Optik 157 (2018) 134–140

Contents lists available at ScienceDirect

Optik

j o ur nal ho me pa ge: www.elsev ier .de / i j leo

Full length article

Improved electron transfer of TiO2 based dye sensitized solarcells using Ge as sintering aid

Muhammad Shakeel Ahmad a, Nasrudin Abd Rahim a,b, A.K. Pandey a,∗

a UM Power Energy Dedicated Advanced Center (UMPEDAC), University of Malaya, 59990, Kuala Lumpur, Malaysiab Renewable Energy Research Group, King Abdulaziz University, Jeddah, 21589, Saudi Arabia

a r t i c l e i n f o

Article history:Received 19 September 2017Accepted 13 November 2017

Keywords:Charge transferDye sensitized solar cellGe nanoparticlesNanocompositeSintering

a b s t r a c t

Low electron mobility of TiO2 semi-conductor and inferior inter-particle contact facilitaterecombination reactions that leads to low performance of dye sensitized solar cells (DSSCs).To improve electron mobility at relatively low sintering temperatures, doping of germa-nium (Ge) nanoparticles with TiO2 have been trailed due to its excellent optoelectronic andlow temperature sintering properties. Anatase TiO2-Ge nanocomposites have been pre-pared by using colloidal suspension process and deposited on conducting glass using doctorblade technique. Four types of nanocomposites i.e. (1) TiO2–0.5 wt%Ge, (2) TiO2–2 wt%Ge,(3) TiO2–5 wt%Ge and (4) TiO2–10 wt%Ge have been prepared and sintered at 400 ◦C witha control specimen fabricated using pure TiO2 nanoparticles (sintered at 450 ◦C) for com-parison purpose. To investigate the morphological and structural characteristics, SEM andXRD have been employed. The UV-vis and impedance spectroscopy have been performedto observe light absorption and electron transfer characteristics respectively. Finally, spec-imens were tested for their photo conversion efficiency. An increase in electron transferability and conversion efficiency have been recorded with increase in Ge nanoparticles evenat 400 ◦C sinter temperature compared to reference TiO2 photoanodessintered at 450 ◦C.

© 2017 Elsevier GmbH. All rights reserved.

1. Introduction

Electron recombination reactions and light scattering ability along with dye pick up capacity are the main parametersto improve the performance of electrochemical dye sensitized solar cells (DSSCs) [1,2]. Efficient charge carrier transport(electrons) from the semiconductor nanostructure to the external load through transparent conducting oxide layer withoutrecombining with electrolyte or dye guarantee the suppression of recombination reactions which is still one of the mainreasons of less efficiency of DSSC device [3]. For efficient electron transfer, it is of utmost importance that the semiconductornanoparticles should be sintered to coble initial sintering stage where inter-particle contact area increases by neck growthfrom 0 to ∼0.2%. Sintering temperatures which are restricted to 400 and 450 ◦C cannot provide this much vibrational energyto increase the thermal entropy (ST) to the extent necessary to lower the surface/boundary energy (�gb) of nanoparticlesby 2/3 compared to solid/vapor surface energy (�sv) of the system. Without enough heat energy the Gibbs free energy of

surfaces will not allow the formation of necks and diffusion of surfaces for proper inter-particle contact [4].

Various approaches have been trialled to improve the incident photon conversion efficiency (IPCE) of DSSCs, such assize quantization [5–7], development of different nano-architectures [8–11] and doping with different cations and anions

∗ Corresponding author at: UM Power Energy Dedicated Advanced Center (UMPEDAC), University of Malaya, 59990, Kuala Lumpur, Malaysia.E-mail addresses: adarsh.889@gmail.com, adarsh@um.edu.my (A.K. Pandey).

https://doi.org/10.1016/j.ijleo.2017.11.0730030-4026/© 2017 Elsevier GmbH. All rights reserved.

[tfschji

p0itofhborp

2

2

EaFsfrapr

2

l4wrFrs

2

hXuPl(

3

T

M.S. Ahmad et al. / Optik 157 (2018) 134–140 135

12–14]. With improvement in light absorption capacity, reduction in recombination reactions and dye loading characteris-ics, the maximum claimed IPCE reached to 12–14% [15,16]. Recently, hetero-junction photoanodes have been investigatedor the efficient transport of the charge carriers from semiconductor oxide to the external load [17]. The idea is to use theo called step approach to facilitate charge transfer using low energy band gap materials such as SnO2 [18], CdS [19] etc,oupled with wide band gap semiconductors such as TiO2 and ZnO. Several groups have claimed improved IPCE throughetero-junctions such as ZnO/TiO2 [20], poly(phenylenevinylene)/C60 [21] and CdTe/TiO2 [22]. It is believed that the hetero-

unction facilitates the movement of charge carriers and suppresses the recombination reactions at the interface, resultingn high efficiency of the hetero-structured photoanodes.

Novel TiO2-Ge nanocomposites and sandwich structures have been developed recently for high performance thin filmhotovoltaic devices [23–25]. Bulk Ge has direct band gap of 0.8 eV (band gap of TiO2 is 3.2ev). And an indirect band gap of.66 eV [23] at room temperature with light absorption characteristics in near infrared spectrum. These characteristics make

t a potential candidate in the field of DSSCs. The effect of Ge quantum dots (QDs) has been studied on TiO2 layer to improvehe optical performance of TiO2 layer. In another related study [26], the effect of functionalized Ge (QDs) has been studiedn the performance of TiO2 based, N719 sensitized DSSC. The study was focused more on the dye loading ability of surfaceunctionalized Ge QDs. A 12% increase in IPCE compared to reference specimen has been reported which was attributed toigh dye loading and more negative conduction band edge of Ge compared to TiO2. In this study, the Ge nanoparticles haveeen employed as sintering aid in DSSCs and its ability to sinter at lower temperatures has been emphasized. Also the effectf second phase sintering mechanism has been discussed to increase inter-particle contact between nanoparticle to impedeecombination reactions. The TiO2-Ge nanocomposite with different wt% of Ge using colloidal suspension method has beenrepared and appliedon conducting glass substrate using doctor blade technique.

. Experimental

.1. Material preparation

Fluorine doped tin oxide (FTO) conducting glass substrates were washed with deionized (DI) water and sonicated (model:lmasonic PH350EL) for 30 min followed by rinsing with DI water. TiO2(supplier by Sigma Aldrich, size range 100–150 nm)nd Ge (supplied by sigma aldrich, size range: 100–150 nm) nanoparticles were weighed using digital weight balance (model:A2104J) and stirred using magnetic stirring (model: C-MAG HS 7) in acetone separately for two hrs with subsequentonication for 1hr and then stirred over night. The two suspensions were mixed together and again sonicated for 1hrollowed by stirring for 12 h. The mixture was then filtered using filter paper and dried at 80 ◦C for 4 h. The process wasepeated for the rest of the specimens. Four series of specimens i.e. (1) TiO2–0.5 wt%Ge, (2) TiO2–2 wt%Ge, (3) TiO2–5 wt%Gend (4) TiO2–10 wt%Ge have been prepared with a control specimen fabricated using pure TiO2 nanoparticles for comparisonurpose. It is worth mentioning here that Ge nanoparticles have been heated to 400 ◦C at ultra-high vacuum condition toemove oxide layer. Furthermore, all the procedures were performed in vacuum glove box.

.2. DSSC assembly

Doctor blade method has been employed to fabricate photoanode. Nanocomposite paste has been made using ethy-ene glycol and coated on to the FTO coated glass. The coated specimens were heated to 80 ◦C for 30 min and sintered at00 ◦Cfor 30 min in vacuum glove box. Pure TiO2photoanode was sinteredat 450 ◦C for comparison. The sinteredsubstratesere immersed in Ru(dcbpy)2(NCS)2 (dcbpy 1/4 2,2-bipyridyl-4,4-dicarboxylato) dye solution (535-bisTBA (N719), Sola-

onix) with acetonitrileover night and rinsed with ethanol. Counter electrodes were prepared using Ptsputtercoating onTO glass. Counter electrodes were drilled and joined with photoanode separated by a thin plastic sheet. Iodide/triiodideedox electrolyte was introduced through drilled holes and then the holes were sealed using hot melt sealant. At least fivepecimens for each wt% have been fabricated for high reproducibility.

.3. Characterization

Scanning electcron microscope (SEM model = Quanta FEG 450) was performed to examine surface morphology andomogeneity of nanoparticles. Secondary electron imaging (SEI) mode was used with an accelerating voltage of 1–5 kV.-ray diffraction (XRD) has been performed for compositional and phase analysis. UV–vis-Nir spectroscopy (modelv2600) has been performed to investigate light absorption characteristics. Electron impedance spectroscopy (EIS) (model:GSTAT101)has been employed to measure charge transfer ability. Finally auto-lab solar simulator (IVT 300) has been uti-

ized to measure I–V curves, open voltage potential (Voc), short circuit current (Jsc), fill factor (FF) and IPCE at AM1.5 sun1000 W/m2 irradiation) condition.

. Results and discussion

Specimens were weighed at each fabrication step i.e. before and after deposition of nanocomposite layer on FTO glass.able 1 is illustrating the weight of the specimens under investigation (weight of deposited layer has been measured after

136 M.S. Ahmad et al. / Optik 157 (2018) 134–140

Table 1Weights and dye loading measurements of specimens.

Sr. No. Specimen ID Weight of FTO glass (g) Weight of deposited FTO glass (g) Weight of deposited layer (g) Dye loading (x10−7/cm2)

1 TiO2 0.888 ± 0.001 0.977 ± 0.001 0.089 ± 0.001 1.022 TiO2-0.5%Ge 0.913 ± 0.001 0.995 ± 0.001 0.082 ± 0.001 1.143 TiO2-2%Ge 0.878 ± 0.001 0.974 ± 0.001 0.094 ± 0.001 1.334 TiO2-5%Ge 0.821 ± 0.001 0.901 ± 0.001 0.080 ± 0.001 1.745 TiO2-10%Ge 0.916 ± 0.001 1.002 ± 0.001 0.086 ± 0.001 1.56

Fig. 1. SEM micrographs of un-sintered and sintered raw TiO nanoparticles (a) low magnification image of coating (b) high magnification un-sintered

2

TiO2 nanoparticles and (c) sintered TiO2 nanoparticles.

sintering at 400 ◦C for TiO2-Ge nanocomposite photoanodes and 450 ◦C for reference TiO2 photoanode). For dye adsorptionthe weight method is not reliable that is why adsorption-desorption method has been performed in the present study.NaOH-ethanol solution was used to desorb the dye loaded on the photoanodes [27].

With increase in wt% of Ge in TiO2 matrix, amount of dye adsorption increased which can be due to lower surface energiesand oxygen vacancies of sintered nanoparticles. Further increase to 10 wt% cause reduction is dye pick up due to excessivenecking and less surface area.

3.1. Morphological study

Fig. 1 is showing SEM images of coated reference TiO2 nanoparticles before and after sintering at 450 ◦C. Particle sizeranges between 100–150 nm and possess irregular shapes. Low magnification TiO2 coating can be seen in Fig. 1a, Inter-particle connects have been improved in sintered coating (Fig. 1c) compared to un-sintered coating (Fig. 1b) due to boundarydiffusion at high temperatures.

Fig. 2 is showing TiO2-Ge nanocomposite coating of specimens before and after sintering at 400 ◦C. Homogenous andagglomeration free mixing of un-sintered nanocomposite photoanodes can be observed in Fig. 1(a,a1-3). Backscatteredimages of specimens (Fig. 1(c,c1-3)) have also been taken to visualize homogenous mixing. Fig. 1(b,b1-3) is showing sinteredTiO2-Ge nanocomposite coatings at 400 ◦C. It can be clearly observed that with the introduction of Ge in TiO2 matrix, theinter-particle contact has been improved (shown using circles) compared to reference TiO2 photoanode specimens sinteredat 450 ◦C (Fig. 1). Furthermore, an increasing trend in inter-particle contact has been observed (Fig. 2(b1-3)) with incrementin the wt% of Ge in TiO2 matrix.

This increase in inter-particle contact (Fig. 2) can be explained using the process of sintering of two different elements ascan be seen in Fig. 3. Fig. 3 is showing development of inter-particle contact between nanoparticles. Fig. 3(a(i)) is illustratingsurface contact between TiO2 nanoparticle and Fig. 3(a(ii)) is showing surface contact between TiO2& Ge nanoparticlesalong with inter-atomic arrangements illustrated in Fig. 3(b(i,ii)). The melting point of TiO2 and Ge is approximately 1843 ◦Cand 938 ◦C respectively. As the sintering temperature is assumed to be 0.5–0.9 of the melting temperature, at 400 ◦C Genanoparticles lose more surface/boundary energy (�gb) compared to TiO2 nanoparticles which facilitates necking betweenthe particles by introducing capillary forces [28,29]. The addition of capillary forces due to low melting point second phasei.e. Ge along with heat energy improved the inter-particle contact which is a decisive parameter to increase the efficiencyof DSSC.

X-ray diffraction (XRD) analysis has also been performed (Fig. 4) to observe any effect of capillary forces on the structureof TiO2. Fig. 4 showed no any observable structural changes in TiO2 matrix. Anatase phase is a dominant phase which ishighly required for DSSC applications due to its superior optoelectronic properties compared to rutile phase. It can also beobserved that the oxide layer of GeO2 has been successfully removed using thermal treatment.

The optical characterization of three constituent materials has also been performed (Fig. 1–3s (supplementary material)).

The optical absorption edge of TiO2 lies in UV region while Ge showed optical absorption in near infrared region (theabsorption edge starts from 900 nm). N719 dye showed optical absorption in visible region of the spectrum.

M.S. Ahmad et al. / Optik 157 (2018) 134–140 137

FbT

3

bGwwwr

ore(om

ig. 2. SEM micrographs of TiO2-Ge nanocomposites (a, a1,2,3) un-sintered TiO2–0.5 wt%Ge, TiO2–2 wt%Ge, TiO2–5 wt%Ge & TiO2–10 wt%Ge respectively (b,1,2,3) Sintered TiO2–0.5 wt%Ge, TiO2–2 wt%Ge, TiO2–5 wt%Ge & TiO2–10 wt%Ge respectively and (c, c1,2,3) backscattered SEM images of TiO2–0.5 wt%Ge,iO2–2 wt%Ge, TiO2–5 wt%Ge & TiO2–10 wt%Ge respectively.

.2. Electron transfer

To study the effect of Ge on the charge transfer capability of TiO2 matrix, EIS has been performed which is supposed toe a powerful tool to study electron transfer resistance in materials. Fig. 5 is showing impedance spectroscopy of the TiO2-e nanocomposites along with band energy diagram. The charge transfer efficiency has been improved with increment int% of ge nanoparticles. With the addition of 0.5 wt% Ge, the charge transfer resistance reduced, which is further reducedith the addition of 2 wt% Ge nanoparticles. Furthermore, a further decrease in recombination reaction has been measuredith the addition of 5 wt% Ge nanoparticles in TiO2 matrix, further addition of Ge nanoparticles to 10 wt% cause increase in

ecombination reactions which is attributed to high concentration of impurities and surface defects.Two operative mechanisms i.e. high inter-particle connection due to capillary forces and thermodynamic suitability

f hetero-structure carrier pathway have improved the charge transfer ability of photoanode material. The neck formationeduced the boundary electron transfer resistance due to more or less ordered atomic arrangement. Another reason is suitablelectronic path of TiO2-Ge nanocomposite which is thermodynamically more suitable compared to reference TiO2 specimen

Fig. 5(b)). Excited electrons from lower unoccupied molecular orbit (LUMO) of N719 dye are injected to the conduction bandf Ge nanoparticles before TiO2 due to the slightly negative conduction band edge of Ge. This hetero-junction arrangementinimizes the recombination reactions and in turn improves the current density of DSSC.

138 M.S. Ahmad et al. / Optik 157 (2018) 134–140

Fig. 3. Sintering mechanism of particles (a(i)) surface diffusion between TiO2 nanoparticle, (a(ii)) surface diffusion between TiO2& Ge nanoparticles, (b(i))absence of atomic arrangement at interface and (b(ii)) ordered atomic arrangement in neck area.

Fig. 4. XRD analysis of as received raw materials and sintered TiO2–10wt%Ge nanocomposite.

Fig. 5. Electrochemical impedance analysis of specimens (a) EIS and (b) Energy band diagram.

M.S. Ahmad et al. / Optik 157 (2018) 134–140 139

Fig. 6. Photocurrent-voltage characteristics of DSSC fabricated using TiO2-Ge nanocomposite photoanodes and compared with reference TiO2 photoanode.

Table 2Summary of detailed parameters of specimens.

Sr. No. Specimen ID Voc (V) Jsc (mA/cm2) FF (%) EFF (%)

1 TiO2 0.63 9.43 73.8 4.032 TiO2-0.5%Ge 0.63 10.38 74.4 4.42

3

Dawwvoii

otaoaeTtis

4

SHDdiiaI

3 TiO2-2%Ge 0.66 11.32 73.7 4.974 TiO2-5%Ge 0.80 12.73 74.7 5.655 TiO2-10%Ge 0.72 12.07 74.1 5.22

.3. I–V curves

Photocurrent verses voltage (I–V) curves have been studied to understand the overall photo conversion properties ofSSCs assemble using TiO2-Ge nanocomposite photoanodes. Fig. 6 is showing I–V curves of specimens under investigationnd other parameters i.e. Voc, Jsc, FF and efficiency are listed in Table 2. An increase in overall performance can be observedith increment in Ge nanoparticles. TiO2–0.5 wt%Ge specimen showed 10.3 mA short current density with IPCE of 4.42%hich is 9.6% higher than the reference TiO2 photoanode. Specimens containing 2 wt% Ge nanoparticles showed 4.97% IPCE

alue which is 23.3% higher than the reference specimens. With the addition of 5 wt% Ge nanoparticle, an increase in IPCEf 5.65% has been recorded which is 40% higher than the reference specimens. Further addition of Ge nanoparticles didn’t

mprove the IPCE rather the efficiency dropped with further addition up to 10 wt% Ge. This is due to high concentration ofmpurity levels which produces defects and electron trapping sites [30].

High current density in nanocomposite specimens is due to high dye loading and high electrical conduction characteristicsf nanocomposite photoanodes. Ge nanoparticles possess more electro-negativity value (1.85) compared to TiO2 (1.45), andhe bond energy of Ge with oxygen (O)atoms is approximately 6657.4 KJ/mol which is double the bond energy of Ti and Otoms (314.5 kJ/mol)[31]. This change in bond energy may have disturbed the bonds between Ti and O atoms and producexygen vacancy which caused higher dye anchoring in nanocomposite photoanode [32,33]. High incident photo energybsorption of nanocomposite photoanodes facilitates dye molecules to absorb high amount of energy and inject morelectrons to conduction band of photoanode. Furthermore, an increase in Voc can be due to upward shift of Fermi level ofiO2 with the addition of Ge nanoparticles [34]. Increased oxygen vacancies, high light absorption and improved electronransfer of TiO2-Ge nanocomposite photoanodes improved the overall IPCE of the DSSC. This investigation revealed that Ges a promising additive in TiO2 photoanode for various types of nanostructured solar devices and photo-electrochemicalystems.

. Conclusions

In this study, TiO2-Ge nanocomposite photoanode materials have been investigated for their performance in DSSCs.pecimens were prepared using colloidal suspension method with 0.5 wt%Ge, 2 wt%Ge, 5 wt%Ge and 10 wt%Ge in TiO2 matrix.igh inter-particle contact at relatively lower temperature has been achieved which improved the commercial viability ofSSC technology. Morphological studies revealed that particles had developed necks on the surfaces due to high surfaceiffusion. Light absorption and impedance spectra showed high light absorption and electron transfer capability. An increase

n overall photo conversion efficiency of 40% compared to reference TiO2 has been measured when doped with 5wt%Ge.Thismprovement can be co-related with high light absorption ability, increased electron transfer ability due to neck formationnd hetero-junction symmetry and high dye loading capacity. Further addition of Ge nanoparticles up to 10 wt% reduces thePCE mainly due to high concentration of impurity levels and reduction in dye loading capacity due to excessive necking.

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[[[[

[[

[

[

140 M.S. Ahmad et al. / Optik 157 (2018) 134–140

Acknowledgements

The authors thank the technical and financial assistance of UM Power Energy Dedicated Advanced Centre (UMPEDAC) andthe Higher Institution Centre of Excellence (HICoE) Program Research Grant, UMPEDAC – 2016 (MOHE HICOE – UMPEDAC).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, athttps://doi.org/10.1016/j.ijleo.2017.11.073.

References

[1] M.S. Ahmad, A. Pandey, N.A. Rahim, Advancements in the development of TiO 2 photoanodes and its fabrication methods for dye sensitized solar cell(DSSC) applications. A review, Renew. Sustain. Energy Rev. 77 (2017) 89–108.

[2] H. Nejadasad, A. piri, H. Zarei, The effect of carbon nanotubes on the efficiency of dye sensitized solar cells based on TiO2 nanorods, Opt. Int. J. LightElectron. Opt. 142 (2017) 211–217.

[3] A. Hegazy, N. Kinadjian, B. Sadeghimakki, S. Sivoththaman, N.K. Allam, E. Prouzet, TiO2 nanoparticles optimized for photoanodes tested in large areaDye-sensitized solar cells (DSSC), Sol. Energy Mater. Sol. Cells 153 (2016) 108–116.

[4] M. Asadikiya, C. Rudolf, C. Zhang, B. Boesl, A. Agarwal, Y. Zhong, Thermodynamic modeling and investigation of the oxygen effect on the sintering ofB4C, J. Alloys Compd. (2017).

[5] S. Rani, P. Suri, P. Shishodia, R. Mehra, Synthesis of nanocrystalline ZnO powder via sol–gel route for dye-sensitized solar cells, Sol. Energy Mater. Sol.Cells 92 (2008) 1639–1645.

[6] M. Rashad, A. Shalan, M. Lira-Cantu, M. Abdel-Mottaleb, Synthesis and characterization of mesoporous anatase TiO2 nanostructures via organic acidprecursor process for dye-sensitized solar cells applications, J. Ind. Eng. Chem. 19 (2013) 2052–2059.

[7] G. Benkö, B. Skårman, R. Wallenberg, A. Hagfeldt, V. Sundström, A.P. Yartsev, Particle size and crystallinity dependent electron injection influorescein 27-sensitized TiO2 films, J. Phys. Chem. B 107 (2003) 1370–1375.

[8] A.M. Bakhshayesh, M.R. Mohammadi, H. Dadar, D.J. Fray, Improved efficiency of dye-sensitized solar cells aided by corn-like TiO2 nanowires as thelight scattering layer, Electrochim. Acta 90 (2013) 302–308.

[9] W.-Q. Wu, H.-S. Rao, Y.-F. Xu, Y.-F. Wang, C.-Y. Su, D.-B. Kuang, Hierarchical oriented anatase TiO2 nanostructure arrays on flexible substrate forefficient dye-sensitized solar cells, Sci. Rep. 3 (2013) 1892.

10] Y. Qiu, W. Chen, S. Yang, Double-layered photoanodes from variable-size anatase TiO2 nanospindles: a candidate for high-efficiency dye-sensitizedsolar cells, Angew. Chem. 122 (2010) 3757–3761.

11] H.M. Upadhyaya, S. Senthilarasu, M.-H. Hsu, D.K. Kumar, Recent progress and the status of dye-sensitised solar cell (DSSC) technology withstate-of-the-art conversion efficiencies, Sol. Energy Mater. Sol. Cells 119 (2013) 291–295.

12] S.G. Kim, M.J. Ju, I.T. Choi, W.S. Choi, H.-J. Choi, J.-B. Baek, H.K. Kim, Nb-doped TiO 2 nanoparticles for organic dye-sensitized solar cells, RSC Adv. 3(2013) 16380–16386.

13] J. Zhang, W. Peng, Z. Chen, H. Chen, L. Han, Effect of cerium doping in the TiO2 photoanode on the electron transport of dye-sensitized solar cells, J.Phys. Chem. C 116 (2012) 19182–19190.

14] A. Latini, C. Cavallo, F.K. Aldibaja, D. Gozzi, D. Carta, A. Corrias, L. Lazzarini, G. Salviati, Efficiency improvement of DSSC photoanode by scandiumdoping of mesoporous titania beads, J. Phys. Chem. C 117 (2013) 25276–25289.

15] S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B.F. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M.K. Nazeeruddin, M. Grätzel,Dye-sensitized solar cells with 13% efficiency achieved through the molecular engineering of porphyrin sensitizers, Nat. Chem. 6 (2014) 242–247.

16] K. Kakiage, Y. Aoyama, T. Yano, K. Oya, J.-i. Fujisawa, M. Hanaya, Highly-efficient dye-sensitized solar cells with collaborative sensitization bysilyl-anchor and carboxy-anchor dyes, Chem. Commun. 51 (2015) 15894–15897.

17] C.W. Kim, S.P. Suh, M.J. Choi, Y.S. Kang, Y.S. Kang, Fabrication of SrTiO3–TiO2 heterojunction photoanode with enlarged pore diameter fordye-sensitized solar cells, J. Mater. Chem. A 1 (2013) 11820–11827.

18] B. Li, L. Luo, T. Xiao, X. Hu, L. Lu, J. Wang, Y. Tang, Zn2 SnO4–SnO2 heterojunction nanocomposites for dye-sensitized solar cells, J. Alloys Compd. 509(2011) 2186–2191.

19] J.H. Park, S.Y. Jung, R. Kim, N.-G. Park, J. Kim, S.-S. Lee, Nanostructured photoelectrode consisting of TiO2 hollow spheres for non-volatileelectrolyte-based dye-sensitized solar cells, J. Power Sources 194 (2009) 574–579.

20] R.T. Ako, D.S.U. Peiris, P. Ekanayake, A.L. Tan, D.J. Young, Z. Zheng, V. Chellappan, DSSCs with ZnO@TiO2 core–shell photoanodes showing improvedVoc: modification of energy gradients and potential barriers with Cd and Mg ion dopants, Sol. Energy Mater. Sol. Cells 157 (2016) 18–27.

21] H. Xin, G. Ren, F.S. Kim, S.A. Jenekhe, Bulk heterojunction solar cells from poly (3-butylthiophene)/fullerene blends: in situ self-assembly ofnanowires, morphology, charge transport, and photovoltaic properties, Chem. Mater. 20 (2008) 6199–6207.

22] J.A. Seabold, K. Shankar, R.H. Wilke, M. Paulose, O.K. Varghese, C.A. Grimes, K.-S. Choi, Photoelectrochemical properties of heterojunction CdTe/TiO2

electrodes constructed using highly ordered TiO2 nanotube arrays, Chem. Mater. 20 (2008) 5266–5273.23] X. Li, F. He, G. Liu, Y. Huang, C. Pan, C. Guo, Fabrication of Ge quantum dots doped TiO2 films with high optical absorption properties via

layer-by-layer ion-beam sputtering, Mater. Lett. 67 (2012) 369–372.24] S. Abe, Y. Adachi, K. Matsuda, M. Nose, Single-phase anatase structure and dominant metallic Ge in Ge/TiO2 multi-layer films using a differential

pumping co-sputtering system, Thin Solid Films 562 (2014) 104–108.25] A.F. Khan, M. Mehmood, T. Ali, H. Fayaz, Structural and optical studies of nanostructured TiO2–Ge multi-layer thin films, Thin Solid Films 536 (2013)

220–228.26] C.H. Kim, E.S. Ha, H. Baik, K.J. Kim, Treatment of TiO2 with COOH-Functionalized germanium nanoparticles to enhance the photocurrent of

Dye-Sensitized solar cells, Chem. An Asian J. 6 (2011) 850–855.27] W. Wang, Y. Liu, J. Sun, L. Gao, Nitrogen and yttrium co-doped mesoporous titania photoanodes applied in DSSCs, J. Alloys Compd. 659 (2016) 15–22.28] L.C. DEJONGHE, M.N. RAHAMAN, 4.1 sintering of ceramics, handbook of advanced ceramics: materials, applications, Process. Prop. (2003) 187.29] W.D. Kingery, Densification during sintering in the presence of a liquid phase. I. Theory, J. Appl. Phys. 30 (1959) 301–306.30] M. Shakeel Ahmad, A.K. Pandey, N.A. Rahim, Towards the plasmonic effect of Zn nanoparticles on TiO2 monolayer photoanode for dye sensitized

solar cell applications, Mater. Lett. 195 (2017) 62–65.31] Y.-R. Luo, J. Kerr, Bond dissociation energies, in: CRC Handbook of Chemistry and Physics, 2012, pp. 89.32] G. Gibbs, M. Boisen, F. Hill, O. Tamada, R.T. Downs, SiO and GeO bonded interactions as inferred from the bond critical point properties of electron

density distributions, Phys. Chem. Miner. 25 (1998) 574–584.33] T. Su, Y. Yang, Y. Na, R. Fan, L. Li, L. Wei, B. Yang, W. Cao, An insight into the role of oxygen vacancy in hydrogenated TiO2 nanocrystals in the

performance of dye-Sensitized solar cells, ACS Appl. Mater. Interfaces 7 (2015) 3754–3763.34] M.-Y. Lu, C.-Y. Tsai, H.-A. Chen, Y.-T. Liang, K.-P. Chen, S. Gradecak, S. Gwo, L.-J. Chen, Plasmonic enhancement of Au nanoparticle—embedded

single-crystalline ZnO nanowire dye-sensitized solar cells, Nano Energy 20 (2016) 264–271.