Observation of Optical Band-Gap Narrowing and Enhanced Magnetic

Article pubs.acs.org/JPCC

Cite This: J. Phys. Chem. C XXXX, XXX, XXX−XXX

Observation of Optical Band-Gap Narrowing and Enhanced Magnetic Moment in Co-Doped Sol−Gel-Derived Anatase TiO2 Nanocrystals V. R. Akshay,†,‡ B. Arun,†,‡ Guruprasad Mandal,§ Geeta R. Mutta,∥ Anupama Chanda,*,⊥ and M. Vasundhara*,†,‡

J. Phys. Chem. C Downloaded from pubs.acs.org by KAOHSIUNG MEDICAL UNIV on 11/08/18. For personal use only.

†

Materials Science and Technology Division, CSIR-National Institute for Interdisciplinary Science and Technology, Trivandrum 695 019, India ‡ Academy of Scientific and Innovative Research (AcSIR), Training and Development Complex, CSIR Campus, CSIR Road, Taramani, Chennai 600 113, India § Centre for Rural and Cryogenic Technologies, Jadavpur University, Kolkata 700032, India ∥ Nano-Materials Laboratory, School of Engineering and Physical Sciences, Heriot Watt University, Edinburgh EH14 4AS, United Kingdom ⊥ Department of Physics, Dr. Hari Singh Gour Central University, Sagar 470003, India S Supporting Information *

ABSTRACT: The magnetic behavior of TiO2 and doped TiO 2 nanocrystals has been a challenge due to the unambiguous nature of defects present in oxide semiconductors. Here, a simple, low-temperature sol−gel method is developed for the synthesis of low-dimensional and highly efficient stable anatase TiO2 nanocrystals. The X-ray powder diffraction pattern and Raman spectra confirm the formation of a single-phase anatase structure of TiO2. High-resolution transmission electron microscopy studies reveal the crystalline nature of the sol−gel-derived nanocrystals. The increase in lattice parameters together with the shifting and broadening of the most intense Eg(1) mode in micro-Raman spectra of Codoped TiO2 nanocrystals indicate the incorporation of Co in TiO2. Shifting of the absorption edge to the visible region in UV−visible spectra indicates narrowing of the band gap due to Co incorporation in TiO2. X-ray photoelectron spectra confirm the presence of Co2+ and Co3+ in Co-doped TiO2 samples. Oxygen vacancy defects lead to the formation of bound magnetic polarons which induces a weak ferromagnetic behavior in air-annealed 3% Co-doped TiO2 at room temperature. It is observed that irrespective of the dopant ion, whether magnetic or nonmagnetic, the overlapping of bound magnetic polarons alone can induce ferromagnetism, while the magnetic impurities give rise to an enhanced paramagnetic moment for higher Co concentrations. A detailed understanding on the variation of these magnetic properties by estimating the concentration of bound magnetic polarons is presented, which is in corroboration with the photoluminescence studies. The observed band-gap narrowing in Co-doped TiO2 nanostructures and the mechanism underlying the magnetic interactions associated with the magnetic impurity concentration are advantageous from an applied perspective, especially in the field of spintronic and magneto-optic devices.

■

Due to its unique properties it finds a wide range of applications in paints, sunscreens, UV-blocking pigments, photocatalysis, sensors, and dye-sensitized solar cells.7−10 It has been largely studied as a photocatalyst for water and air purification, degradation of dyes, pesticides, etc.11−17 However, due to its wide band gap (3.0−3.2 eV), its application as a photocatalyst is limited to the UV region. Considerable efforts

INTRODUCTION

Titanium dioxide (TiO2), a low-cost and nontoxic wide-bandgap semiconductor compound, has received much attention due to its excellent properties like high chemical and thermal stability, biocompatibility, high refractive index, optical transparency in the UV and visible regions, relatively high photocatalytic activities, etc.1−5 It exists in three different polymorphs like rutile, anatase, and brookite,6 out of which anatase is stable at low temperature and rutile is the most stable phase in bulk form and at high temperature, while preparation from solution phase favors the anatase structure. © XXXX American Chemical Society

Received: July 11, 2018 Revised: October 28, 2018

A

DOI: 10.1021/acs.jpcc.8b06646 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C

arises on the defect concentration and necessity of magnetic impurities such as Fe, Co, or Ni for observation of FM. Therefore, in the present investigation, we have systematically studied the Co-doped TiO2 samples to ensure whether oxygen vacancies alone can induce FM in TiO2 nanoparticles or if doping is necessary to enhance magnetization. To explore more on these aspects, we have also carried out a deep analysis of elemental oxidation states and magnetization measurements where a theoretical Langevin fit is employed to estimate the bound magnetic polarons (BMP) concentration which ultimately leads to the magnetic behavior of the material. In the present study, we try to emphasize that doping with magnetic impurities associated with oxygen vacancies could enable coupling between them to enhance BMP overlapping and FM exchange interaction in the system apart from their potential optical properties. To the best of our knowledge, very few studies have reported such significant reduction in band gap due to Co doping in TiO2 as well as observation of overlapping of BMPs in Co-doped TiO2 to explain the magnetic properties.

have been given to reduce its band gap so that it can be active in the visible region by doping with metallic as well as nonmetallic elements,18−25 which can have a wide range of useful applications as energy materials. The electrical and optical properties of the semiconductor material can be significantly modified by introducing appropriate dopants and defects. In particular, nanocrystalline TiO2 has been used as an effective material in photocatalysis, solar cells, sensors, photodegradation of organic pollutants, water splitting, hydrogen generation, and so many other applications due to its much improved physical and chemical properties. Various techniques like sol−gel, hydrothermal, spray pyrolysis, solid-state reaction, metal−organic chemical vapor deposition, ion implantation, etc.,26−33 have been used to produce nanocrystals of TiO2, and the properties of TiO2 nanomaterials are found to depend on the structure, morphology, and mostly on the synthesis techniques. Along with the above-mentioned applications, TiO2 has also been studied largely as a dilute magnetic semiconductor (DMS) material for spintronic and magneto-optic devices34−36 due to its good optical transmission in the UV−visible as well as in the near-infrared regions. DMSs are the materials in which a small fraction of cations in the host lattice is replaced by magnetic impurities. Room-temperature ferromagnetism (RTFM) is required for spintronic devices for their practical utilization, and also the ferromagnetism should be intrinsic. After the discovery of RTFM in Co-doped anatase TiO2 and ZnO,36,37 immense interest has been created on oxide-based DMS for spintronics devices. In particular, TiO2 has been given much importance due to its use as an excellent photocatalyst as well as magneto-optic devices. However, in spite of a good amount of work on TiO2-based DMS, the origin of RTFM in TiO2 is still under debate. There are controversial reports regarding the existence of RTFM in TiO2, i.e., some have reported that the origin of RTFM is due to the intrinsic defects like oxygen vacancy or Ti interstitials, while others have reported the nature as extrinsic arising from magnetic clusters due to transition metal doping.38−46 Even RTFM has been found in undoped and nonmagnetic elementdoped TiO2 and other oxides42,47 which help to understand the role of defects in creating ferromagnetic ordering. However, due to the unambiguous nature of defects, controversies regarding the existence of RTFM remain a challenge for the researchers. Because of both intrinsic as well as extrinsic reasons for the RTFM in TiO2, the magnetic property also depends on fabrication, growth conditions, and doping agents. Cobalt (Co) is an effective dopant to improve the optical as well as the magnetic properties of TiO2.48,49 Incorporation of the Co ions into the TiO2 host lattice involves the replacement of TiO2 atoms/ions either on regular Ti sites or on interstitial sites. However, the optical and magnetic properties of the final system depend on many more parameters including the concentration and distribution of the Co atoms/ions, type and concentration of defects, etc. In the present study we report on the structural, optical, and magnetic properties of Co-doped TiO2 nanocrystals synthesized by a simple, cost-effective sol− gel technique. Here, we have reported a significant reduction in the band gap due to Co doping and the existence of weak ferromagnetism in undoped and a lower concentration of Codoped TiO2 and the appearance of paramagnetism in a higher concentration of Co-doped TiO2. Since the defect-induced TiO2 can generate FM behavior, the controversial question

■

EXPERIMENTAL SECTION Undoped and 3−12 wt % Co-doped TiO2 nanocrystals were synthesized by a simple sol−gel technique where titanium butoxide and cobalt nitrate were used as the Ti and Co ion precursors, respectively. The synthesis approach enables the formation of stable anatase phase in all of the studied samples and the synthesized undoped and Co-doped samples were henceforth named as TP, T3Co, T6Co, T9Co, and T12Co for undoped, 3%, 6%, 9%, and 12% Co-doped samples (all of the results corresponding to undoped TP sample are taken from our previous work which is unpublished). The resultant sol− gel products were calcined at 400 °C for 3.5 h in a muffle furnace under ambient air atmosphere. The structure and phase purity were studied by X-ray diffraction (XRD) at room temperature using a PANalytical X’Pert Pro diffractometer with Cu Kα radiation. The crystal structure of anatase TiO2 was generated using vesta software. Micro-Raman spectroscopy was performed using a Horiba Jobin Yvon LabRAM HR 800 micro-Raman spectrometer using a 1800 grooves mm−1 grating, with a spectral resolution of 2 cm−1. All of the experiments were performed using an excitation wavelength of 405 nm. The morphology and microstructure of nanocrystals were studied using high-resolution-transmission electron microscopy (HR-TEM, FEI Tecnai F20, operated at 200 kV). Again, to confirm the elemental composition, an energydispersive X-ray fluorescence spectrometry (ED-XRFS) study was carried out using PANalytical Epsilon 3. Functional group identification of the prepared nanocrystals was investigated by Fourier transform-infrared (FT-IR) spectra using a Bruker FTIR spectrometer. The UV−vis spectra of the prepared samples were recorded by a Shimadzu UV 2401 PC spectrophotometer. The emission spectra of the prepared samples were obtained from a spectrofluorometer (Cary Eclipse, Varian). The X-ray photoelectron spectroscopy (XPS) measurements were carried out on powder samples at room temperature. The XPS spectrum was acquired using a PHY 5000 Versa Probe II, ULVAC-PHI, Inc. instrument with an Al Kα X-ray source. The pressure in the XPS chamber during the measurements was 5 × 10−10 mbar. The binding energy was corrected by taking C 1s as reference energy (C 1s = 284.60 eV). A wide scan was collected to ensure that no foreign materials were present on the sample surface. The high-resolution scans of the Ti 2p, Co B


Article


Figure 1. (a) XRD patterns of Co-doped TiO2 nanocrystals. (b) Peak shift associated with T3Co, T6Co, T9Co, and T12Co.

Figure 2. Refined XRD patterns of Co-doped TiO2 nanocrystals: (a) T3Co, (b) T6Co, (c) T9Co, and (d) T12Co.

Table 1. Refinement Parameters Obtained for Undoped and Co-Doped TiO2 Nanocrystals TP phase crystal structure space group

anatase tetragonal I41/amd

a (Å) c (Å) volume (Å3)

3.7852 (1) 9.4914 (3) 136.00 (2)

Ti/Cox (4a) Ti/Coy (4a) Ti/Coz (4a) Ox (8e) Oy (8e) Oz (8e) Biso (Ti/Co) (Å2) Biso (O) (Å2)

0.0000 0.7500 0.1250 0.0000 0.2500 0.0835 0.0009 0.0075

Rp Rwp χ2

4.048 5.299 1.634

T3Co anatase tetragonal I41/amd lattice parameters 3.7894 (1) 9.4926 (3) 136.31 (1) atomic positions 0.0000 0.7500 0.1250 0.0000 0.2500 0.0848 0.0006 0.0043 residual parameters 2.326 3.158 1.793 C

T6Co

T9Co

T12Co




3.7906 (2) 9.4942 (1) 136.42 (2)

3.7922 (4) 9.5130 (3) 136.81 (1)

3.7951 (1) 9.5029 (2) 136.87 (2)

0.0000 0.7500 0.1250 0.0000 0.2500 0.0829 0.0004 0.0010

0.0000 0.7500 0.1250 0.0000 0.2500 0.0813 0.0003 0.0007

0.0000 0.7500 0.1250 0.0000 0.2500 0.0867 0.0008 0.0009

2.380 3.307 1.994

3.183 4.052 1.399

3.253 4.487 2.520 DOI: 10.1021/acs.jpcc.8b06646 J. Phys. Chem. C XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry C 2p, and O 1s regions were collected. Curve fitting to the XPS spectrum was done using MultiPak Spectrum: ESCA. Background subtraction was done using the Shirley method. Roomtemperature magnetic studies were performed using a vibrating sample magnetometer attached to the physical property measurement system, Quantum Design Inc. (USA).

■

RESULTS AND DISCUSSION The parent compound is crystallized into a tetragonal crystal structure with anatase form as reported in our earlier studies (unpublished results), the data of which is used for comparison purpose only. The XRD patterns of Co-doped TiO2 powder with varying Co concentrations (T3Co, T6Co, T9Co, T12Co) are shown in Figure 1a. All of the diffraction peaks are well indexed and correspond to the tetragonal anatase phase of TiO2 (space group I41/amd (ICDD 78-2486). The diffraction patterns do not show any secondary phase peak related to rutile or Co or CoO showing that anatase phase is not disturbed due to Co doping in TiO2. A slight shift in the peak position of the most intense peak (101) (shown in Figure 1b) and the change in full-width at half-maximum (fwhm) from undoped to doped samples indicate the incorporation of Co in TiO2.50 The average particle size of undoped and Co-doped TiO2 powder determined from the XRD pattern using Scherrer’s equation taking the (101) peak as reference was found to be around 14 and 14−20 nm, respectively. The Rietveld refinement of the XRD data was done using Fullprof software, which is shown in Figure 2a−d, and the refined parameters are given in Table 1. The Rietveld refinement confirms that TiO2 crystallizes in the anatase tetragonal structure, and no secondary phase has been detected in Codoped samples. Refined data shows an increase in lattice parameter as well as unit cell volume with an increase in Co concentration, which is expected as the ionic radius of Co2+ (0.65 Å) is larger than that of Ti4+ ions (0.61 Å),51 whereas the ionic radius of Co3+ (0.55 Å) is smaller than that of Ti4+ ions, but a still increasing Co2+/Co3+ ratio ultimately leads to the increased lattice parameters. Here, the coordination number of Ti4+ cation is considered to be 6, which is twice the coordination number of the O2− ion. The increase in particle size is also attributed to the difference in ionic radius of dopant and host ions. The difference between the ionic radii of the dopant and host ions is expected to cause a small enhancement of the TiO2 unit cell, and this increase in volume is proportionate with the increase in doping level, which is in accordance with Vegard’s law.52 The crystal structure of anatase TiO2 was generated using Vesta software, which is shown in Figure 3. Rietveld refinement data confirms that Co is incorporated into the TiO2 lattice and has substituted the Ti4+ ions, and due to the small difference in the ionic radii of both ions the local structure might have been little disturbed, causing a change in fwhm. The presence of any other impurity phase is ruled out within the detection limit of XRD. Raman spectroscopy has been used to find the crystallinity, phase, presence of defects, strain, and disorder induced in the crystal due to dopant incorporation in the host lattice. According to symmetry group analysis, tetragonal anatase phase of TiO2 has six active Raman modes and are reported to appear near 144 (Eg(1)), 197 (Eg(2)), 399 (B1g(1)), 516 (A1g+B1g(2)), and 639 cm−1(Eg(3)).53 The Eg band appears due to O−Ti−O symmetric stretching vibration in TiO2, B1g appears due to O−Ti−O symmetric bending vibration, and A1g appears due to O−Ti−O antisymmetric bending vibration.

Figure 3. Crystal structure of Co-doped anatase TiO2 generated using the vesta software.

The Raman spectra of all of the studied nanocrystals were taken at room temperature in the range 100−700 cm−1 are shown in Figure 4a, and the inset shows the position of the corresponding Raman modes of T6Co, T9Co, and T12Co. When Co2+/Co3+ substitutes Ti4+, to balance the charge neutrality, there will be some disorder in the lattice which can be in the form of defects like oxygen vacancy or Ti interstitials. With the increase of the doping concentration, disorder increases and the ideal symmetry of the crystal will be destroyed which results in broadening of the Raman bands.54 The Raman bands at 144, 197, 399, 516, and 639 cm−1 can be assigned as E1g(1), E2g(2), B1g(1), A1g+B1g(2), and E3g(3) modes of anatase phase, respectively; the presence of these confirm the tetragonal anatase phase of TiO 2 . The deconvoluted Raman spectra of all of the doped samples are shown in Figure.S1, and no mode corresponding to any other phase has been observed. The absence of any other mode related to Co, CoO, or Co−Ti species within the detection limit of Raman spectra of doped samples indicates that Co might have gone to the substitutional site replacing Ti in TiO2. The phonon confinement due to nanoscale crystallite size and nonstoichiometry due to defect-induced disorder in the lattice results in shifting and broadening of the Raman bands. Defects like oxygen vacancy in the material strongly affect the Raman modes. In our study, the most intense Eg(1) Raman mode at 144 cm−1 shows blue shifting and broadening with doping (Figure 4b). XRD and TEM (discussed later) results show the crystallite size is in the nanoscale range. Thus, both the phonon confinement and the disorder in the host lattice due to the incorporation of Co atoms/ions creating defects like oxygen vacancies in TiO2 have caused the shifting of the position and broadening of the Raman bands.55−57 To study the microstructure of the samples, TEM analysis was done, and the microscopic images of Co-doped TiO2 powders are shown in Figure 5a−d. The particles observed here are almost spherical in shape, and agglomeration of particles can be seen in all of the samples with an average size in the range 10−20 nm, which agrees well with the particle size derived from XRD data. The high-resolution TEM images (Figure 5i−l) clearly show the crystalline nature of the samples with lattice fringes corresponding to different planes like (101), (200), and (004) planes of tetragonal anatase phase of TiO2. Insets in Figure 5i−l show the FFT pattern taken on all the samples, which again confirms the crystalline nature of all the samples. The selected area electron diffraction (SAED) D


Article


Figure 4. (a) Raman spectra of T3Co, T6Co, T9Co, and T12Co samples. Inset shows the magnified view of Raman spectra of T6Co, T9Co, and T12Co samples, (b) Expanded region of the Eg(1) mode in the range 120−170 cm−1of undoped and Co-doped TiO2 nanocrystals. Parent sample data is taken from our earlier work (unpublished results) for reference purpose only.

Figure 5. Co-doped TiO2 nanocrystals. (a−h) TEM images showing the nanocrystal formation. (i−l) HR-TEM images showing lattice fringes with FFT shown in the inset. (m−p) SAED patterns of T3Co, T6Co, T9Co, and T12Co.

patterns taken on all the Co-doped samples show clear distinct rings corresponding to different planes of tetragonal anatase TiO2 structure. The rings obtained in the SAED pattern are indexed well to different planes of anatase TiO2, which indicate the formation of polycrystalline anatase TiO2 nanopowder. The planes found out from d spacing which were calculated using ImageJ software were similar to the planes obtained in XRD which correspond to tetragonal anatase phase of TiO2. In order to check the chemical compositions of the prepared nanocrystals, we employed ED-XRFS analysis. The chemical compositions of the elemental oxides obtained (shown in

Table 2) are found to be the same as that of the original stoichiometry within the limits of experimental error. FTIR spectra were taken to investigate the vibrations of the TiO2 lattice. Figure 6 shows the FTIR spectra in transmission mode taken on Co-doped TiO2 nanopowders. The IR band around 3600 cm−1 corresponds to the O−H stretching vibration, and the band around 1628 cm−1 is attributed to the H−O−H bending vibration of adsorbed atmospheric water on the sample surface. 58 The band around 847 cm−1 corresponds to anatase TiO2 and is attributed to a Ti−O−Ti stretching vibration.59 The shifting of band position around E


Article


region and an increase in absorption in the visible region in comparison to undoped sample. Also, the doped samples show a broad peak in the visible region. The broad absorption in the visible region in Co-doped samples is due to ligand field transition of Co2+ in octahedral coordination.61 In anatase TiO2, Ti4+ is surrounded by six oxygen ion O2− forming a TiO62− octahedron. In Co-doped TiO2, with the substitution of Co2+/Co3+ on the Ti4+ site, Co2+/Co3+ will experience a strong crystal force due to surrounding oxygen ions. Due to this strong crystal field interaction, the d-band states of Co2+ will split into ground and excited states, giving rise to a d−d electronic transition which falls in the visible region. This is a strong indication of Co2+ substitution for Ti4+ in the TiO2 lattice, whereas Co3+ will not exhibit any d−d electronic transition while splitting into ground and excited states. The band gaps of doped samples were found from Tauc’s plot given in Figure 7b, from which a reduction of the band gap to 2.24 eV in the T12Co sample is observed. In the case of T9Co and T12Co, the band-gap values are close, which corroborates with the intensity variation in PL spectra, which is discussed in the next section. The errors in estimating the band gap for T3Co, T6Co, T9Co, and T12Co are 0.04, 0.05, 0.03, and 0.02 eV, respectively, which is represented in Figure 7b. It is expected that beyond a certain dopant concentration it is very difficult to induce a drastic reduction in the band gap and other associated properties, and hence, a dilute concentration of an impurity can tailor the material properties in a better way. Justica et al.62 theoretically suggested that a high vacancy concentration could induce a vacancy band just below the conduction band, and Zuo et al. reported that due to an oxygen vacancy associated with Ti3+ in TiO2, a miniband can form closely below the conduction band, resulting in narrowing of the band gap.63 A similar observation of band-gap narrowing related to oxygen vacancy is observed in the ZnO system.64 The significant amount of band-gap narrowing as shown in Figure 7c is important in achieving visible light photocatalysis and exploring other practical applications. To further elucidate the nature of defects, PL studies were performed on all of the samples. Figure 8a shows the PL spectra of both undoped and doped TiO2 samples taken at room temperature. All of the PL spectra have been recorded for an excitation wavelength of 325 nm. The undoped sample exhibits emission peaks around 390, 450, 490, 520, and 580 nm, which correspond to near band edge emission of host TiO2, self-trapped excitons, charge transition from Ti3+ to TiO62− linked with oxygen vacancies, F+−center formation,

Table 2. ED-XRFS Data Obtained for Co-Doped TiO2 Nanocrystals compound T3Co T6Co T9Co T12Co

percentage of elemental oxide 96.94% 93.92% 90.89% 87.88%

TiO2 + 3.06% Co3O4 TiO2 + 6.08% Co3O4 TiO2 + 9.11% Co3O4 TiO2+ 12.12% Co3O4

Figure 6. FTIR spectra of Co-doped TiO2 nanocrystals.

847 cm−1 indicates the presence of defects, especially oxygen vacancies due to Co incorporation in the TiO2 lattice. UV−vis absorption spectroscopy is an important tool to study the optical transitions and defect states due to the incorporation of dopants in the host lattice. The absorption spectra of Co-doped samples are shown in Figure 7a. The absorption spectrum of TiO2 sample generally exhibits a sharp absorption around 385 nm which corresponds to the electronic transition from the valence band (O2p state) to the conduction band (Ti3d state).60 The undoped sample shows a strong absorption in the UV region and high transparency in the visible region. The absorption spectra of Co-doped samples show shifting of the absorption edge toward the visible region. The red shift of the absorption edge due to the incorporation of transition metal ions in TiO2 has been reported to be due to sp−d exchange interactions between band electrons and localized d electrons of dopant ions. The s−d and p−d exchange interactions result in downward shifting of the conduction band edge and upward shifting of the valence band edge due to which narrowing of the band gap occurs.50,51,58 The doped samples show a decrease in absorption in the UV

Figure 7. Co-doped TiO2 nanocrystals: (a) solid-state UV spectra and (b) Tauc’s plot. (c) Observation of narrowed band gap with Co substitution. Band gap corresponding to the undoped sample is taken from our earlier work (unpublished results) for reference purpose only. F


Article


Figure 8. PL spectra of (a) undoped and Co-doped TiO2 nanocrystals. (b) Peak-fitted PL spectra corresponding to T3Co. (c) Schematic showing possible PL emissions in Co-doped TiO2.

Figure 9. Magnetic response of Co-doped TiO2 nanocrystals (a) at 300 K (inset shows variation in coercivity at 100 Oe). (b) Representation of weak ferromagnetism in T3Co and paramagnetism in T6Co, T9Co, and T12Co; inset shows M−H behavior of undoped TiO2 taken from unpublished results. (c) Increasing trend in net magnetization with Co substitution at both 2 and 5 Tesla magnetic field.

behavior of all samples could be seen. However, the M−H curves show weak ferromagnetism (FM) in undoped and T3Co sample (Figure 9b) when taken at low fields. There is an increase in the magnetic moment with an increase in the concentration of Co. The inset in Figure 9a shows the variation in coercivity (Hc) of all of the samples at 100 Oe which gives a Hc of 2.94 × 10−5 emu for T3Co sample, which is higher than that of the Hc of 8.93 × 10−6 of the TP sample. There is no Hc in T6Co, T9Co, and T12Co samples; also, a linear increase of magnetization with an increase in H indicates the absence of FM in these samples. RTFM is generally expected in Co-doped TiO2 due to the FM nature of Co. Some works suggested the observation of FM due to Co clusters, while some reported that FM arises due to exchange coupling between substituted Co ions and the charge carriers trapped by oxygen vacancies. 38−46 Other reports 67 have also shown the suppression of FM in TiO2 upon Co doping, resulting in PM or antiferromagnetic (AFM) samples, which may have originated due to segregation of AFM CoO or Co3O4. However, in our study, XRD and Raman analysis exclude the presence of any secondary phases like Co or Co-related oxides. It is interesting to note that undoped TiO2 shows weak FM, which has originated due to the presence of intrinsic defects such as oxygen vacancies/Ti3+ ions or Ti interstitials (unpublished results). The presence of FM and more Hc in T3Co than in undoped sample indicate that the FM is not only intrinsic but also that magnetism can be enhanced due to Co doping, which may have created more oxygen vacancy/Ti3+ ions. However, at higher concentration of Co-doped samples, although the magnetization is increased as shown in Figure 9c,

and hydroxyl (OH) species, respectively (unpublished results). A similar observation has been noticed for all the doped samples, and as a representative of the series, Figure 8b shows the deconvolution of the PL spectrum of the T3Co sample. The emission peak around 390 nm can be ascribed to the near band edge emission of host TiO2. The emission peaks around 400−450 nm are due to self-trapped excitons (STE), oxygen vacancies, surface defects, etc.65,66 The STE originates when a trapped electron captures a hole in TiO2. In an STE emission, the recombination occurs through oxygen vacancies. Since oxygen vacancies are intrinsic defects in TiO2 and Co doping also creates oxygen vacancies due to the substitution of Co2+/ Co3+ on Ti4+, the emissions around 400−450 are supposed to be due to STE. The 489 nm emission is due to charge transition from Ti3+ to TiO62− linked with oxygen vacancies, and the nearly 520 nm emission is due to the F+−center. The emission around 600 nm may be due to the electron transition from F+−center to the acceptor level just above the valence band. The presence of hydroxyl (OH) species is detected in the FTIR spectra as shown in Figure 6 and confirmed from XPS results as well, which are discussed later in this manuscript, which can generate an acceptor level just above the valence band and is contributing to the observed near 600 nm PL emission. The different possible interactions of the excited electron are schematically represented in Figure 8c. The magnetic behavior of undoped and Co-doped TiO2 nanopowders was studied through VSM. The M−H measurements were carried out at room temperature with field varying from −90 to +90 kOe. The M−H curves of all samples are shown in Figure 9a, from which the paramagnetic (PM) G


Article


Figure 10. XPS spectra of T3Co, T9Co, and T12Co samples: (a) Ti 2p, (b) Co 2p, and (c) O 1s.

and a final peak at 797.64 eV corresponding to Co3+ 2p1/2 in Co 2p spectra giving a line spacing of 15.8 eV between the two, along with two strong shakeup satellite peaks at 786.01 and 802.35 eV revealing a Co atom in TiO2 exists in mixed oxidation states of +2/+3. The separation of the Co 2p peak into 2p3/2 and 2p1/2 for both Co2+ and Co3+ indicates that the valence state of Co is 2+/3+. Broadening of the Raman modes, shifting of absorption edge to the visible region, and XPS data strongly support the substitution of Co ions for Ti4+ in the TiO2 lattice which creates an oxygen vacancy. Confirmation of the presence of Ti3+, Co2+, and Co3+ in the XPS analysis indicates that undoped (unpublished results) and Co-doped samples possess a certain amount of oxygen vacancies, which corroborates the Raman and optical absorption spectral data. There could be a slight difference between the number of oxygen vacancies of T3Co, T9Co, and T12Co due to the fact that T3Co consists of Ti3+, which is absent in T9Co and T12Co. This difference in oxygen vacancies is ultimately reflected in the slight peak shifting of the O 1s spectra as shown in Figure 10 c. The deconvoluted O 1s spectra are represented by three symmetric Gaussian curves, similar to that reported in the literature.68 The intense peak at about 529.8 eV arises due to the oxygen in the TiO2 crystal lattice (OL), while the other oxygen peaks are the outcome of Ti−O/ Co−O bonds (OTi3+/OCo2+, 531.0 eV) and the hydroxyl group (OH, 532.0 eV), respectively, as shown in Figure 10 c. The concentration of Ti3+, Ti4+, Co2+, and Co3+ has been estimated from the XPS results,69 which are shown in Table.S1.

there is an absence of Hc which indicates that FM ordering has been lost in these samples, which may be due to the interaction of oxygen vacancy/Ti3+ ions with Co ions in different ways promoting FM/PM, which will be discussed in detail in the coming sections. Raman, UV−visible, and XPS analysis indicate the presence of defects like oxygen vacancies due to Co doping. The oxidation states of Co and Ti were further analyzed by XPS for T3Co, T9Co, and T12Co samples as a representative of the series. The survey spectra of all three samples are shown in Figure.S2, from which the peak of Co can be seen along with the Ti, O, and C peaks, which were expected. The highresolution spectra of Ti 2p, Co 2p, and O 1s were recorded and are shown in Figure 10a−c along with the deconvolution. The Ti spectrum of T3Co sample can be fitted with four peaks corresponding to Ti4+ 2p3/2 at 458.19 eV, Ti4+ 2p1/2 at 463.92 eV, Ti3+ 2p3/2 at 456.03 eV, and Ti3+2p1/2 at 461.67 eV as shown in Figure 10 a. The line separation between Ti4+ 2p1/2 and Ti4+ 2p3/2 is 5.72 eV, which is consistent with the standard binding energy of TiO2.66 Interestingly, T9Co sample exhibits a single oxidation state of Ti4+, and the peak positions are located at Ti4+2p3/2 at 458.54 eV and Ti4+2p1/2 at 464.27 eV, respectively. Again, in T12Co sample, a similar observation as T9Co is noticed with Ti4+ peaks at Ti4+ 2p3/2 at 458.54 eV and Ti4+ 2p1/2 at 464.27 eV. Furthermore, the Co peak appearing in T3Co, T9Co, and T12Co (Figure 10b) can be deconvoluted into four peaks: one at 780.66 eV which corresponds to Co2+2p3/2, a second peak at 782.06 eV which corresponds to Co3+ 2p3/2, another at 796.25 eV corresponding to Co2+2p1/2, H


Article


Figure 11. Langevin-fit to estimate the BMP concentration for (a) T3Co, (b) T6Co, (c) T9Co and T12Co.

of an F−center where one of the electrons associated with the F−center will have a tendency to interact with the adjacent localized Ti4+ and convert Ti4+ ions to Ti3+ ions, resulting in the formation of an F+−center.71 Interestingly, it can be observed that XPS results validate the existence of Ti3+ and associated F+ defect counterparts as shown in Figure 10a. There is a high chance of localization of electrons in the F+− center and could develop bound magnetic polarons (BMPs) by ordering the electron spin corresponding to the Ti3+ (3d1) situated near the oxygen vacancies. Thus, s−d exchange interaction involved with the 1s1 electron spin associated with the F+−center and the 3d1 electron spin of Ti3+ ions could easily favor long-range FM ordering. A similar interaction is possible between the F+−center and the Co2+ ions to develop further BMPs; hence, another possible interaction in lightly Co-doped T3Co is between the F+−centers and the 3d5 electron spin of Co 2+ ions, leading to FM ordering. Interestingly, it is well reported that the doubly occupied F− centers are more likely to form a 1s2 state, which could easily encourage a weak AFM interaction.72 Hence, the BMP formation due to Ti3+ and Co2+ ions with the F+−center, in which the electrons are locally trapped by oxygen site vacancies, where an orbital overlapping with the unpaired 3d1 electron of Ti3+ ions or the unpaired 3d5 electron of Co2+ ions, is explored in detail to investigate the observed RTFM at lower fields in the present case, i.e., in T3Co sample. It is clear from Table S1 that the Ti3+/Ti4+ ratio is 0.65 for the undoped sample and 0.52 for T3Co. Again. the Co2+/Co3+ ratio is observed to be 0.73 for T3Co; at the same time a slight amount of oxygen vacancy is also evident from Table.S1, and these results indicate the s−d exchange interaction associated with the 1s1 electron spin of trapped state and 3d1 electron spin of Ti3+ ions for undoped sample, whereas the 1s1 electron spin of the trapped state, 3d1 electron spin of Ti3+ ions, and 3d5 electron spin of Co2+ ions have induced magnetism in T3Co. To explore more on the specific role of oxygen vacancy and dopant ion concentrations in Co-doped TiO2, we have carried out the BMP fitting on all the doped samples to have a detailed

In the present investigation we have explored the role of oxygen vacancies in undoped TiO2 sample and the effect of Co substitution causing the magnetic response of the TiO2 material. The origin of RTFM observed for the undoped sample (unpublished results) has been mainly attributed to the defect concentration and Ti oxidation state which is shown in the inset of Figure 9 b. The Ti3+ ions have unpaired electrons with a 3d1 electronic configuration, and trapped electrons in the vacancy sites have 1s1 configuration. Therefore, s−d interaction is possible between Ti3+ ions and trapped electrons in the vacancy sites. If only Ti4+ was present in the undoped sample, there would not have been a RTFM behavior since TiO2 with Ti4+ is a perfect diamagnetic material. Hence, depending on the magnetic interactions present between the Ti3+/Ti4+ ions and the Ti3+ interaction with oxygen site vacancies, different magnetic behaviors can be exhibited by the undoped TiO2. Now, let us consider the origin of RTFM observed for the lower doped T3Co sample. Theoretically, it is well reported that a weak FM component in the Ti atom can exist as a result of oxygen vacancy even without the presence of any magnetic impurities at the Ti site.44 The ab initio electronic states calculation demonstrates the impact of oxygen vacancy to act as an electron doping in TiO2 material but which is not sufficient to exhibit a considerable magnetic moment.70 Hence, depending upon the magnetic interactions present between the Ti3+/Ti4+ ions and the dopant ions associated with oxygen site vacancies, magnetic interactions such as diamagnetism, AFM or FM may be possible within the system. The controversial results associated with the different TiO2-based systems indicate that the charge redistribution created by the oxygen vacancies is the key factor determining the magnetic properties of TiO2. The F−center interaction could be predominant when there is the presence of an oxygen vacancy, where the surface of TiO2 could trap the electrons and charge redistribution is caused by the oxygen vacancy. The explanation for this interesting phenomenon can be given in such a way that an oxygen vacancy on the surface of TiO2 locally traps the electrons, ultimately leading to the formation I


Article

The Journal of Physical Chemistry C Table 3. Parameters Obtained from BMP Fitting for Undoped and Co-Doped TiO2 Nanocrystals sample name

M0 (×10−4 emu/g)

Meff (×103 μB)

χm (×10−6 emu/g-Oe)

N (×1017/cm3)

TP T3Co T6Co T9Co T12Co

6.81 148.22 17.23 6.49 3.97

7.21 5.94 4.01 7.16 7.13

0.37 2.50 5.40 7.89 10.48

0.11 2.70 0.46 0.09 0.06

Figure 12. Schematic diagram showing the possible BMP formation involved in (a) T3Co and (b) T12Co for explaining the magnetic behavior of the Co-doped TiO2 nanocrystals. (c) Variation of BMP concentration and Co2+/Co3+ ratio with Co substitution.

Hence, it is clear that BMPs formed as a result of both Co2+ ions and Ti3+ ions are responsible for the weak FM exhibited by T3Co. The underlying mechanism of FM in heavily Co-doped TiO2 samples is rather complicated due to the fact that doubly occupied F−centers and the formation of a 1s2 state can adversely affect the FM interactions, and a few mechanisms are put forward, which includes the Ruderman−Kittel−Kasuya− Yoshida (RKKY)75 model, the FM double exchange and AFM superexchange coupling,76 and the coupling of BMPs formed at the F+−center and the Co2+ surrounding of doped nanocrystals,77 etc. The inherent tendency of defect formation in TiO2 nanocrystals increases the possibility of interactions among defects and carriers along with their coupling that could lead to a variety of magnetic interactions depending upon the defect concentration and dopant ion concentration. Considering the M−H curve represented in Figure 9, it is observed that the net magnetic moment increases with Co substitution and at the same time Hc decreases considerably, leading to a weak FM to PM transition with the increased Co2+ concentration. The BMP concentration, total BMP magnetization, and paramagnetic susceptibility obtained for T6Co are within the permissible limit, and its BMP concentration considerably decreases with Co substitution, whereas paramagnetic susceptibility increases systematically. Now let us consider the reason for the decreased BMP concentration in T6Co in comparison to T3Co. Obviously, more Co2+ ions are present in T6Co when compared with T3Co; hence, the system tends to interact with the increased number of Co2+ sites, and again some amount of Co3+ generation occurs to maintain charge neutrality. The interaction with the Ti4+ site is hindered here due to the fact that the stability of Ti4+ (having noble gas configuration) is more in comparison to Co2+ (can easily oxidize to Co3+). Hence, when more Co2+ ions are present, the charge neutrality concept is bound to the Co2+ site and the Ti4+ site gets undisturbed as shown in Figure 10b. The BMP formation in the T6Co case is only due to the interaction of F+−centers with 3d states of Co2+ ions and obviously will

understanding regarding the suitability of the BMP model and its relationship to the contribution of oxygen vacancy. The fitting has been carried out on the observed M versus H data to the BMP model as discussed in the earlier reports.73,74 According to this model, the measured magnetization can be fitted using the relation M = M 0L(x) + χm H

(1)

In this equation the first term is from the BMP contribution and the second term is due to a paramagnetic matrix contribution. The value of M0 can be obtained as the product of N and ms, where N is the number of BMP involved (per cm3) and ms is the effective spontaneous magnetic moment per BMP. The Langevin function, L(x), is defined as L(x) = coth(x) − 1/x, where x = meffH/kβT. The term meff is the true spontaneous moment per BMP, and it can be approximated to ms. The parameters M0, meff, and χm are the variables in the fitting process. The experimental and fitted data to estimate the BMP concentration are given in Figure 11 for all the doped samples; it can be observed that the fitted data is very well in accordance with the experimental data, and the obtained parameters are tabulated in Table 3. The number of BMPs required to induce a long-range FM is reported to be of the order of 1020/cm3, whereas the undoped sample contains 1016 BMPs, and T3Co sample contains nearly 1017 BMPs as a result of air annealing, and ultimately RTFM is observed only at lower fields. The low concentration of oxygen site deficiency due to air annealing is evident from the approximate low 2 − δ value as shown in Table S1, which is responsible for the low BMP concentrations observed in both undoped and Co-doped samples. When a lower concentration of Co exists in T3Co, the system is bound to interact with Ti4+ defect sites to induce the formation of Ti3+ in order to maintain the charge neutrality. Here, a very few Co2+ ions from Co(NO3)2 will be converted to Co3+ ions, which is the most stable state (but diamagnetic) among different Co ions due to d-orbital splitting into a lower energy t2g state and high-energy eg state. The results are well in agreement with the XPS results. J


Article


results open up a new pathway for synthesizing the defect engineered TiO2 nanostructures for the advanced functional applications such as future spintronic and magneto-optic devices.

result in a decreased concentration of BMPs as the s−d exchange interaction among F+−center and Ti3+ is absent in the T6Co sample. The BMP fitting of T9Co and T12Co is represented in Figure 11c and 11d, where a similar observation is noticed. Hence, T12Co is having the least number of BMPs among the synthesized samples and is found to be least ferromagnetic. It is evident from the XPS results that the ratio of Co2+/Co3+ is increasing with Co substitution, which is represented in Table S1. The increasing trend in net magnetization results from the PM moment contributed by this hiking Co2+ concentration where the isolated Co spins in the trapped centers are responsible for the paramagnetic behavior of T6Co, T9Co, and T12Co. Interestingly, the presence of Ti3+ ions has almost disappeared in these heavily doped samples. Figure 12a and 12b show the schematic representations explaining the possible BMP formation involved in the Co-doped TiO2 nanocrystals. The different possible mechanisms involved here can be explained in detail as discussed below. The higher concentration of Co yields the decreased FM behavior, and a similar observation can be found in most of the recently reported systems.76,78 It could be due to the fact that a large concentration of Co2+ doping (as evident from the XPS analysis) can result in the following possibilities: (i) the AFM ordering between two nearby Ti4+/Co2+ ions in the absence of oxygen vacancy sites by superexchange interaction, (ii) most of the Co2+ spins exist in the isolated PM spin system, or (iii) it may interact ferromagnetically by F+−center−Ti3+ ions/F+−center−Co2+ ions. Even though (i) and (iii) are present in our system, (ii) is dominated by a higher concentration of Co2+ resulting in an increased PM moment. As discussed earlier in the PL results, the concentration of oxygen vacancies is considerably lower in T12Co in comparison to T3Co. Hence, a lower oxygen vacancy concentration is partly responsible for the decreasing FM behavior in T12Co. It is observed that the BMP concentration is the maximum for T3Co, and there is a linear correlation between the increased Co doping and BMP concentration, as shown in Figure 12c. It is extremely challenging to quantitatively estimate the defect concentration from the PL analysis; the observed correlation is in good agreement with the recent report,71 and it is reported that the correlation between BMP concentration and oxygen vacancies is rarely observed. These observations impart an expression that the observed RTFM is contributed significantly by oxygen vacancies created in the Co-doped samples. It is observed that the T3Co samples have numerous oxygen vacancies due to Ti3+ and Co2+, leading to the formation of even small clusters of vacancies which could increase the chance of localization of the charges associated with the exchange interaction due to overlapping of BMPs. Hence, it is understood that in the present samples contribution of magnetization is due to the simultaneous occurrence of oxygen vacancies and BMP formation resulting in FM and PM behavior, respectively. The oxygen vacancies and BMP formation are becoming weaker with the increase of Co ions, resulting in the weakening of FM behavior, and the enhancement in magnetization occurs due to the increase in isolated Co2+ spins with the increase of Co substitution. These results suggest that by fine tuning the oxygen vacancy and the substitution of magnetic ions in TiO2 one can achieve the desired optical and magnetic properties. Therefore, irrespective of the numerous applications of TiO2 nanostructures, these

■

CONCLUSION The effect of Co doping on the structural, optical, and magnetic properties of sol−gel-derived TiO2 nanopowders was investigated. The increase in a and c axis lattice parameters found from Rietveld analysis of XRD patterns and shifting and broadening of the most intense Eg(1) mode in the microRaman study of Co-doped TiO2 nanocrystals indicate the incorporation of Co in TiO2. An increase in absorption in the visible region and shifting of the absorption edge in Co-doped TiO2 to the visible region in the UV−vis spectra indicate narrowing of the band gap due to Co incorporation in TiO2 which can be useful for visible light photocatalysis in practical applications. X-ray photoelectron spectra show the presence of Co2+ and Co3+ in Co-doped TiO2 samples, indicating the incorporation of Co in TiO2. The presence of Ti3+ and Co2+/ Co3+ in the XPS analysis indicates that undoped and 3% Codoped samples possess a certain amount of oxygen vacancies. Magnetic measurement shows weak ferromagnetic behavior in undoped which can be intrinsic in nature and 3% Co-doped TiO2 at room temperature which can be both extrinsic as well as intrinsic, while the M−H curve of higher concentration Codoped TiO2 shows paramagnetic behavior which is explained well by taking into consideration the BMP formation. A deep analysis of elemental oxidation states, magnetization measurements, and Langevin fitting helps to estimate the BMP concentration, ultimately leading to the magnetic behavior of the material. The doping with magnetic impurities associated with oxygen vacancies could enable coupling between them to enhance BMP overlapping and FM exchange interaction in the system apart from their potential optical properties. A significant reduction in band gap associated with its interesting magnetic properties makes Co-doped TiO2 a well-suited material for future spintronic and magneto-optic applications.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcc.8b06646. Deconvoluted Raman spectra; wide scan XPS spectra; estimated concentrations of Ti3+, Ti4+, Co2+, and Co3+ from XPS spectra; complete list of authors for refs 3 and 45 (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected], [email protected]. ORCID

M. Vasundhara: 0000-0002-4004-8186 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the financial support received from the Council of Scientific and Industrial Research (CSIR), Government of India. V.R.A. and B.A. are thankful to the Academy of K


Article


Synthesized by Mechanical Alloying Under Visible Light. J. Am. Ceram. Soc. 2006, 89, 515−518. (20) Kim, S.; Hwang, S.-J.; Choi, W. Visible Light Active PlatinumIon-Doped TiO2 Photocatalyst. J. Phys. Chem. B 2005, 109 (51), 24260−24267. (21) Lee, M. S.; Hong, S.-S.; Mohseni, M. Synthesis of Photocatalytic Nanosized TiO2−Ag Particles with Sol−Gel Method using Reduction Agent. J. Mol. Catal. A: Chem. 2005, 242, 135−140. (22) Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. VisibleLight Photocatalysis in Nitrogen-Doped Titanium Oxides. Science 2001, 293, 269−271. (23) Khan, S. U. M.; Al-Shahry, M. A.; Ingler, W. B., Jr. Efficient Photochemical Water Splitting by a Chemically Modified n TiO2. Science 2002, 297, 2243−2245. (24) Tang, X.; Li, D. Sulfur-Doped Highly Ordered TiO 2 Nanotubular Arrays with Visible Light Response. J. Phys. Chem. C 2008, 112 (14), 5405−5409. (25) Yu, J.; Dai, G.; Xiang, Q.; Jaroniec, M. Fabrication and Enhanced Visible-Light Photocatalytic Activity of Carbon Self-doped TiO2 Sheets with Exposed {001} Facets. J. Mater. Chem. 2011, 21, 1049−1057. (26) Yu, J. G.; Su, Y. R.; Cheng, B. Template-Free Fabrication and Enhanced Photocatalytic Activity of Hierarchical Macro-/Mesoporous Titania. Adv. Funct. Mater. 2007, 17, 1984−1990. (27) Seifried, S.; Winterer, M.; Hahn, H. Nanocrystalline Titania Films and Particles by Chemical Vapor Synthesis. Chem. Vap. Deposition 2000, 6, 239−244. (28) Wang, C.-C.; Ying, J. Y. Sol-Gel Synthesis and Hydrothermal Processing of Anatase and Rutile Titania Nanocrystals. Chem. Mater. 1999, 11 (11), 3113−3120. (29) Pradhan, S. K.; Reucroft, P. J.; Yang, F.; Dozier, A. Growth of TiO2 Nanorods by Metalorganic Chemical Vapor Deposition. J. Cryst. Growth 2003, 256, 83−88. (30) Tian, Z. R.; Voigt, J. A.; Liu, J.; Mckenzie, B.; Xu, H. Large Oriented Arrays and Continuous Films of TiO2-Based Nanotubes. J. Am. Chem. Soc. 2003, 125 (41), 12384−12385. (31) Chanda, A.; Rout, K.; Vasundhara, M.; Joshi, S. R.; Singh, J. Structural and Magnetic Study of Undoped and Cobalt Doped TiO2 Nanoparticles. RSC Adv. 2018, 8, 10939−10947. (32) Zhang, H.; Finnegan, M.; Banfield, J. F. Preparing Single-Phase Nanocrystalline Anatase from Amorphous Titania with Particle Sizes Tailored by Temperature. Nano Lett. 2001, 1 (2), 81−85. (33) Joshi, S. R.; Padmanabhan, B.; Chanda, A.; Malik, V. K.; Mishra, N. C.; Kanjilal, D.; Varma, S. Effect of Cobalt Implantation on Structural and Optical Properties of Rutile TiO2 (110). Appl. Phys. A: Mater. Sci. Process. 2016, 122, 713. (34) Wolf, S. A.; Awschalom, D. D.; Buhrman, R. A.; Daughton, J. M.; Von Molnár, S.; Roukes, M. L.; Chtchelkanova, A. Y.; Treger, D. M. Spintronics: A Spin-Based Electronics Vision for the Future. Science 2001, 294, 1488−1495. (35) Ohno, H. Making Nonmagnetic Semiconductors Ferromagnetic. Science 1998, 281, 951−956. (36) Matsumoto, Y.; Murakami, M.; Shono, T.; Hasegawa, T.; Fukumra, T.; Kawasaki, M.; Ahmet, P.; Chikyow, T.; Koshihara, S.; Koinuma, H. Room-Temperature Ferromagnetism in Transparent Transition Metal-Doped Titanium Dioxide. Science 2001, 291, 854− 856. (37) Dietl, T.; Ohno, H.; Matsukura, F.; Cibert, J.; Ferrand, D. Zener Model Description of Ferromagnetism in Zinc-Blende Magnetic Semiconductors. Science 2000, 287, 1019−1022. (38) Parras, M.; Varela, Á .; Cortés-Gil, R.; Boulahya, K.; Hernando, A.; González-Calbet, J. M. Room-Temperature Ferromagnetism in Reduced Rutile TiO2−δ Nanoparticles. J. Phys. Chem. Lett. 2013, 4 (13), 2171−2176. (39) Zheng, J.-Y.; Bao, S.-H.; Lv, Y.-H.; Jin, P. Activation and Enhancement of Room-Temperature Ferromagnetism in Cu-Doped Anatase TiO2 Films by Bound Magnetic Polaron and Oxygen Defects. ACS Appl. Mater. Interfaces 2014, 6 (24), 22243−22249.

Scientific and Innovative Research and CSIR for granting the fellowship. We also thank the Department of Science and Technology sponsored project no. GAP 232339 for partially supporting this work.

■

REFERENCES

(1) Chen, H.; Nanayakkara, C. E.; Grassian, V. H. Titanium Dioxide Photocatalysis in Atmospheric Chemistry. Chem. Rev. 2012, 112 (11), 5919−5948. (2) Hashimoto, K.; Irie, H.; Fujishima, A. TiO2 Photocatalysis: A Historical Overview and Future Prospects. Jpn. J. Appl. Phys. 2005, 44, 8269. (3) Pelaez, M.; Nolan, N. T.; Pillai, S. C.; Seery, M. K.; Falaras, P.; Kontos, A. G.; Dunlop, P. S. M.; Hamilton, J. W. J.; Byrne, J. A.; O’Shea, K.; et al. A Review on the Visible Light Active Titanium Dioxide Photocatalysts for Environmental Applications. Appl. Catal., B 2012, 125, 331−349. (4) Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconducting Electrode. Nature 1972, 238, 37−38. (5) Roy, N.; Sohn, Y.; Pradhan, D. Synergy of Low-Energy {101} and High-Energy {001} TiO2 Crystal Facets for Enhanced Photocatalysis. ACS Nano 2013, 7, 2532−2540. (6) Reyes-Coronado, D.; Rodríguez-Gattorno, G.; EspinosaPesqueira, M. E.; Cab, C.; de Coss, R.; Oskam, G. Phase-Pure TiO2 Nanoparticles: Anatase, Brookite and Rutile. Nanotechnology 2008, 19, 145605. (7) Medina-Valtierra, J.; Frausto-Reyes, C.; Ramírez-Ortiz, J.; Camarillo-Martínez, G. Self-Cleaning Test of Doped TiO2-Coated Glass Plates under Solar Exposure. Ind. Eng. Chem. Res. 2009, 48 (2), 598−606. (8) Anpo, M.; Takeuchi, M. The Design and Development of Highly Reactive Titanium Oxide Photocatalysts Operating under Visible Light Irradiation. J. Catal. 2003, 216, 505−516. (9) Song, M. Y.; Kim, D. K.; Ihn, K. J.; Jo, S. M.; Kim, D. Y. Electrospun TiO2 Electrodes for Dye-Sensitized Solar Cells. Nanotechnology 2004, 15, 1861−1865. (10) Qiu, J.; Zhang, S.; Zhao, H. Recent Applications of TiO2 Nanomaterials in Chemical Sensing in Aqueous Media. Sens. Actuators, B 2011, 160, 875−890. (11) Guan, K. Relationship between Photocatalytic Activity, Hydrophilicity and Self-Cleaning Effect of TiO2/SiO2 Films. Surf. Coat. Technol. 2005, 191, 155−160. (12) Linsebigler, A. L.; Lu, G.; Yates, J. T. Photocatalysis on TiO2 Surfaces: Principles, Mechanisms, and Selected Results. Chem. Rev. 1995, 95, 735−758. (13) Wang, R.; Hashimoto, K.; Fujishima, A.; Chikuni, M.; Kojima, E.; Kitamura, A.; Shimohigoshi, M.; Watanabe, T. Light-Induced Amphiphilic Surfaces. Nature 1997, 388, 431−432. (14) Zhang, S.; Liu, C.; Liu, X.; Zhang, H.; Liu, P.; Zhang, S.; Peng, F.; Zhao, H. Nanocrystal Cu2O-Loaded TiO2 Nanotube Array Films as High-Performance Visible-Light Bactericidal Photocatalyst. Appl. Microbiol. Biotechnol. 2012, 96, 1201−1207. (15) Lim, S. H.; Luo, J.; Zhong, Z.; Ji, W.; Lin, J. RoomTemperature Hydrogen Uptake by TiO2 Nanotubes. Inorg. Chem. 2005, 44 (12), 4124−4126. (16) Shinde, P. S.; Patil, P. S.; Bhosale, P. N.; Brüger, A.; Nauer, G.; Neumann-Spallart, M.; Bhosale, C. H. UVA and Solar Light Assisted Photoelectrocatalytic Degradation of AO7 Dye in Water using Spray Deposited TiO2 Thin Films. Appl. Catal., B 2009, 89, 288−294. (17) Perron, H.; Vandenborre, J.; Domain, C.; Drot, R.; Roques, J.; Simoni, E.; Ehrhardt, J.-J.; Catalette, H. Combined Investigation of Water Sorption on TiO2 Rutile (1 1 0) Single Crystal Face: XPS vs. Periodic DFT. Surf. Sci. 2007, 601, 518−527. (18) Akpan, U. G.; Hameed, B. H. The Advancements in Sol−Gel Method of Doped-TiO2 Photocatalysts. Appl. Catal., A 2010, 375, 1− 11. (19) Hyun Kim, D.; Sub Lee, K.; Kim, Y.-S.; Chung, Y.-C.; Kim, S.-J. Photocatalytic Activity of Ni 8 wt%-Doped TiO2 Photocatalyst L


Article


and Cobalt-Doped TiO2 Nanoparticles via Sol−Gel Technique. Appl. Nanosci. 2015, 5, 449−456. (61) Solomon, E. I.; Lever, A. B. P. Inorganic Electronic Structure and Spectroscopy; John Wiley and Sons, New York, 1999. (62) Justicia, I.; Ordejón, P.; Canto, G.; Mozos, J. L.; Fraxedas, J.; Battiston, G. A.; Gerbasi, R.; Figueras, A. Designed Self-Doped Titanium Oxide Thin Films for Efficient Visible-Light Photocatalysis. Adv. Mater. 2002, 14, 1399−1402. (63) Zuo, F.; Wang, L.; Wu, T.; Zhang, Z.; Borchardt, D.; Feng, P. Self-Doped Ti3+ Enhanced Photocatalyst for Hydrogen Production under Visible Light. J. Am. Chem. Soc. 2010, 132 (34), 11856−11857. (64) Wang, J.; Wang, Z.; Huang, B.; Ma, Y.; Liu, Y.; Qin, X.; Zhang, X.; Dai, Y. Oxygen Vacancy Induced Band-Gap Narrowing and Enhanced Visible Light Photocatalytic Activity of ZnO. ACS Appl. Mater. Interfaces 2012, 4 (8), 4024−4030. (65) Wu, W.-Y.; Chang, Y.-M.; Ting, J.-M. Room-Temperature Synthesis of Single-Crystalline Anatase TiO2 Nanowires. Cryst. Growth Des. 2010, 10 (4), 1646−1651. (66) Sanjinés, R.; Tang, H.; Berger, H.; Gozzo, F.; Margaritondo, G.; Lévy, F. Electronic Structure of Anatase TiO2 Oxide. J. Appl. Phys. 1994, 75, 2945. (67) Tseng, L.-T.; Luo, X.; Tan, T. T.; Li, S.; Yi, J. Doping Concentration Dependence of Microstructure and Magnetic Behaviours in Co-Doped TiO2 Nanorods. Nanoscale Res. Lett. 2014, 9, 673. (68) Zhuang, J.; Weng, S.; Dai, W.; Liu, P.; Liu, Q. Effects of Interface Defects on Charge Transfer and Photoinduced Properties of TiO2 Bilayer Films. J. Phys. Chem. C 2012, 116 (48), 25354−25361. (69) Schild, D. X-ray Photoelectron Spectroscopy, In: Léon, A. (eds) Hydrogen Technology. Green Energy and Technology. Springer, Berlin, Heidelberg, 2008. (70) Pandey, S. K.; Choudhary, R. J. Effect of Non-Magnetic Impurities on the Magnetic States of Anatase TiO2. J. Phys.: Condens. Matter 2011, 23, 276005. (71) Santara, B.; Giri, P. K.; Imakita, K.; Fujii, M. Evidence of Oxygen Vacancy Induced Room Temperature Ferromagnetism in Solvothermally Synthesized Undoped TiO2 Nanoribbons. Nanoscale 2013, 5, 5476−5488. (72) Coey, J. M. D.; Venkatesan, M.; Fitzgerald, C. B. Donor Impurity Band Exchange in Dilute Ferromagnetic Oxides. Nat. Mater. 2005, 4, 173−179. (73) Chiorescu, C.; Cohn, J. L.; Neumeier, J. J. Impurity Conduction and Magnetic Polarons in Antiferromagnetic Oxides. Phys. Rev. B: Condens. Matter Mater. Phys. 2007, 76, 020404. (74) Madhu, G.; Maniammal, K.; Biju, V. Defect Induced Ferromagnetic Interaction in Nanostructured Nickel Oxide with Core−Shell Magnetic Structure: The Role of Ni2+ and O2− Vacancies. Phys. Chem. Chem. Phys. 2016, 18 (17), 12135−12148. (75) Wang, Z.; Wang, W.; Tang, J.; Tung, L. D.; Spinu, L.; Zhou, W. Extraordinary Hall Effect and Ferromagnetism in Fe-Doped Reduced Rutile. Appl. Phys. Lett. 2003, 83, 518−520. (76) Chen, J.; Rulis, P.; Ouyang, L.; Satpathy, S.; Ching, W. Y. Vacancy-Enhanced Ferromagnetism in Fe-Doped Rutile TiO2. Phys. Rev. B: Condens. Matter Mater. Phys. 2006, 74, 235207. (77) Griffin Roberts, K.; Varela, M.; Rashkeev, S.; Pantelides, S. T.; Pennycook, S. J.; Krishnan, K. M. Defect-Mediated Ferromagnetism in Insulating Co-Doped Anatase TiO2 Thin Films. Phys. Rev. B: Condens. Matter Mater. Phys. 2008, 78, 014409. (78) Dodge, K.; Chess, J.; Eixenberger, J.; Alanko, G.; Hanna, C. B.; Punnoose, A. Role of Oxygen Defects on the Magnetic Properties of Ultra-Small Sn1‑xFexO2 Nanoparticles. J. Appl. Phys. 2013, 113, 17B504.

(40) Wang, S.; Pan, L.; Song, J.-J.; Mi, W.; Zou, J.-J.; Wang, L.; Zhang, X. Titanium-Defected Undoped Anatase TiO2 with p-Type Conductivity, Room-Temperature Ferromagnetism, and Remarkable Photocatalytic Performance. J. Am. Chem. Soc. 2015, 137 (8), 2975− 2983. (41) Karthik, K.; Pandian, S. K.; Kumar, K. S.; Jaya, N. V. Influence of Dopant Level on Structural, Optical and Magnetic Properties of Co-doped Anatase TiO2 Nanoparticles. Appl. Surf. Sci. 2010, 256, 4757−4760. (42) Hong, N. H.; Sakai, J.; Poirot, N.; Brizé, V. Room-Temperature Ferromagnetism Observed in Undoped Semiconducting and Insulating Oxide Thin Films. Phys. Rev. B: Condens. Matter Mater. Phys. 2006, 73, 132404. (43) Yoon, S. D.; Chen, Y.; Yang, A.; Goodrich, T. L.; Zuo, X.; Arena, D. A.; Ziemer, K.; Vittoria, C.; Harris, V. G. Oxygen-DefectInduced Magnetism to 880 K in Semiconducting Anatase TiO2−δ Films. J. Phys.: Condens. Matter 2006, 18, L355. (44) Kim, D.; Hong, J.; Park, Y. R.; Kim, K. J. The Origin of Oxygen Vacancy Induced Ferromagnetism in Undoped TiO2. J. Phys.: Condens. Matter 2009, 21, 195405. (45) Kim, J.-Y.; Park, J.-H.; Park, B.-G.; Noh, H.-J.; Oh, S.-J.; Yang, J. S.; Kim, D.-H.; Bu, S. D.; Noh, T.-W.; Lin, H.-J.; et al. Ferromagnetism Induced by Clustered Co in Co-Doped Anatase TiO2 Thin Films. Phys. Rev. Lett. 2003, 90, 017401. (46) Xu, J.; Shi, S.; Li, L.; Zhang, X.; Wang, Y.; Chen, X.; Wang, J.; Lv, L.; Zhang, F.; Zhong, W. Structural, Optical, and Ferromagnetic Properties of Co-Doped TiO2 Films Annealed in Vacuum. J. Appl. Phys. 2010, 107, 053910. (47) Venkatesan, M.; Fitzgerald, C. B.; Coey, J. M. D. Thin Films: Unexpected Magnetism in a Dielectric Oxide. Nature 2004, 430, 630. (48) Qin, L. Z.; Liang, H.; Liao, B.; Liu, A. D.; Wu, X. Y.; Sun, J. Photocatalytic Performance of Fe-, Ni-, or Cu-ion Implanted TiO2 Films under UV light, Visible light and Sunlight Irradiation. Nucl. Instrum. Methods Phys. Res., Sect. B 2013, 307, 385−390. (49) Yildirim, O.; Cornelius, S.; Butterling, M.; Anwand, W.; Wagner, A.; Smekhova, A.; Fiedler, J.; Böttger, R.; Bähtz, C.; Potzger, K. From a Non-Magnet to a Ferromagnet: Mn+ Implantation into Different TiO2 Structures. Appl. Phys. Lett. 2015, 107, 242405. (50) Das, K.; Sharma, S. N.; Kumar, M.; De, S. K. Morphology Dependent Luminescence Properties of Co Doped TiO2 Nanostructures. J. Phys. Chem. C 2009, 113 (33), 14783−14792. (51) Santara, B.; Pal, B.; Giri, P. K. Signature of Strong Ferromagnetism and Optical Properties of Co Doped TiO 2 Nanoparticles. J. Appl. Phys. 2011, 110, 114322. (52) Vegard, L. The Constitution of the Mixed Crystals and Filling of the Space of the Atoms. Eur. Phys. J. A 1921, 5, 17−26. (53) Ohsaka, T. Temperature Dependence of the Raman Spectrum in Anatase TiO2. J. Phys. Soc. Jpn. 1980, 48, 1661−1668. (54) Bersani, D.; Lottici, P. P.; Ding, X.-Z. Phonon Confinement Effects in the Raman Scattering by TiO2 Nanocrystals. Appl. Phys. Lett. 1998, 72, 73. (55) Zhang, W. F.; He, Y. L.; Zhang, M. S.; Yin, Z.; Chen, Q. Raman Scattering Study on Anatase TiO2 Nanocrystals. J. Phys. D: Appl. Phys. 2000, 33, 912. (56) Choi, H. C.; Jung, Y. M.; Kim, S. B. Size Effects in the Raman Spectra of TiO2 Nanoparticles. Vib. Spectrosc. 2005, 37, 33−38. (57) Liang, L.-H.; Shen, C.-M.; Chen, X.-P.; Liu, W.-M.; Gao, H.-J. The Size-Dependent Phonon Frequency of Semiconductor Nanocrystals. J. Phys.: Condens. Matter 2004, 16, 267. (58) Hamadanian, M.; Reisi-Vanani, A.; Majedi, A. Sol-Gel Preparation and Characterization of Co/TiO2 Nanoparticles: Application to the Degradation of Methyl Orange. J. Iran. Chem. Soc. 2010, 7, S52−S58. (59) Liu, X.-H.; He, X.-B.; Fu, Y.-B. Effects of Doping Cobalt on the Structures and Performances of TiO2 Photocatalyst. Acta Chim. Sin. 2008, 66, 1725−1730. (60) Mugundan, S.; Rajamannan, B.; Viruthagiri, G.; Shanmugam, N.; Gobi, R.; Praveen, P. Synthesis and Characterization of Undoped M


Observation of Optical Band-Gap Narrowing and Enhanced Magnetic

Recommend Documents