Quantitative framework development for understanding the relationship between doping and photoelectrochemical energy conversion of TiO₂

Abstract

Efficient energy harvesting devices are required to achieve sustainable development goals. Titania is a semiconductor that has attracted substantial research attention in pursuit of energy independence since it is an abundant, stable, toxin-free, and energy-prospective substance. Careful tailoring of morphological and electronic properties is required to address the poor conductivity of TiO₂ to achieve competitive Power Conversion Efficiency (PCE). Adding an optimum amount of impurity to TiO₂ is one of the prominent methods to increase its conductivity. This review critically analyzes doping of TiO₂ in light of the power conversion efficiencies of various electrochemical cells. This review suggests an alternative quantitative framework for developing and establishing the relationship between doping and photoelectrochemical energy conversion in TiO₂-based devices. It offers an alternative for accurate data reporting after a careful analysis of the PCE data in research articles and data reporting processes now in use. This study provides information on data analysis, visualization, and contemporary techniques for doping TiO₂.

---

Aparna Markose

Aparna Markose received her Master of Physics degree from the University of Calicut, Kerala, in 2014 and received a Master of Green Energy Technology from Pondicherry University in 2019 with Distinction. She is now a Research Scholar at Madanjeet School of Green Energy Technologies, Pondicherry University. Her research interests are designing and developing heterostructures, one-dimensional and zero-dimensional nano-systems for energy conversion.

Debanita Das

Debanita Das is pursuing her PhD from the Indian Institute of Technology Bombay (IITB) – Monash University Joint Research Academy. She is currently studying the impact of plasmonic structures on optical sensing-based applications. She is working under the joint supervision of Prof. Anshuman Kumar (IITB) and Prof. Alison Funston (Monash University). She had completed her Master's in Green Energy Technology from Pondicherry University and graduated as a gold medalist.

Prasanth Ravindran

Dr. Prasanth Ravindran is professor at Madanjeet School of Green Energy Technologies, Pondicherry University. He received his PhD from the Eindhoven University of Technology, Netherlands, for the thesis entitled “Photonic switching in III–V nanostructures.” He pursued his post-doctoral research at the Peter Debye Institute at Utrecht University. His current research interest is developing one-dimensional and zero-dimensional nano-systems for energy conversion.

---

1. Introduction

Titanium dioxide (titania, TiO₂) is a semiconductor with diverse applications in solar cells, catalysts, sunscreens, medicine, inks, wastewater treatment, water splitting, energy storage, and electrochemical cells^1–4 (Fig. 1). The properties of titanium dioxide such as its wide bandgap, charge injection, extraction-favored band edge alignments, long excited electron lifetime, abundance, chemical inertness, and resistance to photocorrosion make it an excellent material for photovoltaic applications. TiO₂ has three natural polymorphs: anatase, rutile, and brookite (Fig. 2). Anatase and rutile have the same tetragonal structure, while brookite is orthorhombic. Even though brookite is the most stable form, anatase exhibits low recombination and superior charge transfer properties.^5–8 Employing titanium dioxide for photoelectrochemical conversion became a milestone in the history of dye-sensitized solar cells (DSSCs).⁹ Until the discovery of Grätzel cells, DSSCs performed poorly due to their low photon absorption.¹⁰ Other wide bandgap semiconductors such as ZnO, SnO₂, and Nb₂O₅ have been used in DSSCs instead of TiO₂.^11,12 The abundance, stable and trouble-free production of mesoporous particles, non-toxicity, stability, and apposite band alignment with popular dye N719 made TiO₂ the most suitable electrode material for DSSCs.¹²

Fig. 1 Applications of titanium dioxide.

Fig. 2 Structures of TiO₂ natural polymorphs: (a) rutile, (b) anatase and (c) brookite.¹⁷

The limitations of other organic-based solar cells have been an ongoing focus of DSSC research due to the explosive growth of the perovskite solar cell industry over the past five years. Whereas the majority of PV technologies fail to succeed, humidity, oxygen conditions, and an easy production process are favorable for DSSCs.¹³ One significant benefit of DSSCs is their exceptional power production in all lighting circumstances, even indoor lighting. It has been found that under diffused or dim sunlight conditions silicon based solar cells do not perform well.^13–15 This requirement gap is satisfied by DSSCs in an efficient and competitive way. Indoor light harvesting using DSSCs even surpassed 30 percentage without co-sensitizers. For wireless sensor nodes, consumer electronics, wearable technology, and smart meters, DSSCs can be implemented indoors as portable electronic modules.¹⁶ The potential to scale up and that to minimize production costs are key game changers in DSSC research. Because of their potential for indoor lighting, DSSCs can maximize energy efficiency while minimizing their carbon footprint.

In DSSCs, the current is generated when a dye molecule absorbs a photon and injects an electron into the semiconductor's conduction band. The circuit is completed when the dye is degenerated by the electrolyte's redox species. This electrolyte is then reduced at the counter electrode, which is usually platinum. A monolayer dye absorbs approximately 1% of the incident monochromatic light. In order to increase the photon absorption, Grätzel cells were equipped with stable dyes and high surface area TiO₂ films. These transparent, nonporous titanium dioxide films increased the absorption of incident solar energy to 46%.¹⁸ According to Henry J. Snaith, it is possible to attain a maximum efficiency of 20.25 percent in DSSCs. Photovoltaic loss in a DSSC originates from the following (Fig. 3): (a) incomplete light harvesting, (b) inefficient electron transfer, where absorbed photons are diverted to non-injecting channels, (c) heterogeneity of dye and the semiconductor band alignment, which results in an “overpotential” to unify the electron transfer, (d) energy loss when dye relaxation causes a conformational change in the dye structure (this change causes electrons to diffuse into the oxide with lower energy), and (e) dye regeneration which causes a potential drop of around 0.3 eV, eventually leading to the most significant single loss in the DSSC. Considering these, reducing the difference between the optical bandgap of the absorber and the open circuit voltage increases the Power Conversion Efficiency (PCE).

Fig. 3 (a) Illustration of the energy level of the DSSC.²⁰ (b) Charge transfer and recombination mechanisms involved in the DSSC.²⁷

By lowering the bandgap between the semiconductor and sensitizer, it is possible to reduce the loss in potential by a factor of 0.2 eV. When the dye is strongly coupled to TiO₂, it broadens the absorption width and rearranges the optical bandgap of the DSSC. Hence the DSSC's optical bandgap cannot be determined from the sensitizer's absorption alone. Furthermore, the optical bandgap can be estimated only from the sensitized electrode's absorption onset values.¹⁹

Fundamental quantities that affect the PCE are the open circuit voltage (V_oc), short circuit current (J_sc), and fill factor (FF) of DSSCs. J_sc depends on the charge collection efficiency of TiO₂ and the light absorption efficiency of dye. The open circuit voltage depends on the valence band (VB) of the hole transport material (HTM), the band configuration of TiO₂, and recombination inside the DSSC. The fill factor accounts for the resistive losses in the DSSC circuit. The more sophisticated and vigilant the architecture, the better the FF. It is essential to engineer the CB of TiO₂ and the CB/lowest unoccupied molecular orbital (LUMO) of the hole transport material for higher V_oc and improved electron injection to enhance PCE.

Device power output is determined by the TiO₂/dye/electrolyte interface's charge separation and collection. According to the detailed energetics, forward kinetic processes are light absorption, electron injection, dye regeneration and charge transport.^21–26 The thermodynamically downhill loss pathways are excited state decay to the ground state, electron recombination to dye and electron recombination to electrolyte (Fig. 3(b)). The appreciable point in here is that the efficiency of electron injection in DSSCs depends on the magnitude of these injection kinetics relative to excited state decay to the ground state. It is also noteworthy to mention that fast electron injection dynamics requires strong electronic coupling of the dye LUMO to the metal oxide conduction-band states and energetically accessible TiO₂ density of states from the excited state of the dye. The key to achieving maximum efficiency devices lies in minimizing free energy losses, while maintaining high quantum efficiencies.²⁷

The electron injection rate (k_inj) can be expressed as the sum of all electron transfer processes as follows:

where E is the absolute energy relative to NHE of the semiconductor acceptor state, is the redox potential of the dye excited state, ρ(E) is the density of semiconductor acceptor states at energy E relative to NHE, V is the average electronic coupling between the dye excited state and different states in the semiconductor with the same energy E, f(E,E_F) is the Fermi occupancy factor for each semiconductor acceptor state which can be determined from the semiconductor electron Fermi level E_F and λ is the total reorganization energy for electron injection.²⁸ The TiO₂ conduction band density of states increases exponentially in the presence of electrolyte. This exponentially increasing density is explained by the localized intraband states corresponding to the Ti⁴⁺ sites which reduce to Ti³⁺ when subjected to oxygen vacancies or in the presence of more electronegative impurities. The exponentially increasing DOS originates from an inhomogeneous distribution of conduction band edge energies. An example of this type of system is nanoparticle crystals showing inhomogeneous surfaces and charge.²⁹

Doping plays a significant role in PCE by affecting the morphology and particle size of TiO₂ and changing the conduction band edge DOS. These intentional changes induced via doping influence the TiO₂–dye interaction and determine the energetics. If the doping favors the injection dynamics, the PCE increases. If the acceptor DOS potential (Semiconductor, TiO₂) and injection state energy (Dye excitation) potential match, there will be a three-fold increase in the injection kinetics (k_inj).^30–35 Surprisingly, the importance of electron injection and its influence on PCE is not considered widely. One of the reasons for this negligence is the early model systems which indicated an increase in photocurrent with increased proton/lithium concentration. The studies employing ultrafast transient absorption spectroscopy showed that the injection dynamics in N719 sensitized TiO₂ films was two orders of magnitude slower in the presence of typical redox electrolyte.^36–44 However, recent studies have employed time correlated single photon counting to measure injection kinetics and correlated these measurements with photocurrent. In these cases, a correlation was observed between the efficiency of electron injection and photocurrent density. These studies are in perfect agreement with IPCE data, and transient kinetic studies.^{34,35,45–49}

TiO₂ is a natural n-type semiconductor with a 3.2 eV bandgap and an electrical resistivity of 10¹³–10¹⁸ Ohm cm.^50,51 According to DOS calculations, the extreme upper end of the valence band is dominated by O 2p_π orbitals, and the bottom of the conduction band consists of isolated Ti 3d_xy orbitals.⁵² Decreasing the resistivity of TiO₂ is the key to increasing the efficiency of DSSCs. Once the electron is injected, two highly undesirable yet unavoidable reactions occur in DSSCs. Injected photo-induced electrons (i) recombine with oxidized dye molecules and (ii) react with the electrolyte to produce iodide. This recombination reduces the number of electrons traveling through the designated pathways to complete the DSSC circuit; consequently, the PCE is reduced.⁵³ We can resolve these issues by increasing the conductivity of TiO₂, by improving the dye–TiO₂ interaction by refining the band alignment and increasing the surface area of TiO₂ to increase the dye loading.

The most crucial property in increasing the efficiency of a DSSC, the conductivity of TiO₂, depends on its carrier concentration, carrier separation, charge collection efficiency, and charge transport efficiency.^54–57 Conductivity is directly proportional to charge mobility and carrier concentration. Charge mobility indicates a carrier's ability to move through the material under an electric field. This is linked with charge transport and collection efficiency. Oxygen vacancies, titanium interstitials, and the presence of impurities and defects in the structures can improve the conductivity.^50,52,58 Introducing elements into the TiO₂ lattice disturbs the conduction band and forms subbands below the CB. These can be either from the formation of Ti³⁺ or from the impurity which replaced Ti⁴⁺. These n type defects can increase the carrier concentration and boost the conductivity. Moreover, if electrons from deep trap states which lie in the DOS tail are freed, they will also promote conductivity.⁵⁹

HTM and TiO₂ share a delicate interface in DSSCs. This heterojunction creates a space charge region, indicating an energy level difference between the CB of TiO₂ and HTM, causing a band bending. An external voltage that cancels out this band bending is called the flat band potential (V_fb). Negative shift in the existing V_fb will result in an upward shift of the CB and Fermi level and vice versa. The upward shift in the CB reduces the efficiency of electron injection. Nevertheless, an intentional impurity causing a downward shift in the CB by introducing defects makes electron injection efficient.⁶⁰ Throughout the discussion, doping is identifiable as a trivial method for increasing the conductivity of TiO₂, hence improving the PCE of DSSCs.

Electrodes require higher sunlight harvesting efficiency, photoelectron injection efficiency, and electron collection efficiency in order to increase PCE. The surface area and absorption coefficient of TiO₂ determine how effectively sunlight is captured. If doping changes the morphology and increases the surface area without altering the active TiO₂ mass, it can enhance PCE through light absorption. Photoelectron injection efficiency kinetics is dependent on the density of states in the conduction band. The photocurrent density fluctuates as a result of how the flat band potential influences the conduction band density. Electron collection efficiency is determined by electron transport, electron transfer resistance and electron recombination. The crystallite size of doped TiO₂ is connected to the density of states and electron kinetics. Throughout the review, with available data, we explain how doping affects the flat band potential and PCE. The bandgap too plays an influencing role in recombination dynamics. All the recombination pathways are connected to conduction band and surface states. Three major charge recombination pathways are direct transfer from the conduction band to electrolyte, indirect transfer from monoenergetic deep surface states, and indirect transfer from the distribution of bandgap surface states. Understanding how the bandgap changes with doping is the primary step in tackling the recombination.

Doping is an incredible way to engineer the electronic properties of materials. Since the first report of TiO₂ photovoltaic properties, there has been a growing trend in the research industry to increase the conductivity of TiO₂. Fig. 4 shows conclusive evidence for ongoing and booming research in doping TiO₂. Doping TiO₂ for improving the physical and morphological properties is well-known in solar cell research. In DSSCs, TiO₂ acts as a photoanode that facilitates electron transportation from the dye to the conductive oxides. This anode requires high absorbance, a reduced bandgap, and a low recombination rate for better performance. Doping affects all these properties by altering the morphology and structural properties. For example, Ag and Ni doping make TiO₂ more porous, and the size and shape of big particles become irregular. Also, doping creates a sponge-like structure, increasing the surface area and the efficiency of light absorption. Single doping, co-doping, doping into a composite, and growing delicate designs with doped TiO₂ have been tried for efficiency enhancement in DSSCs.⁶¹

Fig. 4 Growing trend of TiO₂ doping (Articles published from 1960 to 2020 using the keywords “TiO₂” and “Doping” on the ScienceDirect website, scrutinized by checking their relevance).

Although TiO₂ is the most suitable candidate as a photoanode material in terms of abundance and non-toxicity, it is a poor conductor. Generations of researchers have doped TiO₂ with various elements in the periodic table. A comprehensive view is essential to understanding doping and its effect on materials properties to engineer a new material. Many reviews have been published based on doping in TiO₂ since the beginning of interest. There are reviews about doping of TiO₂ to increase the efficiency of DSSCs, the fundamental mechanisms and properties of TiO₂ photoanodes, and different TiO₂ fabrication methods.^1,62–64 In doped TiO₂, the cell efficiency varied with the doping strategy. Hence, this review creates a framework for analyzing the effect of doping by normalizing the increase in efficiency. Up until this point, doping has been reported as changing the DSSC output if there is a change in efficiency. However, there are errors involved in this type of analysis. When the preparation method is sophisticated, we get higher efficiency solar cells for undoped TiO₂.⁶⁵ Different researchers have reported different efficiencies for bare TiO₂-based DSSCs. This ambiguity in bare DSSCs’ efficiency makes it difficult to find the effect of doping. There are undoped TiO₂-based DSSCs with less than 1% efficiency and above 5%.^65,66 The efficiency of DSSCs based on undoped TiO₂ changed between 0.019 and 9.05%.^65–75 To understand how a dopant affects the performance of DSSCs, we need to reframe the understanding of how the Power Conversion Efficiency (PCE) changes when we add a tiny amount of dopant into TiO₂. This review provides a detailed summary of the effect of doping of TiO₂ on its photoconversion efficiency based on the performance of the DSSC. However, the trends predicted with this analysis help optimize TiO₂-based thin-film heterojunction solar cells and PEC hydrogen production.

Alkali metals, alkaline earth metals, transition metals, lanthanoids, post-transition metals, metalloids, non-metals, and halogen were doped into TiO₂ to increase the efficiency of electrochemical cells.^76–84 Elements doped into TiO₂ for electrochemical applications are marked in Fig. 5. This review is structured to discuss doping and co-doping separately in different electrochemical systems. Single elemental doping will be discussed based on the periodic element group-wise. Codoping is categorized into three main sections: metal doping, non-metal doping, and metal–nonmetal doping. A detailed discussion of bandgap tuning, size effects with doping, and different preparation methods of doped TiO₂ is included. The review structure is illustrated in Fig. 6.

Fig. 5 Elements doped into TiO₂ included in this review.

Fig. 6 Structure of this review.

2. Framework for analyzing the impact of dopants in DSSCs

The impact of a dopant on the efficiency of a DSSC is estimated from the percentage increase/decrease in efficiency calculated using eqn (1):where η_d is the efficiency of the doped TiO₂ electrode-based DSSC/QDSSC, and η_b is the efficiency of the bare TiO₂-based DSSC (control cell).

Fig. 8 is obtained after calculating the percentage increase in PCE on doping TiO₂ with alkali and alkaline earth metal dopants using eqn (1). The analysis shows that among all the alkali and alkaline earth metals explored to date, Sr dopant has the maximum impact on the efficiency enhancement of the bare TiO₂ photoanode when treated with TiCl₄, and a scattering layer is introduced. However, if we consider only the impact of the dopant alone on a bare TiO₂ photoanode, then Mg and Na are the potential dopants.

3. Single elemental doping

Doping a semiconductor increases the conductivity and tunes its optical properties. Metal dopants block recombination and enhance the driving force for electron injection. They also form complexes that enhance dye binding on the surface of TiO₂. Hence the overall power conversion efficiency can be enhanced. TiO₂ is doped with various elements in the periodic table to adjust its optical and electrical characteristics. This section discusses single elemental doping on TiO₂ in groups. Alkali metals, alkaline earth metals, transition metals, and post-transition metals are all thoroughly discussed. Lanthanoids, metalloids, halogens, and other non-metals are also extensively investigated for doping TiO₂. The link between doping and the bandgap is examined using presented bandgap data.

3.1. Alkali metals

The alkali metal group consists of lithium, sodium, potassium, rubidium, cesium, and francium. All the elements in this group have their outermost electron in the s-orbital. Usually, these metals are all shiny, soft, and highly reactive. Among alkali metals, sodium and lithium were doped into TiO₂ to enhance the PCE of solar cells. Table 1 lists the most recent J_sc, V_oc, and PCE data of alkali metals.

Table 1 Current statistics of alkali and alkaline metal doping in TiO₂

S. no.	Dopant	Treatment/coating	Structure	Synthesis method	V oc (V)	Photocurrent density (mA cm^−2)	FF	PCE (%)	Enhancement in PCE (%)	Bandgap (eV)	Application	Sensitizer/dye	Ref.
1	Lithium (Li)		Nanoparticles	Hydrothermal	0.655	9.81	0.358	2.5	362.96		DSSC	N719	76
2	Sodium (Na)		Nanorods	Hydrothermal	0.55	5.5	0.545	1.65	79.35		DSSC	Extract from petals of Hibiscus sabdariffa (Roselle)	68
3	Calcium (Ca)		Nanorods	Hydrothermal	0.649	19.2	0.67	8.35	13.92		DSSC	N3	96
4	Magnesium (Mg)		Thin films	Aerosol assisted chemical vapor deposition	0.74	11.99	0.67	6.1	81.55	2.8	DSSC	N719	97
5	Magnesium (Mg)		Thin films	Dip coating	0.83	3.02	0.668	1.68		2.98	DSSC	N719	90
6	Magnesium (Mg)		Nanoparticles	Hydrothermal	0.615	19.1	0.605	7.12	12.13		DSSC	N3	89
7	Magnesium (Mg)		Nanoparticles	Solvothermal	1.21	1.8	0.55	1.2	66.67		DSSC	Alkoxysilyl-coumarin dye	91
8	Magnesium (Mg)		Nanoparticles	Solvothermal microwave irradiation (SMI) technique	0.6435	16.536	0.692	7.36	190.91	3.638	DSSC	N3	88
9	Strontium (Sr)	TiCl4	Nanoparticles	Chemical hydrolysis	0.78	18.53	0.66	9.57	249.27		DSSC	N719	98
10	Strontium (Sr)		Nanoparticles	Hydrothermal	0.728	17.43	0.62	7.88	12.73	3.05	DSSC		99

---

Sodium (Na). When Na is doped into the TiO₂ lattice structure, then the Ti⁴⁺ ions are substituted by Na⁺ ions. Such substitution leads to the generation of oxygen vacancies and electrons. As a result, the electron concentration increases, increasing the conductivity. Another impact of Na dopant is that it positively shifts the flat band potential, leading to a high band gap and driving force resulting from incident photons. The enhanced device performance can also be attributed to the large surface area. However, optimizing the dopant concentration is necessary as an excess of dopant can lead to the generation of deep dopant levels or electron–hole recombination centers. Thus, it is clear that excessive dopant can increase the recombination rate, which will be the reason for the reduced percentage absorption and poor performance of the device.⁶⁸Fig. 7(a) and (b) and Table 1 explain the position of sodium among dopants in alkali–alkaline metals.

Fig. 7 (a) Performance of alkali & alkaline earth metal doped TiO₂ based DSSCs. (b) V_oc and J_sc values of alkali & alkaline earth metal doped TiO₂ based DSSCs. Data points are labelled with references.

Fig. 8 Percentage increase in PCE due to alkaline and alkali metal doping. Data points are labelled with references.

Lithium (Li). Electron transport in liquid electrolyte is defined based on carrier diffusion due to the absence of a sizeable electronic field. This can be associated with the formation of traps. When electrons spend more transient time in traps, it will cause slow diffusion. The population of traps has a direct relation to surface area. High concentrations of Li⁺ in the lattice can change the trap population and electron diffusion as well.⁸⁵ Li ions adsorbed onto TiO₂ can intercalate into the interstitial sites and result in the shifting of the TiO₂ conduction band.⁸⁶ Li doping does not increase the effectiveness of the DSSC, despite the fact that it aids in our understanding of structures and processes. A DSSC prepared with Li doped into TiO₂ performed 40% less effectively than a P25-based DSSC.⁸⁷

3.2. Alkaline earth metals

Elements with an oxidation state of +2 and having a full outer s-orbital belong to alkaline earth metals. They have a shiny, silvery appearance. These elements usually exhibit low density, melting point, and boiling point values. Magnesium, calcium, and strontium are some potential dopant candidates that can enhance the photocurrent and efficiency of DSSCs by modifying TiO₂ electrode properties.

Magnesium (Mg). Mg is a readily available, low-cost metal dopant that can enhance the photoanode's performance. Due to the comparable atomic radii of Mg²⁺ (0.72 Å) and the Ti cation (0.64 Å), Mg²⁺ ions occupy quadrivalent Ti⁴⁺ sites via substitutional doping.⁸⁸ This results in a reduced e⁻/h⁺ pair recombination rate and an increased electrical conductivity of the Mg-doped TiO₂ photoelectrode due to the increased excited state electron injection or generation of donor levels.^88,89 Mg doping also positively impacts the flat band potential, which improves the driving force for electrons injected from the LUMO of the dye into the conduction band (CB) of TiO₂.⁸⁹ The enlarged energy difference between the LUMO of the dye and the CB of TiO₂ also plays a vital role in enhancing the short circuit current of the device. Mg doping also shifts the absorption towards the visible region which is attributed to the lowering of the band gap due to the formation of another state between the O 2p VB and Ti 3d CB of the parent lattice. Mg doping increases the electron transport which improves the photocurrent density. Another work has reported a 65% increase in the efficiency of device performance on switching from single layer Mg doping to 7-layer Mg doping due to the increment in grain size and reduction in band gap.⁹⁰ Efforts must be made to optimize the dopant concentration as higher concentrations can affect the crystal growth of the lattice and create defects which, in addition to increasing the charge recombination, also decreases the electron transport proficiency. Mg²⁺ significantly affects V_oc, increasing it by up to 1.2 volts.^91,92 There are works claiming that Mg doping causes a decrease in J_sc as well.⁹³

Calcium (Ca). Doping of the TiO₂ lattice with Ca results in replacing quadrivalent Ti⁴⁺ with the divalent Ca²⁺ ions, resulting in oxygen deficiencies and donor levels. Ca doping is also observed to shift the flat band potential positively. These factors are responsible for enhanced electron density and electrical conductivity. Another notable impact of Ca doping is carrier relaxation. The rate of charge transport in the doped photoanode is faster than the relaxation of the excited state of the sensitizer dye. The recombination time is also higher, thereby increasing the photocurrent density. The reduction in band gap plays a significant role in shifting the absorption spectra to the visible region.^94–96

Strontium (Sr). On increasing the concentration of Sr dopant, the crystallite size of TiO₂ is observed to reduce and cause TiO₂ lattice distortion. The local lattice distortion can be attributed to the mismatch of the sizes of the dopant Sr²⁺ ions (1.18 Å) and host Ti⁴⁺ ions (0.60 Å).⁹⁹ The higher ionic radius of Sr²⁺ than that of Ti⁴⁺ enhances the possibility of local lattice distortions due to the insertion of larger ions into the smaller lattice. Insertion of Sr in the TiO₂ lattice results in oxygen vacancy lattice defects that can trap photogenerated holes, thus prolonging the life of the photo-excited electrons and holes. This results in increased J_sc. The distortion of TiO₆ octahedra and the introduction of oxygen vacancies generate dipole moments. The internal polarization fields in the structure facilitate better charge separation efficiency of photogenerated electrons and holes and decrease the recombination probability. Due to the suppression of the recombination rate and the increment of the electron lifetime, the open circuit voltage of the device increases. Substitution of Ti⁴⁺ with Sr²⁺ has an effect on the electronic energy levels, which is supposedly near the conduction band edge, i.e., Ti⁴⁺ 3d bands, as a result of which the band gap decreases and the absorption curve becomes red-shifted. Additionally, an optimum Sr concentration is also observed to increase the pore diameter, resulting in enhanced dye adsorption. The narrowing of the band gap and enhanced dye adsorption facilitate better photoexcitation, thereby improving the electron density in the device.⁹⁹ Another work has reported an improved dielectric constant on adopting Sr doping. This is attributed to extrinsic and intrinsic effects like the development of Schottky barriers between semiconductor grains at insulating grain boundaries (interfacial/grain boundary effect) and the density of the samples. The work has also suggested the introduction of a light-scattering nanocrystalline TiO₂ layer due to the transparency of the Ti_1−xSr_xO₂ film towards visible light. The TiCl₄ treatment can effectively prolong the charge carrier lifetime of the device, thus claiming its application for a long-term light conversion photovoltaic commercial device.¹⁰⁰

3.3. Lanthanoids

Rare earth elements offer sharp and stable luminescence ideal for optoelectronic uses. Rare earth elements such as lanthanum (La),^101–103 neodymium (Nd),⁸⁰ samarium (Sm),^104,105 europium (Eu),⁷⁰ gadolinium (Gd),¹⁰⁶ terbium (Tb),¹⁰⁷ erbium (Er),^108,109 and ytterbium (Yb)^108–113 have been doped into TiO₂ to improve its optoelectronic properties.^{77,80,104,107} Coupling of weakened ligands arising from 4f orbital shielding by 6s, 5p, and 5d orbitals and forbidden f–f transitions works in favor of stable and sharp luminescence for rare-earth ions. Due to their insolubility and thermal quenching, there were limitations in incorporating rare-earth ions into various semiconductors. However, combustion processing, sol–gel, hydrothermal, solvothermal and ball milling methods have made rare-earth doping possible.

Lanthanoids effectively alter the absorption range of semiconductors by enhancing visible photons’ absorption. All but one lanthanide have significant chemistry associated with 4f orbital filling. 4f orbitals are effective in the adsorption of pollutants and increase photocatalysis. The 5d electronic configuration of these ions contributes to high charge mobility and redox coupling. Lanthanoids are utilized to reduce the bandgap of TiO₂ and increase the surface area. There is an existing conflict regarding the position of lanthanum ions after doping. Some researchers claim that no rare earth elements can penetrate the TiO₂ lattice due to their larger ionic size. They tend to either be in the surface interstitial of TiO₂ or create Ti–O–La bonding on the surface. This bonding enhances the oxygen vacancies and surface area of TiO₂. Doping with rare earth elements improves the photovoltaic performance of TiO₂-based devices. However, there are also reports regarding La³⁺ located in the crystallite interstitials. This penetration can inhibit crystal growth as well.¹¹⁴ The PCE performance of the Lanthanide group elements doped into TiO₂ is illustrated in Fig. 12(a) and (b). The statistics of doping and its effects are listed in Table 2.

Table 2 Current statistics of lanthanoid doping in TiO₂

S. no.	Dopant	Structure	Synthesis method	V oc (V)	Photocurrent density (mA cm^−2)	FF	PCE (%)	Enhancement in PCE (%)	Bandgap (eV)	Application	Sensitizer/dye	Ref.
1	Europium (Eu)	Nanophosphors	Combustion and doctor blade	0.67	18.53	0.709	8.8	5.77	3.32	DSSC	N719	70
2	Europium (Eu)	Nanoparticles	Sol–gel, doctor blade	0.77	9.61	0.69	5.16	21.99		DSSC	N719	105
3	Gadolinium (Gd)	Spheres	Hydrothermal and spray pyrolysis	0.723	7.749	0.736	4.12	73.84		DSSC	N719	106
4	Gadolinium (Gd)	Mesoporous	Solvothermal	0.723	7.749	0.736	4.12	42.48		DSSC	N719	106
5	Neodymium (Nd)	Anatase	Hydrothermal	0.62	12.83	0.66	5.23	34.10	2.6	DSSC	N719	117
6	Neodymium (Nd)	Nanoparticles	Screen printing	0.78	11.5	0.683	6.15	13.89	3.2655	DSSC	N719	80
7	Samarium (Sm)	Nanoparticles	Hydrothermal	0.863	14	0.5032	6.08	72.72		DSSC	N719	104
8	Samarium (Sm)	Nanoparticles	Sol–gel, doctor blade	0.81	10.9	0.67	5.81	37.35		DSSC	N719	105

---

Lanthanum (La). Doping lanthanum (La³⁺) into TiO₂ does not change the binding energy of Ti2p peaks. This steady binding energy can be attributed to absorbed oxygen. Surface treatments with lanthanum promote its absorption into the surface of TiO₂. This absorbed oxygen counteracts oxygen vacancies and band bending effects with lanthanum co-doping. From DRS studies, it is evident that visible light photon absorption is reduced in La-doped TiO₂ in comparison with pure TiO₂. Doping La into TiO₂ distorts the O–Ti–O triad to induce oxygen vacancies. These distortion-generated oxygen vacancies act as deep and near-surface traps. These vacancies change the electron–hole interaction and thus avoid electron–hole recombination. As a result of doping, the absorption edges of the samples shifted to high-energy wavelengths and reduced visible light absorption. Oxygen vacancies generated from the doping, which act as trapping sites, increase the average free electron energy (Fermi energy) towards the conduction band edge minimum and increase the bandgap.^114–116

Neodymium (Nd). Neodymium doping increases the surface area and reduces the pore diameter. Due to the increase in dye adsorption, a Nd-doped TiO₂ photoanode-based DSSC shows significant increases in the corresponding efficiency. Increasing the Nd-doping beyond a particular doping concentration reduces the efficiency due to the shielding of active sites. The high areal coverage by neodymium shields the TiO₂ and prohibits the absorption of photons. Additionally, excessive doping can introduce traps and defects into the TiO₂ lattice. Photogenerated electrons are trapped in defect centers, consequently inhibiting the electron transfer process, thus decreasing the photogenerated current. Nd-doped TiO₂ retained its superiority in all applications such as DSSCs, QDSSCs, and PECs. Nd doping affects the particle size and strain created in the TiO₂. The particle size decreases to a minimum and bounces back with an increase in doping both in QDSSCs (Fig. 10) and in DSSCs (Fig. 11). In this study, the change in the bandgap of TiO₂ according to the doping is portrayed perfectly (Fig. 9). Nd³⁺ doping provides additional energy states in the bandgap of TiO₂ by increasing the oxygen vacancy and enhancing the charge injection rate. This charge injection rate spike is a consequence of the higher dielectric constant exhibited by Nd doping. The 4f suborbitals of TiO₂ enabled by Nd atoms linked to the surface and reduction in oxygen vacancies. The oxygen vacancies prevent charge carrier recombination and consequently increase the charge transport.¹¹⁷

Fig. 9 Crystallite size and bandgap in Nd doped TiO₂.¹¹⁷

Fig. 10 Crystallite size and efficiency in a Nd doped TiO₂ based QDSSC.¹¹⁷

Fig. 11 Crystallite size and efficiency in a Nd doped TiO₂ based DSSC.¹¹⁷

Fig. 12 (a) Percentage increase in PCE with lanthanide doping. (b) V_oc and J_sc values of lanthanoid doped TiO₂ based DSSCs. Data points are labelled with references.

Samarium (Sm). Samarium (Sm³⁺) doping into TiO₂ could transform the incident ultraviolet light into visible light with a down-conversion luminescence effect. The downconversion improves the utilization of sunlight and increases the ability of the dye to excite the electrons effectively. Samarium doping introduces traps into the TiO₂ photoanode. The trap density increases with the doping concentration, and some of the trapped electrons will reach the valence band and cause a nonradiative transition unfavorable to the efficiency of the cell. Samarium doping shifts the absorption edge towards a smaller wavelength. The increase in V_oc explains the observed increase in the band gap. Zala and coworkers claimed that a wide bandgap can make an effective energy barrier to suppress the recombination of electrons. Sm-doped photoanodes show enhanced absorption in the visible region compared with bare TiO₂ samples. Optimal doping significantly increases the efficiency and performance of doped photoanodes in the case of samarium. An increase in samarium doping concentration favors the efficiency until a particular point, and a further increase in doping decreases the efficiency.¹⁰⁴

Europium (Eu). Nanophosphors synthesized through the solution combustion method with different concentrations of europium enhanced the efficiency of DSSCs. Eu³⁺ doping changed the morphology of TiO₂ from tightly packed spheres to a network of larger spheres. The porosity of the samples increased with the Eu doping concentration. These effects wear off when the doping percentage increases above a nominal range. The Eu doping decreases the defect emission in TiO₂ samples. The Ti³⁺ defect level is below and the oxygen defect level is above the conduction band for TiO₂. Eu doping into TiO₂ reduces the defect emission due to their positions. Once UV radiation is absorbed, the energy is released to the defect states. Since the ⁵D₀ emitting state of Eu³⁺ ions is lower than TiO₂, the energy will be transferred to these crystal-field states of Eu³⁺, and this energy transfer improves the PL intensity.^70,107

Gadolinium (Gd). Gadolinium doping leads to a high surface area, interconnected nanostructures, better charge transport, and better light scattering that eventually lead to higher efficiencies in electrochemical cells, including DSSC and photo-related applications. Doping reduces the grain size of TiO₂ significantly and increases the specific surface area. Incorporating Gd in a small percentage into TiO₂ leads to elevated light absorption and shifting in absorption edge towards the red end. Gd has a half-filled 4f electronic configuration, which is stable. Compared to undoped TiO₂, Gd³⁺ doped samples exhibit higher photocatalytic degradation owing to the rearrangements in the 4f orbitals. When a Gd³⁺ ion traps an electron, this electron gets immediately transferred to an absorbed oxygen molecule to regain stability for Gd. In the case of DSSCs, this leads to the decomposition of methylene blue. Gd³⁺ ions at higher concentrations effectively become recombination centers due to the decomposition of methylene blue. However, at optimum doping concentrations, Gd³⁺ doping increases efficiency.¹⁰⁶

Terbium (Tb). According to the literature, terbium isn’t a particularly effective single dopant for TiO₂. Tb³⁺ doping drastically reduces the efficiency in a TiO₂ based DSSC. Tb³⁺ provides a p type doping effect by replacing Ti⁴⁺ with Tb³⁺ ions. This will cause an upward Fermi level shift. This shift contributes to the increment of open circuit voltage. So, whenever the Tb doping increases, the open circuit voltage increases by 6%. The Tb³⁺ doping endorses Ti⁴⁺ substitution, which produces excess charge and deep traps. This creates local recombination centers in the host material. These sink the electrons produced and reduce the current production. Even though there is a high open circuit voltage production, the efficiency is reduced significantly owing to the low current efficiency. Nevertheless, co-doping TiO₂ with terbium with other rare earth elements shows excellent voltage and current output.¹⁰⁷

3.4. Metalloids

Metalloids are also called semi-metals due to their mixed behavior of metals and non-metals. Boron,^118,119 antimony¹²⁰ and germanium¹²¹ have been doped into TiO₂ for attaining better DSSC efficiencies. When doped with metalloids, it creates excess electrons in the system. Increased electron concentration will improve electron transport efficiency, enhancing photocurrent density.

Boron (B). Boron is doped into TiO₂ nanotubes to increase the efficiency of DSSCs. Doping is done via electrochemical anodization using a boric acid precursor. The introduction of boron into the lattice increases the electron lifetime and reduces the recombination rate. Undoubtfully, this is reflected in their high light-harvesting efficiency and short circuit photocurrent density. Additionally, introduction of boron into the interstitial sites of TiO₂ contributes to the conduction band's shift.¹²² The homogeneous and porous layer created by boron doping facilitates the absorption of significant amounts of dye, increasing J_sc.¹²³ Boron (B³⁺) doping also favors the production of oxygen vacancies and conversion of Ti⁴⁺ into Ti³⁺.^124,125 DFT calculations show that boron can exist either in the interstitial sites or substitutionally by replacing Ti⁴⁺ in the TiO₂ lattice.¹¹⁹ The blue shift in the UV-vis absorption spectrum can identify boron interstitial doping.¹²⁶

Antimony (Sb). When doped with Sb, Sb³⁺ ions are incorporated into the TiO₂ lattice. Unlike the theory, the donor density is significantly reduced when Sb is doped into TiO₂. However, IMPS measurements show that 1% Sb doped samples show faster electron transport, contributing to higher J_sc values. A positive shift in the flat band potential of TiO₂ increases the energy gap between the LUMO of dye and the conduction band of TiO₂. This increase in energy gap results in increased injection driving force and improves the electron injection from the LUMO of the dye to the conduction band of TiO₂. Sb doping also follows the trend like other doping methods under study that increase the efficiency to optimum doping and decrease. We can see the doping percentage–efficiency trend in Fig. 13.¹²⁰

Fig. 13 Change in PCE with Sb doping concentration.¹²⁰

3.5. Transition metals

Transition metals are the largest set of cation providers in the periodic table. These elements have a partially filled d subshell or can provide cations with an incomplete d subshell. Transition metals provide a large pool of complex ions with various oxidation states. Doping TiO₂ with transition metals creates additional energy levels in the CB due to the aforementioned partially filled d-orbits. This section discusses the effects of doping these elements into TiO₂ such as scandium,¹²⁷ vanadium,¹²⁸ chromium,^129,130 manganese,^131–133 iron,¹³⁴ cobalt,¹³⁵ nickel,¹³⁶ copper,^137–139 zinc,¹⁴⁰ yttrium,¹⁴¹ zirconium,⁷⁸ niobium,¹⁴² silver,¹⁴³ cadmium,¹⁴⁴ tantalum,¹⁴⁵ and tungsten.¹⁴⁶

Scandium (Sc). On doping TiO₂ with Sc, the Ti⁴⁺ sites are found to be substituted by Sc³⁺. This particular substitution positively affects atomic disorder (morphological alterations such as size and pore size distribution and deformation of the tetragonal lattice) and electronic disorder. The holes generated in the valence band counterbalance the donor levels evolving from the oxygen vacancies of anatase, resulting in a reduction of the energy difference between the conduction band and the LUMO level of the dye, in contrast to W doping. The reduction of the energy difference implies a reduction in the rate of electron injection. Sc doping does not have any positive impact on carrier density. The bandgap varies linearly with Sc concentration on TiO₂. The atomic disorder plays a vital role in governing the transport properties within the semiconductor layer. At an optimum concentration of 0.2 at% Sc, the photocurrent density increases due to the enhancement of electron lifetime. However, beyond this concentration, the transport rate suffers a detrimental impact, leading to a decrease in the photocurrent density. In summary, an increase in Sc concentration can have three effects – an increase in flat band energy and electron lifetime and a reduction in transport rate. In contrast to the expectation, the open circuit voltage decreases due to the voltage drop due to series resistance.¹²⁷

The electron lifetime increases with the substitution of Ti⁴⁺ with trivalent ions such as Y³⁺ and Ga³⁺ but the transport rate decreases. In the VB group of the periodic table, as we move from top to down, the current increases (due to the increase in the concentration of carriers and reduction in film resistance), while the voltage decreases (due to the faster recombination of electrons and reduction of the energy difference between the flat band potential and the I³⁻/I⁻ redox potential). Nb doping is more beneficial among the group VB elements. Group VB elements also improve the electron transport rate, improving the charge collection efficiency. The bandgap undergoes redshift, and there is also an increment in visible light absorption.

Vanadium (V). The advantage of using V is that it can extend the light absorption edge from the UV region to visible region of the TiO₂ semiconductor and increase the carrier lifetime. The increase in visible absorption and photogenerated e⁻/h⁺ is attributed to two facts: (1) the ionic radii of V are much close to Ti and (2) the tetragonal crystal structure of VO₂ is similar to that of TiO₂. The positive shift of the flat band potential resulting from doping with V improves the driving force for injecting electrons from the LUMO of the dye to the conduction band of TiO₂ and the photocurrent. Charge transport was also seen to increase on doping with V. Since V has one electron more than Ti(IV), a donor level was created, which can increase the concentration of the carriers in the V doped TiO₂ material, which results in reduction in the film resistance.¹⁴⁵ In a different work, V is integrated into high surface area porous TiO₂. Lowering the cell's internal resistance increased the conductivity of TiO₂. Additionally, it offered higher chemical stability and improved electron transport compared to the undoped TiO₂.¹⁴⁷

Chromium (Cr). The similar cationic radii of Cr³⁺ (0.755 Å) and Ti⁴⁺ (0.745 Å) have made Cr³⁺ a suitable dopant candidate, ensuring insertion of Cr³⁺ into the TiO₂ without damaging the structure, but causing a slight expansion. Cr doping reduces Ti⁴⁺ to Ti³⁺ by replacing Ti⁴⁺.¹⁴⁸ Ti³⁺ states and oxygen vacancies have suppressed the recombination rate of electron–hole pairs. The oxygen vacancies are also referred to as hole-trapping defects. The generation of oxygen vacancies leads to the trapping of the positive carriers, thus making the carriers less mobile than usual. This results in electron mobility improvement and hence reduces charge recombination. This results in the improvement of the photocurrent efficiency of the device. Due to the different oxidation states, Cr can induce multiple mid-gap energy levels that modify the electronic properties of TiO₂. Cr doping decreases the bandgap energy owing to the formation of Ti³⁺ states and oxygen vacancies. Cr doping also enhances dye adsorption owing to the better binding of the dopant ions and dye molecules. Cr doping enhances the photocurrent density, which is attributed to the formation of mid-gap states. The mid-gap states reduce the movement of photoelectrons to the valence band, reduce the recombination with the oxidized state of dye and therefore strengthen the lifetime of electrons.^149,150

Manganese (Mn). The ionic radius of Mn²⁺ (0.67 Å) almost equal to that of Ti⁴⁺ (0.605 Å) makes Mn a suitable dopant. Owing to the minimum difference between the ionic radii of Mn²⁺ and Ti⁴⁺, Mn²⁺ doping is more prone to agglomeration. The doping mechanism can follow any of the following routes – exchange of parent ions with dopant ions, insertion of dopants into the interstitial sites of the parent lattice or surface doping of the parent lattice. Within a low level of Mn doping, the crystallinity of the parent lattice can be improved by increasing the amount of doping. Mn doping has been observed to follow the route of replacing Ti ions at the interstitial sites of the lattice. Contrary to the general perception of TiO₂ growth inhibition on doping, Mn doping increases the crystallite size. An increase in crystallite size can negatively impact efficiency due to the decrease in the surface area of the photoanode available for dye adsorption. Another suitable impact of Mn doping is the decrease in the bandgap for undoped TiO₂ crystals. Such a property can be attributed to the formation of impurity states between the conduction and valence bands of TiO₂. Additionally, the sp–d interaction between the band electrons of TiO₂ and localized dopant d cations is also responsible for the decrease in the band gap. The trap states induced by doping also influence the recombination process in DSSCs, which can negatively impact efficiency. Therefore, an optimum doping concentration is required to determine the achievable PCE. It has been observed that the structural impact is more dominant than the recombination with high Mn doping on the device's efficiency.¹⁵¹

Iron (Fe). Fe doping increases the crystallite size of the TiO₂ lattice structure, resulting in decreased dye adsorption. Fe doping leads to the generation of Ti³⁺ trap states and induces band edge shift. The Ti³⁺ population is a result of the substitution of Ti⁴⁺ with Fe³⁺. Creation of Ti³⁺ states and oxygen vacancies at deep levels increases electron density and causes a downshift in the quasi-Fermi level. The trap states act as photoactive sites facilitating electron transport via hopping and trapping/de-trapping events. Fe doping is observed to introduce lower defect concentration than Cu doping but higher defect concentration than Sn doping.¹⁵² Fe doping also facilitates an increase in surface-hole concentration, a broad absorption peak and fast transient decay, thus resulting in poor current density at 430 nm compared to a Sn-doped, Cu-doped or undoped TiO₂ lattice. Compared to Cu and Sn, Fe doped TiO₂ photoanodes undergo faster recombination of injected electrons with surface holes, thereby leading to poor performance of the device. Fe doping induces an increase in oxygen vacancies and a reduction in the crystallite size of the TiO₂ lattice structure. The oxygen vacancies lead to defect concentrations of trapped holes. This results in trapped free electrons at sub-energy states (Fe²⁺, Fe⁴⁺, etc.) in the conduction band being quickly excited to the Ti⁴⁺ conduction band, causing higher absorbance. The oxygen vacancies also cause d–d transition within the distorted geometry of the transition metals, resulting in longer lives of excited states in Fe-doped TiO₂ photoanodes. The longer lifetime of excited states can prove detrimental to the device due to recombination. Fe doping has shown less photoelectric conversion efficiency relative to undoped TiO₂.^153,154 Fe-doped TiO₂ samples can be more beneficial for the near-visible photocatalytic reduction process due to their characteristic fast rate of decay of transient absorption kinetics at 430 nm.¹⁵⁴ This fast rate of decay is attributed to the increased concentration of holes in the doped sample. Recombination at the surface states of oxides plays a dominant role in deciding the cell performance with the Fe doped TiO₂ photoanode. Fe doping must be done with caution because excessive Fe³⁺ doping changes n type TiO₂ into p type TiO₂.¹⁵⁵

Cobalt (Co). Co doping decreases TiO₂ lattice size attributed to the hindrance of the crystal growth of TiO₂ nanoparticles. The reduced size implies the availability of more surface area for dye adsorption. M. I. Khan reported the influence of using stacked Co doped TiO₂ multilayer thin films.¹⁵⁶ The exciting result of the work is the phase transformation of anatase TiO₂ to brookite TiO₂ on the introduction of multilayer thin films. Also, using a multilayer of Co doped TiO₂ films shifts the absorption spectra to the visible range due to the d–d electronic transition of Co²⁺ in octahedral coordination. Codoping also decreases the bandgap due to the formation of impurity states in the forbidden bandgap of TiO₂. Increasing layers also increases free carrier concentrations and decreases the multilayer film's resistivity. However, an optimal number of layers must be decided to prevent internal resistance obtained from the increment in thickness.

Nickel (Ni). Ni doping increases the crystallite size of the parent TiO₂ lattice, which can be due to the insertion of Ni ions into the lattice. The increase in size is detrimental to the dye's adsorption capacity due to the reduced surface area. Counter to this effect, the decrease in bandgap induced by Ni doping in the TiO₂ structure results in little increase in the performance of cells.^140,157 Doping Ni into TiO₂ nanowires can enhance charge mobility. EIS studies of Ni-doped samples confirm lower charge transport resistance and reduced charge recombination. When Ni²⁺ replaces Ti⁴⁺, the crystallinity is improved, and certain surface trap states are partially eliminated.^136,158

Copper (Cu). The size similarity of the ionic radius of Cu²⁺ (0.73 Å) with that of Ti⁴⁺ (0.605 Å) has grasped attention among researchers to explore its suitability as a dopant. Cu doping is less likely to form agglomeration in the structure than other dopants like Mn.¹⁵¹ With 1% and 2% Cu doping, there is a probability of the formation of CuO on the lattice. Cu doping can lead to an increase in the crystallite size.^151,159 Such a condition suggests the inclusion of Cu either at the interstitial site or on the surface of the anatase TiO₂ crystals. The formation of additional states between the conduction band and valence band of TiO₂ and the sp–d interaction between the band electrons of TiO₂ and localized dopant d cations decrease the bandgap and improve the optical response. The introduction of trap states impacts the bandgap in addition to the recombination process. Therefore, an optimum doping concentration is highly recommended to balance the positive and negative impacts of trap states. K. Sahu et al. reported an increment in V_oc from 714 to 781 mV with increased Cu doping. The Cu doping facilitates the space charge region owing to increased structural disorder. This increases the conduction band edge of TiO₂ and the upward shift of the Fermi level. Another significant impact of the shift of the Fermi level towards the conduction band edge is a decrease in electron injection efficiency from the dye into TiO₂ and thus a loss in current. Compared to Mn doped TiO₂ photoanodes, Cu doped TiO₂ photoanodes have lower electron lifetimes.¹⁵¹ Another vital feature of Cu is its localized surface plasmon resonance (LSPR). The LSPR effect enhances the light absorption proficiency of dye molecules and boosts the separation of carriers by inducing a collective oscillation of metal free electrons at resonance. The LSPR effect induces a strong electromagnetic field around Cu nanoparticles, facilitating their interaction with dye molecules. This results in the effective resonant energy transfer between the excited state of the dye and surface plasmons.¹⁵¹ Cu doping is responsible for increasing the light-harvesting efficiency of the dye sensitizer by enhancing the far-field scattering, inducing charge separation, or increasing near-field local surface plasmons. Another work¹⁵⁴ compared the concentration of defects introduced on TiO₂ and reported that it was in the order – Cu doped TiO₂ > Fe doped TiO₂ > Sn doped TiO₂ > pure TiO₂. The work also reported that the lower the oxidation number of the dopant, the more significant the broadening of the fundamental vibration modes (Eg₁) in Raman peaks and the greater the generation of oxygen vacancies. Cu²⁺ leads to the generation of Ti³⁺ states due to the availability of more oxygen vacancies. At the oxygen deficient sites, trapped free electrons facilitate charge equilibrium, enhancing the electron density of states and increasing the light absorption. R. T. Ako et al. suggested that the significant factor for choosing a metal dopant for the photocatalytic or photovoltaic device should be governed by the elevation of surface free energy states of photoanodes rather than bulk defect generation. In view of this, their work concluded that the bulk defects in Cu doped TiO₂ can also lead to reduced electron density due to the electron–hole recombination. Along with Cu, the plasmonic behavior of Ag and Au also affects PCE, as illustrated in Fig. 14.

Fig. 14 Impact of the plasmonic behavior on percentage increase in efficiency (note: Au is used only as composites). Data points are labelled with references.

Zinc (Zn. The comparable ionic radii of Zn²⁺ (0.74 Å) and Ti⁺⁴ (0.75 Å) make Zn²⁺ the most effective choice among all the transition metals. Due to this, Zn²⁺ can easily substitute Ti⁴⁺ without altering the TiO₂ crystal structure.¹⁶⁰ Zn doping also decreases the bandgap due to the creation of energy bands between the bandgap of TiO₂. Zn doping also creates pores on the surface film. The porous structure aids dye adsorption and provides a high volume ratio for incrementing charge collection and transportation efficiency on the surfaces of Zn–TiO₂ thin films.^161–164 Zn–TiO₂ exhibits suitable photocatalytic properties owing to its porous surface and introduction of conductive channels. The increment in short circuit current density can be attributed to the improvement in the incident light harvesting, enhancement in conductivity due to the generation of more electron–hole pairs, reduction in internal resistance, and reduction of electron traps.⁷⁴

Yttrium (Y). Y(III) contains one 4d occupied orbital and can quickly produce several electron configurations.¹⁶⁵ The larger ionic radius of Y³⁺ (0.90 Å) relative to the Ti⁴⁺ radius (0.61 Å) prevents the easy penetration of Y into the TiO₂ lattice. The effect of Y doping on the crystallite size is reported differently by different research groups. For easy penetration into a porous film, X. Qu et al. reported the fusing yttrium nitrate solution and observed that Y₂O₃ covered the surface of TiO₂ or, if annealed, formed the sub-grains among grain boundaries.¹⁶⁶ A. K. Chandiran et al. reported the shrinkage of crystallite size along [001] on Y doping and suggested the accommodation of Y³⁺ at the interstitial site of TiO₂.¹⁴¹ B. Zhao et al. reported the expansion of the TiO₂ lattice by incorporating Y³⁺.¹⁶⁵ Such an increment in specific surface area and pore volume helps more extensive dye loading. Such structures can enhance light harvesting capability by increasing the number of excited electrons. When dopant levels of Y-doped anatase TiO₂ are created near the conduction band, the f electrons from Y³⁺ are transferred to the conduction or valence band of TiO₂. This results in a lowering of optical bandgap energy compared to pristine TiO₂. The increased conductivity of the TiO₂ lattice on Y doping can be attributed to the introduction of a point defect. Low electrical resistance is another side effect of moderate trivalent dopant concentration, which prolongs the effects of in-Y doping on the short-circuit current and fill factor while barely affecting the photovoltage. It has also been reported that Y substitution has a negligible impact on the energy distribution and density of the trap states available below the conduction band.¹⁴¹ The formation of the Ti–O–Y complex can lead to the generation of charge imbalance of the lattice. To compensate for the charge imbalance, more recombination of –OH with –COOH radicals can occur when more –OH radicals are absorbed. This results in the improvement of dye loading capacity.

Zirconium (Zr). Similar to titanium, Zr is a quadrivalent element with a slightly larger cation radius. This suggests that Zr doping can either replace Ti⁴⁺ or locate in the interstitial sites. Introducing Zr in TiO₂ reduces the crystallite size and increases the porosity of Zr-doped TiO₂ electrodes, enhancing the dye adsorption capacity. The decrease in crystallite size on Zr doping is attributed to the segregation of Zr⁴⁺ in the grain boundaries of anatase titania grains, implying the location of Zr⁴⁺ at interstitial locations. The negative impact of a highly porous structure on Zr doping can deteriorate the cell performance. Large porous structures can act as electron traps and can be associated with the depth of traps that participate in the electron motion, affecting the diffusion coefficient value. Thus, an optimum porosity is highly required, implying an optimum doping concentration of Zr.

Another exciting feature of Zr doping is that this transition metal accelerates the anatase to rutile transformation.¹⁶⁷ Doping of anatase with elements have valency smaller than 4⁺ provides a charge compensation by forming vacancies, amplifying the ionic transport in anatase, thereby increasing the anatase to rutile transformation. However, ions with a valence higher than 4⁺ form Ti³⁺ cations and hinder the change. This high rutile content for Zr-doped TiO₂ also leads to a slighter difference between the Fermi level and the redox potential of the electrolyte. A recent research study reported that Zr-doped TiO₂ exhibits the lowest bandgap relative to Fe-doped, Ni-doped, and undoped TiO₂. It also suggests that Zr gets doped into TiO₂ on the surface and inside the lattice.⁷² Zr transition metal doping can improve photocatalytic performance at higher rates. The decrease in bandgap implies that it enhances the light absorption capacity of the electrode. A decreasing trend in the V_oc can also be observed at a higher level of Zr concentration, suggesting an increase in recombination processes.

Niobium (Nb). The most extensively researched dopant for TiO₂ is niobium, which is incredibly promising for DSSCs. Because of its superior optical and electrical properties, it is an ideal dopant for TiO₂ in photovoltaic applications.^56,168–170 XPS analysis reveals the shift of the Fermi level away from the conduction band minimum (CBM) for low Nb content attributed to the passivation of oxygen vacancies at the surfaces of the TiO₂ nanoparticles but towards the CBM for high Nb content. T. Nikolay et al. reported that the inclusion of Nb inhibits the formation of rutile phase TiO₂ while supporting the growth of the anatase phase of TiO₂.¹⁷¹ Low Nb doping leads to a slight shift of Fermi energy and a decrease in the concentration of oxygen vacancies on the TiO₂ surface. The passivation of oxygen vacancies by Nb available at the surface site of the lattice occurs because of the attraction between oxygen and Nb. The dangling bonds of the attracted oxygen neutralize the donor defects of Nb. This results in the suppression of recombination at the sturdy junction of the TiO₂/electrolyte interface due to the widening of the space charge region. This also plays a vital role in enhancing the open circuit voltage for low Nb doping levels. Nb doping has been seen to lower the CBM relative to the undoped sample which can impact the electron injection from the dye to the conduction band of TiO₂.

The optimum amount of Nb dopant results in the substitution of Ti⁴⁺ with Nb⁵⁺ so that the Ti³⁺ 3d¹ and Nb⁵⁺ 4d⁰ levels exist in the nanostructure, resulting in the formation of intraband states in the structure. This improves the charge transport and driving force for electron injection from the dye to the TiO₂ conduction band.^171,172 Also, the Ti 3d orbital has one excess electron for each Nb⁵⁺ substitution, thus implying that an increase in electron concentration improves electron conductivity. Such improvements in electron conductivity and increment of electron injection induced by the positive shift of the CBM enhance J_sc in addition to incident photon-to-current conversion efficiency, and the suppression of surface recombination at the TiO₂/electrolyte interface along with a shift in Fermi energy results in the influence of V_oc. Doping TiO₂ with Nb also results in the decrease in bandgap. H. Su et al. demonstrated the long-term stability of Nb doped photovoltaic devices by a 1200 h-aging test.¹⁴² The group showed a 10–20% improvement in efficiency and J_sc of Nb doped cells relative to undoped cells. Fig. 15 illustrates the influence of Nb doping on the average size of TiO₂ nanoparticles. Similarly, Fig. 16 shows the variation of V_oc, J_sc, FF, and efficiency with time of the Nb-doped TiO₂-based DSSCs.

Fig. 15 Influence of Nb doping on the average size of TiO₂ nanoparticles.¹⁷³

Fig. 16 Variation of V_oc, J_sc, FF and efficiency with time for Nb doped DSSCs. Reproduced from ref. 142 with permission from Elsevier.¹⁴²

Nb loading influences the parent structure's morphology, pore volume, and porosity. Parameters like J_sc, V_oc, and FF determine the photoelectron conversion efficiency. To achieve higher efficiency, the primary requirements are higher J_sc and larger V_oc. J_sc can be determined by factors listed below:

(1) Sunlight harvesting efficiency – it is directly related to the surface area of photoanode structures that determines light scattering and dye adsorption abilities. Low Nb doping was found to be suitable for utilizing solar light effectively. However, the absorbance property was reduced, implying a reduction in surface area. With high doping, in addition to the reduction in light scattering, the structure also suffered from the reduction of dye adsorption. Thus, it has been concluded¹³⁸ that the enhanced conversion efficiency should not be determined by morphological modification of Nb-doped TiO₂ composite structures.

(2) Photoelectron injection efficiency – with the increase in Nb dopant, the flat band potential (V_fb) exhibited a positive shift, implying an increase in photoelectron injection (Fig. 17). The active and easy electron injection resulted in an improvement in photocurrent density. The shape and defect of the structure influence the V_fb. A smaller size might induce a positive shift of V_fb, thus increasing the photoelectron injection efficiency.

Fig. 17 Nb dopant crystallite size and V_fb.¹³⁸

(3) Electron collection efficiency – it is determined by the electron transport ability, electron transfer resistance, and electron recombination resistance. Thorough experiments are required to determine optimal doping to attain maximum electron collection efficiency (Fig. 18).

Fig. 18 Nd dopant electron lifetime, recombination resistance and electron transfer resistance correlation.¹³⁸

Another work reported the use of a blocking layer along with a Nb-doped TiO₂ photoanode structure.¹⁷⁴ DSSCs with a blocking layer exhibited pronounced increments in V_oc and J_sc, which is attributed to the suppression of the electron recombination process. The advantage of using a blocking layer is that it prevents backward electron transfer at the FTO/electrolyte interface. An alternative study focusing on annealing revealed an increase in efficiency when annealing was followed by H₂O₂ treatment.¹⁷⁵Fig. 19 shows how the PCE of DSSCs using Ti–Nb photoanodes with and without H₂O₂ treatment after annealing under different conditions and the corresponding improvement in efficiency percentage of Ti–Nb compared to Ti photoanodes.¹⁷⁵

Fig. 19 The conversion efficiency of DSSCs using Ti–Nb photoanodes with and without H₂O₂ treatment after annealing at (a) 300 °C and (b) 450 °C for 30 and 480 min and (c) the improvement in efficiency percentage of Ti–Nb compared to Ti photoanodes.¹⁷⁵.

Silver (Ag). Plasmonic photoanodes have significant benefits, including easy fabrication, minimal use of photoanode materials, and high potential for incredibly high light absorption through the localized surface plasmon resonance (LSPR) effect, making them potential candidates for investigation.¹⁷⁶ The added advantage of the electron sink effect or photocharging effect in metallic nanoparticles improves the electron lifetime, lowers the recombination rate, and enhances electron transfer. A study showed that Ag doping increased the size of TiO₂, indicating the facilitation of interstitial substitution in the tetragonal lattice.¹⁷⁷ Another work reported that despite the decrease in the surface area with the introduction of Ag, the doped photoanode did not negatively impact the dye-loading capacity.¹⁷⁶ In contrast to the usual trend, the unexpected impact is attributed to the modification of trap states. The Ag atom being a donor atom contributes to the shifting of Fermi energy toward the conduction band. As compared to a single layer and an undoped layer of TiO₂, thicker layers of TiO₂ with Ag doping at the bottom layer exhibit a smaller bandgap. An optimum film thickness is highly recommended to facilitate high interfacial surface area, a low recombination rate, and easy mass transfer. With the dopant being a good conductor, it can be ensured that the conductivity of the doped material is enhanced manifold by improving the electron transfer rate. Additionally, the multiple layers acting as blocking layers between the electrolyte and conducting substrates cease the recombination of holes and electrons, thereby facilitating increased conductivity. The dye and plasmonic nanoparticles show a synergistic impact through an improvement in absorption. P. Nbelayim et al. showed in their work that the absorption enhancement effect is indeed complementary; i.e., plasmonic nanoparticles aid in achieving a broadband optical absorption effect with a significant enhancement effect in the regions where the dye exhibits weak absorbance (Fig. 20).¹⁷⁶

Fig. 20 Optical absorption spectra showing enhanced broadband light absorbance of the N719 dye: (a) absorbance of [0.12 mg ml⁻¹ dye] – absorbance of [dye-only]. (b) The absorbance of [0.12 mg ml⁻¹ dye] ÷absorbance of [dye-only].¹⁷⁶

An optimal concentration of Ag aids in enhancing the short circuit current density and reducing in V_oc, which is attributed to the optimal lowering of the Fermi level of the photoanode and efficient electron injection. Optimal doping concentration is necessary to balance the positive and negative effects of plasmonic doping. The advantages are improved charge transport and lower recombination owing to the electron-sink effect. The disadvantage is poor charge injection efficiency due to increasing Fermi levels.

Cadmium (Cd). Compared to undoped TiO₂ samples, cadmium (Cd) doped TiO₂ exhibits a small improvement in UV and visible light absorption. M. Motlak et al. reported that Cd doping does not have any impact on the bandgap. Cd doping, when it leads to the formation of CdO in the lattice structure, significantly reduces the electron–hole recombination rate. Thus, it can be concluded that Cd doping could not strongly influence the electrode's optical properties.¹⁴⁴

Tantalum (Ta). Ta has a 5⁺ valence, and it has one more electron than Ti. Therefore, introducing Ta increases the electron concentration and the conductivity of undoped TiO₂.¹⁷⁸ The doping of TiO₂ with Ta results in the following effects: (1) a positive shift of the flat band potential, (2) an increase in electron density (from (TiO₂) = 8.43 × 10¹⁸ cm⁻³ to ND (TaTiO₂) = 1.89 × 10¹⁹ cm⁻³), (3) a decrease in open circuit voltage, (4) improvement of the driving force of injected electrons from the LUMO of the dye to the conduction band of TiO₂, (5) acceleration of the transfer rate of the electrons and (6) improvement of the fill factor of the solar cell and reduction of the bandgap. Due to the d–d transitions and various gap states, the Ta dopant induces redshift and increment in the absorption of the visible spectral regime. Another work showed inverse opal Ta–TiO₂ electrode structures that could support infiltration of the hole-transport material in films due to large, interconnected cavities/pores and achieved a 23% increment in photoconversion efficiency.¹⁷⁹ These structures also had the added advantage of trapping light due to their sizeable inner scattering. The doping concentration was optimized to 3.4 atom% to avoid the decrease in electron mobility at high defect concentration and to avoid increased charge recombination by the defects and increased electron recombination. With vertically aligned nanostructured films of tantalum-doped TiO₂ nanoparticles, the DSSC achieved a 65% increase in photocurrent and around 39% improvement in overall cell efficiency compared to undoped TiO₂. Doping TiO₂ with Ta does not impact the percentage of dye adsorption.

Tungsten (W). The radius of W(VI) (0.06 nm) is comparable to the radius of Ti⁴⁺ (0.0605 nm), thereby making the doping of W into TiO₂ feasible. The high ionic conductivity of W⁴⁺ suggests it to be a favorable dopant to form titanium dioxide-based photoanodes. Increasing W doping at a low level positively shifts the conduction band of TiO₂ on the one hand and acts as a trapping site for retardation of charge recombination on the other hand. The positive shift of the CB is expected as WO₃ has a lower CB edge relative to TiO₂. W doping suppresses the charge recombination, which, in addition to the downward movement of the conduction band, increases the driving force for electron injection and thus results in the remarkable improvement of the short circuit photocurrent and charge collection efficiency. Therefore, it is preferable to switch to W doping when the dye used for the cell has an insufficiently high LUMO compared to the CB in comparison to the undoped TiO₂. W dopant also significantly improves the electron lifetime. The positive conduction band shift and electron lifetime increment result in almost constant V_oc for a shallow doping level. Nevertheless, for higher doping levels, the V_oc decreases.

Fig. 21(a)–(c) and Table 3 summarize the performance of dopants which increased the efficiency by more than ten percent. However, dopants that increased the PCE by less than 10% are also included in Table 1. As seen in Fig. 21, copper, zinc, and zirconium increased the efficiency drastically when doped.^67,74,180 However, Ag, Cr, Nb, Ta, W, and Y showed an increase in PCE (locked below 50%).^{146,166,181–184} At the same time, in a different study, the highest-performing zirconium and copper showed a decrease in PCE.^154,185,186 Other than these, Ag, Fe, Ni, Mn, and Co also reduced the efficiency of DSSCs compared to the control DSSC when doped into a TiO₂ photoanode.^72,135,187

Fig. 21 (a) Performance of transition metals when doped into TiO₂ for DSSC fabrication. (b) J_sc and V_oc of transition metals when doped into TiO₂ for DSSC fabrication. (c) Reduction in PCE when transition elements are doped into TiO₂ for DSSC fabrication. Data points are labelled with references.

Table 3 Current statistics of transition metal doping in TiO₂

S. no.	Dopant	Treatment/coating	Structure	Synthesis method	V oc (V)	Photocurrent density (mA cm^−2)	FF	PCE (%)	Enhancement in PCE (%)	Bandgap (eV)	Application	Sensitizer/dye	Ref.
1	Chromium (Cr)		Nanoparticles	Hydrothermal and doctor blade	0.705	11.34	0.69	6.35	14.62		DSSC	N719	149
2	Chromium (Cr)		Bilayers, NPs, nanotubes	Microwave assisted hydrothermal	0.69	18.75	0.66	8.69	47.79		DSSC	N719	150
3	Chromium (Cr)		Nanoparticles, nanotubes	Microwave assisted hydrothermal	0.69	26.29	0.6	11.05	22.10		DSSC	N719	65
4	Chromium (Cr)		Nanoparticles, nanotubes	Sol–gel, spin coating	6.98	17.54	0.609	7.47	44.77		DSSC	N719	182
5	Cobalt (Co)		Nanoparticles	Hydrothermal, doctor blade	0.6	3.12	0.571	1.06	−79.89		DSSC	N719	135
6	Cobalt (Co)		Nanoparticles	Microwave assisted hydrothermal	1.201	6.87	0.59	4.85	8.74		DSSC	N719	157
7	Cobalt (Co)		Brookite	Spray pyrolysis	0.87	11.04	0.59	5.66		3.3	DSSC	N719	156
8	Copper (Cu)		Core–shell nanostructures, nanowires	Doctor blade	0.75	18.26	0.68	9.44	23.40		DSSC	N719	75
9	Copper (Cu)		Nanowalls	Liquid phase deposition	0.64	1.8	0.0038	0.44	120.00	3.32	DSSC	N719	67
10	Copper (Cu)		Nanoparticles	Microwave assisted hydrothermal	0.762	13.2	0.689	6.94	19.45	2.96	DSSC	N719	159
11	Copper (Cu)		Nanoparticles	Microwave assisted hydrothermal	1.073	7.34	0.65	5.09	6.71		DSSC	N719	151
12	Copper (Cu)		Nanoparticles	Sol–gel	0.71	18.8	0.642	8.65	34.95	2.7	DSSC	N719	188
13	Copper (Cu)		Nanoparticles	Sol–gel, hydrothermal	0.76	10.68	0.585	5.24	−10.88	3.23	DSSC	N719	189
14	Gold (Au)		Thin shell, hollow sub-microspheres	Hydrothermal	0.63	22.1	0.64	8.13	15.16		DSSC	N719	190
15	Gold (Au)		Nanoparticles		0.7	0.7	0.5611	6.51			DSSC	N719	191
16	Iron (Fe)		Nanoparticles	Microwave assisted hydrothermal	0.939	8.37	0.35	2.74	−38.57		DSSC	N719	157
17	Iron (Fe)		Nanoparticles	Sol–gel	0.29	0.22	0.0027	0.017	−10.53	3.13	DSSC	N719	72
18	Iron (Fe)		Nanoparticles	Sol–gel, hydrothermal	0.71	1.14	0.731	0.67	−88.61	3.24	DSSC	N719	189
19	Manganese (Mn)		Nanoparticles	Hydrothermal, doctor blade	0.656	4.241	0.667	1.85	−64.90		DSSC	N719	135
20	Manganese (Mn)		Nanoparticles	Microwave assisted hydrothermal	1.181	7.52	0.55	4.87	2.10		DSSC	N719	151
21	Nickel (Ni)		Nanoparticles	Microwave assisted hydrothermal	1.126	7.75	0.51	4.47	0.22		DSSC	N719	157
22	Nickel (Ni)		Nanoparticles	Sol–gel	0.31	0.17	0.0029	0.015	−21.05	3.15	DSSC	N719	72
23	Niobium (Nb)		Nanorods	Electrochemical anodization	0.615	14.43	0.583	5.17	33.25	2.8	DSSC	N719	184
24	Niobium (Nb)		Nanoparticles	Horizontal ultrasonic spray pyrolysis deposition (HUSPD)	0.74	16.9	0.6026	7.5	4.75		DSSC	N719	192
25	Niobium (Nb)		Nanoparticles	Hydrothermal	0.7	17.67	0.63	7.8	18.18		DSSC	N719	193
26	Niobium (Nb)		Nanoparticles	Hydrothermal	0.742	15.907	0.717	8.459	19.78		DSSC	N719	142
27	Niobium (Nb)		Nanoparticles	Hydrothermal	0.685	18.9	0.64	8.33	12.26		DSSC	N3	145
28	Niobium (Nb)		Porous microspheres	Hydrothermal method followed by heat treatment	0.599	14.5	0.575	4.99	13.67		DSSC	N719	131
29	Niobium (Nb)		Nanoparticles	Hydrothermal synthesis and screen printing	0.742	15.907	0.717	8.459	19.78		DSSC	N719	142
30	Niobium (Nb)		Nanoparticles	Screen printing	0.74	7.26	0.76	4.14	16.29		DSSC	N719	175
31	Niobium (Nb)		Nanoparticles	Sol–gel, co-hydrolysis, hydrothermal treatment	0.73	16.32	0.68	8	17.65	3.12	DSSC	N719	171
32	Niobium (Nb)		Nanoarrays, nanoparticles	Two-step hydrothermal, spin-coating approach	0.78	13.6	0.68	7.2	45.16		DSSC	N719	138
33	Scandium (Sc)		Mesoporous beads/nanoparticles	Controlled hydrolysis	0.752	19.1	0.675	9.6	6.67		DSSC	N719	127
34	Silver (Ag)		3D flower like microstructures	Hydrothermal, photoreduction	0.72	25.88	0.49	8.98	44.37		DSSC	N719	181
35	Silver (Ag)		Nanoparticles	Polyol, electrospinning	0.764	11.88	0.685	6.23	17.77		DSSC	N719	187
36	Silver (Ag)		Nanoparticles, nanofibres	Polyol, electrospinning	0.752	13.68	0.685	7.05	−1.26		DSSC	N719	187
37	Silver (Ag)		Nanoparticles	Sol process	0.744	8.93	0.591	5	7.30		DSSC	N719	176
38	Silver (Ag)	Ag coating	Nanoparticles		0.71	4.24	0.62				DSSC	N719	177
39	Tantalum (Ta)		Nanoparticles	Atomic layer deposition	0.71	3.69	0.6	1.56	21.88		DSSC	N719	179
40	Tantalum (Ta)		Nanoparticles	Hydrothermal	0.665	19.7	0.65	8.18	10.24		DSSC	N3	145
41	Tantalum (Ta)		Nanoparticles	Hydrothermal	0.665	19.1	0.65	8.18	10.54		DSSC	N3	178
42	Tantalum (Ta)		Nanoparticles	Laser ablation	0.69	15.9	0.61	6.7	39.58		DSSC	N719	183
43	Tantalum (Ta)		Nano spheres	Screen printing method	0.663	13.89	0.77	7.1	31.00		DSSC	N719	194
44	Tungsten (W)		Nanoparticles	Modified sol–gel	0.61	8.94	0.77	4.2	19.76		DSSC	A new coumarin dye	146
45	Vanadium (V)		Nanoparticles	Hydrothermal	0.687	17.6	0.65	7.8	5.12		DSSC	N3	145
46	Yttrium (Y)		Nanoparticles	Gel–sol	0.706	15.74	0.6846	7.61	19.97		DSSC	N719	165
47	Yttrium (Y)		Nanoparticles	Hydrothermal	0.739	15.9	0.77	9	21.62		DSSC, QDSSC	C101 Ru(+II)	141
48	Yttrium (Y)		Nanoparticles	Hydrothermal	0.81	23.9	0.4724	9.18	21.79		DSSC	N719	166
49	Zinc (Zn)		Nanowalls	Liquid phase deposition	0.66	10.68	0.28	1.98			DSSC	N719	195
50	Zinc (Zn)		Nanoparticles	Microwave assisted hydrothermal	1.097	7.85	0.52	4.49	0.67		DSSC	N719	157
51	Zinc (Zn), cobalt (Co)			Sol–gel	0.72	6.44	0.82	3.78	266.99	2.9	DSSC	N719	74
52	Zirconium (Zr)			Ball milling, dissolving, impregnation, mixing, screen printing	0.7067	10.527	0.75	5.42	−1.45		DSSC	N719	186
53	Zirconium (Zr)		Nanoparticles	Sol–gel	0.712	15.5	0.626	6.81	10.37	3.1	DSSC	N719	167
54	Zirconium (Zr)	CNT coating	Nanoparticles	Sol–gel	0.699	19.4	0.602	8.19	32.74		DSSC	N719	167
55	Zirconium (Zr)		Nanoparticles	Sol–gel	0.45	0.13	0.0034	0.02	5.26	3.09	DSSC	N719	72
56	Zirconium (Zr)		Nanofibers	Sol–gel, electrospinning	0.82	7.74	0.71	4.51	180.12	3.22	DSSC	N719	196

---

Even though doping research is thriving, associated bandgap data are relatively less noted. Fig. 22 shows the variation of bandgap and the corresponding effect on PCE with doping. When the bandgap is reduced, there is an increase in PCE and vice versa. Nevertheless, TiO₂ nanofiber does not follow this order. When the bandgap is higher, the PCE increases.

Fig. 22 Bandgap and increase in PCE with doping in transition elements. Data points are labelled with references.

3.6. Post-transition metals

Post-transition metals are poor metals with lower reactivity. They are also brittle and generally have low boiling points than transition metals. These include aluminum, gallium, indium, tin, thallium, lead, and bismuth. Except for thallium and lead, all other post-transition metals were doped into TiO₂ to increase conductivity, change the absorption to the visible region for better charge-hole separation, decrease the recombination, and tailor the surface defects for better charge transfer by changing dimensionality.^81,197–201

Aluminum (Al). Aluminum doping introduces oxygen vacancies into the TiO₂ lattice, increasing the carrier generation and improving the internal conductivity.^202–204 In addition to the increase in conductivity, Al doping introduces electron traps into TiO₂.

These traps effectively reduce the photocurrent through charge recombination. However, aluminum doping has a positive effect on DSSCs despite the drop in the photocurrent.²⁰² Not just in DSSCs, Al³⁺ doped TiO₂ has a positive influence on PCE when used as an electrode material in QDSSCs. But the study by R. Li et al. published in 2016 showed that doping aluminium into the mesoporous TiO₂ followed by 500 °C sintering increased the PCE significantly over the other two studies in 2011²⁰² and 2017.⁷¹ Aluminum doping into the mesoporous layer increased the PCE by around 67 percent. The incorporation of Al ions into the TiO₂ causes a change in the concentration of defect states. The flat band potential plays a vital role by increasing the electron injection rate and charge collection efficiency. Altogether, they contributed to the efficiency in the DSSC. Even though the efficiency is much lower in the study by R. Li et al., the effect of aluminum doping stands out.

Bismuth (Bi). Metal–nanocomposites can enhance the efficiency of DSSCs through electron transfer kinetics at the crystal surface, reducing the probability of electron–hole recombination. Bismuth doping through crystallization produced Bi-doped TiO₂ nanocubes. Doping and crystallization increased the efficiency of a doped TiO₂-based DSSC compared to the undoped TiO₂-based DSSC. In 2016, M. C. Wu et al. doped bismuth into TiO₂ nanofibers and used them as a scattering layer.⁶⁶ This study by M. C. Wu et al. has better efficiency than M. N. An’amt et al. study. Also the effect of doping is comparative in both studies.²⁰¹ Bi³⁺ replaces Ti⁴⁺. The ionic size difference between Ti⁴⁺ (0.061 nm) and Bi³⁺ (0.103 nm) causes a slight shift in the crystalline peak of the Bi-doped sample's XRD pattern. Doping with bismuth almost increases the PCE by 70 percent compared to the undoped TiO₂. Bismuth doping extends the absorption spectra of TiO₂ from the UV to visible region. Nanofibers scatter the light more than the bare TiO₂ scattering layer, contributing to the increase in PCE. It is clear from Fig. 23(a), (b) and Table 4 that all other post-transition metals have a minimal doping impact on increasing the PCE when compared to aluminium and bismuth.

Fig. 23 (a) Percentage increase in PCE in post-transition metals with doping. (b) J_sc and V_oc of post-transition metals when doped into TiO₂ for DSSC fabrication. Data points are labelled with references.

Table 4 Current statistics of transition metal doping in TiO₂

S. no.	Dopant	Treatment/coating	Structure	Synthesis method	V oc (V)	Photocurrent density (mA cm^−2)	FF	PCE (%)	Enhancement in PCE (%)	Bandgap (eV)	Application	Sensitizer/dye	Ref.
37	Aluminium (Al)		Mesoporous	Chemical bath deposition	0.705	16.22	0.6677	7.64	10.09		DSSC	N719	205
38	Aluminium (Al)		Nanoparticles	Doctor blade	0.702	16.5	65.75	7.66	9.12		DSSC	N719	205
39	Aluminium (Al)		Nanoparticles	Screen printing	0.57	11.49	0.318	2.08	45.45		QDSSC	CdS/ZnS	71
40	Bismuth (Bi)		Nanofibres	Hydrothermal	0.787	14.45	78.2	3.57	63.76		DSSC	N719	66
41	Bismuth (Bi)		Nanocubes	Sol–gel hydrothermal	0.59	7.71	0.46	2.11	77.31		DSSC	N3	73
42	Gallium (Ga)		Nanoparticles	Hydrothermal	0.755	13.4	0.79	8.1	9.46		DSSC, QDSSC	C101 Ru(+II)	141
43	Gallium (Ga)		Nanoparticles	Hydrothermal	0.71	12.44	0.51	4.57	45.08		DSSC	N3	206
44	Indium (In)		Nanoparticles	Simple surface doping technique by immersing TiO2 films with In^3+ acidic solution at different soaking times at 70 °C followed by sintering at 450 °C	0.8	14.33	0.63	7.19	18.06		DSSC	C264 triphenylamine dye	199
45	Indium (In)		Nanoparticles	Sol–gel, spin coating	0.735	16.384	0.661	7.96	13.55		DSSC	N719	207
46	Tin (Sn)		Nanoparticles	Hydrothermal	0.714	17.14	0.71	8.66	13.80		DSSC	N3	69
47	Tin (Sn)			Hydrothermal	0.722	16.01	0.707	8.31	11.54		DSSC	N3	208
48	Tin (Sn)	Post-treated with Zr and HNO3	Nanorods	Sol–gel, hydrothermal	0.749	9.81	0.674	4.96		2.99	DSSC	N719	81
49	Tin (Sn)	Post-treated with Zr and HNO4	Nanorods	Sol–gel, hydrothermal	0.747	5.21	0.534	2.09		2.99	DSSC	Leaves of Camellia sinensis	81
50	Tin (Sn)		Nanoparticles	Sol–gel, hydrothermal	0.76	12.76	0.577	6.24	6.12	3.27	DSSC	N719	189

---

Gallium (Ga). Gallium's ability to increase the photocatalytic effect on doping into TiO₂ attracted the attention of J. Chae and his group in 2011. According to J. Chae et al., while doping, Ga³⁺ is inserted into the Ti⁴⁺ sites and also between the lattices. This insertion helps in donating electrons to the neighboring oxygen ions. AFM of doped samples agrees with this observation. The free ions generated due to the charge balance will improve the electron flow more easily to the conduction electrode and FTO glass. The improved electron flow leads to better electron movement compared to the pure TiO₂ electrode, resulting in an increase in the PCE of DSSCs.²⁰⁹ Immediately after this article, in 2011, gallium was again considered for doping. Trivalent elements in the TiO₂ lattice can create intermediate acceptor levels and improve the charge collection efficiency. This can enhance the PCE without compromising the V_oc.

Indium (In). Significant recombination occurs in contacts such as TCO and TiO₂ interfaces, which leads to the introduction of indium into TiO₂. Indium is also well known for its low toxicity and better transparency. Doping TiO₂ with indium leads to better electronic properties and decreases the DSSC's internal resistance when used as an electrode material. B. Baptayev et al. surface doped TiO₂ with indium (In³⁺) using a simple immersion technique. An optimum surface doping in TiO₂ positively affects the V_oc and FF, thus the PCE. At the same time, higher surface doping leads to faster recombination through an increased number of surface trap states. Nevertheless, on comparing these two works, it was found that doping indium into TiO₂ and applying above FTO resulted in a slightly high percentage increase in PCE than doping indium into the mesoporous layer of DSSCs. However, indium surface-doped DSSCs have somewhat lower efficiency than bulk doping.¹⁹⁹ Recent research on indium doped TiO₂ showed a remarkable increase in TiO₂ electronic properties. Indium's ability to be transparent and low toxicity became an added quality to consider it for doping into TiO₂. Sol–gel produced TiO₂ photoanodes showed a 17.4% increase in PCE. This increase is due to the decrease in recombination and internal resistance. An elaborate recombination study was done by Baptayev et al., guiding us through the indium doping and the cell's internal resistances. By setting the dark conditions and voltage to a bare minimum of 0.4 V, they analyzed the voltage decay of doped samples one by one to eliminate the effect of recombination through back contact. The study showed that doping has a positive impact on reducing recombination. According to Baptayev et al., there are three mechanisms for charge recombination:

(1) Direct transfer from the conduction band to electrolyte.

(2) Indirect transfer from monoenergetic deep surface states.

(3) Indirect transfer from the distribution of bandgap surface states.

In the indium doping case, the surface trap states are dominant trap states and lead to decreased recombination centers, eventually benefiting the PCE.¹⁹⁹

Tin (Sn). Tin is the most explored element among post-transition metals doped into TiO₂.^208,210,211 From 2012 to 2016,a series of TiO₂–Tin doping studies were conducted. When Sn increases in the TiO₂ lattice, the oxygen vacancies increase. Subsequently, the oxygen vacancies favor photocatalytic reactions.⁸¹ When Sn⁴⁺ is doped into the TiO₂ lattice, the potential of the flat band (E_fb) shifts depending on the doping concentration. A negative shift in E_fb means a shift in redox potential. However, efficiency increases when the redox potential varies with increasing open-circuit voltage.²⁰⁸ The source of tin doping has a significant effect on efficiency. Y. Duan et al. strongly suggested choosing an appropriate source when doping TiO₂ in DSSCs.⁶⁹ R. T. Ako et al. also agreed with the increase in defect concentration while doping Sn into TiO₂. This lower defect concentration resulted in slower transient states and consequently less recombination. d–d transitions of untrapped electrons also had a longer excitation lifetime.¹⁸⁹ Y. Akhila et al. doped tin into TiO₂ nanorods and prepared nanorods with various Sn concentrations. However, in their study, 1% Sn doping was found to increase oxygen vacancies in TiO₂, favoring photocurrent conversion.⁸¹

Analyzing PCE to understand doping is deceiving because it does not give any information about what it does to efficiency. Meanwhile, an increase in PCE provides vital information regarding the effect of doping. From the graph, it is clear that indium and tin have higher PCEs owing to their preparation methods. Aluminum, bismuth, and gallium when doped into TiO₂ have significantly changed the DSSC's efficiency.

3.7. Halogens

Among halogens, fluorine and iodine showed effective narrowing of the bandgap attributed to their higher ability to absorb visible light. This observation is supported by DFT calculations of n-type iodine-TiO₂ doping and p-type nitrogen-TiO₂ doping. Comparatively, iodine doping improved the conductivity and increased the visible light absorption of anatase TiO₂. From 2011 on, iodine and fluorine alternatively have become the best dopants for TiO₂ from the halogen group. The statistical information about halogen doped TiO₂ based DSSCs is summarized in Table 5.

Table 5 Information about halogen doping in TiO₂

S. no.	Dopant	Treatment/coating	Structure	Synthesis method	V oc (V)	Photocurrent density (mA cm^−2)	FF	PCE (%)	Enhancement in PCE (%)	Bandgap (eV)	Application	Sensitizer/dye	Ref.
1	Fluorine (F)		Nanoparticles	Hydrolysis	0.67	13.89	0.57	5.24	10.08		DSSC	N719	218
2	Fluorine (F)		Nanocuboids	Hydrothermal	0.603	17.621	0.6984	7.463	62.45	3.1	DSSC	N719	84
3	Fluorine (F)		Hollowspheres	Hydrothermal	0.754	11	0.761	6.31			DSSC		215
4	Fluorine (F)		Quasi-cubic structures	Microwave assisted solvothermal	0.745	18.74	0.59	8.2	63.35		DSSC	N719	219
5	Iodine (I)		Nanocrystals	Hydrolysis, hydrothermal	0.715	14.1	0.67	7	42.86	2.4	DSSC	N3	212
6	Iodine (I)		Spheres	Hydrothermal, screen printing	0.616	13.81	0.75	6.38	34.03		Quasi-solid state DSSC	N719	220

---

Iodine (I). In 2011, Hou and coworkers reported a higher photon conversion efficiency of 10 percent by doping TiO₂ nanocrystals with iodine (I-TNC). In the iodine doped samples, recombination of electron–hole pairs may be inhibited due to the ability of iodine to act as a trapping site for photogenerated electrons. Even though the article strongly suggests iodine as a futuristic dopant, it requires focused research to understand the energy conversion mechanism.²¹² Iodine exhibits multiple valencies of I⁻, I⁵⁺, and I⁷⁺. Nevertheless, XPS results indicated that I⁵⁺ (0.062 nm) doped into the titania matrix and substituted Ti⁴⁺ (0.064 nm). Compared to the bare TiO₂ samples, the iodine doped samples exhibited an extended response in visible light absorption. By reducing the recombination and improving the conductivity, I⁵⁺ doping increased the J_sc of DSSCs.^213,214

Fluorine (F). Song and coworkers created a multifunctional electrode with F-doped TiO₂ hollow nanoparticles to reduce charge recombination and contact resistance. However, the lack of details on undoped DSSCs makes it unable to distinguish whether the method stands out. It can be seen that there is no significant difference in surface area in F-doped hollow TiO₂ spheres and undoped samples. This F doping is an example of retaining the morphology and structure of TiO₂ even after doping; thus, there will not be a dramatic change in the dye loading or light scattering. However, there is a 13% increase in the efficiency of a DSSC with F-doped TiO₂.²¹⁵

The idea of high-quality blocking layers that are non-metallic leads to the use of fluorine as a dopant for TiO₂. Noh and coworkers introduced a new concept of dense layers into DSSCs for creating a non-ohmic layer between the nanoporous TiO₂ layer and FTO glass. High interfacial resistance has hindered the efforts to develop adequate blocking layers. For example, Nb doping on the TiO₂ dense layer increased the conductivity of the SrTiO₃ substrate, but the thermal instability and production cost were high. Compared to the undoped samples, this work shows that the doping effect on DSSCs is substantially less.

Nevertheless, considering that this is one of the first works in fluorine-doped TiO₂, this study used a low-cost method and materials for doping and successfully introduced a fluorine-doped dense layer. This layer prevented back electron transfer, a breakthrough in DSSC research.²¹⁶ Fluorine (F⁻) is a superior anionic dopant due to its high electronegativity. It disrupts the CB and inhibits electron recombination by its strong electronegativity. Fluorine ions can penetrate the lattice and induce Ti³⁺ species generation, which helps in charge collection efficiency. Fig. 24 shows that the fluorine content increases in the TiO₂ electrode and the bandgap increases. Consequently, the efficiency of the DSSC increases. On the optimal doping we discuss throughout this paper, at 6% doping, the F-doped TiO₂-based DSSC reaches maximum efficiency. Beyond this doping point, fluorine doping adversely affects the DSSC's efficiency. F-doped TiO₂ nanocubes with a nanovoid/pore morphology play a crucial role in confining light and favoring electron transfer. Nanocubes increase the surface area, thus enhancing dye loading and light trapping. These naturally result in the percentage improvement in the efficiency of the F-doped TiO₂ based DSSC compared to the bare TiO₂-based DSSC. The presence of nanovoids at the surface of TiO₂ influences the grain boundary resistance by increasing the electron concentration. Consequently, this leads to an increase in charge transportation rate. However, the F-doping concentration increases with space charge polarization effects due to intensive charge carriers, resulting in more recombination.²¹⁷

Fig. 24 Correlation of the percentage of F doping with bandgap, particle size and efficiency.

Defects generated by mesoporous TiO₂ result in numerous electronic sub-bandgap trap states.²²¹ Peng Xiang et al. doped TiO₂ with iodine and applied it in a quasi-solid state DSSC. From Fig. 25(a), it is clear that iodine doping is vital in improving the cell performance by 34 percent, owing to the distributed pores, which increase the surface area. Nevertheless, compared to fluorine, the percentage efficiency improvement achieved is insignificant in this study.²²²Fig. 25(b) provides the comparison of V_oc and J_sc data of halogen doped TiO₂ based DSSCs. Recently, microwave-assisted solvothermal doping of fluorine into TiO₂ was explored. This preparation method is very competitive with all other fluorine doping techniques. Compared to other methods, this method is non-toxic and non-corrosive as well. In this study, F-doped TiO₂ forms a quasi-cubic morphology due to the fluorine iron doping. F doping introduces mesoporous structures. The ability of fluorine to separate charges is crucial in DSSC efficiency. Fluorine doping also induces facets that have high surface energy and lead to the formation of heterojunctions. This improves the photogenerated carrier density. The high adsorption capacity of fluorine creates Ti–F and Ti–F–Ti bonding by replacing oxygen. This action favors charge separation.⁸³

Fig. 25 (a) Comparison of the PCEs of different halogen doped TiO₂ based DSSCs from 2011. (b) J_sc and V_oc of halogen doped TiO₂ based DSSCs. Data points are labelled with references.

Since 2011, halogen group elements have been used as prospective candidates for improving the conductivity of photoanodes for DSSCs. Researchers successfully doped TiO₂ into blocking layers into various TiO₂ structures such as hollow spheres, nanocuboids, and mesoporous structures. Since the introduction of hierarchical TiO₂ layers and quasi-solid-state DSSCs, fluorine doping increases the conductivity, improves dye absorption, and quickens electron transport. Around 2020, researchers looked into the possibilities of non-corrosive, non-toxic, green synthesis routes and green precursor materials like polyvinylidene fluoride (PVDF).

3.8. Other nonmetals

Non-metal ions are thermally stable and suitable candidates for engineering the bandgaps by shifting the band edges compared with metal ionic dopants. Nitrogen is the most attractive non-metal for doping into TiO₂ due to its comparable atomic size with oxygen. In 2009, Soon Hyung Kang and co-authors tried to enhance the DSSC photocurrent using a nitrogen-doped TiO₂ anode. They incorporated urea into the sol–gel technique to obtain nitrogen-doped TiO₂ for photovoltaic applications. Compared to the bare TiO₂ samples, these samples were smaller in crystallite size. DSSCs fabricated with optimally doped samples exhibited an efficiency of 4.86%. Even though nitrogen is a frequent dopant, sulphur doped TiO₂ was also used to prepare DSSCs for achieving better efficiencies.^223,224 An increase in PCE of DSSCs when TiO₂ is doped with all other non-metal elements is depicted in Fig. 26. Statistics of other nonmetal doped TiO₂ based DSSCs are provided in Table 6.

Fig. 26 Increase in PCE with other nonmetal category dopants. Data points are labelled with references.

Table 6 Information about doping other non-metals into TiO₂

S. no.	Dopant	Treatment/coating	Structure	Synthesis method	V oc (V)	Photocurrent density (mA cm^−2)	FF	PCE (%)	Enhancement in PCE (%)	Bandgap (eV)	Application	Sensitizer/dye	Ref.
1	Carbonate		Microspheres	Solvothermal	0.76	16.6	0.625	7.65	48.26		DSSC	N719	227
2	Graphidyne (GD)		Nanoparticles	Homocoupling reaction, ball milling	0.8	13.73	0.7434	8.03	20.92	2.89	DSSC	N719	228
3	Nitrogen (N)		Nanoparticles	Doctor blade	0.73	18.76	0.58	7.98	17.01		DSSC	N719	229
4	Nitrogen (N)		Nanofibres	Electrospinning process, hydrothermal treatment	0.75	11.16	0.56	4.7	205.19		DSSC	N719	225
5	Nitrogen (N)		Nano spheres	Hydrothermal and doctor blade	0.474	12.03	0.64	3.67	71.50		QDSSC	CdSe	224
6	Nitrogen (N)		Nanoparticles	Sol–gel	0.726	10.52	0.636	4.86	30.29		DSSC	N719	230
7	Nitrogen (N)		Nanotubes	Solvothermal	0.67	8.82	0.4266	2.53	66.45		DSSC	N719	226
8	Nitrogen (N)		Nanoparticles	Wet chemical	0.7426	9.52	0.6854	4.84	77.94	3.13	DSSC	N719	231
9	Nitrogen (N)			Wet method, screen printing	0.83	12.4	0.7362	7.58	41.68	3.15	DSSC	N719	232
10	Sulphur (S)		Nanofibers	Sol–gel, hydrothermal, electrospinning	0.683	10.66	0.59	4.27	177.27		DSSC		223

---

Nitrogen (N). Recently, there has been interest in nanocrystalline materials such as nanoparticles, nanotubes, nanofibers, and mesoscopic spheres. TiO₂ nanoparticles have roughly 50 nm radii and behave like single atoms but are generally bulk materials called mesoscopic spheres. When doped TiO₂ took the form of mesoscopic spheres, the DSSC efficiency shot up to 10.7%. These nitrogen-doped mesoscopic spheres broadened the absorption from the UV region into the visible region. Applying this technology to QDSSCs achieved a two percent increase in efficiency through fast electron transport in the nitrogen-doped TiO₂ film. This remarkable achievement improved electron transport, charge collection efficiency, and photocurrent density.²²⁴ The direct conduction path for electrons and high surface area for dye absorption made nanotubes and nanorods desirable for DSSC applications. M. Motlak and his group developed nitrogen-doped pristine TiO₂ nanofibers as a photoanode material. They argued that nitrogen doping created a high Schottky barrier that facilitated electron capture. Nitrogen added intermediate electronic levels between the conduction and valence bands of the metal oxide.²²⁵ Nitrogen-doped nanofibers increased the efficiency by 3 percent compared to the normal TiO₂. V_oc and I_sc increased with nitrogen doping, but the increase in current was relatively higher.

Wang and his group incorporated 1D structures and nanoparticles and increased the efficiency percentage by 0.27 times that of a bare TiO₂-based DSSC. Depositing nanoparticles in the walls of nanotubes is effective for increasing efficiency. These aligned nanotubes and dispersed nanoparticles provide a higher surface area. This anode material can be further modified using nitrogen doping. By limiting electron and hole recombination, elements doped into TiO₂ can increase efficiency. Nitrogen-doped nanoparticles decorated with TiO₂ were obtained by the solvothermal method. Small nanocrystals along the wall were confirmed with SEM and HRTEM. This modified structure was treated with Ti(OH)₃, which was found to have a more significant influence on the morphology and photoelectric–chemical properties. The photoanode made with doped-nanoparticle-modified films showed improved light-harvesting efficiency. These nanoparticles that stuck on the nanotubes acted as light scattering centers, which reflected in the higher reflectivity. This Ti–O–N bonding, with nanotubes acting as vertical transport channels for photoinduced electrons and adsorbed nanoparticles together, increased the efficiency of DSSCs by 1.68 times the original efficiency.²²⁶ While the exact position of N remains debatable, Ti–O–N linkages point to the presence of N-ions in the TiO₂ lattice by replacing the oxygen atom.

When the scattering layer came under the spotlight and became an essential part of DSSCs, doping and modifying the scattering layer attracted more comprehensive attention. Mixing nitrogen-doped aggregated sub-micron size particles into commercial TiO₂, and using this as a scattering layer, increases the scattering percentage without any drawback in electrical properties. The IV characteristics of DSSCs with these mixed-phase scattering layers seem to be improved, and the efficiency is increased by 0.14 times that of the commercial scattering layer. Adding a scattering layer introduced the concept of multilayers to DSSC researchers. So, doping and introducing these multilayers into DSSCs has become the next challenge. Multilayers were introduced to increase the J_sc by adding the TiO₂ layer (scattering, P25, Doped P25) into the FTO substrate. Using a compact layer, a P25 layer, and a N-doped TiO₂ layer doubled the J_sc when compared with using an undoped TiO₂ layer in the DSSC. However, this reduced the fill factor slightly and increased the efficiency. The charge transfer resistance of the TiO₂/electrolyte interface is significantly less when a DSSC is prepared with N-doped samples. The resistance–efficiency relationship in these DSSC samples is commendably complex. When the TiO₂ layers use P25, the charge transfer resistance is meager. When undoped TiO₂ is replaced with P25 particles, nearly half of the original charge transfer resistance is increased. Even though doped multilayers considerably improved the solar performance of DSSCs, when N-doped layers are used in place of undoped TiO₂, the charge transfer resistance, ion diffusion resistance, and shunt resistance simultaneously decreased.²²⁹

In general, nitrogen induces a redshift in the absorption band edge due to the positive shift in flat band potential of TiO₂.^233–239 Incorporation of N³⁻ into the TiO₂ lattice can cause replacement of Ti⁴⁺ and fill up interstitial sites. Even when there is a higher J_sc due to increased dye absorption and improved electron injection, charge transport can be affected by lattice distortion.^239,240 However, most studies agree with increased charge transport rates and reduced carrier recombination.^{226,241–244}

Sulfur (S). Considering the fact that nitrogen has been researched thoroughly and is a very successful dopant for TiO₂, sulfur is also exploited to enhance the photoactivity of TiO₂. The presence of 3p orbitals in sulfur, together with TiO₂, has a significant role in narrowing the bandgap and increasing the valence bandwidth.²³⁰ Mahmoud et al. doped sulfur into hydrothermally prepared TiO₂ nanofibers and prepared DSSCs. IV-characteristics show a drastic increase in FF, J_sc, V_oc, and efficiency. Nevertheless, this study requires more data to confirm the effects of sulfur on DSSCs. To understand this doping effect in detail, it is essential to understand the dependency of IV characteristics on sulfur concentration.²²³ Even though sulfur has different oxidation states from S⁶⁺ to S²⁻, there is no evidence of S²⁻ (1.7 Å) replacing O²⁻ (1.22 Å) atoms; instead, S⁶⁺ (0.29 Å) replaces Ti⁴⁺ (0.64 Å). There is also a chance for random replacement of Ti⁴⁺ and O²⁻ by ions with polyoxidation states simultaneously, causing severe lattice distortions.^245,246 Over the past few years since 2009, we have seen a systematic approach to doping and using doped materials for implementing DSSCs. First, it was doping into a single TiO₂ layer, and later researchers started doping into tubes, structures, dense layers, and mesoporous particles.²⁴⁷

4. Co-doping

The light absorption, charge transport, and charge transfer characteristics of an anode determine the efficiency of an electrochemical cell. Successful doping of suitable elements into TiO₂ improves these fundamental properties. However, single elemental doping is inadequate to satisfy the requirements of TiO₂ as an anode material. These monodoped atoms act as recombination centers due to their partially occupied impurity-generated energy bands. Even though contradictory arguments and theoretical calculations exist against passivation theory, many researchers are still on board with passivation theory.²⁴⁸ Investigations by Dhonde et al. show that co-doping with more than one foreign atom can passivate or cause doping asymmetry to these impurity bands. In addition, co-doping effectively modulates the charge equilibrium.⁷⁹ Consequently, co-doping increases the surface area and shifts the band edge in TiO₂.⁸²

Alkaline earth metals, transition metals, post-transition metals, metalloids, and other non-metals investigated for codoping are marked in Fig. 27. Depending on the type of dopant, there are three types of strategies: metal–metal codoping, metal–nonmetal codoping, and nonmetal–nonmetal codoping.

Fig. 27 Elements explored for codoping into TiO₂ for DSSC applications.

4.1. Metal–metal codoping

Mono-metal dopants in TiO₂ create serious recombination centers and deteriorate carrier transport. Metal atom doping of TiO₂ causes the inhomogeneous distribution of dopants and requires high-temperature sintering, leading to particle agglomeration.²⁴⁹ Metal doping also causes rapid recombination of immobilized charge species and thermal instability of TiO₂. However, it improves band structure and shifts absorption towards a longer wavelength. Codoping focuses on increasing V_oc and J_sc simultaneously through the synergetic effect of dopants. The metal dopant combinations are Al–La, Ca–La, Eu–Tb, Mg–La, Ni–Zn, Nb–Fe, Sr–V, Sr–Cr, Zn–Mg, and Zn–Ga. Fig. 28 shows the effects of these metal codoped TiO₂ based DSSC efficiency enhancements.

Fig. 28 Effects of metal-doped TiO₂ in the efficiency profile of DSSCs. Data points are labelled with references.

Amid metal dopants, Zn–Ga combinational doping into TiO₂ exhibits the highest efficiency in DSSCs. Changes in the electronic structure of TiO₂ facilitate oxygen ion movement in the lattice, superior crystallization, and lattice expansion. During doping, Zn²⁺ and Ga³⁺ move into the lattice, facilitating the breaking and formation of new bonds. Zinc, an n-type dopant, causes a shift in Fermi level towards the conduction band, leading to an increase in the redox potential of electrolyte and a consequent increase in open-circuit voltage. Furthermore, the trap state distribution in TiO₂ has been reduced through zinc doping and has abetted the increase in short circuit current.²⁰⁰

A high concentration of trap sites reduced electron transport and formed an impurity-grazed bandgap in Al–La-doped TiO₂. Furthermore, insertion of Al³⁺ ions into the TiO₂ lattice increased the recombination rate and resulted in inefficiency. In Mg–La, Mg–Ca, and Mg–Al doping, the ionic radius plays a vital role in solar cell efficiency. Titanium has smaller ionic radii than La³⁺, Mg²⁺, and Ca²⁺. While doping, these atoms cannot penetrate the lattice. However, Al³⁺ quickly enters the lattice without further problems. Prominent ions, Ca²⁺ and La³⁺, cannot penetrate the TiO₂ lattice after doping.⁷⁷

Consequently, these bigger atoms remain on the surface, facilitating better charge transfer. Relative to these systems, an Al–La codoped system has inefficient charge transfer resulting in low cell efficiency. Mg²⁺ ions are slightly larger than Ti⁴⁺ ions. During doping, Mg²⁺ enters the lattice and causes slight distortions in the parent material lattice. Moreover, while larger atoms induce surface charge trapping, Mg²⁺ doping provides deep charge trapping in the materials. However, Mg²⁺ ion substitution substitutes Ti⁴⁺, resulting in band edge shift.⁷⁷

Ni–Zn doping into TiO₂ achieves more than a 50 percent increase in efficiency than the undoped material. This achievement requires tricky bandgap engineering. Ni produces a negative shift, and Zn causes a positive shift in flat band potential. Consequently, Ni increases the open-circuit voltage and reduces the short-circuit current. It is vice versa for Zn. Ni–Zn doping mainly shifts the fundamental absorption edge and reduces the band energy. However, finding the optimum point for the maximum V_oc and J_sc is challenging. Nb–Fe doping shifts the conduction band edge. Subsequently, the absorption changes from UV to visible light. Fe and Nb dopants serve as trap states and increase the electron lifetime via trapping and de-trapping events over trapping sites of Ti³⁺, Nb⁵⁺and Fe³⁺. Another research study on the same dopants suggests that Fe and Nb at ultralow concentrations generate Ti³⁺ donor trap states. These trap sites facilitate efficient charge diffusion and collection in titania nanobelts.¹⁴⁰

Sr,V codoped TiO₂ shows lower electron transit time, higher electron diffusion, and lower resistance, aiding efficiency. The increased charge separation and electron transformation demonstrated through EIS analysis confirm the doped sample's improved efficiency. The synergistic effect of Sr and Cr doping is vital for increasing efficiency. Cr incorporation into the TiO₂ lattice produces isolated energy levels adjacent to the conduction band, narrowing the band gap. Cr³⁺ has excess valence electrons, leading to the generation of holes near the valence band, further contributing to the reduction in the bandgap.²⁵⁰

When the crystallite size decreases, the efficiency increases in metal codoped TiO₂. Except for the Zn–Ga sample, this observation concludes that large crystallites have an adverse effect on efficiency. Analysing the specific Zn–Ga doping instance, data shows that the DSSC generated with the sample having a crystallite size of around 19 nm has the highest efficiency, in line with the most recent studies. A detailed investigation shows that when the crystallite size increases more than 20 nm, a DSSC exhibits a drastic decrease in DSSC efficiency (Fig. 29). Optimum crystallite size plays a significant role even in the doping-related efficiency profile. The notable point is that doping affects crystallite size in general. An optimum doping amount exists for each dopant, which decreases the crystallite size to an optimum size. Further doping from this point consequently reduces the efficiency and increases agglomeration.

Fig. 29 Codoped TiO₂ efficiency–crystallite size correlation. Data points are labelled with references.

All the doped TiO₂ samples except for Zn–Ga and Nb–Fe doped samples have nanoparticle structures. Nb–Fe doped samples are in the form of nanobelts synthesized via a hydrothermal method. Zn–Ga has a thin-film structure prepared and fabricated by sol–gel and dip coating. These structural differences might have contributed to the efficiency increase in the Zn–Ga doped TiO₂ sample. Compared with other sample preparation methods, the sol–gel technique is extensively used to prepare metal-codoped TiO₂ samples. Doctor blade coating, screen printing, and dip coating are the fabrication methods used to prepare TiO₂ anode materials for DSSCs.¹³⁴

4.2. Metal–nonmetal codoping

The thermal instabilities of metal-doped TiO₂ instigated the search for non-metal dopants for TiO₂. Nitrogen, fluorine, sulfur, and boron are successful non-metal dopants. Doping with these non-metals alone brings changes in photocatalytic activity. However, the latest research indicates that codoping can further improve photocatalytic activity and promote the transportation of photogenerated electrons through the hybridization orbitals. Non-metal codoping is still a feeble area in doping TiO₂. Yuanyuan and co-authors illustrated how doping increases the formation of the pure anatase phase rather than the rutile phase. A set of experiments with different concentrations of mesoporous N, S doped TiO₂ and P25 particles has been conducted to understand the effect of doping into TiO₂. Excess doping always worsens the DSSC's properties among purely doped TiO₂ and undoped TiO₂. This efficiency reduction is due to the larger particle size and lower surface area of the doped TiO₂ sample. While mixing the doped TiO₂ with undoped TiO₂, the cell efficiency increases from 5% to 20%, reaches a peak, and decreases. The 10% N, S doped TiO₂ sample mixed with 90% Degussa P25 shows a 35.6 percent increase in efficiency yield. This work claims that unique mesoporous nanoparticles and hierarchical structures increase dye absorption and facilitate electrolyte diffusion. Su Pei Lim and co-authors obtained similar results when TiO₂ was doped with nitrogen and sulphur. Both the samples achieved a more than 30 percent enhancement in efficiency.^251,252Fig. 30 compares the increase in PCE with non-metal codoping, which is discussed in this section.

Fig. 30 Increase in PCE of nonmetal codoped TiO₂. Data points are labelled with references.

Conductivity is a critical property of TiO₂. Each doping attempt is towards achieving this goal. Fu Lv et al. doped anatase TiO₂ with nitrogen, fluorine, and iodine to induce carriers in the lattice. This process eventually improves conductivity. N–I codoping shows the best blocking effect compared to undoped and single non-metal doped samples. The magnitude of internal charge recombination resistance (R_rec) decides the scale of the blocking effect. Usually, the charge recombination is dominated by the interfacial mesoporous TiO₂, but the compact underlayer also has an influence on charge recombination. The aforementioned blocking effect empties the impurity band and improves the conductivity of TiO₂ with N–I doping. So, the N–I doping inhibits electron recombination, leading to higher current density and efficiency. The N–I codoping increases electron injection between porous TiO₂ and the compact layer by tuning the conduction band. The synergetic effects of the dopants result in a 17.67% increase in efficiency.²⁵³

Trap states in the conduction band of nanostructured TiO₂ play a vital role in its photochemical properties. The current–time profiles of N–F doped samples are greatly influenced by the applied potential. They depend on the trap state density's current decaying behavior. At low trap density, the trap filling is fast; at high trap density, the trap filling is slower. Calculations show that the N–F doping significantly reduces the trap density due to the ability of nitrogen and fluorine to bind at the under-coordinated surface sites of TiO₂.²⁵⁴

Doping improves the performance of DSSCs and has similar effects in other electrochemical systems. B–S doping in QDSSCs is an excellent example of codoping and consequent efficiency increase. Doping TiO₂ with B–S results in bandgap narrowing and redshift in electronic absorption, leading to the increase in photocurrent response. The doping-induced efficiency is 12 percent greater compared to the bare TiO₂₋based QDSSC.¹¹⁸

4.3. Metal–nonmetal codoping

Metal–nonmetal codoping is constantly highlighted as a solution to the difficulties generated by single metal and single non-metal doping. The synergetic effect of both metal and non-metal provides higher thermal stability, lower bandgap values, and enhanced surface area. Furthermore, it reduces the carrier recombination and enhances the short circuit current.²⁵⁵ Doping Cu/S into the TiO₂ lattice minimizes the particle size and scattering effect. In the study by Gupta et al., the concentration of sulfur remains fixed, while the copper concentration is changed from 0 to 0.5 percent.²⁵⁶Fig. 31 demonstrates the increase in PCE with metal–nonmetal codoped TiO₂.

Fig. 31 Increase in PCE with metal–nonmetal codoped TiO₂. Data points are labelled with references.

The synergetic effect of hybridized Cu-3d and S species with Ti-3d orbitals significantly reduced the band gap of TiO₂. Consequently, this dip improved the redshift in favor of efficiency. Fig. 32 shows the varying crystallite size with 0.05 percent sulfur and copper concentration. Optimal doping plays a significant role in the Cu/S doping technique. When the doping level increased, crystals with smaller sizes started to agglomerate, clogged the microspores, and caused a severe drop in efficiency. The probability of an improved efficiency point between 0.2 and 0.3 percent doping is shown in Fig. 32. A more detailed and precise study requires finding the optimum point for maximum efficiency in Cu–S doped TiO₂ anode-based cells.

Fig. 32 Variation of crystallite size and efficiency with copper doping concentration. Data points are labelled with references.

Copper and nitrogen codoping was also investigated in the same way as copper and sulfur codoping. A DSSC made with doped TiO₂ particles (crystallites) shows different efficiencies at the doping concentration of copper. Fig. 33 shows an optimum doping point in Cu–N doping, as discussed in all other examples. In this case, a maximum efficiency is attained when TiO₂ is doped with copper by adding 0 to 20 mM copper nitrate into the paste, keeping the doping amount of nitrogen constant for all experiments. This tendency trails in the current generation through reduced crystallite size and increased surface area. When linking copper doping with nitrogen and sulfur, copper–nitrogen doping exhibits only a 19 percent increase in efficiency due to doping. The absence of data between 1 mM and 20 mM copper–nitrogen doping is probably the reason. According to the observations, a better optimal point for doping exists further to increase the efficiency of anodes.¹³⁷

Fig. 33 Copper doping concentration with (a) particle size, (b) surface area, (c) J_sc and (d) efficiency.¹³⁷

The same trend is observed in nitrogen–aluminum doping as well. However, as the doping increases and aluminum enters the TiO₂ lattice, a gradual fall in the peak intensities of the TiO₂ anatase phase is observed. This fall is an indication of anatase crystal growth obstruction. An appropriate combination of aluminum and nitrogen could effectively contribute to the efficiency. Higher doping causes severe agglomeration of the nanoparticles, leading to sluggish charge transfer and enhanced charge recombination. In this case, the codoping of aluminum and nitrogen leads to a gradual decrease in bandgap energies and an electron–hole lifetime.⁷⁹

Nitrogen and lanthanum codoping leads to a decrease in charge transport resistance, which effectively increases the dye absorption of the photoanode. Besides, the formation of new bonds with N, O, and lanthanum causes an increase in the disorder of the lattice structure of TiO₂ and the generation of trap states. Consequently, J_sc increases, which reflects in efficiency.²⁵⁷

Yttrium–nitrogen doping into TiO₂ shows a significant observation with an XPS scan. The TiO₂ anode shows a discrepancy in the amount of dopant added through precursors and dopant in the nanoparticles synthesized. When the nominal molar ratio of added precursors TiO₂–N–Y for synthesis is 100–100–0.5, the XPS observed concentrations of nitrogen and yttrium are 0.30% and 0.48%, which are different from the one we have prepared. Most studies with added precursors for doped products may have this discrepancy if the dopant ratio is not determined through the characterization technique. Nitrogen–yttrium doping also follows the same trend as other dopants in doping concentration and efficiency. The maximum efficiency is attained for a 100–300–0.5 nominal molar ratio combination sample. This is 18 percent higher compared to pure TiO₂ electrodes.²⁵⁸

Instead of producing an efficient DSSC, silver–sulfur codoped TiO₂ offered an environmentally conscious preparation approach. Even though 0.50% Ag was added for the codoping, XPS analysis showed that the anode had 0.75% silver in the atomic composition. Similar to our previous observations, when the anode had a doping concentration of 0.75% silver and 0.70% sulfur, it had the highest synergy and high-efficiency output.²⁵⁹

Zirconium doping with titanium increases the electron density and uplifts the Fermi level of TiO₂.⁷⁸ Tin is a homovalent atom to zirconium. Similar to zirconium, tin also offers lower transport resistance.⁸¹ The electron transport resistance decreases after tin doping and increases with fluorine doping. Impedance analysis shows that the dual doping increases interfacial charge recombination resistance. This resistance effectively decreases the recombination. A shorter transport time, measured with IMPS, implies a faster electron transport rate. The faster electron transport rate is favorable for charge collection and photocurrent density. The bonding may play a significant role in increasing efficiency as well. F atoms are bonded with the surface fivefold that of Ti, and the binding energy is 1 eV, showing strong chemisorption. Detailed density functional theory implementation gives more robust validation to existing arguments. Stronger bonding with Ti–F than Ti–O lowers the energy of the electrons of the valence band and leads to shifting the Fermi level towards the valence band and extending the bandgap. Both lead to the production of charge-trapping sites on the surface and subdue the charge recombination at the interface by holding trapped electrons, thus ending up in a longer electron lifetime. The longer electron lifetime is confirmed with IMVS measurements.²⁶⁰

Zirconium and nitrogen doping in TiO₂ results in high surface area, enhanced charge transfer, and reduced charge recombination in electrochemical devices. When P25 TiO₂ particles showed an efficiency of 8.32%, codoping with 0.01 zirconium and nitrogen increased the efficiency to 12.62%. While increasing the doping percentage of Zr/N in the anode, the efficiency of the cell decreased drastically. While the nitrogen doping shifts the absorption of TiO₂ into the visible light region, adding zirconium amplifies the short-circuit current. The dip in the efficiency is not discussed in detail in the paper but is ambiguously correlated with the dye absorption. We conclude that an optimum efficiency point exists where various properties such as charge transfer resistance, less electron–hole recombination, and bandgap tuning increase the maximum attainable efficiency. Above this point, the doping could negatively affect the efficiency and properties of the TiO₂ anode.²⁶¹

4.4. Bandgap tuning and codoping

Doping is considered to be the best option to tune the bandgap of any material.⁷⁷ When we introduce foreign materials into the TiO₂ lattice, they can alter the properties in both ways. The Burstein–Moss effect explains the increases in the bandgap when TiO₂ is doped with Cu, Ni, and Cr.²⁶² Here, doping populates the conduction band and pushes the absorption edge to higher energies, resulting in an increased bandgap. At lower doping concentrations, doping can decrease the band gap. Doping with different elements can introduce additional energy levels within the bandgap of the TiO₂. Subsequently, these levels will allow electrons with lower photon energy to reach the conduction band. Fig. 34 shows the quantitative relationship between the doped TiO₂ bandgap and the corresponding increase in efficiency. The V_oc, J_sc, PCE, and bandgap values of codoped TiO₂ are provided in Table 7. A more detailed dataset is provided in the ESI.† While the bandgap changes between 2.5 and 4 electron volts, the PCE increases to 6 percent, an enormous difference. Whenever the TiO₂ bandgap increases above 3, the efficiency tends to decrease. Finding the exact bandgap–efficiency relationship in electrochemical devices required extensive and pointed research in doping of TiO₂. Doping with zirconium and nitrogen tunes the bandgap to 2.83 eV, and the increase in the PCE is 5.65%. Doping TiO₂ with N–La, Al–N, Zr–N, and Cu–S tunes the bandgap around 2.7 to 2.86 eV. In this range, the efficiency is maximum for TiO₂ based devices.^{79,256,257,261}

Fig. 34 The quantitative relationship between the bandgap and increase in PCE. Data points are labelled with references.

Table 7 Quantitative relationship between the bandgap and increase in PCE in numbers

S. no.	Elements doped into TiO2	Bandgap (eV)	Increase in PCE through doping (%)	Ref.
1	Al–La	3.25	1.33	77
2	Nb–Fe	3.08	0.25	263
3	N–F	3.2	1.26	254
4	Cu–N	2.91	1.82	82
5	Ca–La	3.25	1.01	77
6	Mg–La	3.25	−2.7	264
7	Sr–V	2.57	1.62	250
8	Sr–Cr	2.56	1.75	265
9	N–S	3.2	0.78	251
10	N–La	2.83	1.31	257
11	Al–N	2.7	3.21	79
12	Zr–N	2.83	5.65	261
13	Ni–Zn	3.78	0.29	140
14	Cu–S	2.86	4.07	256
15	Zn–Ga	3.2	3.1	200

---

TiO₂ conduction band edges with doping Mg²⁺ ions are well studied, but La doping is not profound. The Mg–La doping increases the bandgap by 0.05 eV with doping. Contradictory to this, Ca–La and Al–La codoping results in the decrease of the bandgap. Lower V_oc values associated with Ca–La codoping in TiO₂ are found to reduce the bandgap. Nevertheless, all these band gap observations are made using Tauc plots, which could affect the accuracy of these measurements. A dopant with a smaller ionic radius than Ti⁴⁺ can enter the titanium lattice. When this dopant goes in, it can create more impurity levels and increase the recombination rate. Consequently, these impurity levels are due to the bandgap changes in the case of Al³⁺ and Mg²⁺. Moreover, Ca²⁺ ions have larger ionic radii than Ti⁴⁺ ions and cannot enter the lattice and get blocked outside. Doping with Ca–La causes a decrease in bandgap and has a positive impact on charge transfer.^77,114

In copper and sulfur codoping into TiO₂, the copper doping concentration changes while keeping the sulfur doping concentration at 0.05%. Their doping ratios are validated with EDAX. When the copper doping increases, the band gap of the doped sample is reduced due to the synergic effect of both dopants. The 3d and s species of copper hybridize with Ti-3d orbitals and produce localized defect levels above the maximum of the valence band. These defect levels effectively shift the valence band towards a longer wavelength. Impurity levels and oxygen doping can widen or narrow the bandgap.²⁵⁶Fig. 35 exemplifies the decrease in bandgap with Cu/S codoping.

Fig. 35 Decrease in bandgap with Cu/S codoping.

Impedance measurements by intensity-modulated photocurrent spectroscopy (IMPS) and intensity-modulated voltage spectroscopy (IMVS) are standard techniques for understanding charge carrier kinetics in oxides. The time constant can be calculated from the impedance semicircle of the Nyquist plot. IMPS and IMVS analysis can provide electron lifetime and transport time details. IMPS spectra consist of two semicircles, one representing the trap limited transport time and the trap free transport time. The effect of doping can be determined from the trap mediated transport time and electron lifetime.

Doping TiO₂ with Fe has a shorter trap limited transport time and a longer electron lifetime than bare TiO₂. Nb–Fe codoped TiO₂ samples take longer trap limited transport times and longer electron lifetimes than the other samples. This observation leads to the guesstimation that during Fe doping of TiO₂, there is the formation of Ti³⁺ donors and Fe traps. Shallow Ti³⁺ donors cause shorter transport times, while Fe has multiple trapping and de-trapping events, causing a longer lifetime. Fe and Nb codoping introduces Nb⁵⁺ into the TiO₂ lattice, creating further shallow Ti³⁺ donors and Nb⁵⁺ traps. These shallow donors extend the electron lifetime through electron excitation and trapping transport, allowing more diffusion under illumination. Oxygen vacancies generated during doping can act in favor of or against the charge collection during transport. However, open circuit photo-potential transient analysis shows that the Nb addition to Fe-doped TiO₂ facilitates charge relaxation through the passivation of deep oxygen vacancies generated by Fe doping. So, considering the OCP analysis, deep oxygen vacancies in the shallow levels may favor electron capture and de-trapping during photoexcitation.¹³⁴

Doping TiO₂ with Fe does not contribute much to bandgap engineering. Nevertheless, doping with Fe and Nb together significantly tunes the valence band position. The Nb doping reduces the bandgap by 0.01 eV, and the valence band position shifts to 0.11 eV. There are also similar effects on the conduction band edge. Even though the final bandgap of TiO₂ does not seem affected, there is an influential band re-tuning with the codoping. Only doping, donor generation, and trap state generation can account for this bandgap tuning.

Nitrogen and sulfur codoping also shows a reduction in the bandgap. Nitrogen is responsible for bandgap reduction, and sulfur facilitates the electron transfer process. Considering strontium and vanadium codoping, V doping into TiO₂ ends up in the stoichiometric reduction of Ti⁴⁺. Vanadium is also responsible for the formation of isolated energy levels near the conduction and valence bands, contributing to the PCE. The reduction facilitated by V leads to increased carrier density and improvement in photoresponse. At the same time, introducing strontium, a relatively high ionic radius ion (Sr²⁺), causes distortions in the TiO₂ lattice. These distortions accumulate oxygen vacancies across the lattice and extend the life of photoexcited electrons and holes. Chromium doping introduces isolated energy levels near conduction and valence bands like V. The reason for these isolated energy levels is excess holes. These holes are responsible for the acceptor band near the TiO₂ valence band.^{250–252,265}

Similar to copper–sulfur doping, zirconium–nitrogen codoping causes an increase and decrease in the bandgap.²⁶¹Fig. 36 shows the Zr/N doping concentration in TiO₂ and the corresponding change in the bandgap.

Fig. 36 Zr/N doping concentration in TiO₂ and the corresponding change in bandgap.

This study argues that the change in bandgap is connected with particle size variation (Fig. 37). When the particle size increases, the bandgap decreases. Simultaneously the reduction in bandgap results in increased cell efficiency. On the other hand, aluminum–nitrogen doping shows a similar bandgap trend to Zr/N.⁷⁹ There is a limitation to concluding the doping concentration and bandgap relationship due to the non-uniformity of reported doping concentration units. Fig. 36 roughly describes the connection between the bandgap and doping concentration. More or less, each dopant pair shows a decrease in the bandgap when materials are added in small amounts. However, adding materials beyond an optimum point decreases the doping effect and increases the bandgap.

Fig. 37 Doping concentration vs. bandgap.

5. Composite doping

Composite materials are macroscopic combinations of two or more distinct materials having a finite interface. They have been introduced into several industries in diverse forms. Doping and adding composites into TiO₂ have attracted considerable attention since they have significantly increased the PCE of DSSCs. Adding various composite nanostructures into TiO₂, including nanoparticles, nanobelts, and nanofibers, through different synthesis methods has advantages in tailoring its properties. Creating TiO₂-elemental composites plays a significant role in the flexibility and performance of the resulting material. In some studies a significantly low amount of material is added into TiO₂ for better performance. This is called composite doping. This composite impurity added to the TiO₂ can alter the photoanode performances and increase the efficiency. Composite dopants include metal composites, MW CNT composites, semiconductor composites, polymer carbon nitride composites, carbonate composites, and amine composites.^266–269

Functional nanocomposites involving metals improve the photocatalytic electron transfer process. Semiconductor–metal composites can sustain charge separation to a great extent. When Pt creates an ohmic type contact with ZnO, Au–ZnO creates a Schottky type contact.^270,271 Introducing Au into bulk TiO₂ and producing a metal nanoparticle composite reduces charge recombination, increases thermal and chemical stability, and inhibits non-radiative quenching.²⁷² In this case, the Au–TiO₂ composite has higher absorption due to the surface plasmon effect provided by Au. In another study, 0 to 0.6 wt% of Ag nanowires coated with SiO₂ were applied to TiO₂. The strength of the material, photon capture ability, and J_sc were significantly increased.²⁷³ F. Zheng and Z. Zhu showed that Au nanoparticles in TiO₂ nanofibers exhibited broad surface plasmonic absorption. The Au concentration in the anode conversely affected the DSSC performance beyond the optimum loading.²⁷⁴ Considering the available metal composite data, a N,S codoped silver nanoparticles decorated photoanode showed an exceptional increase in PCE. When the control sample (bare TiO₂ based) had 2.57 percentage efficiency, a codoped composite sample-based DSSC exhibited 8.22 percentage efficiency. The concentration of Ag NPs plays a dynamic role in the efficiency of the DSSC. When N,S doping reduced the bandgap, shifted the optical absorption towards the visible region, and suppressed the charge recombination, Ag nanoparticles harvested more light owing to the surface plasmon effect. The synergetic effect of codoping and Ag silver nanoparticles amplified the photocurrent generation, and consequently the efficiency of the DSSC.²⁵²Fig. 38 compares the effect of metal composite doping on the percentage increase in PCE.

Fig. 38 Metal composites and percentage increases in PCE. Data points are labelled with references.

TiO₂, dyes, and electrolyte interfaces have been studied since the beginning of the DSSC invention. Significant research is on the electron injection from dyes into the conduction band of TiO₂ since it is a crucial mechanism in determining efficiency. Various semiconductors were developed and deposited over TiO₂ to modulate the interfacial properties. Some include metal oxide semiconductors with wide bandgaps such as NiO, Bi₂S₃:Eu³⁺, Cu₂O, NiO, NiO:Eu+Tb, and ZnO nanobelts. ZnO and TiO₂ exhibit identical physical properties and bandgap values. The electron mobility and crystallization ability of ZnO are better than those of TiO₂. But ZnO-based photoanodes have comparatively lower PCEs than TiO₂ based DSSCs. This is primarily due to the poor dye-ZnO electron injection kinetics. The synergy of ZnO nanobelts and TiO₂ contributes to PCE by improving charge transfer efficiency and large surface area. ZnO nanobelt@TiO₂ also shows an increased carrier lifetime and fast electron transfer.²⁶⁹

Similarly, Bi₂S₃:Eu³⁺ can act as the down-conversion luminescent composite anode material. Bi₂S₃ transfers electrons to TiO₂, and the Eu³⁺ metal ion improves photocatalytic performance. P-type NiO is a wide bandgap semiconductor that is thermally and chemically stable. A photoanode, a combination of NiO and TiO₂, enhances the charge separation and increases the PCE.²⁷⁵ These samples—Bi₂S₃:Eu, NiO, NiO: Eu, and Tb—display excellent efficiency and only a slight increase in PCE due to composite doping. On the other hand, Cu₂O shows a tremendous increase in PCE. Cu₂O studies have been conducted by fabricating QDSSC, CdS, and ZnS layers. Cu₂O is one of the most potential p-type semiconductors for photovoltaic applications. It has high charge mobility and high minority carrier diffusion length. When the control QDDSC shows mere 1.17% efficiency, a Cu₂O incorporated QDDSC offers 3.01% efficiency. Contrary to the PCE, the effect of adding Cu₂O is much higher.²⁷⁶Fig. 39 shows the increase in PCE with each semiconductor composite dopant we have discussed.

Fig. 39 Increase in PCE with different semiconductor composite dopants. Data points are labelled with references.

Carbon nitrides, carbonate groups, amines, and carbon nanotubes have also been used for creating composites. A graphene–C₃N₄–TiO₂ composite exhibits a 29.3% improvement in PCE when used as a photoanode in a QDSSC. It has a suitable band structure and chemical-thermal stability for use as a photoanode. This material increases the roughness and makes the loading of CdS QDs easier.²⁶⁸

When comparing the efficiencies of other composite dopants, the efficiency of g-C₃N₄@TiO₂ seems to be less. There are two studies on carbonate doped TiO₂: one in 2016 and the other in 2018.^227,277 Carbonate doping results in a structure called a carbonate doped mesoporous microstructure (doped MS). A carbonate-doped mesosphere structure-based DSSC has higher PCE than carbonate doped MS nanospheres. Nevertheless, the increase in PCE is more in DSSCs based on carbonate based MS nanospheres.²²⁷ Carbonate doping produces a band tail state which effectively reduces the bandgap. Carbonates increase porosity and decrease electron recombination, and enhance the redox coupling. In the case of a carbonate-MS-NS@TiO₂-based photoanode, the carbonate reduced the internal cell resistance and improved the electron lifetime. The reduced internal cell resistance led to high light absorption and fast electron transport in carbonate based DSSCs.²⁷⁷Fig. 40 shows the increase in PCE with different composite dopants such as carbonates, amines, and carbon nanotubes.

Fig. 40 Increase in PCE with different composite dopants. Data points are labelled with references.

Aniline, an amine, has recently been doped into TiO₂ to achieve better photocurrent generation. Reduction of TiO₂ nanocrystal size using long-chain monoamines decreases the transmission. Amine capping improves efficiency and increases light scattering to a great extent. Amine groups such as aniline and o-phenylenediamine have active sites that can enhance PCE. The ability of nonmetal dopant's N2p and O2p orbitals to modify TiO₂ and reduce the deposition in interstitial sites has drawn attention towards non-metal doping. The operational strategy behind amine group doping is to improve photocurrent generation, increase electron transfer dynamics and reduce recombination in the scattering layer. It has long been known that amines affect morphology (by controlling TiO₂ size and providing an increased surface area) as well as reflectivity and PCE.²⁷⁸ In this study, aromatic amine passivated TiO₂ is prepared through a hydrothermal method. After amine capping, there is a reduction in crystallinity, and broadening in the XRD peak. Due to the formation of coordination bonds, there is a shift observed in the XRD peak. From reflectance studies, it is evident that amines increase the absorbance. By using FTIR, it is clear that the broadened N–H stretching reveals the fermi resonance of N–H and O–H peaks. The overlapping of N–H and Ti–OH bonds confirms the presence of aromatic amines in the TiO₂ lattice. The increase in reflectivity is due to the amines attached to the crystal surface. These amines reduce light penetration into the TiO₂ surface and the pi electrons in aromatic rings absorb light. This effectively reduces the scattering and increases the absorption.²⁶⁷ However, the control cell efficiency and amine-doped photoanode-based cell efficiency are high. The effect of amine capping is insignificant compared to other composite dopants.²⁶⁷

Cr(III) doping into TiO₂/MWCNTs results in an incredible increase in PCE. Impurities in the TiO₂ lattice act as charge trapping sites and improve the rate of charge transport in DSSCs. MWCNT@TiO₂ alone increases the efficiency by 4.6 percent. The percentage increase in PCE is around 300%. When Cr(III) is added to the matrix, the percentage increases from 6.18% to 7.69%. The overall increase in efficiency becomes 6.17 percent, which is around a 400 percent increase.²⁷⁹ Au nanoparticles and MWCNT composites also enhance the PCE by 30 percent. Compared with Cr(III), this is less. However, the control DSSC cell efficiency is higher here (5.09%) than the Cr(III)@MWCNT composite (1.52%). An Au@MWCNT composite-based DSSC has comparable efficiency (6.61%). This improvement in efficiency is attributed to the synergetic effect of Au and MWCNTs through improved charge generation and enhanced light absorption.²⁶⁶

6. Dopant distribution in the TiO₂ lattice – summary

Numerous studies have examined how dopants are distributed and where they are located within the TiO₂ lattice. This positioning information is collected from XPS, FTIR and XRD observations. The TiO₂ lattice can integrate a dopant to:

(a) Substitute Ti⁴⁺,

(b) Substitute Ti⁴⁺ and produce Ti³⁺,

(c) Substitute O²⁻,

(d) Substitute O²⁻ and induce Ti³⁺ formation,

(e) Go into interstitial sites and

(f) Go into surface interstitials.

The ionic radius of the dopant is a critical factor influencing the nature of doping. Dopants with a comparable atomic size of Ti⁴⁺ substitute it and causes slight rearrangements and distortions in the lattice. In rare cases, these substitutions lead to the surge of Ti³⁺ in the lattice. The ionic radius depends on factors such as coordination number, spin state, crystal radius, and electron configuration. Even in detailed ionic radius studies, if there is an uncertainty in coordination number or a deviation from the radius vs. valence plots then the obtained ionic radius values may not be that reliable.²⁸⁰ Keeping that aside, a significant number of dopants replace Ti⁴⁺ such as Mg²⁺, Ca²⁺, Sr²⁺, Sc³⁺, Y³⁺, Zr⁴⁺, V⁵⁺, Nb⁵⁺, Ta⁵⁺, W⁴⁺, Cr³⁺, Mn²⁺, Fe³⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Cd²⁺, In³⁺, Sn⁴⁺, Bi³⁺, and Al³⁺. Among these, bigger ionic radius atoms such as Sr²⁺, Bi³⁺, and Ca²⁺ cause lattice distortions and changes in crystalline information.

It has been suggested that replacing Ti⁴⁺ ions may result in Ti–O–Cr structures and oxygen vacancies. This Ti⁴⁺ withdraws electrons from superficial hydroxyl groups and reduces to Ti³⁺:

Ti⁴⁺ + Cr³⁺ + e(O²⁻) →T³⁺ + Cr³⁺ + O_vacancy

Iron (Fe²⁺) and copper (Cu²⁺) also result in the creation of Ti³⁺ trap states, just like chromium (Cr³⁺) does. In a way analogous to this, fluorine (F⁻) substitutes oxygen and produces Ti³⁺ ions, improving the effectiveness of charge collection. Due to its great electronegativity, it interferes with the CB of TiO₂ and prevents electron recombination. Since oxygen (O²⁻) and nitrogen (N³⁻) have similar ionic sizes, nitrogen can easily replace oxygen. There are some conflicts regarding the oxidation state and position of nitrogen dopant in the TiO₂ lattice. The presence of Ti–O–N bonding indicates the presence of N⁻ ions.^{55,224,229–231,281} FTIR, XPS, and XRD studies show that lithium (Li⁺) and boron (B³⁺) are introduced into the interstitial sites.

Some researchers claim that no rare earth elements can pass through the TiO₂ lattice due to the larger ionic size. They either form Ti–O–La bonding on the surface or are in the surface interstitials of TiO₂. This bonding increases the surface area and oxygen vacancies of TiO₂. When doped with rare earth elements, photovoltaic devices that use TiO₂ perform better. When doped, La³⁺, Nd³⁺, Sm³⁺, Eu³⁺, Gd³⁺, Tb³⁺, Er³⁺, and Yb³⁺ remain in the surface interstitials.

Sulfur shows a peculiar behavior when doped into TiO₂. It is challenging to establish the oxidation state of sulfur when doped.²²³ Although sulphur has oxidation states ranging from S⁶⁺ to S²⁻, there is no evidence that S²⁻ (1.7 Å) would replace O²⁻ (1.22 Å) atoms; instead, S⁶⁺ (0.29 Å) would replace Ti⁴⁺ (0.64 Å). In addition, O²⁻ and Ti⁴⁺ ions with various oxidation states might get randomly substituted, significantly disturbing the lattice.^245,246 The entire discussion is summarized in Fig. 41, which illustrates the distribution of dopants in the TiO₂ lattice.

Fig. 41 Distribution of dopants in TiO₂ (references provided in Table 8).

Table 8 Distribution of dopants in TiO₂ – summary with references

Name of the element	Ionic state	Ionic radius (pm)	Lattice interaction	Ref.
Lithium	Li^+	76	Going into interstitial sites	86
Boron	B^3+	11	Going into interstitial sites	126
Nitrogen	N^3−	146	Substituting O^2−	239
Fluorine	F^−	133	Substituting O^2− and inducing Ti^3+ formation	217
Magnesium	Mg^2+	83	Substituting Ti^4+	88
Aluminium	Al^3+	53.5	Substituting Ti^4+	77
Sulfur	S^6+	12	Complex behavior	245–247
Calcium	Ca^2+	100	Substituting Ti^4+ and distorting the lattice	77
Scandium	Sc^3+	74.5	Substituting Ti^4+	127
Vanadium	V^5+	58	Substituting Ti^4+	147
Manganese	Mn^2+	83	Substituting Ti^4+	151
Chromium	Cr^3+	61	Substituting Ti^4+ and producing Ti^3+	250
Iron	Fe^3+	64.5	Substituting Ti^4+ and producing Ti^3+	153–155
Cobalt	Co^2+	74.5	Substituting Ti^4+	156
Nickel	Ni^2+	69	Substituting Ti^4+	136,158
Copper	Cu^2+	73	Substituting Ti^4+ and producing Ti^3+	151,154,159
Zinc	Zn^2+	74	Substituting Ti^4+	200
Gallium	Ga^3+	62	Substituting Ti^4+	209
Strontium	Sr^2+	118	Substituting Ti^4+	99
Yttrium	Y^3+	90	Substituting Ti^4+	141,165,166
Zirconium	Zr^4+	72	Substituting Ti^4+	72,167
Niobium	Nb^5+	64	Substituting Ti^4+	171,172
Cadmium	Cd^2+	95	Substituting Ti^4+	144
Indium	In^3+	80	Substituting Ti^4+	199
Tin	Sn^4+	69	Substituting Ti^4+	203
Iodine	I^5+	95	Substituting Ti^4+	213,214
Tantalum	Ta^5+	64	Substituting Ti^4+	178
Tungsten	W^4+	66	Substituting Ti^4+	146
Bismuth	Bi^3+	103	Substituting Ti^4+	66,201
Lanthanum	La^3+	103	Surface interstitials	114–116
Neodymium	Nd^3+	98	Surface interstitials	117
Samarium	Sm^3+	95	Surface interstitials	104
Europium	Eu^3+	94.7	Surface interstitials	70,107
Gadolinium	Gd^3+	100	Surface interstitials	106
Terbium	Tb^3+	92	Surface interstitials	106
Erbium	Er^3+	89	Surface interstitials	108,109
Ytterbium	Yb^3+	86	Surface interstitials	107

---

7. Conclusion

This review developed a framework to analyze the doping of TiO₂ based on the PCE of sensitized solar cells. This review helps researchers to identify, analyze and plan doping in TiO₂ electrode-based power sources. Beyond this review, this framework can be used for all electrochemical systems, including hydrogen generation, supercapacitors, and batteries. We identified and summarized significant elements in the periodic table doped into TiO₂ and their effect on the material's PCE, J_sc, V_oc, FF, and morphology.

We compiled information on TiO₂ dopants, such as the treatment/coating used, sub-type of doping, structure, synthesis and fabrication method, V_oc (V), photocurrent density (mA cm⁻²), FF, PCE (%), percentage increase in efficiency due to doping, bandgap (eV), application, sensitizer/dye, and periodic classification of element types (which are included in each section). We have 106 observations where single-element doped TiO₂ is used for developing DSSCs, 34 data points where codoped TiO₂ is used when making DSSCs, and 15 data points where composite doped TiO₂ is employed to fabricate DSSCs. According to our knowledge, this is the largest, complete information data set available until now.

Doping is not merely creating a change in electronic band structure; rather, it is altering the morphology and distribution of atoms throughout the lattice, and how these factors affect PCE. Throughout the review, we discussed how doping impacts the TiO₂ crystal structure, alters the bandgap, and affects PCE. According to electron transfer dynamics studies, the morphology of the crystal surface influences the inhomogeneous distribution of the conduction band.^27,29Fig. 42 features how doping increases and decreases PCE. In general, all the post-transition element dopants, nitrogen, Al–La, Zr–N, and Ca–La (codopants) show high charge transfer resistance compared to other dopants. It is also found that bigger atoms that are not able to penetrate the lattice facilitate the charge transfer process.

Fig. 42 Summary of TiO₂ doping.

Three main factors which affect the PCE have been identified as the concentration of dopant, the size of doped TiO₂ particles, and the band gap of the doped TiO₂ electrode. The concentration of dopant is critical in the doping procedure. The concentration of dopant affects the size of dopant, recombination rate, transport rate, film resistance, diffusion coefficient, and oxygen vacancies. These effects are discussed with examples of Sr, Gd, Sc, V, Zr, and Nb doping.^{99,106,127,145,170,184,228} The efficiency decreases when the dopant concentration is below or above an optimum amount. The challenge involved in doping is finding the optimum point of peak performance. Having a control sample is necessary to understand the research progress in global standards. Many research papers lack data on the bare TiO₂-based control cell to compare the doping effect and understand the role of doping in PCE. V_oc, J_sc, or FF alone is not enough to comprehend the effect of doping. We need to determine the increase in PCE from that of the control-based DSSC sample. However, to summarize, V_oc is increased using Mg, Cu, Nb, W, Ga, In, N, and S-doped TiO₂, whereas Zr, Ag, and Ca–La doped TiO₂ reduce V_oc. The short circuit current increases in Sr, B, Sb, Nb, I, N, S, N–L, Ag, and Au-doped TiO₂, but the J_sc decreases in Mg-doped TiO₂.

The size and morphology of the doped samples are altered by doping. As a result of doping, nanoparticles, nanowalls, nanotubes, nanowires, and nanoflower structures have been created. The doped material has optimum dimensions that provide the highest PCE, similar to the concentration's influence on PCE. When the dimensions increase above this optimum point, the PCE is drastically reduced. The particle/crystallite size and doping concentration impact PCE. A material's bandgap has an impact on its electrical and optical characteristics. The PCE increases as the bandgap narrows. Some scientists contend that the bandgap influences PCE as a result of the bandgap influencing crystallite size.

Among all of the doping methods, codoping might be referred to as sensible doping. Codoping correlates the synergistic effect of two or more dopants, contributes to both V_oc and J_sc simultaneously, and improves PCE. Single dopants contribute to either V_oc or J_sc. Their synergetic effect also contributes to thermal stability, reduction in the bandgap, and enhanced surface area. One element enhances the short-circuit current, boosts electron transfer, and minimizes carrier recombination. The other element contributes to V_oc, provides isolated energy levels close to the conduction band, lowers the bandgap, and participates in codoping. Codoping significantly boosts carrier density and enhances photoresponse.

Regarding green energy technology, doping has a bright future in research and applications. A negligible amount of resources can alter the properties of non-toxic, abundant semiconductors so they serve as excellent light-harvesting energy materials. This review has the highest PCE data to the authors’ best knowledge. It gives a holistic view of doping of TiO₂ based on PCE in sensitized solar cells. This review will help future researchers understand the role of TiO₂ dopants in PCE.

Author contributions

Prasanth Ravindran – supervision and writing (review & editing). Aparna Markose and Debanita Das – conceptualization, data curation, investigation, methodology, writing, and data analysis – (original draft). Aparna Markose – visualization and graphics.

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

We acknowledge the funding from the South Asia Foundation (SAF), the Ministry of New and Renewable Energy (MNRE), the University Grants Commission (UGC), the Council of Science and Industrial Research (CSIR), and the Board of Research in Nuclear Sciences (BRNS). The authors acknowledge Pondicherry University for providing the required infrastructure.

References

1. C. C. Raj and R. Prasanth, Review—Advent of TiO₂ Nanotubes as Supercapacitor Electrode, J. Electrochem. Soc., 2018, 165(9), E345–E358.
2. C. C. Raj, R. Sundheep and R. Prasanth, Enhancement of Electrochemical Capacitance by Tailoring the Geometry of TiO₂ Nanotube Electrodes, Electrochim. Acta, 2015, 176, 1214–1220.
3. M. Grätzel, Photoelectrochemical Cells, Nature, 2001, 414, 338–344.
4. M. Grätzel, Solar Energy Conversion by Dye-Sensitized Photovoltaic Cells, Inorg. Chem., 2005, 44(20), 6841–6851.
5. A. G. Vega Poot, D. Reyes Coronado and G. Oskam, Application of Three TiO₂ Polymorphs in Photoelectrochemical Solar Cells, ECS Meet. Abstr., 2006, MA2006-02(38), 1737.
6. M. Zhang, T. Chen and Y. Wang, Insights into TiO₂ Polymorphs: Highly Selective Synthesis, Phase Transition, and Their Polymorph-Dependent Properties, RSC Adv., 2017, 7(83), 52755–52761.
7. T. Zhu and S. P. Gao, The Stability, Electronic Structure, and Optical Property of TiO₂ Polymorphs, J. Phys. Chem. C, 2014, 118(21), 11385–11396.
8. A. J. Frank, N. Kopidakis and J. V. D. Lagemaat, Electrons in Nanostructured TiO₂ Solar Cells: Transport, Recombination and Photovoltaic Properties, Coord. Chem. Rev., 2004, 248(13–14), 1165–1179.
9. B. O’Regan and M. Graetzel, A Low-Cost, High-Efficiency Solar Cell Based on Dye-Sensitized Colloidal TiO₂ Films, Nature, 1991, 354, 737–740.
10. H. Tributsch and M. Calvin, Electrochemistry of Excited Molecules: Photo-Electrochemical Reactions of Chlorophylls, Photochem. Photobiol., 1971, 14(2), 95–112.
11. B. N. Dimarco, R. N. Sampaio, E. M. James, T. J. Barr, M. T. Bennett and G. J. Meyer, Efficiency Considerations for SnO₂-Based Dye-Sensitized Solar Cells, ACS Appl. Mater. Interfaces, 2020, 12(21), 23923–23930.
12. A. Le Viet, R. Jose, M. V. Reddy, B. V. R. Chowdari and S. Ramakrishna, Nb₂O₅ Photoelectrodes for Dye-Sensitized Solar Cells: Choice of the Polymorph, J. Phys. Chem. C, 2010, 114(49), 21795–21800.
13. D. Devadiga, M. Selvakumar, P. Shetty and M. S. Santosh, Dye-Sensitized Solar Cell for Indoor Applications: A Mini-Review, J. Electron. Mater., 2021, 50(6), 3187–3206.
14. U. Mehmood, Z. Malaibari, F. A. Rabani, A. U. Rehman, S. H. A. Ahmad, M. A. Atieh and M. S. Kamal, Photovoltaic Improvement and Charge Recombination Reduction by Aluminum Oxide Impregnated MWCNTs/TiO₂ Based Photoanode for Dye-Sensitized Solar Cells, Electrochim. Acta, 2016, 203, 162–170.
15. V. Rondán-Gómez, I. Montoya De Los Santos, D. Seuret-Jiménez, F. Ayala-Mató, A. Zamudio-Lara, T. Robles-Bonilla and M. Courel, Recent Advances in Dye-Sensitized Solar Cells, Appl. Phys. A: Mater. Sci. Process., 2019, 125(12), 1–24.
16. M. Freitag, J. Teuscher, Y. Saygili, X. Zhang, F. Giordano, P. Liska, J. Hua, S. M. Zakeeruddin, J.-E. Moser and M. Grätzel, et al., Dye-Sensitized Solar Cells for Efficient Power Generation under Ambient Lighting, Nat. Photonics, 2017, 11(6), 372–378.
17. C. Arrouvel and S. C. Parker, Investigating Surface Properties and Lithium Diffusion in Brookite-TiO₂, J. Braz. Chem. Soc., 2020, 31(1), 51–65.
18. B. O’regan and M. Grätzel, A Low-Cost, High-Efficiency Solar Cell Based on Dye-Sensitized Colloidal TiO₂ Films, Lett. Nat., 1991, 353, 737–740.
19. J. J. Cid, J. H. Yum, S. R. Jang, M. K. Nazeeruddin, E. Martínez-Ferrero, E. Palomares, J. Ko, M. Grätzel and T. Torres, Molecular Cosensitization for Efficient Panchromatic Dye-Sensitized Solar Cells, Angew. Chem., Int. Ed., 2007, 46(44), 8358–8362.
20. H. J. Snaith, Estimating the Maximum Attainable Efficiency in Dye-Sensitized Solar Cells, Adv. Funct. Mater., 2010, 20(1), 13–19.
21. S. Ardo and G. J. Meyer, Photodriven Heterogeneous Charge Transfer with Transition-Metal Compounds Anchored to TiO₂ Semiconductor Surfaces, Chem. Soc. Rev., 2009, 38(1), 115–164.
22. A. Hagfeldt and M. Grätzel, Light-Induced Redox Reactions in Nanocrystalline Systems, Chem. Rev., 1995, 95(1), 49–68.
23. Nanostructured and Photoelectrochemical Systems for Solar Photon Conversion, ed. M. D. Archer and A. J. Nozik, Imperial College Press, 2008.
24. J. R. Durrant, S. A. Haque and E. Palomares, Towards Optimisation of Electron Transfer Processes in Dye Sensitised Solar Cells, Coord. Chem. Rev., 2004, 248(13–14), 1247–1257.
25. S. A. Haque, E. Palomares, B. M. Cho, A. N. M. Green, N. Hirata, D. R. Klug and J. R. Durrant, Charge Separation versus Recombination in Dye-Sensitized Nanocrystalline Solar Cells: The Minimization of Kinetic Redundancy, J. Am. Chem. Soc., 2005, 127(10), 3456–3462.
26. N. A. Andersen and T. Lian, Ultrafast Electron Transfer at the Molecule-Semiconductor Nanoparticle Interface, Annu. Rev. Phys. Chem., 2005, 56(78), 491–519.
27. A. Listorti, B. O’Regan and J. R. Durrant, Electron Transfer Dynamics in Dye-Sensitized Solar Cells, Chem. Mater., 2011, 23(15), 3381–3399.
28. Y. Q. Gao and R. Marcus, On the theory of electron transfer reactions at semiconductor/liquid interfaces. II. A free electron model, J. Chem. Phys., 2000, 113(15), 1–3.
29. R. Katoh, A. Furube, A. V. Barzykin, H. Arakawa and M. Tachiya, Kinetics and Mechanism of Electron Injection and Charge Recombination in Dye-Sensitized Nanocrystalline Semiconductors, Coord. Chem. Rev., 2004, 248(13–14), 1195–1213,  DOI:10.1016/j.ccr.2004.03.017.
30. X. Zhang, J. J. Zhang and Y. Y. Xia, Molecular Design of Coumarin Dyes with High Efficiency in Dye-Sensitized Solar Cells, J. Photochem. Photobiol., A, 2008, 194(2–3), 167–172.
31. K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo, S. Suga, K. Sayama, H. Sugihara and H. Arakawa, Molecular Design of Coumarin Dyes for Efficient Dye-Sensitized Solar Cells, J. Phys. Chem. B, 2003, 107(2), 597–606.
32. T. Santos, A. Dos Morandeira, S. Koops, A. J. Mozer, G. Tsekouras, Y. Dong, P. Wagner, G. Wallace, J. C. Earles and K. C. Gordon, et al., Injection Limitations in a Series of Porphyrin Dye-Sensitized Solar Cells, J. Phys. Chem. C, 2010, 114(7), 3276–3279.
33. K. Hara, Z. S. Wang, T. Sato, A. Furube, R. Katoh, H. Sugihara, Y. Dan-Oh, C. Kasada, A. Shinpo and S. Suga, Oligothiophene-Containing Coumarin Dyes for Efficient Dye-Sensitized Solar Cells, J. Phys. Chem. B, 2005, 109(32), 15476–15482.
34. A. Morandeira, I. López-Duarte, M. V. Martínez-Díaz, B. O’Regan, C. Shuttle, N. A. Haji-Zainulabidin, T. Torres, E. Palomares and J. R. Durrant, Slow Electron Injection on Ru-Phthalocyanine Sensitized TiO₂, J. Am. Chem. Soc., 2007, 129(30), 9250–9251.
35. S. E. Koops, B. C. O’Regan, P. R. F. Barnes and J. R. Durrant, Parameters Influencing the Efficiency of Electron Injection in Dye-Sensitized Solar Cells, J. Am. Chem. Soc., 2009, 131(13), 4808–4818.
36. H. N. Ghosh, J. B. Asbury and T. Lian, Direct Observation of Ultrafast Electron Injection from Coumarin 343 to TiO₂ Nanoparticles by Femtosecond Infrared Spectroscopy, J. Phys. Chem. B, 1998, 102(34), 6482–6486.
37. S. Pelet, M. Grätzel and J. E. Moser, Femtosecond Dynamics of Interfacial and Intermolecular Electron Transfer at Eosin-Sensitized Metal Oxide Nanoparticles, J. Phys. Chem. B, 2003, 107(14), 3215–3224.
38. R. J. Ellingson, J. B. Asbury, S. Ferrere, H. N. Ghosh, J. R. Sprague, T. Lian and A. J. Nozik, Dynamics of Electron Injection in Nanocrystalline Titanium Dioxide Films Sensitized with [Ru(4,4′-Dicarboxy-2,2′-Bipyridine)₂(NCS)₂] by Infrared Transient Absorption, J. Phys. Chem. B, 1998, 102(34), 6455–6458.
39. Y. Tachibana, J. E. Moser, M. Grätzel, D. R. Klug and J. R. Durrant, Subpicosecond Interfacial Charge Separation in Dye-Sensitized Nanocrystalline Titanium Dioxide Films, J. Phys. Chem., 1996, 100(51), 20056–20062.
40. J. B. Asbury, E. Hao, Y. Wang, H. N. Ghosh and T. Lian, Ultrafast Electron Transfer Dynamics from Molecular Adsorbates to Semiconductor Nanocrystalline Thin Films, J. Phys. Chem. B, 2001, 105(20), 4545–4557.
41. J. Kallioinen, G. Benkö, V. Sundström, J. E. I. Korppi-Tommola and A. P. Yartsev, Electron Transfer from the Singlet and Triplet Excited States of Ru(Dcbpy)₂(NCS)₂ into Nanocrystalline TiO₂ Thin Films, J. Phys. Chem. B, 2002, 106(17), 4396–4404.
42. J. E. Moser and M. Grätzel, Observation of Temperature Independent Heterogeneous Electron Transfer Reactions in the Inverted Marcus Region, Chem. Phys., 1993, 176(2–3), 493–500.
43. D. Kuciauskas, J. E. Monat, R. Villahermosa, H. B. Gray, N. S. Lewis and J. K. McCusker, Transient Absorption Spectroscopy of Ruthenium and Osmium Polypyridyl Complexes Adsorbed onto Nanocrystalline TiO₂ Photoelectrodes, J. Phys. Chem. B, 2002, 106(36), 9347–9358.
44. Y. Tachibana, S. A. Haque, I. P. Mercer, J. E. Moser, D. R. Klug and J. R. Durrant, Modulation of the Rate of Electron Injection in Dye-Sensitized Nanocrystalline TiO₂ Films by Externally Applied Bias, J. Phys. Chem. B, 2001, 105(31), 7424–7431.
45. F. Werner, J. F. Gnichwitz, R. Marczak, E. Palomares, W. Peukert, A. Hirsch and D. M. Guldi, Grafting Porphyrins (Face-to-Edge/Orthogonal versus Face-to-Face/Parallel) to ZnO En Route toward Dye-Sensitized Solar Cells, J. Phys. Chem. B, 2010, 114(45), 14671–14678.
46. M. Planells, F. J. Céspedes-Guirao, L. Gonçalves, A. Sastre-Santos, F. Fernández-Lázaro and E. Palomares, Supramolecular Interactions in Dye-Sensitised Solar Cells, J. Mater. Chem., 2009, 19(32), 5818–5825.
47. J. J. Cid, M. García-Iglesias, J. H. Yum, A. Forneli, J. Albero, E. Martínez-Ferrero, P. Vázquez, M. Grätzel, M. K. Nazeeruddin and E. Palomares, et al., Structure-Function Relationships in Unsymmetrical Zinc Phthalocyanines for Dye-Sensitized Solar Cells, Chem. – Eur. J., 2009, 15(20), 5130–5137.
48. A. Reynal, A. Forneli and E. Palomares, Dye Structure-Charge Transfer Process Relationship in Efficient Ruthenium-Dye Based Dye Sensitized Solar Cells, Energy Environ. Sci., 2010, 3(6), 805–812.
49. H. J. Snaith, A. Petrozza, S. Ito, H. Miura and M. Graätzel, Charge Generation and Photovoltaic Operation of Solid-State Dye-Sensitized Solar Cells Incorporating a High Extinction Coefficient Indolene-Based Sensitizer, Adv. Funct. Mater., 2009, 19(11), 1810–1818.
50. H. K. Ardakani, Electrical and Optical Properties of in Situ “Hydrogen-Reduced” Titanium Dioxide Thin Films Deposited by Pulsed Excimer Laser Ablation, Thin Solid Films, 1994, 248(2), 234–239.
51. CRC Handbook of Chemistry and Physics 95th Edition, ed. W. M. Haynes, R. David and T. J. B. Lide, CRC Press, 95th edn, 1942.
52. R. Asahi, Y. Taga and W. Mannstadt, Electronic and Optical Properties of Anatase, Phys. Rev. B: Condens. Matter Mater. Phys., 2000, 61(11), 7459–7465.
53. M. A. Othman, N. F. Amat, B. H. Ahmad and J. Rajan, Electrical Conductivity Characteristic of TiO₂ Nanowires from Hydrothermal Method., J. Phys.: Conf. Ser., 2014, 495(1), 012027,  DOI:10.1088/1742-6596/495/1/012027.
54. M. C. K. Sellers and E. G. Seebauer, Measurement Method for Carrier Concentration in TiO₂via the Mott–Schottky Approach, Thin Solid Films, 2011, 519(7), 2103–2110.
55. G. Murali, M. Reddeppa, C. Seshendra Reddy, S. Park, T. Chandrakalavathi, M. D. Kim and I. In, Enhancing the Charge Carrier Separation and Transport via Nitrogen-Doped Graphene Quantum Dot-TiO₂ Nanoplate Hybrid Structure for an Efficient NO Gas Sensor, ACS Appl. Mater. Interfaces, 2020, 12(11), 13428–13436.
56. A. K. Chandiran, F. Sauvage, M. Casas-Cabanas, P. Comte, S. M. Zakeeruddin and M. Graetzel, Doping a TiO₂ Photoanode with Nb⁵⁺ to Enhance Transparency and Charge Collection Efficiency in Dye-Sensitized Solar Cells., J. Phys. Chem. C, 2010, 114(37), 15849–15856.
57. K. Subalakshmi and J. Senthilselvan, Effect of Fluorine-Doped TiO₂ Photoanode on Electron Transport, Recombination Dynamics and Improved DSSC Efficiency., Sol. Energy, 2018, 171, 914–928.
58. A. Paxton and L. Thiên-Nga, Electronic Structure of Reduced Titanium Dioxide, Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 57(3), 1579–1584.
59. H. J. Snaith and L. Schmidt-Mende, Advances in Liquid-Electrolyte and Solid-State Dye-Sensitized Solar Cells, Adv. Mater., 2007, 19(20), 3187–3200.
60. M. Radecka, M. Rekas, A. Trenczek-Zajac and K. Zakrzewska, Importance of the Band Gap Energy and Flat Band Potential for Application of Modified TiO₂ Photoanodes in Water Photolysis, J. Power Sources, 2008, 181(1), 46–55.
61. V. Kumar, S. K. Swami, A. Kumar, O. M. Ntwaeaborwa, V. Dutta and H. C. Swart, Eu³⁺ Doped down Shifting TiO₂ Layer for Efficient Dye-Sensitized Solar Cells, J. Colloid Interface Sci., 2016, 484, 24–32.
62. B. Roose, S. Pathak and U. Steiner, Doping of TiO₂ for Sensitized Solar Cells, Chem. Soc. Rev., 2015, 44(22), 8326–8349.
63. A. A. Awsha, S. H. Alazoumi and B. Elhub, A Review on the Development of TiO₂ Photoanode for Solar Applications, 2021, 6(2), 1–9.
64. X. Hou, K. Aitola and P. D. Lund, TiO₂ Nanotubes for Dye-Sensitized Solar Cells—A Review., Energy Sci. Eng., 2020, 921–937.
65. H. H. Nguyen, G. Gyawali, J. S. Hoon, T. Sekino and S. W. Lee, Cr-Doped TiO₂ Nanotubes with a Double-Layer Model: An Effective Way to Improve the Efficiency of Dye-Sensitized Solar Cells, Appl. Surf. Sci., 2018, 458, 523–528.
66. M.-C. Wu, W.-C. Chen, T.-H. Lin, K.-C. Hsiao, K.-M. Lee and C.-G. Wu, Enhanced Open-Circuit Voltage of Dye-Sensitized Solar Cells Using Bi-Doped TiO₂ Nanofibers as Working Electrode and Scattering Layer, Sol. Energy, 2016, 135, 22–28.
67. D. Dahlan, S. K. Md Saad, A. U. Berli, A. Bajili and A. A. Umar, Synthesis of Two-Dimensional Nanowall of Cu-Doped TiO₂ and Its Application as Photoanode in DSSCs., Phys. E, 2017, 91, 185–189.
68. S. Shalini, R. Balasundaraprabhu, T. Satish Kumar, N. Muthukumarasamy, S. Prasanna, K. Sivakumaran and M. D. Kannan, Enhanced Performance of Sodium Doped TiO₂ Nanorods Based Dye Sensitized Solar Cells Sensitized with Extract from Petals of Hibiscus SABDARIFFA (Roselle), Mater. Lett., 2018, 221, 192–195.
69. Y. Duan, N. Fu, Q. Zhang, Y. Fang, X. Zhou and Y. Lin, Influence of Sn Source on the Performance of Dye-Sensitized Solar Cells Based on Sn-Doped TiO₂ Photoanodes: A Strategy for Choosing an Appropriate Doping Source, Electrochim. Acta, 2013, 107, 473–480.
70. V. Kumar, S. K. Swami, A. Kumar, O. M. Ntwaeaborwa, V. Dutta and H. C. Swart, Eu³⁺ Doped down Shifting TiO₂ Layer for Efficient Dye-Sensitized Solar Cells, J. Colloid Interface Sci., 2016, 484, 24–32.
71. E. Jalali-Moghadam and Z. Shariatinia, Al³⁺ Doping into TiO₂ Photoanodes Improved the Performances of Amine Anchored CdS Quantum Dot Sensitized Solar Cells, Mater. Res. Bull., 2018, 98, 121–132.
72. F. A. Unal, S. Ok, M. Unal, S. Topal, K. Cellat and F. Şen, Synthesis, Characterization, and Application of Transition Metals (Ni, Zr, and Fe) Doped TiO₂ Photoelectrodes for Dye-Sensitized Solar Cells, J. Mol. Liq., 2020, 299, 112177.
73. M. N. An’amt, S. Radiman, N. M. Huang, M. A. Yarmo, N. P. Ariyanto, H. N. Lim and M. R. Muhamad, Sol–Gel Hydrothermal Synthesis of Bismuth–TiO₂ Nanocubes for Dye-Sensitized Solar Cell, Ceram. Int., 2010, 36(7), 2215–2220.
74. M. I. Khan, M. Sabir, G. M. Mustafa, M. Fatima, A. Mahmood, S. A. Abubshait, H. A. Abubshait and M. Iqbal, 300 keV Cobalt Ions Irradiations Effect on the Structural, Morphological, Optical and Photovoltaic Properties of Zn Doped TiO₂ Thin Films Based Dye Sensitized Solar Cells., Ceram. Int., 2020, 46(10, Part B), 16813–16819.
75. Y. Zhang, N. Zhou, K. Zhang and F. Yan, Plasmonic Copper Nanowire@TiO₂ Nanostructures for Improving the Performance of Dye-Sensitized Solar Cells, J. Power Sources, 2017, 342, 292–300.
76. Q. Ma and Y. M. Huang, Improved Photovoltaic Performance of Dye Sensitized Solar Cell by Decorating TiO₂ Photoanode with Li-Doped ZnO Nanorods, Mater. Lett., 2015, 148, 171–173.
77. A. I. Rafieh, P. Ekanayake, A. L. Tan and C. M. Lim, Effects of Ionic Radii of Co-Dopants (Mg, Ca, Al and La) in TiO₂ on Performance of Dye-Sensitized Solar Cells, Sol. Energy, 2017, 141, 249–255.
78. A. A. Qureshi, H. M. A. Javed, S. Javed, A. Bashir, M. Usman, A. Akram, M. I. Ahmad, U. Ali, M. Shahid and M. Rizwan, et al., Incorporation of Zr-Doped TiO₂ Nanoparticles in Electron Transport Layer for Efficient Planar Perovskite Solar Cells, Surf. Interfaces, 2021, 25, 101299.
79. K. Sahu Dhonde, M. Dhonde and V. V. S. Murty, Novel Synergistic Combination of Al/N Co-Doped TiO₂ Nanoparticles for Highly Efficient Dye-Sensitized Solar Cells, Sol. Energy, 2018, 173, 551–557.
80. V. S. Katta, A. Das, K. R. Dileep, G. Cilaveni, S. Pulipaka, G. Veerappan, E. Ramasamy, P. Meduri, S. Asthana and D. Melepurath, et al., Vacancies Induced Enhancement in Neodymium Doped Titania Photoanodes Based Sensitized Solar Cells and Photo-Electrochemical Cells, Sol. Energy Mater. Sol. Cells, 2021, 220, 110843.
81. Y. Akila, N. Muthukumarasamy, S. Agilan, S. Senthilarasu and D. Velauthapillai, Zirconium Oxide Post Treated Tin Doped TiO₂ for Dye Sensitized Solar Cells, Mater. Sci. Semicond. Process., 2017, 57, 24–31.
82. M. Dhonde, K. Sahu, V. V. S. Murty, S. S. Nemala, P. Bhargava and S. Mallick, Enhanced Photovoltaic Performance of a Dye Sensitized Solar Cell with Cu/N Co-Doped TiO₂ Nanoparticles, J. Mater. Sci.: Mater. Electron., 2018, 29(8), 6274–6282.
83. V. M. Ramakrishnan, Microwave Assisted Solvothermal Synthesis of Quasi Cubic F Doped TiO₂ Nanostructures and Its Performance as Dye Sensitized Solar Cell Photoanode, Int. J. Energy Res., 2021, 45(12), 17259–17268.
84. K. Subalakshmi and J. Senthilselvan, Effect of Fluorine-Doped TiO₂ Photoanode on Electron Transport, Recombination Dynamics and Improved DSSC Efficiency, Sol. Energy, 2018, 171, 914–928.
85. S. Nakade, S. Kambe, T. Kitamura, Y. Wada and S. Yanagida, Effects of Lithium Ion Density on Electron Transport in Nanoporous TiO₂ Electrodes, J. Phys. Chem. B, 2001, 105(38), 9150–9152.
86. D. F. Watson and G. J. Meyer, Cation Effects in Nanocrystalline Solar Cells, Coord. Chem. Rev., 2004, 248(13–14), 1391–1406.
87. A. Subramanian, J. S. Bow and H. W. Wang, The Effect of Li⁺ Intercalation on Different Sized TiO₂ Nanoparticles and the Performance of Dye-Sensitized Solar Cells, Thin Solid Films, 2012, 520(23), 7011–7017.
88. J. Manju and S. M. J. Jawhar, Synthesis of Magnesium-Doped TiO₂ Photoelectrodes for Dye-Sensitized Solar Cell Applications by Solvothermal Microwave Irradiation Method, J. Mater. Res., 2018, 33(11), 1534–1542.
89. Q. Liu, Photovoltaic Performance Improvement of Dye-Sensitized Solar Cells Based on Mg-Doped TiO₂ Thin Films, Electrochim. Acta, 2014, 129, 459–462.
90. M. I. Khan, W. A. Farooq, M. Saleem, K. A. Bhatti, M. Atif and A. Hanif, Phase Change, Band Gap Energy and Electrical Resistivity of Mg Doped TiO₂ Multilayer Thin Films for Dye Sensitized Solar Cells Applications., Ceram. Int., 2019, 45(17, Part A), 21436–21439.
91. K. Kakiage, T. Tokutome, S. Iwamoto, T. Kyomen and M. Hanaya, Fabrication of a Dye-Sensitized Solar Cell Containing a Mg-Doped TiO₂ Electrode and a Br³⁻/Br⁻ Redox Mediator with a High Open-Circuit Photovoltage of 1.21 V, Chem. Commun., 2013, 49(2), 179–180.
92. S. Iwamoto, Y. Sazanami, M. Inoue, T. Inoue, T. Hoshi, K. Shigaki, M. Kaneko and A. Maenosono, Fabrication of Dye-Sensitized Solar Cells with an Open-Circuit Photovoltage of 1 V, ChemSusChem, 2008, 1(5), 401–403.
93. C. Zhang, S. Chen, L. E. Mo, Y. Huang, H. Tian, L. Hu, Z. Huo, S. Dai, F. Kong and X. Pan, Charge Recombination and Band-Edge Shift in the Dye-Sensitized Mg²⁺-Doped TiO₂ Solar Cells, J. Phys. Chem. C, 2011, 115(33), 16418–16424.
94. W. Li, J. Yang, J. Zhang, S. Gao, Y. Luo and M. Liu, Improve Photovoltaic Performance of Titanium Dioxide Nanorods Based Dye-Sensitized Solar Cells by Ca-Doping, Mater. Res. Bull., 2014, 57, 177–183.
95. M. Pan, H. Liu, Z. Yao and X. Zhong, Enhanced Efficiency of Dye-Sensitized Solar Cells by Trace Amount ca-Doping in TiO₂ Photoelectrodes, J. Nanomater., 2015, 2015.
96. Q. Liu, Y. Zhou, Y. Duan, M. Wang, X. Zhao and Y. Lin, Enhanced Conversion Efficiency of Dye-Sensitized Titanium Dioxide Solar Cells by Ca-Doping., J. Alloys Compd., 2013, 548, 161–165.
97. S. Shakir, H. M. Abd-ur-Rehman, K. Yunus, M. Iwamoto and V. Periasamy, Fabrication of Un-Doped and Magnesium Doped TiO₂ Films by Aerosol Assisted Chemical Vapor Deposition for Dye Sensitized Solar Cells, J. Alloys Compd., 2018, 737, 740–747.
98. N. Rajamanickam and K. Ramachandran, Improved Photovoltaic Performance in Nano TiO₂ Based Dye Sensitized Solar Cells: Effect of TiCl4 Treatment and Sr Doping, J. Colloid Interface Sci., 2020, 580, 407–418.
99. H. F. Mehnane, C. Wang, K. K. Kondamareddy, W. Yu, W. Sun, H. Liu, S. Bai, W. Liu, S. Guo and X.-Z. Zhao, Hydrothermal Synthesis of TiO₂ Nanoparticles Doped with Trace Amounts of Strontium, and Their Application as Working Electrodes for Dye Sensitized Solar Cells: Tunable Electrical Properties & Enhanced Photo-Conversion Performance, RSC Adv., 2017, 7(4), 2358–2364.
100. N. Rajamanickam and K. Ramachandran, Improved Photovoltaic Performance in Nano TiO(2) Based Dye Sensitized Solar Cells: Effect of TiCl(4) Treatment and Sr Doping, J. Colloid Interface Sci., 2020, 580, 407–418.
101. X. H. Wu, S. Wang, Y. Guo, Z. Y. Xie, L. Han and Z. H. Jiang, Enhanced Energy Conversion Efficiency of La³⁺-Modified Nanoporous TiO₂ Electrode Sensitized with a Ruthenium Complex, Chin. J. Chem., 2008, 26(10), 1939–1943.
102. S. Yahav, S. Rühle, S. Greenwald, H. N. Barad, M. Shalom and A. Zaban, Strong Efficiency Enhancement of Dye-Sensitized Solar Cells Using a La-Modified TiCl₄ Treatment of Mesoporous TiO₂ Electrodes, J. Phys. Chem. C, 2011, 115(43), 21481–21486.
103. J. Zhang, Z. Zhao, X. Wang, T. Yu, J. Guan, Z. Yu, Z. Li and Z. Zou, Increasing the Oxygen Vacancy Density on the TiO₂ Surface by La-Doping for Dye-Sensitized Solar Cells, J. Phys. Chem. C, 2010, 114(43), 18396–18400.
104. M. Liu, Y. Hou and X. Qu, Enhanced Power Conversion Efficiency of Dye-Sensitized Solar Cells with Samarium Doped TiO₂ Photoanodes, J. Mater. Res., 2017, 32(18), 3469–3476.
105. H. Hafez, M. Saif and M. S. A. Abdel-Mottaleb, Down-Converting Lanthanide Doped TiO₂ Photoelectrodes for Efficiency Enhancement of Dye-Sensitized Solar Cells, J. Power Sources, 2011, 196(13), 5792–5796.
106. K. Singh, S. Harish, J. Archana, M. Navaneethan, M. Shimomura and Y. Hayakawa, Investigation of Gd-Doped Mesoporous TiO₂ Spheres for Environmental Remediation and Energy Applications, Appl. Surf. Sci., 2019, 489, 883–892.
107. E. Akman, S. Akin, T. Ozturk, B. Gulveren and S. Sonmezoglu, Europium and Terbium Lanthanide Ions Co-Doping in TiO₂ Photoanode to Synchronously Improve Light-Harvesting and Open-Circuit Voltage for High-Efficiency Dye-Sensitized Solar Cells, Sol. Energy, 2020, 202, 227–237.
108. J. Wang, J. Lin, J. Wu, M. Huang, Z. Lan, Y. Chen, S. Tang, L. Fan and Y. Huang, Application of Yb³⁺, Er³⁺-Doped Yttrium Oxyfluoride Nanocrystals in Dye-Sensitized Solar Cells, Electrochim. Acta, 2012, 70, 131–135.
109. J. Wu, J. Wang, J. Lin, Z. Lan, Q. Tang, M. Huang, Y. Huang, L. Fan, Q. Li and Z. Tang, Enhancement of the Photovoltaic Performance of Dye-Sensitized Solar Cells by Doping Y_0.78Yb_0.20Er_0.02F₃ in the Photoanode, Adv. Energy Mater., 2012, 2(1), 78–81.
110. J. Yu, Y. Yang, R. Fan, H. Zhang, L. Li, L. Wei, Y. Shi, K. Pan and H. Fu, Er³⁺ and Yb³⁺ Co-Doped TiO₂-XFx up-Conversion Luminescence Powder as a Light Scattering Layer with Enhanced Performance in Dye Sensitized Solar Cells, J. Power Sources, 2013, 243, 436–443.
111. Q. Li, J. Lin, J. Wu, Z. Lan, Y. Wang, F. Peng and M. Huang, Enhancing Photovoltaic Performance of Dye-Sensitized Solar Cell by Rare-Earth Doped Oxide of Lu₂O₃:(Tm³⁺, Yb³⁺), Electrochim. Acta, 2011, 56(14), 4980–4984.
112. G. Xie, Y. Wei, L. Fan and J. Wu, Application of Doped Rare-Earth Oxide TiO₂:(Tm³⁺, Yb³⁺) in Dye-Sensitized Solar Cells., J. Phys.: Conf. Ser., 2012, 339(1), 012010,  DOI:10.1088/1742-6596/339/1/012010.
113. X. Wei-Wei, D. Son-yuan, H. Lin-Hua, L. L. Yun and W. Kong-Jia, Influence of Yb Doped Nanoporous TiO₂ Films on Photovoltaic Performance of Dye Sensitized Solar Cells., Chin. Phys. Lett., 2006, 23(8), 2288–2291.
114. R. T. Ako, P. Ekanayake, A. L. Tan and D. J. Young, La Modified TiO₂ Photoanode and Its Effect on DSSC Performance: A Comparative Study of Doping and Surface Treatment on Deep and Surface Charge Trapping, Mater. Chem. Phys., 2016, 172, 105–112.
115. A. R. Tanyi, A. I. Rafieh, P. Ekaneyaka, A. L. Tan, D. J. Young, Z. Zheng, V. Chellappan, G. S. Subramanian and R. L. N. Chandrakanthi, Enhanced Efficiency of Dye-Sensitized Solar Cells Based on Mg and La Co-Doped TiO₂ Photoanodes, Electrochim. Acta, 2015, 178, 240–248.
116. J. Liqiang, S. Xiaojun, X. Baifu, W. Baiqi, C. Weimin and F. Honggang, The Preparation and Characterization of La Doped TiO₂ Nanoparticles and Their Photocatalytic Activity, J. Solid State Chem., 2004, 177(10), 3375–3382.
117. Neetu, S. Singh, P. Srivastava and L. Bahadur, Hydrothermal Synthesized Nd-Doped TiO₂ with Anatase and Brookite Phases as Highly Improved Photoanode for Dye-Sensitized Solar Cell., Sol. Energy, 2020, 208, 173–181.
118. L. Li, X. Yang, W. Zhang, H. Zhang and X. Li, Boron and Sulfur Co-Doped TiO₂ nanofilm as Effective Photoanode for High Efficiency CdS Quantum-Dot-Sensitized Solar Cells, J. Power Sources, 2014, 272, 508–512.
119. E. Finazzi, C. Valentin and G. Di Pacchioni, Boron-Doped Anatase TiO₂: Pure and Hybrid DFT Calculations, J. Phys. Chem. C, 2009, 113(1), 220–228.
120. M. Wang, S. Bai, A. Chen, Y. Duan, Q. Liu, D. Li and Y. Lin, Improved Photovoltaic Performance of Dye-Sensitized Solar Cells by Sb-Doped TiO₂ Photoanode, Electrochim. Acta, 2012, 77, 54–59.
121. H. Imahori, S. Hayashi, T. Umeyama, S. Eu, A. Oguro, S. Kang, Y. Matano, T. Shishido, S. Ngamsinlapasathian and S. Yoshikawa, Comparison of Electrode Structures and Photovoltaic Properties of Porphyrin-Sensitized Solar Cells with TiO₂ and Nb, Ge, Zr-Added TiO₂ Composite Electrodes, Langmuir, 2006, 22(26), 11405–11411.
122. A. Subramanian and H. W. Wang, Effects of Boron Doping in TiO₂ Nanotubes and the Performance of Dye-Sensitized Solar Cells, Appl. Surf. Sci., 2012, 258(17), 6479–6484.
123. H. Tian, L. Hu, W. Li, J. Sheng, S. Xu and S. Dai, A Facile Synthesis of Anatase N,B Codoped TiO₂ Anodes for Improved-Performance Dye-Sensitized Solar Cells, J. Mater. Chem., 2011, 21(20), 7074–7077.
124. H. Tian, L. Hu, C. Zhang, S. Chen, J. Sheng, L. Mo, W. Liu and S. Dai, Enhanced Photovoltaic Performance of Dye-Sensitized Solar Cells Using a Highly Crystallized Mesoporous TiO₂ Electrode Modified by Boron Doping, J. Mater. Chem., 2011, 21(3), 863–868.
125. J. S. Im, J. Yun, S. K. Lee and Y. S. Lee, Effects of Multi-Element Dopants of TiO₂ for High Performance in Dye-Sensitized Solar Cells, J. Alloys Compd., 2012, 513, 573–579.
126. D. Chen, D. Yang, Q. Wang and Z. Jiang, Effects of Boron Doping on Photocatalytic Activity and Microstructure of Titanium Dioxide Nanoparticles, Ind. Eng. Chem. Res., 2006, 45(12), 4110–4116.
127. A. Latini, C. Cavallo, F. K. Aldibaja, D. Gozzi, D. Carta, A. Corrias, L. Lazzarini and G. Salviati, Efficiency Improvement of DSSC Photoanode by Scandium Doping of Mesoporous Titania Beads, J. Phys. Chem. C, 2013, 117(48), 25276–25289.
128. A. Ray, A. Roy, P. Sadhukhan, S. R. Chowdhury, P. Maji, S. K. Bhattachrya and S. Das, Electrochemical Properties of TiO₂–V₂O₅ Nanocomposites as a High Performance Supercapacitors Electrode Material, Appl. Surf. Sci., 2018, 443, 581–591.
129. C. Kim, K. S. Kim, H. Y. Kim and Y. S. Han, Modification of a TiO₂ Photoanode by Using Cr-Doped TiO₂ with an Influence on the Photovoltaic Efficiency of a Dye-Sensitized Solar Cell, J. Mater. Chem., 2008, 18(47), 5809–5814.
130. L. C. K. Liau and C. C. Lin, Semiconductor Characterization of Cr³⁺-Doped Titania Electrodes with p-n Homojunction Devices, Thin Solid Films, 2008, 516(8), 1998–2002.
131. Y. Liu, H. Ran, J. Fan, X. Zhang, J. Mao and G. Shao, Fabrication and Photovoltaic Performance of Niobium Doped TiO₂ Hierarchical Microspheres with Exposed {001} Facets and High Specific Surface Area, Appl. Surf. Sci., 2017, 410, 241–248.
132. L. B. Patle, V. R. Huse and A. L. Chaudhari, Band Edge Movement and Structural Modifications in Transition Metal Doped TiO₂Nanocrystals for the Application of DSSC., Mater. Res. Express, 2017, 4(10), 105045.
133. B. Yacoubi, L. Samet, J. Bennaceur, A. Lamouchi and R. Chtourou, Properties of Transition Metal Doped-Titania Electrodes: Impact on Efficiency of Amorphous and Nanocrystalline Dye-Sensitized Solar Cells, Mater. Sci. Semicond. Process., 2015, 30, 361–367.
134. C. T. Wang, W. P. Wang and H. S. Lin, Niobium and Iron Co-Doped Titania Nanobelts for Improving Charge Collection in Dye-Sensitized TiO₂ Solar Cells, Ceram. Int., 2018, 44(15), 18032–18038.
135. A. E. Shalan and M. M. Rashad, Incorporation of Mn²⁺ and Co²⁺ to TiO₂ Nanoparticles and the Performance of Dye-Sensitized Solar Cells, Appl. Surf. Sci., 2013, 283, 975–981.
136. P. S. Archana, E. N. Kumar, C. Vijila, S. Ramakrishna, M. M. Yusoff and R. Jose, Random Nanowires of Nickel Doped TiO₂ with High Surface Area and Electron Mobility for High Efficiency Dye-Sensitized Solar Cells, Dalton Trans., 2013, 42, 1024–1032.
137. J. Y. Park, C. S. Kim, K. Okuyama, H. M. Lee, H. D. Jang, S. E. Lee and T. O. Kim, Copper and Nitrogen Doping on TiO₂ Photoelectrodes and Their Functions in Dye-Sensitized Solar Cells, J. Power Sources, 2016, 306, 764–771.
138. W. Liu, H. G. Wang, X. Wang, M. Zhang and M. Guo, Titanium Mesh Supported TiO₂ Nanowire Arrays/Nb-Doped TiO₂ Nanoparticles for Fully Flexible Dye-Sensitized Solar Cells with Improved Photovoltaic Properties, J. Mater. Chem. C, 2016, 4(47), 11118–11128.
139. A. Khlyustova, N. Sirotkin, T. Kusova, A. Kraev, V. Titov and A. Agafonov, Doped TiO₂: The Effect of Doping Elements on Photocatalytic Activity, Mater. Adv., 2020, 1(5), 1193–1201.
140. T. S. Bramhankar, S. S. Pawar, J. S. Shaikh, V. C. Gunge, N. I. Beedri, P. K. Baviskar, H. M. Pathan, P. S. Patil, R. C. Kambale and R. S. Pawar, Effect of Nickel–Zinc Co-Doped TiO₂ Blocking Layer on Performance of DSSCs., J. Alloys Compd., 2020, 817(xxxx), 152810.
141. A. K. Chandiran, F. Sauvage, L. Etgar and M. Graetzel, Ga³⁺ and Y³⁺ Cationic Substitution in Mesoporous TiO₂ Photoanodes for Photovoltaic Applications, J. Phys. Chem. C, 2011, 115(18), 9232–9240.
142. H. Su, Y. T. Huang, Y. H. Chang, P. Zhai, N. Y. Hau, P. C. H. Cheung, W. T. Yeh, T. C. Wei and S. P. Feng, The Synthesis of Nb-Doped TiO₂ Nanoparticles for Improved-Performance Dye Sensitized Solar Cells, Electrochim. Acta, 2015, 182, 230–237.
143. S. Rabhi, H. Belkacemi, M. Bououdina, A. Kerrami, L. Ait Brahem and E. Sakher, Effect of Ag Doping of TiO₂ Nanoparticles on Anatase-Rutile Phase Transformation and Excellent Photodegradation of Amlodipine Besylate, Mater. Lett., 2019, 236, 640–643.
144. M. Motlak, A. M. Hamza, M. G. Hammed and N. A. M. Barakat, Cd-Doped TiO₂ Nanofibers as Effective Working Electrode for the Dye Sensitized Solar Cells, Mater. Lett., 2019, 246, 206–209.
145. J. Liu, Y. Duan, X. Zhou and Y. Lin, Influence of VB Group Doped TiO₂ on Photovoltaic Performance of Dye-Sensitized Solar Cells, Appl. Surf. Sci., 2013, 277, 231–236.
146. X. Zhang, F. Liu, Q. L. Huang, G. Zhou and Z. S. Wang, Dye-Sensitized W-Doped TiO₂ Solar Cells with a Tunable Conduction Band and Suppressed Charge Recombination, J. Phys. Chem. C, 2011, 115(25), 12665–12671.
147. H. Seo, Y. Wang, D. Ichida, G. Uchida, N. Itagaki, K. Koga, M. Shiratani, S. H. Nam and J. H. Boo, Improvement on the Electron Transfer of Dye-Sensitized Solar Cell Using Vanadium Doped TiO₂, Jpn. J. Appl. Phys., 2013, 52(11 PART 2), 2–6.
148. X. Li, Z. Guo and T. He, The Doping Mechanism of Cr into TiO₂ and Its Influence on the Photocatalytic Performance, Phys. Chem. Chem. Phys., 2013, 15(46), 20037–20045.
149. Y. Xie, N. Huang, S. You, Y. Liu, B. Sebo, L. Liang, X. Fang, W. Liu, S. Guo and X. Z. Zhao, Improved Performance of Dye-Sensitized Solar Cells by Trace Amount Cr-Doped TiO₂ Photoelectrodes, J. Power Sources, 2013, 224, 168–173.
150. H. H. Nguyen, G. Gyawali, A. Martinez-Oviedo, Y. K. Kshetri and S. W. Lee, Physicochemical Properties of Cr-Doped TiO₂ Nanotubes and Their Application in Dye-Sensitized Solar Cells, J. Photochem. Photobiol., A, 2020, 397, 112514.
151. B. Ünlü and M. Özacar, Effect of Cu and Mn Amounts Doped to TiO₂ on the Performance of DSSCs, Sol. Energy, 2020, 196, 448–456.
152. S. Shah, N. N. S. Baharun, S. N. F. Yusuf and A. K. Arof, Efficiency Enhancement of Dye-Sensitized Solar Cells (DSSCs) Using Copper Nanopowder (CuNW) in TiO₂ as Photoanode, IOP Conf. Ser.: Mater. Sci. Eng., 2019, 515(1), 012002,  DOI:10.1088/1757-899X/515/1/012002.
153. B. Ünlü, S. Çakar and M. Özacar, The Effects of Metal Doped TiO₂ and Dithizone-Metal Complexes on DSSCs Performance, Sol. Energy, 2018, 166, 441–449.
154. R. T. Ako, P. Ekanayake, D. J. Young, J. Hobley, V. Chellappan, A. L. Tan, S. Gorelik, G. S. Subramanian and C. M. Lim, Evaluation of Surface Energy State Distribution and Bulk Defect Concentration in DSSC Photoanodes Based on Sn, Fe, and Cu Doped TiO₂, Appl. Surf. Sci., 2015, 351, 950–961.
155. L. C. K. Liau and C. C. Lin, Fabrication and Characterization of Fe³⁺-Doped Titania Semiconductor Electrodes with p–n Homojunction Devices, Appl. Surf. Sci., 2007, 253(21), 8798–8801.
156. M. I. Khan, Synthesis, Characterization and Application of Co Doped TiO₂ Multilayer Thin Films, Results Phys., 2018, 9, 359–363.
157. B. Ünlü, S. Çakar and M. Özacar, The Effects of Metal Doped TiO₂ and Dithizone-Metal Complexes on DSSCs Performance, Sol. Energy, 2018, 166, 441–449.
158. A. Malik, S. Hameed, M. J. Siddiqui, M. M. Haque, K. Umar, A. Khan and M. Muneer, Electrical and Optical Properties of Nickel- and Molybdenum-Doped Titanium Dioxide Nanoparticle: Improved Performance in Dye-Sensitized Solar Cells, J. Mater. Eng. Perform., 2014, 23, 3184–3192.
159. K. Sahu, M. Dhonde and V. V. S. Murty, Microwave-Assisted Hydrothermal Synthesis of Cu-Doped TiO₂ Nanoparticles for Efficient Dye-Sensitized Solar Cell with Improved Open-Circuit Voltage, Int. J. Energy Res., 2021, 45(4), 5423–5432.
160. Y. Wang, Y. Hao, H. Cheng, J. Ma and B. I. N. Xu, The Photoelectrochemistry of Transition Metal-Ion-Doped TiO₂ Nanocrystalline Electrodes and Higher Solar Cell Conversion Efficiency Based on Zn²⁺-Doped TiO₂ Electrode, J. Mater. Sci., 1999, 34, 2773–2779.
161. T. N. D. S. Cells, F. Zhu, P. Zhang, X. Wu, L. Fu and J. Zhang, The Origin of Higher Open-Circuit Voltage in Zn-Doped, ChemPhysChem, 2012, 13(16), 1439–4235.
162. P. Taylor, H. Chen, L. Zhao, Y. Xiang, Y. He, G. Song, X. Wang and F. Liang, A Novel Zn–TiO₂/C@SiO₂ Nanoporous Material on Rice Husk for Photocatalytic Applications under Visible Light, Desalin. Water Treat., 2015, 57(21), 9660–9670.
163. K. Wang and H. Teng, Zinc-Doping in TiO₂ Films to Enhance Electron Transport in Dye-Sensitized Solar Cells under Low-Intensity Illumination, Phys. Chem. Chem. Phys., 2009, 11, 9489–9496.
164. J. Zhang, Z. Han, Z. Fu, Y. Wen and F. Zhao, Study of N–Ag–Zn/TiO₂, N–Ag–Zr/TiO₂ with N719 and P3OT Co-Sensitization Effect on the Performance of Dye-Sensitized Solar Cell, J. Sol-Gel Sci. Technol., 2016, 78(1), 207–217.
165. B. Zhao, J. Wang, H. Li, H. Wang, X. Jia and P. Su, The Influence of Yttrium Dopant on the Properties of Anatase Nanoparticles and the Performance of Dye-Sensitized Solar Cells, Phys. Chem. Chem. Phys., 2015, 17(22), 14836–14842.
166. X. Qu, Y. Hou, M. Liu, L. Shi, M. Zhang, H. Song and F. Du, Yttrium Doped TiO₂ Porous Film Photoanode for Dye-Sensitized Solar Cells with Enhanced Photovoltaic Performance, Results Phys., 2016, 6, 1051–1058.
167. M. Moradzaman, M. R. Mohammadi and H. Nourizadeh, Efficient Dye-Sensitized Solar Cells Based on CNTs and Zr-Doped TiO₂ Nanoparticles, Mater. Sci. Semicond. Process., 2015, 40, 383–390.
168. N. Tsvetkov, L. Larina, O. Shevaleevskiy and B. T. Ahn, Effect of Nb Doping of TiO₂ Electrode on Charge Transport in Dye-Sensitized Solar Cells, J. Electrochem. Soc., 2011, 158(11), B1281.
169. L. Long, L. Wu, X. Yang and X. Li, Photoelectrochemical Performance of Nb-Doped TiO₂ Nanoparticles Fabricated by Hydrothermal Treatment of Titanate Nanotubes in Niobium Oxalate Aqueous Solution, J. Mater. Sci. Technol., 2014, 30(8), 765–769.
170. N. Yamada, T. Hitosugi, J. Kasai, N. L. H. Hoang, S. Nakao, Y. Hirose, T. Shimada and T. Hasegawa, Direct Growth of Transparent Conducting Nb-Doped Anatase TiO₂ Polycrystalline Films on Glass, J. Appl. Phys., 2009, 105(12), 123702,  DOI:10.1063/1.3148267.
171. T. Nikolay, L. Larina, O. Shevaleevskiy and B. T. Ahn, Electronic Structure Study of Lightly Nb-Doped TiO₂ Electrode for Dye-Sensitized Solar Cells, Energy Environ. Sci., 2011, 4(4), 1480–1486.
172. Y. Liu, H. Ran, J. Fan, X. Zhang, J. Mao and G. Shao, Fabrication and Photovoltaic Performance of Niobium Doped TiO₂ Hierarchical Microspheres with Exposed {001} Facets and High Specific Surface Area, Appl. Surf. Sci., 2017, 410, 241–248.
173. W. Liu, H. Wang, X. Wang, M. Zhang and M. Guo, Titanium Mesh Supported TiO₂ Nanowire Arrays/Nb-Doped TiO₂ Nanoparticles for Fully Flexible Dye-Sensitized Solar Cells with Improved Photovoltaic Properties, J. Mater. Chem. C, 2016, 4(47), 11118–11128.
174. B.-R. Koo, D.-H. Oh and H.-J. Ahn, Influence of Nb-Doped TiO₂ Blocking Layers as a Cascading Band Structure for Enhanced Photovoltaic Properties, Appl. Surf. Sci., 2018, 433, 27–34.
175. J.-M. Song, P.-J. Wang, L.-H. Chan, C.-M. Chen, W.-F. Ho and S.-Y. Chen, Efficiency Enhancement of Dye-Sensitized Solar Cells Using Ti–Nb Alloy Photoanodes with Mesoporous Oxide Surface, J. Electrochem. Soc., 2020, 167(4), 046501.
176. P. Nbelayim, G. Kawamura, W. Kian Tan, H. Muto and A. Matsuda, Systematic Characterization of the Effect of Ag@TiO₂ Nanoparticles on the Performance of Plasmonic Dye-Sensitized Solar Cells, Sci. Rep., 2017, 7(1), 1–12.
177. K. Usha, P. Kumbhakar and B. Mondal, Effect of Ag-Doped TiO₂ Thin Film Passive Layers on the Performance of Photo-Anodes for Dye-Sensitized Solar Cells, Mater. Sci. Semicond. Process., 2016, 43, 17–24.
178. J. Liu, H. Yang, W. Tan, X. Zhou and Y. Lin, Photovoltaic Performance Improvement of Dye-Sensitized Solar Cells Based on Tantalum-Doped TiO₂ Thin Films, Electrochim. Acta, 2010, 56(1), 396–400.
179. J.-H. Choi, S.-H. Kwon, Y.-K. Jeong, I. Kim and K.-H. Kim, Atomic Layer Deposition of Ta-Doped TiO₂ Electrodes for Dye-Sensitized Solar Cells, J. Electrochem. Soc., 2011, 158(6), B749.
180. I. M. A. Mohamed, V.-D. Dao, N. A. M. Barakat, A. S. Yasin, A. Yousef and H.-S. Choi, Efficiency Enhancement of Dye-Sensitized Solar Cells by Use of ZrO₂-Doped TiO₂ Nanofibers Photoanode, J. Colloid Interface Sci., 2016, 476, 9–19.
181. L. Zhao, C. Zhong, Y. Wang, S. Wang, B. Dong and L. Wan, Ag Nanoparticle-Decorated 3D Flower-like TiO₂ Hierarchical Microstructures Composed of Ultrathin Nanosheets and Enhanced Photoelectrical Conversion Properties in Dye-Sensitized Solar Cells, J. Power Sources, 2015, 292, 49–57.
182. N. Massihi, M. R. Mohammadi, A. M. Bakhshayesh and M. Abdi-Jalebi, Controlling Electron Injection and Electron Transport of Dye-Sensitized Solar Cells Aided by Incorporating CNTs into a Cr-Doped TiO₂ Photoanode, Electrochim. Acta, 2013, 111, 921–929.
183. R. Ghosh, Y. Hara, L. Alibabaei, K. Hanson, S. Rangan, R. Bartynski, T. J. Meyer and R. Lopez, Increasing Photocurrents in Dye Sensitized Solar Cells with Tantalum-Doped Titanium Oxide Photoanodes Obtained by Laser Ablation, ACS Appl. Mater. Interfaces, 2012, 4(9), 4566–4570.
184. P. Zhong, X. Chen, B. Niu, C. Li, Y. Wang, H. Xi, Y. Lei, Z. Wang and X. Ma, Niobium Doped TiO₂ Nanorod Arrays as Efficient Electron Transport Materials in Photovoltaic, J. Power Sources, 2020, 450, 227715,  DOI:10.1016/j.jpowsour.2020.227715.
185. S. K. M. Saad, A. A. Umar, M. Y. A. Rahman and M. M. Salleh, Porous Zn-Doped TiO₂ Nanowall Photoanode: Effect of Zn²⁺ Concentration on the Dye-Sensitized Solar Cell Performance., Appl. Surf. Sci., 2015, 353, 835–842.
186. R. Bendoni, E. Mercadelli, N. Sangiorgi, A. Strini, A. Sangiorgi and A. Sanson, Alternative Route for the Preparation of Zr-Doped TiO₂ Layers for Energy and Environmental Applications, Ceram. Int., 2015, 41(8), 9899–9909.
187. B. Sebo, N. Huang, Y. Liu, Q. Tai, L. Liang, H. Hu, S. Xu and X. Z. Zhao, Dye-Sensitized Solar Cells Enhanced by Optical Absorption, Mediated by TiO₂ Nanofibers and Plasmonic Ag Nanoparticles, Electrochim. Acta, 2013, 112, 458–464.
188. M. Dhonde, K. Sahu, V. V. S. Murty, S. S. Nemala and P. Bhargava, Surface Plasmon Resonance Effect of Cu Nanoparticles in a Dye Sensitized Solar Cell, Electrochim. Acta, 2017, 249, 89–95.
189. R. Tanyi, P. Ekanayake, D. James, J. Hobley, V. Chellappan, A. Ling, S. Gorelik and G. Sandhya, Evaluation of Surface Energy State Distribution and Bulk Defect Concentration in DSSC Photoanodes Based on Sn, Fe, and Cu Doped TiO₂, Appl. Surf. Sci., 2015, 351, 950–961.
190. J. Du, J. Qi, D. Wang and Z. Tang, Facile Synthesis of Au@TiO₂ Core-Shell Hollow Spheres for Dye-Sensitized Solar Cells with Remarkably Improved Efficiency, Energy Environ. Sci., 2012, 5(5), 6914–6918.
191. Y. Y. Li, J. G. Wang, X. R. Liu, C. Shen, K. Xie and B. Wei, Au/TiO₂ Hollow Spheres with Synergistic Effect of Plasmonic Enhancement and Light Scattering for Improved Dye-Sensitized Solar Cells, ACS Appl. Mater. Interfaces, 2017, 9(37), 31691–31698.
192. B. R. Koo, D. H. Oh and H. J. Ahn, Influence of Nb-Doped TiO₂ Blocking Layers as a Cascading Band Structure for Enhanced Photovoltaic Properties, Appl. Surf. Sci., 2018, 433, 27–34.
193. X. Lü, X. Mou, J. Wu, D. Zhang, L. Zhang, F. Huang, F. Xu and S. Huang, Improved-Performance Dye-Sensitized Solar Cells Using Nb-Doped TiO₂ Electrodes: Efficient Electron Injection and Transfer, Adv. Funct. Mater., 2010, 20(3), 509–515.
194. P. Xiang, W. Ma, T. Xiao, L. Jiang, X. Tan and T. Shu, Ta-Doped Hierarchical TiO₂ Spheres for Dye-Sensitized Solar Cells, J. Alloys Compd., 2016, 656, 45–50.
195. S. K. M. Saad, A. A. Umar, M. Y. A. Rahman and M. M. Salleh, Porous Zn-Doped TiO₂ Nanowall Photoanode: Effect of Zn²⁺ Concentration on the Dye-Sensitized Solar Cell Performance, Appl. Surf. Sci., 2015, 353, 835–842.
196. I. M. A. Mohamed, V. Dao, N. A. M. Barakat, A. S. Yasin, A. Yousef and H. Choi, Efficiency Enhancement of Dye-Sensitized Solar Cells by Use of ZrO₂-Doped TiO₂ Nanofibers Photoanode, J. Colloid Interface Sci., 2016, 476, 9–19.
197. E. Jalali-moghadam and Z. Shariatinia, Al³⁺ Doping into TiO₂ Photoanodes Improved the Performances of Amine Anchored CdS Quantum Dot Sensitized Solar Cells, Mater. Res. Bull., 2018, 98, 121–132.
198. A. K. Chandiran, F. Sauvage, L. Etgar and M. Graetzel, Ga³⁺ and Y³⁺ Cationic Substitution in Mesoporous TiO₂ Photoanodes for Photovoltaic Applications, J. Phys. Chem. C, 2011, 115(18), 9232–9240.
199. B. Baptayev, S. Adilov and M. P. Balanay, Surface Modification of TiO₂ Photoanodes with In³⁺ Using a Simple Soaking Technique for Enhancing the Efficiency of Dye-Sensitized Solar Cells, J. Photochem. Photobiol., A, 2020, 394, 112468,  DOI:10.1016/j.jphotochem.2020.112468.
200. M. I. Khan, G. Hassan, M. S. Hasan, S. A. Abubshait, H. A. Abubshait, W. Al-Masry, Q. Mahmood, A. Mahmood and S. M. Ramay, Investigations on the Efficiency Variation of Zinc and Gallium Co-Doped TiO₂ Based Dye Sensitized Solar Cells, Ceram. Int., 2020, 46(16), 24844–24849.
201. M. N. An, S. Radiman, N. M. Huang and M. A. Yarmo, Sol–Gel Hydrothermal Synthesis of Bismuth–TiO₂ Nanocubes for Dye-Sensitized Solar Cell, Ceram. Int., 2010, 36(7), 2215–2220.
202. F. Huang, B. Y. Cheng and R. A. Caruso, Al-Doped TiO₂ Photoanode for Dye-Sensitized Solar Cells, Aust. J. Chem., 2011, 64, 820–824.
203. A. M. Bakhshayesh and N. Farajisafiloo, Anatase-Stabilised Al_xTi_1−xO₂ Photoanodes Containing Uniform Spherical Particles for Efficient Dye-Sensitized Solar Cells, Appl. Surf. Sci., 2015, 331, 58–65.
204. K. Manoharan and P. Venkatachalam, Photoelectrochemical Performance of Dye Sensitized Solar Cells Based on Aluminum-Doped Titanium Dioxide Structures, Mater. Sci. Semicond. Process., 2015, 30, 208–217.
205. R. Li, Y. Zhao, R. Hou, X. Ren, S. Yuan, Y. Lou, Z. Wang, D. Li and L. Shi, Enhancement of Power Conversion Efficiency of Dye Sensitized Solar Cells by Modifying Mesoporous TiO₂ Photoanode with Al-Doped TiO₂ Layer, J. Photochem. Photobiol., A, 2016, 319–320, 62–69.
206. J. Chae, D. Y. Kim, S. Kim and M. Kang, Photovoltaic Efficiency on Dye-Sensitized Solar Cells (DSSC) Assembled Using Ga-Incorporated TiO₂ Materials, J. Ind. Eng. Chem., 2010, 16(6), 906–911.
207. X. Sun, Q. Zhang, Y. Liu, N. Huang, P. Sun, T. Peng, T. Peng and X. Z. Zhao, Photovoltaic Performance Improvement of Dye-Sensitized Solar Cells through Introducing In-Doped TiO₂ Film at Conducting Glass and Mesoporous TiO₂ Interface as an Efficient Compact Layer, Electrochim. Acta, 2014, 129, 276–282.
208. Y. Duan, N. Fu, Q. Liu, Y. Fang, X. Zhou, J. Zhang and Y. Lin, Sn-Doped TiO₂ Photoanode for Dye-Sensitized Solar Cells, J. Phys. Chem. C, 2012, 8–13.
209. J. Chae, D. Y. Kim, S. Kim and M. Kang, Photovoltaic Efficiency on Dye-Sensitized Solar Cells (DSSC) Assembled Using Ga-Incorporated TiO₂ Materials., J. Ind. Eng. Chem., 2010, 16(6), 906–911.
210. Y. Duan, N. Fu, Q. Zhang, Y. Fang, X. Zhou and Y. Lin, Influence of Sn Source on the Performance of Dye-Sensitized Solar Cells (Based on Sn-Doped TiO₂ Photoanodes: A Strategy for Choosing an Appropriate Doping Source), Electrochim. Acta, 2013, 107, 473–480.
211. S. P. Berglund, S. Hoang, R. L. Minter, R. R. Fullon and C. B. Mullins, Investigation of 35 Elements as Single Metal Oxides, Mixed Metal Oxides, or Dopants for Titanium Dioxide for Dye-Sensitized Solar Cells, J. Phys. Chem. C, 2013, 117(48), 25248–25258.
212. Q. Hou, Y. Zheng, J. F. Chen, W. Zhou, J. Deng and X. Tao, Visible-Light-Response Iodine-Doped Titanium Dioxide Nanocrystals for Dye-Sensitized Solar Cells, J. Mater. Chem., 2011, 21(11), 3877–3883.
213. Z. He, L. Zhan, F. Hong, S. Song, Z. Lin, J. Chen and M. Jin, A Visible Light-Responsive Iodine-Doped Titanium Dioxide Nanosphere, J. Environ. Sci., 2011, 23(1), 166–170.
214. G. Liu, C. Sun, X. Yan, L. Cheng, Z. Chen, X. Wang, L. Wang, S. C. Smith, G. Q. Lu and H. M. Cheng, Iodine Doped Anatase TiO₂ Photocatalyst with Ultra-Long Visible Light Response: Correlation between Geometric/Electronic Structures and Mechanisms, J. Mater. Chem., 2009, 19(18), 2822–2829.
215. J. Song, H. B. Yang, X. Wang, S. Y. Khoo, C. C. Wong, X. W. Liu and C. M. Li, Improved Utilization of Photogenerated Charge Using Fluorine-Doped TiO₂ Hollow Spheres Scattering Layer in Dye-Sensitized Solar Cells, ACS Appl. Mater. Interfaces, 2012, 4(7), 3712–3717.
216. S. In, K. Bae, H. Ahn and T. Seong, Improved Efficiency of Dye-Sensitized Solar Cells through Fluorine-Doped TiO₂ Blocking Layer, Ceram. Int., 2013, 39(7), 8097–8101.
217. K. Subalakshmi and J. Senthilselvan, Effect of Fluorine-Doped TiO₂ Photoanode on Electron Transport, Recombination Dynamics and Improved DSSC Efficiency, Sol. Energy, 2018, 171, 914–928.
218. S. I. Noh, K. N. Bae, H. J. Ahn and T. Y. Seong, Improved Efficiency of Dye-Sensitized Solar Cells through Fluorine-Doped TiO₂ Blocking Layer, Ceram. Int., 2013, 39(7), 8097–8101.
219. V. Madurai Ramakrishnan, M. Natarajan, S. Pitchaiya, A. Santhanam, D. Velauthapillai and A. Pugazhendhi, Microwave Assisted Solvothermal Synthesis of Quasi Cubic F Doped TiO₂ Nanostructures and Its Performance as Dye Sensitized Solar Cell Photoanode, Int. J. Energy Res., 2021, 45(12), 17259–17268.
220. P. Xiang, F. Lv, T. Xiao, L. Jiang, X. Tan and T. Shu, Improved Performance of Quasi-Solid-State Dye-Sensitized Solar Cells Based on Iodine-Doped TiO₂ Spheres Photoanodes, J. Alloys Compd., 2018, 741, 1142–1147.
221. C. Y. Neo and J. Ouyang, LiF-Doped Mesoporous TiO₂ as the Photoanode of Highly Efficient Dye-Sensitized Solar Cells, J. Power Sources, 2013, 241, 647–653.
222. P. Xiang, F. Lv, T. Xiao, L. Jiang, X. Tan and T. Shu, Improved Performance of Quasi-Solid-State Dye-Sensitized Solar Cells Based on Iodine-Doped TiO₂ Spheres Photoanodes, J. Alloys Compd., 2018, 741, 1142–1147.
223. M. S. Mahmoud, M. S. Akhtar, I. M. A. Mohamed, R. Hamdan, Y. A. Dakka and N. A. M. Barakat, Demonstrated Photons to Electron Activity of S-Doped TiO₂ Nanofibers as Photoanode in the DSSC, Mater. Lett., 2018, 225, 77–81.
224. T. Shu, P. Xiang, Z. M. Zhou, H. Wang, G. H. Liu, H. W. Han and Y. D. Zhao, Mesoscopic Nitrogen-Doped TiO₂ Spheres for Quantum Dot-Sensitized Solar Cells, Electrochim. Acta, 2012, 68, 166–171.
225. M. Motlak, M. S. Akhtar, N. A. M. Barakat, A. M. Hamza, O. B. Yang and H. Y. Kim, High-Efficiency Electrode Based on Nitrogen-Doped TiO₂ Nanofibers for Dye-Sensitized Solar Cells, Electrochim. Acta, 2014, 115, 493–498.
226. H. Wang, H. Li, J. Wang, J. Wu, D. Li, M. Liu and P. Su, Nitrogen-Doped TiO₂ Nanoparticles Better TiO₂ Nanotube Array Photo-Anodes for Dye Sensitized Solar Cells, Electrochim. Acta, 2014, 137, 744–750.
227. Z. S. Seddigi, S. A. Ahmed, S. Sardar and S. K. Pal, Carbonate Doping in TiO₂ Microsphere: The Key Parameter Influencing Others for Efficient Dye Sensitized Solar Cell, Sci. Rep., 2016, 6, 1–9.
228. M. Zhu, Y. Dong, J. Xu, B. Zhang and Y. Feng, Enhanced Photovoltaic Performance of Dye-Sensitized Solar Cells (DSSCs) Using Graphdiyne-Doped TiO₂ Photoanode, J. Mater. Sci., 2018, 54(6), 4893–4904.
229. S. Shogh, R. Mohammadpour, A. Iraji zad and N. Taghavinia, A New Strategy on Utilizing Nitrogen Doped TiO₂ in Nanostructured Solar Cells: Embedded Multifunctional N-TiO₂ Scattering Particles in Mesoporous Photoanode, Mater. Res. Bull., 2015, 72, 64–69.
230. S. H. Kang, H. S. Kim, J. Y. Kim and Y. E. Sung, Enhanced Photocurrent of Nitrogen-Doped TiO₂ Film for Dye-Sensitized Solar Cells, Mater. Chem. Phys., 2010, 124(1), 422–426.
231. M. A. K. L. Dissanayake, J. M. K. W. Kumari, G. K. R. Senadeera, C. A. Thotawatthage, B. E. Mellander and I. Albinsson, A Novel Multilayered Photoelectrode with Nitrogen Doped TiO₂ for Efficiency Enhancement in Dye Sensitized Solar Cells, J. Photochem. Photobiol., A, 2017, 349, 63–72.
232. Y. Gao, Y. Feng, B. Zhang, F. Zhang, X. Peng, L. Liu and S. Meng, Double-N Doping: A New Discovery about N-Doped TiO₂ Applied in Dye-Sensitized Solar Cells, RSC Adv., 2014, 4(33), 16992–16998.
233. T. Lindgren, J. M. Mwabora, E. Avandaño, J. Jonsson, A. Hoel, C. G. Granqvist and S. E. Lindquist, Photoelectrochemical and Optical Properties of Nitrogen Doped Titanium Dioxide Films Prepared by Reactive DC Magnetron Sputtering, J. Phys. Chem. B, 2003, 107(24), 5709–5716.
234. X. Wang, Y. Yang, Z. Jiang and R. Fan, Preparation of TiN_xO_2−x Photoelectrodes with NH₃ under Controllable Middle Pressures for Dye-Sensitized Solar Cells, Eur. J. Inorg. Chem., 2009, 3481–3487.
235. F. Guo, H. Hu, Q. D. Tai, B. L. Chen, B. Sebo, C. H. Bu, J. H. Xu and X. Z. Zhao, Facile Preparation of Nanofibrous Polyaniline Thin Film as Counter Electrodes for Dye Sensitized Solar Cells, J. Renewable Sustainable Energy, 2012, 4(2), 023109,  DOI:10.1063/1.3699618.
236. W. Liu, Z. Feng and W. Cao, Preparation of Large-Area Dye-Sensitized Solar Cells Based on Hydrothermally Synthesized Nitrogen-Doped TiO₂ Powders, Res. Chem. Intermed., 2013, 39(4), 1623–1631.
237. D. A. Duarte, M. Massi and A. S. Da Silva Sobrinho, Development of Dye-Sensitized Solar Cells with Sputtered N-Doped TiO₂ Thin Films: From Modeling the Growth Mechanism of the Films to Fabrication of the Solar Cells, Int. J. Photoenergy, 2014, 2014.
238. S. P. Lim, A. Pandikumar, N. M. Huang, H. N. Lim, G. Gu and T. L. Ma, Promotional Effect of Silver Nanoparticles on the Performance of N-Doped TiO₂ Photoanode-Based Dye-Sensitized Solar Cells, RSC Adv., 2014, 4(89), 48236–48244.
239. W. Mekprasart, S. Suphankij, T. Tangcharoen, A. Simpraditpan and W. Pecharapa, Modification of Dye-Sensitized Solar Cell Working Electrode Using TiO₂ Nanoparticle/N-Doped TiO₂ Nanofiber Composites., Phys. Status Solidi A, 2014, 211(8), 1745–1751.
240. A. Bjelajac, R. Petrović, M. Popović, Z. Rakočević, G. Socol, I. N. Mihailescu and D. Janaćković, Doping of TiO₂ Nanotubes with Nitrogen by Annealing in Ammonia for Visible Light Activation: Influence of Pre- and Post-Annealing in Air, Thin Solid Films, 2019, 692, 137598.
241. W. Guo, Y. Shen, G. Boschloo, A. Hagfeldt and T. Ma, Influence of Nitrogen Dopants on N-Doped TiO₂ Electrodes and Their Applications in Dye-Sensitized Solar Cells, Electrochim. Acta, 2011, 56(12), 4611–4617.
242. L. Qi, C. Li and Y. Chen, Dye-Sensitized Solar Cells Based on Nitrogen-Doped TiO₂-B Nanowire/TiO₂ Nanoparticle Composite Photoelectrode, Chem. Phys. Lett., 2012, 539–540, 128–132.
243. C. K. Lim, H. Huang, C. L. Chow, P. Y. Tan, X. Chen, M. S. Tse and O. K. Tan, Enhanced Charge Transport Properties of Dye-Sensitized Solar Cells Using TiN_xO_y Nanostructure Composite Photoanode, J. Phys. Chem. C, 2012, 116(37), 16659–19664.
244. J. Zhang, Z. Fu, Q. Lv, X. Yang and W. Cao, Technology of Preparing Anode Films and the Effect of Co-Sensitization in Dye-Sensitized Solar Cells, J. Sol-Gel Sci. Technol., 2012, 63(3), 554–562.
245. T. Sano, N. Mera, Y. Kanai, C. Nishimoto, S. Tsutsui, T. Hirakawa and N. Negishi, Origin of Visible-Light Activity of N-Doped TiO₂ Photocatalyst: Behaviors of N and S Atoms in a Wet N-Doping Process., Appl. Catal., B, 2012, 128, 77–83.
246. T. Ohno, M. Akiyoshi, T. Umebayashi, K. Asai, T. Mitsui and M. Matsumura, Preparation of S-Doped TiO₂ Photocatalysts and Their Photocatalytic Activities under Visible Light, Appl. Catal., A, 2004, 265(1), 115–121.
247. Z. Xing, Z. Li, X. Wu, G. Wang and W. Zhou, In-Situ S-Doped Porous Anatase TiO₂ Nanopillars for High-Efficient Visible-Light Photocatalytic Hydrogen Evolution, Int. J. Hydrogen Energy, 2016, 41(3), 1535–1541.
248. B. A. D. Williamson, J. Buckeridge, N. P. Chadwick, S. Sathasivam, C. J. Carmalt, I. P. Parkin and D. O. Scanlon, Dispelling the Myth of Passivated Codoping in TiO₂, Chem. Mater., 2019, 31(7), 2577–2589.
249. B. S. Constantino, C. H. Gonzalez, N. L. Gomes, N. D. G. Silva, R. O. Domingues, R. A. S. Ferreira and Y. P. Yadava, Effect of TiO₂ Doping on the Sintering Behavior of Zr-Based Ceramic for Use as TBC, J. Aust. Ceram. Soc., 2021, 57(5), 1425–1434.
250. A. M. Bakhshayesh and N. Bakhshayesh, Enhanced Short Circuit Current Density of Dye-Sensitized Solar Cells Aided by Sr,V Co-Doped TiO₂ Particles, Mater. Sci. Semicond. Process., 2016, 41, 92–101.
251. Y. Li, L. Jia, C. Wu, S. Han, Y. Gong, B. Chi, J. Pu and L. Jian, Mesoporous (N, S)-Codoped TiO₂ Nanoparticles as Effective Photoanode for Dye-Sensitized Solar Cells, J. Alloys Compd., 2012, 512(1), 23–26.
252. S. P. Lim, A. Pandikumar, H. N. Lim, R. Ramaraj and N. M. Huang, Boosting Photovoltaic Performance of Dye-Sensitized Solar Cells Using Silver Nanoparticle-Decorated N,S-Co-Doped-TiO₂ Photoanode, Sci. Rep., 2015, 5, 1–14.
253. F. Lv, Y. Ma, P. Xiang, T. Shu, X. Tan, L. Qiu, L. Jiang, T. Xiao and X. Chen, N–I Co-Doped TiO₂ Compact Film as a Highly Effective n-Type Electron Blocking Layer for Solar Cells, J. Alloys Compd., 2020, 837, 155555.
254. S. Yang, H. Xue, H. Wang, H. Kou, J. Wang and G. Zhu, Improved Efficiency of Dye-Sensitized Solar Cells Applied with Nanostructured N–F Doped TiO₂ Electrode, J. Phys. Chem. Solids, 2012, 73(7), 911–916.
255. Y. Duan, J. Zheng, M. Xu, X. Song, N. Fu, Y. Fang, X. Zhou, Y. Lin and F. Pan, Metal and F Dual-Doping to Synchronously Improve Electron Transport Rate and Lifetime for TiO₂ Photoanode to Enhance Dye-Sensitized Solar Cells Performances, J. Mater. Chem. A, 2015, 3(10), 5692–5700.
256. A. Gupta, K. Sahu, M. Dhonde and V. V. S. Murty, Novel Synergistic Combination of Cu/S Co-Doped TiO₂ Nanoparticles Incorporated as Photoanode in Dye Sensitized Solar Cell., Sol. Energy, 2020, 203, 296–303.
257. Y. F. Jiang, Y. Y. Chen, B. Zhang and Y. Q. Feng, N, La Co-Doped TiO₂ for Use in Low-Temperature-Based Dye-Sensitized Solar Cells, J. Electrochem. Soc., 2016, 163(10), F1133–F1138.
258. W. Wang, Y. Liu, J. Sun and L. Gao, Nitrogen and Yttrium Co-Doped Mesoporous Titania Photoanodes Applied in DSSCs, J. Alloys Compd., 2016, 659, 15–22.
259. L. Wang, L. Jia and Q. Li, A Novel Sulfur Source for Biosynthesis of (Ag, S)-Modified TiO₂ Photoanodes in DSSC, Mater. Lett., 2014, 123, 83–86.
260. Y. Duan, J. Zheng, M. Xu, X. Song, N. Fu, Y. Fang, X. Zhou, Y. Lin and F. Pan, Metal and F Dual-Doping to Synchronously Improve Electron Transport Rate and Lifetime for TiO₂ Photoanode to Enhance Dye-Sensitized Solar Cells Performances, J. Mater. Chem. A, 2015, 3(10), 5692–5700.
261. J. Y. Park, K. H. Lee, B. S. Kim, C. S. Kim, S. E. Lee, K. Okuyama, H. D. Jang and T. O. Kim, Enhancement of Dye-Sensitized Solar Cells Using Zr/N-Doped TiO₂ Composites as Photoelectrodes, RSC Adv., 2014, 4(20), 9946–9952.
262. S. Munir, S. M. Shah, H. Hussain and R. Ali Khan, Effect of Carrier Concentration on the Optical Band Gap of TiO₂ Nanoparticles., Mater. Des., 2016, 92, 64–72.
263. C. T. Wang, H. S. Lin and W. P. Wang, Hydrothermal Synthesis of Fe[Sbnd] and Nb-Doped Titania Nanobelts and Their Tunable Electronic Structure toward Photovoltaic Application, Mater. Sci. Semicond. Process., 2019, 99, 85–91.
264. A. R. Tanyi, A. I. Rafieh, P. Ekaneyaka, A. L. Tan, D. J. Young, Z. Zheng, V. Chellappan, G. S. Subramanian and R. L. N. Chandrakanthi, Enhanced Efficiency of Dye-Sensitized Solar Cells Based on Mg and La Co-Doped TiO₂ Photoanodes, Electrochim. Acta, 2015, 178, 240–248.
265. A. M. Bakhshayesh and N. Bakhshayesh, Enhanced Performance of Dye-Sensitized Solar Cells Aided by Sr,Cr Co-Doped TiO₂ Xerogel Films Made of Uniform Spheres, J. Colloid Interface Sci., 2015, 460, 18–28.
266. M. Mohammadnezhad, G. S. Selopal, O. Cavuslar, D. Barba, E. G. Durmusoglu, H. Y. Acar, Z. M. Wang, G. P. Lopinski, B. Stansfield and H. Zhao, et al., Gold Nanoparticle Decorated Carbon Nanotube Nanocomposite for Dye-Sensitized Solar Cell Performance and Stability Enhancement, Chem. Eng. J., 2021, 421, 127756.
267. T. R. Naveen Kumar, S. Yuvaraj, P. Kavitha, V. Sudhakar, K. Krishnamoorthy and B. Neppolian, Aromatic Amine Passivated TiO₂ for Dye-Sensitized Solar Cells (DSSC) with ∼9.8% Efficiency, Sol. Energy, 2020, 201, 965–971.
268. J. Men, Q. Gao, S. Sun, X. Zhang, L. Duan and W. Lü, Carbon Nitride Doped TiO₂ Photoelectrodes for Photocatalysts and Quantum Dot Sensitized Solar Cells, Mater. Res. Bull., 2017, 85, 209–215.
269. H. Lu, W. Tian, J. Guo and L. Li, Interface Engineering through Atomic Layer Deposition towards Highly Improved Performance of Dye-Sensitized Solar Cells, Sci. Rep., 2015, 5, 1–12.
270. V. Subramanian, E. Wolf and P. V. Kamat, Semiconductor-Metal Composite Nanostructures. To What Extent Do Metal Nanoparticles Improve the Photocatalytic Activity of TiO₂ Films?, J. Phys. Chem. B, 2001, 105(46), 11439–11446.
271. A. Wood, M. Giersig and P. Mulvaney, Fermi Level Equilibration in Quantum Dot-Metal Nanojunctions, J. Phys. Chem. B, 2001, 105(37), 8810–8815.
272. P. V. Kamat and B. Shanghavi, Interparticle Electron Transfer in Metal/Semiconductor Composites. Picosecond Dynamics of CdS-Capped Gold Nanoclusters, J. Phys. Chem. B, 1997, 101(39), 7675–7679.
273. S. P. Lim, A. Pandikumar, H. N. Lim and N. M. Huang, Essential Role of N and Au on TiO₂ as Photoanode for Efficient Dye-Sensitized Solar Cells, Sol. Energy, 2016, 125, 135–145.
274. F. Zheng and Z. Zhu, Preparation of the Au@TiO₂ Nanofibers by One-Step Electrospinning for the Composite Photoanode of Dye-Sensitized Solar Cells, Mater. Chem. Phys., 2018, 208, 35–40.
275. B. Xu, G. Wang and H. Fu, 23327Enhanced Photoelectric Conversion Efficiency of Dye-Sensitized Solar Cells by the Incorporation of Flower-like Bi₂S₃:Eu³⁺ Sub-Microspheres., Sci. Rep., 2016, 6, 1–9.
276. Z. Zolfaghari-Isavandi and Z. Shariatinia, Enhanced Efficiency of Quantum Dot Sensitized Solar Cells Using Cu₂O/TiO₂ Nanocomposite Photoanodes, J. Alloys Compd., 2018, 737, 99–112.
277. R. S. Ganesh, M. Navaneethan, S. Ponnusamy, C. Muthamizhchelvan, S. Kawasaki, Y. Shimura and Y. Hayakawa, Enhanced Photon Collection of High Surface Area Carbonate-Doped Mesoporous TiO₂ Nanospheres in Dye Sensitized Solar Cells, Mater. Res. Bull., 2018, 101, 353–362.
278. N. Mir and M. Salavati-Niasari, Effect of Tertiary Amines on the Synthesis and Photovoltaic Properties of TiO₂ Nanoparticles in Dye Sensitized Solar Cells, Electrochim. Acta, 2013, 102, 274–281.
279. A. G. Dhodamani, K. V. More, S. M. Patil, A. R. Shelke, S. K. Shinde, D. Y. Kim and S. D. Delekar, Synergistics of Cr(III) Doping in TiO₂/MWCNTs Nanocomposites: Their Enhanced Physicochemical Properties in Relation to Photovoltaic Studies., Sol. Energy, 2020, 201, 398–408.
280. R. D. Shannon, Revised Effective Ionic Radii and Systematic Studies of Interatomic Distances in Halides and Chalcogenides, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32(5), 751–767.
281. K. H. Eom, T. K. Yun, J. Y. Hong, J. Y. Bae, S. Huh and Y. S. Won, Effect of Nitrogen Doping on the Performance of Dye-Sensitized Solar Cells Composed of Mesoporous TiO₂ Photoelectrodes, J. Nanosci. Nanotechnol., 2014, 14(12), 9362–9367.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ma00059a

---