Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Contents lists available at ScienceDirect

Optik
journal homepage: www.elsevier.com/locate/ijleo

Original research article

TiO2-based dye-sensitized solar cells prepared with bixin and

T
norbixin natural dyes: Effect of 2,2’-bipyridine additive on the
current and voltage
I.C. Carvalhoa, M.L. Barbosaa, M.J.S. Costaa, E. Longob, L.S. Cavalcantea,*,
V.G.F. Vianac, R.S. Santosa
a
PPGQ-GERATEC-DQ-Universidade Estadual do Piauí, Rua: João Cabral, N. 2231, P.O. Box 381, 64002-150, Teresina, PI, Brazil
b
CDMF - Universidade Estadual Paulista, P.O. Box 355, Araraquara, 14801-907, SP, Brazil
c
Programa de pós-graduação em Engenharia de Materiais PPGEM, IFPI campus central, Teresina, Piauí, Brazil

A R T IC LE I N F O ABS TRA CT

Keywords: In this paper, we discovered that the 2,2′-bipyridine (bipy) was able to increase the voltage and
Bixin short circuit current of dye-sensitized solar cells (DSSCs) prepared using bixin and norbixin
Norbixin natural dyes that are environmentally friendly. Titanium oxide (TiO2) film was deposited on FTO-
TiO2 nanoparticles-based DSSC glass. Morphological analysis revealed that the film was porous and consisted of almost spherical
Additive electrolyte
particle with an average size of 35 nm. The photoelectric behavior of the DSSCs was evaluated
under polychromatic radiation adjusted to 1.5 AM standard. The analysis revealed that the
presence of bipy in the electrolyte significantly increased the open circuit voltage (VOC) values
from 299 to 519 mV for the DSSCs prepared using bixin dye, whereas the VOC values increased
from 293 to 311 mV for the DSSCs prepared using norbixin dye. Also, it was observed that the
bipy additive positively influenced the short-circuit current (JSC) values of these devices. The
simultaneous increase in VOC and JSC rendered this bipyridinic compound a good additive can-
didate for TiO2 nanoparticles-based DSSC using natural dyes.

1. Introduction

In recent years, several studies indicate a growing demand for clean and renewable energy has guided the search for new
alternative energy sources. The use of fossil fuels as one of the main means of generating energy can cause significant damage to the
environment and human health. In addition, fossil fuels are nonrenewable, which suggests that they will become scarce in the future
[1]. Many researchers studied alternative, renewable energy sources, such as solar, wind, and geothermal energy [2]. In this context,
photovoltaic devices are great alternatives for converting sunlight into electrical energy. The most used solar cells rely on crystalline
or polycrystalline silicon, which can reach a useful life of approximately 20 years [3]. However, silicon cells are still expensive and
require relatively high cost investments for their initial installation. Dye-sensitized solar cells (DSSCs), which are considered among
the third generation of photovoltaic cells, have been presented as low-cost alternatives for the production of electrical energy [4,5].
O'Regan and Grätzel, first introduced DSSCs as cells consisting of a TiO2 nanostructured semiconductor electrode, which featured
Ru bipyridyl dye-based complexes adsorbed on its surface, redox electrolyte, and Pt counter electrode [6]. The electrolyte generally
consists of the iodide/triiodide (I−/ I−3 ) pair and the counter electrode are formed by depositing a Pt layer on the semiconductor glass.

⁎
Corresponding author.
E-mail address: [email protected] (L.S. Cavalcante).

https://doi.org/10.1016/j.ijleo.2020.165236
Received 25 April 2020; Received in revised form 24 June 2020; Accepted 11 July 2020
Available online 15 July 2020
0030-4026/ © 2020 Elsevier GmbH. All rights reserved.
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Fig. 1. 3D chemical structure of (a) cis-Bx and (b) cis-NBx molecules.

The dye molecules adsorbed on the mesoporous oxide surface are excited, and then electrons are injected into the conduction band
(CB) of the semiconductor. The injected electrons travel through the oxide layer through the external circuit toward the conducting
substrate and then reach the counter electrode [7]. Lastly, I− is converted into I−3 to regenerate the dye, which allows the oxidation
process to be restarted [8].
Dyes used as sensitizers for solar cells are usually organometallic complexes that contain Ru, and are reported to exhibit high light
absorption capacity and stability [9]. Also, aldimine derivative were investigated as used as photosensitizers for DSSC, reaching an
efficiency of 0.575 %. [10] In addition, studies have demonstrated that DSSCs can achieve efficiencies higher than 10 %. Natural dyes
extracted from vegetable species have been investigated as economical and non-toxic options for preparing DSSCs. Natural dyes or
pigments can be obtained from fruits, seeds, leaves, roots, and flowers [11–15]. Recently, Orona-Navar et al. reported the use of
Haematococcus pluvialis pigments as sensitizers for solar cells, which presented the efficiency of 0.1 % [16,17]. Also, Batniji et al.
investigated natural dye extracted from Trigonella seeds as sensitizer for DSSCs, which displayed efficiency of ca. 021 % [18].
Annatto (Bixa orellana) fruit seeds are rich in red pigment, which consists mainly of two carotenoid molecules: (a) Bixin (Bx) and
(b) norbixin (NBx) [19]. Fig. 1 presents the molecular structures of these pigments.
Bixa orellana is a tropical shrub native to the American continent, and its inedible fruit contains approximately 50 red seeds
[20,21]. The seeds can be extracted from the fruit and crushed; and, the obtained powder is commonly used as pigment [22]. Annatto
is also widely used for staining food, perfumes, textile products, varnishes, cosmetics, tattoos, and for medicinal purposes [16]. The
molar absorptivity coefficients of Bx and NBx pigments are high: 1.9 × 105 and 1.4 × 104 L mol−1 cm−1, respectively [23], and
guarantee that the pigments exhibit large photon capture capacities for generating electricity. However, compared to DSSCs prepared
using inorganic complexes, those obtained using natural pigments exhibit lower efficiencies and shorter shelf lives.
Studies have indicated that DSSCs prepared using carotenoid pigments presented low efficiency (0.37 and 0.17 % for Bx and NBx,
respectively), despite their high absorptivity coefficient values [11,24]. This suggested that photon absorptivity was not the only
variable that could have affected the efficiency of DSSCs prepared using natural pigments, and other factors such as the semi-
conductor oxide, type of counter electrode, and electrolyte composition should be considered. In this study, it was investigated the
additive influence present in the electrolyte. Kim et al. showed that pyridine-based additives can increase the efficiency of DSSCs by
positively influencing the open circuit voltage (VOC) value [25]. The effect of 4-tert-butylpyridine (4TBP) as additive on VOC has been
long debated in the literature, both for Ru dye- and natural pigments-sensitized cells. Also, Nguyen et al. observed an enhancement in
photocurrent values for DSSC with Ru complexes prepared in presence of 2,2′-bipyridine (bipy) as new electrolyte additive [26].

2
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

However, no reports have been published yet of the use of bipy in DSSC sensitized using natural pigments such as Bx and NBx.
Therefore, the objective of this study was to evaluate the effect of bipy as additive on the voltage and current density of cells
sensitized using Bx and NBx pigments. Both Bx- and NBx-sensitized cells were prepared, and their current and voltage values before
and after the addition of bipy to the electrolyte were compared. In addition, experiments were conducted to evaluate the constituents
of DSSCs (semiconductor and electrolyte), including X-ray diffraction (XRD) analysis, field emission-scanning electron microscopy
(FE-SEM) investigations and electrochemical characterization.

2. Experimental

2.1. Synthesis of mesoporous TiO2 films and preparation of DSSC

The reagents used for this study were of pure analytical grade, and were used without further purification. Anatase phase TiO2
was prepared following the methodology already reported by our research group [27]. In brief, 30 mL deionized H2O (Mil-Q) was
added to a round-bottomed flask and the pH was adjusted to 1.2 by adding 20 % v/v HNO3 (Dynamics). Then, 2.5 mL titanium (IV)
isopropoxide was added under magnetic stirring.
The flask containing the suspension was pipetized at 86 °C for 12 h, to deagglomerate the particles. The resulting colloidal
suspension was placed into a stainless-steel autoclave featuring an internal Teflon vessel. This system was maintained at 200 °C for 8
h. After being cooled at room temperature, oxides particles were collected and stored. Subsequently 2.44 g suspension, 0.895 g
polyethylene glycol with average molecular weight of 20,000 (Mw 20,000, Sigma Aldrich), and 30 μL Triton-X (40 wt.% oxide,
Vertec) were mixed, and the mixture was stirred for 12 h. To obtain the TiO2 photoanode and Pt electrodes, fluorine-doped tin oxide
(FTO) glass (TCO22-7, Sigma Aldrich, ∼7 Ω/sq) was cut into 1.5 × 2.5 cm samples. Subsequently, the glass was cleaned in an
ultrasonic bath in three 15 min successive stages, using liquid detergent and water, deionized water, and lastly, isopropyl alcohol. The
TiO2 photoanode was prepared by delimiting the geometric area of 0.28 cm2 on the surface of the FTO glass using adhesive tape. A
volumetric pipette was used to add 30 μL TiO2 suspension onto the delimited area, and a glass rod was used to spread the suspension
(doctor-blade method). After drying, the photoanode was placed in a furnace for sintering at two temperatures: 400 and 500 °C for 30
min each, using the heating rate of 1 °C min−1. The mass density of the deposited TiO2 oxide was 1.6 ± 0.3 mg cm−2.
The TiO2 structure was investigated using a LabX XRD-6000 diffractometer (Shimadzu, Japan) equipped with Cu Kα radiation (λ
= 0.15406 nm) in the 2θ range from 10° to 70° at a scanning rate of 1°/min. The diffraction patterns were compared with the data
from the Joint Committee on Powder Diffraction Standards (JCPDS) database. Also, the film morphology was evaluated using a field-
emission scanning electron microscope (FE-SEM, FEI quanta FEG 250). Optical properties of the films were studied using a UV-Vis
spectrophotometer (Shimadzu, model UV-2600). From this data, in the transmittance mode, the optical band gap energy, EBG was
determined using the Wood-Tauc method [28,29].
The Bx and NBx were extracted following methodology described early [22]. The sensitization solution was obtained using Bx and
NBx pigments as follows: 0.197 g Bx and 10 mL chloroform (CHCl3, 99.5 % purity, Synth) were added to a 100 mL beaker, and the
system was ultrasonicated for 30 min. The mixture was transferred to a 25 mL volumetric flask where sufficient CHCl3 was added to
obtain a 2 × 10−2 mol L−1 solution. The same procedure was followed for NBx, however 0.190 g of NBx were used to obtain the
same final concentration.
The counter electrode was prepared by adding 10 μL of 15 mmol L−1 hexachloroplatinic acid solution onto the conductive surface
of the FTO-glass. Then, the film was placed into a muffle where it was heated at 400 °C for 30 min at the heating rate of 1 °C
min−1.The electrode was then washed using detergent and isopropanol, and was further activated at 400 °C for 30 min in a muffle,
before its use.
To prepare the electrolyte, 0.25 g iodine (I2) and 1.596 g lithium iodide (LiI) were added to a beaker that contained 10 mL
acetonitrile. The system was ultrasonicated for 1 h. Afterward, the mixture was transferred to a 25 mL volumetric flask, and acet-
onitrile was added to it to form a solution that contained 0.04 mol L−1 I2 and 0.48 mol L−1 LiI. To obtain a 0.02 mol L−1 bipy
solution, 0.03 g bipy was added to 10 mL electrolyte. The next step was the assembly of the solar cell. After the photoanode
preparation, adhesive tape was again used to delimit the TiO2 film area. A volumetric pipette was used to add 30 μL sensitizer
solution onto the delimited TiO2 area for the sensitization process. After the sensitizer solution dried, the electrolyte was added onto
the same area, the Pt counter electrode was placed onto the photoelectrode, and then, the system was held in sandwich format.

2.2. Photoelectrochemical analysis of electrolytes and cell

The flat band potential (Efb) of the semiconductor was determined using linear scanning voltammetry utilizing a three-electrode
electrochemical cell configuration, and following the methodology reported earlier in the literature [30]. The work electrode con-
sisted of the TiO2 film featuring the active area of 1 cm2, and Pt wire and Ag/AgCl were used as counter and reference electrodes,
respectively. In addition, 0.100 mol L−1 Na2SO4 solution (pH ≈ 6.20) was used as supporting electrolyte. For comparison, all
potential values obtained using the Ag/AgCl reference electrode were converted into reversible hydrogen electrode (RHE) potentials
and electron volt values, using Eqs. (1) and (2), respectively.

E (vs RHE) = E (vs Ag/AgCl) + 0.0591 V × pH + 0.199 V (1)

E(eV) = − 4.5 eV − e ERHE(V) (2)

3
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Fig. 2. XRD patterns of thermally treated TiO2 deposited on FTO glass.

where e is the electron charge and ERHE is the potential with respect to the reversible hydrogen electrode.
Electrochemical studies were performed using a potentiostat/galvanostat (PGSTAT302 N Metrohm Autolab) coupled with the
NOVA 1.7 software. Cathodic linear voltammetry at the scan rate of 2 mVs−1 was used to characterize the DSSCs. The cells were
analyzed under irradiation from a metal vapor lamp (HQI-TS, 150 W NDL) set at 100 mW cm−2 (1.5 AM). It was possible to
determine the shot-circuit current (JSC), VOC, fill factor (FF), maximum power (Pmax), and efficiency (η) of the cells using experi-
mental data.

3. Results and discussion

3.1. Structural analysis of TiO2 film

The crystallinity of the TiO2 sample prepared on the FTO glass substrate was analyzed using XRD measurements, as illustrated in
Fig. 2.
Fig. 2 presents the XRD pattern of TiO2 nanoparticles calcined at 400 °C. Considering the information from the JCPDS Card No.
21-1272, the presence of the (101) and (200) planes in the XRD pattern of the sample revealed that anatase was the majority phase.
Brookite phase signals were also recorded, considering that the signal attributed to the (221) plane was ascribed to brookite [31,32].
The part of the spectrum that is highlighted in green, corresponds to the crystalline phase of SnO2:F present in the substrate (FTO-
glass) [33,34].
Most studies on TiO2-based DSSCs used anatase TiO2 [35]. However, studies on brookite TiO2-based DSSCs have also been
reported, and the results indicated that this metastable TiO2 phase would also be satisfactory for the preparation of DSSCs [36,37].
Therefore, the synthesized TiO2 could be used for DSSCs, even if it presented both brookite and anatase phases.

3.2. Optical characterization of pigment solutions and TiO2 film

The optical band gap energy (EBG) of the heated TiO2 film was estimated using the Tauc plot method, assuming direct forbidden
transitions, according to Eq. (3):

(αhν)2 = C(hν― EBG) (3)

where hν is the incident photon radiation, α is the absorption coefficient of the film, C is a proportionality constant, and EBG is the
band gap energy.
The EBG value was obtained by extrapolating the linear part of the (αhν)2 vs hv graph for αhν = 0, as depicted in Fig. 3.
As illustrated in Fig. 3, EBG of TiO2 was estimated to be 3.24 eV. This result was in good agreement with the data reported in the
literature [38]. Optical analysis of the Bx and NBx pigment solutions (Fig. 4) revealed two well-defined absorption regions featuring
the maximum wavelengths of approximately 470 and 502 nm, respectively. The absorbance of Bx was higher than that of NBx at the
same concentration, which suggested that the absorption capacity of Bx was superior to that of NBx [11]. Using the calibration curve,
the molar absorptivity coefficients of Bx with wavelength of maximum absorption in 471 nm and NBx in 470 nm were estimated to be
6.1 × 104 and 5.3 × 104 L mol−1 cm −1, considering that pigment solutions were prepared using CHCl3 (See Inset in Fig. 4).
The UV-Vis transmittance spectra of films with Bx and NBx (inset Fig. 4) indicated that oxide sensitization extended their ab-
sorption regions of the spectrum. For the 600−450 nm range, the photon absorption of Bx was higher than that of NBx, and this was
attributed to the molar absorptivity coefficient of Bx being higher than that of NBx. This result was in concordance with registered
before for pigment solution absorbance. Also, the above-mentioned maximum wavelengths almost disappeared. Moreover, the results
suggested that the pigment interacted effectively with the TiO2 nanoparticles, and therefore, the positions of its highest occupied and
lowest unoccupied molecular orbital (HOMO and LUMO, respectively) energy levels were altered, which allowed the orbitals of the

4
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Fig. 3. Tauc plot method for determining EBG of 450 °C-annealed TiO2 film.

Fig. 4. UV–vis absorption spectra of 1 × 10−5 mol L−1 CHCl3 solutions of Bx and NBx.

pigments to overlap with the CB of the TiO2 nanoparticles [39]. Thus, in DSSCs, the electrons generated when the pigment was
excited using solar radiation could be injected into the CB of the semiconductor [40].

3.3. Morphological characterization of TiO2 films

The TiO2 films were analyzed using FE-SEM to evaluate their morphological properties, porosity, and average particle size.
Fig. 5(a) and (b) illustrate the FE-SEM images and the inset with superior view of TiO2 films.
The FE-SEM images in Fig. 5 confirmed that the particles were slightly spherical and presented irregular morphology owing to
their agglomeration. It was also observed that the obtained film was mesoporous and the coalescence of the nanoparticles hindered
the estimation of both their diameters and pore sizes. The analysis revealed that the crystallite diameters ranged from 24.76 to 45.81

Fig. 5. Surface morphology of TiO2 film prepared on FTO glass after thermal treatment at 450 °C: (a) high magnification FE-SEM images and (b) FE-
SEM images at surface of TiO2 film.

5
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Fig. 6. (a) Linear voltammetry of TiO2 film in 0.1 mol L−1 Na2SO4 solution at pH ≈ 6.20. (b) Energy levels diagram of cell constituents and cyclic
voltammetry data of electrolyte in inset.

nm and the average crystallite diameter of the sample was 35.14 nm.

3.4. Photoelectrochemical properties of solar cells

Perfect synchronisms between the various elements comprising solar cells are required for the cells to work properly. Therefore,
an adequate relationship should exist between the energy levels of the oxide, dye, and electrolyte. The relative positions of these
energy levels are responsible for generating the driving force for electron injection in the semiconductor and dye regeneration by the
electrolyte [41]. The relative energy levels positions of TiO2 and electrolyte used in this work were estimated following the meth-
odology presented in previously published papers [23,41,42], and the results are displayed in Fig. 6.
Fig. 6(a) illustrates the chopped dark-light linear voltammetry curve of the TiO2 film. From this curve it was possible to estimate
the flat band potential (Efb) of TiO2 using the Gärtner–Butler model which assumes that the electric field within the depletion region
is not influenced by irradiation. In this model, the Efb value is that one where the first photocurrent signs appear.
[16]. For n-type semiconductors, the energy of the CB (ECB) lies close to the Fermi level (EF), and the relationship between Efb and
CB is defined by Eq. (4):

k T ⎛ NC ⎞
ECB = E fb − ln
⎜ ⎟

e ⎝ ND ⎠ (4)

where k is the Boltzmann constant, T is the absolute temperature, e is the electron charge, NC is the effective density of states in the
CB, and ND is the effective charge density [43,44]. Considering that the values of ND and NC are similar, the second term in Eq. (4)
should be relatively small and then ECB can be approximated to Efb [45,46]. In the setting presented in Fig. 6(b) inserted, the
voltammogram indicates that an increase in photocurrent values occurred under irradiation, owing to the electron injection into the
CB. However, when the light was turned off, an abrupt reduction in photocurrent was registered. This behavior is typically observed
for n-type semiconductors during anodic scans. As illustrated in Fig. 6(a), Efb was estimated to be −0.172 V (vs Ag/AgCl), which
corresponded to 4.89 eV
The optical energy transitions of Bx and NBx pigments were estimated using UV-Vis spectra and theoretical density functional
theory (DFT) data, which have been previously reported in the literature (functional MO5−2X and base set 6-31 + g (d,p)) [47].
Using this information, it was possible to determine the HOMO and LUMO energy levels of the pigments. Also, UV-Vis spectra and
cyclic voltammetry analysis were utilized to calculate the energy level values of the I−/ I−3 electrolyte utilized in this study (See the
inserting in Fig. 6(c)). All energy levels in the diagram in Fig. 6(c) are expressed in electron volt.
The process of converting photons of solar radiation into electric current was favored for both pigments, since the LUMO levels of
the pigments were more positive than the CB of the semiconductor, while the HOMO levels were more negative than the redox
potential of the electrolyte. Thus, the electron injection force into the CB of the semiconductor (Table 1) was higher for the Bx-

6
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Table 1
Electrons injection and regeneration forces of pigments based on energy diagram.
Strength of: Bixin Norbixin

Electron Injection (eV) 1.15 0.82

Electron Regeneration (eV) 0.88 1.24

sensitized cell (1.15 eV) compared to the NBx-sensitized cell one, while the regeneration force of the NBx-sensitized cell was higher
(1.24 eV) than the Bx-sensitized cell. The electron injection and regeneration capabilities influenced the cell parameters. The DSSC
featuring the higher electron injection force at the TiO2/CB interface would generate a higher current density, whereas the DSSC
presenting the higher regeneration force would exhibit greater pigment lifetime and the effect of recombination would be minimized.
The electrochemical cells were characterized using the JSC vs VOC curves, under a polychromatic light source adjusted for 100 mW
cm−2 (1.5 AM standard). Eqs. (5,6) were used to calculate FF and η, respectively:
Jmax × Vmax
FF =
JSC × VOC (5)

JSC × FF × VOC
η=
Iinc (6)
where Jmax is the maximum current density and Vmax is the maximum voltage [48]. In addition, the performance of natural pigments
as sensitizers in DSSCs was determined considering other parameters, such as JSC, VOC, and Pmax (maximum power).
Fig. 7(a) and (b) display the photoelectric analysis of the Bx- and NBx-sensitized cells, respectively. The cells were analyzed when
bipy was used as additive as well as in the absence of this additive.
From Fig. 7 it was possible to verify that the bipy additive increased JSC and VOC of the DSSCs. The initial JSC of the Bx-sensitized
cell was at least four times higher and its VOC value was almost twice as big for the DSSCs where bipy was used as additive. The

Fig. 7. JSC in function VOC curves of (a) Bx- and (b) NBx-sensitized cells.

7
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Table 2
DSSC parameters (JSC, VOC, FF, PMAX, η) extracted from the graphs displayed in Fig. 7.
Parameters BX without Byp BX with Byp NBX without Byp NBX with Byp

JSC (mA) 0.057 0.195 0.011 0.125

VOC (mV) 299 519 305 432
FF 0.398 0.700 0.293 0.311
PMAX (μW) 6.76 71 0.98 16.8
η (%) 0.006 0.071 0.001 0.017

increase in JSC occurred regardless of the use of the additive. This can be explained in part by the higher absorptivity coefficient of Bx
compared to NBx [49], which allowed Bx to capture more photons. Therefore, the probability of converting those photons into
electricity would be higher for Bx than for NBx. Moreover, the electron injection force into the CB of TiO2of Bx was higher than that
of NBx, as already observed in the energy levels diagram (Fig. 6). However, the potential analysis indicated that the difference in
voltage between the Bx- (299 mV) and NBx-sensitized cells (305 mV) was very small when no bipy was added to the electrolyte.
The most significant difference in voltage between the Bx-(519 mV) and NBx-sensitized cells (432 mV) was observed in the
presence of the additive. Analyzing the cells sensitized using the same pigments (Bx with and without additive and NBx with and
without additive), the current and voltage gains of the devices using the bipy additive were more pronounced. This increase was most
significant, again, for the Bx-sensitized cell, (Table 2).
The results extracted from the photoelectric analysis in this study indicated that the bipy additive, when added to the electrolyte,
optimized the performance of the DSSCs. This was related to the phenomena that occurred at the TiO2/pigment/electrolyte interfaces
[50], since the voltage and current of the cells depend on the relative positions of the energy levels of each component. Significant
improvement in the performance of DSSCs can be achieved by adding certain compounds to the electrolyte. The most frequently used
additive was 4TBP [51], which significantly improved the VOC of DSSCs, while JSC was not significantly affected or slightly decreased
[52]. The VOC values of DSSCs depend on the position of the EF of the semiconductor in relationship to the potential of the electrolyte,
and the difference between the two potentials is the maximum voltage supplied by the cell. The elevation of the CB of the semi-
conductor allowed it to be approximated with the LUMO level of the pigment. This approach would reduce the injection force of the
electrons into the conduction band by the pigment, and would decrease the current density [41].
Since the bipy and 4TBP molecules present structural and chemical similarities, we were able to determine the increase in voltage
of the Bx- and NBx-sensitized cells in this study. However, the basic character of bipy is lower than 4TBP, owing to the low avail-
ability of the electron pairs of nitrogen, and therefore, the interaction between bipy and TiO2in the CB of TiO2 was less intense than
that between 4TBP and TiO2. Hence, the DSSCs using bipy as additive exhibited smaller VOC variations than those using 4TBP as
additive [44]. On the other hand, the larger number of free electron pairs (even not available) of bipy led to the conclusion that bipy
interacted with more I−3 triiodide ions than 4TBP. In addition, the use of bipy in the Bx- and NBx-sensitized cells caused their current
values to significantly increase. The current increase was not predicted since pyridine-based compounds would approximate the CB of
the semiconductor to the LUMO of the pigment, and therefore, would decrease the injection force of electrons. Therefore, the additive
in this study was not limited to interacting only with the semiconductor, it could have also interacted with the electrolyte and/or
pigment, which would justify the increase in current density.

4. Conclusion

In summary, TiO2 nanoparticles-based solar cells were sensitized using Bx and NBx natural dyes and their performance was
analyzed in the presence and absence of bipy additive in the electrolyte. The results demonstrated that the bipy additive promoted the
significant increase in the voltage of the cells. This effect was more pronounced for the Bx-sensitized cell, as its potential increased
from 299 to 519 mV. In addition, the current density of the cells also increased owing to the additive. Pyridine-based additives
increased the CB of the semiconductor by bringing it closer to the LUMO of the natural pigments; and therefore, the current density
would either stay constant or decrease owing to the reduction in the electron injection force. This could be explained by the decrease
in the recombination effect. The literature states that the 4TBP additive (which is similar to bipy) interacts with I−3 triiodide in the
electrolytic solution and prevents its access to the surface of TiO2 nanoparticles by reducing the recombination effect. Assuming that
bipy, owing to its chemical structure, is able to interact with more I−3 ions than 4TBP, this would lead to a much smaller re-
combination effect, and therefore, the current would be much larger. It was concluded that the additive in this study could be very
promising because it simultaneously promoted the increase in voltage and current density of the analyzed cells.

Declaration of Competing Interest

The authors declare that the article is original, has been written by the stated authors who are all aware of its content and approve
its submission. The paper has not been published previously and not under consideration for publication elsewhere.

8
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

Acknowledgements

The authors would like to thank to the CAPES, CNPq (312318/2017-0 and 408036/2018-4) and personnel at the Laboratory of
Advanced Materials (LIMAV) at Federal University of Piauí for their help with the FE-SEM measurements.

References

[1] J. Wu, Z. Lan, J. Lin, M. Huang, Y. Huang, L. Fan, G. Luo, Electrolytes in dye-sensitized solar cells, Chem. Rev. 115 (2015) 2136–2173.
[2] G. Richhariya, A. Kumar, P. Tekasaku, B. Gupta, Natural dyes for dye sensitized solar cell: a review, Renew. Sustain. Energy Rev. 69 (2017) 705–718.
[3] M. Zarmai, N.N. Ekere, C.F. Oduoza, E.H. Amalu, A review of interconnection technologies for improved crystalline silicon solar cell photovoltaic module
assembly, Appl. Energy 154 (2015) 173–182.
[4] T. McDonnell, S. Korsmeyer, Nature publishing group, Nature 354 (1991) 56–58.
[5] P.P. Kumavat, P. Sonar, D.S. Dalal, An overview on basics of organic and dye sensitized solar cells, their mechanism and recent improvements, Renew. Sustain.
Energy Rev. 78 (2017) 1262–1287.
[6] M. Zannotti, C.J. Wood, G.H. Summers, L.A. Stevens, M.R. Hall, C.E. Snape, R. Giovannetti, E.A. Gibson, Ni Mg mixed metal oxides for p-type dye-sensitized solar
cells, ACS Appl. Mater. Interfaces 7 (2015) 24556–24565.
[7] S. Shalini, R. Balasundara Prabhu, S. Prasanna, T.K. Mallick, S. Senthilarasu, Review on natural dye sensitized solar cells: operation, materials and methods,
Renew. Sustain. Energy Rev. 51 (2015) 1306–1325.
[8] S. Thomas, T.G. Deepak, G.S. Anjusree, T.A. Arun, S.V. Nair, A.S. Nair, A review on counter electrode materials in dye-sensitized solar cells, J. Mater. Chem. A 2
(2014) 4474–4490.
[9] Z.-S. Wu, X.-C. Song, Y.-D. Liu, J. Zhang, H.-S. Wang, Z.-J. Chen, S. Liu, Q. Weng, Z.-W. An, W.-J. Guo, New organic dyes with varied arylamine donors as
effective co-sensitizers for ruthenium complex N719 in dye sensitized solar cells, J. Power Sources 451 (2020) 227776.
[10] A.Y. Batniji, R. Morjan, M.S. Abdel-Latif, T.M. El-Agez, S.A. Taya, H.S. El-Ghamri, Aldimine derivatives as photosensitizers for dye-sensitized solar cells, Turk. J.
Phys. 38 (2014) 86–90.
[11] D.J. Godibo, S.T. Anshebo, T.Y. Anshebo, Dye sensitized solar cells using natural pigments from five plants and quasi-solid state electrolyte, J. Braz. Chem. Soc.
26 (2015) 92–101.
[12] M.S. Abdel-Latif, M.B. Abuiriban, T.M. El-Agez, S.A. Taya, Dye-sensitized solar cells using dyes extracted from flowers, leaves, parks, and roots of three trees, Int.
J. Renew. Energy Res. 1 (2015) 294–298.
[13] H.S. El-Ghamri, T.M. El-Agez, S.A. Taya, M.S. Abdel-Latif, A.Y. Batniji, Dye-sensitized solar cells with natural dyes extracted from plant seeds, Mater. Sci. 32
(2014) 547–554.
[14] S.A. Taya, T.M. El-Agez, M.S. Abdel-Latif, H.S. El-Ghamri, A.Y. Batniji, I.R. El-Sheikh, Fabrication of dye-sensitized solar cells using dried plant leaves, Int. J.
Renew. Energy Res. 4 (2014) 384–388.
[15] I.M. Radwana, S.A. Taya, T.M. El-Ageza, M.S. Abdel-Latifb, H.S. Ghamria, Improvement of the performance of purple carrot sensitized solar cells by acidic
treatment of FTO glass substrate and TiO2 film, Acta Phys. Pol. A 130 (2016) 795–799.
[16] A. Orona-Navar, I. Aguilar-Hernández, A. Cerdán-Pasarán, T. López-Luke, M. Rodríguez-Delgado, D.L. Cárdenas-Chávez, E. Cepeda-Pérez, N. Ornelas-Soto,
Astaxanthin from Haematococcus pluvialis as a natural photosensitizer for dye-sensitized solar cell, Algal Res. 26 (2017) 15–24.
[17] S.A. Taya, T.M. El-Agez, M.S. Abdel-Latif, H. Ghamri, A. Batniji, W.A. Tabaza, Dyes extracted from safflower, medicago sativa, and Ros Marinus oficinalis as
photosensitizers for dye-sensitized solar cells, J. Nano- Electron. Phys. 8 (2016) 01026.
[18] A. Batniji, M.S. Abdel-Latif, T.M. El-Agez, S.A. Taya, H. Ghamri, Dyes extracted from Trigonella seeds as photosensitizers for dye-sensitized solar cells, J. Theor.
Appl. Phys. 10 (2016) 265–270.
[19] T. Taham, F.A. Cabral, M.A.S. Barrozo, Extraction of bixin from annatto seeds using combined technologies, J. Supercrit. Fluids 100 (2015) 175–183.
[20] S.C. Alcázar-Alay, J.F. Osorio-Tobón, T. Forster-Carneiro, M.A.A. Meireles, Obtaining bixin from semi-defatted annatto seeds by a mechanical method and
solvent extraction: process integration and economic evaluation, Food Res. Int. 99 (2017) 393–402.
[21] N.M. Gómez-Ortíz, I.A. Vázquez-Maldonado, A.R. Pérez-Espadas, G.J. Mena-Rejón, J.A. Azamar-Barrios, G. Oskam, Dye-sensitized solar cells with natural dyes
extracted from achiote seeds, Sol. Energy Mater. Sol. Cells 94 (2010) 40–44 G.
[22] L.P. Fontinele, R.C. de Sousa, V.G.F. Viana, E.A. de O. Farias, E.L. Queiroz, C. Eiras, Norbixin extracted from urucum (Bixa orellana L.) for the formation of
conductive composites with potential applications in electrochemical sensors, Surf. Interfaces 13 (2018) 92–100.
[23] G. Calogero, J. Barichello, I. Citro, P. Mariani, L. Vesce, A. Bartolotta, A. Di Carlo, G. Di Marco, Photoelectrochemical and spectrophotometric studies on dye-
sensitized solar cells (DSCs) and stable modules (DSCMs) based on natural apocarotenoids pigments, Dyes Pigm. 155 (2018) 75–83.
[24] H. Hug, M. Bader, P. Mair, T. Glatzel, Biophotovoltaics: Natural pigments in dye-sensitized solar cells, Appl. Energy 115 (2014) 216–225.
[25] J.-Y. Kim, J.Y. Kim, D.-K. Lee, B. Kim, H. Kim, M.J. Ko, Importance of 4- tert-butylpyridine in electrolyte for dye-sensitized solar cells employing SnO2 electrode,
J. Phys. Chem. C 116 (2012) 22759–22766.
[26] P.T. Nguyen, T.A.P. Phan, N.H.T. Ngo, T.V. Huynh, T. Lund, 2,2′-Bipyridine – a new electrolyte additive in dye-sensitized solar cells, Solid State Ion. 314 (2018)
98–102.
[27] R.S. Santos, G.A. Faria, C. Giles, C.A.P. Leite, H.S. Barbosa, M.A.Z. Arruda, C. Longo, Iron insertion and hematite segregation on Fe-doped TiO2 nanoparticles
obtained from sol-gel and hydrothermal methods, ACS Appl. Mater. Interfaces 4 (2012) 5555–5561.
[28] B.D. Viezbicke, S. Patel, B.E. Davis, D.P. Birnie, Evaluation of the Tauc method for optical absorption edge determination: ZnO thin films as a model system,
Phys. Status Solidi B 252 (2015) 1700–1710.
[29] G.S. Costa, M.J.S. Costa, H.G. Oliveira, L.C.B. Lima, G.E. Luz Jr., L.S. Cavalcante, R.S. Santos, Effect of the applied potential condition on the photocatalytic
properties of Fe2O3|WO3 heterojunction films, J. Inorg. Organomet. Polym. Mater. 30 (2020) 2851–2862.
[30] M.J.S. Costa, G.S. Costa, A.E.B. Lima, G.E. Luz Jr., E. Longo, L.S. Cavalcante, R.S. Santos, Photocurrent response and progesterone degradation by employing
WO3 films modified with platinum and silver nanoparticles, Chempluschem 83 (2018) 1153–1161.
[31] R.S. Dubey, Temperature-dependent phase transformation of TiO2 nanoparticles synthesized by sol-gel method, Mater. Lett. 215 (2018) 312–317.
[32] M.V. Chittan, C.M. Kumar, B.R. Kumar, X-ray peak profile analysis and microstructural characterization of solid state sintered TiO2 doped ZnO ceramics, Mater.
Today Proc. 4 (2017) 2879–2886.
[33] E. Şilik, S. Pat, S. Özen, R. Mohammadigharehbagh, H.H. Yudar, C. Musaoğlu, Ş. Korkmaz, Electrochromic properties of TiO2 thin films grown by thermionic
vacuum arc method, Thin Solid Films 640 (2017) 27–32.
[34] J. Li, H. Zhang, W. Wang, Y. Qian, Z. Li, Improved performance of dye-sensitized solar cell based on TiO2 photoanode with FTO glass and film both treated by
TiCl4, Phys. B Condens. Matter. 500 (2016) 48–52.
[35] J. Gong, K. Sumathy, Q. Qiao, Z. Zhou, Review on dye-sensitized solar cells (DSSCs): advanced techniques and research trends, Renew. Sustain. Energy Rev. 68
(2017) 234–246.
[36] J. Xu, S. Wu, J. Jin, T. Peng, Preparation of brookite TiO2 nanoparticles with small sizes and the improved photovoltaic performance of brookite-based dye-
sensitized solar cells, Nanoscale 8 (2016) 18771–18781.
[37] C. Magne, S. Cassaignon, G. Lancel, T. Pauporté, Brookite TiO2 nanoparticle films for dye-sensitized solar cells, ChemPhysChem 12 (2011) 2461–2467.
[38] A.J. Haider, R.H. Al-Anbari, G.R. Kadhim, C.T. Salame, Exploring potential environmental applications of TiO2 nanoparticles, Energy Procedia 119 (2017)
332–345.
[39] N.A. Ludin, A.M.A.-A. Mahmoud, A.B. Mohamad, A.A.H. Kadhum, K. Sopian, N.S.A. Karim, Review on the development of natural dye photosensitizer for dye-

9
I.C. Carvalho, et al. Optik - International Journal for Light and Electron Optics 218 (2020) 165236

sensitized solar cells, Renew. Sustain. Energy Rev. 31 (2014) 386–396.

[40] G.G. Sonai, M.A. Melo Jr., J.H.B. Nunes, J.D. Megiatto Jr., A.F. Nogueira, Solar cells sensitized with natural dyes: an introductory experiment about solar energy
for undergraduate students, Química Nova 38 (2015) 1357–1365.
[41] H. Nan, H.-P. Shen, G. Wang, S.-D. Xie, G.-J. Yang, H. Lin, Studies on the optical and photoelectric properties of anthocyanin and chlorophyll as natural co-
sensitizers in dye sensitized solar cell, Opt. Mater. 73 (2017) 172–178.
[42] T.-D. Nguyen, Y.-P. Lan, C.-G. Wu, High-efficiency cycloruthenated sensitizers for dye-sensitized solar cells, Inorg. Chem. 57 (2018) 1527–1534.
[43] I.E. Paulauskas, J.E. Katz, G.E. Jellison Jr., N.S. Lewis, L.A. Boatner, G.M. Brown, Growth, characterization, and electrochemical properties of doped n-type KTaO
[sub 3] photoanodes, J. Electrochem. Soc. 156 (2009) B580–B587.
[44] A.E.B. Lima, M.J.S. Costa, R.S. Santos, N.C. Batista, L.S. Cavalcante, E. Longo, G.E. Luz Jr., Facile preparation of CuWO4 porous films and their photoelec-
trochemical properties, Electrochim. Acta 256 (2017) 139–145.
[45] P.K. Pandey, N.S. Bhave, R.B. Kharat, Spray deposition process of polycrystalline thin films of CuWO4 and study on its photovoltaic electrochemical properties,
Mater. Lett. 59 (2005) 3149–3155.
[46] B. Xu, Z. Zhu, J. Zhang, H. Liu, C.-C. Chueh, X. Li, A.K.-Y. Jen, 4-Tert-butylpyridine free organic hole transporting materials for stable and efficient planar
perovskite solar cells, Adv. Energy Mater. 7 (2017) 1–10.
[47] T. Ruiz-Anchondo, N. Flores-Holguín, D. Glossman-Mitnik, Natural carotenoids as nanomaterial precursors for molecular photovoltaics: a computational DFT
study, Molecules 15 (2010) 4490–4510.
[48] P. Gu, D. Yang, X. Zhu, H. Sun, J. Li, Fabrication and characterization of dye-sensitized solar cells based on natural plants, Chem. Phys. Lett. 693 (2018) 16–22.
[49] G. Calogero, A. Bartolotta, G.D. Marco, A.D. Carlo, F. Bonaccorso, Vegetable-based dye-sensitized solar cells, Chem. Soc. Rev. 44 (2015) 3244–3294.
[50] V. Gondane, P. Bhargava, Tuning flat band potential of TiO2 using an electrolyte additive to enhance open circuit voltage and minimize current loss in dye
sensitized solar cells, Electrochim. Acta 209 (2016) 293–298.
[51] J. Gao, A.M. El-Zohry, H. Trilaksana, E. Gabrielsson, V. Leandri, H. Ellis, L. D’Amario, M. Safdari, J.M. Gardner, G.G. Andersson, L. Kloo, Light-induced
interfacial dynamics dramatically improve the photocurrent in dye-sensitized solar cells: an electrolyte effect, ACS Appl. Mater. Interfaces 10 (2018)
26241–26247.
[52] L. Boschloo, A. Häggman, Hagfeldt, Quantification of the effect of 4-tert-butylpyridine addition to I−/I3- redox electrolytes in dye-sensitized nanostructured TiO2
solar cells, J. Phys. Chem. B 110 (2006) 13144–13150.

10