Effect of ZnO Nanomaterial and Red and Green Cabbage Dyes on the Performance of Dye-Sensitised Solar Cells

: Visible light can be converted into electricity using dye sensitised solar cells (DSSCs), with their performance mainly based on the type of dye used as a sensitiser. Currently, dyes extracted from natural sources are highly preferred by researchers in this ﬁeld. Natural dyes reduce the high cost of metal complex sensitisers and replace expensive processes of chemical synthesis with simple extraction processes. Natural dyes are environmentally friendly, abundant, easily extractable, and safe. Their application has become a promising development in DSSC technology. In this study, two natural dyes extracted from the plant leaves of green cabbage (GC) and red cabbage (RC) that were used as sensitisers. The performance characteristics of RC and GC extracts were investigated using both cyclic voltammetry and amperometry methods for solar cell detection. At an extraction temperature of 60 ◦ C maintained for 8 h under optimum conditions, the measured values of maximum power ( P m ), ﬁll factor (FF), and efﬁciency ( η ) were 1.36 mW/cm 2 , 92.34%, and 0.161% for RC, and 0.349 mW/cm 2 , 44.19%, and 0.095% for GC, respectively. The RC and GC extracts exhibited excellent electrochemical performance with respect to current density potential and good cycling stability.


Introduction
Fossil fuels such as oil, natural gas, and coal are the main sources of energy used all over the world. Moreover, the burning of these fossil fuels emits tons of carbon dioxide, which pollutes the environment and changes the climatic conditions. The development of renewable energy sources may lead to sustainable technologies for a clean environment. Solar, wind, hydro, biomass, and geothermal energy generation technologies constitute the main sources of renewable energy, and all of them have their own merits and shortcomings [1]. Solar energy provides energy to all humans, plants, and animals for their growth and development; however, it varies geographically across the Earth. Solar energy can be directly converted into heat or electricity at commercial and domestic levels, and it is beneficial to individual owners as well as the environment. Today, scientists and engineers seek to utilise solar radiation directly to generate electricity through economic devices [2]. A photovoltaic cell converts solar radiation directly into electrical energy [3]. Silicon is usually employed for the fabrication of crystalline solar cells because it is abundant and an environmentally safe material. These solar cells are manufactured from silicon wafers, and their efficiency is higher than that of titanium dioxide (TiO 2 ), zinc oxide (ZnO), and iron oxide. The overall commercial cost of crystalline solar cells is high due to their high cost of fabrication. Tobnaghi et al. [4] used one PV (photovoltaic) cell where the upper or n-type layer is crystalline silicon doped with phosphorus with 5 valence electrons while the lower or p-type layer is doped with boron, which has 3 valence electrons. The performance of crystalline solar cells is dependent on temperature; an increase of 67% in power conversion efficiency is observed at 36 • C for concentrator-coupled cells.
tions. Sadasivuni et al. [26] presented a review of recent research progress in flexible solar cell devices. Bahman et al. [27] used a surface methodology to extract alkaloid compounds. Divya et al. [22] used an anthocyanin-rich extract with TiO 2 thin films for the photocatalytic degradation of methylene blue. Harnchana et al. [24] used poly (N-vinylcarbazole) (PVK) as an additive in the I − /I −3 redox electrolyte and studied the photoelectrochemical effect of PVK with various concentrations, in terms of the corresponding photovoltaic performance. Bahman et al. [27] explored various approaches for improving the current-voltage potential of solar cells. Ahmad and Kan [28] developed visible-light-driven methods by coating cotton fabrics with dye-sensitised TiO 2 . Liang and Chen [29] used arylamine organic dyes, demonstrated the advantages/disadvantages of each class, compared the cell performance of each dye, and pointed out that the field that needs to reinforce the research direction towards the further application of DSSCs.
Calogero et al. [30] compared the solar energy conversion efficiency of synthetic dyes with that of vegetable dyes. Salinas and Ariza [31] focused on the selection, extraction, and characterization of natural dyes. Narayan [32] enhanced the sustainability strategies of classical DSSC materials. Anbarasan et al. [33] presented a detailed summary and evaluation of DSSCs. Pratiwi et al. [34] discussed the effect of adding synthetic dye to anthocyanin dye. Sharma et al. [35] outlined the relation between the construction principles of DSSCs and their efficiencies. Muliani et al. [36] described the fabrication of DSSCs based on gel electrolytes. Prabavathy et al. [37] studied DSSCs operating with iodide-based electrolytes and platinum as counter electrodes. Krawczak [38] reviewed the published literature on the influence of photosensitisers on DSSC efficiency. Hamid et al. [39] used ethanol to extract natural sensitisers for the fabrication of a DSSC device.
Alami et al. [40] presented the synthesis and deposition of CuI, Fe, and Cu materials on copper substrates for DSSC applications. Narayan [32] discussed the emergence, operation, and components of DSSCs together with the prior work done on natural dye-based DSSCs. Hosseinnezhad [41] developed tandem DSSCs using perovskite materials on the top and organic dyes at the bottom. Shaikh et al. [42] and Kumar et al. [43] focused on photo-anodes counter electrodes of DSSCs and nanostructured materials. Maurya et al. [44] used natural dyes extracted from four flowers as sensitisers for TiO 2 -based DSSCs. Ruhane et al. [45] explored the formulation of nanocrystalline TiO 2 semiconductor paste for DSSCs. Bandy et al. [46] used electrophoretic deposition of TiO 2 nanocrystalline particles for applications in DSSCs. Kotta et al. [47] used a non-metallic doping procedure for developing a new material with a performance similar to that of DSSCs. Licklederer et al. [48] used hightemperature H 2 S treatment for synthesising Nano tubular TiS 2 structures of anodised TiO 2 nanotube layers and by extension for use as counter electrodes in DSSCs. Selvaraja et al. [49] studied the effect of changing the TiO 2 electrode thickness on the performance of DSSCs. Desai et al. [50] summarised and discussed the reasons for the exceptional performance of ZnO superstructures. Le et al. [51] reported ZnO urchin-like nanostructures synthesised by a one-step hydrothermal method, with the morphology optimised via the hydrothermal temperature. Habibi et al. [52] discussed the effect of coating the working electrodes with zinc titanate (ZT), ZnO, and TiO 2 on a DSSC's optical efficiency. Solano et al. [53] synthesised TiO 2 and ZnO nanoparticles using a green chemistry approach. Susanti and Wicaksana [54] investigated the usage of a purple cabbage (Brassica oleracea var. capitata f.) and duwet fruit (Eugenia cumini) as natural dyes for constructing a DSSC prototype.
The performance of a dye solar cell is generally evaluated by determining the different parameters of the cell, such as the open circuit voltage, short circuit current, fill factor, maximum voltage, and maximum current of the cell [55,56]. Each of these parameters is discussed in detail as follows:

Current-Voltage (I-V) Characteristics
Under constant illumination and a potential scan application from zero voltage (short circuit conditions) to the open-circuit potential, I-V curves can be obtained in which all possible working points in the considered range are plotted. DSSCs have a relatively Coatings 2021, 11, 1057 4 of 21 slow electrical response due to their high capacity. Therefore, the voltage scan must be performed sufficiently slowly. I sc is determined at the zero voltage measurement point and V oc obtained from the zero current measurement point of the I-V curve. The maximum output power of a solar cell is obtained when the product of |V × I| reaches a maximum. At this characteristic point the slope of the I-V curve is highlighted, as shown in Figure 1.
factor, maximum voltage, and maximum current of the cell [55,56]. Each of these parameters is discussed in detail as follows:

Current-Voltage (I-V) Characteristics
Under constant illumination and a potential scan application from zero voltage (short circuit conditions) to the open-circuit potential, I-V curves can be obtained in which all possible working points in the considered range are plotted. DSSCs have a relatively slow electrical response due to their high capacity. Therefore, the voltage scan must be performed sufficiently slowly. Isc is determined at the zero voltage measurement point and Voc obtained from the zero current measurement point of the I-V curve. The maximum output power of a solar cell is obtained when the product of |V × I| reaches a maximum. At this characteristic point the slope of the I-V curve is highlighted, as shown in Figure 1.

Open-Circuit Voltage
The open circuit voltage (Voc) of the solar cell is defined at the cell open terminals. With the cell temperature increment, the Voc decreases and is expressed as follows: (1)

Short-Circuit Current
The short circuit current (ISC) increases with an increasing temperature. It is expressed by the following equation: (2)

Fill Factor
The fill factor (FF) of a solar cell can be defined as the ratio of the actual power (product of voltage Vm and current Im at the maximum power point) to the dummy power (product of Voc and Isc).

Maximum Power
Maximum power can be obtained through the following equation:

Open-Circuit Voltage
The open circuit voltage (V oc ) of the solar cell is defined at the cell open terminals. With the cell temperature increment, the V oc decreases and is expressed as follows: (1)

Short-Circuit Current
The short circuit current (I SC ) increases with an increasing temperature. It is expressed by the following equation: (2)

Fill Factor
The fill factor (FF) of a solar cell can be defined as the ratio of the actual power (product of voltage V m and current I m at the maximum power point) to the dummy power (product of V oc and I sc ).

Maximum Power
Maximum power can be obtained through the following equation:

Efficiency
The Current-Voltage (I-V) characteristics of a solar cell under illumination are used to determine the Incident Photon to Current Conversion Efficiency η: where V oc is the open-circuit voltage, I sc is the short-circuit current, I o is the dark saturation current, V m is the voltage at the maximum power point, I m is the current at the maximum power point, V t is the terminal voltage of the cell, and η is the efficiency of the cell. Pin is the incident optical power. Solar Power Constant = 650 W/cm 2 , [35], Surface Area = 3.0 cm × 3.0 cm = 9.0 cm 2 . This study aimed to use two natural dyes, extracted from plant leaves of green cabbage (GC) and red cabbage (RC), as sensitisers with zinc oxide nanoparticles (NZnO) using a green chemistry approach. The performances of RC and GC extracts were investigated using both cyclic voltammetry and amperometry methods for solar cell evaluation. The optimum extracting temperature and application period are discussed with respect to the photovoltaic parameters, that is, P m , FF, and η values. The effects of the RC and GC extracts on the electrochemical performance, current density potential, and cycling stability of the cells were also determined, since they are promising alternative sensitisers for DSSCs for absorbing, transforming, and converting solar energy into electricity.

Preparation of NZnO
Aqueous solutions of zinc sulphate heptahydrate and sodium hydroxide were combined drop-wise in a molar ratio of 1:2 under vigorous stirring for almost 18 h. A large amount of white precipitate was formed. This precipitate was filtered, washed with distilled water, and dried using a muffle furnace at a temperature of 100 • C [57].

Preparation of NZnO Paste
NZnO pastes were prepared using NZnO powder in ethanol. The NZnO suspension was stirred at 300 rpm for 30 min. NZnO pastes that had been formed were inserted into a bottle covered with aluminium foil and stored in an area free from direct sunlight.

Preparation of Natural Dye Sensitisers
Green cabbage (GC) and red cabbage (RC) leaves were cut into smaller pieces, washed several times, dried at 80 • C, and crushed into a fine powder. Subsequently, 10 g of GC ( Figure 2a

Preparation of Photo-Anode Electrode, Counter Electrode, and DSSC
NZnO powder (500 mg) was mixed with 3.5 mL concentrated acetic acid. A transparent conducting glass was cut into 3 cm × 3 cm pieces using a diamond glass cutter and used for both photo-anode and counter electrodes. A fluorine-doped tin oxide (FTO) conductive glass surface was scribed carefully and cleaned using a tissue paper. NZnO paste was sensitised by immersing it in the natural dye solutions, dried using a drier, deposited onto the FTO to form a photo-electrode (see Figure 3a), preheated at 120 °C for 10 min, and sintered on a conveyor-belt furnace at 500 °C for 15 min. The iodide electro-

Preparation of Photo-Anode Electrode, Counter Electrode, and DSSC
NZnO powder (500 mg) was mixed with 3.5 mL concentrated acetic acid. A transparent conducting glass was cut into 3 cm × 3 cm pieces using a diamond glass cutter and used for both photo-anode and counter electrodes. A fluorine-doped tin oxide (FTO) conductive glass surface was scribed carefully and cleaned using a tissue paper. NZnO paste was sensitised by immersing it in the natural dye solutions, dried using a drier, deposited onto the FTO to form a photo-electrode (see Figure 3a), preheated at 120 • C for 10 min, and sintered on a conveyor-belt furnace at 500 • C for 15 min. The iodide electrolyte composed of 2 g of iodine (I 2 ), 3 g of potassium iodide (KI), and 50 mL of deionised water and stirred to dissolve. Deionised water 100 mL diluted the dissolved solution. Two drops of this iodide electrolyte dropped onto the photo-anode electrode.

Preparation of Photo-Anode Electrode, Counter Electrode, and DSSC
NZnO powder (500 mg) was mixed with 3.5 mL concentrated acetic acid. A transparent conducting glass was cut into 3 cm × 3 cm pieces using a diamond glass cutter and used for both photo-anode and counter electrodes. A fluorine-doped tin oxide (FTO) conductive glass surface was scribed carefully and cleaned using a tissue paper. NZnO paste was sensitised by immersing it in the natural dye solutions, dried using a drier, deposited onto the FTO to form a photo-electrode (see Figure 3a), preheated at 120 °C for 10 min, and sintered on a conveyor-belt furnace at 500 °C for 15 min. The iodide electrolyte composed of 2 g of iodine (I2), 3 g of potassium iodide (KI), and 50 mL of deionised water and stirred to dissolve. Deionised water 100 mL diluted the dissolved solution. Two drops of this iodide electrolyte dropped onto the photo-anode electrode. Another glass plate was coated with carbon soot to act as a counter electrode (Figure 3b). Then the DSSC was assembled by clipping the photo-anode and the counter electrode together ( Figure 3c) using a pair of binder clips (Figure 3d). A voltammeter was used to characterise the complete DSSC properties under sunlight ( Figure 3e). The interaction of the dye molecules with the photons from the sunlight caused their photoexcitation, generating electron hole pairs that were injected into the conduction band of the semiconductor NZnO. The electrons travelled through the semiconductor to the transparent electrode and eventually reached the counter electrode, regenerating the electrolyte via energy conversion ( Figure 3f). A three-electrode cell assembly consisting of an Ag/AgCl reference electrode, a Pt wire counter electrode, and a glassy carbon-working electrode were used for the electrochemical measurements.

Basic Characterization of the Fabricated DSSC
The DSSC was constructed using: (a) support (FTO) coated with Transparent Conductive Oxides; (b) semiconductor film, (NZnO); (c) GC or RC sensitiser absorbed onto the surface of a semiconductor; (d) an electrolyte containing a redox mediator, and (e) a counter electrode that is capable of regenerating the redox mediator. The basic operating principle for NZnO DSSC consists of three steps: Absorption, separation, and collection. These steps were optimized to attain better efficiency. Thus, absorption occurs at the first step of the reactions in DSSC. Under illumination, GC or RC sensitiser dye (D) absorbs a photon that leads to an excited sensitiser state (D*). Photoexcitation the sensitiser was followed by the electron injection into the conduction band of the semiconductor (mesoporous). This caused the sensitiser to enter an oxidized state D+. The electrolyte donate electron containing a redox couple and the original dye state were restored. Notably, the relevant iodide couple is Iodide/triiodide. Iodide regenerates the sensitiser and the sensitiser is regenerated itself by the reduction of triiodide at the counter electrode. In this way, the circuit is completed by transferring electrons via the external load, as shown in Figure 3f in agreement with Divya et al. [22].

UV-Visible Absorption Spectra of Natural Dyes
The absorption spectra of the dye solutions prepared from the RC and GC leaves showed maximum absorption peaks at 330 nm and 280 nm, respectively (see Figure 3). It is evident that these natural extracts absorbed light and hence fulfilled the primary criterion for their use as sensitisers in DSSCs. The absorption spectra of the RC dye showed a higher photosensitive efficiency for DSSCs, which is in agreement with prior results [58,59].

Fourier Transform Infrared (FT-IR) Analysis for GC and RC Dyes
The FT-IR device (IRPrestige21machine from Shimadzu, Japan) used to obtain the FT-IR spectrum of NZnO, RC dye, GC dye, RC dye with NZnO, and GC dye with NZnO at wave numbers ranging from 400-4000 cm −1 , Figure 4, and the interpretation of the spectra appears in Table 1. The FT-IR spectrum of the ZnO nanoparticles ( Figure 4a) includes the characteristic absorption bands at 510.61 cm −1 for NZnO, at 511.98 cm −1 for RC dye without NZnO, at 516.85 cm −1 for RC dye with NZnO, 516.54 for GC without NZnO, and 509.81 for GC dye with NZnO cm −1 . The absorption band in the range of 3200-3500 cm −1 was assigned to the OH groups on the surface of the NZnO [60]. The spectra shown in Figure 3b,c corresponded to the functional groups of the dye extracted from the GC, as detailed in Table 1  The GC dyes contain a pigment called chlorophyll that absorbed sunlight in the red (long wavelength) domain and in the blue (short wavelength) domain of the visible light spectrum. The chlorophyll strongly absorbed blue and orange-red light. The RC dyes contain anthocyanin pigments that make them purple. These dyes also contain chloroplasts with chlorophyll, only in lesser quantities than the GC dyes. The absorption spectrum of the anthocyanin pigments (a carotenoid pigment) includes features in violet light and blue-green light. This is because red light has a longer wavelength and lower energy than violet light. Figure 5 and Table 1 indicate that GC dye has a longer wavelength compared to RC dye with a short wave length, so RC dye showed a higher electrical conversion of energy compared to the GC dye. dye without NZnO, at 516.85 cm for RC dye with NZnO, 516.54 for GC without NZnO, and 509.81 for GC dye with NZnO cm −1 . The absorption band in the range of 3200-3500 cm −1 was assigned to the OH groups on the surface of the NZnO [60]. The spectra shown in Figure 3b,c corresponded to the functional groups of the dye extracted from the GC, as detailed in Table 1. The CH3 vibrations were observed in the range of 2930-2919 cm −1 , and the C-H2 and C-N vibrations were observed at 2855 cm −1 and 2360 cm −1 , respectively. Moreover, the vibrations associated with C=O stretching in esters appear at 1735 cm −1 and 1946 cm −1 in the GC, whereas the C=C vibrations appear in the ranges of 1604-1610 cm −1 and 1606-1615 cm −1 in the RC and the GC, respectively [61]. The C-O-C ether vibrations appear at 1019 cm −1 and 1021 cm −1 for the RC and at 1018 cm −1 and 1023 cm −1 for the GC. As observed from the functional groups of the dye extracted from the RC, detailed in Table 1

Scanning Electron Microscopy (SEM) Imaging of NZnO
The SEM images obtained for NZnO are shown at a high magnification in Figure 6. It can be seen that particles smaller than 20 nm formed. The image shows that the particles separated smoothly and were not significantly affected by agglomeration. The ZnO nanoparticles took the form of rectangles and triangles. Agglomerations of individual zinc nanoparticles can also be seen in the SEM images. The agglomerated lump showed the presence of several nanoparticle aggregates, while some individual crystals are also

Scanning Electron Microscopy (SEM) Imaging of NZnO
The SEM images obtained for NZnO are shown at a high magnification in Figure 6. It can be seen that particles smaller than 20 nm formed. The image shows that the parti-cles separated smoothly and were not significantly affected by agglomeration. The ZnO nanoparticles took the form of rectangles and triangles. Agglomerations of individual zinc nanoparticles can also be seen in the SEM images. The agglomerated lump showed the presence of several nanoparticle aggregates, while some individual crystals are also visible.

Scanning Electron Microscopy (SEM) Imaging of NZnO
The SEM images obtained for NZnO are shown at a high magnification in Figure 6. It can be seen that particles smaller than 20 nm formed. The image shows that the particles separated smoothly and were not significantly affected by agglomeration. The ZnO nanoparticles took the form of rectangles and triangles. Agglomerations of individual zinc nanoparticles can also be seen in the SEM images. The agglomerated lump showed the presence of several nanoparticle aggregates, while some individual crystals are also visible.

Energy Dispersive Spectroscopy (EDS) of NZnO
The EDS plots obtained for NZnO are illustrated in Figure 7. The ZnO nanoparticles' optical absorption peaks were attributed to the surface plasmon resonance of the ZnO nanoparticles. The origin of these elements lies in the biological components, mostly aligning along with the ZnO nanoparticles and containing SKa, SKb VKa, VKb, and Zn, which provide chemical information of the synthesised NPs (Nanoparticles) in agreement with [63,64]. The values of S and V elements found in the EDS plots are traces from

Energy Dispersive Spectroscopy (EDS) of NZnO
The EDS plots obtained for NZnO are illustrated in Figure 7. The ZnO nanoparticles' optical absorption peaks were attributed to the surface plasmon resonance of the ZnO nanoparticles. The origin of these elements lies in the biological components, mostly aligning along with the ZnO nanoparticles and containing SK a , SK b VK a , VK b , and Zn, which provide chemical information of the synthesised NPs (Nanoparticles) in agreement with [63,64]. The values of S and V elements found in the EDS plots are traces from the aqueous solutions of zinc sulphate heptahydrate and sodium hydroxide that were combined drop-wise in a molar ratio of 1:2 to obtain NZnO as given in item 2.1. Preparation of NZnO.

Electrochemistry of Red and Green Cabbage Extracts
The voltammogram curves for the RC and GC extracted films are shown in Figure  8a-f. The current-potential relations for the RC and GC extracts are shown individually in Figure 8a,b, respectively. For the RC, there were two anodic peaks: at 0.5 V with 0.19 mA and at 0.8 V with 0.21 mA, coupled with two cathodic peaks, at 0.2 V with −0.18 mA and at 0.17 V with 0.18 mA, respectively. For the GC, there were also two anodic peaks, at 0.2 V with 0.17 mA and 0.65 V with 0.195 mA, coupled with two cathodic peaks, at 0.3 V with −0.19 mA and at 0.15 V with 0.15 mA. The two cabbage extracts displayed higher reversibility and faster redox reactions due to the Faradaic pseudo capacitance (see Figure 8c). With an increase in the scan rate, the distances between the peaks increased for both the anode and the cathode, owing to the higher limit of the ion diffusion rate required to satisfy electronic neutralization during the redox reaction. The RC extract outperformed the GC extract (see Figure 8d). The CV curves have similar shapes for both

Electrochemistry of Red and Green Cabbage Extracts
The voltammogram curves for the RC and GC extracted films are shown in Figure 8a-f. The current-potential relations for the RC and GC extracts are shown individually in Figure 8a  ). With an increase in the scan rate, the distances between the peaks increased for both the anode and the cathode, owing to the higher limit of the ion diffusion rate required to satisfy electronic neutralization during the redox reaction. The RC extract outperformed the GC extract (see Figure 8d). The CV curves have similar shapes for both extracts, implying that both the extracts possessed reversibility as well as faster redox reactions, which is consistent with the results in the literature [64][65][66]. These observations describe the near-surface layer storage charge pseudo-capacitance. The RC-derived cells possessed higher values of pseudo-capacitance, in agreement with previously reported results [67].  Figure 9c illustrates the J/V relation of both GC and RC trends where the RC trend is shown to be higher than the GC trend, indicating the higher absorption of sun energy for RC compared to GC. The amperometric determination of the charge and discharge curves in terms of current and time for the RC and the GC is shown in Figure 8e,f, respectively. The nonlinear charge and discharge curves represent pseudo-capacitive processes, confirming the CV results. There were two charge trends: the first began at −0.02 mA and ended at −0.005 mA for the RC, and the second began at 0.079 mA and ended at −0.011 for the GC (see Figure 8e). The discharge profiles of the RC and GC are illustrated in Figure 8f, showing two sloping potential regions of −0.02 mA to −0.2 mA for the RC, and −0.02 mA to −0.05 mA for the GC.

J/V Relation of GC and RC
These observations describe the near-surface layer storage charge pseudo-capacitance. The RC-derived cells possessed higher values of pseudo-capacitance, in agreement with previously reported results [67]. Figure 9 illustrates the J/V relation of both GC and RC. Figure 9a shows the J/V relation of GC with the following values: J sc GC, V oc GC, Jmax. GC, V max. GC of 0.223 mAcm 2 , 0.456 V, 0.157 mAcm 2 , and 0.416 V, respectively. Figure 9b showed the J/V relation of RC with the following values: J sc GC, V oc GC, J max. GC, V max. GC of 0.797 mAcm 2 , 1.306 V, 0.531 mAcm 2 , and 0.59 V, respectively. Figure 9c illustrates the J/V relation of both GC and RC trends where the RC trend is shown to be higher than the GC trend, indicating the higher absorption of sun energy for RC compared to GC.

Effect of Test Temeratures and Periods on the Photovolatic Properties of GC and RC Solar Cells
The effects of temperature on both the short and open circuit current densities and the voltages of the RC and GC-based cells are shown in Figures 10 and 11, respectively. It is shown that the RC-based cells' short and maximum circuit current and open-circuit voltage trends are higher than those of the GC-based cells. This is attributed to the strong bonds between the dye and ZnO molecules through which electrons can be transported from the excited dye molecules to the nano ZnO film, which is opposite to the results reported in [16]. Also, the lower trends of the GC-based cells are due to the reduced stability of chlorophyll at higher temperatures, in agreement with previously reported results [66].

Effect of Test Temeratures and Periods on the Photovolatic Properties of GC and RC Solar Cells
The effects of temperature on both the short and open circuit current densities and the voltages of the RC and GC-based cells are shown in Figures 10 and 11, respectively. It is shown that the RC-based cells' short and maximum circuit current and open-circuit voltage trends are higher than those of the GC-based cells. This is attributed to the strong bonds between the dye and ZnO molecules through which electrons can be transported from the excited dye molecules to the nano ZnO film, which is opposite to the results reported in [16]. Also, the lower trends of the GC-based cells are due to the reduced stability of chlorophyll at higher temperatures, in agreement with previously reported results [66].

Effect of Test Temeratures and Periods on the Photovolatic Properties of GC and RC Solar Cells
The effects of temperature on both the short and open circuit current densities and the voltages of the RC and GC-based cells are shown in Figures 10 and 11, respectively. It is shown that the RC-based cells' short and maximum circuit current and open-circuit voltage trends are higher than those of the GC-based cells. This is attributed to the strong bonds between the dye and ZnO molecules through which electrons can be transported from the excited dye molecules to the nano ZnO film, which is opposite to the results reported in [16]. Also, the lower trends of the GC-based cells are due to the reduced stability of chlorophyll at higher temperatures, in agreement with previously reported results [66].   The relationships between circuit current density and voltage as functions of test periods and temperature are shown in Figures 12 and 13 for the RC and GC-based solar cells. For both test periods and temperatures, the GC-based cells' maximum current trends were higher than those of the RC-based cells. This was due to the presence of larger amount of the green matter necessary for photosynthesis in green cabbage than in red cabbage, which is in agreement with the results in previous reports [68].  The relationships between circuit current density and voltage as functions of test periods and temperature are shown in Figures 12 and 13 for the RC and GC-based solar cells. For both test periods and temperatures, the GC-based cells' maximum current trends were higher than those of the RC-based cells. This was due to the presence of larger amount of the green matter necessary for photosynthesis in green cabbage than in red cabbage, which is in agreement with the results in previous reports [68]. The relationships between circuit current density and voltage as functions of test periods and temperature are shown in Figures 12 and 13 for the RC and GC-based solar cells. For both test periods and temperatures, the GC-based cells' maximum current trends were higher than those of the RC-based cells. This was due to the presence of larger amount of the green matter necessary for photosynthesis in green cabbage than in red cabbage, which is in agreement with the results in previous reports [68].  The relations between the solar cell temperature, incident power, and the conversion energy efficiency are shown in Figures 14 and 15. Although the incident power for the GC-based cells was higher than that of the RC-based cells, the energy conversion efficiency for the RC-based cells was higher, owing to the weak bonds between the dye and the nano-ZnO molecules in the GC-based cells, meaning the electrons could not be as readily transported from the excited dye molecules to the nano ZnO film [69].  The relations between the solar cell temperature, incident power, and the conversion energy efficiency are shown in Figures 14 and 15. Although the incident power for the GC-based cells was higher than that of the RC-based cells, the energy conversion efficiency for the RC-based cells was higher, owing to the weak bonds between the dye and the nano-ZnO molecules in the GC-based cells, meaning the electrons could not be as readily transported from the excited dye molecules to the nano ZnO film [69]. The relations between the solar cell temperature, incident power, and the conversion energy efficiency are shown in Figures 14 and 15. Although the incident power for the GC-based cells was higher than that of the RC-based cells, the energy conversion efficiency for the RC-based cells was higher, owing to the weak bonds between the dye and the nano-ZnO molecules in the GC-based cells, meaning the electrons could not be as readily transported from the excited dye molecules to the nano ZnO film [69].  The maximum power variations of the solar cells as functions of temperature and test periods are shown in Figures 15 and 16. It is clear that the RC trends lie at higher values than the GC trends in both cases. This is because the RC carboxylic compounds exhibit a better anchoring between the dye and the nano ZnO surface, with an efficient dye regeneration indicating that the interaction between the sensitiser and the nano ZnO film is significant in enhancing the energy conversion efficiency of DSSCs. This is in agreement with the results reported previously [70]. Therefore, the DSSCs made from RC have fewer parasitic losses than the other tested cells, and can be considered as good-quality cells. Figures 15 and 16 also show that the maximum solar cell power can be obtained at a temperature of 60 °C and a test period of 8 h for both RC and GC, which is in agreement with the results of the previous investigations [71,72]. In the present work: the measured values of maximum power (Pm), fill factor (FF), and efficiency (η) were 1.36 mW/cm 2 , 92.34%, and 0.161% for RC, and 0.349 mW/cm 2 , The maximum power variations of the solar cells as functions of temperature and test periods are shown in Figures 15 and 16. It is clear that the RC trends lie at higher values than the GC trends in both cases. This is because the RC carboxylic compounds exhibit a better anchoring between the dye and the nano ZnO surface, with an efficient dye regeneration indicating that the interaction between the sensitiser and the nano ZnO film is significant in enhancing the energy conversion efficiency of DSSCs. This is in agreement with the results reported previously [70]. Therefore, the DSSCs made from RC have fewer parasitic losses than the other tested cells, and can be considered as good-quality cells. Figures 15 and 16 also show that the maximum solar cell power can be obtained at a temperature of 60 • C and a test period of 8 h for both RC and GC, which is in agreement with the results of the previous investigations [71,72]. The maximum power variations of the solar cells as functions of temperature and test periods are shown in Figures 15 and 16. It is clear that the RC trends lie at higher values than the GC trends in both cases. This is because the RC carboxylic compounds exhibit a better anchoring between the dye and the nano ZnO surface, with an efficient dye regeneration indicating that the interaction between the sensitiser and the nano ZnO film is significant in enhancing the energy conversion efficiency of DSSCs. This is in agreement with the results reported previously [70]. Therefore, the DSSCs made from RC have fewer parasitic losses than the other tested cells, and can be considered as good-quality cells. Figures 15 and 16 also show that the maximum solar cell power can be obtained at a temperature of 60 °C and a test period of 8 h for both RC and GC, which is in agreement with the results of the previous investigations [71,72]. In the present work: the measured values of maximum power (Pm), fill factor (FF), and efficiency (η) were 1.36 mW/cm 2 , 92.34%, and 0.161% for RC, and 0.349 mW/cm 2 , In the present work: the measured values of maximum power (P m ), fill factor (FF), and efficiency (η) were 1.36 mW/cm 2 , 92.34%, and 0.161% for RC, and 0.349 mW/cm 2 , 44.19%, and 0.095% for GC, respectively ( Figure 17). By comparison, Amadi et al. [72] produced the green cabbage's greatest photoelectric conversion efficiency (η) at up to 0.1%, an opencircuit voltage (VOC) of 532 mV, and a short-circuit current density (J sc ) of 1.2 mA/cm 2 . Meanwhile, Pratiwi et al. [34] achieved a conversion efficiency of 0.024%, while the DSSC by combination dyes achieved a conversion efficiency of 0.054% by anthocyanin dye of red cabbage. Hamid et al. [39] found the efficiency of natural sensitisers for avocado peel, blueberry, and pomegranate to be 48.25 × 10./0%, 24.62 × 10./0%, and 15.75 × 10./0%, respectively. For natural dyes extracted from four different flowers, namely, Cassia surattensis, Cassia tora, Cassia alata, and Cassia occidentalis, Maurya et al. [44] found that: the short circuit current density (I SC ) values ranged from 0.06 mA/cm 2 to 0.20 mA/cm 2 ; the open circuit voltage (V OC ) varied from 0.292 V to 0.833 V; the fill factor (FF) from 0.7 to 0.9; and the efficiencies (η) from 0.013% to 0.15%.  44.19%, and 0.095% for GC, respectively ( Figure 17). By comparison, Amadi et al. [72] produced the green cabbage's greatest photoelectric conversion efficiency (η) at up to 0.1%, an open-circuit voltage (VOC) of 532 mV, and a short-circuit current density (Jsc) of 1.2 mA/cm 2 . Meanwhile, Pratiwi et al. [34] achieved a conversion efficiency of 0.024%, while the DSSC by combination dyes achieved a conversion efficiency of 0.054% by anthocyanin dye of red cabbage. Hamid et al. [39] found the efficiency of natural sensitisers for avocado peel, blueberry, and pomegranate to be 48.25 × 10./0%, 24.62 × 10./0%, and 15.75 × 10./0%, respectively. For natural dyes extracted from four different flowers, namely, Cassia surattensis, Cassia tora, Cassia alata, and Cassia occidentalis, Maurya et al. [44] found that: the short circuit current density (ISC) values ranged from 0.06 mA/cm 2 to 0.20 mA/cm 2 ; the open circuit voltage (VOC) varied from 0.292 V to 0.833 V; the fill factor (FF) from 0.7 to 0.9; and the efficiencies (η) from 0.013% to 0.15%.

Conclusions
Based on our findings, the following conclusions were obtained: (1) Red and green cabbage dyes are promising alternative sensitisers for DSSCs; (2) As the measured values of maximum power (Pm), fill factor (FF), and efficiency (η) were 1.36 mow/cm 2 , 92.34%, and 0.161% for RC, and 0.349 mow/cm 2 , 44.19%, and 0.095% for GC, respectively. Therefore, RC extract is better than GC extract as sensitiser for DSSCs; (3) Cyclic voltammetry and amperometry techniques can be used to analyse the spectrum absorption and power conversion of the prepared DSSCs; (4) Preparation of DSSCs using nanocrystalline ZnO and two local natural dyes extracted from red cabbage and green cabbage was feasible; (5) The test sensitisation temperatures and the test period had the most significant effect on the spectrum absorption at a temperature of 60 °C and a test period of 8 h; (6) The colour has the most marked effect on the light absorption; (7) The light absorption characteristics of both red cabbage and green cabbage exhibited a broad absorption with peak absorbances at 340 nm and 300 nm, respectively; and (8) The red cabbage and green cabbage extracts achieved solar energy conversion efficiencies of 0.161% and 0.082%, respectively, at a 60 °C extraction temperature and 8 h extraction time, which are higher than the values previously reported in the literature.

Conclusions
Based on our findings, the following conclusions were obtained: (1) Red and green cabbage dyes are promising alternative sensitisers for DSSCs; (2) As the measured values of maximum power (P m ), fill factor (FF), and efficiency (η) were 1.36 mow/cm 2 , 92.34%, and 0.161% for RC, and 0.349 mow/cm 2 , 44.19%, and 0.095% for GC, respectively. Therefore, RC extract is better than GC extract as sensitiser for DSSCs; (3) Cyclic voltammetry and amperometry techniques can be used to analyse the spectrum absorption and power conversion of the prepared DSSCs; (4) Preparation of DSSCs using nanocrystalline ZnO and two local natural dyes extracted from red cabbage and green cabbage was feasible; (5) The test sensitisation temperatures and the test period had the most significant effect on the spectrum absorption at a temperature of 60 • C and a test period of 8 h; (6) The colour has the most marked effect on the light absorption; (7) The light absorption characteristics of both red cabbage and green cabbage exhibited a broad absorption with peak absorbances at 340 nm and 300 nm, respectively; and (8) The red cabbage and green cabbage extracts achieved solar energy conversion efficiencies of 0.161% and 0.082%, respectively, at a 60 • C extraction temperature and 8 h extraction time, which are higher than the values previously reported in the literature.