High Substitution Rate in TiO2 Anatase Nanoparticles with Cationic

Jun 24, 2015 - Doping is generally used to tune and enhance the properties of metal oxides. However, their chemical composition cannot be readily modi...
0 downloads 11 Views 2MB Size
Download PDF
Article pubs.acs.org/cm

High Substitution Rate in TiO2 Anatase Nanoparticles with Cationic Vacancies for Fast Lithium Storage Wei Li,†,‡ Dario Corradini,†,‡ Monique Body,§ Christophe Legein,§ Mathieu Salanne,†,‡,∥ Jiwei Ma,†,‡ Karena W. Chapman,⊥ Peter J. Chupas,⊥ Anne-Laure Rollet,†,‡ Christian Julien,†,‡ Karim Zhagib,∇ Mathieu Duttine,†,‡,# Alain Demourgues,# Henri Groult,†,‡ and Damien Dambournet*,†,‡ †

Sorbonne Universités, UPMC Univ Paris 06, UMR 8234 PHENIX, 75005 Paris, France CNRS, UMR 8234 PHENIX, 75005 Paris, France § LUNAM Université, Université du Maine, UMR CNRS 6283, Institut des Molécules et des Matériaux du Mans (IMMM), Avenue Olivier Messiaen, 72085 Le Mans Cedex 9, France ⊥ X-ray Science Division, Advanced Photon Source, Argonne National Laboratory, Argonne, Illinois, United States ∇ Energy Storage and Conversion, Institut de Recherche Hydro-Québec, Varennes, Quebec, Canada J3X 1S1 # CNRS, Univ Bordeaux, ICMCB, UPR 9048, F-33600 Pessac, France ∥ Réseau sur le Stockage Electrochimique de l’Energie (RS2E), FR CNRS 3459, 80039 Amiens Cedex, France ‡

S Supporting Information *

ABSTRACT: Doping is generally used to tune and enhance the properties of metal oxides. However, their chemical composition cannot be readily modified beyond low dopant amounts without disrupting the crystalline atomic structure. In the case of anatase TiO2, we introduce a new solution-based chemical route allowing the composition to be significantly modified, substituting the divalent O2− anions by monovalent F− and OH− anions resulting in the formation of cationic Ti4+ vacancies (□) whose concentration can be controlled by the reaction temperature. The resulting polyanionic anatase has the general composition Ti1−x−y□x+yO2−4(x+y)F4x(OH)4y, reaching vacancy concentrations of up to 22%, i.e., Ti0.78□0.22O1.12F0.4(OH)0.48. Solid-state 19F NMR spectroscopy reveals that fluoride ions can accommodate up to three different environments, depending on Ti and vacancies (i.e. Ti3-F, Ti2□1-F, and Ti1□2-F), with a preferential location close to vacancies. DFT calculations further confirm the fluoride/vacancy ordering. When its characteristics were evaluated as an electrode for reversible Li-ion storage, the material shows a modified lithium reaction mechanism, which has been rationalized by the occurrence of cationic vacancies acting as additional lithium hosting sites within the anatase framework. Finally, the material shows a fast discharging/charging behavior, compared to TiO2, highlighting the benefits of the structural modifications and paving the way for the design of advanced electrode materials, based on a defect mediated mechanism.

■

INTRODUCTION Transition-metal oxides are an important class of materials, whose properties are dependent on many factors, including their composition, structure, and morphology. Among them, titanium dioxide (TiO2) is a multifunctional material used for a broad range of applications. The low toxicity and the abundance of titanium have favored the emergence of Tibased compounds for photocatalytic hydrogen production by water splitting, rechargeable batteries/supercapacitors, dyesensitized solar cells, sensors, and biomedical devices.1−6 Over the years, several approaches have been developed to improve its properties. For instance, crystal facets engineering and structural modifications have been widely employed.7−9 The latter approach comprises the introduction of reduced titanium or heteroatoms within the lattice. Although the use of dopants has been proved as an effective way to tune this material’s properties, changing the chemical composition of © XXXX American Chemical Society

TiO2 with a degree of substitution that goes beyond the doping level appears as a challenging and promising way to tailor its properties. On more general grounds, knowing how strongly a metal oxide can be modified while maintaining its original network is of fundamental interest. With the goal of modifying the composition of anatase, which is one of the polymorphs of TiO2, we propose an approach where divalent oxide anions are substituted by monovalent ones such as fluoride and hydroxide. The anatase structure consists of a zigzag assembly of both corner- and edge-sharing TiO6 octahedra (Figure 1a) where the oxide anions are 3-fold coordinated to Ti4+ ions. The substitution of O2− by monovalent anions generally occurs through the Received: April 17, 2015 Revised: June 23, 2015

A

DOI: 10.1021/acs.chemmater.5b01407 Chem. Mater. XXXX, XXX, XXX−XXX

Article

Chemistry of Materials

scattering data, from which it would be suitable to obtain the PDF data, were measured at the 11-ID-B beamline at the Advanced Photon Source located at Argonne National Laboratory. High-energy X-rays (λ = 0.2128 Å) were used, in combination with a large amorphoussilicon-based area detector to collect data to high values of momentum transfer Q ≈ 22 Å−1.14,15 The diffraction images were integrated within fit2D to obtain the one-dimensional diffraction data.16 The G(r) function was extracted from the data using PDFgetX2,17 after correcting for background and Compton scattering. The refinement of the PDF data was performed using the PDFgui software.18 Refined parameters were the instrument parameters, the lattice parameters, the atomic displacement parameters, the anion position, and the cationic site occupancy. 19 F solid-state MAS NMR experiments were performed on a Bruker Avance 300 spectrometer operating at 7.0 T (19F Larmor frequencies of 282.2 MHz), using a 1.3 mm CP-MAS probe head. The roomtemperature 19F MAS spectra of Ti1−x−y□x+yO2−4(x+y)F4x(OH)4y were recorded using a Hahn echo sequence with an interpulse delay equal to one rotor period. The 19F MAS spectra were recorded at various spinning frequencies up to 64 kHz in order to discriminate between isotropic peaks and spinning sidebands. The 90° pulse length was set to 1.55 μs and the recycle delay was set to 20 s. 19F spectra are referenced to CFCl3, and they were fitted by using the DMFit software.19 In order to quantify the fluorine content on the Ti1−x−y□x+yO2−4(x+y)F4x(OH)4y sample, 19F solid-state MAS NMR (Hahn echo) spectra were also recorded for YF3 and LaF3 and the masses of each sample in the rotor were measured. The fits of the spectra allow one to determine the integrated intensities (I) for each sample. Since, for each sample, the recycle delays were chosen to ensure that the amount of signal detected is maximum (420 s for YF3 and 120 s for LaF3), we assume that the integrated intensities are proportional to the number of scans (256 for Ti1−x−y□x+yO2−4(x+y)F4x(OH)4y and 16 for YF3 and LaF3) and to the molar quantity of F atoms (n) in the rotor. This assumption is verified since the calculated I/n ratio for YF3 and LaF3 are equal. The intensities of the NMR signals of Ti1−x−y□x+yO2−4(x+y)F4x(OH)4y (x + y = 0.22), denoted as I1, and YF3 (or LaF3), denoted as I2, allow one to calculate x, using the following formula:

Figure 1. (a) Structure representation of TiO2 anatase (gray = Ti, red = O). (b) TEM micrograph of isolated particle and (c) experimental and calculated PDF profiles (Rw = 19.6%) of Ti0.78□0.22O1.12F0.4(OH)0.48.

formation of reduced Ti3+ in the lattice.10,11 Most of the synthesis methods used so far involved a thermal posttreatment, which prevents a high substitution rate to be achieved.12 Low-temperature chemical methods, as opposed to the ceramic route, favor metastable compositions.13 Here, we report a new solution-based method, allowing concomitant cationic and anionic sublattice modifications of anatase nanoparticles. We demonstrate that, for every four substitutions of oxides by monovalent anions such as fluoride and hydroxide, one cationic vacancy (□) is created. This defect-mediated mechanism, where the defect is either a vacancy or a substitution, can be rationalized by the general chemical formula Ti1−x−y□x+yO2−4(x+y)F4x(OH)4y.

■

⎛ 4xmTi1 − x − y□x + yO2 − 4(x + y)F4xOH4y ⎞ ⎜ M ⎟ ⎝ Ti1 − x − y□x + yO2 − 4(x + y)F4xOH4y ⎠ I1 = ⎛ 3m YF3 ⎞ I2 ⎜ ⎟ ⎝ M YF3 ⎠

MATERIALS AND METHODS

Synthesis of Ti 1− x− y □ x+ y O 2− 4( x+ y ) F 4 x (OH) 4y Anatase. Ti1−x−y□x+yO2−4(x+y)F4x(OH)4y was synthesized using a mild solvothermal process. In a typical experiment, a solution containing 27 mmol of HF (40%) and 25 mL of isopropanol was added to 13.5 mmol of titanium isopropoxide (4 mL) in a Teflon line container. After sealing, the solution was heated at a selected temperature for 12 h. After cooling to room temperature, the white precipitate was separated from the solution using centrifugation (4400 rpm) and washed three times with ethanol. The solid was dried at 80 °C overnight. Note that Ti0.78□0.22O1.12F0.4(OH)0.48 was further outgassed at 150 °C overnight under primary vacuum prior to structural, electrochemical, and thermogravimetric analyses. For comparative study, TiO2 anatase was synthesized using the same solvothermal process without HF. A post-heat treatment was applied at 400 °C under air for 2 h to obtain a pure TiO2 anatase. [Caution: HF is a highly corrosive acid and proper protective equipments are mandatory.] Characterization Methods. Powder X-ray diffraction (XRD) data were collected using a Rigaku diffractometer, equipped with Cu Kα radiation, in a Bragg−Brentano geometry. The transmission electron microscopy (TEM) analysis was performed using a JEOL Model 2010 UHR microscope operating at 200 kV equipped with a TCD camera. The thermogravimetric analysis (TGA) was performed using Setaram Setsys equipment that was coupled with a mass spectrometry (MS) system (Pfeiffer Ominstar), allowing the detection of HF and water departures. Samples were heated from room temperature up to 600 °C under helium atmosphere (heating rate = 5 °C/min). For pair distribution function (PDF) analysis, samples were loaded inside in a Kapton capillary and sealed prior to measurements. X-ray

x=

I1 ⎛ 3m YF3 ⎞ ⎜ ⎟(0.78M Ti I2 ⎝ M YF3 ⎠

4mTi1 − x − y□x + yO2 − 4(x + y)F4xOH4y +

+ 2MO + 0.88MH) I1 ⎛ 3m YF3 ⎞ ⎜ ⎟4(M O I2 ⎝ M YF3 ⎠

+ MH − MF)

where m and M are the mass and the molar mass, respectively. XPS measurements were performed by a VG 220 iXL ESCALAB with a non-monochromatized Mg source (1253.6 eV) at 100 W. The analyzed area had a diameter of ∼200 μm. The insulating character of the powder needed low energy (4−6 eV) electron compensation. Surveys and high-resolution spectra were recorded with pass energies of 150 and 20 eV, respectively. Electrochemical Characterization. The electrodes were prepared by hand-milling of active material (80 wt %), acetylene black (10 wt %) as a conductive agent, and polyvinylidene difluoride (10 wt %) as binder. Electrodes with a diameter of 1 cm and a typical mass of 2 mg were assembled into a Swagelok cell inside a glovebox. Lithium (Aldrich) was used as the second electrode. Copper was used as the current collector. A solution of LiPF6 dissolved in ethylene carbonate and ethyl methyl carbonate (LP30, Merck) was used as the electrolyte. The cells were tested in the voltage range of 1−3 V using different current densities. The GITT measurements were performed using a BST8-WA battery analyzer. Evolution of the equilibrium voltage was collected during the second discharge to eliminate irreversible B

DOI: 10.1021/acs.chemmater.5b01407 Chem. Mater. XXXX, XXX, XXX−XXX

Article

Chemistry of Materials

The resulting Ti0.78□0.22O1.12F0.4(OH)0.48 chemical composition represents a significant deviation from the stoichiometric TiO2 composition, indicating that the anatase framework can accommodate strong local substitutional disorder, while keeping its long-range structural order. The absence of significant deviation on the long-range order observed in the PDF refinement suggests that the vacancies are not locally ordered within the lattice. Nevertheless, on the atomic scale, the occurrence of cationic vacancies induces additional coordination modes for anions. Within pure TiO2 anatase, three Ti atoms surround an O atom. In the presence of vacancies, different cationic environments for the anions become possible. We note them Tii□j-X, where i and j are, respectively, the number of Ti atoms and vacancies surrounding the central anion X (with X = O2−, F−, and OH−). Using highresolution 19F magic-angle-spinning (MAS) solid-state NMR spectroscopy, we were able to probe the local environment of F atoms. The 19F NMR spectrum shows three distinct lines (see Figure 2), indicating the existence of three different environ-

reactions. The cells were intermittently discharged at a C-rate of C/10 (33.5 mA/g) for 20 min, followed by 20 h of relaxation. Density Functional Theory (DFT) Calculations. The DFT ab initio simulations were performed with the CP2K software,20 using the Quickstep algorithm. We used the GGA PBE21 exchange-correlation functional, and we employed the DZVP-MOLOPT-SR-GTH basis set. The Goedecker−Teter−Hutter22−24 pseudo-potentials were used. The electronic orbitals were explicitly represented as follows: for Ti atoms, 3s23p63d24s2; for O atoms, 2s22p4; and for F atoms, 2s22p5. We set a plane wave cutoff of 400 Ry. We added dispersive interactions through the use of the DFT-D3 correction,25 with a cutoff radius of 30 Å. We applied isotropic periodic boundary conditions. Two types of calculations were performed: on one hand, a 10 ps molecular dynamics simulations was run in the NVT ensemble with a target temperature of 300 K on a typical Ti0.78□0.22O1.12F0.88 structure. On the other hand, 0 K cell optimizations with relaxation of atomic positions and cell vectors were performed, starting from a 4 × 4 × 2 anatase TiO2 structure (128 TiO2 units) in which 4 O were replaced by 4 F, and one Ti was removed. The case of two vacancies and 8 O replaced by 8 F was also tested.

■

RESULTS AND DISCUSSION A high substitution rate was achieved by a solution-based method, in which a solution containing Ti(OiPr)4, isopropanol, and aqueous hydrofluoric acid was heated under mild solvothermal conditions (i.e., 90 °C for 12 h). This results in the formation of crystalline nanoparticles (see Figure 1b). A typical powder XRD pattern was indexed using a tetragonal symmetry (space group: I41/amd), which is characteristic of the anatase crystal structure (see Figure S1 in the Supporting Information). The atomic structure was investigated using the Pair Distribution Function (PDF), which yields direct information on the real-space structure.26 The refinement of the PDF was successfully performed using a single phase of anatase (Figure 1c), ruling out the presence of a second amorphous phase. The extracted crystal parameters are similar to those of pure anatase (see Figure S2 and Table S1 in the Supporting Information). The refinement of the Ti site occupancy yields to ca. 74(4)%, indicating the presence of a large amount of cationic vacancies. Strikingly, the concentration of these vacancies is dependent on the synthesis temperature (Figure S3 in the Supporting Information) and can thus be controlled. Attempt to determine an accurate chemical composition was performed by using a combination of analytical and physical characterization tools, namely, PDF, 19F solid-state nuclear magnetic resonance (NMR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and thermogravimetric analysis (see Figures S4−S6 in the Supporting Information). 19F NMR was used to quantify the fluorine content by using reference samples. Thermogravimetric analysis, combined with mass spectrometry, was used to quantify the amount of OH groups. Both NMR and TGA quantitative data were combined with PDF refinement to yield a precise content of cationic vacancies. Finally, XPS allows to identify the tetravalent state of Ti ions. Hence, the following chemical composition was established: Ti0.78□0.22O1.12F0.4(OH)0.48. The large fluorine content indicates that HF reacts with titanium alkoxide through a fluorolysis reaction,27,28 such as Ti(OR)4 + xHF → Ti(OR)4−xFx + xROH. Such a reaction very likely occurs concomitantly with hydrolysis. The presence of OH groups indicates that condensation reactions yielding oxo-bridge were uncompleted. This can be due to the presence of fluoride ions, which renders Ti ions more Lewis acidic, thus strengthening Ti−OH bonds.28

Figure 2. Experimental (in blue) 19F solid-state MAS (60 kHz) NMR spectrum of Ti0.78□0.22O1.12F0.4(OH)0.48 anatase, fitted (in red) with three NMR lines (in black, indicated by vertical dashed lines), which are assigned to Ti1□2-F, Ti2□1-F, and Ti3−F (from left to right) environments. The asterisks indicate the spinning sidebands. Corresponding structures are also shown (gray = Ti, green = F, and red = O; on each snapshot, the illustrated bonds are shown as thick solid cylinders, while the other bonds are shown as thin transparent cylinders).

ments with isotropic chemical shifts of approximately −88, −4, and 98 ppm, assigned to Ti3−F, Ti2□1-F, and Ti1□2-F, respectively (see Part II in the Supporting Information). Moreover, their respective concentrations of 4%, 64%, and 32% point toward a preferential localization of F atoms close to vacancies with 96% of the F atoms being localized close to at least one vacancy. The broad nature of the three NMR lines indicates that Ti−F distances are as widely distributed as in glassy materials, showing that the local structure is strongly disordered. We note that this effect increases with the number of cationic vacancies surrounding the F atoms (see Table S2 in the Supporting Information). C

DOI: 10.1021/acs.chemmater.5b01407 Chem. Mater. XXXX, XXX, XXX−XXX

Article

Chemistry of Materials

rates of 1, 5, and 10 C, the Ti0.78□0.22O1.12F0.4(OH)0.48 electrode can deliver capacities of 204, 185, and 134 mAh/g, respectively, as compared to that obtained for nanoporous anatase (190, 142, and 118 mAh/g, respectively). To test the ability of lithium diffusion within Ti0.78□0.22O1.12F0.4(OH)0.48 at extremely high current density, cells were cycled at a C-rate of 50 C for 300 cycles (Figure 4b). The Coulombic efficiency improves rapidly to almost 100% after few cycles, which is due to the progressive activation of the electrode. After 300 cycles, Ti0.78□0.22O1.12F0.4(OH)0.48 still delivers a specific capacity of 75 mAh/g, proving fast discharging/charging ability. Further electrochemical characterizations of charge/discharge profiles can be found in the Supporting Information (Figure S8). The lithium reaction mechanism that takes place in Ti0.78□0.22O1.12F0.4(OH)0.48 was further investigated to gain insights into its fast discharging/charging ability. The quasiequilibrium voltage obtained by the galvanostatic intermittent titration technique (GITT) shows a smooth curve (Figure 4c), indicating that lithium is inserted into Ti0.78□0.22O1.12F0.4(OH)0.48 via a solid solution mechanism. This is in contrast with the TiO2 phase, which presents a potential−composition plateau (see inset in Figure 4c), corresponding to the well-established two-phase transition yielding the orthorhombic (space group: Imma) Li-rich titanate Li0.5±αTiO2.32,33 The monophasic behavior of Ti0.78□0.22O1.12F0.4(OH)0.48 vs Li was further confirmed by PDF analysis of the discharged electrode (Figure 4d) for which the best refinement was obtained by using the parent anatase model (space group: I41/amd). Details regarding the PDF refinement can be found in the Supporting Information. Furthermore, the lithium insertion might imply cationic vacancies as hosting sites, providing an additional chemical state for the Li ion. The rationalization beyond the observed solid-solution behavior remains challenging as defect chemistry and size effect can both modify the intercalation mechanism.34 Reducing particle size of TiO2 anatase was shown to favor lithium insertion via a solid-solution mechanism.35−37 Nevertheless, for similar particle sizes (i.e., 6−9 nm) reported in this work, TiO2 anatase maintains a two-phase mechanism.35,38 This indicates that cationic vacancies play a central role in the lithium reaction mechanism that takes place in Ti0.78□0.22O1.12F0.4(OH)0.48. Such an hypothesis is supported by recent results obtained from first-principle calculations, demonstrating that the origin of the phase transition occurring in anatase TiO2 arises from a pairing effect of edge-sharing LiO6 octahedra occupying interstitial sites.39 In the present case, cationic vacancies acting as additional vacant sites for hosting lithium can minimize the formation of these edge-shared octahedra, thus causing the absence of the two-phase transition. In addition, the presence of several anions within the network affecting the chemical environment of lithium might also affect the pairing effect. The origin of the observed high rate capability very likely arises from a combination of factors, including size effect and particle orientations.40 Defect chemistry with the incorporation of cationic vacancies very likely plays a major role on the observed high rate capability through increasing the mobility of Li ions by providing additional Li-diffusion paths.41 Moreover, the change of the reaction mechanism from a two-phase transition to a solid-solution behavior might contribute to the improved performance.42

To gain more insights on the stabilization of cationic vacancies through a possible vacancy/fluorine ordering, static DFT calculations (see the Supporting Information) were performed on a Ti127□1F4O252 supercell, i.e., where four O2− are replaced by four F−, creating only one vacant Ti site. We then compared the relative energies of five different O/F arrangements. The configuration where the vacancy is surrounded by six O2− was used as an energy reference. The effect of the local anionic environment on the energy was then monitored by progressively substituting O2− by F−. The results (Figure 3) indicate a continuous stabilization of the vacancy by

Figure 3. DFT calculations performed on a Ti127□1F4O252 supercell containing one vacancy □ (light gray) was built by replacing four O atoms (red) with four F atoms (green). The anionic environment was tuned by adding F in the vicinity of the vacancy. In all cases, the F atoms that were not placed around the vacancy substituted for random O atoms in the lattice and, thus, had the Ti3-F environment. The sticks in the snapshots link the central vacancies to their neighboring atoms (only the six atoms around the vacancy are drawn). When F is progressively localized around the vacancy, we observed a strong decrease in the energy, with respect to the case where only O surrounds the vacancy.

F atoms; the most stable configuration corresponding to the four F anions is located within the equatorial plane of the vacancy. This explains the preferential location of F close to vacancies, as suggested by solid-state 19F NMR. Consistent results are found for the Ti1□2-F environment (see Figure S7 in the Supporting Information). A widely studied potential application of TiO2 is an anode material for lithium-ion batteries.29 Ti-based electrodes can provide enhanced safety, compared to graphite, and must be able to sustain high discharge/charge rates to be used in highpower applications such as electric vehicles.30 Figure 4a displays the comparison of the rate capability between pure TiO2 anatase and Ti0.78□0.22O1.12F0.4(OH)0.48 electrodes from 0.1 to 10 C (note that, for comparison purposes, 1 C = 335 mAh/ g, with regard to the theoretical capacity of TiO2). It is shown that Ti0.78□0.22O1.12F0.4(OH)0.48 exhibits higher capacities at different current densities and a better rate capability, compared to TiO2 anatase, especially at high current densities. At the current density of 1 C, Ti0.78□0.22O1.12F0.4(OH)0.48 electrode delivers a capacity of 204 mAh/g, which is three times higher than that of TiO2. When a 10-fold higher current density (10 C) was applied, only a modest drop in capacity to 134 mAh/g was observed for Ti0.78□0.22O1.12F0.4(OH)0.48, whereas TiO2 anatase is almost inactive toward lithium insertion, suggesting a highly efficient kinetics of lithium diffusion at high current density within Ti0.78□0.22O1.12F0.4(OH)0.48. A comparison with literature data reveals that Ti0.78□0.22O1.12F0.4(OH)0.48 shows an even superior rate capability than nanoporous anatase.31 At CD

DOI: 10.1021/acs.chemmater.5b01407 Chem. Mater. XXXX, XXX, XXX−XXX

Article

Chemistry of Materials

Figure 4. (a) The rate capability of the stoichiometric pure TiO2 anatase and Ti0.78□0.22O1.12F0.4(OH)0.48 electrodes. (b) Cycling behavior of Ti0.78□0.22O1.12F0.4(OH)0.48 at high rate of 50 C (16750 mA/g). (c) Second discharge profiles obtained by GITT measurement for Ti0.78□0.22O1.12F0.4(OH)0.48. Circles represent the equilibrium voltage after 20 h of relaxation. Inset: Open circuit voltage (OCV) obtained by GITT measurement for TiO2 and Ti0.78□0.22O1.12F0.4(OH)0.48. (d) Refinement of the PDF of discharged Ti0.78□0.22O1.12F0.4(OH)0.48 electrode using the anatase-type structure (I41/amd). Inset: structural representation of anatase TiO2 and LiTiO2 (gray: Ti, red: O, green: Li).

■

Notes

CONCLUSION In summary, our results show that a large concentration of cationic vacancies can be synthetically introduced and tuned in anatase. The presence of such vacancies is induced by the negative charge deficiency created by the F−/OH− substitution of O2−. The resulting structural disorder is highlighted by the three coordination modes of F with Ti, with a preferential location of F around the vacancies. This work therefore shows that titanium-based compounds can be drastically modified, achieving metastable composition, while keeping the same crystallographic structure. The ability to tune the vacancy content will likely enable the adjustment of material properties, with respect to the targeted applications. Further synthetic work should allow for an even finer tuning of the chemical composition, providing a new platform to study the impact of the chemical composition on the material properties. Finally, our method is very likely applicable to other technologically relevant metal oxides.

■

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The research leading to these results has received funding from the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (No. FP7/20072013), under REA Grant Agreement No. [321879] (FLUOSYNES). We also thank Hydro-Québec and UPMC for support. C. Labrugère and S. Casale are acknowledged for XPS and HRTEM measurements. The work done at the Advanced Photon Source, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science by Argonne National Laboratory, was supported by the U.S. DOE under Contract No. DE-AC02-06CH11357.

■

(1) Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37−38. (2) O’regan, B.; Grätzel, M. A low-cost, high-efficiency solar-cell based on dye-sensitized colloidal TiO2 films. Nature 1991, 353, 737− 740. (3) Grätzel, M. Photoelectrochemical cells. Nature 2001, 414, 338− 344. (4) Kavan, L.; Grätzel, M.; Gilbert, S. E.; Klemenz, C.; Scheel, H. J. Electrochemical and Photoelectrochemical Investigation of SingleCrystal Anatase. J. Am. Chem. Soc. 1996, 118, 6716−6723. (5) Khan, S. U. M.; Al-Shahry, M.; Ingler, W. B. Efficient photochemical water splitting by a chemically modified n-TiO2. Science 2002, 297, 2243−2245.

ASSOCIATED CONTENT

S Supporting Information *

Additional information involving structural and electrochemical characterization, as well as DFT simulations. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.5b01407.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. E

DOI: 10.1021/acs.chemmater.5b01407 Chem. Mater. XXXX, XXX, XXX−XXX

Article

Chemistry of Materials (6) Chen, X.; Mao, S. S. Titanium dioxide nanomaterials: Synthesis, properties, modifications, and applications. Chem. Rev. 2007, 107, 2891−2959. (7) Yang, H. G.; Sun, C. H.; Qiao, S. Z.; Zou, J.; Liu, G.; Smith, S. C.; Cheng, H. M.; Lu, G. Q. Anatase TiO2 single crystals with a large percentage of reactive facets. Nature 2008, 453, 638−641. (8) Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. VisibleLight Photocatalysis in Nitrogen-Doped Titanium Oxides. Science 2001, 293, 269−271. (9) Chen, X.; Liu, L.; Yu, P. Y.; Mao, S. S. Increasing Solar Absorption for Photocatalysis with Black Hydrogenated Titanium Dioxide Nanocrystals. Science 2011, 331, 746−750. (10) Yu, J. C.; Yu, J. G.; Ho, W. K.; Jiang, Z. T.; Zhang, L. Z. Effects of F− doping on the photocatalytic activity and microstructures of nanocrystalline TiO2 powders. Chem. Mater. 2002, 14, 3808−3816. (11) De Angelis, F.; Di Valentin, C.; Fantacci, S.; Vittadini, A.; Selloni, A. Theoretical Studies on Anatase and Less Common TiO2 Phases: Bulk, Surfaces, and Nanomaterials. Chem. Rev. 2014, 114, 9708−9753. (12) Czoska, A. M.; Livraghi, S.; Chiesa, M.; Giamello, E.; Agnoli, S.; Granozzi, G.; Finazzi, E.; Di Valentin, C.; Pacchioni, G. The nature of defects in fluorine-doped TiO2. J. Phys. Chem. C 2008, 112, 8951− 8956. (13) Feldmann, C. Metastable SolidsTerra Incognita Awaiting Discovery. Angew. Chem., Int. Ed. 2013, 52, 7610−7611. (14) Chupas, P. J.; Qiu, X. Y.; Hanson, J. C.; Lee, P. L.; Grey, C. P.; Billinge, S. J. L. Rapid-acquisition pair distribution function (RA-PDF) analysis. J. Appl. Crystallogr. 2003, 36, 1342−1347. (15) Chupas, P. J.; Chapman, K. W.; Lee, P. L. Applications of an amorphous silicon-based area detector for high-resolution, highsensitivity and fast time-resolved pair distribution function measurements. J. Appl. Crystallogr. 2007, 40, 463−470. (16) Hammersley, A. P.; Svensson, S. O.; Hanfland, M.; Fitch, A. N.; Hausermann, D. Two-dimensional detector software: From real detector to idealised image or two-theta scan. High Pressure Res. 1996, 14, 235−248. (17) Qiu, X.; Thompson, J. W.; Billinge, S. J. L. PDFgetX2: A GUIdriven program to obtain the pair distribution function from X-ray powder diffraction data. J. Appl. Crystallogr. 2004, 37, 678. (18) Farrow, C. L.; Juhas, P.; Liu, J. W.; Bryndin, D.; Bozin, E. S.; Bloch, J.; Proffen, T.; Billinge, S. J. L. PDFfit2 and PDFgui: Computer programs for studying nanostructure in crystals. J. Phys.: Condens. Matter 2007, 19, 335219. (19) Massiot, D.; et al. Modelling one- and two-dimensional solidstate NMR spectra. Magn. Reson. Chem. 2002, 40, 70−76. (20) CP2K Developers Group; see http://www.cp2k.org. (21) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (22) Goedecker, S.; Teter, M.; Hutter, J. Separable dual-space Gaussian pseudopotentials. Phys. Rev. B 1996, 54, 1703−1710. (23) Hartwigsen, C.; Goedecker, S.; Hutter, J. Relativistic separable dual-space Gaussian pseudopotentials from H to Rn. Phys. Rev. B 1998, 58, 3641−3662. (24) Krack, M. Pseudopotentials for H to Kr optimized for gradientcorrected exchange-correlation functionals. Theor. Chem. Acc. 2005, 114, 145−152. (25) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H−Pu. J. Chem. Phys. 2010, 132. (26) Billinge, S. J. L.; Levin, I. The problem with determining atomic structure at the nanoscale. Science 2007, 316, 561−565. (27) Kemnitz, E.; Groß, U.; Rüdiger, S.; Shekar, C. S. Amorphous Metal Fluorides with Extraordinary High Surface Areas. Angew. Chem., Int. Ed. 2003, 42, 4251−4254. (28) Rudiger, S.; Kemnitz, E. The fluorolytic sol−gel route to metal fluoridesA versatile process opening a variety of application fields. Dalton Trans. 2008, 1117−1127.

(29) Wagemaker, M.; Mulder, F. M. Properties and Promises of Nanosized Insertion Materials for Li-Ion Batteries. Acc. Chem. Res. 2013, 46, 1206−1215. (30) Ren, Y.; Hardwick, L. J.; Bruce, P. G. Lithium Intercalation into Mesoporous Anatase with an Ordered 3D Pore Structure. Angew. Chem., Int. Ed. 2010, 49, 2570−2574. (31) Shin, J.-Y.; Samuelis, D.; Maier, J. Sustained Lithium-Storage Performance of Hierarchical, Nanoporous Anatase TiO2 at High Rates: Emphasis on Interfacial Storage Phenomena. Adv. Funct. Mater. 2011, 21, 3464−3472. (32) Cava, R. J.; Murphy, D. W.; Zahurak, S.; Santoro, A.; Roth, R. S. The crystal structures of the lithium-inserted metal oxides Li0.5TiO2 anatase, LiTi2O4 spinel, and Li2Ti2O4. J. Solid State Chem. 1984, 53, 64−75. (33) Wagemaker, M.; Kearley, G. J.; van Well, A. A.; Mutka, H.; Mulder, F. M. Multiple Li Positions inside Oxygen Octahedra in Lithiated TiO2 Anatase. J. Am. Chem. Soc. 2003, 125, 840−848. (34) Gibot, P.; Casas-Cabanas, M.; Laffont, L.; Levasseur, S.; Carlach, P.; Hamelet, S.; Tarascon, J. M.; Masquelier, C. Room-temperature single-phase Li insertion/extraction in nanoscale LixFePO4. Nat. Mater. 2008, 7, 741−747. (35) Wagemaker, M.; Borghols, W. J. H.; Mulder, F. M. Large Impact of Particle Size on Insertion Reactions. A Case for Anatase LixTiO2. J. Am. Chem. Soc. 2007, 129, 4323−4327. (36) Shen, K.; Chen, H.; Klaver, F.; Mulder, F. M.; Wagemaker, M. Impact of Particle Size on the Non-Equilibrium Phase Transition of Lithium-Inserted Anatase TiO2. Chem. Mater. 2014, 26, 1608−1615. (37) Patra, S.; Davoisne, C.; Bruyère, S.; Bouyanfif, H.; Cassaignon, S.; Taberna, P.-L.; Sauvage, F. Room-Temperature Synthesis of High Surface Area Anatase TiO2 Exhibiting a Complete Lithium Insertion Solid Solution. Part. Part. Syst. Charact. 2013, 30, 1093−1104. (38) Sudant, G.; Baudrin, E.; Larcher, D.; Tarascon, J.-M. Electrochemical lithium reactivity with nanotextured anatase-type TiO2. J. Mater. Chem. 2005, 15, 1263−1269. (39) Morgan, B. J.; Watson, G. W. Role of Lithium Ordering in the LixTiO2 Anatase → Titanate Phase Transition. J. Phys. Chem. Lett. 2011, 2, 1657−1661. (40) Liu, G.; Yin, L.-C.; Pan, J.; Li, F.; Wen, L.; Zhen, C.; Cheng, H.M. Greatly Enhanced Electronic Conduction and Lithium Storage of Faceted TiO2 Crystals Supported on Metallic Substrates by Tuning Crystallographic Orientation of TiO 2 . Adv. Mater. 2015, DOI: 10.1002/adma.201500198. (41) Hahn, B. P.; Long, J. W.; Rolison, D. R. Something from Nothing: Enhancing Electrochemical Charge Storage with Cation Vacancies. Acc. Chem. Res. 2013, 46, 1181−1191. (42) Liu, H.; Strobridge, F. C.; Borkiewicz, O. J.; Wiaderek, K. M.; Chapman, K. W.; Chupas, P. J.; Grey, C. P. Capturing metastable structures during high-rate cycling of LiFePO4 nanoparticle electrodes. Science 2014, 344, 1252817.

F

DOI: 10.1021/acs.chemmater.5b01407 Chem. Mater. XXXX, XXX, XXX−XXX