smooth surface of CNTs with deficient functional groups. To further examine the architecture of the [email protected] and [email protected] coaxial

3

nanotubes, the samples were investigated by TEM and HRTEM. From Fig. 4a-b, all

4

the particles are anchored on the surfaces of CNTs without the presence of free

5

particles in solution, but the naked CNTs exist in the sample of [email protected] (Fig. 4b).

6

By contrast, the [email protected] presents to be a perfect composite with uniform CFO

7

particles covered on the whole surface of p-CNTs completely (Fig. 4c-e). Furthermore,

8

the HRTEM image (Fig. 4f) demonstrates the anchored CFO particles are

9

well-textured with high crystallinity. The lattice fringe with lattice spacing of 0.34 nm

10

corresponds to the (002) crystalline planes of the multi-walled CNTs, and the other

11

one with lattice spacing of 0.29 nm corresponds to the (220) plane of CFO.

12

Additionally, the elemental mapping images of the as-prepared [email protected] on C,

13

O, Fe and Co (Fig. 4e) also demonstrate the uniform dispersion of CFO on p-CNTs.

14

Moreover, both of the samples show similar CFO size, that is, 12 nm in diameter (the

15

insets of Fig. 4b and 4d). It is well-known that the nanosized particles tend to

16

agglomerate because of the overwhelming driving force to reduce their surface energy.

17

However, the agglomeration of CFO is adequately prevented in this work, indicating

19

SC

M AN U

TE D

EP

AC C

18

RI PT

2

that the scCO2-expanded ethanol is the key factor in controlling the morphology of nanoparticles on the substrate.

20

In order to analyze the porous structure of p-CNTs and [email protected], nitrogen

21

adsorption-desorption isotherms were measured with CNTs and [email protected] as

22

contrast. Fig. 5a shows that the CNTs have only one hysteresis at about 0.95 P/P0, 11

ACCEPTED MANUSCRIPT

while p-CNTs (Fig. 5b) presents two hystereses at about 0.8 and 0.95 P/P0, indicating

2

the presence of new mesopores on p-CNTs. Besides, the p-CNTs give rise to the

3

special surface area of 149.4 m2 g-1 and pore volume of 1.24 cm3 g-1, much higher

4

than those of CNTs (62.45 m2 g-1, 0.95 cm3 g-1), which mainly contributes to the

5

porosity of the p-CNTs. The pore size distribution of CNTs shows a dominant peak at

6

around 2.5 nm, besides that the p-CNTs shows an extra kind of pore centered at about

7

8.0 nm, which is caused by the acid vapor treatment, and the result is in accordance

8

with the above TEM image in Fig. 3d. It is believed that such a porous structure can

9

provide

efficient

transport

M AN U

SC

RI PT

1

pathways

of

lithium

ions

and

increase

the

electrolyte/electrode contact area, which is critical for high-rate LIB applications. In

11

addition, after the incorporation of CFO particles in the matrix (Fig. 5c, d), both the

12

BET surface area and pore volume of [email protected] (52.13 m2 g-1, 0.25 cm3 g-1) and

13

[email protected] (69.9 m2 g-1, 0.28 cm3 g-1) composites significantly decrease.

14

Meanwhile, the dominant peak of [email protected] present at around 4.5 nm, as the

15

mesopores are partly blocked by the anchored CFO particles.

16

3.2. Electrochemical performances

18

EP

AC C

17

TE D

10

The electrochemical performance of the two kinds of CNTs and the composites

were evaluated by cyclic voltammetry (CV) and galvanostatic charge/discharge

19

cycling. Fig. 6a and 6c show the cyclic voltammograms of the CNTs and p-CNTs,

20

respectively. During the first cathodic scan, both types of CNTs show two

21

characteristic reduction peaks at the voltages of 0.6 and 0.05 V. The peak centered at 12

ACCEPTED MANUSCRIPT

0.6 V is ascribed to the electrolyte decomposition and solid electrolyte interface (SEI)

2

film formation on CNTs surface. And the peak at about 0.05 V should be attributed to

3

the intercalation of lithium ions between carbon layers of nanotubes [13]. During the

4

first anode scan, two oxidation peaks at 0.2 and 1.25 V are observed for both CNTs

5

and p-CNTs, which can be attributed to the extraction of Li+ ions from the graphitic

6

layers and the inner channels of the CNTs, respectively [31]. Although the reduction

7

and oxidation peaks are similar, the peak width and area of cyclic voltammograms for

8

p-CNTs are obviously bigger than those of CNTs, indicating that p-CNTs provide

9

more sites for lithium diffusion and storage in comparison with pristine CNTs.

10

Galvanostatic charge/discharge cycling measurements also exhibit the similar results

11

in Fig. 6b, d. The first discharge capacity of p-CNTs (1160 mAh g-1) is much higher

12

than that of CNTs (556 mAh g-1), especially in the voltage range from 0.2 to 0.01 V.

13

The extra discharge capacity of p-CNTs could be attributed to the presence of pores

14

and structural defects of p-CNTs as well as abundant oxygen functional groups, which

15

can not only make the interlayer space of CNTs accessible for lithium-ion insertion

16

and extraction, but also prompt the electrolyte decomposition, leading to the increase

17

of lithium-storage capacity [31, 32]. Fig. 6e and 6g show the CV curves of the

19

SC

M AN U

TE D

EP

AC C

18

RI PT

1

[email protected] and [email protected] for the first five cycles. During the first cycle, both samples present two obvious reduction peaks located at about 1.5 and 0.6 V. The first

20

peak could be associated with the irreversible reduction of electrolyte [33], and the

21

second one is associated with the reduction of CoFe2O4 as followings: CoFe2O4 +

22

8Li+ + 8e- → Co + 2Fe + 4Li2O, which is consistent with the conversion reaction of 13

ACCEPTED MANUSCRIPT

other transition metal oxides [34]. The oxidation peak locates at around 1.5-2.2 V,

2

which is ascribed to the reversible oxidation of Fe and Co to Fe3+ and Co2+,

3

respectively [35]. Similarly, the peak width and area of cyclic voltammograms for

4

[email protected] are bigger than those of [email protected] Additionally, according to the

5

results of the first discharge/charge voltage profiles (Fig. 6f, h), the initial discharge

6

capacity of [email protected] is 1246 mAh g-1; it is much larger than that of

7

[email protected] electrode (1080 mAh g-1).

SC

RI PT

1

Fig. 7a shows the cycling performance of all samples at 0.1 A g-1 in the voltage

9

range of 0.01-3.0 V. Obviously, the p-CNTs show a little higher discharge capacity

10

than that of CNTs, which is consist with the CV results. Meanwhile, the

11

[email protected] also shows more stable cycle performance, and the discharge capacity

12

still can be retained at 1077 mAh g-1 after 100 cycles, while the [email protected] shows a

13

relatively lower value (940 mAh g-1). Fig. 7b shows the rate performance of the

14

[email protected] and [email protected] composites at different current densities. When

15

cycled at the current densities of 0.2, 0.4, 0.6, 0.8, 1 and 2 A g-1, the [email protected]

16

electrode can deliver discharge capacities of 1125, 1032, 996, 963, 932 and 806 mAh

17

g-1, respectively, much higher than those of [email protected] electrode, i.e., 962, 901, 823,

19

TE D

EP

AC C

18

M AN U

8

758, 670 and 436 mAh g-1. It is worth mentioning that a capacity of about 694 mAh g-1 can be retained for [email protected] even at a large current density of 3 A g-1,

20

suggesting its excellent high-rate capability. Besides, it is also observed that the

21

capacity difference between the two samples increases with the increasing of current

22

density. In other words, the advantage of the [email protected] over [email protected] 14

ACCEPTED MANUSCRIPT

1

becomes more significant at higher rates. Fig. 7c shows the long-life cycling stability of [email protected] and [email protected] at a

3

high current density of 2 A g-1 up to 200 cycles. As seen, the capacity of [email protected]

4

drops rapidly before 50 cycles and retains only 47.4% after 200 cycles, while the

5

[email protected] presents better stability with about 87.7% capacity retention after 200

6

cycles, indicating its excellent long-life stability. In addition, based on the unique

7

structural advantages of [email protected], we compare the result in this work with our

8

previously reported graphene-based CFO composite ([email protected]) with similar CFO

9

content (Fig. 7d) [36]. Generally, graphene is an excellent supporting material for

10

nanocomposite because of its superior electronic conductivity, high surface-to-volume

11

ratio, ultrathin thickness, and structural flexibility [37, 38]. In this work, however, the

12

as-prepared [email protected] shows unexpected better performance than that of

13

[email protected], about 40-70 mAh g-1 higher than that of [email protected] at each rate. The

14

result is explained as: in the case of [email protected], the presence of mesopores in

15

p-CNTs enables the mass transport through the tubes in a 3D model, therefore,

16

enhancing the performance as the mass transfer path is shorten. However, for the

17

[email protected] composite, mass transport is mainly in a 2D model along the space

19

SC

M AN U

TE D

EP

AC C

18

RI PT

2

between the graphene sheets, since the direction vertical to the graphene sheets is blocked by the micrometer-scaled graphene sheets [39, 40]. Furthermore, the

20

performances of [email protected] at high rates (i.e., 806 mAh g-1 at 2 A g-1; 694 mAh

21

g-1 at 3 A g-1) are significantly higher than those previously reported materials on

22

CFO-based and Fe3O4-based composites (as shown in Fig. S1) [41-47]. Based on the 15

ACCEPTED MANUSCRIPT

above results, the suitable porous structure is beneficial for the efficient diffusion of

2

electrolyte and transfer of electrons, which is benefit for the enhancement of the rate

3

capability.

RI PT

1

Electrochemical impedance spectroscopy (EIS) measurements were also conducted

5

to further confirm the relationship of structure and electrode kinetics. The Nyquist

6

plots of [email protected] and [email protected] electrodes after 100 cycles at 2 A g-1 are

7

compared in Fig. 8a-b. They are composed of a depressed semicircle in the medium

8

frequency region followed by a slanted line in the low-frequency region. The

9

high-frequency semicircle is attributed to the charge transfer process at

10

electrode/electrolyte interface. The inclined line corresponds to the lithium-diffusion

11

process into the bulk of the electrode, the so-called Warburg diffusion. The impedance

12

spectra for both electrodes can be modeled by the equivalent circuit in the inset of Fig.

13

8a. Rs is the electrolyte resistance, and Cf and Rf are the capacitance and resistance of

14

the passivation film formed on the surface of the electrode, respectively. Cdl and Rct

15

are the double-layer capacitance and charge-transfer resistance, respectively [48]. Zw

16

is the Warburg impedance related to the diffusion of lithium ions into the bulk of the

17

electrode. Additionally, the exchange current density (i0), a parameter to indicate the

19

M AN U

TE D

EP

AC C

18

SC

4

reversibility of the electrode, can be calculated with the equation: i0 = RT/nFRct. The kinetic parameters of two electrodes are summarized in Table 1. From Table 1, it can

20

be seen that the Rct of [email protected] is significantly smaller than that of [email protected],

21

while the lithium ion diffusion coefficient (DLi) of [email protected] is two times of that

22

of [email protected], indicating the unique porous structure indeed enhances the charge 16

ACCEPTED MANUSCRIPT

transfer across the electrode/electrolyte interface and the solid migration velocity of

2

lithium ions. Moreover, the i0 value of [email protected] is higher than that of

3

[email protected], suggesting that the electrochemical reversibility of [email protected] is

4

enhanced by oxidation treatment and porous structure.

RI PT

1

Additionally, the electrode morphology after cycling was analyzed (Fig. 8c-d). The

6

TEM image of the cycled [email protected] shows that most CFO particles have peeled off

7

from the CNTs backbones, while the [email protected] can preserve its original

8

morphology to some extent. It is speculated that the closely packed CFO particles of

9

[email protected] cause severe volume changes because of lack of space to accommodate

10

during charge/discharge processes, resulting in mechanical disintegration and

11

electrical disconnection from the CNTs backbones. Moreover, it is known that the

12

stability of SEI is extremely important for good cyclability of LIBs, since SEI layer

13

allows Li+ ions to diffuse through it, while blocking the electrolyte molecules and

14

ensuring reversible lithiation and delithiation during cycling [49]. While in the case of

15

[email protected], some CNTs are not fully covered instead appearing large amounts of

16

naked surfaces, which could cause pronounced surface reactions between CNTs and

17

electrolyte, forming more SEI layers and leading to the electrical disconnection from

19

M AN U

TE D

EP

AC C

18

SC

5

the CNTs backbones, rapid deterioration and the low retention of the capacity. Considering the complementary properties of the two components in energy storage,

20

a synergistic effect may explain the superior high-rate and cycling performance of

21

[email protected] Firstly, the porous structure increases the electrode/electrolyte

22

interfaces and shortens the Li+ ions diffusion pathways to nanoscale, which improve 17

ACCEPTED MANUSCRIPT

the diffusion kinetic process (i.e., reduced charge transfer resistance, increased the

2

diffusion coefficient of Li+ ions and exchange current density), resulting in excellent

3

rate capability. In addition, the existence of pores could also benefit the buffer of the

4

volume expansion/contraction upon cycling [8, 11, 14], thus improving the cycle

5

stability. Secondly, the rich oxygen-containing functional groups on p-CNTs provide

6

more nucleation sites and form tight confinement of CFO on p-CNTs backbone,

7

which further reduce the self-aggregation of active materials and pulverization of the

8

electrode, thereby maintaining the structural integrity and cycle stability of the

9

electrode. Meanwhile, the intimate entanglement between the conductive p-CNTs and

10

CFO particles through strong metal-oxygen bridges ensures the electron fast capture

11

and transfer as well as the structural robustness [50], and thus resulting in a high

12

reversible capacity and excellent rate performance. Thirdly, in contrast with the

13

existence of naked CNTs in [email protected], the p-CNTs are completely and uniformly

14

covered with CFO particles, which can enhance the utilization efficiency of

15

electroactive materials by providing more CFO particles to react with lithium ions

16

more quickly in high current densities.

18 19

SC

M AN U

TE D

EP

AC C

17

RI PT

1

4. Conclusions

In summary, we adopted a straightforward, highly-efficient and green synthetic

route to successfully anchor nanosized CoFe2O4 on porous carbon nanotubes. The

20

p-CNTs with numerous interconnected mesopores in the tubes possess largely

21

increased specific surface area, and rich oxygen-containing functional groups

22

compared to pristine multi-walled CNTs. Compared to the non-porous [email protected] 18

ACCEPTED MANUSCRIPT

composite, the [email protected] composites exhibit excellent lithium storage properties

2

in terms of cycling performance (1077 mAh g-1 after 100 cycles) and rate capability

3

(694 mAh g-1 at 3 A g-1), making it a promising anode material for high-energy LIBs.

4

The excellent electrochemical performance can be mainly attributed to the synergistic

5

advantages of the uniform distribution of nanosized CFO on p-CNTs and the porous

6

p-CNTs support, which provide bicontinuous electron/ion pathways, large

7

electrode/electrolyte contact area, facile strain relaxation and firm immobilization of

8

active materials. As the notable capacity advantage of the present [email protected], the

9

strategy provides in this work direct access to hierarchical porous carbon materials,

M AN U

SC

RI PT

1

which have immense potential applications in the design of high performance LIBs.

11

Acknowledgements

TE D

10

This work was supported by the NSFC-JSPS international cooperation project

13

(21311140166) and NSFC (21273222), Shandong Province Science and Technology

14

Program (2014GGX102020) and Higher Educational Science and Technology

15

Program (J14LC08).

17 18

AC C

16

EP

12

Appendix A. Supplementary data Supplementary

data

related

to

this

article

can

be

found

at

http://dx.doi.org/10.1016/j.jpowsour.2015.××.×××.

19

References

20

[1] Y.X. Tang, Y.Y. Zhang, J.Y. Deng, D.P. Qi, W.R. Leow, J.Q. Wei, S.Y. Yin, Z.L. 19

ACCEPTED MANUSCRIPT

Dong, R. Yazami, Z. Chen, X.D. Chen, Angew. Chem. Int. Ed., 53 (2014)

2

13488-13492.

3

[2] L. Qie, W.M. Chen, Z.H. Wang, Q.G. Shao, X. Li, L.X. Yuan, X.L. Hu, W.X.

4

Zhang, Y.H. Huang, Adv. Mater., 24 (2012) 2047-2050.

5

[3] Z.S. Wu, W.C. Ren, L. Wen, L.B. Gao, J.P. Zhao, Z.P. Chen, G.M. Zhou, F. Li,

6

H.M. Cheng, ACS Nano, 4 (2010) 3187-3194.

7

[4] Y. Li, Z.Y. Fu, B.L. Su, Adv. Funct. Mater., 22 (2012) 4634-4667.

8

[5] L.Y. Wang, L.H. Zhuo, C. Zhang, F.Y. Zhao, ACS Appl. Mater. Interfaces, 6 (2014)

9

10813-10820.

M AN U

SC

RI PT

1

[6] L.Y. Wang, L.H. Zhuo, C. Zhang, F.Y. Zhao, Chem. Eur. J., 20 (2014) 4308-4315.

11

[7] E. Kang, Y.S. Jung, A.S. Cavanagh, G.H. Kim, S.M. George, A.C. Dillon, J.K.

12

Kim, J. Lee, Adv. Funct. Mater., 21 (2011) 2430-2438.

13

[8] X. Huang, H. Yu, J. Chen, Z.Y. Lu, R. Yazami, H.H. Hng, Adv. Mater., 26 (2014)

14

1296-1303.

15

[9] L.W. Yin, Z.W. Zhang, Z.Q. Li, F.B. Hao, Q. Li, C.X. Wang, R.H. Fan, Y.X. Qi,

16

Adv. Funct. Mater., 24 (2014) 4176-4185.

17

[10] F. Zhou, S. Xin, H.W. Liang, L.T. Song, S.H. Yu, Angew. Chem., 126 (2014)

19

EP

AC C

18

TE D

10

11736-11740.

[11] L.C. Zeng, W.C. Zeng, Y. Jiang, X.L. Wei, W.H. Li, C.L. Yang, Y.W. Zhu, Y. Yu,

20

Adv. Energy Mater., (2014) 1401377.

21

[12] J.K. Hwang, C.S. Jo, K.Y. Hur, J. Lim, S.S. Kim, J.W. Lee, J. Am. Chem. Soc.,

22

136 (2014) 16066-16072. 20

ACCEPTED MANUSCRIPT

[13] K.Z. Cao, L.F. Jiao, H.Q. Liu, Y.C. Liu, Y.J. Wang, Z.P. Guo, H.T. Yuan, Adv.

2

Energy Mater., (2014) 1401421.

3

[14] Q.Y. An, F. Lv, Q.Q. Liu, C.H. Han, K.N. Zhao, J.Z. Sheng, Q.L. Wei, M.Y. Yan,

4

L.Q. Mai, Nano Lett., 14 (2014) 6250-6256.

5

[15] F. Zou, X.L. Hu, Z. Li, L. Qie, C.C. Hu, R. Zeng, Y. Jiang, Y.H. Huang, Adv.

6

Mater., 26 (2014) 6622-6628.

7

[16] J.C. Ye, A.C. Baumgaertel, Y.M. Wang, J. Biener, M.M. Biener, ACS Nano,

8

(2014).

9

[17] C.N. He, S. Wu, N.Q. Zhao, C.S. Shi, E.Z. Liu, J.J. Li, ACS Nano, 7 (2013)

M AN U

SC

RI PT

1

4459-4469.

11

[18] K. Fu, O. Yildiz, H. Bhanushali, Y.X. Wang, K. Stano, L.G. Xue, X.W. Zhang,

12

P.D. Bradford, Adv. Mater., 25 (2013) 5109-5114.

13

[19] M. Mazloumi, S. Shadmehr, Y. Rangom, L.F. Nazar, X.W. Tang, ACS Nano, 7

14

(2013) 4281-4288.

15

[20] Y. Wu, Y. Wei, J.P. Wang, K.L. Jiang, S.S. Fan, Nano Lett., 13 (2013) 818-823.

16

[21] Z.Y. Wang, D.Y. Luan, S. Madhavi, Y. Hu, X.W. Lou, Energy Environ. Sci., 5

17

(2012) 5252-5256.

19

EP

AC C

18

TE D

10

[22] X.L. Jia, Z. Chen, X. Cui, Y.T. Peng, X. Wang, G.L. Wang, F. Wei, Y.F. Lu, ACS Nano, 6 (2012) 9911-9919.

20

[23] R.A. Dileo, A. Castiglia, M.J. Ganter, R.E. Rogers, C.D. Cress, R.P. Raffaelle,

21

B.J. Landi, ACS Nano, 4 (2010) 6121-6131.

22

[24] S.J. Ding, J.S. Chen, X.W. Lou, Adv. Funct. Mater., 21 (2011) 4120-4125. 21

ACCEPTED MANUSCRIPT

[25] M. Bedewy, E.R. Meshot, M.J. Reinker, A.J. Hart, ACS Nano, 5 (2011)

2

8974-8989.

3

[26] J. Ming, Y.Q. Wu, Y.C. Yu, F.Y. Zhao, Chem. Commun., 47 (2011) 5223-5235.

4

[27] L. Sun, M.Y. Li, Y. Jiang, W.B. Kong, K.L. Jiang, J.P. Wang, S.S. Fan, Nano Lett.,

5

14 (2014) 4044-4049.

6

[28] S.H. Ren, X.Y. Zhao, R.Y. Chen, M. Fichtner, J. Power Sources, 260 (2014)

7

205-210.

8

[29] L.H. Zhuo, Y.Q. Wu, J. Ming, L.Y. Wang, Y.C. Yu, X.B. Zhang, F.Y. Zhao, J.

9

Mater. Chem. A, 1 (2013) 1141-1147.

M AN U

SC

RI PT

1

[30] J.F. Li, S.L. Xiong, Y.R. Liu, Z.C. Ju, Y.T. Qian, ACS Appl. Mater. Interfaces, 5

11

(2013) 981-988.

12

[31] X.X. Wang, J.N. Wang, H. Chang, Y.F. Zhang, Adv. Funct. Mater., 17 (2007)

13

3613-3618.

14

[32] S.B. Yang, H.H. Song, X.H. Chen, A.V. Okotrub, L.G. Bulusheva, Electrochim.

15

Acta, 52 (2007) 5286-5293.

16

[33] Q.Q. Xiong, J.P. Tu, S.J. Shi, X.Y. Liu, X.L. Wang, C.D. Gu, J. Power Sources,

17

256 (2014) 153-159.

19

EP

AC C

18

TE D

10

[34] L.L. Luo, J.S. Wu, J.M. Xu, V.P. Dravid, ACS Nano, 8 (2014) 11560-11566. [35] A.K. Rai, J. Gim, T.V. Thi, D. Ahn, S.J. Cho, J. Kim, J. Phys. Chem. C, 118

20

(2014) 11234-11243.

21

[36] L.Y. Wang, L.H. Zhuo, C. Zhang, F.Y. Zhao, J. Power Sources, 275 (2015)

22

650-659. 22

ACCEPTED MANUSCRIPT

[37] Y.J. Gong, S.B. Yang, Z. Liu, L.L. Ma, R. Vajtai, P.M. Ajayan, Adv. Mater., 25

2

(2013) 3979-3984.

3

[38] N.O. Weiss, H.L. Zhou, L. Liao, Y. Liu, S. Jiang, Y. Huang, X.F. Duan, Adv.

4

Mater., 24 (2012) 5782-5825.

5

[39] D. Zhou, Y. Cui, P.W. Xiao, M.Y. Jiang, B.H. Han, Nat. Commun., 5 (2014)

6

4716-4722.

7

[40] C. Uthaisar, V. Barone, Nano Lett., 10 (2010) 2838-2842.

8

[41] Z.L. Zhang, Y.H. Wang, M.J. Zhang, Q.Q. Tan, X. Lv, Z.Y. Zhong, F.B. Su, J.

9

Mater. Chem. A, 1 (2013) 7444-7450.

M AN U

SC

RI PT

1

[42] Y.Z. Su, S. Li, D.Q. Wu, F. Zhang, H.W. Liang, P.F. Gao, C. Cheng, X.L. Feng,

11

ACS Nano, 6 (2012) 8349-8356.

12

[43] W. Wei, S.B. Yang, H.X. Zhou, I. Lieberwirth, X.L. Feng, K. Muellen, Adv.

13

Mater., 25 (2013) 2909-2914.

14

[44] J.L. Cheng, B. Wang, C.M. Park, Y.P. Wu, H.T. Huang, F.D. Nie, Chem. Eur. J.,

15

19 (2013) 9866-9874.

16

[45] G.M. Zhou, D.W. Wang, F. Li, L.L. Zhang, N. Li, Z.S. Wu, L. Wen, G.Q. Lu,

17

H.M. Cheng, Chem. Mater., 22 (2010) 5306-5313.

19

EP

AC C

18

TE D

10

[46] J.E. Lee, S.H. Yu, D.J. Lee, D.C. Lee, S.I. Han, Y.E. Sung, T. Hyeon, Energy Environ. Sci., 5 (2012) 9528-9533.

20

[47] H. Fu, Z.J. Du, W. Zou, H.Q. Li, C. Zhang, Carbon, 65 (2013) 112-123.

21

[48] N.N. Wang, H.Y. Xu, L. Chen, X. Gu, J. Yang, Y.T. Qian, J. Power Sources, 247

22

(2014) 163-169. 23

ACCEPTED MANUSCRIPT

[49] A.V. Cresce, S.M. Russell, D.R. Baker, K.J. Gaskell, K. Xu, Nano Lett., 14 (2014)

2

1405-1412.

3

[50] G.M. Zhou, D.W. Wang, L.C. Yin, N. Li, F. Li, H.M. Cheng, ACS Nano, 6 (2012)

4

3214-3223.

AC C

EP

TE D

M AN U

SC

RI PT

1

24

ACCEPTED MANUSCRIPT

Figure captions

2

Scheme 1. Schematic illustration of the forming process of [email protected] composite.

3

Fig. 1. (a) XRD and (b) TG curves of the (i) [email protected] and (ii) [email protected]

4

composites.

5

Fig. 2. (a) C1s XPS spectra of the pristine CNTs and p-CNTs. High-resolution (b) Co

6

2p, (c) Fe 2p and (d) O 1s XPS spectra of the [email protected] and [email protected]

7

composites.

8

Fig. 3. SEM images of (a) CNTs, (b) [email protected], and (c) enlarged SEM image of

9

the selected area marked by a box in (b). SEM images of (d) p-CNTs (The inset shows

M AN U

SC

RI PT

1

a typical TEM image of one individual p-CNT.), and (e, f) [email protected]

11

Fig. 4. TEM and high resolution TEM (HRTEM) images of (a, b) [email protected] and

12

(c-f) [email protected] Insets in b and d are the size distribution of CFO. (g) STEM and

13

corresponding elemental mappings of the as-prepared [email protected] composite.

14

Fig. 5. Nitrogen adsorption-desorption isotherms of (a) CNTs, (b) p-CNTs, (c)

15

[email protected] and (d) [email protected] Inset: Pore size distribution from desorption

16

branch through the BJH method.

18 19

EP

AC C

17

TE D

10

Fig. 6. CVs and the first two discharge/charge voltage profiles of (a, b) pristine CNTs, (c, d) p-CNTs, (e, f) [email protected] and (g, h) [email protected] Fig. 7. (a) Cycle stability of all samples at 0.1 A g-1 with the Coulombic efficiency of

20

[email protected], (b) rate performance, (c) high-rate performance at 2 A g-1, and (d)

21

comparison of capacity at different rates for [email protected] electrode with our

22

previously reported graphene-based CFO composite ([email protected]). 25

ACCEPTED MANUSCRIPT

Fig. 8. Nyquist plots and TEM images of the electrodes after 100 cycles at 2 A g-1 for

2

(a, c) [email protected] and (b, d) [email protected] The insets in a and b show the

3

equivalent circuit model of the system and their structural models.

4

Table 1. Impedance parameters of the electrodes.

AC C

EP

TE D

M AN U

SC

RI PT

1

26

ACCEPTED MANUSCRIPT

Rs/Ω Rct/Ω

DLi/cm2 s-1

i0/A cm-2

[email protected]

8

220

1.51*10-13

4.78*10-5

52

3.03*10-13

7.22*10-5

AC C

EP

TE D

M AN U

SC

[email protected] 5

RI PT

Sample

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights



Nano-sized CoFe2O4 particles are uniformly deposited on the surface of porous



RI PT

CNTs.

The [email protected] electrode shows outstanding high capacity and good cycle stability.

The [email protected] shows excellent high-rate capability of 694 mAh g-1 at 3 A

SC



g-1.

The pore and rich functional groups on p-CNTs contribute the excellent

M AN U



AC C

EP

TE D

performance.

ACCEPTED MANUSCRIPT

Supporting information for Porous Carbon Nanotubes Decorated with Nanosized Cobalt

RI PT

Ferrite as Anode Materials for High-Performance Lithium-Ion Batteries

EP

TE D

M AN U

SC

Lingyan Wang, Linhai Zhuo, Haiyang Cheng, Chao Zhang, Fengyu Zhao*

AC C

Figure S1. Comparison of capacity at different rates for [email protected] electrode with those of CoFe2O4-based and Fe3O4-based composite anodes reported.

1