Introduction

The use of semiconductor oxides with solar incidence is a promising technology to energy conversion [1,2,3]. Recent application of chenodeoxycholic acid-substituted dyes, in the third generation of dye-sensitized solar cells (DSSCs) raised up a new study area in PV systems [4]. The capability of suppressing dye-dye interactions and anchoring in the oxide surface become an excellent alternative in DSSCs development.

Dye-sensitized solar cells are made by the use of an electron transport material on transparent conductor oxide (TCO), acting as a scaffold material. There is also the presence of a dye, a photosensitive molecule that inject electron by solar irradiation, providing a charge flow responsible for the solar energy conversion [1,2,3].

Simply, the solar cell operating is based on the dye ability to inject electrons in the conduction band of scaffold oxide, usually TiO2 [5], ZnO [6], and SnO2 [7] with solar energy incidence. The injected electrons, moves through a TCO to a cathode catalyst, generating a charge flow as demonstrated in Reaction 1, (where SO is the semiconductor oxide).

$${\text{SO }}\left( {{\text{e}}^{ - } } \right) + {\text{Cathode}} \to {\text{SO}} + {\text{Cathode }}\left( {{\text{e}}^{ - } } \right)$$
(Reaction 1)
$${\text{SO }}\left( {{\text{e}}^{ - } } \right) + {\text{electrolyte}}^{ + } \to {\text{SO}} + {\text{Electrolyte}}$$
(Reaction 2)

The intermediation of redox species is provided by an electrolyte that has been able to capture the electron in the cathode, reducing the dye [8, 9]. However, another reaction has been able to occur in the oxide/dye/electrolyte interface, the back-electron recombination (Reaction 2) [5, 10].

Suppressing the oxide-electrolyte recombination is one of the most studied ways to produce DSSC with improved photoelectrochemical parameters [11]. From this standpoint, many technologies have been used to reduce Reaction 2: push–pull dyes [12], using mix oxides [13, 14], and electrolyte additives [15]. Recently, Rodrigues and co-workers, [11] applied the poly (4-vinyl pyridine), observing a TiO2 surface passivation, able to suppress the back-electron recombination and increasing the current, even at low dye loading [16, 17]. This related study shows that is possible to use organic additives in dye solar cells, such as benzotriazole (BTAH).

Benzotriazole is used as an effective inhibitor for several metal alloys and demonstrates high solubility in many aqueous media, good thermal stability, and has been able to chemical adsorbed in metal surface, due to nitrogen presence. In other words, some impressive advantages over the use of chenodeoxycholic acid. Furthermore, the equilibrium involving BTAH− and BTAH2− can contribute to physisorption and modulation of a space charge layer in a semiconductor/electrolyte interface [15, 16].

With a dye, BTAH might act efficiently avoiding dye aggregation, and shifting the oxide fermi level (Ef), influencing the charge separation and photoelectrochemical values of the cells [15, 17, 18]. Regarding toxicity and recent applications, concentrations lower than 2.10–2 mol L−1 are not toxic and relate to its use, beyond the corrosion inhibitor, BTAH has been used as protease inhibitor to several virus classes, such as Corona Virus (SARS-CoV) [19, 20].

This work aims to first report the application of BTAH as an additive to DSSC and understand their role in photoanode/electrolyte interface, as a new alternative to produce systems with superior photoelectrochemical parameters.

Experimental

The oxide paste was produced with 3 g the TiO2 (Aldrich® 25 nm 100% anatase), 1 ml of polyethylene glycol, 0.1 mL of acetylacetone, 4 mL of deionized water, and 0.1 mL of Triton-X. TiO2 suspension was coated by Doctor Blading method under FTO (~ 7 Ω sq−1 Aldrich®). After deposition, the films were sintered at 450 °C for 30 min followed by immersion for 12 h in N719 commercial dye solution (2.5 × 10–4 mol L in ethanol) containing 0.148 mg mL−1; 2 mg mL−1 and 20 mg mL−1 of BTAH [10].

The electrolyte was fabricated by mixing 0.5 mol L−1 tert-butylpiridine, 0.6 mol L−1 tetrabutylammonium, 0.1 mol L−1 lithium iodide, and 0.1 mol L−1 of resublimate iodine, in methoxypropionitrile [5]. The cathode catalyst was produced by Pt electrodeposition under FTO, using a three-electrode system, with FTO as a working electrode, Pt as a counter electrode, and Ag/AgCl as a reference electrode, using K2PtCl6 dissolved in HCl as electrolyte [18].

The solar cells were assembled in sandwich format, in an active area of 0.2 cm2. The electrochemical measurements of j-V curves were taken in Zhenium Zahner@ potentiostat with a solar simulator provided by a Xenon lamp at 100 mW cm−2. Electrochemical Impedance Spectroscopy (10 mHz–10 kHz at 10 mV of perturbation), Intensity Modulated Photovoltage Spectroscopy (IMVS), and Intensity Modulated Photocurrent Spectroscopy (IMPS) (100 mHz–1 kHz at 10 mW of perturbation) were performed in an auto lab potentiostat from Metrohm®, coupled with a led lamp of 530 nm at 50 mW cm−2.

Results and discussion

In Fig. 1 it is depicted the j-V curves to dye solar cells analyzed, and the photoelectrochemical results in Table 1.

Fig. 1.
figure 1

j-V curves to solar cells with different BTAH amounts

Full size image
Table 1. Photoelectrochemical parameters to solar cells with different BTAH amounts
Full size table

It is well observed a reduced open circuit potential (Voc) to the cells with the BTAH amounts. It can be explained by the reduction of TiO2 conduction band edge, due to the protonation of oxide by N group contained in BTAH molecules [19, 20]. The reduced Voc can be seen as a negative point to solar cell working, by potentiated charge recombination reaction [21]. On the other hand, some papers demonstrated no relation between lower Voc with back-electron recombination suppression, as showed by Guimaraes and co-workers [12].

As the Voc is the difference between the quasi-Fermi level and the standard redox potential to the electrolyte, it is noted that BTAH amount in N719 dye has been able to shift the Ef level to more positive potentials [22]. One possible key to improve Voc values is synthesizing new dyes containing BTAH in their own composition, not acting only as a spacer, even to composing the sensitizing material [4].

The photoconversion energy efficiency (ɳ) was obtained using Eq. 1, where jsc is the short circuit current, FF fill factor, and Pin the incident power [23].

$$\eta =\frac{{j}_{sc}{V}_{oc}FF}{{P}_{in}}x100$$
(1)

It has been observed superior ɳ value (4.26 ± 0.03) to the cell prepared with BTAH in 2.0 mg mL−1 condition when compared to the standard cell (3.78 ± 0.09). It demonstrated an efficiency boost of 12.8% in the energy conversion, due to superior photocurrent to the same BTAH condition. The ɳ values, when compared to literature were reduced since it was not made TiCl4 treatment and the scattering TiO2 layer was not used [24].

It is also observed, that BTAH in the concentration of 0.148 mg mL−1 has not demonstrated a different behavior when compared to bare cells (due to deviation values). In higher concentrations of 20.0 mg mL−1 and superior, it is expected an inhibition process of the charge collection, because BTAH molecules must occupy all the TiO2 active layer, reducing the current and PCE values [25]. The space charge region in TiO2/electrolyte interface (SCR) should be undergoing changes, improving electron collection rate (Reaction 1), and reducing the recombination reaction demonstrated by Reaction 2 [26]. For a better analysis of its supposed behavior, EIS measurements were taken and Bode diagram is depicted in Fig. 2.

Fig. 2.
figure 2

EIS Bode diagram to solar cells with different BTAH amounts

Full size image

In Bode diagram of Fig. 2, it is observed two different regions. Higher frequencies are related to Pt/Electrolyte interface and in lower frequencies, SCR interface [12]. It is noted a shift to reduced frequency values to the region of working electrode. As frequency and time are inversely proportional, lower frequencies mean a slowness in the charge transfer process, and to confirm this resistance, Nyquist plots (Fig. 3) were fitted using a simplified equivalent circuit (Fig. 1.SI) [11, 27,28,29]. The resistance values of TiO2/Dye, BTAH interface (RWE), and Pt/Electrolyte with the chemical capacitance (Cμ) were estimated in Table 2.

Fig.3.
figure 3

Nyquist plots with fitted values from DSSC with different BTAH amounts (In fitted diagrams, the Zr was shifted to zero values to better visualization)

Full size image
Table 2. - Resistances and chemical capacitance of the cells obtained by fitting the Nyquist EIS plots
Full size table

By adding BTAH in 2.0 mg mL−1 concentration, it is noted a great minimization of charge transfer resistance in TiO2/dye, BTAH/electrolyte interface. This behavior demonstrates a more conductive and less resistive system which should facilitate the charge transport mechanism, being a great advantage when compared to similar additives applications, that showed an increase in RWE [8]. At higher values of added benzotriazole, RWE starts to rise up, proving that at high amounts (> 20 mg mL−1) it is obtained a solar system with a charge process transfer minimized. Nyquist diagrams in Fig. 3 also demonstrate the same reduction in Zi values, with BTAH addition, which corroborates the jsc and ɳ values obtained from j-V curves (Table 1).

For the chemical capacitance (Cμ), it is observed proportional decreased values with the amounts of BTAH added, as also verified in Cμ versus f curves of Fig. 2.SI. This behavior has been shown a reduce in the electrical double layer formed at working electrode interface.

With equal or higher amounts of BTAH (> 20.0 mg mL−1), it is supposed to obtain an increased resistance value, due to TiO2 pores filling that would be filled by the photosensitive molecule, decumulating the Cμ. It is clear that there is a concentration threshold that potentiates the solar cell photoelectrochemical parameters which must be well analyzed. A schematic overview about these discussions is shown in Fig. 4.

Fig. 4.
figure 4

Schematic overview of BTAH influence in TiO2, dye/electrolyte interface

Full size image

As shown in Fig. 4, a vacancy site, due to dye absence is an active site of recombination reaction, caused by electrolyte filling [30]. With BTAH insertion, there is a deaggregation of the dye, due to a competition for a TiO2 surface site. As benzotriazole molecules present a reduced size, it is able to fill the vacancy sites, avoiding by a steric hindrance and oxide adsorption effect, the contact of TiO2 and oxidized electrolyte. As demonstrated in some papers, BTAH has been able to easily adsorb in several materials [19, 20]. The charge recombination process should still occur, due to conjugated duple bonds, however with a slower speed.

As Reaction 1 and 2 occurs in TiO2/electrolyte interface, it is not feasible to study the processes separately by EIS diagram. To well analyze the back-electron recombination and charge collection the IMVS and IMPS diagram were performed on cells and are depicted in Fig. 5A-B.

Fig. 5.
figure 5

a–b. IMVS diagram in 5A and IMPS in 5B to solar cells with different BTAH amounts.

Full size image

As the IMVS operates in open circuit condition and IMPS in short circuit, it is possible to calculate the electron lifetime ( e) and collection lifetime ( c) using Eq. 2 [29, 31, 32]. The calculated results are shown in Table 3.

(2)
Table 3. - Extracted data from the IMPS and IMVS diagrams with their respective time calculations
Full size table

In Fig. 5A, it is observed a shift to lower frequencies of the maximum frequency point to 2.0 mg mL−1 of BTAH when compared to an unmodified device. This condition suggested a superior electron lifetime, which is a positive point to dye solar device and was confirmed by values in Table 2 [10, 32]. It is seen a e = 0.041 s and a e = 0.034 s to 2.0 mg mL−1 and 0 mg mL−1 of BTAH, respectively, showing a prevention of charge recombination between I3− with TiO2 [5]. The presence of the organic structure in BTAH difficult to access of I3− species and these molecules also can occupy the vacancy sites in TiO2 semiconductors, develo** a slow back-electron recombination. On the other hand, the 20 mg mL−1 BTAH addition has not been presented the best ɳ values to conditions tested, and despite superior e, its present low Cμ.

In the IMPS diagram, it is also noted a shift to higher frequencies with BTAH insertion, suggesting faster collection reactions. Using Eq. 2c was calculated and an improvement in collection time was observed when BTAH was added to the dye. This behavior, evidence that the used additive has been able to boost the collection in TiO2 semiconductor oxide [18]. Again, the steric effect is caused by benzene presence in a molecule, and also the N719 dye deaggregation, as the BTAH molecules might be positioning themselves between the dye molecules, rising up the charge separation and transportation in the semiconductor electrolyte interface. To 20 mg mL−1 of BTAH, despite amazing e and c presumably reduce the number of dye molecules anchored in TiO2 surface, supplying less electron under solar energy incidence, supported by jsc value in Table 1 [4, 11].

Then, it is important to emphasize that BTAH inserted in N719 dye solution is an amazing additive to be used in dye solar devices to enhance photoelectrochemical parameters.

Conclusions

BTAH molecules are effective additives in dye solar devices, able to enhance the photoelectrochemical parameters of the cell.

The back-electron recombination (e) can be restricted and the collection time (c) can be raised by BTAH presence. e and c enhancing are proportional to additive concentration, due to decreasing in recombination TiO2/electrolyte sites. On the other hand, higher BTAH amounts, difficult the dye anchoring, reducing the photoelectrochemical parameters.

The best solar cell was achieved at concentration of 2.0 mg mL−1 of BTAH, reaching a jsc = 9.92 ± 0.26, Voc = 0.770 ± 0.022, FF = 0.557 ± 0.014, ɳ = 4.26 ± 0.13, e = 0.034 s and c = 0.082 s with lower charge transfer resistance.