Microwave Assisted Preparation of Barium Doped Titania (Ba/TiO2) as Photoanode in Dye Sensitized Solar Cells

Pure TiO2 and barium (0.5 wt%) doped TiO2 (Ba/TiO2) nanostructures have been synthesized via facile microwave irradiation method. The pure anatase phase of synthesized photoactive material was confirmed by X-ray diffraction. Ba doping in the TiO2 host structure influenced the optical band gap as confirmed by UV-visible spectroscopy. The optical band gap increased from 3.21 eV for the TiO2 to 3.26 eV for Ba/TiO2. Morphological analysis of synthesized TiO2 and Ba/TiO2 was conducted using scanning electron microscopy. Energy dispersive X-ray spectroscopy confirmed the formation of Ba/TiO2 and no impurities were observed. Electrochemical impedance spectroscopy showed that the charge transfer resistance increased for Ba/TiO2, which reduced dark current creation in a dye-sensitized solar cell. The highest power conversion efficiency (3.24%) was achieved for Ba/TiO2 photoanode compared to 2.1% for a pure TiO2 photoanode-based device.


Introduction
Society needs green and cost-effective technologies for the energy sector owing to the global greenhouse effect and depletion of fossil fuels in recent years. Among all existing green energy technologies such as wind turbines, fuel cells, hydropower, tidal power, solar thermal and biomass, solar cells are considered among the most promising and prominent energy sources [1][2][3][4][5]. Dye-sensitized solar cells (DSSCs) constitute an advanced generation of solar cells that mimic natural photosynthesis. With fabrication techniques including solution processable methods, the use of cheap materials and decent power conversion efficiency (PCE), DSSCs are promising candidates for the conversion of solar light into electricity [6][7][8]. The highest PCE of 13% has been achieved in DSSCs by employing ruthenium as a light-absorbing material [9]. Techniques employed to improve the performance of DSSCs aim to minimize recombination of the photoinduced electron hole pair while improving the charge transfer rate of photo-generated electrons in the transparent conducting substrate [10]. Reducing the charge recombination and enhancing charge transfer can be achieved by optimizing the band gap of the semiconducting metal oxide (MOS). The MOS in a DSSC serves as a substrate for dye uplifting as well as provides a path for photoinduced charge generation [11,12].
TiO 2 nanostructures are one of the significant photoanode materials owing to their superior properties, including wide band gap, large surface area, non-toxicity, low cost and high stability [13,14]. The morphology of MOS plays a significant role in DSSCs; therefore, a fine-tuned morphology is required to deliver improved PCE in the device [15]. The photocatalytic efficiency of TiO 2 is affected by its ability to respond to radiation (UV along with visible light), and thus could be improved by shifting its optical absorption from UV to the visible region. It has been observed that the photocatalytic efficiency of TiO 2 is enhanced by modifying its band gap with different dopants [16]. A variety of TiO 2 nano-compositions involving nanoparticles, carbon spheres, TiO 2 nanoparticle composites, TiO 2 micro tablets, hollow spheres, nanotube composites, and nano-fiber nanoparticle composites have been successfully synthesized and used as DSSC photoanode materials [17][18][19][20]. Doping with transition metals [21,22], rare-earth elements [23,24] and noble metals [25] into TiO 2 has been widely investigated to reduce TiO 2 particle size and improve its (photo)(electro)catalytic performance. Co-doping along with different doping agents including metal-nonmetal, nonmetal-nonmetal, and metal-metal synergistically improves the optical absorption towards the visible region and reduces the recombination process [26]. Cu-N, B-N, Fe-C, Fe-N, Ni-N, and V-N have been studied as co-dopants with TiO 2 [27][28][29][30][31][32][33]. Co-doping of TiO 2 with N-C has been reported to modify its band gap, enhancing its photocatalytic activity towards the visible spectrum [34]. Co-doping of titania nanowires with tungsten carbon improves their electrochemical activity towards the oxygen-evolution reaction (OER) [35]. Ag/V-TiO 2 was also studied to photo-catalyze the degradation of rhodamine B dye in aqueous solutions. These co-doped materials exhibited the highest photocatalytic activity in the UV-visible region [36], Similarly, it has also been previously reported that barium doping in TiO 2 nanoparticles using microwave irradiation increases the optical band gap and reduces the particle size of TiO 2 nanoparticles for microbiological activity [37]. Several studies have reported co-doping of TiO 2 to enhance its optoelectrical properties.
The present work deals with the preparation of TiO 2 and Ba/TiO 2 composites using a simple microwave irradiation method. Microwave-assisted swift heating has achieved substantial significance as an emerging and favorable method for one-pot preparation of different metal nanostructures. Structural, optical, morphological, elemental, and electrochemical behaviors of synthesized photoanodes materials are investigated by X-ray diffraction (XRD), UV-visible spectroscopy, scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and electrochemical-impedance spectroscopy (EIS). Moreover, the photovoltaic performance of synthesized TiO 2 and Ba/TiO 2 photoanodes has been studied. Increasing the optical band gap and surface area of Ba/TiO 2 photoanodes led to a reduction of dark current and increased dye loading capacity [38]. DSSCs prepared using Ba/TiO 2 composite photoanode delivered a PCE of 2.34%.

Materials
Titanium (IV) isopropoxide TTIP (95% purity), ruthenium dye (N3 dye), polyethylene glycol (PEG), barium chloride and Triton X-100 were purchased from Alfa Aesar Company India. Ethanol was purchased from Merck Company India. Chloroplatinic acid (99.9%) was purchased from Himedia Company India. All of these chemicals were utilized as received, without any further treatments. All of the solutions were prepared using deionized water (DI).

Preparation of TiO 2 and Ba/TiO 2 Photoanode
A facile microwave irradiation method was employed to prepare Ba (0 . 5) TiO 2 photoanode materials. An initial solution was prepared using 2 g of TTIP added with 60 mL of deionized water and stirred for 1 h at 100 • C. A separate solution was prepared using barium chloride dissolved in DI water. After that, both solutions were mixed and kept under stirring for 1 h at room temperature. The solution was kept under microwave irradiation for 5 min at 450 W. The advantage of using a microwave oven relates to a reduced synthesis time and energy consumption as compared to conventional heating. Microwaves provide homogeneous and rapid heating [39]. Afterwards, the obtained photoanode materials were 3 of 11 calcined at 500 • C for 1 h to enhance the crystalline nature of the photoanode material. The same procedure was adopted for the preparation of TiO 2 photoanode.

Photoanode Preparation
Fluorine-doped tin oxide (FTO) glass substrates were cleaned in a soap solution followed by distilled water and ethanol. Hereafter, the synthesized photoanode material was prepared using binder solutions of PEG and Triton X100. As prepared, Ba/TiO 2 was deposited onto FTO using the doctor blade technique. The coated substrate was sintered at 300 • C for 1 h to evaporate the binder as well as to generate porosity in the materials. The sintered substrate was immersed into 0.5 mM of ruthenium dye solution for 24 h to absorb the dye molecule. A thin layer of platinum was deposited onto the FTO glass using 5 mM chloroplatinic acid via doctor blade technique. The photo-electrode was clamped (using binder clips) with the counter-electrode to produce a sandwich-type formation. The liquid redox electrolyte composed of 0.5M potassium iodide (KI) and 0.02 M iodine (I 2 ) dissolved in acetonitrile was filled between the photoelectrode and counter-electrode. The active surface area of the as-prepared DSSC was about 0.25 cm 2 and the cross-sectional area of TiO 2 and Ba/TiO 2 , approx. 10 µm.

Characterization
X-Ray diffraction (PANalytical-X'pert Pro powder diffractometer λ = 1.54 Å using Cu Kα radiation) analysis was conducted to determine the crystal structure of the MOS with 2θ range values from 10 • to 80 • . The measurement was carried out by step size of 0.0170 • up to four decimal accuracies. SEM (ZEIS EVO 18 instrument with 20 kV acceleration voltage and 1.30 K magnification) was used for the morphological studies of the of photoanode materials. UV-visible spectrophotometry (Perkin-Elmer-Lambda-35 spectrophotometer) was used to investigate the optical properties. EIS was conducted by VSP300 biological instrument with a 10 mV voltage amplitude in the frequency range of 1 MHz-1 Hz. The current voltage (I-V) measurement was observed by means of an electrochemical computercontrolled workstation as well as a SAN-EI solar-stimulator. A 150-watt xenon lamp was used as a light source with an AM 1.5 G filter and an intensity of 100 mW cm −2 .  [26]. No other characteristic peaks were reflected in the XRD pattern. Hence, both samples exhibited tetragonal structure, and corresponding lattice constant values are shown in Table 1. In the case of Ba/TiO 2 nanomaterials, an increase in intensity was evidenced due to the incorporation of Ba 2+ (1.42 nm) into the TiO 2 host lattice because the ionic radius of Ba 2+ is greater than that of Ti 4+ (0.74 nm), which resulted in an increase in crystallinity. This substitution caused lattice distortion, inflation of the TiO 2 nanocrystal, and internal stress due to Ba doping [40], which reduced nucleation growth (particle size controller) [41]. Another study showed that doping of substrates with Ba caused shrinkage of the crystallite size of modified nanomaterials [42]. The structural constraints of synthesized TiO2 and Ba/TiO2 nanomaterials were calculated using the following formulas [43]:

Structural Analysis
where d is the d-spacing between two consecutive lattices observed from XRD, (h,k,l) indicate the miller indices, t is the crystallite size, β is the full with half maximum, a, b, and c display as lattice constant for a = b in tetragonal structure, λ indicates the wavelength of the X-ray radiation used, and θ refers to the diffraction angle; where, ρ is the density of the prepared sample, n is 4 for anatase phase TiO2 and 2 for rutile phase TiO2, M is the molecular weight of the prepared nanomaterials, N corresponds to Avogadro's constant value, V is the volume of the prepared nanomaterials, and Sa states the specific surface of the prepared sample. Table 1 shows the calculated structural values The structural constraints of synthesized TiO 2 and Ba/TiO 2 nanomaterials were calculated using the following formulas [43]: where d is the d-spacing between two consecutive lattices observed from XRD, (h,k,l) indicate the miller indices, t is the crystallite size, β is the full with half maximum, a, b, and c display as lattice constant for a = b in tetragonal structure, λ indicates the wavelength of the X-ray radiation used, and θ refers to the diffraction angle; where, ρ is the density of the prepared sample, n is 4 for anatase phase TiO 2 and 2 for rutile phase TiO 2, M is the molecular weight of the prepared nanomaterials, N corresponds to Avogadro's constant value, V is the volume of the prepared nanomaterials, and Sa states the specific surface of the prepared sample. Table 1 shows the calculated structural values derived from the XRD patterns of TiO 2 and Ba/TiO 2 nanomaterials. The average crystallite sizes estimated were 25 nm for single TiO 2 and 16 nm for Ba/TiO 2 nanomaterials, respectively. When Ba ions were doped into the TiO 2 lattice, a decrease in nucleation (and subsequent growth rate) was observed for TiO 2 nanomaterials, leading to a reduction in particle size of Ba/TiO 2 . The reduction in particle size increases the surface area of TiO 2 . The specific surface area of TiO 2 (61 m 2 /g) was smaller than that of Ba/TiO 2 (95 m 2 /g). This favored dye adsorption in the photo electrode, thereby improving photovoltaic performance in the DSSC. Figure 2 represents the UV-visible spectra of TiO 2 and Ba/TiO 2 photoanode materials. Ba/TiO 2 exhibited a significant visible absorption above 400 nm, which was an indication of changes occurring in the optical absorption of TiO 2 photoanode material due to metal doping. Peak edges of Ba/TiO 2 photoanode showed some regular and trivial blue shifts comparative to TiO 2 photoanode. The blue shift was attributable to the Burstein-Moss effect, leading to the movement of Fermi level towards the conduction band due to an increase in electron concentration because of Ba doping [44,45]. It is very useful to increase the photocurrent density and also the charge transfer resistance of synthesized photoanodes. The absorption of TiO 2 and Ba/TiO 2 were at 415 nm and 400 nm, respectively. mance in the DSSC. Figure 2 represents the UV-visible spectra of TiO2 and Ba/TiO2 photoanode materials. Ba/TiO2 exhibited a significant visible absorption above 400 nm, which was an indication of changes occurring in the optical absorption of TiO2 photoanode material due to metal doping. Peak edges of Ba/TiO2 photoanode showed some regular and trivial blue shifts comparative to TiO2 photoanode. The blue shift was attributable to the Burstein-Moss effect, leading to the movement of Fermi level towards the conduction band due to an increase in electron concentration because of Ba doping [44,45]. It is very useful to increase the photocurrent density and also the charge transfer resistance of synthesized photoanodes. The absorption of TiO2 and Ba/TiO2 were at 415 nm and 400 nm, respectively. Optical band gap values were determined by using Tauc plots as presented in Figure  3a. Ba/TiO2 photoanode (3.26 eV) ha a higher band gap value compared to TiO2 photoanode (3.21 eV) because of quantum confinement effects observed due to the smaller size of as-synthesized samples. According to the theory of quantum efficiency, the holes of valence band and conduction band electrons are confined by the potential barrier of the surface. As a result, the band gap energy is increased with decreasing size of synthesized particles [46,47]. The increase in band gap of doped TiO2 can be interpreted in terms of Mott's transition effect and Burstein-Moss (BM) effects. According to Mott' criteria, the band gap of the donor combines with the conduction band of the host material [48], while BM effects present a relationship between the band gap and concentration of carriers [49,50], as given below: Optical band gap values were determined by using Tauc plots as presented in Figure 3a. Ba/TiO 2 photoanode (3.26 eV) ha a higher band gap value compared to TiO 2 photoanode (3.21 eV) because of quantum confinement effects observed due to the smaller size of assynthesized samples. According to the theory of quantum efficiency, the holes of valence band and conduction band electrons are confined by the potential barrier of the surface. As a result, the band gap energy is increased with decreasing size of synthesized particles [46,47]. The increase in band gap of doped TiO 2 can be interpreted in terms of Mott's transition effect and Burstein-Moss (BM) effects. According to Mott' criteria, the band gap of the donor combines with the conduction band of the host material [48], while BM effects present a relationship between the band gap and concentration of carriers [49,50], as given below:

Optical Properties
where λ, h, and m e are the effective reduced mass of TiO 2 , Planck's constant and the electron free mass, respectively. According to the BM effect, the carrier density of semiconductors increases with Ba doping, which causes the Fermi level to shift into the conduction band [51]. Due to this doping, the low energy transition is clogged, and broadening in the band gap takes place [52]. This novel effect of Ba doping on optical properties can pave the way to its extended use in optoelectronic and photovoltaic applications. The value of n coefficient is determined by the Tauc relationship (Equation (6)). The n coefficient is related to the possible transition of electrons. For indirect allowed electronic transition, its value is 2 and 3 for indirect forbidden transitions.
Appl. Sci. 2022, 12, 9280 6 of 11 where α is the absorption coefficient, h is Plank's constant, and A is absorption constant [53]. From the slope of ln F ® vs. ln (−ν − E g ), the value of n is determined. The result listed in Table 2 represents the approximated value of n = 2 for TiO 2 and Ba/TiO 2, which is in agreement with values in the literature for TiO 2 , which is an indirect semiconductor [54] Appl. Sci. 2022, 12, x 6 of 12 where λ, h, and me are the effective reduced mass of TiO2, Planck's constant and the electron free mass, respectively. According to the BM effect, the carrier density of semiconductors increases with Ba doping, which causes the Fermi level to shift into the conduction band [51]. Due to this doping, the low energy transition is clogged, and broadening in the band gap takes place [52]. This novel effect of Ba doping on optical properties can pave the way to its extended use in optoelectronic and photovoltaic applications.  (7), inset relates to Ba/TiO2. The value of n coefficient is determined by the Tauc relationship (Equation (6)). The n coefficient is related to the possible transition of electrons. For indirect allowed electronic transition, its value is 2 and 3 for indirect forbidden transitions.
where α is the absorption coefficient, h is Plank's constant, and A is absorption constant [53]. From the slope of ln F® vs. ln (-ν − Eg), the value of n is determined. The result listed in Table 2 represents the approximated value of n = 2 for TiO2 and Ba/TiO2, which is in agreement with values in the literature for TiO2, which is an indirect semiconductor [54] Figure 4 presents the UV-visible absorption spectrum of dye molecule-adsorbed TiO2 and Ba/TiO2 photoanodes. The Ba/TiO2 photoanode had high absorption compared to the TiO2 photoanode. The increased absorption was due to the increasing surface area of the Ba/TiO2 photoanode, which was also confirmed by XRD results. This behavior enhances photocurrent density in DSSCs. Moreover, the shift in wavelength in the visible region  (7), inset relates to Ba/TiO 2 .  Figure 4 presents the UV-visible absorption spectrum of dye molecule-adsorbed TiO 2 and Ba/TiO 2 photoanodes. The Ba/TiO 2 photoanode had high absorption compared to the TiO 2 photoanode. The increased absorption was due to the increasing surface area of the Ba/TiO 2 photoanode, which was also confirmed by XRD results. This behavior enhances photocurrent density in DSSCs. Moreover, the shift in wavelength in the visible region further supported our findings as productive in DSSCs, because the photocatalytic efficiency of TiO 2 is affected by its ability to respond to UV radiation along with visible light, which could be improved by shifting its optical absorption from UV to the visible region. Ba doping increased the absorption in the visible region.
Appl. Sci. 2022, 12, x 7 of 12 further supported our findings as productive in DSSCs, because the photocatalytic efficiency of TiO2 is affected by its ability to respond to UV radiation along with visible light, which could be improved by shifting its optical absorption from UV to the visible region. Ba doping increased the absorption in the visible region.

Morphological Studies
The surface morphologies of the TiO2 and Ba/TiO2 nanostructures were subsequently examined by SEM as shown in Figure 5. Both TiO2 (5a) and Ba/TiO2 (5b) exhibited similar agglomerated spherical shape particles. In terms of chemical composition, TiO2 nanostruc-

Morphological Studies
The surface morphologies of the TiO 2 and Ba/TiO 2 nanostructures were subsequently examined by SEM as shown in Figure 5. Both TiO 2 (5a) and Ba/TiO 2 (5b) exhibited similar agglomerated spherical shape particles. In terms of chemical composition, TiO 2 nanostructures contained Ti and O, only, which confirmed that the material was free from impurities, as observed from Figure 5c. The presence of Ba was confirmed in the Ba/TiO 2 nanostructures from Figure 5d.

Morphological Studies
The surface morphologies of the TiO2 and Ba/TiO2 nanostructures were subsequently examined by SEM as shown in Figure 5. Both TiO2 (5a) and Ba/TiO2 (5b) exhibited similar agglomerated spherical shape particles. In terms of chemical composition, TiO2 nanostructures contained Ti and O, only, which confirmed that the material was free from impurities, as observed from Figure 5c. The presence of Ba was confirmed in the Ba/TiO2 nanostructures from Figure 5d.

Electrochemical Behavior of TiO 2 and Ba/TiO 2
The reaction kinetics linked with DSSCs were investigated using EIS results. EIS spectra of TiO 2 and Ba/TiO 2 photoanodes under dark conditions with 0.6 V open circuit voltage are depicted in Figure 6. The EIS of DSSC was explained by three semicircles in the high, middle, and low frequency range, corresponding to the electrochemical output in platinized counter-electrode and electrolyte-interface (R 1 ), WE/electrolyte (R 2 ), and diffusion resistance of electrolyte (R d ), respectively. The displacement of the graph from origin represents the series resistance due to the contact resistance of the cell (R s ) [55]. Associated charge transfer in the electrolyte influenced the low frequency region, which was evident from the third semicircle. From the fitted data, the Ba/TiO 2 photoanode charge transfer resistance value was higher than the TiO 2 photoanode charge transfer resistance value. Barium doping increased the band gap value (which was also confirmed by optical analysis), and this reduces electron recombination in the Ba/TiO 2 /electrolyte interface. This behavior is attributed to the charge transfer value increment. The increment of charge transfer resistance was helpful to reduce charge recombination, which was evident for the photocurrent increment.
was evident from the third semicircle. From the fitted data, the Ba/TiO2 photoanode charge transfer resistance value was higher than the TiO2 photoanode charge transfer resistance value. Barium doping increased the band gap value (which was also confirmed by optical analysis), and this reduces electron recombination in the Ba/TiO2/electrolyte interface. This behavior is attributed to the charge transfer value increment. The increment of charge transfer resistance was helpful to reduce charge recombination, which was evident for the photocurrent increment.

Photovoltaic Studies
The photovoltaic performance of the fabricated DSSC is shown in Figure 7. Photocurrent density of Ba/TiO2 was higher than that of the TiO2 photoanode. This increment was due to the increased photon absorption of dye molecule as compared to the TiO2 counterpart. The Ba/TiO2 photoanode had higher surface area than the TiO2 photoanode, which led to enhanced dye-loading capacity. These results were also confirmed by the XRD and UV analysis. Another important factor was that the increment in photocurrent density was supported by the recombination rate or dark current reduction. The Ba/TiO2 photoanode had the high recombination resistance compared to the TiO2 photoanode, which reduce dark current production between the MOS and electrolyte.

Photovoltaic Studies
The photovoltaic performance of the fabricated DSSC is shown in Figure 7. Photocurrent density of Ba/TiO 2 was higher than that of the TiO 2 photoanode. This increment was due to the increased photon absorption of dye molecule as compared to the TiO 2 counterpart. The Ba/TiO 2 photoanode had higher surface area than the TiO 2 photoanode, which led to enhanced dye-loading capacity. These results were also confirmed by the XRD and UV analysis. Another important factor was that the increment in photocurrent density was supported by the recombination rate or dark current reduction. The Ba/TiO 2 photoanode had the high recombination resistance compared to the TiO 2 photoanode, which reduce dark current production between the MOS and electrolyte. The Ba/TiO2 photoanode yielded higher PCE (3.4%) compared to the TiO2 photoanode (2.1%). The calculated photovoltaic parameters are summarized in Table 3. The higher PCE of the Ba/TiO2 photoanode (3.4%) corresponds to a maximum output power density (Pmax) of approximately 34.4 μW/cm 2 as compared to the TiO2 photoanode's Pmax of 26.5 μW/cm 2 . These findings suggest that DSSCs can be produced with the appropriate The Ba/TiO 2 photoanode yielded higher PCE (3.4%) compared to the TiO 2 photoanode (2.1%). The calculated photovoltaic parameters are summarized in Table 3. The higher PCE of the Ba/TiO 2 photoanode (3.4%) corresponds to a maximum output power density (P max ) of approximately 34.4 µW/cm 2 as compared to the TiO 2 photoanode's P max of 26.5 µW/cm 2 . These findings suggest that DSSCs can be produced with the appropriate power rating to operate electronic devices such as tablets and Internet-of-Things gadgets that require low power P max < 100 µW/cm 2 [56].

Conclusions
Incorporation of barium (0.5 wt%) into TiO 2 was successfully accomplished using a microwave irradiation method. With respect to Ba content, its incorporation modulated the band gap of TiO 2 by decreasing particle size, leads to increased specific surface area and consequently, superior dye uptake and high photocurrent, resulting in higher DSSC power conversion efficiency. The higher band gap value of Ba/TiO 2 (3.26 eV) as compared to TiO 2 photoanode (3.21 eV) was attributable to the quantum confinement effect, which was observed due to the reduced size of the as-synthesized samples. EDS results confirmed the presence of barium in Ba/TiO 2 without any impurities; this was further confirmed by XRD results. A higher efficiency of Ba/TiO 2 photoanode was observed as compared to pure TiO 2 (3.4 vs 2.1%). Barium doped titania (Ba/TiO 2 ) could be employed in DSSCs owing to its relevant photoactive properties, with further prospects for other applications including photocatalysis as well as wastewater treatment.