Published on Web Date: March 17, 2010

r 2010 American Chemical Society 1134 DOI: 10.1021/jz1002555 |J. Phys. Chem. Lett. 2010, 1, 1134–1138

pubs.acs.org/JPCL

Quantum Dot-Dye Bilayer-Sensitized Solar Cells:Breaking the Limits Imposed by the Low Absorbanceof Dye MonolayersMenny Shalom, Josep Albero, Zion Tachan, Eugenia Martínez-Ferrero, Arie Zaban,* andEmilio Palomares*

Institute of Nanotechnology & Advanced Materials, Department of Chemistry, Bar-Ilan University, 52900 Ramat Gan, Israel,Institute of Chemical Research of Catalonia (ICIQ), Avda. Països Catalans 16, E-46007 Tarragona, Spain, and Instituci�o Catalanade Recerca I Estudis Avanc-ats (ICREA), Avda. Lluís Companys 23, E-08010 Tarragona, Spain

ABSTRACT Here, we present a new DSSC design, consisting of sequential QDsand dye sensitization layers, that opens the path toward high optical density DSSCsthat cover a significant part of the solar spectrum. Thenewconfiguration is enabledby the application of an amorphous TiO2 layer between the two sensitizers,allowing both electron injection from the outer absorber and fast hole extractionfrom the inner sensitizing layer. Utilizing two sensitizing layers, we obtain a 250%increase in cell efficiency compared to a QD monolayer cell.

SECTION Energy Conversion and Storage

C urrent conversion efficiency of dye-sensitized solarcells (DSSCs) reflects the limits imposed by the lowabsorbance of dye monolayers and the low efficiency

of dye multilayers. The use of stronger absorbers such asquantum dots1-10 (QDs) has not led to an increase in effi-ciency due to the slow hole removal that enhances the recom-bination efficiency and instability of the nanocrystals wheniodine-based electrolytes are used.11 In addition, the spectralproperties of the relevant sensitizers require the use of morethan one absorber in order to match the solar spectrum.12

Record conversion efficiencies of DSSCs have reached nearly12%, following optimization of the various cell componentsand fabrication methods.13 Thus, further improvement re-quires new concepts to overcome the limitations inherent inthe standard cell.

A well-known restriction is associated with the fact thatefficient photoinduced charge separation requires direct con-tact between the absorber and themetal-oxide surface.14 Thegeometryof the nanoporous oxide electrode and thediffusionlength of electrons15 in the electrode limit the surface areathat can be used for dye molecule adsorption. Consequently,extending the spectral response of the cell by utilizing severalsensitizers comes at the expense of lower optical density.12 Inother words, the effective surface area of the nanoporouselectrodes defines the portion of the solar radiation that isutilized by the cell. Thus, the injection of electrons fromabsorbers that are not attached directly to the nanoporousoxide should lead to DSSCs of enhanced optical density andhigher overlap with the solar spectrum, without damag-ing the electron collection efficiency.16,17 In this Letter, wereport a method designed to overcome the dye-monolayerrestriction.

Recently, we showed that it is possible to stabilize CdS QD-basedDSSCsbycoating theQD-sensitizednanoporouselectrodewith a thin amorphous TiO2 layer.

18 The coating enables the useof various QD sensitizers in the presence of iodine-basedelectrolytes. On the basis of this knowledge, we developed thenew bisensitizer layer configuration shown in Figure 1. Theresulting cell consists of a nanoporous TiO2 electrode, a layerof CdS quantum dots as the sensitizer, an amorphous TiO2

coating (aTiO2) that serves as both a stabilizer for theQDs and asubstrate for the second absorber, a N719 dye layer, and finallythe standard iodine-based electrolyte. Current-voltage andincident photon-to-current efficiency (IPCE) measurementsshow that both absorbers contribute to the performance of thecell. Similar conclusions are drawn fromphotoinduced transientabsorption spectroscopy (TAS) measurements that directlyprobe the interfacial electron injection and recombination dy-namicsof thedevice.Furthermore, the results reveal significantlyfaster hole extraction from the oxidized QDs by the secondsensitizer layer, which contributes to the overall cell efficiency.

Figure 1 presents a schematic illustration and an energyband diagram of the bisensitizer nanoporous electrode(nanocrystalline TiO2/CdS QD/amorphous TiO2/N719 dye).This bisensitizer structure has been confirmed by high-resolu-tion transmission electron microscopy (HRTEM) images of thescraped electrode (see Supporting Information, Figure S1),where the amorphous TiO2 coating of the crystalline TiO2

decorated with the CdS QDs can be easily observed.

Received Date: February 23, 2010Accepted Date: March 11, 2010

r 2010 American Chemical Society 1135 DOI: 10.1021/jz1002555 |J. Phys. Chem. Lett. 2010, 1, 1134–1138

pubs.acs.org/JPCL

Figure 2 shows IPCE curves of three DSSCs that are basedon the bisensitizer electrodewhere, in one case, the N719 dyewas not adsorbed onto the electrode and, in another case,QDs were not present. In the absence of the outer dye layer,the IPCE resembles theabsorption of theCdSQDs (Figure 2a).When the dye is adsorbed on the amorphous TiO2 coating(Figure 2b), we observe an increase of the IPCE valuesthroughout the measured spectrum. The calculated contribu-tion of the dye, presented in Figure 2c (subtraction of curve 2afrom curve 2b), is similar to the IPCE contribution of the N719dye, presented in Figure 2d. This similarity reveals that bothsensitizer layers participate in thephotovoltaic process. There-fore, despite the sequential position of the two-sensitizerlayers, electrons injected from the outer dye layer arrive atthe front contact, while hole transfer from the inner QD layerto the electrolyte is possible.

The contribution of the two absorbers to the overall cellperformance is also evident in the I-V curves presented inFigure 3. Comparing the bisensitizer nanoporous configura-tion (nanocrystalline TiO2/CdS QD/amorphous TiO2/N719dye) with a monosensitizer cell (nanocrystalline TiO2/CdSQD), we obtained a significant increase in all cell parameters.The short-circuit current, Isc, increased from 1.27 to 3.1 mA/cm2 (þ240%), the open-circuit voltage, Voc, increased from715 to 775 mV (þ60 mV), and the fill factor (FF) remainunchanged (0.65), despite the increase in current. Conse-quently, under operating conditions (1 sun illumination), the

bisensitizer configuration improved the conversion efficiencyfrom 0.57 to 1.51% (þ263%).

In order to learn themechanism of electron and hole pathsin the bisensitizer solar cell, we describe the cell operation tothe inner and outer sensitizers dynamics (Figure 4 andextended in Figure S2 (Supporting Information)). The firstscientific question was whether the outer layer can injectelectrons thought the QD (Figure 4, process b) or via theamorphous TiO2 coating to the mesoporous TiO2 (Figure 4,process a).

Photoinduced transient absorption spectroscopy (TAS) hasbeen used to study the charge-transfer dynamics of thebisensitizer layer configuration since the time-resolved spec-tral differences enable understanding of the charge-transfermechanisms associated with each sensitizer.19 The UV-visible absorption spectra of QD-sensitized solar cells areshown in Figure 5a. The excitonic peak of the QD appearsas a broad band due to the high absorption rate of theQD intoTiO2 and is centered at 439 nm, while the maximum absor-bance of theN719 dye is at 535nm. The nanocrystalline TiO2/CdS QD/amorphous TiO2/N719 dye spectrum is the result ofthe contribution of both the CdS and the dye. When measur-ing TAS spectra, wewill excite at themaximumwavelength ofthe species to obtain the absorbance spectra of the corre-sponding transient species formed after charge separation.Figure 5b shows the transient absorbance spectra of the

Figure 1. Interfacial charge recombination processes. (A) Uponlight irradiation, electrons are injected directly (solid lines) fromthe QD into the TiO2 films and from the dye to the amorphousTiO2, which is in contact with the TiO2 film (dotted line). (B) Afterthe charge separation, the electrons are transported to the contact,and the QDs are regenerated, leaving the holes in the dye.

Figure 2. Incident photon-to-current efficiency (IPCE) of the (a)QD-sensitized solar cell, (b) bisensitizer electrode, (c) calculatedcontribution of the dye, and (d) N719/aTiO2/TiO2 solar cell.

Figure 3. I-V measurement under illumination of 1 sun (100mW/cm2) of the (a) QD sensitizing solar cell and (b) bisensitizerelectrode.

Figure 4. The mechanism processes in the bisensitizer cell.Electrons injection can occur from the inner layer of CdS QDs tothe nanocrystalline TiO2 (process d) or from the outer dye to theamorphous TiO2 (process a). Process d occurs together with holeextraction from the QD to the dye (process c). Electron injectionfrom the dye to the TiO2 through the QD (process b) is thusdiscarded.

r 2010 American Chemical Society 1136 DOI: 10.1021/jz1002555 |J. Phys. Chem. Lett. 2010, 1, 1134–1138

pubs.acs.org/JPCL

different molecules involved in our devices. The as-obtainedabsorbancemaxima (560 nm for the CdS and 760 nm for theN719•þ) indicate where to probe when recording time-re-solveddecays of the transients. It has to be noted that theCdS-containing samples give two different types of signals. TiO2/CdS shows a positive maximum after excitation at 439 nmthat corresponds to the transient CdSþ formed after electroninjection into TiO2. On the other hand, TiO2/CdS/a-TiO2/N719or TiO2/CdS/N719 show a bleaching at the same position thatcorresponds to the CdS due to the fast hole regeneration (videinfra).

Figure 6 shows the dynamics of the outer dye layer, whichreveal electron photoinjection into the nanocrystalline TiO2

via the amorphous TiO2 coating (Figure 4, process a).Wenotethat the spectral window of the dye and the CdS dots enablesole excitation of the dyewhile the innerQDs layer remains inthe ground state (see Figure 5b). In a reference system inwhich the dye is directly attached to the QDs (i.e., without the

amorphous TiO2), both the steady-state (not shown) andtransient measurements (Figure 6) show no charge transferfrom the dye to the nanocrystalline TiO2. Consequently, thereis no electron injection across the CdS QDs into the nano-crystalline TiO2, that is, we discard process b of Figure 4. Ourresults show that the amorphous TiO2 opens a transport pathto the underlying nanocrystalline TiO2 (Figure 4, process a).Further support to this conclusion is drawn from the rate ofelectron return to the dye. Durrant and co-workers havedemonstrated that back-electron-transfer kinetics from theTiO2 to the oxidized dye follow distance-dependent behaviorfor a wide range of dyes, including ruthenium polypyridilcomplexes.20 Thus, in the presence of the amorphous TiO2/N719 outer layer, retardation on the back electron transfershould be observed. Indeed, Figure 6 shows that the electronrecombination half-lifetime in the presence of the amorphousTiO2 (360 s coating)/N719 outer layer is on the order of 0.25 s,which is a hundred times slower than for the standard TiO2/N719 system (0.001 s).21 In other words, the conformalamorphous TiO2 layer not only creates the possibility forsequential cosensitization (bisensitizer layer) but also slowsthe back electron transfer to the dye, thus improving thecharge separation associated with the outer sensitizer layer.

The second question regards the inner sensitizer (QD) inthe presence of the outer layer (dye). Due to spectral overlap,it is not possible to excite the QDs without exciting the dye.However, the cation dynamics of both theQDand the dye canstill be separately monitored (see Figure 5b). Figure 7 com-pares the back-electron-transfer kinetics from the TiO2 to theoxidized CdS in the mono- and bisensitizer configurations.The monosensitizer system gives a positive decay with arecombination half-lifetime of 1.2 ms (see also Figure S3(Supporting Information)). In contrast, while the secondsensitizer is being anchored, a negative signal is observed,indicating that holes havebeenextractedandmoved fromtheinner CdS layer to the outerdye layer (Figure 4, process c), likepreviously reported.22,23 The electronic energy levels be-tween CdS and N719 dye are aligned such that the N719

Figure 5. (a) UV-visible spectrum of the TiO2/CdS, TiO2/CdS/a-TiO2, TiO2/N719, and TiO2/CdS/a-TiO2/N719 together with TiO2 shown as areference; (b) TAS absorption spectra taken 100 μs after excitation at 439 or 535 nm (see legend).

Figure 6. N719 recombination dynamics in the TiO2/CdS/amor-phous TiO2(360 s)/N719 (green) and TiO2/CdS/N719 (black) devi-ces. Back-electron-transfer decay after selective excitation at 535nm and monitoring at 760 nm.

r 2010 American Chemical Society 1137 DOI: 10.1021/jz1002555 |J. Phys. Chem. Lett. 2010, 1, 1134–1138

pubs.acs.org/JPCL

HOMO is above the CdS valence band, facilitating holeextraction from the CdS quantum dot to the dye (Figure 4,process c). The negative signal feature has been previouslyreported for dye-cosensitized TiO2 films as a feature of holetransfer between the dyes.23 Consequently, adsorption ofN719 on the amorphous TiO2, forming a bisensitizer elec-trode, significantly improves the CdS regeneration rate, whichis a key factor in achieving an efficient QD-based solar cell.

We have demonstrated the ability to utilize two sequentialsensitizer layers in dye-sensitized solar cells. The new bisen-sitizer layer cell consists of a nanocrystalline TiO2/CdS QD/amorphous TiO2/N719 dye. Consequently, it is possible towiden the spectral response of the cell whilemaintaining highoptical density, with a theoretical limit equal to twice a givenelectrode area.While improving thecell's ability to harvest thesolar radiation, the new configuration offers a method ofslowing the interfacial charge recombination processes andthus increasing the charge-separation efficiency. Further ex-periments are underway to extend the bisensitizer layerconcept to the solid-state analogue in which back-electron-transfer kinetics is a major limit.

EXPERIMENTAL SECTION

Mesoporous TiO2 films were prepared by electrophoreticdeposition (EPD)24 of Degussa P25 particles with an averagediameter of 25 nm onto a fluorine-doped tin oxide (FTO)-covered glass substrates (Pilkington TEC 15) with a 15 Ω/square sheet resistance. Films were deposited in two con-secutive cycles for 30 s at a constant current density of0.4 mA/cm2 (which corresponds to ∼70 V at an electrodedistance of 50 mm) and dried at 120 �C for ∼5 min inbetween the cycles. Following the EPD process, all of theelectrodes were dried in air at 150 �C for 30min and sinteredat 550 �C for 1 h. For CdS QD deposition, the electrodes wereimmersed in a mixture of 2.35 mL of 0.5 M CdSO4 and2.65mLof0.7Mpotassiumnitrilotriacetate (K3NTA) at pH8.5adjusted by 10%KOH. This solution wasmixed with 4.25mLof 0.4 M thiourea and then diluted with 7.55 mL of distilled

water. Finally, thepHwas readjusted to pH11using again10%KOH. After the electrodewas immersed in the solution, it washeated to 80 �C for 2 h,25 resulting in CdS coating of themesoporous TiO2 electrode. The final step involved a thinTiO2 coatingof theCdS-sensitizedmesoporous TiO2 electrodeby electrophoretic deposition of stabilized a TiO2 precursor(Ti(OiC3H7)4) for60s (current2mA) and360s (current0.2mA).Mild heat treatment of the coated electrodes at 80 �C for 10minwas used to stabilize the amorphous TiO2. For the bilayerstructure, the QD-sensitized electrodes were immersed in10 mM N719 in ethanol for 12 h. An I-/I3

- redox electrolytewas used in the CdSQD-sensitized solar cells consisting of 0.1Mlithium iodide, 0.05 M iodine, 0.6 M 1-propyl-2,3-dimethylimi-dazolium iodide, and 0.5 M 4-tertbutylpyridine dissolved inpropylenecarbonate. A Pt-coated FTO glass was used as acounterelectrode. I-Vmeasurements were performed with anEco-Chemie Potentiostat. A 300 W xenon arc lamp (Oriel)calibrated to 100 mW/cm2 served as a light source.

The transient absorbance experiments were recorded byexcitation of the sensitized films with pulses from a nitrogen-pumped dye laser PTI GL-301 (<1 ns pulse duration, 1 Hz,intensity of 0.05 mJ cm-2 when exciting at 535 nm and0.09 mJ cm-2 when exciting at 439 nm). The resultingphotoinduced changes in optical density were monitored byemploying a 150 W tungsten lamp, with 1 nm bandwidthmonochromators before and after the sample, a home-builtphotodiode-baseddetection system, andaTDS-200Tektronixoscilloscope. The N719 was excited at 535 nm and its decaymonitored at 760nm,while the CdSwas studied by excitationat 439 nmand recording at 560 nm. It should be noted that at439 nm, the dye was also excited.

SUPPORTING INFORMATION AVAILABLE Supporting fig-ures and illustration of electron and hole paths in the bisensitizedsolar cell. This material is available free of charge via the Internet athttp://pubs.acs.org.

AUTHOR INFORMATION

Corresponding Author:*To whom correspondence should be addressed. E-mail: zabana@mail.biu.ac.il (A.Z.); epalomares@iciq.es (E.P.).

ACKNOWLEDGMENT M.S., Z.T., and A.Z. acknowledge thesupport of the Israel Science Foundation. E.M.F. and J.A. thank theSpanish MICINN for the Juan de la Cierva and Torres Quevedofellowships, respectively. E.P. expresses gratitude to the MICINNfor the HOPE CSD-2007-00007 (Consolider-Ingenio 2010) andCTQ2007-60746/BQU projects as well as the European ResearchCouncil for the ERC fellowship PolyDot. The ICIQ and ICREA supportis also gratefully acknowledged.

REFERENCES

(1) Kamat, P. V. Quantum Dot Solar Cells. Semiconductor Nano-crystals as Light Harvesters. J. Phys. Chem. C 2008, 112,18737–18753.

(2) Kongkanand, A.; Tvrdy, K.; Takechi, K.; Kuno, M.; Kamat, P. V.QuantumDot Solar Cells. Tuning Photoresponse through Sizeand Shape Control of CdSe-TiO2 Architecture. J. Am. Chem.Soc. 2008, 130, 4007–4015.

Figure 7. Dye-mediated recombination decay dynamics. Afterselective excitation of the CdS QDs at 439 nm, in the absence ofN719, the photoinjected electrons recombine with oxidized CdS(squares). In the presence of amorphous TiO2/N719, we observe theregenerationof theCdSgroundstate independentlyof the timeused tosimilarly coat the TiO2/CdS film. The signalswere recorded at 560nm.

r 2010 American Chemical Society 1138 DOI: 10.1021/jz1002555 |J. Phys. Chem. Lett. 2010, 1, 1134–1138

pubs.acs.org/JPCL

(3) Lee, H. J.; Chen, P.; Moon, S. J.; Sauvage, F.; Sivula, K.; Bessho,T.; Gamelin, D. R.; Comte, P.; Zakeeruddin, S. M.; Il Seok,S.; et al. Regenerative PbS and CdS Quantum Dot SensitizedSolar Cells with a Cobalt Complex as HoleMediator. Langmuir2009, 25, 7602–7608.

(4) Lee, H. J.; Yum, J. H.; Leventis, H. C.; Zakeeruddin, S. M.;Haque, S. A.; Chen, P.; Seok, S. I.; Grazel, M.; Nazeeruddin,M. K. CdSe Quantum Dot-Sensitized Solar Cells ExceedingEfficiency 1%at Full-Sun Intensity. J. Phys. Chem. C2008, 112,11600–11608.

(5) Lee, Y. L.; Lo, Y. S. Highly Efficient Quantum-Dot-SensitizedSolar Cell Based on Co-Sensitization of CdS/CdSe. Adv. Funct.Mater. 2009, 19, 604–609.

(6) Mora-Sero, I.; Gimenez, S.; Moehl, T.; Fabregat-Santiago, F.;Lana-Villareal, T.; Gomez, R.; Bisquert, J. Factors Determiningthe Photovoltaic Performance of a CdSe Quantum Dot Sensi-tized Solar Cell: The Role of the Linker Molecule and of theCounter Electrode. Nanotechnology 2008, 19, 424007.

(7) Shalom, M.; Ruhle, S.; Hod, I.; Yahav, S.; Zaban, A. EnergyLevel Alignment in CdS Quantum Dot Sensitized Solar CellsUsing Molecular Dipoles. J. Am. Chem. Soc. 2009, 131, 9876–9877.

(8) Gimenez, S.; Mora-Sero, I.; Macor, L.; Guijarro, N.; Lana-Villarreal,T.; Gomez, R.; Diguna, L. J.; Shen, Q.; Toyoda, T.; Bisquert, J.Improving thePerformanceofColloidalQuantum-Dot-SensitizedSolar Cells. Nanotechnology 2009, 20, 295204.

(9) Guijarro, N.; Lana-Villarreal, T.; Mora-Sero, I.; Bisquert, J.;Gomez, R. CdSe Quantum Dot-Sensitized TiO2 Electrodes:Effect of Quantum Dot Coverage and Mode of Attachment.J. Phys. Chem. C 2009, 113, 4208–4214.

(10) Hodes, G. Comparison of Dye- and Semiconductor-SensitizedPorous Nanocrystalline Liquid Junction Solar Cells. J. Phys.Chem. C 2008, 112, 17778–17787.

(11) Lee, Y. L.; Chang, C. H. Efficient Polysulfide Electrolyte for CdSQuantum Dot-Sensitized Solar Cells. J. Power Sources 2008,185, 584–588.

(12) Lee, H. B.; Leventis, H. C.; Moon, S. J.; Chen, P.; Ito, S.; Haque,S. A.; Torres, T.; Nuesch, F.; Geiger, T.; Zakeeruddin, S.M.; et al.PbS and CdS Quantum Dot-Sensitized Solid-State Solar Cells:Old Concepts, New Results. Adv. Funct. Mater. 2009, 19, 1–8.

(13) Luo, Y.; Li, D.;Meng,Q. TowardsOptimization ofMaterials forDye-Sensitized Solar Cells. Adv. Mater. 2009, 21, 4647–4651.

(14) Gratzel, M. Dye-Sensitized Solar Cells. J. Photochem. Photo-biol., C 2003, 4, 145–153.

(15) Sodergren, S.; Hagfeldt, A.; Olsson, J.; Lindquist, S. E. Theo-retical-Models for the Action Spectrum and the Current-Voltage Characteristics of Microporous Semiconductor-Filmsin Photoelectrochemical Cells. J. Phys. Chem. 1994, 98, 5552–5556.

(16) Hardin, B. E.; Hoke, E. T.; Armstrong, P. B.; Yum, J. H.; Comte,P.; Torres, T.; Frechet, J. M. J.; Nazeeruddin, M. K.; Gratzel, M.;McGehee, M. D. Increased Light Harvesting in Dye-SensitizedSolar Cells with Energy Relay Dyes. Nat. Photonics 2009, 3,406–411.

(17) Shankar, K.; Feng, X.; Grimes, C. A. Enhanced Harvesting ofRed Photons in Nanowire Solar Cells: Evidence of ResonanceEnergy Transfer. ACS Nano 2009, 3, 788–794.

(18) Shalom, M.; Dor, S.; Ruhle, S.; Grinis, L.; Zaban, A. Core/CdSQuantum Dot/Shell Mesoporous Solar Cells with ImprovedStability and Efficiency Using an Amorphous TiO2 Coating.J. Phys. Chem. C 2009, 113, 3895–3898.

(19) Palomares, E.; Clifford, J. N.; Haque, S. A.; Lutz, T.; Durrant,J. R. Control of Charge Recombination Dynamics in DyeSensitized Solar Cells by the Use of Conformally Deposited

Metal Oxide Blocking Layers. J. Am. Chem. Soc. 2003, 125,475–482.

(20) Haque, S. A.; Palomares, E.; Cho, B. M.; Green, A. N. M.;Hirata, N.; Klug, D. R.; Durrant, J. R. Charge Separation versusRecombination inDye-SensitizedNanocrystalline Solar Cells:The Minimization of Kinetic Redundancy. J. Am. Chem. Soc.2005, 127, 3456–3462.

(21) Sanchez-Diaz, A.; Martinez-Ferrero, E.; Palomares, E. ChargeRecombination Studies in Conformally coated Trifluoroace-tate/TiO2modifiedDye Sensitised Solar Cells (DSSC). J. Mater.Chem. 2009, 19, 5381–5387.

(22) Mora-Ser�o, I.; Gross, D.; Mittereder, T.; Lutich, A. A.; Susha,A. S.; Dittrich, T.; Belaidi, A.; Caballero, R.; Langa, F.; Bisquert,J.; et al. Nanoscale Interaction Between CdSe or CdTe Nano-crystals and Molecular Dyes Fostering or Hindering Direc-tional Charge Separation. Small 2010, 6, 221–225.

(23) Leventis, H. C.; Haque, S. A. Control of Charge Recombina-tion at Nanostructured Quantum-Dot Sensitized TiO2 Inter-faces Employing a Multi-Step Redox Cascade. EnergyEnviron. Sci. 2009, 2, 1176–1179.

(24) Grinis, L.; Dor, S.; Ofir, A.; Zaban, A. Electrophoretic Deposi-tion and Compression of Titania Nanoparticle Films for Dye-Sensitized Solar Cells. J. Photochem. Photobiol., A 2008, 198,52–59.

(25) Gorer, S.; Hodes, G. Quantum Size Effects in the Study ofChemical Solution Deposition Mechanisms of Semiconduc-tor Films. J. Phys. Chem. 1994, 98, 5338–5346.