Porous carbon nanofiber webs derived from bacterial cellulose as an anode for high performance lithium ion batteries

Porous carbon nanofiber webs derived from bacterial cellulose as an anode for high performance lithium ion batteries

Accepted Manuscript Porous carbon nanofiber webs derived from bacterial cellulose as an anode for high performance lithium ion batteries Wei Wang, Yin...

4MB Sizes 2 Downloads 31 Views

Accepted Manuscript Porous carbon nanofiber webs derived from bacterial cellulose as an anode for high performance lithium ion batteries Wei Wang, Ying Sun, Bing Liu, Shuguang Wang, Minhua Cao PII: DOI: Reference:

S0008-6223(15)00332-2 http://dx.doi.org/10.1016/j.carbon.2015.04.041 CARBON 9859

To appear in:

Carbon

Received Date: Accepted Date:

3 April 2015 13 April 2015

Please cite this article as: Wang, W., Sun, Y., Liu, B., Wang, S., Cao, M., Porous carbon nanofiber webs derived from bacterial cellulose as an anode for high performance lithium ion batteries, Carbon (2015), doi: http://dx.doi.org/ 10.1016/j.carbon.2015.04.041

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.

Porous carbon nanofiber webs derived from bacterial cellulose as an anode for high performance lithium ion batteries

Wei Wang, Ying Sun, Bing Liu, Shuguang Wang, Minhua Cao* Key Laboratory of Cluster Science, Ministry of Education of China, Beijing Key Laboratory of Photoelectronic/Electrophotonic Conversion Materials, Department of Chemistry, Beijing Institute of Technology, Beijing 100081, P. R. China

* Corresponding author. Fax: +86 10 68912631; Tel: +86 10 68918468 E-mail: [email protected] (M. H. Cao) 1

Abstract Carbon nanofiber webs with a hierarchically porous structure and a large surface area were developed by KOH activation of the pyrolyzed bacterial cellulose (denoted as pBC), which was extracted from the low-cost, eco-friendly and industrially producible coconut juice. The activated-pBC (denoted as A-pBC-x, x represents the mass ratio of KOH to pBC, x = 5, 6 and 7) is composed of inter-welded homogeneous carbon nanofibers, which construct a mechanically robust three-dimensional (3D) conductive porous network. As an anode material for the lithium ion batteries (LIBs), the A-pBC-x exhibits significantly improved electrochemical performance compared to the pBC and current commercial graphite. Remarkably, the A-pBC-6 electrode can deliver a specific capacity of over 857.6 mAh g-1 after 100 cycles at 100 mA g-1 and retain high capacity of 325.38 mAh g-1 even cycled at high current density of 4000 mA g-1. The significant improvement for the lithium storage performance of the A-pBC-6 sample could be attributed to its hierarchical micropore-mesopore structure and high surface area, which can greatly enhance the contact area of the electrode-electrolyte, decrease the diffusion resistance of lithium ions, shorten the diffusion length of lithium ions and provide a solid and continuous pathway for electron transport.

2

1. Introduction

To meet the ever-increasing demands for portable electronics and electric vehicles, researchers are devoting significant attention to developing high capacity electrode materials for lithium-ion batteries (LIBs) [1]. Many novel anode materials have been explored for higher specific capacity and better cycle performance [2,3]. Among them, carbonaceous materials have been holding the predominant position and demonstrated to be the most common anode materials for LIBs. Especially, graphite, with its desirable electrochemical attributes, is the dominant commercialized electrode materials in current LIBs. First, the closely spaced voltage plateau is close to Li+/Li, which will afford the widest possible voltage window for a given cathode material [4]. Second, the minor and highly reversible dilation of the graphite lattice during the lithiation-delithiation process ensures the stability of the electrodes upon extensive cycling [5]. Moreover, the relatively low level of cycling-induced solid electrolyte interface (SEI) films in the graphite electrodes can provide superior Coulombic efficiency (CE) [6]. However, the lithium storage capacity in the graphite was greatly limited by the intercalation between graphite layers [7], which leads to a low theoretical capacity of 372 mA h g-1. Besides, the low interlayer diffusion coefficient in this intercalation reaction results in the limited rate capability [8]. These two aspects are main obstacles for achieving the prospective for industrialization. Thus, intense attentions have been paid on other carbon-based materials to improve lithium ion storage capacity. 3

To avoid the disadvantage of graphite, one of the most promising tactics is to develop carbonaceous anode materials with regular microstructure, such as carbon nanotubes, ordered mesoporous carbon, hollow nanospheres, and carbon nanosheets [9-12]. The nanostructured carbon materials could provide higher lithium storage in micropores, at defects or as a result of nanosize effects, and at the same time could shorten the transport length for lithium ion and electrons and increase the carbon materials/electrolyte contact areas, leading to the enhancement of the lithium ion and electron transport kinetics in electrodes. Thus excellent lithium storage performance can be achieved. Here, It is worth mentioning that three-dimensional (3D) interpenetrating networks are proved to have particularly high efficiency in terms of shortening the transport pathway of both electrons and lithium ion upon charging-discharging due to the excellent electrical connection between the network and the current collector, improving kinetics of lithium ion transport, and alleviating stress induced by volume changes [13]. For example, Huang et al. reported the synthesis of an interconnected 3D carbon framework via chemical activation of the polypyrrole nanofiber webs, which provided about 943 mAh g-1 even at 2 A g-1 after 600 cycles [14]. Another practical strategy to improve the electrochemical performance of the carbon-based anode is introducing additional pores, especially mesopores, to carbon frameworks. The large amount of mesopores in LIB anodes are desirable for facilitating the charge transfer at the electrode-electrolyte interface, and could offer efficient inner-pore mass transport and alleviate the stress induced by 4

volume changes. Therefore, constructing mesopores in activated carbon have been an intensively investigated method for application in LIBs and high lithiation capability and excellent cycling stability can be achieved. Depending on the size and porous structure, improved capacities of 300-1000 mAh g-1 have been achieved when employing meso-macroporous carbon fibrous mats, hollow carbon cages, and mesoporous carbon for LIB anodes [7,15,16]. However, the carbon precursors in many

reports

were

non-renewable

and

high

cost,

such

as

polypyrrole,

polyacrylonitrile and phenol. Therefore, exploring a facile, economic, and inexhaustible source to produce carbon-based nanostructured anodes with both large reversible capacity and long cycle life are highly desirable but still very challenging. Bacterial cellulose (BC), extracted from low-cost coconut juice via a microbial fermentation process, can be produced on industrial scale. The BC pellicles are composed of interconnected 3D network of nanofibers, and the pyrolyzed BC (pBC) generally holds a high conductivity and extraordinary electromechanical stability even under high stretching and bending strain [17]. Thus, it has been widely used as precursor material to prepare graphitized films, macroscopic-scale carbon-nanofiber aerogels and flexible solid-state supercapacitor [17-19]. Such a mechanically robust 3D network structure can be used as a favorable electrode specimen. Inspired by the abundance of BC and combining with its low cost and eco-friendliness, we propose to construct 3D conductive carbon nanofiber networks from BC pellicles and introduce mesopores in its carbonized structure to yield an active material for the application in 5

LIBs. Herein, taking full advantage of both the cellular structure and the high degree of cross-linking, we report a 3D activated-pBC (denoted as A-pBC-x, x represents the mass ratio of KOH to pBC, x = 5, 6 and 7) nanofiber network with good electrical conductivity, high surface area and abundant porosity through a facile tailored synthesis process of carbonization and activation. When evaluated as an anode material in LIBs, the resultant sample A-pBC-6 can deliver a reversible capacity of 857.6 mAh g-1 after 100 cycles at 100 mA g-1, and around 325.38 mAh g-1 can be still retained even at high current density of 4000 mA g-1. The high reversible capacity, superior rate performance as well as excellent cycling stability indicate their promise as an anode material for high rate LIBs.

2. Experimental Section

2.1. Preparation of the pBC aerogel The BC pellicles were provided by Ms C. Y. Zhong (Hainan Yeguo Foods Limited Company, China). BC fiber content in the pellicles was approximately 1% in volume ratio. The gel-like pellicles of BC with about 15 mm thick were rinsed several times by deionized water and soaked in deionized water for one week. After freeze-dried with the temperature and pressure of -50 °C and 20 Pa for 24 h, the frozen sample was then annealed at 700 °C for 2 h under a flowing N2 atmosphere with a heating rate of 5 °C min-1 to yield black pBC [19]. 2.2. Preparation of the activated pBC (A-pBC) aerogel 6

The black pBC aerogels were activated by KOH under various conditions by adjusting the mass ratio of KOH to pBC (5, 6 and 7). For a typical activation, 85 mg of black pBC aerogels were impregnated with KOH solution (0.425 g in 10 g ethanol), followed by an evaporation step at room temperature. The activation process was carried out in a tube furnace under flowing nitrogen with a flow rate of 80 mL min-1 at a rate of 5 °C min-1 up to 700 °C for 1 h. Then, the resultant mixture was removed with a 1 M HCl solution followed by deionized water to obtain the final A-pBC. 2.3. Characterization X-ray diffraction (XRD) was performed on a Bruker D8 (Cu-Kα radiation, λ = 1.5418 Å) from 10 to 80° (2θ) with a scanning step of 12° min-1. Raman spectra were measured using an Invia Raman spectrometer, with an excitation laser wavelength of 514.5 nm. The field emission scanning electron microscopy (FE-SEM) along with energy dispersive X-ray spectroscopy (EDS) and the element mapping were taken on Hitachi S-4800 SEM unit. The microstructures were studied by transmission electron microscopy (TEM, JEOL 2100F) in high resolution mode operating at 200 kV. X-Ray photoelectron spectroscopy (XPS) measurements were performed on an ESCALAB 250 spectrometer (Perkin-Elmer) to characterize the surface composition. The Brunauer-Emmett-Teller (BET) surface area of as-synthesized samples was measured using a Belsorp-max surface area detecting instrument by N2 adsorption-desorption at 77 K. The pore size distribution was calculated via a non-local density functional theory (NLDFT) method using nitrogen adsorption data with a slit pore model. 7

2.4. Electrochemical measurements The electrochemical measurement was performed at room temperature by using coin cells (CR2025) on LAND CT2001A with a cutoff voltage of 0.01-3.00 V versus Li+/Li. The samples obtained above were used as active materials. The working electrode was produced by coating the mixture of the active materials (pBC or A-pBC, 80 wt%), carbonaceous additive (acetylene black, 10 wt%), and polyvinylidene fluoride (10 wt%) binder in N-methylpyrrolidone (NMP) on a copper foil, which was first dried in a vacuum furnace at 80 °C for 4 h and then at 120 °C for 12 h. The lithium metal was used as both the counter and reference electrode, and a Celgard 2400 polypropylene film was used as the separator. 1 M LiPF6 dissolved in ethylene carbonate, dimethyl carbonate and diethyl carbonate mixture (1:1:1 in volume) was used as the nonaqueous electrolyte. The loading of the active materials was around 0.9-1.05 mg cm-2. Cell assembly was carried out in circulating argon glove box where both the moisture and oxygen contents were below 1 ppm. Cyclic voltammetry (CV) was performed using an electrochemical workstation (CHI660D) with a scanning rate of 0.1 mV s-1 at room temperature. A full Li-ion battery is assembled by coupling the A-pBC-6 anode with commercially available LiFePO4 cathode. The LiFePO4 electrode film was prepared by pasting a slurry of the active material (80 wt%), acetylene black (10 wt%) and PVDF (10 wt%) in NMP on aluminum foil, and drying overnight under vacuum at 120 °C. In order to match the cathode/anode capacity, there is a slight excess capacity 8

for the anode compared with the cathode, and the mass ratio of A-pBC-6 to LiFePO4 was adjusted to 1:4 (3.6-4.2 mg cm-2 for the cathode loading). Prior to full cell assembly, the A-pBC-6 electrode was pre-lithiated to reduce the initial irreversible capacity. This ex-situ lithiation process was performed by directly contacting the electrode with a lithium metal foil wetted by the electrolyte solution for 30 min. The LiFePO4 film was employed as the counter electrode and Celgard 2400 polypropylene film was used as the separator. 1 M LiPF6 dissolved in ethylene carbonate, dimethyl carbonate and diethyl carbonate mixture (1:1:1, in volume) was used as nonaqueous electrolyte. The full cell of LiFePO4/A-pBC-6 was assembled in a high purity argon filled glove box. The cells were cycled at 0.1C (17mA g-1, referred to the cathode mass) in the voltage range of 0.9-3.9 V [20]. 3. Results and Discussion 3.1. Synthesis process of A-pBC The synthesis process of A-pBC is schematically illustrated in Fig. 1, which mainly includes three steps, freeze-drying, pyrolysis, and activation. The BC, extracted from fresh coconut juice (shown in Fig. 1a), can be produced on industrial scales via the microbial fermentation process. The BC exhibits strong hydrophilic properties because of a large number of hydroxyl groups in its backbone structure (Fig. S1), and can be easily cut into any shape as necessary (Fig. 1d). The size of purified BC pellicle used in our experiment was 30×20×0.15 cm3. In a typical synthesis of pBC, we first cut the BC pellicle into rectangular or cubic shape, which was freeze-dried to 9

forrm BC C aero a ogell (F Fig. 1bb). Thhen,, th he obta o aineed whhite BC C aaeroogel was w subbjectedd too pyyrollysiis undeer N2 atm a mosppheere to t form f m bblack ppBC C prrodu uct (Fiig. 1c)). Fiinally, to furrtheer im mpro ovee thhe surffacee arrea of pB BC and a d inntroducce mes m sopporees iin thhe pBC, wee acctivated thee pBC p C ussingg KOH K H and a d the reesuultannt A-p A pBC C has h abu unddantt pooroous struuctuuree inn thhe naanoffibeers (Fiig. 1g and h). h Con C nsidderiingg that the t strructturaal ffeatuurees and a siggnifficaantlyy ennhan nceed poro p ous proopeertiees of o tthe pBC nan n nofibberrs, we w bellievve thhatt thee A-pB A BC cann be ussed ass a nooveel LIB L B ano a ode. The T deetaiiledd faabricattionn pproccess is i desscribbedd inn Exxpeerim menttal Secctioon.

Figg. 1 Thhe synthessis proc p cesss off A-pB BC. Phootoggrapph of o thhe pris p stinee cooconnut juicce (a) ( andd BC C pelliclle (d); Phootoggrapphs of the obbtainned origginaal BC B aeroogeel affter freeeze-dry yingg (bb) and pBC C a r and pyyrollysis (cc) [((e) aand (f) aree thee coorrespondinng SEM S M im magges]; (gg, h)) Scchem matiic aerrogeel after illuustrratioon of o innter--welldedd pB BC and a A-ppBC C naanoffiberrs.

3.22. Mat M terial chaaraacteerizzatiionss 100

The chemical composition and morphology of the original BC and pBC were first studied by XRD, Raman and SEM. As shown in Fig. 2a, three characteristic peaks centered at 14.78, 16.98, and 22.78 were observed in the XRD pattern of the original —

BC and they can be assigned to the typical ( 1 1 0 ), (110), and (020) planes of cellulose I, respectively [17,19]. However, after pyrolyzing the original BC precursor at 700 °C in N2 atmosphere for 2 h, these three diffraction peaks disappeared as depicted in Fig. 2b, thus indicating that the crystalline structure of original BC has been destroyed and amorphous carbon was obtained [18]. To further confirm the composition of the pyrolyzed sample, Raman spectrum was conducted. As shown in Fig. 2c, two characteristic carbon bands at 1343.3 and 1583.2 cm-1 can be clearly observed, which can be assigned to D-band (defect and disordered carbon) and G-band (graphitic carbon). SEM images of the original BC are shown in Fig. 2d and e, from which it can be seen that the original BC possesses a highly porous network structure constructed by numerous inter-twinned homogeneous nanofibers of about 20 to 30 nm in diameter. Fig. 2f shows the SEM images of pBC and clearly, the pBC maintains the same porous network structure as the BC precursor in the whole sample, indicating that the pBC has good mechanical stability during the thermal treatment process. Moreover, the small size of the pBC nanofibers will provide large surface area to facilitate the fast diffusion of electrolyte. As well known, KOH activation is widely used for tailoring the pore texture of carbon materials and that the mass ratio of KOH to the used carbon materials has an 11

important effect on the final porous properties [7,14,23]. Therefore, we here also activate the resultant pBC sample with KOH using different mass ratios of KOH to

(c) Intensity (a.u.)

(b) Intensity (a.u.)

Intensity (a.u.)

(a)

10

20

30

40

50

2θ (degree)

60

70

(d)

80

10

20

(e)

30

40

50

2θ (degree)

60

70

80

900

1200

1500

1800

-1

Wavelength (cm )

(f)

Fig. 2 (a) XRD patterns of the origin BC; (b,c) XRD pattern and Raman spectrum of the pBC; (d,e) SEM images of the origin BC at different magnifications; (f) SEM image of pBC nanofibers and the inset is a magnified image.

carbon. According to references [14,23], we adjust three mass ratios of KOH to pBC, including 5:1, 6:1, and 7:1, and the resultant samples were denoted as A-pBC-x, x = 5, 6 and 7. XRD measurement was first performed to study the structure of the prepared A-pBC-5, A-pBC-6 and A-pBC-7. As shown in Fig. 3a, two broad diffraction peaks at around 25° and 44° were observed in all samples, which correspond to the amorphous carbon [18]. The Raman spectra of the activated products are shown in Fig. 3b, which display two peaks at around 1340 and 1590 cm-1, corresponding to the well-documented D and G bands. The intensity ratio of D to G (ID/IG) suggests the graphitization degree and defects of the carbon materials [24]. Generally speaking, the higher the ratio, the lower the degree of graphitization but the larger amount of defects. The ID/IG values for pBC were 1.09. After the KOH activation, the ID/IG values 12

increased from 1.11 for A-pBC-5 to 1.13 for A-pBC-6 and then 1.28 for A-pBC-7, respectively. As can be seen, for all samples, the ID/IG values all are larger than 1.0, confirming their abundant defects and poor graphitization degree [24]. Moreover, the ID/IG values of the activated products are higher than that of pBC, indicating a decreased graphitization degree and more amount of defects in the resultant carbon nanofibers. In addition, the ID/IG value increases with the increase of the KOH/pBC molar ratio [24,25]. This indicates that the increase of the KOH amount used in the chemical activation helps to create more defects in the resultant carbon nanofibers [24]. And these effects would give rise to higher intensity of the D-band, implying that the prepared carbon materials consisted mainly of amorphous carbon [24]. This is consistent with the XRD results. However, the decrease of the graphitization degree will lead to the decrease of the conductivity of the activated carbons. Thus finely tailoring the porous microstructure and the degree of graphitization of porous carbons is crucial for KOH activation [24]. The morphology of the samples was then investigated by SEM measurements. Like the pBC, all of the activated samples still consist of nanofibers, which have uniform diameters in the range of 20-30 nm and have a large aspect ratio with the length up to tens of micrometers (Fig. 3c-e). In other words, the activated sample completely maintains the morphology of the pBC during this KOH activation process. Thus, we choose A-pBC-6 to perform the TEM measurement. As shown in Fig. 3f,g, homogeneous nanofibers of about 20 to 30 nm in diameter are observed. More 13

importantly, in most cases the thus-prepared nanofibers are welded to each other to form a 3D cross-linking structure. This welding character may provide a microscopic explanation for the outstanding mechanical and conducting properties of the pBC along with a large interfacial area [13,18,19].

10

A-pBC-7 A-pBC-6 A-pBC-5

20

30

40

50

60

70

(b)

Intensity (a.u.)

Intensity (a.u.)

(a)

A-pBC-7 A-pBC-6 A-pBC-5

900

80

1200

2θ (degree)

1500

(d)

(e)

(f)

(g)

(h)

Counts (a.u.)

Counts (a.u.)

C 1s O 1s

0

300

600

900

Binding energy (eV)

(j)

1200 280

C 1s

C-C& C=C

C-OH C=O& COOH C-O-C

282

284

286

288

290

Binding Energy (eV)

Counts (a.u.)

(c)

(i)

1800

2100

Wavelength (nm)

(k)

292 396

N 1s

Pyridinic N

Pyrrolic N

398

400

402

Binding energy (eV)

404

Fig. 3 (a) XRD patterns and (b) Raman spectra of A-pBC-x samples (x = 5, 6, and 7); (c-e) SEM images of the A-pBC-x (x = 5,6, and 7) at different magnifications; (f-h) TEM and HRTEM images of a local region of the A-pBC-6 and its Fast Fourier Transform; (i-k) XPS spectra of the typical A-pBC-6: (i) survey spectrum, (j) C 1s and (k) N 1s.

The high resolution TEM image further reveals that these nanofibers exhibited fairly homogeneous texture and only typical worm-like porous structure of graphite

14

with high curvature texture was observed (Fig. 3h). Furthermore, the selected area electron diffraction pattern of the A-pBC-6 did not show any clear diffraction rings or dots (the inset in Fig. 3h), indicating an amorphous state for the nanofibers. Furthermore, XPS has also been employed to investigate the composition and surface chemical state of the A-pBC-6 (Fig. 3i-k). The survey XPS spectrum in Fig. 3i revealed that the as-obtained sample was mainly composed of C and O. And the characteristic peak was observed at ca. 540 eV, which can be ascribed to O in the carbonyl group and various other O groups bound to the A-pBC-6 surface [1]. Fig. 3j shows the high resolution C 1s XPS spectrum and four types of carbon with different chemical states were detected. The obvious peaks appearing at 284.42, 285.14, 286.40 and 288.79 eV can be assigned to the graphite C-C/C=C, C-OH, C-O-C and C=O/COOH, respectively [26]. It should be pointed out that the nitrogen element has also been discovered in the high resolution XPS spectrum (Fig. 3k), and the strong peak at 400.41 eV can be assigned to pyrrolic N, whereas the weaker peak at 398.86 eV suggests the presence of pyridinic N [27]. However, the N component has failed to be detected in the survey spectrum, which is probably due to its low content in this sample. This can be demonstrated by CHN elemental analysis (Table S1). Besides, the EDS measurement also revealed signals of C, N and O elements, in agreement with the XPS results. Furthermore, the corresponding element mapping images proved the homogeneous distribution of C, N and O elements in the as-prepared sample. (Fig. S2,S3). With such a structure, the as-synthesized A-pBC could have potential for use 15

in LIBs since the unique microstructures are crucial in providing a short transport pathway for both electrons and lithium ions during cycling.

200

350

(a)

pBC

300

A-pBC-5

0.0

0.2 0.4 0.6 0.8 Relative pressure (P/P0)

2 -1

Sbet= 710.63 m g

100

1.0

100 0

(b)

0.0

0.2 0.4 0.6 0.8 Relative pressure (P/P0)

1.0

0.0

0.2 0.4 0.6 0.8 Relative pressure (P/P0)

30

5

0.0

10 15 20 Pore size (nm)

0.2 0.4 0.6 0.8 Relative pressure (P/P0)

25

30

1.0

(h) A-pBC-7

dvp/drp 0

0

A-pBC-6

dvp/drp 25

1.0

(f) A-pBC-5

dvp/drp

10 15 20 Pore size (nm)

2 -1

Sbet= 970.32 m g

200 100

(d) pBC

5

2 -1

Sbet= 1235.58 m g

200

0

A-pBC-7

300

300

50

(g)

400

400

150

50

0

A-pBC-6

200 2 -1

Sbet= 497.77 m g

100

0

(e)

500

250 150

500

600

(c)

dvp/drp

3 -1

Volume adsorbed (cm g )

250

0

5

10 15 20 Pore size (nm)

25

30

0

5

10 15 20 Pore size (nm)

25

30

Fig. 4 Nitrogen adsorption isotherms and pore size distribution curves of the pBC (a,b), A-pBC-5 (c,d), A-pBC-6 (e,f) and A-pBC-7 (g,h).

To further understand the porous nature of the pBC and A-pBC samples, nitrogen adsorption-desorption measurements were carried out. The nitrogen sorption isotherms and corresponding pore-size distribution curves are displayed in Fig. 4. Evidently, all the nitrogen sorption isotherms can be classified as type I according to International Union of Pure and Applied Chemistry classification, in other words they are typical microporous structure. The micropore diameter of pBC is determined to be 1.8 nm by the NLDFT method from the adsorption branch (Fig. 4b). Such a porous structure gives rise to the BET specific surface area as high as 497.77 m2 g-1. In contrast, the calculated BET specific surface areas of A-pBC-5, A-pBC-6 and A-pBC-7 are 710.63, 1235.58 and 970.32 m2 g-1, respectively (Fig. 4b-d), which are far larger than that of un-activated sample. Unlike the pore size distribution curve of pBC that exhibits one peak at 1.8 nm, the activated samples also display a large 16

number of mesopores besides micropores in their pore size distributions (Fig. 4d, f and h). Moreover, the total pore volume of pBC is 0.5754 cm3 g-1, whereas after activation, the total pore volume increases to 0.634 cm3 g-1 for A-pBC-5, 1.029 cm3 g-1 for A-pBC-6 and 0.7955 cm3 g-1 for A-pBC-7. It can be clearly observed that BET specific surface area and total pore volume were significantly enhanced by KOH activation. In addition, both the surface area and total pore volume of the products were affected significantly by the mass ratio of KOH to pBC in the activated samples. It is well demonstrated that in the activation process, the KOH could induce the formation of micropores and mesoporous in the pBC framework via etching action between KOH and carbon, thus increasing the surface area and the pore volume [23]. The results show that the surface area and pore volume of all activated samples are larger than that of pristine pBC, indicating the effectiveness of KOH in developing new pores in carbon frameworks. However, with the increase of the KOH/pBC mass ratio, on the one hand, the micropores have dramatically increased for the severe etching action by large amount of metallic potassium, attaching with the fairly larger extent of enlarging surface area and total pore volume (as depicted in Fig. 4). On the other hand, the sharp competition between the etching in the KOH activation and mesopore shrinkage make the reaction becomes harsher than that at low KOH/pBC ratio [23]. Moreover, if the KOH/pBC mass ratio is too high, inner ultra-small micropores without the connecting by mesopores are difficult for N2 adsorption, inducing a 17

decrease of surface area and pore volume. From these results we can see that the mild KOH activation treatment (KOH/carbon = 6:1) helpfully enriches the microporous structure, which could provide large surface area and pore volume. The high specific surface area will ensure a higher electrolyte-electrode contact area and a large number of

active

sites

for

charge-transfer

reactions,

while

the

hierarchical

micropore-mesopore structure could significantly decrease the mass transport resistance. With such a porous structure, the A-pBC-6 was expected to exhibit enhanced lithium ion storage capacity and improved rate capability. 3.3. Electrochemical characterizations In view of its unique composition, the resultant A-pBC-6 is expected to have potential application as an anode material in LIBs. Herein, we evaluate the electrochemical properties of A-pBC-6 by galvanostatic charge-discharge cycling over the voltage range 0.01-3.0 V vs. Li+/Li. For comparison, the un-activated sample pBC, A-pBC-5 and A-pBC-7 were selected as references. Fig. 5 shows the discharge and charge curves of all the products for the 1st, 5th, 10th, 30th, 50th and 100th cycles at a current density of 100 mA g-1, respectively. All the profiles gave a voltage plateau at around 0.7 V at the first lithiation process, which can be assigned to the electrolyte decomposition and SEI [7]. For the A-pBC-6 electrode (Fig. 5a), the initial discharge-charge delivers specific capacities of 1948.5 and 1091.4 mAh g-1, respectively, with an initial CE of 56% based on the total mass of the A-pBC-6. The

18

initial capacity loss may result from the formation of SEI layer on the electrode surface and irreversible lithium ion insertion into the carbon network [8]. Despite the

3.5

3.0

Charge

1 st 5 th 10th 30 th 50 th 100 th

2.5 2.0 A-pBC-6

1.5

Voltage (V vs. Li+/Li)

(a)

1.0 0.5 0.0

Discharge 0

1000

1500

2000

(c)

3.0

Charge

1 st 5 th 10th 30 th 50 th 100 th

2.5 2.0 A-pBC-5

1.5 1.0 0.5 0.0

Discharge 0

400

800

1200

-1

1600

Specific capacity (mAh g )

Charge

1 st 5 th 10th 30 th 50 th 100 th

2.5 2.0 pBC

1.5 1.0 0.5

3.5

3.5

Voltage (V vs. Li+/Li)

500

Specific capacity (mAh g-1)

(b)

3.0

0.0

Voltage (V vs. Li+/Li)

Voltage (V vs. Li+/Li)

3.5

3.0

Discharge 0

300

600

900

Specific capacity (mAh g-1)

(d)

Charge

1 st 5 th 10th 30 th 50 th 100 th

2.5 2.0 A-pBC-7

1.5

1200

1.0 0.5 0.0

Discharge 0

400

800

1200

Specific capacity (mAh g-1)

1600

Fig. 5 Voltage profiles of as-prepared samples at a current density of 100 mA g−1: (a) A-pBC-6, (b) pBC, (c) A-pBC-5 and (d) A-pBC-7.

large irreversible capacity loss in the first cycle, a perfect reversible capacity was obtained with an average CE of higher than 98% for up to 100 cycles after the second cycle. In contrast, for the pBC electrode (Fig. 5b), the initial discharge and charge capacities are only 1042.6 and 464.3 mAh g-1. Meanwhile, A-pBC-5 exhibits initial discharge and charge capacities of 1392.6 and 625.6 mAh g-1 (Fig. 5c) and A-pBC-7 shows initial discharge and charge capacities of 1402.1 and 625.8 mAh g-1 (Fig. 5d). Clearly, the A-pBC-6 sample delivered a superior reversible capacity compared with 19

other samples, which might be associated with its unique hierarchical structure and high surface area. 1500

Capacity (mAh g-1 )

1200 900

Capacity (mAh g-1)

A-pBC-6 A-pBC-7 A-pBC-5 pBC

(a)

600 300 0

0

20

60

80

(c)

100 mAg

900

300 600

600 300 0

0 0.2

100

Cycle number (n)

1500 1200

40

A-pBC-6 A-pBC-7 A-pBC-5 100 pBC

(b) 1200 100 mAg-1

15

1000 2000 40008000 10000

30

45

300 mAg

-1 -1

1000 mAg

900

-1

2000 mAg 300

4000 mAg 0

20

40

75

90

0.0 1 st 2 nd 3 rd 4 th

-0.2

600

0

60

Cycle number (n)

(d)

-1

Current (mA)

Capacity (mAh g-1)

1500

60

80

Cycle number (n)

-0.4

-1

100

-0.6

0.0

0.5

1.0

1.5

2.0 + 2.5

Voltage (V vs. Li /Li)

3.0

Fig. 6 (a) Cycling performance and (b) Rate capability behavior of pBC and A-pBC-x at a current density of 100 mA g−1, (c) Cycling performance of the A-pBC-6 at 100, 300, 1000, 2000 and 4000 mA g-1, (d) CV curves of the A-pBC-6 electrodes in the range of 0.01-3.00 V at the scan rate of 0.1 mV s-1.

The cycling performance of A-pBC-6 sample is shown in Fig. 6a. For comparison, those other three samples are also provided. At a constant current density of 100 mA g-1, A-pBC-6 outperforms the other three samples in cycling stability and a high capacity of 857.6 mAh g-1 can be maintained even after 100 cycles. In sharp contrast, pBC shows a much inferior cycling performance. Although it delivers a high initial discharge capacity (1042.6 mAh g-1), the capacity decreases to 440.4 mAh g-1 after 100 cycles, which is half of the value for A-pBC-6. The cycling stability for A-pBC-5

20

and A-pBC-7 is somewhat better than that of pBC, but still inferior to that of A-pBC-6. They deliver an initial discharge capacity of 1392.6 and1402.1 mAh g-1, and only 547.1 and 654.5 mAh g

-1

by 100th cycle can be maintained. Considering that all

samples have almost the same 3D network structure, the difference in capacity and stability contributed from the carbon network should be negligible. Thus, the difference in these two aspects should come from the hierarchical structure and high surface area rather than the carbon 3D networks. Activated sample A-pBC-6 is expected to display good rate performance under a large current intensity. As shown in Fig. 6b, as expected, a reversible capacity of 1067.78, 769.16, 607.54, 505.27, 435,66, 334.42 and 261.3 mAh g-1 can be obtained at 0.1, 0.3, 0.6, 1.0, 2.0, 4.0, and 8.0 A g-1, respectively, owing to the good electrical conductivity of carbon matrix. Even at high rate up to 10.0 A g-1, a capacity of 249.87 mAh g-1 can still be obtained. Meanwhile, the capacity of about 975.62 mAh g-1 can be retained when the rate is reduced back to 0.1 A g-1 after 80 cycles, indicating its good reversibility and stability. Moreover, the A-pBC-6 electrode shows a much enhanced rate performance compared to those of the A-pBC-5, A-pBC-7 and pBC at various current densities, as shown in Fig. 6b. The cycling stability of the A-pBC-6 electrode at higher current densities was futher investigated in Fig. 6c. The recorded discharge capacities were 913.86 (100 mA g-1), 712.51 mAh g-1 (300 mA g-1), 549.15 mAh g-1 (1000 mA g-1) and 413.61 mAh g-1 (2000 mA g-1) even after 100 cycles, respectively. All of capacities are much higher than the theoretical capacity of graphite. 21

Furthermore, a capacity of 325.38 mAh g-1 can still be obtained when the current density rate was up to 4000 mA g-1 after 100 cycles. Clearly, the A-pBC-6 anode might be suitable for very high power density. In order to fully understand the electrochemical performance of A-pBC-6 as anode material for LIBs, CV was then performed to clearly distinguish the electrode reaction processes in the range of 0.01-3.0 V at a sweep rate of 0.1 mV s-1. As shown in Fig. 6d, a peak at about 0.7 V is observed only during the first discharge, in accordance with the formation of SEI layer [7,9], which is consistent with the behavior of the charge-discharge profiles (Fig. 5). After the first cycle, the CV profiles are overlapped well, indicating high stability and reversibility of the as-synthesized A-pBC-6 for lithium-ion insertion and extraction.

(a)

(b)

(c)

(d)

Fig. 7 (a) Schematic illustration of electron transfer and lithium-ion storage in the A-pBC-6. (b-d) SEM images of A-pBC-6 electrode after 100 charge-discharge cycles at different magnifications.

The high capacity and super rate capability for A-pBC-6 can be explained with its novel hierarchical structure, high surface area and good mechanical strength. First, as 22

illustrated in Fig. 7a, a large number of hierarchical pores were introduced in pBC during the KOH activation process, which not only provide high surface area but also act as reservoirs for storage of lithium-ion. On the one hand, the high surface area of the A-pBC-6 leads to sufficient electrode-electrolyte interface to absorb lithium ion and promote rapid charge-transfer reaction. On the other hand, the nanosized fibers can reduce the transport length of lithium ion, while the nanopores on their surface can supply facile transport channels for lithium ion. Moreover, the interpenetrating carbon webs provide a solid and continuous pathway for electron transport [10,14]. Second, the interconnected carbon network could effectively accommodate the strains caused by the volume change during cycling and maintain the structure to provide short electron and ion transfer paths upon long-term cycling. These deductions got solid support by the SEM measurements of the electrode after 100 discharge-charge cycles. As shown in Fig. 7b-d, the continuous and interconnected structure of carbon frameworks maintained well and no significant pulverization and fracture were observed. Notably, the rough surface of the A-pBC-6 electrode after cycles might result from the formation of SEI [28], and the small lumps of particles observed are super P carbon additive during electrode fabrication [29]. To further demonstrate practical application of the as-prepared porous fibers in LIBs, a full Li-ion battery was investigated by coupling the A-pBC-6 anode with a commercially available LiFePO4 cathode. According to Scrosati’s report [20], in assembling a full battery, the 1:1 ratio of the cell capacity balance of A-pBC-6 anode 23

and LiFePO4 cathode with a slight excess of anode capacity is the optimization of the cell itself. To realize this optimal capacity balance, we first obained the specific capacity ratio of A-pBC-6 anode to LiFePO4 cathode based on their respective half Li-ion battery tests at the same cycling rate (628.8 mAh g-1 for A-pBC-6 at 1C, 1C = 372 mA g-1; 144.7 mAh g-1 for LiFePO4 at 1C, 1C = 170 mA g-1), which was determined to be 4.3:1. According to above principle of capacity balance, the mass ratio of A-pBC-6 to LiFePO4 thus was determined to be about 1:4 and the loadings of A-pBC-6 and LiFePO4 in our experiment are 0.9-1.05 and 3.6-4.2 mg cm-2, respectively. Fig. 8a shows the full-cell voltage profiles after 1, 5, 10, 20, 30 and 40 cycles with the voltage window of 0.9-3.9 V at 0.1C. The initial discharge capacity was 159.3 mAh g-1 based on the cathode mass, with an initial CE of 85% (this initial irreversibility is due to SEI film formation at the LiFePO4 cathode side) [20]. After the first cycle, the CE can maintain at about 98.5% and a high overlapping in the potential trends indicated good operation reversibility. Fig. 8b presents the cycle performance and CE of the as-developed full cell. It is evident that, the LiFePO4/A-pBC-6 full cell holds a stable discharge capacity of 122.4 mAh g−1 after 40 cycles. Moreover, LiFePO4/A-pBC-6 full cell has good capacity retention of 77.5%, and the CE increases to over 99% and maintains for the measured 40 cycles. From these results, we can see that the LiFePO4/A-pBC-6 full cell exhibits a better performance in terms of specific capacity, cyclability, and Coulombic efficiency, thus confirming the potentiality of A-pBC-6 as an anode materials for LIBs. 24

3.5

Charge 1 st 5 th 10th 20 th 30 th 40 th

3.0 2.5 2.0 1.5

Discharge

1.0 0.5

0

50

100

150 -1

Specific capacity (mAhg )

250

120

(b)

100

200

80

150

60 100

Charge 40 Discharge 20

50 0

200

0

Coulombic efficiency (%)

(a)

Capacity (mA h g-1)

Voltage (V vs. Li+/Li)

4.0

10

20

30

Cycle number (n)

0 40

Fig. 8 (a) Voltage profiles of LiFePO4/A-pBC-6 full cell for 1, 5, 10, 20, 30 and 40 cycles; (b) Corresponding specific capacity versus cycle number of the battery (left) and CE (right). Cycling rate 0.1C (1C=170 mA g−1 vs LiFePO4).

4. Conclusions In summary, versatile 3D conductive porous carbon nanofiber networks, derived from a low-cost, eco-friendly and scalable biomaterial (coconut juice), have been successfully fabricated via chemical activation of the pBC nanofibers by KOH. The product (A-pBC-6) possesses a hierarchically porous structure with a unique interconnected 3D framework and its surface area is as high as 1235.58 m2 g-1. As an anode material for LIBs, its specific capacity is as high as 857.6 mAh g-1 at 100 A g-1 after 100 cycles and it also exhibits super rate capacity even at a high rate of 10 A g-1. The remarkably enhanced electrochemical performance can be attributed to its porous structural features, which can enlarge the active materials/electrolyte contact area, shorten the lithium ion diffusion length, ensure fast mass transport for lithium ion, and accommodate volume expansion during lithiation reaction. Furthermore, the LiCoO2/A-pBC-6 full cell demonstrated good cycle performance and relative large capacity. Thus A-pBC-6 nanofibers show promising properties as an anode alternative 25

for commercial graphite for LIBs.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21471016 and 21271023) and the 111 Project (B07012).

References

[1] Gao Q, Zhao XY, Xiao Y, Zhao D, Cao MH. A mild route to mesoporous Mo2C-C hybrid nanospheres for high performance lithium-ion batteries. Nanoscale 2014;6(11):6151-7. [2] Wang C, Ju J, Yang YQ, Tang YF, Lin JH, Shi ZJ, et al. In situ grown graphene-encapsulated germanium nanowires for superior lithium-ion storage properties. J Mater Chem A 2013;1(31):8897-902. [3] Song T, Jeon Y, Samal M, Han H, Park H, Ha JW, et al. A Ge inverse opal with porous walls as an anode for lithium ion batteries. Energy Environ Sci 2012;5(10):9028-33. [4] Peled E, Menachem C, Bar Tow D, Melman A. Improved graphite anode for lithium-ion batteries chemically bonded solid electrolyte interface and nanochannel formation. J Electrochem Soc 1996;143(1):L4-L7. [5] Marcinek M, Hardwick LJ, Richardson TJ, Song X, Kostecki R. Microwave plasma chemical vapor deposition of nano-structured Sn/C composite thin-film 26

anodes for Li-ion batteries. J Power Sources 2007;173(2):965-71. [6] Ding J, Wang H, Li Z, Kohandehghan A, Cui K, Xu Z, et al. Carbon Nanosheet frameworks derived from peat moss as high performance sodium ion battery anodes. ACS Nano 2013;7(12):11004-15. [7] Zhou GM, Wang DW, Shan XY, Li N, Li F, Cheng HM. Hollow carbon cage with nanocapsules of graphitic shell/nickel core as an anode material for high rate lithium ion batteries. J Mater Chem 2012;22(22):11252-6. [8] Yang

SB,

Feng

XL,

Zhi

LJ,

Cao

Q,

Maier

J,

Müllen

K.

Nanographene-constructed hollow carbon spheres and their favorable electroactivity with respect to lithium storage. Adv Mater 2010;22(7):838-42. [9] Li SS, Luo YH, Lv W, Yu WJ, Wu S, Hou PX, et al. Vertically aligned carbon nanotubes grown on graphene paper as electrodes in lithium-ion batteries and dye-sensitized solar cells. Adv Energy Mater 2011;1(4):486-90. [10] Guo BK, Wang XQ, Fulvio PF, Chi M, Mahurin SM, Sun XG, et al. Soft-templated mesoporous carbon-carbon nanotube composites for high performance lithium-ion batteries. Adv Mater 2011;23(40):4661-6. [11] Han FD, Bai YJ, Liu R, Yao B, Qi YX, Lun N, et al. Template-free synthesis of interconnected hollow carbon nanospheres for high-performance anode material in lithium-ion batteries. Adv Energy Mater 2011;1(5):798-801. [12] Fang Y, Lv YY, Che RC, Wu HY, Zhang XH, Gu D, et al. Two-dimensional mesoporous carbon nanosheets and their derived graphene nanosheets: synthesis and 27

efficient lithium ion storage. J Am Chem Soc 2013;135(4):1524-30. [13] Wang B, Li XL, Luo B, Yang JX, Wang XJ, Song Q, et al. Pyrolyzed bacterial cellulose: a versatile support for lithium ion battery anode materials. Small 2013;9(14):2399-404. [14] Qie L, Chen WM, Wang ZH, Shao QG, Li X, Yuan LX, et al. Nitrogen-doped porous carbon nanofiber webs as anodes for lithium ion batteries with a super high capacity and rate capability. Adv Mater 2012;24(15):2047-50. [15] Wang SX, Yang LP, Stubbs LP, Li X, He CB. Lignin-derived fused electrospun carbon fibrous mats as high performance anode materials for lithium ion batteries. ACS Appl Mater Interfaces 2013;5(23):12275-82. [16] Li Z, Xu ZW, Tan XH, Wang HL, Holt CM, Stephenson T, et al. Mesoporous nitrogen-rich carbons derived from protein for ultra-high capacity battery anodes and supercapacitors. Energy Environ Sci 2013;6(3):871-8. [17] Liang HW, Guan QF, Song LT, Yao HB, Lei X, Yu SH. Highly conductive and stretchable conductors fabricated from bacterial cellulose. NPG Asia Mater 2012;4(6):e191-e6. [18] Chen LF, Huang ZH, Liang HW, Yao WT, Yu ZY, Yu SH. Flexible all-solid-state high-power supercapacitor fabricated with nitrogen-doped carbon nanofiber electrode material derived from bacterial cellulose. Energy Environ Sci 2013;6(11):3331-8. [19] Wu ZY, Li C, Liang HW, Chen JF, Yu SH. Ultralight, flexible, and fire-resistant carbon nanofiber aerogels from bacterial cellulose. Angew Chem Int Ed 28

2013;52(10):2925-9. [20] Hassoun J, Bonaccorso F, Agostini M, Angelucci M, Betti MG, Cingolani R, et al. An Advanced Lithium-Ion Battery Based on a Graphene Anode and a Lithium Iron Phosphate Cathode. Nano Lett 2014;14(8):4901-6. [21] Lv YY, Zhang F, Dou YQ, Zhai YP, Wang JX, Liu HJ, et al. A comprehensive study on KOH activation of ordered mesoporous carbons and their supercapacitor application. J Mater Chem 2012;22(1):93-9. [22] Zhou M, Pu F, Wang Z, Guan SY. Nitrogen-doped porous carbons through KOH activation with superior performance in supercapacitors. Carbon 2014;68:185-94. [23] Cai JJ, Qi JB, Yang CP, Zhao XB. Poly(vinylidene chloride)-Based Carbon with Ultrahigh Microporosity and Outstanding Performance for CH4 and H2 Storage and CO2 Capture. ACS Appl Mater Interfaces 2014;6(5):3703-11. [24] Lin ZY, Waller G, Liu Y, Liu ML, Wong CP. Facile Synthesis of Nitrogen-doped graphene via pyrolysis of graphene oxide and urea, and its electrocatalytic activity toward the oxygen-reduction reaction. Adv Energy Mater 2012;2(7):884-8. [25] Zou F, Hu XL, Sun YM, Luo W, Xia FF, Qie L, et al. Microwave-Induced In Situ Synthesis

of

Zn2GeO4/N-Doped

Graphene

Nanocomposites

and

Their

Lithium-Storage Properties. Chem Eur J 2013;19(19):6027-33. [26] Hu CG, Xiao Y, Zhao Y, Chen N, Zhang ZP, Cao MH, et al. Highly nitrogen-doped carbon capsules: scalable preparation and high-performance applications in fuel cells and lithium ion batteries. Nanoscale 2013;5(7):2726-33. 29

[27] Cherian CT, Sundaramurthy J, Reddy MV, Suresh Kumar P, Mani K, Pliszka D, et al. Morphologically robust NiFe2O4 nanofibers as high capacity Li-ion battery anode material. ACS Appl Mater Interfaces 2013;5(20):9957-63.

30