Journal of Applied Electrochemistry

, Volume 46, Issue 1, pp 47–58 | Cite as

Efficiency enhancement in plasmonic dye-sensitized solar cells with TiO2 photoanodes incorporating gold and silver nanoparticles

  • M. A. K. L. Dissanayake
  • J. M. K. W. Kumari
  • G. K. R. Senadeera
  • C. A. Thotawatthage
Research Article
Part of the following topical collections:
  1. Solar Cells


Plasmonic dye-sensitized solar cells (DSSCs) were fabricated using TiO2 photoanodes incorporating gold nanoparticles (AuNPs) with 15–20-nm size and silver nanoparticles (AgNPs) with 40–60-nm size. These were characterized by UV–Vis spectroscopy, JV characteristic, IPCE spectroscopy, EIS analysis and dark I–V measurements. Under the illumination of 100 mW cm−2 (AM 1.5), the efficiency of the reference DSSC without Au and Ag NPs in TiO2 was 5.12 %, while the efficiencies of plasmonic DSSCs with TiO2:AuNP and TiO2:AgNP were 6.23 and 6.51 %, respectively, representing an efficiency enhancement by 21.6 % for AuNPs and 27 % for AgNPs. The increased efficiencies of the two plasmonic DSSCs appear to be due to the increased short-circuit photocurrent density by enhanced light harvesting caused by the localized surface plasmon resonance effect. The IPCE spectra of the two plasmonic DSSCs suggest the narrowing of the energy band gap of TiO2 due to the presence of Au and Ag nanoparticles.


Dye-sensitized solar cells Efficiency enhancement TiO2 photoanodes Gold and silver nanoparticles Surface plasmon resonance effect 

1 Introduction

Since the dye-sensitized solar cell (DSSC) was first reported by Grätzel et al. [1], it has been a focus of intensive research activity aimed at developing low-cost, high-performance solar cells. In the past few decades, this has been achieved largely by developing better sensitizers [2, 3, 4], novel electrolytes and improved photoanodes [5, 6, 7, 8, 9]. Recently, some unconventional approaches for enhancing dye-sensitized solar cell performance have also been actively demonstrated. In particular, the effect of surface plasmons, arising from noble metal nanoparticles (such as Au and Ag), has been used to increase the optical absorption and photocurrent in plasmonic DSSCs [10, 11, 12, 13, 14, 15].

Gold and silver nanoparticles have been extensively used for applications both in biology (e.g., bio-imaging) and technology (e.g., photonics) due their unique optical properties. These properties are conferred by the interaction of light with electrons on the nanoparticle surface. At a specific wavelength of light, collective oscillations of electrons on the nanoparticle surface cause a phenomenon called Local Surface Plasmon Resonance (LSPR) resulting in strong extinction of light (absorption and scattering). The particular wavelength, or frequency, of light where this occurs is strongly dependent on the nanoparticle type, size, shape, surface, and agglomeration state. The surface plasmon resonance effect is manifested by the constructive interference between electromagnetic light fields and the surface plasmons of metal NPs. The LSPR effect amplifies and intensifies the electromagnetic fields near the metal NPs, leading to the precise control of optical fields and the enhanced absorption band in the UV–Vis spectrum. The use of the plasmonic effect for light trapping, therefore, provides an efficient way to make the full use of the dyes for enhancing solar cell efficiency.

In the case of DSSCs, Ihara et al. demonstrated an enhanced absorption coefficient of the Ru-dye using an Ag island film [16]. Jeong et al. investigated the photocurrent improvement due to the increased effective absorption cross section of the dye as a result of the neighboring Ag NPs [17]. More recently, various plasmonic structures have been used to increase the photovoltaic (PV) performance of DSSCs [18, 19, 20, 21]. Extending the LSPR effect further, plasmon-enhanced light absorption of the dye and hence the increased photocurrent have been demonstrated in DSSCs by incorporating metal (Au, Ag) particles and more often by introducing the metal@TiO2 [22], metal@SiO2 [23], or metal@SiO2@TiO2 core–shell structures [24] into flat or porous TiO2 anode.

In the present study, Au and Ag NPs were synthesized using conventional citrate reduction method and incorporated into TiO2 photoanode. The volume effects of Au and Ag NPs on the PV performance of the plasmonic DSSCs were studied in detail. The UV–Vis absorption spectra and incident photon to current conversion efficiency (IPCE) spectra were measured. The results of the electrochemical impedance spectroscopy (EIS), Dark I–V characteristics, and Flat Band Potential measurements were analyzed to support the short-circuit current density enhancement caused by the surface plasmon effect in the plasmonic DSSCs.

2 Experimental

2.1 Preparation of gold and silver nanoparticle suspensions

Au and Ag nanoparticles can be synthesized using the citrate reduction method. According to the literature, this method produces gold NPs with diameters ranging from 15 to 20-nm and relatively larger-sized silver nanoparticles with diameters ranging from 40 to 60 nm. In this study, gold nanoparticles were prepared following the procedure described by Tabrizi et al. [25] and Huang et al. [26]. Hydrogen tetrachloroaurate (III) and trisodium citrate purchased from Fluka were used as starting materials. A 20-ml solution of 1 mM of HAuCl4 was prepared by adding distilled water, and the solution was boiled under continuous stirring. 1 % Na3C6H5O7 solution was prepared by dissolving 0.1 g of the salt in 10 ml of distilled water. 2 ml of this solution was added to the boiling solution of HAuCl4. The color of the solution mixture gradually changed from blue to deep red. The heating was stopped when the solution has turned deep red (Fig. 1). The mechanism of the chemical reaction could be expressed as follows:
$$2 {\text{HAuCl}}_{ 4} + {\text{ 3C}}_{ 6} {\text{H}}_{ 8} {\text{O}}_{ 7} \to {\text{ 2Au }} + {\text{ 3C}}_{ 5} {\text{H}}_{ 6} {\text{O}}_{ 5} + {\text{ 8HCl }} + {\text{ 3CO}}_{ 2} .$$
Fig. 1

Photographs of (a) Colloidal Au NPs (deep red color) and (b) Colloidal Ag NPs (orange color) prepared in this work. (Color figure online)

Silver nanoparticles were prepared following the procedure described by Chang et al. [27] and Ratyakshi et al. [28]. Silver nitrate was purchased from BDH and trisodium citrate was purchased from Fluka. 1 mM of AgNO3 25 ml solution was prepared by adding distilled water and the solution was boiled with continuous stirring. 1 % Na3C6H5O7 solution prepared by dissolving 0.1 g in 10 ml of distilled water and 2.5 ml of this solution was added to the boiling solution of AgNO3. The solution mixture gradually changed from colorless to yellow. The heating was stopped when the solution has turned yellow (Fig. 1). The mechanism of the chemical reaction could be expressed as follows:
$$4 {\text{Ag}}^{ + } + {\text{ C}}_{ 6} {\text{H}}_{ 5} {\text{O}}_{ 7} {\text{Na}}_{ 3} + {\text{ 2H}}_{ 2} {\text{O}} \to 4 {\text{Ag }} + {\text{ C}}_{ 6} {\text{H}}_{ 5} {\text{O}}_{ 7} {\text{H}}_{ 3} + {\text{ 3Na}}^{ + } + {\text{ H}}^{ + } + {\text{ O}}_{ 2} \uparrow .$$

2.2 Optical absorption measurements on colloidal solutions

Optical absorption spectra of the Au NP and Ag NP colloidal solutions were taken with a Shimadzu 2450 UV–Vis spectrophotometer in the wavelength range from 300 nm to 800 nm.

2.3 TiO2 electrode preparation and characterization

Nanocrystalline TiO2 photoelectrodes were prepared by using the following procedure. 0.20 g of TiO2 powder (Degussa P-25) was ground for 30 min with 12 drops (about 0.15 g) of glacial acetic acid, one drop (about 0.02 g) of Triton X-100 and about 2 ml of ethanol until the mixture became a creamy paste. Appropriate amounts of Au nanoparticle colloidal suspension (by volume) were added to the TiO2 paste and ground again for 15 min to obtain a homogeneous material. The same procedure was repeated by adding Ag nanoparticle colloidal suspension to obtain TiO2 paste with Ag nanoparticles. The paste was doctor bladed on a precleaned Fluorine-doped conducting Tin Oxide (FTO) glass (Solaronix sheet glass 8 Ω/sq) keeping the active cell area of 0.25 cm2. In order to compare the effects of Au and Ag nanoparticles in the TiO2 films, pristine TiO2 films were also prepared without adding Au and Ag nanoparticles. FTO/TiO2 plates were sintered at 450 °C for 45 min and slowly cooled down to room temperature. They were subsequently dipped in an ethanolic dye (0.3 Mm) solution containing Ruthenium dye N 535 [RuL2 (NCS)2: 2TBA where, L = 2,2′-bipyridyl-4′, dicarboxylic acid; TBA = tetrabutyl ammonium] for 24 h [5].

The XRD spectra of the TiO2, TiO2 Au NP, and TiO2 Ag NP were taken in order to confirm the presence of Au NPs and Ag NPs in the respective plasmonic photoanodes.

2.4 Plasmonic solar cell fabrication and I–V characterization

The electrolyte solution for the DSSCs was prepared by adding 0.738 g of tetrapropyl ammonium iodide (Pr4NI) and 0.060 g of I2 to a precleaned 10-ml volumetric flask containing 3.6 ml of molten (MP 40 °C) ethylene carbonate (EC) and 1.0 ml of acetonitrile. The solution mixture was stirred overnight. DSSCs were fabricated by sandwiching the above electrolyte solution containing I/I3 redox couple in the configuration FTO/TiO2/dye/electrolyte/Pt/FTO. The photocurrent–voltage (I–V) characteristics of the cells were measured under the illumination of 100 m W cm−2 (AM 1.5) simulated sunlight using a homemade computer-controlled setup coupled to a Keithley 2000 multi-meter and a potentiostat/galvanostat HA-301. A Xenon 500 W lamp was used with AM 1.5 filter to obtain the simulated sunlight with above intensity.

2.5 Flat band potential measurement

Flat band potential of the photoanodes can be used to understand the movement of the Fermi Level of the metal NP-incorporated TiO2 electrodes with respect to the reference photo electrode. In order to find the flat band potential of the three TiO2 electrodes, Mott-Schottky measurements were taken by immersing TiO2, TiO2–Au NP, and TiO2–Ag NP electrodes in a 0.5 M Na2SO4 electrolyte solution, one at a time. The stabilized voltage was measured with respect to a standard calomel electrode (SCE) using a Metrohm Autolab PGSTAT 128 N potentiostat. The scanned frequencies were 1.0 kHz and 2.0 kHz.

2.6 IPCE and optical absorption measurements

IPCE is defined as the number of electrons in the external circuit produced by an incident photon at a given wavelength divided by the number of incident photons. IPCE measurements were taken for the above three DSSC systems, (A) TiO2/dye/liquid electrolyte/Pt, (B) TiO2–Au NP/dye/liquid electrolyte/Pt, and (C) TiO2–Ag NP/dye/liquid electrolyte/Pt. Experimental setup consisted of monochromatic light illumination from a Bentham PVE300 unit equipped with a TMC 300 monochromator-based IPCE system with a 150 W xenon arc lamp covering the 300–800-nm wavelength range. A calibrated Si photo detector (type DH) was used as the reference. For comparison purposes, the UV–Vis absorption spectra of the metal NP incorporated TiO2 powder samples of the three dyed photoanode materials was also taken. For these measurements, a paste of (a) TiO2, (b) TiO2–Au NP, or (c) TiO2–Ag NP was applied on FTO glass plate. After sintering at 450 °C for 45 min., the powder was scratched off from glass plates. 0.1 g of each powder was dipped in 2 ml of Ru dye solution. After leaving for 24 h for dye loading, these dyed powder samples were washed and dried, and then used to get the optical absorption curves. UV–Vis Absorption spectrum was also taken for powder samples of TiO2, TiO2–Au NP, and TiO2–Ag NP without dye for the determination of the corresponding energy band gap values. All optical absorption measurements were done using a Shimadzu 2450 UV–Vis spectrophotometer in the wavelength range of 300–800 to nm.

2.7 EIS measurements

In order to see the effects of metal NP-incorporated TiO2 on the electron life time and the interfacial charge-transfer resistance, R 2CT, electrochemical Impedance Spectroscopy (EIS) measurements were performed on the DSSCs using a Metrohm Autolab Potentiostat/Galvanostat PGSTAT 128 N with a FRA 32 M Frequency Response Analyzer (FRA) covering the 1 MHz–0.01 Hz frequency range. These measurements were carried out under the illumination of 100 mW cm−2 using the same solar simulator that was used for I–V measurements.

2.8 Dark I–V measurements

Dark I–V measurements were performed for the three different types of DSSCs using Autolab PGSTAT 128 N in order to see the effect of metal NP-incorporated TiO2 on the reverse breakdown voltage of the DSSCs.

3 Results and discussion

3.1 Optical absorption by gold and silver nanoparticles’ colloidal suspensions

The colloidal suspensions of gold NPs and silver NPs prepared in this study exhibit a deep red color and an orange color, respectively, as shown in Fig. 1. The UV–Vis optical absorption spectra of the two metal NPs’ colloidal suspensions are shown in Fig. 2. As seen from Fig. 2, the absorption spectra of the Au and Ag colloids exhibit a surface plasmon absorption band with the maxima around 520 and 420 nm, respectively. From the positions of these absorption maxima, it has been estimated that the diameters of the gold nanoparticles are in the 15–20-nm range [29], and the diameters of the silver nanoparticles are in the 40–60-nm range [30].
Fig. 2

Absorbance spectra of (a) Colloidal Au NPs and (b) Colloidal Ag NPs. (Color figure online)

As the sizes or the shapes of the nanoparticles change, the observed color also changes. Gold nanospheres prepared in this work have a characteristic red color, while the silver nanospheres are orange as shown in Fig. 1. It is generally accepted that the color is due to the collective surface plasmon oscillations of the electrons in the conduction band. The oscillation frequency is usually in the visible region for gold and silver giving rise to the strong surface plasmon resonance absorption [31]. Although the conduction and valence bands of semiconductors such as TiO2 are separated by a well-defined band gap, metal nanoparticles have close-lying bands where electrons can move quite freely. These free electrons give rise to surface plasmon-absorption bands in metal NPs, which depends on both the particle size and the chemical surroundings.

The free electrons in the metal (d electrons in Au and Ag) are free to travel through the material. The mean free path in Au and Ag is nearly 50 nm, and therefore, in particles smaller than this, no scattering is expected from the bulk [31]. Thus, all interactions are expected to be within the surface. When the wavelength of light is much larger than the nanoparticle size, it can set up standing resonance conditions. Light in resonance with the surface plasmon oscillation causes the free-electrons in the metal to oscillate. As the wave front of the light passes, the electron density in the particle gets polarized to one surface and oscillates in resonance with the light’s frequency causing a standing oscillation. Under this surface plasmon resonance condition, enhanced light absorption occurs which depends on the types, sizes, shapes, and the dielectric constants of the metal and the surrounding medium.

3.2 Optimization of DSSC efficiency by incorporating different amounts of Gold and Silver NP suspensions into TiO2 electrodes

Different amounts (M Au, moles) of gold NP colloidal suspension were added while preparing the TiO2 paste, and plasmonic DSSCs were fabricated as described in Sect. 2.4 using TiO2 Au NP and TiO2 Ag NP photoanodes, and their efficiencies were determined from I–V characteristics (Tables 1).
Table 1

Efficiency variations with the added amounts of gold NPs (left) and silver NPs (right)

M Au/(×10−7 mol)

η  %


5.00 ± 0.09


5.22 ± 0.05


5.47 ± 0.07


5.84 ± 0.02


5.59 ± 0.08


5.35 ± 0.02

MAg/(×10−7 mol)

η  %


5.00 ± 0.09


5.24 ± 0.06


5.34 ± 0.04


5.82 ± 0.03


6.12 ± 0.07


5.38 ± 0.04


5.05 ± 0.04

The bold numbered rows correspond to the DSSCs with highest efficiency

The XRD patterns of the TiO2, TiO2–Au NP, and TiO2–Ag NP powders, taken after sintering are shown in Fig. 3. Although the peak intensity due to the NPs are very weak and most of the characteristic peaks of both Au NPs and Ag NPs overlap with peaks of the pristine TiO2 (P25) powder, the weak peaks at 2 theta = 44o gives clear evidence for the presence of the Au and Ag nanoparticles.
Fig. 3

XRD spectra of pristine TiO2, TiO2 Au NP- and TiO2 Ag NP-sintered powders. (Color figure online)

Figure 4 shows the efficiency variations with the number of moles of gold and silver NPs incorporated into TiO2 in the plasmonic DSSCs. In each case, the efficiency vs the amount of metal NPs goes through a maximum, and the highest efficiency values are obtained for DSSCs with M Au = 2.36 × 10−7 mol and M Ag = 2.91 × 10−7 mol, respectively. The initial increase in the efficiency is clearly due to the enhanced light absorption by the electrons in the metal NPs due to local surface plasmon resonance effect which is proportional to the number of metal NPs incorporated into TiO2 electrode. This increases the short-circuit photocurrent density, J sc of the two plasmonic DSSCs and hence their efficiencies. Concurrently, a competing mechanism starts to dominate when the amount of incorporated metal NPs increases. A large number of these NPs could concentrate inside the TiO2 crystallites rather than arranging them in the surface of TiO2 crystallites and thereby increase the parasitic absorption, which does not generate an electron/hole pair. This parasitic absorption significantly hampers the plasmonic effect and eventually decreases the photocurrent and the efficiency of the DSSC after reaching to a maximum favorable incorporation of NPs [32].
Fig. 4

Variations of efficiency of DSSCs with the amounts of gold and silver NPs incorporated into TiO2 photoanodes

3.3 Photovoltaic performance of DSSCs

The comparison of photocurrent density versus voltage (I–V) curves for the three types of DSSCs with photoanodes (A) TiO2 only, (B) TiO2 with Au NPs, and (C) TiO2 with Ag NPs under the same illumination conditions is shown in Fig. 5. For these of DSSCs, the fill factor (FF) is calculated using Eq. (1), and the overall energy conversion efficiency (η) was estimated using Eq. (2). In Eq. (1), the V m and J m are the voltage and the current density, respectively, for the maximum power output;
Fig. 5

Photocurrent density–voltage curves for DSSCs with (a) normal TiO2, (b) TiO2–Au NP, and (c) TiO2–Ag NP. (Color figure online)

$$FF \, = \, {{J_{\text{m}} V_{\text{m}} } \mathord{\left/ {\vphantom {{J_{\text{m}} V_{\text{m}} } {J_{\text{sc}} V_{\text{oc}} }}} \right. \kern-0pt} {J_{\text{sc}} V_{\text{oc}} }}$$
$$\eta = \, {{J_{\rm{sc}} V_{\rm{oc}} \, FF} \mathord{\left/ {\vphantom {{J_{\rm{sc}} V_{\rm{oc}} \, FF} {\text{Total incident power density}}}} \right. \kern-0pt} {\text{Total incident power density}}}$$
The short-circuit photocurrent density, J sc open-circuit voltage, V oc, the fill factor, FF, and the efficiency, η, derived from photo current–voltage measurements (Fig. 4) of the three types of DSSCs are summarized in the Table 2. When preparing these DSSCs, it was ensured to have the same TiO2 film thickness of ~5 μm and the same amount of dye loading.
Table 2

Photovoltaic parameters of the DSSCs with (A) TiO2, (B) TiO2–Au NP, and (C) TiO2–Ag NP

DSSC photoanode

J sc (mA cm−2)

V oc (mV)

FF (%)

η (%)

(A) TiO2





(B) TiO2–Au NP





(C) TiO2–Ag NP





When the photovoltaic parameters of the three DSSCs A, B, and C shown in Table 2 are compared, one can clearly see that the short-circuit photocurrent density (J sc), the fill factor (FF), and the energy conversion efficiency (η) values of the B and C plasmonic DSSCs have increased, but their open-circuit voltage (V oc) values have decreased.

As seen from Table 2, the conversion efficiency for DSSCs made with pristine TiO2, TiO2–Au NP, and TiO2–Ag NP are 5.12, 6.23, and 6.51 %, respectively, showing clearly that the enhanced short-circuit photocurrent density, J sc in the two plasmonic DSSCs is largely responsible for the enhanced efficiency in these cells. The photocurrent density enhancement evidently takes place due to the improved light absorption by the metal NP containing TiO2 electrode in the two plasmonic DSSCs by the local plasmon resonance effect as already described in the Introduction section. The overall solar cell efficiency has shown a dramatic increase by 21.6 % for the solar cells with Au NPs and by 27.1 % for the solar cells with Ag NPs. However, it should be noted that the V oc values of the two plasmonic DSSCs have decreased relative to the reference DSSC. A possible mechanism for this effect will be discussed later after the IPCE spectra and results of the Flat Band Potential measurements are analyzed. Clearly, the contributions from the enhancement in J sc to the overall energy conversion efficiency of the two plasmonic DSSCs grossly outweigh the reduction in V oc resulting in the observed efficiency enhancement. Very similar results of enhanced J sc values due to plasmon-induced charge transfer are reported for the plasmonic DSSCs incorporating Ag nanoparticles in TiO2 nanofiber-based photoanodes [33].

3.4 IPCE analysis

The effectiveness of a dye-sensitized cell to convert light of various wavelengths into electrical energy is measured by the incident monochromatic light current conversion efficiency (IPCE) defined as the number of electrons generated by light per number of photons incident on the cell as formulated by IPCE  % = 1240 J sc/λ W iwhere J sc is the short-circuit current density (mA cm−2), λ the excitation wavelength (nm), and W i is the photon flux (W m−2).

Results of IPCE measurements taken for the three DSSC systems: (A) TiO2/dye/liquid electrolyte/Pt, (B) TiO2–Au NP/dye/liquid electrolyte/Pt, and (C) TiO2–Ag NP/dye/liquid electrolyte/Pt, using Ru N719 as the sensitizing dye, are shown in Fig. 6. The IPCE peak height for TiO2–Ag DSSC is 56.1 % at 539 nm, and for TiO2–Au NP DSSC, it is 52.4 % at 533 nm. Both these are much higher than the peak height of 46.3 % at 531 nm for reference TiO2 DSSC. The UV–Vis absorption spectra of the metal NP-incorporated, sintered, and dyed TiO2 powder samples of the three photoanode materials: (a) TiO2-dye, (b) TiO2–Au NP-dye, or (c) TiO2–Ag NP-dye are also shown in the same figure for easy comparison with IPCE spectra.
Fig. 6

Comparison of optical absorption curves for sintered and dyed powders (a) TiO2-dye, (b) TiO2–Au NP-dye, and (c) TiO2–Ag NP-dye; and spectral responses (IPCE) of DSSCs with photoanodes: (A) TiO2, (B) TiO2–Au, and (C) TiO2–Ag. (Color figure online)

There are two major effects that can be expected due to the presence of the Au and Ag NPs in TiO2 photoelectrodes: (1) increased light absorption due to surface plasmon effect as described earlier under the Introduction; and (2) narrowing of the energy band gap in metal NP-incorporated TiO2 electrodes. We can observe the results of both these effects in the IPCE spectra of the two plasmonic DSSCs.

As seen from Fig. 6, the amounts of absorption by all the wavelengths in the visible region as estimated from the total integrated area under the absorption curve also increases substantially from 0.45 (Reference TiO2 powder) to 0.54 (TiO2–Au NP powder) and to 0.65 (TiO2 Ag NP powder) (Table 3). This supports the absorption enhancement by plasmon resonance effect in the two plasmonic solar cells. The strongly localized electromagnetic field around the Au and Ag NPs will strongly enhance the optical absorption efficiency of the dye molecules, resulting in the observed increase in the photocurrent of the Au and Ag NP-incorporated DSSC systems [15, 33] . Therefore, apart from the electron injection from the dye molecules due to the photoexcitation, the gold and silver nanoparticles are also photoexcited due to the plasmon resonance effect, and charge separation is accompanied by the transfer of photoexcited electrons from the metal particle to the TiO2 conduction band and the simultaneous transfer of compensative electrons from the donor in the electrolyte to the metal particle.
Table 3

Wavelengths of the maxima in the powder absorption spectra and the maxima of the IPCE spectra of the reference DSSC and the DSSCs with Au and Ag NPs. Data taken from Fig. 5

Powder sample used for photoanode

Absorbance peak (nm)

Absorption (Relative)


IPCE peak wavelength (nm)

(a) TiO2-dye



(A) TiO2


(b) TiO2–Au-dye



(B) TiO2–Au


(c) TiO2–Ag-dye



(C) TiO2–Ag


As can be seen from the figure, there is a good agreement between the wavelength maxima in the IPCE plots (IPCEmax) of these three DSSC systems and the dye absorption maxima (Absmax) of the corresponding dyed TiO2 anode powders. For the reference D-TiO2 system, the IPCEmax (=531 nm) is slightly red shifted from Absmax (=528 nm). For D-TiO2–Au NP plasmonic system, the IPCEmax (=533 nm) is red shifted from Absmax (=527 nm) and for D-TiO2–Ag NP plasmonic system, the IPCEmax (=539 nm) is also red shifted from Absmax (=522 nm). This phenomenon was described by Standridge et al. [11] where it was observed that the red shift in the IPCEmax occurs relative to D-TiO2 (powder) system when Au and Ag NPs were incorporated. Similar behavior was reported for solar cells containing Au NPs by Hagglund et al. [31, 32] and for solar cells containing Ag NPs by Rand et al. [33].

The observed red shift in the IPCEmax of the two plasmonic DSSCs can be easily understood by referring to the apparent Fermi level shift obtained from Mott-Schottky measurements as discussed below under the flat band potential.

3.5 Flat band potential analysis

Results of the flat band potential measurements of the three types of TiO2 electrodes are shown in Fig. 7. This figure clearly shows the reduction in the flat band potential V fb values of the TiO2 Au NP and TiO2 Ag NP electrodes compared to that of reference TiO2 electrode, in agreement with the observed reduction of the V oc following the same trend.
Fig. 7

Flat band potential (Vfb) values of the three different photoelectrodes: (a) Reference TiO2, (b) TiO2 Au NP, and (c) TiO2 Ag NP and their comparison with V oc values of the three corresponding DSSCs

In a DSSC kept in the dark, the Fermi level equilibrates with the redox couple present in the electrolyte. Upon excitation of TiO2 films by light, one observes a rise in the photovoltage as the electrons accumulate within the TiO2 film. As the semiconductor TiO2 and metal NP are in contact, the photogenerated electrons are distributed between TiO2 and metal NPs. The transfer of electrons from the excited TiO2 into metal NPs continues until the two systems attain equilibrium. This electron accumulation shifts the resultant pseudo Fermi level (\(E_{{_{\text{F}} }}^{\prime }\)) of the composite system (TiO2 and metal NPs) to more negative values. When this happens, the system attains a new photo equilibrium. For an intrinsic semiconductor, the value of \(E_{{_{\text{F}} }}^{\prime }\) is given by the expression:
$$E_{{_{\text{F}} }}^{\prime } = E_{\text{CB}} + kT{ \ln }\left( {{{n_{\text{c}} } \mathord{\left/ {\vphantom {{n_{\text{c}} } {N_{\text{c}} }}} \right. \kern-0pt} {N_{\text{c}} }}} \right)$$
where E CB is conduction band energy, N c is the effective state density, and n c is the carrier density including the accumulated electrons [32]. Therefore, the position of the pseudo Fermi level \(E_{{_{\text{F}} }}^{\prime }\) will depend on both the position of ECB and the total concentration of electrons in the TiO2. Any shift in \(E_{{_{\text{F}} }}^{\prime }\) thus represents increased accumulation of electrons in the TiO2 film. The magnitude of the photovoltage (V oc) represents the energy difference (∆E) between the apparent Fermi level (\(E_{{_{\text{F}} }}^{\prime }\)) of the TiO2 film and the reduction potential (\(E_{\text{redox}}^{{^{\text{o}} }}\)) of the redox couple in the electrolyte,
$$V_{\text{OC}} = \, \Delta E \, = E_{{_{\text{F}} }}^{\prime } - E_{\text{redox}}^{{^{\text{o}} }}.$$

According to the results of our flat band potential measurements (Fig. 7), the apparent Fermi level (\(E_{{_{\text{F}} }}^{\prime }\)) has shifted to more negative values for the two metal NP-incorporated TiO2 films compared to the reference TiO2 film. This is consistent with the reduced V OC values of the two plasmonic DSSCs as observed from their I–V characteristics (Table 2].

A similar shift in apparent Fermi level and a decrease in the photovoltage as has been reported for the DSSCs with silver nanoparticle-incorporated TiO2 nanofibers [34] and for the case of single-crystal TiO2 electrode that was deposited with thin Pt film [35].

3.6 Energy band gap from optical absorption spectra

The energy band gaps of reference TiO2 powder as well as the TiO2 powder incorporated with Au and Ag NPs were determined from UV–Vis absorption measurements as shown in Fig. 8. The E g values are 3.26 eV for the reference TiO2 powder, 3.14 eV for the TiO2–Au NP powder, and 3.01 eV for the TiO2–Ag NP powder, providing direct evidence for the band gap narrowing due to the presence of metal nanoparticles in TiO2.
Fig. 8

Comparison of optical absorption curves for sintered powders, without dye for determination of the energy band gap values: (a) TiO2 (b) TiO2–Au NP, and (c) TiO2–Ag NP. (Color figure online)

3.7 EIS analysis

Figure 9 shows the Nyquist plots of the electrochemical impedance spectra of DSSCs with TiO2, TiO2–Au NP, and TiO2–Ag NP in the frequency range from 1 × 10−2 to 1 × 106 Hz. Each Nyquist plot in Fig. 8 displays two semicircles: the larger semicircle in the low-frequency range is related to the charge transport/accumulation at dye-attached TiO2/electrolyte interface resistance (R2 CT), and the smaller semicircle in high-frequency region is attributed to the charge-transfer resistance of the Pt counter electrode/electrolyte interface (R1 CT) [36, 37]. The impedance parameters were extracted using the equivalent circuit model (inset of Fig. 9), and the fitted lines are shown as solid lines in the Nyquist plots. The calculated values of the series resistance (RS), charge-transfer resistance of the Pt/electrolyte interface (R1 CT) and charge-transfer resistance of the TiO2/electrolyte interface (R2 CT) based on the equivalent circuits for the three types of DSSCs are tabulated in Table 4.
Fig. 9

Impedance plots taken for the three different DSSCs with TiO2, TiO2–Au NP, and TiO2–Ag NP in the frequency range of 1 × 10−2–1 × 106 Hz

Table 4

Series resistance (Rs), charge-transfer resistance of the Pt/electrolyte (R1CT), charge-transfer resistances of the TiO2/electrolyte (R2CT) of DSSCs with TiO2, TiO2–Au NP, and TiO2–Ag NP

Cell configuration

J sc (mA cm−2)

V oc (mV)

R s (Ω)

R1 CT (Ω)

R2 CT (Ω)

(A) TiO2






(B) TiO2–Au NP






(C) TiO2–Ag NP






As seen from Table 4, the R S, R1 CT, and R2 CT values of the two plasmonic DSSCs are sia velue gnificantly lower than the corresponding quantities of the reference DSSC. The decrease in the R1 CT value of the NP-incorporated DSCCs is an indication of the increase in the oxidation rate of at the counter electrode and the redox mediator favoring the electron collection efficiency at the counter electrode. From the charge-transfer resistance values (R2 CT) in Table 4, one can see that the two plasmonic DSSCs exhibit lower resistance values implying more efficient charge-transfer processes across the TiO2 metal NP/electrolyte interface, while the reference DSSC with TiO2 shows the highest R2 CT. R2 CT has a value of 27.60 ohm for the reference cell, 6.08 ohm for the TiO2–Au NP cells, and 6.06 ohm for the TiO2–Ag NP cells indicating that the presence of metal NPs in TiO2 in the two plasmonic DSSCs has improved the electron transport across the TiO2/electrolyte interface. This also contributes to the improved overall electron-transport process in the two plasmonic devices.

The results in Table 4 further show that the short-circuit current density J sc of the three DSSCs increases with the decreasing R2 CT values in the order: TiO2, TiO2–Au NP, TiO2–Ag NP. The lower R 2CT values in the two plasmonic DSSCs clearly imply that by incorporating Au NP and Ag NPs into TiO2 film, the electron-transfer mechanism at the TiO2/electrolyte interface has been improved. The better connectivity and more conducting pathways created by the presence of metal nanoparticles within the TiO2 nanoporous structure (photoanode) is likely to facilitate this charge-transfer process, and it would be easier for the electrons released by the redox reaction at the interface to reach the dye cation across the Au/Ag TiO2 electrode.

The Bode phase plots of EIS spectra, as shown in Fig. 10, display the frequency peaks of the charge-transfer process at the TiO2/electrolyte interface of the three DSSCs. The characteristic low-frequency peak f max is located at 12.73 Hz for the reference TiO2 cell, at 18.14 Hz for the TiO2–Au NP cell, and at Hz for the TiO2–Ag NP cell. The electron lifetime (τ r) in the DSSCs can be determined from the value of f max [34]. Table 5 shows the relationship between V oc, J sc, and τ r for the three types of DSSCs.
Fig. 10

Complex impedance data plotted as phase angle vs frequency (Bode Plots) for the three DSSCs with TiO2, TiO2–Au NP, and TiO2–Ag NP in the frequency range of 1 × 10−2–1 × 106 Hz

Table 5

Electron recombination life times (τr), extracted from the Bhode plots (Fig. 7)

DSSC anode

η  %

F max (Hz)

τ r (ms)

J sc (mA cm−2)

V oc (mV)

(A) TiO2-dye






(B) TiO2–Au NP -dye






(C) TiO2–Ag NP -dye






From Table 5, it can be seen that the electron life time τ r is highest for the reference cell A (TiO2), and the V oc of this cell is also the highest. τ r values of Au NP and Ag NP-incorporated cells (B and C) are significantly lower, and their V oc values are also lower compared to the reference cell A (TiO2). Shorter electron life time means faster recombination. Therefore, the presence of Au and Ag nanoparticles in TiO2 in the two plasmonic DSSCs can promote nonradiative recombination either with dye cations or with \(I_{{_{\rm{3}} }}^{-}\) redox species leading to a reduced number of electrons arriving at the collecting electrode. The reduction in V oc of the two plasmonic DSSCs might be related to the increased electron recombination. However, the enhancement in J sc surpasses the drop in V oc so that the net result has been the enhancement in efficiency. Very similar results, namely, decreases in τ r and V oc for Ag NP-incorporated TiO2 nanofiber-based DSSCs have been reported by Li et al. [34].

3.8 Dark I–V analysis

Comparison of dark I–V characteristics for the three DSSCs based on TiO2, TiO2–Au NP, and TiO2–Ag NP is shown in Fig. 11. The onset under reverse bias occurs at three different voltages for the three cells as shown in Table 6.
Fig. 11

Dark I–V characteristics of the three DSSCs with TiO2, TiO2–Au NP, and TiO2–Ag NP photoanodes

Table 6

Onset of the reverse breaking voltages of the three DSSCs with three different photoanodes

DSSC anode

Reverse breaking down voltage (onset) (mV)

V oc (mV)

(A) TiO2 -dye



(B) TiO2–Au NP -dye



(C) TiO2–Ag NP -dye



Interfacial electron-transfer processes are central for the working principle of the DSSC device. In a typical DSSC under illumination, the following processes occur: (a) Dye excitation, (b) Electron injection, (c) Electron recombination with dye cations, (d) Electron recombination with redox electrolyte, and (e) Dye regeneration. The recombination with redox species in the electrolyte ((d) above) is often referred to as dark current, as it is essentially independent of light. In contrast, the recombination pathway to the oxidized dye requires light and does not occur in the dark. The charge recombination at the TiO2–redox electrolyte interface is expected to play a significant role in lowering the photovoltage. The V oc value kinetically manifests the degree of back electron transfer from the TiO2 conduction band to the \(I_{{_{\rm{3}} }}^{-}\) ions in the electrolyte. The recombination of TiO2 conduction band electrons with the holes in the electrolyte, i.e., \(I_{{_{\rm{3}} }}^{-}\), is the most significant loss mechanism in a DSSC [38, 39].

The dark current measurement of DSSCs provides a qualitative measure for the current leakage due to the back electron transfer. In a typical DSSC, the origin of the dark current stems from the existence of naked FTO sites due to the porous nature of the TiO2 structure. The porous structure provides pathways for the commonly used liquid redox electrolyte (e.g., I 3 species) to penetrate through the TiO2 film and contact the bare conductive FTO surface. At these bare sites, the potential is thermodynamically favorable for the reduction of the oxidative species, mainly I 3 . This causes electron recombination and results in the loss of photovoltage.

In general, in a typical DSSC, the electron back injection from the conduction band of TiO2 to electrolyte is due to trapping of conduction band electrons by deep states followed by reduction processes at the interface. In a plasmonic DSSC, this process is facilitated due to the presence of metal nanoparticles in the TiO2 matrix and due to the lower charge-transfer resistance, R2 CT. In a plasmonic DSSC, there are more recombination pathways within the TiO2 semiconductor due to the presence of metal NPs which could provide additional transient sites for electron migration. The rate of recombination or the rate of reaction of \(I_{{_{\rm{3}} }}^{-}\)ions with the electrons at the surface of the dyed TiO2 electrode will increase with the conduction band electron concentration resulting in back current and the accompanying decrease of the open-circuit voltage. In the present study, higher electron concentration in the composite TiO2 electrode (TiO2 semiconductor with Au or Ag nanoparticles) would lead to a higher back current in the two plasmonic DSSCs compared to the reference DSSC.

As seen from Fig. 11 and Table 6, the breakdown voltages at the negative region in the two plasmonic DSSCs are lower compared with the reference DSSC indicating that the electron recombination of the plasmonic DSSCs is greater. This indicates that the rate of the reaction of I 3 with electrons at the surface of the dyed TiO2 electrode is greater for the two plasmonic DSSCs, leading to a reduction in their V OC value. This is consistent with the lower electron lifetimes derived from EIS measurements for the two plasmonic DSSCs as discussed earlier. The reduced V OC, of course, will make a negative contribution to the observed efficiency enhancement in the two plasmonic DSSCs [40]. Similar results have been reported by Li et al. for plasmonic DSSCs fabricated with Ag NP-incorporated TiO2 nanofiber-based photoanodes [34].

It should be mentioned that, our initial attempts were focused on incorporating a mixture of Au NPs and Ag NPs in different molar ratios, to benefit from the wider spectral range seen in the optical absorption spectrum. However, despite many attempts, the efficiencies of the corresponding plasmonic DSSCs were lower than the DSSC with pristine TiO2. One could speculate that, this could possibly be due to a destructive interference effect resulting from the two resonance plasma frequencies.

4 Conclusion

plasmon-enhanced energy conversion in DSSCs with Au- and Ag NP-modified TiO2 photoanode has been investigated. It was observed that the efficiency of the plasmonic DSSCs for a selected size range of nanoparticles depends strongly on the added amount of Au and Ag NPs. For each plasmonic cell, the efficiency showed a maximum with the increasing amount of NPs. Compared with the normal DSSC with pristine TiO2 photoanode, the short-circuit current density (J sc) and the fill factor (FF) of the DSSCs with Au and Ag NP-incorporated TiO2 electrode were improved, enhancing the overall solar cell efficiency by 21.6 % for the Au NPs and 27.1 % for the Ag NPs. Based on consistent results from several types of electrochemical and optical measurements, it is clear that the Au- and Ag NP-incorporated plasmonic DSSCs exhibit two processes: (a) high optical absorption characterized by a red shift in the IPCE max peak and a significantly-enhanced optical absorption induced by the surface plasmon resonance effect, enhancing the short-circuit current density, J sc and the efficiency; and (b) low open-circuit voltage, V oc resulting from the narrowing of the energy band gap due to the presence of metal NPs in TiO2 and the reduced electron life timesincreasing the electron recombination with redox species at the TiO2/electrolyte interface.



Authors wish to thank Dr. LP Teo and the Centre for Ionics, University of Malyaya (CIUM) for carrying out the XRD measurements.


  1. 1.
    O’Regan B, Grätzel M (1991) A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO2 films. Nature 353:737–740CrossRefGoogle Scholar
  2. 2.
    Hara K, Arakawa H (2003) Dye-sensitized solar cells. In: Tachibana Y, Hara K, Sayama K, Arakawa H (eds) Handbook of photovoltaic science and engineering. John Wiley & Sons Ltd, Chichester, pp 664–670Google Scholar
  3. 3.
    Koumura N, Wang Z, Mori S, Miyashita M, Suzuki E, Hara K (2006) Alkyl-functionalized organic dyes for efficient molecular photovoltaics. J Am Chem Soc 128:14256–14257CrossRefGoogle Scholar
  4. 4.
    Grätzel M (2005) Solar energy conversion by Dye-sensitized photovoltaic cells. Inorg Chem 44:6841–6851CrossRefGoogle Scholar
  5. 5.
    Dissanayake MAKL, Thotawatthage CA, Senadeera GKR, Bandara TMWJ, Jayasundera WJMJSR, Mellander B-E (2012) Efficiency enhancement by mixed cation effect in dye-sensitized solar cells with PAN based gel polymer electrolyte. J Photochem Photobiol A 246:29–35CrossRefGoogle Scholar
  6. 6.
    Wang H, Zhang X, Gong F, Zhou G, Wang Z (2012) Novel ester-functionalized solid-state electrolyte for highly efficient all-solid-state dye-sensitized solar cells. Adv Mater 24:121–124CrossRefGoogle Scholar
  7. 7.
    Dissanayake MAKL, Rupasinghe WNS, Seneviratne VA, Thotawatthage CA, Senadeera GKR (2014) Optimization of iodide ion conductivity and nano filler effect for efficiency enhancement in polyethylene oxide (PEO) based dye sensitized solar cells. Electrochim Acta 145:319–326CrossRefGoogle Scholar
  8. 8.
    Nazeeruddin M, Kay A, Rodicie I, Humphry-Baker R, Müller E, Liska P (1993) Conversion of light to electricity bycis-XzBis(2,2′-bipyridyl-4,4′-dicarboxylate)ruthenium(11) charge-transfer sensitizers (X = C1-, Br-, I-, CN-, and SCN-) on nanocrystalline TiO2 electrodes. J Am Chem Soc 115:6382–6390CrossRefGoogle Scholar
  9. 9.
    Dissanayake MAKL, Divarathne HKDMNR, Thotawatthage CA, Dissanayake CB, Senadeera GKR, Bandara BMR (2014) Dye-sensitized solar cells based on electrospun polyacrylonitrile (PAN) nanofibre membrane gel electrolyte. Electrochim Acta 130:76–81CrossRefGoogle Scholar
  10. 10.
    Chandrasekharan N, Kamat P (2000) Improving the photoelectrochemical performance of nanostructured TiO2 films by adsorption of gold nanoparticles. J Phys Chem B 104:10851–10857CrossRefGoogle Scholar
  11. 11.
    Standridge S, Schatz G, Hupp J (2009) Distance dependence of plasmon enhanced photocurrent in dye-sensitized solar cells. J Am Chem Soc 131:8407–8409CrossRefGoogle Scholar
  12. 12.
    Ding K, Zhu J, McGehee M, Cai W, Moon S, Cai N et al (2011) Plasmonic dye-sensitized solar cells. Adv Energy Mater 1:52–57CrossRefGoogle Scholar
  13. 13.
    Zhang DZ, Wang M, Brolo A, Shen J, Li X, Huang S (2013) Enhanced performance of dye-sensitized solar cells using gold nanoparticles modified fluorine tin oxide electrodes. J Phys D Appl Phys 46:024005–024013CrossRefGoogle Scholar
  14. 14.
    Meen T, Tsai J, Chao S, Lin Y, Wu T, Chang T (2013) Surface plasma resonant effect of gold nanoparticles on the phototelectrodes of dye-sensitized solar cells Nanoscale Research Lett 8:450–456Google Scholar
  15. 15.
    Jeong N, Prasittichai C, Hupp J (2011) Photocurrent enhancement by surface plasmon resonance of Silver nanoparticles in highly porous dye-sensitized solar cells. Langmuir 27(23):14609–14614CrossRefGoogle Scholar
  16. 16.
    Ihara M, Tanaka K, Sakaki K, Honma I, Yamada K (1997) Enhancement of the -absorption coefficient of cis-(NCS)2 Bis(2,2‘-bipyridyl-4,4‘-dicarboxylate)ruthenium(II) dye in dye-sensitized solar cells by a silver Island film. J Phys Chem B 101:5153–5157CrossRefGoogle Scholar
  17. 17.
    Jeong NC, Prasittichai C, Hupp J (2011) Photocurrent enhancement by surface plasmon resonance of silver nanoparticles in highly porous dye-sensitized solar cells. Langmuir 27:14609–14614CrossRefGoogle Scholar
  18. 18.
    Ding IK, Zhu J, Cai W, Moon SJ, Cai N, Wang P, Zakeeruddin S, Gratzel M, Brongersma MML, Cui Y, mcgehee MD (2011) Plasmonic dye-sensitized solar cells. Adv Energy Mater 1:52–57CrossRefGoogle Scholar
  19. 19.
    Hou W, Pavaskar P, Liu Z, Theiss J, Aykol M, Cronin SB (2011) plasmon resonant enhancement of dye-sensitized solar cells. Energy Environ Sci 4:4650–4655CrossRefGoogle Scholar
  20. 20.
    Chang S, Li Q, Xiao X, Wong KY, Chen T (2012) Enhancement of low energy sunlight harvesting in dye-sensitized solar cells using plasmonic gold nanorods. Energy Environ Sci 5:9444–9448CrossRefGoogle Scholar
  21. 21.
    Zhang X, Liu J, Li S, Tan X, Yu M, Du J (2013) Bioinspired Synthesis of Ag@TiO2 plasmonic nanocomposites to enhance the light harvesting of dye-sensitized solar cells. RSC Adv 3:18587–18595CrossRefGoogle Scholar
  22. 22.
    Chang H, Chen C, Kao M, Hsiao H (2014) Effect of core-shell Ag@TiO2 volume ratio on characteristics of TiO2-based DSSCs. J Nanomater 2014:264108–264116Google Scholar
  23. 23.
    Brown M, Suteewong T, Kumar R, D’Innocenzo V, Petrozzo A, Lee M (2011) Plasmonic dye-sensitized solar cells using core-shell metal-insulator nanoparticles. Nano Lett 11:438–445CrossRefGoogle Scholar
  24. 24.
    Sheehan S, Noh H, Brudvig G, Cao H, Schmuttenmaer C (2013) Plasmonic enhancement of dye-sensitized solar cells using core − shell − shell nanostructures. J Phys Chem C117:927–934Google Scholar
  25. 25.
    Tabrizi A, Ayhan F, Ayhan H (2009) Gold nanoparticle synthesis and characterisation. Hacettepe J Biol Chem 37(3):217–226Google Scholar
  26. 26.
    Huang H, Yang X (2003) Chitosan mediated assembly of gold nanoparticles multilayer. Coll Surf A 226:77–86CrossRefGoogle Scholar
  27. 27.
    Chang H, Chen D (2009) Fabrication and photocatalytic activities in visible and UV light regions of Ag@TiO2 and NiAg@TiO2 nanoparticles. Nanotechnology 20:105704–105714CrossRefGoogle Scholar
  28. 28.
    Ratyakshi I, Chauhan R (2009) Colloidal synthesis of Silver Nanoparticles. Asian J Chem 21(10):113–116Google Scholar
  29. 29.
    Njoki PN, Lim IIMS, Mott D, Park HY, Khan B, Mishra S, Sujakumar R, Luo J, Zhong CJ (2007) Size correlation of optical and spectroscopic properties for gold nanoparticles. J Phys Chem C 111:14664–14669CrossRefGoogle Scholar
  30. 30.
    Eustis S, El-Sayed NA (2006) Why Gold Nanoparticles are more precious than pretty gold: noble metal surface plasmon resonance and its enhancement of the radiative and nonradiative properties of nanocrystals of different shapes. Chem Soc Rev 35:209–217CrossRefGoogle Scholar
  31. 31.
    Hagglund C, Zach M, Kasemo B (2008) Enhanced charge carrier generation in dye sensitized solar cells by nanoparticle plasmons. Appl Phys Lett 92:013113–013113CrossRefGoogle Scholar
  32. 32.
    Chandrasekharan N, Kamat P (2000) Improving the Photoelectrochemical Performance of Nanostructured TiO 2 Films by Adsorption of Gold Nanoparticles. The J Phys Chem B 104(46):10851–10857CrossRefGoogle Scholar
  33. 33.
    Rand B, Peumans P, Forrest S (2004) Long-range absorption enhancement in organic tandem thin-film solar cells containing silver nanoclusters. J Appl Phys 96:7519–7526CrossRefGoogle Scholar
  34. 34.
    Li J, Chen X, Ai N, Hao J, Chen Q, Strauf S (2011) Silver nanoparticle doped TiO2 nanofiber dye sensitized solar cells. Chem Phys Lett 514:141–145CrossRefGoogle Scholar
  35. 35.
    Nakato Y, Shioji M, Tsubomura H (1982) Photoeffects on the potentials of thin metal films on a n-TiO2 crystal wafer The mechanism of semiconductor photocatalysts. Chem Phys Lett 90:453–456CrossRefGoogle Scholar
  36. 36.
    Subramanian V, Wolf E, Kamat PV (2001) Semiconductor − metal composite nanostructures. to what extent do metal nanoparticles improve the photocatalytic activity of TiO2 films. J Phys Chem B 105(46):11439–11446CrossRefGoogle Scholar
  37. 37.
    Pan M (2013) Enhanced efficiency of dye-sensitized solar cell by high surface area anatase-TiO2-modified P25 paste. J Nanomater 2013:760685–760691Google Scholar
  38. 38.
    Park NG, Lagemaat JVD, Frank AJ (2000) Comparison of dye-sensitized rutile and anatase-based TiO2 solar cells. J Phys Chem B 104:8989–8994CrossRefGoogle Scholar
  39. 39.
    Tang Xiao, Wang Yuxun, Cao Guozhong (2013) Effect of the adsorbed concentration of dye on charge recombination in dye-sensitized solar cells. J Electroanal Chem 694:6–11CrossRefGoogle Scholar
  40. 40.
    Sathiya Priya AR, Subramania A, Jung Young-Sam, Kim Kang-Jin (2008) High- performance quasi-solid-state dye-sensitized solar cell based on an electrospun PVdF-HFP membrane electrolyte. Langmuir 24(17):9816–9819CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2015

Authors and Affiliations

  • M. A. K. L. Dissanayake
    • 1
    • 2
  • J. M. K. W. Kumari
    • 1
    • 2
  • G. K. R. Senadeera
    • 1
    • 2
    • 3
  • C. A. Thotawatthage
    • 1
    • 2
  1. 1.National Institute of Fundamental StudiesKandySri Lanka
  2. 2.Postgraduate Institute of ScienceUniversity of PeradeniyaPeradeniyaSri Lanka
  3. 3.Department of PhysicsThe Open University of Sri LankaNugegodaSri Lanka

Personalised recommendations