Facile One-Step Route for the Development of in ... - ACS Publications

Subscriber access provided by AUSTRALIAN NATIONAL UNIV

Article

Facile One-Step Route for the Development of In-Situ Co-Catalyst Modified Ti3+Self-Doped TiO2 for Improved Visible-Light Photocatalytic Activity Raju Kumar, Sivakumar Govindarajan, Reddy Kunda Siri Kiran Janardhana, Tata N. Rao, Shrikant Vishwanath Joshi, and Srinivasan Anandan ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b07000 • Publication Date (Web): 26 Sep 2016 Downloaded from http://pubs.acs.org on September 26, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 44

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

ACS Applied Materials & Interfaces

Facile One-Step Route for the Development of In-Situ Co-Catalyst Modified Ti3+Self-Doped TiO2 for Improved Visible-Light Photocatalytic Activity Raju Kumar,a Govindarajan Sivakumar, b Reddy Kunda Siri Kiran Janardhana,a Tata Narasinga Rao,a Shrikant Vishwanath Joshi,b Srinivasan Anandana* a

Centre for Nano Materials, bCentre for Engineered Coatings. International Advanced Research Centre for Powder Metallurgy and New Materials, Hyderabad-500005, India

ABSTRACT Development of visible-light-driven photocatalysts by employing a relatively simple, efficient and cost-effective one-step process is essential for commercial applications. Herein, we report for the first time the synthesis of in-situ Cu-ion modified Ti3+ self-doped rutile TiO2 by such a facile one-step solution precursor plasma spray (SPPS) process using a water soluble titanium precursor. In the SPPS process, Ti3+ self-doping on Ti4+ of rutile TiO2 is found to take place because of electron transfer from the created oxygen vacancies to Ti4+-ions. In-situ Cumodification of the above Ti3+ self-doped rutile TiO2 by additionally introducing a Cu solution into plasma plume is also demonstrated. While the Ti3+ self-doping induces broad absorption in the visible-light region, the addition of Cu-ion leads to even broader absorption in the visible region owing to resulting synergistic properties. The above materials have been evaluated for various self-cleaning photocatalytic applications under visible-light illumination. Cu-ion modified Ti3+ self-doped rutile TiO2 is noted to exhibit a remarkably enhanced visible-light activity in comparison with Ti3+ self-doped rutile TiO2, with the latter itself outperforming commercial TiO2 photocatalysts, thereby suggesting the suitability of the material for indoor applications. The broad visible-light absorption by Ti3+ self-doping, the holes with strong oxidation power generated in the VB, and electrons in Ti3+ isolated states that are effectively separated into the high reductive sites of Cu-ions upon visible-light irradiation, accounts for improved photocatalytic activity. Moreover, the synthesis process (SPPS) provides a valuable alternative to orthodox multi-step processes for the preparation of such visible-light-driven photocatalysts. Keywords: Solution precursor plasma spray, co-catalyst, self-doping, rutile TiO2, visible-light, self-cleaning

1

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

1. INTRODUCTION Titanium dioxide (TiO2) is an attractive semiconductor photocatalyst owing to its large band gap, chemical stability, non-toxicity, low-cost and has already found wide spread use in photocatalysis, photovoltaics, photonics, and photo-sensors. However, the large band gap of TiO2 (3.2 eV for anatase; and 3.0 eV for rutile) renders it active only in the UV region, thereby limiting its practical applications under visible-light illumination. To extend the light absorption of TiO2 into the visible region, several approaches such as cationic1 & anionic2 doping, surface modification3 and semiconductor composites4 have been investigated. Despite extensive research efforts to modify the optical properties of TiO2, the quantum efficiency (QE) under visible-light irradiation has remained too low to support photocatalytic reactions for practical use5. It has been understood from literature that high oxidation power of holes in the VB and high reduction power of the electrons in the conduction band (CB) are essential to achieve high photocatalytic activity in semiconductors.6 Consistent with the above requirements, co-catalyst such as Cu2+ or Fe3+ modified on TiO2 surface has been identified as promising material, in which the co-catalyst induces efficient interfacial charge transfer (IFCT) of VB electrons upon visible-light irradiation and causes multi-electron reduction of oxygen via trapping the photo-generated electrons.6-8 However, the visible-light absorption capacity of Cu2+ or Fe3+modified on TiO2 is relatively weak, as interfacial charge transfer (IFCT) occurs only at TiO2 particles/nanoclusters interfaces.9-11 Later, co-catalyst modification on doped TiO2 (CB controlled metal doped TiO2/ZnO) have been designed to improve the QE when compared to the pristine co-catalyst modified TiO2.12-14 However, Cu2+-ions modified metal doped TiO2 exhibited QE’s that were significantly lower than that of undoped co-catalyst modified TiO2 because of slow charge 2


Page 2 of 44

Page 3 of 44

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


transfer kinetics from doped level to co-catalyst on TiO2. Thus in order to achieve a high QE, the concept of energy level matching was presented, wherein, it was proposed that the dopant energy levels and surface energy states must be in similar levels for efficient transfer of electrons.15,16 Particularly, Cu2+ ion modified Ti3+ self-doped TiO2 has been reported with very high efficiencies and shows excellent photocatalytic properties.15 In this material, the strong visible-light absorption by Ti3+ self-doping and efficient electron transfer from doped level to the co-catalyst (Cu2+-ions) on account of energy level matching16 are attributed for the improved efficiency of TiO2. Though the above mentioned visible-light-driven photocatalyst is effective in exhibiting superior photocatalytic properties, the process involved to synthesize the same is tedious and in the need of multiple steps. For example, the fabrication of co-catalyst modified Ti3+ self-doped TiO2 includes the requirement of already synthesized rutile TiO2 & Ti2O3, oxidation treatment at high temperatures followed by Cu2+ grafting, which are expected to be time consuming and expensive processes. Hence, it becomes essential to synthesize visible-light-driven photocatalysts (co-catalyst grafted Ti3+ self-doped rutile TiO2) by employing relatively simple, efficient and cost-effective processes. Various attempts have been made to synthesize Ti3+ self-doped TiO2 by using chemical methods.17-20 For example, Ren et.al17 and Mao et.al18 have prepared Ti3+ self-doped TiO2 by reducing TiO2 powders using NaBH4 as a reducing agent and strong acid namely HCl, respectively. Pan et.al19 prepared Ti3+ self-doped TiO2 by Nd:YAG laser assisted synthesis. Though these processes are not complicated they require pre-synthesized TiO2 for Ti3+ doping into Ti4+ lattice by a separate process, along with additional requirement for strong acid/alkali conditions. Grabstanowicz et.al20 have synthesized Ti3+ self-doped TiO2 by oxidizing TiH2 wherein it required the preparation of raw materials (TiH2) and utilization of reducing gas

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

(hydrogen) followed by the requirement of oxidizing agent to synthesize Ti3+ self-doped TiO2 from TiH2. Shah et.al21 reported a synthesis of defective TiO2-x nanocrystal by using TiCl3 as the Ti precursor with visible-light photocatalytic activity. In addition, Zhang et.al22 also reported a microwave assisted synthesis of self-doped TiO2 which showed good photo-electrochemical water splitting properties under visible-light irradiation. The reported literature to date on the synthesis of Ti3+ self-doped TiO2 involves complicated experimental conditions (hazardous Ti precursors, reducing atmosphere like providing hydrogen gas, reducing agents, high vacuum conditions, requirement of microwaves and post heat treatment) and multi-step processing. In addition, efforts have also been made to synthesize Ti3+ self-doped TiO2 by plasma process, notable among them is the synthesis of Ti3+ self-doped TiO2 using RF plasma.23 Zhou et.al24 reported an enhanced H2 production on account of the presence of Ti3+ in the material. Zhou et.al25 synthesized a Ti3+ and nitrogen co-doped TiO2 which showed enhanced visible light absorption. However considering from an economic point of view it becomes essential to develop methods that are quick, easy and hassle free. Solution precursor plasma spray (SPPS) is one such simple and rapid processing technique to synthesize a wide range of nanostructured metal oxide materials from precursor solutions.26 The potential advantages of utilizing SPPS are (i) avoiding costly powder feedstock preparation (ii) better control over chemistry of prepared precursors and deposits (iii) ability to form nano-sized microstructures, and (iv) processing versatility and rapid exploration of novel precursor compositions. SPPS has been used to synthesize a wide range of oxides, including anatase, rutile and mixed phases of TiO2.27, 28 Hence in the present study we have focused on the development of a facile, one step process under ambient conditions to synthesize Ti3+ self-doped rutile TiO2 and subsequently to realize in-situ copper modification on the surface of Ti3+ self-doped rutile TiO2. With the method being

4


Page 4 of 44

Page 5 of 44

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


reported here, it is possible to not only have both self-doping and copper modification in a single step process but also ensure a continuous supply of required materials as long as there is supply of precursors. Secondly, the precursors used for the experimentation are water soluble and hence avoid the usage of other solvents making it cost effective. Unlike conventional TiO2 precursors which are volatile, hazardous and which require meticulous handling, water soluble Ti precursor used in the present study increases safety, with enhanced ease of handling and transportation. The adopted process in the present study is quick and easy without the need for any post annealing and can be used to obtain either coating or powder depending on requirement. Moreover, this is an easily scalable process i.e. the material is continuously produced as long as there is supply of precursors and overcomes short comings of batch type reactors where consistency can be a serious issue during production. Further, the photocatalyst can be directly applied on the desired surfaces like tiles; floor etc. by this process, thereby eliminating of multiple steps for photocatalyst coating. This paper is the first example to show that the formation of in-situ Cu2+ ion modified self-doped rutile TiO2 is possible in a single step. The highlights of the study are summarized herein and reveal interesting physico-chemical characteristics (Table 1) in terms of structural, optical, morphological properties etc. and exhibit considerable promise for various photocatalysis assisted self-cleaning applications under visiblelight illumination. 2. EXPERIMENTAL SECTION 2.1 Chemicals Commercially available Titanium (IV) bis (ammonium lactato) dihydroxide solution (TBLAH) (50 wt. % in H2O, Aldrich) and Copper (II) chloride dihydrate, 99% (CuCl2. 2H2O, Alfa Aesar) was used without any further purification. 5



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

2.2 Synthesis of Ti3+ self-doped rutile TiO2 and Cu-ion modified Ti3+self-doped rutile TiO2 The Ti3+self-doped rutile TiO2 was prepared by SPPS using water soluble TBLAH solution as Ti precursor. The TBLAH solution was transferred to a pressurized liquid reservoir and the resulting solution was directly injected through an atomizer nozzle into the plasma jet at a flow rate of 0.9 - 1.8 liters per hour in the form of droplets. The experiment was conducted using a direct current plasma torch (9 MBHE & 9 MBM Plasma Spray Gun, Sulzer Metco, USA) which was connected to a multi-axis robotic arm. The plasma power used was in the range 30 - 50 kW. Argon served as primary gas and hydrogen as secondary gas, whereas compressed air was used as the atomizing gas. The plasma torch was scanned across the stainless steel substrate by a robotic arm at the rate of 0.05 - 0.2 m/s. Unlike the more prevalent use of SPPS for forming coatings,29 in the present study the same SPPS techniques was used to collect material from a stainless steel substrate, which was then ground to obtain a fine powder (referred to as Ti3+ selfdoped rutile TiO2 based on subsequent characterization). The distance between spray gun and substrate was maintained at 0.04 - 0.150 meters. When the precursor solution is introduced into the plasma plume, it undergoes various transformations including evaporation, gelation, pyrolysis and sintering before it is finally deposited on the substrate. The aforementioned process was also repeated using suitable Cu forming precursor (with different atomic percentage of copper ions) dissolved in the Ti precursor as starting feedstock solution. As subsequent results reveal, this was to achieve in-situ formation of Cu-ion modified Ti3+self- doped rutile TiO2. 2.3 Characterization XRD analysis of the above powder samples was carried out using Cu-Kά (λ=1.5406 Å) radiation over two – theta range varying from 100 to 800 at room temperature (XRD, Bruker D8-advance, Germany). The crystal size of the SPPS derived materials were calculated using Scherrer’s

6


Page 6 of 44

Page 7 of 44

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


equation.30 UV-DRS analysis was carried out by a spectrophotometer (Varian Carry 5000, UVvis NIR). The band-gap of developed material was estimated by Kubelka Munk equation.31 FESEM (S-4300-SE/N microscope, Hitachi, Japan) and HR-TEM (TECNAI-200 kV, FEI Netherlands) were used to determine the morphology and size distribution of the prepared materials. XPS (Omicron XPS system) was used to investigate the surface chemistry of titanium and oxygen in Ti3+self-doped rutile TiO2. Electron Paramagnetic Resonance (JEOL/JES-FA200) analysis was done at room temperature in the range of 0-1000 mT with a sweep width of 500 mT at an X band frequency of 9.44 GHz. The concentration of Cu was measured by rf-ICP (Varian 720-ES). For ICP analysis, the solution was prepared as follows: 0.1g of material (Ti3+ selfdoped rutile TiO2 or Cu-ion modified Ti3+ self-doped rutile TiO2) and 2 g of KHSO4 crystal (SDFCL) were placed in a 10 ml platinum crucible and heated in a high temperature flame. The resulting suspension was transferred into 50 ml of diluted H2SO4 and placed on a hot plate at 400 °C until the material was completely dissolved. Finally, a homogenous solution was obtained by adding 50 mL of water into the above solution and the resulting solution was used for the elemental analysis of Cu and Ti. A known concentration of Cu (1000 ppm ICP Multi-element Standard Solution IV, Merck) and Ti (1000 ppm, Accu Trace, Traceable to natural Institute of standards and technology reference materials, USA) solution were used as standard solution to quantify the unknown concentration of Cu and Ti in Ti3+self-doped rutile TiO2 and Cu-ion modified Ti3+ self-doped rutile TiO2. 2.4 Photocatalytic activity 2.4.1 Decomposition of acetaldehyde: Photocatalytic oxidation activities of Ti3+ self-doped rutile TiO2 and Cu-ion modified Ti3+ selfdoped rutile TiO2 nanomaterials were evaluated by the degradation of gas phase acetaldehyde.

7



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 photocatalyst was loaded in a static glass reactor (500 mL). Prior to photocatalytic measurements, the photocatalyst was pre-treated for 1 h using LED light illumination in air. After loading the sample into the glass reactor, 100 ppmv of acetaldehyde was injected and the entire system was placed in dark for 2 h to ensure adsorption-desorption equilibrium. A blue LED lamp, which emits light of wavelength between 450 to 600 nm, was used as the light source for photocatalytic reaction. The intensity of the blue LED lamp, measured by a spectroradiometer (USR-40D, Ushio Ltd), was 13 mW/cm2. The concentrations of acetaldehyde and the generation of CO2 were measured using a 4-channel Micro gas chromatograph (Agilent 490 Micro GC), in which each channel was equipped with an electronic gas control, injector, column, and universal TCD detector. Pora plotQ (10 m), and CP-Sil 5CB (8 m) columns were used to separate carbon dioxide and acetaldehyde from the gas mixture. 2.4.2 Anti-bacterial activity and Decomposition of methylene blue: Anti-bacterial activity as well as decolorization of dye efficiency of the developed material was evaluated by the in-activation of Escherichia coli (E.coli) and the decomposition of methylene blue (MB) using the experimental procedure as reported elsewhere.32 Briefly, E.coli was used as a test pathogen to assess the photocatalytic efficiency of the photocatalysts. Initially a bacterial suspension was prepared by cultivating the bacteria overnight. 50 mL buffer (phosphate) solution was added to 50 mg of photocatalyst and sonicated for 1h. The suspension was transferred to a Petri dish in which 0.5 mL of cultivated E. coli was added. The resulting suspension was illuminated under blue-LED light. At regular intervals, samples were withdrawn from the Petri dish, diluted and later cultivated at 370C for 24h. The percentage of surviving bacteria was quantified to determine the efficiency of the photocatalyst. For MB degradation studies, 50 ml of aqueous MB solution with a concentration of 0.01 mM was added to 50 mg of photocatalyst and

8


Page 8 of 44

Page 9 of 44

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 suspension was sonicated in the dark for 30 min to reach adsorption−desorption equilibrium before illumination. Sun light with/without UV cut-off filter was used as a visible-light/UV light source, respectively, for photocatalytic experiment. At certain time intervals during the experiment, the absorbance values of MB solution were measured by a spectrophotometer. The peak absorbance value of MB appeared at 664 nm and the change of absorbance was measured which was plotted against time. 3. RESULTS AND DISCUSSION SPPS is a versatile technique to achieve Ti3+ self-doping and, further, in-situ Cu-ion modification of the same in one step process. The prepared materials were characterized using various analysis techniques to assess our claim. The mechanism for the formation of Ti3+ self-doped rutile TiO2 by SPPS is illustrated in Figure 1. During the SPPS process, precursor droplets are exposed to high plasma temperature (~8000 K) for a short residence time (few milliseconds).33 The entrained precursor droplets undergo many physico-chemical transformations upon exposure to the high temperature plasma, before it is converted into Ti3+ self-doped rutile TiO2 deposits (Figure 1A). According to prior literature,34 depending upon use of appropriate precursor properties and spray variables, the injected precursor droplets have been shown to eventually form molten splats upon impact with a substrate. Considering that the droplets pass through high temperature zones of the plasma plume, the in-situ formed TiO2 is likely to undergo complete melting (melting point of TiO2 is ~1840 °C) followed by crystallization. The TiO2 precursor forming droplets, entering into high temperature zones are expected to have oxygen vacancies within the TiO2 lattice, where one oxygen vacancy results in formation of two excess electrons.35 For charge compensation, these electrons move from oxygen vacancies to Ti4+ sites of TiO2, thereby forming Ti3+ self-doped rutile TiO2. According to prior literature, oxygen vacancies are

9



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 10 of 44

reported to form in bulk of TiO2 either by treating the powders at elevated temperatures under reduced atmosphere or subjecting to heat treatment above 700°C.36 During SPPS deposition, the repeated heating by the plasma plume over the previously deposited layer typically acts akin to a secondary heat treatment process assisting in the formation of oxygen vacancies. In order to promote formation of the above oxygen vacancies, the traverse speed was optimized to induce high heat flux during rastering. Moreover, since the temperature is in excess of 8,000 K in the core of the plasma plume, it is reasonable to expect significant Ti3+ rich rutile TiO2 formation on account of more oxygen vacancies created at high temperature as shown in Figure 1B. Thus, Ti3+ self-doped rutile TiO2 preparation is achieved by a one-step SPPS process through a proper combination of spray parameters and precursor formulation, without the need of a reducing agent or any secondary treatment. Following the formation of in-situ Ti3+ self-doped rutile TiO2 by the SPPS method as above, Cu-ion modified Ti3+ self-doped rutile TiO2 was also successfully prepared using mixed precursor solutions comprising of Ti and Cu precursors before being introduced into the plasma plume. The resulting deposited materials are referred to as Cu-ion modified Ti3+ self-doped rutile TiO2, since results discussed subsequently will indeed establish Cu-modification. Powder X-ray diffraction (XRD) was used to measure the crystal structure of Ti3+ self-doped rutile TiO2 (Figure 2). The resulting powder shows strong diffraction peaks {Figure 2A (a)} at 2θ = 27.47o, 36.03o, 39.22o, 41.31o, 44.10o, 54.26o, 56.65o, 62.72o, 64.02o, 69.0o, and 69.20o, which can be indexed to (110), (101), (200), (111), (210), (211), (220), (002), (310), (301), and (112) planes of standard rutile TiO2 {Figure 2A (c)}, respectively, (in agreement with ICDD-00021-1276). The sharp peaks indicate that the rutile TiO2 formed is highly crystalline. The formation of pure rutile phase TiO2 without any impurities, similar to commercial rutile TiO2

10


Page 11 of 44

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 2A (b)}, has been achieved in this work by carefully controlling the experimental conditions. Such a pure rutile phase could be an advantage for photocatalytic applications as it is more photoactive than anatase under visible light illumination. However, enlarged XRD patterns of Ti3+ self-doped rutile TiO2 {Figure 2A (a)} and commercial rutile TiO2 {Figure 2A (b)} show difference as shown in Figure 2B. It is intersting to note that the (110) and (101) planes of Ti3+ self-doped rutile TiO2 prepared in this study shows a shift compared to commercial rutile TiO2 similar to XRD peak shift by metal ions in metal doped TiO2,13,14 demonstrating that the Ti3+ formed during the process may have been doped on Ti4+ sites of rutile TiO2. The XRD pattern of Cu-ion modified Ti3+ self-doped rutile TiO2 (Figure S1A-a) is in good agreement with the patterns of commercial (Figure S1A-b) and standard rutile TiO2 (Figure S1A-c) (ICDD-00-0211276) and these results conclude that Cu-ion modified Ti3+ self-doped rutile TiO2 photocatalysts retain the crystal structure similar to rutile TiO2, reinforcing the fact that Cu-ion modification does not change the crystal structure of rutile TiO2. The XRD patterns of Cu-ion modified Ti3+self-doped rutile TiO2 with different concentrations of Cu-ion (Figure S1B) reveal that there is no change in XRD even at high concentration of Cu-ion. One can expect Cu-doping into TiO2 when a Cu precursor is introduced in a high temperature plasma plume; however, the XRD results are not conclusive to confirm Cu doping into TiO2. In general, when Cu-ion is doped into the TiO2 lattice,37 a shift of (110) peak of rutile TiO2 occurs. The shift in peak position with increasing dopant concentration indicates that Cu2+ ions replaced some Ti4+ ions in the TiO2 lattice. In the present study, such shift is being observed for photocatalyst prepared without the addition of any copper precursor i.e. Ti3+ self-doped TiO2 {presence of Ti3+ in the bulk of TiO2 confirmed in the later section (EPR and XPS)} is itself giving a shift compared to commercial rutile TiO2 which is reinforcing a fact that Ti3+ doping is responsible for the shift. Secondly, the

11



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 12 of 44

Ti3+ self-doped TiO2 after Cu2+ modification exhibit similar XRD pattern like Ti3+ self-doped TiO2 without any shift indicating that the lattice parameters did not change concluding that doping into the crystal lattice did not occur. Thirdly, many studies by previous authors12-14 have concluded that copper modification on the surface of TiO2 does not lead to a change in the XRD pattern as there is no doping of copper into the lattice rather the particles stay in the form of Cu2+ nanoclusters on the surface. The above points lead us to claim that the Cu-ions are not getting doped into the TiO2 lattice. Further, it was reported previously37 that CuO like entities may be incorporated in TiO2 by scavenging process or condensation onto the surface of TiO2 particles when Cu and Ti precursor solutions are introduced at high temperatures into the flame. The latter observation, i.e. the formation of CuO like entities i.e. Cu (II) nanoclusters is highly possible in the present study since kinetics parameters such as residence time and precursor feed concentration are similar to latter experimental conditions, which indicates that Cu-ions exist on the surface of rutile TiO2 particles. Moreover, there are no XRD peaks corresponding to metallic copper and, thus, we can conclude that copper ion may be present on the surface of Ti3+ selfdoped rutile TiO2 in an amorphous form. This conclusion is based on the fact that, as shown later, the presence of copper is conclusively established by EELS, rf-ICP, and Raman spectroscopic analysis. The field emission scanning electron microscope (FE-SEM) images of Ti3+ self-doped rutile TiO2 and Cu-ion modified Ti3+ self-doped rutile TiO2 are shown in Figure 3. The particles are disordered in shape with different sizes and agglomeration as being evident from FE-SEM images (Figure 3 - A, B, C and D). A wide particle size distribution, with clear grain boundaries are observed in case of both pristine and Cu-ion modified Ti3+ self-doped rutile TiO2. HR-TEM analysis (Figure S2 and Figure 4) shows distorted spherical TiO2 particles of sub-micron size,

12


Page 13 of 44

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


whereas the selected area electron diffraction (SAED) pattern indicates that the material is highly crystalline. The lattice fringes show the growth of rutile TiO2 along the (101) direction and the lattice spacing of approximately 2.5 Å corresponds to the d-spacing between the adjacent (101) crystallographic planes of rutile TiO2 as given in Figure 4. The dispersion of Cu nanoclusters (indicated by white arrows) on Ti3+ self-doped rutile TiO2 are shown in Figure 4B and 4C. The semicircular dotted black lines in Figure 4B indicate the presence of amorphous coppernanoclusters and their attachment to the Ti3+ self-doped rutile TiO2 surface. In the present study, we employed electron energy loss spectroscopy (EELS) to investigate the low concentrations of Cu-ion (particles inside semicircle dotted black lines as shown in Figure 4B in Cu-ion modified Ti3+ self-doped rutile TiO2, since EELS is one of the most sensitive and versatile method for the investigation of the outermost surface layer. Figure 4D shows the energy loss spectrum of Cuion modified Ti3+ self-doped rutile TiO2, in which M23 and M1 ionization edges of Cu-ions are observed at binding energies of ~80 and 100 eV, respectively, indicating the presence of Cuion38 on the surface of Ti3+self- doped rutile TiO2, although the oxidation state of Cu-ion is not clear. Similar kinds of surface nanoclusters on TiO2 particles have recently been reported.15, 37 Based on HR-TEM analysis, antibacterial activity (described ahead) and the results of similar systems, we are anticipating that in the present study the Cu2+ nanoclusters are present on the surface of Ti3+ self-doped rutile TiO2. XPS survey spectrum and the wide scan spectra of Ti3+ self-doped rutile TiO2 reveals that the material is composed of titanium and oxygen as shown in Figure 5A. The wide scan spectrum (Figure 5B) of Ti2p exhibits a doublet at 460.8 eV for Ti2p3/2 and at 466.4 eV for Ti2p1/2. Generally, the binding energy difference of Ti2p1/2 and Ti2p3/2 is approximately 5.8 eV, which corresponds with +4 oxidation state of Ti in TiO2. Though a shift in binding energy is observed when compared to standard rutile the same

13



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 14 of 44

difference of approximately 5.8 eV between Ti2p1/2 and Ti2p3/2 is still maintained, confirming the presence of Ti4+ in Ti3+ self-doped rutile TiO2. Secondly, there are additional peaks at 458.6 eV & 464.3 eV which can be ascertained to the 2p3/2 and 2p1/2 of Ti3+ ions.39, 40 The shift in XPS spectrum of our material and the presence of Ti3+ (2p3/2 and 2p1/2) peaks confirm the doping of Ti3+ into Ti4+ ions, which were created by high temperature induced oxygen vacancies in TiO2. To determine the binding states of oxygen in Ti3+ self-doped rutile TiO2, the O1s spectrum was de-convoluted into three peaks centered at binding energies of 529.7 eV, 530.94 eV and 532.6 eV. The XPS peaks of O1s spectrum (Figure 5C) at 529.7 eV and 530.94 eV could be assigned to lattice oxygen in TiO2 and Ti2O341 while the peak at 532.6 eV could be ascribed to adsorbed oxygen on the surface. XPS results of Ti2p and O1s conclude in-situ self-doping of Ti3+ onto Ti4+ sites during the formation of rutile TiO2. The XPS analysis of Ti2p and O1s of Cu-ion modified Ti3+ self-doped rutile TiO2 (Figure not included) also reveals in-situ self-doping of Ti3+ onto Ti4+ sites during the formation of rutile TiO2 and the presence of oxygen vacancies which is similar to the trend observed in Ti3+ self-doped rutile TiO2, indicating the existence of Ti3+ in Cuion modified Ti3+ self-doped rutile TiO2. Cu-ions in Cu-ion modified Ti3+ self-doped rutile TiO2 were not detected either in survey or wide scan spectrum as the concentration of Cu-ions used for the modification of TiO2 is maybe below the detection limit of XPS. However, the EELS analysis presented previously is conclusive and clearly shows the presence of copper on the surface of rutile TiO2. Further, elemental analysis by rf-ICP confirms the presence of 0.14 % Cu in Cu-ion modified Ti3+self- doped rutile. Raman analysis (Figure S3) of Cu (II)-ions modified Ti3+self-doped rutile TiO2 also shows characteristic peak of Cu2+ of CuO at 632 cm-1, which is consistent with the results reported previously for CuO42 indicating the presence of copper ions in Cu-ion modified Ti3+ self-doped rutile TiO2. Though the Cu-ions were not detected by XPS

14


Page 15 of 44

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


analysis, HR-TEM and rf-ICP analysis clearly show the presence of Cu-ions as nanoclusters on the surface of TiO2 with appropriate concentration of Cu-ions respectively. Further, the results of XRD and anti-bacterial activity in dark43 (described later) well support the presence of Cu-ions in the Cu (II) state by ruling out the existence of metallic copper and Cu1+ ions in the Cu (II)-ions modified Ti3+self-doped rutile TiO2. In addition, the UV-DRS results of Cu (II)-ions modified Ti3+self-doped rutile TiO2 discussed later clearly indicates a broad absorption spectrum in the visible region which is due to the grafting of Cu (II) ion nanoclusters onto the surface of Ti3+selfdoped rutile TiO2 consistent with results reported previously.6,13,32 Furthermore, Irie et. al.44 have extensively studied the copper ion nanoclusters existing on the surface of TiO2 similar to the present study with the help of XANES & EXAFS and concluded that the Cu-ions are grafted as Cu (II)-ions which is having distorted CuO like structure. Considering all our experimental results as well as the available literature, herein we are concluding that the Cu (II) nanoclusters are grafted onto the surface of Ti3+ self-doped TiO2 as Cu2+ ion as in similar to the structure of CuO. For confirming the presence of Ti3+ in our sample we further characterized our sample with the help of EPR analysis. Generally, an EPR peak at g~ 1.98 confirms the existence of Ti3+ in the bulk whereas the EPR peak at g ~ 2.02 indicates the presence of Ti3+ on the surface.39 When we analyzed our Ti3+ self-doped rutile TiO2 by EPR analysis (Figure 5D), it exhibits a major peak at g ~ 1.98 corresponding to Ti3+ states in the bulk of TiO2 consistent with previous literatures39, 45 and shows no peaks at g ~ 2.02. Thus from this data we can confirm that Ti3+ is present in a large fraction in the bulk confirming our claim that Ti3+ is doped on the Ti4+ sites in TiO2. EPR analysis of the Cu-ion modified Ti3+ self-doped rutile TiO2 also shows similar peaks at g ~ 1.98. Thus the presence of Ti3+ is confirmed both by XPS and EPR analysis.

15



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 44

The UV−Vis. diffuse reflectance spectra of Ti3+ self-doped rutile TiO2, Cu-ion modified Ti3+ selfdoped rutile TiO2 and commercial rutile TiO2 are given in Figure 6A. Ti3+ self-doped rutile TiO2 exhibits broad visible light absorption in the range of 420-700 nm compared to commercial rutile TiO2 due to Ti3+ doping onto Ti4+ of rutile TiO2.46 The calculated band gap energy of Ti3+ selfdoped rutile TiO2 is 2.87 eV, which is lower than that of commercial rutile TiO2 (Ebg= 2.97 eV) as given in Figure 6B, indicating that high concentration of Ti3+ doping could be responsible for the above. It was observed that the Cu-ion addition onto Ti3+ self-doped rutile TiO2 induces small visible light absorption (450-500 nm) due to interfacial charge transfer (IFCT) whereas the broad absorption (700-800 nm) is due to d-d transition47 of Cu-ion. There is no change in band gap energy value (Figure 6B) of Cu-ion modified Ti3+ self-doped rutile TiO2 demonstrating that Cu-ion addition only induces surface states rather than band gap narrowing. It is interesting to note that the absorption spectrum of Cu ion modified Ti3+ self-doped rutile TiO2 in the wavelength region ~400-500 nm matches well with the spectrum of LED lamp. Hence it is expected that only the absorption at ~ 400-500 nm (IFCT) contributes for the visible-light activity of Cu ion modified Ti3+ self- doped rutile TiO2. Standard rutile TiO2 did not show any significant absorption spectrum in the visible region due to its large band gap. The absorption spectra of Cu-ion modified Ti3+self-doped rutile TiO2 with different concentrations of Cu-ions (Figure S4A) reveals the broadening of absorption spectrum with increase in Cu-ion concentration. There is negligible change in band gap energy value observed for different concentrations of Cu-ion modified Ti3+ self-doped rutile TiO2 (Figure S4B). We have also provided an image of the various photocatalysts prepared by the SPPS in Figure S5. Thus, Cuion modified Ti3+ self-doped rutile TiO2, with improved visible light absorption which has been

16


Page 17 of 44

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


achieved by one-step process as demonstrated in the present study will be a promising candidate for self-cleaning photocatalytic applications under visible light illumination. The self-cleaning photocatalytic activity of aforementioned visible light photocatalytic materials were assessed by decomposition of gas phase acetaldehyde (Figure 7). The decrease in concentration of acetaldehyde and increase in concentration of carbon dioxide during decomposition of acetaldehyde by Ti3+ self-doped rutile TiO2 and Cu-ion modified Ti3+ selfdoped rutile TiO2 is shown in Figure 7A and Figure 7B, respectively. The generation of carbon dioxide by Cu-ion modified Ti3+ self-doped rutile TiO2 with different concentrations of Cu-ion is shown in Figure 7C. For the comparison, the results of photocatalytic decomposition of acetaldehyde without material (as a control), with a commercially procured TiO2, with commercial Degussa P25, with Ti3+ self-doped rutile TiO2 and with Cu-ion modified Ti3+ selfdoped rutile TiO2 are included in Figure 7D. Neither decrease in concentration of acetaldehyde nor an increase in concentration of carbon dioxide is observed for photocatalytic reaction without material, indicating that light illumination alone is not enough for acetaldehyde decomposition. Commercial rutile TiO2 showed negligible visible light photocatalytic activity, which is due to the large band-gap of TiO2 and the lack of absorption in the visible region. Degussa P25 TiO2 does not show any degradation as it does not absorb visible light on account of its large bandgap. In contrast, Ti3+ self-doped rutile TiO2 exhibited visible light photocatalytic activity due to the broad visible light absorption which is induced by Ti3+ doping onto Ti4+ of rutile TiO2. Although the photocatalytic activity of Ti3+ self-doped rutile TiO2 is better than that of the commercial photocatalyst, the decrease in concentration of acetaldehyde is found to be less when compared to Ti3+ self-doped rutile TiO2 modified with Cu-ion. The reason for the above is that the isolated states formed below the CB edge due to Ti3+ self-doping localize the electrons, leading to a

17



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 44

decrease in the mobility of photo-generated electrons and a resultant decrease in photocatalytic activity under visible light. Moreover, the isolated energy levels formed between the band gap due to Ti3+ self-doping decrease the reduction potential of electrons and, hence, it is not feasible to reduce the oxygen molecules to produce O2- radicals through single electron reduction (E= 0.064 Vs SHE, pH=0). On the other hand, the in-situ addition of Cu-ion during the formation of Ti3+ self-doped rutile TiO2 greatly improved the decomposition of acetaldehyde in comparison with pristine Ti3+self-doped rutile TiO2 because of the overlap between the potential energy of Ti3+ isolated states and that of Cu2+/Cu+ redox couples (0.16eV), the excited electrons in Ti3+ isolated states can be transferred to Cu-ion upon visible light irradiation, thus improving photocatalytic activity through efficient charge separation of photo-generated charge carriers (holes/electrons) and having a high quantum efficiency.15 After 2h of illumination, the Cu-ion modified Ti3+ self-doped rutile TiO2 completely decomposed acetaldehyde, whereas only 30% decomposition of acetaldehyde was observed in case of Ti3+ self-doped rutile TiO2 without Cuion modification. Further, the generation of

Facile One-Step Route for the Development of in ... - ACS Publications

Recommend Documents