Synthesis and Characterization of Cu and N Codoped RF-Sputtered

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Article

Synthesis and Characterization of Cu and N CoDoped RF-Sputtered TiO2 Films: Photoluminescence Dynamics of Charge Carriers Relevant for Water Splitting. Zakaria El Koura, Massimo Cazzanelli, Nicola Bazzanella, Nainesh Patel, Rohan Fernandes, Georgios A. Arnaoutakis, Anna Gakamsky, Andrew Dick, Alberto Quaranta, and Antonio Miotello J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.6b03058 • Publication Date (Web): 18 May 2016 Downloaded from http://pubs.acs.org on May 23, 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.

The Journal of Physical Chemistry C 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 27

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

Synthesis and Characterization of Cu and N Co-Doped RF-Sputtered TiO2 Films: Photoluminescence Dynamics of Charge Carriers Relevant for Water Splitting.

Z. El Koura1, M. Cazzanelli1*, N. Bazzanella1, N. Patel2, R. Fernandes2, G. E. Arnaoutakis3, A. Gakamsky3, A. Dick3, A. Quaranta4, and A. Miotello1

1.

Laboratorio IdEA, Dipartimento di Fisica, Università di Trento, via Sommarive, 14 Povo (Trento)

Italy. 2.

Department of Physics and National Centre for Nanosciences & Nanotechnology, University of

Mumbai, Vidyanagari, Santacruz (E), Mumbai 400098, India 3.

Edinburgh Instruments Ltd., Livingston, EH54 7DQ, United Kingdom

4.

Department of Industrial Engineering, University of Trento, Via Sommarive 9, I-38132 Povo, Trento,

Italy

*

authors to whom the correspondence should be addressed: [email protected] Tel. +39-

0461-282954.

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

Page 2 of 27

Abstract

Cu-N co-doped TiO2 anatase thin films with a nano-columnar architecture have been synthesized by RF-magnetron sputtering and characterized by Raman, scanning electron spectroscopy, and Xphotoelectron spectroscopy. Absorption, photoluminescence, and photoluminescence lifetimes of the prepared samples have been investigated to understand the dynamics of the photo-generated carriers in connection to both introduced defects and the modified TiO2 band-structure. At low concentrations Cu is mainly present as Cu+, while at higher concentrations the Cu2+ oxidation state prevails. Nitrogen, at low concentration and without the presence of copper dopant, substitutionally replaces oxygen to form O-Ti-N linkage. With increasing concentration, interstitial nitrogen and Ti-O-N and Ti-O-N-O linkages are observed. In all co-doped samples nitrogen is present both as interstitial and substitutional dopant. From photoluminescence spectra it is observed that nitrogen, in cooperation with Cu, more heavily affects the oxide structure, through Ti-N linkages, in such a way to quench the TiO2 exciton luminescence through charge trapping or energy transfer mechanisms. Time resolved PL analysis evidenced that Cu-N co-doping hinders the exciton radiative recombination in the anatase network, giving rise to increase of both the mean lifetime and recombination probability on defects at the nanocolumn surface.


Page 3 of 27

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

TiO2 is the most studied material for hydrogen production by photocatalytic water splitting and has also important applications in water treatment, air purification, self-cleaning coatings and self-sterilising coatings1. Indeed, its high resistance to corrosion and photo-corrosion, the energy band edges wellmatching the redox level of water, its low cost and widespread availability make it a very promising photocatalyst material for water splitting. However, the photocatalytic performance of TiO2 is limited by its wide band gap (laying in the UV) and by the rapid (∼µs) recombination rate of the photogenerated charge carriers. Many efforts have been spent in the last decades to solve these drawbacks2-8. One of the main challenges in TiO2 photocatalysis is to shift its optical absorption towards the visible region to exploit a wider portion of sunlight. The main adopted strategy to enhance the visible absorption of TiO2 is by doping either with metal cations or with non-metal anions4-7. However, the high concentration of dopants required to obtain an effective narrowing of the band gap, increases the recombination rate of the charge carriers. Thus, an enhanced absorption does not always result in a satisfactory photocatalytic activity. Recently, the concept of second-generation TiO2-based materials was introduced, and the co-doping with two dopant elements was proposed to produce a synergetic effect enhancing the visible light absorption and reducing the charge recombination rate6. For example, effective narrowing of the band gap of TiO2 was achieved by V and N co-doping8.This is due to the fact that the V-dopant creates energy levels below the conduction band minimum (CBM) and N creates levels just above the valence band maximum (VBM). In addition, V and N species present in low concentrations act as electron and hole trapping centres, respectively, thus limiting the recombination of photogenerated electron – hole pairs.



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 4 of 27

Zhu et al.9, described a new concept of non-compensated n–p co-doping generating an intermediate band (IB) in the mid gap of TiO2. This method increases both the thermodynamic stability and the kinetics of charge carriers. Cr and N co-doping was presented as a model for non-compensated n-p codoping, showing an enhanced visible light absorption and charge separation. However, localized dstates of Cr below the CBM make this material a bad candidate for proton reduction in the water splitting process because of the non favorable position of the new conduction band level for electron transfer to reduce proton. Therefore, Dashora et al.10 looked for a new dopant element pair, which does not form impurity states near the CBM and, at the same time, generates a band in the middle of TiO2 band gap. First principles calculations on the electronic and optical properties of Cu and N co-doped TiO2 revealed the presence of an isolated intermediate band deeply located in the band gap due to strong hybridization between Cu 3d and N 2p orbitals. This intermediate band has the property to enhance the visible light absorption of TiO2. At the same time, Cu and N co-doping only causes a small shift of the conduction band and preserves the ability of TiO2 to drive the water reduction reaction. The enhancement of the photocatalytic activity of TiO2 prepared by sol-gel by and Cu and N co-doping was confirmed by experiments on degradation of organic pollutants11-13.

However, an effort to

correlate the optical features (like absorption, photoluminescence and photoluminescence lifetimes) of this important photocatalytic material with its crystalline structure and with the interaction of the photogenerated carriers with both defects and the modified TiO2 band-structure is still lacking. Radio-frequency magnetron sputtering, is known to be a versatile physical-vapour-deposition technique enabling the deposition over a wide range of different substrates. In particular, it offers a precise control of the growth parameters, allowing a fine tuning of the morphology and crystallinity of the deposited material. Its employment in the production of visible-absorbing co-doped TiO2 photocatalysts is very useful for possible industrial application of such environmental devices.


Page 5 of 27

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


In this work we study the structural, morphological and optical properties of sputtered TiO2 thin films doped with Cu and/or N at different concentrations to understand the effect of co-doping on the recombination dynamics of photoexcited charge carriers and to try to increase the relaxation time before charges recombination.

2. Experimental Pure, Copper, and Nitrogen mono- and co-doped TiO2 samples were deposited by radio frequency (RF) magnetron sputtering starting from a ceramic TiO2 target (purity 99.99%). Argon (purity 99.9%) as a working gas and 150 W of RF-power was employed. Copper doping was obtained by placing metal copper 2x2 mm2 pieces (purity 99.99%) over the 180 mm2 TiO2 target, and nitrogen doping by inletting a controlled flux of nitrogen gas into the deposition chamber. Co-doping was obtained by combining the two strategies. Before the deposition, a base pressure of ca. 10-7 mbar was achieved, then Argon gas was introduced in the deposition chamber by a mass-flow controller such that a constant pressure of 8×10-3 mbar of Argon was maintained. The target-substrate distance was kept constant at 6.0 cm, and the substrate holder was kept oscillating in the plasma to improve the homogeneity of the deposited films. The substrate temperature was raised up to 350 °C before the deposition and maintained constant (± 2 °C) throughout the deposition process. Before each deposition, the targets were pre-sputtered for 20 minutes to remove all possible contaminants from their surface. Samples with different dopants concentrations were obtained by fixing a number of Cu pieces (0, 2, 3, 4, and 5) on the target and by varying the nitrogen flux to 0, 10, 20, and 30 sccm (the argon flux was 20 sccm for all the depositions). 20 thick samples (ca. 800 nm) with different N and/or Cu concentrations were deposited on silicon (100) and quartz, to study the morphological, structural and optical properties of the films. The samples were designated as XYY, where X reports the number of copper pieces used during the deposition, while YY reports the flux (in sccm) of nitrogen in the chamber. For example the 5 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

Page 6 of 27

sample 310 was deposited by placing 3 pieces of copper on the TiO2 target and fluxing 10 sccm of nitrogen during the whole deposition process. Structural characterization of the samples was carried out by Raman spectroscopy, using a LabRam Aramis Jobin-Yvon Horiba microRaman system equipped with a 100X objective and He-Ne laser source (632.8 nm). Scanning Electron Microscopy (SEM) was used to analyse the morphology and the thickness of the films. The surface of the films was analysed by a JEOL JSM-7001F field effect gun (FEG-SEM) apparatus at 20 keV beam energy. SEM apparatus is equipped with an Energy Dispersive X-ray Spectroscopy detector (EDXS, Oxford INCA PentaFETx3), that was used to estimate the concentration of copper. X-ray Photoelectron Spectroscopy (XPS) was performed to examine the surface composition and chemical states of the samples using a “PHI 5000 VersaProbe II” instrument equipped with a mono-chromatic Al Kα (1486.6 eV) X-ray source and a hemispherical analyser. Appropriate electrical charge compensation was employed to perform the analysis. The binding energy was referenced to the C1s peak at 284.8 eV. Photoluminescence (PL) measurements were performed using the third-harmonic output (355 nm) of a Nd:YAG laser operating at a repetition rate of 3 KHz and having a pulse duration nominally equal to 5 ns. The spectrum was recorded with a thermoelectrically-cooled Ocean Optics FL-Q65000 fiberpigtailed Spectrometer. Integration time was chosen equal to 5 sec per spectrum. The scattered laser light was filtered with a Semrock edge filter mod. BLP01-364R-25. Absorption spectra were measured with a Cary5000 spectrophotometer having a resolution of 0.4 nm and an integration time of 0.1 sec per point. Time-resolved emission measurements were performed with a spectrometer (Edinburgh Instruments, FLS980) equipped with two monochromators and a photomultiplier tube (PMT) detector (Hamamatsu, R928P). The samples were excited at 355 nm by the 3rd harmonic of a Q-switched Nd:YAG pulsed laser (Continuum, Minilite) coupled with the fluorescence spectrometer. To minimize scattered 6 ACS Paragon Plus Environment

Page 7 of 27

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


excitation light, the samples were tilted at 45ο in respect to the excitation, while a 395 nm long wavepass filter was additionally used in the emission arm. The time-resolved emission spectra were obtained in the instrument’s F980 software by data slicing the emission maps at 6 bins between 32-42 µs. Time decay curves were extracted by selecting a fixed wavelength from the PL maps.



3. Results and Discussion 4.1 Structural characterization

A)

300 310 320 330

Eg

Eg

0

B1g

A1g

B2g

100 200 300 400 500 600 700 800 900 -1

Raman Shift (cm )

300 310 320 330

1.0 0.8 0.6 0.4 0.2 0.0 120

130

140

150

160

C) Intensity (Arb. Units)

B) 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

Page 8 of 27

1.0

000 300 500

0.8 0.6 0.4 0.2 0.0

170

120

130

140

150

160

170

-1

-1

Raman Shift (cm )

Raman Shift (cm )

Fig. 1: A) Normalized Raman spectra of TiO2 films doped with Cu and N using 3 copper pieces and different nitrogen fluxes (0, 10, 20 and 30 sccm). All the spectra present the typical five Raman peaks of Anatase TiO2. B) Magnification showing the shift of the main peak of Anatase TiO2 (Eg at 144 cm-1) towards higher wavenumbers due to Cu and N codoping. C) Effect of Cu doping on the position of the main peak of Anatase TiO2.

Raman spectra of all the samples, shown in Fig. 1A, present five characteristic peaks, at 144 cm-1, 197 cm-1, 399 cm-1, 513 cm-1 and 639 cm-1, which are the fingerprint of anatase TiO2 and are attributed to 8 ACS Paragon Plus Environment

Page 9 of 27

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 Eg, Eg, B1g, A1g, and B2g modes, respectively14. This indicates the anatase phase is maintained even after doping in thin films. Moreover, the presence of sharp peaks indicates a good crystalline quality. Fig. 1B and 1C show the normalized anatase main peak for samples with different concentration of N and Cu, respectively. It is observed that the peak broadens and shifts to higher wavenumbers as the dopant concentration increases. This behaviour can be ascribed either to the lowering of the crystallite size, thus affecting the force constants and the vibrational amplitudes of the nearest neighbour bonds15, or to internal strain fields induced by the dopants and affecting the Raman peak positions16.

4.2 SEM and EDXS characterization

Fig. 2: SEM micrographs of the sample 310 in a top-view (left) and cross section configuration (right), representative of the whole set of samples morphology. The scale is reported in the figures.

Fig. 2 shows, on the left panel, the surface morphology of the sample named 310 and, on the right panel, the cross-sectional SEM image of the same sample, where the film exhibits a nanometric columnar structure with a width lower than 50 nm. This morphology was detected for all the films regardless to the dopant types and concentrations.



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 27

Fig. 3 reports the results of EDXS/SEM characterization. The thickness of the films was measured from cross-sectional SEM images. Films deposited with the same number of Cu pieces but with different nitrogen fluxes present a lowering of thickness as the nitrogen flux is increased. This is expected because the presence of nitrogen changes both the gas pressure and the sputtering rate of the process. In fact, the plasma energy is dissipated in ionizing and accelerating Ar atoms, on one hand, and in dissociating N2 molecules on the other hand7. Thus, as the N2 ratio is increased in the carrier gas, an increasing part of the plasma energy available serves to dissociate N2 molecules, leaving thereby less energy to produce Ar ions which in turn translates into less efficient sputtering process. EDXS allowed to estimate the copper concentration [Cu] for all the samples, but not the nitrogen content (see Fig. 3). In particular, the Cu concentration is around 1% with 2 metal fragments and increases with the number of Cu pieces reaching 2.3 at.% in sample 400 and 6.8 at.% in 500. As can be clearly seen from Fig. 4, for fixed number of fragments, Cu concentration decreases when N is introduced. The highest variation was detected in the set with 5 copper pieces where Cu concentration goes from 6.8 at.% (sample 500) to 4.6 at.% (sample 530).


1200 1100 1000 900 800 700 600 500 400 300 200 100 0

Cu5

Cu4

Cu3

Cu2

Cu0

12 11 10 9 8 7 6 5 4 3 2 1 0

[Cu] [%]

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


Thickness [nm]

Page 11 of 27

520 500 420 400 320 300 220 200 020 000

Sample name Fig. 3: Left (black) axis refers to the thickness of the samples whose label is reported on the X-axis. Right (red) axis reports the copper concentration as estimated from EDXS. The error bars for the thicknesses are smaller than the symbols used. All the films were deposited by keeping constant the deposition time.

4.3 XPS characterization The chemical states of the elements on the surface of the samples were examined by x-ray photoelectron spectroscopy (XPS). Pure TiO2 film display two peaks centred at 458.5 eV and 464.2 eV(not shown here) assigned to Ti 2p3/2 and Ti 2p1/2 core levels of Ti4+, respectively. These two peaks shift positively by 0.2 eV with Cu or/and N doping, thus indicating a variation of Ti4+ chemical environment in the matrix due to the incorporation of doping species. A small peak at 457 eV indicates the presence of Ti3+, probably as a defect at the surface of the nano-columnar structure. The O 1s core level of all samples present a broad peak that can be de-convoluted into two peaks: a small peak



centred at 531.4 eV, attributed to O-H bond, and the main peak centred at 529.7 eV (for pure TiO2), related to the O-Ti bond inside TiO2. This peak shifts by 0.2 eV in Cu/N doped and co-doped samples.

Cu 2p

520

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

Page 12 of 27

320

300

955

950

945

940

935

930

Binding Energy (eV) Fig. 4: XPS spectra of Cu 2p core levels of Cu3-TiO2, Cu3-N20-TiO2, and Cu5-N20-TiO2.

Samples with low Cu concentrations (ca. 2 at.%) show two peaks centred at 932.3 eV and 952.0 eV (see Fig. 4), which are attributed to Cu 2p3/2 and Cu 2p1/2 core levels, respectively, in Cu2O (Cu+)17, 18. N incorporation does not alter Cu XPS features, as can be inferred comparing 300 and 320 curves in Fig. 5. Both signals show a faint contribution at higher energies (934.2 and 954.3 eV), which become very intense in films with higher Cu content (about 5 at.%), as can be observed in the spectrum corresponding to sample 520. The peak at 934.2 eV is attributed to Cu 2p3/2 and the one at 954.3 eV to Cu 2p1/2 in CuO (Cu2+) [18]. In addition, Cu2+ has three Cu 2p3/2 satellite peaks at 941.9, 943.0, and 944.8 eV19, giving rise to the broad feature between 940.0 and 945.0 eV. This analysis evidences that at low concentrations, the copper is present mainly in Cu+ state while the concentration increases, the amount of Cu2+ species increases. In particular, by means of convolution analyses we evaluated a distribution of 94% for Cu+ and 6% for Cu2+ in sample 300 while in sample 520 the two values are 32% and 51%, respectively. 12 ACS Paragon Plus Environment

Page 13 of 27

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


Fig. 5: XPS spectra of N 1s core levels of N10-TiO2, N20-TiO2, N30-TiO2, and Cu5-N20-TiO2.

Fig. 5 shows the XPS spectra of N 1s core levels of selected samples. A different nature of nitrogen doping was found for different nitrogen contents. At low nitrogen concentrations, only one peak is observed, at B.E. ~ 399.8 eV (Fig. 5 sample 010). This peak can be assigned to N substitutionally replacing oxygen to form O-Ti-N linkage. With increasing the nitrogen doping, a second peak for N1s peak at 396 eV (Fig. 5 sample 020) assigned to Ti-N bond is also observed. With further increasing the N content (Fig. 5 sample 030), a third peak arises at 402 eV, due to interstitial occupation of N and oxidized N species (NO) to form Ti-O-N and Ti-O-N-O linkages. These three peaks of nitrogen are



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 27

present in all co-doped samples, suggesting that Nitrogen is present both as interstitial and substitutional dopant (Fig. 5 sample 520). XPS analysis was used also to quantify the concentration of the dopants. Copper concentrations obtained by XPS are in good agreement with those given by EDXS analysis. In addition, XPS allowed the quantification of the nitrogen concentration in the samples. The nitrogen content increases with the nitrogen flux, giving 0.7 N at.% for 10 sccm, 1.1 N at.% for 20 sccm, and 1.3 N at.% for 30 sccm. It is known that the nitrogen incorporation efficiency depends on the flow rate ratio of Ar and nitrogen. Three N incorporation regimes were reported in literature7. For very low N2/Ar flux ratios, the amount of nitrogen in the plasma is low and as a consequence also the number of dissociated N atoms. As the nitrogen flux is increased, more and more molecules become available, and the probability of N2 dissociation increases. In this case a clear dependence of the nitrogen concentration from N2/Ar ratio was observed. In general, if the ratio is further increased a saturation regime is observed, since the increment of N2 flux decreases the deposition rate but does not change significantly the incorporation rate of N. We are very near to the saturation regime, as confirmed by the weak dependence of the nitrogen content in the films on the nitrogen flux increase and by the decrement of the films thickness discussed in the previous section.


Page 15 of 27

4.4 Optical characterization 10

000

a)

Absorbance (Arb.Units)


10

1

0.1

0.01

000 010 020 030

b) 1

0.1

0.01 300

400

500

600

700

800

300

400

500

600

700

800

Wavelength (nm)

Wavelength (nm)

10

10

300 310 320 330

c) 1



0.1

500 510 520 530

d) 1

0.1

0.01

0.01 300

400

500

600

700

300

800

Wavelength (nm)

400

500

600

700

800

Wavelength (nm)

Cu=0 N=0 Cu=3 N=0 Cu=5 N=0

10

Absorbance (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


e) 1

0.1

0.01 200

300

400

500

600

700

800

Wavelength (nm)

Fig. 6:Figure a) reports the absorbance of pure TiO2. Figure b) shows the absorbance versus the nitrogen concentration. c) Refers to a [Cu] = 1.5 at.% and panel d) to the highest copper concentration and with varying [N]. Panel e) refers to the variation of absorption with copper content. Cu=0 N=0 is the sample 000, Cu=3 N=0 is the sample 300 while Cu= 5 N=0 is the sample 500. The film thicknesses are nominally identical to those reported in Fig. 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

Page 16 of 27

In Fig. 6 are shown the optical absorbance spectra of a pure TiO2 deposited thin film and of doped samples with different concentrations of Cu. As can be observed, the films are transparent, exhibiting fringes related to multiple interface reflections. At high copper concentrations, the sample absorbs in the visible region between 400 and 600 nm, as evidenced by the fringes damping. When the copper dopant is concentrated enough, the nitrogen content affects significantly the optical properties of the titania films. It seems that a good co-doping region falls around [Cu] ≈ 5 at.% and [N]≈1 at.%. This effect could be related to the co-doping of copper and nitrogen, which in turn affects the bandgap region in such a way that a relatively wide defect-related visible absorption appears where usually TiO2 is optically transparent13. Fig. 7 shows the photoluminescence (PL) spectra of thin films grown on silicon substrates with 0%, 1%, 1.7% and 6-8% of copper, at different N concentrations. The typical wide photoluminescence spectrum of anatase titania has been recorded and we observe the superimposition of the interference fringes due to multiple reflections at the film interfaces of the PL emission. The observed PL envelope is compatible for its width as well as for its lineshape with radiative recombination of self-trapped excitons and impurity trapped excitons20.


Page 17 of 27

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


Fig. 7:Panel a)-d) refer respectively to [Cu] = 0, 1, 1.7, 6 at.%. The insertion of nitrogen quenches the PL in the region 400650 nm. The 532 nm peak is the laser residual transmitted through the UV filter.

In pure TiO2 and in samples doped only with Cu, the PL spectrum is constituted by a very broad feature centered at around 500 nm, whose intensity lowers by increasing the dopant concentration. The insertion of nitrogen in the structure quenches the main emission band and in particular when it is interstitially present, its effect on the film crystalline structure reflects detrimentally on its light emission properties, leaving a faint component in the NIR region, around 750-800 nm. From these results, it appears that Cu and N play two different roles in the network. In particular, copper is dispersed in the matrix in such a way to leave an anatase connected structure allowing the exciton formation and radiative recombination, giving rise to the characteristic PL feature. On the other hand, nitrogen more heavily affects the oxide structure, through Ti-N linkages, in such a way to quench the 17 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

Page 18 of 27

TiO2 exciton luminescence through charge trapping or energy transfer mechanisms. Quenching effects from Ti-O-N and Ti-O-N-O too, cannot be excluded at this stage of the results. It is important to notice the cooperative effect of Cu and N co-doping: when both dopants are present, the photoluminescence suppression is more effective. We also think that copper generates Ti3+ -bound traps influencing the photoluminescence yield and lifetime. Time-resolved PL, performed on the samples grown on silicon (Fig. 8), gave interesting information about the carrier recombination dynamics. In particular it can be seen that pure TiO2 shows two different PL centers. At around 400-450 nm a typical single exponential decay, with a recombination lifetime of 93±6µs and ascribed to the exciton recombination, can be observed. For longer wavelengths, the decay becomes more and more complex, and at around 600 nm the temporal law which better explains the photoluminescence decay is the so-called stretched exponential decay

= () , where I0 is the photoluminescence intensity at time t = 0 sec, τ the time decay parameter and β a pure number between 0 and 1. So, at higher wavelengths, the PL components are related to intrinsic TiO2 defects, probably present at the nano-column surfaces21. In 530 doped sample, again the single exponential decay with 96±6 µs characteristics lifetime is present only close to the bandgap spectrum region, whilst at lower energies the decay curve again becomes a stretched exponential (see Fig. 8), indicating a recombination path involving carrier diffusion/hopping through the defects matrix before their radiative or non-radiative recombination. In particular, sample 510 shows no bandgap emission and the recombination is purely a stretched one, with a lifetime of 3.8±0.6µs and β=0.5±0.1, corresponding to a mean relaxation time =7.6±0.6µs22. It is clear that the increasing concentration of centers related to the doping elements creates nonradiative recombination-centers hindering the emission feature of pure titania and leaving the components related to network defects. Fig. 8 gives a glance of such a behavior.


Page 19 of 27

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


We think that the quite long lifetimes observed, which are very promising for photocatalytic applications and are –as far as we know- the longest in literature, are probably due to an increased (>0.4 eV) surface barrier height due to the nanometric TiO2 columnar (see Fig. 2-right) architecture23. Actually, in 23 it is proposed a model for nano-sized TiO2 structures where the carrier lifetime τ depends from the potential barrier height V experienced at the semiconductor surface via the relation =

where is a characteristic lifetime depending on the thermal velocity of the photo-excited

carriers and the density of the surface recombination centers, while expresses a deviation from the ideal thermionic emission of carriers. K is the Boltzmann constant and T is the temperature. To the extent that the photocarrier lifetime contribute to photocatalysis efficiency the surface potential barrier control could indicate a viable route to enhance the photocatalysis of titania, and we are programming future temperature resolved photoluminescence measurements in order to deepen such aspect. The lifetimes observed in this work are compatible with those observed by 24,25, in particular when they used high-quality (colorless) anatase crystals and approached 300 K, but it has to be taken into account that in literature carriers lifetimes in TiO2 present values ranging from nanoseconds to microseconds, due to the different microstructures26-30 allowing different paths to photocarrier recombination and affecting the lifetimes of charge-layers on the surface. An overview of physical chemistry topics related to this work can be found in 31-36



1000

000 τ=7.3±0.2µs β=0.64±0.02 530 τ=7.2±0.3µs β=0.59±0.03 510 τ=7.2±0.2µs β=0.60±0.02 at 600nm

100

PL 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

Page 20 of 27

10

1 20

1000

40

60

80

100

120

140

100 000 τ=93±6µs 530 τ=96±6µs 510 τ=3.8±0.6µs β=0.5±0.1 at 410nm

10

1 0

20

40

60

80

Time (µs)

100

120

140

Fig. 8:Top panel shows the photoluminescence decay for samples 000, 530 and 510 in the defects region around 600 nm. Bottom panel the decay in the bandgap recombination related region. The fitting curves are reported above the data and the results of the fits are in the figure legend.

Time resolved spectra, shown in Fig. 9, are reconstructed from decays measured across the PL of samples 000, 530 and 510 and. enlighten the different contributions of TiO2 matrix and defect related features. It can be seen that carrier recombination is completely quenched at t=0 µs in the Cu-N doped samples 530 and 510. On the contrary, 120 µs after the arrival of the excitation pulse the long exciton lifetimes at 410 nm observed in pure TiO2 are quenched in the Cu-N doped samples. It can be clearly seen also how the Cu-N co-doping, changes the film optical properties when the dopant concentration is high; in fact, the bandgap related recombinations are completely quenched in favor of defect ones. 20 ACS Paragon Plus Environment

Page 21 of 27

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 spectra are so-called time-resolved spectra and this means that they are spectral snapshots of the emission at fixed delays after the laser excitation pulse.

Fig. 9: The panels report, for the samples indicated in the insets, the time-resolved spectra at different delays. It can be clearly seen how the bandgap UV emission (the right part of it at above 400 nm) becomes dominating at wider time windows.

5 Conclusions

In this work we grew via RF magnetron sputtering Cu-N co-doped TiO2 thin films to study the effect of co-doping on the recombination dynamics of photoexcited charge carriers. The produced films have Cu concentrations ranging from 1% to 7%, while a maximum concentration of 1.3% has been achieved for N. Anatase thin film have been grown with a nano-columnar architecture, with a high surface to volume aspect ratio. XPS measurements evidenced that at low concentrations (ca. 2 at.%) Cu is present mainly as Cu+, while at higher concentrations the Cu2+ 21 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

Page 22 of 27

oxidation state prevails. At low nitrogen concentrations (about 0.7 at%), the only one observed peak is assigned to N substitutionally replacing oxygen to form O-Ti-N linkage. With increasing the nitrogen doping (about 1.1 at%) a second peak is observed and assigned to Ti-N bond. With further increasing of the N content (about 1.3 at%) a third peak arises due to interstitial occupation of N and oxidized N species (NO) to form Ti-O-N and Ti-O-N-O linkages. These three peaks of nitrogen are present in all co-doped samples, suggesting that Nitrogen is present both as interstitial and substitutional dopant. All the films are quite transparent in the visible range, with the exception of the samples with high Cu concentrations. Time resolved PL analysis evidenced that Cu-N co-doping hinders the exciton radiative recombination in the anatase network, giving rise to increase of the mean lifetime and increasing the recombination probability on defects at the nano-column surface which indicates that the best codoping region is where sample 510 falls (i.e. 10 sccm of nitrogen flux and with the use of 5 Cu pieces in the sputtering chamber).” We think that a control of the surface potential barrier and an increase of the N amount by means of suitable deposition conditions could be a viable route to further increase the carrier lifetime and enhance the photocatalytic properties of such systems.

Acknowledgments The work was financially supported by Provincia Autonoma di Trento in cooperation with IPCB (CNR) in the framework of the ENAM project.


Page 23 of 27

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


References 1.

Hanaor, D.A.H., C.C. Sorrell Review of the anatase to rutile phase transformation. J. Mater. Sci., 2011, 46, 855-874.

2.

Binetti, E., El Koura Z., Bazzanella N., Carotenuto G., Miotello A. Synthesis of mesoporous ITO/TiO2 electrodes for optoelectronics. Mater. Lett., 2015, 139, 355-358.

3.

Binetti, E., El Koura Z., Patel N., Dashora A., Miotello A. Rapid hydrogenation of amorphous TiO2 to produce efficient H-doped anatase for photocatalytic water splitting. Appl. Catal., A, 2015, 500, 69-73.

4.

Choi, W., Termin A., Hoffmann M. R. The role of metal ion dopants in quantum-sized TiO2: Correlation between photoreactivity and charge carrier recombination dynamics. J. Phys. Chem., 1994, 98, 1366913679.

5.

Wang, S., Meng K. K., Zhao L., Jiang Q., Lian J.S. Superhydrophilic Cu-doped TiO2 thin film for solardriven photocatalysis. Ceram. Int., 2014, 40, 5107-5110.

6.

Dozzi, M.V, Selli E. Doping TiO2 with p-block elements: Effects on photocatalytic activity. J. Photochem. Photobiol., C, 2013, 14, 13-28.

7.

Delegan, N., Daghrir R., Drogui P., El Khakani M. A. Bandgap tailoring of in-situ nitrogen-doped TiO2 sputtered films intended for electrophotocatalytic applications under solar light. J. Appl. Phys., 2014, 116, 153510-1 - 153510-8.

8.

El Koura, Z., Patel N., Edla R., Miotello A. Multilayer films of indium tin oxide/TiO2 codoped with vanadium and nitrogen for efficient photocatalytic water splitting. Int. J. Nanotechnol., 2014, 11, 10171027.

9.

Zhu, W., Qiu X., Iancu V., Chen X.-Q., Pan H., Wang W., Dimitrijevic N. M., Rajh T., Meyer III H. M., Paranthaman P., Stocks G. M., Weitering H. H., Gu B., Eres G., Zhang Z. Band gap narrowing of titanium oxide semiconductors by noncompensated anion-cation codoping for enhanced visible-light photoactivity. Phys. Rev. Lett., 2009, 103, 226401-1 – 226401-4.



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

10.

Page 24 of 27

Dashora, A., Patel N., Kothari D.C., Ahuja B.L., Miotello A. Formation of an intermediate band in the energy gap of TiO2 by Cu–N-codoping: First principles study and experimental evidence. Sol. Energy Mater. Sol. Cells, 2014, 125, 120-126.

11.

Wang, S., Yang X. J., Jiang Q., Lian J. S.. Enhanced optical absorption and photocatalytic activity of Cu/N-codoped TiO2 nanocrystals. Mater. Sci. Semicon. Process., 2014, 24, 247-253.

12.

Song, K., Xhou J., Bao J., Feng Y.. Photocatalytic activity of (copper, nitrogen)‐codoped titanium dioxide nanoparticles. J. Am. Ceram. Soc., 2008, 91, 1369-1371.

13.

Jaiswal, R., Barambe J., Patel N., Dashora A., Kothari D.C., Miotello A. Copper and Nitrogen co-doped TiO2 photocatalyst with enhanced optical absorption and catalytic activity Appl. Catal., B, 2015, 168– 169, 333-341.

14.

Ohsaka, T. Temperature dependence of the Raman spectrum in Anatase TiO2. J. Phys. Soc. Jpn., 1980, 48, 1661-1668.

15.

Choi, H.C., Jung Y. M., Kim S.B. Size effects in the Raman spectra of TiO2 nanoparticles. Vib. Spectrosc., 2005, 37, 33-38.

16.

Bianco, F., Fedus K., Enrichi F., Pierobon R., Cazzanelli M., Ghulinyan M., Pavesi L. Two-dimensional micro-Raman mapping of stress and strain distributions in strained silicon waveguides. Semicond. Sci. Technol., 2012, 27, 085009-1 - 085009-8.

17.

Ertl, G., Hierl R., Knoezinger H., Thiele N., Urbach H.P. XPS study of copper aluminate catalysts. Appl. Surf. Sci., 1980, 5, 49-64.

18.

Jolley, J.G., Geesey G.G., Hankins M.R., Wright R.B., Wichlacz P.L. Auger electron and X-ray photoelectron spectroscopic study of the biocorrosion of copper by alginic acid polysaccharide. Appl. Surf. Sci., 1989, 37, 469-480.

19.

Parmigiani, F., Pacchioni G., Illas F., Bagus P.S. Studies of the CuO bond in cupric oxide by X-ray photoelectron spectroscopy and ab initio electronic structure models. J. Electron Spectrosc. Relat. Phenom., 1992, 59, 255-269.


Page 25 of 27

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

20.


Tang, H., Berger H., Schmid P.E., Levy P. Photoluminescence in TiO2 anatase single crystals. Solid State Commun., 1993, 87, 847-850.

21.

Vial, J., Bsiesy A., Gaspard F., Herino R., Ligeon M., Muller F., Romestain M., Macfarlane M.M. Mechanisms of visible-light emission from electro-oxidized porous silicon. Phys. Rev. B, 1992, 45, 14171-14176.

22.

Gradshteyn, I., Ryzhik I. Table of integrals, series, and products: corrected and enlarged edition. Academic Press: San Diego, U.S., 1980.

23.

Ozawa, K., Emori M., Yamamoto S., Yukawa R., Yamamoto S., Hobara R., Fujikawa K., Sakama H., Matsuda I. Electron–Hole Recombination Time at TiO2 Single-Crystal Surfaces: Influence of Surface Band Bending. J. Phys. Chem. Lett., 2014, 5, 1953-1957.

24.

Wakabayashi, K., Yamaguchi Y., Sekiya T., Kurita S. Time-resolved luminescence spectra in colorless anatase TiO2 single crystal. J. Lumin., 2005, 112, 50-53.

25.

Sekiya, T., Tasaki M., Wakabayashi, K., Kurita S. Relaxation process in anatase TiO2 single crystals with different colors. J. Lumin., 2004, 108, 69-73.

26.

Xu, M., Gao Y., Martinez Moreno E., Kunst M., Muhler M., Wang Y., Idriss H., Woell C. Photocatalytic activity of bulk TiO2 anatase and rutile single crystals using infrared absorption spectroscopy. Phys. Rev. Lett., 2011, 106, 138302-1 – 138302-4.

27.

Yamada, Y. and Kanemitsu Y. Determination of electron and hole lifetimes of rutile and anatase TiO2 single crystals. Appl. Phys. Lett., 2012, 101, 133907-1 – 133907-4.

28.

Bahnemann, D.W., Hilgendorff M., Memming R. Charge carrier dynamics at TiO2 particles: reactivity of free and trapped holes. J. Phys. Chem. B, 1997, 101, 4265-4275.

29.

Cavigli, L., Bogani F., Vinattieri A., Faso V., Baldi G. Volume versus surface-mediated recombination in anatase TiO2 nanoparticles. J. Appl. Phys., 2009, 106, 053516-1 – 053516-8.

30.

Fujihara, K., Izumi S., Ohno T., Matsumura M. Time-resolved photoluminescence of particulate TiO2 photocatalysts suspended in aqueous solutions. J. Photochem. Photobiol., A, 2000, 132, 99-104.



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

31.

Page 26 of 27

Jagadale T. C., Takale S.P., Sonawane R.S., Joshi H.M., Patil S.I., Kale B.B., Ogale S.B. N-Doped TiO2 Nanoparticle Based Visible Light Photocatalyst by Modified Peroxide Sol-Gel Method, J. Phys. Chem. C, 2008, 112, 14595-14602

32.

Liu Y., Zhou W., Liang Y., Cui W., Wu P. Tailoring Band Structure of TiO2 To Enhance Photoelectrochemical Activity by Codoping S and Mg, J. Phys. Chem. C, 2015, 119, 11557-11562

33.

Marquez A. M., Plata J.J., Ortega Y., Sanz J.F. Making Photo-selective TiO2 Materials by Cation−Anion Codoping: From Structure and Electronic Properties to Photoactivity, J. Phys. Chem. C, 2012, 116, 18759-18767

34.

Patel N., Dashora A., Jaiswal R., Fernandes R., Yadav M., Kothari D.C., Ahuja B.L., Miotello A. Experimental and Theoretical Investigations on the Activity and Stability of Substitutional and Interstitial Boron in TiO2 Photocatalyst, J. Phys. Chem. C, 2015, 119, 18581-18590

35.

Tobaldi D. M., Rozman N., Leoni M., Seabra M.P., Skapin A.S., Pullar R.C., Labrincha J.A. Cu−TiO2 Hybrid Nanoparticles Exhibiting Tunable Photochromic Behavior, J. Phys. Chem. C, 2015, 119, 2365823668

36.

Yamanaka K., Ohwaki T., Morikawa T. Charge-Carrier Dynamics in Cu- or Fe-Loaded Nitrogen-Doped TiO2 Powder Studied by Femtosecond Diffuse Reflectance Spectroscopy, J. Phys. Chem. C, 2013, 117, 16448-16456


Page 27 of 27

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


TOC Image


Synthesis and Characterization of Cu and N Codoped RF-Sputtered

Recommend Documents