- Nano Express
- Open Access
Nanostructured titania films sensitized by quantum dot chalcogenides
Nanoscale Research Lettersvolume 6, Article number: 266 (2011)
The optical and structural properties of cadmium and lead sulfide nanocrystals deposited on mesoporous TiO2 substrates via the successive ionic layer adsorption and reaction method were comparatively investigated by reflectance, transmittance, micro-Raman and photoluminescence measurements. Enhanced interfacial electron transfer is evidenced upon direct growth of both CdS and PbS on TiO2 through the marked quenching of their excitonic emission. The optical absorbance of CdS/TiO2 can be tuned over a narrow spectral range. On the other side PbS/TiO2 exhibits a remarkable band gap tunability extending from the visible to the near infrared range, due to the distinct quantum size effects of PbS quantum dots. However, PbS/TiO2 suffers from severe degradation upon air exposure. Degradation effects are much less pronounced for CdS/TiO2 that is appreciably more stable, though it degrades readily upon visible light illumination.
In recent years, nanostructured materials and quantum dots (QDs) light harvesting assemblies have emerged as highly promising building blocks for the development of and third generation solar cells affording efficient conversion of solar energy to electricity. Among different technologies, dye sensitized solar cells (DSCs)  hold great promise as an alternative renewable energy system with the advantages of low cost, transparency and flexibility . DSCs make use of nanocrystalline semiconducting electrodes (the most common being TiO2) sensitized with molecular dyes (the most efficient being polypyridyl ruthenium(II) complexes) in order to harvest solar light. In contrast to conventional p-n type devices, charge separation in DSCs takes place at the photoelectrode/sensitizer interface via electron injection from the dye into the conduction band of the semiconductor, followed by diffusive electron transport through the interpenetrated mesoporous network of the TiO2 semiconductor to the charge collector, while dye regeneration occurs via a redox electrolyte. Even though such devices have reached high performance and stability standards , the prospect of developing inorganic hybrid heterojunctions with enhanced selectivity, efficiency and robustness offering cost reduction and simplification in the DSCs manufacturing is attracting a great deal of attention.
One of the most attractive approaches for the utilization of inorganic heterojunctions in DSCs is the exploitation of the exceptional electronic properties of chalcogenide such as CdS, CdSe, PbSe, PbS and CdTe nanocrystals as light harvesting antennas [4–6]. Based on the unique quantum confinement effects, QDs offer unique high extinction coefficients and band gap tunability from the visible to the infrared spectral range by size control. Moreover, they can form favourable QDs/TiO2 as well as QDs/dye/TiO2 heterojunctions for efficient charge extraction [7–11]. A major drawback underlying the relatively low light harvesting ability and the concomitant reduced photocurrents in quantum dot sensitized solar cell devices is the amount of QDs adsorbed on the TiO2 electrode. Two main approaches have been so far exploited for the sensitization by QDs: in situ growth of QDs on TiO2 by chemical bath deposition (CBD) [7, 12] and successive ionic layer adsorption and reaction (SILAR) [13, 14] or attachment of preformed colloidal QDs to the TiO2 mesoporous structure by means of bifunctional linker molecules or direct adsorption using a suitable solvent in the colloidal solution [8, 11]. Linker-assisted and direct QD adsorption onto TiO2 allows fine control of the QD size, exploiting colloidal synthesis. However these systems suffer from rather low QD loading and relatively weaker electronic coupling between QDs and TiO2. On the other hand, CBD permits enhanced electron transfer to the wide band gap TiO2 electrode and significantly higher loading at the cost of appreciable QD aggregation that finally deteriorates solar cell performance [5, 6]. On the contrary, direct growth of QDs by SILAR has recently emerged as a promising deposition route combining high QD loading together with low degree of aggregation and efficient electron transfer to TiO2[14, 15].
In this work, we report a comparative investigation on the direct growth of chalcogenide CdS and PbS nanocrystals spanning a wide spectral range for light absorption on mesoporous TiO2 films employing the SILAR method. Reflectance and transmittance together with micro-Raman measurements were exploited to identify the optical and structural properties as well as quantum size effects of the sulfide nanocrystals and their stability upon air and light exposure. The electron injection efficiency of the sensitized films was accessed by photoluminescence (PL) measurements and the variation of the QD emission signal upon grafting onto TiO2.
Mesoscopic TiO2 films of a thickness of 15 μm were prepared using a TiO2 paste made of Degussa P25 nanoparticles on glass substrates, followed by sintering at 450°C . Films present excellent adherence to the glass substrate. For the CdS SILAR deposition , the TiO2 films were pretreated with a quick soaking in 1 M NH4F aqueous solution. Then, they were dipped into 0.05 M Cd(NO3)2, ethanol solution, rinsed in pure ethanol to remove excess of the precursor and dried in air. The same process was followed for depositing S2-, by successive dipping the films in 0.05 M Na2S solution, rinsing in pure methanol and drying. Each individual step lasted for 1 min and a total of 9 SILAR cycles were employed. PbS deposition was likewise carried out by sequential immersing the TiO2 film initially in a 0.02-M Pb(NO3)2 methanol solution, and then to a 0.02-M Na2S methanol solution. The process starts and terminates with Pb2+ deposition accomplishing 5.5 SILAR cycles .
Diffuse reflectance (R) and transmittance (T) measurements were carried out employing a Hitachi 3010 spectrophotometer equipped with a 60-mm diameter integrating sphere. The absorbance (A) spectra were derived as A = 1 - R - T. Surface morphology was examined with a digital Instruments Nanoscope III atomic force microscope (AFM), operating in the tapping mode. Micro-Raman and PL measurements were performed at room temperature employing a vacuum cell equipped with an optical window. For Raman, a Renishaw inVia spectrometer was employed, using an Ar+ ion laser (λ = 514.5 nm) and a high power near infrared (NIR) diode laser (λ = 785 nm) as excitation sources for CdS and PbS QDs, correspondingly. The spectra were recorded by focusing the laser beam on the film surface and controlling the light power to give 0.01 to 0.2 mW/μm2 at about 1.5 μm diameter spot. For PL experiments in PbS, the above facility was used, while for CdS, excitation of the film was done by focusing the 476.5-nm line of an Ar+ laser at 20 mW on the sample surface with an 8-cm focal length cylindrical lens. The emitted radiation was analyzed through a SPEX double monochromator, followed by photomultiplier detection.
Results and discussion
Figure 1a shows the evolution of the CdS/TiO2 absorbance, calculated from the corresponding transmittance and reflectance spectra, for successive SILAR cycles compared to that of the bare TiO2 films. Significant absorption in the visible range is thus observed, indicating the formation of CdS nanocrystals with gradually increasing concentration with the SILAR cycles. However, the distinct excitonic peaks, commonly observed for colloidal CdS QDs with a narrow size distribution, cannot be resolved, implying rather broad size dispersion for the SILAR deposited QDs. Moreover, the CdS/TiO2 absorption edge reached 585 nm upon completion of the ninth coating cycle. This value is close to that expected for bulk CdS, whose energy gap is approximately 2.4 eV, complying with the formation of nanocrystals with size exceeding 6 nm, above which quantum size effects essentially cease for CdS QDs . On the other hand, an appreciable increase of the mean CdS particle size can be inferred from the gradual red-shift of the absorption edge, most prominent for the initial SILAR cycles. This is indicative of weak quantum size effects, pertaining for CdS nanocrystals with diameters slightly below 6 nm.
Figure 1b shows the corresponding evolution of the PbS/TiO2 absorbance spectra with the SILAR cycles. In that case, the absorption edge of the sensitized system extended well in the NIR spectral region, presenting a marked shift from 690 nm for the first SILAR cycle up to 840 nm for the last PbS coating. These wavelengths are much shorter than the absorption edge (approximately 3000 nm) of bulk PbS that possess a narrow band gap of only 0.41 eV. This distinct variation of the PbS/TiO2 absorbance reflects essentially the large exciton Bohr radius (approximately 18 nm) of PbS QDs, affording wide tunability through the pronounced quantization effects for PbS nanocrystals over an extended particle size . Even though the broad spectral absorption of PbS/TiO2 is expected to comprise appreciable contributions from the whole electronic spectrum of the underlying PbS nanocrystals, its strong dependence on the coating cycles verifies that direct growth of PbS QDs on TiO2 and their optical response can be efficiently tuned by the SILAR technique through a broad size/spectral range.
However, storage of the PbS/TiO2 films under ambient conditions produced rapid degradation of their optical response. Specifically, brief exposure of the PbS/TiO2 to air for 90 min resulted in the drastic decrease of the absorbance and the shift of the absorption edge to shorter wavelengths, indicative of the reduction of the PbS size, as shown by the dashed line in Figure 1b. This variation can be associated with the prominent tendency of lead sulfide towards surface oxidation at ambient conditions, which is especially detrimental for the larger PbS nanocrystals . Storage under vacuum conditions in evacuated cells was accordingly found to be necessary to retain the PbS/TiO2 spectral characteristics intact. Similar degradation effects were also observed for the CdS/TiO2 films upon air exposure, though much less severe than those on PbS/TiO2, indicating their higher resistance to air oxidation that can be largely prevented by storage under inert atmosphere.
QD nanoparticles can be hardly identified in SEM and AFM images of the films, due to the rough characteristics of the TiO2 nanostructured substrate film. However, a morphological evidence of the CdS QDs came from 1 × 1 μm AFM surface images (not shown) on nanoparticulate sol-gel anatase TiO2 (chosen as a reference substrate) and comparing it with the surface of the CdS/TiO2 film corresponding to the full set of the 9 SILAR cycles. Thus, significant enhancement of the surface roughness was observed (Rms = 15.9 nm for CdS/TiO2 vs. 6.6 nm for bare TiO2), due to the CdS QDs growth on the surface, in agreement with literature .
The structural characteristics of the QD sensitized TiO2 films were investigated by resonance Raman measurements under vacuum in order to avoid air degradation. Figure 2 shows the Raman spectrum of CdS/TiO2 (9 SILAR cycles) at 514.5 nm, which is close to the absorption edge of the CdS nanocrystals and thus allows their resonant excitation. The characteristic Raman-active phonons of the underlying TiO2 substrate can be readily identified in comparison with the bare TiO2 electrode, the most intense being the low frequency anatase E g mode at approximately 142 cm-1, together with the resonantly excited longitudinal optical (LO) phonon of CdS QDs at approximately 300 cm-1. Spectral analysis reveals a slight asymmetric broadening of the CdS LO mode at the low frequency side, which can be effectively fitted to the superposition of two peaks, the LO mode at 301 cm-1 with full width at half maximum (FWHM) of 25 cm-1 and a broad low frequency mode at 277 cm-1 with FWHM of approximately 109 cm-1. Moreover, resonant excitation allows identifying the first (2 LO) and second (3 LO) overtones of the CdS nanoctystals at 604 and approximately 900 cm-1, respectively. The frequency of the LO peak matches bulk CdS (301 cm-1), whereas its width is considerably larger than the corresponding bulk value (approximately 12 cm-1) . The broadening of the LO peak together with its asymmetric lineshape corroborates the presence of a broad size distribution of CdS nanocrystals and the absence of strong phonon confinement effects , in agreement with the features of the CdS/TiO2 optical absorbance.
Raman measurements under NIR excitation (785 nm) were applied to identify the structural integrity of the lead sulfide nanocrystals through resonance excitation on the PbS/TiO2 films. A composite band comprising two bands at 202 and 260 cm-1 could be accordingly resolved on the sensitized PbS/TiO2, as shown in the inset of Figure 2. Lead sulfide crystallizes in rock salt structure precluding first-order Raman scattering from phonons near the centre of the Brillouin zone (k = 0). However, the formally 'forbidden' LO scattering at 200 to 215 cm-1 may become allowed under conditions of resonant or quasi-resonant Raman excitation via the Fröhlich interaction, while appreciable contributions may also arise at these frequencies from two-phonon scattering of longitudinal acoustic and transverse optical modes in PbS . A characteristic broad Raman band has been also reported at approximately 430 cm-1 due to 2 LO scattering in PbS , which, however, cannot be safely discriminated in the PbS/TiO2 spectra due to the additional contribution of the rutile TiO2 phonon at approximately 447 cm-1.
Degradation effects were also observed in the CdS Raman signal when acquired in ambient conditions, though considerably less pronounced than those of PbS/TiO2. Most importantly, an intriguing photodegradation effect on the CdS Raman intensity was evidenced by varying the laser irradiation time in ambient conditions. Figure 3 shows characteristic resonance Raman spectra of CdS/TiO2 acquired in air under variable laser power density and different acquisition times so that the total irradiation dose (product of laser power × acquisition time) remains constant. In that case, a marked increase of the CdS LO Raman intensity relative to that of the E g anatase TiO2 mode occurred by decreasing the spectral acquisition time (inset of Figure 3). Ordinary local heating effects are excluded since the relative CdS LO intensity was found to increase with the laser power and no appreciable shift and broadening of the LO mode or variation of the I 2LO/I LO intensity ratio were identified , indicating that the observed behavior is related to the duration of exposure of the CdS/TiO2 films to the laser beam. This variation was completely suppressed when Raman experiments were conducted in an isolated cell compartment under vacuum conditions, pointing to a photodegradation effect of the CdS nanocrystals under ambient conditions. A similar result was recently reported for CdSe QDs anchored to TiO2 following visible light irradiation under atmospheric conditions . In that case, time resolved transient absorbance and emission measurements revealed that electrons injected from CdSe to TiO2 may be scavenged by surface adsorbed oxygen leaving behind reactive holes, which cause anodic corrosion of the CdSe QDs. An analogous mechanism can be accordingly proposed for the CdS/TiO2 system upon resonant laser irradiation at 514.5 nm, causing electron injection to TiO2 and the surface oxidation of CdS nanocrystals through the remaining valence band holes.
Figure 4 shows the PL spectra acquired simultaneously with the Raman signal of the CdS/TiO2 under anaerobic conditions. To explore the charge injection efficiency for the QDs to the TiO2 substrate, CdS nanocrystals were deposited on microscopic glass employing 9 SILAR cycles, leading to a film with similar optical and Raman spectroscopic characteristics to that grown on TiO2. Comparison of the corresponding PL spectra, after subtraction of the relatively weak emission of the glass substrate, reveals significant changes between the CdS/TiO2 and CdS/glass films. The PL spectra of CdS/glass exhibits a strong component at about 530 nm, which is close to the band gap emission of bulk CdS arising from radiative excitonic recombination, while a rather broad emission band occurs at 625 nm most likely due to the recombination of trapped carriers by defect states . The frequency of the former emission band indicates the absence of significant quantum size effects, further supporting the growth of nanocrystals with size appreciably larger than the Bohr exciton radius of CdS (approximately 2.8 nm). Moreover, the width of the CdS excitonic peak (FWHM ~ 80 nm) in the CdS/glass film exceeds largely that of bulk CdS (FWHM ~ 20 nm) , indicative of a broad size distribution of the SILAR deposited CdS nanocrystals. However, upon CdS deposition on TiO2, the PL intensity of the excitonic emission is drastically suppressed, verifying the effective quenching of the radiative recombination of photoexcited carriers by electron transfer from CdS to TiO2.
In the case of PbS/TiO2, the PL emission spectra could be detected simultaneously with the Raman signal at 785 nm excitation. A very weak and broad PL band could be thus traced at 955 nm after subtraction of the glass background, as shown in the inset of Figure 4. This emission band emerges at wavelengths just above the absorption edge of the PbS/TiO2 (approximately 840 nm), complying with the excitonic PL of an ensemble of PbS QDs with a broad size distribution around 3 nm . Moreover, the PL emission band could be resolved only for freshly sensitized films PbS/TiO2, while it degraded rapidly upon air exposure verifying the great sensitivity of the system to surface oxidation. The drastic reduction of excitonic emission evidenced for both CdS and PbS nanocrystals upon direct growth on TiO2 by SILAR, markedly weaker than the emission colloidal QDs adsorbed on TiO2[11, 23], verifies the great potential of this deposition technique to enhance electronic coupling and the concomitant charge transfer between QDs and the underlying TiO2 substrate.
CdS and PbS nanocrystals can be efficiently deposited as sensitizers on mesoporous TiO2 substrates via the SILAR method. Enhanced electronic coupling and interfacial electron transfer are confirmed upon direct growth of the chalcogenide nanocrystals on TiO2 through the marked quenching of their excitonic emission. The optical absorbance of CdS/TiO2 can be tuned over a narrow spectral window in the visible range, reflecting essentially the small exciton Bohr radius of CdS QDs that inhibits utilization of quantum size effects for light harvesting. On the other hand, PbS/TiO2 exhibits pronounced band gap tunability spanning the visible to the NIR range, due to the prominent quantum size effects of PbS QDs. However, PbS/TiO2 degrades severely upon air exposure requiring a protection layer for application in solar cell devices. In contrast, CdS/TiO2 is appreciably more stable under ambient conditions, though it degrades readily under visible light irradiation.
atomic force microscope
chemical bath deposition
dye sensitized solar cells
full width at half maximum
successive ionic layer adsorption and reaction.
O'Regan B, Grätzel M: A low-cost, high-efficiency solar-cell based on dye-sensitized colloidal TiO 2 films. Nature 1991, 353: 737–740.
Meyer GJ: The 2010 Millennium Technology Grand Prize: Dye-Sensitized Solar Cells. ACS Nano 2010, 4: 4337–4343. 10.1021/nn101591h
Likodimos V, Stergiopoulos T, Falaras P, Harikisun R, Desilvestro J, Tulloch G: Prolonged Light and Thermal Stress Effects on Industrial Dye-Sensitized Solar Cells: A Micro-Raman Investigation on the Long-Term Stability of Aged Cells. J Phys Chem C 2009, 113: 9412–9422. 10.1021/jp901185f
Kamat PV: Quantum Dot Solar Cells. Semiconductor Nanocrystals as Light Harvesters. J Phys Chem C 2008, 112: 18737–18753.
Hodes G: Comparison of Dye- and Semiconductor-Sensitized Porous Nanocrystalline Liquid Junction Solar Cells. J Phys Chem C 2008, 112: 17778–17787. 10.1021/jp803310s
Mora-Sero I, Gimenez S, Fabregat-Santiago F, Gomez R, Shen Q, Toyoda T, Bisquert J: Recombination in Quantum Dot Sensitized Solar Cells. Acc Chem Res 2009, 42: 1848–1857. 10.1021/ar900134d
Lee YL, Lo Y-S: Highly Efficient Quantum-Dot-Sensitized Solar Cell Based on Co-Sensitization of CdS/CdSe. Adv Funct Mater 2009, 19: 604–609. 10.1002/adfm.200800940
Guijarro N, Lana-Villarreal T, Mora-Seró I, Bisquert J, Gómez R: CdSe Quantum Dot-Sensitized TiO 2 Electrodes: Effect of Quantum Dot Coverage and Mode of Attachment. J Phys Chem C 2009, 113: 4208–4214. 10.1021/jp808091d
Shalom M, Dor S, Rühle S, Grinis L, Zaban A: Core/CdS Quantum Dot/Shell Mesoporous Solar Cells with Improved Stability and Efficiency Using an Amorphous TiO 2 Coating. J Phys Chem C 2009, 113: 3895–3898. 10.1021/jp8108682
Lee HJ, Wang M, Chen P, Gamelin DR, Zakeeruddin SM, Grätzel M, Nazeeruddin MK: Efficient CdSe Quantum Dot-Sensitized Solar Cells Prepared by an Improved Successive Ionic Layer Adsorption and Reaction Process. Nano Lett 2009, 9: 4221–4227. 10.1021/nl902438d
Mora-Seró I, Likodimos V, Giménez S, Martínez-Ferrero E, Albero J, Palomares E, Kontos AG, Falaras P, Bisquert J: Fast Regeneration of CdSe Quantum Dots by Ru Dye in Sensitized TiO 2 Electrodes. J Phys Chem C 2010, 114: 6755–6761.
Diguna LJ, Shen Q, Kobayashi J, Toyoda T: High efficiency of CdSe quantum-dot-sensitized TiO 2 inverse opal solar cells. Appl Phys Lett 2007, 91: 023116. 10.1063/1.2757130
Chang CH, Lee YL: Chemical bath deposition of CdS quantum dots onto mesoscopic TiO 2 films for application in quantum-dot-sensitized solar cells. Appl Phys Lett 2007, 91: 053503. 10.1063/1.2768311
Lee HJ, Chen P, Moon SJ, Sauvage F, Sivula K, Bessho T, Gamelin DR, Comte P, Zakeeruddin SM, Seok SI, Grätzel M, Nazeeruddin MK: Regenerative PbS and CdS Quantum Dot Sensitized Solar Cells with a Cobalt Complex as Hole Mediator. Langmuir 2009, 25: 7602–7608. 10.1021/la900247r
Barea EM, Shalom M, Giménez S, Hod I, Mora-Seró I, Zaban A, Bisquert J: Design of Injection and Recombination in Quantum Dot Sensitized Solar Cells. J Am Chem Soc 2010, 132: 6834–6839. 10.1021/ja101752d
Kantonis G, Stergiopoulos T, Katsoulidis AP, Pomonis PJ, Falaras P: Electron dynamics dependence on optimum dye loading for an efficient dye-sensitized solar cell. J Photochem Photobiol A Chem 2011, 217: 236–241. 10.1016/j.jphotochem.2010.10.015
Yu W, Qu L, Guo W, Peng XG: Experimental determination of the extinction coefficient of CdTe, CdSe, and CdS nanocrystals. Chem Mater 2003, 15: 2854–2860. 10.1021/cm034081k
Cademartiri L, Montanari E, Calestani G, Migliori A, Guagliardi A, Ozin GA: Size-dependent extinction coefficients of PbS quantum dots. J Am Chem Soc 2006, 128: 10338–10346. 10.1021/ja063166u
Tang J, Brzozowski L, Barkhouse DAR, Wang X, Debnath R, Wolowiec R, Palmiano E, Levina L, Pattantyus-Abraham AG, Jamakosmanovic D, Sargent EH: Quantum Dot Photovoltaics in the Extreme Quantum Confinement Regime: The Surface-Chemical Origins of Exceptional Air- and Light-Stability. ACS Nano 2010, 4: 869–878. 10.1021/nn901564q
Sahoo S, Arora AK: Laser-Power-Induced Multiphonon Resonant Raman Scattering in Laser-Heated CdS Nanocrystal. J Phys Chem B 2010, 114: 4199–4203. 10.1021/jp912103t
Vasilevskiy MI, Rolo AG, Gomes MJM, Vikhrova OV, Ricolleau C: Impact of disorder on optical phonons confined in CdS nano-crystallites embedded in a SiO2 matrix. J Phys Condens Matter 2001, 13: 3491–3509. 10.1088/0953-8984/13/14/320
Etchegoin PG, Cardona M, Lauck R, Clark RJH, Serrano J, Romero AH: Temperature-dependent Raman scattering of natural and isotopically substituted PbS. Phys Status Solidi B 2008, 245: 1125–1132. 10.1002/pssb.200743364
Tvrdy K, Kamat PV: Substrate Driven Photochemistry of CdSe Quantum Dot Films: Charge Injection and Irreversible Transformations on Oxide Surfaces. J Phys Chem A 2010, 113: 3765–3772. 10.1021/jp808562x
Orii T, Kaito SI, Matsuishi K, Onari S, Arai T: Photoluminescence of CdS nanoparticles suspended in vacuum and its temperature increase by laser irradiation. J Phys Condens Matter 2002, 14: 9743–9752. 10.1088/0953-8984/14/41/329
Peterson JJ, Krauss TD: Fluorescence spectroscopy of single lead sulfide quantum dots. Nano Lett 2006, 6: 510–514. 10.1021/nl0525756
This work is financially supported by the "Sensitizer Activated Nanostructured Solar Cells -SANS"/FP7-NMP-2009-SMALL3-246124 project. The authors thank Ivan Mora-Seró and Juan Bisquert for valuable suggestions.
The authors declare that they have no competing interests.
AGK participated in the design and implementation of the work and help to draft the manuscript. VL carried out the Raman characterization and analysis. EV carried out the preparation of CdS QDs on TiO2. IK carried out the preparation of PbS QDs on TiO2. YSR participated in the realization of the photoluminescence experiments. CR have been involved in revising the manuscript critically for important intellectual content. PF conceived the study, participated in its design and coordination, and helped to draft and finalize the manuscript. All authors read and approved the final manuscript.