Thermally activated delayed fluorescence (TADF) compounds in energy transfer processes. A general alternative to metal-based complexes

Abstract

Recently, TADF (thermally activated delayed fluorescence) compounds have expanded their role beyond materials chemistry, enriching the field of photocatalysis and offering a cost-effective and environmentally friendly alternative to classical metal-based complexes. However, while their role in photoredox catalysis is well-established, research on their use in EnT reactions is sparse. In this tutorial review, we discuss and summarise the observed reactivity trends for some leading families of TADF compounds. Further, we examine their structure–property relationships and explore the development of synthetic methods for greener, more sustainable protocols in which TADF molecules serve as crucial catalytic systems.

---

Katy Medrano-Uribe

Katy Medrano-Uribe obtained her Master's and PhD degrees at the Universidad Nacional Autónoma de México (UNAM) under the mentorship of Prof. Luis Demetrio Miranda Gutiérrez. The main focus of her projects was the derivatization of multicomponent dehydroamino acids via polar and free-radical reactions. In 2022, she was awarded the MSCA Seal of Excellence and the Young Researchers-SoE (MUR-PNRR) funding through the NextGenerationEU program. She joined the Dell'Amico group at the University of Padua as a junior assistant professor (RTDa). Her research interests involve studying and controlling the photophysical behavior of organic molecules to discover new photochemical reactivity.

Luca Dell'Amico

Luca Dell'Amico completed his PhD in 2014 at the University of Parma under the supervision of Prof. Franca Zanardi, including a spell with Prof. Karl Anker Jørgensen at Aarhus University. Then, he was a Marie-Curie COFUND fellow with Prof. Paolo Melchiorre at ICIQ. In 2017, he began his independent career as an assistant professor at the University of Padova. In 2019, he was awarded the Giacomo Ciamician Medal from the Italian Chemical Society. In 2022, he received an ERC StG to investigate new pathways in organophotoredox catalysis. His research focuses on developing and elucidating the mechanisms of novel photochemical processes.

---

Green foundation

1. We discuss the use of organic molecules that present TADF as photosensitizers. The discussion is focused on the structure-properties relationship for the rational design, and also considering green parameters like the replacement of metal-based photosensitizers obtaining similar or highest yields, the use of safer solvents, and less energetic light sources.

2. The use of TADF compounds in photocatalysis has been studied more for the photoinduced electron transfer process than for the photoinduced energy transfer process. Therefore, it is essential to understand the potential of these molecules when analyzing overall trends in the design process. This knowledge will help synthesize new photosensitizers to modernize old methods more sustainably and to develop new ones aligned with green chemistry principles.

3. Despite the increasing application of TADF compounds in photocatalysis, their potential as photosensitizers continues to be a fascinating research area. This tutorial review aims to enhance understanding of how to rationally design new TADF-based photosensitizers that could replace metal-based alternatives, offering comparable or higher efficiency in a more environmentally friendly manner.

1. Introduction

In recent years, photocatalysis has emerged as a tool to perform innovative chemical transformations in a more sustainable manner.¹ However, when looking at light-driven energy-transfer (EnT) processes, many reactions are catalyzed by transition-metal based complexes. While these molecules are important for developing new transformations, they can often be replaced by metal-free alternatives, such as purely organic photosensitizers (PSs). In this context, some molecules that exhibit thermally activated delayed fluorescence (TADF) have emerged as promising candidates due to their ability to mimic metal-to-ligand charge transfer (MLCT) via a long-lived charge-transfer (CT) state (Fig. 1A and B). Generally, the photoexcitation of TADF compounds can lead to two distinct mechanisms: electron transfer (ET) or energy transfer (EnT), with the latter being much less used and less investigated.

Fig. 1 Design strategies in TADF molecules used in photocatalysis.

TADF molecules are characterized by a small gap between S₁ and T₁ (commonly lower than 0.3 eV), which favors ISC (intersystem crossing) and RISC (reverse intersystem crossing) while suppressing internal conversion (Fig. 1E). In this regard, the synthesis and design of organic TADF compounds is based on donor and acceptor units (D–A, Fig. 1C) or molecules that incorporate heteroatoms with lone pair electrons (Fig. 1D).² Both strategies are related to Spin–Orbit Coupling (SOC) values, which depend on the nature of the excited states, orbital types, and the heavy-atom effect.³ Based on this, a common strategy for the design of D–A molecules is characterized by highly twisted structures (Fig. 1C) that promote a well-defined delocalization of the HOMO and LUMO orbitals, decreasing their overlap and the ΔE_ST. Additionally, substitutions in donor and acceptor structures tune the molecular energy levels, enabling a rational study of structure–property relationships. On the other hand, incorporating heavy atoms (Fig. 1D) can also induce SOC, facilitating the ISC (Fig. 1E).

However, the ΔE_ST character of TADF molecules is more peculiar than only considering the spatially separated HOMO–LUMO. In particular, has been studied that in carbazolycyanobenzene-based compounds, the energy of the S₁ state is reduced more significantly than T₁ state, while incorporating carbazole units. Moreover, the T₁ state has both contributions of the locally excitation (LE) and charge transfer (CT), the latter can be increased as the ΔE_HOMO–LUMO decreases.⁴ This duality CT/LE provides them with interesting properties for applications in materials chemistry and photocatalysis.

For example, TADF compounds can undergo energy transfer processes via the Förster resonance energy transfer (FRET) and Dexter energy transfer (DET) mechanisms. The first one is related to the dipole–dipole coupling mediated over long distances. While DET is based on simultaneous intermolecular electron exchange of the excited state of the photosensitizer (PS) and the ground state of the substrate.

In contrast, the heavy-metal-based complexes are more limited, typically achieving only DET, exemplifying the importance and versatility of organic TADF-PS for future applications. The FRET pathway mediated by TADF molecules and facilitated by a higher energy triplet has recently been explored in the development of hyperfluorescent OLEDs.⁵

In organic synthesis, the ability of TADF compounds to act as photosensitizers has been more extensively exploited via DET processes (e.g., alkene photoisomerization). This reactivity can be attributed to their ISC quantum yields and long delayed-triplet lifetimes, making the Triplet–Triplet Energy Transfer (TT-EnT) thermodynamically feasible.

However, there are certain target properties of metal-based structures that the design of more effective TADF photosensitizers must aim to outperform:

(1) An efficient SOC, facilitating an ultrafast ISC from ¹MLCT to ³MLCT.⁶

(2) Structural rigidity through the incorporation of ligands, preventing the distortion, the chemical decomposition and increasing the excited states lifetime.⁷

(3) An extended wavelength window is achieved by modulating the electronic nature of the ligands, facilitating the use of less energetic light sources.⁷

In response to this, the synthetic community has been focused on the study and design of TADF compounds that can offer more sustainable methodologies, emulating or improving the photophysical and photochemical properties of metal-based compounds. These efforts are leading to new synthetic pathways in line with the need for greener chemistry.

1.1. About this tutorial review

While several review articles have discussed the theoretical aspects of TADF molecules and their synthetic importance based on photocatalysis,⁸ this tutorial review highlights the most common trend found in the structure–reactivity relationship on different families of TADF compounds that efficiently catalyze EnT manifold. We will present reports on the design of TADFs to achieve the desired triplet energy for EnT mechanisms, and we will compare the most recent methodologies that utilize TADFs with previous reports that used metal-based photosensitizers or harsh conditions. This review is organized with particular emphasis on the structure-properties relationships in TADF compounds that have been utilized in EnT processes. The work is divided into two main chapters: (i) TADF photosensitizers with donor–acceptor (D–A) structure and (II) other types of TADF compounds.

TADF photosensitizers with donor–acceptor (D–A) structure. This D–A configuration has been widely used in materials chemistry to develop new organic emitters and luminescent materials. In particular, the family of carbazolycyanobenzene-based compounds has seen remarkable use in photocatalysis, mediating EnT processes (section 2.1). The ability to modify the number and positions of cyano and N-based groups has enabled multiple combinations that can modulate the reactivity of the PS, facilitating the attainment of the target products.

Recently, the exploration of triplet energy in nitrogen-based cores has emerged as a novel alternative. In imidazoacridine- and imidazophenothiazine-based compounds (section 2.2), the dependence of ET values on structural modifications in the donor and acceptor moieties has been highlighted. Studies of these D–A compounds have examined the presence of various nitrogen-based heterocycles (modifying donor strength), and the influence of spin–orbit coupling SOC on PS reactivity. The triplet energy of more complex structures, such as diindoloquinoxaline-based compounds, has been studied (section 2.2). The structural diversification facilitates, for example, improvements in excited-state lifetimes. Also, an excellent example of how the design of new photosensitizers can be addressed is presented in section 2.2, where the PS was specifically designed for the development of a new methodology, rather than its application in benchmark reactions.

While sulfur-based acceptors have been extensively reported in materials chemistry, only a few examples have been applied in photocatalysis. Interestingly, a dual role as a photocatalyst/photosensitizer has also been explored (section 2.3), expanding the application field of this type of D–A compounds.

Other TADF compounds. TADF behavior is not limited to donor–acceptor compounds. Other examples include molecules in which a small energy gap between the singlet and triplet states arises from factors such as heavy atoms in the structure. For instance, thioxanthone (section 3.1), a core used in optoelectronic materials exhibiting TADF, has been studied for asymmetric induction. Additionally, xanthene-based compounds (section 3.2), and multiresonant structures (section 3.3) used in oxidation and isomerization processes are included.

2. TADF photosensitizers with D–A structure used in EnT processes

2.1. Carbazolycyanobenzene-based compounds

The rational design of efficient PSs is challenging. The difficulties mainly stem from the fact that several characteristics can influence the required physicochemical properties, such as the T₁ lifetime or the ISC quantum yield. However, Zhang and co-workers reported in 2018 an excellent example of how the modulation of the triplet excited states can be achieved.⁹ The authors synthesized carbazole-cyanobenzene fluorophores (D–A type), varying the number and position of cyano and carbazole groups. A clear tendency can be discerned by analyzing compounds within the same family. For example, the introduction of more electron-donating groups (EDGs) decreases the E_T from 3.03 eV to 2.68 eV (Table 1).

Table 1 Impact of increasing the number of EDGs in carbazolylcyanobenzene-based compounds

---

The E_T and lambda values served as initial criteria for selecting the fluorophores involved in the isomerization of E-stilbene (E_T = 2.2 eV) to Z-stilbene (E_T = 2.5 eV). D–A compounds with a high E_T showed a lower isomerization rate, possibly due to the simultaneous activation of the Z-isomer (see the tendency in 6, 7, and 5 in Table 2). Interestingly, compound 8 and its 1,4-dicyanobenzene regioisomer, 2CzTPN (9), do not follow this trend, showing a Z/E ratio similar to that of compounds with higher E_T values (Table 1, for molecule 2, Z/E ratio = 1.86). The authors identified strong ³LE (locally excited triplet state) character in compounds 8 and 9 based on theoretical calculations. This characteristic aligns with various reports describing that this type of TADF emitter with a D–A structure generally exhibits mixed local excitation (LE) and intramolecular charge transfer (CT) character.¹⁰

Table 2 Trend and counter-trend found in the study of carbazolylcyanobenzene-based compounds

---

Spectroscopic analysis using femtosecond and nanosecond transient absorption spectroscopy confirmed the existence of two triplet states (³LE and ³CT). The ³LE state was more significant in those PSs with an unexpectedly low Z/E ratio. With this key information, the authors proposed two different strategies to modulate the ³LE/³CT population by tuning the energy of ¹CT: (i) change the solvent polarity and analyze the effect on the ³LE depopulation. (ii) Increase the electron-donating strength in the D–A structure, improving the charge transfer process. For molecule 8, the Z/E ratio obtained in different solvents varied slightly due to a consistently lower-energy absorption band (¹CT_abs) with higher energy than ³LE. The addition of the tert-butyl group in 10, which makes the carbazole core more electron-rich, shifted the ¹CT_abs below ³LE, increasing the Z/E ratio from 2.1 to 4.6. Ultimately, both strategies demonstrated that the ³CT population can influence the efficiency of TADF fluorophores in stilbene isomerization (Fig. 2).

Fig. 2 Solvent and substitution effect in molecules 8 and 10.

It is worth mentioning that this transformation was originally carried out using higher temperatures or under UV light irradiation.¹¹ In this methodology, the authors showed that an organic PS and a CFL lamp can be efficiently used at ambient temperature.

The understanding of the origin of the triplet excited state and how to control the ³LE/³CT population allows the use of D–A fluorophores in the cross-coupling reaction of carboxylic acids and aryl halides. The synthetic strategy involves modifying the redox potential by replacing the carbazole groups with better donor units, such as diphenylamines (see molecules 16 and 17). This structural modification makes the PS less reactive toward dehalogenation of 12 (E_red = −1.58 V vs. SCE) via a singlet-electron transfer (SET) process, while enhancing the EnT manifold that leads to product 13. PS 17 represents a suitable alternative to Ir(ppy)₃ that was used in the seminal publication by Macmillan and co-workers,¹² in which the same compound 13 was obtained in 85% yield (Fig. 3).

Fig. 3 Chemoselectivity in the synthesis of 13 and 14 and results obtained with Ir(ppy)₃.

In 2020, Miyake and colleagues employed Cinnamyl bromide as a bromine radical precursor to initiate a cascade via the EnT mechanism. This process involves the β-fragmentation of 20, which produces a bromine radical that adds to vinyl or ethynylcyclopropane 18, leading to ring opening with the subsequent addition to olefin 19 to produce the key intermediate (see Fig. 4).¹³ The involvement of an EnT mechanism is supported by the limited ability of the PS to promote a SET process, considering cinnamyl bromide's redox potential (E_ox = +2.01 V) and 4CzIPN (15) . During the development of this methodology, three Iridium-based photosensitizers were tested. Notably, only Ir(dF-CF₃ppy)₂(dtbby)PF₆ exhibited results comparable to 15, with yields of 95% and 93%, respectively. The authors then continued to optimize the reaction conditions using the latter to establish a metal-free approach.

Fig. 4 Synthesis of 21 through bromine radical catalysis.

The carbazolyl cyanobenzene compound 15 has been used for the [2 + 2] photodimerization of chalcone 22, which has an E_T of 2.13 eV. Remarkably, the authors found that the tBu-4CzIPN photosensitizer 25 showed the highest conversion among the IPN derivatives. They also emphasized that 4CzIPN (15) outperformed Ir(ppy)₃ upon irradiation at 425 or 455 nm. The reaction's success depends on the PS E_T value: the TADF photosensitizer 26 with the lowest E_T (2.27 eV), shows the lowest activity (see Fig. 5).¹⁴

Fig. 5 Comparison of the activity of the photosensitizers 25 and 26 for the synthesis of compound 24.

All [4 + 2] cycloaddition reactions have been successfully photocatalyzed using 4CzIPN (15) with 2,3-dibromonaphtoquinone (27) and phenylbenzofurans (28).¹⁵ One of this methodology's most significant challenges was promoting the DA (Diels–Alder) reaction instead of the competitive [2 + 2] cycloaddition. Importantly, the authors found that adding a Lewis acid (SnCl₄) helps to diminish the competition between the two cycloadditions. Also, extending the aromatic system can help stabilize the 1,6-diradical intermediate 30.

Although a clear trend in the triplet energy of the PSs cannot be identified, the influence of solvent polarity is clear. The less polar solvent (benzene) provides the highest yield, whereas the more polar solvent (acetonitrile, MeCN) shows a decrease in yield (Fig. 6), probably due to a better stabilization of the excited state species.

Fig. 6 Effect of the solvent in the [4 + 2] cycloaddition of 27 and 28.

The same D–A fluorophore (4CzIPN, 15) has been used in the 6π-photocyclization/defluorination process of N-aryl enamines 31 (E_T = 2.6 eV) in the synthesis of indoles 32 (Fig. 7).¹⁶ Due to their similar triplet energies, the TADF compound demonstrated an equivalent performance in this reaction compared with some metal-based triplet sensitizers, making this protocol a greener option.

Fig. 7 Synthesis of indoles 32 through a 6π-photocyclization/defluorination protocol.

Different photosensitizers with varying triplet energies were used to standardize the reaction conditions. In general, the photosensitizers with a small E_T value showed a lower yield or no reaction progress (Table 3).

Table 3 Dependence of the molecule's yield 32 on the E_T value

---

The study of the EnT mechanism for the obtention of 32 was based on comparing the mismatched redox potentials of the substrate (31) and the photosensitizer during luminescence-quenching experiments. The latter showed that the N-aryl enamine 31 could quench the excited 4CzIPN. Also, using a triplet quencher like 2,5-dimethyl-hexa-2,4-diene inhibited the reaction progress.

In 2022, Kano and co-workers reported that the triplet energy of 4[Cz(CF₃)₂]IPN (37) is very similar to that of Ir–F (35) (a highly oxidizing and expensive photosensitizer, and priced at 152€ per 100 mg, Table 3). The purely organic photosensitizer performed well in the photocycloaddition of quinolinium triflate derivatives (33) with alkenes (34) (Fig. 8).¹⁷ The authors reported the triplet energy for 37 as 2.81 eV (64.8 kcal mol⁻¹), capable of sensitizing molecule 33 (E_T = 2.44 eV) to produce 35 with a 92% yield, slightly less than the 98% yield obtained using Ir–F (36).

Fig. 8 Intermolecular dearomative [4 + 2] cycloaddition for the obtention of 35.

The PS 8, with a lower E_T value, performed poorly in achieving the transformation (Table 4).

Table 4 Dependency of product yield in the synthesis of 35

---

In the same context, the use of 36 in the dearomative photocyclization can be replaced with the carbazolycyanobenzene compound 8. Adachi reported an E_T of 2.63 eV for this molecule, a value similar to the 2.68 eV assigned to Ir–F (36). This similarity inspired Koenig and co-workers to explore the use of the organic TADF molecule 8 in previously reported dearomatization protocols for naphthol (38, E_T ∼ 2.59 eV) and indole derivatives (40).¹⁸ Additionally, compound 8 shows better solubility in the most commonly used solvents and can be recycled more efficiently than the Iridium complex (Fig. 9).

Fig. 9 Dearomative cycloaddition photocatalyzed by 8.

Interestingly, replacing 1,4-dioxane with chloroform led to the selective formation of molecule 39, with the ratio increasing from 1.2 : 1 to 16 : 1. The involvement of the EnT process in this reaction can be attributed to the redox potential mismatch between the selected photosensitizer and the aromatic substrates, thereby ruling out an operative SET pathway.

2.2. N-heterocycles-based compounds

Hudson reported the synthesis of a new type of TADF emitter in 2020 based on an imidazole/acridine chromophore 42.¹⁹ This report established that using N-phenylbenzimidazole constrained in a coplanar fashion is an efficient acceptor to allow a well-localized LUMO that, combined with a heterocyclic donor, can enhance a small ΔE_ST.

In 2022, the author also reported using these D–A compounds in the [2 + 2] cycloaddition via an EnT process as a more sustainable alternative to Ir-based photosensitizers.²⁰ In this family of TADF compounds, the relationship between the strength of EDG and the E_T values remains unclear (Table 5): the worst EDG (c) is not the one with the highest E_T value.

Table 5 Impact of the EDGs on the E_T values of imidazoacridine-based photosensitizer

---

The PS 42c was shown to be highly efficient at activating compound 43 (E_T = 2.54 eV), thereby facilitating an intramolecular [2 + 2] cycloaddition to construct the tetracyclic 44 (Fig. 10). The D–A compound 42c showed similar yields compared with the metal-based PS [Ir(dF(Me)ppy)₂(dtbbpy)]PF₆, which is also one of the most expensive commercially available Ir-based photosensitizers (300€ per 100 mg). The authors proposed that the higher performance of compound 42c is related to a more efficient ISC process.

Fig. 10 Intramolecular [2 + 2] cycloaddition using 42 derivatives.

In 2023, an extensive study was published to expose the impact of the SOC phenomenon in a family of new donor–acceptor molecules. The investigation included the study of three different sulfur-based acceptors (sulfide, sulfoxide, and sulfone) and diverse donors.²¹

In the acceptor unit, the presence of the sulfur atom in the imidazophenotiazine core increases the conjugation, allowing a red-shifted absorption compared to the imidazoacridine family before commented. In general, sulfur oxidation resulted in lower E_T parameter and SOC character values. Notably, this trend continued in the proportion of ³LE in the T₁ (Table 6).

Table 6 Effect of sulfur oxidation on the imidazophenothiazine family

---

This value was essential for effective performance as a photosensitizer across nearly all tested reactions, including thiolation of alkenes (Fig. 11a) and Ni-catalyzed esterification (Fig. 11b), in which molecule 46a (higher E_T and SOC) delivered higher yields. In general, these new photosensitizers outperformed the reported 43 derivatives in all reactions studied, achieving better efficiency in each transformation.

Fig. 11 Reactivity of 45 derivatives in representative examples.

In 2023, Sengupta and collaborators synthesized four indoloquinoxaline-based TADF compounds, including phenyl carbazole 53 and diindolocarbazole (DiICz) 54, as donors.²² The complete characterization was performed to explore their various properties. Specifically, the D–A compounds reported in this study exhibited similar E_T values. However, when the photosensitization protocol for the isomerization of E-stilbene is used, a better performance was generally observed when the reaction time was extended to 30 hours. Interestingly, although compounds 53a and 53b have the same E_T value, their conversion (%) differed significantly (69% and 90%, respectively) (Fig. 12). The authors did not investigate the reason for this difference. Nonetheless, this exemplifies that the E_T parameter alone does not fully explain EnT efficiency.

Fig. 12 E/Z isomerization using indoloquinoxalines-based compounds.

The same group reported two more DiICz-based molecules with a donor–acceptor-donor structure using dibenzo[a,c]phenazine (56a) and phenanthrol[9,10-d]imidazole (56b) as donors (Fig. 13). In general, these molecules exhibited long excited-state lifetimes compared to 52 and 53.²³ These compounds were tested in the photoisomerization of molecule 55a, where 56b presented the lowest yield, due to its capacity to activate the cis-isomer.

Fig. 13 Other TADF photosensitizers reported by Sengupta and colleagues.

An interesting example of the synthesis and design of photosensitizers aimed at achieving challenging reactivity was reported by Dell'Amico and colleagues.²⁴ The goal of this synthesis was to produce azetidines 60 from azabicyclo[1.1.0]butanes 58. During the reaction with compound 59, the authors observed the formation of an imine dimer caused by the higher concentration of the iminyl radical in the reaction mixture. Controlling its formation was key to increasing the yield of the target product 60. This was achieved by promoting RISC in molecule 57 by lowering the ΔE_ST to 0.05 eV, which was done by adding a tert-butyl group (compound 57b) to increase steric hindrance (Fig. 14). In this particular case, it can be counterintuitive for the reader to see the benefit of RISC improvement when it decreases energy transfer efficiency. However, the reader must visualize that, for the purposes of this synthetic methodology, it was a strategic way to access product 60 efficiently while avoiding dimerization.

Fig. 14 Design of PS 57 for the synthesis of 60.

This methodology offers an alternative approach for designing new photosensitizers based on the reactivity requirements needed for challenging chemical transformations. Unlike the traditional method of studying the reactivity of new photosensitizers in benchmark reactions such as cycloadditions and isomerizations, this approach emphasizes specific reactivity needs.

2.3. Sulfone-based compounds

In 2023, Zysman-Colman and colleagues reported sulfur-based molecule 61, which can participate in ET and EnT processes.²⁵ The manuscript specifically compares the properties and photocatalytic reactivity of molecule 61 with the well-known 4CzIPN (15). Molecule 61 has a higher triplet energy (E_T = 2.97 eV) than 4CzIPN (E_T = 2.59 eV). The authors did not discuss the reason for this difference, such as whether a possible variance in the spin–orbit coupling nature in the two species might affect the E_T value. The benchmark reaction used to study EnT behavior was the isomerization of E-stilbene (54a). When using compound 61, the Z-isomer (55a) was obtained in 63% yield, whereas 4CzIPN yielded 87%. The authors claimed that the lower performance is related to the higher E_T value, which can activate Z-stilbene (E_T = 2.5 eV), thus facilitating the reverse process. This selectivity issue is avoided when using an alkene with higher E_T values, such as diisopropyl fumarate 54b (2.7 eV and 3.1 eV for the E and Z-isomers, respectively) (Fig. 15). Notably, the yield with 61 (81%) is the highest among those using 4CzIPN and [Ir(dF(CF₃)ppy)₂(dtbbpy)]PF₆ (6% and 58%, respectively).

Fig. 15 Selectivity obtained for the E/Z isomerization using molecule 61.

3. Other TADF photosensitizers used in EnT processes

3.1. Thioxanthone-based compounds

The thioxanthone core has been used as an acceptor moiety in the construction of OLED devices. These TADF compounds are characterized by having high triplet energy and long triplet lifetimes.²⁶ These characteristics have opened the possibility of using them as organic photocatalysts in hydrogen atom transfer (HAT), single electron transfer (SET), and different EnT protocols. Furthermore, the literature includes examples illustrating the importance and versatility of the thioxanthone core, achieving different processes, even without any structural modification.²⁷

In general, modifying the thioxanthone core has focused more on enabling asymmetric reactions. The importance of adding a chiral auxiliary has allowed these thioxanthone-based derivatives to be used in different transformations with good yields and enantioselectivity. For the purposes of this tutorial review, we will highlight the most representative examples of this structural modification.

Inspired by their pioneering work on the cycloaddition [2 + 2] assisted by a chiral photosensitizer based on xanthone,²⁸ Bach and colleagues investigated the use of chiral thioxanthone 63, which includes a 1,5,7-trimethyl-3-azabicyclo[3.3.1]nonan-2-one skeleton to promote hydrogen bonding interactions. Replacing xanthone with thioxanthone allows the use of the latter as a photosensitizer in the visible region, reducing the energy needed for substrate sensitization. The enantioselective reaction was mainly focused on modifying 2(1H)-Quinolones 62, which served as model substrates for various transformations (Fig. 16).

Fig. 16 Derivatization of compound 62 through [2 + 2] cycloaddition using alkenes (a and b), allenes (c) and allyl quinolone (d).

The first report involves synthesizing tetracyclic product 65 through an intramolecular reaction with extraordinary regio- and enantioselectivity. This work established, for the first time, the hydrogen-bonding interaction between molecules 62 and 63 in complex 64, where the thioxanthone serves as an antenna that can donate energy to the absorbed photon (Fig. 16a).²⁹

The intermolecular reaction was reported using a variety of electron-deficient alkenes 66. The successful conversion of the starting materials to product 67 was attributed to the hydrogen-bonding interaction in complex 64, whose dissociation is responsible for the loss of enantioselectivity (Fig. 16b). Importantly, from a green perspective, the authors carried out this transformation using solar irradiation without compromising the ee.³⁰

Another tetracyclic 68 variant was obtained from 3-alkylquinolinones substituted with 4-O-tethered alkenes and allenes. Notably, the quinolinones α-substituted at the carbonyl showed a decrease in the E_T value, allowing the use of photosensitizer 63 and the reaction to proceed (Fig. 16c).³¹

The same group explored the possibility of using compound 62 not only in cycloaddition reactions. In 2019, they reported the synthesis of cyclopropanes 69 using allyl quinolone. Also, the authors demonstrated the capacity of thioxanthone 63 to racemize cyclopropanes enantiomerically pure (Fig. 16d).³² Even when this type of cycloaddition can be achieved by direct excitation of the substrates, the use of 63 facilitates the visible light irradiation (420 nm) and stereocontrol. Some years later, the EnT mechanism was corroborated using DFT calculations. It was also possible to establish that the first C–C bond formation is the critical step in the rate and selectivity performance.³³

To the best of our knowledge, there is only one report on the effect of structural changes on the capacity of thioxanthone to act as a photosensitizer. The work presented by Booker-Milburn and collaborators examines the structure–property relationships modifying the substitution patterns of the aforementioned core.³⁴ The UV-Vis absorption values of the synthesized molecules ranged between 300 and 450 nm, enabling their activation with various energy sources like UV and visible light. Additionally, the UV-Vis absorption and E_T values are inversely proportional, as shown in Table 7.

Table 7 Relationship between the absorption and E_T in thioxanthones

---

The study of the potential of these molecules as triplet energy photosensitizers involves intra- and intermolecular [2 + 2] cycloadditions. A general trend in reactivity is observed when a 125 Hg lamp is used as the light source; the molecule with the highest E_T value (70) performs better than molecule 74 (lowest E_T). However, molecule 74 could still be used in a visible light protocol to achieve the [2 + 2] cycloaddition with different substrates.

3.2. Xanthene-based compounds

The family of xanthene has not been used as donor–acceptor compounds with TADF or in material chemistry, unlike other families previously studied. Nevertheless, this type of compound was initially explored in the study of this photoluminescence phenomenon. In this case, SOC, which is necessary for efficient ISC, is facilitated by the heavy-atom effect present in the structure. The most common approach to induce this effect is modifying the chromophore structure by adding metals (such as Ru, Ir, Pt) or halogens (such as Br, I). These elements are often located at the core of these traditional chromophores.

The synthetic application of this type of compound has been less explored using the EnT mechanism. Although Eosin Y was the first compound reported to exhibit TADF,³⁵ its synthetic applications following an EnT mechanism mainly involve generating singlet oxygen to mediate various oxidation reactions.

In this context, Ando and colleagues described the aerobic photooxidation of bulky tellurides to telluroxide using Eosin Y 75 and Rose Bengal 76, replacing traditional methods such as 3-chloroperbenzoic acid (mCPBA). The weaker Te–O bond in mCPBA can transfer to various organic substrates, facilitating the oxidation of alcohols³⁶ and phosphite esters,³⁷ as shown in Fig. 17. This approach is notable for its environmental friendliness, as it uses ethanol—a safer solvent—compared to the solvents previously discussed. Additionally, it enables the in situ generation of toxic telluroxides.

Fig. 17 Use of photosensitizers 75 and 76 in oxidation reactions mediated by telluroxides.

Similarly, compound 75 has been used to synthesize phosphine oxide 78, a versatile motif in organic, medicinal, and supramolecular chemistry. The reaction involves generating arylphosphadioxane 77, which can produce the target compound 78 in the presence of a second molecule of the triarylphosphine (Fig. 18).³⁸

Fig. 18 Synthesis of phosphine oxide through aerobic oxidation.

Another interesting reaction using Eosin Y (75) as a photosensitizer is the synthesis of Michael acceptors from the oxidation of the corresponding silyl enol ethers 79. The TADF molecule facilitates the generation of O₂, which undergoes an ene reaction with compound 79 to produce the cyclic intermediate 80. The subsequent ring opening yields the hydroperoxy silyl hemiacetal 81, followed by an intramolecular silyl transfer to form the α,β-unsaturated final product 82 (Fig. 19).³⁹

Fig. 19 Use of photosensitizer 75 in the synthesis of Michael acceptors.

Furthermore, the photosensitizers 75 and 76 are widely used in the photooxidation reaction of furan 83. Typically, this process involves a [4 + 2] cycloaddition to form the key intermediate 84. This bicyclic compound (84) can react in two main ways: with a nucleophile, producing the molecule 85, or with a base, forming the furanone core 86. These two products can then be used in further transformations, increasing structural complexity and diversity (Fig. 20).⁴⁰ Notably, some of these transformations are performed using safer, less hazardous solvents such as water or ethanol.

Fig. 20 Representative reactivity of furans in the presence of oxygen and xanthene-based photosensitizers.

3.3. MR-TADF compounds

The Zysman-Colman research group reported using the multi-resonant TADF compounds 87 and 88, which yielded good results for ET and EnT. These molecules were studied as photosensitizers for the isomerization of the previously mentioned substrates (54), now using visible light, and decreasing the reaction time.⁴¹ In particular, these compounds have a small positive solvatochromism and higher absorptivity. Both characteristics were expected to enhance photocatalytic activity. In fact, due to their higher absorptivity, these molecules can be used at a low concentration (0.7 mol%) for alkene isomerization, improving process efficiency and catalytic performance. Like formerly reported Donor–Acceptor PS 61 (see section 2.3), better performance was achieved when using diisopropyl fumarate 54b (90% yield) (Fig. 21).

Fig. 21 E/Z isomerization in the presence of MR-TADF compounds 87 and 88.

4. Conclusions and future perspectives

In this tutorial review, we have discussed the primary reactions mediated by energy transfer (EnT) processes in the presence of thermally activated delayed fluorescence (TADF) compounds, emphasizing their emerging role as efficient and sustainable photosensitizers in modern synthetic chemistry. These systems have demonstrated the capacity to promote a broad range of photochemical transformations under mild conditions, often relying on visible light irradiation and avoiding the need for precious metal complexes. A valuable feature of some of these molecules is their ability to also participate in electron transfer (ET) processes, which provides them with exceptional versatility for use in modern and sustainable synthetic protocols. This dual reactivity profile, in which a single photosensitizer can engage in either EnT or ET pathways depending on reaction conditions and substrate properties, represents a powerful conceptual and practical advance in photocatalyst design. Therefore, designing new molecules with this type of reactivity is a crucial focus in advancing photochemistry.

The coexistence of EnT and ET manifolds in TADF compounds originates from their distinctive electronic structure, typically based on donor–acceptor architectures that promote small singlet–triplet energy gaps (ΔE_ST). This feature enables efficient reverse intersystem crossing (RISC) and long-lived excited states, which are key parameters for productive bimolecular interactions, dictating exciton dynamics and providing long lifetimes that facilitate PS-substrate interactions (crucial for the DET mechanism), but not so long as to cause non-productive quenching. It is essential to study in depth the structure–property relationships influencing the fine-tuning of electronic and photophysical parameters that facilitate this interesting duality, ET/EnT. Subtle structural modifications, such as varying donor strength, acceptor identity, torsional angles, or regioisomerism, can alter charge-transfer (CT) character, locally excited (LE) character, and triplet energies (E_T). The former parameter has been utilized to analyze, in a general and easy way, if the EnT process between a donor (excited PS) and an acceptor (substrate) is thermodynamically feasible. Anyway, it is clear that the E_T value is not able to explain some inconsistencies in the reactivity (see discussion in section 2.1, Table 2), therefore, a deeper understanding and discussion of the physical properties of the reported TADF PS must be included in future publications. In other words, a systematic understanding of how these parameters correlate with reactivity will allow chemists to rationally design next-generation photosensitizers with predictable and programmable behavior.

Furthermore, by utilizing the structural derivatization of TADF PS, these molecules can both facilitate photoreactivity and incorporate chiral auxiliaries, leading to successful outcomes in certain asymmetric transformations. The modular nature of donor–acceptor scaffolds offers a unique opportunity to introduce stereochemical elements without compromising photophysical performance. However, only one family has been modified for these applications, leaving the potential to extend this reactivity to other TADF molecules. Expanding asymmetric variants to structurally diverse TADF platforms could unlock new catalytic modes, particularly in enantioselective energy transfer catalysis, where precise control over substrate orientation and excited-state interactions is required.

The field is still in its early stages of development. Despite the rapid progress achieved in recent years, many mechanistic aspects remain insufficiently explored, particularly regarding the quantitative prediction of energy transfer efficiencies. It is vital to understand how we can connect the photochemical properties of these structures to the existing literature on TADF molecules in material chemistry to unlock new reactivity in EnT protocols. The extensive knowledge accumulated in the context of organic light-emitting diodes (OLEDs) and optoelectronic materials provides a valuable foundation for translating photophysical principles into synthetic applications.

In the coming years, we expect TADF compounds to become increasingly prominent, replacing traditional, costly metal-based PS and enhancing the performance of organic-based alternatives. Their tunability, structural diversity, and compatibility with visible light irradiation position them as highly attractive candidates for sustainable catalysis. Continued interdisciplinary collaboration between synthetic chemists, photophysicists, and materials scientists will be essential to fully exploit the potential of TADF-based photosensitizers and to establish them as central tools in energy transfer-driven synthetic methodologies.

Conflicts of interest

There are no conflicts to declare.

Data availability

No primary research results, software or code have been included and no new data were generated or analysed as part of this review.

Acknowledgements

This work was supported by the European Research Council (ERC-Starting Grant 2021 SYNPHOCAT 101040025) (L.D.). K.M.U. thanks MUR for a Young Researchers-SoE fellowship.

References

1. C. K. Prier, D. A. Rankic and D. W. C. MacMillan, Chem. Rev., 2013, 113, 5322; S. P. Pitre, C. D. McTiernan and J. C. Scaiano, ACS Omega, 2016, 1, 66; R. C. McAtee, E. J. McClain and C. R. J. Stephenson, Trends Chem., 2019, 1, 111.
2. D. J. Volz, J. Photonics Energy, 2016, 6, 020901; W. Zhang, H. Song, J. Kong, Z. Kuang, M. Li, Q. Guo, C. F. Chen and A. Xia, J. Phys. Chem. C, 2019, 123, 19322; J. M. Teng, Y. F. Wang and C. F. Chen, J. Mater. Chem. C, 2020, 8, 11340; S. J. Zou, Y. Shen, F. M. Xie, J. De Chen, Y. Q. Li and J. X. Tang, Mater. Chem. Front., 2020, 4, 788; Y. Z. Shi, H. Wu, K. Wang, J. Yu, X. M. Ou and X. H. Zhang, Chem. Sci., 2022, 13, 3625.
3. J. M. dos Santos, D. Hall, B. Basumatary, M. Bryden, D. Chen, P. Choudhary, T. Comerford, E. Crovini, A. Danos, J. De, S. Diesing, M. Fatahi, M. Griffin, A. K. Gupta, H. Hafeez, L. Hämmerling, E. Hanover, J. Haug, T. Heil, D. Karthik, S. Kumar, O. Lee, H. Li, F. Lucas, C. F. R. Mackenzie, A. Mariko, T. Matulaitis, F. Millward, Y. Olivier, Q. Qi, I. D. W. Samuel, N. Sharma, C. Si, L. Spierling, P. Sudhakar, D. Sun, E. Tankelevičiūtė, M. Duarte Tonet, J. Wang, T. Wang, S. Wu, Y. Xu, L. Zhang and E. Zysman-Colman, Chem. Rev., 2024, 124, 13736.
4. K. Lee and D. Kim, J. Phys. Chem. C, 2016, 120, 28330.
5. For recent bibliography about the use of TADF compounds in hyperfluorescence materials, see: T.K. Stavrou, L. G. Franca, A. Danos and A. P. Monkman, Nat. Photonics, 2024, 18, 554; U. Deori, G. P. Nanda, C. Murawski and P. Rajamalli, Chem. Sci., 2024, 15, 17739; M. Fatahi, D. Barman, Y. Tsuchiya, C. Adachi and E. Zysman-Colman, Adv. Opt. Mater., 2026, e71254; A. Aarti, B. P. Veettil, A. Rodger and K. Venkatesan, Adv. Opt. Mater., 2025, 13, e01648.
6. M. H. Shaw, J. Twilton and D. W. C. MacMillan, J. Org. Chem., 2016, 81, 6898.
7. D. Kim, V. Q. Dang and T. S. Teets, Chem. Sci., 2024, 15, 77.
8. M. A. Bryden and E. Zysman-Colman, Chem. Soc. Rev., 2021, 50, 7587.
9. J. Lu, B. Pattengale, Q. Liu, S. Yang, S. Li, J. Huang and J. Zhang, J. Am. Chem. Soc., 2018, 140, 13719.
10. Y. Olivier, M. Moral, L. Muccioli and J.-C. Sancho-García, J. Mater. Chem. C, 2017, 5, 5718; B. H. Drummond, N. Aizawa, Y. Zhang, W. K. Myers, Y. Xiong, M. W. Cooper, S. Barlow, Q. Gu, L. R. Weiss, A. J. Gillett, D. Credgington, Y. J. Pu, S. R. Marder and E. W. Evans, Nat. Commun., 2021, 12, 4532.
11. S. Yamashita, Bull. Chem. Soc. Jpn., 1961, 34, 490.
12. E. R. Welin, C. Le, D. M. Arias-Rotondo, J. K. McCusker and D. W. C. MacMillan, Science, 2017, 355, 380.
13. D. F. Chen, C. H. Chrisman and G. M. Miyake, ACS Catal., 2020, 10, 2609.
14. Q. A. Wu, F. Chen, C. C. Ren, X. F. Liu, H. Chen, L. X. Xu, X. C. Yu and S. P. Luo, Org. Biomol. Chem., 2020, 18, 3707.
15. Z. Mao, A. Huang, L. Ma and M. Zhang, RSC Adv., 2021, 11, 38235.
16. Y. F. Song, X. e. Niu, J. Zhao, S. Shen and X. L. Yang, Org. Chem. Front., 2022, 9, 3499.
17. T. Morofuji, T. Kurokawa, Y. Chitose, C. Adachi and N. Kano, Org. Biomol. Chem., 2022, 20, 9600.
18. B. Rolka and B. Koenig, Org. Lett., 2020, 22, 5035.
19. E. R. Sauvé, J. Paeng, S. Yamaguchi and Z. M. Hudson, J. Org. Chem., 2020, 85, 108.
20. E. R. Sauvé, D. M. Mayder, S. Kamal, M. S. Oderinde and Z. M. Hudson, Chem. Sci., 2022, 13, 2296.
21. R. Hojo, K. Bergmann, S. A. Elgadi, D. M. Mayder, M. A. Emmanuel, M. S. Oderinde and Z. M. Hudson, J. Am. Chem. Soc., 2023, 145, 18366.
22. S. Sharma and S. Sengupta, Chem. – Eur. J., 2024, 30, e202303754.
23. S. Sharma and S. Sengupta, Org. Chem. Front., 2023, 10, 6087.
24. R. I. Rodríguez, V. Corti, L. Rizzo, S. Visentini, M. Bortolus, A. Amati, M. Natali, G. Pelosi, P. Costa and L. Dell’Amico, Nat. Catal., 2024, 7, 1223.
25. M. A. Bryden, F. Millward, T. Matulaitis, D. Chen, M. Villa, A. Fermi, S. Cetin, P. Ceroni and E. Zysman-Colman, J. Org. Chem., 2023, 88, 6364.
26. For examples of Thioxanthones used in materials chemistry; N. Kanno, Y. Ren, Y. Kusakabe, K. Suzuki, K. Shizu, H. Tanaka, Y. Wada, H. Nakagawa, J. Geldsetzer and H. Kaji, Appl. Phys. Express, 2023, 16, 011006; Y. Mu, L. Liao, W. Li, Z. Yang, Y. Zhang, Y. Huo and Z. Chi, Chem. – Eur. J., 2023, 29, e202300867; Z. Wang, Y. Li, X. Cai, D. Chen, G. Xie, K. K. Liu, Y. C. Wu, C. C. Lo, A. Lien, Y. Cao and S. J. Su, ACS Appl. Mater. Interfaces, 2016, 13, 8627.
27. For reviews where the diverse reactivity of Thioxanthone is widely discussed, see: N. F. Nikitas, P. L. Gkizis and C. G. Kokotos, Org. Biomol. Chem., 2021, 19, 5237; V. P. Demertzidou, E. Skolia and C. G. Kokotos, ChemCatChem, 2025, 17, e00760.
28. M. M. Maturi, M. Wenninger, R. Alonso, A. Bauer, A. Pöthig, E. Riedle and T. Bach, Chem. – Eur. J., 2013, 19, 7461.
29. R. Alonso and T. Bach, Angew. Chem., Int. Ed., 2014, 53, 4368.
30. A. Tröster, R. Alonso, A. Bauer and T. Bach, J. Am. Chem. Soc., 2016, 138, 7808.
31. X. Li, C. Jandl and T. Bach, Org. Lett., 2020, 9, 3618–3622.
32. A. Tröster, A. Bauer, C. Jandl and T. Bach, Angew. Chem., 2019, 58, 3538.
33. Y. Yang, Y. Wen, Z. Dang and H. Yu, J. Phys. Chem. A, 2017, 121, 4552.
34. L. D. Elliott, S. Kayal, M. W. George and K. Booker-Milburn, J. Am. Chem. Soc., 2020, 142, 14947.
35. C. A. Parker and C. G. Hatchard, Trans. Faraday Soc., 1961, 57, 1894.
36. M. Oba, M. Endo, K. Nishiyama, A. Ouchi and W. Ando, Chem. Commun., 2004, 14, 1672.
37. M. Oba, Y. Okada, K. Nishiyama and W. Ando, Org. Lett., 2009, 11, 1879.
38. Y. Zhang, C. Ye, S. Li, A. Ding, G. Gu and H. Guo, RSC Adv., 2017, 7, 13240.
39. J. Zhang, L. Wang, Q. Liu, Z. Yang and Y. Huang, Chem. Commun., 2013, 49, 11662.
40. For an extensive discussion about furan reactivity in the presence of oxygen, see; T. Montagnon, D. Kalaitzakis, M. Triantafyllakis, M. Stratakis and G. Vassilikogiannakis, Chem. Commun., 2014, 50, 15480.
41. C. Prentice, J. Morrison, A. Smith and E. Zysman-Colman, Chem. – Eur. J., 2023, 29, e202202998.

---