Mn2+ Substitutional Doping of TiO2 Nanoribbons: A Three-Step

Aug 21, 2014 - ... to load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... Analysis of the manganese content by X-ray photoel...
0 downloads 0 Views 6MB Size
Article pubs.acs.org/JPCC

Mn2+ Substitutional Doping of TiO2 Nanoribbons: A Three-Step Approach Polona Umek,*,† Carla Bittencourt,‡ Peter Guttmann,§ Alexandre Gloter,∥ Srečo D. Škapin,† and Denis Arčon†,⊥ †

Jožef Stefan Institute, Jamova cesta 39, SI-1000 Ljubljana, Slovenia Chimie des Interactions Plasma Surface, CIRMAP, University of Mons, 23 Place du Parc, B-7000 Mons, Belgium § Helmholtz-Zentrum Berlin für Materialien und Energie GmbH, Institute for Soft Matter and Functional Materials, Albert-Einstein-Str. 15, D-12489 Berlin, Germany ∥ Laboratoire de Physique des Solides, Université Paris Sud, CNRS UMR 8502, F-91405 Orsay, France ⊥ Faculty of Mathematics and Physics, University of Ljubljana, Jadranska cesta 19, SI-1000 Ljubljana, Slovenia ‡

S Supporting Information *

ABSTRACT: An in situ doping approach was successfully employed for synthesis of Mn2+-doped sodium titanate nanoribbons, which were used as a precursor for preparation of TiO2 nanoribbons with homogeneous distribution of Mn2+ ions. The comprehensive structural characterization using powder X-ray diffraction (XRD) and electron paramagnetic resonance (EPR) provided compelling evidence that the Mn2+ ion predominantly substitutes the Ti4+ ion at octahedral coordination sites in bulk. Measurements performed on individual nanoribbons using near edge X-ray absorption fine structure spectromicroscopy revealed that the strong alkaline environment required for the formation of sodium titanate nanoribbons did not affect the manganese oxidation state. In the next two steps, the ion exchange process in HCl(aq) solution followed by the thermal treatment in air, lead to the formation of Mn2+ doped TiO2 nanoribbons. Analysis of the manganese content by X-ray photoelectron spectroscopy of several TiO2 nanoribbon samples calcined in the temperature range from 400 to 700 °C as well as analysis performed at the Ti L2,3 and Mn L2,3 edges with electron energy loss spectroscopy (EELS) showed that calcination at elevated temperatures induced the diffusion of manganese ions toward the nanoribbons’ surface. However, transformation of anatase nanoribbons to rutile nanoparticles, this process started at around 580 °C, was also accompanied by the partial oxidation of Mn2+ to Mn3+ and Mn4+. Manganese atoms that diffused to the TiO2 surface preferentially formed MnOx clusters as observed from characteristic electron paramagnetic resonance spectra and EELS measurements. In addition, the presence of Mn2+ reduced the beginning of phase transformation from anatase to rutile to near 120 °C. protons and/or dopant ions.17−19 Considering the impact on the nanostructures’ morphology, these two doping methods are far less destructive compared to some more established methods, e.g., high temperature treatment or ion implantation. When aiming for substitutional doping in combination with morphology preservation of 1D TiO2 nanostructures, the in situ doping method is the optimal choice. This method was already employed in the past for synthesis of sodium titanate and TiO2 nanotubes/nanoribbons substitutionally doped with transition metal ions Cu2+, Cr3+, and Co2+.12,13,20 However, in alkaline environment required for the growth of titanate nanostructures this approach may not work for some transition metal ions as they form insoluble hydroxides and/or they

1. INTRODUCTION Since the beginning of commercial production of titanium dioxide (TiO2) in the early twentieth century, TiO2 remains one of the most investigated materials owing to its remarkable functional properties, nontoxicity, and chemical stability.1−3 The efficiency of TiO2 in many applications, especially in heterogeneous catalysis and photocatalysis, is closely related to particle size, shape, and crystalline phase dictating the electronic properties.2−6 The particle morphology, size, and crystalline phase can be tailored by careful selection of synthesis procedures and post treatments, while the fine-tuning of electronic properties is usually achieved by chemical doping.7,8 One-dimensional (1D) alkali titanates represent an attractive precursor for preparation of 1D TiO2 nanostructures in gram quantities.2,9−11 Furthermore, this pathway also enables doping at different stages: (i) at the very beginning when the synthesis of alkali titanates takes place (in situ doping)12−16 or (ii) during an ion exchange process when alkali cations are exchanged with © XXXX American Chemical Society

Received: June 27, 2014 Revised: August 4, 2014

A

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

oxidize to a higher oxidation state, e.g., Co2+ ions forming oxidehydroxide compounds.20 Among transition metal ions manganese in the oxidation state 2+ is an especially attractive dopant for titania materials. For instance, by occupation of Ti4+ substitutional places Mn2+ can induce oxygen vacancies that facilitate formation of the rutile phase and thus significantly alter the thermal stability of the anatase phase.21,22 Furthermore, the presence of Mn2+ in the anatase framework has a strong impact on the material’s photocatalytic activity.22,23 However, since the charge of Mn2+ is different from that of Ti4+ and its ionic radius is 0.83 Å significantly larger compared to Ti4+ (0.605 Å), it makes substitutional Mn2+ doping challenging.24 Thus, the synthesis of Mn2+-doped TiO2 one-dimensional nanostructures proved to be very elusive in the past. Here we report on the efficient three step approach for preparation of 1D TiO2 nanoribbons (NRs) doped with Mn2+: (i) in the first step, in order to avoid clustering of Mn2+ species in strong alkaline environment during the synthesis of sodium titanate NRs, anatase powder doped with Mn2+ was used as a precursor. Then, in the second step (ii) by an ion-exchange process sodium titanate NRs were transformed to the protonated form which was (iii) by thermal treatment converted to TiO2 NRs. The temperature at which phase transformation from anatase to rutile started decreased for at least 120 °C (from 700 to 580 °C). A comprehensive study of the manganese oxidation state and its local environment in TiO2 NR samples calcined at different temperatures was done using diverse characterization techniques (near edge X-ray adsorption fine structure− transmission X-ray microscopy (NEXAFS-TXM), electron paramagnetic resonance (EPR), energy-dispersive X-ray spectrometry (EDXS), X-ray photoelectron spectroscopy (XPS), and scanning transmission electron microscopy−high angle annular dark field (STEM-HAADF) combined with electron energy loss spectroscopy (EELS) mapping). Obtained results suggest diffusion of manganese atoms and their aggregation on the TiO2 surface in the form of MnOx nanoclusters with increasing calcination temperature.

prepared dispersion was stirred at room temperature for 1 h, and then solid material was isolated by centrifugation. This was repeated for two more times. At the end, the solid material was washed twice with 300 mL of distilled water and dried overnight at 100 °C. The samples were labeled Mn@HTiNRs and Mn@HTiNTs (labels HTiNRs and HTiNTs refer to the undoped protonated titanate nanoribbons and nanotubes, respectively). Protonated titanate nanostructures doped with Mn2+ were subsequently calcined in air at 400 (Mn@TiO2NRs400 and Mn@TiO2NTs-400), 580 (Mn@TiO2NRs-580), and 700 °C (Mn@TiO2NRs-700) for 10 h in air. 2.2. Characterization Techniques. The phase analysis was performed on the cut surface by X-ray powder diffraction (XRD) using the diffractometer with Cu Kα radiation (λ = 1.5406 Å) and a Sol-X energy-dispersive detector (Endeavor D4, Bruker AXS, Karlsruhe, Germany). The angular range 2θ was from 5 to 60° with a step size of 0.02° and a collection time of 3 s. The morphology of synthesized products and samples was investigated with a field emission scanning (FE-SEM, Jeol, 7600F) and transmission (TEM, Jeol 2100) electron microscopes. The elemental compositions of the sample was investigated with a FE-SEM (Carl Zeiss, Supra 35 LV) equipped with an EDXS (energy-dispersive X-ray spectrometer) element analysis system and X-ray photoelectron spectroscopy (XPS). The XPS measurements were performed in a VERSAPROBE PHI 5000 from Physical Electronics, equipped with a Monochromatic Al Kα X-ray source. The energy resolution was 0.7 eV. For the compensation of built up charge on the sample surface during the measurements a dual beam charge neutralizer composed of an electron gun (∼1 eV) and an argon ion gun (≤10 eV) was used. The samples for XPS measurements were prepared by pressing the sample into a pellet. A conductive double face tape was used to attach the pallet to a sample holder. Atomically resolved scanning transmission electron microscopy−high angle annular dark field (STEM-HAADF) images have been acquired using a C3/C5 spherical aberrationcorrected microscope Nion USTEM working at 100 keV. Electron energy loss spectra (EELS) were recorded with a modified GATAN EELS system with a back-illuminated chargecoupled device camera. The X-band electron paramagnetic resonance (EPR) measurements at the Larmor frequency νL ≈ 9.6 GHz were conducted on a commercial Bruker E580 spectrometer equipped with a dielectric ring resonator ER 4118X-MD5 and an Oxford cryogenics liquid helium flow cryostat. Typically, low microwave powers of 1 mW and modulation fields of 1 mT were used in continuous wave (cw) mode at 150 K. For TEM and NEXAFS−TXM (near edge X-ray adsorption fine structure−transmission X-ray microscopy) analysis the samples were sonically dispersed in ethanol and a drop of the solution was deposited onto a lacey carbon film supported by a copper grid. Note that due to electron-beam damage effects on the sodium titanate nanostructures, different grids were used for the electron microscopy and for the NEXAFS−TXM analysis. The Ti L-edge and Mn L-edge spectra were recorded with the TXM installed at the undulator beamline U41-XM at the electron storage ring BESSY II, Helmholtz-Zentrum Berlin (HZB).26−29 In the best case the HZB TXM provides a high spatial resolution close to 10 nm (half-pitch)30 and a spectral resolution up to E/ΔE ≈ 104. Measurements can be performed at room or liquid nitrogen temperature in a vacuum of 10−7

2. EXPERIMENTAL SECTION 2.1. Synthesis. Sodium titanate nanoribbons/nanotubes doped with Mn2+ (Mn@NaTiNRs/Mn@NaTiNTs) were synthesized in two steps. First, TiO2 doped with 3 wt % (4.3 at. %) of Mn2+ (Mn@TiO2) was prepared by the sol−gel method using tetraisopropyl titanate [Ti(OCH(CH3)2)4] (Aldrich) as a titanium source and manganese(II) acetat tetrahydrate [Mn(CH3COO)2 × 4H2O] (Aldrich) as a source of Mn2+. The detailed procedure is described elsewhere.20 In the second step, 2 g of Mn@TiO2 was dispersed into 20 mL of 10 M NaOH (aq), using a procedure analogous to that described in ref 25. In brief, a Teflon-lined autoclave was filled with 18 mL of a reaction solution and held for 72 h at 175 °C (nanoribbons: Mn@NaTiNRs) and 125 °C (nanotubes: Mn@ NaTiNTs). When the autoclaves were cooled to room temperature a thin brown surface layer was removed and pink to gray products were dispersed into ∼300 mL of deionized water and filtered. The products were collected on a filter and were subsequently washed first with 300 mL of distilled water and then with 50 mL of EtOH and finally dried overnight at 100 °C. Next, sodium ions were removed by an ion-exchange process in 0.1 M HCl(aq): in general, 1 g of Mn@NaTiNRs/Mn@NaTiNTs was dispersed into 300 mL of 0.1 M HCl(aq) using an ultrasonic bath (30 min). The B

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

Table 1. Summary of Sample Preparation, Morphology Type, Manganese Oxidation State and Manganese Content for Manganese Doped Titanate/TiO2 Samplesa sample

precursor

preparation

morphology

Mn@NaTiNRs

Ti(OCH(CH3)2)4, Mn(CH3COO)2 × 4H2O Mn@TiO2

Mn@HTiNRs

Mn@NaTiNRs

sol−gel, calcination at 400 °C, 10 h NaOH(aq), in autoclave at 175 °C ion exchange, 0.1 M HCl

Mn@TiO2NRs400 Mn@TiO2NRs580 Mn@TiO2NRs700

Mn@HTiNRs

calcination at 400 °C, 10 h

nanoribbons

Mn@HTiNRs

calcination at 580 °C, 10 h

nanoribbons

Mn@HTiNRs

calcination at 700 °C, 10 h

nanoribbons and nanoparticles

Mn@TiO2

a

phase

manganese oxidation state

Mn wt %

nanoparticles

anatase

2+

3.0

nanoribbons

sodium trititanate

2+

1.3

nanoribbons

protonated trititanate TiO2−B

2+

1.5

2+

1.6

2+

1.6

2+, 3+, 4+

1.7

TiO2−B, anatase, rutil anatase, rutil

Manganese content was determined using EDXS.

Figure 1. (A) Powder XRD patterns of manganese doped protonated titanate nanoribbons (Mn@HTiNRs, 100 °C) and TiO2 nanoribbon samples calcined at 400 (Mn@TiO2NRs-400), 580 (Mn@TiO2NRs-580), and 700 °C (Mn@ TiO2NRs-700). Peaks belonging to rutile, anatase and TiO2− B are marked with R, A and B, while H denotes peaks of H2Ti3O7. Expanded views of the anatase (101) peak of Mn@TiO2NRs-580 (B) and Mn@ TiO2NTs-400 (C) are compared with the measured profiles of undoped anatase nanoribbon and nanotube samples.

temperature (125 °C) SEM image (Figure S2B) shows that as grown nanostructures form clusters characteristic of sodium titanate nanotubes.34 However, detailed HRTEM investigation revealed that beside nanotubes also a larger quantity of partially rolled titanate nanosheets were formed (Figure S3A). Similar morphology was observed when in the reaction mixture copper ions in the oxidation state 2+ were present.12 Therefore, we propose that the formation of these nanostructures is most likely promoted by the presence of transition metal ions, Mn2+ ions in this case, in the reaction mixture. The color of the isolated products, Mn@NaTiNRs and Mn@NaTiNTs, were gray-pink and dark pink. In addition, the manganese content in both products determined with EDXS was about 1.3 wt % (Table 1 and Supporting Information, Table S1). Sodium ions were efficiently exchanged with protons by an ion-exchange process (Table S1). Isolated Mn2+-doped protonated titanate nanoribbons (Mn@HTiNRs) were subsequently calcined at 400, 580, and 700 °C. Detailed experimental conditions for samples preparation, morphological properties, manganese content and manganese oxidation state are summarized in Table 1. 3.1. Structural Determination of TiO2 Polymorphs. A powder X-ray diffraction (XRD) pattern of Mn@HTiNRs corresponds to H2Ti3O7 and is in agreement with those reported in the literature9 (Figure 1A). Broadened peaks in the XRD pattern reflect the presence of a local disorder in the structure, which can be associated with the presence of water molecules.28 After sample heating at 400 °C (Mn@TiO2NRs-

Torr. The spectra for this study were recorded at room temperature in a transmission mode by taking a sequence of images over a range of photon energies covering the investigated absorption edges with E/ΔE ≥ 4500. The exit slit of the monochromator was set to 5 μm, which corresponds to a calculated spectral resolution of E/ΔE = 2 × 104. For the study presented here a zone plate objective with an outermost zone width of 25 nm was used to image the sample onto a cooled back-illuminated soft X-ray CCD camera (Roper Scientific, PI SX1300). The NEXAFS spectra presented show the optical density OD(E) = −log [I(E)/I0(E)].31,32 The I(E) were recorded on individual nanostructures and the I0 was recorded in a bare region of the sample support close to nanostructure (Figure S1). During the measurements (changing the photon energy) the nanostructures were kept in focus. The alignment of the images was performed using a cross-correlation method from the IMOD tomography package.33

3. RESULTS AND DISCUSSION Mn2+ doping of sodium titanate nanoribbons (Mn@NaTiNRs) and nanotubes (Mn@NaTiNTs) was achieved by in situ doping method. In agreement with the established temperature range of nanoribbon formation,34 nanoribbons were formed with high morphological yield at higher (175 °C) reaction temperature (Figure S2A). Their average length is about 2 μm, whereas diameters vary from 20 to 200 nm. At lower reaction C

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

results indicate that the observed shift is not caused by changes in morphology but in the structure due to the Mn2+substitutional doping. A close inspection of (101) peak in Figure 1B reveals anisotropic peak-broadening which is more pronounced for doped than undoped nanoribbons. This might originate/arise from the TiO2−B (110) diffraction peak, strain broadening caused by partial substitution or by particle morphology with large aspect ratio. The latter explanation is less probable since both samples, Mn@TiO2NRs-580 and TiO2NRs-580, have similar nanoribbon morphology and should thus exhibit similar anisotropic broadening. Most likely this originates from the presence of TiO2−B phase which is one of the three TiO2 phases detected in Mn@TiO2NRs-580 (Figure 1A). The TiO2−B (110) diffraction peak is positioned at 24.992° (JCPDS No. 35-0088) while the anatase (101) peak is positioned at 25.332° (JCPDS No. 86-1157) and therefore causes the observed anistropic broadening. This broadening is not observed in the diffractogram of Mn@TiO2NTs-400 (Figure 1C) because the sample is only composed of anatase (Figure S4). 3.2. Morphology. Figure 2 summarizes the effect of calcination temperature on morphological changes. Nanoribbon morphology remains preserved up to 580 °C although at this temperature first changes in the nanoribbons’ shape can already be noticed. For instance, nanoribbons’ edges (insets in Figure 2C) become more rounded when compared to the corresponding nanoribbon edges in parent Mn@NaTiNRs and Mn@HTiNRs. Moreover, a direct comparison of nanoribbons’ surface of Mn@NaTiNRs and the samples calcined at 400 (Figure 2 B) and 580 °C (Figure 2D) shows that the surface with increasing calcination temperature becomes rougher. In Mn@TiO2NRs-400 this is a direct consequence of interlayered OH groups leaving the structure37 which results in the pore formation. The average pore diameter is ∼5 nm (Figure 2B). Their presence disrupts the crystallinity of nanoribbons and thus is responsible for the peak broadening in the Mn@ TiO2 NRs-400 X-ray diffractogram (Figure 1A). When calcination temperature is increased to 580 °C, the average pore size increases while their number decreases (Figure 2D). Not surprisingly, the surface reconstruction is reflected in peak narrowing in the XRD pattern (Figure 1). Finally, in the sample calcined at 700 °C in addition to nanoribbons also nanoparticles are observed (Figure 2E,F). Nanoribbons found in Mn@TiO2NRs-700 are crystalline, the measured atomic plane distances of all checked nanoribbons correspond to anatase structure (inset in Figure 2F: the interplanar spacing of 0.35 nm is corresponding to the (101) reflection of anatase (JCPDS No. 86-1157)). The nanoribbons in Mn@TiO2NRs-700 are in anatase crystalline phase while the observed rutile peaks in the XRD study must arise from the nanoparticles. Anatase and rutile do not have structural similarities (like H2Ti3O7, TiO2−B and anatase) and therefore transformation from anatase to rutile is necessarily accompanied by the collapse of nanoribbon morphology explaining the formation of nanoparticles in the latter case. Typically, rutile nanoparticles are bridged by forming longer chains as shown in Figure 2E. In some cases due to the partial sintering nanoparticles do not have a specific shape, as exemplified by the nanoparticles shown in Figure 2F. 3.3. Determination of Manganese Content. EDXS (energy-dispersive X-ray spectroscopy) and XPS (X-ray photoelectron spectroscopy) analytical techniques were used to determine the relative content of manganese and sodium in

400) new peaks emerge in the XRD pattern that correspond to the TiO2−B phase (JCPDS No. 35-0088). The XRD peaks are still broadened, most likely due to the formation of porous structure (vide infra, Figure 2A,B) and the related structural

Figure 2. TEM images of manganese doped TiO2 nanoribbon samples calcined at 400 (A and B), 580 (C and D), and 700 °C (E and F).

disorder.35 On the other hand, in the XRD patterns of Mn@ TiO2NRs-580 and Mn@TiO2NRs-700 a coexistence of TiO2 phases was detected. In Mn@TiO2NRs-580 anatase prevails over rutile and TiO2−B. On the other hand, the dominant phase in Mn@TiO2NRs-700 is rutile. At this point, we stress that none of the XRD patterns (Figure 1A) display any distinguished peaks characteristic of manganese(II) hydroxide, manganese(III) oxidehydroxide, and/or manganese oxides. We also stress that transitions to anatase and further to rutile in the case of undoped protonated titanate nanoribbons occur at higher temperatures. Presence of Mn2+ ions in the TiO2 nanoribbons considerably reduced the beginning of anatase to rutile phase transformation (Figure S4). Protonated titanate nanotubes doped with Mn2+ (Mn@ HTiNTs) transformed to anatase nanotubes already at 400 °C as shown by suitable changes in the XRD pattern (Figure S5). The lower temperature needed for the transformation of protonated titanate into anatase nanotubes/nanosheets, in comparison to the nanoribbon sample, arises from the difference in morphology. The ionic radius of Mn2+ is 0.82 Å and is larger than that of 4+ Ti (0.61 Å).24 In the case of partial substitution of Ti4+ with Mn2+ the anatase unit cell size should thus increase, e.g., as reported by Choudhury et al.36 in the case of Mn2+-doping of anatase nanoparticles. Figure 1B,C shows expanded regions around the anatase (101) peak for Mn2+ doped and undoped anatase nanoribbon and nanotube samples. Indeed, in the case of the manganese doped anatase NRs a slight shift of the peak by 0.06° (Figure 1B) toward lower angles is clearly observed. On the other hand, in the XRD of anatase sample with nanotubular morphology (Mn@TiO2NTs-400) no such shift is noticed (Figure 1C) within our experimental resolution. These D

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

2E,F) and only NRs for the samples thermally treated at 400 and 580 °C. The titanium L-edge NEXAFS spectrum of Mn2+@NaTiNRs (Figure 3B) share common features with the spectrum reported for anatase:39,40 it is composed of distinguishable peaks in the photon energy range between 455 and 470 eV corresponding to excitations of the Ti 2p states into the empty Ti 3d states.3 The Ti L-edge shows two groups of peaks arising from the spin−orbit splitting of the Ti 2p core level into 2p1/2 (L2-edge) and 2p3/2 levels (L3-edge).39 The most prominent difference between the spectra of the studied titanium oxides is the value of the energy splitting of the fine structure in the L3-eg band. The line-shape of the egband is highly sensitive to the local symmetry around metal cations. Krüger et al. showed that the L3-eg peak splitting in TiO2 is a band structure effect which mainly reflects the connectivity of the TiO6 octahedrons rather than local distortions of the individual octahedra.40 In fact, as we can observe from Figure 3B for increasing calcination temperature, i.e., ordering the nanostructure, the splitting increases until it reach the same value of the splitting for anatase. This is in accordance with the XRD results which show that the structure of manganese doped TiO2 nanoribbon sample post treated at 700 °C (Mn@TiO2NRs-700) is very similar to anatase. These observations strongly support our HRTEM results (Figure 2E,F) showing that in Mn@TiO2NRs-700 sample the nanoribbons have anatase structure whereas the nanoparticles have rutile structure. It is important to note that the NEXAFS spectra recorded on the nanoparticles found in the Mn@ TiO2NRs-700 sample (Figure 3B) are characteristic of rutile structure, the feature at 461 eV is more intense, than the one at 459 eV (just the opposite of the anatase phase). The differences in the intensity ratio of these peaks are due to symmetric variation in the octahedral TiO2 structure. In nanosized titanates the local manganese coordination can strongly vary, depending on the exact location of manganese atom. For instance, manganese doped into bulk titanate structure will have a significantly different local environment compared to those manganese atoms located at the surface, where even the under-coordinate geometry may take place. We thus decided to employ local probe electron paramagnetic resonance (EPR) technique, which has proved in the past to be an extremely powerful tool to investigate Mn2+ (S = 5/2) local coordination. At the same time it is insensitive to Mn3+ (S = 2) in the X-band (Larmour frequency of 9.6 GHz) due to its large zero-field splitting. In Figure 4 we show the X-band EPR spectrum of as-prepared Mn2+doped sodium titaante nanoribbons (Mn@NaTiNRs), which is centered at the g-factor of ∼2. The hyperfine sextet of central −1/2↔1/2 transition superimposed on a broad background of the other unresolved m+1↔ m transitions is clearly observable. The hyperfine sextet arises from the coupling to 55Mn nuclear spin (I = 5/2) characterized by the hyperfine coupling constant a. Simulations of the spectrum yield a = 8.7(1) mT, which is very characteristic of Mn2+ oxidation state and thus unambiguously rule out any EPR contribution from the other manganese oxidations states, e.g., Mn4+ has typically smaller a ∼ 7.5 mT.21 Extracted a also matches the expected value for the Mn2+ octahedrally coordinated sites in the bulk of anatase nanoparticles. For the under coordinated surface Mn2+ sites a larger (a ≈ 9.6 mT) hyperfine coupling constant is expected.18 Moreover, the large estimated zero-field splitting parameter of about D ≈ 70 mT is also corroborating with the suggestion that these sites are more

the nanoribbon and nanotube samples. In all TiO2 NR samples manganese content determined by EDXS was ∼1.6−1.7 wt % (Table 1). Surprisingly, the manganese content in the NRs samples determined by XPS showed a sudden increase of manganese for the samples calcined at temperatures above 550 °C compared to the related EDXS results (Figure S6). This may be associated with the inhomogeneous distribution of manganese that most likely arises from the diffusion of manganese ions toward the NRs’ surface with increasing calcination temperature. The regions of the sample probed by EDXS and XPS are different. The XPS performed at 1486 eV photon energy probes only a surface layer of about 60 Å thicknesses while the EDXS is a bulk sensitive analytical technique. 3.4. Manganese Oxidation State and Its Local Environment Study. Determination of the manganese oxidation state was not possible by the XPS technique due to the low manganese content. For this reason NEXAFS (near edge X-ray adsorption fine structure) measurements were employed. In combination with TXM (transmission X-ray microscopy), the NEXAFS-TXM (near edge X-ray adsorption fine structure−transmission X-ray microscopy) allows the analysis of isolated nanostructures. The Mn L3-edge spectra recorded on the samples were performed using X-ray photon energies between 638 and 645 eV. Comparing them with reported NEXAFS results for different oxidation states,38 the measurement revealed (Figure 3A) that the oxidation state of

Figure 3. NEXAFS-TXM spectra recorded on individual nanoribbons: (A) Mn L-edge and (B) Ti L-edge spectra for as synthesized manganese doped sodium titanate nanoribbons and manganese doped TiO2 nanoribbons calcined at 400, 580, and 700 °C.

manganese in the as synthesized sodium titanate nanoribbons, and in TiO2 NRs post-treated up to 580 °C is 2+ (Mn@ TiO2NRs-400 and Mn@TiO2NRs-580). Only in the sample calcined at 700 °C other oxidation states, Mn3+ and Mn4+, can also be observed. Furthermore, the NEXAFS-TXM method also allows investigation of the electronic structure of individual nanoribbons. Figure S1 (Supporting Information) shows X-ray images of the image stack used to record the NEXAFS spectra. These images show a large number of nanostructures that can be analyzed individually. In accordance with the images recorded by electron microscopy the X-ray images show nanoparticles and NRs for the sample thermally treated at 700 °C (Figure E

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

more symmetric. The spectral changes reflect the growth of the component with a nearly Lorentzian line shape and the peakto-peak line width of about ΔBpp ≈ 100 mT. This is most evident for the sample treated at 700 °C, where this line is completely dominating the EPR spectrum. Such Lorentzian-like EPR line is typical for MnOx nanoparticles41,42 and indicates that during the high-temperature treatment manganese ions migrate toward the TiO2 surface where they tend to nanoclusters. For the sample treated at 700 °C some of the Mn2+ ions may be still left inside in the nanoribbon structure, judging from the very weak but at the same time also very sharp hyperfine sextet of central −1/2↔1/2 transition. 3.5. Manganese Oxide Phase Segregation. The difference in manganese content obtained from EDXS and XPS analysis and the results of the EPR study indicate that with increasing calcination temperature manganese atoms migrate toward the NRs surface. To further obtain information where on the surface a concentration of manganese atoms is increasing with increasing calcination temperature, we performed EELS mapping on individual nanoribbons on the same set of samples. Figure 5 shows, respectively, HAADF-STEM (high angle annular dark field−scanning transmission electron microscopy) images of individual nanoribbons, and titanium intensity profiles and a ratio of Mn/Ti content obtained from the EELS (electron energy loss spectroscopy) analysis of the Ti L and Mn L edges for the same samples, respectively. Despite the low manganese content, which is in all investigated samples ∼1.6 wt % (Table 1), EELS mapping results revealed that manganese is relatively homogeneously distributed within the investigated area in all of the samples (Figure 5). Nevertheless, changes in manganese distribution can be observed for the samples calcined above 400 °C (Figure 5C,D). A high resolution HAADF-STEM image (Figure 5C) shows a part of the nanoribbon calcined at 580 °C. The pore present in the nanoribbon is easily observed as a darker area. Actually, from the images taken in transmission mode we cannot deduce whether pores are located at the surface or inside the nanoribbon. The pore’s diameter is found in the range 5−20 nm. A white rectangle over one of the pores marks the area where EELS mapping analysis was performed. A calculated Mn/Ti ratio unambiguously shows that manganese concentration is locally increased around the pore edges thus corroborating with the EPR conclusions. In addition, the same was observed for the sample calcined at 700 °C.

Figure 4. Comparison of X-band EPR spectra measured in manganese doped sodium titanate and TiO2 nanoribbons: as prepared nanoribbons (Mn@NaTiNRs), nanoribbons after ion exchange (Mn@ HTiNRs) and after calcination at 400 (Mn@TiO2NRs-400), 580 (Mn@TiO2NRs-580) and (Mn@TiO2NRs-700) 700 °C. The spectra were recorded at T = 150 K. Sharp lines centered at around 170 and 410 mT and marked by asterisk originate from the EPR resonator.

likely located in the bulk than on the surface of nanoribbons. Therefore, our preparation method does not favor the initial formation of Mn2+ surface centers that were observed in TiO2 nanoparticles prepared under hydrothermal conditions from a suspension of protonated titanate nanotubes in Mn 2+ solution.18 In our nanotube samples a broad Lorentzian component on top of the characteristic Mn2+ hyperfine splitting is observed (Figure S7). This is reminiscent to spectra reported in ref 19 on thermally annealed Mn2+-doped TiO2 nanotubes samples where a broad component was attributed to the strongly distorted Mn2+ octahedral sites. This component is missing in the nanoribbon samples because of their much better crystallinity. A single-electron-trapped oxygen vacancy signal was observed in Mn2+-doped TiO2 prepared by calcination of MnxH2‑xTi3O7 nanotubes at 400 °C in N2/H2 atmosphere.19 However, in our case even the maximum calcination temperature of 700 °C was not enough to observe such a center. Ion-exchange results in some intriguing changes in the EPR spectra that reflect dramatic evolution in the Mn2+ local coordination. For the protonated titanate nanoribbon sample (Mn@HTiNRs, sample was thermally treated at 100 °C), we notice that the zero-field splitting and hyperfine coupling constants do not change appreciably. This implies that the ionexchange and the low-temperature treatment do not trigger the rearrangements of Mn2+ ions within the nanoribbons. However, we do notice the significant broadening of the hyperfine sextet, which we attribute to the strong electron−nuclear dipolar interactions. Since the nuclear magnetic moment of 1H is by a factor of 3.8 larger than that of 23Na the observed broadening quantitatively agrees with the ion exchange of Na+ with H+ ions. More pronounced changes of the EPR spectra are observed for the NR samples calcinated at higher temperatures. Namely, the central −1/2↔1/2 transition is barely discernible already for the sample treated at 400 °C and the spectrum becomes

4. CONCLUSIONS The hydrothermal synthesis of sodium titanate nanoribbons in the presence of Mn2+ ions resulted in successful incorporation and homogeneous distribution of Mn2+ ions in the bulk occupying octahedral sites. Furthermore, strong alkaline environment at which synthesis of sodium titanates nanostructures took place did not affect manganese oxidation state. Morphology and manganese valence were retained on transformation from protonated titanate to TiO2−B and further to anatase. The dopant effect on the temperature required for the phase transformations from anatase to rutile was observed. The presence of Mn2+ in anatase nanoribbons reduced the onset of anatase to rutile phase transformation for at least 120 °C. However, diffusion of manganese ions toward the nanoribons’ surface with increasing calcinations temperature was noticed. A change in morphology from anatase to rutile phase transformation was observed. We presumed that at this phase F

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

different temperatures, and EPR spectra of Mn@NaTiNTs and Mn@TiO2NTs-400. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Tel: +386-1-4773-500. E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The financial support from the Slovenian Research Agency for Project J2-4034 and bilateral Project BI-FR/12-13-PROTEUS0 is gratefully acknowledged. The research leading to these results has received funding from the European Union Seventh Framework Program under Grant Agreement 312483 − ESTEEM2 (Integrated Infrastructure Initiative − I3) and Grant Agreement 312284 − CALIPSO. We thank HZB for the allocation of synchrotron radiation beam time. The authors also acknowledge financial support from the COST action MP0901 “NanoTP”.



Figure 5. HAADF-STEM images of manganese doped sodium titanate nanoribbons (A, Mn@NaTiNRs) and manganese doped TiO2 nanoribbons calcined at 400 (B, Mn@TiO2NRs-400), at 578 (C, Mn@TiO2NRs-578), and at 700 °C (D, Mn@TiO2NRs-700) with corresponding Ti-L intensity profiles (red line) and calculated Mn/Ti ratios. The chemical profiles were obtained from the EELS analysis of the Ti L and Mn L edges in the areas marked with white rectangles in the HAADF-STEM images.

transformation oxidation of Mn2+ to oxidation states 3+ and 4+ occurs simultaneously. The EPR results showed that at this stage manganese was in the form of MnOx nanoclusters located on the surface of the nanoparticles.



REFERENCES

(1) Carp, O.; Huisman, C. L.; Reller, A. Photoinduced Reactivity of Titanium dioxide. Prog. Solid State Chem. 2004, 32, 33−177. (2) Chen, X.; Mao, S. S. Titanium Dioxide Nanomaterials: Synthesis, Properties, Modifications, and Applications. Chem. Rev. 2007, 107, 2891−2959. (3) Diebold, U. The Surface Science of Titanium Dioxide. Surf. Sci. Rep. 2003, 48, 53−229. (4) Mao, Y.; Wong, S. S. Size- and Shape-Dependent Transformation of Nanosized Titanate into Anatase Titania Nanostructures. J. Am. Chem. Soc. 2006, 128, 8217−8226. (5) Hanaor, A. H.; Sorell, C. S. Review of the Antase to Rutile Phase Transformation. J. Mater. Sci. 2011, 46, 855−874. (6) Ohno, T.; Sarukawa, K.; Matsumura, M. Crystal Faces of Rutile and Anatase TiO2 Particles and their Roles in Photocatalytic Reactions. New J. Chem. 2002, 26, 1167−1170. (7) De los Santos, D.; Aguilar, T.; Sánchez-Coronilla, A.; Navas, J.; Cruz Hernández N.; Alcántara, R.; Fernández-Lorenzo, C.; MartínCalleja, J. Electronic and Structure Properties of Highly Aluminum Ion Doped TiO2 Nanoparticles: A Combined Experimental and Theoretical Study. ChemPhysChem. doi: 10.1002/cphc.201402071. (8) Dozzi, M. V.; Selli, E. Doping TiO2 with p-block Elements: Effects on Photocatalytic Activity. J. Photochem. Photobiol. C 2013, 14, 13−28. (9) Kolen’ko, Y. V.; Kovnir, K. A.; Gavrilov, A. I.; Garshev, A. V.; Frantti, J.; Lebedev, O. I.; Churagulov, B. R.; Van Tendeloo, G.; Yoshimura, M. Hydrothermal Characterization of Nanorods of Various Titanates and Titanium Dioxide. J. Phys. Chem. B 2006, 110, 4030− 4038. (10) Bavykin, D.; Friedrich, J. M.; Walsh, F. C. Protonated Titanates and TiO2 Nanostructured Materials: Synthesis, Properties and Applications. Adv. Mater. 2006, 18, 2807−2824. (11) Peng, C.-W.; Richard-Plouet, M.; Ke, T.-Y.; Lee, C.-Y.; Chiu, H.T.; Marhic, C.; Puzenat, E.; Lemoigno, F.; Brohan, L. Chimie Douce Route to Sodium Titanate Nanowires with Modulated Structure and Conversion to Highly Photoactive Titanium Dioxides. Chem. Mater. 2008, 20, 7228−7236. (12) Umek, P.; Pregelj, M.; Gloter, A.; Cevc, P.; Jagličić, Z.; Č eh, M.; Pirnat, U.; Arčon, D. Coordination of Intercalated Cu2+ Sites in Copper Doped Sodium Titanate Nanotubes and Nanoribbons. J. Phys. Chem. C 2008, 112, 15311−15319. (13) Santara, B.; Giri, P. K.; Dhara, S.; Imakita, K.; Fujii, M. Oxygen Vacancy-Mediated Enhanced Ferromagnetism in Undoped and Fe-

ASSOCIATED CONTENT

S Supporting Information *

Characteristic TXM images of Mn@TiO2NRs-580 and Mn@ TiO2NRs-700, SEM images of as synthesized Mn2+-doped sodium titanate nanoribbon and nanotube samples, TEM images of Mn@NaTiNTs and Mn@TiO2NTs-400, sodium and manganese content for Mn@NaTiNRs, Mn@HTiNTs, Mn@NaTiNTs, and Mn@HTiNTs, an XRD pattern of Mn@ TiO2NTs-400, comparison of manganese content determined using EDXS and XPS for Mn@TiO2NR samples calcined at G

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry C

Article

doped TiO2 Nanoribbons. J. Phys. D: Appl. Phys. 2014, 47, 235304 (14pp). (14) Chong, S. V.; Suresh, J. X.; Yamaki, K.; Kadowaki, K. Tailoring the Magnetization Behavior of Co-doped Titanium Dioxide Nanobelts. Solid State Commun. 2008, 148, 345−349. (15) Díaz-Guerra, C.; Umek, P.; Gloter, A.; Piquears, J. Synthesis and Characterization of Undoped and Cr3+-doped Sodium Titanate Nanotubes and Nanoribbons. J. Phys. Chem. C 2010, 114, 8192−8198. (16) Ji, T.; Yang, F.; Lv, Y.; Zhou, J.; Sun, J. Synthesis and VisibleLight Photocatalytic Activity of Bi-doped TiO2 Nanobelts. Mater. Lett. 2009, 63, 2044−2046. (17) Sun, X.; Li, Y. Synthesis and Characterisation of IonExchangeable Titanate Nanotubes. Chem.Eur. J. 2003, 9, 2229− 2238. (18) Saponjic, Z. V.; Dimitrijevic, N. M.; Poluektov, O. G.; Chen, L. X.; Wasinger, E.; Welp, U.; Tiede, D. M.; Zuo, X.; Rajh, T. Charge Separation and Surface Reconstruction: A Mn2+ Doping Study. J. Phys. Chem. B 2006, 110, 25441−25450. (19) Szirmai, P.; Horváth, E.; Náfrádi, B.; Micković, Z.; Smajda, R.; Djokić, D. M.; Schenk, K.; Forró, L.; Magrez, A. Synthesis of Homogenous Manganese-Doped Titanium Oxide Nanotubes from Titanate Precursors. J. Phys. Chem. C 2013, 117, 697−702. (20) Umek, P.; Bittencourt, C.; Gloter, C.; Dominko, R.; Jagličić, Z.; Cevc, P.; Arčon, D. Local Coordination and Valence States of Cobalt in Sodium Titanate Nanoribbons. J. Phys. Chem. C 2012, 116, 11357− 11363. (21) Arroyo, R.; Córdoba, G.; Padilla, J.; Lara, V. H. Influence of Manganese Ions on the Anatase-Rutile Phase Transition of TiO2 prepared by the Sol-gel process. Mater. Lett. 2002, 54, 397−402. (22) Devi, L. G.; Kottam, N.; Kumar, S. G. Preparation and Characterization of Mn-doped Titanates with a Bicrystalline Framework: Correlation of the Crystallite Size with the Synergistic Effect on the Photocatalytic Activity. J. Phys. Chem. C 2009, 113, 15593−15601. (23) Gomathi Devi, L.; Kottam, N.; Kumar, S. G.; Anantha Raju, K. S. Mechanism of Charge Transfer in the Transition Metal Ion Doped TiO2 with Bicrystalline Framework of Anatase and Rutile: Photocatalytic and Photoelectrocatalytic Activity. Catal. Lett. 2009, 131, 612−617. (24) Shannon, R. D. Revised Effective Ionic Radii and Systematic Studies of Interatomic Distances in Halides and Chalcogenides. Acta Crystallogr. 1976, A32, 751−767. (25) Umek, P.; Cevc, P.; Jesih, A.; Gloter, A.; Ewels, C. P.; Arčon, D. Impact of Structure and Morphology on Gas Adsorption of TitanateBased Nanotubes and Nanoribbons. Chem. Mater. 2005, 17, 5945− 5950. (26) Guttmann, P.; Zeng, X.; Feser, M.; Heim, S.; Yun, W.; Schneider, G. J. Energy-Tunable Full-Field X-Ray Microscopy: CryoTomography and Nano-Spectroscopy with the New BESSY TXM. Phys.: Conf. Ser. 2009, 012064−012066. (27) Schneider, G.; Guttmann, P.; Heim, S.; Rehbein, S.; Mueller, F.; Nagashima, K.; Heymann, J. B.; Muller, W. G.; McNally, J. G. The Three-Dimensional Cellular Ultrastructure Resolved by X-Ray Microscopy. Nat. Methods 2010, 7, 985−987. (28) Guttmann, P.; Bittencourt, C.; Rehbein, S.; Umek, P.; Ke, X.; Van Tendeloo, G.; Ewels, C. P.; Schneider, G. Nanoscale Spectroscopy with Polarized X-Rays by NEXAF-TXM. Nat. Photon. 2012, 6, 25−29. (29) Zeng, X.; Duewer, F.; Feser, M.; Huang, C.; Lyon, A.; Tkachuk, A.; Yun, W. Ellipsoidal and Parabolic Glass Capillaries as Condensers for X-ray Microscopes. Appl. Opt. 2008, 47, 2376−2381, DOI: 10.1364/AO.47.002376. (30) Rehbein, S.; Guttmann, P.; Werner, S.; Schneider, G. Characterization of the Resolving Power and Contrast Transfer Function of a Transmission X-Ray Microscope with Partially Coherent illumination. Opt. Express 2012, 20, 5830−5839. (31) Bogomolova, L. D.; Krasil’nikova, N. A.; Bodanov, V. L.; Khaliev, V. D.; Mitrofanov, V. V. Electron Paramagnetic Resonance of Transition Metal Ions in Fluorogermanate Glasses. J. Non-Cryst. Solids 1995, 188, 130−135.

(32) Jacobsen, C.; Wirick, S.; Flynn, G.; Zimba, C. Soft X-ray Spectroscopy from Image Sequences with sub-100 nm Spatial Resolution. J. Microscopy 2000, 197, 173−184. (33) Kremer, J. R.; Manostronarde, D. N.; McIntosh, J. R. Computer Visualization of Three-Dimensional Image Data Using IMOD. J. Struct. Biol. 1996, 116, 71−76. (34) Umek, P.; Cerc Korošec, R.; Jančar, B.; Dominko, R.; Arčon, D. The Influence of the Reaction Temperature on the Morphology of Sodium Titanate 1D Nanostructures and a Study of Their Thermal Stability. J. Nanosci. Nanotechnol. 2007, 7, 3502−3508. (35) Zhao, B.; Chen, F.; Liu, H.; Zhang, J. Mesoporous TiO2-B Nanowires Synthesized from Tetrabutyl Titanate. J. Phys. Chem. Solids 2011, 72, 201−206. (36) Choudhury, B.; Choudhury, A. Oxygen Vacancy and Dopant Concentration Dependent Magnetic Properties of Mn Doped TiO2 Nanoparticle. Curr. Appl. Phys. 2013, 113, 1025−1031. (37) Zhang, M.; Jin, Z.; Zhang, J.; Guo, X.; Yang, J.; Li, W.; Wang, X.; Zhang, Z. Effect of Annealing Temperature on Morphology, Structure and Photocatalytic Behavior of Nanotubes H2Ti2O4(OH)2. J. Mol. Catal. A: Chem. 2004, 217, 203−210. (38) Saywell, A.; Magnano, G.; Satterley, C. J.; Perdigão, L. M. A.; Britton, A. J.; Taleb, N.; Del Carmen Giménez-López, M.; Champness, N. R.; O’Shea, J. N.; Beton, P. H. Self-Assembled Aggregates Formed by Single-Molecule Magnets on a Gold Surface. Nat. Commun. 2010, 8 DOI: 10.1038/ncomms1075. (39) Bittencourt, C.; Krüger, P.; Lagos, M. J.; Ke, X.; Van Tendeloo, G.; Ewels, C.; Umek, P.; Guttmann, P. Towards Atomic Resolution in Sodium Titanate Nanotubes Using Near-edge X-Ray-Absorption Finestructure Spectromicroscopy Combined with Multichannel Multiplescattering Calculations. Beilstein J. Nanotechnol. 2012, 3, 789−797. (40) Krüger, P. Multichannel Multiple Scattering Calculation of L2,3 -edge Spectra of TiO2 and SrTiO3: Importance of Multiplet Coupling and Band Structure. Phys. Rev. B, Cond. Matter. Mater. Phys. 2010, 81, art. no. 125121 (pp 6). (41) Golosovsky, I. V.; Arčon, D.; Jagličić, Z.; Cevc, P.; Sakhnenko, V. P.; Kurdyukov, D. A.; Kumzerov, Y. A. ESR Studies of MnO Embedded into Silica Nanoporous Matrices with Different Topology. Phys. Rev. B 2005, 72, 144410 (6 pages). (42) Chen, Z. S.; Zhang, S.; Tan, S.; Li, F.; Wang, J.; Jin, S.; Zhang, Y. Preparation and Electron Spin Resonance Effect of Nanometer-sized Mn2O3. J. Cryst. Growth 1997, 180, 280−283.

H

dx.doi.org/10.1021/jp5063989 | J. Phys. Chem. C XXXX, XXX, XXX−XXX