Enhanced lithium ion storage in TiO2 nanoparticles, induced by sulphur and carbon co-doping

Journal of Power Sources 326 (2016) 270e278

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Enhanced lithium ion storage in TiO2 nanoparticles, induced by sulphur and carbon co-doping Svetlozar Ivanov a, *, Adriana Barylyak b, Khrystyna Besaha c, Anna Dimitrova d, Stefan Krischok d, Andreas Bund a, Jaroslav Bobitski c, e €t Ilmenau, Gustav-Kirchhoff-Straße 6, 98693, Ilmenau, Germany Electrochemistry and Electroplating Group, Technische Universita Danylo Halitsky Lviv National Medical University, Pekarska Str.69, 79010, Lviv, Ukraine c Department of Silicate Engineering, Lviv Polytechnic National University, S. Bandery Str. 12, 79013, Lviv, Ukraine d €t Ilmenau, PF 100565, 98684, Ilmenau, Germany Institute of Physics and Institute of Micro- and Nanotechnologies, Technische Universita e Faculty of Mathematics and Natural Sciences, University of Rzeszow, Pigonia Str.1, 35959, Rzeszow, Poland a

b

h i g h l i g h t s
Anatase nanoparticles are simultaneously doped by sulphur and carbon.
In SC-TiO2 material carbon exists in elemental and oxide forms.
Sulphur is integrated in form of sulfate compounds in þ6 oxidation state.
SC-TiO2 shows significant capacity increase and very fast lithiation kinetics.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 April 2016 Received in revised form 20 June 2016 Accepted 27 June 2016

Sulphur and carbon codoped anatase nanoparticles are synthesized by one-step approach based on interaction between thiourea and metatitanic acid. Electron microscopy shows micrometer-sized randomly distributed crystal aggregates, consisting of many 25e40 nm TiO2 nanoparticles. The obtained phase composition and chemical states of the elements in the structure are analyzed by means of X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). XRD shows that after doping the tetragonal anatase structure is preserved. Further data assessment by Rietveld refinement allows detection of a slight increase of the c lattice parameter and volume related to incorporation of the doping elements. XPS confirms the coexistence of both elemental and oxide carbon forms, which are predominantly located on the TiO2 particle surface. According to XPS analysis sulphur occupies titanium sites and the element is present in S6þ sulfate environment. Analysis based on cyclic voltammetry and galvanostatic intermittent titration (GITT) suggests an accelerated Liþ transport in the doped TiO2 structure. The synthesized S and C co-doped anatase has an excellent electrochemical performance in terms of capacity and very fast lithiation kinetics, superior to the non-doped TiO2. The material displays 83% capacity retention for 500 galvanostatic cycles and nearly 100% current efficiency. © 2016 Elsevier B.V. All rights reserved.

Keywords: TiO2 nanoparticle Methatitanic acid Thiourea Li ion battery Sulphur-carbon doping X-ray photoelectron spectroscopy

1. Introduction Anatase, the most extensively used polymorph modification of TiO2, has been widely researched as an alternative anode material for lithium-ion batteries [1e10]. Due to its excellent electrochemical stability, high operating voltage and low cost TiO2 based anode materials attracted high attention. A considerable number of

* Corresponding author. E-mail address: [email protected] (S. Ivanov). http://dx.doi.org/10.1016/j.jpowsour.2016.06.116 0378-7753/© 2016 Elsevier B.V. All rights reserved.

studies underlined the positive impact of TiO2 for battery safety [5,8,9]. The higher TiO2 lithiation potential reduces the risk of metallic Li dendrites formation and thermal runaway [5,8]. Furthermore, the minimal heat exchange upon charge-discharge [9] and lack of solid-electrolyte interfacial (SEI) layer formation enhance the positive effect of TiO2 for the battery safety and reliability [9]. It is generally accepted that the main weakness of TiO2 concerns its relatively low conductivity and hindered kinetics of Liion transport. These drawbacks result in a poor rate capability and a limited capacity of TiO2 based electrodes [10].

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

Three basic approaches, including particle size reduction and nanostructuring [11], application of conducting coatings and composites with conducting additives [12], and appropriate doping of the structure [13e30] have been established as the most promising for boosting TiO2 electrochemical performance. Significant improvement of TiO2 functional parameters based on the reduction of Li-ion diffusion length has been achieved by nanostructuring [11]. Further enhancement of Li ion mobility and storage was realized on atomic level by means of integration of carbon [13,14,28], nitrogen [15e17,28], hydrogenation [18] and introducing other elements [19e30]. Sulphur doping during TiO2 chemical synthesis was initially applied to modify the photocatalytic properties of the material [24e26]. The narrowing of the electronic band gap of TiO2 is of great importance for improving the optical and photocatalytic properties of this material in the visible wavelength range [25,26]. It was concluded that the effective doping of TiO2 with non-metal elements like sulphur, creates strongly delocalized impurity states essential for a good mobility of the photogenerated holes and minimization of electron-hole recombination [26]. Furthermore, the doping of TiO2 with additional elements, including sulphur, considerably increases the conductivity of the material, required as well for the improvement of lithiation/delithiation kinetics [27] and for the enhancement of its electrochemical performance in water based media [26]. Along with several nonmetallic dopants (including C, N, F), sulphur doping has resulted in an effective electrochemical storage and improvement of lithiation kinetics [27]. The synthetic approach involves a three step method, comprising nitriding in gaseous NH3 of already synthesized TiO2 nanoparticles, followed by oxidation in air and final S doping in gaseous H2S. Even though a rather complex synthetic procedure has been used, a visible improvement of the capacity and rate capability of the material was achieved [27]. It was further observed that the distribution of dopant elements in the TiO2 lattice changes the electronic structure and therefore gives an impact to the electron and ion transport in the solid state. The homogeneous distribution of dopants in the entire TiO2 particle is essential for attaining a good conductivity and improved rate performance of the material. Considering the occupation site of the doping element in the TiO2 structure, it is very important that nonmetallic dopants to be substitutional for a lattice atom instead to have an interstitial position in TiO2. The dopants, located in the interstitial spaces between TiO6 octahedra, can block the intercalation of lithium ions and hence inhibit the electrochemical kinetics of lithium exchange [27]. Using first principles calculations, structural, electronic and optical properties of S-doped anatase are studied [31]. Different states of sulphur (anionic and cationic) are considered depending on their position in TiO2 lattice: in interstitial site and in substitution for either oxygen or titanium atoms. It was experimentally observed that among the explored structures, two anionic and one cationic configurations induce an improved light absorption in the visible region. It was found that the cationic sulphur configurations S4þ and S6þ are strongly stabilized in a wide range of oxygen chemical potential, while anionic species exist only at very low chemical potential values. Systems involving sulphur on Ti position, incorporated in the form of SO2 units are expected to be thermodynamically more stable and with improved optical properties. Considering possible synergetic effect caused by a second doping element, DFT analysis revealed a positive influence of the carbon doping on the electronic and photocatalytic properties of TiO2. The observed carbon effect is attributed to the inhibition of impurity states localized in the middle of the band gap by the single sulphur doping [32]. Taking into consideration the complex structural aspects of the

271

sulphur TiO2 doping, their impact on the lithiation mechanism and kinetics of the structure remains still controversial. Current work concentrates on sulphur and carbon co-doped TiO2 nanoparticles (SC-TiO2), obtained by a direct procedure based on interaction between thiourea and metatitanic acid. The synthesis method is an example of a simple and effective one-step strategy for boosting the capacity and cycling rate performance of anatase electrodes. The study aims at revealing the structural characteristics of the doping induced lithium ion storage in anatase. 2. Experimental 2.1. Chemicals and materials Ethylene carbonate (EC), dimethyl carbonate (DMC), LiPF6 and Li metal foil were supplied by Alfa Aesar. Thiourea was provided by Wako Pure Chemical Industry. The solvents and electrolytes necessary for electrochemical experiments were dried until a value of 15 ppm H2O was reached. The moisture in the electrolytes was controlled by Karl e Fischer titration (831 KF Coulometer from Metrohm). A Cu foil as a current collector was supplied by Li Tec. Other chemicals were obtained from commercial sources and were used without further purification. 2.2. Synthesis of TiO2eSC nanoparticles A mixture of 10.4 g metatitanic acid and 3.6 g thiourea was triturated in an agate mortar to obtain a homogeneous mass, which was heated up at 500  C for 1 h. Thus, a yellow powder was synthesized using a solid phase method. 2.3. Scanning electron microscopy (SEM), Transmission electron microscopy (TEM), X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) A high-resolution scanning electron microscope Hitachi S-4800 II was used for the surface morphology characterisation. JEOL-JEM1011 TEM microscope, operated at an accelerating voltage of 80 kV with a resolution of 0.2 nm, was used to obtain images of individual nanoparticles. The powder samples were prepared by air-drying a drop of a sonicated suspension onto copper grids. The phase identification of sulphur doped TiO2 structure was carried out by powder X-ray, using a Siemens D5000 diffractometer in reflection mode with Cu Ka radiation. XRD patterns were recorded in the 2Ɵ range of [15; 100 ] with a step size of 0.02 and a stay time of 1s/step. XPS measurements were carried out in normal emission using monochromatic AlKa (hn ¼ 1486.7 eV) radiation. More details about the experimental setup can be found in Ref. [33]. Core level spectra were recorded at constant pass energy with a total energy resolution of 0.6 eV and at absence of charge neutralization and further binding energy (BE) correction. All core level spectra were analyzed by subtracting a linear background and peak positions. The areas were obtained by a least-squares fitting of model curves (70% Gaussian, 30% Lorentzian) to the experimental data. The data were fitted using CasaXPS (Version 23.16 Dev52, Casa Software Ltd., www.casaxps.com). For calculation of the atomic concentration a homogeneous distribution of the elements within the XPS detected volume was assumed. The data for photoemission cross sections and asymmetry factors were obtained from the work of Yeh and Lindau [34]. The error in the quantitative analysis is estimated in the range of ±10%, while the accuracy for BE assignment is ±0.2 eV.

272

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

2.4. Electrode preparation and electrochemical measurements In order to improve the electrochemical performance of the lithium ion cells, the slurry composition and mixing were optimized. The best composition used to prepare slurries for qualitative coatings with stable performance is 80 wt % active material, 10 wt % polyvinylidene fluoride (PVDF) dissolved in n-methyl-2 pyrrolidinone (NMP) used as a binder and 10 wt % carbon black used as a conductive additive. The obtained slurry was deposited on Cu current collector by automated slurry applicator instrumentation. Afterwards, the obtained deposit was dried at 50  C for 4 h in air. As a next step, the electrodes were cut in size compatible for a standard CR2016 coin cells. Prior to cell assembly, the electrodes were weighed and dried under vacuum at 100  C for 12 h. The electrolyte solution used for the electrochemical experiments consists of 1 M LiPF6 dissolved in EC: DMC (1:1 w.%). Li foil was used as counter and reference electrode. Celgard polyethylene foil was used as separator. The water content in the electrolyte (15 ppm H2O) was monitored by means of Karl Fischer titration. The coin cells were assembled in a glove box under a dry and high purity argon atmosphere. All electrochemical measurements were carried out using a BioLogic multichannel potentiostat/galvanostat VMP3 in three electrode Swagelok® cell configuration with lithium counter and reference electrodes, under argon atmosphere. The charge and discharge performance of the material was characterized galvanostatically over a potential range between 2.5 and 1.0 V vs. Li/Liþ. GITT was performed in a three-electrode Swagelok cell, equipped with Li metal counter and reference electrodes. The procedure combined a multiple constant current (I ¼ 7.14 mA; t ¼ 600 s) and open circuit (I ¼ 0; t ¼ 6600 s) measurement steps. 3. Results and discussion 3.1. Formation and structural characterisation of SC-TiO2 nanoparticles The synthesis of SC-TiO2 was performed by heat treatment of a mixture of thiourea and metatitanic acid. Thermal analysis completed in our previous work revealed the temperature conditions for the material synthesis [35]. The formation of SC-doped anatase phase involves an endothermic process at 175  C corresponding to the dehydration of methatitanic acid, followed by a relatively long crystallization taking place at temperatures above 205  C. In parallel to the anatase formation thiourea decomposes forming NH3, H2S, CS2 and other products. The decomposition is accompanied by a continuous weight loss, lasting up to 500  C and beyond [36]. Based on these results the optimum temperature for the sintering of the samples was set to 500  C, which is the finishing point of crystallization process. The selected thermal conditions allowed an efficient modification of the TiO2 structure by sulphur and carbon provided by the decomposition products of thiourea. The phase composition of the obtained after thermal treatment material was analyzed by means of XRD (Fig. 1). X-ray analysis for both samples showed the existence of only one crystalline phase the tetragonal modification of TiO2 e anatase. The cell unit parameters for the powders of pure TiO2 and SCTiO2 were refined by the Rietveld method (Fig. S1). The results are summarized in Table 1. Compared to pure TiO2, the alterations of the a and c lattice parameters of SC-TiO2 structure were determined to be Da ¼ 0.0007(1) Å and Dc ¼ 0.0022(0) Å. The analysis demonstrates that the lattice parameters remain practically unchanged along a and b axis, however, a slight increase of c lattice parameter and volume due to incorporation of additional elements has been observed for the doped sample. The obtained results for

Fig. 1. XRD patterns of SC-TiO2 (red) and non-doped (blue) nanopowder samples

Table 1 TiO2 crystal lattice parameters assessed by the Rietveld refinement method. Sample

TiO2 SC-TiO2

Lattice parameters a, Å

c, Å

c/a

V, Å3

3.7862(6) 3.7855(5)

9.505(2) 9.527(2)

2.5105 2.5167

136.25(7) 136.52(6)

the doping induced behavior of the lattice parameters are consistent with the state of the art [37]. Nevertheless, the TiO2 doping at atomic level and consequently the induced functional modification of the material depend to a high extent on the location and chemical state of the integrated element [31,38]. In order to further clarify these structural aspects of the doping process analysis of the chemical states and their elemental environment in the SC-TiO2 nanomaterial by means of XPS was performed. The core level spectra of C1s, O1s, Ti2p and S2p are shown in Fig. 2, and the results of the quantitative analysis are reported in Table 2. The presence of Ti4þ is demonstrated by the XPS spectrum of Ti2p, which two peaks at 464.7 eV (2p1/2) and 459.0 eV (2p3/2) are characteristic for TiO2 (Fig. 2a). The corresponding O1s peak appears at 530.2 eV and correlates well with available TiO2 spectroscopic data [39,40]. The O1s component at 531.8 eV can be attributed to CeOeTi or SeOeTi fragments, resulting from carbon and sulphur codoping. This is confirmed by C1s peak at 289.0 eV, which suggests a presence of CeO bonds due to the substitution of Ti atom by C in the TiO2 lattice and formation of TieOeC structure [41e43]. The other two C1s peaks at 285,1 eV and 286,5 eV are characteristic for CeC/H and CeO groups, respectively [34,44,45]. The main maximum at 285.1 eV is attributed to elemental disordered carbon [40], which might originate from in vacuo or external post synthesis contamination and chemical doping during synthesis. Studies in the field have shown that the in vacuo contamination rates are very low compared with the detected carbon amount in our work, displaying in most of the cases a nanometer thin island-type deposit [46]. Therefore, the detected high amount of surface carbon (13.2 at.%) originates probably not from in vacuo contamination. In addition, no peak around 282 eV was detected, indicating the absence of Ti e C bond due to substitution of oxygen atoms by carbon [41,42]. The presence of sulphur was confirmed by two S2p peaks with 2p3/2 levels at 168.5 eV and 169.8 eV, (Fig 2d), which indicates that

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

273

Fig. 2. XPS spectra of the composition elements (Ti, O, C and S) of the codoped SC-TiO2 material. Ti e (A), O e (B), C e (C) and S e (D).

Table 2 Binding energy and atomic elemental concentration evaluated from XPS spectra of CS co-doped TiO2 nanoparticles. Element

BE/eV

Group

At.%.

C1s

285.1 286.5 289.0 530.2 531.8 533.2 459.0 168.5 169.8

CeC/H CeO CeOeTi TiO2 CeOeTi, SeOeTi (C]O), OeCeO TiO2 6þ SO2 4 (S ) 6þ SO2 4 (S )

13.2 2.4 2.4 43.4 9.5 1.3 26.2 1.1 0.5

O1s

Ti2p3/2 S2p3/2

18.0

54.2

26.2 1.6

sulphur is integrated in the form of S6þ [47]. As the previous reports in the literature point out the

substitution of Ti4þ by S6þ is chemically more favorable than replacing O2 by S2, because S2 (1.7 Å) has a considerably larger ionic radius compared to that of O2 (1.2 Å) and as evidenced by XRD there is a small lattice expansion caused by the doping elements [48,49]. Moreover, the energy of TieS bond (418.0 kJ/mol) is less than that of the existing TieO bond (672.4 kJ/mol) [50]. This indicates that sulphur incorporation in CS-TiO2 is favored in the TieOeS configuration, as the O1s core peaks at 528,1 eV suggests, as well. The presence of two sulphur peaks can be assigned to a difference in the nearest neighbor atom. The peak at 169.8 eV originates from S-dopants in the bulk of the material (i.e., TieO-Sd1þ-O-Ti), while the peak at 168.5 eV can be attributed to S-atoms with higher partial positive charge and are incorporated on the surface of the TiO2 nanoparticles (i.e., TieO-Sd2þ-OH), d2þ> d1þ. The quantitative analysis (Table 2) reveals that more sulphur is located at the surface

274

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

then in the bulk of the material. In addition, no peaks around 160e163 eV in the S2p region were detected, suggesting absence of TieS bond formation. The analytical results on the sulphur TiO2 doping are well supported by a recent theoretical investigation of the structural aspects of sulphur incorporation performed by means of first principle calculations [31]. Among the variety of possible sulphur locations and chemical states it was suggested that the stabilization of S6þ in the TiO2 lattice involves a substitution at Ti sites, where SO2 4 structural units with four short SeO bonds are formed. It was further shown that in the case of S6þ stabilization in TiO2 structure, a shift of the absorption edge to the visible spectral range and enhancement of the photocatalytic and electrical properties are not expected [31]. In contrast to these observations, the synthesized in our work material exhibits pronounced photoactivity in the visible wavelength and marked photocatalytic properties [35]. One possible interpretation of this phenomenon is the formation of defects in the TiO2 lattice, induced by the doping. In support of this assumption is the well expressed paramagnetic activity of the doped sample [35,51]. Electron Paramagnetic Resonance (EPR) spectra showed a narrow line with g-factor of 2.004, associated with a single electron trapped oxygen vacancy responsible for the doping induced changes in the electronic properties of TiO2 [51]. Analytical information from the volume of SC-TiO2 powder is acquired by EDX spectroscopy, confirming the presence of much lower amount of the doping elements (0.45% S and 2.38% C), compared to XPS [35]. The observed difference points out that the doping elements are predominantly concentrated at the periphery of the nanoparticles. The morphology of the synthesized SC-TiO2 powder is characterized by electron microscopy. SEM imaging showed micrometersized randomly distributed crystal aggregates (Fig. 3). The high magnification imaging revealed that the observed crystal aggregates consist of many 15e40 nm sized individual TiO2 nanoparticles. High resolution TEM confirmed the size of well-shaped nanocrystals and their affinity to form aggregates (Fig. 4). 3.2. Electrochemical characterisation The electrochemical performance of SC-TiO2 and non-doped TiO2 material in Li-ion containing organic carbonate electrolyte was investigated by means of cyclic voltammetry at different scan rates in a broad potential range (Fig. 5). It is well established that the process of electrochemical Liþ intercalation in the TiO2 lattice can be described as one-electron transfer Liþ insertion reaction by the following equation:


TiO2 þ x Liþ þ e 4Lix TiO2

Fig. 3. SEM images of the SC-TiO2 nanopowder taken at different magnifications: 4500 e A; 80,000 e B; 180,000 e C.

(1)

The highest level of Liþ insertion that can be practically reached for anatase corresponds to an insertion coefficient x z 0.5 [3]. It was shown that upon lithiation anatase undergoes phase transition from lithium-poor tetragonal phase Li0$01TiO2 (space group I41/ amd) to an orthorhombic lithium titanate phase, Li0$5TiO2 (space group Imma) [3]. The voltammetric cycling of SC-TiO2 material shows a stable behavior, exhibiting the characteristic well-shaped electrochemical couple, typical for the anatase TiO2 structure. The observed voltammetric peaks reflect the electrochemical lithiation-delithiation of this type of material, corresponding to the above described phase transition. In general, the peak separation for both TiO2 structures is much larger than for a Nernstian (reversible) redox reaction, indicating (i) a potential drop in the system due to slow electron transport in the nanoporous network or/and ion transport

Fig. 4. TEM image of SC-TiO2 taken at 80 kV accelerating voltage.

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

275

Fig. 5. Cyclic voltammetry of the SC-TiO2 (blue) and non-doped reference material (red), measured at 40 (A) and 80 mV s1 (B). Voltammetric measurements performed at different scan rates. The numbers indicate the scan rate in mV s1. (C) and peak currents linearization vs. scan rate square root (D). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

problems; (ii) slow electrochemical reaction, which follows quasireversible or irreversible kinetics [52]. The comparison between the cyclic voltammetry of SC-TiO2 and non-doped TiO2 at 40 mV s1 shows similar main voltammetric features (Fig. 5a,b). However, detectable decrease of the peak currents due to fading and higher peak separation (e 300 mV) characterize the non-doped material. This effect is even more pronounced at scan rate 80 mV s1, where the peak separation rises to 380 mV. If we ascribe the currents of the voltammetric couple as only due to the depletion/enrichment of Ti4þ states, it is possible to estimate the amount of TiO2 active for energy storage. The integrated charge under anodic delithiation peak corresponds to a mole fraction of TiO2 active for the Li ion intercalation. Whereas the calculated for SC-TiO2 insertion coefficient is about 0.4 (x ¼ 0.384 at v ¼ 40 mV s1 and x ¼ 0.359 at v ¼ 80 mV s1), for the non-doped material much lower values were observed (x ¼ 0.312 at v ¼ 40 mV s1 and x ¼ 0.188 at v ¼ 80 mV s1). Therefore, the obtained from voltammetric data results display more efficient Li ion insertion in the doped SC-TiO2 structure. Further analysis of the Liþ transport kinetics suggests that the ion insertion-extraction is governed in a broad scan rate interval by diffusion phenomena. The latter is confirmed by the linearization of the current maxima obtained from the cyclic voltammograms in coordinates J e v0.5 (Fig. 5c,d). The voltammetric data interpretation allowed extracting the chemical diffusion coefficient D for both SC-TiO2 and the nondoped reference material [49]. The data assessment resulted in about one order of magnitude higher diffusion coefficient for the doped TiO2 (DLiþ SC-TiO2 ¼ 3,1.1016 cm2 s1 and DLiþ 17 cm2 s1), indicating a visible acceleration of the Li TiO2 ¼ 1,8.10 ion transport in the TiO2 solid state, caused by the chemical doping. The results showed a good correlation with previously obtained literature data by means of cyclic voltammetry [52] and chronoamperometry [53], supporting the analysis of Li insertion kinetics in SC-TiO2.

Voltammetric techniques are widely applied for studying solid state insertion reactions. However, the quantitative data interpretation requires a number of approximations related to the high ohmic resistance, pseudocapacitive surface charge accumulation and complicated reaction kinetics. Galvanostatic intermittent titration, a technique combining multiple constant current and open circuit measurement steps, offers an experimental assessment of a number of kinetic and thermodynamic parameters determined with considerable accuracy [54]. In general, the information obtained by GITT can be also attained by using other transient methods. However, by means of GITT the above described experimental issues in other voltammetric techniques can be taken into account and furthermore the kinetic parameters can be analyzed with high precision as a function of composition. GITT measurements for both SC-TiO2 and non-doped TiO2 samples have been carried out and the results for chemical diffusion coefficient as a function of potential are summarized in Fig. 6. The experimental details on GITT can be found in Fig. S2 (Supplementary Information). The diffusion coefficient was determined by using the approach introduced by Weppner and Huggins [54], assuming onedimensional diffusion in a solid solution electrode without consideration of phase transformation. It can be observed that the chemical doping considerably influenced the Li ion transport in the solid state. Depending on the lithiation degree and correspondingly electrochemical potential, diffusion coefficient value varies in the range 8.1011 e 2.1014 cm2 s1 for SC-TiO2 and 5.1011 e 1.1015 cm2 s1 for the non-doped sample. In a broad potential interval the chemically doped SC-TiO2 sample displays a higher diffusion coefficient than the non-doped material. Similar effect was observed for N-doped TiO2, correlating well with our results [16]. Special attention deserves the broad two phase region, where the diffusion coefficient displays a sharp minimum at 1.85 V. This stoichiometric range is characteristic with the coexistence of

276

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

Fig. 6. Apparent diffusion coefficient of SC-TiO2 (red) and non-doped TiO2 (blue), obtained by GITT as a function of open circuit potential at steady-state. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

lithium-poor tetragonal phase and orthorhombic lithium titanate phase. It can be observed that the chemical doping induces a potential range narrowing of the two-phase region and a considerable increase of diffusion coefficient. Nevertheless, the transport phenomena in the two-phase region cannot be interpreted by the approach of Weppner and Huggins, since it refers a solid state diffusion in a single phase media. Therefore, it has to be underlined that the diffusion coefficients evaluated for the mixed-phase region have apparent rather than actual values.

The galvanostatic cycling of the doped and non-doped anatase nanopowders was performed at different constant currents in 1 mol/L LiPF6 EC: DMC (1:1 w/w) in the potential limits 1e2.6 V (Fig. 7a,b). The results showed a good electrochemical stability and flattype galvanostatic potential profile for both doped and nondoped anatase samples. The potential behavior during charge displays an initial maximum, which is more pronounced at higher Crates. The potential maximum at the early stage of charging has been attributed to the high overpotential related to metallic Li dendrites formation [55]. At low C-rates (C/20 e C/5) SC-TiO2 sample displayed discharge capacities in the range of (220e175) mAh g1, a considerably higher values than the theoretically predicted one. This outcome demonstrates that TiO2 in the doped nano-particulate form is able to exchange more than 0.5 lithium ions per structural unit (x > 0.5; C > 168 mAhg1). It can be further seen that at moderate and high cycling rates doped TiO2 sample displays high capacities (152 mAhg1 at 1  C, down to 78 mAhg1 at 20 C) (Fig. 7c). The nondoped sample showed an inferior rate performance in comparison with doped one (60 mAhg1 at 1 C and 9 mAhg1 at 20 C). The similarity of the cycling behavior of both samples in terms of capacities at low rates and the big difference at high currents can be explained by taking into account the two main factors influencing their electrochemical behavior e particles size and chemical doping. The doped and pristine TiO2 nanopowders have a small and comparable particle size, displaying similar electrochemical behavior at low cycling rates. At intermediate and high C-rates however, the higher discharge capacity of SC-TiO2 indicates the positive influence of the chemical doping as a dominating factor for the lithiation performance.

Fig. 7. Galvanostatic curves of SC-TiO2 (A) and non-doped reference material (B), measured at different cycling rates. Charge and discharge capacities of SC-TiO2 (red) and nondoped material (blue) measured at different cycling rates (C). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

277

detectable acceleration of lithiation/delithiation kinetics of the doped SC-TiO2 sample. This effect was displayed in terms of high voltammetric currents and decreased peak separation, characteristic for the doped sample. The evaluated by GITT apparent diffusion coefficient demonstrated increased values for the SC-TiO2 nanopowder, confirming that the S and C doping positively influences the ionic transport in the material. The synthesized S and C codoped anatase showed an excellent electrochemical performance in terms of capacity (initial values of 210 mAh g1) and very fast lithiation kinetics, superior to the nondoped TiO2. The material displayed 83% capacity retention for 500 galvanostatic cycles and nearly 100% current efficiency. Acknowledgements

Fig. 8. Discharge capacities and current efficiencies during long term cycling of SCTiO2 (red) and non-doped material (blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Comparison with the literature data showed that SC-TiO2 nanoparticles present a similar rate performance to boron doped TiO2 nanomaterial [56] and even superior to N/Cr [15] and N/S [27] co-doped TiO2 nanopowders. The capacity improvement of the nanostructured TiO2 was discussed in a number of studies in terms of reduction of Li ion diffusion length, increase of the specific surface area and pseudocapacitive surface charge storage [18,57]. However, the assessment of the volume and surface charge storage from voltammetric data for the doped and non-doped TiO2 material suggested a negligible surface storage impact. The latter indicates that the improvement of the material performance is related to the acceleration of the Li ion transport in the solid state by the SeC chemical doping. In order to test the long term electrochemical behavior of the doped TiO2 nanopowder layers we have performed electrochemical multiple galvanostatic cycling (Fig. 8). The material displayed 83% capacity retention for 500 galvanostatic cycles and nearly 100% current efficiency. 4. Conclusions Simultaneous sulphur and carbon codoping of anatase has been achieved by a direct synthesis based on the interaction between thiourea and methatitanic acid. The effects of C and S chemical doping on the structure of the obtained nanopowder material were analyzed by XRD and XPS. SEM imaging showed micrometer-sized randomly distributed crystal aggregates, consisting of many 25e40 nm TiO2 nanoparticles. The results from XRD showed that after the C/S codoping the stability of the tetragonal anatase phase is preserved. Further data assessment by means of Rietveld refinement demonstrates that the lattice parameters remain practically unchanged along a and b axis, however, a slight increase of c lattice parameter and volume due to incorporation of additional elements has been observed for the doped sample. By means of XPS it was confirmed that carbon exists in both elemental and oxide forms, which are predominantly adsorbed on the TiO2 particle surface. According to XPS analysis sulphur is present in S6þ sulfate environment, evidenced by two S2p3/2 peaks at 168.5 eV and 169.8 eV. The two-component S spectral structure can be assigned to S-atoms located in two bulk and surface positions. XPS of the doped sample do not indicate existence of TieS and CeS chemical bonds. The voltammetric behavior of TiO2 nanopowders showed a

Financial support by the DFG funded priority program WeNDeLIB (project 6: “Linking of model and commercial active materials for lithium ion batteries by in situ determination of thermodynamic and kinetic data”) is gratefully acknowledged. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2016.06.116. References [1] W. Macklin, R. Neat, Performance of titanium dioxide-based cathodes in a lithium polymer electrolyte cell, Solid State Ionics 53 (1992) 694. [2] R. van de Krol, A. Goossens, J. Schoonman, Spatial extent of lithium intercalation in anatase TiO2, J. Phys. Chem. B 103 (1999) 7151e7159.  , I. Exnar, V. Lorenzen, R. Nesper, M. Graetzel, [3] L. Kavan, M. Kalba c, M. Zukalova lithium storage in nanostructured TiO2 made by hydrothermal growth, Chem. Mater. 16 (2004) 477e485. [4] M. Wagemaker, A. van Well, G. Kearley, F. Mulder, The life and times of lithium in anatase TiO2, Solid State Ionics 175 (2004) 191e193. [5] G.-N. Zhu, Y.-G. Wang, Y.-Y. Xia, Ti-based compounds as anode materials for Li-ion batteries, Energy Environ. Sci. 5 (2012) 6652e6667. [6] Y. Jun, J. Park, M. Kang, The preparation of highly ordered TiO2 nanotube arrays by an anodization method and their applications, Chem. Commun. 48 (2012) 6456e6471. [7] P. Bruce, B. Scrosati, J.-M. Tarascon, Nanomaterials for rechargeable lithium batteries, Angew. Chem. Int. Ed. 47 (2008) 2930e2946. [8] T. Djenizian, I. Hanzu, P. Knauth, Nanostructured negative electrodes based on titania for Li-ion microbatteries, J. Mater. Chem. 21 (2011) 9925e9937. [9] D. Choi, D. Wang, V. Viswanathan, I.-T. Bae, W. Wang, Z. Nie, J.-G. Zhang, G. Graff, Jun Liu, Z. Yang, T. Duong, Li-ion batteries from LiFePO4 cathode and anatase/graphene composite anode for stationary energy storage, Electrochem. Commun. 12 (2010) 378e381. [10] H. Han, T. Song, E.-K. Lee, A. Devadoss, Y. Jeon, J. Ha, Y.-C. Chung, Y.-M. Choi, Y.-G. Jung, U. Paik, Dominant factors governing the rate capability of a TiO2 nanotube anode for high power lithium ion batteries, ACS Nano 6 (2012) 8308e8315. [11] D. Deng, M. Kim, J. Lee, J. Cho, Green energy storage materials: nanostructured TiO2 and Sn-based anodes for lithium-ion batteries, Energy Environ. Sci. 2 (2009) 818e837. [12] S.-J. Park, H. Kim, Y.-J. Kim, H. Lee, Preparation of carbon-coated TiO2 nanostructures for lithium-ion batteries, Electrochim. Acta 56 (2011) 5355e5362. [13] J. Xu, Y. Wang, Z. Li, W. Zhang, Preparation and electrochemical properties of carbon-doped TiO2 nanotubes as an anode material for lithium-ion batteries, J. Power Sources 175 (2008) 903e908. [14] H. Kim, S.-H. Yu, Y.-E. Sung, S. Kang, Carbon treated self-ordered TiO2 nanotube arrays with enhanced lithium-ion intercalation performance, J. Alloys Compd. 597 (2014) 275e281. [15] Z. Bi, M. Paranthaman, B. Guo, R. Unocic, H. Meyer, C. Bridges, X.-G. Suna, S. Daia, J. Mater. Chem. A 2 (2014) 1818e1824. [16] Y. Zhang, F. Du, X. Yan, Y. Jin, K. Zhu, X. Wang, H. Li, G. Chen, C. Wang, Y. Wei, Improvements in the electrochemical kinetic properties and rate capability of anatase titanium dioxide nanoparticles by nitrogen doping, Appl. Mater. Interfaces 6 (2014) 4458. [17] G. Hasegawa, T. Sato, K. Kanamori, K. Nakanishi, T. Abe, Synthesis and electrochemical performance of hierarchically porous N-doped TiO2 for Li-ion batteries, New J. Chem. 38 (2014) 1380e1384. [18] Y. Yan, B. Hao, D. Wang, G. Chen, E. Markweg, A. Albrecht, P. Schaaf, Understanding the fast lithium storage performance of hydrogenated TiO2 nanoparticles, J. Mater. Chem. A 1 (2013) 14507e14513.

278

S. Ivanov et al. / Journal of Power Sources 326 (2016) 270e278

[19] Y. Wang, B. Smarsly, I. Djerdj, Niobium doped TiO2 with mesoporosity and its application for lithium insertion, Chem. Mater. 22 (2010) 6624e6631. [20] X. Lü, W. Yang, Z. Quan, T. Lin, L. Bai, L. Wang, F. Huang, Y. Zhao, Enhanced electron transport in Nb-doped TiO2 nanoparticles via pressure-induced phase transitions, J. Am. Chem. Soc. 136 (2014) 419e426. [21] N. Kyeremateng, F. Vacandio, M.-T. Sougrati, H. Martinez, J.-C. Jumas, P. Knauth, T. Djenizian, Effect of Sn-doping on the electrochemical behaviour of TiO2 nanotubes as potential negative electrode materials for 3D Li-ion micro batteries, J. Power Sources 224 (2013) 269e277. [22] C. Yu, Y. Bai, D. Yan, X. Li, W. Zhang, Improved electrochemical properties of Sn-doped TiO2 nanotube as an anode material for lithium ion battery, J. Solid State Electrochem. 18 (2014) 1933e1940. [23] L. Anh, A. Rai1, T. Thi, J. Gim, S. Kim, E.-C. Shin, J.-S. Lee, J. Kim, Improving the electrochemical performance of anatase titanium dioxide by vanadium doping as an anode material for lithium-ion batteries, J. Power Sources 243 (2013) 891e898. [24] H. Wang, J. Lewis, Second-generation photocatalytic materials: anion-doped TiO2, J. Phys. Condens. Matter 18 (2006) 421e434. [25] T. Umebayashi, T. Yamaki, H. Itoh, K. Asai, Band gap narrowing of titanium dioxide by sulfur doping, Appl. Phys. Lett. 81 (2002) 454e456. [26] K. Reddy, B. Baruvati, M. Jayalakshmi, M. Rao, S-, N and C e doped titanium dioxide nanoparticles: synthesis, characterization and redox charge transfer study, J. Solid State Chem. 178 (2005) 3362e3368. [27] W. Jiao, N. Li, L. Wang, L. Wen, F. Li, G. Liu, H.-M. Cheng, High-rate lithium storage of anatase TiO2 crystals doped with both nitrogen and sulfur, Chem. Commun. 49 (2013) 3461e3463. [28] T. Song, U. Paik, TiO2 as an active or supplemental material for lithium batteries, J. Mater. Chem. A 4 (2016) 14e31. [29] J.-Y. Shin, J. Joo, K. Adepalli, D. Samuelis, J. Maier, The effects of n-type doping on lithium storage in TiO2, Phys. Chem. Chem. Phys. 18 (2016) 8963e8970. [30] M. Lübke, I. Johnson, N. Makwana, D. Brett, P. Shearing, Z. Liu, J. Darr, High power TiO2 and high capacity Sn-doped TiO2 nanomaterial anodes for lithium-ion batteries, J. Power Sources 294 (2015) 94e102. [31] M. Harb, P. Sautet, P. Raybaud, Anionic or cationic S-Doping in bulk anatase TiO2: insights on optical absorption from first principles calculations, J. Phys. Chem. C 117 (2013) 8892e8902. [32] Q. Liu, Z. Zhao, Q. Liu, Synergistic effects of nonmetal co-doping with sulfur in anatase TiO2: a DFT þ U study, Phys. Chem. Chem. Phys. 17 (2015) 3426e3434. [33] M. Himmerlich, S. Krischok, V. Lebedev, O. Ambacher, J. Schaefer, Morphology and surface electronic structure of MBE grown InN, J. Cryst. Growth 306 (2007) 6e12. [34] J. Yeh, I. Lindau, Atomic subshell photoionization cross sections and asymmetry parameters: 1 Z 103, Atomic Data Nucl. Data Tables 32 (1985) 1e155. [35] S. Ivanov, A. Barylyak, K. Besaha, A. Bund, Y. Bobitski, R. Wojnarowska-Nowak, I. Yaremchuk, M. Kus-Liskiewicz, Synthesis, characterization, photocatalytic properties of sulfur- and carbon-codoped TiO2 nanoparticles, Nanoscale Res. Lett. 11 (2016) 140, http://dx.doi.org/10.1186/s11671-016-1353-5. [36] M. Fieser, L. Fieser, Reagents for Organic Synthesis, vol. 4, Wiley, New York, 1974, p. 456. [37] H. Sun, Y. Bai, Y. Cheng, W. Jin, N. Xu, Preparation and characterization of visible-light-driven carbon-sulfur-codoped TiO2 photocatalysts, Ind. Eng. Chem. Res. 45 (2006) 4971e4976. [38] Q. Liu, Z. Zha, Q. Liu, Analysis of sulfur modification mechanism for anatase and rutile TiO2 by different doping modes based on GGA þ U calculations, RSC Adv. 4 (2014) 32100e32107. [39] X. Chen, C. Burda, Photoelectron spectroscopic investigation of nitrogen-

doped titania nanoparticles, J. Phys. Chem. B 108 (2004) 15446e15449. [40] M. Sathish, B. Viswanathan, R. Viswanath, C. Gopinath, Synthesis, characterization, electronic structure, and photocatalytic activity of nitrogen-doped TiO2 nanocatalyst, Chem. Mater. 17 (2005) 6349e6353. [41] C. Reilley, D. Everhart, Chemical derivatization in electron spectroscopy for chemical analysis of surface functional groups introduced on low-density polyethylene film, Anal. Chem. 53 (1981) 665e676. [42] Y. Zhang, Z. Zhao, J. Chen, L. Cheng, J. Chang, W. Sheng, C. Hu, S. Cao, C-doped hollow TiO spheres: in situ synthesis, controlled shell 2 thickness, and superior visible-light photocatalytic activity, Appl. Catal. B Environ. 165 (2015) 715e722. [43] H. Wang, Z. Wu, Y. Liu, A simple two-step template approach for preparing carbon-doped mesoporous TiO2 hollow microspheres, J. Phys. Chem. C 113 (2009) 13317e13324. [44] J. Lv, T. Sheng, L. Su, G. Xu, D. Wang, Z. Zheng, Y. Wu, N, S co-doped-TiO2/fly ash beads composite material and visible light photocatalytic activity, Appl. Surf. Sci. 284 (2013) 229e234. [45] F. Beguin, I. Rashkov, N. Manolova, R. Benoit, R. Erre, S. Delpeux, Fullerene core star-like polymersd1. Preparation from fullerenes and monoazidopolyethers, Eur. Polym. J. 34 (1998) 905e915. [46] F. Mangolini, J. McClimon, F. Rose, R. Carpick, Accounting for nanometer-thick adventitious carbon contamination in x-ray absorption spectra of carbonbased materials, Anal. Chem. 86 (2014) 12258e12265. [47] S. Abu Bakar, C. Ribeiro, An insight toward the photocatalytic activity of S doped 1-D TiO2 nanorods prepared via novel route: As promising platform for environmental leap, J. Mol. Catal. A Chem. 412 (2016) 78e92. [48] J. Yu, G. Dai, Q. Xiang, M. Jaroniec, Fabrication and enhanced visible-light photocatalytic activity of carbon self-doped TiO2 sheets with exposed {001} facets, J. Mater. Chem. 21 (2011) 1049e1057. [49] T. Ohno, M. Akiyoshi, T. Umebayashi, K. Asai, T. Mitsui, M. Matsumura, Preparation of S-doped TiO2 photocatalysts and their photocatalytic activities under visible light, Appl. Catal. A Gen. 265 (2004) 115e121. [50] Y. Luo, Comprehensive Handbook of Chemical Bond Energies, CRC Press, Boca Raton, FL, 2007. [51] Y. Wang, C. Feng, M. Zhang, J. Yang, Z. Zhang, Enhanced visible light photocatalytic activity of N-doped TiO2 in relation to single-electron-trapped oxygen vacancy and doped-nitrogen, Appl. Catal. Environ. 100 (2010) 84e90. € m, S. So €dergren, A. Solbrand, H. Rensmo, J. Hjelm, A. Hagfeldt, S.[52] H. Lindsto E. Lindquist, Liþ insertion in TiO2 (Anatase). 2. Voltammetry on nanoporous films, J. Phys. Chem. B 101 (1997) 7717e7722. € m, S. So €dergren, A. Solbrand, H. Rensmo, J. Hjelm, A. Hagfeldt, S.[53] H. Lindstro E. Lindquist, Liþ ion insertion in TiO2 (anatase). 1. Chronoamperometry on CVD films and nanoporous films, J. Phys. Chem. B 101 (1997) 7710e7716. [54] W. Weppner, R. Huggins, Determination of the kinetic parameters of mixedconducting electrodes and application to the system LiSb, J. Electrochem. Soc. 124 (1977) 1569e1578. [55] D. Lin, Y. Liu, Z. Liang, H.-W. Lee, J. Sun, H. Wang, K. Yan, J. Xie, Y. Cui, Layered reduced graphene oxide with nanoscale interlayer gaps as a stable host for lithium metal anodes, Nat. Nanotechnol. (2016), http://dx.doi.org/10.1038/ nnano.2016.32. [56] J.-H. Jeong, D. Jung, E. Shin, E.-S. Oh, Boron-doped TiO2 materials for high-rate lithium ion batteries, J. Alloys Compd 604 (2014) 226e232. [57] S. Ivanov, L. Cheng, H. Wulfmeier, D. Albrecht, H. Fritze, A. Bund, Electrochemical behavior of anodically obtained titania nanotubes in organic carbonate and ionic liquid based Li ion containing electrolytes, Electrochim. Acta 104 (2013) 228e235.