Experimental and Theoretical Investigations on the ... - ACS Publications

Page 1 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Experimental and Theoretical Investigations on the Activity and Stability of Substitutional and Interstitial Boron in TiO2 Photocatalyst N. Patel*,a,b, Alpa Dashoraa, R. Jaiswala, R. Fernandesa, M. Yadava, D.C. Kotharia, B. L. Ahujac, A. Miotellob

a

Department of Physics and National Centre for Nanosciences & Nanotechnology,

University of Mumbai, Vidyanagari, Santacruz (E), Mumbai 400 098, India b

Dipartimento di Fisica, Università degli Studi di Trento, I-38123 Povo ( Trento), Italy

c

Department of Physics, M.L. Sukhadia University, Udaipur 313001, India

* Corresponding author: Nainesh Patel, Tel. No.:

+39-0461882012

Fax No.:

+39-0461881696

e-mail address: [email protected]

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: Effects of boron doped in TiO2 at a) interstitial site (Bint); b) substitutional site (Bsub) and combination of both the sites (Bint+sub), have been investigated experimentally and theoretically to understand the origin of enhanced photocatalytic activity and stability. Bdoped TiO2 powder was synthesized by sol-gel method with different concentrations of boron. XPS results indicate that boron first prefers Bint site when doped with low concentration (upto 1 at.% B), but as the concentration increases (2 at.% and above) B also occupies substitutional O position in addition to Bint to form TiO2 containing Bint+sub (TiO2Bint+sub). Higher absorption of visible light is achieved for TiO2-Bint+sub due to the presence of two absorption edges (2.4 and 2.2 eV) as observed in the absorption spectra, while insignificant narrowing of band gap is observed for TiO2-Bint. Electronic structure calculated by DFT for TiO2 with Bint, Bsub, and Bint+sub revealed that the two localized deep levels are formed in the mid gap which are responsible for these optical transitions for TiO2-Bint+sub. Photoluminescence (PL) emission spectra showed that the shallow level (as inferred from the DFT calculations) created below the conduction band is able to decrease the radiative recombination process in TiO2-Bint by trapping electrons and prolonging the lifetime of charge carriers as observed in the time resolved PL decay curve. Furthermore, lower effective mass ratio of charge carriers calculated using DFT for TiO2-Bint also suggests better charge mobility and low recombination rate. Photocatalytic degradation rate of organic pollutant in water was significantly higher after B-doping with higher performance obtained with TiO2 containing Bint as compared to Bint+sub. By imposing the destabilizing circumstances it was established that TiO2-Bsub is metastable and collapses under mild conditions, whereas TiO2Bint is highly stable and retains all its properties. All these unprecedented findings disclose that higher activity of TiO2-Bint as compared to that of TiO2-Bint+sub is mainly because of the delayed recombination processes even though the optical band gap is not significantly varied.


Page 2 of 35

Page 3 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION: Titanium dioxide (TiO2) remains the best candidate for the solar-driven photocatalysis reactions owing to its intrinsic properties such as chemical inertness, long-term stability, high resistance against corrosion, nontoxicity and low cost [1]. It also has possible usage in large number of applications in environmental purification, energy production, and self sterilization [1]. However, inability to utilize the visible part of the solar spectrum due to its large band gap, and recombination of photogenerated charge carriers are the two main limitations associated with TiO2 for its practical applications. Doping with foreign elements with appropriate concentration can address both these issues by forming defect levels within the band gap. These levels not only cause intraband transitions for absorption of higher number of photons but also act as the trapping sites for photoexcited charges for its prolonged lifetime. Metal ion doping significantly improves the light absorption ability by forming deep level in the band gap but at the same time introduces recombination centers which affects the charge transfer towards the surface for photocatalytic reaction. On the other hand, by introducing shallow levels, non-metal or anion dopants do not cause major narrowing of the band gap, yet these impurity states participate in trapping the charges to cause improvement in the photocatalytic activity. Among the various anions (N [2], C [3], S [4], F [5], B [6]) doped in TiO2, boron is very efficient in band gap narrowing and in decreasing the recombination of charges. However, it is not well understood that which of these factors is more effective for photocatalytic activity. Chen et al. [6] observed blue shift in the band-edge of TiO2 after boron doping in interstitial site due to quantization effects which improves the photocatalytic activity by intense absorption in the UV range. It was also tentatively suggested that boron in interstitial site reduces Ti4+ to Ti3+ to facilitate separation of photoexcited electron and hole pairs and slow down their recombination [6]. In et al. [7] were able to decrease the band gap



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

by substitutionally replacing oxygen with boron in the TiO2 lattice to enhance the visible light photoactivity. The electronic variation upon B-doping was theoretically studied by Yang et al. [8] and Geng et al. [9]. Using density function theory (DFT) calculations, they showed that boron in the substitutional site of oxygen can create mid-gap states and eventually causes the red shift in the absorption edge, while interstitial occupancy of B does not cause any significant variation in the band gap. By similar type of calculations Finazzi and co-worker [10] suggested that substitutional B is metastable species and decomposes into interstitial boron or B2O3 by forming oxygen vacancy. They also predicted that interstitial boron is the preferred site in TiO2. The experimental evidence for the stability of these boron species was provided by Artiglia et al. [11] using different synthesizing techniques to produce four components; substitutional B with Ti (Ti-B), substitutional B with O (Ti-B-O), interstitial B (Ti-O-B) and B2O3. On annealing, both the substitutional sites of B disappear to form B2O3 while the interstitial B is highly stable at all temperatures. However, the effect of these sites of B on its photocatalytic activity was not studied. Various techniques were adopted by Zaleska et al. [12, 13] to dope B in TiO2 and observed that activity depends on the preparation conditions, method and type of TiO2 used, but failed to explain the role of B in improvement in the photocatalytic activity. Boron presence simultaneously at both the interstitial and substitutional locations was first reported by Xue et al. [14] and was later also confirmed by Wu et al. [15]. The optical properties studied by the latter group [15] using theoretical calculation showed the presence of mid gap states at different location than that found in TiO2 doped with substitutional boron. These states are formed by the charge balancing due to the interaction between interstitial and substitutional boron. Coexistence of both these B-species shifts the ability of H2 generation found in pure TiO2 to O2 generation in the photocatalytic water splitting [15]. Few attempts were carried out to codoped B with N [16], Ag [17], and Ni2O3 [18] to improve the visible light photocatalytic activity.


Page 4 of 35

Page 5 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Looking at the above studies it is still not clear if B in substitutional site (Bsub), creating band gap narrowing, is more catalytically active than B in interstitial site (Bint) assisting in charge separation, or if combination of both these positions provides synergic effect in increasing photoactivity. Although theoretically the variation in the band gap is well studied with respect to boron position in TiO2, experimentally reproducing these band structures has not been reported until now. Speculation about better charge separation is reported for TiO2-Bint due to Ti3+ formation but no experimental or theoretical proof has been presented for the reduced recombination rate in TiO2-Bint or TiO2-Bsub. Doping concentration playing the central role in controlling the photocatalytic activity is also under studied. In et al. [7] and Zaleska et al. [12] found discrepancy in the photocatalytic performance upon varying the boron concentration but failed to explain the physical reasoning behind it. Although metastable nature of TiO2-Bsub is known, the consequence upon destabilizing TiO2-Bsub has not been reported yet. Understanding these factors will provide pathways for tuning this material for the optimum photocatalytic activity. Thus the present work focuses on the experimental and theoretical studies of TiO2-Bint and TiO2-Bint+sub for determining the origin of enhanced photocatalytic activity and the stability of doped TiO2.

EXPERIMENTAL AND THEORETICAL SECTIONS:

Photocatalyst Preparation: Pure TiO2 and boron-doped TiO2 were synthesized by sol-gel method by using titanium butoxide [Ti(OC4H9)4] and boric acid (H3BO3) as precursors for TiO2 and B doping, respectively. During the synthesis of pure TiO2 the solution mixture of de-ionized water, ethanol and HNO3 (used as catalyst) was added dropwise to the homogeneous mixture of ethanol and titanium butoxide under constant stirring. The resultant solution was further stirred for 1 h to increase the homogeneity of the mixture. Molar ratio of



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Ti(OC4H9)4/H2O/ethanol/HNO3 was kept constant at 1/30/20/0.1. In the case of B-doped TiO2, boric acid was added to the acidic solution before adding it to the Ti precursor solution. The concentration of boron was varied from 1 to 4 at % by varying the amount of boric acid. Gelation of the final mixture was performed overnight at room temperature, followed by drying at 100 oC for 3 h to remove the excess solvent. Subsequent calcination was performed at 400 oC for 2 h for crystallization of the TiO2 powders. Characterizations: Scanning Electron Microscope (SEM-FEG, JSM 7001F, JEOL) equipped with Energy-Dispersive Spectroscopy analysis (EDS, INCA PentaFET-x3) was used to study the surface morphology and composition of all the samples. The structural characterization was performed by X-ray diffraction (XRD) using Cu Kα radiation (λ = 1.5414 Å) and by Raman spectroscopy using Renishaw micro-Raman spectrometer (RE-04) using solid state laser with the diode pumped at 514 nm. The band gap of the doped TiO2 was determined by measuring diffuse reflectance absorption spectra, using Cary 500 UV-Vis-NIR spectrophotometer, in the range of 200 nm to 800 nm. The surface composition and chemical states of each element in the samples were examined by X-ray photoelectron spectra (XPS) using a PHI 5000 VersaProbe II instrument equipped with a monochromatic Al Kα (1486.6eV) X-ray source and a hemispherical analyzer. Appropriate electrical charge compensation was employed to perform the analysis and binding energy was referenced to the C 1s peak. Photoluminescence (PL) emission spectra were collected, by exciting the photocatalyst with wavelength of 385 nm, using Fluorescence Spectrophotometer (Varian; Cary Eclipse). Time-resolved photoluminescence spectroscopy was performed using ISS Chronos BH fluorometer. For excitation, pulse diode laser (Hamamatus) of wavelength 405 nm with pulse width of 70 ps operating with a peak power of 100 mW was used. Both normal and time-resolved PL spectra were acquired by dispersing 3 mg of powder photocatalyst in aqueous medium of fixed amount and transferred into 1 cm x 1 cm cuvette for the


Page 6 of 35

Page 7 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


measurement. BET surface area of the powder photocatalysts was determined by nitrogen absorption at 77 K (Smart SORB 93) after degassing at 120 ºC for 2 hrs.

Photocatalytic Activity measurement: The photocatalytic activity of all the samples was evaluated by photodegradation of p-nitrophenol (p-NP) aqueous solution (10 ppm) and methylene blue (MB) dye (0.01mM) solution under light irradiation. For the source of light, 150 W Xenon lamp (Philips) was used having a spectrum nearly similar to solar spectrum. 20 mg of photocatalyst was dispersed in 50 ml of aqueous p-NP or MB solution for each photocatalytic degradation experiment. The powder suspension was first stirred for 30 min in dark to attain adsorption-desorption equilibrium of the molecules in the solution. The distance between the beaker and the light source was kept constant at 10 cm. For a given time intervals, 1 ml of p-NP or MB aqueous solution was filtered out from the reactor vessel and transferred to the 1 cm x 1 cm quartz cuvette. The photocatalytic activity was determined by measuring the normalized intensity of the absorption band of p-NP at 320 nm and of MB at 665 nm using spectrophotometer, and plotting it as a function of irradiation time. All the photocatalysis experiments were performed at room temperature and the pH of the solution was maintained neutral during all the photocatalytic measurements.

Theoretical Calculations: To understand the effects of B doping in TiO2, we have performed the first principles calculations. For this purpose we have used a full-potential linearized augmented plane wave (FP-LAPW) method within the DFT as implemented in WIEN2k code [19]. In the FP-LAPW prescription, wave functions, charge density, and potential are expanded in spherical harmonics within non-overlapping muffin-tin (MT) spheres, while plane waves are used in the remaining interstitial region of the unit cell. Core and valence states are, respectively,



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

treated

within a

multiconfiguration

relativistic

Dirac–Fock

Page 8 of 35

approach and

scalar

relativistically. Different from the previous theoretical studies [8, 9, 10], the exchange and correlation of PBE-sol given by Perdew el al. [20] within the generalized gradient approximation with modified Backe-Johnson (mBJ) [21] correction were used for the present computations. For pure anatase TiO2, lattice parameters were considered as a=3.838 and c=9.642 Å as optimized by Rubio-Ponce et al. [22] using FP-LAPW scheme. As suggested in existing literatures [8, 9, 10], that B atom in the TiO2 lattice occupies various positions. In the present study, we have considered B atom at interstitial site and substitution at O site henceforth referred as Bint and Bsub respectively. There are various possible sites for interstitial B as suggested by Finazzi et al. [10]. Authors [10] have considered three different interstitial sites and finally concluded that the electronic structure of all the interstitial B-doped TiO2 samples changes in the same way and introduces the same kind of defects. Therefore, we have considered a single interstitial doping of B in TiO2. For this we have constructed a 2x2x2 supercell of TiO2 which generates 48 atoms in the supercell. We have added 1 interstitial and 1 substitutional B (substituted at O site) in the supercell. With increasing B concentration, we have also studied the effect of both interstitial and substitutional doping of B simultaneously in the TiO2 lattice, referred as Bint+sub. For the present computations, radii for MT spheres were selected as 1.92 a.u. for Ti, 1.74 a.u. for O and 1.62 a.u. for B atom. The k-point sampling of the Brillouin zone (BZ) was set 14x14x14 and 5x5x5 for pure TiO2 and other studied compounds, respectively. Other computational parameters were set as RMTKmax=7, lmax=10 and Gmax=12 [16]. Finally all the structures were fully relaxed for atomic forces until the forces on each atom reach a value less than 5 mRy per a.u. To ensure sufficient accuracy in convergence, total energy of all the studied compounds was converged to 0.01 mRy.


Page 9 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RESULTS AND DISCUSSION: XRD (Fig 1a) and Raman spectroscopy (Fig 1b) were used to study the structural modification induced both at surface and in the bulk of TiO2 after boron doping respectively. All the major peaks assigned to anatase phase are clearly visible in pure TiO2. In addition to anatase phase, the peaks due to rutile (27.4o, 36o, 41.1o, 56.5o) and brookite (30.8o) are also observed after boron doping in TiO2. The sharpness and intensity of peaks due to rutile increases with the doping concentration, while that of brookite remains unchanged. Most importantly the main diffraction peak of anatase at 25.3o shifts to the lower angle (inset of 1a) indicating the expansion of TiO2 lattice by introduction of B. In the case of boron substitutionally replacing O, the unit cell volume is expected to decrease since the radius of B3+ (0.023nm) is smaller than O2- (0.132nm) while the occupancy of the interstitial position will induce the lattice expansion to shift the diffraction to the lower angle. This result shows that boron is dominantly present in the interstitial sites rather than in substitutional position. However, Bsub cannot be neglected because of the existence of rutile phase formed by rearrangement of atoms caused by oxygen deficiency created by the occupancy of boron atom through oxygen substitution. In agreement with XRD, Raman spectra (Fig. 1b) also show the mode of vibrations of only anatase phase for pure TiO2. On the other hand, B-doped TiO2 shows the mixture of anatase, rutile and brookite phase with major contribution from anatase phase. No signal due to boron or boron oxide (B2O3) is detected in the XRD pattern and Raman spectra. The average crystal size measured through the width of XRD peaks is in the range of 6-9 nm for all the samples. This value is consistent with the particle size measured using SEM images (Fig. S1 of supporting information). However the nanoparticles are in agglomerated state with lower surface area than that expected from the nanoparticles



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

below 10 nm size. No major variation is detected after B-doping in BET surface area (Table S1) which is in the range of 50-80 m2/g for all samples. UV-Vis absorption spectra were measured for pure and B-doped TiO2 in diffuse reflectance mode and are reported in Fig 2a. Irrespective of boron concentration the absorption edge is red shifted for B-doped TiO2 as compared to pure anatase TiO2. The shift is small in the case of 1 at% B-doped TiO2 while the maximum narrowing of the band gap is obtained with 2 at% B. Further increase in concentration (3 at% and 4 at%) causes the absorption edge to shift to lower wavelength. Tauc plot (Fig. 2b) was used to determine the band gap of these samples (Table S1). At low concentration of B-doping (1 at%), a minor decrease in band gap is observed, reducing from 3.15 eV for pure TiO2 to 3.0 eV. However, in the case of 2 at% B-doping the presence of two absorption edges (2.4 and 2.2 eV) is clearly evident from Tauc plot thus suggesting a possible existence of localized state deep in the band gap causing two step transition. At higher concentrations (3 at% and 4 at%) similar characteristic of the absorption edge is observed but with higher energy transition. These new energetic features in the band gap at higher B concentration are mainly due to the possibility of different B doping sites (Bint or Bsub or Bint+sub) in TiO2 and the interaction of B atoms with the host elements in the TiO2 lattice which will be discussed in the next section. XPS was used to investigate the chemical state of boron in TiO2 (Fig. 3). For both pure TiO2 and 1 at% B-doped TiO2, two peaks centered at 458.4 eV and 464.1 eV, assigned to Ti 2p3/2 and Ti 2p1/2 of Ti4+ states, are observed in Ti 2p core level. Similar peaks were detected for TiO2 doped with higher boron concentrations (2 at%, 3 at% and 4 at%) but with positive shift in binding energy by 0.2 eV as compared to those of pure TiO2. This signifies that boron is present in the TiO2 lattice, in agreement with the XRD data, to influence the local chemical states of Ti4+ ion. The signal of Ti3+ was detected on the surface only for the 1 at% B-doped TiO2 which is in the form of shoulder at around 456.8 eV. In oxygen 1s core


Page 10 of 35

Page 11 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


level, a single peak centered at 529.8eV is displayed in the spectra of all the samples. However in B-doped TiO2, after deconvolution, an extra peak at 530.7 eV attributed to B-O bond is present in addition to the major peak at 529.8 eV related to oxygen bonded in TiO2. For 1 at% B-doped TiO2, broad signal in B 1s level was deconvoluted into two peaks with binding energy of 191.7 eV and 193.0 eV. The former peak with major contribution (95%) in overall signal is attributed to the interstitial boron while the latter peak is due to the presence of B-O bond of the boron oxide (B2O3). Besides these two peaks, a small peak centered at 189.0 eV assigned to substitutional B at oxygen site was clearly observed for B-doping above 1 at%. The contribution due to Bsub is about 20% in overall boron signal with major signal recorded from Bint (above 75%) and rest (about 5%) from B2O3. By increasing the concentration of boron content an increase of the signal due to B2O3 is observed while no change in the intensity ratio of Bint/Bsub was detected. This XPS result shows that boron is mainly present at interstitial site forming B-O-Ti bond when doped with low concentrations (1 at%). In this chemical bond, boron atom reduces Ti4+ in the lattice to form B3+ and 3Ti3+ [10] thus explaining the presence of Ti3+ species in 1 at% B-doped TiO2. Such ionization is compensated by forming bonds between B and neighboring O to obtain B-O-Ti species. As concentration increases to 2 at%, boron is not only present at the interstitial site but also occupies substitutional site by replacing oxygen atom in the lattice forming B-Ti-O bonds. At higher concentrations, after saturating the interstitial and substitutional sites, the extra boron might contribute in forming B2O3 species. Further evidence of boron present at substitutional site along with the interstitial site (for B concentration 2 at% and above) is provided by the positive shift of Ti4+ peak. Boron is more electronegative than Ti atom, so the presence of boron in the substitutional site will cause electron transfer from Ti to B to create positive shift of Ti2p binding energy. In addition, the separation between the B 1s peak of Bsub and Bint is



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

about 2.7 eV which is in good agreement with that reported by the calculations (2.6 eV) performed by Finazzi et al. [10]. To further understand the effect of B doping at various sites (Bint or Bsub or Bint+sub) on the electronic structure of TiO2, theoretical simulations were performed using DFT with mBJ potential. Boron replacing Ti is energetically less favorable and not stable thus the calculations were performed a) by substitutionally replacing one O atom with B atom (Bsub), b) by placing boron in the interstitial position (Bint) and c) by introducing boron as both substitutional as well as interstitial sites (Bint+sub). The selected interstitial site is the most stable site according to the calculations performed by Finazzi et al. [10] and the electronic properties of B-doped TiO2 at different interstitial sites remains unchanged. In Fig. 4 (a-d), we have presented the total density of states (TDOS) [left panel] and crystal structure [right panel] of TiO2, TiO2-Bint, TiO2-Bsub and TiO2- Bint+sub. In Fig. 4 (a), zero energy represents the Fermi level (EF) for pure TiO2. Band gap of 2.9 eV was observed for pure TiO2 (Fig. 4a) which is within 10% of the experimental band gap (3.2 eV) (DFT with mBJ potential produces results within 10% of experimental value). In Figs. 4 (b-d), the zero energy has been shifted to the conduction band minima (CBM) of TiO2, and EF is marked by dotted line. Fig. 4b presents TDOS and crystal structure (right panel in the Fig) of Bint located in the center of anatase cell where B forms bond with surrounding Ti and O atoms [3]. The present theoretical result for Bint is in good agreement with the earlier reported data [23]. The impurity bands are formed below the conduction band (CB) of TiO2 and reduce the band gap of TiO2 with EF shifted towards the CB. In TDOS TiO2-Bsub (Fig. 4c), B-atom forms impurity states around 1 eV above the valance band maxima (VBM) in the band gap of TiO2 and again the EF is shifted towards the CB. Fig. 4d shows the TDOS and crystal structure of Bint+sub in which the concentration of B is double than that in Bint and Bsub. In crystal structure (right panel), the B interstitial and substitutional sites are similar as considered in the case of


Page 12 of 35

Page 13 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Bint and Bsub, respectively. The localized impurity states of Bint+sub are formed below the CBM and these states are not identical to the impurity states in Bint and Bsub. EF is shifted towards CB even in this case. A minor change in the band edges of TiO2 is observed on B doping while impurity states of B are formed in the band gap of TiO2. Figs. 5 (a-c) present the energy bands (left panel) and PDOS (right panel) of Bint, Bsub and Bint+sub with EF at 0.0 eV. Impurity states by Bint doping (Fig. 5a) are formed just below the CBM to reduce the band gap to 2.7 eV from 2.9 eV as seen in pure TiO2. This value of band gap reconciles well with UV-vis spectra of 1 % B-doping where B in the interstitial position, as confirmed by XPS, is able to reduce the gap by ~0.15 eV. This shallow level (Fig. 5a) is formed by strong hybridization of Ti, O and B states as observed from the overlapping of the Ti, O and B orbitals (in anti-bonding state) in PDOS which is also seen as B-O-Ti bond in the XPS results. In Bsub (Fig. 5b), a localized state is formed deep in the band gap located at 0.96 eV (1.86 eV) above (below) the VB (CB) of TiO2. Unlike Bint, solely Bsub orbital is responsible for the formation of this intermediate level. Small contribution of B states is also found around EF and in CB region. In Fig. 5c, where B atoms are placed at both interstitial and substitutional sites, the impurity states generated by B in the mid gap are located at different positions than those created by Bint or Bsub. Two localized impurity states separated by 0.33 eV at Γ position of BZ (mainly due to Bsub) are observed in the band gap placed at about 2.24 eV and 2.57 eV above the VBM (at Γ position of BZ). This new electronic configuration is able to create two additional optical transitions by absorbing visible light as confirmed by UV-vis spectra of 2% B-doped TiO2 in which Bint+sub is present as suggested by the XPS spectra. The overlapping of B orbitals at interstitial and substitutional sites is also observed around EF. The absence of shallow level in Bint+sub near the CBM due to Bint is mainly attributed to the interaction between the Bint and Bsub as indicated by the overlapping of B orbitals. Thus the theoretical results are in good agreement



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

with the experimental results and show that for Bint a small narrowing of the band gap is due to the formation of shallow level formed below the CBM due to B-O-Ti bond as in case of 1% B. Whereas when Bint+sub is present two extra optical transitions, occurring within the band gap by the formation of two localized state, are responsible for high visible light absorption for 2 at% B-doped TiO2. As the concentration of B increases above 2 at% the increase in formation of B2O3 is also observed which will contribute in the absorption edge of TiO2. Since the band gap of B2O3 (6.2 eV) is higher than that of TiO2, so as the concentration of B increases the absorption edge is blue shifted for B-doped TiO2 at 3 at% and 4 at%. The recombination process of photogenerated electron and hole pair is as important as the absorption of photon to determine the photocatalytic performance. Photoinduced charges can recombine by either radiative transition or by non-radiative phonon generation. To study the radiative transition, PL spectra acquired by exciting the samples with wavelength of 385 nm (matching the band edge of the TiO2) are reported in Fig. 6a for pure and doped TiO2. Three distinct emission peaks at 425, 486, and 527 nm are recorded in the visible range. First emission peak (425 nm) in the visible range arises by relaxation of self trapped exciton localized on TiO6 octahedral while the other two emission peaks (486 and 527nm) are due to the intraband transitions within the energy levels due to traps or surface defects related to oxygen vacancies. All three peaks are associated with the recombination of charges either at shallow trap level (425 nm) or deep trap levels (486 and 527nm). The intensity of these peaks are either unchanged or increased nominally by boron doping except for 1 at% B concentration. Time resolved PL spectra (Fig. 6b) are acquired to determine the lifetime of the charge carriers for pure and B-doped TiO2 (the equations and parameters used to fit the decay curve are available in the supporting information). The average lifetime, obtained by fitting the decay curve is about 1.39 ±0.07 ns for pure TiO2. In the case of B-doped TiO2, the average lifetime is increased to 1.68 ±0.08 ns for low concentration B-doping (1 at%)


Page 14 of 35

Page 15 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


showing prolonged time period of the charge carriers as compared to that of pure TiO2. At higher concentrations of B-doping, the lifetime reduces to the value below that of pure TiO2 with minimum value recorded for 2 at% B-doped TiO2 (1.17 ±0.06 ns). The reduced PL intensity with longer lifetime for 1 at% B-doped TiO2 suggests the decrease in radiative recombination of charge carriers in comparison to that for pure TiO2, while the phenomenon is reversed for higher boron concentration. The reduction in radiative recombination may result due to various factors such as non-radiative Auger type recombination, grain boundary and surface defects, mobility of carriers, doping concentration quenching effect, band structure and presence of trap centres and their locations. The first four factors depend on the dopants and their role increases with doping concentration. In the present case, since the radiative recombination is higher after boron doping, except for 1 at% B-doping, so these factors do not play any role. Due to indirect energy band in the anatase phase, the electrons from the conduction band have to traverse an indirect path for undergoing recombination with the holes thus delaying the recombination of charges.

Conversely, in rutile phase the

direct band gap allows the easy pathway for recombination processes. According to this model, in the case of 1% B-doped TiO2 the resulting mixed phase of anatase and rutile will produce higher recombination rate than pure TiO2 containing pure anatase phase. The present contradictory results with 1 at% B-doped TiO2 clearly rules out the contribution of band structure in improving the charge separation. Defect centers introduced by the dopant species can play huge role in trapping the charges depending upon the location of these centers in the band gap. For 1 at% B-doped TiO2, boron is mainly positioned at the interstitial site forming a shallow level below the CBM due to B-O-Ti bond. This shallow level containing Ti3+ seems to act as the trapping site for the photogenerated electrons to increase the lifetime and decrease the recombination process. Nevertheless, at higher concentrations the boron is present at interstitial as well as substitutional sites with latter forming the defect level deep in



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 35

the gap. Looking at the PL results, the presence of these deep levels act as the recombination centers to produce the emission peak intensity and lifetime of the boron doped at higher concentration nearly similar to that of pure TiO2. Similar findings were reported by Chaudary et al. [24] for Cu-doped TiO2, in which shallow trap centers due to oxygen defects are able to capture photo-excited electrons. These trap centers transfer the electron to the surface to enhance the photocatalytic activity.

On the contrary, the deep trap formed by Cu d-states

were not suitable for photacatalytic applications due to the increase in recombination as indicated by the increase in emission intensity. To check the charge carrier mobility and separation using first-principles calculations we have calculated the relative effective masses ratio of electron and hole. The effective mass of charge carriers is calculated by taking the coefficient of second-order term in a quadratic fit of energy band diagrams (Figs. 5 a-c, left panel) at the edges of VB, CB and intermediates bands. Lower charge carrier effective mass corresponds to a higher charge carrier mobility and large value of ratio of electron to hole effective mass

(m

* e

/ m*h ) suggests a lower

recombination rate of the electron-hole pair. The energy bands at VBM, CBM and intermediate bands at various k-points of BZ (mainly, Γ, H and K points) were fitted to calculate the effective masses of electrons and holes for all the B-doped samples and finally taking the average of the ratio of effective masses of electron to hole at the same k-point. The average values of the ratio of effective masses of electron to hole are obtained as 3.2 ±0.16, 2.1 ±0.11 and 2.7 ±0.14 for Bint, Bsub and Bint+sub, respectively. High value of m*e / m*h for Bint shows low recombination rates of the charge carriers which is in good agreement with above PL results. The effect of boron doping on the photocatalytic performance was investigated by degrading organic pollutants such as P-Nitrophenol (p-NP) and Methylene blue (MB) dye.


Page 17 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The catalytic activity for degradation was determined by examining the characteristic absorption peak of these organic pollutants as a function of time (Fig. 7). In absence of photocatalyst, negligible amount of self degradation of p-NP and MB dye was observed under the light irradiation after 4 h. In contrast to pure TiO2 the boron doped TiO2 showed significantly improved degradation rate for both the pollutants. Among the B-doped TiO2, 1 at% B-doped TiO2 showed the highest photocatalytic activity with 75% and 65% degradation achieved in 4 h for p-NP and MB dye in comparison to 20% and 25% obtained with pure TiO2, respectively. The data points in figure 7 were fitted linearly to calculate the apparent rate constant and the p-NP and MB degradation rates per gram of the photocatalyst (Table S2). Low concentration (1 at %) Boron doped TiO2 is able to degrade p-NP and MB dye with the rate 4 and 2.5 times higher than that with pure TiO2 respectively. However as the concentration of boron increases to 2 % the rate decreases but is still higher than that of pure TiO2 with values nearly constant for further increase in doping concentrations. Photogenerated holes oxidize the absorbed H2O on the surface of TiO2 to produce strong .

oxidizing agent in the form of OH radicals which participate in the degradation of organic pollutants like p-NP and MB dye. This process is mainly affected by the surface area of the photocatalyst, light absorption capability, and charge separation and transfer. The surface area of all the photocatalysts is approximately same, so it cannot be responsible for the significant difference in photocatalytic activity. The enhanced activity of 1 at% B-doped TiO2 is mainly attributed to the reduced recombination processes. The boron located in the interstitial site develops a shallow level, below CBM due to the formation of B-O-Ti bond. This level hardly affects the band gap but plays a major role in trapping the photogenerated electrons, due to the presence of Ti3+, to prolong the time for the hole to travel to the surface . and form OH radical. On the contrary, for higher doping of boron (2 at%, 3 at% and 4 at%)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 35

the increase in the degradation rate is related to the enhanced light absorption achieved by the deep levels formed within the band gap by Bsub. These deep levels do not participate in charge separation by trapping. Although at these higher concentrations interstitial boron is also present it is however not able to form shallow level due to the interaction with substitutional boron. This shows that interstitial boron is far more active than the boron present in both the sites (Bint

+ sub).

The present results highlight the importance of the

recombination of photogenerated charges over light absorption ability in photocatalytic processes. To check the relative stability of B-doped TiO2 with B at various sites, we have computed the formation energies of all the studied compounds (equations used to formulate the formation energies are reported in the supporting information). The values of formation energy per unit cell are calculated as 2.7 ±0.13, 4.7 ±0.23 and 3.2 ±0.16 eV for TiO2 containing Bint, Bsub and Bint

+ sub,

respectively. These values suggest that Bint is the highly

preferred B-doping site in TiO2, while Bsub is difficult to obtain experimentally because of the high formation energy. The formation energy value of TiO2-Bsub + int is in between that of TiO2-Bint and TiO2-Bsub thus signifying that as the doping concentration increases B will occupy substitutional sites but only after the interstitial sites are

saturated, to produce

mixture of both the sites at the same time. Accordingly to the calculated formation energy, TiO2-Bint is more stable than TiO2Bsub. Finazzi et al. [10] also showed through DFT calculations that TiO2-Bint is preferred over TiO2-Bsub and that the latter is a metastable structure. In order to verify this finding experimentally, 1 at% and 2 at% B-doped TiO2 containing Bint and Bint+sub respectively, were treated thermally and mechanically in order to destabilize the structure. Former treatment was done by calcinating the finally prepared samples for 4 h at 450 oC (the temperature selected is higher than that used for as-prepared sample) and later was achieved by ball milling for 6 h at


Page 19 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


500 rpm. No major variation in structural, optical and in chemical state of boron was observed in 1 at% B-doped TiO2 as confirmed by XRD pattern, UV-Vis and XPS spectra respectively. On the contrary, for 2 at% B-doped TiO2 the peak due to substitutional boron at 189 eV completely disappeared from the XPS spectra (Fig. 8a) after both the treatment conditions. The absorption edge was also relocated near to that of pure TiO2 indicating the disappearance of deep levels within the band gap which was responsible for higher light absorption in the untreated sample (Fig. 8b). Even visually the color of 2 at% B-doped TiO2 was transformed from dark yellow to white, like pure TiO2, after both the treatment conditions. All these results suggest that substitutional boron is not stable and collapses upon mild destabilizing conditions. As suggest by the Finazzi et. al. [10] the Bsub can convert into Bint and form oxygen vacancy. By analyzing Ti core level, the ball milled sample showed the presence of small peak at 456.5 eV attributed to Ti3+ species formed by oxygen vacancy, while this peak is not observed in the thermally treated sample. Ball milling is performed at room temperature in the closed chamber while the thermal treatment is carried out in open air at high temperature. Thus even if the oxygen vacancies are formed upon Bsub moving to Bint the oxygen from the air can refill the vacancy at high temperature which is not possible during ball milling process. Photocatalytic degradation of MB dye performed using these treated samples (Fig. S2) shows the reduction in the activity in relation to the as prepared 2 at% B-doped TiO2. Lack of visible light absorption due to disappearance of deep levels within the band gap was mainly responsible for detrimental photocatalytic performance. Shallow level, accountable for reduction in recombination of photogenerated charges in TiO2, was absent in the as-prepared 2 at% B-doped TiO2 and does not seems to form even after the conversion from Bsub to Bint upon treatment. Presently, the location and chemical state of boron when moved out of the substitutional position is not well understood. Further experimental and theoretical work is underway in order to clarify this aspect. However the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

major point is well understood, namely that the boron in the interstitial site is stable and active even though the optical band gap of the doped TiO2 is not majorly affected. On the other hand, substitutional boron does participate in the high number of photons absorption but the boron doped in TiO2 in this case is metastable and inactive towards photocatalytic reaction.

CONCLUSIONS: B-doped TiO2 powder was synthesized by sol-gel method with different B atomic concentrations. Several analyses of the samples have been performed both experimentally and theoretically, and the photocatalytic properties of the doped TiO2 were tested by measuring the degradation rate of organic pollutant in water. XPS results proved that at low concentrations (1 at%) B stays in interstitial position (Bint) , but as the concentration increases (2 at% or more) B also occupies lattice sites by substituting O position (Bsub) thus having both positions, Bint+sub. Higher absorption of visible light is achieved with Bint+sub where two absorption edges (2.4 and 2.2 eV) are observed in the absorption spectra. DFT calculations made for TiO2 containing Bint, Bsub, and Bint+sub revealed that two localized deep levels are formed in the mid energy gap of TiO2 which explain the optical features of TiO2-Bint+sub. In the case of TiO2-Bint, a shallow energy level is formed below the conduction band minimum that is consistent with the measured photoluminescence emission spectra. This new level is able to decrease the efficiency of the radiative recombination process in TiO2 by trapping electrons and thus increasing the lifetime of charge carriers as observed in the time resolved PL decay curve. In addition, the experimentally evaluated lifetime of


Page 20 of 35

Page 21 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


charge carriers seems consistent with the calculated ratio of the effective masses of electrons and holes. The relative stability of the B-doped TiO2, with B at various lattice sites, was established by calculating the formation energies of all the studied compounds. The result indicates that TiO2-Bint is the more stable configuration as we indeed experimentally confirmed by thermal and mechanical destabilizing processes of the doped samples. Among the B-doped TiO2, 1 at% B-doped TiO2 showed the highest photocatalytic activity with 75% and 65% degradation achieved in 4 h of p-Np and MB dye in comparison to 20% and 25% obtained with pure TiO2 respectively. This feature is explained, according to all our analyses, on the basis of the time-delayed charge recombination processes which were found to be more important for B doped TiO2 photocatalysts than the value of the optical band gap that indeed was found to be significantly reduced in the case of TiO2-Bint+sub.

ACKNOWLEDGEMENT: The research activity is partially supported by UGC-UPE Green Technology Project in India and by the PAT (Provincia Autonoma di Trento) project ENAM in cooperation with Istituto PCB of CNR (Italy). The authors A.Dashora and R.Fernandes acknowledge DST and UGC for providing financial support through SERB-DST, New Delhi, India for Young Scientist Project and Dr. D. S. Kothari post-doctoral fellowship programme, respectively.

SUPPORTING INFORMATION: The equations and parameters used to fit the time resolved PL decay curve. Equations used to formulate the formation energies of B-doped TiO2 with boron at various sites in lattice. Figure S1 showing SEM image of TiO2 and B-doped TiO2. Figure S2 representing Comparison of photocatalytic degradation of p-nitrophenol blue dye under light irradiation in



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

presence of 1 at% B-doped TiO2 and 2 at% B-doped TiO2 after calcinating at 450 oC for 4 h and ball milled at 600 rpm for 6 h. Table S1 summerizing BET surface area and band gap values of the pure and B-doped TiO2. Table S2 comparing the photocatalytic activity for the degradation of p-NP and MB dye solution.

REFERENCES: [1]

Fujishima, X; Zhang, C.R. Titanium Dioxide Photocatalysis: Present Situation and Future Approaches. Chimie 2006, 9, 750–760.

[2]

Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Visible Light Photocatalysis in Nitrogen-Doped Titanium Oxides. Science 2001, 293, 269−271.

[3]

Di Valentin, C.; Pacchioni, G.; Selloni, A. Theory of Carbon Doping of Titanium Dioxide. Chem. Mater. 2005, 17, 6656−6665.

[4]

Umebayashi, T.; Yamaki, T.; Itoh, H.; Asai, K. Band Gap Narrowing of Titanium Dioxide by Sulfur Doping. Appl. Phys. Lett. 2002, 81, 454−456.

[5]

Yu, J. C.; Yu, J. G.; Ho, W. K.; Jiang, Z. T.; Zhang, L. Z. Effects of F-Doping on the Photocatalytic Activity and Microstructures of Nanocrystalline TiO2 Powders. Chem. Mater. 2002, 14, 3808−3816.

[6]

Chen, D.; Yang, D.; Wang, Q.; Jiang, Z. Effects of Boron Doping on Photocatalytic Activity and Microstructure of Titanium Dioxide Nanoparticles. Ind. Eng. Chem. Res. 2006, 45, 4110-4116.

[7]

In, S.; Orlov, A.; Berg, R.; Garcı´a, F.; Pedrosa-Jimenez, S.; Tikhov, M.S.; Wright, D.S.;. Lambert, R. M. Effective Visible Light-Activated B-Doped and B,N-Codoped TiO2 Photocatalysts. J. AM. Chem. Soc. 2007, 129, 13790-13791.


Page 22 of 35

Page 23 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[8]

Yang,K.; Dai,Y.; Huang, B.; Density Functional Study of Boron-Doped Anatase TiO2. J. Phys. Chem. C 2010, 114, 19830–19834.

[9]

Geng, H.; Yin, S.; Yang, X.; Shuai, Z.; Liu, B. Geometric and Electronic Structures of the Boron-Doped Photocatalyst TiO2. J. Phys.: Condens. Matter. 2006,18, 87–96.

[10] Finazzi, E.; Valentin, C.D.; Pacchioni, G. Boron-Doped Anatase TiO2: Pure and Hybrid DFT Calculations. J. Phys. Chem. C 2009, 113, 220–228. [11] Artiglia, L.; Lazzari, D.; Agnoli, S.; Rizzi, G.A; Granozzi, G. Searching for the Formation of Ti−B Bonds in B‑Doped TiO2−Rutile. J. Phys. Chem. C 2013, 117, 13163−13172. [12] Zaleska, A.; Grabowska, E.; Sobczak, J.W.; Gazda, M.; Hupka, J. Photocatalytic Activity of Boron-Modified TiO2 Under Visible Light: The Effect of Boron Content, Calcination Temperature and TiO2 Matrix Appl. Catal. B: Environ. 2009,89, 469–475. [13] Zaleska, A.; Sobczak, J.W.; Grabowska,E.; Hupka, J.; Preparation and Photocatalytic Activity of Boron-Modified TiO2 Under UV and Visible Light. Appl. Catal. B: Environ. 2008, 78, 92–100. [14] Xue, X.; Wang, Y.; Yang, H. Preparation and Characterization of Boron-Doped Titania Nano-Materials With Antibacterial Activity. Appl. Surf. Sci. 2013, 264, 94– 99. [15] Wu, T.T.; Xie, Y-P.; Yin, L-C.; Liu, G.; Cheng, H-M. Switching Photocatalytic H2 and O2 Generation Preferences of Rutile TiO2 Microspheres With Dominant Reactive Facets by Boron Doping. J. Phys. Chem. C 2015, 119, 84-89. [16] Feng, N.; Zheng A.; Wang Q.; Ren P.; Gao X.; Liu S.B.; Shen Z.; Chen T.; Deng F. Boron Environments in B-Doped and (B, N)-Codoped TiO2 Photocatalysts: A Combined Solid-State NMR and Theoretical Calculation Study. J. Phys. Chem. C 2011, 115, 2709-2719.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[17]

Feng, N.; Wang Q.; Zheng A.; Zhang Z.; Fan J.; Liu S.B.; Amoureux J.P.; Deng F. Understanding the High Photocatalytic Activity of (B, Ag)-Codoped TiO2 Under Solar-Light Irradiation with XPS, Solid-State NMR, and DFT Calculations. J. AM. Chem. Soc. 2013, 135, 1607-1616.

[18]

Zhao W.; Ma W.; Chen C.; Zhao J.; Shuai Z. Efficient Degradation of Toxic Organic Pollutants with Ni2O3/TiO2-xBx under Visible Irradiation. J. AM. Chem. Soc. 2004, 126, 4782-4783.

[19]

Blaha, P.; Schwarz, K.; Luitz, J. Wien2K Code, An Augmented Plane Wave Plus Local Orbitals Program for Calculating Crystal Properties, Vienna University of Technology, Vienna, Austria, 2001.

[20] Perdew, J. P.; Ruzsinszky, A.;

Csonka, G.I.; Vydrov, O.A,;. Scuseria, G.E.;

Constantin, L.A,; Zhou, X.; Burke, K. Restoring the Density-Gradient Expansion for Exchange in Solids and Surfaces. Phys. Rev. Lett. 2008, 100, 136406-1-4. [21] Tran, F.; Blaha, P. Accurate Band Gaps of Semiconductors and Insulators with a Semilocal Exchange-Correlation Potential. Phys. Rev. Lett. 2009, 102, 226401-1-4. [22] Rubio-Ponce, A.; Conde-Gallardo,A.; Olguín,D. First-Principles Study of Anatase and Rutile TiO2 Doped with Eu Ions: A Comparison of GGA And LDA+U Calculations. Phys. Rev. B 2008, 78, 035107-1-9. [24] Yang, K.; Dai, Y.; Huang, B.; Origin of the Photoactivity in Boron-doped Anatase and Rutile TiO2 Calculated from First Principles. Phys. Rev. B 2007, 76, 195201-1-6. [23] Choudhary, B.; Choudhary, A. Tailoring Luminescence Properties of TiO2 Nanoparticles by Mn Doping. J. Lumin. 2013, 136, 339-346.


Page 24 of 35

Page 25 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


FIGURE CAPTIONS: Figure 1: (a) XRD patterns and (b) Raman spectra of pure and B-doped TiO2 powders with different dopant concentrations. Inset shows the shift in peak position of anatase phase at 25.3o. Figure 2: (a) UV-Vis absorption spectra (taken in diffuse reflectance mode) and (b) Tauc plot, of pure and B-doped TiO2 powders with different dopant concentrations. Figure 3: XPS spectra of Ti2p, O1s and B1s levels of pure and B-doped TiO2 powders with different dopant concentrations. Figure 4: Total density of states (TDOS) [left panel] and crystal structure [right panel] of TiO2, TiO2-Bint, TiO2-Bsub and TiO2-Bsub + int. Figure 5: Energy bands (left panel) and partial DOS (right panel) of TiO2, TiO2-Bint, TiO2Bsub and TiO2-Bsub + int. Figure 6: (a) Photoluminescence emission spectra and (b) Time-resolved photoluminescence decay curve of pure and B-doped TiO2 powders with different dopant concentrations. Figure 7: Comparison of photocatalytic degradation of (a) p-nitrophenol and (b) methylene blue dye under light irradiation in presence of pure and B-doped TiO2 powders with different dopant concentrations, plotted in terms of the normalized intensity of the absorption band of p-NP and MB at 320 nm and 665 nm, respectively, in UV-vis measurements vs irradiation time (lines are drawn to guide the eye). Figure 8: (a) XPS spectra of Ti2p, and B1s levels and (b) UV-Vis absorption spectra (taken in diffuse reflectance mode) of 2 at% B-doped TiO2 after calcinating at 450 oC for 4 h and ball milled at 600 rpm for 6 h.



FIGURES:

Intensity (a.u.)

(a)

Pure TiO2 1% B-TiO2

A = Anatase R = Rutile B = Brookite

2% B-TiO2 3% B-TiO2 4% B-TiO2

A

23

24

25

26

R B

20

R

30

A R

A

A

40

R

50

A

A

60

A

70

80

2θ (Degree)

(b)

A = Anatase R = Rutile B = Brookite

Pure TiO2 1% B-TiO2 2% B-TiO2

Intensity (a.u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3% B-TiO2 4% B-TiO2

A

A

A

B B B

A R

A

100

200

300

400

500

600

700

800

-1

Raman shift (cm ) Figure 1:


Page 26 of 35

Page 27 of 35

1.0

(a)

Pure TiO2 1% B-TiO2 2% B-TiO2

Absorbance

0.8

3% B-TiO2 4% B-TiO2

0.6 0.4 0.2 0.0 200

300

400

500

600

700

800

Wavelength (nm)

2.0

2.0

(b)

1/2

(ev/cm)

1/2

1.5

(αhν)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1%B-TiO2

2%B-TiO2

1.5

1.0

1.0

0.5

0.5

2.0 0.0

0.0 2.0

1

2

3

4

1

2

3

4

1.6

1.5

3%B-TiO2

4%B-TiO2

1.2

1.0 0.8 0.5 0.4 0.0

1

2

3

4

0.0

1

2

3

hν (eV) Figure 2:


4


Intensity (Arb Units)

B1s

2%B-TiO2

1%B-TiO2

198 196 194 192 190 188 186 184 198 196 194 192 190 188 186 184

3%B-TiO2

4%B-TiO2

198 196 194 192 190 188 186 184 198 196 194 192 190 188 186 184

Binding Energy (eV) Pure TiO2

Ti 2p

1% B-TiO2

Intensity (Arb. units)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


468

466

464

462

460

458

456

454

Binding energy (eV)


Page 28 of 35

Page 29 of 35

Pure TiO2

O1s

1% B-TiO2

Intensity (Arb. units)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



536

534

532

530

528

526

524

Binding energy (eV)

Figure 3:



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4:


Page 30 of 35

Page 31 of 35

Energy (eV)

(a) TiO2-B (Int)

k

Energy (eV)

(b) TiO2-B (Sub)

k (c) TiO2-B (Int+Sub)

Energy (eV)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


k

Figure 5:



Fluorescence intensity (Arb. Units)

1000

(a)

Pure TiO2 1% B-TiO2

800

2% B-TiO2 3% B-TiO2

600

4% B-TiO2 400

200

0

400

420

440

460

480

500

520

540

560

580

Wavelength (nm)

Pure TiO2

Intensity (Arb. Units)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1% B-TiO2

(b)


1.0

1.1

1.2

Time (ns) Figure 6:


Page 32 of 35

Page 33 of 35

1.0

(a)

0.9

C/C0

0.8 0.7 0.6

Pure TiO2

0.5

1% B-TiO2 2% B-TiO2

0.4

3% B-TiO2

0.3

4% B-TiO2

0.2 0

50

100

150

200

250

Time (min)

1.1

(b)

1.0 0.9 0.8

C/C0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.7 0.6

Pure TiO2 1% B-TiO2

0.5

2% B-TiO2 3% B-TiO2

0.4

4% B-TiO2

0.3 0

50

100

150

200

Time (min.) Figure 7:


250


(a)

Intensity (Arb. Units)

B 1s

198

196

194

B 1s

Ball-Milled

192

190

188

186

184

198

Ball-Milled

Ti 2p

468

465

462

459

456

453

450

196

194

Heat-Treated

192

190

186

0.9

465

462

459

456

453

(b) 2% B-TiO2-As Preapared

0.8

2% B-TiO2-Thermal treament

0.7

2% B-TiO2-Ball milled

0.6

TiO2

0.5 0.4 0.3 0.2 0.1 0.0 200

300

400

500

600

184

Heat-Treated

Ti 2p

468

188

Binding Energy (eV)

Absorbance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 35

700

Wavelength (nm) Figure 8:


800

450

Page 35 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table of Content image


Experimental and Theoretical Investigations on the ... - ACS Publications

Recommend Documents