Use of BODIPY and BORANIL Dyes to Improve Solar Conversion in the Fabrication of Organic Photovoltaic Cells Through the Co-Sensitization Method

Abstract

This work addresses the implementation of the co-sensitization technique to increase the energy efficiency of organic dye-sensitized solar cells (DSSCs). Fluorescent dyes derived from boron complexes— (BORANIL) and (BODIPY)— were successfully synthesized and used as co-sensitizers in different volume percentage ratios to verify the most effective concentration for photon capture through these sensitizers. The dyes were optically characterized using ultraviolet–visible spectroscopy (UV-VIS) and Fourier transform infrared (FTIR), analyzing them through the optical performance of each hybrid combination of dyes, an optimization of the photon collection capacity in the tests performed in a volume percentage ratio of 25:75 or 1:3. The morphology and surface roughness of the electrodes were analyzed using scanning electron microscopy (SEM) and atomic force microscopy (AFM), respectively. Through electrochemical characterizations, it was found that the highest photovoltaic conversion efficiency was obtained with the ATH1005 (D) dye mixed with ATH032 (G) in the proportion of 25%:75% or DG 1:3, with efficiency (η) of 3.45%, against 2.43% and 1.90% for DG 1:1 and DG 3:1 cells, respectively. Cells with BODIPY dyes also present higher conversion efficiencies compared to BORANIL cells. The results corroborate the presentation of organic solar cells as a viable option for electricity generation.

1. Introduction

Currently, any type of technological development is directly or indirectly linked to the production of electrical energy. In addition, with population growth, the demand for electricity tends to increase over the years. To have a promising future, we must think about conscious development, producing energy without causing major environmental impacts. As a result of the facts mentioned above, we know that the demand for electrical energy will always be increasing, so we must concern ourselves with how we can generate energy in a sustainable way and satisfy this same demand [1,2,3,4]. One of the ways to convert solar energy into electrical energy is through photovoltaic cells. Specifically, this work deals with solar cells sensitized by dyes [5]. Dye-sensitized solar cells (DSSCs) were developed in 1991 by Michael Grätzel and O’Regan at the École Polytechnique Féderale de Lausanne [6].

The generation of electrical energy in these devices occurs through an electrochemical process involving two electrodes. This process is characterized by the interaction between semiconductor oxides, such as TiO₂, and photosensitizing dyes, in addition to the application of an electrolytic solution between the anode and cathode, which facilitates the transfer of electrical current from the oxidizing electrode to the platinum counter electrode. Initially, before exposure to solar radiation, the dye molecules are in their ground state, with the energy level of the HOMO (highest occupied molecular orbital). During this period, the TiO₂ semiconductor oxide located on the anode’s surface is also in a similar energy state, close to the valence band, which is non-conductive. When sunlight hits the DSSC (oxidizing region), the dye molecules are excited, causing a transition to energy levels of the LUMO (lowest unoccupied molecular orbital) [7]. As a result, the excited molecules overcome the difference in the semiconductor valence bandwidth. Photoexcitation of the dye, characterized by high absorption, is essential for the maximum use of the sunlight incident on the DSSCs.

The use of molecules containing nitrogen, oxygen, and conjugated π-bonds in the development of heterocyclic compounds has enabled the creation of several new families of fluorescent dyes. The distinct fluorescent coloration of these chromophores is attributed to the extent of the conjugated π-bonds system present in these molecules. In particular, tetracoordinated organic boron difluoride complexes are widely recognized for their fluorescent properties. The study of these organic boron difluoride complexes has aroused significant interest in the scientific community [8].

In recent years, BODIPY derivatives have gained increasing attention from researchers in the field of photoelectric materials due to their exceptional properties, such as delocalized molecular orbitals, flexible structure modification for suitable bandgaps and absorption profiles, metal-free organic sensitizers, a high molar absorption coefficient, and excellent photochemical and thermal stability [9,10,11,12]. As a result, there is a pressing need and challenge to explore new solar cells sensitized by BODIPY-based dyes [13,14,15,16,17]. Given their similar structural and photophysical properties, we also decided to investigate BORANIL dyes [18,19].

Electron-rich and electron-deficient substituents were introduced to create a donor–acceptor (D-A) configuration, which promotes intermolecular dipole–dipole interactions and enhances molecular ordering. This flexible tuning of the HOMO and LUMO energy levels facilitates stronger molecular interactions and improves charge transport through directed molecular self-assembly [20]. Furthermore, the introduction of iodine into BODIPY-based dyes significantly red-shifts the absorption spectrum and lowers the HOMO energy level [21].

The current research explores the co-sensitization technique to enhance the energy efficiency of organic DSSCs. It involves synthesizing fluorescent dyes from boron complexes (BORANIL and BODIPY) and testing various volume ratios to optimize photon capture. Characterization methods, including UV-VIS and FTIR spectroscopy, alongside SEM and AFM analysis, reveal that a 25%:75% ratio of ATH1005 (D) and ATH032 (G) yields the highest photovoltaic efficiency of 3.45%. The findings indicate that BODIPY dyes outperform BORANIL, highlighting the potential of organic solar cells for sustainable electricity generation.

2. Experimental Details

2.1. Reagents and Instruments

Transparent thin films coated with fluorine-doped tin oxide (FTO) were purchased from Sigma Aldrich and cut into 2.5 cm × 2.5 cm dimensions for use in the experiments. Other materials, such as semiconductor TiO₂ (size ≅ 20 nm), polyethylene glycol, potassium iodide, iodine, and hexachloroplatinic acid, were also obtained from Sigma Aldrich. To ensure the integrity of the anode–cathode junction, TEKBOND superglue was used. The dyes were evaluated using Agilent Cary 630 FTIR and SHIMADZU UV-2600i spectrometers. The electrodes were annealed using a QUIMIS Q318S25T muffle furnace. The morphological and structural analysis of the TiO₂ and Pt surfaces was conducted with a JEOL JSM-7100F field emission scanning electron microscope (FESEM) and a NanosurfEasyScan 2 atomic force microscope (AFM), respectively. The electrochemical properties, such as short-circuit voltage (J_SC) and open-circuit voltage (V_OC) of the solar cells, were determined by characterizations performed on the IVIUM Compactstat multipotentiostat, coupled to the Ivisun^® IVIUM Technologies (Eindhoven, The Netherlands) solar simulator, using an incident power of 100 mW/cm². The chemical reactions were monitored by thin-layer chromatography, using silica gel chromate plates (Merck) with a layer thickness of 0.2 mm, and flash chromatography was performed on Merck silica gel of 35–70 mesh. Melting points were measured using the PF1500 FARMA apparatus (Gehaka, Brazil), and the values were uncorrected. NMR spectra were recorded on a VARIAN 500 MHz spectrometer, using tetramethylsilane (TMS) as an internal standard for the 1 H and 13 C nuclei and trifluoroacetic acid (TFA) as an external standard for the 19 F nucleus. Chemical shifts (δ) were presented in parts per million (ppm) and coupling constants (J) in Hertz (Hz). The multiplicity of the signals was designated as follows: s, singlets; d, doublets; t, triplets; q, quartets; quintets; m, multiplets; dd, doublets; and dt, doublets. High-resolution mass spectra were recorded using a Bruker Daltonics micrOTOF-Q II (ESI+ mode) or Thermo Scientific Exactive Plus Orbitrap MS with atmospheric pressure chemical ionization (APCI) and electrospray ionization (ESI). For morphological and structural characterization, attenuated total reflection spectra were collected with a Bruker ALPHA FT-IR spectrometer, equipped with a single-reflection diamond ATR, covering the mid-infrared range (400–4000 cm⁻¹) with a resolution of 4 cm⁻¹, with the average spectrum obtained from 64 scans performed over 3 min.

2.2. Preparation of BORANIL and BODIPY

The synthesized boron complex dyes were obtained according to the literature [22,23,24,25] and shown in Figure 1 and Figure 2. The synthesized sensitizers were chosen considering the degree of excitation in the visible region of the electromagnetic spectrum, a region which favors the conversion of photons into electricity in DSSCs, in addition to the cataloging in the literature of the mixing of this class of dyes. Thus, the BORANIL dyes, more precisely ATH017, ATH019, and ATH024, were classified as A, B, and C, respectively. The BODIPY dyes, specifically ATH1005, ATH1006, ATH031, and ATH032, were also classified as D, E, F, and G, respectively. The structures of the compounds were confirmed by ¹H and ¹⁹F NMR spectroscopy and ATR–FTIR spectroscopy (see Supplementary File).

General procedure for the synthesis of BORANIL ATH017 (A), ATH019 (B), and ATH024 (C).

In a round-bottomed flask, 2-hydroxy-1-naphthaldehyde (361.6 mg, 2.1 mmol) was add to a solution of 50 mL of dichloromethane and the corresponding aryl amine (2.0 mmol). The resulting solution was then stirred at room temperature for 1 h, monitored by thin-layer chromatography. After a period of 1 h, triethylamine (TEA) (2.8 mL, 20.0 mmol) was added via a syringe. Subsequently, boron trifluoride diethyl etherate (BF₃OEt₂) (5.1 mL, 40.0 mmol) was added slowly, dropwise, via a syringe. The resulting solution was then stirred at room temperature for 18 h. The mixture was poured in 40 mL of water and then extracted with dichloromethane three times. The organic layer was dried over Na₂SO₄, and the solvent was evaporated under a vacuum. The resulting residue was purified by a chromatographic column using a dichloromethane gradient as the eluent.

General procedure for the synthesis of BODIPYs ATH1005 (D) and ATH031 (F).

2,4-dimethyl-1H-pyrrole (0.9990 g, 10.5 mmol) and the corresponding aryl aldehyde (5.0 mmol) were mixed with a pestle and mortar. Trifluoroacetic acid (TFA) (5 drops) was added via a pipette, while the mixture was ground with the pestle for about 2 min. To the resulting paste, CHCl₃ (2.0 mL) was added, followed immediately by the addition of p-chloranil (1.81 g, 7.4 mmol). The purple paste was ground for 2 min, after which triethylamine (TEA) (6.0 mL, 43.0 mmol) was added via a syringe. The resulting dark-brown paste was ground with the pestle for 3 min. Subsequently, boron trifluoride diethyl etherate (BF₃OEt₂) (6.0 mL, 47.4 mmol) was added slowly, dropwise, via a syringe, and the mixture was ground for 3 min until a thick dark-red paste was formed. The reaction mixture was dissolved in CHCl₃, transferred to a separation funnel, and washed with saturated Na₂CO₃, followed by brine. The solvent was evaporated under reduced pressure, and the crude solid was purified with a chromatographic column using an hexane–chloroform gradient as the eluent to give the desired product.

General procedure for the synthesis of the iodinated BODIPYs ATH1006 (E) and ATH032 (G).

In a round-bottomed flask, iodic acid (HIO₃) (0.276 mmol) and iodine (I₂) (0.276 mmol) were added to a solution of 10 mL of ethanol and the corresponding BODIPY (0.138 mmol). The resulting solution was then stirred at room temperature for 12 h. The mixture was poured in 40 mL of water and then extracted with chloroform three times. The organic layer was dried over Na₂SO₄, and the solvent was evaporated under a vacuum. The resulting residue was purified with a chromatographic column using an hexane–chloroform gradient as the eluent.

2.3. Preparation of the Cathode

To minimize errors arising from the internal resistance of the solar cell when controlling the potential of the working electrode, platinum (Pt) was used as the counter electrode. The choice of Pt is justified by its proven superiority over other materials, such as graphite, especially regarding catalytic efficiency and interaction with the oxidizing electrode, as demonstrated in comparative studies. Platinum stands out for its high electrical conductivity and excellent chemical stability, characteristics which are crucial for optimizing the performance of DSSCs [26]. The Pt counter electrode was prepared using 5 mM of hexachloroplatinic acid (H₂Cl₆Pt₆H₂O) solution. Initially, the ethanolic solution of H₂Cl₆Pt₆H₂O was subjected to an ultrasonic bath, a process which promotes the uniform dispersion of the particles and removes possible agglomerations; this was carried out in distilled water for a period of 30 min. The deposition method adopted was drop casting, where the solution was applied in thin and uniform layers on the substrates using a micropipette, ensuring precise control of the amount of deposited material. In the final step, the FTO substrates coated with the Pt layer were subjected to an annealing process in a muffle furnace at a temperature of 350 °C for 30 min. This step is essential to promote the adhesion of platinum to the substrate and improve the catalytic properties of the counter electrode, contributing to the overall efficiency of DSSCs. Careful selection of materials and strict control of the counter electrode’s preparation conditions are essential for the development of devices with a high photovoltaic performance.

2.4. Preparation of the Photoanode

The photoanode was prepared using the following compounds: polyethylene glycol (0.3 g), titanium dioxide (TiO₂) powder (1 g), distilled water (6 mL), and acetic acid (6 mL). The photoanode was prepared by sequentially depositing two layers of TiO₂ on the fluorine-doped tin oxide (FTO) thin film, similarly to the process used in the counter electrode. However, due to the higher density of the semiconductor solution, it was necessary to apply the spin coating technique to ensure a uniform and controlled deposition. This technique was regulated by an Arduino system, which maintained the rotation at 1000 RPM for 10 s, providing a homogeneous layer of TiO₂ on the substrate. The use of spin coating is essential for the formation of thin and uniform films, which are crucial for the efficiency of the photoanode in DSSCs. The uniformity of the TiO₂ layer directly influences the electron collection capacity and the minimization of structural defects, factors which significantly impact the photovoltaic efficiency of the device [27,28,29]. After deposition, the photoanodes were subjected to an annealing process in a muffle furnace at 450 °C for 30 min. Annealing is a critical step in promoting the crystallization of TiO₂, improving its electronic properties and adhesion to the FTO substrate. In addition, the furnace was programmed to allow gradual cooling to room temperature to avoid thermal stresses and ensure a stable crystalline semiconductor structure. The quality of the annealing directly affects the power conversion efficiency of the device, making this step fundamental in the DSSC fabrication process [30,31,32,33,34].

2.5. Preparation of Different Sensitizers and Co-Sensitizes

Initially, 10 milligrams of the prepared BORANIL (A, B, and C) and BODIPY (D, E, F, and G) dyes samples was individually dispersed in 100 mL of ethanolic solution. Based on the wide range of absorption, the co-sensitizers were prepared with BODIPY dyes (D and E with G dye) as follows: (i) ATH1005:ATH032 (DG) and (ii) ATH1006:ATH032 (EG) in different volume ratios, such as 1:3, 1:1, and 3:1. Finally, the dyes were stored in hermetically sealed containers in the refrigerator at a temperature of 5 °C and were later used as photosensitizers in the construction of DSSCs [35].

2.6. Construction of DSSCs

Figure 3 shows the assembly diagram of the DSSCs developed in this work. On each electrode (sensitized anode and cathode), a small external strip is cleaned on the coated surfaces, which will serve as a contact point for electrochemical characterization, and these strips are positioned equidistantly. For the sealing process, a small amount of instant glue (TEKBOND) is applied to the two faces adjacent to those that will be used for contact. Then, the electrolyte is introduced between the two faces, and the cathode and anode are joined. The redox electrolyte used to evaluate the solar cell performance was prepared with the following composition: 2.075 g of potassium iodide (KI), 0.12 g of iodine (I₂) dissolved in 5 mL of acetonitrile, and 0.02 g of polyethylene glycol [36,37].

3. Results and Discussion

3.1. FTIR Spectroscopy

In FTIR, the identification and correlation of vibrational bands with their respective wavenumbers is essential for the structural characterization of compounds such as BORANIL and BODIPY. These bands are directly associated with the vibrational modes of the chemical bonds present in the compounds and provide information about the presence of functional groups, molecular interactions, and structural changes after complexation with boron. Figure 4a shows out-of-plane bending vibrations of C-H in the aromatic BORANIL samples around 590 and 722 cm⁻¹ [18,38]; for the BODIPY sample, these vibrations occur around 700 and 900 cm⁻¹, as indicated in Figure 4b. These bands are associated with the deformation of the pyrrole ring and are also influenced by complexation with boron [39]. It is possible to observe B-O and B-N stretching vibrations around 675–1090 cm⁻¹ for the BORANIL samples. The presence of these bands in the spectrum confirms the formation of the boron complex and the coordination of the ligand nitrogen with the boron atom, while, for the BODIPY samples, B-F vibrational bonds are observed around 1100 cm⁻¹. It is also possible to identify C=N stretching vibrations in the region comprising 1600 to 1750 cm⁻¹ in both samples, an indication of the strong π-conjugated interactions in the system, which are amplified by the presence of boron [40,41]. The shift in the vibrational bands in the FTIR spectra of BORANIL and BODIPY compounds after complexation with boron is one of the main indicators of electronic and structural changes. In BORANIL, the vibrations associated with the B-O and B-N bonds, in addition to the deformations of the aromatic ring, confirm the formation of the BF₂ ring and its stabilization [41]. In BODIPY, the B-F vibrational bands are characteristic and point to the efficient interaction of boron with the π-conjugated system [38,39,40,41].

3.2. UV-Vis Spectrophotometry

The compounds derived from BORANIL ATH017 (A), ATH019 (B), and ATH024 (C) and the compounds derived from BODIPY ATH1005 (D), ATH1006 (E), ATH031 (F), and ATH032 (G) were subjected to an analysis of the ultraviolet absorption spectra, providing essential information about the absorption transition between the ground state of the dye molecule, the excited state, and the range of solar energy absorbed by the molecule of the active layer. In Figure 5a, the BORANIL dyes exhibit absorption peaks in the range of 315~363 nm [18,38,42], which comprises the near-ultraviolet region; in addition to these, there are peaks in the visible ultraviolet region of 449~460, but with a low intensity. The BODIPY dyes exhibit absorption peaks in the ranges of 365~388 nm and 499~535 nm (Figure 5b), covering a wavelength range with greater intensity than BORANIL dyes in the ultraviolet–visible region. The absorption spectra peaks of the dyes align with the HOMO-LUMO energy difference characteristic of the class of boron-derived dyes [43,44,45].

It can be concluded that the BODIPY dyes behave better from the point of view of absorbance, since these dyes, in addition to being excited in a broader region of the spectrum, have greater vibrational intensity compared to BORANIL dyes. Therefore, to apply the co-sensitization technique to the synthesized dyes, the hypothesis of using BORANILs was discarded in favor of comparing the data obtained with the BODIPYs. Analyzing the results obtained with the characterization of the absorbance spectra of the BODIPY dyes, ATH031 (F) was discarded for the method, considering the vibrational properties presented by the dyes, so that the dyes whose co-sensitization window covered the largest differential level among the observed wavelengths were selected [46,47,48,49,50]. Consequently, this greater difference is seen between dyes ATH1005 (D) and ATH032 (G), respectively. It can be observed that, in ATH032 (G), the first peak obtained in the near-ultraviolet region is presented at 224 nm, while, in ATH1005 (D), the peak is at 260 nm. Furthermore, in the visible region of the spectrum, in ATH1005 (D), there is a peak at the lowest wavelength value among the samples, located at 499 nm, while, in ATH032 (G), the highest value is recorded, at 535 nm; therefore, it is expected that the co-sensitization between these will have the most satisfactory result. ATH1006 (E) was also included in the co-sensitization method for the comparison and validation of the results.

In Figure 6a, the absorption spectra of three mixed dyes, labeled DG 1:1, DG 1:3, and DG 3:1, are shown. The main purpose of combining these dyes is to broaden the absorption spectrum of solar radiation. Often, mixtures of several dyes with varying absorption spectra are prepared to achieve maximum absorption in the near-visible, visible, and infrared regions. In the context of DSSC applications, these mixtures co-sensitize the device, improving the overall absorption by utilizing the broadest possible wavelength range and, thus, maximizing efficiency [51,52]. The DG sample in a 1:3 volume ratio exhibits an enhanced absorption spectrum. Peaks are observed at 259 nm, 380 nm, 499 nm, and 536 nm, with a particularly high absorbance in the visible wavelength range. When the volume ratio is adjusted to 1:1, the absorption peak remains within the same wavelength range but with reduced absorbance. On the other hand, the sample at a 3:1 ratio exhibits an even lower absorption peak within the same wavelength range. Notably, the 1:3 volume ratio represents the optimized configuration for the DG sample, exhibiting the highest absorption spectrum. It is noteworthy that all DG samples share the same absorption peaks but with varying intensities.

Figure 6b shows the hybrid spectra between samples E and G, labeled EG, performed at the same ratio and volumetric concentrations as the previous assay. Absorption peaks are evident in all EG sample mixtures. In the 1:3 volume mixture, absorption peaks are observed at 312 nm, 371 nm, and 511 nm. The 1:1 mixture exhibits an absorbance peak at a smaller wavelength range, characterized by less absorption. On the other hand, the mixture with a volume ratio of 3:1 presents an increased absorption peak in relation to the 1:1 analysis, within the same wavelength range. Given the results obtained, it can be concluded that the behavior of the co-sensitizers was as expected, considering that the highest degree of absorbance was achieved with a mixture between dyes D and G.

3.3. FESEM Morphological Characterization

The field emission scanning electron microscope (FESEM) was used to evaluate the morphological properties of the electrodes (anode and cathode). This analysis was performed in partnership with the Multiuser Laboratory of Nanofabrication and Characterization of Nanomaterials (NANOFAB) of the State University of Rio de Janeiro (UERJ). The morphological characterization allowed us to verify the uniformity and porosity of the electrodes, which were subjected to deposition techniques, such as “spin coating” for the anode coating [53] and “drop casting” for the cathode coating [54]. In addition, it is noteworthy that this nanostructural characterization can provide detailed information on the granulometry, the thickness of the samples, and the possible presence of defects or discontinuities on the surfaces [55,56,57]. This information is crucial for optimizing manufacturing parameters and understanding the electrochemical properties of devices, directly influencing their performance in technological and industrial applications.

Figure 7a shows the desirable mesoporosity of the TiO₂ semiconductor nanostructures deposited on the FTO thin film, analyzed at a magnification of 50,000× and with an emission of 15 kV. In this image, it is possible to identify the grain size and uniformity of the film, which vary between 20 nm and 25 nm, which demonstrates the correct execution of the spin coating deposition technique. This uniformity is essential, as it favors the electrode-electrolyte interaction, optimizing adsorption on the surface of the anode that will later be photoexcited [58]. Mesoporosity also contributes to a larger specific surface area, promoting the efficiency of the charge transfer processes in the device. Figure 7b shows an image of the Pt cathode, also evaluated under the same magnification and emission voltage conditions. The grain size varies between 20 nm and 30 nm, also showing homogeneity in the platinum thin film. This homogeneity is crucial to improve light interactions and reflections in the DSSC device, which can result in increased photocatalytic activity [59]. Platinum, acting as an intermediary between the electron donor and acceptor electrodes, plays a key role in the kinetics of the solar cell [60]. Furthermore, the good performance of the drop casting deposition technique was also confirmed, reinforcing the coating quality and the overall efficiency of the device. The correct execution of these techniques is crucial for the electrochemical performance of DSSCs, directly influencing their energy efficiency and stability.

3.4. AFM Structural Characterization

This study incorporated the root mean square (RMS) roughness analysis for two layers, an approach supported by the literature due to its potential for improved photocatalytic performance [28]. Detailed evaluation of the electrode surface is essential for the characterization of TiO₂ thin-film-based substrates, since crucial parameters for functional performance, such as the effectiveness of the semiconductor film and counter electrodes, are directly influenced by surface roughness. Figure 8a shows the topographic image of a thin film composed of TiO₂ on a micrometric scale of 25 µm, which demonstrates a satisfactory result, since its surface shows uniformity and, simultaneously, mesoporosity, contributing to good dye absorption and good electron flow in the semiconductor. The platinum cathodes were also subjected to evaluation with AFM, as shown in Figure 8b. The platinum nanoparticles were uniformly distributed on the surface of the FTO electrode, forming a densely distributed film. The topographic image of the platinum thin film appears more uniform when compared to the TiO₂ thin film, being suitable for the flow of diffusion electrons [61].

In addition, the topographic surface roughness of the photoanodes deposited on the individual BORANIL and BODIPY dyes was examined, as shown in Figure 9. The BODIPY dyes showed the best topographic distribution out of the films produced. When we analyzed the electrodes fabricated with the BORANIL dyes (Figure 9a), we could see that they had a flatter distribution compared to the electrodes fabricated with the BODIPY dyes (Figure 9b). These presented the formation of sharper peaks and valleys, but in a uniform manner, which favored the absorption of the active layer by the semiconductor electrode.

3.5. Electrochemical Characterization

For the electrochemical characterization of the dye-sensitized solar cells (DSSCs), a multipotentiostat (IVIUM Compactstat) was employed in conjunction with a solar simulator (Ivisun^® IVIUM Technologies). The solar simulator was calibrated to deliver an incident power of 100 mW/cm², maintaining 15 cm between the photovoltaic cell and the light source. These conditions ensured accurate measurement of the electrical parameters: short-circuit current density (J_SC) and open-circuit voltage (V_OC).

From these electrical parameters, the fill factor (FF), maximum power (P_max), and solar conversion efficiency (η) were calculated. The J_SC represents the maximum current produced by the DSSC under zero applied voltage (V = 0). It is influenced by factors such as the number and energy of incident photons, the efficiency of light absorption by the dye, and the charge transport properties of the semiconductor and electrolyte [62]. The V_OC is the maximum potential generated by the DSSC when no external current flows (J = 0). It is determined by the difference in potential between the conduction band of the semiconductor and the redox potential of the electrolyte [63].

The fill factor (FF) quantifies the deviation of a real DSSC from an ideal one, with a value ranging from 0 to 1, where J_max and V_max are the current and voltage at the maximum power point, respectively. A higher FF indicates a more efficient DSSC in terms of utilizing the generated current and voltage. It is calculated using Equation (1):

$$FF={\displaystyle \frac{I_{MAX}·V_{MAX}}{J_{SC}·V_{OC}}}$$

(1)

After obtaining the calculated photovoltaic parameters, it is possible to verify the solar conversion efficiency (η) through Equation (2) [51], considering that the experimental area of the DSSCs is 6.25 cm²:

$$\eta (\%)={\displaystyle \frac{{\mathrm{J}}_{\mathrm{SC}}·V_{OC}·FF}{P_{in}}}.100$$

(2)

It can be observed that all cells made under the co-sensitization method (Figure 10c,d) reached higher levels of current density compared to individual cells (Figure 10a,b). In particular, the DG 1:3 cell presented a higher short-circuit current density with 2.413 mA/cm², while the EG cell in the same proportion presented a J_SC of 1.607 mA/cm². The same occurred with the open-circuit voltage, where the respective results were 0.59 V and 0.55 V. To evaluate the maximum output power, Equation (3) [46] was used:

$$P_{MAX}=I_{MAX}·V_{MAX}$$

(3)

The maximum power provided by the DSSC, shown in Figure 11, sensitized by the DG 1:3 hybrid method with 0.55195 mW/cm², has an utilization that is almost twice that of its combinations individually (D and G), with these respective values being 0.3618 mW/cm² and 0.36036 mW/cm².

To evaluate the intrinsic resistance of each photosensitizer and understand the impact of the dye mixture on the performance of the solar cells, electrochemical impedance analyses were performed. The results, presented in Nyquist plots (Figure 12c,d), reveal that the solar cells with mixed photosensitizers exhibit a significantly smaller resistive semicircle compared to the cells made with individual dyes (Figure 12a,b). For a more detailed analysis, the equivalent electrical circuit of the experimental and target photosensitizers is presented in Figure 12e. The values of the electronic components of this circuit, which represent physical elements of the solar cell (such as load resistances, Helmholtz layer capacitances, etc.), are summarized in

Table 1. Values of each element of the equivalent circuit with their associated errors.

Dyes	RS (Ω)	R1 (Ω)	R2 (Ω)	R3 (Ω)	C1 (μF)	C2 (μF)	C3 (μF)
BORANIL ATH017 (A)	36.39 ± 1.22	119.0 ± 2.44	46.88 ± 6.57	38.94 ± 3.02	6.32 ± 6.14	2.65 ± 3.19	7.14 ± 2.17
BORANIL ATH019 (B)	38.87 ± 1.14	89.91 ± 3.75	48.84 ± 6.79	40.07 ± 4.71	5.98 ± 8.03	2.14 ± 3.03	6.55 ± 3.81
BORANIL ATH024 (C)	62.70 ± 0.92	86.28 ± 4.94	66.25 ± 6.33	49.52 ± 3.72	7.44 ± 10.39	1.90 ± 2.91	6.49 ± 3.02
BODIPY ATH1005 (D)	32.02 ± 0.30	27.23 ± 14.6	34.31 ± 11.6	34.14 ± 2.78	82.49 ± 23.71	3.62 ± 1.01	6.94 ± 2.88
BODIPY ATH1006 (E)	100.1 ± 0.14	38.63 ± 8.01	40.98 ± 7.58	35.93 ± 4.80	43.33 ± 13.3	13.35 ± 3.49	7.13 ± 4.55
BODIPY ATH031 (F)	46.61 ± 0.31	35.48 ± 7.57	48.56 ± 4.62	8.62 ± 8.98	39.54 ± 12.7	8.38 ± 2.90	8.19 ± 4.40
BODIPY ATH032 (G)	31.77 ± 0.30	33.46 ± 10.9	35.70 ± 10.3	40.29 ± 8.16	43.64 ± 17.9	14.52 ± 5.35	7.78 ± 4.42
DG 1:1	35.23 ± 1.75	22.93 ± 5.77	60.28 ± 3.13	47.36 ± 3.75	54.96 ± 13.6	6.27 ± 7.57	1.10 ± 3.18
DG 1:3	62.05 ± 0.13	18.27 ± 1.73	23.97 ± 1.85	42.34 ± 2.26	97.70 ± 3.63	8.21 ± 5.15	2.54 ± 3.23
DG 3:1	98.93 ± 0.16	39.43 ± 9.48	33.68 ± 11.1	45.91 ± 3.77	33.08 ± 15.4	13.33 ± 5.30	3.93 ± 4.01
EG 1:1	34.75 ± 1.69	22.06 ± 5.82	59.95 ± 3.10	846.17 ± 3.82	52.61 ± 13.9	6.27 ± 7.49	1.14 ± 3.10
EG 1:3	32.23 ± 0.35	26.56 ± 14.3	42.40 ± 9.05	41.02 ± 2.95	10.17 ± 24.4	20.65 ± 4.39	4.14 ± 2.27
EG 3:1	91.55 ± 0.15	18.50 ± 13.3	44.46 ± 5.55	39.22 ± 3.43	61.80 ± 22.3	8.18 ± 2.48	7.23 ± 5.73

. The comparison of these values allows for a deeper interpretation of the impedance results, providing insights into the charge transfer processes and the limitations of each photosensitizer.

To deepen our understanding of the charge transport processes in solar cells, Nyquist diagrams were modeled using an electrical equivalent circuit. This circuit, commonly used in the literature [58], considers a resistive component (R₁) associated with the charge transfer resistance at the counter electrode/electrolyte interface. The mid-frequency region in the Nyquist diagrams is attributed to electron transport at the semiconductor/dye/electrolyte interface and is represented by resistance R₂ in the equivalent circuit. Resistance R₃, in turn, is related to the internal diffusion process of the electrolyte, while the serial ohmic resistance (R_S) is attributed to the intrinsic resistance of the electrolyte [63]. Figure 12 represents a comparative analysis of the electrical parameters obtained from the photovoltaic tests and the impedance modeling for each photosensitizer. This analysis allows one to establish correlations between the resistance values (R₁, R₂, R₃, and R_S) and the overall performance of the solar cells, providing valuable insights into the limiting mechanisms of each device.

Table 2. Photovoltaic performance of DSSCs with various natural dyes.

Dye	JSC (mA/cm2)	VOC (V)	FF	η (%)	Ref.
BODIPY	1.42	0.43	0.55	0.34	[64]
BODIPY 1	0.73	0.26	0.55	0.11	[8]
BODIPY 2	3.10	0.37	0.54	0.63	[8]
BODIPY 3	1.66	0.49	0.52	0.42	[50]
BODIPY 9	1.32	0.18	0.51	0.49	[65]
BODIPY 1/KI/I2	3.40	0.67	0.51	1.70	[47]
BODIPY 2/KI/I2	4.00	0.69	0.52	2.10	[47]
BODIPY 3/KI/I2	4.70	0.72	0.52	2.50	[47]
H-T-BO:PC71BM 1:2	6.80	0.67	34.3	1.56	[52]
Br-T-BO:PC71BM 1:2.5	7.62	0.72	35.7	1.96	[52]
H-T-BO:PC71BM 1:2.5	11.3	0.74	35.5	3.13	[52]
ATH017 (A)	0.99	0.49	0.48	1.46	[this work]
ATH019 (B)	0.96	0.60	0.39	1.40	[this work]
ATH024 (C)	0.95	0.60	0.44	1.57	[this work]
ATH1005 (D)	1.57	0.59	0.39	2.26	[this work]
ATH1006 (E)	1.46	0.60	0.38	2.09	[this work]
ATH031 (F)	1.25	0.60	0.45	2.10	[this work]
ATH032 (G)	1.26	0.60	0.48	2.25	[this work]
DG 1:1	1.44	0.60	0.45	2.43	[this work]
DG 1:3	2.41	0.59	0.39	3.45	[this work]
DG 3:1	1.26	0.52	0.46	1.90	[this work]
EG 1:1	1.41	0.60	0.45	2.41	[this work]
EG 1:3	1.60	0.55	0.49	2.71	[this work]
EG 3:1	1.24	0.60	0.45	2.08	[this work]

presents the solar conversion efficiency obtained for the two sensitizers used in this work compared with previous studies.

4. Conclusions

In this work, fluorescent dyes derived from BORANIL and BODIPY were used as sensitizers for DSSCs. The structural analyses revealed a homogenous surface, attesting to the effectiveness of the material deposition process in electrodes by AFM and SEM, respectively. This finding reinforces the robustness and quality of the substrates used in this study. The results indicate that the increase in the absorption spectrum has a direct impact on improving the adsorption of the active layer, favoring the generation of photoexcited electrons. The presence of boron and fluorine molecules in the extracts was optically investigated and confirmed by FTIR and UV-Vis. The UV results showed that the absorption was higher for DG 1:3, compared to the other samples. It was also observed that this mixture presented a greater absorption capacity, essential for good photoexcitation and interaction with the semiconductor oxide, corroborating the satisfactory electrical parameters of solar conversion efficiency (η) of 3.45% and P_max of 0.55195 mW/cm². It can thus be concluded that the dyes used in this work for the development of DSSCs can be considered good sensitizers, creating promising perspectives in the field of emerging solar cells. Thus, solar cells sensitized by organic dyes, or DSSCs, emerge as a prospect for producing electrical energy in the coming years.