Solution-processed CdS quantum dots on TiO₂: light-induced electrochemical properties

Abstract

Mesoporous titanium dioxide (TiO₂) nanostructured thin films, formed by a simple chemical route, namely, the successive ionic layer adsorption reaction (SILAR) technique, were efficiently used to grow a cadmium sulphide (CdS) nanoparticle architecture for device grade applications. CdS nanoparticle have a highly symmetric size and shape that could be varied in a controlled manner (<10 nm size) depending on the number of SILAR immersions without the use of an organic linker molecule. XRD and HR-TEM studies showed the formation of CdS nanoparticles on TiO₂ nanoparticles. Such a nanomorphology showed good optical coverage over the visible region of light, and it revealed a red-shift with respect to the number of CdS layers, which was confirmed by both optical and external quantum efficiency studies. The photoelectrochemical performance of the TiO₂/CdS photoanodes showed that as the particle size of CdS increases with the number of SILAR immersions, the photocurrent density and photovoltage were found to be directly concurrent with the charge transport mechanism, which increases the cell efficiency.

---

Introduction

Nowadays, the use of three dimensional (3D) confined nanomaterials of inorganic metal oxides and chalcogenide semiconductors have driven a remarkable advancement in device grade applications. The nanoparticles can be easily tuned to alter their optical, mechanical and electrical behaviour according to their shape and size, and can also be fabricated in different shape-controlled assemblies by various deposition methods.^1,2 Additionally, nanoparticles possess higher excitation coefficients^3,4 compared with metal organic dye molecules, as well as a large intrinsic dipole moment leading to rapid charge transportation.⁵ These advantages make inorganic nanoparticles of metal chalcogenides and metal oxides ideal materials for development for various device grade applications. In spite of a use of individual inorganic semiconducting nanomaterials, the combined structure of narrow band gap metal chalcogenides with wide band gap metal oxides materials is more beneficial for the device level applications. Regarding this, large numbers of reports have been successfully published^6–17 and there has been great research interest in developing and refining a more efficient clean energy production device that might satisfy future energy demand. Considering this, nanomaterials will likely play a dynamic role in future technologies.

In photoelectrodes, narrow band gap CdS,^6–9 CdSe^10–15 and CdTe^16,17 are used as light harvesters, mostly coupled with wide band gap semiconducting materials, such as TiO₂, ZnO and SnO₂. One of the subjects investigated recently is the linker-assisted attachment of these metal chalcogenides with wide band gap materials for solar cell applications, such as in the quantum dot sensitized solar cell (QDSSC).^6–21 In the current research, the use of bifunctional organic linker molecules for the attachment of nanoparticles to the electrode material reduces the electron tunnelling injection effect.¹⁸ Also, different bifunctional linker molecules directly influence the kinetics of charge separation and transportation,^11,19–21 and this has been considered a key reason to limit the conversion efficiencies in QDSSC. Hence, there is a prevailing need to find a suitable sensitizer to harvest more visible-light utilizing a linker-free approach. There are many reports where the conformational growth of the CdS quantum dots has been achieved without any linker over different morphologies of TiO₂. Among these, Chen et al. demonstrated the growth of CdS QDs by the CBD method, which involved the anchoring of CdS over TiO₂ nanotubes.^22,23 Recently, it was found that the attachment of the CdS QDs can be done more efficiently through use of the SILAR method. In this perspective, Pawar et al. formed a uniform growth of CdS QDs over TiO₂ nanorods.²⁴ Apart from this, Tau et al. synthesized CdS nanoparticles over vertically aligned TiO₂ nanorod arrays.²⁵ However, in all of the above-stated linker-free approaches the basic TiO₂ film was formed either by anodic oxidation of the titanium foil or by the hydrothermal method. Therefore, to ease the fabrication, an easy, convenient and efficient method for the growth of the TiO₂ film is desired. Consequently, in the present study we fabricated TiO₂ through a spin-coating method at ambient temperature.

Apart from the fabrication of TiO₂, this article also reports on the room temperature controlled growth of a CdS assembly over mesoporous TiO₂ by a successive ionic layer adsorption reaction (SILAR) method utilizing a linker-free approach. The direct chemisorption involved in the reaction kinetics facilitates the chemical reaction, resulting in the growth of particles and, hence, the growth can be controlled at an ionic level. van der Waals and electrostatic forces took part in the heterogeneous nucleation, resulting in the growth of quantum-sized CdS nanoparticles on TiO₂ film surface by varying the number of SILAR immersions. This method enables the systematic monitoring of the nucleation and growth of accumulation of CdS quantum dots on the TiO₂ electrode by an ion-by-ion reaction mechanism. The systematic growth can be modulated in the evolution of nanoparticles, which is depicted in the schematic in Fig. 1 showing the growth at 8, 13, 15 and 20 SILAR immersions, respectively. Additionally, the inset in Fig. 1 clearly demonstrates the TiO₂ as bigger-sized particle, while the yellow dots represent the CdS quantum dots formed. The surface morphological studies confirmed that the growth of CdS nanoparticles films resulted in a cauliflower-like morphology with the increasing number of SILAR immersions, as shown in the schematic in Fig. 1. Diverse characterizations were performed to support the size tuning and the results are reported and discussed herein. The synthesized nanostructure (TiO₂/CdS) could function as a photoanode for solar cell applications. A sandwich-type solar cell device was compiled and its photovoltaic performance tested under light illumination (AM 1.5G, 50 mW cm⁻²) and the photovoltaic behaviour is also discussed herein.

Fig. 1 Schematic evolution of the growth of CdS nanoparticles onto mesoporous TiO₂ film with the different number of SILAR cycles. The inset clearly demonstrates the TiO₂ with bigger size particles, whereas the yellow dots represent the CdS quantum dots formed.

Results and discussion

In order to control the growth of CdS nanoparticles, we utilized low concentrations of cationic and anionic sources and methanol as a solvent instead of the use of surfactants or ligands. Here, using methanol as a solvent offers a number of advantages: (1) it stabilizes the Cd²⁺ and S²⁻ ions in separate cationic and anionic solutions, (2) it diminishes the probability of cadmium hydroxide formation compared to the use of an aqueous solution. In this reaction, a stable CdS nano to quantum architecture is formed on the TiO₂ surface by an ion-by-ion reaction mechanism. A 1 min vertical immersion of the mesoporous TiO₂ film in the cationic solution provides the surface for the chemical attachment of Cd²⁺ ions. The attachment of Cd²⁺ ions on the TiO₂ electrode was achieved by van der Waals and electrostatic forces, forming a cationic Helmholtz double layer on the surface of the TiO₂ in the first step. In the second step, loosely bound or excess Cd²⁺ ions were washed out by rinsing the substrate meticulously in an ample amount of methanol for 2 min. Further, the substrate was immersed in anionic solution for 1 min, where S²⁻ ions reacted with pre-adsorbed Cd²⁺ ions to form a CdS nanoparticles monolayer. Finally, the substrate was thoroughly rinsed in methanol to eliminate excess anions and CdS powdery particles in the solid liquid interface. This completed one cycle, but by repeating the number of immersions we could control the quantum dot-to-cauliflower CdS architecture evolution on the TiO₂ surface by heterogeneous nucleation. The growth of CdS particles depended on an immersion time, the number of immersion and the rinsing time, which are the key aspects discussed in this article for the controlled growth of CdS nanoparticles.

Fig. 2 shows the X-ray diffraction (XRD) patterns of FTO, TiO₂ and TiO₂/CdS nanostructures for different number of immersions. The most intense diffraction peak at 2θ = 25.3° is indexed to the tetragonal crystal structure of TiO₂ with an anatase phase (JCPDS card 84-1285) and lattice orientation along the (101) plane. Additionally, some small peaks appear along the (004), (200) and (105) planes related to the seminal crystal structure and phase of TiO₂, denoted by the (T) symbol. Due to the high intensity of the TiO₂ and FTO peaks, the effect of the CdS nanoparticles coating on TiO₂ is diminished, even though, some tiny peaks appear at 27.8°, 36.1° and 43.8° in Fig. 2, confirming the presence of another layer coated on the TiO₂ electrode. In the XRD patterns, the diffraction peaks along the (420), (305) and (444) crystal planes correspond to the orthorhombic structure of CdS (JCPDS: 47-1179) and are denoted in the figure by the (C) symbol. The broadness in tiny peaks itself revealed, the nanocrystalline nature of the CdS material deposited on TiO₂ surface. Thus, the formation of TiO₂ along with CdS was confirmed from the XRD patterns. This result confirms the successful deposition of CdS on the TiO₂ film. The other peaks denoted by the triangles (Δ) symbol are due to the FTO substrates. The average crystallite size was calculated by the well-known Debye–Scherrer equation:

where, λ is 0.15418 nm (X-ray wavelength), θ is the Bragg angle in degrees, D is the crystallite size for the individual peak of the crystal in nanometres, β is the full width at half maxima in radians and K is the shape factor and was assigned a value of 0.89. The crystallite sizes were thus estimated to be 28 nm for TiO₂ and 4, 6, 7 and 10 nm for CdS for 8, 13, 15 and 20 immersions, respectively, with a correction factor of 10%.

Fig. 2 X-ray diffraction patterns of the FTO, TiO₂ and CdS coated FTO/TiO₂ for 8, 13, 15 and 20 SILAR cycles.

Fig. 3 shows the high-magnification SEM images of mesoporous TiO₂ electrode at different stages of modifications with a scale bar of 100 nm. These images provide a perspective of the electrode morphology and its ability to anchor the CdS nanoparticles on the mesoporous TiO₂ electrode. Fig. 3(a) shows the growth of a few small grains on and in inside 20–30 nm pores on the TiO₂ surface for eight immersions. The TiO₂ surface was further modified by coating CdS nanoparticles for 13, 15 and 20 immersions, as depicted in Fig. 3(b)–(d), respectively. From Fig. 3(b), it can be seen that the CdS quantum dots (size < 10 nm) are clearly visible on the TiO₂ surface after 13 immersions. The inset in Fig. 3(b) shows a bigger white circle corresponding to TiO₂, while the small dots on the TiO₂ belong to CdS. Further increases in the number of immersions causes the aggregation of CdS nanoparticle on the TiO₂ mesoporous network. Most of the TiO₂ surface areas are covered by CdS nanoparticles, resulting in a cauliflower-like morphology (Fig. 3(c) and (d)). This may be due to the growth of the CdS grain not being on to the TiO₂ surface but rather on pre-grown CdS particles, which act as secondary nucleation centres and finally lead to the cauliflower-like morphology. These results demonstrate that the synthesised mesoporous TiO₂ thin film provides a high surface area on which the CdS particles grow, resulting in the nano to quantum structure. The surface area was calculated by using the following formula:

where, W = weight of dry/regenerated sample, p = partial pressure of nitrogen in the cylinder, p₀ = total partial pressure of the gas mixture and t = temperature (°C), while the other terms have their usual meanings. BET measurement of the mesoporous TiO₂ film gives a surface area of 56.42 m² g⁻¹. Clearly, the CdS coating on the mesoporous film reduces the surface area of the film, as depicted in the Table 1. This result agrees well support with the observations from the surface morphological studies.

Fig. 3 SEM images of the CdS-coated FTO/TiO₂ in (a) 8, (b) 13, (c) 15 and (d) 20 SILAR cycles. The inset in (b) with the bigger circle corresponds to TiO₂, while the small dots on the TiO₂ belong to CdS.

Table 1 Photovoltaic's parameters of the devices with the structure FTO/TiO₂/CdS/electrolyte/contact with the variation in the number of SILAR immersions of CdS on TiO₂

TiO2/CdS SILAR cycles	Jsc (mA cm^−2)	Voc (mV)	FF (%)	η (%)	VFB (mV)	Rs (Ω)	Rsh (kΩ)	Surface area (m^2 g^−1)
8	0.398	266	43.4	0.091	−155	276	1.19	33.09
13	0.707	336	35.1	0.166	−256	265	1.21	29.87
15	1.114	384	34.8	0.297	−301	261	1.34	26.84
20	2.026	402	35.8	0.579	−423	176	1.47	17.64

---

For the TEM images, the nanostructure architecture was scratched without disturbing the FTO substrate and placed on a copper-coated carbon grid. Fig. 4(a) shows a typical TEM image of the TiO₂ coated with CdS quantum dots (size < 10 nm) for 13 immersions. The surface coating in LHS and the surface coverage with uniformity in the RHS image can be clearly seen with a size of less than 10 nm. The inset of Fig. 4(a) shows the evidence for the presence of Cd, Ti, S, O, C and Cu elements in the deposited material. The presence of Cu and C is due to the copper-coated carbon mesh. Fig. 4(b) depicts the HR-TEM image of the CdS-coated TiO₂ from 13 SILAR immersions. The inter-planar spacing (‘d’ value) was estimated as 0.34 nm and was assigned to the anatase phase of tetragonal TiO₂ oriented along the (101) plane. This supports well the plane observed at 25.3° in the XRD studies for TiO₂ (JCPDS card 84-1285). As a typical example, the clear existence of CdS quantum dots on TiO₂ can be seen from the high-quality HR-TEM image shown in Fig. 4(c). The inter-planar distance of 0.24 nm was calculated and assigned to the orthorhombic crystal structure of CdS, which corresponds to the (305) plane as revealed from the XRD analysis (JCPDS card 47-1179). The particle size for 13 immersions was estimated to be 4–6 nm, which is responsible for the red-shift due to the size quantization effect observed in the optical absorption curve. Thus, the successful formation of CdS nanoparticles on mesoporous TiO₂ nanoparticles was confirmed.

Fig. 4 (a) TEM image of CdS coated on FTO/TiO₂. The inset shows the EDAX analysis of TiO₂/CdS scratched powder placed on the copper grid (b) HR-TEM image of the CdS coated on FTO/TiO₂ in 13 cycles. (c) High-quality HR-TEM image of the CdS crystal structure.

The absorbance spectra were recorded in the 300 to 700 nm wavelength regime of incident light. The absorption spectra of bare TiO₂ and four different-sized CdS nanoparticles-coated TiO₂ networks are illustrated in Fig. 5, with FTO as a reference. A sharp increase in the absorption spectra at shorter wavelengths (below 380 nm) can be observed for TiO₂, suggesting that the absorption band edge would facilitate its application as a wide band gap material. Fig. 5 depicts the absorption spectra of the CdS-coated TiO₂ film for the different number of immersions in the SILAR technique. Swelling in the absorption curves between 450 and 500 nm is clearly evident and is boosted with the increase in the number of SILAR immersions; this clearly indicates a red-shift in the absorption edge. By noting the absorption maxima at 451, 461, 472 and 485 nm for the 8, 13, 15 and 20 SILAR immersions, the respective band edge transition energies were calculated at the respective absorption maxima and were found to be 2.74, 2.68, 2.62 and 2.55 eV.²⁶ The particle sizes of CdS with the different number of SILAR immersions were calculated from the absorption maxima by utilizing the empirical fitting function reported by Yu et al.³ for CdS and were found to be 5.52, 5.66, 6.18 and 6.81 nm for the respective maxima of the absorption peaks. The particle size distribution should result in sharp absorption peaks appearing in the absorption spectra measured for the particles in solvents; however, this is not seen in Fig. 5 and instead small swellings in the absorption spectra were observed. This may be attributed to the fact that the increase in the number of immersions causes the agglomeration of small-sized particles resulting in bigger ones, which are responsible for the shift in the optical properties towards the bulk system.

Fig. 5 UV-Vis absorption spectra for FTO/TiO₂ and CdS coated on the FTO/TiO₂ electrode in different numbers of SILAR immersions.

A sandwich cell was prepared by using TiO₂ and TiO₂/CdS as a photoanode and a platinum-coated conducting glass electrode as the counter electrode. The two electrodes were separated by a thin polyethylene film (50 μm) as a spacer. The empty cell was tightly held and then the edges were heated to 130 °C to seal the two electrodes together. The composition of the polysulphide electrolyte was 1 M Na₂S, 1 M sulphur powder and 1 M NaOH. The electrolyte was introduced into all the fabricated cells by using capillary action. The active surface area of the photoanode was 0.16 cm². In addition to enhanced light absorption, the formation of a TiO₂ and TiO₂/CdS nano-heterojunction in the hybrid structure also facilitated separation of the photogenerated electrons and holes, as shown in Fig. 6(a).

Fig. 6 (a) Schematic illustration of the working of the FTO/TiO₂ coated with CdS electrode in the photoelectrochemical cell. (b) EQE spectra of bare TiO₂ and CdS coated on the FTO/TiO₂ electrode in different numbers of SILAR immersions. (c) Current density vs. voltage characteristics under dark and under illumination conditions for different numbers of SILAR immersions of CdS. (d) Mott–Schottky plots of bare TiO₂ and CdS coated on the FTO/TiO₂ electrode in different numbers of SILAR immersions.

The effectiveness of the CdS over TiO₂ for different rinsing cycles and the use of bare TiO₂ as a photoanode in photoelectrochemical QD-sensitized solar cells were assessed utilizing EQE measurements. On increasing the number of immersion cycles due to the enhanced absorption of a number of photons, the increasing growth of CdS results in a steady growth of the short-circuit current density. This gradually increases the value of the EQE for the different number of immersions of the TiO₂/CdS samples, which may be due to the charge injection of highly efficient light harvesting CdS QDs into porous TiO₂, as shown in Fig. 6(b), in which the highest EQE value of 16% was obtained for 20 SILAR cycles of CdS on TiO₂. This leads to more electron collection, which inhibits electron recombination and enhances the electron scattering ability. Furthermore, this may result in the easy transfer of electrons to the conduction band of TiO₂ and to the electrolyte.

The experimental current density–voltage (J–V) characteristics measured from the TiO₂ and TiO₂/CdS photoelectrode are shown in Fig. 6(c). For all the cases, the photoelectrochemical cells' significant activities in the dark were shown by the illustrated heterojunction properties. In the third quadrant of the figure, a sharp increase in current density can be observed, which might be due to penetration of the liquid electrolyte through the pores of TiO₂ and it then touch the back platinum contact. This is rather a good indication of the formation of a nano- or mesoporous TiO₂ network, which is necessary to coat CdS particles from the top to the bottom. Under illumination, TiO₂ showed a slight photovoltaic effect (photocurrent density of ∼88 μA cm⁻²) due to the absorption of UV light in the nano- or mesoporous network. For comparison, the photovoltaic curves for CdS with different numbers of layers over TiO₂ are illustrated. A significant enhancement in the short-circuit current can be seen with the increase in the number of SILAR immersions of CdS along with an enhancement in the open-circuit voltage. This gives a maximum short-circuit current value of 2.02 mA cm⁻² for 20 immersions, with an efficiency of 0.579%. This enhancement is strictly attributed to not only the enhancement in the absorption values with respect to the increase in layer thickness but also to the agglomeration of small-sized particles to form bigger ones, which are responsible for the shift in the optical band gap to lower energies due to the quantum size effect. The more filling of CdS inside the nanoporous TiO₂ layer is also seen by the decrease in the tendency of sharp increase in reverse current, which mainly avoids direct contact between the liquid electrolyte and back contact. The increase in the photovoltage is well attributed to the covering of the TiO₂ layer with more CdS with the increase in the number of immersions. The photovoltage generated in the semiconductor sensitizer solar cell is characteristic of the recombination centres in the semiconductor sensitizer solar cell. The increase in the thickness of the CdS layer on TiO₂ with the increasing number of SILAR immersions simultaneously provides more recombination's centres and hence the photovoltage and photocurrent increase with the number of SILAR immersions.^27,28 A straightforward technique was used to estimate R_s and R_sh from the solar cell photovoltaic output curves by using the following relations²⁹ and the results are given in Table 1

The decrease in R_s is direct evidence for the photogenerated charge carriers. The photovoltaic output increases and the obtained values of R_s decrease with the increase in the number of SILAR immersions (i.e. the increasing thickness of the absorber layer), which supports well the separation of charge carriers across the TiO₂/CdS interface, whereby CdS can act as a sensitizer to generate the charge carriers. Hence, a boost in device efficiency is observed with increasing the number of SILAR cycles. Also, R_sh is affected by both the charge carrier recombination and short circuiting within the cell. Therefore, a higher coating layer for the heterostructure electrodes can result in a pinhole-free surface, thus providing less access for electrolyte contacts to the FTO substrate. The increase in R_sh values with the increasing number of SILAR immersions i.e. with the coating of the CdS layer, reduces the participation of FTO in PEC performances and increases the number of available recombination centres. The photovoltaic parameters, such as the current density (J_sc), open-circuit voltage (V_oc), fill factor (FF), power conversion efficiencies (η%) series resistance (R_s) and shunt resistance (R_sh), are depicted in Table 1. The photovoltaic conversion efficiency was calculated using the following formula:

Beyond 20 SILAR cycles of CdS over TiO₂ (as shown in Fig. S1†), it was found that the thickness of the electrode increases, and as a result of the light is transmitted into the depth of the electrode, which leads to a decrease in the light intensity. This phenomenon directly lowers the excessive electron density, resulting in the lowering of V_oc, as shown in (Table T1†).³⁰ This then results in the increase in R_s. Now, due to the greater thickness of the electrode than the light penetration depth, the recombination centres increase and electron loss takes place. This effect directly decreases the value of J_sc of the cell. The measurement of capacitance as a function of applied voltage provides useful information about photoelectrodes, such as about the flat band potential (V_FB). V_FB is an important factor for photoelectrochemistry and facilities the energetic position of semiconductor materials. The value of V_FB can be obtained by using the Mott–Schottky relation:

where, C_s is the space charge capacitance per unit area and V_FB are the flat band potentials (vs. SCE) and the other terms have their usual meanings. In the present study, the C–V measurement was performed for all cells using a polysulphide as the electrolyte. From Fig. 6(d), extrapolation of the Mott–Schottky plot to 1/C² = 0 yields the “flat band potentials”, which are tabulated in Table 1. V_FB was found decrease with increasing the CdS layer, which well supports the covering of the CdS layer over TiO₂ and reduces the electrolyte contact towards the FTO-coated glass substrate. Hence, the flat band potential is reduced with increases in the CdS layer coating and thus leads to an improvement in device efficiency with the increasing number of SILAR immersions.³¹ This favourable negative shift enhances the V_oc and as a result electron injection from the CdS nanoparticles to TiO₂ becomes easier.³² The changed value of the flat band potential indicates the change in carrier concentration.³³ Furthermore, the change can also be attributed to the increased amount of surface adsorption of electrolyte species and the creation of new donor levels.

In order to investigate the kinetics of the photoelectrochemical process, electrochemical impedance spectroscopy (EIS) was performed at a forward bias potential of −0.4 V under dark conditions. Under this condition, the QDSSC behaves as a leaking capacitor.³⁴ A frequency scan of 0.01 to 10⁴ Hz was chosen for the investigations. The nature of the photoelectrochemical plots, such as the Nyquist and Bode plots, for bare TiO₂ and TiO₂/CdS QDSSC was quiet similar those reported by Sudhagar et al. where they demonstrated self-assembled CdS quantum dot sensitised TiO₂ nanospheroidal solar cells without using any linker.³⁵ Fig. 7(a) shows that the Nyquist plots were first semicircular arcs in the high frequency range, reflecting the charge transfer resistances (R₁) and the constant phase element (Q₁) at the Pt/electrolyte interface. The second semicircle towards the higher frequency region can be ascribed to the interfacial charge transfer (R₂) and constant phase element (Q₂) at the TiO₂/CdS/electrolyte interface, while (R) represents the serial resistance, which is determined by the sheet resistance between the TiO₂ layer and FTO substrate. Q₁ and Q₂ basically describe a non-ideal frequency dependent capacitance due to the non-uniform distribution of the current by the material heterogeneity. The inset of the figure is the simple equivalent electric circuit that was used to fit the EIS data to explain the parameters such as the electron lifetime and the recombination of the TiO₂/QDs film.^36,37 For the TiO₂/CdS system, Lee et al. also showed the same equivalent circuit diagram and controlled the rate of recombination by the incorporation of a ZnS layer.³⁸ It can be seen from Table 2 that the value of (R) decreases from the bare TiO₂ layer to the TiO₂/CdS layer (at 20 SILAR cycles). The value of (R₁) at the high frequency region was found to decrease as the number of SILAR cycles increases. Besides this, the second semicircular arc of the Nyquist plot was found to decrease along the radius, which apparently decreases R₂, which suggests that the QDs are coated on the surface of the highly porous TiO₂ with a high photocurrent density.³⁹ This might effectively decrease the ohmic polarization and impart a better interfacial charge transfer. The lowest value of R₂ for TiO₂/CdS at 20 SILAR cycles indicates the easy transport and effective charge collection, with a very low chance of recombination and thus an enhanced electron transfer rate and high power conversion efficiency.⁴⁰

Fig. 7 (a) Nyquist plot in the dark and (b) Bode plot of bare TiO₂ and CdS coated on the FTO/TiO₂ electrode in different numbers of SILAR immersions.

Table 2 Electrochemical parameters of bare TiO₂ and TiO₂/CdS with different number of SILAR cycles as photo anode

Sl. no.	Label	R1 (kΩ)	R2 (Ω)	τe (ms)	keff (s^−1)
1	TiO2	10	1038	126	7.93
2	8 cycles CdS/TiO2	8.85	946	201	4.93
3	13 cycles CdS/TiO2	8.49	813	253	3.95
4	15 cycles CdS/TiO2	7.25	725	323	3.09
5	20 cycles CdS/TiO2	5.59	676	496	2.01

---

Fig. 7(b) shows the Bode plot for the bare TiO₂ and TiO₂/CdS at different SILAR cycles. Interestingly, it was been found that on increasing the number of SILAR cycles over bare TiO₂, degradation of the TiO₂ surface traps can occur, which passivates the Bode plot towards the low frequency region. Again from Table 2, the values of the electron lifetime (τ_e = 1/2πf_max) increases, which in turn decreases the effective recombination rate (k_eff = 1/τ_e).⁴¹ This may be the cause for the increase in the value of the open-circuit voltage (V_oc), since the increase in the carrier recombination is the prominent factor for the reduction of the V_oc. Further recombination rate increases lead to high FF values, which is in line with the photovoltaic performance data.

Conclusion

The facile room temperature successive ionic layer adsorption reaction (SILAR) was successfully used for the controlled growth of CdS nanoparticles onto an interconnected network of TiO₂. The size of the CdS nanoparticles was controlled by varying the number of SILAR immersions of the TiO₂ electrode. XRD and HR-TEM results confirmed the formation of anatase TiO₂ along with various sizes of orthorhombic CdS nanoparticles. SEM analysis revealed a uniform coverage of quantum- to nano-sized CdS particles on the mesoporous TiO₂ electrode. A clear red-shift was manifested in the absorption studies, which was strongly supported by the XRD and HR-TEM analyses. Assessment of the requisite device grade solar cell parameters along with the electrochemical analysis clearly showed that the CdS nanoparticles contribute significantly to the improvement in photovoltaic response and, hence, the cell efficiency with the deposition of the CdS layer in the linker-free approach. Overall, we successfully demonstrated the controlled growth of CdS with size tuning with our novel process, which could also be applicable for other metal chalcogenides as a low cost alternative towards device grade applications.

Experimental section

Synthesis of TiO₂ thin films

A fluorine-doped tin oxide (FTO)-coated glass substrate as a transparent conducting oxide (TCO) with 15 Ω □⁻¹ sheet resistance was used as the substrate to cast TiO₂ film by a spin-coating method. The TiO₂ gel was prepared by mixing 2 g of P-25 TiO₂ powder with mixture of 8 ml ethanol, 2 ml acetyl acetone and 1 ml titanium isopropoxide solutions. The resultant solution was subsequently stirred and probe sonicated at the 20 kHz frequency for 30 min to form the TiO₂ gel. The resultant gel was kept in an ultrasonic bath prior to use in order to avoid further agglomerations of the separated nanoparticles. The TiO₂ electrodes were fabricated by casting the slurry of TiO₂ (Degussa, P-25) gel onto the TCO by a simple low cost spin-coating technique, followed by air annealing at 450 °C for 60 min. This removes the organic ingredient and helps to enhance the adherence and porosity of the film with the substrate surface. The synthesised mesoporous TiO₂ thin film provides a high surface area for the growth of CdS nanoparticles at room temperature.

Synthesis of CdS nanoparticles onto TiO₂ thin films

The SILAR method was used for the controlled growth of CdS nanoparticles onto mesoporous TiO₂ film. This involved a vertical immersion of the substrate (i.e. spin-coated mesoporous TiO₂ film) into separate cationic and anionic precursor solutions and rinsed in between each step. Here, the cationic solution of pH 6.5 was prepared by dissolving 0.066 g of cadmium acetate into 50 ml of methanol. This formed a polar solution containing polar cadmium and acetate ions in the solution, as shown in the following reaction:

Cd(CH₃COO)₂ + CH₃OH → Cd²⁺ + 2CH₃COO⁻ + CH₃OH (7)

Meanwhile, an anionic solution was prepared by dissolving 2.4 mg of sodium sulphide flakes into 50 ml methanol, giving a resultant solution of pH 7.6. This formed a polar solution containing polar S²⁻ ions in solution, as shown in the following reaction:

Na₂S + CH₃OH → S²⁻ + 2Na⁺ + CH₃OH (8)

The simplified version of the ion-by-ion mechanism of the chemical reaction for the growth of CdS nanoparticles onto the surface of TiO₂ is as follows:

Cd²⁺ + S²⁻ → CdS (9)

The growth of the nano-sized CdS nanoparticles on the mesoporous TiO₂ electrode was controlled by varying the number of vertical immersions of the TiO₂ film into the separate anionic and cationic solutions. In the present work, the growth of CdS nanoparticles, which constitutes the stages of the thin film growth depending on number of SILAR immersions as 8, 13, 15 or 20, is discussed herein.

Characterization techniques

X-ray diffraction (XRD) patterns were recorded at the scanning angles of 20° to 60° with a Rigaku Rotalflex RU-200B diffractometer using Cu K_α radiation (λ = 1.5418 Å). Surface morphological studies were performed by using a scanning electron microscopy (SEM) unit (JEOL JSM-7500F) operated at 10 kV. Transmission electron microscopy (TEM), high-resolution TEM (HR-TEM), and energy dispersive X-ray spectroscopy (EDAX) analyses were performed by using a JEOL JEM-2100 unit operated at 200 kV. The optical absorption spectra were recorded between 300 and 700 nm by a UV-Vis spectrophotometer (Shimadzu model-1601). External quantum efficiency (EQE) measurements were performed by using a Bentham-TM 300 monochromator. The photovoltaic measurements were performed by using simulated sunlight (1 Sun) under the standard condition of 1.5 AM 50 mW cm⁻² light illumination by a solar simulator [Sciencetech, Model SS150W AAA Solar Simulator]. The current–voltage characteristics were recorded by using a Gamry Reference 600™ potentiostat/galvanostat. The electrochemical impedance spectroscopy properties were studied using PARSTAT 4000/Princeton Applied Research.

Acknowledgements

BRS acknowledges DAE-BRNS, Mumbai, India for financial support through the DAE-BRNS Project no. 2010/37C/5/BRNS/830 for 2010–2013 and Director VNIT for URL grant.

References

1. T. Trindade and P. O'Brien, Nanocrystalline Semiconductors:  Synthesis, Properties, and Perspectives, Chem. Mater., 2001, 13, 3843–3858.
2. C. Z. Yao, B. H. Wei, L. X. Meng, H. Li, Q. J. Gong, H. Sun, H. X. Ma and X. H. Hu, Controllable electrochemical synthesis and photovoltaic performance of ZnO/CdS core–shell nanorod arrays on fluorine-doped tin oxide, J. Power Sources, 2012, 207, 222–228.
3. W. W. Yu, L. Qu, W. Guo and X. Peng, Experimental Determination of the Extinction Coefficient of CdTe, CdSe, and CdS Nanocrystals, Chem. Mater., 2003, 15, 2854–2860.
4. P. Wang, S. M. Zakeeruddin, J. E. Moser, R. Humphry-Baker, P. Comte, V. Aranyos, A. Hagfeldt, M. K. Nazeeruddin and M. Gratzel, Stable New Sensitizer with Improved Light Harvesting for Nanocrystalline Dye-Sensitized Solar Cells, Adv. Mater., 2004, 16, 1806–1811.
5. R. Vogel, P. Hoyer and H. Weller, Quantum-Sized PbS, CdS, Ag₂S, Sb₂S₃, and Bi₂S₃ Particles as Sensitizers for Various Nanoporous Wide-Bandgap Semiconductors, J. Phys. Chem., 1994, 98(12), 3183–3188.
6. J. C. Lee, T. G. Kim, W. Lee, S. H. Han and Y. M. Sung, Growth of CdS Nanorod-Coated TiO₂ Nanowires on Conductive Glass for Photovoltaic Applications, Cryst. Growth Des., 2009, 9(10), 4519–4523.
7. R. S. Dibbell, D. G. Youker and D. F. Watson, Excited-State Electron Transfer from CdS Quantum Dots to TiO₂ Nanoparticles via Molecular Linkers with Phenylene Bridges, J. Phys. Chem., 2009, 113, 18643–18651.
8. M. E. Harakeh and L. Halaoui, Enhanced Conversion of Light at TiO₂ Photonic Crystals to the Blue of a Stop Band and at TiO₂ Random Films Sensitized with Q-CdS: Order and Disorder, J. Phys. Chem. C, 2010, 114, 2806–2813.
9. L. Li, X. Yang, J. Gao, H. Tian, J. Zhao, A. Hagfeldt and L. Sun, Highly Efficient CdS Quantum Dot-Sensitized Solar Cells Based on a Modified Polysulfide Electrolyte, J. Am. Chem. Soc., 2011, 133, 8458–8460.
10. Y. L. Lee, B. M. Huang and H. T. Chien, Highly Efficient CdSe-Sensitized TiO₂ Photoelectrode for Quantum-Dot-Sensitized Solar Cell Applications, Chem. Mater., 2008, 20, 6903–6905.
11. A. Kongkanand, K. Tvrdy, K. Takechi, M. Kuno and P. V. Kamat, Quantum Dot Solar Cells. Tuning Photoresponse through Size and Shape Control of CdSe–TiO₂ Architecture, J. Am. Chem. Soc., 2008, 130, 4007–4015.
12. J. H. Bang and P. V. Kamat, Quantum Dot Sensitized Solar Cells. A Tale of Two Semiconductor Nanocrystals: CdSe and CdTe, ACS Nano, 2009, 3(6), 1467–1476.
13. J. B. Sambur, S. C. Riha, D. J. Choi and B. A. Parkinson, Influence of Surface Chemistry on the Binding and Electronic Coupling of CdSe Quantum Dots to Single Crystal TiO₂ Surfaces, Langmuir, 2010, 26, 74839–74847.
14. V. Chakrapani, D. Baker and P. V. Kamat, Understanding the Role of the Sulfide Redox Couple (S²⁻/S_n²⁻) in Quantum Dot-Sensitized Solar Cells, J. Am. Chem. Soc., 2011, 133, 9607–9615.
15. H. Wang, C. Luan, X. Xu, S. V. Kershaw and A. L. Rogach, In Situ versus ex Situ Assembly of Aqueous-Based Thioacid Capped CdSe Nanocrystals within Mesoporous TiO₂ Films for Quantum Dot Sensitized Solar Cells, J. Phys. Chem., 2012, 116, 484–489.
16. X. F. Gao, H. B. Li, W. T. Sun, Q. Chen, F. Q. Tang and L. M. Peng, CdTe Quantum Dots-Sensitized TiO₂ Nanotube Array Photoelectrodes, J. Phys. Chem. C, 2009, 113, 7531–7535.
17. X. Wang, H. Zhu, Y. Xu, H. Wang, Y. Tao, S. Hark, X. Xiao and Q. Li, Aligned ZnO/CdTe Core–Shell Nanocable Arrays on Indium Tin Oxide: Synthesis and Photoelectrochemical Properties, ACS Nano, 2010, 4(6), 3302–3308.
18. E. P. A. M. Bakkers, A. L. Roest, A. W. Marsman, L. W. Jenneskens, L. I. de, J. V. Steensel, J. J. Kelly and D. Vanmaekelbergh, Characterization of Photoinduced Electron Tunneling in Gold/SAM/Q-CdSe Systems by Time-Resolved Photoelectrochemistry, J. Phys. Chem. B, 2000, 104, 7266–7272.
19. T. W. Zeng, I. S. Liu, K. T. Huang, H. C. Liao, C. T. Chien, D. K. P. Wong, C. W. Chen, J. J. Wu, Y. F. Chen and W. F. Su, Effects of bifunctional linker on the optical properties of ZnO nanocolumn-linker-CdSe quantum dots heterostructure, J. Colloid Interface Sci., 2011, 358, 323–328.
20. I. M. Seró, S. Gimènez, T. Moehl, F. F. Santiago, T. L. Villareal, R. Gómez and J. Bisquert, Factors determining the photovoltaic performance of a CdSe quantum dot sensitized solar cell: the role of the linker molecule and of the counter electrode, Nanotechnology, 2008, 19, 424007.
21. N. Guijarro, T. L. Villarreal, I. M. Seró, J. Bisquert and R. Gómez, CdSe Quantum Dot-Sensitized TiO₂ Electrodes: Effect of Quantum Dot Coverage and Mode of Attachment, J. Phys. Chem. C, 2009, 113, 4208–4214.
22. C. Chen, G. Ali, S. H. Yoo, J. M. Kum and S. O. Cho, Improved conversion efficiency of CdS quantum dot-sensitized TiO₂ nanotube-arrays using CuInS₂ as a co-sensitizer and an energy barrier layer, J. Mater. Chem., 2011, 21, 16430–16435.
23. C. Chen, F. Li, G. Li, F. Tan, S. Li and L. Ling, Double-sided transparent electrodes of TiO₂ nanotube arrays for highly efficient CdS quantum dot-sensitized photoelectrodes, J. Mater. Sci., 2014, 49, 1868–1874.
24. S. A. Pawar, D. S. Patil, A. C. Lokhande, M. G. Gang, J. C. Shin, P. S. Patil and J. H. Kim, Chemical synthesis of CdS onto TiO₂ nanorods for quantum dot sensitized solar cells, Opt. Mater., 2016, 58, 46–50.
25. J. Tao, Z. Gong, G. Yao, Y. Cheng, M. Zhang, J. Lv, S. Shi, G. He, X. Chen and Z. Sun, Enhanced photocatalytic and photoelectrochemical properties of TiO₂ nanorod arrays sensitized with CdS nanoplates, Ceram. Int., 2016, 42, 11716–11723.
26. R. Loef, A. J. Houtepen, E. Algorn, J. Schoonman and A. Goossens, Study of Electronic Defects in CdSe Quantum Dots and Their Involvement in Quantum Dot Solar Cells, Nano Lett., 2009, 9(2), 856–859.
27. M. Seró and J. Bisquert, Breakthroughs in the Development of Semiconductor-Sensitized Solar Cells, J. Phys. Chem. Lett., 2010, 1, 3046–3052.
28. M. Seró, D. Gross, T. Mittereder, A. A. Lutich, A. S. Susha, T. Dtrich, A. Belaidi, R. Caballero, F. Langa, J. Bisquert and A. L. Rogach, Nanoscale Interaction Between CdSe or CdTe Nanocrystals and Molecular Dyes Fostering or Hindering Directional Charge Separation, Small, 2010, 6, 221–225.
29. R. S. Mane, B. R. Sankapal and C. D. Lokhande, Photoelectrochemical (PEC) characterization of chemically deposited Bi₂S₃ thin films from non-aqueous medium, Mater. Chem. Phys., 1999, 60, 158–162.
30. R. Gomez and P. Salvador, Photovoltage dependence on film thickness and type of illumination in nanoporous thin film electrodes according to a simple diffusion model, Sol. Energy Mater. Sol. Cells, 2005, 88, 377–388.
31. R. S. Mane, B. R. Sankapal and C. D. Lokhande, Photoelectrochemical cells based on chemically deposited nanocrystalline Bi₂S₃ thin films, Mater Chem. Phys., 1999, 60, 196–203.
32. S. F. Shaikh, R. S. Mane and O.-S. Joo, Mass scale sugar-mediated green synthesis and DSSCs application of tin oxide nanostructured photoanode: Effect of zinc sulphide layering on charge collection efficiency, Electrochim. Acta, 2014, 147, 408–417.
33. X. Yang, A. Wolcott, G. Wang, A. Sobo, R. C. Fitzmorris, F. Qian, J. Z. Zhang and Y. Li, Nitrogen-Doped ZnO Nanowire Arrays for Photoelectrochemical Water Splitting, Nano Lett., 2009, 6(9), 2331–2336.
34. J. Bisquert, Theory of the Impedance of Electron Diffusion and Recombination in a Thin Layer, J. Phys. Chem. B, 2002, 106, 325.
35. P. Sudhagar, J. H. Jung, S. Park, R. Sathyamoorthy, H. Ahn and Y. S. Kang, Self-assembled CdS quantum dots-sensitized TiO₂ nanospheroidal solar cells: Structural and charge transport analysis, Electrochim. Acta, 2009, 55, 113–117.
36. K. Park, Q. Zhang, B. B. Garcia and G. Cao, Effect of Annealing Temperature on TiO₂–ZnO Core–Shell Aggregate Photoelectrodes of Dye-Sensitized Solar Cells, J. Phys. Chem. C, 2011, 115, 4927–4934.
37. K. Park, Q. Zhang, B. B Garcia, X. Zhou, Y.-H. Jeong and G. Cao, Effect of an Ultrathin TiO₂ Layer Coated on Submicrometer-Sized ZnO Nanocrystallite Aggregates by Atomic Layer Deposition on the Performance of Dye-Sensitized Solar Cells, Adv. Mater., 2010, 22, 2329–2332.
38. Y.-S. Lee, C. V. V. M. Gopi, M. V. Haritha and H.-J. Kim, Recombination control in high-performance quantum dot-sensitized solar cells with novel TiO₂/ZnS/CdS/ZnS heterostructure, Dalton Trans., 2016, 45, 12914–12923.
39. W. Hu, T. Liu, Y. Guo, S. Luo, H. Shen, H. He, N. Wang and H. Lin, Highly Efficient Solid-State Solar Cells Based on Composite CdS–ZnS Quantum Dots, J. Electrochem. Soc., 2015, 162, H747–H752.
40. D. V. Shinde, D. Y. Ahn, V. V. Jadhav, D. Y. Lee, N. K. Shrestha, J. K. Lee, H. Y. Lee, R. S. Mane and S. H. Han, A coordination chemistry approach for shape controlled synthesis of indium oxide nanostructures and their photoelectrochemical properties, J. Mater Chem. A, 2014, 2, 5490–5498.
41. P. K. Baviskar, D. P Dubal, S. Majumder, A. Ennaoui and B. R. Sankapal, “Basic idea, advance approach”: Efficiency boost by sensitization of blended dye on chemically deposited ZnO films, J. Photochem. Photobiol., A, 2016, 318, 135–141.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra14275c

---