Photosensitization of nanostructured TiO 2 electrodes with CdSe quntum dots: effects of microstructure in substrates Q. Shen 1,2) and T. Toyoda 1,2) Department of Applied Physics and Chemistry 1), and Course of Coherent Optical Science 2) The University of Electro- Communications Chofu, Tokyo 182-8585, Japan
BACKGROUNDS Nanocrystalline TiO 2 has received much attention in numerous fields of applications (e.g., photocatalysis, photoelectrochemical solar cell and gas sensor). Dye-Sensitized Solar Cell (DSSC) with high energy conversion efficiency exceeding 10 has been developed by (a) preparation of highly porous, nanostructured TiO 2 electrodes to increase the surface area; (b) dye or semiconductor quantum dot sensitization to extend the photoresponse of TiO 2 to the visible region. For higher efficiency, further research on the nanostructured TiO 2 electrodes (morphology, optical absorption, charge injection, transport and recombination) is essential.
RESEARCH OBJECTIVES TiO 2 electrodes Anatase-type TiO 2 nanoparticles made from TiCl 4 hydrolysis and oxidation processes. Composition of different size anatase-type TiO 2 nanoparticles. Composition with rutile-type content on anatase-type TiO 2 nanoparticles. TiO 2 nanotubes and nanowires. TiO 2 photonic crystals. Sensitizer CdSe quantum dots.
Advantages of semiconductor quantum dots Quantum confinement allows for energy gap tunable across the solar spectrum. Large extinction coefficient resulting from quantum confinement. Large intrinsic dipole moment which may lead to rapid charge separation. Robust inorganic nature. Impact ionization occurs with the possibility of high efficiency.
MOTIVATIONS In order to investigate morphological dependence, two types of nanostructured TiO 2 electrodes were prepared and CdSe quantum dots with various sizes were adsorbed on TiO 2 as sensitizer. TiO 2 electrodes were characterized using SEM, XRD, photoacoustic (PA) and photoelectrochemical current (PEC) methods. PA method is powerful for the investigation of optical absorption of optically opaque or scattering solid materials.
TiO 2 electrode preparation Nanocrystalline TiO 2 powder (30 wt) Pure water Acetylacetone 10 wt%) Polyethylene glycolpeg) (MW:500000) (40 w vs TiO 2 ) A ( 15 nm) B ( 27 nm) TiO 2 paste Stirring for one hour Depositing TiO 2 paste on FTO using squeegee method. Heating in air at 450 for 30 min. Two types of TiO 2 electrodes
Chemical solution deposition (CD) method of CdSe quantum dots on the TiO 2 electrode S. Gorer, G. Hodes, J. Phys. Chem. 98 (1994) 5338. 80mM CdSO 4 120mM N(CH 2 COONa) 3 (NTA) 80 mm Na 2 SeSO 3 mixed (1:1:1) Chemical Deposition (CD) Solution TiO 2 Solution For some Time Sample CD Solution 1h 3h 5h 9h 23h 49h
SEM micrographs of TiO 2 electrodes Electrode A Electrode B 500 nm Size of TiO 2 in the paste: 15 nm 500 nm Size of TiO 2 in the paste: 27 nm
SEM micrographs of CdSe-quantum dots adsorbed on TiO2 electrodes CdSe 5h CdSe CdSe 13h µm CdSe µm CdSe 60h CdSe 28h µm µm
X-Ray diffraction patterns Intensity (Arb. Units) CdSe (111) TiO 2 FTO FTO TiO 2 CdSe/ TiO 2 /FTO TiO 2 /FTO CdSe (220) TiO 2 CdSe (311) 25 30 35 40 45 50 2θ (Degrees)
Intensity (Arb. Units) CdSe/ TiO 2 /FTO (220) 40 41 42 43 44 45 2θ (Degrees) Scherrer equation: D = 0.9λ β cosφ D: CdSe quantum dot size β: half-width of the diffraction peak Φ: diffraction angle λ: 0.154 nm D5-6 nm (CdSe120h)
Photoacoustic Spectroscopy ( PAS ) Xe-Lamp Monochromator 050 0 Optical Lens PA Cell sample DATA Chopper Microphone Pre-Amplifier Computer Lock-in Amplifier Light source : 300W xenon arc lamp Normalization : carbon black sheet Wavelength range250nm800nm Modulation frequency : 33Hz
Wavelength (nm) 900 750 600 450 300 PA Intensity (Arb. Units) 1.5 1.0 0.5 0.0 E g E 1 A (TiO 2 : 15nm) bulk 120h 23h 9h 5h 0h 1.5 2.0 2.5 3.0 3.5 4.0 Photon Energy (ev) PA spectra of TiO 2 electrodes A (TiO 2 :15nm) deposited with CdSe quantum dots for various times.
Effective mass approximation : 2 h E = E1 Eg = ( D = 2a) 2 8µ a 10 Diameter (nm) 8 6 4 2 0 0 20 40 60 80 100 120 Deposition Time (Hours) Dependence of CdSe quantum dot size on deposition time.
PA Intensity (Arb. Units) 1.5 B (TiO 2 : 27nm) 1.0 0.5 0.0 120h 49h 23h 9h 5h 3h 0h Wavelength (nm) 900 750 600 450 300 1.5 2.0 2.5 3.0 3.5 4.0 Photon Energy (ev) PA spectra of TiO 2 electrodes B (TiO 2 :27nm) deposited with CdSe quantum dots for various times.
Photoelectrochemical Current (PEC) Spectroscopy Lock-in Amplifier Ammeter Computer Chopper Electrolyte Xe-Lamp 123 4 Sample Monochromator Optical Lens PEC Cell Pt Light source : 300W xenon arc lamp Normalization : carbon black sheet Material of the PEC cell quartz Wavelength range250nm800nm Electrolyte : 1M KCl + 0.1M Na 2 S
PEC Intensity (Arb. Units) 400 300 200 100 0 750 600 450 300 120h 49h 23h 9h 5h 3h 0h A (TiO 2 : 15nm) 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Photon Energy (ev) PEC spectra of TiO 2 electrodes A (TiO 2 :15nm) deposited with CdSe quantum dots for various times.
PEC Intensity (Arb. Units) 800 600 120h 49h 23h 9h 5h 3h 0h Wavelength (nm) 750 600 450 300 B (TiO 2 : 27nm) 400 200 0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Photon Energy (ev) PEC spectra of TiO 2 electrodes B (TiO 2 :27nm) deposited with CdSe quantum dots for various times.
IPCE: incident photon-to-current conversion efficiency for monochromatic radiation IPCE(%) = = injected electron numbers incident photon numbers 1240 (ev nm) photocurrent density ( µ A cm -2 wavelength (nm) photoflux ( µ W cm ) -2 )
IPCE (%) 28 24 20 16 12 8 4 0 CdSe120h CdSe49h CdSe23h CdSe9h CdSe5h CdSe3h CdSe0h Wavelength (nm) 800 600 400 A (TiO 2 : 15nm) 1.5 2.0 2.5 3.0 3.5 Photon Energy (ev) IPCE spectra of TiO 2 electrodes A (TiO 2 :15nm) deposited with CdSe quantum dots for various times.
IPCE (%) 28 24 20 16 12 8 4 0 Wavelength (nm) 800 600 400 CdSe120h B (TiO 2 : 27nm) CdSe120h CdSe49h CdSe23h CdSe9h CdSe5h CdSe3h CdSe0h 1.5 2.0 2.5 3.0 3.5 Photon Energy (ev) IPCE spectra of TiO 2 electrodes B (TiO 2 :27nm) deposited with CdSe quantum dots for various times.
Summary Two types of nanostructured TiO 2 electrodes A (15nm) and B (27nm) deposited with CdSe quantum dots by chemical deposition, have been characterized using PA, PEC spectra and IPCE measurements. Photosensitization by CdSe quantum dots was demonstrated and red shift of the spectra with increasing CdSe sizes can be clearly observed. PEC and IPCE spectra in the visible region are quite different for the two types of TiO 2 electrodes, even for the same deposition time of CdSe quantum dots.
The electron diffusion coefficient of the TiO 2 electrode A was found to be two times larger than that of electrode B using transient photocurrent responses. PEC and IPCE in the visible region in the CdSesensitized TiO 2 nanostructured electrodes largely depend on both the microstructure and electron transport property in the TiO 2 electrodes as well as the size and quantity of CdSe nanoparticles. Future studies: TiO 2 /CdSe interfacial property; Energy conversion efficiency; Ultrafast relaxation dynamics etc.
Ultrafast Carrier Dynamics of CdSe Quantum Dots Adsorbed on Nanostructured TiO 2 Films The ultrafast carrier dynamics were investigated by using lens-free heterodyne detection transient grating (LF-HD-TG) technique. One kind of optical pump and probe technique: Measurements of the decay of excited carrier densities after pulsed injection.
LF-HD HD-TG method Excitation of sample Probe Detection of signal Pump Signal Reference Detector Transmission grating Sample Heterodyne detection By approaching a sample to the grating, 1: Grating excitation 2: Heterodyne detection Easy TG method without lenses (No daily setup) K. Katayama et al., APL 82, 2775 (2003).
Lens-free heterodyne detection transient grating technique ( LF-HD-TG ) Titanium/sapphire laser Wavelength: 800 nm; Pulse width: 150 fs; Intensity: 5 µj/pulse SHG Probe beam Optical Delay Line Chopper Grating Sample Lock-in Amp. Advantages 1) Easy and compact. 2) High sensitivity. Measurements under low pump intensity. 3) Suitable for rough surfaces or optically scattering samples.
LF- HD- TG Signal (arb. units) 1.0 0.8 0.6 0.4 0.2 0.0 0 10 20 30 40 50 60 70 Delay Time (ps) LF-HD-TG response of nanostructured TiO 2 electrode without CdSe quantum dots.
LF- HD- TG Signal (arb. units) 1.0 5h 20h 0.8 44h 125h 0.6 0.4 0.2 0.0 0 10 20 30 40 50 60 70 Delay Time (ps) LF-HD-TG responses of nanostructured TiO 2 electrodes B (TiO 2 : 27 nm) with CdSe quantum dots for different deposition times.
Signal intensity (arb. units) 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Pump intensity 0 20 40 60 80 100 Time (ps) Pump intensity dependence of the LF-HD-TG response of nanostructured TiO 2 electrode B (TiO 2 : 27 nm) deposited with CdSe quantum dots for 49h. The pump intensity is ranging from 2.5 to 16.5 µj/pulse.
Signal intensity (arb. units) 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 10 20 30 40 50 60 70 Relative pump intensity (arb. units) The pump intensity dependence of the maximum signal intensity for nanostructured TiO 2 electrode B (TiO 2 : 27 nm) deposited with CdSe quantum dots for 49 h.
Analyses of LF-HD-TG response y = y 0 + A e 1 t / τ 1 t / + A 2 e τ 2 Transient Grating Signal 1.0 cdse20h fit t ing 0.5 y0 0.11043 }0.00715 A1 1.00703 }0.03463 τ 1 2.55526 }0.12042 A2 0.39276 }0.00834 τ 2 27.48046 }1.86301 0.0 0 10 20 30 40 50 60 70 Delay Time (ps)
Results of the Fitting Parameters CdSe deposition time (h) CdSe size (nm) τ 1 (ps) τ 2 (ps) A 1 A 2 y 0 5 4.8 1.8 16 0.74 0.34 0.15 9 5.1 2.1 19 0.68 0.37 0.16 20 5.8 2.1 20 0.65 0.31 0.20 44 6.5 2.2 29 0.63 0.28 0.18 125 6.7 1.9 30 0.60 0.31 0.27
tr: trapped et: electron transfer E g hυ e - tr tr et Eg: energy gap hυ: excitation energy nonradiative radiative excitation Energy h + (1) Fast decay process (τ 1 ): decrease of hole carrier numbers by trapping at the CdSe QD surface states; (2) Slow decay process (τ 2 ): photoexcited electron relaxation process, i.e. recombination and/or transfer to the TiO 2 electrode; (3) Base line (y 0 ): thermal diffusion signal owing to thermal grating.
Summary Nanostructured TiO 2 electrodes adsorbed with CdSe quantum dots by chemical deposition, have been characterized using LF-HD-TG measurements for the first time. Three decay processes can be shown with decay time of 2 ps (τ 1 ), 16-30 ps (τ 2 ) and larger than a few hundreds ps, respectively. It is found that τ 1 is independent of quantum dot size, but τ 2 depends on quantum dot size.
Acknowledgements Financial supports: D. Arae (now at Anritsu) M. Hayashi (now at CANON) I. Tsuboya (now at SONY) Y. Kumagai K. Katayama (University of Tokyo, now at MIT) T. Sawada (University of Tokyo, now at Tokyo University of Agriculture and Technology) A Grant-in Aid for Scientific Research (No. 14750645 and No. 15510098) and one on on Priority Area 417 (No. 15033224) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of the Japanese Government. The 21 st Century COE program on Coherent Optical Science.