DSpace at VNU: Application of nitrogen-doped TiO2 nano-tubes in dye-sensitized solar cells

Accepted Manuscript Title: Application of Nitrogen-doped TiO2 nano-tubes in Dye-sensitized solar cells Author: Vy Anh Tran Trieu Thinh Truong Thu Anh Pham Phan Trang Ngoc Nguyen Tuan Van Huynh Antonio Agresti Sara Pescetelli Tien Khoa Le Aldo Di Carlo Torben Lund So-Nhu Le Phuong Tuyet Nguyen PII: DOI: Reference: S0169-4332(16)32853-7 http://dx.doi.org/doi:10.1016/j.apsusc.2016.12.125 APSUSC 34672 To appear in: APSUSC Received date: Revised date: Accepted date: 21-10-2016 13-11-2016 15-12-2016 Please cite this article as: Vy Anh Tran, Trieu Thinh Truong, Thu Anh Pham Phan, Trang Ngoc Nguyen, Tuan Van Huynh, Antonio Agresti, Sara Pescetelli, Tien Khoa Le, Aldo Di Carlo, Torben Lund, So-Nhu Le, Phuong Tuyet Nguyen, Application of Nitrogen-doped TiO2 nano-tubes in Dye-sensitized solar cells, Applied Surface Science http://dx.doi.org/10.1016/j.apsusc.2016.12.125 This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain Application of Nitrogen-doped TiO2 nano-tubes in Dye-sensitized solar cells Vy Anh Trana, Trieu Thinh Truonga, Thu Anh Pham Phana, Trang Ngoc Nguyena, Tuan Van Huynhb, Antonio Agrestic, Sara Pescetellic, Tien Khoa Lea, Aldo Di Carloc, Torben Lundd, So-Nhu Lea, Phuong Tuyet Nguyena* a Faculty of Chemistry, University of Science, Vietnam National University, Ho Chi Minh City, Vietnam Faculty of Physics and Engineering Physics, University of Science, Vietnam National University, Ho Chi Minh City, Vietnam c CHOSE (Centre for Hybrid and Organic Solar Energy), Department of Electronic Engineering, University of Rome Tor Vergata, Italy d Department of Science and Environment, Roskilde University, DK-4000, Denmark b Graphical abstract Highlights      N-doped TiO2 nanotubes are synthesized by alkaline hydrothermal and reflux method Formation of TiO2 nanotube morphology and anatase phase is shown by TEM, XRD, Raman Coordination of NH4+ via Ti─O─NH4+ or H4N+─Ti─O bonds is indicated by FT-IR and XPS Blocking effect of N-doped toward electron transfer on TiO2 anode is studied by CV N-doped TiO2 improved DSC performance up to 30-40% mainly due to an increase in Jsc Abstract Our research aimed to improve the overall energy conversion efficiency of DSCs by applying nitrogen-doped TiO2 nano-tubes (N-TNT) for the preparation of DSCs photo-anodes The none-doped TiO2 nano-tubes (TNTs) were synthesized by alkaline hydrothermal treatment of Degussa P25 TiO2 particles in 10M NaOH The nanotubes were N-doped by reflux in various concentrations of NH4NO3 The effects of nitrogen doping on the structure, morphology, and crystallography of N-TNT were analyzed by transmission electron microscopy (TEM), infrared spectroscopy (IR), Raman spectroscopy, and X-ray photoelectron spectra (XPS) DSCs fabricated with doped N-TNT and TNT was characterized by J-V measurements Results showed that nitrogen doping significantly enhanced the efficiency of N-TNT cells, reaching the optimum value (η = 7.36%) with 2M nitrogen dopant, compared to η = 4.75% of TNT cells The high efficiency of the N-TNT cells was attributed to increased current density due to the reduction of dark current in the DSCs Keywords: Dye-sensitized solar cell, TiO2 nano-tubes, N-doped TiO2, TiO2 surface modification, dark current Introduction In the last decade, the dye-sensitized solar cell (DSC) has extensively been studied all over the world because of its significantly lower manufacturing cost in contrast to other solar devices such as silicon cells [1-3] However, the light to electricity conversion efficiency of DSCs are still too low compared with silicon solar cells and need to be improved [3] Dark current, which is the recombination of the injected electrons from the TiO2 semi- conductor to the oxidized dye or oxidized form of the mediator in the electrolyte, is one of factors limiting the efficiency of DSCs Dark current thus should be minimized in order for the performance of DSCs to improve Many groups have applied different methodologies to enhance the efficiency of DSCs, such as preparation of photo-anodes from various nano-structured materials (nano-particles, nano-wires, nano-fibers and nano-tubes) [4, 5] Another approach is to modify the TiO2 surface using metals or non-metal elements [6-8] C Kim et al.[6] reported that Cr-doped TiO2 photo-anodes decreased the dark current in DSCs They obtained an energy conversion efficiency of 8.4% in Cr-doped TiO2 DSCs incomparison with 7.1% in non-doped DSCs In 2010, M Grätzel et al.[7] modified TiO2 photo-anodes by doping it with Nb The DSCs with the dopant concentration of 0.5 M achieved an increase in the energy conversion efficiency from 7.4 to 8.1% They attributed this improvement to the retardation of the electron back reaction to tri-iodide Investigations by T Ma et al [9] have shown that the efficiencies of DSCs using N-doped TiO2 electrodes have been significantly improved by 44% and 17% compared to commercial TiO2 P25 and pure anatase TiO2 electrodes, respectively They used dry and wet methods with ammonia and nitrogen flow as nitrogen dopants to synthesize the nano-particles for the preparation of N-doped TiO2 electrodes Differently, H Y Kim et al.[5] used an electro-spinning process and a hydrothermal method to synthesize N-doped TiO2 nano-fibers with urea as nitrogen dopant Their N-doped TiO2 nano-fibers achieved an energy conversion efficiency of 4.7%, markedly higher than the efficiency of 1.6% obtained from DSCs prepared with none-doped nano-fibers Both T Ma and H Y Kim proposed that the improvement in photovoltaic performance was due to the enhanced electron injection, longer electron lifetime, and the retardation of the charge recombination However, both studies did not investigate the influence of various dopant concentrations and optimum conditions of product synthesis were not reported Moreover, the presence of doped Nitrogen in the TiO2 was not identified clearly Differently to the aforementioned work, we have introduced NH4NO3 as a new nitrogen dopant at various concentrations, ranging from to M, to synthesize N-doped TiO2 nano-tubes By combination of alkaline hydrothermal treatment of Degussa P25 TiO2 particles followed by reflux of the synthesized nano-tubes in aqueous solutions of NH4NO3, we have successfully synthesized N-doped nanotubes TiO2 according to the synthetic scheme illustrated in Scheme The morphology, TiO2 phases and N-doping of the nano-tubes after undergoing the synthetic steps in Figure were characterized by transmission electron microscopy (TEM), infrared spectroscopy (IR), Raman spectroscopy and X-ray photoelectron spectra (XPS) Cyclic voltammetry (CV) was used to measure the electrode kinetics between the N-doped TiO2 photo-anodes and 1,4-dicyanophthalene to determine the charge resistance and electron life times in the TNT and N-TNT DSC Our results showed that N-TNT significantly enhanced the performance of the DSCs Our data suggest that this improvement is due to a reduced dark current and longer electron life-time in the photo-anode TiO2 layer Optimum NH4NO3 dopant concentration of 2M was found to produce the highest performing DSCs Experimental section 2.1 Preparation of TNT and N-TNT powder The alkaline hydrothermal treatment was utilized to prepare the TNT About 2.0 g of Degussa P25 TiO2 were dispersed in sodium hydroxide solution (10M) and ultra-sonicated for 30 The nano-particles develop first into Titanate sheets followed by rolling into nano-tubes [10] The alkaline was transferred to a 70mL stainless steel autoclave container, tightly sealed, and heated at 140 0C for 18 h Finally, the mixture was filtered by filter papers and repeatedly washed with distilled water until the pH of the solution was The N-TNT and TNT were prepared by reflux method as follows The nano-tubes obtained from the first step were dispersed in an aqueous NH4NO3 solution at various concentrations of 0, 0.5, 1, 2, and M followed by hours reflux Finally, the solid was filtered by filter papers and repeatedly washed with distilled water until constant pH was achieved, and then was dried at 1000C for 24 h The obtained powders were labeled as follows: TNT powder (TO), N-doped TNT 0.5M (NTO-0.5), N-doped TNT 1M (NTO-1), N-doped TNT 2M (NTO-2), N-doped TNT 3M (NTO-3) 2.2 Preparation of TNT and N-TNT pastes A mixture of polyethylene glycol (PEG) (Merck, average MW of 200,000 g/mol), terpinol (Fluka), ethyl cellulose 10% (Fluka) and Triton X-405 (DOW) were added to 0.4 g of N-doped or none-doped TNT in absolute ethanol The solution was simultaneously stirred and ultra-sonicated for h, followed by stirring for 96 h Finally, the solvent was slowly evaporated by rotary evaporation The obtained pastes were named as follows: TNT paste (PTO), N-doped TNT 0.5M (PNTO-0.5), N-doped TNT 1M (PNTO-1), N-doped TNT 2M (PNTO2), N-doped TNT 3M (PNTO-3) 2.3 Characterization of the nano-tubes The surface structure and morphology of the TNT and N-TNT were studied by transmission electron microscopy (TEM, JEM-1400 (Japan)) The crystalline phases of the TNTs electrodes were analyzed by X-ray diffraction (XRD-d8 Advance) at a scanning rate of 2.250/min from 2θ = 200 to 800 To determine the presence of nitrogen, the Fourier transmission infrared spectroscopy (FTIR) measurements were performed in transmission mode (Bruker Tensor 37) The surface area of N-TNT was measured by a BET (Brunauer-EmmettTeller) analysis performed on a Quantachrome Instrument version 11.0 with an outgas time of h Raman spectra were performed using a Jobin-Yvon-Horiba micro-Raman system (LabRAM ARAMIS) equipped with Ar + ion laser (514 nm) as the excitation source The Horiba micro-spectrometer was coupled with a confocal microscope that allows the spatial resolution of the sample through a detector pinhole aperture The cut-off from the notch filters in the spectrometer is less than 100 cm -1 The laser light reached the sample surface at normal incidence by means of ultra-long working distance (50X) objective with 10.5 mm focal distance The scattered radiation was collected in a backscattering geometry [11] Subtraction of the fluorescence background on the Raman spectra was performed by a polynomial fitting, while spectral de-convolution was carried out by nonlinear least-squares fitting of the Raman peaks to a mixture of Lorentzian and Gaussian line shapes, providing the peak position, width, height, and integrated intensity of each Raman band The spectrometer is equipped with a diffraction grating of 1800 lines/mm coupled to a CCD camera To characterize the surface atomic composition and chemical environment of elements at the surface of TNT, Xray photoelectron spectra (XPS) was obtained by a Kratos Axis Ultra-DLD spectrometer (Kratos Analytical Ltd, UK) equipped with a hemispherical analyzer and a microfocussed (analysis area was 300 × 700 μm2) monochromatized radiation Al Kα line (1486.6 eV) operating at 150 W The spectrometer pass energy was set to 160 eV for survey spectrum and to 20 eV for high resolution spectra of the elements of interest Surface charging was minimized using an electron flood gun All the binding energies were referenced to the C 1s peak at 285.0 eV originating from the surface carbon contamination A nonlinear Shirley-type background [12] was used to analyse core peaks The quantification of surface composition was based on Scofield’s relative sensitivity factors [13] 2.4 Cyclic voltammetry The conductive glass plates (Pilkington – 8Ωcm-2) were cleaned with ethanol and were dried at 100 °C A layer of TNT or N-TNT paste with a circular surface area of 0.38 cm2 (d = 6.1 mm) was doctor bladed on the conducting side of the glass substrate The plates were sintered for 30 at 500°C Electrodes used for CV were carefully covered with Kapton polyimide tape (DuPont, USA) on the bare FTO areas surrounding the active electrode area The electrodes were used as working electrodes in a standard three-electrode electrochemical setup in combination with a VersaSTAT3-100 potentiostate (Princeton Applied Research) to obtain cyclic voltammograms of 1,4-dicyanonaphthalene (0.5 mM) in acetonitrile with tetrabutylammonium tetrafluoroborate (0.1 M) as supporting electrolyte [14] 2.5 Fabrication of dye-sensitized solar cells (DSC) The FTO conducting glass (Solaronix, sheet resistance: Ω/cm 2) was first cleaned in a detergent solution, and then rinsed with distilled water and ethanol The FTO glass plates were immersed in a 40 mM aqueous TiCl4 solution at 70oC for 30 and washed with distilled water The prepared TiO2 paste was coated onto the FTO glass with the area of 0.25 cm2 by doctor blading Then the thin film was sintered at 500oC for 30 After that, the TiO2 electrode was immersed into a dye solution consisting of 0.5 mM ruthenium 535 bis-TBA (N719, Solaronix) in methanol at room temperature for 24 h The dye-adsorbed TiO2 electrodes were rinsed with methanol and dried in an argon flow A counter electrode was formed on the opposite glass plate prepared by spreading Pt-catalyst T/SP (Solaronix) by doctor blading The counter electrode was sintered 30 at 450 oC The two glass substrates were assembled into a cell using the polymeric sealant Surlyn ® 1707 (DupontTM) Finally, the cell was filled with the electrolyte comprised of 0.05 M I2, 0.1 M LiI, 0.6 M tetrabutylammonium iodide, and 0.5 M 4-tertbutylpyridine in 3-methoxypropionitrile, and the filling holes were sealed with Surlyn® Each experiment produced to DSCs 2.6 J-V characterization of DSC Photovoltaic measurements of DSCs were performed under the solar simulator AM 1.5 (100 mWcm -2) using a 450-W xenon lamp and a power source calibrated with amorphous-Si standard DSCs were masked with a 0.25 cm2 active area of the anode electrode The current–voltage (J–V) characteristic of the cell was analyzed using a model 2400 digital source meter (Keithley, USA) The method includes applying an external bias under the given illumination conditions and measuring the generated photocurrent Results and discussions 3.1 Morphologies of TNT and N-TNT The morphologies of the TNT, N-TNT powder and paste were characterized by transmission electron microscopy (TEM) As shown in Figure 1, all the nanocrystal TNT, N-TNT tubes before and after paste fabrication are of equal size with diameters of 7-10 nm As seen from Figure 1, the concentration of the NH4NO3 dopant did not influence the diameters of the nano-tubes The lengths of the N-doped tubes, however(Figure 1b, 1c and 1d) had nano-tubes with shorter length compared to the TNT, most likely resulted from the refluxing process The PNTO-2 sample consists of TiO2 in its anatase phase as described later, and the BET surface area was found to be 496 m2g-1 The surface area of PNTO-2 was higher than normal surface area of TiO2 nano-tubes of 300 – 350 m2g-1 [15] 3.2 FTIR analysis The FTIR spectra of TO and NTO-2 are showed on Figure Consistent with previous reports of other groups, our result showed the stretching vibration of the hydroxyl groups from 3000 to 3600 cm-1 , the bending vibration of adsorbed H2O molecules 1600 cm-1 [5, 9, 16] The spectrum of NTO-2 showed a strong peak at 1400cm-1 which may be assigned to the NH4+dopant ion [17] while the peak is absent in TO These data clearly showed that the nano-tubes were nitrogen doped by the NH4NO3 reflux treatment by coordination of NH4+ ion on the surface of TiO2 3.3 Raman analysis Raman spectroscopy was used to investigate the phase transformation of the titanium dioxide after the hydrothermal base and NH4NO3 treatments shown in Scheme and to verify the effect of nitrogen doping after the calcination process As expected, Raman spectrum acquired on calcinated undoped PTO sample (Figure 3A) greatly differs from the pure anatase spectrum carried out on calcinated doped samples (Figure 3B) In particular, Raman spectrum of the calcinated PTO sample (Figure 3A) showed nine bands that can be attributed to sodium trititanate, Na2Ti3O7, as the major phase for temperatures up to 500°C [16] The band at 143 cm-1, 195 cm-1 and 380 cm-1 can be confidently ascribed to the Eg(1), Eg(2) and B1g(1) vibrations of TiO2 anatase [18] The presence of anatase bands is not expected for pure Na 2Ti3O7 nanotubes and can be explained by considering the effect of washing steps with deionized water and/or the effect of refluxing process in pure water after the hydrothermal treatment In fact, as detailed by E Morgado Jr and co-.worker [19], titanate nanotubes synthesized by hydrothermal treatment (TiO2 powder dispersed in 10 M sodium hydroxide solution) consisted of a trititanate structure with general formula Na xH2−xTi3O7·nH2O, where 0