Review of Bio-Inspired Green Synthesis of Titanium Dioxide for Photocatalytic Applications

Abstract

Titanium dioxide (TiO₂) is an important photocatalyst that is widely studied for environmental applications, especially for water treatment by degradation of pollutants. A range of methods have been developed to produce TiO₂ in the form of nanoparticles and thin films. Solution-based synthesis methods offer the opportunity to tune the synthesis through a choice of reagents, additives and reaction media. In particular, the use of biomolecules, such as proteins and amino acids, as bio-inspired additives in TiO₂ synthesis has grown over the last decade. This review provides a discussion of the key factors in the solution-based synthesis of titania, with a focus on bio-inspired additives and their interaction with Ti precursors. In particular, the role of bio-inspired molecular and biomolecular additives in promoting the low-temperature synthesis of titania and controlling the phase and morphology of the synthesised TiO₂ is discussed, with a particular focus on the interaction of TiO₂ with amino acids as model bio-inspired additives. Understanding these interactions will help address the key challenges of obtaining the crystalline TiO₂ phase at low temperatures, with fast kinetics and under mild reaction conditions. We review examples of photocatalytic applications of TiO₂ synthesised using bio-inspired methods and discuss the ways in which bio-inspired additives enhance photocatalytic activity of TiO₂ nanomaterials. Finally, we give a perspective of the current challenges in green synthesis of TiO₂, and possible solutions based on multi-criteria discovery, design and manufacturing framework.

1. Introduction

Photocatalysis plays a key role in the pursuit to replace fossil fuels with renewable resources, such as sunlight, by harvesting energy from light to drive a variety of industrially and environmentally important processes, such as the degradation of pollutants in air and water, hydrogen production from water splitting, CO₂ reduction and organic synthesis [1,2,3,4,5]. In particular, photocatalysis has emerged as an effective technique to destroy pollutants via photoinduced oxidation processes [5,6]. In photocatalytic water treatment, photocatalyst materials absorb light to form photogenerated electrons and holes, which then react with water molecules to produce reactive oxygen species, such as hydroxyl radicals, which are able to oxidise organic pollutants to small non-toxic fragments. Photocatalytic water treatment complements conventional water treatment techniques, which have limitations and are unable to completely eliminate harmful pollutants from water [6,7]. However, advanced water treatment methods themselves have environmental impacts, such as high energy use, infrastructure costs [8,9,10] and the large amount of energy required for the synthesis of photocatalyst materials [9,11,12]. Hence, it is important to minimise the environmental impact of advanced oxidation processes (AOPs), e.g., by using renewable energy in water treatment processes [10]. In particular, the problem of the energy-intensive synthesis of photocatalysts can be solved by developing methods for the green synthesis of photocatalysts [12].

Titanium dioxide is the most widely used photocatalyst [13,14]. Since 1972, when Honda and Fujishima first investigated the photocatalytic properties of TiO₂, titanium dioxide has been a popular choice as a photocatalyst material, due to its photocatalytic efficiency, photostability when being reused, cost effectiveness and environmental sustainability [13,15,16,17]. TiO₂ can exist in three polymorphs: anatase, rutile and brookite. Rutile is thermodynamically more stable, but anatase and brookite display a higher photocatalytic efficiency [18,19,20]. Furthermore, the metastable monoclinic bronze phase of TiO₂ was found to have high photocatalytic activity [21]. Photocatalytic efficiency also depends on the exposed crystal facets, and the less stable (001) facet of anatase is photocatalytically more active than the more stable (101) facet [22].

The photocatalytic degradation of organic pollutants in water using TiO₂ photocatalysts is restricted by the limitations of this material, such as its large bandgap, light scattering due to the small particle size, and reduction in the number of active sites due to the agglomeration of particles [23]. To address these limitations, controlled synthesis strategies are needed to tailor the crystal structure and morphology and to broaden the photoresponse of TiO₂ using surface functionalisation.

Titanium dioxide photocatalysts are utilised in suspension, as thin films or coatings or as powders in catalytic beds. The synthesis of the powder or particle form of titanium dioxide photocatalysts can be generally carried out by solution-based synthesis [24] or solid-state methods [25], whereas thin-film deposition and coatings can be achieved by gas-phase synthesis [26]. Gas-phase or vapour-phase thin-film coating and deposition techniques include thermal evaporation [27], thermal plasma technology [28], sputtering [29,30] and chemical vapour deposition (CVD) [31,32,33]. However, gas-phase synthesis can also produce the powder form of TiO₂, while the powder obtained from liquid-phase synthesis can be blade-coated to make thin films.

Solution-based synthesis methods promise better control over morphology compared to solid-phase and gas-phase synthesis methods. The solution synthesis route includes methods such as sol-gel, solvothermal, hydrothermal and precipitation syntheses [24,34]. These methods produce crystalline titanium dioxide after post-synthetic heat treatment.

Using solution-based techniques, the structural and functional properties of nanocrystalline particles or coatings can be controlled by selecting the synthesis parameters, which are summarised in Figure 1. The key parameters in the solution synthesis are precursors, chemical reagents and conditions. The selection of reagents for synthesis includes the type of precursor, type of additive, type of solvent, the precursor to additive ratio and the pH of the solution. Additional factors are the reaction temperature, calcination temperature, reaction time and aging time.

The conventional techniques of solution-based or liquid-phase synthesis require high-precision equipment [28], a high energy input, toxic chemical additives and capping or chelating agents [34] and involve high processing costs [35]. Since the principles of sustainable and green chemistry aim to minimise the energy, cost and environmental impacts of synthesis processes, low-temperature synthesis methods are desirable for green synthesis [36].

The least energy-intensive choices for solution synthesis would be the preferred route for the sustainable synthesis of TiO₂, maintaining the product quality without increasing the resource intensity. Ideally, a technique is needed to synthesize crystalline titanium dioxide at room temperature. One possible route to achieve this is through the use of biomolecules or bio-inspired additives [37].

Bio-mineralisation processes have been occurring in nature for millennia and involve the assistance of biomolecules in the formation of inorganic materials in living organisms [38,39]. In the past two decades, various biomolecules associated with biological and non-biological minerals have been identified and employed successfully for the in vitro synthesis and growth of inorganic nanostructures of metals, oxides and semiconductors [37,40,41]. Biomolecules, such as amino acids, peptides and proteins, have a critical role in mineralisation due to the specificity of their interactions with minerals, and they can also act as templates to produce nanostructures [37,40,41,42,43]. There are two ways to tap into the potential of biomineralization in material synthesis: (i) the direct use of biomolecules that exist in nature; and (ii) developing molecules in the lab that mimic or are inspired by natural biomolecules [44].

Besides the challenges associated with synthesis processes, the scale and sustainability associated with the synthesised materials also need to be considered. By addressing multiple goals at the same time, through the approach of ‘multicriteria thinking’, the full potential of bio-inspired additives for nanomaterials synthesis can be tapped [45].

The focus of this review is exploring the role of bio-inspired molecular additives in the synthesis of TiO₂. Specifically, this review provides a comprehensive and critical analysis of the effects of molecular additives and links them to chemical interactions underlying the reactions of TiO₂ formation. While there have been several recent reviews on TiO₂ synthesis using nature-based additives, such as plants [46,47,48,49], there is a lack of reviews focussed on the role of molecular and biomolecular additives, such as short peptides, amino acids and other “simple” molecules. The reviews of plant-based additives have largely focussed on synthesis procedures and precursors and on applications of TiO₂ products, rather than on mechanisms and interactions responsible for TiO₂ formation. The focus on “simple” additives in this review will help identify key future directions, e.g., investigating TiO₂–additive chemical interactions, which ultimately can lead to TiO₂ synthesis in a controllable manner instead of trial and error.

This review discusses the use of bio-inspired additives for synthesising titanium dioxide by considering multiple reaction design parameters. The key factors affecting the synthesis of titanium dioxide (Figure 1) are discussed, with a particular focus on the bio-inspired synthesis of TiO₂. The aim is to provide an overview of the state-of-the-art solution synthesis methods for titanium dioxide, including various bio-inspired methods, and to discuss their merits and limitations to serve as a platform for further research in this area. Finally, as a key application, the photocatalytic performance of TiO₂ synthesized by green synthesis methods is reviewed.

2. Choice of Synthesis Parameters

2.1. Selection of Precursor

Titanium (IV) compounds, such as titanium tetrachloride (TiCl₄), titanium alkoxides and titanium bis (ammonium lactate) dihydroxide (TiBALDH), are the most commonly used precursors for the synthesis of titanium dioxide via solution synthesis. Titanium tetrachloride (TiCl₄) reacts with water exothermically to form titanium dioxide and HCl. Table S1 (Supplementary Materials) presents several examples of TiO₂ synthesis using a TiCl₄ precursor. However, one of the downsides of using TiCl₄ is the formation of orthotitanic acid (H₄TiO₄) as a byproduct. As the formation of orthotitanic acid cannot be avoided, it may hamper the homogeneity of the titanium dioxide synthesised using TiCl₄ [50]. The TiCl₄ precursor is also volatile and can react with atmospheric moisture and thus requires a fume cupboard during its use [51].

An alternative precursor, TiF₄, is more stable than TiCl₄ and can be used for the selective formation of {001} facets of anatase, which are highly reactive and therefore beneficial for applications such as photocatalysis [52]. For example, a mixture of TiCl₄ and TiF₄ precursors were used to achieve the controlled synthesis of {101} and {001} facets [53]. Although TiF₄ is good for facet-specific synthesis [22], its dissociation to form extremely harmful HF at higher temperatures makes TiF₄ a less desirable choice.

Titanium (IV) alkoxides are another type of commonly used precursors for sol-gel synthesis of titanium dioxide, which go through hydrolysis and a condensation reaction to form TiO₂. Table S2 lists examples of studies that have used Ti isopropoxide and Ti butoxide precursors. Due to their poor solubility, alkoxides hydrolyse slowly. Their rate of hydrolysis can be controlled by the pH and the type of catalyst used (e.g., hard vs. soft acid) [54]. However, due to the electronegativity of alkoxy groups, titanium alkoxides are very sensitive to nucleophilic attack by water [55] and can be hydrolysed even by reacting with moisture in the atmosphere. Precursors that are less sensitive to environmental conditions, such as moisture in the air, are required in order to avoid early hydrolysis, which could hinder the degree of control over the synthesis [34].

In order to avoid the use of non-aqueous solvents, an ideal precursor needs to be water soluble. Ti (IV) sulfates, glycolates and lactates are examples of water-soluble precursors. For example, titanium sulfate Ti(SO₄)₂ precursors were used for TiO₂ synthesis by solvothermal or hydrothermal methods [56,57,58,59]. Ti oxysulfate was used as a precursor and exhibited a co-operative effect in the amino acid lysine-assisted synthesis of TiO₂, to yield single crystals of anatase TiO₂ with exposed {101} and {100} facets [60].

Titanium (IV) bis (ammonium lactato) dihydroxide (TiBALDH), also known as TALH or ALT, is a water-soluble titanium precursor which is stable at neutral pH conditions at room temperature [37,61,62,63]. TiBALDH does not hydrolyse under ambient conditions, but it was found that an aqueous solution of urea can slowly hydrolyse it at 90 °C [64]. The hydrolysis of TiBALDH at a high pH was reported to be better controlled compared to the hydrolysis of Ti alkoxides, with aqueous NaOH decomposing TiBALDH to TiO₂, NH₃ and sodium lactate [61]. The conversion of TiBALDH to TiO₂ was found to occur due to a coordination equilibrium, as shown in Equation (1), where TiBALDH in the form of ammonium oxo-lactato-titanate (NH4)₈Ti₄O₄(Lactate)₈ in solution is in equilibrium with ammonium tris-lactato-titanate, (NH₄)₂Ti(Lactate)₃ (with its structure shown in Figure 2) and TiO₂, so that uniform crystalline TiO₂ anatase nanoparticles are formed even at room temperature and are stabilized by surface-capping lactate ligands [62].

3 [Ti₄O₄(Lactate)₈]⁸⁻ + 8 ΝH₄⁺ ⇌ 8 [Ti(Lactate)₃]²⁻ + 4 TiO₂ + 8 ΝH₃ + 4 H₂O

(1)

In summary, the TiBALDH precursor stands out as an ideal precursor for bio-inspired synthesis of titania based on the criteria listed in Figure 1, and it has become a popular precursor choice for bio-inspired synthesis, used in multiple studies as shown in Table 1.

Besides these main types of Ti precursors, other precursors have been used: for example, TiCl₃ was used to produce the rutile [65] and anatase phases of TiO₂ [65,66], as well as the “bronze” phase of TiO₂ [21]. The rutile, brookite and bronze phases of TiO₂ were also obtained using Ti metal powder dissolved in hydrogen peroxide [21,67,68].

Table 1. Titanium dioxide synthesis using TiBALDH precursor.

Solvent	Additives	Reaction Temperature	Reaction Time	Calcination Temperature	Product	Ref.
Deionized water	Urea	RT, then 100 °C	1 h at RT, 20 h at 100 °C	500 °C	Amorphous thin-film coating	[64]
Aqueous solution	Urea	160 °C	Overnight	300–550 °C	Pure anatase, pure brookite or biphasic anatase/brookite mixtures	[69]
Water	Urea	95 °C	24 h	-	Anatase TiO2 sol	[70]
Water	L-arginine	RT	30 min	480 °C	Anatase	[71]
Tris–HCl buffer	Arginine	RT	0.5–10.0 min	-	Anatase	[72]
Water, phosphate buffer	Spermidine or spermine	RT	Overnight	200, 400, 600, 800 °C	Anatase after annealing at 800 °C	[73]
Aqueous solution	Poly(allylamine hydrochloride), poly(diallyldimethyl-ammonium chloride)	RT	5–60 days	-	Aggregated nanoparticles of anatase (anatase was observed after 30 days)	[74]
Phosphate-citrate buffer solution	c-terminal tetra peptide Gly-Gly-Gly-Trp	RT	10 min	-	Nanoparticles <50 nm in size contained very fine (<10 nm) anatase and monoclinic TiO2 domains	[75]
Tris buffer	Serine-lysine (S-K) peptides KSSKK, SKSK3SKS	RT	24 h	-	Amorphous or crystalline particles, 150–1200 nm diameter	[76]
Water	KIIIIKYWYAF peptide	70 °C	48 h	580 °C	Anatase after 580 °C	[77]
Phosphate buffer or water	R5 peptide or poly-L-lysine-hydrobromide	RT	5 min	600–900 °C	Anatase at 600 °C. Anatase to rutile transition was at 700 °C	[78]
Tris or phosphate buffer	R5 peptide and its truncated analogues	RT	24 h	600 °C	Amorphous TiO2 at RT; anatase formed after annealing at 600 °C	[79]
Tris buffer, phosphate buffer or distilled water	Titanium dioxide binding peptides Ti-1, Ti-2 and R5	RT	2–72 h	-	<10 nm TiO2 sols, mostly amorphous with some anatase and monoclinic phases	[41]
Phosphate buffer	R5 peptide	RT		-	TiO2 nanosheets several μm in size, amorphous with <10 nm anatase domains	[80]
Citrate buffer	Car9 peptide fused to superfolder green fluorescent protein (sfGFP)	RT	120 min	-	Mixture of amorphous, anatase and monoclinic (bronze) TiO2 phases	[81]
Deionized water	Silicatein protein	20 °C	24 h (at 20 °C), 1 h (calcination)	27–927 °C in steps of 100 °C	Mixture of amorphous and nanocrystalline anatase; transition to rutile was at 850 °C	[37]
Tris-HCl buffer	Proteins protamine, lysozyme, gelatin, haemoglobin, yeast alcohol dehydrogenase and bovine serum albumin	RT	5 min	600–700 °C	Amorphous at RT; transition to anatase at 600–700 °C and to rutile at 800 °C	[82]
Phosphate-buffered saline (PBS)	Bioengineered silicatein α and β and scaffold protein silintaphin-1	RT	12 h	-	Amorphous and anatase phases	[83]
Water, phosphate/ citrate buffer	Silaffin protein	RT	20 min	-	Rutile	[40]

Table 1. Titanium dioxide synthesis using TiBALDH precursor.

Solvent	Additives	Reaction Temperature	Reaction Time	Calcination Temperature	Product	Ref.
Deionized water	Urea	RT, then 100 °C	1 h at RT, 20 h at 100 °C	500 °C	Amorphous thin-film coating	[64]
Aqueous solution	Urea	160 °C	Overnight	300–550 °C	Pure anatase, pure brookite or biphasic anatase/brookite mixtures	[69]
Water	Urea	95 °C	24 h	-	Anatase TiO2 sol	[70]
Water	L-arginine	RT	30 min	480 °C	Anatase	[71]
Tris–HCl buffer	Arginine	RT	0.5–10.0 min	-	Anatase	[72]
Water, phosphate buffer	Spermidine or spermine	RT	Overnight	200, 400, 600, 800 °C	Anatase after annealing at 800 °C	[73]
Aqueous solution	Poly(allylamine hydrochloride), poly(diallyldimethyl-ammonium chloride)	RT	5–60 days	-	Aggregated nanoparticles of anatase (anatase was observed after 30 days)	[74]
Phosphate-citrate buffer solution	c-terminal tetra peptide Gly-Gly-Gly-Trp	RT	10 min	-	Nanoparticles <50 nm in size contained very fine (<10 nm) anatase and monoclinic TiO2 domains	[75]
Tris buffer	Serine-lysine (S-K) peptides KSSKK, SKSK3SKS	RT	24 h	-	Amorphous or crystalline particles, 150–1200 nm diameter	[76]
Water	KIIIIKYWYAF peptide	70 °C	48 h	580 °C	Anatase after 580 °C	[77]
Phosphate buffer or water	R5 peptide or poly-L-lysine-hydrobromide	RT	5 min	600–900 °C	Anatase at 600 °C. Anatase to rutile transition was at 700 °C	[78]
Tris or phosphate buffer	R5 peptide and its truncated analogues	RT	24 h	600 °C	Amorphous TiO2 at RT; anatase formed after annealing at 600 °C	[79]
Tris buffer, phosphate buffer or distilled water	Titanium dioxide binding peptides Ti-1, Ti-2 and R5	RT	2–72 h	-	<10 nm TiO2 sols, mostly amorphous with some anatase and monoclinic phases	[41]
Phosphate buffer	R5 peptide	RT		-	TiO2 nanosheets several μm in size, amorphous with <10 nm anatase domains	[80]
Citrate buffer	Car9 peptide fused to superfolder green fluorescent protein (sfGFP)	RT	120 min	-	Mixture of amorphous, anatase and monoclinic (bronze) TiO2 phases	[81]
Deionized water	Silicatein protein	20 °C	24 h (at 20 °C), 1 h (calcination)	27–927 °C in steps of 100 °C	Mixture of amorphous and nanocrystalline anatase; transition to rutile was at 850 °C	[37]
Tris-HCl buffer	Proteins protamine, lysozyme, gelatin, haemoglobin, yeast alcohol dehydrogenase and bovine serum albumin	RT	5 min	600–700 °C	Amorphous at RT; transition to anatase at 600–700 °C and to rutile at 800 °C	[82]
Phosphate-buffered saline (PBS)	Bioengineered silicatein α and β and scaffold protein silintaphin-1	RT	12 h	-	Amorphous and anatase phases	[83]
Water, phosphate/ citrate buffer	Silaffin protein	RT	20 min	-	Rutile	[40]

2.2. Role of Solvent, pH and Buffer

The choice of solvent affects the equilibrium between the TiBALDH precursor and TiO₂, as described in Equation (1). This equilibrium can be shifted towards TiO₂ by using a less polar solvent, such as methanol or ethanol, diluting the solution, introducing salts or raising the temperature. In contrast, the use of polar and strongly solvating media, such as dimethyl sulfoxide, shifts the equilibrium towards reactants [62]. The ability to reverse the direction of the reaction is essential for producing nanocrystalline and monodisperse TiO₂ at room temperature. This is because once TiO₂ crystallites have nucleated, if the reaction is able to proceed backwards towards the reactants, then it restricts further agglomeration, leading to a smaller particle size.

It is well established that anatase formation is preferred to rutile at higher (basic) pH levels in aqueous media, whereas rutile is more prone to forming in acidic media [84]. This is explained by the mechanism of condensation of precursor [Ti(OH)_x(H₂O)_y]ⁿ⁺ complexes: at high pH levels, the Ti complexes contain more OH groups, resulting in the edge sharing of octahedra, leading to the anatase phase. In contrast, at lower pH levels with few OH ligands, rutile formation takes place due to the corner bonding of the octahedra [84,85]. The conversion between TiO₂ phases can be achieved by controlling the pH of the solution: for example, amorphous anatase was reported as the initial product of the hydrolysis of the titanium isopropoxide precursor, which was converted to anatase and then to rutile upon the addition of acid [86]. It was hypothesized that the anatase-to-rutile conversion was accelerated in the acidic medium thanks to the formation of an intermediate metastable ionic phase [Ti(OH)₂]²⁺, which lowered the energy barrier for the anatase-to-rutile transformation [86].

The pH of a solution can be controlled by adding a buffer solution. For example, phosphate, citrate or tris(hydroxymethyl)aminomethane (Tris) HCl buffers are commonly used. The influence of pH on TiO₂ synthesis was investigated using buffers with the TiBALDH precursor [63] in bio-inspired syntheses using different biomolecule additives, such as peptides [76,78], polyamines [73] and amino acids [87]. Complex trends were observed, depending both on the pH and on the nature of the biomolecule additives. For example, a study of TiO₂ formation using a peptide additive with phosphate buffers ranging between pH levels of 5.5 to 7.5 found that the TiO₂ production rate reached the maximum in the pH range of 6.0–7.5, and a pH of 7.5 yielded well-defined particles [78]. A study of TiO₂ formation using polyamine additives showed different pH dependence trends depending on the length of the polyamine chain: long-chain polyamine spermine enabled TiO₂ formation in a wide range of pH levels between 2.9 and 12.6, while a medium-chain polyamine spermidine enabled TiO₂ formation only at acidic and neutral pH levels, up to a pH of 9.5, as depicted in the scanning electron microscopy (SEM) images in Figure 3 [73].

Furthermore, the effect of pH on the crystalline phase of the final TiO₂ product was investigated in depth in amino-acid-assisted synthesis [87]. Syntheses using a series of amino acids produced anatase as the dominant phase at pH levels of 1 to 6, while only amorphous titania was formed at a pH of 8. However, in addition to the anatase phase, a secondary TiO₂ crystalline phase was obtained in some cases, with the amount and nature of this secondary phase dependent both on the pH and on the nature of the amino acid additives. For example, glycine, lysine and arginine produced a mixture of anatase and rutile at a pH of 1 and a mixture of anatase and brookite with pH levels from 2 to 6 [87]. This is consistent with the previously reported preference for the rutile phase at an acidic pH and the anatase phase at a more basic pH [84,85,86]. In contrast, aspartic acid, glutamic acid and serine produced only the anatase phase in the range of pH levels from 1 to 6, while histidine and proline produced a mixture of anatase and brookite with pH levels from 1 to 4 and pure anatase at a pH of 6. This study revealed a complex interplay of the effects of pH levels and amino acid additives, which are analysed in more detail in Section 3.1, in which the effects of additives are discussed.

Beyond investigating the effect of the pH, the effect of the chemical nature of the buffer on the mechanism of peptide-assisted TiO₂ mineralisation was investigated by synthesizing TiO₂ sol particles at a pH of 7.4 maintained in NaOH solution, Tris buffer and phosphate buffer [41]. Tris buffer and non-buffered aqueous solutions produced monodisperse crystalline TiO₂ particles, while the phosphate buffer solution produced polydisperse, poorly crystalline particles with the presence of some adsorbed phosphate. This co-precipitation of TiO₂ with phosphate indicated that the nature of the buffer can affect the TiO₂ condensation reaction and in particular showed that titanium phosphates are formed instead of pure TiO₂ in Ti–polymer complexation, thus disrupting TiO₂ crystallisation [41].

3. Bio-Inspired Additives

Research on bio-inspired additives for titanium dioxide synthesis has emerged over the last two decades. A wide range of biomolecules, such as amines, amino acids, peptides and proteins, as well as organic matter and organic waste, have been employed as additives for the synthesis of TiO₂ nanomaterials. Biomaterials originating directly from nature have also been explored for the synthesis of TiO₂, including plant derivatives such as pomelo peel [88], jatropha leaves [89], pollen grains [90], green tea extract, aloe vera gel [91], leaf extracts of morinda citrofolia [92], parthenium hysterophorus [93] and eucalyptus globulus [94] as nature-based additives. Besides the plant-based extracts, microbes such as fungi and bacteria have been used for TiO₂ synthesis [46,47,48,95]. Recent reviews on the bio-inspired synthesis of TiO₂ list various plant-based and microbe-based TiO₂ synthesis studies [46,47,48,49]. However, the direct use of biomolecules from nature has limitations in terms of up-scaling due to their availability. Thus, a trend of using synthetic analogues of natural biomolecules is emerging. These bio-inspired additives have been employed for the synthesis of nanomaterials [39,44].

The use of biomolecules as additives for TiO₂ synthesis was pioneered by Morse et al., who synthesised TiO₂ using silicatein, a protein derived from marine silica sponges, as a template [37]. Further TiO₂ nanoparticle synthesis studies have used a variety of proteins [40,81,82,83,96] and peptides [41,43,75,76,77,78], as well as enzymes such as protease and lipase [97]. Following on from peptides and proteins, biomolecular additives with simpler chemical structures were used, such as amines and polyamines [74,98,99] and amino acids [60,87,100,101,102,103].

For example, Cole and Valentine investigated several naturally occurring polyamines as additives for TiO₂ synthesis and found that the chemical nature of the polyamines was a significant factor: the polyamines spermine and spermidine reacted with Ti precursors to form TiO₂, while the shorter diamines putrescine and cadaverine did not result in TiO₂ mineralisation [73]. They concluded that the number of amine functionalities in the polyamines played a significant role in the titanium mineralisation: while diamines could bind to a single Ti atom by bidentate chelation, polyamines such as spermine and spermidine with three or more amine groups could bind to multiple Ti atoms and induce condensation to form TiO₂.

Amino acids, as the simplest building blocks of proteins, have been widely used for the bio-inspired synthesis of TiO₂ (Table 2) [60,100,101,102,103]. The use of amino acids enables the testing of the effect of variables, such as the charge of the biomolecule and its degree of protonation at a particular pH. Amino-acid-assisted synthesis was found to produce various phases of TiO₂ nanoparticles both after annealing [71,100,102,103,104] and in room-temperature synthesis [60,72,87]. For example, the synthesis of TiO₂ from TiCl₄ precursors using a series of amino acids as additives at a range of pH levels resulted in mixtures of anatase, rutile and brookite [87]. Another study using TiBALDH as a precursor and arginine as an additive produced anatase particles, with 3,4-dihydroxy-L-phenylalanine (dopa) then added to terminate the reaction and to control the size of the synthesized nanoparticles between 35 and 350 nm [72]. Solvothermal synthesis at 160 °C using titanium oxysulfate as a precursor and lysine as an additive in an acidic medium provided anatase single crystals with facet control [60]. Beyond TiO₂, amino acids have also been employed in the synthesis of various metal oxides, such as ZnO [105], WO₃ [106], SnO₂ [107], perovskite nanoparticles [108] and metal nanoparticles, such as Ag [109], Pd nanocrystals [110] and bimetallic PtCo nanospheres [111]. Besides synthesis, mechanistic studies of the bio-assisted synthesis of TiO₂ have been carried out. These mechanistic aspects are discussed below, with a particular focus on the role of bio-inspired additives as catalysts and templates.

Table 2. Amino-acid-assisted titanium dioxide synthesis.

Ti Precursor	Types of Amino Acids	Process	Calcination	TiO2 Phase	TiO2 Morphology	Ref.
Titanium n-butoxide (Ti(OBu)4	Glycine	Hydrothermal synthesis at 120 °C for 48 h	500 °C; 3.5 h	Anatase	Flower-like hierarchical spheres with a 2 μm diameter assembled on 20 nm thick nanosheet	[102]
Titanium isopropoxide	Glycine, DL-alanine, β-alanine, DL-valine, proline, serine, DL-aspartic acid, L-glutamic acid	Gel formation after 12 h at room temperature, drying at 100 °C	500 °C; 3 h	Anatase	10–15 nm cubic particles	[101]
TiBALDH	Arginine	g-C3N4 + distilled water; 30 min at room temperature	480 °C; 2 h	Anatase	Uniformly distributed TiO2 nanoparticles, d < 10 nm on g-C3N4 nanosheets	[71]
Titanium n-butoxide (Ti(OBu)4	Glycine	200 °C for 20 h	450 °C; 5 h	Anatase	Hollow microspheres, with a crystallite size of 4.8 nm	[103]
TiCl4	Glycine, alanine, serine, threonine, β-alanine	Seeded growth of TiO2 nanorods in HCl on pre-annealed FTO glass. Seeds grown at 95 °C	450 °C; 1 h	Rutile	300–900 nm nanorods on FTO glass	[100]
Titanium isopropoxide (TTIP)	L-alanine	TTIP, L-alanine and dodecylamine in ethyl alcohol reacted at 60 °C for 24 h	400 °C; 4 h	Anatase	200 nm nanoparticles	[112]
Titanium isobutoxide	L-lysine	60 °C 20 h; 100 °C 24 h	350 °C	Mixed phase anatase + brookite	Mesoporous nanocrystals	[113]
Titanium (IV) oxysulfate	Lysine	Solvothermal synthesis in precursor in diluted H2SO4 at 160 °C 24 h	No further calcination	Anatase with exposed {101} and {111} facets	Single-crystal-like hierarchical spheres	[60]
TiCl4	Glycine, glutamic acid, aspartic acid, serine, histidine, proline, lysine, arginine	Thermo-hydrolysis at 60 °C, from 1 day to 1 week, at a pH of 1 to 8	No further calcination, but long reaction time	Anatase, brookite, anatase + brookite, anatase + rutile, amorphous	Nanoparticles with controlled shapes and sizes	[87]
TiBALDH	Arginine, serine, lysine, histidine, glycine	10 min at room temperature	No further calcination	Surface functionalised anatase only with arginine	35–350 nm nanoparticles	[72]

Table 2. Amino-acid-assisted titanium dioxide synthesis.

Ti Precursor	Types of Amino Acids	Process	Calcination	TiO2 Phase	TiO2 Morphology	Ref.
Titanium n-butoxide (Ti(OBu)4	Glycine	Hydrothermal synthesis at 120 °C for 48 h	500 °C; 3.5 h	Anatase	Flower-like hierarchical spheres with a 2 μm diameter assembled on 20 nm thick nanosheet	[102]
Titanium isopropoxide	Glycine, DL-alanine, β-alanine, DL-valine, proline, serine, DL-aspartic acid, L-glutamic acid	Gel formation after 12 h at room temperature, drying at 100 °C	500 °C; 3 h	Anatase	10–15 nm cubic particles	[101]
TiBALDH	Arginine	g-C3N4 + distilled water; 30 min at room temperature	480 °C; 2 h	Anatase	Uniformly distributed TiO2 nanoparticles, d < 10 nm on g-C3N4 nanosheets	[71]
Titanium n-butoxide (Ti(OBu)4	Glycine	200 °C for 20 h	450 °C; 5 h	Anatase	Hollow microspheres, with a crystallite size of 4.8 nm	[103]
TiCl4	Glycine, alanine, serine, threonine, β-alanine	Seeded growth of TiO2 nanorods in HCl on pre-annealed FTO glass. Seeds grown at 95 °C	450 °C; 1 h	Rutile	300–900 nm nanorods on FTO glass	[100]
Titanium isopropoxide (TTIP)	L-alanine	TTIP, L-alanine and dodecylamine in ethyl alcohol reacted at 60 °C for 24 h	400 °C; 4 h	Anatase	200 nm nanoparticles	[112]
Titanium isobutoxide	L-lysine	60 °C 20 h; 100 °C 24 h	350 °C	Mixed phase anatase + brookite	Mesoporous nanocrystals	[113]
Titanium (IV) oxysulfate	Lysine	Solvothermal synthesis in precursor in diluted H2SO4 at 160 °C 24 h	No further calcination	Anatase with exposed {101} and {111} facets	Single-crystal-like hierarchical spheres	[60]
TiCl4	Glycine, glutamic acid, aspartic acid, serine, histidine, proline, lysine, arginine	Thermo-hydrolysis at 60 °C, from 1 day to 1 week, at a pH of 1 to 8	No further calcination, but long reaction time	Anatase, brookite, anatase + brookite, anatase + rutile, amorphous	Nanoparticles with controlled shapes and sizes	[87]
TiBALDH	Arginine, serine, lysine, histidine, glycine	10 min at room temperature	No further calcination	Surface functionalised anatase only with arginine	35–350 nm nanoparticles	[72]

3.1. Influence of Bio-Inspired Additives on Reaction Kinetics and Phase Control

Several studies have demonstrated the role of amino acid, peptide and protein additives as catalysts, which influence reaction kinetics by lowering the time and/or the temperature required for the TiO₂ synthesis reaction [72,74,75,76,81,87,100,114]. For example, in the glycine-assisted synthesis of TiO₂ nanorods, the rate of hydrolysis of the titanium tetrachloride precursor was found to depend on the amino acid concentration: an increase in the concentration of glycine increased the rate of the reaction, confirming the catalyst-like behaviour of glycine in synthesizing TiO₂ nanoparticles [100]. In another example of catalytic action of amino acid additives, arginine was found to act as a catalyst for TiO₂ nanoparticle nucleation and growth, with the size of synthesized TiO₂ nanoparticles increasing both with time and with the increased concentration of arginine [72]. These examples from the literature suggest that amino acids act as catalysts by lowering the temperature of crystalline-phase formation, while the reaction time in these studies varies between several days and several minutes depending on the additive and processing conditions (Table 2) [72,87].

Besides the initial precursor hydrolysis, the use of bio-inspired additives affects the next step of synthesis—calcination. As seen in Table 1 and Table 2 and in Tables S1 and S2 in the Supplementary Materials, the conventional synthesis procedure generally involves two steps, in which the first step is the precursor’s hydrolysis, and the second step is calcination at a high temperature, which is required to obtain crystalline TiO₂.

It can be seen from the compilation of studies from the literature in Table 1 that amorphous TiO₂ is commonly formed from various Ti precursors at room temperature, while post-synthesis heat treatment at 700 °C or 800 °C is required to produce the crystalline anatase or rutile phase, respectively. However, as seen in Table 1 and Table 2, biomolecule-assisted syntheses can produce crystalline TiO₂ at low temperatures below 100 °C [40,72,74,75,83,87,100]. In particular, some amino acids, amines, peptides and protein additives can promote the synthesis of crystalline TiO₂ at room temperature without calcination [41,72,74,75,76,81,87].

The calculation temperatures required for anatase and rutile phase formation are highly additive-dependent: for example, syntheses using silicatein and protamine protein additives required higher temperatures for the anatase-to-rutile transformation compared to the alkali-catalysed hydrolysis of TiO₂ precursors without additives [37,82]; this was attributed to intermediate formation of composites between the Ti precursors and proteins, which may have prevented the crystallisation of amorphous titania, and to the strain imposed by the proteins on the crystal surfaces [37,82]. While silicatein was the protein of choice in the early studies because of its known ability to biomineralize silica [37], other proteins and peptides have been found to be more effective in the biomineralization of TiO₂ under mild conditions. For example, a composite of bioengineered silicatein protein with silintaphin-1 as a scaffold protein was found to produce a mixture of the amorphous and anatase phase of TiO₂ at room temperature [83]. The protein silaffin and its analogues have been widely explored recently because of their ability to drive the formation of crystalline TiO₂. In particular, the use of silaffin as a suitable protein additive was reported to produce the rutile phase at room temperature (Figure 4, left panel) [40]. However, the R5 peptide derived from silaffin produced mostly amorphous TiO₂ particles containing domains of the anatase and monoclinic phase at room temperature (Figure 4, right panel) [41,78,80], with annealing required to form anatase [78,79]. A composite of a Car9 silica-binding peptide fused with superfolder green fluorescent protein (sfGFP) was found to produce a mixture of amorphous and crystalline TiO₂ [81]. Interestingly, a similar composite of R5 fused with sfGFP in the same study was not effective at precipitating TiO₂ at room temperature, which led the authors to suggest that both the N- and C-termini of R5 must be free to induce TiO₂ mineralization [81], thus highlighting the key role of the chemical nature of the peptides and their interaction with Ti precursors.

Similarly, several studies using small bio-additives, such as amino acid additives, have produced crystalline TiO₂ nanoparticles at room temperature [72,87]. For example, Yan et al. observed the conversion of amorphous titania to anatase after 30 days of aging at room temperature (without high-temperature annealing) and the appearance of rutile after 60 days of aging (Figure 5) [74]. Thus, while the goal of a fully controllable synthesis of crystalline TiO₂ phases at room temperature and within short timeframes has not yet been attained, the choice of the amino acid or protein additives offers a promising avenue to achieve this aim and is expected to be one of the key directions of future research.

Studies have also shown that the choice of amino acid additives can direct the formation of the crystalline phase of TiO₂. For example, a study by Shi et al. using TiBALDH as a precursor obtained the anatase phase at room temperature with arginine as an additive, but not with other amino acids such as serine, glycine, histidine or lysine [72].

Durupthy at al. found that the interactions of a TiCl₄ precursor with different amino acids resulted in different TiO₂ phases (Figure 6) [87]. In particular, the use of amino acids such as serine, glutamic acid or aspartic acid led to pure anatase phase formation. In contrast, histidine and proline additives produced mixtures of anatase with brookite, while glycine, lysine and arginine produced mixtures of anatase with rutile or with brookite, depending on the reaction pH. The initial formation of amorphous titania, prior to anatase formation, was attributed to preferential adsorption of the amino acids on the embryos of TiO₂, thus stabilising the amorphous particles and causing a delay in crystallisation. The formed amorphous titania then slowly converted to anatase after 1 week of aging [87]. The absence of the rutile phase in those cases was tentatively explained by the formation of [amino acid–Ti⁴⁺] complexes which favoured anatase formation. The nucleation of anatase in preference to rutile or brookite in reactions with aspartic acid, glutamic acid and serine was attributed to the specific attachment of these amino acids to the growing anatase facets, thus lowering the energies of these facets. The differences in adsorption of the amino acids on TiO₂ could be explained by the difference in the amino acids’ pKa values [87]. This study opened the possibility of controlling the phase of crystalline TiO₂ particles through the choice of a suitable amino acid additive.

In another synthesis carried out under hydrothermal conditions, the amino acids aspartic acid and tyrosine yielded a pure anatase phase, whereas pure rutile was synthesised with alanine, glycine, proline, glutamic acid, serine, threonine, cystine and methionine. However, lysine and arginine resulted in 70–80% of the brookite phase, while a higher concentration of lysine gave rise to pure brookite nanoparticles [68]. Thus, the choice of amino acid additives can direct TiO₂ synthesis towards desired phases, e.g., the anatase phase or the brookite phase, which are less stable but often display higher photocatalytic activity [18,19].

A protein-assisted TiO₂ synthesis study by Hellner et al. [81] similarly found that the nature and the ratio of the crystalline phases was dependent on the chemical nature of the peptide additives. This study considered several mutations to the Car9 peptide fused to the sfGFP protein. While the TiO₂ products were only partly crystalline, the amount of the crystalline phases increased when Car9 or its mutants were added to sfGFP; the mutations also tuned the ratio of the anatase and monoclinic bronze phase from 20%:80% to 65%:35% (Figure 7). The authors hypothesized that the catalytic action of the peptides was due to the peptides acting as ligands and displacing lactate ligands in the TiBALDH precursor, thus promoting polycondensation between adjacent Ti complexes. The tendency to form monoclinic crystals was attributed to favourable electrostatic interactions between positively charged side chains in peptides and negatively charged titania precursors, promoting the formation of compact monoclinic crystallites [81].

The reasons for the preferential formation of particular TiO₂ phases were discussed in these studies using qualitative arguments. To gain a deeper understanding of how specific amino acids or amino acid sequences may drive the growth of different titania phases, future studies need to investigate the interactions between amino acids and growing TiO₂ particles; in particular, computational modelling studies are needed to obtain quantitative insights into the strength of amino acid–titania interactions.

3.2. Role of Bio-Inspired Additives as Templates and Capping Agents: Effect on Morphology

Amino acids, peptides and proteins can act as templates for oxide particle formation [101]. Beyond crystallinity and the crystal phase, properties such as the particle size, shape and surface area of synthesised TiO₂ can also be controlled by the selection of an appropriate additive. This is attributed to additives acting as capping agents, which form bonds to the growing crystal surfaces to control the growth in a particular direction of a crystal facet on the surface of a particle. This allows for the minimisation of interfacial tension and stabilisation of a particular facet [68,115].

For example, Kanie and Sugimoto first reported in 2004 that amino acids could be used as shape controllers in TiO₂ synthesis using Ti(OH)₄ gel at 140 °C (Figure 8) [115]. The shape of the synthesised titanium dioxide particles relates to the abundance of particular crystal facets [22]. The dominance of a particular crystal facet in a synthesis was controlled by the concentration of the amino acids [115]. Moreover, the acidic or basic nature of the side groups of amino acids also affected the shape of the synthesised nanoparticles, as shown in Figure 8. For example, the basic amino acid lysine and neutral amino acids, such as glycine and serine, preferred to generate TiO₂ nanoparticles with rod-like shapes, whereas acidic amino acids such as aspartic or glutamic acid produced spindle-like nanostructures of TiO₂ [115].

Similarly, more complex bio-inspired additives involving peptides fused to the superfolder green fluorescent protein produced a variety of morphologies of TiO₂ nanoparticles, such as needles, threads, plates and peapods [96]. Furthermore, the morphology of TiO₂ products could be changed from large bulk-like monoliths to networks of small interconnected particles by controlling the diffusion in the reaction medium by increasing the amount of agarose hydrogel in the solution [96].

The templating effect of additives was also demonstrated in a study which used L-alanine and dodecylamine (DDA) additives with a titanium tetraisopropoxide precursor [112]. While the addition of alanine alone produced small (apx. 10 nm diameter) TiO₂ particles, the simultaneous addition of alanine and DDA resulted in particles with diameters between 300 and 700 nm. DDA was believed to act as a neutral surfactant controlling the size and shape of spherical TiO₂ particles (Figure 9), while alanine reacted with the Ti precursor and acted as a dopant to introduce nitrogen in the synthesized titania [112].

The position of the amino group in the chain was also found to affect the particle size and photocatalytic performance of the synthesised nanoparticles; e.g., a comparison of α-alanine and β-alanine showed the formation of a dense forest of fine TiO₂ nanorods with α-alanine, whereas coarser nanorods with a lower density were observed with β-alanine [100]. Different amino acids were also reported to result in different mean particle sizes of as-synthesised TiO₂ between 2.5 and 8 nm, with the particle sizes dependent both on the nature of the amino acids and on the pH [87].

Moreover, a larger diameter of synthesised TiO₂ nanoparticles was obtained when using the glycylglycine peptide additive consisting of two glycine units compared to the glycine additive [100]. Furthermore, Bakre et al. reported that amino-acid-assisted synthesised TiO₂ anatase had a smaller particle size and crystallite size compared to commercial P25. Specifically, proline, valine and aspartic acid yielded smaller particle sizes of TiO₂ nanoparticles than other amino acids, such as glycine, alanine, glutamic acid and serine; these smaller particle sizes then resulted in an improved photocatalytic performance [101].

A pronounced effect on TiO₂ particle size was achieved using peptide additives that differed only in the ratio of serine (S) and lysine (K) residues [76]. The shorter KSSKK peptide precipitated spherical TiO₂–peptide composite particles with a mean diameter of 200 nm, while a longer peptide SKSK₃SKS precipitated much larger spherical particles with a mean diameter of 510 nm (Figure 10). The precipitation kinetics also differed for the two peptide additives, saturating at very low peptide concentrations for the former peptide and occurring over a large range of concentrations for the latter. Complementary density functional theory calculations did not reveal strong differences in the binding of the S and K amino acid residues to TiO₂; therefore, the difference in precipitation was instead attributed to the different self-aggregation behaviours of the peptides themselves [76].

These studies demonstrate that the size and shape of TiO₂ particles can be controlled through the choice of amino acid or peptide additives. Different nanoparticle sizes are likely to be required for different applications: for example, small nanoparticle sizes of tens of nm are beneficial for photocatalysis [101], while applications as pigments require large sizes of about 250 nm [116]. In biomedical applications, small nanoparticle sizes are useful for drug delivery, as a small size facilitates cell membrane penetration; however, a larger nanoparticle size helps avoid cytotoxicity [117]. Therefore, developing targeted synthesis procedures that controllably produce nanoparticles of specific sizes is an important direction for technological applications of TiO₂.

Finally, the removal of the template after the synthesis is a crucial step for obtaining pure TiO₂ products when conventional templates are used for the synthesis of hierarchical nanomaterials [34,118,119]. One of the benefits of calcination is that it helps to remove the organic templates. In case of bio-inspired additives, centrifuging and washing can be sufficient to remove the additives and can help avoid the calcination step [72].

3.3. Interactions of Bio-Additives with TiO₂

As summarised in Figure 1, bio-inspired additives, such as amino acids, play a multifunctional role: they initialise hydrolysis, and they can be used as catalysts, templates and capping agents in TiO₂ synthesis. They can influence the phase, shape, size and surface area of the synthesised TiO₂ nanoparticles. Going beyond the successful synthesis of TiO₂ with bio-inspired additives, a number of studies have investigated the nature of the interaction of amino acids with TiO₂ and their role in the mechanism of amino-acid-assisted TiO₂ synthesis [120,121,122,123,124,125,126,127,128,129,130]. Interactions of amino acids with Ti precursors and TiO₂ play a key role in understanding nanoparticle nucleation and growth mechanisms during amino-acid-assisted synthesis. These interactions are believed to be predominantly electrostatic [131], although other interactions, such as hydrogen bonds, can also be present [87]. The formation of hydrogen bonds between amine groups of amino acids and surface hydroxyl groups of the TiO₂ colloid was confirmed by infrared spectroscopy studies [101]. The adsorption of amino acids on TiO₂ crystal facets can affect the surface energy (interfacial tension) of these facets and stabilise particular crystal facets and/or particular polymorphs, such as anatase and brookite [87].

To understand the nature of interactions of amino acids with TiO₂, a number of experimental and theoretical studies on the adsorption of amino acids on TiO₂ have been carried out [120,121,122,123,124,125,126,127,128,129,130]. Since amino acids contain at least two functional groups (an amine group, a carboxylic group and a specific side group), their adsorption on TiO₂ is complex and depends on the nature of the amino acid, on the TiO₂ crystallographic surface and the extent of surface hydroxylation, and on the solution pH [120,122,123,126,127,128,129,130]. The studies showed that the binding of amino acids on dry (non-hydroxylated) TiO₂ surfaces occurs mainly via the carboxylic group binding to the surface Ti atoms [121,123,130], while amino acids in solution bind to the hydroxyl groups present on TiO₂ surfaces [120,128,129].

To gain a molecular-level understanding of the nature and strength of interactions of amino acids with TiO₂ surfaces, computational studies of amino acid adsorption were carried out using molecular dynamics [128,129] and density functional theory (DFT) [127,130] to compare the behaviour of basic and acidic amino acids (which are expected to be protonated or deprotonated, respectively, at a neutral pH). The adsorption of protonated amino acids, such as arginine and lysine, was found to be stronger as compared to that of deprotonated aspartic acid, both on the rutile (110) surface [128] and on the TiO₂ anatase (101) surface [127], both on the dry surfaces and in an aqueous environment [127]. A DFT study of multiple amino acids on dry rutile (110) and anatase (101) surfaces also found that polar amino acids were adsorbed more strongly than non-polar ones on both surfaces [130]. DFT calculations of amino acids on the anatase (101) surface (structures shown in Figure 11) found that the adsorption was 0.1–1.0 eV weaker on the hydrated surface compared to the dry surface; the reason for this large energy variation is that the calculated adsorption energies on the hydrated surface depended on the number of water molecules assigned to the amino acids’ solvation shells [127]. Molecular dynamics calculations of free energies (incorporating both enthalpic and entropic effects) of arginine, lysine and aspartic acid on the dry and hydrated rutile (110) surfaces showed only slightly more favourable binding on the dry surfaces, showing that the effect of entropy must be taken into account when evaluating the strength of binding [128].

Understanding the interactions of amino acids with TiO₂ is insightful for determining the role of more complex bio-additives, such as peptides that comprise multiple amino acids. For example, Puddu et al. compared the performance of two different peptides, Ti-1 (QPYLFATDSLIK) and Ti-2 (GHTHYHAVRTQT), by analysing surface interactions of the constituent amino acids of the peptides with amphoteric surface groups on the titania surface [41]. They found that the Ti-1 peptide provided faster kinetics in TiO₂ formation; this was explained by Ti-1 containing oppositely charged aspartic acid and lysine groups, which have a greater affinity towards the TiBALDH precursor than the Ti-2 peptide during the initial nucleation stage. However, Ti-2 was found to have a higher affinity towards the TiO₂ surface, because it contains several basic amino acids, which bind more strongly to the negatively charged TiO₂ surface groups at a neutral pH [41].

Thus, specific interactions between amino acids and TiO₂ govern the mechanisms, kinetics and thermodynamics of bio-inspired synthesis reactions. Furthermore, amino acid–TiO₂ interactions are relevant when amino acids are used as surface modifiers for TiO₂ functionalisation, e.g., to improve the photocatalytic activity of TiO₂ [132,133]. Moreover, bio-inspired additives, e.g., pre-polymerised dopamine, can also be used as growth inhibitors of TiO₂ nanoparticles, which enable size control as well as further functionalisation with more complex organic molecules [72].

The production of TiO₂ nanoparticles in a controllable manner rather than by trial and error requires a molecular-scale understanding of interactions at all stages of TiO₂ synthesis, from nucleation to growth and the termination of synthesis reactions. Theoretical modelling is indispensable to achieving this understanding. While calculations have already been used to investigate the strength of the binding of amino acids and peptides to TiO₂ [127,128,129,130], further insights from modelling are needed: for example, exploring the interactions of Ti precursors and growing nuclei with molecular additives will help us understand the key structural factors responsible for TiO₂ phases, shapes and reaction kinetics.

4. Photocatalytic Performance of TiO₂ Synthesised via Bio-Inspired Route

Photocatalysis is one of the key applications for titanium dioxide. By considering the relationship between material, process, property and performance (MP³), synthesis process parameters have a direct influence on the photocatalytic performance of synthesised TiO₂. For example, the selection of a suitable precursor material can improve the photocatalyst’s performance: e.g., TiO₂ made from the TiBALDH precursor showed better levels of stability and photocatalytic performance compared to TiO₂ made from Ti isopropoxide [64,134].

Table 3 provides examples of TiO₂ photocatalysts synthesised by bio-inspired routes and their photocatalytic performance. Industrial pollutants, including textile dyes such as rhodamine B, methyl orange, methylene blue, alizarin red, malachite green and crystal violet, have been degraded using bio-inspired synthesised TiO₂ on the laboratory scale [71,83,88,93,101,102]. Although bio-inspired additives do not directly participate in the photocatalytic degradation process, bio-inspired synthesis processes enable the control of the phase, morphology and size of as-synthesised TiO₂ particles and therefore can indirectly influence the photocatalytic performance. For example, the photocatalytic activity of TiO₂ produced by bio-inspired synthesis was found to be superior compared to a commercial P25 TiO₂ photocatalyst and TiO₂ synthesized without bio-additives [88,101,135,136].

Table 3. Photocatalytic performance of bio-inspired synthesised TiO₂.

Material	Precursor	Additive	Photocatalytic Process	Photocatalytic Performance	Ref.
TiO2 nanofibers	TiCl4	Pomelo peel	Degradation of methyl orange (MO), rhodamine B, reactive brilliant blue, malachite green	Better photocatalytic activity than commercial P25 TiO2. Up to 99% degradation of MO in 30 min	[88]
TiO2 nanoparticles	TiCl4	Jatropha leaf extract	Degradation of tannery wastewater	82% removal of chemical oxygen demand COD; 76% removal of Cr+6	[137]
Mesoporous TiO2 photocatalysts	Tetra butyl titanate (Ti(OBu)4)	Pollen grains	Degradation of rhodamine B	95% degradation after 120 min	[90]
TiO2 nanoparticles	Ti isopropoxide (Ti(Oi-Pr)4)	Aloe vera gel	Degradation of picric acid	Complete degradation in 120 min	[91]
TiO2 nanohybrids	TiO4	Parthenium hysterophorus	Degradation of methylene blue (MB), crystal violet (CV), methyl orange (MO), alizarin red (AR)	In 6 h: degradation in (%) 92.5 MB, 81.5 MO, 79.7 CV, 77.3 AR	[93]
Indium-modified TiO2 composite with tobacco stem silk	Tetra butyl titanate	Tobacco stem silk	Degradation of tetracycline hydrochloride (TCH)	92.9% removal efficiency in 90 min under visible light	[138]
Graphene-supported g-C3N4/TiO4 hetero-aerogels	Ti-BALDH	KIIIIKYWYAF peptide	Degradation of methylene blue, rhodamine B (RhB)	MB: 97% in 120 min; RhB: 60% in 120 min	[77]
g-C3N4/TiO2	Ti-BALDH	Arginine	Degradation of rhodamine B, phenol	Rhodamine B 84% degraded in 5 h; Phenol 76% degradation in 120 min	[71]
Mesoporous nano TiO2	Ti isopropoxide	Various amino acids	Degradation of methylene blue, calmagite	Almost complete degradation. Samples prepared with proline, valine and aspartic acid resulted in better degradation activity than P25 TiO2.	[101]
TiO2 hierarchical spheres	Tetra butyl titanate	Glycine	Degradation of methyl orange (MO)	98% degradation of MO in 30 min	[102]
L-hydroproline modified TiO2	TiCl4	L-hydroproline	Degradation of rhodamine B	97% degradation in 4 h, better performance than pure TiO2, visible light activity	[135]
Amino-acid-modified TiO2	Tetra butyl titanate	L-proline, L-arginine, L-methionine	Degradation of methyl orange (MO), direct red 16 DR16	MO removal: 95% DR16 removal: 97% in 60 min	[139]
Amino-acid-modified TiO2	Tetra butyl titanate	L-proline, L-arginine, L-methionine	Degradation of metronidazole, cephalexin	Metronidazole removal: 99.9% (TOC removal: 81%) Cephalexin removal: 97.2% (TOC removal: 75%)	[140]
CdS/Au/N-doped TiO2 heterostructure	TiCl3	Cherry blossom leaves	Hydrogen production	H2 evolution activity higher than P25 or TiO2 synthesized without template	[141]
Templated TiO2	TiCl3	Olive leaves	Hydrogen production	H2 evolution activity 64% higher than P25 under solar light and 144% higher under UV light	[65]
Templated mesoporous TiO2	TiCl3	Camellia tree leaves	CO2 reduction	Higher yield of CO + CH4, higher selectivity towards CH4 than towards P25	[66]
TiO2 rutile and brookite nanoparticles	Peroxo-titanic acid	Various amino acids	CO2 reduction	Brookite synthesised in the presence of Lys showed the highest photocatalytic activity	[68]

Table 3. Photocatalytic performance of bio-inspired synthesised TiO₂.

Material	Precursor	Additive	Photocatalytic Process	Photocatalytic Performance	Ref.
TiO2 nanofibers	TiCl4	Pomelo peel	Degradation of methyl orange (MO), rhodamine B, reactive brilliant blue, malachite green	Better photocatalytic activity than commercial P25 TiO2. Up to 99% degradation of MO in 30 min	[88]
TiO2 nanoparticles	TiCl4	Jatropha leaf extract	Degradation of tannery wastewater	82% removal of chemical oxygen demand COD; 76% removal of Cr+6	[137]
Mesoporous TiO2 photocatalysts	Tetra butyl titanate (Ti(OBu)4)	Pollen grains	Degradation of rhodamine B	95% degradation after 120 min	[90]
TiO2 nanoparticles	Ti isopropoxide (Ti(Oi-Pr)4)	Aloe vera gel	Degradation of picric acid	Complete degradation in 120 min	[91]
TiO2 nanohybrids	TiO4	Parthenium hysterophorus	Degradation of methylene blue (MB), crystal violet (CV), methyl orange (MO), alizarin red (AR)	In 6 h: degradation in (%) 92.5 MB, 81.5 MO, 79.7 CV, 77.3 AR	[93]
Indium-modified TiO2 composite with tobacco stem silk	Tetra butyl titanate	Tobacco stem silk	Degradation of tetracycline hydrochloride (TCH)	92.9% removal efficiency in 90 min under visible light	[138]
Graphene-supported g-C3N4/TiO4 hetero-aerogels	Ti-BALDH	KIIIIKYWYAF peptide	Degradation of methylene blue, rhodamine B (RhB)	MB: 97% in 120 min; RhB: 60% in 120 min	[77]
g-C3N4/TiO2	Ti-BALDH	Arginine	Degradation of rhodamine B, phenol	Rhodamine B 84% degraded in 5 h; Phenol 76% degradation in 120 min	[71]
Mesoporous nano TiO2	Ti isopropoxide	Various amino acids	Degradation of methylene blue, calmagite	Almost complete degradation. Samples prepared with proline, valine and aspartic acid resulted in better degradation activity than P25 TiO2.	[101]
TiO2 hierarchical spheres	Tetra butyl titanate	Glycine	Degradation of methyl orange (MO)	98% degradation of MO in 30 min	[102]
L-hydroproline modified TiO2	TiCl4	L-hydroproline	Degradation of rhodamine B	97% degradation in 4 h, better performance than pure TiO2, visible light activity	[135]
Amino-acid-modified TiO2	Tetra butyl titanate	L-proline, L-arginine, L-methionine	Degradation of methyl orange (MO), direct red 16 DR16	MO removal: 95% DR16 removal: 97% in 60 min	[139]
Amino-acid-modified TiO2	Tetra butyl titanate	L-proline, L-arginine, L-methionine	Degradation of metronidazole, cephalexin	Metronidazole removal: 99.9% (TOC removal: 81%) Cephalexin removal: 97.2% (TOC removal: 75%)	[140]
CdS/Au/N-doped TiO2 heterostructure	TiCl3	Cherry blossom leaves	Hydrogen production	H2 evolution activity higher than P25 or TiO2 synthesized without template	[141]
Templated TiO2	TiCl3	Olive leaves	Hydrogen production	H2 evolution activity 64% higher than P25 under solar light and 144% higher under UV light	[65]
Templated mesoporous TiO2	TiCl3	Camellia tree leaves	CO2 reduction	Higher yield of CO + CH4, higher selectivity towards CH4 than towards P25	[66]
TiO2 rutile and brookite nanoparticles	Peroxo-titanic acid	Various amino acids	CO2 reduction	Brookite synthesised in the presence of Lys showed the highest photocatalytic activity	[68]

One of the limitations of the pure TiO₂ photocatalyst is that it can only function under UV light irradiation. However, with the bio-inspired route, further enhancements in its photocatalytic activity and biocompatibility can be achieved by modifying the surface of TiO₂ with bio-inspired additives, such as the amino acids glycine, hydroxyproline or tyrosine [133,135,136], which can reduce the bandgap of the TiO₂–amino acid composite, as shown in Figure 12 for a TiO₂–tyrosine composite [136].

Surface coating with bio-inspired materials hinders the agglomeration of TiO₂ nanoparticles. Thus, TiO₂ nanoparticles coated with L-valine, L-leucine, L-isoleucine and L-methionine displayed reduced agglomeration [142]. This provides a greater surface area for interactions with pollutants and light absorption.

In summary, there are several ways in which bio-inspired additives can contribute to the photocatalytic activity of TiO₂ nanomaterials: (1) their use as an additive in the nano-TiO₂ synthesis reaction; and (2) their use for the surface functionalisation of TiO₂.

In the first aspect, when the bio-inspired additives are added to the synthesis reaction, they provide process benefits, such as carrying out the synthesis reaction at mild conditions and controlling TiO₂ particles’ phase, size and shape. TiO₂ synthesised with bio-inspired additives can form nanoparticles with a higher surface area to volume ratio, which are suitable for photocatalytic applications. In the second aspect, when the bio-additives are used as surface modifiers for TiO₂ nanoparticles, they can increase light absorption and improve the photocatalytic activity by enabling light harvesting in the visible region, thus improving the pollutant degradation performance of the TiO₂ photocatalyst.

In both of these aspects, the interactions of bio-additives, such as amino acids and proteins, with the surface of TiO₂ particles are important. In bio-inspired synthesis, the influence of these interactions on the properties is indirect, as bio-additives help control the growth of the synthesised nucleus of TiO₂ by acting as surfactants, templates and capping agents and, ultimately, as size and shape controllers. Thus, they determine the surface-area-to-volume ratio and can lead to a larger surface area for better light absorption and pollutant adsorption. In comparison, in surface functionalisation, the interactions are directly influencing the properties of the produced composite material by creating additional energy levels to enable the absorption of light in a broader spectral range and thus to facilitate photocatalysis activated by visible light.

5. Conclusions

This review highlights the role of bio-inspired additives as crucial ingredients in the synthesis of titanium dioxide, which have great potential for controlling the phase, shape and size of TiO₂ nanoparticles for specific applications, in particular for the photocatalytic degradation of pollutants. Various biomolecules, such as proteins, peptides and amino acids, have been used with different Ti precursors to successfully synthesise crystalline or amorphous TiO₂. Amino acids, as relatively simple bio-inspired additives, have also been the subject of mechanistic studies to elucidate their role in the synthesis of TiO₂. However, the key process challenges to design a green synthesis method still remain unresolved.

The key challenges and hence potential opportunities for future research in the green synthesis of TiO₂ can be summarised as follows, which are broadly based on a recently developed multi-criteria discovery, design and manufacturing framework [45]:

• Obtaining desired critical quality attributes (CQAs) such as the crystallinity and morphology of titanium dioxide suitable for desired applications, such as photocatalysis;
• Carrying out synthesis under mild conditions, ideally at room temperature;
• Assessing the economics and sustainability of bio-inspired synthesis methods;
• Up-scaling the methods of the green synthesis of TiO₂ for industrial production.

We conclude the review by discussing potential solutions to these challenges.

(1) As seen from the examples discussed in this review, progress has already been made in addressing the first of these challenges: bio-inspired additives have enabled the synthesis of specific phases and morphologies of titania nanoparticles. However, the investigations so far have used the trial-and-error approach rather than systematic explorations informed by an understanding of titania–biomolecule interactions. Further, these studies have not focussed on the CQAs needed for desired applications. To identify molecular additives that controllably produce the desired CQAs, such as TiO₂ phase and morphology, several complementary approaches should be undertaken.

First, computational investigations of amino-acid-assisted synthesis reactions are needed in order to understand the mechanisms of bio–inorganic interactions and design new efficient bio-inspired additives. The computational studies so far have investigated the binding of amino acids on TiO₂ surfaces [127,128,129,130]. Future directions should go beyond simple binding to explore reaction processes, similar to the multiscale modelling studies of the bio-inspired synthesis of silica [143].

Second, the design of experiments approach should be utilised, which seeks relationships between process variables (e.g., concentrations of reactants) and response variables (e.g., product yield) as well as performance-oriented CQAs [144,145,146].

Third, machine learning, which has emerged as a powerful technique for identifying trends in data and developing predictive models for synthesis planning [147,148], should be applied to TiO₂ synthesis. For example, design of experiments has been applied to the sol-gel synthesis of TiO₂ [149,150], while machine learning has been used to optimize the synthesis of doped TiO₂ for photocatalytic applications [151] and to investigate the effect of synthesis parameters on the size and shape of TiO₂ obtained by hydrothermal synthesis [152]. This multi-pronged approach would help reach the final goal of developing standardised synthesis protocols (Ti precursor, molecular additives, solvent, pH and temperature) for producing TiO₂ nanoparticles with well-defined sizes and phase compositions.

(2) The key challenge in the state-of-the-art solution synthesis of titanium dioxide is obtaining the desired crystallinity and morphology of the titanium dioxide at room temperature. Studies that report the synthesis of titanium dioxide at room temperature require a long time (up to several days [74,87]). Thus, reducing the reaction time while keeping the reactions at mild conditions remains an important goal. Titanium dioxide synthesised and obtained at room temperature still does not necessarily have the required crystallinity, and post-synthesis heat treatment is required in most cases to achieve crystalline TiO₂. These post-synthesis heat treatments are carried out at high temperatures, typically above 450 °C. The aim for the sustainable synthesis of TiO₂ remains to precipitate crystalline titanium dioxide at low temperatures and mild conditions.

The same three strands of exploration—computational modelling of TiO₂–biomolecule interactions to guide the choice of additives, the use of the design of experiments approach and machine learning analyses—should be applied in the search for additives that enable the low-temperature synthesis of crystalline TiO₂. At present, this area is underdeveloped, with very few examples of crystalline TiO₂ obtained at room temperature [40,72,74,87] and few examples of mixtures of amorphous and crystalline TiO₂ [41,75,76,81,83]. Based on the current knowledge, it is impossible to say conclusively whether small-molecule additives, such as amino acids, or larger additives, such as peptides, are more effective for the low-temperature synthesis of TiO₂. Finding an answer to this question requires a combination of “bottom-up” studies involving large-scale screening of multiple amino acids, and short peptide additives and “top-down” studies starting with protein additives, such as silaffin (the most successful additive for the room-temperature synthesis of TiO₂ [40]), and systematically exploring smaller fragments of these proteins. Such systematic studies can then be combined with machine learning analyses to identify peptide sequences that are effective at driving TiO₂ synthesis at low temperatures.

(3) An important point to consider is how economical, sustainable or green the bio-inspired synthesis methods are [77]. Compared to the conventional surfactants and capping agents used in the synthesis of nanomaterials, bio-inspired additives promise environmentally benign alternatives [39]. To assess this, it will be necessary to perform quantitative assessments of the bio-inspired additives and bio-assisted synthesis approaches. While a full life-cycle analysis is ideal, it can present significant practical barriers due to the lack of desired data on yields, conversions, reaction rates, etc. Simpler alternatives are being developed, such as the quantification based on the 12 principles of green chemistry. For example, a tool called DOZN^TM has been recently developed [153] and applied to nanomaterial synthesis [154]—such tools can rapidly provide an early-stage evaluation of “greenness”, which can direct the researchers in selecting and identifying greener routes to titania synthesis.

(4) Finally, the challenge for the industrial production of TiO₂ is to scale-up the methods of green synthesis. During the synthesis stage, it is often believed that scale-up challenges are either separate to the discovery stage or trivial to solve, or both. Actually, it is neither of these, because decisions made at the discovery stage, e.g., the choice of solvents and conditions, can have profound effects on scalability and can even render some syntheses inherently non-scalable. Further, while reaction rates/chemical kinetics do not change with up-scaling, transport properties (mixing, heat and mass transfer) change with scale and in a non-linear fashion [155]. Here, lessons learned from the large-scale synthesis of other relevant materials can be used, considering factors such as reactor design, reagent mixing and product separation [156,157,158], as well as evaluating the technoeconomics, availability and toxicity of reagents [158,159,160,161].

In conclusion, resolving these challenges offers the tantalising prospect of the low-cost, large-scale bio-inspired synthesis of high-value TiO₂ with desired structures and properties that are suitable for diverse applications. As described above, future research should take a holistic approach and embrace multidisciplinary collaborations spanning experiments, modelling and novel mathematical and statistical tools.