The magic of semi-perfluorinated mixtures: engineering MR-TADF emission from host–guest mixtures of liquid crystals

Abstract

Two liquid crystalline emitter–host mixtures have been developed, where each of the emitter and host self-assembles into a columnar hexagonal (Col_h) mesophase, as does the mixture. The emitter consists of a modified 7-(tert-butyl)-5-oxa-8b-aza-15b-bora-benzo[a]naphtho[1,2,3-hi]-aceanthrylene B–O–Cz multi-resonant thermally activated delayed fluorescence (MR-TADF) core containing lateral semiperfluorinated side chains, while the host is derived from a modified MR-TADF 5,9-dioxa-13b-boranaphtho[3,2,1-de]anthracene (DOBNA) core that also contains lateral semiperfluorinated side chains. Unlike the previously used alkoxy side chains in TADF liquid crystalline materials, the semiperfluorinated side chains only minimally perturb the optoelectronic properties of the luminescent cores, this despite the strong aggregation within the ordered columnar hexagonal (Col_ho) phase. This was achieved by shielding the emitter moieties from each other within the host mesophase, which mitigates excimer emission from aggregates and by engineering an efficient Förster resonance energy transfer from the host material to the emitter in the blend.

---

Introduction

Emissive liquid crystals (LCs) are attractive materials for various applications such as organic light-emitting diodes (OLEDs), organic field effect transistors (OFETs), organic photovoltaic devices (OPVs),^1,2 linear polarized emission,³ and luminescent concentrators,⁴ among others.⁵ In particular, discotic LCs are promising in that respect. The strong π–π interactions of their (hetero)aromatic cores and van der Waals interactions of the flexible side chains promote self-assembly into columnar mesophases with long-range orientational order,^5–10 resulting in increased charge carrier mobility along the columnar axis.¹¹ Moreover, the electronic properties of the (hetero)aryl core can be tailored so that either hole or electron-conducting materials are obtained. Functionalization of the (hetero)aryl core leads to improved solution processability.^11–14 Further advantages are the self-healing of defects through the local fluidity of the molecules within the mesophase.^5,6,9 An important parameter for OLEDs is the alignment of the emitter, which is frequently correlated with the orientation of the transition dipole moment that controls the light outcoupling. The alignment of the molecule can be controlled by the type of columnar mesophase (i.e., hexagonal or rectangular)⁷ or surface nanopatterning,¹⁵ and monitored by 4D scanning tunneling electron microscopy (STEM).¹⁶ This results in controlled alignment of transition dipoles.^14,17 For OLEDs to be efficient, the device must harvest both singlet and triplet excitons to produce light, typically managed by the emissive material. One possible solution is LC emitters that are thermally activated delayed fluorescent (TADF).^18,19 TADF emitters possess a small singlet–triplet energy gap (ΔE_ST), enabling the thermal upconversion of triplet excitons into singlets by reverse intersystem crossing (RISC), thus permitting up to 100% of the excitons to be converted into light in the device.^18,19

Several TADF-LCs have been reported (Fig. 1), such as carbazole-benzonitrile donor–acceptor (D–A)–TADF compounds 1a–c displaying columnar hexagonal (Col_h) mesophases.^12,13 the substitution pattern and symmetry have a strong impact on the photophysical properties. For example, quantum yields Φ decreased from 11% (for 1b) to 1.5% (for 1c) despite the small singlet–triplet gaps ΔE_ST of 0.01 and 0.04eV, respectively. In addition, several TADF emitters that self-assemble into lamellar phases have been reported^11,17 (for details see SI, Fig. S64 and discussion).

Fig. 1 Overview of literature-known TADF-LCs; (b) our previous work on MR-TADF LCs (DiKTaLC, BON-LC, LCn); (c) our new strategy towards liquid crystalline emitter host mixtures features MR-TADF emission at room temperature.

A 1 : 1 mixture of columnar DOBNA derivative 2a (Y = B) and its N-analogue 2b (Y = N) forms a supramolecular copolymer via B–N frustrated Lewis pairs.²⁰ The copolymer showed yellow emission in decaline (λ_PL = 550 nm; τ_p = 96 ns, τ_d = 6 µs) and circular polarized emission (λ_PL = 550 nm, g_PL = 7 × 10⁻³). Very recently, the dual fluorescent dye 4,7-CzPhen consisting of phenanthroline with two peripheral carbazole units, was reported, which showed broad Col_h mesophases already at ambient temperature.²¹ The substitution pattern controlled the photophysical properties. Within the series of regioisomers, only 4,7-CzPhen displayed TADF in PS matrix (≪1 wt%) (Φ_PL = 25%). Columnar self-assembly of perylene imides ColLC-A, ColLC-B led to mesophase-assisted aggregation-induced delayed fluorescence (AIDF) by triplet–triplet annihilation.²² While the large ΔE_ST = 0.83, 1.16 eV, respectively, precluded TADF emission, the close contact of the stacked molecules of the two different mesogens in the Col_h phase promoted delayed fluorescence by triplet–triplet annihilation (TTA) from imide aggregates (i.e., ColLC-A), assisted by energy transfer from the imidoester ColLC-B (Φ_PL = 37%). All blends show similar emission (λ_PL = 575 nm, Φ_PL = 16% for 1 : 1 blend) as neat ColLC-A (Φ_PL = 17%). OLEDs with ColLC-A showed EQE_max = 1.8%.²²

However, the strong intermolecular interactions governing LC self-assembly impair the photophysical properties. For example, the formation of columnar stacks of emitter molecules often leads to aggregation-caused quenching (ACQ) and thus the Φ_PL is significantly reduced in the liquid crystalline bulk phase as compared to in solution.^{11–14,17,23} To address this challenge, one strategy involves designing TADF-LC materials that self-assemble into lower-order structures. Therefore, we developed DiKTaLC (Fig. 1), a nematic discotic (N_D) LC having a ΔE_ST of 0.20 eV.¹⁴ Neat thin films showed green emission (λ_PL = 512 nm, Φ_PL = 52%, τ_p = 7 ns, τ_d = 70.2 µs). Comparison with the non-mesogenic counterpart revealed that the mesogenic 4-cyanobiphenyl units control the emitter orientation, resulting in homeotropic alignment in thin films and a preferentially horizontally aligned TDM. Doped blue-green emitting OLEDs (λ_PL = 492 nm) with DiKTaLC showed an EQE_max of 13.6%.¹⁴

A second common approach to reduce excimer formation utilizes doping of the TADF-LC emitter (1–20 wt%) into a low molecular weight or polymeric host material,^{11–14,17,23} a strategy that is also employed for non-LC emitters.^24,25 For example, BON-LC12 self-assembles into a broad Col_h phases (Fig. 1), which is stable even at sub-ambient temperatures. However, the strong interactions in the neat Col_h mesophase suppressed TADF emission.²³ In contrast, when BON-LC12 was diluted into a polystyrene host matrix (1 wt% BON-LC12), TADF emission is observed (λ_PL = 467 nm, Φ_PL = 90%, τ_p = 9.85 ns, τ_d = 107.9 µs).²³

Despite the improved photophysical properties, in most cases, such host–guest mixtures no longer show any LC behavior, which effectively defeats the purpose of incorporating the mesogenic units.¹¹ Recently, we disclosed host–emitter mixtures consisting of a columnar DOBNA host LCn²⁶ and a non-mesogenic “disk-like” MR-TADF emitter BCzBN,²⁷ where the columnar mesophase of the host was maintained in the mixture, enabling efficient Förster resonance energy transfer (FRET) from the LC host to the emitter, and thus, narrowband MR-TADF emission (e.g. LC8/BCzBN: λ_PL = 476 nm, Φ_PL = 41%, τ_p = 8.19 ns, τ_d = 108.5 µs), was observed, even in the tightly packed mesophases at room temperature.²⁶ The XRD data of the LC8/BCzBN mixtures in the Col_h mesophase suggest preferential localization of BCzBN in the aliphatic chain network.²⁶ While this arrangement is beneficial for photophysics alone (narrowband emission spectrum), it is not so helpful for controlling exciton dynamics.¹⁹ Thus, the formation of mixed columnar stacks would be much more interesting. Unfortunately, mixing of LCn (as host) and BON-LC (as emitter) was not photophysically compatible due to the too low T₁ energy level of LCn (resulting from partial long-range charge transfer, LRCT, character of the emissive excited state due to the presence of the electron-rich alkoxy side groups). Thus, we wanted to implement alkyl-linked side chains on the central core to increase the T₁ energy of the host.

To minimize the electronic influence of the side chain on the emissive cores of both host and emitter, and to simultaneously enhance nanosegregation and thus promote liquid crystalline phase stability, we considered the use of semi-perfluorinated side chains instead of alkoxy chains as mesogenic groups attached to the MR-TADF core of the host and/or emitter, making use of the so-called fluorophobic effect.²⁸ The high polarity and low polarizability of C–F bonds lead to helical conformations of perfluorinated chains, and together with the larger size of F as compared to H, the lipophilicity and rigidity of the chains are increased.²⁸ It was shown for naphthalenetetracarboxylic diimide-derived organic transistors that the use of semi-perfluorinated side chains resulted in high mobility, and environmental and bias stress stability.²⁹ Improved transistor performance by fluorophobic self-organization was also reported for polymers with semi-perfluorinated side chains.^30,31 Moreover, the fluorophobic effect has been successfully exploited to induce and stabilize LC self-assembly.^32,33 Such an approach has yet to be used in TADF LC materials design.

Thus, here we designed columnar host–emitter LC mixtures showing MR-TADF emission in the columnar mesophase. We anticipated that the attachment of semiperfluorinated side chains on both host and emitter should facilitate LC self-assembly by fluorophobic interactions, while maintaining the singlet and triplet energy levels of the parent unsubstituted system. As the semi-perfluorinated side chains require more volume than alkoxy chains, only 6 instead of 9 side chains were attached in both hosts, LC3-F, LC7-F, and emitter BON-LC3-F, to avoid steric clashing that would perturb the self-assembly process. Our results demonstrate that this two-pronged approach indeed results in all columnar host–guest mixtures and enables FRET from the host to the emitter, resulting in narrowband MR-TADF emission.

Results and discussion

Theoretical studies

To understand the influence of the semi-perfluorinated alkyl chains on the hosts LC3-F and LC7-F, and the emitter BON-LC3-F, time-dependent density functional (TD-DFT) calculations at the PBE0/6-31G(d,p) level in the gas phase were carried out. Both hosts, LC3-F and LC7-F, have similar localizations of the HOMO and LUMO on the aromatic core (Fig. S1 and S2). The HOMO and LUMO are localized on alternating atoms, which is illustrative of a short-range charge transfer (SRCT) S₀–S₁ transition (HOMO–LUMO) that is typical of MR-TADF emitters.^34–36 As expected, the semi-perfluoroalkyl chains have very little influence on the HOMO and LUMO coefficients and energies. The HOMO (−5.75 eV) and LUMO (−1.64 eV) of the host LC3-F are slightly destabilized as compared to the HOMO (−5.81 eV) and LUMO (−1.70 eV) of the homologue LC7-F that has longer side chains, while the energy gap (E_g = 4.11 eV) remains identical. The energy values are slightly stabilized as compared to those of DOBNA derivatives containing alkoxy side chains, like LC8 (HOMO−5.65 eV, LUMO−1.60 eV, E_g = 4.05 eV), reflecting the modest impact that the electronics of the mesogenic groups have on the energies of the frontier molecular orbitals.²⁶ Thus, we assumed that the MR-TADF character of the parent DOBNA system would be conserved in the derivatives containing semi-perfluorinated chains. Similarly, emitter BON-LC3-F possesses HOMO (−5.37 eV) and LUMO (−1.67 eV) levels that show the same alternating pattern of their electron density, and has an E_g = 3.70 eV, again in line with the SRCT character of the MR-TADF emitter B-O-Cz (Fig. S2). Comparison of BON-LC3-F with the known dodecyloxy-substituted derivative BON-LC12²³ revealed that the former has more stabilized orbitals than the latter (HOMO = −5.21 eV; LUMO = −1.67 eV; E_g = 3.54 eV for BON-LC12).

Synthesis of LC3-F, LC7-F and BON-LC3-F

The synthesis of host and emitter compounds commenced with the preparation of the semiperfluorinated side chains (Scheme 1). Williamson etherification of semiperfluorinated alcohols 4 with propargyl bromide 5 gave the ethers 6a, b. Due to the high volatility of 6a, it was used without isolation. In contrast, the corresponding higher homologue 6b was obtained in 88% yield. Subsequent Sonogashira cross-coupling of ethers 6a, b with 3,4-diiodobromobenzene 7 gave 3,4-bisalkinylbromobenzenes 8a, b in 76 and 85% yield, which were then submitted to a sequential lithiation, borylation, and final catalytic hydrogenation of the triple bond to the pinacol borolanes 9a, b carrying semiperfluorinated alkyl chains in the 3,4-position in 48 and 45% yield over 3 steps.

Scheme 1 Synthesis of hosts LC3-F, LC7-F and emitter BON-LC3-F.

Known DOBNABr₃²⁶ was treated with pinacol borolanes 9a, b under Suzuki–Miyaura cross-coupling conditions to afford the desired hosts LC3-F, LC7-F in 41 and 57% yield, respectively. The corresponding Suzuki–Miyaura cross-coupling of known BONBr₃²³ gave the short-chain derivative BON-LC3-F in 25% yield, while the corresponding larger homologue BON-LC7-F could not be isolated due to severe solubility and purification issues.

Mesomorphic properties of LC3-F, LC7-F and BON-LC3-F

Upon cooling under the polarized optical microscope (POM), host LC3-F displayed large homeotropic areas with line defects (Fig. S42b) as well as fan-shaped textures (Fig. S42a), which were shearable and are characteristic of a columnar mesophase.^37–39 Similar POM textures were observed for the host LC7-F, also suggesting a columnar mesophase (Fig. S44).

In the 2nd heating cycle of the DSC of LC3-F, an endothermic clearing transition at 175.6 °C was detected (Fig. 2a and Table S2). Upon the 2nd cooling cycle, the isotropic phase to the Col_ho transition was detected at 175.0 °C, and a glass transition was found at −68.6 °C. Thus, the phase range of the host LC3-F (ΔT = 244 K) is significantly broadened as compared to the DOBNA alkoxy derivative LC8 (ΔT = 175 K),²⁶ and the Col_h phase was stabilized (Fig. 2c). During the 2nd cooling cycle, LC7-F displayed a higher clearing point (237.1 °C) and a glass transition at −61.7 °C (Fig. S45a and Table S2), resulting in an expanded mesophase range (ΔT = 299 K), which is much broader compared to that of the alkoxy derivative LC12 (ΔT = 106 K). These results demonstrate the stabilizing effect of the semiperfluorinated side chains resulting from the fluorophobic effect.^40,41

Fig. 2 (a) DSC thermogram (heating/cooling rate: 10 K min⁻¹) of compound LC3-F. G = glassy state, Col_ho: columnar hexagonally ordered mesophase, I = isotropic melt; (b) DSC thermogram (heating/cooling rate: 10 K min⁻¹) of LC3-F doped with 1 wt% of BON-LC3-F; (c) phase ranges of the semi-perfluorinated derivatives LC3-F, LC7-F, BON-LC3-F as well as mixtures LC3-F : BON-LC3-F, LC7-F : BON-LC3 compared with the known alkoxy derivatives LC8, LC12 and BON-LC12 respectively, determined from the second cooling cycle of the DSC measurements. Values of LC8, LC12, BON-LC12, were taken from ref. 23 and 26 for comparison.

In the WAXS, LC3-F showed four sharp reflections in the small-angle section and two diffuse wide-angle reflections, namely the halo caused by the distance of fluid-like semiperfluorinated side chains (d_halo = 5.07 A) and the π–π reflection caused by the intracolumnar distance (d_π–π = 3.47 Å) (Fig. 3a and Table S3). The higher intensity and broad appearance of the halo can be explained by the higher electron density and layer space filling of the semiperfluorinated side chains as compared to the alkoxy side chains.^40,42,43 The sharp reflections of the SAXS (Fig. 3b) were assigned as (10), (11), (20), and (21) reflections of a Col_ho phase (p6mm) due to their characteristic ratio of 1 : 1/√3 : 1/2 : 1/√7. The small intensity of the (10) peak as compared to the higher order peaks (11), (20), (21) is probably caused by the high electron density of the semiperfluorinated side chain.^40,41,44

Fig. 3 (a) WAXS diffractogram of LC3-F recorded at 33 °C with the 2D WAXS pattern shown as inset. The red trace represents the fit of the wide-angle region with two Lorentzian functions (grey dashed traces); (b) SAXS diffractogram recorded at 33 °C with the 2D SAXS pattern shown as inset; (c) WAXS diffractogram of LC3-F doped with BON-LC3-F (1 wt%) recorded at 33 °C with the 2D WAXS pattern shown as inset. The red trace represents the fit of the wide-angle region with two Lorentzian functions (grey dashed traces); (d) SAXS diffractogram recorded at 33 °C with the 2D SAXS pattern shown as inset.

The calculated molecular diameter of LC3-F d_calc = 31.6 Å is very similar to the experimentally observed lattice parameter a = 32.3 Å, indicating that only little interdigitation of the semiperfluorinated chains takes place. The longer homologue LC7-F displayed similar WAXS and SAXS behavior, and thus a Col_ho phase (p6mm) with a = 40.5 Å and d_π–π = 3.50 Å was assigned (Fig. S47 and Table S3).

In contrast to the hosts LC3-F and LC7-F, the emitter BON-LC3-F showed only uncharacteristic textures under the POM (Fig. S43), which were shearable and thus indicative of a mesophase. As we anticipated potential problems with stability, transition temperatures, and enthalpies were extracted from the 1st heating and cooling cycle of the DSC. Upon the 1st cooling, BON-LC3-F displayed a clearing transition at 343.6 °C. Upon subsequent cooling, the mesophase reappeared at 343.6 °C, and a glass transition at −17.5 °C was detected, resulting in a large phase width (ΔT = 364 K, Fig. S45b and Table S2). In the WAXS, at 116 °C, BON-LC3-F displayed three distinct small-angle reflections and two diffuse wide-angle reflections, i.e., the halo (d_halo = 5.30 Å) and the π–π reflection (d_π–π = 3.53 Å) (Fig. S48 and Table S3). As discussed above, the small-angle reflections were assigned as (10), (11), and (20) reflections of a Col_ho phase (p6mm) with a lattice parameter a = 29.0 Å.^40,42,43 Interestingly, when temperature-dependent XRD experiments were carried out (Table S4 and Fig. S48c, d), the intensity of the (10) reflection decreased with decreasing temperature and disappeared below 69 °C, whereas the (11), (20) reflections and the two diffuse wide-angle reflections remained. Such behavior was rather unexpected for Col_ho. However, it should be noted that Donnio and co-workers reported that ionic mesogens having a Col_ho mesophase had a missing (10) reflection when large, perfluorinated anions were used.⁴⁵ Presumably, the molecular form factor in BON-LC3-F was reduced and finally disappeared.

Mesomorphic properties of mixtures

We next explored mixtures of materials containing 1 wt% of emitter BON-LC3-F. Under POM, the mixture LC3-F : BON-LC3-F (99 : 1, w/w) displayed shearable textures with homeotropic areas and line defects, which are characteristic for the host LC3-F (Fig. S49).^37–39 The mixture LC7-F : BON-LC3-F (99 : 1, w/w) containing the larger host LC7-F also showed homeotropic areas with line defects as well as fan-shaped textures upon heating (or cooling) (Fig. S50). Both mixtures displayed distinct phase transitions without phase separation.

Investigation of the mixture LC3-F : BON-LC3-F (99 : 1, w/w) by DSC revealed that upon 2nd cooling, there were clearing and glass transitions at 176.1 and −52.9 °C, respectively (Fig. 2b and Table S2). Both transition temperatures are higher as compared to those of LC3-F (Fig. 2c). The phase width (ΔT = 229 K) of the mixture is slightly smaller than the phase widths of the host LC3-F. The mixture LC7-F : BON-LC3-F (99 : 1, w/w) possesses a higher clearing temperature (237.2 °C) and a somewhat lower glass transition (−60.8 °C) (Fig. S51 and Table S2) resulting in a phase width (ΔT = 298 K), which that is identical with the temperature range of LC7-F yet much larger than for the mixture LC3-F : BON-LC3-F (99 : 1, w/w).

The 2D WAXS pattern of an oriented fiber showed orthogonal wide- and small-angle reflections for LC3-F doped with 1 wt% BON-LC3-F (Fig. 3c and Table S3). The two broad reflections in the wide-angle region were fitted by Lorentz functions, which provided d_halo = 4.51 Å and d_π–π = 3.47 Å. In the SAXS, characteristic (10), (11), (20), and (21) reflections of a Col_ho phase (p6mm) with a lattice parameter of a = 32.1 Å were detected.^46–48 Thus, the mixture LC3-F : BON-LC3-F (99 : 1, w/w) possesses very similar mesophase geometries as compared to the pure host LC3-F (Fig. S52 and Table S3). The mixture LC7-F : BON-LC3-F (99 : 1, w/w) also showed a Col_ho phase (p6mm) (Fig. S53 and Table S3). Again, the lattice parameter of the mixture and the pure host LC7-F were identical (a = 40.5 Å). These results indicate that the mesomorphism of the host remains almost untouched by the guest.

Photophysical properties of LC3-F and LC7-F

The absorption spectrum of host LC3-F in toluene displays two intense bands at λ_abs = 327 nm (ε = 3.9 × 104 M⁻¹ cm⁻¹) and 391 nm (ε = 3.1 × 104 M⁻¹ cm⁻¹) (Fig. 4 and Table S5). The low-energy band was assigned to the SRCT transition of the substituted DOBNA core and is situated similarly to those of DOBNA-Ph (λ_abs = 395 nm)³⁴ and LC8 (λ_abs = 393 nm),²⁶ but red-shifted as compared to DOBNA (λ_abs = 376 nm).³⁴ The high-energy band was assigned to a locally excited state of the mesogenic groups of LC3-F, which was also found in LC8 (λ_abs = 340 nm) and DOBNA-Ph (λ_abs = 320 nm), but is absent in DOBNA. Upon excitation of LC3-F at 330 nm in degassed toluene, a deep blue narrowband emission at λ_PL = 408 nm (full-width at half maximum, FWHM = 26 nm) was observed. The Stokes shift is small (18 nm), reflecting the small degree of structural reorganization in the excited state. The Φ_PL is 56% (Table 1 and Table S5).^49–51 The PL band is nearly identical to that of LC8 (λ_PL = 408 nm, Φ_PL = 63%),²⁶ and DOBNA-Ph (λ_PL = 410 nm, Φ_PL = 60%),³⁴ and is red-shifted compared to that of DOBNA (λ_PL = 398 nm, Φ_PL = 72%). Time-resolved PL measurements revealed monoexponential decay kinetics, with a τ_p = 4.47 ns (Fig. S54b and Table S5), which is slightly slower than for LC8 (τ_p = 3.65 ns) and for DOBNA-Ph (τ_p = 5.61 ns) and DOBNA (τ_p = 1.07 ns). The photophysical properties of LC3-F are similar to the alkoxy analogue LC8.²⁶

Fig. 4 Steady-state absorption (black trace) and emission (blue trace) spectra of LC3-F in degassed toluene (c = 0.02 mM, λ_exc = 320 nm).

Table 1 Photophysics of hosts, emitter and host–emitter mixtures

Compound	Conditions	λPL/nm	FWHM/nm (FWHM/eV)	ΦPL/%	τp,avg/ns	τd/µs
a No delayed emission observed.b The quantum yield could not be determined due to the low absorption of the film.c LC8 in doped (1 wt% BCzBN) polystyrene (PS) films.d BON-LC12 in PS film (1 wt%).e BON-LC3-F in PS film (0.1 wt%).f BON-LC3-F in a PS film in LC3-F (1 wt%) or LC7-F (1 wt%).
LC8	Toluene	408	24 (0.18)	63	3.65	n.d.^a
LC3-F	Toluene	408	26 (0.19)	56	4.47	n.d.^a
LC7-F	Toluene	408	27 (0.20)	59	4.14	n.d.^a
BON-LC12	Toluene	466	23 (0.13)	77	6.75	n.d.^a
BON-LC3-F	Toluene	461	23 (0.13)	76	5.85	n.d.^a
LC8	Neat	472	70 (0.40)	19	16.62	n.d.^a
LC3-F	Neat	460	63 (0.36)	19	15.42	n.d.^a
LC7-F	Neat	457	59 (0.34)	20	13.29	n.d.^a
BON-LC12	Neat	544	77 (0.33)	39	31.96	n.d.^a
BON-LC3-F	Neat	528	72 (0.31)	30	30.18	35.0
LC8^c	PS	476	14 (0.08)	41	8.19	n.d.^a
BON-LC12^d	PS	467	45 (0.25)	90	9.85	107.8
BON-LC3-F^e	PS	463	65 (0.35)	n.d.^b	5.37	142.6
BON-LC3-F:LC3-F^f	PS	477	41 (0.22)	54	15.20	52.05
BON-LC3-F:LC7-F^f	PS	478	43 (0.23)	49	16.99	56.1

---

Solvent-dependent steady-state absorption and PL measurements revealed that there is effectively no change in the absorption spectrum with solvent polarity, reflecting the non-polar ground state of LC3-F (Fig. S54c and Table S6), while there is a small bathochromic shift from λ_PL = 402 nm (MCH) to 410 nm (butyronitrile) that is characteristic of an excited state of SRCT character (Fig. 5a). In comparison with the PL spectrum of LC8 (Fig. 5b), there is no low-lying LRCT state that becomes stabilized in polar media, and the emissive SRCT excited state is retained across all of these solvents in LC3-F.

Fig. 5 (a) Solvatofluorochromism of LC3-F in different solvents (c = 0.01 mM, λ_exc = 330 nm); (b) solvatofluorochromism of LC8 in different solvents (c = 0.02 mM, λ_exc = 350 nm).

Next, we determined the S₁ and T₁ levels of LC3-F from the onsets of the steady-state PL and phosphorescence spectra in frozen 2-MeTHF glass at 77 K (Fig. S54e). The obtained energies (S₁ = 3.12 eV, T₁ = 2.79 eV) result in a ΔE_ST of 0.32 eV (Table S5).⁴⁹ Compared to LC8, the T₁ state of LC3-F is destabilized by 0.09 eV. Moreover, comparison of the S₁ and T₁ levels (determined from the maxima) of LC3-F (S₁ = 3.05 eV, T₁ = 2.68 eV) with alkoxy derivative LC8 (T₁ = 2.41 eV) and DOBNA-Ph (T₁ = 2.71 eV, EtOH, 77 K) clearly indicated that the T₁ level of LC3-F is only destabilized by 0.03 eV compared to DOBNA-Ph, whereas LC8 is destabilized by 0.29 eV. The derivative LC7-F has very similar photophysical properties in degassed toluene solution (λ_PL = 408 nm, FWHM = 27 nm, Φ_PL = 59%) and identical S₁ and T₁ levels to those of LC3-F (Table 1, Fig. S55, Table S5). Thus, the nature of the chain exerts no influence on the photophysical properties, in agreement with the previous observations made for the DOBNA alkoxy series LCn.²⁶ The solution studies demonstrated that monomolecular species of LC3-F and LC7-F maintain the SRCT character of the S₁ state in contrast to derivatives carrying alkoxy side chains, where the S₂ state of LRCT character is stabilized. Further, LC3-F and LC7-F possess higher T₁ levels, thus increasing their suitability as host materials for blue emitters.

Next, the photophysical properties of the bulk phase of LC3-F, LC7-F were examined to determine whether the observed trends in solutions of LC3-F, LC7-F are also maintained in the bulk LC phase in the presence of columnar aggregates in thin films. It should be noted that thin films of LC3-F and LC7-F were prepared by drop-casting CHCl₃ solutions rather than spin-coating because the latter method did not produce stable films, presumably due to the fluorophobic effect of the side chains with the quartz and sapphire substrates. The absorption spectrum of the thin film of LC3-F showed two bands at λ_abs = 330 and 396 nm, respectively (Table S5, entry 4), which are blue-shifted by 5 and 4 nm compared to the solution spectrum (Table S5, entry 1). Despite the strong aggregation in the Col_ho phase, very little electronic interaction in the ground state was found.^52,53 Compared to the solution-state PL spectrum, that in the solid state is bathochromically shifted and broadened (λ_PL = 460 nm, FWHM = 63 nm, Fig. 6a and Table 1). This implies excimer emission.^20,27,54 Compared to the Φ_PL in solution, the value of the film is much lower at 19%, presumably due to significant ACQ.⁵⁴ The results are in good agreement with those observed for LC8.²⁶

Fig. 6 (a) Absorption (black trace) and steady-state emission (blue trace, λ_exc = 320 nm) of LC3-F in a neat film, inset shows the film under UV light; (b) steady-state absorption (black trace) and emission (blue trace) (λ_exc = 410 nm) spectra of BON-LC3-F in degassed toluene; (c) photoluminescence spectrum of BON-LC3-F in polystyrene; (d) absorption (black trace) and steady-state emission (blue trace, λ_exc = 440 nm) of BON-LC3-F in a neat film, the inset shows the film under UV light.

Time-resolved PL measurements of LC3-F by TCSPC (Fig. S57b and Table S8) showed a triexponential decay of the PL emission, with an average prompt lifetime τ_p,avg = 15.42 ns (Table 1) that is slightly shorter than that of LC8 (τ_p,avg = 16.62 ns).²⁶ Excited-state levels of LC3-F in the bulk (S₁ = 2.94 eV, T₁ = 2.62 eV) are stabilized by 0.18 and 0.17 eV, respectively, as compared to the energy levels in solution, reflecting the state energies of the excimer. Unfortunately, the resulting ΔE_ST of 0.32 eV is large and thus no delayed emission was detected. In comparison to LC8, the T₁ level of LC3-F in the bulk is destabilized by 0.05 eV. The photophysical properties of the homologous derivative LC7-F (λ_PL = 457 nm, FWHM = 59 nm, Φ_PL = 20%, Table 1 and Fig. S58) also reflect excimer formation in the Col_ho phase.^20,27,54 The T₁ level of LC7-F is destabilized by 0.11 eV as compared to that of LC8. Thus, the thin film studies with LC3-F and LC7-F revealed a higher energy T₁ level in the bulk Col_ho phase (T₁ = 2.62 and 2.65 eV, respectively), indicating that LC3-F and LC7-F are suitable host materials for the MR-TADF emitter BON-LC3-F (T₁ = 2.51 eV) in contrast to their LCn DOBNA counterparts, which contain alkoxy chains.

Photophysical properties of BON-LC3-F

In dilute toluene solution, BON-LC3-F exhibits a strong absorption at λ_abs = 446 nm and a sharp sky-blue PL emission at λ_PL = 461 nm (FWHM = 23 nm) associated with a Φ_PL of 76% (Table 1, Fig. S56, Table S5).^49–51 Comparison to the parent B-OCz emitter⁵⁵ revealed that both the absorption and emission maxima are red-shifted, by 21 and 20 nm, respectively. Time-resolved PL measurements in degassed toluene showed a τ_p,avg of 5.86 ns (Fig. S56b). Comparison to BON-LC showed that the emission is slightly blue-shifted (λ_PL = 466 nm, FWHM = 23 nm, Φ_PL = 77%),²³ but otherwise this compound shows very similar behavior.

Solvatochromism measurements support the assignment of the SRCT character to the emissive excited state (Fig. S56c, d and Table S6). To investigate whether BON-LC3-F is a suitable MR-TADF emitter, S₁ and T₁ energies were determined from the onsets of the steady-state PL and phosphorescence spectra in frozen 2-MeTHF glass at 77 K (S₁ = 2.75 eV, T₁ = 2.57 eV) (Fig. S56e). Thus, the ΔE_ST is 0.18 eV (Table S5), which is 0.11 eV larger than that for BON-LC12 (ΔE_ST = 0.07 eV).^23,56,57 These results suggest that BON-LC3-F can act as a suitable dopant with the hosts LC3-F and LC7-F. No delayed fluorescence was detected for BON-LC3-F in solution, similar to many reported MR-TADF emitters that do not show delayed emission in solution, presumably due to competing non-radiative decay.^18,23,26 Therefore, a doped film of BON-LC3-F in polystyrene (PS, 0.1 wt%) was studied. In this matrix, there is an intense PL signal at λ_PL = 463 nm that is significantly broadened (FWHM = 65 nm, Fig. S60a), presumably caused the strong tendency of the dye molecules to aggregate (Fig. 6c and Table 1). From the energy levels (S₁ = 2.82 eV, T₁ = 2.58 eV), a slightly larger ΔE_ST of 0.24 eV was determined, due to the S₁ level being destabilized (Fig. S60c and Table S5). Prompt lifetime measurements by TCSPC showed a triexponential decay with τ_p,avg = 5.37 ns, while MCS TR PL measurements revealed delayed emission with a lifetime of τ_d = 142.6 µs, which was completely quenched by oxygen (Table 1 and Fig. S62b, d). Temperature-dependent time-resolved PL measurements by MCS indicated that the magnitude of the delayed emission increases with increasing temperature, a hallmark of TADF behavior (Fig. S60e). Thus, BON-LC3-F is an MR-TADF emitter.^58–60

In comparison with BON-LC12, the PL band of BON-LC3-F is broadened due to the increased tendency of the semiperfluorinated derivative towards aggregation. The delayed lifetime of BON-LC3-F (τ_d = 142.6 µs) is significantly longer than for BON-LC12 (τ_d = 107.8 µs), reflecting its larger ΔE_ST.²³ These results demonstrate that the photophysical properties of the MR-TADF emitter are maintained in the presence of the semiperfluorinated side chains.

Drop-cast thin films of BON-LC3-F showed an absorption band at λ_abs = 452 nm (Fig. 6d and Table 1), which is slightly bathochromically shifted and somewhat broadened compared to that in toluene, which indicates that there are only few intermolecular interactions in the ground state despite the strongly aggregated Col_ho mesophase.^52,53 In contrast to the sky-blue emission in toluene (λ_PL = 461 nm) or PS (λ_PL = 463 nm), a green emission (λ_PL = 528 nm) was observed in the neat film (Fig. 6d and Table S5). The large Stokes shift (82 nm), the significant PL broadening (FWHM = 72 nm), and the much reduced Φ_PL of 30% compared to the data acquired in toluene solution all point to excimer emission. This is similar behavior to that observed for the neat film photophysics of other MR-TADF emitters.^20,27,54 Time-resolved PL measurements by TSCPC reveal a triexponential decay with a τ_p,avg = 30.18 ns (Fig. S59b and Table S7). The ΔE_ST is reduced to 0.08 eV, and the τ_d is 35 µs (Table S5). Not surprisingly, the PL is sensitive to O₂, the temperature-dependent time-resolved PL measurements also document TADF behavior (Fig. S59d and Table S5). Above 300 K, both the intensity of the delayed lifetime as well as the PL intensity decrease, which reflects increased temperature-dependent non-radiative decay that may originate from increased vibrational relaxation in the mesophase (T_g = 256 K) of BON-LC3-F.⁶¹ In comparison with BON-LC12 (λ_PL = 544 nm, FWHM = 77 nm, Φ_PL = 39%), BON-LC3-F showed similar photophysical properties as a neat film, with a small observed blue-shift of the PL maximum. Thus, in both molecules, the photophysical properties of the neat film reflect excimer formation in the strongly aggregated Col_ho phase. Despite these similarities, it should be emphasized that a neat film of BON-LC12 does not show TADF, while that of BON-LC3-F does.

Photophysical properties of the host–emitter mixtures

A qualitative comparison of the absorption spectra of the host LC3-F in the bulk state and the absorption of the emitter BON-LC3-F in toluene (Table 1) revealed that at the chosen excitation wavelengths λ_exc = 379 nm, only the host LC3-F should absorb and thus a direct excitation of the emitter does not take place (Fig. 7a).⁶² Moreover, the PL band of the host LC3-F and the absorption band of the emitter BON-LC3-F show sufficient overlap to permit an efficient energy transfer from host to emitter.⁶³ The film of the mixture LC3-F:BON-LC3-F (containing 1 wt% of the emitter BON-LC3-F) displays a sky blue emission (λ_PL = 477 nm, FWHM = 41 nm, Φ_PL = 54%, Fig. 7b and Table 1), which was assigned to the emission of BON-LC3-F. In addition, the emission spectrum of the mixture shows a small shoulder at λ_PL = 450 nm, which was assigned to emission from the host LC3-F. It should be noted that there is no excimer emission at λ_PL = 530 nm resulting from aggregation of the emitter BON-LC3-F. Presumably, the emitter is well integrated in the liquid crystalline matrix of the host, which effectively suppresses emitter aggregation. In comparison to the PL spectrum of BON-LC3-F in toluene (λ_PL = 461 nm, FWHM = 23 nm), the PL band of the mixture is bathochromically shifted by 16 nm and somewhat broadened, indicating weak interactions between host and emitter, as well as emitter/emitter interactions, in the mixture.^64,65 Thus, in the mixture, there is almost complete energy transfer from the host LC3-F to the emitter BON-LC3-F. Based on the significantly increased Φ_PL of the mixture to 54% as compared to the host LC3-F (Φ_PL = 19%), the energy transfer occurs almost certainly via FRET.^{62,63,66–68}

Fig. 7 (a) Comparison of the absorption spectra of the host material LC3-F (dashed line) and the emitter BON-LC3-F (in toluene, c = 0.02 mM, black) with the emission spectrum of neat LC3-F (λ_exc = 350 nm, blue); (b) PL-spectrum of a LC3-F : BON-LC3-F (99 : 1, w/w) blend (λ_exc = 375 nm, light blue) compared to that of the pure host LC3-F (λ_exc = 350 nm, blue dashed line), the inset shows the film under UV light; (c) possible Förster resonance energy transfer (FRET) from the host LC3-F to the emitter BON-LC3-F; (d) temperature-dependent MCS-lifetime measurements.

From the PL and phosphorescence spectra of the mixture at 77 K, the energy levels S₁ = 2.74 eV and T₁ = 2.60 eV were extracted. The ΔE_ST of the mixture of 0.14 eV is similar to the ΔE_ST value of the emitter BON-LC-F₃ (ΔE_ST = 0.18 eV in toluene) and significantly different from that of the neat host LC3-F (ΔE_ST = 0.32 eV in the bulk), further supporting the assignment of the PL bands (for details, see Table S5). Lifetime measurements showed a biexponential decay of the prompt fluorescence (τ_p,avg = 15.20 ns) and a delayed fluorescence with single exponential decay (τ_d = 52.05 µs), which was quenched by O₂ (Fig. 8a and b). Temperature-dependent time-resolved PL measurements confirm that the mixture exhibits TADF^58–60 (Fig. 8c).^61,69–71 Upon increasing the emitter concentration in the host LC3-F (Fig. 8d and Table S8), the PL bands bathochromically shifted and broadened from λ_PL = 477 nm, FWHM = 41 nm (1 wt%) to λ_PL = 501 nm, FWHM = 68 nm (20 wt%), resulting from the increased intermolecular emitter–emitter interactions in the columnar mesophase of the mixture (for further details, see Fig. S62).

Fig. 8 Photophysical properties of the BON-LC3-F : LC3-F (99 : 1, w/w) blend (λ_exc = 375 nm; for TCSPC and MCS measurements: λ_exc = 379 nm): (a) time-correlated single photon counting (TCSPC) lifetime measurement (blue curve) with a triple-exponential decay fit (red curve) and instrument response function (IRF, grey); (b) microsecond-scale time-resolved emission (MCS) lifetime measurement in air (black curve) and under vacuum (blue curve) with a mono-exponential decay fit (red curve) and IRF (grey); (c) temperature-dependent PL-spectra, the inset shows the integrated intensity of the BON-LC3-F emission band (integration range: 450–650 nm); (d) steady state emission of BON-LC3-F (1, 2, 5, 10, 20 wt%) doped into LC3-F. Steady-state emission of neat BON-LC3-F (green dashed trace) and neat LC3-F (blue dashed traces) is given for comparison.

The analogous mixture LC7-F : BON-LC3-F (99 : 1, w/w) shows very similar photophysical behavior, with a distinct PL band from the emitter at λ_PL = 478 nm (FWHM = 43 nm, Φ_PL = 49%, τ_d = 56.1 µs) (Table 1 and Fig. S62). Thus, the excitation energy is likewise transferred via FRET from the host LC7-F to the emitter BON-LC3-F.

Packing model of the mixtures

We surmised that the observed photophysical and mesomorphic properties may be rationalized by a distinct packing model of the mesophase. In principle, for binary mixtures of columnar liquid crystals with similar molecular scaffolds and lattice parameters, three different packing geometries are possible (Fig. 9a). The first is one where host and emitter molecules may undergo a macroscopic phase separation similar to the arrangement of block copolymers,^72,73 if host and emitter are non-miscible. The second is one where the host and emitter are partially miscible and intercolumnar phase separation may occur, i.e., columns of host are mixed with columns of emitter, resulting in intercolumnar mixed phases.^20,72,73 Finally, if host and emitter are fully miscible, each column contains mixtures of host and emitter, either in an alternating manner or blockwise separated.^20,74 Because the mixtures contain only 1 wt% of the emitter BON-LC3-F, the lattice parameters of the mixtures LC3-F: BON-LC3-F and LC7-F:BON-LC3-F are mostly governed by the host LC3-F or LC7-F. That there is the absence of any host excimer emission in the mixture implies that the first two packing models are highly unlikely and strongly supports a mixed columns model (Fig. 9b). The relatively broad PL bands of BON-LC3-F in these mixtures can be rationalized by weak interactions between emitter BON-LC3-F and hosts LC3-F (or LC7-F) or weak interactions between two emitter molecules closely located within a column. It should be noted that the semiperfluoroalkyl side chains have two distinct functions: (i) they are “electronically neutral” compared to the strongly electron-donating alkoxy chains. The electronic isolation is caused by the alkyl subunit. Thus, the S₁ and T₁ states are only slightly affected, and the ΔE_ST of both hosts LC3-F, LC7-F and emitter BON-LC3-F, respectively, remain similar to the parent non-mesogenic dyes DOBNA and B-O-Cz, respectively, whereas the ΔE_ST values for LC8 and BON-LC12 increased (Table S7). (ii) The semiperfluoroalkyl chains strengthen nanoseggregation through the perfluorinated subunit. This results in increased mesophase stabilities and temperature ranges for both neat films of these compounds as well as mixtures containing semiperfluorinated side chains rather than alkoxy chains (see Fig. 2c).

Fig. 9 (a) Possible arrangements for binary mixtures of two columnar liquid crystals, A and B; (b) proposed packing model of the mixtures BON-LC3-F : LC3-F (99 : 1, w/w) and BON-LC3-F : LC7-F (99 : 1, w/w) based on X-ray diffraction data and the photophysical properties of the mixtures.

Conclusion

The first room-temperature columnar MR-TADF host–guest system is reported. This is based on a columnar DOBNA-derived host LC3-F (or LC7-F), and a columnar emitter BON-LC3-F. Semiperfluorinated side chains decorating both host and emitter improved nanosegregation and stabilized the columnar mesophases as compared to the corresponding host and emitter with alkoxy side chains. Temperature ranges of the columnar mesophase increased by 69 K for LC3-F, 193 K for LC7-F, and 238 K for BON-LC3-F as compared to the counterparts LC8 and BON-LC12 that contain alkoxy side chains. In the mixtures, no phase separation was observed by POM, and almost identical XRD data of neat hosts and host emitter mixtures revealed that the emitter BON-LC3-F blends perfectly into the host LC3-F (or LC7-F) mesophase.

Moreover, the semiperfluorinated side chains gratifyingly do not stabilize the excited states of the host, resulting in higher T₁ levels and smaller ΔE_ST as compared to analogous hosts carrying alkoxy side chains. Whereas the neat host LC3-F (or LC7-F) showed no TADF and the neat emitter BON-LC3-F showed only very weak TADF, mixtures LC3-F : BON-LC3-F (99 : 1, w/w) and LC7-F : BON-LC3-F (99 : 1, w/w) showed MR-TADF, with delayed lifetimes of τ_d = 52.0 µs and τ_d = 56.1 µs, respectively, due to energy transfer from host to emitter via FRET. Despite the large excess of the host material in the mixture (99 wt%), the PL bands of the mixtures originate from the 1 wt% of emitter. Consequently, with increasing relative amount of emitter, the Φ_PL increases from 54% (1 wt%) to 68% (5 wt%). The perfect integration of the emitter in the host was also deduced from the fact that no excimer emission of the emitter BON-LC3-F was detected in these mixtures.

Such “turn-on” TADF by mixing columnar hosts with columnar emitters provides a versatile approach by choosing suitable combinations of host (determining the LC properties) and emitter (determining the photophysical properties), while ACQ of columnar self-assembled dye molecules was overcome by high dilution of the emitter in the host. Such approach should be amenable to other compound classes of host and emitter useful for various optoelectronic applications. Future work must demonstrate whether the beneficial effect of fluorophobic self-assembly reported for organic transistors is also true for OLED devices.

Author contributions

J. A. K. conceived and supervised the project and performed the synthesis as well as characterization. T. G. performed part of the synthesis. T. M. helped with characterization, computation and manuscript preparation. S. L. and A. Z. wrote the manuscript. A. Z. checked the data. E. Z.-C. and S. L. supervised the project and the manuscript preparation.

Conflicts of interest

There are no conflicts of interest.

Data availability

Additional raw data files are available from the corresponding author upon reasonable request. The research data supporting this publication can be accessed at https://doi.org/10.17630/45e47c51-fa07-455e-98d5-22f2328a37ba.

The data supporting the findings of this study are available within the article and its supplementary information (SI). Supplementary information: ¹H NMR and ¹³C NMR spectra, HRMS, and GPC traces; details of X-ray crystallography, DSC, and POM; supplemental photophysical data. See DOI: https://doi.org/10.1039/d5tc03809j.

Acknowledgements

Generous financial support by the Studienstiftung des deutschen Volkes (PhD fellowship for J. A. K.), the Ministerium für Wissenschaft, Forschung und Kunst des Landes Baden Württemberg, the Carl Schneider Stiftung Aalen (shared instrumentation grant), the Universität Stuttgart (Global glimpse travel grant for J. A. K.) and the DFG (INST 41/897-1 FUGG for 700 MHz-NMR, INST 41/1136-1 FUGG for LC-Orbitrap-MS: Exactive Plus Orbitrap MS System and INST 41/1135-1 FUGG for GC-Orbitrap-MS: Exactive GC Orbitrap MS System) are gratefully acknowledged. The St Andrews team would also like to thank EPSRC (EP/Z535291/1, EP/W007517/1, and EP/W015137/1) for financial support.

Notes and references

1. M. E. Prevot, J. P. Vanegas and E. Hegmann, Emissive Nanomaterials and Liquid Crystals, in 21st Century Nanoscience – A Handbook, CRC Press, Boca Raton, 2020.
2. A. Kumar and G. Singh, J. Mol. Liq., 2023, 386, 122607.
3. Y. Wang, J. Shi, J. Chen, W. Zhu and E. Baranoff, J. Mater. Chem. C, 2015, 3, 7993–8005.
4. A. Surampudi, G. Zhang, R. Singh, G. Faulkner, D. C. O’Brien, M. J. Booth and S. M. Morris, Crystals, 2023, 13, 1615.
5. J. Uchida, B. Soberats, M. Gupta and T. Kato, Adv. Mater., 2022, 34, 2270171.
6. T. Wöhrle, I. Wurzbach, J. Kirres, A. Kostidou, N. Kapernaum, J. Litterscheidt, J. C. Haenle, P. Staffeld, A. Baro and F. Giesselmann, Chem. Rev., 2016, 116, 1139–1241.
7. S. Laschat, A. Baro, N. Steinke, F. Giesselmann, C. Hägele, G. Scalia, R. Judele, E. Kapatsina, S. Sauer and A. Schreivogel, Angew. Chem., Int. Ed., 2007, 46, 4832–4887.
8. R. De and S. K. Pal, Chem. Commun., 2023, 59, 3050–3066.
9. S. Sergeyev, W. Pisula and Y. H. Geerts, Chem. Soc. Rev., 2007, 36, 1902–1929.
10. R. K. Gupta and A. A. Sudhakar, Langmuir, 2018, 35, 2455–2479.
11. Y. Zhu, S. Zeng, B. Li, A. J. McEllin, J. Liao, Z. Fang, C. Xiao, D. W. Bruce, W. Zhu and Y. Wang, ACS Appl. Mater. Interfaces, 2022, 14, 15437–15447.
12. A. F. Suleymanova, M. Z. Shafikov, A. C. Whitwood, R. Czerwieniec and D. W. Bruce, J. Mater. Chem. C, 2021, 9, 6528–6535.
13. A. F. Suleymanova and M. Z. Sha, Phys. Chem. Chem. Phys., 2022, 24, 22115–22121.
14. D. Chen, F. Tenopala-Carmona, J. A. Knöller, A. Mischok, D. Hall, S. Madayanad Suresh, T. Matulaitis, Y. Olivier, P. Nacke and F. Gießelmann, Angew. Chem., Int. Ed., 2023, 62, e202218911.
15. G. Schweicher, G. Gbabode, F. Quist, O. Debever, N. Dumont, S. Sergeyev and Y. H. Geerts, Chem. Mater., 2009, 21, 5867–5874.
16. R. I. Gearba, D. V. Anokhin, A. I. Bondar, W. Bras, M. Jahr, M. Lehmann and D. A. Ivanov, Adv. Mater., 2007, 19, 815–820.
17. B. He, Q. Zhong, Q. Dong, X. Yang, S. J. Cowling, W. Qiao, D. W. Bruce, W. Zhu, P. Duan and Y. Wang, Mater. Horiz., 2024, 11, 1251–1260.
18. J. M. Dos Santos, D. Hall, B. Basumatary, M. Bryden, D. Chen, P. Choudhary, T. Comerford, E. Crovini, A. Danos and J. De, Chem. Rev., 2024, 124, 13736–14110.
19. M. Y. Wong and E. Zysman-Colman, Adv. Mater., 2017, 29, 1605444.
20. B. Adelizzi, P. Chidchob, N. Tanaka, B. A. Lamers, S. C. Meskers, S. Ogi, A. R. Palmans, S. Yamaguchi and E. W. Meijer, J. Am. Chem. Soc., 2020, 142, 16681–16689.
21. A. F. Suleymanova, J. R. Fortwengler, X. Chen, R. Czerwieniec, Y. Wang, M. Z. Shafikov and D. W. Bruce, J. Mater. Chem. C, 2025, 13, 12348–12364.
22. L. G. Franca, P. L. Dos Santos, P. Pander, M. G. Cabral, R. Cristiano, T. Cazati, A. P. Monkman, H. Bock and J. Eccher, ACS Appl. Electron. Mater., 2022, 4, 3486–3494.
23. J. A. Knöller, B. Sönmez, T. Matulaitis, A. K. Gupta, E. Zysman-Colman and S. Laschat, Chem. Commun., 2024, 60, 4459–4462.
24. N. Li, F. Ni, X. Lv, Z. Huang, X. Cao and C. Yang, Adv. Opt. Mater., 2022, 10, 2101343.
25. T. Chatterjee and K.-T. Wong, Adv. Opt. Mater., 2019, 7, 1800565.
26. J. A. Knöller, F. Müller, T. Matulaitis, J. M. Dos Santos, A. K. Gupta, E. Zysman-Colman and S. Laschat, Chem. Sci., 2024, 15, 18022–18030.
27. S. Xu, Q. Yang, Y. Zhang, H. Li, Q. Xue, G. Xie, M. Gu, J. Jin, L. Huang and R. Chen, Chin. Chem. Lett., 2021, 32, 1372–1376.
28. C. Tschierske, Top. Curr. Chem., 2012, 318, 1–108.
29. B. J. Jung, K. Lee, J. Sun, A. G. Andreou and H. E. Katz, Adv. Funct. Mater., 2010, 20, 2930–2944.
30. K. Miao, G. J. Chae, X. Wu, Q. Shu, X. Zhu, B. Sun, J. Fan and S. Cho, RSC Adv., 2016, 6, 29164–29171.
31. H.-G. Jeong, B. Lim, D. Khim, M. Han, J. Lee, J. Kim, J.-M. Yun, K. Cho, J.-W. Park and D.-Y. Kim, Adv. Mater., 2013, 25, 6416–6422.
32. J. C. Haenle, Y. Stöckl, R. Forschner, E. Haenle and S. Laschat, ChemPhysChem, 2018, 19, 2758–2767.
33. P. Ehni, S. M. Bauch, P. M. Becker, W. Frey, A. Zens, J. Kästner, Y. Molard and S. Laschat, Phys. Chem. Chem. Phys., 2022, 24, 21617–21630.
34. H. Hirai, K. Nakajima, S. Nakatsuka, K. Shiren, J. Ni, S. Nomura, T. Ikuta and T. Hatakeyama, Angew. Chem., Int. Ed., 2015, 54, 13581–13585.
35. X. Cai, J. Xue, C. Li, B. Liang, A. Ying, Y. Tan, S. Gong and Y. Wang, Angew. Chem., Int. Ed., 2022, 61, e202200337.
36. T. Wang, X. Yin, X. Cao and Y. Yang, Angew. Chem., Int. Ed., 2023, 62, e202312451.
37. R. Forschner, J. Knelles, K. Bader, C. Müller, W. Frey, A. Köhn, Y. Molard, F. Giesselmann and S. Laschat, Chem. – Eur. J., 2019, 25, 12966–12980.
38. J. A. Lugger, D. J. Mulder, S. Bhattacharjee and R. P. Sijbesma, ACS Nano, 2018, 12, 6714–6724.
39. G. Zucchi, B. Donnio and Y. H. Geerts, Chem. Mater., 2005, 17, 4273–4277.
40. U. Dahn, C. Erdelen, H. Ringsdorf, R. Festag, J. H. Wendorff, P. A. Heiney and N. C. Maliszewskyj, Liq. Cryst., 1995, 19, 759–764.
41. V. Percec, G. Johansson, G. Ungar and J. Zhou, J. Am. Chem. Soc., 1996, 118, 9855–9866.
42. M. Campos-Vallette and M. Rey-Lafon, J. Mol. Struct., 1983, 101, 23–45.
43. N. Terasawa, H. Monobe and K. Kiyohara, Liq. Cryst., 2007, 34, 311–324.
44. V. Percec, D. Schlueter, Y. K. Kwon, J. Blackwell, M. Moeller and P. J. Slangen, Macromolecules, 1995, 28, 8807–8818.
45. J. Dai, K.-Q. Zhao, B.-Q. Wang, P. Hu, B. Heinrich and B. Donnio, J. Mater. Chem. C, 2020, 8, 4215–4225.
46. N. Godbert, A. Crispini, M. Ghedini, M. Carini, F. Chiaravalloti and A. Ferrise, Appl. Crystallogr., 2014, 47, 668–679.
47. P. Ehni, K. Guy, M. Ebert, S. Beardsworth, K. Bader, R. Forschner, A. Bühlmeyer, N. Dumait, C. Roiland and Y. Molard, Dalton Trans., 2018, 47, 14340–14351.
48. S. K. Prasad, D. S. Rao, S. Chandrasekhar and S. Kumar, Mol. Cryst. Liq. Cryst., 2003, 396, 121–139.
49. K. Shizu and H. Kaji, Commun. Chem., 2022, 5, 53.
50. T. Hatakeyama, K. Shiren, K. Nakajima, S. Nomura, S. Nakatsuka, K. Kinoshita, J. Ni, Y. Ono and T. Ikuta, Adv. Mater., 2016, 28, 2777–2781.
51. A. Pershin, D. Hall, V. Lemaur, J.-C. Sancho-Garcia, L. Muccioli, E. Zysman-Colman, D. Beljonne and Y. Olivier, Nat. Commun., 2019, 10, 597.
52. F. Würthner, T. E. Kaiser and C. R. Saha-Möller, Angew. Chem., Int. Ed., 2011, 50, 3376–3410.
53. S. Ma, S. Du, G. Pan, S. Dai, B. Xu and W. Tian, Aggregate, 2021, 2, e96.
54. D. Hall, S. M. Suresh, P. L. dos Santos, E. Duda, S. Bagnich, A. Pershin, P. Rajamalli, D. B. Cordes, A. M. Slawin and D. Beljonne, Adv. Opt. Mater., 2020, 8, 1901627.
55. J. Park, J. Lim, J. H. Lee, B. Jang, J. H. Han, S. S. Yoon and J. Y. Lee, ACS Appl. Mater. Interfaces, 2021, 13, 45798–45805.
56. R. Keruckiene, A. A. Vaitusionak, M. I. Hulnik, I. A. Berezianko, D. Gudeika, S. Macionis, M. Mahmoudi, D. Volyniuk, D. Valverde and Y. Olivier, J. Mater. Chem. C, 2024, 12, 3450–3464.
57. X. Wu, B.-K. Su, D.-G. Chen, D. Liu, C.-C. Wu, Z.-X. Huang, T.-C. Lin, C.-H. Wu, M. Zhu and E. Y. Li, Nat. Photonics, 2021, 15, 780–786.
58. H. Uoyama, K. Goushi, K. Shizu, H. Nomura and C. Adachi, Nature, 2012, 492, 234–238.
59. D. Sun, S. M. Suresh, D. Hall, M. Zhang, C. Si, D. B. Cordes, A. M. Slawin, Y. Olivier, X. Zhang and E. Zysman-Colman, Mater. Chem. Front., 2020, 4, 2018–2022.
60. D. Sun, E. Duda, X. Fan, R. Saxena, M. Zhang, S. Bagnich, X. Zhang, A. Köhler and E. Zysman-Colman, Adv. Mater., 2022, 34, 2110344.
61. J. M. dos Santos, D. Sun, J. M. Moreno-Naranjo, D. Hall, F. Zinna, S. T. Ryan, W. Shi, T. Matulaitis, D. B. Cordes and A. M. Slawin, J. Mater. Chem. C, 2022, 10, 4861–4870.
62. A. C. Chen, S. W. Culligan, Y. Geng, S. H. Chen, K. P. Klubek, K. M. Vaeth and C. W. Tang, Adv. Mater., 2004, 16, 783–788.
63. C.-P. Han, E. H.-C. Shi, C. A. Chen, C.-H. Hsu, C.-H. Chen, C.-C. Wu, Y.-T. Lee, T.-L. Chiu, J.-H. Lee and M. Leung, Adv. Opt. Mater., 2024, 12, 2303295.
64. M. Yang, I. S. Park and T. Yasuda, J. Am. Chem. Soc., 2020, 142, 19468–19472.
65. K. Stavrou, A. Danos, T. Hama, T. Hatakeyama and A. Monkman, ACS Appl. Mater. Interfaces, 2021, 13, 8643–8655.
66. T. Förster, Ann. Phys., 1948, 437, 55–75.
67. S.-H. Liu, M.-S. Lin, L.-Y. Chen, Y.-H. Hong, C.-H. Tsai, C.-C. Wu, A. Poloek, Y. Chi, C.-A. Chen and S. H. Chen, Org. Electron., 2011, 12, 15–21.
68. Y. Wu, C. Yan, X.-S. Li, L. H. You, Z.-Q. Yu, X. Wu, Z. Zheng, G. Liu, Z. Guo and H. Tian, Angew. Chem., 2021, 133, 24754–24762.
69. S. Wu, Y.-N. Hu, D. Sun, K. Wang, X.-H. Zhang and E. Zysman-Colman, Chem. Commun., 2024, 60, 2489–2492.
70. S. Wu, Y.-N. Hu, J. Wang, D. Sun, K. Wang, X.-H. Zhang and E. Zysman-Colman, J. Mater. Chem. C, 2024, 12, 6177–6184.
71. S. Wu, A. Kumar Gupta, K. Yoshida, J. Gong, D. Hall, D. B. Cordes, A. M. Slawin, I. D. Samuel and E. Zysman-Colman, Angew. Chem., Int. Ed., 2022, 61, e202213697.
72. W. Pisula, M. Kastler, D. Wasserfallen, J. W. Robertson, F. Nolde, C. Kohl and K. Müllen, Angew. Chem., Int. Ed., 2006, 45, 819–823.
73. T. Wöhrle, H. Taing, C. Schilling, S. H. Eichhorn and S. Laschat, Liq. Cryst., 2019, 46, 1973–1984.
74. E. O. Arikainen, N. Boden, R. J. Bushby, O. R. Lozman, J. G. Vinter and A. Wood, Angew. Chem., 2000, 112, 2423–2426.

---