FIELD OF THE INVENTION

The invention relates to spiro-(fluorene-9,9’-(thio)xanthene)-based compounds and a process for the preparation of these. The compounds are useful as thermally activated delayed fluorescence material in electroluminescent devices such as organic light-emitting diodes (OLEDs). The invention extends to electroluminescent devices comprising one or more of these compounds as an emitter and a host material.

BACKGROUND OF THE INVENTION

The broadcasting of information in modern society is heavily dependent on displays and OLEDs are becoming increasingly desirable for curved TVs, high-contrast- ratio smart phones and virtual reality (VR) technologies. In comparison with traditional liquid crystal displays (LCDs), OLEDs have been found to provide improved image quality (better contrast, higher brightness, fuller viewing angle, a wider colour range and much faster refresh rates), have lower power consumption, and can be of simpler design (and thus more suited to enabling ultra-thin, flexible, foldable and transparent displays). The OLED market is expected to be valued at almost $50 billion by 2023, having grown significantly in the last five years.

OLEDs are light emitting diodes in which the emissive electroluminescent layer is a film of an organic compound or of organic compounds that emits light in response to an electric current.

A parameter used to characterise an OLED is its external quantum efficiency (EQE), the ratio of the number of photons emitted from the OLED to the number of electrons passing through the device. The EQE is a measure of how efficiently the device converts electrons to photons and allows them to escape. It is desirable for an OLED to have a high EQE. Prior to the introduction of phosphorescent materials based on platinum and iridium complexes, the maximum efficiency for OLED devices was limited to 25%. These new materials increased the internal quantum efficiency to 100% but are not without their challenges: the materials are expensive, often challenging to process from solution and have low efficiency in emitting blue light.

Thermally activated delayed fluorescence (TADF) has emerged as a very popular mechanism for harvesting excitons in OLEDs. Metal-free TADF materials, like state-of-the-art phosphorescent materials, can harvest 100% of the generated excitons and convert them into light. Electroluminescent devices using TADF emitters exhibit comparable efficiencies to state-of-the-art phosphorescent OLEDs. TADF materials have performance advantages over, e.g. conventional fluorescent materials since they can harvest both singlet and triplet excitons for light emission and thus can have a high internal quantum efficiency (up to 100%).

An efficient TADF mechanism is observed from charge transfer (CT) systems that possess a small singlet-triplet energy gap (ΔE _ST ) as a consequence of the small exchange integral. In donor-acceptor TADF emitter architectures, hole and electron densities in the lowest singlet and triplet excited states are spatially separated. This is frequently due to a large torsion between donor and acceptor fragments. A small ΔE _ST allows for efficient reverse intersystem crossing (RISC), resulting in an observed TADF upon photo- or electrical excitation. However, unlike phosphorescent OLEDs, TADF OLEDs frequently suffer from high efficiency roll-off at high current densities owing to the relatively longer delayed exciton lifetimes (τ _d ) of the emitters. Many efforts have been devoted to combat this limitation by modification of TADF emitter architecture. For example, it has been established both theoretically and experimentally that a highly twisted donor-acceptor structure with a correspondingly small ΔE _ST is advantageous when seeking high RISC rates. However, this approach leads to low oscillator strength, and the resultant low radiative decay rates of the emissive S ₁ state usually lead to low photoluminescence quantum yields (Φ _PL ). There is therefore a challenge in emitter design to develop a system that possesses both small ΔE _ST and high Φ _PL in order to realize efficient TADF OLEDs.

Small ΔE _ST values can be obtained in spiro-type compounds where donor and acceptor moieties are disposed orthogonally with respect to each other. Owing to their three-dimensional structure, excimer emission is suppressed in the solid state. Spiro compounds also possess excellent thermal stability, high glass transition temperatures, and have been widely studied as optoelectronic materials. Despite this, there have been limited reports of emitter design based on spiro architectures. Fig. 1 depicts reported spiro-based TADF emitters.

The first example of a spiro-based TADF compound, spiro-CN (Nakagawa, T.; Ku, S. Y.; Wong, K. T.; Adachi, C., Electroluminescence based on thermally activated delayed fluorescence generated by a spirobifluorene donor-acceptor structure. Chem. Commun. 2012, 48, 9580-9582), is a yellow-emitting material (λ _PL = 545 nm) with a ΔE _ST and τ _d of 0.056 eV and 14 μs, respectively, in 6 wt% mCP doped films. The Φ _PL , however, remained only 27% and the resulting OLEDs displayed a maximum external quantum efficiency, EQE _max , of only 4.4% with λ _EL of 550 nm.

Upon incorporation of the donor fragment within the spiro framework as in ACRFLCN, (Mehes, G.; Nomura, H.; Zhang, Q.; Nakagawa, T.; Adachi, C., Enhanced electroluminescence efficiency in a spiro-acridine derivative through thermally activated delayed fluorescence. Angew Chem Int Ed Engl 2012, 51, 11311-11315), the photophysical properties and the device performances were improved significantly. ACRFLCN exhibited sky blue emission (λ _PL = 485 nm) with a small ΔE _ST of 0.01 eV but with a rather long τ _d of 3.9 ms in 6 wt % doped TPSI-F films. The Φ _PL in the solid state was nearly 67% and the OLEDs possessed an improved EQE _max of 10.1% with λ _EL = 500 nm; however, the devices showed rather high efficiency roll-off.

When both the donor and acceptor units are intrinsically a part of the spiro framework as in ACRSA (Nasu, K.; Nakagawa, T.; Nomura, H.; Lin, C. J.; Cheng, C. H.; Tseng, M. R.; Yasuda, T.; Adachi, C., A highly luminescent spiro-anthracenone-based organic light-emitting diode exhibiting thermally activated delayed fluorescence. Chem. Commun. 2013, 49, 10385-10387), sky blue emission was obtained (λ _PL = 490 nm). In the 20 wt% DPEPO doped films, a small ΔE _ST of 0.030 eV, short τ _d of 5.3 μs and highΦ _PL of 81% were obtained. The resulting OLEDs showed an EQE _max of 16.5% with a λ _EL of 490 nm. However, this high EQE _max was obtained at a very low current density of 0.01 mA cm ^-2 .

The emitter DPAA-AF (Ohkuma, H.; Nakagawa, T.; Shizu, K.; Yasuda, T.; Adachi, C., Thermally Activated Delayed Fluorescence from a Spiro-diazafluorene Derivative. Chem. Lett. 2014, 43, 1017-1019), where the acridine core donor unit is substituted with peripheral electron-donating diphenylamines, exhibited blue-green emission (λ _PL = 490 nm), a ΔE _ST of 0.021 eV, τ _d of 4.3 μs and Φ _PL of 70% in 6 wt% mCP doped films. The OLEDs showed an EQE _max of 9.6% and an λ _EL of 499 nm.

Two emitters OSTFCN and OSTFB (Wang, Y.-K.; Wu, S.-F.; Yuan, Y.; Li, S.-H.; Fung, M.-K.; Liao, L.-S.; Jiang, Z.-Q., Donor-σ-Acceptor Molecules for Green Thermally Activated Delayed Fluorescence by Spatially Approaching Spiro Conformation. Org. Lett. 2017, 19, 3155-3158) were based on a modified design of the emitter ACRSA. OSTFCN and OSTFB both showed green-yellow emission with λ _PL of 530 and 550 nm, accompanied by relatively larger ΔE _ST of 90 and 0.110 eV, longer τ _d of 150 and 180 μs and Φ _PL of 80% and 60%, respectively, in 10 wt% mCP doped films. The OLEDs displayed high EQE _max values of 20.4% and 18.8% with λ _EL of 530 and 550 nm for OSTFCN and OSTFB, respectively.

The spiro-blocking strategy has recently been demonstrated to realize deep blue TADF emission in non-doped devices of the emitter TXADO spiro DMACF (Rao, J.; Zhao, C.; Wang, Y.; Bai, K.; Wang, S.; Ding, J.; Wang, L., Achieving Deep-Blue Thermally Activated Delayed Fluorescence in Nondoped Organic Light-Emitting Diodes through a Spiro-Blocking Strategy. ACS Omega 2019, 4, 1861-1867). This emitter possessed a λ _PL of 445 nm, a Φ _PL of 54%, a much larger ΔE _ST of 0.280 eV resulting in a longer τ _d of 160 μs in neat films. Non-doped OLEDs based on TXADO spiro DMACF showed deep blue emission [λ _EL = 444 nm; CIE of (0.16, 0.09)]. However, the EQE _max of the OLED was limited to 5.3% and device suffered from severe efficiency roll-off at high current densities.

Recently, emitter SAF-3CN was reported (Zhu, X.-D.; Peng, C.-C.; Kong, F.-C.; Yang, S.-Y.; Li, H.-C.; Kumar, S.; Wang, T.-T.; Jiang, Z.-Q.; Liao, L.-S., Acceptor modulation for improving a spiro-type thermally activated delayed fluorescence emitter. J. Mater. Chem. C 2020, 8, 8579-8584), a slightly modified design of ACRFLCN where an additional electron-withdrawing cyano group was introduced at the C4 site of the fluorene moiety. SAF-3CN showed green emission with λ _PL of 540 nm, Φ _PL of 65%, a small ΔE _ST of 0.01 eV and τ _d of 21.5 μs in 2 wt% doped films in CBP host. The OLEDs exhibited green emission [λ _EL = 532 nm; CIE of (0.36, 0.57)] and an EQE _max of 19.4%, which reduced to 14.4% at 1000 cd m ^-2 .

The ACRFLCN moiety was modified in emitter QAFCN to generate a more rigid donor structure in order to enhance both the Φ _PL and RISC in the solid state (Wang, Y.- K.; Huang, C.-C.; Ye, H.; Zhong, C.; Khan, A.; Yang, S.-Y.; Fung, M.-K.; Jiang, Z.-Q.; Adachi, C.; Liao, L.-S., Through Space Charge Transfer for Efficient Sky-Blue Thermally Activated Delayed Fluorescence (TADF) Emitter with Unconjugated Connection. Adv. Optical Mater. 2020, 8, 1901150). QAFCN showed sky-blue emission with λ _PL = 490 nm,Φ _PL of 71%, and ΔE _ST of 0.19 eV in 12 wt% doped films in DPEPO host. The resulting sky-blue OLEDs [λ _EL = 488 nm; CIE of (0.19, 0.35)] showed an EQE _max of 17.9%, which reduced significantly to 9.5% at high current density of 100 cd m ^-2 .

From a synthetic perspective, all the previously reported structures are very similar in that they have either nitrile acceptors (Spiro-CN, ACRFLCN, OSTFCN, SAF- 3CN and QAFCN), or the acceptor substituent is incorporated into one of the spiro π- systems (ACRSA, OSTFB, DPAA-AF and TXADO spiro DMACF). This may simplify, in some cases, the synthesis of these materials but it also restricts molecular design to a small number of donor-acceptor pairs, thereby limiting the utilization of the great potential associated with the spiro architecture for designing highly efficient TADF emitters.

Spiro-CN and ACRFLCN are based on spirobifluorene (SBF) cores. A number of further TADF materials based on SBF cores have been reported. In CN 107011184 (AAC Microtech (Changzhou) Co. Ltd.), several SBF-based compounds are described, in addition to light-emitting devices utilizing these compounds.

In WO 2013/011954 and WO 2013/011956 (Kyushu University National University Corporation), an organic electroluminescence element using SBF-based compounds in a light emitting layer is described. The SBF-based compounds are synthesised by first preparing each half of the compound before coupling or by carrying out a stepwise substitution of a tetrabrominated SBF core.

Ken-Tsung Wong et al., in J. Org. Chem. 2006, 71, 2, 456-465, describe the synthesis of SBF-based compounds in order to investigate photoinduced electron transfer (PET). The compounds are synthesised from analogous SBF-based compounds comprising cyano acceptor groups, the cyano groups of which are further functionalised in order to build up different acceptor groups.

The introduction of TADF materials enables the incorporation of purely organic molecules, which are low cost, easily modifiable and processable. Whilst it is evident that spiro-configured architectures incorporated into emitter design can contribute towards small ΔE _ST values, concurrently maintaining desirable Φ _PL and τ _d values and/or allowing access to a variety of donor- and acceptor- substituted spiro compounds remains a challenge. Perhaps the most significant commercial challenge facing the sector is the development of efficient deep blue OLED emitters. Furthermore, out- coupling light from TADF devices remains a limiting factor. The present invention seeks to address these challenges.

SUMMARY OF THE INVENTION

The inventors have demonstrated for the first time how one can use compounds with a spirofluoroxanthene (SFX) core to obtain high-efficiency OLEDs. The SFX core used by the inventors is inexpensive (about 30 to 40 times less expensive than SBF) and each half of the SFX core may be substituted independently from the other half. The colour of the emission of the SFX-based compounds is tunable by selecting different pairs of donors and acceptors. This technology could potentially contribute to the development of efficient deep blue OLED emitters.

The present invention provides novel spiro-(fluorene-9,9’-(thio)xanthene)-based compounds suitable for use as TADF materials in electroluminescent devices such as OLEDs. In addition to use as TADF material in electroluminescent devices, the compounds described herein may also function as ambipolar hosts for phosphorescent OLEDs.

The invention is premised on harnessing the electronic separation of the two halves of spiro-(fluorene-9,9'-(thio)xanthene) by the quaternary sp ³ hybridised carbon atom forming part of both halves of the spiro compound. The inventors reasoned that a consequence of this electronic separation is that the substitution of both halves of the molecule, that is substitution of the fluorene part of the molecule on the one hand, and substitution of either xanthene or thioxanthene on the other, may be effected independently. For example, and in particular, introduction of electron-deficient substituents on one half of the molecule does not decrease the capability of the other half to electrophilic substitution. Accordingly, the present invention allows spiro- (fluorene-9,9’-(thio)xanthene)-based compounds as an intact unit to be derivatised, both on the fluorene and (thio)xanthene sides, rather than introducing desired substituents into precursor molecules before generation of the spiro centre. This approach allows generation of the compounds described and claimed herein, and provides significant flexibility to the generation of spiro-based TADFs, and in particular TADF emitters with small ΔE _ST , high Φ _PL , short τ _d , which may be expected to translate to OLEDs with high EQEs and improved efficiency roll-off. Exemplified compounds have been tested as emitters in OLEDs where they show high efficiencies.

The quaternary sp ³ hybridised carbon atom forms part of both halves of the spiro compound. The inventors reasoned that a consequence of this electronic separation is that the substitution of both halves of the molecule, that is substitution of the fluorene part of the molecule on the one hand, and substitution of either xanthene or thioxanthene on the other, may be effected independently. For example, and in particular, introduction of electron-deficient substituents on one half of the molecule does not decrease the capability of the other half to electrophilic substitution. Accordingly, the present invention allows spiro-(fluorene-9,9’-(thio)xanthene)-based compounds to be derivatised as an intact unit, both on the fluorene and (thio)xanthene sides, rather than introducing desired substituents into precursor molecules before generation of the spiro centre. This approach allows generation of the compounds described and claimed herein, and provides significant flexibility to the generation of spiro-based TADFs

A series of spiro-configured xanthene-based TADF emitters has been synthesised and found to exhibit a confluence of desirable photophysical properties. Three highly efficient TADF emitters have been developed by incorporating a spiro- (fluorene-9,9’-xanthene) (SFX) system, where a fluorene unit decorated with electron- accepting phosphine oxides is spiro-linked to a xanthene core containing pendant diarylamine donors. As with most spiro compounds, the SFX-based emitters exhibit very weak spatial overlap between the frontier molecular orbitals due to the mutually orthogonal donor (D) and acceptor (A) orientation, so that the lowest electronic excitation should be ‘dark’. However, vibronic coupling in the lowest excited state can result in intensity borrowing from higher-lying bright states, imparting the emissive state with non- negligi ble oscillator strength and radiative decay rates.

The degree of overlap between HOMO and LUMO moieties was modulated through the choice of donor system in order to achieve efficient radiative decay and short τ _d (<10 μs) while maintaining a small ΔE _ST (ca. 0.10 eV) in doped films in a mCP host matrix. Further, the inherent rigidity of a spiro centre also restricts both intramolecular non-radiative decay to the ground state from either the first singlet (S ₁ ) or triplet (T ₁ )) excited states and intermolecular interactions in the solid state, which limits bimolecular recombination processes such as triplet-triplet and triplet-polaron annihilations, contributing to the high Φ _PL (~70%).

OLEDs incorporating these emitters exhibited sky-blue to green emission. EQE _max values approaching 11% and 16% were obtained for SFX-PO-DPA and SFX- PO-DPA-OMe, respectively. Impressively, the OLED with SFX-PO-DPA-Me as the emitter showed an EQE _max of 23% combined with low efficiency roll-off (EQE ₁₀₀ = 19% at 100 cd m ^-2 ). This device performance validates the SFX spiro design and places it as the most efficient OLED among those employing spiro-configured TADF emitters to date. Diphenylphosphine oxide was used as an acceptor and focus was on the influence of the different diphenylamine substituents. However, the results indicate great influence of the type of the substituent, and hence a wide opportunity for fine-tuning and perspective for further development. The optoelectronic properties of the emitters can be modulated by appropriate choice of donor-acceptor combinations.

Molecular modelling reveals that the emission in these compounds occurs through a Herzberg-Teller mechanism where some intramolecular vibrations promote larger overlapping electron and hole density in vicinity of the sp ³ carbon atoms, thereby enhancing the radiative decay rate. The designed compounds exhibit non-radiative decay rates an order of magnitude slower than the rate of reverse intersystem crossing. The resulting OLEDs exhibited high EQE _max values with only modest efficiency roll-off at 100 and 1000 cd m ^-2 as a result of the short delayed lifetimes and reduced triplet-triplet and triplet-polaron annihilation associated with the expected reduced triplet diffusion due to the bulky shape of the compounds. These results clearly illustrate that SFX blocking units are able to produce highly efficient TADF systems.

Viewed from a first aspect, the invention provides a compound of formula (I) wherein:

-X- is -O- or -S-; each D is independently an aromatic electron-donating group; each A is independently an electron-accepting group; each of o, p, m and n is independently 0, 1 or 2 with the proviso that both o + p is not 0 and m + n is not 0; and each of the optionally D- and A-substituted aromatic rings is independently optionally substituted with 1 to 3 C ₁ -C ₆ alkyl groups.

Viewed from a second aspect, the invention provides a method of preparing a compound of formula (I) comprising:

(i) halogenating a spiro-(fluorene-9,9’-(thio)xanthene) compound of formula (la) wherein:

-X- is -O- or -S-; each A is independently an electron-accepting group, each of m and n is independently 0, 1 or 2 with the proviso that m + n is not 0; and each of the optionally A-substituted aromatic rings and the other two aromatic rings depicted in formula (la) is independently optionally substituted with 1 to 3 C ₁ -C ₆ alkyl groups, to obtain a compound of formula (lb) wherein: each Y is independently a halo group; m and n are the same as the compound of formula (la) from which a compound of formula (lb) is obtained; each of o and p is independently 0, 1 or 2 with the proviso that o + p is not 0; and each of the optionally Y- and A-substituted aromatic rings is independently optionally substituted with one to three C ₁ -C ₆ alkyl groups; and

(ii) substituting each halo group Y on the compound of formula (lb) with an electron- donating group D.

Viewed from a third aspect, the invention provides a light-emitting device comprising any one or a combination of compounds according to the invention’s first aspect.

Further aspects and embodiments of the invention will be evident from the discussion that follows below.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 depicts chemical structures and performances of spiro-based TADF compounds in the prior art.

Fig. 2 shows cyclic voltammograms (CVs) and differential pulse voltammograms (DPVs) of three compounds of the invention (SFX-PO-DPA (bottom), SFX-PO-DPA-Me (middle) and SFX-PO-DPA-OMe (top)).

Fig. 3a shows the UV-Vis absorption and photoluminescence (PL) spectra of the three emitters in toluene ((SFX-PO-DPA (λ _PL = 490 nm), SFX-PO-DPA-Me ((λ _PL = 492 nm) and SFX-PO-DPA-OMe (λ _PL = 514 nm)).

Fig. 3b shows the PL spectra in PMMA and mCP thin films of SFX-PO-DPA (10 wt% in both), SFX-PO-DPA-Me (10 wt% in PMMA and 15 wt% in mCP) and SFX-PO- DPA-OMe (10 wt% in both). λ _exc = 360 nm.

Fig. 4 shows normalized time-resolved PL spectra of vacuum-deposited mCP films with 10 wt% SFX-PO-DPA, 15 wt% SFX-PO-DPA-Me and 10 wt% SFX-PO-DPA- OMe at room temperature. λ _exc = 378 nm.

Fig. 5 shows temperature-dependent time-resolved PL decay traces of doped films in mCP with (a) 10wt% SFX-PO-DPA (b) 15 wt% SFX-PO-DPA-Me and (c) 10 wt% SFX-PO-DPA-OMe ( λ _exc = 378 nm).

Fig. 6 shows prompt and delayed (by 1 ms) spectra (at 77 K) of doped films in mCP with (a) 10 wt% SFX-PO-DPA (b) 15 wt% SFX-PO-DPA-Me and (c) 10 wt% SFX- PO-DPA-OMe ( λ _exc = 355 nm).

Fig. 7 shows chemical structures and energy levels of materials used for device fabrication.

Fig. 8 shows (a) current density-voltage characteristics; (b) luminance vs voltage; (c) EQE vs Luminance; and (d) normalized EL spectra of SFX-PO-DPA (λ _EL of 500 nm), SFX-PO-DPA-Me ( λ _EL of 502 nm) and SFX-PO-DPA-OMe ( λ _EL of 520 nm). DETAILED DESCRIPTION OF THE INVENTION

The definitions set out below apply herein, unless a context dictates to the contrary.

Throughout this specification, one or more aspects of the invention may be combined with one or more features described in the specification to define distinct embodiments of the invention.

The nomenclature used herein for defining compounds, in particular the compounds described herein, is intended to be in accordance with the rules of the International Union of Pure and Applied Chemistry (IUPAC) for chemical compounds, specifically the "IUPAC Compendium of Chemical Terminology (Gold Book)” (see A. D. Jenkins et al., Pure & Appl. Chem., 1996, 68, 2287-2311). For the avoidance of doubt, if a rule of the IUPAC organisation is contrary to a definition provided herein, the definition herein is to prevail.

References herein to a singular of a noun encompass the plural of the noun, and vice-versa, unless the context implies otherwise. For example, “a compound of formula (I)” refers to one or more compounds of formula I.

Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps. The term “comprising” includes within its ambit the terms “consisting essentially of’ and "consisting of.” “Consisting essentially of’ permits the inclusion of substances that do not materially affect the basic and novel characteristics of the composition under consideration.

The term "consisting” or variants thereof is to be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, and the exclusion of any other element, integer or step or group of elements, integers or steps.

The term "stereoisomer” is used herein to refer to isomers that possess identical molecular formulae and sequence of bonded atoms, but which differ in the arrangement of their atoms in space.

The term "enantiomer" defines one of a pair of molecular entities that are mirror images of each other and non-superimposable, i.e. cannot be brought into coincidence by translation and rigid rotation transformations. Enantiomers are chiral molecules, i.e. are distinguishable from their mirror image.

The term “racemic” is used herein to pertain to a racemate. A racemate defines a substantially equimolar mixture of a pair of enantiomers. The term “diastereoisomers” (also known as diastereomers) defines stereoisomers that are not related as mirror images.

The term “solvate" is used herein to refer to a complex comprising a solute, such as a compound or salt of the compound, and a solvent. If the solvent is water, the solvate may be termed a hydrate, for example a mono-hydrate, di-hydrate, tri-hydrate etc, depending on the number of water molecules present per molecule of substrate.

The term "isotope” is used herein to define a variant of a particular chemical element, in which the nucleus necessarily has the same atomic number but has a different mass number owing to it possessing a different number of neutrons.

The term “about" herein, when qualifying a number or value, is used to refer to values that lie within ± 5% of the value specified. For example, if a temperature is specified to be about 100 °C, temperatures of 95 to 105 °C are included.

The term “electron-donating group” refers to a group that donates electron density into the u-system of the spiro-(fluorene-9,9’-(thio)xanthene)-based compound. Electron donation may be via resonance or inductive effect. The electron-donating group is typically more electron-donating than a proton at the same position.

The term “electron-withdrawing group” refers to a group that withdraws electron density from the u-system of the spiro-(fluorene-9,9'-(thio)xanthene)-based compound. Electron withdrawal may be via resonance or inductive effect. The electron-withdrawing group is typically more electron-withdrawing than a proton at the same position.

By alkyl is meant a saturated hydrocarbyl radical, which may be straight-chain or branched. Typically, alkyl groups will comprise from 1 to 25 carbon atoms, more usually 1 to 10 carbon atoms, more usually still 1 to 6 carbon atoms. Thus, for example, “alkyl" can mean methyl, ethyl, propyl, isopropyl, butyl, isobutyl, tert-butyl, pentyl or hexyl.

By cyclo(alkyl) is meant a saturated cyclic hydrocarbyl radical. Typically, cyclo(alkyl) groups will comprise from 3 to 10 carbon atoms, more usually 3 to 8 carbon atoms. Examples include cyclopropyl, cyclobutyl, cyclopentyl and cyclohexyl. A common cyclo(alkyl) is cyclohexyl.

By phosphinoyl is meant -P(=O)R ₂ where R is a substituent such as an aryl group or a (cyclo)alkyl group.

By sulfonyl is meant -S(=O) ₂ -R where R is a substituent such a hydrogen or a (cyclo)alkyl group or an aryl group.

By aryl is meant a radical of C ₆ -C ₁₂ hydrocarbyl aromatic group, i.e. where the aromatic group has had a hydrogen atom abstracted. Aryl includes monocyclic and bicyclic aromatic groups. Examples include phenyl and naphthyl. Heteroaryl moieties are aromatic moieties that comprise one or more heteroatoms (that may be the same or different), such as oxygen, nitrogen or sulfur, in place of one or more carbon atoms in a corresponding aryl moiety. Examples of suitable heteroaryl groups include thienyl, furanyl, pyrrolyl, pyridinyl, oxazolyl, thiazolyl, imidazolyl, pyrazolyl, isoxazolyl, isothiazolyl, oxadiazolyl, triazolyl, tetrazolyl, thiadiazolyl etc. Typically, heteroaryl groups are oxadiazolyl and tetrazoyl.

By alkyloxy or alkoxy is meant -OR where R is an alkyl group.

By alkylthio is meant -SR where R is an alkyl group.

By aryloxy is meant -OR, where R is aryl.

By alkylcarboxy is meant -R-COOH, where R is alkylene.

By alkanoyl is meant -C(O)R, where R is alkyl.

By alkylamido is meant -C(O)NR ₂ , where R is hydrogen or alkyl.

By alkylsulfonamido is meant -S(O) ₂ NR ₂ , where R is hydrogen or alkyl.

By alkylcarbamoyl is meant -NHC(O)R, where R is alkyl.

By sulfonato is meant -S(=O) ₂ O-.

By phosphinato is meant -P(=O) ₂ O-.

By halo is meant chloro, fluoro, bromo or iodo.

By haloalkyl is meant an alkyl group substituted with one or more halo substituents.

According to the first aspect of the invention there is provided a compound of formula (I) wherein:

-X- is -O- or -S-; each D is independently an aromatic electron-donating group; each A is independently an electron-accepting group; each of o, p, m and n is independently 0, 1 or 2 with the proviso that both o + p is not 0 and m + n is not 0; and each of the optionally D- and A-substituted aromatic rings is independently optionally substituted with 1 to 3 C ₁ -C ₆ alkyl groups.

The compounds of formula (I) are ssuubbssttiittuutteedd spiro-(fluorene-9,9’- (thio)xanthene)-based compounds. Spiro-(fluorene-9,9’-(thio)xanthene)s denotes herein compounds that are either spiro-(fluorene-9,9’-xanthene)s or spiro-(fluorene-9,9’- thioxanthene)s and thus the term substituted spiro-(fluorene-9,9’-(thio)xanthene)- compounds denotes compounds that are either substituted spiro-(fluorene-9,9’- xanthene) compounds or substituted spiro-(fluorene-9,9’-thioxanthene) compounds.

According to particular embodiments of all aspects and embodiments of the invention, the substituted spiro-(fluorene-9,9’-(thio)xanthene)-based compounds are substituted spiro-(fluorene-9,9’-xanthene)-based compounds, i.e. X in formula (I) is oxygen, however the invention is not limited in this regard.

In formula (I), each of o, p, m and n is independently 0, 1 or 2 with the proviso that both o + p is not 0 and m + n is not 0. In other words, there must be at least one aromatic electron-donating group D and at least one electron-accepting group A in the compounds of the invention. According to particular embodiments of the invention, each of o, p, m and n is independently 1 or 2 and more typically, each of o, p, m and n is 1 . However, each one of these may be 0. For example, according to some embodiments of the invention, the compound of formula (I) comprises one A group and one D group (i.e. one of o and p is 0 and the other is 1 ; and one of m and n is 0 and the other is 1.

According to particular embodiments of the invention, where a compound of formula (I) has one A group and one D group, these groups may be on opposite sides of the substituted spiro-(fluorene-9,9’-(thio)xanthene), by which is meant that either (i) o and n are both 0 and p and m are both 1 or (ii) o and n are both 1 and p and m are both 0.

Typically, but not necessarily, the total number of A and D groups is more than 2, for example there are 2 A groups and 1 D group, 1 A group and 2 D groups, 2 A groups and 2 D groups, or more than 4 A and D groups in total, up to a maximum of 8 of such groups (when each of o, p, m and n is 2).

For ease of reference, the following numbering scheme is adopted throughout herein in relation to spiro-(fluorene-9,9’-(thio)xanthene)-based compounds, denoting the number of aromatic carbon atoms of the spiro-(fluorene-9,9’-(thio)xanthene) system susceptible to substitution (omitting the optional D and A substituents in formula (I) for clarity): Typically, but not necessarily, D and A substituents will be present at the 2' and 7’ positions and 2 and 7 positions respectively. Where present, optional alkyl groups will therefore typically be at other positions.

According to particular embodiments, compounds of formula (I) have substitution patterns selected from the group consisting of two D groups at positions 2’ and 7’ and two A groups at positions 2 and 7; two D groups at positions 2’ and 7’ and one A group at position 2 (or 7, which is equivalent to position 2 where the two D groups are the same and where there is no further substitution of the spiro-(fluorene-9,9’-(thio)xanthene)- (e.g. with alkyl groups); two A groups at positions 2 and 7 and one D group at position 2’ (or 7', which is equivalent to position 2’ where the two A groups are the same and where there is no further substitution of the spiro-(fluorene-9,9’-(thio)xanthene)- (e.g. with alkyl groups); and one A group at position 2 and one D group at position 7’ (or one A group at position 7 and one D group at position 2’, which is equivalent where there is no further substitution of the spiro-(fluorene-9,9’-(thio)xanthene)- (e.g. with alkyl groups).

Typically, where more than one D group is present, D groups will be the same. Likewise, where more than one A group is present, such A groups will typically be the same.

As the skilled person is aware, donor and acceptor moieties are well-known to those acquainted with TADF technology and allow the energy gap between the lowest singlet and triplets states to be small and variable.

Each A group is independently an electron-accepting, i.e. electron withdrawing, group, also referred to herein as an acceptor group, the meaning of acceptor groups being well understood in the field of TADF compounds.

There is no particular limit as to the electron-accepting groups the compounds described herein (e.g. of formula (I)): these may be of the types conventionally employed in TADF molecules.

Acceptor groups may be selected from the group consisting of: cyano (-CN), trifluoromethyl, ketones, esters, amides, aldehydes, sulfones, sulfoxides, phosphine oxides and optionally substituted 1,3,5 triazine, benzonitrile, phthalonitrile and 1 ,3,4 oxadiazole moieties. Other examples of acceptor groups include optionally substituted pyridines, pyrimidines, pyrazines and 1 ,2,4-triazoles. In general, electron-poor heterocycles, for example electron-poor 5- and 6-membered heterocycles, or other aromatic or heteroaromatic groups (e.g. cyano-, trifluoromethyl- fluoro- or nitro- substituted aromatic or heteroaromatic groups, such as phenyl substituted one or more times with these substituents), may be used as acceptor groups. For example the following ketone, ester, amide, aldehyde, sulfone, sulfoxide and phosphine oxide moieties may be used as acceptor groups A, the dotted line representing the bonding position to the remainder of formula (I) (or (la) or (lb)):

In these moieties, each -R ¹ may be, independently for each occurrence, selected from the group consisting of optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties

Heteroaryl groups A may be selected from the following: wherein: the dotted line representing the bonding position to the remainder of formula (I) (or (la) or (lb));

X represents O, S or NR ³ and groups R ³ are, independently for each occurrence, selected from the group consisting of -H and optionally substituted aryl, heteroaryl and C ₁ -io non-aromatic hydrocarbyl moieties; and groups R ² are, independently for each occurrence, selected from the group consisting of -H; optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; and halo, -NO ₂ , amino, hydroxyl, alkoxy, alkylthio, carboxy, cyano, thio, formyl, ester, acyl, thioacyl, amido, sulfonamido, carbamate, -SF ₅ , -CF ₃ , phosphine oxide, phosphine sulphide and the like, wherein amino may be NH ₂ , NHR or NR ₂ , where the substituents R may be C _1-6 alkyl, aryl or heteroaryl.

According to particular embodiments, the A groups in compounds of formulae (I), (la) and (b) may be, or each A in such compounds may be independently selected from:

wherein: the broken line represents the bonding position to the remainder of the compound of formula (I), (la) or (lb); each R ¹ , independently for each occurrence, is selected from the group consisting of optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; each R ² , independently for each occurrence, is selected from the group consisting of -H; optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; and halo, -NO ₂ , amino, hydroxyl, alkoxy, alkylthio, carboxy, cyano, thio, formyl, ester, acyl, thioacyl, amido, sulfonamido, carbamate, -SF ₅ , -CF ₃ , phosphine oxide, phosphine sulphide and the like, wherein amino may be NH ₂ , NHR or NR ₂ , where the substituents R may be C _1-6 alkyl, aryl or heteroaryl. each R ³ , independently for each occurrence, is selected from the group consisting of -H and optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; and each R ⁴ , independently for each occurrence, is independently cyano, trifluoromethyl, fluoro or nitro. Herein, unless a context defines otherwise, a group or groups stated to be optionally substituted may be unsubstituted or substituted one or more times, e.g. unsubstituted or substituted once, with one or more groups independently selected from the group consisting of halo, -NO ₂ , amino, hydroxyl, alkoxy, alkylthio, carboxy, cyano, thio, formyl, ester, acyl, thioacyl, amido, sulfonamido, carbamate, -SF ₅ , -CF ₃ , phosphine oxide, phosphine sulphide and the like. Where a substituent is amino it may be NH ₂ , NHR or NR ₂ , wherein amino may be NH ₂ , NHR’ or N(R’) ₂ , where the substituents R’ may be C _1-6 alkyl, aryl or heteroaryl. Often, amino is NH ₂ .

Typically, where more than one R ¹ , R ² or R ⁴ group is present in an A group, commonly labelled groups are the same. For example, where more than one R ⁴ group is present in an A group such groups are often but not necessarily cyano (nitrile (-CN)).

According to particular embodiments, the or each A is independently a diarylphosphinoyl or arylsulfonyl group (preferably a di C _5- C ₈ arylphosphinoyl and C _5- C ₈ arylsulfonyl, often a diphenylphosphinoyl and phenylsulfonyl group.

According to particular embodiments discussed hereinafter, each A group is diphenylphosphinoyl. However, it will be understood that the exemplification of the invention with this particular A group is illustrative only and the method of the invention (discussed in more detail hereinafter) allows modification of the nature of the A groups in the compounds and allows wide opportunities for fine-tuning and further development.

Each D group is independently an aromatic electron-donating group, also referred to herein as a donor group, the meaning of donor groups (as with the meaning of acceptor groups) being well understood in the field of TADF compounds.

Illustrative D groups in compounds of formula (I) may be, or each D in such compounds may be independently selected from the following moieties:

wherein: the broken line represents the bonding to the remainder of the compound of formula (I);

X ¹ is selected from the group consisting of O, S, NR ⁵ , Si(R ⁵ ) ₂ , PR ⁵ and C(R ⁵ ) ₂ ; each R ⁵ is independently selected from the group consisting of -H; C ₁ -C ₆ alkyl; and C ₅ -C ₈ aryl and C ₃ -C ₇ heteroaryl optionally substituted with any one or a combination selected from the group consisting of C ₁ -C ₆ alkyl, C ₁ -C ₆ fluoroalkyl, C ₁ -C ₆ alkyloxy, C ₁ -C ₆ alkylthio, C ₅ -C ₈ aryl, C ₃ -C ₇ heteroaryl, -SF ₅ , nitro, fluoro, C _5- C ₈ aryloxy, cyano, C ₁ -C ₆ alkylcarboxy, C ₁ -C ₆ alkanoyl, C ₁ -C ₆ alkylamido, C ₁ - C ₆ alkylsulfonamido, C ₁ -C ₆ alkylcarbamoyl, diC ₅ -C ₈ arylphosphinoyl, C _5- C ₈ arylsulfonyl, phosphinato and sulfonato; represents a fused aromatic ring; each Ar and is independently selected from the group consisting of C _5- C ₈ aryl and C ₃ -C ₇ heteroaryl, optionally substituted with any one or a combination selected from the group consisting of C ₁ -C ₆ alkyl, C ₁ -C ₆ fluoroalkyl, C ₁ -C ₆ alkyloxy, C ₁ -C ₆ alkylthio, C _5- C ₈ aryl, C ₃ -C ₇ heteroaryl, -SF ₅ , nitro, fluoro, C _5- C ₈ aryloxy, cyano, C ₁ -C ₆ alkylcarboxy, C ₁ -C ₆ alkanoyl, C ₁ -C ₆ alkylamido, C ₁ -C ₆ alkylsulfonamido, C ₁ -C ₆ alkylcarbamoyl, diC _5- C ₈ arylphosphinoyl, C _5- C ₈ arylsulfonyl, phosphinato and sulfonato; wherein, when the broken line is from a fused aromatic ring, which is benzene, the broken line is positioned para to the nitrogen atom of the ring to which the fused aromatic ring is attached; and

( ) _n indicates the optional presence of saturated -CH ₂ - groups in the rings annelated to the benzene ring, wherein n is independently for each occurrence, 0, 1 , or 2.

According to particular embodiments, the or each D is independently chosen from the following group: wherein the broken line, X ¹ , Ar and R ⁵ are as immediately hereinbefore defined.

According to more particular embodiments, the or each D is independently chosen from the following group:

wherein:

R ⁵ is as immediately hereinbefore defined; and each R ⁶ is independently selected from the group consisting of -H, C ₁ -C ₆ alkyl, C ₁ -C ₆ fluoroalkyl, C ₁ -C ₆ alkyloxy, C ₁ -C ₆ alkylthio, C ₅ -C ₈ aryl, C ₃ -C ₇ heteroaryl, -SF ₅ , nitro, fluoro, C ₅ -C ₈ aryloxy, cyano, C ₁ -C ₆ alkylcarboxy, C ₁ -C ₆ alkanoyl, C ₁ -C ₆ alkylamido, C ₁ - C ₆ alkylsulfonamido, C ₁ -C ₆ alkylcarbamoyl, diC ₅ -C ₈ arylphosphinoyl, C ₅ -C ₈ arylsulfonyl, phosphinato and sulfonato.

According to more specific embodiments, the or each D is chosen from the group consisting of: wherein R ⁶ is as immediately hereinbefore defined.

According to particular embodiments, R ⁶ is independently selected from the group consisting of -H, C ₁ -C ₆ alkyl, and C ₁ -C ₆ fluoroalkyl, such as -H or C ₁ -C ₆ alkyl.

According to particular embodiments discussed hereinafter, the or each D group is chosen from the group consisting of:

However, it will be understood that the exemplification of the invention hereinafter with these particular D group is illustrative only and the method of the invention (discussed in more detail hereinafter) allows modification of the nature of the D groups (as it does the A groups) in the compounds and allows wide opportunities for fine-tuning and further development.

According to more particular embodiments of the invention, the compounds described herein have the following formulae:

According to still more particular illustrative embodiments of the invention, the compounds described herein, may have the following formulae:

It will be understood from the discussion above that these formulae define 5 spiro- (fluorene-9,9’-xanthene) compounds and 5 spiro-(fluorene-9,9’-thioxanthene) compounds. According to specific embodiments, the spiro-(fluorene-9,9’-xanthene) compounds may be mentioned.

The compounds of the invention may exist in different stereoisomeric forms and mixtures thereof, including enantiomers and racemic mixtures, which are all included within the scope of the invention. Such stereoisomeric forms include enantiomers and diastereoisomers. Individual stereoisomers of compounds of the invention, i.e., associated with less than 5%, preferably less than 2% and in particular less than 1% of the other stereoisomer, are included. Mixtures of stereoisomers in any proportion, for example a racemic mixture comprising substantially equal amounts of two enantiomers, are also included within the invention.

Also included are solvates and isotopically-labelled compounds of the invention. Isotopically-labelled compounds are identical to those recited herein, but for the fact that one or more atoms are replaced by an atom having an atomic mass or mass number different from the atomic mass or mass number predominantly found in nature. Examples of isotopes that can be incorporated into compounds of the invention include isotopes of hydrogen, carbon, oxygen and sulfur, such as ² H, ³ H, ¹³ C, ¹⁴ C, ¹⁸ O, ¹⁷ O and ³⁵ S, respectively.

The method of the invention allows access to the compounds described herein. According to this, there is a method of preparing a compound of the invention comprising: (i) halogenating a spiro-(fluorene-9,9’-(thio)xanthene) compound of formula (la) wherein:

-X- is -O- or -S-; each A is independently an electron-accepting group, each of m and n is independently 0, 1 or 2 with the proviso that m + n is not 0; and each of the optionally A-substituted aromatic rings and the other two aromatic rings depicted in formula (la) is independently optionally substituted with 1 to 3 C ₁ -C ₆ alkyl groups, to obtain a compound of formula (lb) wherein: each Y is independently a halo group; m and n are the same as the compound of formula (la) from which a compound of formula (lb) is obtained; each of o and p is independently 0, 1 or 2 with the proviso that o + p is not 0; and each of the optionally Y- and A-substituted aromatic rings is independently optionally substituted with one to three C ₁ -C ₆ alkyl groups; and

(ii) substituting each halo group Y on the compound of formula (lb) with an electron- donating group D.

The particular consideration and embodiments described in relation to the first aspect of the invention apply mutatis mutandis to the method of the invention, in particular in relation to the nature of X (according to particular embodiments being O), the number m and n of the A groups and the typical positions on the six-membered rings of the fluorene moiety at which these are located. In this regard it will be understood that the number o and p of the Y (halo) groups corresponds to the number o and p of the D groups prepared from compounds of formula (lb) in the substituting reaction. Furthermore, the embodiments relating to the nature of D (including the number o and p of the D groups and the typical positions on the six-membered rings of the fluorene moiety at which these are located) described in the first aspect of the invention apply mutatis mutandis to the nature of D in step (ii) of the second aspect of the invention.

As noted above, the compounds provided in accordance with the present invention are accessible by harnessing the electronic separation of the two halves of spiro-(fluorene-9,9’-(thio)xanthene). In this regard, the compounds of formula (la) described herein may be accessed by chemistry conducted on appropriate spiro- (fluorene-9,9’-(thio)xanthene)-based compounds. For example, precursor compounds to those of formula (la) having bromine substituents (and the optional alkyl groups present in the compounds of formula (la)) corresponding to the number and position of the A groups in the compounds of formula (la) (and (lb) and then (I)) may be used to prepare compounds of formula (la).

For example, according to particular embodiments the compound 2,7- dibromospiro[fluorene-9,9'-xanthene] (abbreviated SFX-Br) may be prepared in accordance with a modified protocol (Xie, L.-H.; Liu, F.; Tang, C.; Hou, X.-Y.; Hua, Y.- R.; Fan, Q.-L.; Huang, W., Unexpected One-Pot Method to Synthesize Spiro[fluorene- 9,9‘-xanthene] Building Blocks for Blue-Light-Emitting Materials. Org. Lett. 2006, 8, 2787-2790) reported for the large-scale synthesis of spiro[fluorene-9,9'-xanthene] (Maciejczyk, M.; Ivaturi, A.; Robertson, N., SFX as a low-cost ‘Spiro’ hole-transport material for efficient perovskite solar cells. J. Mater. Chem. A 2016, 4, 4855-4863) involving the reaction between 2,7-dibromofluorenone, phenol and methane sulfonic acid:

It will be understood, as indeed is reported by Xie, L.-H. et al. (supra) that modification of the starting fluorenone will allow access to differently substituted dibromospiro[fluorene-9,9’-xanthenes, for example use of 2-dibromofluorenone allows access to 2-bromospiro[fluorene-9,9’-xanthene]. Other brominated fluorenones are known in the academic and patent literature and/or are commercially available (for example a synthesis of 3-bromospiro[fluorene-9,9’-xanthene] is described in WO 2017/105041 A1 (Doosan Corporation) from the corresponding reaction of 3- bromofluorenone prepared from permanganate oxidation of the corresponding 9,10- phenanthrenequinone).

Access to the corresponding spiro[fluorene-9,9’-thioxanthenes is likewise possible for the skilled person, for example through use of thiophenol in place of phenol in the syntheses described above, with other synthetic possibilities within the routine ability of the skilled person.

Access to suitable spiro[fluorene-9,9’-(thio)xanthene derivatives, typically but not necessarily functionalised with bromine on the fluorene half of the molecule, permits access to the compounds of formula (la) described herein. For example, access to compounds of formula (la) comprising one or more diphenylphosphine oxide (diphenylphosphinoyl) W groups may be achieved using chemistry of the type described by J Zhao et al. (Zhao, J.; Xie, G.-H.; Yin, C.-R.; Xie, L.-H.; Han, C.-M.; Chen, R.-F.;

Xu, H.; Yi, M.-D.; Deng, Z.-P.; Chen, S.-F.; Zhao, Y.; Liu, S.-Y.; Huang, W.,

Harmonizing Triplet Level and Ambipolar Characteristics of Wide-Gap Phosphine Oxide Hosts toward Highly Efficient and Low Driving Voltage Blue and Green PHOLEDs: An Effective Strategy Based on Spiro-Systems. Chemistry of Materials 2011, 23 (24), 5331- 5339). In this publication, treatment of SFX-Br with ⁿ BuLi followed by treatment with chlorodiphenylphosphine affords an intermediate crude product oxidised with aqueous hydrogen peroxide to forward the desired bis diphenylphosphinoyl compound.

Whilst the exemplification of the present invention is demonstrated with diphenylphosphine oxide as the acceptor moiety, the skilled person will understand how to construct other acceptor- (i.e. A group-containing) spiro[fluorene-9,9’-(thio)xanthene derivatives. For example, for such syntheses and others useful in connection with the present invention, the skilled person may refer to standard texts concerning synthetic organic chemistry, for example those illustrative of means to build or attach groups to a ring system, well known to him or her (for example Name reactions in heterocyclic chemistry (2005), Jie Jack Li, editor; Wiley; and Strategic Applications of Organic Named Reactions in Organic Synthesis (2005), Laslo Kurti and Barbara Czako; Academic Press.

In accordance with the method of the invention, an A-containing compound of formula (la) is halogenated to obtain a compound of formula (lb), having one or more halogen atoms Y on the benzene rings of the (thio)xanthene system. As noted above, a particular feature of the present invention is that selective substitution of these benzene rings (as opposed to those of the fluorene system or those present in the fluorene system) is possible owing to electronic separation of the fluorene and xanthene π - systems.

Advantageously, electrophilic aromatic halogenation may be effected to introduce the halogen atoms Y present in the compounds of formula (lb). Electrophilic aromatic substitution is well known to the skilled person, being fully described in relevant text with which (s)he is familiar (see, for example M. B. Smith, March’s Advanced Organic Chemistry: Reactions, Mechanisms and Structure, (8th ed. 2013); Carey and Sundberg, Advanced Organic Chemistry, Sth ed. (2007)).

According to some embodiments, step (i) of the method of the invention involves a the compound of formula (la) with a source of halo groups Y, wherein the source preferably comprises a source of bromo, chloro or iodo groups, typically chloro or bromo groups, most typically bromo groups. According to particular embodiments, the source may be N-bromosuccinimide, bromine, N-iodosuccinimide, iodine, N-chlorosuccinimide, chlorine, 1-chloromethyl-4-fluoro-1,4-diazoniabicyclo[2.2.2]octane bis(tetrafluoroborate), N-fluoro-o-benzenedisulfonimide, N-fluorobenzenesulfonimide and fluorine.

According to a specific embodiment, the source is a source of bromo groups, which is N-bromosuccinimide. Conveniently, the reaction between this reagent and A- containing spiro-(fluorene-9,9’-(thio)xanthenes) was found to be effective when conducted in acidic solvents, for example, carboxylic acid solvents, such as acetic acid, in particular glacial acetic acid at elevated temperature, for example between about 100 °C and its boiling point (118 °C), for example about 110 °C. However, the skilled person will be aware, for example by consulting M. B. Smith (supra) other means of effecting halogenation, in particular bromination, and appropriate conditions for such reactions, for example the stoichiometry of such reactions, use of Lewis acids, extent of heating, duration of reaction etc. According to particular embodiments of the invention, halo groups Y are introduced at the 2’ and 7‘ positions of the spiro-(fluorene-9,9’-(thio)xanthenes) of formula (la) is to provide spiro-(fluorene-9,9’-(thio)xanthenes) of formula (lb).

Step (ii) of the method of the invention substitutes each of the one or more halo groups Y in the compounds of formula (lb) with an electron-donating group D, to provide a compound of formula (I). The skilled person is aware of a multitude of suitable coupling reactions for introducing D groups, typically catalysed by transition metal catalysts, in particular those containing palladium. The skilled person will therefore be able to select an appropriate reaction for the particular combination of D group and compound of formula (lb), for example the nature of halo groups Y in the latter. Detailed reviews of palladium-based couplings are available to the skilled person in this regard, for example A Biffis et al., Pd Metal Catalysts for Cross-Couplings and Related Reactions in the 21st Century: A Critical Review, Chem. Rev., 118(4), 2249-2295 (2018).

Conveniently, where the D groups are connected to the remainder of the compound of formula (I) through a nitrogen atom, Buchwald-Hartwig coupling between bromine-containing compounds of formula (lb) and sources of such D groups was found to be convenient. Use of cesium carbonate as a base, Pd ₂ (dba) ₃ (tris(dibenzylideneacetone)dipalladium(0)) catalyst and P( ^t Bu) ₃ as ligand convenient but other reagents and reactions will present themselves to the skilled person.

The compounds of the first aspect of the invention can be employed as light- emitting materials and thus useful in affording light-emitting devices (i.e. devices capable of, or suitable for, emitting light) in accordance with the third aspect of the invention. Such electroluminescent devices may be OLEDs or light-emitting electrochemical cells (LEECs), which may comprise one or more of the compounds of the first aspect of the invention.

The devices of the invention comprises an emissive layer (also referred to as light-emitting layer), the emissive layer comprising one or more compounds of the first aspect of the invention. The emissive layer may comprise other materials such as host materials. In one embodiment, the emissive layer is non-doped, i.e. it contains no host materials. In this embodiment, the emissive layer can consist essentially of one or more compounds of the first aspect of the invention, i.e. the device of the invention can comprise such an emissive layer.

In another aspect, the invention provides a method of preparing an electroluminescent device such as an OLED comprising depositing an emissive layer on a substrate using a solution processing technique, wherein said emissive layer comprises one or more compounds according to the first aspect of the invention. The basic structure of an electroluminescent device such as an OLED is a thin film of organic material sandwiched between two electrodes. The electroluminescent device may be an OLED which comprises an anode, a cathode with the emissive layer situated between the anode and the cathode. For bottom emitting OLEDs, the anode, which is commonly Indium tin oxide (ITO), generally rests on top of a support such as glass but, metal foils and plastic substrates (such as polyethylene terephthalate (PET) or polyethylene naphthalate (PEN)) ccaann be used. Poly(3,4- ethylenedixoythiophene): polystyrene sulfonate (PEDOT:PSS) is often deposited on top of the ITO anode as the hole injection layer the skilled person will be aware that other materials can be used. For example, in the device exemplified below, N,N'- bis(naphthalen-1-yl)- N,N'-bis(phenyl)benzidine (NPB) was used, followed by a layer of tris(4-carbazoyl-9-ylphenyl)amine (TCTA), which acts as a hole transporting layer, and a layer of 1,3-bis(N-carbazolyl)benzene (mCP), which acts as an electron blocker.

The next layer is an emissive layer and contains the emitter material which can be doped in a host material. In the device exemplified below, the emissive layer is made up of both mCP and the emitter material. A layer of Bis[2- (diphenylphosphino)phenyl]ether oxide (DPEPO) was used as a hole blocker and a layer of 1,3,5-Tri(m-pyridin-3-ylphenyl)benzene (TmPyPB) was used as an electron transporting layer to transport electrons to the cathode (the final layer). The cathode is reflective and is made of metal such as aluminum, gold or silver, often combined with a very thin layer of lithium fluoride (LiF) or calcium to enhance the electron injection. The thickness of the organic layers in the basic structure is unusually between 100 and 150 nm.

For bottom emitting OLEDs, the light emitting layer is commonly deposited on the hole injection layer (such as NPB in Fig. 7) and then, typically, exciton blocking layer, electron transporting layer, electron injection layer and cathode are deposited subsequently, i.e. a hole transporting layer (such as TCTA in Fig. 7) need not be used. For top emitting OLEDs, the light emitting layer is commonly deposited on the electron transporting layer, and then, typically, the hole transporting layer, hole injection layer, and anode are deposited subsequently, i.e. a hole or electron blocking layer (such as DPEPO and mCP in Fig. 7) need not be used.

The compounds of the invention are TADF emitters, thus use of the compounds of the invention as emitter materials and/or as TADF materials is included.

In a further aspect, the invention provides a compound of formula (lb) wherein:

-X- is -O- or -S-; each A is independently an electron-accepting group, each of m and n is independently 0, 1 or 2 with the proviso that m + n is not 0; each Y is independently a halo group; each of o and p is independently 0, 1 or 2 with the proviso that o + p is not 0; and each of the optionally Y- and A-substituted aromatic rings is independently optionally substituted with one to three C ₁ -C ₆ alkyl groups.

For the avoidance of doubt, the particular consideration and embodiments described in relation to the first aspect of the invention apply mutatis mutandis to the a compounds of formula (lb), in particular in relation to the nature of X (according to particular embodiments being O), the number m and n of the A groups and the typical positions on the six-membered rings of the fluorene moiety at which these are located. In this regard it will be understood that the number o and p of the Y (halo) groups corresponds to the number o and p of the D groups prepared from compounds of formula (lb) in the substituting reaction. Thus, the embodiments relating to the number o and p of the D groups and the typical positions on the six-membered rings of the fluorene moiety at which these are located described in the first aspect of the invention apply mutatis mutandis to the nature of Y.

The invention may be further understood with reference to the following non-limiting numbered clauses:

1. A compound of formula (I) wherein:

-X- is -O- or -S-; each D is independently an aromatic electron-donating group; each A is independently an electron-accepting group; each of o, p, m and n is independently 0, 1 or 2 with the proviso that both o + p is not 0 and m + n is not 0; and each of the optionally D- and A-substituted aromatic rings is independently optionally substituted with 1 to 3 C ₁ -C ₆ alkyl groups.

2. The compound of clause 1, wherein each of m, n, o and p is 1.

3. The compound of clause 1 or 2, wherein the D groups are at positions 2’ and 7’ and the A groups are at positions 2 and 7.

4. The compound of any one of clauses 1 to 3, wherein each A is independently chosen from the group consisting of: cyano, trifluoromethyl, ketones, esters, amides, aldehydes, sulfones, sulfoxides, phosphine oxides and heteroaryl groups selected from: wherein: the dotted line represents the bonding position to the remainder of formula (I), X represents O, S or NR ³ and groups R ³ are, independently for each occurrence, selected from the group consisting of -H and optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; and groups R ² are, independently for each occurrence, selected from the group consisting of -H; optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; and halo, -NO ₂ , amino, hydroxyl, alkoxy, alkylthio, carboxy, cyano, thio, formyl, ester, acyl, thioacyl, amido, sulfonamido, carbamate, -SF ₅ , -CF ₃ , phosphine oxide, phosphine sulphide and the like, wherein amino may be NH ₂ , NHR or NR ₂ , where the substituents R may be C _1-6 alkyl, aryl or heteroaryl.

5. The compound of any one of clauses 1 to 3, wherein each A is independently chosen from: wherein: the broken line represents the bonding position to the remainder of the compound of formula (I); each R ¹ , independently for each occurrence, is selected from the group consisting of optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; each R ² , independently for each occurrence, is selected from the group consisting of -H; optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; and halo, -NO ₂ , amino, hydroxyl, alkoxy, alkylthio, carboxy, cyano, thio, formyl, ester, acyl, thioacyl, amido, sulfonamido, carbamate, -SF ₅ , -CF ₃ , phosphine oxide, phosphine sulphide and the like, wherein amino may be NH ₂ , NHR or NR ₂ , where the substituents R may be C _1-6 alkyl, aryl or heteroaryl. each R ³ , independently for each occurrence, is selected from the group consisting of -H and optionally substituted aryl, heteroaryl and C _1-10 non-aromatic hydrocarbyl moieties; and each R ⁴ , independently for each occurrence, is independently cyano, trifluoromethyl, fluoro or nitro.

6. The compound of any one of clauses 1 to 5, wherein each A is independently chosen from a diarylphosphinoyl or arylsulfonyl group.

7. The compound of any one of clauses 1 to 5, wherein A is independently chosen from a diC ₅ -C ₈ arylphosphinoyl and C ₅ -C ₈ arylsulfonyl.

8. The compound of any one of clauses 1 to 5, wherein A is independently chosen from diphenylphosphinoyl and phenylsulfonyl.

9. The compound of any one of clauses 1 to 5, wherein A is diphenylphosphinoyl.

10. The compound of any one of the preceding clauses, wherein each D is independently chosen from:

wherein: the broken line represents the bonding position to the compound;

X ¹ is selected from the group consisting of O, S, NR ⁵ , Si(R ⁵ ) ₂ , PR ⁵ and C(R ⁵ ) ₂ ; each R ⁵ is independently selected from the group consisting of -H; C ₁ -C ₆ alkyl; and C ₅ -C ₈ aryl and C ₃ -C ₇ heteroaryl optionally substituted with any one or a combination selected from the group consisting of C ₁ -C ₆ alkyl, C ₁ -C ₆ fluoroalkyl, C ₁ -C ₆ alkyloxy, C ₁ -C ₆ alkylthio, C ₅ -C ₈ aryl, C ₃ -C ₇ heteroaryl, -SF ₅ , nitro, fluoro, C ₅ - C ₈ aryloxy, cyano, C ₁ -C ₆ alkylcarboxy, C ₁ -C ₆ alkanoyl, C ₁ -C ₆ alkylamido, C1- C ₆ alkylsulfonamido, C ₁ -C ₆ alkylcarbamoyl, diC _5- C ₈ arylphosphinoyl, C ₅ - C ₈ arylsulfonyl, phosphinato and sulfonato; represents a fused aromatic ring; each Ar and is independently selected from the group consisting of C ₅ - C ₈ aryl and C ₃ -C ₇ heteroaryl, optionally substituted with any one or a combination selected from the group consisting of C ₁ -C ₆ alkyl, C ₁ -C ₆ fluoroalkyl, C ₁ -C ₆ alkyloxy, C ₁ -C ₆ alkylthio, C _5- C ₈ aryl, C ₃ -C ₇ heteroaryl, -SF ₅ , nitro, fluoro, C _5- C ₈ aryloxy, cyano, C ₁ -C ₆ alkylcarboxy, C ₁ -C ₆ alkanoyl, C ₁ -C ₆ alkylamido, C ₁ -C ₆ alkylsulfonamido, C ₁ -C ₆ alkylcarbamoyl, diC ₅ - C ₈ arylphosphinoyl, C _5- C ₈ arylsulfonyl, phosphinato and sulfonato; wherein, when the broken line is from a fused aromatic ring, which is benzene, the broken line is positioned para to the nitrogen atom of the ring to which the fused aromatic ring is attached; and

( ) _n indicates the optional presence of saturated -CH ₂ - groups in the rings annelated to the benzene ring, wherein n is independently for each occurrence, 0, 1, or 2. 12. The compound of clause 10, wherein each D is independently chosen from the following group: wherein: each R ⁶ is independently selected from the group consisting of -H, C ₁ -C ₆ alkyl, C ₁ -C ₆ fluoroalkyl, C ₁ -C ₆ alkyloxy, C ₁ -C ₆ alkylthio, C _5- C ₈ aryl, C ₃ -C ₇ heteroaryl, -SF ₅ , nitro, fluoro, C _5- C ₈ aryloxy, cyano, C ₁ -C ₆ alkylcarboxy, C ₁ -C ₆ alkanoyl, C ₁ -C ₆ alkylamido, C ₁ - C ₆ alkylsulfonamido, C ₁ -C ₆ alkylcarbamoyl, diC _5- C ₈ arylphosphinoyl, C _5- C ₈ arylsulfonyl, phosphinato and sulfonato.

12. The compound of clause 11 , wherein each D is chosen from the group consisting of:

13. The compound of clause 12, wherein each D is independently chosen from the group consisting of:

14. The compound of any one of the preceding clauses chosen from the group consisting of (Va) to (Vr):

15. The compound of any one of the preceding clauses chosen from the group consisting of (Va to Ve):

16. The compound of any one of the preceding clauses, wherein X is O.

17. A method of preparing a compound according to any one of the preceding claims comprising:

(i) halogenating a spiro-(fluorene-9,9’-(thio)xanthene) compound of formula (la) wherein:

-X- is -O- or -S-; each A is independently an electron-accepting group, each of m and n is independently 0, 1 or 2 with the proviso that m + n is not 0; and each of the optionally A-substituted aromatic rings and the other two aromatic rings depicted in formula (la) is independently optionally substituted with 1 to 3 C ₁ -C ₆ alkyl groups, to obtain a compound of formula (lb) wherein: each Y is independently a halo group; m and n are the same as the compound of formula (la) from which a compound of formula (lb) is obtained; each of o and p is independently 0, 1 or 2 with the proviso that o + p is not 0; and each of the optionally Y- and A-substituted aromatic rings is independently optionally substituted with one to three C ₁ -C ₆ alkyl groups; and

(ii) substituting each halo group Y on the compound of formula (lb) with an electron- donating group D.

18. The method of clause 17, wherein step (i) involves contacting the compound of formula (la) with a source of halo groups Y. 19. The method of clause 17, wherein step (i) involves contacting the compound of formula (la) with N-bromosuccinimide, bromine, N-iodosuccinimide, iodine, N- chlorosuccinimide, chlorine, 1-chloromethyl-4-fluoro-1 ,4-diazoniabicyclo[2.2.2]octane bis(tetrafluoroborate), N-fluoro-o-benzenedisulfonimide, N-fluorobenzenesulfonimide and fluorine.

20. The method of clause 17, wherein step (i) involves contacting the compound of formula (la) with a source of bromo groups.

21. The method of clause 17, wherein step (i) involves contacting the compound of formula (la) with N-bromosuccinimide.

22. The method of any one of clauses 17 to 21, wherein step (ii) involves reacting the compound of formula (lb) to introduce the one or more electron-donating groups D in the presence of a transition metal (preferably palladium).

23. The method of any one of clauses 17 to 21 wherein A is as described in any one of clauses 4 to 8.

24. The method of any one of clauses 17 to 23, wherein D is as described in any one of clauses 9 to 12.

25. The method of any one of clauses 17 to 24, comprising a step of preparing a compound of formula (la) prior to step (i).

26. A compound of formula (lb) wherein:

-X- is -O- or -S-; each A is independently an electron-accepting group, each of m and n is independently 0, 1 or 2 with the proviso that m + n is not 0; each Y is independently a halo group; each of o and p is independently 0, 1 or 2 with the proviso that o + p is not 0; and each of the optionally Y- and A-substituted aromatic rings is independently optionally substituted with one to three C ₁ -C ₆ alkyl groups.

27. The compound of clause 26, wherein A is as described in any one of clauses 4 to 8.

28. A light-emitting device comprising any one or a combination of compounds of any one of clauses 1 to 16.

29. The light-emitting device of clause 28, wherein said light emitting device is a light- emitting diode or a light-emitting electrochemical cell.

30. Use of any one or a combination of compounds of any one of clauses 1 to 15 as a TADF compound.

Each and every patent and non-patent reference referred to herein is hereby incorporated by reference in its entirety, as if the entire content of each reference were set forth herein in its entirety.

The invention is further described by way of the following non-limiting examples,

EXEMPLIFICATION

SYNTHESIS

General methods

Elemental analyses were carried out by Stephen Boyer at London Metropolitan University using a Carlo Erba CE1108 Elemental Analyser. NMR spectroscopy was conducted using Bruker (400, 500 or 600 MHz) spectrometers. ¹ H, ¹³ C and ³¹ P NMR spectra were all recorded in deuterated chloroform. Chemical shifts are reported in parts per million. Chemical shifts multiplicities are reported as s: singlet, d: doublet, t: triplet, q: quartet, quint: quintet, and m: multiplet. Mass spectra were recorded with Xevo QTOF (Waters) high resolution, accurate mass tandem mass spectrometer equipped with Atmospheric Solids Analysis Probe (ASAP) and Bruker MicroToF 2. All spectra were recorded using electrospray (ESI) or fast atom bombardment (FAB) ionisation. Synthesis

All preparations were carried out using standard Schlenk line and air-sensitive chemistry techniques under nitrogen atmosphere. The starting material, 2,7- dibromospiro[fluorene-9,9'-xanthene] (SFXBr) was prepared according to modified recipe (Xie, L.-H. et al. (supra)) reported for large scale synthesis of spiro[fluorene-9,9'- xanthene] (Maciejczyk, M et al. (supra)). Spiro[fluorene-9,9’-xanthene]-2,7- diylbisfdiphenylphosphine oxide) (SFX-PO) was synthesized by the protocol reported by J Zhao et al. (supra). In this ((Xie, L.-H. et al. (supra)) described heating at 150 °C a mixture of 2,7-dibromofluorenone (6.523 g, 19.3 mmol, 1 equiv), phenol (18 g, 191.4 mmol, 10 equiv), and methane sulfonic acid (MeSO3H, d =1.48 g/ml, 5.0 ml, 7.41 g, 77.2 mmol, 4 equiv) under nitrogen for 24 h before adding the reaction mixture slowly into water (50 ml) and extracting with dichloromethane, drying the combined extracts over MgSO ₄ , evaporating, and purified by column chromatography (hexane/ethyl acetate, 10:1) to afford colorless solid (7.346 g).

The selective electrophilic aromatic dibromination of the xanthene ring of the SFX-PO scaffold was achieved by utilization of N-bromosuccinimide (NBS) in glacial acetic acid at elevated temperature (110 °C) in a good yield of 72%. The successful conditions left the deactivated fluorene and diphenylphosphine oxide rings untouched by taking advantage of electronic separation from the more reactive xanthene half, even in the presence of excess (4 equiv.) NBS. The total yield after three steps for this SFX-PO- Br intermediate was 53%.

The emitters were obtained by Buchwald-Hartwig coupling of SFX-PO-Br and three different diphenylamines - diphenylamine, 4,4'-dimethyldiphenylamine and 4,4'- dimethoxydiphenylamine - using cesium carbonate as a base with Pd ₂ (dba) ₃ as the catalyst and P(t-Bu) ₃ as the ligand. This successfully yielded the desired products with moderate yields of 45%, 53%, 44% for SFX-PO-DPA, SFX-PO- DPA- Me and SFX-PO- DPA-OMe, respectively. As has been previously reported, the use of a weak base like cesium carbonate is advantageous in the presence of base-sensitive functional groups (Sadighi, J. P.; Harris, M. C.; Buchwald, S. L., A highly active palladium catalyst system for the arylation of anilines. Tetrahedron Lett. 1998, 39, 5327-5330.).

All other materials were purchased from commercial suppliers and used without further purification. Toluene was dried using a solvent purification system. Column chromatography was carried out by Dry Column Vacuum Chromatography (DCVC) (Pedersen, D. S.; C., R., Dry Column Vacuum Chromatography. Synthesis 2001, (16), 2431) with dry Silica 60A (particle size 6-35 pm, Davisil) or Silica gel 60 (particle size 15- 40 pm, Merck) as the stationary phase. TLC was performed on pre-coated silica gel plates (0.25 mm thick, 60 F254, Merck, Germany) and observed under UV light.

The synthetic procedures for the preparation of the SFX-based emitters exemplified are presented in Scheme 1 :

Scheme 1 : Synthetic routes for SFX-PO-Br-starting material and three studied emitters: SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA-OMe.

2', 7’-Dibromo-spiro[fluorene-9,9’-xanthene]-2, 7-diylbis(diphenylphosphine oxide (SFX-PO-Br)

SFX-PO (0.700 g, 0.955 mmol) and N-bromosuccinimide (0.680 g, 3.821 mmol, 4 equiv.) were dissolved in 150 mL of glacial acetic acid and heated at 110 °C overnight. After that, TLC analyses indicated no starting material and it was poured into water and extracted with DCM. Dried over magnesium sulfate, filtrated and concentrated on rotavap with Celite. Placed on DCVC column and eluted with EtOAc/MeOH 100/1, starting with pure EtOAc. The product was recrystallized from acetone to give 0.614 g (72%) of a white powder.

¹ H NMR (601 MHz, Chloroform-d) δ 7.92 (ddd, J = 7.8, 2.4, 0.7 Hz, 2H), 7.71 - 7.49 (m, 16H), 7.44 (ddd, J = 8.8, 5.4, 2.3 Hz, 8H), 7.34 - 7.24 (m, 2H), 7.04 (d, J = 8.8 Hz, 2H), 6.43 (d, J = 2.3 Hz, 2H). ¹³ C NMR (151 MHz, CDCb) δ 54.23, 115.74, 119.26,

120.97, 121.06, 124.60, 128.52, 128.61, 129.75, 129.81, 129.87, 131.81, 131.89,

131.98, 132.05, 132.09, 132.11, 132.50, 132.64, 132.71, 133.75, 134.43, 141.82, 141.84, 150.16, 154.41, 154.50. Spiro[fluorene-9,9’-xanthene]-2, 7-diylbis(diphenylphosphine oxide)-2’, 7’-diylbis (N,N-di (4-methyldiphenylamine)) (SFX-PO-DPA)

SFX-PO-Br (0.500 g, 0.561 mmol, 1 equiv.), diphenylamine (0.285 g, 1.684 mmol, 3 equiv.), Pd ₂ (dba) ₃ (0.031 g, 0.034 mmol, 0.06 equiv.) and Cs ₂ CO ₃ (0.549 g, 1.684 mmol, 3 equiv.) has been dried under vacuum for 20 minutes. Then toluene (20 mL) and P(t-Bu) ₃ 1M (0.11 mL, 0.112 mmol, 0.2 equiv.) has been added and it was heated at 110 °C overnight. After overnight stirring it was brought to room temperature and filtered through Celite and washed with EtOAc. Solvent was removed on rotary evaporator after the addition of Celite. Obtained solid was placed on 3 cm long, 1 cm dia. DCVC column and eluted with DCM/EtOAc solvent system. Parameters of gradient elution: 15 mL fractions with 0.5 mL increase of EtOAc from 0% to 25% contents. Pure product has been obtained as a yellow powder 0.267 g (45%).

¹ H NMR (601 MHz, Chloroform-d) δ 6.11 (d, J= 2.6 Hz, 2H), 6.73-6.79 (m, 8H), 6.79 - 6.87 (m, 6H), 6.96 - 7.02 (m, 8H), 7.06 (d, J = 8.8 Hz, 2H), 7.44 (td, J = 7.8, 3.0 Hz, 8H), 7.50 (ddd, J = 11.6, 7.9, 1.4 Hz, 2H), 7.53 - 7.57 (m, 4H), 7.61 (ddt, J = 12.0, 6.8, 1.4 Hz, 8H), 7.68 (ddd, J = 7.8, 2.5, 0.7 Hz, 2H), 7.83 (ddd, J = 11.7, 1.3, 0.7 Hz, 2H). ¹³ C NMR (151 MHz, CDCb) δ 54.60, 118.36, 120.61, 120.69, 122.06, 122.19, 122.82, 123.71, 126.00, 128.53, 128.61, 128.94, 129.37, 129.43, 131.83, 131.92, 131.99, 132.02, 132.36, 132.90, 133.05, 133.59, 141.76, 141.77, 142.47, 147.07, 147.32, 156.43, 156.51. ³¹ P NMR (162 MHz, Chloroform-d) δ 28.38 - 28.26 (m). MS (ESI): m/z (%)= 1067.3 [(M)+, 100], Elemental analysis (%). Calculated for C ₇₃ H ₅₂ N ₂ O ₃ P ₂ : C 82.16, H 4.91, N 2.63. Result: C 81.96, H 5.05, N 2.65. Spiro[fluorene-9,9’-xanthene]-2,7-diylbis(diphenylphosphin eoxide)-2’,7’-diylbis(N,N-

4,4’-dimethyl-diphenylamine (SFX-PO-DPA-Me)

SFX-PO-Br (0.600 g, 0.674 mmol, 1 equiv.), 4,4'-dimethyldiphenylamine (0.399 g, 2.021 mmol, 3 equiv.), Pd ₂ (dba) ₃ (0.037 g, 0.040 mmol, 0.06 equiv.) and Cs ₂ CO ₃ (0.659 g, 2.021 mmol, 3 equiv.) has been dried under vacuum for 20 minutes. Then toluene (15 mL) and P(t-Bu) ₃ 1M (0.13 mL, 0.135 mmol, 0.2 equiv.) has been added and it was heated at 110 °C overnight. It turns dark green at higher temperature. After overnight stirring it was brought to room temperature and filtered through Celite and washed with 70 mL of EtOAc. Solvent was removed on rotary evaporator after the addition of Celite. Obtained solid was placed on 5cm long, 3 cm dia. DCVC column and eluted with Hexane/EtOAc solvent system. Parameters of gradient elution: 20 mL fractions with 0.5 mL increase of EtOAc from 0% to 20% contents. Pure product has been obtained as a yellowish powder 0.404 g (53%).

¹ H NMR (400 MHz, Chloroform-d) δ 2.18 (s, 12 H, CH ₃ ), 6.06 (d, J = 2.6 Hz, 2H, Ar-H), 6.58 - 6.68 (m, 8H, Ar-H), 6.70 - 6.81 (m, 10H, Ar-H), 6.98 (d, J = 8.9 Hz, 2H, Ar- H), 7.42 (tdd, J = 8.2, 2.9, 1.3 Hz, 8H, Ar-H), 7.47 - 7.55 (m, 6H, Ar-H), 7.56 - 7.69 (m, 10H, Ar-H), 7.73 - 7.83 (m, 2H, Ar-H). ¹³ C NMR (126 MHz, CDCb) δ 20.68, 54.62, 118.13, 120.59, 120.70, 121.85, 122.79, 122.92, 125.19, 128.52, 128.61, 129.37, 129.45, 129.51, 131.41, 131.71, 131.80, 131.93, 131.98, 132.01, 132.35, 132.72, 133.18, 133.54, 141.80, 141.82, 142.72, 145.05, 146.59, 156.57, 156.67. ³¹ P NMR (162 MHz, Chloroform-d) δ 28.37 - 28.52 (m). MS (ESI): m/z (%)= 1124.4 [(M)+1, 60], Elemental analysis (%). Calculated for C ₇₇ H ₆₀ N ₂ O ₃ P ₂ : C 82.33, H 5.38, N 2.49. Result: C 82.05, H 5.32, N 2.57. Spiro[fluorene-9,9’-xanthene]-2, 7-diylbis(diphenylphosphine oxide)-2’, 7’-diylbis (N,N-di (4-methoxdiphenylamine) ) ( SFX-PO-DPA-OMe)

SFX-PO-Br (0.150 g, 0.168 mmol, 1 equiv.), 4,4'-dimethoxydiphenylamine (0.116 g, 0.505 mmol, 3 equiv.), Pd ₂ (dba) ₃ (0.009 g, 0.010 mmol, 0.06 equiv.) and Cs ₂ CO ₃ (0.165 g, 0.505 mmol, 3 equiv.) has been dried under vacuum for 20 minutes. Then toluene (5 mL) and P(t-Bu) ₃ 1M (0.034 mL, 0.034 mmol, 0.2 equiv.) has been added and it was heated at 110 °C overnight. It turns green at higher temperature. After overnight stirring it was brought to room temperature and filtered through Celite and washed with 20 mL of EtOAc. Solvent was removed on rotary evaporator after the addition of Celite. Obtained solid was placed on 3 cm long, 1 cm dia. DCVC column and eluted with Hexane/AcOEt solvent system. Parameters of gradient elution: 6 mL fractions with 1 mL increase of AcOEt from 0% to 100% contents. Pure product has been obtained as a yellow powder 0.087 g (44%).

¹ H NMR (601 MHz, Chloroform-d) δ 3.70 (s, 12H), 5.96 (d, J = 2.7 Hz, 2H), 6.54 - 6.61 (m, 8H), 6.66 - 6.72 (m, 8H), 6.73 (dd, J = 8.9, 2.7 Hz, 2H), 6.97 (d, J = 8.9 Hz, 2H), 7.44 (td, J = 7.8, 2.9 Hz, 8H), 7.49 - 7.57 (m, 6H), 7.61 (m, 8H), 7.65 - 7.70 (m, 2H), 7.75 (ddd, J = 11.6, 1.4, 0.7 Hz, 2H). ¹³ C NMR (151 MHz, CDCb) δ 25.38, 55.49, 114.36, 117.91, 120.52, 120.60, 120.95, 121.77, 123.55, 124.70, 128.52, 128.60, 129.40, 129.46, 131.66, 131.73, 131.91, 131.98, 132.00, 132.01, 132.39, 132.76, 133.08, 133.44, 141.05, 141.79, 141.80, 143.35, 146.01, 154.97, 156.56, 156.65. ³¹ P NMR (162 MHz, Chloroform-d) δ 28.58 - 28.43 (m). MS (FAB): m/z (%)= 1187.4 [(M)+, 100], Elemental analysis (%) Calculated for C ₇₇ H ₆₀ N ₂ O ₇ P ₂ : C 77.90, H 5.09, N 2.36. Result: C 77.80, H 5.21, N 2.45.

Crystal Structures

The structures of SFX-PO-DPA and SFX-PO-DPA-Me were studied by single- crystal X-ray diffraction. SFX-PO-DPA formed light-yellow rod-shaped crystals and was found to crystallize in the monoclinic space group C2/c when recrystallised from a mixture of methanol and dichloromethane by solvent layering. SFX-PO-DPA-Me formed pale-yellow block-shaped crystals in the monoclinic /2/a space group when recrystallised from a mixture of hexane and dichloromethane by evaporation. Recrystallization of SFX- PO-DPA-Me from the same mixture as for SFX-PO-DPA (i.e., methanol and dichloromethane by solvent layering) led to formation of pale-yellow block-shaped crystals in the same spaced group C2/c.

Co-crystallized solvent molecules have an influence on the dihedral angle between fluorene and xanthene in the spiro-structure. When dichloromethane and hexane are present in the SFX-PO-DPA-Me molecule, the dihedral angle is 86.7°. However, switching to a more polar solvent system (dichloromethane and methanol), the angle increases to 89.5°. With the same solvent system (dichloromethane and methanol) applied to the crystallization of SFX-PO-DPA only two methanol molecules were incorporated in the structure leading to a dihedral angle of 87.6°. The variation in the value of the dihedral angle as a function of the environment will have an influence on through bond interactions between the orthogonally disposed donor and acceptor units, thereby affecting the photophysical properties of these spiro-based structures.

Electrochemical Properties.

The HOMO and LUMO levels energies of the three emitters were inferred from an analysis of the cyclic voltammetry (CV) and differential pulse voltammetry (DPV) measurements obtained in dichloromethane (for oxidation) and N,N-dimethylformamide (for reduction).

CV and DPV analyses were performed on an Electrochemical Analyzer potentiostat model 620D from CH Instruments. Samples for the oxidation measurements were prepared as dichloromethane (DCM) solutions, degassed by sparging with DCM- saturated nitrogen gas for 10 minutes. For reduction measurements, dimethylformamide (DMF) was used as a solvent and the solution was bubbled with nitrogen gas for 10 minutes prior to the measurements. All measurements were performed in 0.1 M DCM/DMF solution of tetrabutylammonium hexafluorophosphate, which was used as the supporting electrolyte. An Ag/Ag ⁺ electrode was used as the reference electrode while a glassy carbon electrode and a platinum wire were used as the working electrode and counter electrode, respectively. The redox potentials are reported relative to a saturated calomel electrode (SCE) at a scan rate of 100 mV s ^-1 , with a ferrocene/ferrocenium (Fc/Fc ⁺ ) redox couple as the internal standard (0.46/0.45 V (DCM/DMF) vs SCE).

The CVs and DPVs are shown in Fig. 2 (scan rate: 100 mV s ^-1 ). The data are summarized in Table 1.

Each of the emitters exhibits two highly reversible oxidation waves, reflecting sequential oxidation of each of the two donor moieties. As predicted by the theoretical calculations (see Molecular Modelling section, infra), the HOMO in all three compounds is delocalized across the two donor moieties through the xanthene bridge. Such delocalization can significantly stabilize the radical cation formed upon the first oxidation. As a result, removal of the second electron becomes less difficult than for a less- delocalized HOMO due in part to lower Coulombic repulsion and this is manifested in the fairly close spacing of the two CV oxidation waves. For SFX-PO-DPA, E _ox , the peak oxidation potential as determined from DPV, is 0.76 V. The oxidation potentials shifted cathodically with increasing donor strength to 0.73 V and 0.60 V for SFX-PO-DPA-Me and SFX-PO-DPA-OMe, respectively, consistent with the picture obtained by DFT calculations (see again Molecular Modelling section, infra). These values are almost identical to the oxidation potentials of related tetrasubstituted hole transport materials, such as SFX-MeOTAD (0.52 V vs SCE) with four dimethoxydiphenylamine groups and SFX-TAD (0.68 V vs SCE) with four diphenylamine groups, reported previously (Maciejczyk, M. et al., supra). This indicates no influence on the oxidation potential from the diphenylphosphine oxide-modified fluorene fragment. The HOMO energies were determined to be -5.56 eV, -5.53 eV and -5.40 eV for SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA-OMe, respectively. The decrease in HOMO energy is associated with the increase in mesomeric effect with increasing electron-donating strength of the amine donor groups. Only one irreversible reduction wave was observed for all three compounds. LUMO energies were inferred from the reduction potentials obtained by DPV measurements in DMF and were found to be -2.94 eV, -3.03 eV and -3.15 eV for SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA-OMe, respectively. Unexpectedly, there is an observed small stabilization of the LUMO with increasing donor strength associated with an increase in stabilizing inductive effect by substituting DPA successively with Me and OMe substituents, the mesomeric effect being inexistent due to the complete localization of the LUMO on the acceptor (see molecular modelling section infra).

Photophysical Properties

Characterization

Optically dilute solutions of concentrations on the order of 10 ^-5 or 10 ^-6 M were prepared in HPLC grade solvent for absorption and emission analysis. Absorption spectra were recorded at room temperature on a Shimadzu UV-1800 double beam spectrophotometer. Aerated solutions were bubbled with compressed air for 5 minutes whereas degassed solutions were prepared via three freeze- pump-thaw cycles prior to emission analysis using an in-house adapted fluorescence cuvette, itself purchased from Starna. Steady-state emission and time-resolved emission spectra were recorded at 298 K using an Edinburgh Instruments FLS980 fluorimeter. Samples were excited at 360 nm for steady-state measurements and at 378 nm for time-resolved measurements. Photoluminescence quantum yields for solutions were determined using the optically dilute method (Crosby, G. A.; Demas, J. N., Measurement of photoluminescence quantum yields. Review. J. Phys. Chem. 1971, 75 (8), 991-1024) in which four sample solutions with absorbance of ca. 0.100, 0.080, 0.060 and 0.040 at 360 nm were used. Their emission intensities were compared with those of a reference, quinine sulfate, whose quantum yield (Φ _r ) in 1 N H ₂ SO ₄ was determined to be 54.6% using the absolute method (Melhuish, W. H., Quantum Efficiences Of Fluorescence Of Organic Substances: Effect Of Solvent And Concentration Of The Fluorescent Solute 1. J. Phys. Chem. 1961, 65 (2), 229-235). The quantum yield of the sample, Φ _PL , can be determined by the equation: Φ _PL = Φ _r (A _r /A _s )((l _s /l _r )(n _s /n _r ) ² , where: A stands for the absorbance at the excitation wavelength (λ _exc = 360 nm), I is the integrated area under the corrected emission curve and n is the refractive index of the solvent with the subscripts “s” and “r” representing sample and reference respectively. An integrating sphere was employed for quantum yield measurements for thin film samples (Greenham, N. C.; Samuel, I. D. W.; Hayes, G. R.; Phillips, R. T.;

Kessener, Y. A. R. R.; Moratti, S. C.; Holmes, A. B.; Friend, R. H., Measurement of absolute photoluminescence quantum efficiencies in conjugated polymers. Chem. Phys. Lett. 1995, 241 (1-2), 89-96). Doped thin films were prepared by mixing the sample (10 wt.%) and PMMA in toluene followed by spin-casting on a quartz substrate. The Φ _PL of the films were then measured in air and by purging the integrating sphere with N ₂ gas flow. Time-resolved PL measurements of the thin films were carried out using the time- correlated single-photon counting technique. The samples were excited at 378 nm by a pulsed laser diode (Picoquant, model PLS 370) and were kept in a vacuum of < 10 ^{- 4} mbar. The singlet-triplet splitting energy ΔE _ST was estimated by recording the prompt fluorescence spectra and phosphorescence emission at 77 K in 10 and 15 wt.% doped films in mCP. The films were excited by a Q-switched Nd:YAG laser emitting at 355 nm (Laser-export). Emission from the samples was focused onto a spectrograph (Chromex imaging, 250is spectrograph) and detected on a sensitive gated iCCD camera (Stanford Computer Optics, 4Picos) having sub nanosecond resolution.

Fig. 3a shows the UV-Vis absorption and photoluminescence (PL) spectra of the three emitters in toluene. The data are summarized in Table 2:

All three compounds exhibit similar optical absorption and possess a series of highly absorbing bands from 290 nm to 320 nm, attributed from Time-Dependent Density Functional Theory (TD-DFT) calculations to high-oscillator-strength locally excited electronic transitions to S _n (n>1 ) states, while, as expected from its strong charge- transfer character, the oscillator strength for the promotion to S ₁ is much smaller (Tables 5 and 6, vide infra).

The PL spectra in degassed toluene are broad and unstructured, typical of emission from a CT state. A red-shift in the emission from SFX-PO-DPA (λ _PL = 490 nm) to SFX-PO-DPA-Me (λ _PL = 492 nm) and to SFX-PO-DPA-OMe (λ _PL = 514 nm) is consistent with the decrease in the HOMO-LUMO gap, an observation corroborated by theoretical calculations (see again Molecular Modelling section, infra). The Φ _PL values of 56%, 80% and 68% for SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA-OMe, respectively, in degassed toluene varied significantly as a function of the pendant groups on the donor amine. Emission was dramatically quenched, with much reduced □PL of 19%, 38% and 25% SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA-OMe, respectively, upon exposure to air, an indication of triplet harvesting in the absence of oxygen.

Next, the PL behavior of the three compounds was investigated in the solid state as doped films in both PMMA and mCP (Fig. 3b). Irrespective of the host, the emission profiles remained broad and unstructured. In solution-processed 10 wt% doped films in PMMA, all three compounds exhibited blue-green to green emission with λ _PL of 490 nm for SFX-PO-DPA, 495 nm for SFX-PO-DPA-Me and 515 nm for SFX-PO-DPA-OMe, results in line with those observed in toluene. Φ _PL values of 10%, 38% and 25% were obtained in the PMMA films for SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA- OMe, respectively. Time-resolved PL measurements in PMMA revealed bi-exponential decay profiles for all three emitters. SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA- OMe possessed prompt lifetimes, τ _p , of 92 ns, 126 ns and 156 ns, respectively, and delayed lifetimes, τ _d , of 10 μs, 9.4 μs and 6.4 μs, respectively. Based on the emission energies of the three emitters, mCP was chosen as the host matrix for OLEDs. Optimum doping concentrations of 10 wt% for SFX-PO-DPA and SFX-PO-DPA-OMe, and 15 wt% for SFX-PO-DPA-Me were determined based on a concentration-dependence study of absolute Φ _PL values of vacuum-deposited films (Table 3).

At these concentrations, slightly red-shifted emission maxima of 500 nm for SFX- PO-DPA and SFX-PO-DPA-Me, and 520 nm for SFX-PO-DPA-OMe were observed along with much enhanced Φ _PL values of 50%, 70% and 58%, respectively, compared to the measurements in PMMA. Time-resolved PL measurements again showed bi- exponential decay behavior for all three emitters, with an enhanced contribution from the delayed emission (Fig. 4). In comparison to the non-polar host PMMA, faster transient PL dynamics were observed in mCP both in terms of prompt and delayed lifetimes. Prompt lifetimes of 81 ns (43%), 49 ns (18%) and 100 ns (40%), and delayed lifetimes of 8.4 μs (57%), 8.3 μs (82%) and 6.2 μs (60%) were obtained for SFX-PO-DPA, SFX- PO-DPA-Me and SFX-PO-DPA-OMe, respectively, showing that delayed fluorescence also occurs in the mCP host. The long radiative lifetimes measured for the prompt component of these compounds support the emission from CT states with limited overlap between hole and electron densities.

To further confirm the TADF character of these emitters, the temperature dependence of the time-resolved emission decays of the mCP doped films containing the three emitters was then studied. An increase in the intensity of the delayed component of the lifetime with increasing temperature provides direct evidence to support the TADF character of these spiro-based emitters (Fig. 5a-c). The ΔE _ST values in mCP of the three emitters were determined from the difference in the energies of the fluorescence and phosphorescence spectra obtained from the onset of prompt and delayed emission, respectively, at 77 K (Fig. 6a-c). Very small ΔE _ST values of 0.05 eV, 0.02 eV and 0.01 eV and S ₁ energies of 2.85 eV, 2.80 eV and 2.75 eV were measured for SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA-OMe respectively, in line with theory (Table 5, vide infra) and confirming their strong potential as efficient TADF emitters in the solid state.

In order to understand the origin of the short τ _d and high Φ _PL , the rate constants for the radiative decay from the singlet state (k _r ^s ), RISC (k _rlSC ) and non-radiative decay from the triplet state ( k _nr ^T ) were calculated. It should be noted that the rate constants for radiative and non-radiative processes can only be explicitly calculated assuming a monoexponential decay of the emission. For bi-exponential decay, the corresponding rates were calculated making a set of assumptions as described previously (Masui, K.; Nakanotani, H.; Adachi, C., Analysis of exciton annihilation in high-efficiency sky-blue organic light-emitting diodes with thermally activated delayed fluorescence. Org. Electron. 2013, 14, 2721-2726).

The absolute rate constants for radiative and non-radiative processes can only be explicitly calculated for a monoexponential decay. For a bi- or multiexponential decay, it was assumed that the k _nr ^s approaches zero and therefore the intersystem crossing can be defined as Φ _ISC = 1- Φ _p , following the method described by Masui et a/._(Masui, K. et al. (ibid) The rate constants (k _r ^s , k _nr ^T , k _ISC , k _rISC ) associated with all the three emitters were evaluated as follows: k _p = 1/τ _p , k _d = 1/ τ _d , where k _p and k _d represent the prompt and delayed fluorescence rates which were calculated from the experimentally measured prompt and delayed lifetimes.

The prompt and delayed fluorescence quantum efficiencies, Φ _p and Φ _d , were determined by integrating the transient PL signal from 0 to 500 ns as the prompt components and from 500 ns to 50 μs as the delayed components. Therefore: k _r ^s = Φ _p k _p , ; k _ISC = (1- Φ _p ) k _p . k _rISC = (k _p k _d I k _ISC ) * Φ _d / Φ _p . k _nr ^T = k _d - Φ _p k _Risc , where k _v ^S is the radiative decay rate of the singlet state, k _ISC is the intersystem crossing rate, k _rISC is the reverse intersystem crossing rate, and k _nr ^T is non-radiative decay rate of the triplet state.

All three emitters exhibited similar k _r ^S on the order of 2.3-2.6 x 10 ⁶ s ^-1 (Table 4).

Importantly, k _RISC , which is a crucial parameter responsible for an efficient TADF mechanism, remained higher than the competing k _nr ^T rate constant for all three emitters, implying that the thermal upconversion of excitons from Ti to S ₁ is preferred over non- radiative decay from the Ti state. Furthermore, the relatively higher value of k _RISC coupled with a very small ΔE _ST in SFX-PO-DPA-Me is indicative of a faster and more efficient RISC mechanism in this emitter, which results in the highest Φ _PL values observed across all media. Such short τ _d values and large have been shown to play a useful role in realizing low efficiency roll-off in the devices by reducing the probability of triplet exciton quenching mechanisms (Chen, Z.; Wu, Z.; Ni, F.; Zhong, C.; Zeng, W.; Wei, D.; An, K.; Ma, D.; Yang, C., Emitters with a pyridine-3,5-dicarbonitrile core and short delayed fluorescence lifetimes of about 1 .5 μs: orange-red TADF-based OLEDs with very slow efficiency roll-offs at high luminance. J. Mater. Chem. C 2018, 6, 6543-6548).

Molecular Modelling

Characterization of the electronic structure and optoelectronic properties of the three SFX spiro compounds has been carried on with the help of DFT calculations. More specifically, previously developed methodology that offers a particularly accurate description of the electronic structure of materials for OLED applications was used, namely the PBEO functional was adopted and the excited-state properties calculated within the Tamm-Dancoff approximation (Moral, M.; Muccioli, L.; Son, W. J.; Olivier, Y.; Sancho-Garcia, J. C., Theoretical Rationalization of the Singlet-Triplet Gap in OLEDs Materials: Impact of Charge-Transfer Character. J. Chem. Theory Comput. 2015, 11, 168-177). The increasingly shallower HOMO level along the series SFX-PO-DPA (-4.95 eV), SFX-PO-DPA-Me (-4.81 eV) and SFX-PO-DPA-OMe (-4.49 eV) is consistent with the increasing donor strength of the diarylamine donor. For each emitter, the LUMO is localized exclusively on the fluorene core with no contribution from the pendant phosphine oxides and is only slightly affected by stabilizing inductive effects of the side groups attached to the DPA core. S ₁ energies decrease along the series from 2.77 eV for SFX-PO-DPA through 2.65 eV for SFX-PO-DPA-Me to to 2.34 eV for SFX-PO-DPA- OMe. Very small ΔE _ST values of 0.0038 eV for SFX-PO-DPA and SFX-PO-DPA-Me and 0.00051 eV for SFX-PO-DPA-OMe reflect the near orthogonal arrangement of the donor and acceptor groups.

The nature of the excited states was determined using the Φ _S metric that quantifies the overlap between the hole and electron densities as obtained in the attachment-detachment formalism (Table 5) (Olivier, Y.; Moral, M.; Muccioli, L.; Sancho- Garcia, J.-C., Dynamic nature of excited states of donor-acceptor TADF materials for OLEDs: how theory can reveal structure-property relationships. J. Mater. Chem. C 2017, 5, 5718-5729).

DFT calculations were performed with the Gaussian 09 revision D.018 suite (Frisch, M. J et al. Gaussian 09, Revision D.01, Gaussian Inc.: Wallingford, CT, 2013).

Initially the geometries of both emitters in the ground state in the gas phase were optimized employing the PBEO (Adamo, C.; Barone, V., Toward reliable density functional methods without adjustable parameters: The PBEO model. J. Chem. Phys. 1999, 110 (13), 6158-6170) functional with the standard Pople 6-31G(d,p) basis set (Pople, J. A.; Binkley, J. S.; Seeger, R., Theoretical Models Incorporating Electron Correlation. Int. J. Quant. Chem. Symp. 1976, 10, 1). Tim x 10 x 10dependent DFT calculations were performed within the Tamm-Dancoff approximation (TDA) (Moral, M et al., supra) The molecular orbitals were visualized using GaussView 5.0 software.

Computationally predicted (Mol. Phys., 116, 19 (2018)) fluorescence rate (Ay), where f is oscillator strength (unitless), n is the refractive medium and ΔE is vertical energy (cm ^-1 ).

Table 5: Transition energies, contributions of canonical orbitals to the excited states transitions, Φ _S index values, oscillator strengthand nature associated with the two lowest singlet and triplet excited states of SFX-PO-DPA, SFX-PO-DPA- Me and SFX-PO-DPA- OMe compounds using the PBEO functional with the 6-31G(d,p) basis set in the gas phase

Using this methodology, full charge-transfer (CT) [completely localized (LE)] excited states feature Φ _S = 0 [ Φ _S = 1] can be inferred. Φ _S values were calculated using the NANCY_EX package (Etienne, T.; Assfeld, X.; Monari, A., Toward a Quantitative Assessment of Electronic Transitions’ Charge-Transfer Character. J. Chem. Theory Comput. 2014, 10, 3896-3905) and the hole and the electron densities were visualised with DrawMol (V. Liegeois, DrawMol, UNamur, https://www.unamur.be/drawmol, 2021). Very low values, Φ _S < 0.12 (Table 5), were calculated for both the S ₁ and Ti states for all three emitters, implying excited states with strong CT character. This is mirrored by the very small calculated oscillator strengths ranging from 0 to 10 ^-4 and translates into radiative decay rates of ~4.8 x 10 ⁴ s ^-1 roughly two orders of magnitude smaller than the experimental values (Table 6).

Table 6: Transition wavelength, contributions of canonical orbitals to the excited states transitions, Φ _S index values, oscillator strength and electron density reorganization upon transition in these highly absorbing singlet excited states in SFX-PO-DPA, SFX-PO- DPA-Me and SFX-PO-DPA-OMe using the PBE0 functional with the 6-31G(d,p) in the gas phase.

A range of other functionals were applied to SFX-PO-DPA-Me, with oscillator strength remaining very low (Table 7).

Optimization of the singlet excited state was also performed for SFX-PO-DPA-

Me and offered a slight increase in the predicted oscillator strength with a value of 2× 10 ^-4 resulting in a radiative rate of 9.7 x 10 ⁴ s ^-1 . The discrepancy between the theoretical and the experimental radiative decay rates for the ground state and S ₁ equilibrium geometries of SFX-PO-DPA-Me suggests that the Franck-Condon approximation breaks down. Experimental and simulated emission are thus reconciled by invoking Herzberg-Teller intensity borrowing, mediated via vibronic coupling. Within this framework, perturbative interactions between the S ₁ electronic wave function and nuclear motions result in a combined electronic and vibronic state that can mix with nearby purely electronic states. We thus sum, consistently with the Thomas-Kuhn-Reiche sum-rules, the oscillator strengths associated with each vibronic transition from the ground state of vibration of state S ₁ to the n-th vibrational levels of the ground state, S ₀ . Considering the undistorted-undisplaced harmonic oscillator approximation and the same energy for all vibronic transitions corresponding to the transition energy from the S ₁ optimized geometry to the ground state, the oscillator strength for the transition between the S ₁ and the ground states is: where m _e and e are the electron mass and charge, is the reduced Planck constant, is the transition dipole moment matrix element between the ground state of vibration of S ₁ and the n-th vibrational level of component of the transition dipole moment computed at the S ₁ optimized geometry, is the first derivative of the electronic transition dipole moment along the normal coordinates Q _i evaluated at the equilibrium geometry of the reference S ₁ state, ω _i , is the angular frequency of the ith normal mode; are the oscillator strengths computed at the S ₁ optimized geometry and the Herzberg-Teller contribution of the i-th normal mode, respectively.

In practice, the transition dipole moment derivative is obtained by displacing the S ₁ -optimized excited-state geometry along its normal modes by 0.01 A in an interval going from -0.1 to 0.1 and fitting the transition dipole moment evolution as a function of the normal coordinate, Q _i , with a linear regression. To reduce computational cost, vibrational modes leading to a concomitant bending of the electron- donating and electron-accepting units were selected, as these sensibly increase the hole and electron densities near the sp ³ hybridized carbon and therefore could prompt larger oscillator strength. Accounting for these vibrations indeed results in an increase in oscillator strength up to 3.7 × 10 ^-4 . This brought the calculated radiative decay rate (considering a refractive index of 1.73 for mCP) to 1.81 × 10 ⁵ s ^-1 , which is within the same order of magnitude as the measured value of k _r = 2.3 × 10 ⁶ s ^-1 , highlighting the importance of Herzberg-Teller effects in spiro-based compounds. The (dynamic, i.e. induced by vibrations) increase in electron-hole overlap and in the ensuing radiative decay rate was expected to be sensitive to the nature of the electron-active substituents. To verify this hypothesis, compound SFX-PO-PTZ was designed, where the DPA-Me side groups are replaced by stronger electron-donating phenothiazine (PTZ) groups, and carried out the same analysis, i.e., S ₁ excited-state optimization followed by sampling along the relevant normal modes. As expected, SFX-PO-PTZ exhibits an oscillator strength in the relaxed S ₁ excited-state geometry that amounts to 7.13 x 10 ^-5 when Herzberg-Teller corrections are included, which is one order of magnitude smaller than SFX-PO-DPA-Me. This result can be readily explained by the removal of the hole density away from the sp ³ carbon in presence of the stronger PTZ donor moieties.

Device Fabrication

The photophysical picture points to spiro-based compounds that are TADF emitters and have high Φ _PL in mCP and short τ _d . These compounds were therefore assessed as emitters in OLEDs.

OLED devices were fabricated in bottom emitting architecture via vacuum sublimation in high vacuum at a base pressure of 2-5×10 ^-7 mbar. The organic layer sequence and the metal cathode were deposited onto pre-cleaned glass substrates coated with indium tin oxide (ITO) which has a sheet resistance of around 30 Ω/sq. A pre-patterned ITO glass substrate was treated by ultrasonic cleaning in acetone and isopropanol consecutively and then treated by oxygen plasma before the transfer to the vacuum chamber. Organic layers were deposited at a rate of 0.3-0.6 Å/s, which was controlled in situ using the quartz crystal monitors. Doping of the emission layers was achieved through co-evaporation of the emitter and host materials. The electron injection layer LiF was deposited at a rate of 0.10 Å/s while the Al cathode was deposited at a rate of 0.5 Å/s through the shadow mask defining the top electrode. The spatial overlap of the anode and cathode electrodes determined the active area of the OLED which was estimated to be 2 mm ² . All the devices were encapsulated with glass lids and UV epoxy resin inside the inert atmosphere. The luminance-current-voltage characteristics were measured in an ambient environment using a Keithley 2400 source meter and Keithley 2000 multimeter connected to a calibrated Si photodiode. The external quantum efficiency was calculated assuming Lambertian emission distribution. The electroluminescence spectra were recorded by an Andor DV420-BV CCD spectrometer.

The OLED device stack architecture is shown in Fig. 7 and consists of: ITO / NPB (30 nm) / TCTA (20 nm) / mCP (10 nm) / Emitter: mCP (20 nm) / DPEPO (10 nm) / TmPyPB (40 nm) / LiF (1 nm) / Al (100 nm), where N,N' -bis(naphthalen-1-yl)-N,N' - bis(phenyl)benzidine (NPB) was used as a hole injection layer (HIL), tris(4-carbazoyl-9- ylphenyl)amine (TCTA) was used as a hole transporting layer (HTL), mCP (1 ,3-bis(N- carbazolyl)benzene) and DPEPO were used as an electron and hole blockers, respectively. 1 ,3,5-tris(3-pyridyl-3-phenyl)benzene (TmPyPB) was chosen as the electron transporting layer (ETL) as it possesses a high electron mobility of 10 ^-4 cm ² V ^-1 s ^-1 and a high triplet energy of 2.75 eV along with a deep HOMO energy of 6.7 eV (Su, S.-J . ; 39, T.; Takeda, T.; Kido, J., Pyridine-Containing Triphenylbenzene Derivatives with High Electron Mobility for Highly Efficient Phosphorescent OLEDs. Adv. Mater. 2008, 20, 2125-2130). The emissive layer (EML) consisted of 10 wt% doped films of SFX-PO- DPA, SFX-PO-DPA-OMe and 15 wt% doped film of SFX-PO-DPA-Me in mCP.

The electroluminescence properties are shown in Fig. 8 and data are summarized in Table 8: The three emitters exhibited blue-green to green electroluminescence with systematic red-shifting of the EL spectra as a function of donor strength, in parallel to the trends and maxima observed in their respective PL spectra. The OLED with SFX-PO- DPA showed a λ _EL of 500 nm with CIE coordinates of (0.20, 0.47), the OLED with SFX- PO-DPA-Me showed a slight red-shift with a λ _EL of 502 nm and CIE coordinates of (0.23, 0.50) and SFX-PO-DPA-OMe displayed the most red-shifted EL spectrum with a λ _EL of 520 nm and CIE coordinates of (0.29, 0.56). All OLEDs showed steep current-voltage- luminance behavior (Figs 8a and 8b) with low turn-on voltages of 3.2 V for devices within SFX-PO-DPA and SFX-PO-DPA-Me and 3.8 V for the device with SFX-PO-DPA-OMe.

Fig. 8c shows the EQE vs luminance behavior of the three devices. The OLED with SFX-PO-DPA-Me showed an excellent device performance with an EQE _max of 23% at a brightness of 2.5 cd m ^-2 and relatively low roll-off, with an EQE ₁₀₀ of 19% at a display- relevant brightness of 100 cd m ^-2 . Similarly, the OLEDs with SFX-PO-DPA and SFX- PO-DPA-OMe also maintained high EQE _max values of 11% and 16%, respectively, at low luminance, and likewise showed low efficiency roll-off with EQE ₁₀₀ of 9% and 15%, respectively. Even at a luminance of 1000 cd m ^-2 , EQE ₁₀₀₀ of 7%, 16% and 12% were maintained for the OLEDs with SFX-PO-DPA, SFX-PO-DPA-Me and SFX-PO-DPA- OMe, respectively. The high observed EQEs are indicative of efficient triplet harvesting while the low efficiency roll-offs are due to the short τ _d values, which are beneficial in reducing the various triplet exciton quenching mechanisms such as triplet-triplet annihilation (TTA) and triplet polaron annihilation (TPA) in the devices. The poorer device performance of the OLEDs with SFX-PO-DPA and SFX-PO-DPA-OMe can be correlated to their lower Φ _PL values (Table 1). Compared to the reported OLEDs with spiro-based emitters shown in Fig. 1 , the devices based on SFX-PO-DPA-Me show the best performance in terms of EQE _max and reduced efficiency roll-off. Further, SFX-PO- DPA and SFX-PO-DPA-OMe emitters show comparable EQE _max values and improved efficiency roll-off characteristics compared to literature devices.

Conclusions

A series of spiro-configured xanthene-based TADF emitters have been synthesised, which exhibit a confluence of desirable photophysical properties. An efficient synthetic strategy to incorporate an SFX bridge has been demonstrated in order to produce highly sterically demanding structures that translate into materials that possess short delayed lifetimes of < 10 μs, very small ΔE _ST values coupled with high Φ _PL in the solid state. Molecular modelling reveals that the emission in these compounds occurs through a Herzberg-Teller mechanism where some intramolecular vibrations promote larger overlapping electron and hole density in vicinity of the sp ³ carbon atoms, thereby enhancing the radiative decay rate. The designed compounds exhibit non- radiative decay rates an order of magnitude slower than the rate of reverse intersystem crossing. The resulting OLEDs exhibited EQE _max values as high as 23% and with only modest efficiency roll-off at 100 and 1000 cd m ^-2 as a result of the short delayed lifetimes and reduced triplet-triplet and triplet-polaron annihilation associated with the expected reduced triplet diffusion due to the bulky shape of our compounds. These results clearly illustrated how useful an SFX blocking unit is in realizing highly efficient TADF systems.