New natural dyes extracted by ultrasonic and soxhlet method: Effect on dye-sensitized solar cell photovoltaic performance

The increase in the need for energy and the rapid depletion of fossil fuels have directed many researchers to alternative and renewable energy sources (Kumar et al. 2022; Alinejad et al. 2023). Photovoltaic technology is particularly appealing among renewable energy sources because it turns sunlight directly into electrical energy (Rajkumar et al. 2019). The high cost and non-environmental nature of silicon-based solar cells have made it inevitable to develop low-cost devices (Rajkumar et al. 2022). Until now, many remarkable studies have been carried out on the production of low-cost and high-efficiency solar cells (Nnorom et al. 2022; Zatirostami 2021). Due to their low cost and ease of production, dye-sensitized solar cells (DSSCs) are very important in photovoltaic technology (Nguyen et al. 2020; Ahmad et al. 2023; Senthamarai et al. 2022). To catalyze the reaction, a typical DSSC includes a dye molecule, counter electrode, redox electrolyte, and photoanode. Dye molecules easily absorb artificial light or sunlight and excite electrons by collecting the emitted photon flux on the surface of the oxide layer. Excited electrons are injected into the metal oxide semiconductor layer's conduction band. These electrons are transferred to the current collector via the metal oxide thin film, and the dye molecule is renewed in the redox solution by an electron donor. Finally, a complete regeneration takes place by transferring electrons to the electron acceptor on the counter electrode of the DSSC (Peiris et al. 2021; Prakash et al. 2022). The working principle of a typical DSSC is given in Fig. 1. The first study on DSSCs was conducted by Gratzel et al. They have been successful in creating DSSCs with photoanodes based on titanium dioxide (TiO₂) that have a power conversion efficiency of more than 7% (Richhariya et al. 2022). In subsequent studies, efficiency values above 15% were achieved (Teja et al. 2023). These findings hold promise for DSSC commercialization in the future. However, more thorough research is now required because DSSC efficiency is still below the target level. Researchers have attempted to improve the dye, counter electrode, electrolyte, and photoanode of DSSCs to boost their efficiency (Sen et al. 2023).

Dye sensitizers have a great impact on the efficiency of DSSC as they greatly affects the light harvesting (Rahman et al. 2023). A typical DSSC's performance depends on the sensitizing dye's molecular structure, which separates the electron–hole pairs when excited by light. The task of the dye in DSSCs is to provide charge transfer to the semiconductor surface by converting the light into electronic excitation. The properties of a natural dye that can be used as a sensitizer in DSSCs are given below (Sánchez-García et al. 2018).

Many studies have been conducted on metal-based dyes that exhibit high performance on DSSCs. Metal complexes and different synthetic dye derivatives showed the highest power conversion efficiencies (Zatirostami 2020). Especially in DSSCs produced using N719 sensitizer, efficiency values over 14% were achieved. However, due to the high cost and toxic effects of such metal-based dyes, more environmentally friendly and natural dyes have been considered as alternatives. Natural plant dyes are non-toxic and can be easily synthesized at a low cost (Singh et al. 2021). However, the efficiency of DSSCs sensitized with natural dyes has not yet reached the desired level (Prakash and Janarthanan 2023). As a result, numerous research projects were carried out to raise the stability and effectiveness of DSSCs sensitized with natural dyes, using various tactics, including mixing various natural dyes (Obi et al. 2020), purification via column chromatography (Wang et al. 2006), and different extraction techniques (Calogero et al. 2015). In a study by Kristoffersen et al., a significant increase in efficiency was detected after the removal of non-pigment organic compounds from DSSCs using anthocyanin as a sensitizing dye (P. N., B. R., A.S. Kristoffersen, B. G., P. S., S. K., K. M.D., S.R. Erga, D. Velauthapillai 2021). Furthermore, an enhancement in the performance of solar cells sensitized with the betalain-anthocyanin pair was noted in the research carried out by Ramomoorthy et al. (Ramamoorthy et al. 2016). In another study by Calogero et al., the performance of DSSCs was improved by up to 1.6% by using different extraction methods to obtain sensitizing dye. (Calogero et al. 2018). Using various solvents and extraction techniques, Inbarajan et al. reported achieving a power conversion efficiency of 1.17% in another study (Inbarajan et al. 2022).

Among the dyes leached by the extraction of natural plants, the most commonly used ones are carotenoids, anthocyanins, betalains, flavonoids, and chlorophylls (Safaei-Ghomi and R.* Masoomi, M. Hosseinpour, H. Batooli 2020). The main classification of plant pigments is given in Fig. 2. Plant pigmentation and chemical composition are key factors in the dye classification process. The dye's molecular makeup makes it more apt to stick to the TiO₂ electrode and makes it easier for electrons to flow toward the TiO₂ conduction band. Additionally, the dye structure helps slow charge recombination at the interface. Several chemical functional groups bind the sensitizing dye to the semiconductor metal oxide surface. Among these functional groups, there are carbonyl or/and hydroxyl groups. Dyes containing carbonyl or/and hydroxyl groups not only establish a stronger electronic connection with the photoanode but also provide a better interaction with the electrolyte solution (Tabacchi et al. 2019). Therefore, some physical and chemical properties of sensitizers used in DSSCs can significantly affect the performance of DSSCs (Kabir et al. 2022a).

Organic-based dyes extracted from natural plants are used extensively in many areas such as the textile industry, anti-viral drug research, and pharmaceuticals (Sen et al. 2023). Among these plants, Rubia fruticosa fruits, Rhamnus tinctoria seeds, and Pinus pinea bark are also used as dyestuffs. Pinus pinea bark, Rhamnus tinctoria seeds and Rubia fruticosa fruits used in this study are flavonoids (Sönmezolu et al. 2012; Gómez-Ariza et al. 2006; Espinosa et al. 2012). Rhamnus tinctoria is a thorny shrub plant that is up to 3 m in length and grows endemic in Central Anatolia. The seeds of this plant remain green for a long time before, they turn black and brownish. Dyes produced from the Rhamnus tinctoria plant were used as textile dyes from ancient times to the nineteenth century (Akin et al. 2016). Another plant species, Rubia fruticosa, is an endemic species that grows in the central archipelago of Macaronesia. This plant species is known for its fleshy fruits and can reach half a meter in length (Vieito et al. 2018). In addition, the Pinus pinea plant accounts for about 7% of the biomass above ground. It is an evergreen coniferous tree that can grow up to 30 m in height in the form of an umbrella, which is abundant in the Mediterranean basin. A brown-reddish dye is leached by extraction of the bark of this plant (Vieito et al. 2018; Agrícolas et al. xxxx). Plants contain many dyestuff compounds. The ratio of these compounds varies according to the type of plant, the region where it grows, its age, and the harvest season. The most common dyestuffs in the structure of Rhamnus tinctoria seeds, Rubia fruticosa fruits, and Pinus pinea bark are emodin, alizarin, and quercetin, respectively. The main classification of plant pigments used in this study can be seen in Fig. 2.

To extract maximum dye from a plant, an effective extraction is required. The factors such as economy, efficiency and ease of access play important roles in choosing the extraction method. It is also important that the chemical used in extraction has a minimum impact on the environment (Teja et al. 2023). It is known that dyes obtained by different extraction methods change the photovoltaic properties of DSSCs (Kocak et al. 2019). The most commonly used method to study the effect of the extraction method is trial and error. It is assumed that there is a strong relationship between the extraction method and the absorbance of the dye in the visible region (Mphande 2014).

Sol–gel method, Precipitation, Spray pyrolysis, Template method and Hydrothermal method are generally used in TiO₂ synthesis. Among these methods, the hydrothermal method creates an effective reaction environment for metal oxide production due to controllable particle size, high nanocrystallinity and high purity (Yeoh et al. 2023).

TiO₂ is the most widely used semiconductor metal oxide in DSSCs due to its features such as being environmentally friendly, abundant and providing the necessary surface area for good dye loading (Calogero et al. 2010; Senthamarai et al. 2021). The nanoporous TiO₂ layer with a large surface area coated on the conductive glass surface allows high absorption of the sensitizing dye (Weldekidan et al. 2018). The most important factor that encourages the use of TiO₂ in DSSCs is its ability to absorb a large portion of incoming light. TiO₂ used in DSSCs is generally in the rutile or anatase phase. Anatase TiO₂ is used more due to its wide band gap and chemical stability (Mozaffari et al. 2015).

In this study, it was investigated to what extent dyes leached from different plant species by soxhlet and ultrasonic methods affect the performance of DSSCs. In the literature review, no studies were found on the use of dyes extracted from Pinus pinea bark, Rhamnus tinctoria seeds and Rubia fruticosa fruits in DSSCs. Experimental studies have shown that natural dyes obtained by different extraction methods significantly affect the photovoltaic properties of DSSCs.

The XRD pattern of TiO₂ nanoparticles produced by the hydrothermal process is shown in Fig. 5. Reflection peaks in the XRD pattern validate the anatase phase and excellent crystalline structure of TiO₂. Furthermore, the anatase phase and 141:I41/amd unit cell symmetry of TiO₂ nanoparticles match card number DB 5000223. The Williamson-Hall formula given in Eq. 1 was used to determine the average crystal sizes of the produced TiO₂ nanoparticles (Sethy et al. 2023).

The symbols "λ", "ɛ", "D", and "θ" in this formula indicate the peak's reflection angle, crystal size, lattice strain, and X-ray wavelength, respectively. FWHM, or the full width at half the peak height, is denoted by the symbol βs. The large crystal sizes improve the photovoltaic conversion efficiency of cells. Because the increase in crystal size reduces the grain boundaries in the structure and allows electrons to move more easily within the crystal structure (Moazami et al. 2015). The average particle size of the synthesized TiO₂ nanopowders was calculated as 13.54 nm. This average particle size is consistent with the average particle sizes of TiO₂ used in high-performance DSSC applications (Akman et al. 2020; Huang et al. 2022; Richhariya et al. 2023).

Surface photographs of the synthesized TiO₂ were examined with FE-SEM. The SEM photograph given in Fig. 6 reflected close-contact spherical particles with an average diameter of 100 mn. Such close contact facilitates electron transport at the interface and allows greater dye absorption at the surface (Ge et al. 2021).

Figure 7 shows the EDS graphs of TiO₂ particles produced by the hydrothermal method. The presenceof titanium and oxygen in the structure is clearly understood from the EDS peaks.

When it comes to redox cycles and dye loading of DSSCs, the surface area and porosity of TiO₂ nanopowders are crucial. We examined the pore size and specific surface area of the produced TiO₂ nanoparticles using the BET technique. The isotherm curve of N₂ gas based on the principle of adsorption and desorption was shown in Fig. 8. The generated TiO₂ nanoparticles demonstrated the presence of mesoporous structure and represented isotherm type IV by the Brunauere Deminge DemingeTeller (BDDT) classification (Xu et al. 2015). The produced TiO₂ particles were found to have a specific surface area of 6.36 m²/g. The porosity of the TiO₂-based photoanode was determined by the BJH (Barret-Joyner-Halenda) method. The TiO₂ thin film's mean pore radius and pore volume were determined to be 61.95 nm and 0.189 cm³/g, respectively. Large surface area and pore size photoanodes were known to enhance electrolyte transport pathways and dye loading in DSSCs (Wali et al. 2016). BET studies showed that the obtained parameters are suitable for high-performance DSSCs (Sayahi et al. 2021; Anitha et al. 2015).

The absorption spectra of dyes extracted from Pinus pinea bark, Rhamnus tinctoria seeds, and Rubia fruticosa fruits by soxhlet and ultrasonic method were shown in Fig. 9. Dyes extracted from natural plants exhibited different absorption spectra. Variations and discrepancies in sensitizing dyes' absorption behavior were indicative of various pigment structures. Absorption spectra ranging from about 300 to 550 nm were detected in the dyes quercetin, emodin, and alizarin extracted from natural plants. Absorption bands in the near-UV–vis region (300–350 nm) reflect the π−π* transition, while bands in the visible blue region (about 450 nm) correspond to intramolecular charge transfer between donor–acceptor (Sen et al. 2023). Extraction methods led to differences in the maximum absorption wavelengths of the dyes. The maximum absorption wavelengths of the dyes extracted from the bark of the Pinus pinea plant by soxhlet and ultrasonic methods were observed at 397 nm and 361 nm, respectively. The peaks in the UV region of the dyes leached by two different methods indicate the presence of quercetin in the dyes. It can be said that the absorption of quercetin dye extracted by the Soxhlet method shifts towards the visible regiosn. The dyes extracted from the fruits of the Rubia fruticosa plant using soxhlet and ultrasonic methods had maximum absorption wavelengths of 445 and 476 mn, respectively. These observed peaks indicate the presence of alizarin pigment in the dyes. Additionally, Alizarin dye extracted by the Soxhlet method exhibited absorption in a wider wavelength range in the visible region, from 435 to 571 nm. The soxhlet method yielded two strong absorption peaks at 310 and 500 nm, while the ultrasonic method yielded peaks at 290 and 440 nm for the dye extracted from Rahamnus tinctoria. These peaks observed in the UV and visible regions confirm the presence of emodin pigment in the dye. In addition, emodin dye extracted by the soxhlet method exhibited absorption in a wider wavelength range (between 420–575 nm) in the visible region compared to other dyes. The presence of the pigments quercetin, alizarin and emodin in natural dyes was confirmed by the references to peaks made in previously reported studies (Prakash and Janarthanan 2023; Inbarajan et al. 2022; Kabir et al. 2022a). Here it seems that soxhlet is the best method to extract quercetin, alizarin and emodin pigments, because all dyes extracted by the soxhlet method showed stronsger absorption behavior. When six dyes made from real plants using two different processes were compared, the dye emodin that was leached from Rhamnus tinctoria seeds absorbed the biggest amount of light. This higher absorbance led to higher efficiency of emodin dye-sensitized DSSC leached by the soxhlet method. It is predicted that intense absorptions seen in the visible region improve cell performance (Safie et al. 2017). These absorptions in different energy regions helped the DSSC to capture photons and hence improve cell performance (Das et al. 2020). UV–vis absorptions of quercetin, emodin, and alizarin dyes extracted from natural plants were measured in this study, indicating that these dyes can be used as photoanode sensitizers.

Infrared (IR) spectroscopy application is used especially in the identification of organic molecules by considering certain functional groups (Thangavel et al. 2022). This spectroscopy helps to detect functional groups in dyes that play an active role in DSSCs. The Fourier transform infrared spectra (FTIR) of quercetin, emodin, and alizarin dyes extracted by different methods from Pinus pinea bark, Rhamnus tinctoria seeds, and Rubia fruticosa fruits, respectively, in the frequency range of 4000–400 cm⁻¹ are given in Fig. 10. Each spectrum displays the distinctive absorption bands that correlate with the chemical makeup of the dyes extracted using the soxhlet and ultrasonic methods, as shown in Fig. 10. The dyes leached by different extraction methods showed similar spectral properties among themselves, and the pigments (quercetin, emodin, and alizarin) in the structure confirmed the existence of their functional groups (Safaei-Ghomi et al. 2020; Akin et al. 2016; Xavier et al. 2021). The chemical structures of the extracted sensitizing dyes are given in Fig. 11.

Dye molecules can stick to the surface of the metal oxide layer through various binding mechanisms such as hydrogen bonding, covalent bonding, and Van-der Vaals forces (Zhang and Cole 2015). However, a strong interaction with TiO₂ surface atoms is only possible through covalent bonding. Because other interactions are weak and can make the device unstable. The strong interaction of the carbonyl and hydroxyl groups in natural dye molecules with the hydroxyl group on the surface of the porous photoanode (TiO₂) provides a better electron transfer to the conduction band of TiO₂. In other words, these groups on the metal oxide surface increase the electron injection rate in cells (Subramanian and Wang 2012).

The O–H (hydroxyl) group is present in the extracts made using the two distinct extraction techniques, as shown by the wide bands between 3321 and 3338 cm⁻¹ (Rajan and Cindrella 2019). The asymmetric and symmetric vibrational modes corresponding to the C-H stretch vibrations are responsible for the peaks that appear between 2848 and 2976 cm⁻¹, indicating the presence of the aromatic C-H group (Ammar et al. 2019). The C = O stretching mode vibrations are thought to be responsible for the small shoulder peak that corresponds to a frequency of 1743 cm⁻¹ (Singh et al. 2021). Of > C = O, the stretching vibrational mode is represented by the peaks at 1656, 1653, and 1654 cm⁻¹ (Montagni et al. 2023). According to previous research (Güzel et al. 2018), vibrations of the aromatic ring and aromatic C = C bond are responsible for the peaks seen at 1596 and 1456 cm⁻¹, respectively. Strong bands seen at 1038 cm⁻¹ are caused by the C-O stretching vibrational modes, while the peak at 1205 cm⁻¹ is attributed to the amine groups' C-N bending (Safaei-Ghomi and R.* Masoomi, M. Hosseinpour, H. Batooli 2020). The bands at 987 and 918 cm⁻¹ are a result of out-of-plane bending of the C-H atom. Moreover, the planar bending vibration peaks of O–C–O are corresponding to the 718 and 875 cm⁻¹ peaks, respectively (Hajji et al. 2017). It is well known that functional groups in dye molecules, such as hydroxyl or/and carbonyl, can more effectively form bonds between the dye molecules and the TiO₂ surface. Better electron transfer to the TiO₂ conduction band is made possible by the strong interaction between the hydroxyl or/and carbonyl groups in natural dye molecules and the hydroxyl group on the TiO₂ surface (Singh et al. 2021). The functional groups in the sensitizing dyes affected the performance of the produced DSSCs. Compared to other dyes, the best sensitizer was obtained from Rhamnus tinctoria seeds by the soxhlet method (ɳ = 0.47, J_sc = 0.92 mA/cm², FF = 0.72 and V_oc = 0.71 V). This could be because the dye extracted from Rhamnus tinctoria seeds contains hydroxyl groups, which can bind strongly to the TiO₂ surface (Safaei-Ghomi et al. 2020). In addition, the dye extracted from Rubia fruticosa by the soxhlet method exhibited better sensitizing properties than the dye extracted from Rhamnus tinctoria by the ultrasonic method. This showed that pigment diversity and extraction methods in plants may exhibit different sensitization performances. On the other hand, dyes extracted from Pinus pinea bark by two different methods showed low V_oc (ultrasonic-0.36 V, soxhlet-0.41 V) and J_sc (ultrasonic-0.06 mA/cm², soxhlet-0.11 mA/cm²) values in DSSCs. The reason for this low V_oc and J_sc compared to other extracts may have been due to the low concentration of some functional groups that have high performance in DDSCs (Hamadanian et al. 2012). FTIR test results of dyes extracted from Pinus pinea bark, Rhamnus tinctoria seeds, and Rubia fruticosa fruits by different methods in this study showed that these natural pigments with carbonyl or/and hydroxyl groups can be used as suitable sensitizers in DSSCs.

Figure 12 shows the J_sc-V curves of DSSCs made with natural dyes, and Table 1 summarizes their photovoltaic parameters. Photovoltaic measurements were carried out under 100 mW/cm² illumination (AM 1.5G). Equation 2 and 3 were used to calculate the power conversion efficiency (ɳ) and filling factor (FF) of the prepared DSSCs (Franchi et al. 2023). While P_max and P_in in these equations represent the maximum output power and input power, respectively, V_oc, FF and J_sc correspond to the open circuit voltage, filling factor and short circuit current density. J_max and V_max reflect the maximum current density and maximum voltage, respectively.

The cell using the sensitizer extracted from Rhamnus tinctoria seeds by the soxhlet method showed the highest power conversion efficiency (0.47%) with J_sc 0.92 mA/cm², V_oc 0.71 V and FF 0.72. DSSCs using dyes leached by the Soxhlet method exhibited superior cell performance compared to the ultrasonic method. This can be explained by the fact that the dyes extracted by the Soxhlet method show absorption at a wider wavelength in the visible region and the high interaction of carbonyl or/and hydroxyl groups in the sensitizers with the TiO₂ surface provides a better charge transfer (Kabir et al. 2022b). The low power conversion efficiency of ultrasonically leached dyes in DSSCs may be due to the degradation of some high-performance functional groups during extraction. This situation restricts the dye's interaction with the TiO₂ surface (Hamadanian et al. 2012). Moreover, DSSCs using emodin and alizarin dyes extracted from Rhamnus tinctoria and Rubia fruticosa by the soxhlet method showed maximum V_oc values of 0.71 and 0.65 V, respectively. The high V_oc values in DSSCs are due to the high interaction of the alkyl chains in the dye with the TiO₂ surface. Thanks to this strong steric barrier, electrons in the TiO₂ thin film were prevented from leaking into the electrolyte solution (Lin et al. 2015). On the other hand, it was determined that the photovoltaic parameters of DSSCs sensitized with quercetin dye extracted from Pinus pinea bark by soxhlet method were quite low, such as V_oc 0.36 and J_sc 0.06 mA/cm². These low V_oc and J_sc values seen in DSSCs may be due to the low concentration of some high-performance pigments and insufficient electron injection into the cells, respectively (Narayan 2012). FF has a significant impact on the performance of DSSCs. DSSCs sensitized with dyes using the ultrasonic method have a high recombination rate and, accordingly, they have lower FF values than cells sensitized with dyes using the soxhlet method, because low charge recombination indicates maximum photoelectron lifetime and minimum dark current. This allows the cell to have higher FF values (Tian et al. 2011; Yedidi et al. 2014). As a result, natural dyes obtained by different extraction methods may exhibit different sensitization performances. With sensitizers leached by various methods, photovoltaic parameters such as V_oc and J_sc can be improved in DSSCs. In addition, thanks to the practical applications of these methods, more economically suitable solar energy devices can be produced.

Interfacial charge transfer of DSSCs sensitized with natural dyes was investigated using electrochemical impedance spectroscopy (EIS). Under 100 mW/cm² light intensity, the response of the prepared cells to 10 mV open circuit voltage was measured in the frequency range from 0.1 Hz to 100 kHz. Figure 13a shows the Nyquist curves of various DSSCs produced. Additionally, the resistance parameters calculated from the equivalent circuit are summarized in Table 2 using Zview software. The charge transfer resistance at the Pt/counter electrode/electrolyte interface is defined as the high-frequency region. The recombination resistance (R_ct2) at the TiO₂/sensitizing dye/electrolyte interface is reflected by larger semicircles in the mid-frequency region. The low-frequency region corresponds to the Warburg resistance and redox couple (I⁻³/I⁻) diffusion properties (Chandra Maurya et al. 2019). A Nyquist diagram typically has three semicircles, but in Fig. 13a only two semicircles are visible in the high and mid-frequency regions. The undetectable third semicircle may be due to Warburg diffusion of redox couples in the electrolyte or time constant difference (Bella et al. 2016). R_s in Table 2 represents the series resistance, R_ct1 represents the electron transport resistance at the counter electrode, and R_ct2 represents the electron transfer resistance between TiO₂/electrolyte, which is directly proportional to the diameter of the arc at medium frequency. DSSCs sensitized with dyes extracted from natural plants by soxhlet and ultrasonic method showed charge transfer resistance ranging from 105.7 to 1593.6 Ω. It is understood from Table 2 that the plant variety and extraction methods significantly affected the load transfer resistance of the cells produced. Devices sensitized with quercetin dyes leached from the bark of the Pinus pinea plant by soxhlet and ultrasonic method showed high load transfer resistance. High charge transfer resistance indicated a higher rate of degradation and detachment, resulting in lower cell efficiency (Lohrasbi et al. 2013; Qi et al. 2018). In addition, as the R_ct2 value increased in the produced samples, it was observed that the V_oc values decreased from 0.71 to 0.36. High R_ct2 values caused slower electron recombination between TiO₂/dye/redox, which negatively affects the V_oc values (Rajan and Cindrella 2019). On the other hand, it was observed that DSSC sensitized with emodin dye extracted from Rhamnus tinctoria seeds by the soxhlet method had the lowest charge transfer resistance (R_ct2 = 105.7 Ω) and longer electron lifetime (\(\tau_{e}\) = 590 µs). The lower R_ct2 and higher \(\tau_{e}\) values observed in this cell are consistent with the higher power conversion efficiency (ɳ = 0.47) of this native dye-sensitized cell.

The Bode phase plot corresponding to the charge transfer process of cells sensitized with natural dyes is shown in Fig. 13b. In low-frequency regions, Bode curves are closely related to electron lifetime. Long electron lifetime reflects improved electron collection ability and low recombination resistance. The electron recombination lifetime (\(\tau_{e}\)) is calculated using Eq. 4 (Akman and Karapinar 2022); in this equation, f_max is the Bode peak. The highest electron lifetime (\(\tau_{e}\) = 590 μs) was observed in DSSC sensitized with emodin dye extracted from Rhamnus tinctoria seeds by the soxhlet method. In a study by Akman, it was emphasized that cells with a long electron lifetime limit the recombination of photo-injected electrons and redox pair I⁻/I³⁻ ions, thus leading to a significant improvement in the efficiency of DSSCs (Akman 2020). On the other hand, the f_max of the cell sensitized with quercetin dye extracted from Pinus pinea barks by the ultrasonic method is in the high-frequency region; therefore, \(\tau_{e}\) has the lowest value with 39.3 µs. The low \(\tau_{e}\) value observed in this cell negatively affected the efficiency of this device (ɳ = 0.013). When Table 1 and 2 are examined, it is seen that electron lifetimes and photoelectrical parameters of DSSCs support each other.

In this, natural dyes leached from Rhamnus tinctoria seeds, Rubia fruticosa fruits, and Pinus pinea bark by soxhlet and ultrasonic method were used as sensitizers in TiO₂-based DSSCs. FTIR spectra revealed the presence of functional groups found in natural dyes. It was concluded that natural pigments containing carbonyl or/and hydroxyl groups are good sensitizers for DSSCs. It was observed that the dyes extracted by various methods exhibited different absorption behaviors in the UV–vis region. Different extraction methods caused differences in the maximum absorption wavelengths of the dyes. Dyes extracted by the Soxhlet method showed absorption in a wider wavelength range in the UV–vis region. DSSC sensitized with dye extracted from Rhamnus tinctoria seeds by the soxhlet method showed the highest cell performance (ɳ = 0.47). This situation was explained by the strong interaction of the carbonyl and/or hydroxyl groups in the emodin dye extracted by the Soxhlet method with the TiO₂ surface and the absorption behavior in a wider wavelength range in the visible region. It was also concluded that the resistance parameters obtained from EIS analyses and the photovoltaic parameters of the cells supported each other. It was understood that the DSSC with the highest power conversion efficiency (ɳ = 0.47) had lower R_ct2 and higher τ_e value. Recent advances in DSSCs produced using different natural sensitizers have led to the use of dyes with high absorbency in the UV–vis region. The extraction method used to extract dyes has a major impact on the efficiency of DSSCs. Because natural dyes are inexpensive and environmentally friendly, using them in DSSCs is a good alternative. In addition, light harvesting natural dyes for DSSCs could provide a sustainable solution to meet future energy needs.