Carbon treated self-ordered TiO2 nanotube arrays with enhanced lithium-ion intercalation performance

Journal of Alloys and Compounds 597 (2014) 275–281

Contents lists available at ScienceDirect

Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom

Carbon treated self-ordered TiO2 nanotube arrays with enhanced lithium-ion intercalation performance Hyun Sik Kim a, Seung-Ho Yu b, Yung-Eun Sung b,⇑, Soon Hyung Kang c,⇑ a

Energy Material Group, Lotte Chemical, 115, Gajeongbuk-ro, Yuseong-gu, Daejeon 305-726, South Korea School of Chemical & Biological Engineering and Research Center for Energy Conversion & Storage, Seoul National University, San 56-1, Sillim-dong, Gwanak-gu, Seoul 151-744, South Korea c Department of Chemical Education, Chonnam National University, Gwangju 500-757, South Korea b

a r t i c l e

i n f o

Article history: Received 17 January 2014 Received in revised form 3 February 2014 Accepted 3 February 2014 Available online 10 February 2014 Keywords: Li ion battery Carbon doping Titanium oxide nanotubes Anodization

a b s t r a c t Vertically aligned TiO2 nanotube (TONT) arrays on titanium substrate developed by facile electrochemical anodization in an aqueous solution of 0.5 M Na2SO4, 0.5 M H3PO4, 0.2 M sodium citrate, and 0.5 wt% NaF were prepared having a pore diameter and thickness of 100 nm and 1.2 lm, respectively. The undoped (u-doped) TONT arrays possessing an anatase phase were again annealed at 500 °C under a mixed gas flux of nitrogen (N2) and acetylene (C2H2), to induce the enhancement of electrical conductivity. It was designated as carbon-doped (c-doped) TONT arrays. Undoped and c-doped TONT arrays were compared using various characterization tools, including X-ray diffraction (XRD), field-emission scanning electron microscopy (FE-SEM), and X-ray photoelectron spectroscopy (XPS). Furthermore, based on several electrochemical tests (galvanostatic charge/discharge tests, cyclic voltammetry (CV), and electrochemical impedance spectroscopy (EIS)), it was observed that c-doped TONT arrays revealed improved charge/discharge capacity, cycle stability, and rate capability, due to the enhanced electrical conductivity of c-doped TONT arrays. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction Among the reliable, quickly rechargeable power sources available, lithium ion batteries (LIB) are the most promising technologies because of their high energy storage density, stable cycle life, little memory effect, absence of poisonous metals and so on. At present, graphite is the commercially used anode material, which has a theoretical specific capacity of 372 mAh/h and quite stable durability. However, several limitations such as electrical disconnection and structural deformation hinder compact LIBs in higher energy density and higher stable cell applications [1,2]. To avoid these limitations, many advanced LIB anode materials, including Si, Ge, and Sn having a high capacity have been examined. However, the decomposition of lithium salt-based liquid electrolyte at low operating voltage (<1 V vs. Li/Li+) and the formation of an unstable solid electrolyte interface (SEI) on the electrode surface have been found [3,4]. Therefore, as a new anode material, titanium dioxide (TiO2) can be suggested due to its high operating voltage (1.75 V vs. Li/Li+) in which the electrolyte solution is stable ⇑ Corresponding authors. Tel.: +82 62 530 2497 (S.H. Kang), +82 2 880 1889 (Y.-E. Sung). E-mail address: [email protected] (S.H. Kang). http://dx.doi.org/10.1016/j.jallcom.2014.02.013 0925-8388/Ó 2014 Elsevier B.V. All rights reserved.

and has high safety, good cyclability, low self-discharge rates, and small volume change (<3%) during lithiation/delithiation, even though the theoretical capacity of TiO2 is lower than the Sn, Si, and graphite [5,6]. In addition, TiO2 shows chemical stability and negligible toxicity, and has relatively simple methods. In terms of the nanostructure, it is well known that the lithium intercalation activity and cycling stability rely significantly on the morphology of electrode materials, and that their properties have dramatically improved with the modification of nanoscale features [7–10]. Above all, one-dimensional (1-D) TiO2 nanoarchitecture has been of significant interest due to the promotion of ionic and electronic diffusion, increasing the electrode/electrolyte interfacial area and ensuring easy accommodation of strain suffered from the lithium-ion insertion/extraction. Recently, several groups have synthesized 1-D anatase TiO2 nanostructures such as nanorods, nanowires, and nanotubes (NTs), and assessed their electrochemical properties [11–13]. Especially, anatase TiO2 nanotubes (TONTs) have shown superior electrochemical properties compared to TiO2 nanoparticles (such as the discharging capacity maintaining 168 mAh/g in the 30th cycle at 210 mAh/g), representing a high-rate cycling performance, columbic efficiency approaching 98%, and excellent reversibility. TONTs are thus recognized as very attractive anode

276

H.S. Kim et al. / Journal of Alloys and Compounds 597 (2014) 275–281

materials for LIB [14,15]. Subsequently, considerable attention has been given to the chemical synthesis of various nanostructure 1-D TiO2 materials, presenting good electrochemical performance [16]. This still shows that 1-D nanoparticles exist that have a randomly oriented property and poor contact with the current collector, reducing the lithium ionic and electrical conductivities. Consequently, self-ordered 1-D nanoarchitectures grown directly on a current collector such as Cu, Pt, and Ti have been widely studied as a negative electrode for LIB [17–19]. Recently, the oriented anatase TONT arrays, prepared using facile electrochemical anodization, have attracted much attention because the several disadvantages induced in the chemically synthesized TiO2 nanomaterials were overcome and several strong advantages emerged [20]. First, the oriented pore structure of NT arrays is expected to facilitate 1-D electronic/ionic conduction and to accommodate volume change during charging/discharging cycling. Second, the Li+ ions in the electrolyte are highly accessible to the interior and exterior surface of the NT walls. Third, the length of the Li+ diffusion path in the thin wall of NT arrays is short. Fourth, the electron transport in the environment where the NT arrays are directly grown on the titanium substrate is fast. With these potential advantages associated with oriented NT arrays, it was found that the low electrical conductivity of NT arrays results in unstable capacity retention and poor rate capability, leading to rapid capacity fading during charging/discharging cycling. To increase the electrical conductivity of NT arrays, the doping process can be suggested and applied for various devices such as photocatalysts, photoelectrochemical water splitting, and lithium ion batteries [21–23]. Among various dopants, carbon-doped (c-doped) TiO2 showed improved electrical conductivity, leading to better photocatalytic effect. However, to our knowledge, few reports have been presented on the electrochemical performance of c-doped TONT arrays. Herein, c-doped TONT was fabricated using a nondestructive thermal-treatment at 500 °C under a 20% C2H2/N2 gas mixture ambient. The c-doped TONT arrays showed high capacity and cycling stability from the discharging/charging cycling, cyclic voltammetry, and electrochemical impedance spectroscopy.

Electrochemical measurements were executed in two electrode cells using Li foil as both the reference and counter electrodes. The coin-type half cells were assembled using Li foil, an electrolyte of 1 M LiClO4 in a 50:50 (w/w) mixture of ethylene carbonate and dimethyl carbonate, and polyethylene film (Celgard 2300) as the separator in an Ar-filled glove box. The cells were charged and discharged at a constant current density of 42, 84, 168, and 336 mA/g between 1.0 V and 2.5 V vs. Li+/Li on a TOSCA-3100 battery cycler (Toyo Co., Japan). Cyclic voltammetry (CV) was carried out in the potential range of 1.0 V–2.5 V vs. Li+/Li at a scan rate of 0.1 mV/s on a Solartron multi-state instrument (Model 1480, UK). Electrochemical impedance spectroscopy (EIS; IM6, Zahner, Germany) was carried out by applying an alternating current (ac) voltage of 5 mV in the frequency from 100 kHz to 0.01 Hz.

3. Results and discussion Fig. 1 compares the typical top and cross-sectional FE-SEM images of the u-doped and c-doped TONT arrays with the length of approximately 1.2 lm, which enables uniform carbon doping through the inner and outer walls. The pore diameter and wall thickness were observed to be 100(±25) nm and 25(±5) nm, respectively. Furthermore, it was observed that the irregular shaped pores were not perfectly round in shape. After a 20% C2H2/N2 gas treatment, no morphological modification was observed, confirming that the acetylene thermal treatment did not induce the morphological change of the TONT arrays. In order to determine the crystallinity of the u- and c-doped TONT arrays, XRD measurements were performed as shown in Fig. 2. The as-anodized TONT arrays exhibit an amorphous structure without any detectable crystalline phases, since only the peaks associated with the Ti substrate were evident. This is consistent with reports by other researchers that only amorphous titania can be produced via anodization without high-temperature annealing. Annealing the as-prepared TONT arrays at 450 °C for

2. Experimental Ti foil (Goodfellow, England) of 0.1 mm thickness and 99.6% purity was used as a substrate. Before each experiment, the Ti foil was cleaned by sonicating sequentially in acetone, isopropanol, and methanol for 10 min each, followed by rinsing in distilled (DI) water and then drying in a nitrogen stream. The electrochemical anodization system was composed of a two-electrode configuration with a working electrode of Ti foil and a counter electrode of Pt mesh. The electrolyte used consisted of 0.5 M Na2SO4, 0.5 M H3PO4, 0.2 M sodium citrate, and 0.5 wt% NaF in an aqueous bath. Experimental conditions were kept constant with an applied voltage of 25 V, duration of 50 min, temperature of approx. 25 °C, and slow magnetic agitation of 150 rpm to identify the reproducible results. The as-prepared TONT arrays were ultrasonically cleaned in DI water for 1 min to remove the remnants of the anodic reaction and were immediately annealed at 450 °C (heating/cooling rate of 1.5 °C/min) for 3 h under air ambient. For the c-doping into TONT arrays, a second thermal treatment was carried out. During the heating process to 500 °C, a heating rate of 5 °C/min was adapted in a N2 gas flow and during the holding process at 500 °C for 10 min, a 20% C2H2/N2 gas mixture was introduced for c-doping. Finally, the samples were naturally cooled in an N2 atmosphere. The crystal structures of the as-prepared samples were investigated by X-ray diffraction (XRD) using a Rigaku diffractometer operated with a Cu Ka radiation source (k = 1.541 Å) at an operating voltage of 50 kV and current of 200 mA. To survey the surface and cross-sectional morphology of TONT arrays, field-emission scanning electron microscopy (FE-SEM; JSM-6330F, JEOL, Japan) was employed. To validate the change of chemical state in the TONT and c-doped TONT arrays, X-ray photoelectron spectroscopy (XPS) analyses were performed in a UHV multipurpose surface analysis system (SIGMA PROBE, Thermo, UK) operating at a base pressure of <109 mbar. The photoelectron spectra were excited by an Al Ka (1486.6 eV) anode operating at a constant power of 100 W (15 kV and 10 mA). The binding energy (BE) scale was calibrated from the hydrocarbon contamination using the C 1s peak at 284.6 eV.

Fig. 1. FE-SEM images of (a) u- and (b) c-doped TONT arrays annealed at 500 °C in a C2H2/N2 atmosphere for 10 min.

277

Ti(102)

Ti(101)

A(211) A(105)

A(200)

A(004)

Ti(100)

Intensity (a.u.)

A(101)

Ti(002)

H.S. Kim et al. / Journal of Alloys and Compounds 597 (2014) 275–281

c-doped TiO2 nantube

TiO2 nanotube

As-grown Ti substrate

20

30

40

50

60

2θ (deg.) Fig. 2. XRD patterns of u- and c-doped TONT arrays carbonized at 500 °C in a C2H2/ N2 atmosphere.

3 h leads to the formation of anatase phase with the planes of (1 0 1), (0 0 4), (0 0 2), (2 0 0), (2 1 1), and (1 0 5). Additional annealing of u-doped TONT arrays at 500 °C for 30 min to induce the c-doping in the TiO2 elements results in no additional formation of crystalline peaks and a little shift of anatase TiO2 peaks to a high degree of about 0.05°. This was attributed to the carbon incorporation in the oxygen site in the TiO2 lattice, which subsequently led to the reduction of the lattice distance, together with a shift toward the high degree, following Bragg’s equation. However, it can be seen that the extent is extremely minimal. In addition, it was found that the anatase superstructure peaks of the c-doped TONT arrays

(b)

1000

800

600

400

200

0

290

Binding energy (eV)

462

460

Binding energy (eV)

284

529.7

(d)

Ti 2p

458.4

464

286

282

Binding energy (eV)

464.2

466

288

Intensity (a.u.)

Intensity (a.u.)

(c)

C 1S

288.3

Ti 3s Ti 3p

1200

284.7

Intensity (a.u.)

C 1s

Ti 2s

Ti 2p1/2

Intensity (a.u.)

Ti 2p3/2

O 1s

(a)

exhibit larger heights and smaller full width at half-maximum compared to those of the TONT arrays. This observation indicates that the mean crystallite size decreases with acetylene thermal treatment. The change of carbon-doping that induced a chemical binding state on the TiO2 lattice was surveyed using the XPS technique, as shown in Fig. 3. The quantitative atomic ratios of C, O, and Ti elements, as measured in the wide scan range (Fig. 3(a)), were 33.93, 48.9 and 17.16 at%, respectively. To survey thoroughly the chemical states of each element, the C 1s, Ti 2p, and O 1s core level peaks were summarized as shown in Fig. 3(b–d). In the case of the C 1s core level peak, two peaks at binding energies (BEs) of 284.7 and 288.3 eV with a shoulder at 286 eV were observed and the main peaks were assigned as the adventitious elemental carbon and the advent of TiAC bond [24], respectively. It was well known that the shoulder at 286 eV originated from the C@O bond or TiAC bond. In the case of the Ti 2p core-level peak assigned as 458.2 eV (Ti 2p3/2) and 464.0 eV (Ti 2p1/2), a shift toward a low binding energy site was noticed at about 0.4 eV, compared to the BEs of Ti 2p3/2 (458.6 eV) and Ti 2p1/2 (463.6 eV) in the u-doped TONT arrays. This shift was attributed to the main substitutional incorporation of carbon elements on the TiO2 lattice, leading to the formation of Ti3+ species [25]. Furthermore, the O 1s core-level peak with the BE of 531.1 eV definitely confirmed that the above result was due to the oxygen bonding to Ti3+ [26]. Thus, based on UV–vis spectra and XPS data, it was determined that the carbon elements were substitutionally incorporated in the TiO2 lattice to make c-doped TONT arrays. Fig. 4 shows the first five cycles of the galvanostatic charge/discharge curves of (a) u- and (b) c-doped TONT arrays conducted at 50 mAh/g between the cutoff voltages from 2.5 V to 1.0 V (vs. Li/ Li+). The discharge curve, corresponding to the Li+ insertion

458

456

O 1S

531.1

534

533

532

531

530

529

528

527

Binding energy (eV)

Fig. 3. XPS survey spectra of c-doped TONT arrays over a wide scan range and C 1s, Ti 2p and O 1s core level peaks.

H.S. Kim et al. / Journal of Alloys and Compounds 597 (2014) 275–281

2.6

(a)

+

Voltage (V vs. Li /Li)

2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0

20

40

60

80

100

120

140

160

180

200

220

160

180

200

220

Capacity (mAh/g) 2.6

(b)

+

Voltage (V vs. Li /Li)

2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0

20

40

60

80

100

120

140

Capacity (mAh/g) Fig. 4. Charge/discharge curves of (a) u- and (b) c-doped TONT arrays between 1.0 V and 2.5 V at 0.5 C rate.

process, can be divided into three consecutive potential regions. The first region, where the potential reduced rapidly and monotonously from 3 to about 1.75 V, is associated with the formation of a solid solution domain (LixTiO2 with x up to 0.15). In the second region, the potential plateau was formed at 1.75 V, signifying the biphasic region (e.g., coexistence of TiO2 and Li0.5TiO2) from the Li+ insertion process. In other words, the phase equilibrium between the Li-poor and Li-rich phases is formed at these plateaus, where half of the Ti4+ states were converted to Ti3+ states during the insertion/desertion process of Li+ ions [26]. Therefore, the electrochemical insertion/desertion of Li+ ions into/from the anatase lattice can be written as TiO2 + x(Li+ + e-) ? tLixTiO2, where the insertion coefficient, x, is usually close to 0.5 in anatase, corresponding to a capacity of 168 mAh/g. In the third region from 1.75 V to 1.0 V, the potential of TONT arrays decreases linearly with increasing discharge capacity from 100 to 160–200 mAh/g, indicating a distinctive pseudocapacitive behavior. Moreover, a short voltage plateau appeared near 1.3 V (u-doped TONT) and 1.4 V (c-doped TONT) within the steep region from 1.73 V to 1.0 V, closely correlated to the surface imperfections or defects of the anatase TONT arrays. Similarly, the charge curves can also be separated into three consecutive potential regions, corresponding to the reverse processes (i.e., Li+ extraction) of the three regions of discharge curves. In the charge process, the potential plateau during Li+ extraction is reached at about 1.95 V, which is typical for the Li+ extraction from anatase TiO2 electrodes [27,28].

Typically, the first discharge and charge capacities of the c-doped TONT arrays were 207 mAh/g and 181 mAh/g, respectively, which were higher than those previously reported for nanostructured TiO2 [29,30]. The inevitably large irreversible capacity in the first cycle was attributed to the insertion of lithium in irreversible sites and side reactions induced by a trace of water resulting from the large surface area of NT arrays. However, after one large capacity drop, the average capacity loss was no longer observed as second to the fifth cycle. On the other hand, u-doped TONT arrays show the fast capacity fading as due to repetitive cycling, poor electrochemical performance with lower initial discharge and charge capacities of 193 mAh/g and 167 mAh/g, respectively, a shorter voltage plateau, and a wider polarization (Vch – Vdis) compared to those of c-doped TONT arrays. Therefore, these results show that carbon doping results in not only longer discharge/charge voltage plateaus but also the minimization of first irreversible capacity loss. Fig. 5 shows the cyclic stability of u- and c-doped TONT arrays measured using a chronopotentiometric method for more than 30 continuous charge/discharge cycles at a current density of 50 mAh/g. The large irreversible lithium intercalation capacities in the first and second cycles can be ascribed to the lithium insertion in irreversible sites and various side reactions. The irreversible Li insertion sites are almost filled and a trace of water is consumed gradually during the next several cycles; thus, the discharging/ charging stabilizes in the following cycles. Generally, c-doped TONT arrays clearly showed better cycle performance and higher charge capacity (about 180 mAh/g) than those of u-doped TONT arrays (about 155 mAh/g). That is, relative to the reversible capacity of u-doped TONT arrays, that of c-doped TONT arrays is large at 181 mAh/g after the first charge process while retaining a high reversible capacity of 174 mAh/g even after 30 cycles. On the other hand, u-doped TONT arrays show poor cycle stability as the initial reversible capacity of 167 mAh/g was steadily reduced to 151 mAh/g after 30 cycles. Furthermore, c-doped TONT arrays maintain high capacity retention, still achieving 96% of the second capacity after 30 cycles seven at the rate of 50 mAh/g, which was much higher than 44% of u-doped TONT arrays. As a result, it was concluded that c-doped TONT arrays exhibit good cycling stability and complete reversibility, showing that no structural change occurs during lithium insertion and extraction. This may be due to the carbon doping, in which no correspondence occurred with structural deformation during the lithium intercalation/

220

c -doped TiO2 discharge c -doped TiO2 charge u -doped TiO2 discharge

200

Capacity (mAh/g)

278

u -doped TiO2 charage

180

160

140

0

5

10

15

20

25

Cycle number (n) Fig. 5. Cycle performance of u- and c-doped TONT arrays.

30

279

H.S. Kim et al. / Journal of Alloys and Compounds 597 (2014) 275–281

0.02

(a)

Current (mA)

0.01

0.00

-0.01

c-doped TiO2

-0.02

u-doped TiO2 -0.03 1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

+

Voltage (V vs. Li /Li) 0.06

0.04

0.8 mV/s 0.6 mV/s 0.4 mV/s

0.02

0.2 mV/s 0.1 mV/s

0.00 0.06

-0.02

0.05

Ip / mA

Current (mA)

(b)

-0.04

0.04 0.03

C-doped TiO 2

0.02

-0.06 0.01 0.2

undoped TiO U-doped TiO22 0.3

0.4

0.5

0.6 1/2

(Scan rate, ν)

-0.08 1.0

1.2

1.4

1.6

1.8

2.0

2.2

0.7

0.8

0.9

1.0

1/2

/ (mV/s)

2.4

2.6

2.8

+

Voltage (Li vs. Li /Li) Fig. 6. Cyclic voltammograms of (a) u- and c-doped TONT arrays at a scan rate of 0.1 mV/s and (b) c-doped TONT arrays at a scan rate of 0.1, 0.2, 0.4, 0.6, 0.8 mV/s. The inset shows the relationship between the cathodic peak currents of voltammogram (Ip) and the square root of scan rate (m1/2).

extraction process and the enhanced conductivity of NT arrays was attained. The electrochemical behavior of u- and c-doped TONT arrays was investigated from the first cyclic voltammograms from 2.5 V to 1.0 V with a scan rate of 0.1 mV/s, as displayed in Fig. 6a. Finally, both curves exhibit a similar shape with only one pair of cathodic/ anodic peaks corresponding to lithium insertion and extraction near 1.69 V and 2.04 V (Li+/Li), respectively, which are associated with electrolytic lithium insertion and extraction from the TiO2 anatase structure, according to the schematic reaction; þ

TiO2 þ xLi þ xe ! Lix TiO2 It can be seen that the position of these peaks was in accordance with the potential plateaus of the charge/discharge curves presented in Fig. 4. It should also be noted that a slight asymmetry in the CV curve here reveals the difference in lithium ion insertion and extraction capabilities. The magnitude of peak current in the cdoped TONT arrays is higher than that of the u-doped TONT arrays, representing a slightly greater capacity. Also, the extent of separa-

tion between the anodic and cathodic peak potential demonstrates that the extent of reversibility is smaller in the c-doped TONT arrays than in the u-doped TONT arrays, reflecting a better reversibility from carbon doping. Fig. 6(b) displays the typical CVs of c-doped TONT arrays measured at scan rates from 0.1 to 0.8 mV/s. The intensities of both cathodic (discharge) and anodic (charge) currents increased significantly at higher scan rates over the entire potential window. The inset of Fig. 6(b) shows the dependence of the peak discharge current (Ip) on the scan rate (m1/2). Similar scan rate dependence is observed for the charge current. Herein, a linear relationship between peak current (Ip) and the square root of scan rate (m1/2) was noticed, showing that the reaction kinetics are totally controlled by the diffusion process. This behavior described by i – m1/2 is typical for Li insertion in an ordinary anatase lattice. The CV curves also mean that this electrode reveals good cycling stability and reversibility, in agreement with the results of the cycling performance shown in Fig. 5. To calculate the diffusion coefficient for Li+ in the anatase TiO2, the following equation was used [31,32]:

Ip ¼ 2:69  105 An3=2 C o D1=2: m1=2

280

H.S. Kim et al. / Journal of Alloys and Compounds 597 (2014) 275–281

Relative charge capacity (%)

c -doped TiO2 100

u -doped TiO2

80

60

40

20 0.5

1.0

1.5

2.0

C-rate Fig. 7. Comparison of current rate versus the reversible charge capacities of u- and c-doped TONT arrays. (The data were obtained by normalizing the third charge capacities at various C rates to vs. 0.25 C.)

As a powerful tool to evaluate the kinetic parameters of the Liion transfer process, the ac EIS of u- and c-doped TONT arrays was measured as shown in Fig. 8. Both Nyquist plots consisted of a semicircle in the high frequency region, attributed to the absence of an SEI layer on the TiO2 surface and a sloping straight line in the low frequency region. Referring to the preceding literature survey [23,33], the high-frequency semicircle is related to the Li-ion transfer resistance from the electrolyte into the TONT arrays, while the sloping line at the low frequency domain represents Li-ion diffusion in the bulk electrode, known as the Warburg impedance (W). From these plots, it is found that the semicircle diameter of c-doped TONT arrays is larger than that of u-doped TONT arrays, implying a smaller charge transfer resistance (Rct) and suggesting that carbon doping improves the charge transfer rate of Li-ion and thus the electrical conductivity of c-doped TONT arrays. On the basis of these results, it was concluded that the self-aligned c-doped TONT arrays as an anode for LIB shows high capacity, good cyclability, and excellent reversibility, as a consequence of its superior electrochemical performance compared with that of u-doped TONT arrays. 4. Conclusion

where Ip is the peak current in the voltammogram, A is the surface area of TONT arrays, n is the number of electrons per molecule during the intercalation, Co is the concentration of lithium ions, D is the diffusion coefficient of lithium ions, and m is the scan rate. The diffusion coefficient of c-doped TONT arrays was 1.10  107 cm2/s, which was higher than the 6.78  108 cm2/s of u-doped TONT arrays. It can be clearly seen that the carbon doping in the TONT arrays enhanced the diffusivity to the lithium ion in the electrolyte. Fig. 7 shows the relative charge capacity as a function of various current rates to confirm the rate capability of u- and c-doped TONT arrays. The charge capacities in the third charge process with a current density of 42 mAh/g (0.25 C) were then adapted as a standard value, showing 182 mAh/g and 163 mAh/g in the case of u- and cdoped TONT arrays, respectively. By increasing the current rate from 0.5 C to 2 C, the third charge capacities were reduced from 169 mAh/g to 126 mAh/g and the capacity retention was also decreased from 92.9% to 69.3% for c-doped TONT arrays. On the other hand, the charge capacities were reduced from 140 mAh/g to 82 mAh/g and the capacity retention was decreased from 85.3% to 48.8% for u-doped TONT arrays at the same current rates. Based on these results, it was confirmed that c-doped TONT arrays exhibit superior rate capabilities relative to those of u-doped TONT arrays.

Self-organized c-doped TONT arrays were successfully prepared by simple electrochemical anodization, subsequently followed by acetylene heat treatment at 500 °C. Relative to u-doped TONT arrays, c-doped TONT arrays exhibited superior cycle performance and favorable electrochemical kinetics during the lithium ion insertion/desertion process. The main discrepancy in electrochemical behavior between the u- and c-doped TONT arrays was considered to be due to the different electrical conductivities of the intrinsic electrode materials during repetitive cycling. Also, considering the superior capacity retention on cycling in the c-doped TONT arrays, the carbon doping on the self-ordered TONT arrays can be regarded as a promising modification tool to alter the materials’ electrical and optical features, validating a status for the potential anode materials for high power LIB. Acknowledgments This work was supported by the Institute for Basic Science (IBS) and Fundamental R&D Program for Technology of World Premier Materials funded by the Ministry of Knowledge Economy, Korea (10037919). References

600

c -doped TiO2

500

u -doped TiO2

-Z'' (ohm)

400

300

200

100

0 0

100

200

300

400

500

600

Z' (ohm) Fig. 8. Electrochemical impedance spectra (EIS) of u- and c-doped TONT arrays.

[1] K.M. Abraham, Electrochim. Acta 38 (1993) 1233. [2] A.N. Jansen, A.J. Kahaian, K.D. Kepler, P.A. Nelson, K. Amine, D.W. Dees, D.R. Vissers, M.M. Thackeray, J. Power Sources 81 (1999) 902. [3] E. Peled, D. Golodnitsky, G. Ardel, V. Eshkenazy, Electrochim. Acta 40 (1995) 2197. [4] V. Eshkenazi, E. Peled, L. Burstein, D. Golodnitsky, Solid State Ionics 170 (2004) 83. [5] S.-J. Bao, Q.-L. Bao, C.-M. Li, Z.-L. Dong, Electrochem. Commun. 9 (2007) 1233. [6] V. Subramanian, A. Karki, K.I. Gnanassekar, F.P. Eddy, B. Rambabu, J. Power Sources 159 (2006) 186. [7] A.S. Arico, P.G. Bruce, B. Scrosati, J.M. Tarascon, W.V. Schalkwijk, Nat. Mater. 4 (2005) 366. [8] J. Schoonman, Solid State Ionics 129 (2000) 38. [9] N.C. Li, D.T. Mitchell, K.P. Lee, C.R. Martin, J. Electrochem. Soc. 150 (2005) A979. [10] A.H. Whitehead, J.M. Elliott, J.R. Owen, J. Power Sources 81 (1999) 33. [11] H. Zhang, G.R. Li, L.P. An, T.Y. Yan, X.P. Gao, H.Y. Zhu, J. Phys. Chem. C 111 (2007) 6143–6148. [12] A.R. Armstrong, C. Armstrong, J. Canales, R. Garcia, P. Brue, Adv. Mater. 17 (2005) 862. [13] X.P. Gao, Y. Lan, H.Y. Zhu, J.W. Liu, Y.P. Ge, F. Wu, D.Y. Song, Electrochem. SolidState Lett. 8 (2005) A26. [14] G. Sudat, E. Baudrin, D. Larcher, J.-M. Tarascon, J. Mater. Chem. 15 (2005) 1263. [15] C. Jiang, M. Wei, Z. Qi, T. Kudo, I. Honma, H. Zhou, J. Power Sources 166 (2007) 239.

H.S. Kim et al. / Journal of Alloys and Compounds 597 (2014) 275–281 [16] [17] [18] [19] [20] [21] [22] [23] [24]

J. Li, Z. Tang, Z. Zhang, Electrochem. Solid-State Lett. 8 (2005) A316. J.M. Mack, H. Tsuchiya, P. Schmuk, Angew. Chem., Int. Ed. 44 (2005) 2100. C.J. Patrissi, C.R. Martin, J. Electrochem. Soc. 146 (1999) 3176. J. Liu, Y. Li, R. Ding, J. Jiang, Y. Hu, X. Ji, Q. Chi, Z. Zhu, X. Huang, J. Phys. Chem. C 113 (2009) 5336. G.F. Ortiz, I. Hanzu, P. Lavela, P. Knauth, J.L. Tirado, T. Djenizian, Chem. Mater. 22 (2010) 1926. R. Hahn, A. Ghicov, J. Salonen, V.-P. Lehto, P. Schmuki, Nanotechnology 18 (2007) 105604. J.H. Park, S. Kim, A.J. Bard, Nano Lett. 6 (2006) 24. J. Xu, Y. Wang, Z. Li, W.F. Zhang, J. Power Sources 175 (2008) 903. W. Ren, Z. Ai, F. Jia, L. Zhang, X. Fan, Z. Zou, Appl. Catal. B – Environ. 69 (2007) 138.

281

[25] Y. Li, D.-S. Hwang, N.H. Lee, S.-J. Kim, Chem. Phys. Lett. 404 (2005) 25. [26] Q. Xiao, J. Zhang, C. Xiao, Z. Si, X. Tan, Sol. Energy 82 (2008) 706. [27] Y.-K. Zhou, L. Cao, F.-B. Zhang, B.-L. He, H.-L. Li, J. Electrochem. Soc. 150 (2003) A1246. [28] Z. Wang, S. Liu, G. Chen, D. Xia, J. Electrochem. Soc. 10 (2007) A77. [29] P. Kubiak, J. Geserick, N.H. Husing, M. Wohlfahrt-Mehrens, J. Power Sources 175 (2008) 510. [30] S.W. Oh, S.-H. Park, Y.-K. Sun, J. Power Sources 161 (2006) 1314. [31] R.V.D. Krol, A. Goossens, J. Schoonman, J. Phys. Chem. B 103 (1999) 7151. [32] J. Wang, J. Polleux, J. Lim, B. Dunn, J. Phys. Chem. C 111 (2007) 14925. [33] B.-L. He, B. Dong, H.-L. Li, Electrochem. Commun. 9 (2007) 425.