Abstract 1. Introduction

Repsol Technology Center, Agustín de Betancourt, s/n, 28935 Móstoles, Madrid, Spain. *Contact authors: [email protected]; ... Some studies have proven t...
4 downloads 8 Views 6MB Size
Subscriber access provided by READING UNIV

Article 2

Hydrogenation and Structuration of TiO Nanorods Photoanodes: Doping Level and the Effect of Illumination in Trap-States Filling Carles Ros, Cristian Fabrega, Damián Monllor-Satoca, María Dolores HernándezAlonso, Germán Penelas-Pérez, Joan R Morante, and Teresa Andreu J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.7b12468 • Publication Date (Web): 23 Jan 2018 Downloaded from http://pubs.acs.org on January 25, 2018

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 34 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

Hydrogenation and Structuration of TiO2 Nanorods Photoanodes: Doping Level and the Effect of Illumination in Trap-states Filling. a

C. Ros, * C. Fàbrega, a T. Andreu * a.

b.

c.

d.

e.

a,b,c

D. Monllor-Satoca,

a,d

e

e

M.D. Hernández-Alonso, G. Penelas-Pérez, J. R. Morante

a,b

and

Catalonia Institute for Energy Research (IREC). Jardins de les Dones de Negre 1, 08930 Sant Adrià del Besòs, Barcelona, Spain. MIND-Departament of Electronics, Universitat de Barcelona (UB), c/Martí i Franquès 1, E-08028 Barcelona, Spain. 2 Institute of Nanoscience and Nanotechnology (IN UB), Universitat de Barcelona (UB), c/Martí i Franquès 1, E-08028 Barcelona, Spain. Department of Analytical and Applied Chemistry, IQS School of Engineering, Universitat Ramon Llull, Via Augusta 390, Barcelona, Spain. Repsol Technology Center, Agustín de Betancourt, s/n, 28935 Móstoles, Madrid, Spain.

*Contact authors: [email protected]; [email protected]

Abstract Both electronic properties and light absorption are key features in materials engineering to achieve efficient photoelectrodes for water splitting. Adjusting the potential drop inside the nanostructured semiconductor material through modification of the electronic properties is mandatory to drive efficiently the photogenerated charges. In this work, a hydrogen reduction treatment on titanium dioxide rutile nanorods based photoanodes has been performed in order to adjust the donor density to maximize electron-hole separation. Also, incident photon-to-current efficiency (IPCE) measurements and the effect of a light bias have been elucidated, finding relevant differences in low illumination conditions due to non-filled trap states. A physical model is proposed to show the role of the donor density on the overall performance of the photoanodes under study. The obtained productivity was enhanced by structuring the conductive glass substrates where TiO 2 nanorods were grown, resulting in a 20% increase of the photocurrent density.

1. Introduction In the 70s, Fujishima and Honda demonstrated the capability of TiO 2 to perform photoelectrochemical (PEC) water splitting, absorbing UV light and dissociating water into hydrogen and oxygen in a single step 1. Since then, much research has been done towards photoinduced water splitting 2–4, but PEC requirements are still so stringent that there is not yet any single cost-effective material capable of overcoming the 10% solar to hydrogen (STH) efficiency needed to make the technology marketable 4,5. Among these requirements, at least a 1.8 eV band gap material, no corrosion in aqueous environments and appropriate conduction band and valence band positions that straddle water oxidation and reduction potentials are conditions which have reduced the possible materials to a few 4,6. Among them, metal oxides such as TiO 27–11, Fe2O312–14, BiVO415–18 and WO319–21 have demonstrated to be some of the most reliable materials to be implemented in PEC systems, where the former one presents the highest stability and reduced cost 22. However, rutile titanium dioxide has a 3.0 eV band gap, limiting the maximum STH efficiency to a mere 2.2%23 (corresponding to 1.8 mA·cm -2), as only 5% of the solar spectrum is harvested 24. Nonetheless, TiO2 is still attractive because, thanks to its excellent performance as protective layer and good photoelectrochemical properties towards oxygen evolution reaction (OER), it is being used in the most advanced photoelectrodes recently developed 11,25–28. Among the plethora of nanostructured TiO 2 synthetic routes developed during the last years29, dense and vertically aligned rutile nanorods grown by hydrothermal methods on fluorinated tin oxide (FTO) coated glass substrates have proven to be one of the most successful process in terms of STH efficiency30. This configuration maximizes the active

ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

surface area and reduces the number of TiO 2/substrate interface recombination centers 31, as the substrate (i.e. fluorinated tin oxide, FTO) acts as a seed layer for an epitaxial growth. Furthermore, relying on the high temperature and pressure involved in the hydrothermal process, the crystallinity of the obtained TiO 2 is of extreme quality, which hinders the amount of defects along the nanorods. In an attempt to enhance the electronic and absorptive properties of TiO 2, scientists have tried to tune its band gap using additives or dopants such as Nb, W or Fe32–34, or by ammonia treatments11,35 but the results obtained so far have not been conclusive and their performance underwhelming. Other authors have also tried to modify electronic properties of titania by electroreducing the material 36. Some studies have proven that hydrogenated TiO 2 nanopowders shows an enhanced visible light absorption by adding disorder into the crystal lattice 2,37–40, thus creating intra-band gap energy levels41–43, however, the role of these levels has not been completely elucidated. In fact, the increased efficiency of hydrogenated TiO2 photoanodes have been attributed to the increment of charge carrier density by oxygen vacancies 44,45 and the introduction of surface states auspicious for the OER 38,46. In this work, we present an optimization of TiO 2 nanorods photoanodes growth process and heating treatment under a hydrogen reductive atmosphere of different duration and at different temperatures, together with substrate patterning for higher active area. A full photoelectrochemical characterization was carried out, paying special attention to the incident photon-to-current efficiency (IPCE) measurements and the effect of a light bias, which is analysed to be significantly influenced by the extra donors created by the hydrogenation treatment. Working under 1 sun light bias presented noteworthy differences, pointing out to the need of a light bias for IPCE measurements due to non-filled states in low light conditions. These findings should encourage other authors to study the need of a light bias in their experiments, or results not representative of the device’s working conditions may be obtained. Proper IPCE measurements are especially important for large band gap materials, as these measurements are independent of the used spectra and commercial light sources can have significant deviations from AM1.5G spectra in the UV range 47,48. Substrate structuration is known from photovoltaics49, but has been analysed for the first time in water splitting photoanodes up to our knowledge, and we report it’s enhancement of device’s productivity. According to the obtained results, we also proposed a model to explain the relation between the increase of charge carrier density, and the efficiency improvement of the treated photoanodes. Combining the strategies of optimization of the growth process, application of a reductive hydrogen treatment and substrate structuration, we were capable to fabricate a TiO 2 photoanode with photocurrents up to 1.2 mA·cm-2 and IPCE conversion efficiencies over 65%.

2. Materials and methods 2.1 Photoelectrodes fabrication TiO2 nanorods were grown by hydrothermal synthesis on fluorine doped tin oxide (FTO) conductive substrates of 20 cm 2 (4x5 cm2). Substrates were previously cleaned by sonication for 15 min with a deionized (DI) water, acetone and isopropanol (1:1:1) solution followed by abundant rinsing in DI water, and finally dried under a nitrogen stream. Electrical contacts were made by partially covering the substrates with a Teflon film (1x5 cm2) prior to TiO2 deposition. Teflon was chosen due to its resistance under acidic and moderate temperature conditions50. Teflon-covered FTO substrates were set at an inclination of 45 degrees with the conductive side facing down in a 250 ml Teflon cylindrical reaction chamber. A 100 ml solution was introduced, 18.5% in volume HCl with either 1.36 or 1.53 ml of titanium butoxide to achieve 40 and 45 mM precursor concentrations, respectively. The Teflon container was sealed into a stainless steel reactor (Parr Instruments Co.), heated up to 200 oC at 2 oC/min, maintained at

ACS Paragon Plus Environment

Page 2 of 34

Page 3 of 34 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

200 oC for 120 min, and cooled down to room temperature. Samples were then sonicated in DI water for 10 min and annealed at 500 oC during 120 min under air conditions to remove any residuals from the synthetic process 11. We grew two sets of samples using a 40 mM titanium butoxide precursor solution and two more with a 45 mM one. Higher concentrations created thicker TiO 2 layers that detached from the FTO substrate. These samples were then cut into 40 smaller samples (~1 x 2 cm2) with part of the FTO exposed for electrical contact. All samples were fully characterized and measured in a quartz PEC cell before being subjected to further treatments, in order to compare each sample before and after the modifications. Ten of the samples were annealed under a pure H 2 atmosphere in a sealed Alumina furnace, which was purged with Ar for 15 min, filled with H 2 for 15 min and heated at different temperatures ranging from 210 to 430 oC with 5 oC/min ramp rate and two different dwell times of 30 or 60 min. Samples were labelled as: TNR - precursor concentration (mM) - dwell time (min) temperature treatment ( oC) (i.e: TNR-40-30-320). To obtain the structured substrate, 500 nm of polycrystalline silicon was deposited by chemical vapor deposition (CVD) on a quartz wafer, followed by deposition of a photoresist exposed with a mask having a pattern of 10 µm in diameter circles separated 90 µm in a square configuration. A wet etching procedure with a HNO 3 and HF solution was used to transfer the pattern to the polysilicon. A second wet etching process with HF (40% in volume) for 24 min was conducted to pattern the quartz prior to FTO and TiO2 deposition. The structured substrate was covered with a 550 nm thick layer of FTO (10 Ω/square) by CVD. 2.2 Characterization TiO2 nanorods morphology was observed with a Zeiss Series Auriga field effect scanning electron microscope (FESEM). Structural characterization was carried out by X-ray diffraction (XRD) in a D8 Advance Bruker equipment with a Cu Kα radiation source working at 40 kV and 40 mA. The XRD spectrum has been normalized to the TiO 2 peak. X-ray photoelectron spectroscopy (XPS) measurements were recorded on a Perkin Elmer 5600ci spectrometer at a pressure lower than 10 -9 mbar, using a non-monochromatized Al Kα excitation source (1486.6 eV). Transmission spectrum was obtained with a Pekin-Elmer UV-VIS-NIR Lambda 950 spectrometer. The photoelectrical measurements were obtained with a Princeton Applied Research PARSTAT 2273 potentiostat using saturated Ag/AgCl/KCl as reference electrode and a platinum mesh as counter electrode. The quartz PEC cell was filled with 100 ml of 1 M NaOH electrolyte. Mott-Schottky measurements were collected with a Biologic VMP-300 potentiostat at a frequency of 500 Hz in dark conditions. Cyclic voltammograms and (at a scan rate of 20 mV/s), saturation, and Faradaic efficiency measurements were obtained using a 300 W Xenon Lamp with an AM 1.5 solar spectrum filter under the appropriate distance to receive 100 mW·cm-2 (1 sun). Faradaic efficiency is measured collecting the generated H 2 bubbles with and inverted burette and calculated as presented in supplementary information (S.I.). IPCE measurements were performed with an Abet 150 W Xenon Lamp coupled with an Oriel Cornerstone 260 monochromator, using light pulses of 100 s with 20 s dark, and a light spot smaller than the sample. Light biased IPCE measurements were performed with monochromatic light reaching the sample from the TiO 2 side and 1 sun illumination from the substrate side, as detailed in Fig. S.1. IPCE is calculated as: 𝐼𝑃𝐶𝐸(%) =

1.24·10−3 (𝑛𝑚×µ𝑊×𝐴−1 )×𝑗𝑝ℎ (𝐴) 𝜆(𝑛𝑚)×𝑃𝑙𝑎𝑚𝑝 (µ𝑊)

× 100

(1)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

where jph is the photocurrent generated by the monochromatic light, λ is the wavelength and Plamp is the light intensity absorbed by the sample coming from the monochromator. The half-cell solar-to-hydrogen (HC-STH)51 conversion efficiency is given by: 𝐻𝐶 − 𝑆𝑇𝐻 (%) =

𝑗𝑝ℎ ×(𝐸𝐻2𝑂/𝑂2 −𝐸) 𝑃𝑠𝑢𝑛

× 100

(2)

where 𝑗𝑝ℎ is the photocurrent density obtained under an applied bias (𝐸), 𝐸𝐻2𝑂/𝑂2 is the equilibrium redox potential of oxygen evolution reaction (1.23 V vs. NHE) and 𝑃𝑠𝑢𝑛 is the standard solar irradiation, 100 mW·cm-2.

3. Results and discussion 3.1 Characterization prior to hydrogen treatment SEM images of the samples grown from 40 mM and 45 mM precursor solution concentrations presented TiO 2 tetragonal nanorods over FTO substrates (Fig. 1). The increasing precursor concentration resulted in a longer (5 μm and 6 μm for TNT-40 and TNT45, respectively) and denser concentration of nanorods. In both cases we observed a dispersion in the nanorods diameter, from tens of nanometers up to few hundreds. Looking closer, the nanorods seem to be formed by a mesh of thinner square-section fibres. The nonverticality and disperse thickness can be attributed to the rough FTO substrate 30. The rutile phase is more favourable to grow on the FTO substrate due to lower lattice parameter mismatch (