Published in completed and revised form in Chemistry – a European 1

Journal DOI: 10.1002/chem.201303283, Copyright Wiley VCH 2

3

Mesoporous anatase TiO2 nanorods as thermally robust 4

anodes for Li-ion batteries – a detailed insight into 5

formation mechanism 6

7

Gulaim A. Seisenbaeva[a]*, Jean-Marie Nedelec[b,c], Geoffrey Daniel[d], Carmen 8

Tiseanu[e,f], Vasile Parvulescu[e], Vilas G. Pol[g], Luis Abrego[g], Vadim G. Kessler[a]* 9

10

╫ Department of Chemistry, Biocenter, SLU, Box 7015, SE-75007, Uppsala, Sweden; ╩ 11

Clermont Université, ENSCCF, Institut de Chimie de Clermont-Ferrand, BP 10448, F-12

63177 Clermont-Ferrand, France, † CNRS, UMR 6296, ICCF, F-63171 AUBIERE, 13

France, φ Department of Forest Products/Wood Science, SLU, Box 7008, 75007 Uppsala, 14

Sweden, � National Institute for Laser, Plasma and Radiation Physics, P.O.Box MG-36, 15

RO 76900, Bucharest-Magurele, Romania, ‡ Department of Chemistry, Univ. Bucuresti, 16

B-dul Regina Elisabeta, nr. 4-12, Sector 3, 030018 Bucuresti, Romania, 17

╪Electrochemical Energy Storage Department, Chemical Sciences and Engineering 18

Division, Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, IL 60439, USA 19

20

21

22

Corresponding author: Pr. Vadim Kessler 23

Tel: +46-18 67 15 41 24

E-mail: vadim.kessler@slu.se25

Abstract 26

Uniformly mesoporous and highly thermally robust anatase nanorods were produced with 27

quantitative yield by a new simple and efficient one-step approach. The mechanism of 28

this process was revealed via insertion of Eu3+ cations from the reaction medium as 29

luminescent probe. The obtained structure displays unusually high porosity and active 30

surface area of about 300 m2/g making this material attractive as an anode electrode for 31

Li-ion batteries with specific capacity of 167 mAh/g at C/3 rate. An additional attractive 32

feature is its remarkable thermal stability – heating to 400°C results in decrease of the 33

active surface area to a still relatively high value of 110 m2/g with conservation of open 34

mesoporosity. Thermal treatment at 800°C or higher, however, causes transformation into 35

a non-porous rutile monolith as it commonly observed with nano-titania. 36

37

Keywords : Mesoporous titania, Anatase, Formation mechanism, Thermal stability, 38

Electrochemical performance, Li-ion battery 39

1. Introduction 40

Rational design of more efficient devices for energy storage is one of the most important 41

quests set for modern materials science.1 Li-ion batteries are broadly recognized as one of 42

the most attractive answers to this quest,2-4 combining such attractive features as 43

generally high capacity and low weight and volume, permitting the creation of small and 44

highly efficient energy sources. The key to building up an attractive battery is the design 45

of electrode materials. The matrix for incorporation of Li-ions needs to be thermally and 46

chemically stable, have relatively low density (and molecular mass per unit) and, of 47

course, display high capacity in relation to the stored charge in combination with rapid 48

charging-up kinetics. Among materials, potentially capable to meet these challenges, an 49

important place is occupied by the titanium dioxide5,6. All the crystalline modifications of 50

TiO2 in nano form such as nanoparticles,7 nanotubes,8 nanowires9 and nanoribbons10 were 51

considered as prospective for application as anodes for the Li-ion batteries. They have 52

revealed, however, also certain disadvantages such as poor electron conductivity because 53

of potential aggregation and microsintering of the nanopowders together with poor 54

contact between individual particles and low packing density as discussed in5. Solutions 55

to these problems were sought in application of easily packed microparticles with 56

improved contacts between nano building blocks in their structure11,12. Relatively 57

complex technical solutions for preparation of spherical nanostructured microparticles 58

involving template synthesis in combination with spray-drying have been proposed 59

offering TiO2 with high capacity and superior rate performance with 165 mAh g−1 at 60

10°C and 116 mAh g−1 at 60°C 5. For enhancing electronic conductivity, the autogenic 61

technique was used to synthesize carbon encapsulated, anatase (TiO2–C) anode 62

nanoparticles for lithium-ion batteries, providing a reversible capacity of 200mAhg 63

with C/17 rate 13. 64

We have recently reported an extremely facile rapid hydrothermal synthesis approach 65

based on quick immersion of solid crystalline titanium alkoxide precursors such as 66

commercially available titanium methoxide into boiling water.14 In the present work we 67

have applied the in situ doping of the thus forming nanostructure with Eu3+-cations in 68

order to produce advanced insight into the mechanism of this process and get better 69

understanding for the structural transformation of this nanomaterial in the course of 70

thermal treatment and its functioning as an anode for Li-ion batteries. 71

72

2. Experimental section 73

The synthetic procedure for anatase nanorods has been described by us recently 74

elsewhere. 14 In a typical experiment about 1 g of titanium methoxide, Ti(OCH3)4 75

(Aldrich [992-92-7], Cat. No. 463582), was quickly immersed into 50 ml of water 76

brought to boiling and the mixture was then kept on stirring and boiling in an open air for 77

30 min. It was then cooled down to the room temperature and the product, separated by 78

decantation was dried in air at room temperature. 79

SEM-EDS studies were carried out with a Hitachi TM-1000-μ-DeX tabletop scanning 80

electron microscope. TEM investigations was made using CM12 transmission electron 81

microscope (Philips, Holland) with an accelerating voltage of 100 kV. X-ray powder 82

patterns were obtained using a Bruker SMART Apex-II diffractometer operating with 83

MoKα radiation (λ = 0.71073 Å). Bruker Apex-II and EVA software were used for 84

integration and data treatment. Textural characteristics (specific surface area, mean pore 85

size, porous volume) were measured using nitrogen sorption at 77 K on a Quantachrome 86

Autosorb 1 apparatus. 87

The photoluminescence (PL) measurements were carried out using a Fluoromax 4P 88

spectrofluorometer (Horiba) operated in both the fluorescence and the phosphorescence 89

mode. The repetition rate of the xenon flash lamp was 25 Hz, the integration window 90

varied between 300 ms and 3 s, the delay after flash varied between 0.03 and 3ms, and up 91

to 100 flashes were accumulated per data point. The slits were varied from 0.1 nm to 10 92

nm in excitation as well as emission. Time resolved emission spectra (TRES) were 93

recorded using a wavelength tunable NT340 Series EKSPLA OPO (Optical Parametric 94

Oscillator) for samples excitation at 340-360 nm, 457, 488, 460 - 470, and 514 nm 95

operated at 10 Hz as excitation light source and an intensified CCD camera (Andor 96

Technology) coupled to a spectrograph (Shamrock 303i, Andor) as detection system. The 97

TRES were collected using the box car technique. The initial gate delay, (delay after laser 98

pulse, δt) was set to 1 µs and the gate width δt was varied between 10 to 100 µs. The 99

number of accumulations per pulse laser was varied form 50 up to 500. The PL was 100

detected in the spectral range of 500 nm < λem < 750 nm with a spectral resolution of ~0.3 101

nm. PL decays were measured either by using the “decay by delay” feature of the 102

phosphorescence mode of Fluormax 4P either by integrating the time-resolved emission 103

within certain wavelengths interval. The PL decays were analyzed by fitting with a 104

multi-exponential function f(t) using the commercial software (OriginPro 8): f (t) =∑ Ai 105

exp(− t/τi )+ B (Eq.1), where Ai is the decay amplitude, B is a constant (the baseline 106

offset) and τi is the time constant of the decay i. The average PL lifetime was calculated 107

as amplitude-weighted using the formula: ∑∑

=

ii

iii

A A

Aττ . 108

3. Results and discussion 109

Rational design of titania nanostructures represents one of the most importat challenges 110

in the modern physical chemistry. 15 Recent efforts in the development of titania-based 111

anode materials for Li-ion batteries have been concentrated on several promising tracks: 112

in addition to the already mentioned mesoporous architectures, 11,12 arising from 113

macromolecular templation, and the surfactant-free long-time hydrothermal approaches 114

commonly followed by firing, 16 the techniques involving agents facilitating 115

crystallization, such as fluoride ions have been employed. 17 It has also been noticed that 116

especially nanorod and nanobelt micro morphologies were beneficial for the performance 117

and stability of the titania anodes. 18,19 Even the amorphous nanorods were showing good 118

performance, demonstrated recently for Na-ion batteries. 20 In a number of cases the 119

introduction of conductive filler like graphite or, especially, graphene or graphene oxide 120

18,19,21,22 appeared extremely beneficial, possibly, due to improvement of the electric 121

contacts between the oxide grains. Even after the removal of the carbonaceous template, 122

like in the cases, where collagen fibers of paper filter materials were used to insure 123

nanorod-like morphology, the material remained more efficient and robust in its 124

electrochemical performance. 19,23 125

It was thus plausible that creation of a stable nanorod morphology could act as key factor 126

in producing an efficient TiO2-anode. In the view of the earlier observed topotactic 127

transformation of solid metal alkoxides into highly porous mesostructured 128

nanoarchitectures when the precursors were immersed into boiling water, it appeared 129

attractive to investigate the mechanism of this transformation. It appeared also highly 130

realistic that obtaining such porous nanorod constructions could offer an attractive and 131

very facile approach to highly stable performant, and, due to their high crystallinity, also 132

highly stable anode materials for Li-ion batteries. 133

The material investigated in this work was produced by an extremely simple one-step 134

approach using just a short time boiling in pure water of a metal-organic precursor – 135

titanium methoxide that naturally crystallizes in the form of nanorods. This technique 136

provides easily packed TiO2 nanorods composed of strongly aggregated fully crystalline 137

anatase nanoparticles (Fig. 1) through topotactic hydrolytic transformation14 of the 138

precursor alkoxide crystals. 139

140

[Fig 1] 141

142

In order to provide deeper insight in the formation mechanism of these nanostructures we 143

have carried out an attempt to dope the material with Eu3+-cations to be used as 144

luminescence probe. Insertion was carried out via hydrolysis of precursor using aqueous 145

solution of europium nitrate with the total amount corresponding to 2 mol% content in 146

relation to the sum of cations. The doping did not influence the general appearance of the 147

samples in SEM. XRD showed the presence of anatase as the only distinguishable 148

crystalline phase. Figure 2 illustrates the emission and excitation properties of the non-149

heated and 500 °C heated Eu doped TiO2. From the relative intensities of the broad titania 150

absorption band at ~ 340 nm and f-f absorption lines of Eu measured before and after 151

heating to 500°C, it appears that the titania sensibilization of europium emission is 152

practically absent, although some tiny enhancement was observed for the non-heated 153

sample relative to the heated one (Figure FS1, Upper Inset). 154

155

[Fig 2] 156

157

The non-heat-treated samples displayed rather low degree of sensibilization for the 158

europium bands compared to Eu:TiO2 produced by conventional sol-gel technology 24, 159

indicating that the degree of substitution for Ti atoms in the structure was minor, much 160

less than the total content of the introduced dopant (compare 25). The bands were 161

relatively broad, indicating multiple positions with different arrangement to be occupied 162

by the Eu3+-cations. The heat treatment at 500°C, leads to longer lifetime for 163

luminescence resulting most probably from “burning out” of the OH-groups causing non-164

radiant transformation after excitation (see Fig. 3, TS1). 165

Heating normally also leads to improvement of both intensity and the line width 166

(narrowing) but had in this case the opposite effect: the 5D0-7F0 emission band used as 167

control is clearly broadened (from 1.4 to 2.8 nm for used spectral resolution of 0.3 nm) 168

and shifted to a shorter wavelength (Figures FS2, FS3). 169

170

[Fig 3] 171

172

Such transformation would be in agreement with the transfer of the cations to an even 173

larger number of sites with in general slightly improved symmetry. Thus we conclude 174

that the dopant is incorporated exclusively into the periphery of the TiO2 crystallites 175

(surface sites or the residual amorphous shell) and is transferred on heat-treatment into 176

better crystallized domains of, presumably, impurity phases like Eu2O3 or Eu2Ti2O7. The 177

intensity of luminescence is considerably decreased after heat-treatment also confirming 178

that the Eu cations are mostly not situated within the TiO2 lattice but in the parasite Eu-179

rich phases, where quenching of the radiant transition takes place. It is well known that 180

the cross- relaxation process between identical Eu3+ ions results in reducing the 181

5D1 emission with increasing Eu concentration. The narrowing of the emission spectra at 182

selected excitation wavelenghts for the heated sample (spectra not shown) as well as the 183

comparison of the 5D1-related emission and decays (Fig. 4) support for more 184

heterogeneous distribution of Eu3+ in the heated sample, including partial substitution of 185

Ti4+ by Eu3+ dopants and parasite Eu-rich phases (please, compare 25). 186

187

[Fig 4] 188

189

The obtained photoluminescence data reveal the formation mechanism for the produced 190

mesoporous anatase material, confirming the hypothesis presented in14: the anatase 191

particles are nucleating through a topotactic hydrolysis-polycondensation process rapidly 192

developing from the surface of the precursor alkoxide crystals inside them. The activity 193

of water attacking the faces of the crystal submerged into an aqueous medium is greatly 194

higher than that of the Eu3+-cations. No heterometallic species are formed on the 195

nucleation step and thus the primary oxometallate-type nuclei with anatase core 196

(MTSALs) 26,27 are simply free from europium dopant. The incorporation of the dopant 197

into the structure occurs later when Eu-cations start to diffuse into the amorphous shells 198

of the already formed primary particles. Thermal treatment initiates improved 199

crystallization and phase separation of the Eu-rich inter-particle domains and results in 200

the Eu-rich parasite phases with decreased intensity of luminescence as the result. The 201

same effect can be traced in the photoluminescence decay graphs (Figs.3a and 3b, the 202

decay times are provided in TS1). The graph corresponding to non-heat-treated samples 203

(Fig 3a) shows distinctly non-exponential decay, indicating low symmetry and, most 204

probably, multiplicity of the occupied sites. The one for the heat-treated sample (Fig.3b) 205

is more resembling an exponential decay due, supposedly, to improved symmetry of the 206

positions occupied by Eu3+ dopant cations. 207

An important feature that needs to be controlled to insure the possibility of long term 208

exploitation for a battery is the electrochemical performance of the electrode material. 209

The latter indicates potential capacity of the material to resist the loss of performance that 210

can be caused by coalescence of the grains and loss of the active surface in charge-211

discharge cycles. Special efforts have recently been directed at creation of thermally 212

stable TiO2-based electrode materials 28, and, in particular, the hydrothermal stability has 213

recently been a topic of intensive studies. 29 Energy transfer during charge-discharge 214

cycles can in time result in further sintering of the particles with considerable changes in 215

porosity and capacity as a result,5 – the effect much analogous to those of thermal 216

annealing. We have thus investigated the effects of thermal annealing on the crystal 217

structure, surface area and pore diameter in the material produced by rapid hydrothermal 218

synthesis (see Fig. 5). Heating to 500°C with speed of 10°C/min preserved the anatase 219

structure and was associated with the loss of active surface area from on average 250-300 220

m2/g to 110-120 m2/g (see TS2 and FS5) – still an impressive value typical for the best 221

analogs described in recent literature30-32. Heating to 800°C results not unexpectedly in 222

transition into the rutile phase and complete loss of active surface area (to about 2 m2/g) 223

because of sintering of the particles. 224

225

[Fig. 5] 226

227

The non-doped material obtained directly after rapid hydrothermal synthesis from solid 228

Ti(OMe)4 is practically fully crystalline, is completely free from organic impurities and 229

contains strongly aggregated anatase crystallites. Combination of these factors made it 230

potentially attractive as anode material for Li-ion batteries. It is worth noting that the 231

oxide matrix after synthesis contains up to 30 wt% adsorbed water and needs to be 232

dehydrated in vacuum in 6h at 120°C before lithium insertion. 233

The electrodes comprised of 85 wt% TiO2 nanoparticles, 8 wt% acetylene black, and 7 234

wt% polyvinylidene difluoride (PVDF) binder coated on copper foil. The electrochemical 235

properties of the prepared TiO2 anode were evaluated using coin-type cells (2032, 236

Hohsen) assembled in a helium-filled glove box. The Li/TiO2 half cells were cycled 237

between 1 and 2.5V at various current densities. The electrolyte was 1.2M LiPF6 in a 3:7 238

(by volume) mixture of ethylene carbonate and ethyl methyl carbonate. The reported 239

charge/discharge capacities are based on the active mass of TiO2 particles used in the 240

electrodes. 241

Fig. 6a displays the initial two discharge/charge curves of a Li/TiO2 half-cell, cycled 242

between 1 and 2.5 V at various current densities (22 mAg−1 to 112 mAg−1) and Fig 6b 243

shows corresponding capacity vs. cycle number plot. The first cycle discharge capacity 244

for the TiO2 particles is 299 mAh g−1 and the corresponding charging capacity is 225 245

mAh g−1 at a rate of C/15 (22mAg−1). A specific charge capacity of ~186 mAh g–1 was 246

obtained at C/10 rate; this value dropped to 177 and 167 mAh g–1 at C/6, and C/3, 247

respectively. Xu et al. hydrothermally synthesized anatase TiO2 nanotubes and reported 248

reversible capacity of 168 mAh g−1 at 210 mAg−1 current density with 98% coulombic 249

efficiency. 33 Wang et al. already demonstrated that aligned 1D-nanostructured materials 250

have a high surface to- volume ratio, meso–macroporous structure, and excellent activity 251

towards electrochemical lithium storage reactions. 34 The produced TiO2 electrode 252

demonstrated high capacity and stability, matching the best results for non-encapsulated 253

TiO2 reported in recent literature. 30 254

255

[Fig 6] 256

257

4. Conclusions 258

The results of this work permit to demonstrate that the proposed new rapid synthetic 259

approach, whose mechanism was confirmed here by photoluminescence studies, offers a 260

material with superior porosity and thermal stability, consisting of fully crystalline TiO2 261

grains even resistant to the thermal diffusion of Eu3+-cations. Particularly in view of the 262

simplicity of its preparation, it is highly interesting for preparation of anode materials for 263

Li-ion batteries. 264

265

Acknowledgements. The authors express their gratitude to the Swedish Research 266

Council (Vetenskapsrådet) for the Support to the grant “Molecular Precursors and 267

Molecular Models of Nanoporous Materials”. VGP and LA were supported by the Center 268

for Electrical Energy Storage: Tailored Interfaces, an Energy Frontier Research Center 269

funded by the US Department of Energy, Office of Science, Office of Basic Energy 270

Sciences. CT acknowledges the support from the Romanian National Authority for 271

Scientific Research (CNCS-UEFISCDI) (project number PN-II-ID-PCE-2011-3-0534). 272

273

Electronic supplementary includes the detailed information on the photoluminescence 274

(figures FS1-FS3, Table TS1) and thermal stability studies of the produced TiO2 material 275

(Table TS2 and figure FS4). This material is available free of charge via the Internet at 276

http://pubs.acs.org/ 277

References 278

279

(1) Arico, A.S.; Bruce, P.; Scrosati, B.; Tarascon, J.M.; van Schalkwijk, W. Nature 280

Mater., 2005, 4, 366. 281

(2) Goodenough, J.B., Park., K.S. J. Am. Chem. Soc., 2013, 135, 1167. 282

(3) Etacheri, V.; Marom, R.; Elazari, R.; Salitra, G.; Aurbach, D. Energy. Environ. Sci., 283

2011, 4, 3243. 284

(4) Marom, R.; Amalraj, S.F.; Leifer, N.; Jacob, D.; Aurbach, D. J. Mater. Chem., 285

2011, 21, 9938. 286

(5) Liu, H.; Bi, Z.H.; Sun, X.G.; Unocic, R.R.; Paranthaman, M.P., Dai, S.; Brown, 287

G.M. Adv. Mater., 2011, 23, 3450. 288

(6) Fröschl, T.; Hörmann, U.; Kubiak, P.; Kucerova, G.; Pfanzelt, M.; Weiss, C.K.; 289

Behm, R.J.; Hüsing, N.; Kaiser, U.; Landfesterd, K.; Wohlfahrt-Mehrens M. Chem. 290

Soc. Rev., 2012, 41, 5313. 291

(7) Wessel, C.; Zhao, L.A.; Urban, S.; Ostermann, R.; Djerdj, I.; Smarsly, B.M.; Chen, 292

L.Q.; Hu, Y.S.; Sallard, S. Chem. Eur. J., 2011, 17, 775. 293

(8) Armstrong, G.; Armstrong, A.R.; Canales, J.; Bruce, P.G. Chem. Commun., 2005, 294

2454. 295

(9) Armstrong, A.R.; Armstrong, G.; Canales, J.; Garcia, R.; Bruce, P.G. Adv. Mater., 296

2005, 17, 862. 297

(10) Beuvier, T.; Richard-Plouet, M. ; Mancini-Le Granvalet, M. ; Brousse, T. ; 298

Crosnier, O. ; Brohan, L. Inorg. Chem., 2010, 49, 8457. 299

(11) Ren, Y. ; Hardwick, L.J.; Bruce, P.G. Angew. Chem. Int. Ed., 2010, 49, 2570. 300

(12) Saravanan, K.; Ananthanarayanan, K.; Balaya, P. Energy Environ. Sci., 2010, 3, 301

939. 302

(13) Pol, V.G.; Kang, S.H.; Calderon-Moreno, J.M.; Johnson, C.S.; Thackeray, M.M. J. 303

Power Sources 2010, 195, 5039. 304

(14) Seisenbaeva, G.A.; Daniel, G.; Nedelec, J.M.; Gun’ko, Y.K.; Kessler, V.G. J. 305

Mater. Chem., 2012, 22, 20374. 306

(15) Kamat, P. J. Phys. Chem. C, 2012, 116, 11849. 307

(16) Yoon, S.; Manthiram, A. J. Phys. Chem. C, 2011, 115, 9410. 308

(17) Jung, H.G.; Yoon, C.S.; Prakash, J.; Sun, Y.K. J. Phys. Chem. C, 2009, 113, 309

21258. 310

(18) Zhang, X.; Kumar, P.S.; Aravindan, V.; Liu, H.H.; Sundaramurthy, J.; 311

Mhaisalkar, S.G.; Duong, H.M.; Ramakrishna, S.; Madhavi, S. J. Phys. Chem. 312

C, 2012, 116, 14780. 313

(19) Ryu, M.H.; Jung, K.N.; Shin, K.H.; Han, K.S.; Yoon, S. J. Phys. Chem. C, 2013, 314

117, 8092. 315

(20) Xiong, H.; Slater, M. D.; Balasubramanian, M.; Johnson, C. S.; Rajh, T. J. Phys. Chem. 316

Letters, 2011, 2, 2560. 317

(21) Qiu, J.X.; Zhang, P.; Ling, M.; Li, S.; Liu, P.R.; Zhao, H.J.; Zhang, S.Q. ACS Appl. 318

Mater. Interfaces, 2012, 4, 3636. 319

(22) Xin, X.; Zhou, X.F.; Wu, J.H.; Yao, X.Y.; Liu, Z.P. ACS Nano, 2012, 6, 11035. 320

(23) Zhao, B.; Shao, Z.P. J. Phys. Chem. C, 2012, 116, 17440. 321

(24) Luo, M.; Cheng, K.; Weng, W.J.; Song, C.L.; Du, P.Y.; Shen, G.; Xu, G.; Han, 322

G.R. Nanoscale Res. Lett., 2009, 4, 809. 323

(25) Li, J.G.; Wang, X.H.; Watanabe, K.; Ishigaki, T. J. Phys. Chem. B, 2006, 110, 324

1121-1127. 325

(26) Kessler, V.G.; Spijksma, G.I., Seisenbaeva, G.A.; Håkansson, S.; Blank, D.H.A.; 326

Bouwmeester, H.J.M. J. Sol-Gel Sci. Tech., 2006, 40, 163. 327

(27) Seisenbaeva, G.A., Kessler, V.G.; Pazik, R.; Strek, W. Dalton Trans, 2008, 3412. 328

(28) Ortiz, G.F.; Hanzu, I.; Djenizian, T.; Lavela, P.; Tirado, J.L.; Knauth, P. Chem. 329

Mater., 2009, 21, 63. 330

(29) Assaker, K.; Carteret, C.; Durand, P.; Aranda, L.; Stebe, M.J.; Blin, J.L. J. Phys. 331

Chem. C, 2013, in press DOI: 10.1021/jp405517w. 332

(30) Yang, M.C.; Leea, Y.Y.; Xu, B.; Powers, K.; Meng, Y.S. J. Power Sources, 2012, 333

207, 166. 334

(31) Yoon, S.; Manthiram, A. J. Phys. Chem. C, 2011, 115, 9410. 335

(32) Zhang, F.; Zhang, Y.; Song, S.Y.; Zhang, H. J. Power Sources, 2011, 196, 8618. 336

(33) Xu, J.W.; Jia, C.H.; Cao, B.; Zhang W.F. Electrochim. Acta, 2007, 52, 8044. 337

(34) Wang, D.W.; Fang, H.T.; Li, F.; Chen, Z.G.; Zhong, Q.S.; Lu, G.Q.; Cheng H.M. 338

Adv. Funct. Mater., 2008, 18, 3787. 339

340

341

342

Figure legends 343

344

Fig. 1. SEM and TEM views of the rod-shaped anatase microparticles produced by rapid 345

hydrothermal synthesis from rod-shaped Ti(OMe)4 microcrystals. 346

347

Fig. 2. Luminescence spectra for non-heat-treated sample (TE) and the sample heat-348

treated at 500°C for 1 h (TE-500) in the full spectral range (a) and specifically for the 349

5D0-7F0 transition (b). 350

351

Fig. 3 Luminescence lifetimes for non-heat-treated sample (TE) and the sample heat-352

treated at 500°C for 1 h (TE-500) – (a) and (b) respectively. The legend format is 353

λ(excitation)/λ(emission). 354

355

Fig. 4 Comparion between 5D1- related emission spectra and decays of Eu- TiO2 before 356

and after thermal treatment at 500 °C (λex / λem = 464.8 nm / 536 nm). 357

358

Fig. 5 Nitrogen sorption isotherms for the dehydrated material after synthesis (a), the 359

material heat treated in dynamic regime up to 500°C (b) and the material heat-treated at 360

800°C. 361

362

Fig. 6. Cycle performance of the TiO2-based anode material. 363

TOC graphic 364

365

Figure 1 366

367

Figure 2 368

369

Figure 3 370

371

372

Figure 4 373

374

Figure 5 375

0.0 0.2 0.4 0.6 0.8 1.00

2

4

6

8

10

12

14

16a

b

Volu

me

sorb

ed (c

m3 .g

-1)

relative pressure p/p0

c

376

Figure 6. 377

378

379

380

381

C/15 C/10 C/6

C/3C/6 C/10

b

a