Optimizing the deposition of CdSe colloidal quantum dots on TiO₂ film electrode via capping ligand induced self-assembly approach

Abstract

The deposition method for immobilizing quantum dots on a TiO₂ film electrode is crucial for the photovoltaic performance of the resulting quantum dot sensitized solar cells (QDSCs). The recently developed capping ligand-induced self-assembly approach has been demonstrated to be an effective deposition route with fast, dense and uniform immobilization of the pre-prepared colloidal QDs on the TiO₂ film electrode. In order to make this technique more reproducible and applicable, the influences of variable experimental parameters, including pH, free ligand concentration, deposition temperature, and QD concentration, on the loading amount and photovoltaic performance of the resulting cell devices have been systematically investigated with the use of CdSe QDs as a model. The intrinsic mechanism of the effects from these investigated variables has been elaborated primarily. An average power conversion efficiency of 6.49% (J_sc = 15.91 mA cm⁻², V_oc = 0.618 V, FF = 0.660) under the irradiation of full one sun (AM 1.5G) has been obtained for the CdSe-based QDSCs under the optimum QD deposition conditions.

---

1. Introduction

Quantum dot sensitized solar cells (QDSCs) have been regarded as a promising candidate for low-cost third-generation solar cells because of the distinguished advantages of colloidal QDs, such as their tunable band gap, high extinction coefficient, multiple exciton generation (MEG), large dipole moment, and solution processability.^1–5 However, the current best power conversion efficiency (PCE) of QDSCs is only at the level of 7–8%,^6–8 which is inferior to those of its analogue dye sensitized solar cells (DSCs, at the level of 12–13%).^9,10 The current moderate performance of QDSCs is partially ascribed to the low QD loading on the TiO₂ film electrode, which plays an crucial role in determining the light-harvesting capability, and therefore directly relates to the photocurrent and PCE of the resultant cell devices.

Tethering QDs on a TiO₂ mesoporous film electrode at monolayer or sub-monolayer coverage is a prerequisite for high light-harvesting capacity and therefore a high photocurrent and high PCE of the resultant cell device.¹¹ Traditionally, two general approaches have been developed to tether QDs on TiO₂ film electrodes: (1) the direct growth of QDs through chemical bath deposition (CBD),^12–15 successive ionic layer adsorption and reaction (SILAR) techniques,^16,17 or electrodeposition,¹⁸ and (2) post-synthesis assembly using pre-synthesized colloidal QDs via direct attachment,^19–21 electrophoretic deposition (EPD),^22–24 or a linker-assisted assembly route.^{6–8,25–27} The linker-assisted assembly approach can be divided into two methods.^28,29 In the first method, the TiO₂ film electrode is firstly functionalized with bifunctional linker molecules and then immersed into a QD dispersion in nonpolar solvents. In the second method, i.e. the capping ligand-induced self-assembly (CLIS) method, bifunctional ligand molecules (mainly thiolcarboxylic acid) capped water-soluble QDs are immobilized onto the untreated TiO₂ film electrode surface induced by the affinity between the terminal carboxylic group of the capping molecules and TiO₂.^30–33 In this method, the deposition of QDs on the TiO₂ film electrode is similar to the case of the carboxylic terminated molecular dyes absorbed onto a TiO₂ electrode in a DSC system. The CLIS method for the sensitization of the TiO₂ electrode has become more and more popular in the construction of high efficiency QDSCs in recent years because the superior electronic properties of the pre-synthesized colloidal QDs after being bound on the TiO₂ substrate can be retained by this method.²⁹ Furthermore, the interfacial properties between the QDs and the TiO₂ substrates can be flexibly designed by the CLIS method. Benefitting from the CLIS method for electrode sensitization, the PCE of QDSCs has improved steadily from 4–5% to 8–9%.^{6–8,34–38}

Although the CLIS method has been adopted for the construction of a series of record efficiency QDSCs, the influence of experimental parameters in the QD deposition process on the performance of the resulting cell devices has not been investigated and optimized systematically. Furthermore, some divergent viewpoints and even paradoxical conclusions exist in this sensitization method. For example, some previous literatures have demonstrated that the CdSe QD loading amount could be effectively enhanced when the pH value matches the point of zero charge of TiO₂ nanoparticles (low to pH 7),^31,39 while in some other reports,^40,41 a higher pH (10–12) has been demonstrated to favor the loading of CdTe QDs on TiO₂ film electrodes. Based on these results, it is imperative that the experimental parameters for QD deposition via the CLIS method should be investigated and optimized systematically, since these parameters affect the amount of QD loading and the photovoltaic performance of the resultant cell devices.

Herein, we focus on the investigation of the influences of experimental parameters in the process of CdSe QD deposition with use of the CLIS method, on the amount of QD loading and the photovoltaic performance of the resultant cell devices. The investigated experimental variables include: the pH of the QD solution, the free ligand, the deposition temperature, and the concentration of the QDs. The experimental results demonstrate that both the deposition temperature and the concentration of the QD solution have nearly no influence, while the free capping ligand and pH value have major effects on the amount of QD loading and the performance of the resultant cell devices. Under the optimum deposition conditions, an average PCE of 6.49% has been obtained for CdSe based QDSCs. It is expected that this work could pave the way for the construction of high efficiency QDSCs.

2. Experimental section

Chemicals

Oleic acid (90%), and 3-mercaptopropionic acid (MPA, 98%) were received from Alfa Aesar. Oleylamine (OAm, 95%), trioctylphosphine (TOP, 90%), selenium powder (200 mesh, 99.99%), 1-octadecene (ODE, 90%), and cadmium oxide (CdO, 99.99%) were obtained from Aldrich. All reagents were used as received without further treatment.

Preparation of CdSe QDs and sensitization of photoanode

The TiO₂ mesoporous film electrode with use of a fluorine doped SnO₂ (FTO) glass substrate consisting of a 9.0 ± 0.5 μm transparent layer and a 6.0 ± 0.5 μm scattering layer was prepared according to our previous report.⁴² The original oil-soluble OAm-capped CdSe QDs with the first excitonic absorption peak at ∼618 nm were synthesized through a hot-injection method.⁴³ The optical properties and the TEM image of the OAm-capped CdSe QDs are shown in Fig. S1 of ESI.† The water soluble MPA-capped CdSe QDs were obtained from the original oil-soluble OAm-capped CdSe QDs via a ligand exchange procedure with use of MPA as the phase transfer reagent.⁴⁴ The purified MPA-capped CdSe QDs were re-dispersed in deionized water with a resulting QD concentration of 3.6 μM (corresponding absorbance at 618 nm is 2.0). Then, MPA was added into the above solution with an MPA concentration of 0.1 M and the pH value of the solution was adjusted to 10.0 through the use of 10% NaOH solution. The obtained QD solution can be used for the sensitization of the TiO₂ film electrode by dipping the mesoporous films into the QD solution and keeping for 2 h at room temperature (25 °C). Then, the film was rinsed by water and ethanol and dried. The resultant films were further overcoated with a ZnS layer by dipping them into 0.1 M Zn(OAc)₂ and Na₂S solutions alternately for 1 min per dip for four cycles, and a SiO₂ layer by immersing in a tetraethoxysilane (TEOS) solution at 35 °C for 2 h.⁷ CdSe-sensitized photoanodes were obtained. In the controlled experiments, QD solutions with different compositions were adjusted accordingly.

Assembly of solar cells

The solution of polysulfide electrolyte was prepared by adding Na₂S (2.0 M), S powder (2.0 M), and KCl (2.0 M) into deionized water. The Cu₂S/brass counter electrodes were obtained by immersing a brass foil in HCl solution (1.0 M) at 70 °C for 5 min, followed by immersing in polysulfide electrolyte solution (2.0 M) for 10 min. The cell devices were assembled by clamping the Cu₂S counter electrode and QD-sensitized photoanode using a binder clip followed by injecting 10 μL of electrolyte.

Characterization

The UV-vis absorption and PL emission spectra were measured on an UV-visible spectrophotometer (Shimadzu UV-3101 PC) and a fluorescence spectrophotometer (Cary Eclipse Varian) respectively. Current–voltage characterization (J–V curves) was carried out on a Keithley 2400 source meter and an AM 1.5G solar simulator (Oriel, Model no. 94022A) was used with the illumination intensity at 100 mW cm⁻². The photoactive area was settled at 0.237 cm² by covering a black metal disk. Incident photon-to-current conversion efficiency (IPCE) spectra were measured using a Keithley 2000 multimeter, and a spectral product DK240 monochromator with a 300 W tungsten lamp. Transmission electron microscopy (TEM) images were obtained on a JEOL JEM-2100 microscope with an accelerating voltage of 200 kV. The analysis of dynamic light scattering (DLS) of the QD aqueous solution was carried out using a Zeta Sizer nano series laser light scattering system (Malvern Instrument Corporation).

3. Results and discussion

In order to achieve high efficiency QDSCs, a dense and uniform distribution of QDs on the TiO₂ film electrode is crucial. Due to the relative chemical stability and short carbon chain length, MPA is the most used bifunctional molecule linker molecule in the capping ligand-induced self-assembly (CLIS) method to tether QDs onto a mesoporous TiO₂ film in the construction of high efficiency QDSCs. To demonstrate the influence of the experimental variables in the CLIS process on the performance of the resulting CdSe cells, and therefore obtain the optimal experimental conditions and best photovoltaic performance, the deposition experimental parameters, such as the pH of the QD solution, the presence of the free ligand, the deposition temperature and the QD concentration, have been systematically investigated. A CdSe QD with the first absorption peak at 618 nm was selected as the sensitizer due to the excellent photovoltaic performance of the resulting QDSCs and the reproducible nature of its synthesis.⁴⁵ All adopted CdSe QDSCs were constructed employing polysulfide electrolyte as the redox media and Cu₂S/brass as the counter electrode. To minimize the sampling deviation, the average photovoltaic performance of five cells in parallel under each studied condition was used to evaluate the effect from the studied variables.

Influence of pH value in the absence of free ligand

Since the protonation and deprotonation of the capping ligand around the QD surface and the surface electronic charge of the TiO₂ substrate are heavily dependent on pH value of the solution media,⁴⁶ the pH value of the QD aqueous solution is expected to influence the deposition of QD on the TiO₂ electrode, and consequently the performance of the resultant cell devices. The effect of pH of the QD aqueous solution in the absence of the free MPA ligand was studied, while all the other parameters such as the TiO₂ film electrode thickness and configuration, the compositions in each QD solution except for the pH value, and the deposition temperature were kept constant as specified in the Experimental section in the process of QD sensitization via CLIS method. Acknowledged from the literature report,^43,47 MPA-capped CdSe QDs are well-dispersed in weak basic media, while they show a progressive agglomeration and precipitation in strong basic or acidic media. Our own experimental results also indicate that the obtained MPA-capped CdSe QD aqueous dispersion can be stable only in the pH range of 8–12, therefore in this study, the investigated pH value of the QD solution was set at the range of 8.0 to 12.0. In the experiment, 5 pieces of TiO₂ film electrodes from the same batch were immersed in QD aqueous solutions with different pH values (8.0, 9.0, 10.0, 11.0 and 12.0) for a certain period, and the corresponding absorption spectra were measured to reflect the loading amount of the CdSe QDs on the TiO₂ film electrode. It is noted that in the absorption spectra study, in order to avoid the scattering effect from large sized particles, the used TiO₂ film electrodes, which are different from those used in the construction of cell devices, contain only a 5.0 μm transparent layer with dimensions of 2.0 cm × 1.0 cm. The temporal evolution of the absorption spectra of the sensitized film electrodes under CdSe QD solutions with different pH values were measured to monitor the amount of CdSe QD loading. From these absorption spectra, the temporal evolution of absorbance at the first absorption excitonic peak (herein 618 nm) was then obtained and is illustrated in Fig. S2a of ESI.† It was found that the absorbance (i.e. the loading amount of CdSe QD sensitizer) approached a saturation value in a period of 2 h under each pH condition and no further enhancement was observed upon extending the deposition time. It was found that the saturation adsorption time is 2 h and no more QDs can be anchored onto the TiO₂ film after then. The absorption spectra with the maximum loading amounts under different pH conditions are shown in Fig. 1a with corresponding photographs of the film electrode in the inset. The absorption peaks of the five sensitized film electrodes stay at nearly the same position as that of the colloidal QD solution at around 618 nm. This indicates that the agglomeration of QD particles has not occurred. As can be seen from Fig. 1a, the intensity of the absorption peaks of the different pH based QD-sensitized TiO₂ films are in the sequence of pH 12.0 > pH 11.0 > pH 10.0 > pH 9.0 > pH 8.0, with a normalized ratio of 3.04 : 2.89 : 2.67 : 2.28 : 1. Due to the fact that the absorbance is directly related to the loading amounts of the QDs onto the TiO₂ films, the amount of QD loading also obeys the same sequence of pH 12.0 > pH 11.0 > pH 10.0 > pH 9.0 > pH 8.0 with a relative ratio of 3.04 : 2.89 : 2.67 : 2.28 : 1. Furthermore, the systematically deepening coloration of the sensitized electrodes, as shown in the inset of Fig. 1a, also demonstrates this variation trend. From this result, it is clearly demonstrated that a high pH value in the QD solution favors QD loading on the TiO₂ electrode.

Fig. 1 (a) Absorption spectra of CdSe QD-sensitized TiO₂ films under different pH conditions. Inset: photographs of the corresponding TiO₂ films. (b) Dynamic light scattering graphs of three CdSe QDs aqueous solutions of different pH values.

Due to the large size of QD particles compared to the mesoporous channel in the film electrode, the prerequisite for high QD loading is the availability of a well-dispersible QD aqueous solution, i.e. QDs are dispersed in the form of isolated particles without agglomeration. The average hydrodynamic diameters of the QDs in different pH solutions were measured by dynamic light scattering (DLS), and the results are shown in Fig. 1b. It was found that the average hydrodynamic diameter of the QDs becomes larger with a decreasing pH value of the QD solutions. This means that the agglomeration of QDs in solution occurs with decreasing pH in solution. This result primarily indicates that the dispersibility of the QDs determines the amount of QD loading on the TiO₂ film electrode. A high pH value of the QD solution favors the dispersibility of QDs, which consequently benefits the QD loading. The reason is ascribed to the deprotonation of the carboxylic group from the QD surface binding MPA ligand at high pH, which renders the surface of the QD negatively charged. The mutual electrostatic repulsion of the QD particles therefore undoubtedly improves the colloidal stability and dispersibility. Another factor relating to pH is that a high pH value favors the deprotonation of the thiol group in MPA and causes the formation of a thiolate group. The interaction between the MPA ligand and the CdSe QDs could be considered as a coordination bond. According to previous reports,^47–50 the binding strength for the thiolate group (deprotonated products of thiols) to a Cd atom at the QD surface is about 40 times higher than that for the thiol group to Cd. The strengthened interaction between the capping ligand and the QD surface would improve the colloidal stability and dispersibility of the resultant QD, and therefore enhance the QD loading amount on the TiO₂ film electrode. A high pH value of QD solution favors the dispersibility of QDs, which consequently benefits the QD loading. A high QD loading means that more incident photons can be harvested and a higher photocurrent and PCE are expected in the resulting cell devices. To confirm this, the corresponding photovoltaic performance of the resultant solar cells was investigated below.

After the saturation of QD sensitization, regenerated sandwich-type QDSCs were assembled with use of the sensitized film electrode as the photoanode, Cu₂S/brass as the counter electrode, and a S_n²⁻/S²⁻ aqueous solution as the electrolyte media. The J–V curves of five champion QDSCs corresponding to photoanode deposition under the different pH conditions as discussed above were measured under the illumination of a AM 1.5G solar simulator at the intensity of one full sun (100 mW cm⁻²). The results are presented in Fig. 2a and the corresponding average photovoltaic parameters are summarized in Table 1 and details for each cell are listed in Table S1 of the ESI.† Both the average open-circuit voltage (V_oc) and fill factor (FF) show no significant differences among the five cells, while the average short-circuit density (J_sc) displays a significant enhancement from 5.26 to 15.30 mA cm⁻² with the increase in the pH value of the QD solution, as expected. The ratio of J_sc of each QDSC device is 1 : 2.26 : 2.67 : 2.82 : 2.91 from pH 8 to 12, which is highly consistent with the ratio of absorbance corresponding to the sensitized photoanodes under different pH conditions. Benefitting from the higher J_sc, the cell corresponding to the QD deposition at pH 12 exhibits the best PCE of 6.03 with a V_oc value of 0.614 V, a J_sc value of 15.30 mA cm⁻² and a FF of 0.642.

Fig. 2 (a) J–V curves and (b) IPCE curves of CdSe QDSCs prepared at different pH values in the absence of free ligands.

Table 1 The average photovoltaic performance of solar cells based on different pH values of CdSe QD aqueous solutions

pH	Voc (V)	Jsc (mA cm^−2)	FF	PCE (%)
8	0.613	5.26	0.653	2.14 ± 0.07
9	0.614	11.90	0.643	4.70 ± 0.10
10	0.611	14.04	0.640	5.49 ± 0.11
11	0.610	14.85	0.645	5.83 ± 0.06
12	0.614	15.30	0.642	6.03 ± 0.09

---

The generation of a higher photocurrent under a higher pH value is further verified by the incident-photo-to-carrier conversion efficiency (IPCE) spectra of the five pH based QDSCs, as shown in Fig. 2b. The curves of the IPCE spectra show an identical spectral profile and photoresponse range for the five different cell devices. In consistency with the trend of absorbance, the highest IPCE value for each QDSC increases from 27.42 to 74.65% with the increase in pH, which can be explained by the superior light-harvesting efficiency due to the greater loading amount of QD sensitizers under conditions of high pH.

Influence of the free ligand in QD aqueous solution at pH 12.0

The above results indicate that the QD dispersity has a heavy effect on the amount of QD loading. Previous reports on colloidal QD solutions tell us that the presence of free capping ligands can improve the dispersibility of QD solutions remarkably.^51–53 The instability of the QD dispersion is mainly caused by ligand desorption from the surface of the QD. When the free ligand is present in the QD solution, the free ligand combines with the QD and keeps the nanocrystals stable in solution for a longer period of time. Therefore, the effect of the free ligand MPA in the QD solution was studied. In this experiment, the effect of the concentration of the free ligand MPA in the QD solution with a pH of 12 was investigated. Our experimental results indicate that stable QD aqueous solutions can be obtained with the free ligand MPA in the concentration range of 0–0.3 M. Therefore, the investigated MPA concentration range was set at 0–0.3 M. In similarity to the above experiment, except for the concentration of MPA, all the other experimental variables were the same. 3 pieces of TiO₂ film electrodes from the same batch were immersed in each QD aqueous solution with different MPA concentrations (0, 0.05, 0.1, 0.3 M), and the relative loading amounts were characterized by the absorption spectra of the sensitized film electrode, which are shown in Fig. 3a. The saturation adsorption time of the CdSe QDs onto the TiO₂ electrode is 2 h and the absorbance can not be enhanced by extending the deposition time. The curves of time evolution of the absorption spectra at the first absorption peak of the corresponding samples are shown in Fig. S2b of ESI.† All of the absorption spectra have similar profiles but with different absorption intensities. The intensities of the absorption peaks follows the sequence of 0.3 M ≈ 0.1 M > 0.05 M > 0 M with a normalized ratio of 1 : 1 : 0.95 : 0.87. This result is also verified by the coloration in the corresponding photograph of the sensitized film electrodes, which is shown in the inset of Fig. 3a. This result clearly indicates that with the presence of the free ligand in the QD solution (herein the proper concentration in the range of 0.1–0.3 M), the amount of QD loading can be enhanced by ∼10%. In order to further determine whether the enhancement in QD loading is ascribed to the improvement of the QD dispersibility in solution, DLS measurements were also carried out for QD solutions in the presence of different concentrations of MPA, and the results are shown in Fig. 3b. It is found that the average hydrodynamic diameter of the QDs in solution with different MPA concentrations follows the sequence of 0.3 M ≈ 0.1 M < 0 M. Clearly, the presence of the free ligand MPA in the QD dispersion favors the reduction of the hydrodynamic diameter of the QDs, and this results in higher QD loading on the film electrode. Therefore, we can conclude that the enhancement of QD loading benefits from the improvement of QD dispersibility. The –S–Cd bond between the capping ligand MPA and the surface of QD is not strong enough, and the dissociation and re-coordination of the ligand in the QD dispersion is a dynamic process. With detachment of MPA from the QD, the agglomeration of QDs would occur, whereas with the presence of free MPA, the dissociation of the ligand from the QD surface would be hindered, and therefore the colloidal stability and dispersibility of QD would be improved.

Fig. 3 (a) Absorption spectra of CdSe-sensitized TiO₂ films with different concentrations of free MPA. Inset: photographs of the corresponding TiO₂ films. (b) Dynamic light scattering graphs of three CdSe aqueous solutions with different free MPA concentrations.

Similarly, the photovoltaic performances of the cell devices derived from the photoanode deposited under different free ligand concentrations were measured under standard conditions. The J–V curves for the champion cells are shown in Fig. 4a, the average performance from 5 samples in parallel are listed in Table 2, and the photovoltaic parameters for each cell are available in Table S2 of the ESI.† As found in Table 2, the cell devices corresponding to photoanodes with higher QD loading under conditions of free MPA concentration of 0.1 to 0.3 M have higher PCE values in comparison with the cells with lower QD loading in the absence of MPA ​and MPA concentration of 0.05 M. The higher PCE is derived from the higher J_sc (15.87, 15.91 vs. 15.63, 15.21 mA cm⁻²), while V_oc and FF remain almost constant for all of the cell devices. The IPCE results, as shown in Fig. 4b, give the same observation. Cell devices corresponding to the absence of free MPA and 0.05 M MPA give lower IPCE values in comparison with cell devices corresponding to 0.1–0.3 M MPA. Both the optical and photovoltaic results indicate that the presence of free MPA with a concentration in the range of 0.1–0.3 M in the CdSe QD solution favors the loading of QDs on the TiO₂ film electrode, and consequently benefits the performance of the resulting cell devices.

Fig. 4 (a) The J–V curves and (b) resulted IPCE curves of QDSCs prepared under different concentrations of the free MPA ligand.

Table 2 Average photovoltaic performance of solar cells based on different concentrations of free MPA ligand

MPA (M)	Voc (V)	Jsc (mA cm^−2)	FF	PCE (%)
0	0.612	15.21	0.642	5.96 ± 0.06
0.05	0.618	15.63	0.649	6.27 ± 0.06
0.1	0.618	15.91	0.660	6.49 ± 0.07
0.3	0.617	15.87	0.661	6.48 ± 0.08

---

Influence of pH value in the presence of 0.1 M free ligand

In the above experiment, the effect of pH was studied in the absence of the free ligand. However, the presence of the free ligand also has a significant influence on the QD loading. Therefore, an experiment to investigate the influence of the pH value in the presence of 0.1 M free MPA was designed and employed. In similarity to the above experiments, except for the pH, all the other experimental variables are the same. TiO₂ film electrodes were immersed in CdSe aqueous solutions with the presence of 0.1 M free MPA but with different pH values (8, 9, 10, 11, 12) for 2 h, and then the absorption spectra of the sensitized film electrodes were measured, which are shown in Fig. 5a. In contrast to the case with the absence of free MPA, wherein the pH of the QD solutions has a remarkable influence on the QD loading, herein the pH has almost no effect on the QD loading with nearly identical absorption peak intensities in the absorption spectra. The J–V curves for the champion cells are shown in Fig. 5b, the average performances from 5 samples in parallel are listed in Table 3, and the photovoltaic parameters for each cell are available in Table S3 of the ESI.† Correspondingly, the J–V measurement results from all the five groups of cells show also the same performance for the champion cells in each group of cells. These experimental results suggested that with the presence of 0.1 M free MPA in the QD solution, the loading amount of QDs on the TiO₂ film electrode is insensitive to the pH of the QD solution, and the appropriate pH value range can be expanded to a wide range of 8–12. This brings about a high reproducibility for the CLIS deposition approach.

Fig. 5 (a) UV-vis absorption spectra of CdSe QD-sensitized films which contain only a transparent layer prepared under different pH values with 0.1 M of free MPA ligands, (b) the J–V curves of five CdSe-sensitized QDSCs prepared at different pH values.

Table 3 Average photovoltaic performance of CdSe QDSCs based on different pH values with the presence of 0.1 M free MPA ligand

pH	Voc (V)	Jsc (mA cm^−2)	FF	PCE (%)
8	0.617	15.96	0.652	6.43 ± 0.07
9	0.618	15.93	0.655	6.45 ± 0.05
10	0.619	15.93	0.658	6.49 ± 0.06
11	0.618	15.91	0.655	6.45 ± 0.03
12	0.617	15.94	0.656	6.46 ± 0.07

---

Influence of deposition temperature and QD concentration

In the above experiments, the optimum pH value and the concentration of free MPA in the CdSe QD solution have been ascertained. Herein, the effect of other variables, such as the temperature for the deposition process and the concentration of QDs in solution, on the QD loading were also exploited. In the temperature experiments, a QD solution with a pH of 10.0 and 0.1 M of free MPA was stored in a fridge with a temperature of 5 °C, or in an oven with a temperature of 25, or 50 °C, respectively. It is noted that the container for the QD solution is closed in the whole deposition process to avoid evaporation of the solvent. The experimental results indicate that the period for reaching the saturation loading amount is extended to about 4 h when the solution temperature is lowered to 5 °C, while only 1.0 h is needed to reach the saturation loading amount when the solution temperature is raised to 50 °C. This observation is reasonable since both the diffusion rate of the QDs and the coordination reaction rate for the terminal carboxyl and TiO₂ are temperature-dependent. However, the absorption spectra for the sensitized film electrodes under different deposition temperatures, as shown in Fig. S3 of the ESI,† give a nearly identical peak intensity corresponding to an equal amount of QD loading. Furthermore, the photovoltaic measurement results indicate that all the cell devices corresponding to different deposition temperatures show nearly the same performance as listed in Tables 4 and S4 of the ESI.† Hence, we can conclude that the deposition temperature can influence the deposition rate, but has no observable effect on the loading amount and on the performance of the resultant cell devices. Therefore, from the view of convenience, room temperature of 25 °C is suggested to be adopted in the CLIS method.

Table 4 The average photovoltaic performances of CdSe QDSCs based on different QD deposition temperatures, and concentrations of CdSe QD solution

		Voc (V)	Jsc (mA cm^−2)	FF	PCE (%)
Temperature (°C)	5	0.619	15.93	0.649	6.41 ± 0.06
	25	0.615	15.91	0.650	6.41 ± 0.09
	50	0.615	15.92	0.654	6.40 ± 0.08
Concentration (μM)	3.6	0.618	15.87	0.658	6.45 ± 0.10
	1.8	0.616	15.94	0.658	6.46 ± 0.05
	0.72	0.617	15.94	0.654	6.43 ± 0.06
	0.36	0.616	15.83	0.656	6.40 ± 0.10

---

Meanwhile, the influence of the QD solution concentration was also studied. The investigated QD concentrations were: 3.6 μM, 1.8 μM, 0.72 μM, and 0.36 μM, which correspond to absorbance values at the excitonic peak of 2.0, 1.0, 0.4 and 0.2, respectively. In similarity to the influence of temperature, with a decrease in QD concentration, the saturation deposition time is extended to about 24 h for a QD concentration of 0.36 μM from the 2 h corresponding to a QD concentration of 3.6 μM. However, from the absorption spectra of the four CdSe-sensitized TiO₂ electrode films shown in Fig. S3 of ESI,† the maximum loading amount of the QDs on the TiO₂ film electrode is almost insensitive to the QD concentration. Correspondingly, the photovoltaic performances for the resulting cell devices are also insensitive to the QD concentration during the CLIS process in uploading the QDs on the TiO₂ electrode, as can be seen from Tables 4 and S5 of ESI.† Therefore, the concentration of the QD solution has no significant effect on the amount of QD loading, and on the performance of the resultant cell devices. For purpose of time saving, the a QD concentration of 3.6 μM was selected for the CLIS process.

4. Conclusion

The influence of experimental parameters during the CLIS process, including pH value, free ligand MPA concentration, temperature and concentration of QD solution on the amount of QD loading on a TiO₂ film electrode and on the photovoltaic performance of the resulting cell devices has been systematically investigated and optimized. The experimental results show that the adsorption temperature and concentration of QD have nearly no effects, while the pH value and the presence of the free ligand MPA have heavy influences. The optimum conditions for the CLIS process are the QD aqueous solutions containing 0.1–0.3 M free MPA with a pH value in the range of 8–12. Under the optimum deposition conditions, CdSe based QDSCs with an average PCE of 6.49% under 1 full sun intensity can be obtained with high reproducibility. It is expected that this work could provide a guide to the effective deposition of QDs for high efficiency QDSCs with high reproducibility.

Acknowledgements

We thank the National Natural Science Foundation of China (no. 91433106, 21421004, 21175043), and the Fundamental Research Funds for the Central Universities for financial support.

References

1. P. V. Kamat, K. Tvrdy, D. R. Baker and J. G. Radich, Chem. Rev., 2010, 110, 6664.
2. I. J. Kramer and E. H. Sargent, Chem. Rev., 2014, 114, 863.
3. J. B. Sambur, T. Novet and B. A. Parkinson, Science, 2010, 330, 63.
4. O. E. Semonin, J. M. Luther, S. Choi, H. Y. Chen, J. Gao, A. J. Nozik and M. C. Beard, Science, 2011, 334, 1530.
5. G. Zhang, S. Finefrock, D. Liang, G. G. Yadav, H. Yang, H. Fang and Y. Wu, Nanoscale, 2011, 3, 2430.
6. Z. Pan, I. Mora-Seroó, Q. Shen, H. Zhang, Y. Li, K. Zhao, J. Wang, X. Zhong and J. Bisquert, J. Am. Chem. Soc., 2014, 136, 9203.
7. K. Zhao, Z. Pan, I. Mora-Seroó, E. Canovas, H. Wang, Y. Song, X.-Q. Gong, J. Wang, M. Bonn, J. Bisquert and X. Zhong, J. Am. Chem. Soc., 2015, 137, 5602.
8. S. Jiao, Q. Shen, I. Mora-Seroó, J. Wang, Z. Pan, K. Zhao, Y. Kuga, X. Zhong and J. Bisquert, ACS Nano, 2015, 9, 908.
9. A. Yella, H. W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W. G. Diau, C. Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629.
10. S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, Nat. Chem., 2014, 6, 242.
11. R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojuma, A. Kistamure, M. Shimohigoshi and T. Watanabe, Nature, 1997, 388, 431.
12. W. T. Sun, Y. Yu, H. Y. Pan, X. F. Gao, Q. Chen and L. M. Peng, J. Am. Chem. Soc., 2008, 130, 1124.
13. L. Li, X. Yang, J. Gao, H. Tian, J. Zhao, A. Hagfeldt and L. Sun, J. Am. Chem. Soc., 2011, 133, 8458.
14. Q. Zhang, X. Guo, X. Huang, S. Huang, D. Li, Y. Luo, Q. Shen, T. Toyoda and Q. Meng, Phys. Chem. Chem. Phys., 2011, 13, 4659.
15. X. Huang, S. Huang, Q. Zhang, X. Guo, D. Li, Y. Luo, Q. Shen, T. Toyoda and Q. Meng, Chem. Commun., 2011, 47, 2664.
16. P. K. Santra and P. V. Kamat, J. Am. Chem. Soc., 2012, 134, 2508.
17. J. W. Lee, J. D. Hong and N. G. Park, Chem. Commun., 2013, 49, 6448.
18. C. Lévy-Clément, R. Tena-Zaera, M. A. Ryan, A. Katty and G. Hodes, Adv. Mater., 2005, 17, 1512.
19. A. Zaban, O. I. Micic, B. A. Gregg and A. J. Nozik, Langmuir, 1998, 14, 3153.
20. S. Giménez, I. Mora-Seró, L. Macor, N. Guijarro, T. Lana-Villarreal, R. Gómez, L. J. Diguna, Q. Shen, T. Toyoda and J. Bisquert, Nanotechnology, 2009, 20, 295204.
21. N. Guijarro, T. Lana-Villarreal, I. Mora-Seroó, J. Bisquert and R. Goómez, J. Phys. Chem. C, 2009, 113, 4208.
22. N. J. Smith, K. J. Emmett and S. J. Rosenthal, Appl. Phys. Lett., 2008, 93, 043504.
23. P. K. Santra and P. V. Kamat, J. Am. Chem. Soc., 2013, 135, 877.
24. N. P. Benehkohal, V. González-Pedro, P. P. Boix, S. Chavhan, R. Tena-Zaera, G. P. Demopoulos and I. Mora-Seró, J. Phys. Chem. C, 2012, 116, 16391.
25. A. Kongkanand, K. Tvrdy, K. Takechi, M. Kuno and P. V. Kamat, J. Am. Chem. Soc., 2008, 130, 4007.
26. H. Zhang, K. Cheng, Y. M. Hou, Z. Fang, Z. X. Pan, W. J. Wu, J. L. Hua and X. H. Zhong, Chem. Commun., 2012, 48, 11235.
27. Z. Ning, H. Tian, C. Yuan, Y. Fu, H. Qin, L. Sun and H. Ågren, Chem. Commun., 2011, 47, 1536.
28. D. F. Watson, J. Phys. Chem. Lett., 2010, 1, 2299.
29. J. B. Sambur, S. C. Riha, D. Choi and B. A. Parkinson, Langmuir, 2010, 26, 4839.
30. W. Li and X. Zhong, J. Phys. Chem. Lett., 2015, 6, 796.
31. W. J. Li and X. H. Zhong, Acta Physica Sinica–Chinese Edition, 2015, 64, 038806.
32. W. C. W. Chan and S. Nie, Science, 1998, 281, 2016.
33. J. T. Margraf, A. Ruland, V. Sgobba, D. M. Guldi and T. Clark, Langmuir, 2013, 29, 2434.
34. J. Wang, I. Mora-Seroó, Z. Pan, K. Zhao, H. Zhang, Y. Feng, G. Yang, X. Zhong and J. Bisquert, J. Am. Chem. Soc., 2013, 135, 15913.
35. Z. Pan, K. Zhao, J. Wang, H. Zhang, Y. Feng and X. Zhong, ACS Nano, 2013, 7, 5215.
36. K. Zhao, H. Yu, H. Zhang and X. Zhong, J. Phys. Chem. C, 2014, 118, 5683.
37. Z. Pan, H. Zhang, K. Cheng, Y. Hou, J. Hua and X. Zhong, ACS Nano, 2012, 6, 3982.
38. W. Li, Z. Pan and X. Zhong, J. Mater. Chem. A, 2015, 3, 1649.
39. J. Chen, D. W. Zhao, J. L. Song, X. W. Sun, W. Q. Deng, X. W. Liu and W. Lei, Electrochem. Commun., 2009, 11, 2265.
40. F. Bo, C. L. Wang, S. H. Xu, C. F. Zhang, Z. Y. Wang and Y. P. Cui, J. Phys. D: Appl. Phys., 2014, 47, 305101.
41. S. Xu, C. Wang, H. Zhang, Z. Wang, B. Yang and Y. Cui, Nanotechnology, 2011, 22, 315703.
42. Z. Du, H. Zhang, H. Bao and X. Zhong, J. Mater. Chem. A, 2014, 2, 13033.
43. X. Zhong, Y. Feng and Y. Zhang, J. Phys. Chem. C, 2007, 111, 526.
44. L. Liu and X. Zhong, Chem. Commun., 2012, 48, 5718.
45. X. G. Peng, L. Manna, W. D. Yang, J. Wickham, E. Scher, A. Kadavanich and A. P. Alivisatos, Nature, 2000, 404, 59.
46. A. N. Jumabekov, F. Deschler, D. Böhm, L. M. Peter, J. Feldmann and T. Bein, J. Phys. Chem. C, 2014, 118, 5142.
47. J. Aldana, N. Lavelle, Y. Wang and X. Peng, J. Am. Chem. Soc., 2005, 127, 2496.
48. Y. Q. Liang, J. E. Thorne and B. A. Parkinson, Langmuir, 2012, 28, 11072.
49. Y. Zhang, L. Mi, P.-N. Wang, J. Ma and J.-Y. Chen, J. Lumin., 2008, 128, 1948.
50. W. Lin, W. Zou, Z. J. Du, H. Q. Li and C. Zhang, J. Nanopart. Res., 2013, 15, 1629.
51. M. J. Mulvihill, S. E. Habas, I. Jen-La Plante, J. Wan and T. Mokari, Chem. Mater., 2010, 22, 5251.
52. J. Aldana, Y. A. Wang and X. Peng, J. Am. Chem. Soc., 2001, 123, 8844.
53. O. M. Primera-Pedrozo, Z. Arslan, B. Rasulev and J. Leszczynski, Nanoscale, 2012, 4, 1312.

---

Footnote

† Electronic supplementary information (ESI) available: The TEM image and optical properties of the OAm-capped CdSe QDs. The details of the photovoltaic performance of each tested solar cell based on different pH values of the QD aqueous solution, different concentrations of free MPA ligands, different pH of QD solution with free ligands, different deposition temperatures and different concentrations of QD solution. Temporal evolution of the absorbance at first excitonic absorption peak of CdSe-sensitized TiO₂ film deposition under QD aqueous solution with different pH values and different MPA concentrations. UV-vis absorption spectra of CdSe-sensitized films under different deposition temperatures and concentrations of QD solution. See DOI: 10.1039/c5ra17412k

---