Development of free standing anodes of high aspect ratio carbon materials for rechargeable Li-ion batteries

Development of free standing anodes of high aspect ratio carbon materials for rechargeable Li-ion batteries

Accepted Manuscript Title: Development of free standing anodes of high aspect ratio carbon materials for rechargeable Li-ion batteries Authors: Priyan...

1MB Sizes 1 Downloads 31 Views

Accepted Manuscript Title: Development of free standing anodes of high aspect ratio carbon materials for rechargeable Li-ion batteries Authors: Priyanka H. Maheshwari, C. Nithya, Shilpa Jain, R.B. Mathur PII: DOI: Reference:

S0013-4686(13)00050-9 doi:10.1016/j.electacta.2013.01.031 EA 19838

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

18-10-2012 3-1-2013 8-1-2013

Please cite this article as: P.H. Maheshwari, C. Nithya, S. Jain, R.B. Mathur, Development of free standing anodes of high aspect ratio carbon materials for rechargeable Li-ion batteries., Electrochimica Acta (2010), doi:10.1016/j.electacta.2013.01.031 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.

Development of free standing anodes of high aspect ratio carbon materials for rechargeable Li-ion batteries. Priyanka H. Maheshwari1*, C. Nithya2, Shilpa Jain1, R. B. Mathur1. 1

ip t

CSIR–Network Institutes for Solar Energy and CSIR– National Physical Laboratory, New Delhi, India. 2 CSIR–NISE and CSIR– Central Electrochemical Research Institute, Karaikudi, Tamil Nadu, India

Abstract

cr

Carbon materials of various types have been extensively used as negative electrode

us

materials for rechargeable Li-ion batteries because of their consistent performance and potentialities. High aspect ratio (>1000) carbons like carbon fibers and multiwalled

an

carbon nanotubes (MWCNTs) of different dimensions have been employed to fabricate free standing anode materials. Various characterization techniques like SEM, TEM, TGA,

M

XRD, Raman spectroscopy, mercury intrusion porosimetry has been carried out to evaluate the structure of the anode that was further correlated to its performance in Li-ion

d

cell. MWCNTs prepared under specified conditions not only exhibits high purity and

te

crystallinity in structure but also shows exceptional electrochemical behavior of

Ac ce p

increasing capacity with successive cycling. This is probably due to the formation of a very constructive SEI with negligible charge transfer resistance as shown by the Nyquist plots.

Keywords

Carbon fiber; Electrode; multiwalled carbon nanotubes; solid-electrolyte interface; specific capacity.

*Corrosponding author E-mail: [email protected]; Tel: +91 11 45608508; fax: +91 11 45609310 1

Page 1 of 40

1. Introduction With ever increasing demand of energy and depleting energy sources, rechargeable Liion batteries are emerging as promising candidates to power electronic devices and

ip t

electric vehicles. Because of the safety and life-cycles issues with the use of elemental lithium and lithium-metal composites [1-5], lithium-carbon alloys are being used as

cr

battery anodes. Various forms of carbon materials like natural graphite [6-12], hard

us

carbons from various polymer precursors [13-16], petroleum coke [17-20], mesocarbon microbeads (MCMB) [21-25] have been studied extensively for improved battery

an

performance in terms of reversible capacity, cycle ability and rate capacities. In the present study we have compared the properties of free standing anode developed with

M

high aspect ratio (>1000) carbon materials (like carbon fibers and multiwalled carbon

d

nanotubes) and their corresponding electrochemical performance. The high aspect ratio

te

of the materials not only favors the fabrication of strong and flexible self supporting electrode but also reduces the doping levels required to achieve the percolation threshold

Ac ce p

to sustain connective pathways for electrons to move. Extraordinary strength and resilience further safeguards the cracking of electrode during operation and vibration environments resulting in increased battery lifetime and extended cycling. Carbon fibers obtained from different precursor materials have been studied (in powdered form) by various groups [26-29]. Among the new materials studied, carbon nanotubes (CNTs) are of special interest due to their unique structure and properties. The high thermal conductivity helps in effective heat dissipation thus enhancing the safety of the electrode and the battery. Multiwalled carbon nanotubes combine structural features of graphite and fullerenes and open up different ways of interaction with foreign atoms and

2

Page 2 of 40

molecules; specifically intercalation between the graphene layers [30], insertion inside the central tube [31] and appending on the surface. The self supported anode has an added advantage that it can be used as electrodes without binder (unlike other

ip t

carbonaceous powder materials) this helps us to elucidate precise electrochemical properties and surface area of the carbonaceous materials themselves. Moreover

cr

electrodes consisting of nano particles can possess high rate capabilities because of short

us

diffusion lengths.

The electrochemical performance of carbon nanotubes strongly depends on their structure,

an

morphology, disorder and purity. Since CNT is a recent discovery, methods of CNT production with desired structure (an important factor determining its electrochemical

M

performance) is not yet established. This has resulted in large variation in the reported

d

capacities. A lithium storage capacity of 950 mAh/g was obtained from MWCNT

te

produced by catalytic chemical vapor deposition (CVD) of camphor using nanocrystalline iron as the catalyst [32]. MWCNT obtained by catalytic decomposition of

Ac ce p

acetylene at 900°C over silica supported cobalt also shows a capacity 952 mAh/g in the first cycle that reduces to 273 mAh/g just after 5 cycles due to large irreversible capacities [33]. The MWCNTs prepared by catalytic decomposition of ferrocene and xylene shows a reversible capacity of ~ 200 mAh/g [34]. Camphor based CNTs using Fe catalyst shows discharge capacity of 175 mAh/g owing to more graphitic structure as compared to arc-made and boron doped CNTs showing a capacity of 114 and 98 mAh/g resp. The capacity was however reported for only 5 cycles [35]. All the above and other studies show varying capacities of MWCNTs based electrodes but the performance are reported for limited no. of cycles [36-38]. We report synthesis of MWCNT and

3

Page 3 of 40

corresponding development of self supporting anode with consistent performance for 100 cycles which in turn is attributed to its dimensions, crystallinity, purity and alignment.

ip t

2. Experimental 2.1. Materials

cr

Three different types of high aspect ratio (AR) carbon materials with highly varying

us

dimensions were used to make the free standing anode materials for Li-ion batteries, i.e. (i) commercial grade Nanocyl 7000 multiwalled carbon nanotubes (CNT-n) with

an

diameter 20-30 nm and AR >1000, (ii) multiwalled CNTs synthesized in laboratory using chemical vapor deposition (CVD) technique (CNT-q) with diameter 80-120 nm and AR

M

>5000, and (iii) PAN based T-300 grade carbon fiber with diameter 0.7 µm and AR

d

~10000.

te

MWCNTs were synthesized in the laboratory by CVD technique. The experimental set up consisted of multiple quartz reactors placed inside a specially designed steel reactor

Ac ce p

(of length 110 cm and diameter 10 cm) which in turn was kept inside a single zone split furnace. The effective reaction zone of the furnace was 50 cm. Ferrocene was used as a precursor for Fe catalyst to initiate the growth of carbon nanotubes. A solution of ferrocene (8% by weight) in toluene was injected in the furnace along with a steady flow of nitrogen (0.5 lit/min), with a feed rate of 10 ml/hour for 3 hours. The furnace temperature was maintained at 750°C. MWCNTs produced on the walls of the quartz reactor were scraped off as highly aligned bundles as shown in the SEM image (fig. 1). The TEM image of the samples (fig. 2) show long length curved tubes with almost uniform diameter and open ends (inset of fig. 2). The MWCNTs produced were further

4

Page 4 of 40

purified by heating the sample to 2000°C in an inert atmosphere. The process not only removes most of the catalytic impurity but also improves the crystallinity of the material

CNT sample so prepared has been referred to as CNT-q.

ip t

by annealing away the defects and improves the thermal stability of the sample [39] . The

The TEM image of the CNT-n sample however shows closed end and open ended

2.2. Preparation of free standing carbon anodes

us

cr

tubes as is clear from fig. 3

an

The anodes of MWCNT were prepared by bucky paper route. MWCNTs (CNT-q and CNT-n) were dispersed in isopropanol by ultra-sonication. The well dispersed tubes were

M

vacuum filtered through a filter paper to form an optically opaque MWCNT preforms

d

that was peeled off the support after drying. The samples so formed were found to be

te

mechanically robust and flexible. The SEM images of the anodes prepared using CNT-n and CNT-q are shown in fig. 4 and 5 and designated as samples C1 and C2 respectively.

Ac ce p

Polyacrylonitrile (PAN) based T-300 grade carbon fiber (density 1.76 g/cc) was used for making carbon fiber preforms, by the well known paper making technique. Carbon fibers having a diameter of nearly 0.7 µm and chopped in length of 0.7 cm were dispersed in an aqueous media with the help of suitable surfactants by using high speed mechanical stirrer. The slurry containing well dispersed carbon fibers was vacuum filtered and dried to obtain the carbon fiber preform as shown in fig. 6 and will be designated as sample C3 in the following text.

2.3. Characterization

5

Page 5 of 40

The structural details of the MWCNT samples were studied with the help of high resolution transmission electron microscopy (HRTEM). The MWCNTs were dispersed uniformly in dimethyl formamide and the solution was drop casted on the carbon coated

ip t

copper grid and allowed to dry. The samples were characterized using Tecnai G2 F30 STwin instrument. Raman spectroscopy was carried out using a Renishaw InVia Reflex

cr

Micro Raman Spectrometer equipped with the CCD detector at room temperature and in

us

air. Green laser (excitation line 514 nm) was used to excite the samples. One scan per sample was recorded wherein the samples were exposed to the laser power of 25 mW for

an

10 sec. The analysis of the carbon fiber preform was however carried out with red laser (excitation line 785 nm at a laser power of 25 mW). The X-ray diffraction examination of

M

the samples was performed on Rikagu powder X-ray diffractometer model: XRG 2KW

d

using Cu Kα radiation. The mean crystallite size and lattice parameters was calculated

te

from line broadening and d spacing measurements using the (002) and (100) reflections. Thermal gravimetric analysis of the electrode samples was carried out on TGA/DSC

Ac ce p

1600 by Mettler Todedo. The experiments were carried out in air at the rate of 10°C/min. The connected pore volume was determined using mercury porosimetry analyzer (model: Poremaster (33/60), P/N 05060) obtained from Quantachrome instruments, USA. In this method, mercury (Hg) with its very high surface tension (4.80 x 10-5 J cm-2) is forced into the pores of the sample. The amount of Hg uptake as a function of pressure allows one to calculate the various parameters related to the porous network. The free standing samples prepared as anode (with diameter 18 mm) were dried in vacuum oven and transferred into the gloves box and placed into the half cell consisting of lithium foil as a counter electrode, a separator (polypropylene film) and organic

6

Page 6 of 40

electrolyte (1M LiPF6 in 1:1 ratio of EC+DEC). The weight of the three electrodes C1, C2 and C3 is 0.0018 g, 0.0015 g and 0.0064 grams respectively. The cell was allowed to age for 24 hours. Galvanostatic charge – discharge was carried out at current densities of

ip t

37.2 mA/g and 74.4 mA/g. Electrochemical Impedance Spectra were measured by using an EG & G instruments Model 5210 with an AC voltage signal of 5 mV and the

us

cr

frequency range was between 100 KHz and 5 mHz.

3. Results and Discussion

an

The XRD pattern of the three samples is shown in fig. 7. A (002) peak is observed for samples C1 and C3 corresponding to an interlayer spacing of 0.3462 nm and 0.3545 nm

M

respectively, much higher than that of graphite. The (002) peak of sample C3 is quite

d

broad which signifies the disordered/ polycrystalline nature of the material. Sample C2 is

te

highly graphitic with interlayer spacing close to that of graphite and long range stacking order of the hexagonal planes (as shown by the Lc values) as compared to the other two

Ac ce p

samples as shown by the data in table 1. The long range ordering is also confirmed by the appearance of peaks at 2Θ = 77.56° and 81.28° corresponding to (110) and (112) planes respectively indicating the onset of 3D inter planar site correlation of the CNT structure. Raman spectroscopy is another useful tool which has been widely used to characterize the quality and type of carbon materials. Fig. 8 shows the Raman spectra of the samples C1, C2 and C3. Sample C3 did not show any peak with the green laser. The Raman spectrum of C3 was therefore repeated with red laser. Three bands: D (defect/ disorder induced) band, G (graphite like) band, and 2D (second order harmonic to D) band have been identified. The intensities of the bands were determined as the area under the

7

Page 7 of 40

spectral curve. The intensity of the G-band (IG) depends on the structure of the material, incident laser beam intensity, detector characteristics and other parameters. It is therefore used as a reference in determining the relative intensities of the D band (ID) and 2D band

ip t

(I2D).

As shown in table 2, the ratio ID /IG is least for sample C2. This signifies long range

cr

ordering of the hexagonal lattice of carbon atoms owing to its purity level and lesser

us

structural defects [40]. The large value of ID /IG for sample C1 (as compared to C2) despite almost similar values of crystallite size (La) (as shown by the XRD data) is

an

possibly due to the large number of pentagons and heptagons that form the closed ends of the small sized CNT-n tubes. The 2D band like the G band of the sample is a measure of

M

the crystallinity of sample and its atomic arrangement. There is almost three times

d

increase in the relative intensity of the 2D band (i.e. I2D/IG ratio) of sample C2 over C1.

te

This indicates long range stacking order in sample C2 [41]. Sample C3 do not show the 2D peak at all. This confirms the observation by the XRD data of the samples.

Ac ce p

The full width at half maximum (FWHM) of the Raman peak is also a criterion in evaluating the structural disorder of CNTs, which arises from bond angle and bond length distortions along with other impurities [42]. The FWHM of the D band, G band and 2D band for samples are shown in table 2. The lower FWHM values of the sample C2 are also attributed to the decrease in defects and improvement in its graphitic ordering. The TGA and its first derivative (DTG) curves of the electrode samples are compared in fig. 9. The TGA curve of sample C1 and C2 registers a weight loss of 88.23 % and 92.9 % respectively. The residual content is the catalyst used for CNT growth, i.e. cobalt and nickel in case of sample C1 (as analyzed by the XRD of the residue) and iron or its

8

Page 8 of 40

oxide in case of sample C2. There is no left over in case of sample C3. However the curve shows other small peaks in low temperature range which can be attributed to the loss of either the functional groups or the sizing on the fiber surface.

ip t

As shown by the DTG curves (shown by dashed lines), the decomposition of C1 takes place at around 610°C whilst that of sample C2 and C3 takes place at nearly 780°C and

cr

725°C respectively. The increase in the stability can be attributed to the improved

us

structure and crystallinity of the sample C2.

In order to find the porosity and related parameters of the electrode, the carbonaceous

an

preforms were subjected to mercury intrusion under the pressure range 0 – 6E+04 psi. Table 3 shows the effective porosity of the samples. The table also illustrates the pore

M

tortuosity and the exposed surface area of the different samples as provided by the Hg

d

intrusion studies. The percentage porosity of the anode samples increases with increasing

te

dimensions of the carbonaceous material. Thus smaller the size, larger is the packing density. The percentage intra-particle porosity is however higher for sample C1 (nearly

Ac ce p

50% of the total porosity) and zero for sample C3. The data suggests strongly tangled structure of CNT-n due to the wander wall forces highly effective at nano-scale and also some open ended MWCNTs as is also clear from the inset of TEM fig. 2 and fig. 3. The pore size distribution in the different samples is shown in fig. 10, where the pore radius computed is for hypothetical cylindrical pores. Sample C1 shows non uniformity in the pores with peaks appearing at 100 µm and 0.2 µm. sample C2 shows a peak at 10 µm. Carbon fiber preform has a very broad distribution with all the pores laying in the range of more that 50 µm.

9

Page 9 of 40

3.1. Charge discharge characteristics of the carbonaceous anodes The charge – discharge characteristics of the three samples were initially carried out at a current density of 37.2 mA/g. Fig. 11 shows the charge-discharge characteristics of the

ip t

sample C1. As the reaction starts, there is a drop in potential so as to overcome the activation energy barrier. At 2.0 volts a potential plateau is observed which is probably

cr

due some contamination or the conversion reactions as a result of alloying reactions of Li

us

with Fe & Co/ oxide. The steps appearing at 0.8 V is probably due to the lithium consumption in the formation of SEI in the first cycle. The first cycle shows high charge

an

capacity of 1740 mAh/g and also a very high irreversible capacity of 1587 mAh/g. The high irreversibility is not only due to the high surface area of the sample (that increases Li

M

consumption in SEI formation) but also due to defects (as pointed by the XRD and

d

Raman analysis), such as unorganized graphitic structures, micro cavities from which Li

te

can not be retrieved. The irreversibility decreased dramatically after the second cycle. The second cycle insertion curve of the sample show a capacity of 213 mAh/g while the

Ac ce p

extraction shows about 144 mAh/g that decreases very slightly, to nearly 137 mAh/g after 50 cycles (fig. 14). Fig. 12 shows the charge-discharge characteristics of the half cell with sample C2 as anode. The first cycle irreversibility is reduced to nearly 734 mAh/g (less than 50% as compared to sample C1). The second cycle insertion and extraction curves of the sample shows a capacity of about 213 mAh/g and 169 mAh/g resp. Surprisingly the discharge capacity increases gradually from 161 mAh/g in the first cycle to 193 mAh/g after 50 cycles as shown in fig. 14. However in case of sample 3 the first charge capacity is nearly 273 mAh/g and the irreversibility is nearly 188 mAh/g. the insertion and extraction capacities in the second

10

Page 10 of 40

cycle is nearly 92 mAh/g and 96 mAh/g suggesting the extraction of some of the Li ions entrapped during the first charge cycle. A discharge capacity reduces to 47 mAh/g after 30 cycles (fig. 13 and 14).

ip t

The decrease in the irreversible capacity from sample C1 to sample C3 is probably due to the reduced surface area that reduces the irreversible side reactions that leads to the

cr

formation of SEI. Irreversibility due to high surface area has also been reported elsewhere

us

[34, 43, 44]. Increase in porosity and decrease in the pore tortuosity of the anode (as shown in table 3) further accounts for the drop in the irreversibility. In order to

an

understand the above behavior impedance studies of the cell were carried out before and after cycling.

M

The Nyquist plots for the three samples before and after the charge – discharge cycles is

d

shown in fig. 15 and 16 respectively. The electrolyte resistance (RS) for the C1 and C3

te

sample cells increases from 6.9 and 7.1 to 10.5 and 12.8 respectively indicating a decrease in the ion concentration and/ or ion mobility in the electrolyte solution.

Ac ce p

However for the sample cell C2, RS changes very slightly from 8.3 before cycling to 8.4 ohm after 50 cycles that signifies negligible change in solution configuration. The semicircle obtained in the Nyquist plots corresponds to the properties related to the Li ion migration through the solid/electrolyte interface (SEI) layer. The charge transfer resistance (RCT) was calculated by extrapolating the semicircles in fig. 15 and was found to be 7000, 11500 and 14000 ohm for the sample cells C1, C2 and C3 respectively. However after cycling there was a marked reduction in the values of RCT as shown in fig. 16. The RCT was minimum (44 ohm) for cell C2 as compared to 1000 and 2600 ohm for

11

Page 11 of 40

sample cells C1 and C3 respectively. This signifies minimum resistance to the transport of Li-ions through the interface layer of sample cell C2. The straight line portion in the curves represents the dominance of the solid state

ip t

diffusion in the electrode. Extrapolating the line where it meets the abscissa shows minimum diffusion resistance for sample C2 and maximum for sample C3.

cr

After the 50 cycles, the charge – discharge characteristics of the three samples were

us

further carried out at increased current density of 74.4 mA/g and are shown in fig. 17. Almost similar trend was achieved with slightly lower capacities. The performance of

an

sample C2 increases, whereas that of C1 was consistent, for further 50 cycles and still continuing as shown in fig. 14.

M

The insertion sites in carbon materials can broadly be classified into two types (i)

d

graphite like, where Li is intercalated in between the hexagonal cluster at the potential of

te

0-0.25 V and (ii) in surface sites and defects of disordered hexagonal planes where localized radicals may exist which function as acceptors for the lithium ions (in the

Ac ce p

potential range 0.25-2 V). The ratio of the capacities in the potential range 0 - 0.25 V to that in the range 0.25- 2 V was evaluated and tabulated in table 4. The above results suggest that there is an increase in graphite like intercalation mechanism at low current densities. Moreover above results also confirm the high crystallinity and graphitization degree of sample C2 and that more than 50% of the Li ion have intercalated either in between the graphitic interlayer spacing of the MWCNTs or in certain capillaries formed between aligned undispersed MWCNT bundles (as they are scraped out of the reactor as shown in fig. 1). The insertion of Li in sample C1 and C3 on the other hand is dominated by the surface sites and defects in the hexagonal

12

Page 12 of 40

structures. The increase of the graphitic content of C2 sample might also originate from the larger diameter of the tubes, which implies more graphene layer stacked in each tube, thus more ordering possible. Similarly, a larger diameter would allow for more volume

ip t

changes and graphite-like insertion.

Despite the highly graphitic structure of CNT-q, sample C2 shows lower capacity as

cr

compared to the theoretical capacity of graphite. This is probably due to the long length

us

and curved structure of the tubes that probably slows down the graphitic intercalation process leading to improper utilization of the entire anode material. Another reason could

an

be the constrained nature of the graphene sheets, which cannot easily increase their

M

interlayer distance to host lithium as graphite would do.

d

4. Conclusions

te

The structure and crystallinity of the carbon material plays a major role in determining the discharge capacity, reversibility, Li-ion insertion mechanism and rate capacity.

Ac ce p

MWCNTs have been prepared for the first time by indirect heating of the quartz reactor and have proved successful in getting aligned nanotube bundles that shows quite a perfect structure in terms of purity and crystallinity. Moreover the SEI formed in the reaction is just the ideal one leading to improved capacity with successive cycling.

Acknowledgement The authors are grateful to Prof. R.C. Budhani, Director, NPL for his support, encouragement and permission to publish the results. Thanks are also due to Mr. K.N. Sood, Dr. Vidyanand Singh for carrying out the SEM and TEM studies and also to Mr.

13

Page 13 of 40

R.K. Seth for carrying out the TGA analysis. The authors are thankful to CSIR, New

References

ip t

Delhi for funding the research under the CSIR-TAPSUN program.

Rechargeable Lithium Batteries, Adv. Mater. 10 (1998) 725.

cr

[1] M. Winter, J. Besenhard, M. Spahr, P. Novak, Insertion Electrode Materials for

us

[2] J. Wang, I.D. Raistrick, R.A. Huggins, Behavior of Some Binary Lithium Alloys as Negative Electrodes in Organic Solvent-Based Electrolytes, J. Electrochem. Soc.

an

133 (1986) 457.

M

[3] L.Y. Beaulieu, K.W. Eberman, R.L. Turner, L.J. Krause, J.R. Dahn, Colossal Reversible Volume Changes in Lithium Alloys , Electrochem. Solid-State Lett. 4 (2001)

d

A137.

te

[4] J.M. Tarascon, M. Armand, Issues and challenges facing rechargeable lithium

Ac ce p

batteries, Nature 414 (2001) 359.

[5] P. Poizot, S. Laruelle, S. Grugeon, L. Dupont, J.M. Tarascon, Searching for new anode materials for the Li-ion technology: time to deviate from the usual path, J. Power Sources 97-98 (2001) 235.

[6] J. Shim, K.A. Striebel, Electrochemical characterization of thermally oxidized natural graphite anodes in lithium-ion batteries, J. Power Sources 164 (2007) 865. [7] H. Buqa, P. Golob, M. Winter, J. O. Besenhard, Modified carbons for improved anodes in lithium ion cells, J. Power Sources 97-98 (2001) 122. [8] Y.P. Wu, C. Jiang, C. Wan, R. Holze, Anode materials for lithium ion batteries by oxidative treatment of common natural graphite, Solid State Ionics 156 (2003) 283.

14

Page 14 of 40

[9] Y.P. Wu, C. Jiang, C. Wan, R. Holze, Modified natural graphite as anode material for lithium ion batteries, J. Power Sources 111 (2002) 329. [10] H. Wang, M. Yoshio, Carbon-coated natural graphite prepared by thermal vapor

ip t

decomposition process, a candidate anode material for lithium-ion battery, J. Power Sources. 93 (2001) 123.

cr

[11] S. Flandrois, B. Simon, Carbon materials for Lithium-ion rechargeable batteries,

us

Carbon 37 (1999) 165.

[12] C. Menachem, Y. Wang, J. Flowers, E. Peled, S.G. Greenbaum, Characterization of

an

lithiated natural graphite before and after mild oxidation, J. Power Sources 76 (1998) 180. [13] J.S. Xue, J.R. Dahn, Dramatic Effect of Oxidation on Lithium Insertion in.

M

Carbons Made from Epoxy Resins, J. Electrochem. Soc. 142 (1995) 3668.

te

Mater. Chem. 8 (1998) 2223.

d

[14] Y. Wu , S. Fang, Y. Jiang, Carbon anode materials based on melamine resin, J.

[15] W. B. Xing, J. S. Xue, T. Zheng, A. Gibaud, J. R. Dahn, Correlation Between

Ac ce p

Lithium Intercalation Capacity and Microstructure in Hard Carbons, J. Electrochem. Soc. 143 (1996) 3482.

[16] T. Zheng, Q. Zhong, J. R. Dahn, High-Capacity Carbons Prepared from Phenolic Resin for Anodes of Lithium-Ion Batteries, J. Electrochem. Soc. 142 (1995) 212. [17] S. Yoon, H. Kim, S.M. Oh, Surface modification of graphite by coke coating for reduction of initial irreversible capacity in lithium secondary batteries, J. Power Sources 94 (2001) 68. [18] R. Yazami, K. Zaghib, M. Deschamps, Carbon fibers and natural graphite as negative electrodes for lithium ion-type batteries, J. Power Sources 52 (1994) 55.

15

Page 15 of 40

[19] T. Zheng, W. Xing, J.R. Dahn, Carbon prepared from coals for anodes of lithium-ion cells, Carbon 34 (1996) 1501. [20] Y.C. Chang, H.J. Sohn, Y. Korai, Anodic performances of coke from coals, Carbon

ip t

36 (1998) 1653.

[21] A. Mabuchi, K. Tokumitsu, H. Fujimoto, T. Kasuh, Charge Discharge

cr

Characteristics of the Mesocarbon Miocrobeads Heat Treated at Different Temperatures,

us

J. Electrochem. Soc. 142 (1995) 1041.

[22] T. Kasuh, A. Mabuchi, K. Tokumitsu, H. Fujimoto, Recent trends in carbon negative

an

electrode materials, J. Power Sources 68 (1997) 99.

[23] C. Yuqin, L. Hong, W. Lie, L. Tianhong, Irreversible capacity loss of graphite

M

electrode in lithium-ion batteries, J. Power Sources 68 (1997) 187.

d

[24] W. Xing, J.R. Dahn, Study of irreversible capacities for Li insertion in hard and

te

graphitic carbons, J. Electrochem. Soc. 144 (1997) 1195. [25] N. Imanishi, Y. Ono, K. Hanai, R. Uchiyama, Y. Liu, A. Hirano, Y. Takeda,

Ac ce p

O.Yamamoto, Surface-modified mesocarbon microbeads anode for dry polymer lithiumion batteries, J. Power Sources 178 (2008) 744. [26] T. Ohsaki, M. Kanda, Y. Aoki, H. Shiroki, S. Suzuki, High-capacity lithium-ion cells using graphitized mesophase-pitch-based carbon fiber anodes, J. Power Sources 68 (1998) 102.

[27] J.K. Lee, K.W. An, J.B. Ju, B.W. Cho, W.I. Cho, D. Park , K.S. Yun, Electrochemical properties of PAN-based carbon fibers as anodes for rechargeable lithium ion batteries, Carbon 39 (2001) 1299.

16

Page 16 of 40

[28] M. Endo, Y. Nishimura, T. Takshashis, K. Takeuchi, M.S. Dressehauss, Lithium storage behavior for various kinds of carbon anodes in Li ion secondary battery, J. Phys. Chem. Solids 51 (1996) 725.

ip t

[29] N. Takami, A. Satoh, M. Hara, T. Ohsaki, Rechargeable Lithium-Ion Cells Using Graphitized Mesophase-Pitch-Based Carbon Fiber Anodes, J. Electrochem. Soc. 142

cr

(1995) 2564.

us

[30] G. Maurin, Ch. Bousquet, F. Henn, P. Bernier, R. Almairac, B. Simon, Electrochemical lithium intercalation into multiwall carbon nanotubes: a micro-Raman

an

study, Solid State Ionics 136–137 (2000) 1295.

[31] E. Dujardin, T.W. Ebbesen, H. Hiura, K. Tanigaki, Capillarity and Wetting of

M

Carbon Nanotubes, Science 262 (1994) 1850.

d

[32] G.X. Wang, J. Yao, H.K. Liu, S.X. Dou, J.H. Ahn, Growth and Lithium Storage

te

Properties of Vertically Aligned Carbon Nanotubes, Met. Mater. Int. 12 (2006) 413. [33] E. Frackowiak, S. Gautier, H. Gaucher, S. Bonnamy, F. Beguin, Electrochemical

Ac ce p

storage of lithium multiwalled carbon nanotubes, Carbon 37 (1999) 61. [34] H.C. Shin, M. Liu, B. Sadanadan, A.M. Rao, Electrochemical insertion of lithium into multi-walled carbon nanotubes prepared by catalytic decomposition, J. Power Sources 112 (2002) 216.

[35] M. Sharon, W.K. Hsu, H.W. Kroto, D.R.M. Walton, A. Kawahara, T. Ishihara, Y. Takita, Camphor based carbon nanotubes as an anode in lithium secondary batteries, J. Power Sources 104 (2002) 148.

17

Page 17 of 40

[36] J. Wang, C.Y. Wang, C.O. Too, G.G. Wallace, Highly-flexible fiber battery incorporating polypyrrole cathode and carbon nanotubes anode, J. Power Sources 161 (2006) 1458.

ip t

[37] G.T. Wu, C.S. Wang, X.B. Zhang, H.S. Yang, Z.F. Qi, P.M. He, W.Z. Li, Structure and Lithium Insertion Properties of Carbon Nanotubes, J. Electrochem. Soc. 146 (1999)

cr

1696.

us

[38] Lu Yue, H. Zhong, L. Zhang, Enhanced reversible lithium storage in a nanoSi/MWCNT free-standing paper electrode prepared by a simple filtration and post

an

sintering process, Electrochem. Acta 76 (2012) 326.

[39] P.H. Maheshwari, R. Singh, R.B. Mathur, The effect of heat treatment on the

d

Chem. and Phys. 134 (2012) 412.

M

structure and stability of multiwalled carbon nanotubes produced by CVD technique, Mat.

te

[40] R.A. Dileo, B.J. Landi, R.P. Raffaelle, Purity assessment of multiwalled carbon nanotubes by Raman spectroscopy, J. Appl. Phys. 101 (2007) 064307.

Ac ce p

[41] S. Santangelo, G. Messina, G. Faggio, M. Lanza, C. Milone, Evaluation of crystalline perfection degree of multi-walled carbon nanotubes: correlations between thermal kinetic analysis and micro-Raman spectroscopy, J. Raman Spectrosc. 42 (2011) 593.

[42] H.M. Heise, R. Kuckuk, A.K. Ojha, A. Srivastava, V. Srivastava, B.P. Asthana, Characterisation of carbonaceous materials using Raman spectroscopy: a comparison of carbon nanotube filters, single- and multi-walled nanotubes, graphitised porous carbon and graphite, J. Raman Spectrosc. 40 (2009) 344.

18

Page 18 of 40

[43] B. Gao, C. Bower, J.D. Lorentzen, L. Fleming, A. Kleinhammes, X.P. Tang, L.E. McNeil, Y. Wu, O. Zhou, Enhanced saturation lithium composition in ball-milled singlewalled carbon nanotubes, Chem. Phy. Lett. 327 (2000) 69.

ip t

[44] A.S. Claye, J.E. Fischer, C.B. Huffman, A.G. Rinzler, R.E. Smalley, Solid-State Electrochemistry of the Li Single Wall Carbon Nanotube System, J. Electrochem Soc.

us

cr

147 (2000) 284.

Figure Captions

SEM image of aligned bundles of multiwalled carbon nanotubes scraped out of the

an

1.

quartz reactor.

TEM image showing long length curved CNT-q sample. Inset showing open ended

M

2.

d

tube. TEM image of CNT-n sample.

4.

SEM image of free standing sample C1.

5.

SEM image of free standing sample C2.

7. 8. 9. 10.

Ac ce p

6.

te

3.

SEM image of free standing sample C3 X-ray diffraction curves of samples C1, C2 and C3. Raman intensity vs. wave number of samples C1, C2 and C3. TGA and DTA curves of samples C1, C2 and C3. Pore size distribution as a function of normalized volume for samples C1, C2 and C3.

11.

Charge – discharge characteristics of the Li-ion half cell with sample C1 as anode for first two cycles.

19

Page 19 of 40

12.

Charge – discharge characteristics of the Li-ion half cell with sample C2 as anode for first two cycles. Charge – discharge characteristics of the Li-ion half cell with sample C3 as anode for first two cycles.

ip t

13.

Cycling behavior of Li-ion half cell with samples C1, C2 and C3 as anodes.

15.

Nyquist plots of samples C1, C2 and C3 before cycling.

16.

Nyquist plots of samples C1, C2 and C3 after cycling.

17.

Charge – discharge characteristics of the Li-ion half cell with sample C1, C2 and

us

M

an

C3 at current density of 74.4 mA/g.

Table Captions

cr

14.

d

1. XRD analysis of the anode samples C1, C2 and C3.

te

2. Raman analysis of the anode samples C1, C2 and C3. 3. Mercury intrusion porosimetry results of the anode samples C1, C2 and C3.

Ac ce p

4. Ratio of the capacities in the potential range 0-0.25 V to that in the range 0.25 to 2 V.

20

Page 20 of 40

Research Highlights ! ! High aspect-ratio carbon materials favor the formation of self supporting anodes

ip t

! ! Structure of CNTs is an important factor determining its electrochemical performance ! ! Increasing capacity with successive cycling was obtained for CNT sample.

Ac ce p

te

d

M

an

us

cr

! ! Constructive SEI is the key to reversibility and cycle ability.

21

Page 21 of 40

Table 1 S. No. 1

Parameter 2-Theta corresponding to

C1

C2

C3

25.71°

26.34°

25.10°

d 002/ Å

3.462

3.381

3

FWHM corresponding to

3.31°

1.62°

24.589

us

13.845

1.62°

1.59°

-

52.774

-

C1

C2

C3

1.168

0.185

2.776

I2D /IG

0.657

1.855

-

FWHM corresponding to D

65.48

32.74

266.19

69.57

28.65

138.72

102.32

49.11

-

Crystallite size, Lc/ Å

5

FWHM corresponding to (100) XRD peak Crystallite size, La/ Å

51.630

Table 2

2 3

ID /IG

Ac ce p

1

Parameter

te

d

XRD results of the anode samples C1, C2 and C3.

81.539

an

4

S. No.

5.88°

M

(002) XRD peak

6

3.545

cr

2

ip t

(002)

band

4

FWHM corresponding to G band

5

FWHM corresponding to 2D band

Raman results of the anode samples C1, C2 and C3.

22

Page 22 of 40

Table 3 C2

C3

1

Mercury intrusion porosity/ %

28.26

40.85

2

Inter-particle porosity/ %

13.93

24.10

3

Intra-particle porosity/ %

14.33

4

Pore tortuosity

1.911

5

Exposed surface area/ m2 g-1

203.07

ip t

C1

cr

S. No. Parameter

an

us

16.75

62.39 62.39 0

1.76

1.525

105.92

0.39

M

Mercury intrusion porosimetry results of the anode samples C1, C2 and C3.

Table 4 Sample C1

Sample C2

Sample C3

37.2 mA/g

0.358

0.645

0.208

74.4 mA/g

0.109

0.515

0.162

Ac ce p

te

d

Current density

Ratio of the capacities in the potential range 0-0.25 V to that in the range 0.25 to 2 V

23

Page 23 of 40

an

us

cr

ip t

Figure(s)

Ac ce p

te

d

M

Fig. 1

Page 24 of 40

an

us

cr

ip t

Figure(s)

Ac ce p

te

d

M

Fig. 2

Page 25 of 40

an

us

cr

ip t

Figure(s)

Ac ce p

te

d

M

Fig. 3

Page 26 of 40

an

us

cr

ip t

Figure(s)

Ac ce p

te

d

M

Fig.4

Page 27 of 40

an

us

cr

ip t

Figure(s)

Ac ce p

te

d

M

Fig. 5

Page 28 of 40

an

us

cr

ip t

Figure(s)

Ac ce p

te

d

M

Fig. 6

Page 29 of 40

Figure(s)

(002)

C1 C2 C3

ip t

300

cr

200

100 (100)

(101)

us

X ray signal intensity/ a. u.

400

(004)

0 10

20

30

40

an

(110)

50

60

70

(112)

80

M

2 theta

Ac ce p

te

d

Fig. 7

Page 30 of 40

Figure(s)

4000

C1 C2 C3

2D G

3000 D

ip t

2500 2000

cr

1500 1000 500

1500

2000

2500

an

0 1000

us

Raman signal intensity/ a. u.

3500

Raman shift/ cm

3000

-1

Ac ce p

te

d

M

Fig. 8

Page 31 of 40

Figure(s)

0.001 100 0.000

ip t

60

-0.003 -0.004

40

20

us

-0.005

0 0

200

400

an

C1 C2 C3 600

800

-0.006

Weight loss/ mg s

-0.002

cr

% Weight

-1

-0.001

80

-0.007

-0.008 1000

M

Temperature/ °C

Ac ce p

te

d

Fig. 9

Page 32 of 40

ed

ce pt

Ac

us

an

M

cr

Normalized volume/ cc g-1

C2

ip t

Figure(s)

C1

C3

Diameter/ µm

Page 33 of 40

Figure(s)

3.0

2

1.5

ip t

2.0

1

cr

+ Potential vs (Li/Li )/ V

2.5

1.0

2

-200

0

200

1

400

600

800

1000

1200

1400

1600

an

0.0

us

0.5

Specific capacity/ mA h g

1800

-1

Ac ce p

te

d

M

Fig. 11

Page 34 of 40

Figure(s)

3.0

ip t

2.0

2

2

1

cr

1.5

1

1.0

us

+ Potential vs (Li/Li )/ V

2.5

0.5

0

200

400

an

0.0 600

1000

-1

M

Specific Capacity/ mA h g

800

Ac ce p

te

d

Fig. 12

Page 35 of 40

Figure(s)

3.0 1

1.5

2

cr

1

ip t

2.0

1.0 2

us

+ Potential vs (Li/Li )/ V

2.5

0.5

0

50

an

0.0 100

150

200

300

-1

M

Specific Capacity/ mA h g

250

Ac ce p

te

d

Fig. 13

Page 36 of 40

Figure(s)

210 -1 37.2 mA g

-1 74.4 mA g

150

120

ip t

-1 37.2 mA g

-1 74.4 mA g

cr

90 -1 37.2 mA g

60 30

-1 74.4 mA g

0 20

40

60

80

an

0

C1 C2 C3

us

Specific Capacity/ mA h g

-1

180

100

Cycle number

Ac ce p

te

d

M

Fig. 14

Page 37 of 40

Figure(s)

20000

C1 C2 C3

ip t

10000

cr

Z(im)/ ohm

15000

0 5000

10000

15000

20000

an

0

us

5000

25000

Z(re)/ ohm

Ac ce p

te

d

M

Fig. 15

Page 38 of 40

Figure(s)

400

C1 C2 C3

4000

100

cr

3000

Z(im)/ ohm

Z(im)/ ohm

200

2000

0

us

1000

0

ip t

C1 C2 C3

300

-500

0

500

1000

1500

2000

2500

3000

3500

4000

4500

Z(re)/ ohm

0

100

200

300

500

600

an

Z(re)/ ohm

400

Ac ce p

te

d

M

Fig. 16

Page 39 of 40

Figure(s)

C1 C2 C3

1.6

ip t

1.2 1.0

cr

0.8 0.6 0.4 0.2 0.0 20

40

60

80

100

120

140

an

0

us

+ Potential vs (Li/Li )/ V

1.4

Specific Capacity/ mA h g

160

180

-1

Ac ce p

te

d

M

Fig. 17

Page 40 of 40