Photoelectric Characteristics of Nanocrystalline TiO
2Film Prepared from TiO
2Colloid Sol for Dye-Sensitized Solar Cell
Kyung-Jun Hwang, Jae-Wook Lee, Ho-Sung Yoon,t Hee-DongJang「Jin-Ge이Kim,* JinSuk Yang,§ andSeung-Joon Yoo#，*
DepartmentofChemical and Biochemical Engineering, Chosun University, Gwangju501-759, Korea
‘KoreaInstitute of Geoscienceand Miner시 Resources,Daejeon305-350,Korea
土 DepartmentofChemical Engineering, Soonchunhyang University, Asan, Chungnam 561-756,Korea
§ Department of Chemical and Biomolecular Engineering, Sogang University, 1,Sinsoo,Seoul 121-742, Korea 큐Department ofEnvironmental andChemical Engineering, Seonam University, Namwon590-711,Korea
* E-mail: sjyoo001@hanmail.net
ReceivedMay 13, 2009, Accepted September 1, 2009
A working electrode in dye-sensitized solar cells was fabricated using T1O2 colloidal sol prepared from titanium isopropoxide used as a starting material by applying the sol-gel method. The effect of aging times and temperatures on physical and chemical properties of TiO2sol particles was systematically investigated. Results showed that the crystallinity and average particle size of TiO2 colloidal sol can be successfully controlled by the adjustment of aging time and temperature. The conversion efficiency of the repetitive dry coating films fabricated using the dried TiO2
colloidal sol particles and hydroxypropyl cellulose binder (15%) was 10.31% with a high transparency.
KeyWords:Dye-sensitized solar cell, TiO2 colloidal sol, Aging
Introduction
Dye-sensitizedsolarcells(DSSCs)are the most promising alternative to conventional solar cells conceived in recent years.1Nanosized TiO2 particles havereceived great attention for useasa photoelectrode in dye-sensitized solarcell systems.
In general,the TiO2 photoelectrodeconsists of nanosized coll
oids that are sinteredon atransparentconducting substrate, thus having a porous geometry.2,3 Various preparationmethods such assoftchemistry, hydrothermal, and the sol-gelprocess have been suggested for nanosized TiO2 colloids.4-6 It has been reported that the sol-gel process hasmany advantages for the fabrication of thin film inDSSC. The function of nano crystalline thin films fabricatedfromcolloidal sol hasbeen found tobestronglydependent onthe sol preparation process.
Suchsynthetic procedureshave been shown tocontrol a wide range of material properties of the resultingfilms,in including particle size and crystallographic phase, film porosity,surface structure, electrontransport properties, and opticallightscat- tering.7 Recently, the synthesis of TiO2 colloidal paste for making of photoelectrode in DSSCs is muchattention from all over the world.8 Unfortunately,there is no report on thesyste matic controls of microstructured TiO2 thin film to induce a high efficiency of DSSCs uptonow.
In this study, the effect of agingconditions in theTiO2 sol preparation wasinvestigatedto make TiO2 thin film havinga high transparency and a high efficiency in DSSCs. For the control of nanostructuredTiO2 thin films, TiO2 colloidalsol was prepared through the aging process accompanied by hydrolysis/polycondensationusingtitanium-tetraisopropoxide (TTIP)asa starting material.We found that theagingprocess among the sol preparation steps is themostimportantstepin controlling the size and structure of TiO2 colloidal sol. We
alsofound that theperformance of DSSC using theresultant TiO2 film is highly affected by the physical and chemical properties of TiO2 solparticles.
Experiment시
Synthesisof T1O2 colloidl sols (i.e., SN-T1O2 sol).Figure 1 showsaschematic diagram for the synthesis of TiO2 soland particles from astartingmaterial. The reaction between TTIP (Titaniumtetra-isopropoxide, Junsei Chemical Co. > 98%) andH2O was performed under the condition of 1000 rpmand temperatures of 40oC and 90 oC The amount of water was fixedat 100H2O/TI molar ratio.9Thefirstchemical reaction wasahydrolysis betweenTTIP andwaterand it wastoofast tocontrol the reaction rate. Ontheotherhand, after theinitial hydrolysis, the reaction was easily controlledbecause of the slowrate. The characteristic of TiO2 solparticles was controlled byoptimizingthetime and temperature of aging. The initial fastreactionbetween thewater and TTIPreachedthe equili brium temperature within 30 min. The reactionafter 30 min was named“aging”.The effect of aging wasinvestigated accord
ing to the variation of agingtimes (0, 24, 72 h) under the condition of twodifferentaging temperatures of40 oCand 90 oC. Thecharacteristics of TiO2 sol particles wasmeasuredby the effectof theagingtimeunderthecondition of aging tem peratures with 40 oCand90 oC and a mixing rate of 1000 rpm.
After theaging,TiO2 solutionwastransformedintoa solution phaselikepaste, andapeptizing agentneeded acidicor basic electrolyte to peptize thesolution effectively.Although various electrolytes such asHCl, HNO3 and NH4OH were available forpeptizing agent,thisexperiment wasperformed underthe condition ofHCl toderive the ionic repulsive force requiredat thecondition of lowconcentration. An acidic electrolyte was
Soc. , 30, No. 10 Kyung-Jun al
Figure 1. Experimental flowchart for the synthesis of T1O2.
added in the solution with TiO2 aggregatedparticles deter
mined byaging toinducetherepulsive force of interparticles and preventcontinuousaggregation. At thispoint, the acidic electrolyte was selected with a nitric acid (HNO3,Daejung Co. 60%) as an inorganicacid. Theamount of the HNO3 was optimizedat 0.25 HNO3/Ti molarratio condition determined by the preliminary experiment. Thehydrous TiO2 sol was dried at80 oC for 1day and calcined at 500 oC by2 oC/min for1 hr atthe final temperatureto obtaina photoelectrode thinfilm.
Pnepaiation of T1O2 thin film. As mentionedbefore, SN- TiO2 powders were obtained under the variation of aging times(0, 24, 72 h) at twodifferentaging temperatures (40 oC and90 oC). For theTiO2 thin film, TiO2 slurry was prepared by mixing of 2 g SN-TiO2 particles,0.68 mL 10% [V/V] acetyl acetone, 1 g hydroxypropyl cellulose (i.e., HPC, Mw. 80,000, Aldrich), and10.68 mL water for 30 min at 300rpmusing a pastemixer (PDM-300, Korea mixing technology Co.). Thus, SN-TiO2 filmwas fabricated by coating blendedpaste onto thefluorine-dopedSnO2 conducting glass plates (FTO, 10 Q/
cm2, Asahi glass Co.) using a squeeze printing technique (adhesive tape was used as spacer ofca. 65“m thickness. The TiO2 film was heat-treated at 500 oC for 1h. TheTiO2 film formed on the FTOglass was 7“m thick and 0.5 cm乂 0.5 cm.
The TiO2 thin film was prepared underthe different mixing ratio of the dried SN-TiO2 (5 〜25 wt%) andHPC binder (10 〜25 wt%) for photoelectrode with high efficiency and high trans
parency. In addition, repetitive dry coatingof theTiO2 film was conductedto enhancethe conversion efficiency of DSSCs.10 Tofabricate theDSSCs,thepreparedTiO2 thin film electrode was immersed in the N719dye(Solaronix Co.) solution of 5x
10-4Mat 50oCfor 4 h, rinsed withanhydrous ethanol and dried.
A Ptcoatedglass-SnO2： F (FTO, 10Q/cm2, AsahiglassCo.) electrodewas prepared as a counter electrode with an active area of0.25 cm2. The Pt electrodewas placed over the dye- adsorbed TiO2 thin film electrode, and theedgesof the cell were sealedwith 5 mmwide stripers of 60 “m thick sealing sheet (SX 1170-60, Solaronix). Sealingwasaccomplished by hot-pressing the twoelectrodes together at 120 oC. Theredox electrolytewas injectedintothe cellthrough thesmallholes and sealed with a small square of sealing sheet. The redox electrolyte consistsof 0.3 M 1,2-dimethyl-3-propylimidazo- lium iodide (Solaronix), 0.5M LiI (Aldrich), 0.05 MI2 (Aldrich), and 0.5M4-tert-butylpyridine (4-TBP, Aldrich) and 3-metoxy- propionitrilenitrile (3-MPN,Fluka) as a solvent.
Characteiization of SN-T1O2 soland SN-T1O2 thin film. To investigate tochange in crystallinity causedbyaging conditon, the crystallinity of the manufactured SN-TiO2 powder was characterizedbyanX-raydiffraction(Rigaku,D/MAX-1200) using a Cuka X-ray andNi filterat35kV and 15 mA. In addi tion, to probethe particlesize and formation process of TiO2
solparticle,the driedTiO2 solpowderswereinvestigated by transmissionelectron microscopy (HR-TEM,TECNAI,F20).
Moreover,the Fourier transform infrared spectrophotometer (FT-IR, Jasco, FT/IR-410) was used toanalyzemicrostructure changes in SN-TiO2 particle causebyagingconditions. The thickness and surfacemorphology of TiO2 thin film prepared from dried SN-TiO2 as well as the calcined SN-TiO2 were measured by field-emission scanning electron microscopy (FE-SEM, S-4700,Hitachi). Thecapacities of fabricated DSSCs and the current-voltage (I-V) curveswere measured using a source measure unitunder irradiation of white lightfrom a 1000W Xenon lamp(Thermo Oriel Instruments,USA).The incident light intensity and theactivecellarea were 100mW/
cm2 and 0.25cm2, respectively. TheI-Vcurveswereusedto calculate the short-circuit current (Isc), open-circuit voltage (V>c),fill factor (FF), and overall conversion efficiency(neff) of DSSCs. Thus, theelectrochemical impedancespectroscopy (EIS) measurementswere performed using the AC impedance (CHI 660A Electrochemical Workstation, USA) over the frequency ranging from 1 to 106 Hz with amplitudes of士 5 mV over the Voc.
Results and Discussion
Chaiacteiizationof SN-T1O2 sol particlesand SN-T1O2 thin film. Figures 2(a)〜(f) show HR-TEM images of SN-TiO2 sol particlesprepared under different aging times at two tem peratures.Figures 2(a) 〜 (c) presentthesizes and shapes ofsol particles according to theaging times(0, 24, 72 h) at lower 40 oC. At0 h, TiO2 sol particlesare 3〜5 nmlike a primary par
ticle.Ontheotherhand,after 24h, solparticles areaggregated tolargeparticles. Finally, after72 h, particlesarenecked and aggregated to thestructure of a bunch of grapes.Figures 2 (d)〜
(f)showtheaggregated particles undertheagingtemperature of higher90 oC. At0 h, manysolparticlesareaggregatedto larger bunchesof grapes than in the caseof Figure 1 (c).11,12 After 72 h at 90 oC, the average particlesize aggregated is about 50 nm showing the formationof continuousnecking of
Figure 2. HR-TEM image of SN-T1O2 sol particles according to the aging times (0 h (a,d), 24 h (b,e) and 72 h (c,f)) at 40 oC (left side) and at 90 oC (right side).
primary solparticles. These findings ledustoconclude that aging temperaturesrather than agingtimesare promoting the aggregation of particles. Inthe case of 72 h at 90 oC (right side), however,the particle size wasthebiggest,the color of TiO2 solwas opaque, and theviscosityincreasedmore than in thecase of 0hat 40 oC. We found fromthis result that the aging time and temperatureare importantprocess variables in con
trolling thesize particle size. The crystallinity of TiO2 sol par
ticles and calcined TiO2 particles was analyzed by XRD.
Figures3 (a, b) show theXRDpatterns of driedTiO2 sol pre
pared under different aging times attwo different tempera
tures of at 40 oCand 90 oC. Beforecalcination,all samples had thebrookitecrystalline phase mainly.With increasing aging timesat 40 oC, an increment in the brookitecrystallinephase was observed. Also, TiO2 sol prepared at 72 hand 90 oC (Figure 3 (b)) showsanintensitybrookitepeakat25o higher thanthat of theTiO2 solprepared at72 h and40 oC (Figure 3 (a)). It may besafetoconcludethat thecrystallinityincreased with theagingtemperatures and times. In addition, theXRD patterns of calcined SN-TiO2 (500 oC) particles prepared underdifferent aging conditions were investigated. Numerous changes wereobserved with theaging timesfor thesample of
40oC(Figure 2 (c)). With increasing aging times, TiO2 particles having rutile crystalline phase as (110) were changed into anatase crystallinephase as (101). For the sample of72h, a greater amount of rutilephasechanged to ananatasecrystal line phase with the small existence of brookite crystalline phase (121) and also small rutilecrystalline phaseas (110).
However, anatase crystalline phases for the samples (24h, 72 h) at 90 oC wasobservedmosltly (Figure 3 (d)). Figure 4 shows the result of FT-infrared spectra (FT-IR)for different aging temperatures of 40 oC and 90oC.Thetype and theamount of bonding group existing in the driedTiO2 sol particles were analyzed using the KBr method in the range of 400to 4,000 cm-1 wavelength number. Ingeneral,thecharacteristicabsor ption peaks of Ti-O-bending vibration and Ti-OHstretching vibration in thedriedTiO2 sol particlesappear at 500 cm-1and 1,630 cm-1. As expected, the unique vibrationpeak of Ti-O- bending and Ti-OH stretching was observed as shown in Figure 4, and the intensity of Ti-O-absorption peaknear 500 cm-1 increased with the progression of aging. In the result of FT-IRas shown Figure4(a),Ti-OH groups are largely formed withthe progressionof aging, andthen they turn intoTi-O-Ti bond from Ti-OHbondsbypolymerization/condensationas follows; Ti-OH+Ti-OH 一 Ti-O-Ti+ H2O.
On the other hand,allthechemical bondsreachthe equili
brium state in thecase of theaging of72 h.Eventhough there is nochemicalchange anymore,theparticles are aggregated byphysical bondsaccordingto aging. As aresult, crystalline phases are affectedby particle size eventhe calcinations of 500oC as showninFigure 3(c, d). Therefore,the bigger the particle sizes, the slower the transition of the crystalline phase. Thus, theaging is animportant process for the increase of particle size irrespective of chemical orphysicalprocesses.
Photoelectiic characteristics ofSN-T1O2 films.Theproper ties of the nanostructured TiO2 particle synthesized in this work were investigatedtoprepare for feasibleDSSCs. The role of aging conditionin the making of photoelectrode for DSSCs was assessed on the basic of photovoltaic perfor mance. Figure 5 shows the FE-SEM images of calcined SN-TiO2 thinfilms. Inthe caseof the samplesof different agingtimes(a,c) for 0 h,rutilephase with big particle size (>
50 nm) was mostlyobserved after calcination at 500 oC. On the otherhand, in the caseof 72 h (b, d), anatasephase with small particle size(av. 20 nm)was mostlyobserved under the same calcination temperature of 500oC. An examination of the photovoltaic performance of calcined SN-TiO2 particle films with respect to their aging conditions revealed the currentdensity (厶。),open circuitvoltage(Voc), fill factor(FF), and energy conversion efficiency (從任)as summarizedin
Table 1.When theaging times of 0h and 72 h at 40 oC are compared, the sample of 0hgave a low current density because of itsamount of rutilecrystallinephaserelatively larger than the anatasephase. This maybe attributedtothe fact thatthe dye adsorption capacity of thephotoelectrode consisting of high rutilecrystalline was lowerthanthat ofhigh anatase cry
stalline. Thus, the high current density of photoelectrode was notedinthecaseof thesample having high anatasecrystalline phase.13-15 With increasing agingtemperature,theefficiency increased from 3.96%to4.63%underthesameagingtime of
Soc. , 30, No. 10 Kyung-Jun al
8 0 7 0 6 0 5 0 4 0 3 0 2 0 0
(b)-
은 으 -
(d)-
은
으
-
3
A - S U
으으
8 0 7 0 6 0 5 0 4 0 3 0 2 0
40 r,24h
R(110)
40 r, 0 h A-
us
으-
0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80
20, deg. 20, deg.
⑺
% 一8
§
틍 흔
Figure 3. XRD patterns of dried SN-T1O2 sol particles (above) and calcined SN-T1O2 sol particles (bellow) with various aging time at 40 oC (a,c) and at 90 oC (b,d).
3750 3450 3150 2850 2550 2250 1950 1650 1350 1050 750 450
Wavenumbers, cm-1
⑶
% 一8
§
틍 드
」一
3750 3450 3150 2850 2550 2250 1950 1650 1350 1050 750 450
Wavenumbers, cm-1
Figure 4. FT-IR spectra of dried SN-TiO2 sol particles with various aging time at 40 oC (a) and at 90 oC (b).
72 h. Figure 6 showsthe Nyquist plots of calcined SN-TiO2
thin film with different agingtimes at 40oC. A large semi
circleatlowfrequencies and a small oneat highfrequencies wereobserved for threesolarcells.Responses in the frequency
regions 103 ~106, 1 ~103, and 0.1 ~1Hz are assignedtocharge transfer processes occurring atthePt/electrolyte interfaceand TiO2/dye/electrolyteinterface, in Nernstdiffusion withinthe electrolyte, respectively.TheresistanceattheTiO2/dye/elec-
Table 1. Photocurrent-voltage of DSSCs for calcined SN-T1O2 thin films under different aging times at 40 oC and 90 oC.
Type of temperature
Aging time
Isc, mA/cm2
Voc, V
Fill factor, FF neff,
%
0 h 6.68 0.64 0.62 2.64
40 oC 24 h 10.01 0.64 0.62 3.94
72 h 10.32 0.63 0.61 3.96
0 h 10.63 0.62 0.55 3.66
90 oC 24 h 12.56 0.60 0.54 4.06
72 h 12.77 0.62 0.59 4.63
Figure 5. FE-SEM image of calcined SN-TiO2 thin films with aging times (0 h (a,c) and 72 h (b,d)) at 40 oC (upside) and at 90 oC (down
side).
Table 3. Photocurrent-voltage of DSSCs using dried SN-TiO2 thin films depending on HPC binder concentrations.
HPC concentration in the paste
Isc, mA/cm2
Voc, V
Fill factor, FF
neff,
%
10% HPC 16.33 0.67 0.59 6.46
15% HPC 17.69 0.70 0.62 7.80
20% HPC 13.95 0.65 0.66 5.98
25% HPC 10.16 0.63 0.66 4.20
Figure 6. Nyquist plots of calcined SN-TiO2 thin film at aging time of 40 oC.
Table 2. Photocurrent-voltage of DSSCs using dried SN-TiO2 thin films depending on TiO2 content.
TiO2 content in the paste
Isc, mA/cm2
Voc, V
Fill factor, FF
neff,
%
5% SN-TiO2 5.61 0.62 0.63 2.19
10% SN-TiO2 8.56 0.63 0.63 3.41
15% SN-TiO2 11.81 0.63 0.62 4.58
20% SN-TiO2 14.15 0.64 0.57 5.16
25% SN-TiO2 8.18 0.63 0.63 3.24
trolyteinterface decreasedwithincreasingaging times, which result from the particle size and surfacemorphology as shown inFigures5 (a, b). Thisresultimpliesthat the electrontrans
port inthe rutile layeris slowerthan in the anatase layer.16 Similarresultswere reported in our preliminary studies that electron diffusioncoefficient (Dn) for therutile film is about one order of magnitude lowerthan that of theanatase film.
Determination ofOptimal Condition for HighEfficient DSSCs
Theoptimalcondition for thehighestefficiency of DSSCs
Figure 7. FE-SEM image of dried SN-TiO2 thin film: (a) surface, (b) thickness, and (c) thickness of repetitive dry coating.
Soc. , 30, No. 10 Kyung-Jun al
30
cm 25
20 Eo <E
"_ s u
①Q r
은」
n
° 15
10
5
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Potential, V
Figure 8. I-V curves of dried SN-T1O2 thin films with repetitive dry coating method fabricated using TiO2 colloidal sols prepared under optimal aging conditions of 90 oC and 72 h.
was chosen based on the experimental results. The optimal time and temperatureunder our experimentalconditionswas 72 hand 90 oC for thesynthesis of TiO2 colloidal sol. After calcination at 500 oC, TiO2 particles mainly showedanatase phase.Theaverage particle size wasinthe range of20 nm. To increase the efficiency ofDSSCwith high transparency,the optimal amount of thedried SN-TiO2 particle wasdetermined in the range of (5 〜25%) and 10% ofHPC as a binder. The resultsof conversionefficiencyare listedin Table 2. Although the highest efficiency (5.16%) for20% SN-TiO2 samplewas obtained, the fill factor was lowercompared with the other samples. Thus, we reducedthe TiO2 contents from 20% to 18%.Todetermine theoptimal amount of binder, the contents of HPCwasadjusted in therange of 10 〜25% for thesample of 18% SN-TiO2.As shown in Table3, a higher conversion effi
ciency of 7.80%was obtained when the contents of TiO2 and HPCwere 18% and 15%, respectively.Figure 6showstheFE- SEM image of SN-TiO2 thin film prepared fromthe SN-TiO2
colloidalsolobtained underthe optimal aging conditions(i.e., 72 h, 90 oC). As 아iown in Figure 7 (a), thedriedSN-TiO2 thin filmshowedsmallaggregated and homogeneouslydispersed particles. The average particlesizewasabout 20 nm and the thickness of thin film wasabout7 “m(Figure 7 (b)). For the manufacturing ofhighly efficient DSSC, wecoatedthe TiO2
filmagainusing therepetitivedrycoating method.17Figure 7 (c)showsthe FE-SEM imageof SN-TiO2 thin film. The thick
ness ofSN-TiO2 thinfilm is about 14“m. Onthebasis of the resultsobtained under our experimental conditions,themax imumenergyconversionefficiency wassetat10.31% (Figure 8)for theTiO2 thin filmspreparedfrom theTiO2 colloidal sol synthesized under the optimal aging conditions (i.e.,72 h, 90 oC).
Conclusions
TiO2 colloidal solwasprepared through theagingprocess accompaniedbyhydrolysis/poly condensation.We found from theanalysis of XRD,FT-IR and TEM that theaging step inthe colloidal sol preparation isthe mostimportant in controlling thesize and structure of TiO2 particles. It was alsoobserved surprisingly thatthe physical and chemicalproperties of the TiO2 particleswas not changedeven after thecalcinations of TiO2 sol. DSSCfabricated by the dried TiO2 sol particles (18%) andHPC binder (15%) witha repetitivedrycoatingmethod, hasa high conversion efficiency of 10.31% with high trans
parency.
Acknowledgments.This work wassupported bytheKorea ResearchFoundation Grant funded by the Korean Govern ment (MOEHRD)(KRF-2007-H00023) and the Ko^aInstitute of Geoscience and MineralResources ofSouth Korea.
References
1. Hagfeldt, A.; Gratzel, M. Acc. Chem. Res. 2000, 33, 269.
2. Hwang, K. J.; Yoo, S. J.; Jung, S. H.; Park, D. W.; Kim, S. I.; Lee, J. W. Bull. Korean Chem. Soc. 2009, 30(1), 172.
3. Kang, M. G.; Ryu, K. S.; Chang, S. H.; Park, N. G.; Hong, J. S.;
Kim, K. J. Bull. Korean Chem. Soc. 2004, 25(5), 742.
4. Yin, H.; Wada, Y; Kitamura, T.; Sumida, T.; Hasegawa, Y;
Yanagida, S. J. Mater. Chem. 2002, 12, 378.
5. Park, S. D.; Cho, Y. H.; Kim, W. W.; Kim, S. J. J. Solid State Chem. 1999, 146, 230.
6. Wang, C. C.; Ying, J. Y. Chem. Mater. 1999, 11, 3113.
7. Barbe, C. J.; Arendse, F.; Comte, P; Jirousek, M.; Lenzmann, F.;
Shklover, V.; Gratzel, M. J. Am. Ceram. Soc. 1997, 80(12), 3157.
8. Jung, H. S.; Lee, S. W.; Kim, J. Y; Hung, K. S.; Lee, Y. C.; Ko, K. H. J. Colloid Interface Sci. 2004, 279,479.
9. Yoldas, B. E. Amer. Ceram. Soc. Bull. 1975, 54(3), 289.
10. Hwang, K. J.; Yoo, S. J.; Kim, S. S.; Kim, J. M.; Shim, W. G.;
Kim, S. I.; Lee, J. W. J. Nanosci. andNanotechnol. 2008, 8,4976.
11. Brinker, C. J.; Scherer, G. W. Sol-Gel Science; Academic press:
San Diego, U. S. A., 1990; p 362.
12. Yoo, S. J.; Lee, S. I.; Kwak, D. H.; Kim, K. G.; Hwang, K. J.; Lee, J. W.; Hwang, U. Y. H.; Park, H. S.; Kim, J. O. Korean J. Chem.
Eng. 2008, 25(5), 1232.
13. Lee, J. W.; Hwang, K. J.; Shim, W. G.; Park, K. H.; Gu, H. B.;
Kwun, K. H. Korean J. Chem. Eng. 2007, 24(5), 847.
14. Lee, J. W.; Hwang, K. J.; Park, D. W.; Park, K. H.; Shim, W. G.;
Kim, S. C. J. Nanosci. and Nanotechnol. 2007, 7, 3717.
15. Park, D. W.; Park, K. H.; Lee, J. W.; Hwang, K. J.; Choi, Y. K. J.
J. Nanosci. and Nanotechnol. 2007, 7, 3722.
16. Park, N. G.; van de Lagemaat, J.; Frank, A. J. J. Phys. Chem. B 2000, 104, 8989.
17. Ito, S.; Kitamura, T.; Wada, Y.; Anagida, S. Sol. Energy Mater. Sol.
Cells 2003, 76, 3.
