Introduction

In recent years, chiral lanthanide supramolecules with circularly polarized luminescence (CPL) properties have attracted great attention due to their potential applications in three-dimensional (3D) displays1,2,3,4, information storage5,6,7,8,9, CPL probes10,11,12,13,14, asymmetric synthesis15,16,17. The luminescence dissymmetry factor (glum) along with the luminescence quantum yields (QYs) constitutes two crucial parameters when it comes to evaluating the performance of these materials. Due to the magnetic dipole transition characteristics of some lanthanide ions Ln(III), the glum values of the lanthanide luminescent materials are obviously higher than those of other luminescent systems, such as organic molecules18,19, polymers20,21, supramolecules22,23, and transition-metal complexes24,25. For instance, in 2008, Muller et al. reported a mononuclear chiral lanthanide complex Cs[Eu((+)-hfbc)4], whose glum value was as high as + 1.3826. This work surpassed the g-value of circularly polarized materials at the time; however, its luminescence efficiency in chloroform was only 0.6%. Therefore, the inability to simultaneously achieve high glum values and high luminescence quantum yields limits its application in lanthanide-based CPL materials. Subsequently, researchers developed a series of Eu-based circularly polarized luminescent materials with excellent luminescent performance. For example, Md.J. Islam et al27. reported the circularly polarized properties of [Eu(tfc)3(DPO)2] (H2O)2, with a g-value as high as 0.025, but a relatively low PLQY of 4.09%. They attributed the low PLQY to the metal-to-ligand back electron transfer (LMBET) process. Francesco et al28. reported three new Eu-based circularly polarized materials with excellent performance, all of which exhibited PLQYs of 30% in solution and glum of approximately 0.77, 0.75, and 0.22, respectively. In many materials, a high glum is typically associated with a low PLQY. This phenomenon is primarily attributed to the competitive mechanisms involved in the emission process: as the g-value increases, non-radiative recombination processes (such as LMBET, thermal excitation) may be enhanced, leading to a reduction in PLQY. Consequently, the interplay between the glum and PLQY has become a central topic in the current research on circularly polarized luminescent materials.

The pursuit of high PLQY remains a central focus in the design of CPL materials29. According to existing studies, β-diketones are extensively employed as ligands in luminescent lanthanide complexes due to their ability to efficiently sensitize lanthanide ions, including visible-emitting Eu(III)30,31,32 and Sm(III)33, as well as near-infrared-emitting Nd(III)34, Er(III)35,36, and Yb(III)37. In addition to PLQY, another crucial factor to consider in the development of CPL materials is the enhancement of the glum value. Although no universally applicable relationship has been established between glum and the structures of chiral lanthanide complexes, it is generally believed that high glum values are associated with the rigidity and high symmetry of the complexes. For instance, our research group reported a glum value of 0.81 for a pair of ytterbium helicates, [Yb2(R/S-BTHP)4]2−, the highest value reported for Yb-based CPL materials to date38. Additionally, the Hasegawa research group reported a glum value of 1.54 for chiral tetrakis-Eu(III) complexes, the highest value reported for chiral luminescent molecules39.

Building upon the aforementioned considerations, we designed and synthesized two pairs of enantiopure bis-β-diketones, L1R/S and L2R/S, which self-assemble with Eu(III) ions in a 2:1 molar ratio, resulting in the formation of quadruple-stranded helicates, (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] (Scheme 1). Comprehensive spectroscopic analysis confirmed the successful formation of these quadruple-stranded helicates. Single-crystal X-ray diffraction analysis revealed that the coordination environments of the two Eu(III) metal centers in both (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] exhibit a fully symmetric square antiprismatic geometry. Chiroptical property measurements demonstrated that (NEt4)2[Eu2(L1S)4] exhibits remarkable CPL activity, with a glum value of + 1.34 and a QYs of 37.19%. In contrast, (NEt4)2[Eu2(L2S)4] shows relatively weaker CPL activity, with a glum value of + 1.14 and a QYs of 30.30%.

Scheme 1
Scheme 1
Full size image

Synthetic routes of the L1R/S, L2R/S and corresponding lanthanide complexes, (NEt4)2[Ln2(L1R/S)4] and (NEt4)2[Ln2(L2R/S)4] (Ln = Eu and Gd).

Materials and methods

NMR and ESI-TOF–MS

The 1H NMR, 19F NMR and 1H-1H DOSY spectra were recorded on a Bruker AVANCE III 400 MHz spectrometer. 1H NMR chemical shifts are in ppm relative to tetramethylsilane (TMS): CDCl3 (7.26 ppm for 1H), CD3CN (1.94 ppm for 1H).

High-resolution electrospray ionization mass spectrometry (ESI-TOF–MS) were recorded by using a Bruker maXis mass spectrometer. Data analysis was conducted with the Mass-Lynx Data Analysis software (Version 4.1) and simulations were performed with the MassLynx Isotope Pattern software.

FT-IR spectra measurements

FT-IR spectra of all samples were performed with a Perkin Elmer Spectrum One spectrophotometer by using KBr disks in the range of 4000 − 370 cm−1.

Photophysical behavior

UV–vis spectra were recorded in CH3CN (c = 1.0 × 10−5 M) at room temperature in 10 mm light path quartzcuvettes on a PerkinElmer Lambda 25 spectrometer.

Excitation and emission spectra were recorded using an Edinburgh FLS 980 fluorescence spectrometer equipped with a red-sensitive photomultiplier detector (Hamamatsu R928). Excitation spectra were measured in CH3CN (c = 1.0 × 10−5 M) with quartz cuvettes of 10 mm path length. Emission spectra were measured in CH3CN (c = 1.0 × 10−5 M) with quartz cuvettes of 10 mm path length.

Luminescence lifetimes were recorded on a single photon counting spectrometer from Edinburgh Instruments (FLS 980) with a microsecond pulse lamp as the excitation source.

The luminescence quantum yields (Φ) of the samples were recorded at room temperature through an absolute method using an Edinburgh Instruments integrating sphere coupled to the modular Edinburgh FLS 980 fluorescence spectrometer. The absolute quantum yield was calculated using the following expression:

(1)

Where Lemission is the emission spectrum of the sample, collecting using the sphere, Esample is the spectrum of the incident light used to excite the sample, collected using the sphere, and Ereference is the spectrum of the light used for excitation with only the reference in the sphere. The method is accurate within 10%.

(2)

The radiative rate constant (kr) is proportional to the intensity ratio of total integrated emission of the 5D0 → 7FJ transitions (Itot) to the integrated emission of the 5D0 → 7F1 transitions (IMD). AMD,0 (14.65 s–1) is the spontaneous emission probability of the 5D0 → 7F1 transition and n is the refractive index of the medium.

The radiative transition (kr) values and non-radiative transition (knr) determine the intrinsic quantum yield (ΦLn) of Eu3+ ion emission as shown in Eq. (3).

(3)

Τobs is the observed lifetimes. On the basis of the emission decay curves monitored within the 5D0 → 7F2 transition. The sensitization efficiencies (ηsen) can be calculated.

(4)

Chiraloptical measurements

CD and CPL experiments were performed on an Olis DM245 spectrometer at room temperature. All samples were dissolved in CH3CN (c = 1.0 × 10−5 M), and quartz cuvettes with optical pathway of 10 mm were employed. CD spectra were recorded in the range of 250 − 450 nm in increments of 1 nm, and a slit width of 2 mm for the excitation was utilized. CPL spectra were recorded with a 375 nm laser as light source. The emission of left- and right-handed polarized light were collected in the range of 550 − 720 nm with the integration time of 1 s and the emission slit width of 0.6 mm.

X-ray crystallography

Crystallographic data of (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] are given in Table S3 and Table S4. Single crystals of suitable dimensions of (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] were selected for single-crystal X-ray diffraction analysis. Crystallographic data were collected at 140 K and 150 K on a Xcalibur, Eos, Gemini diffractometer with Mo Kα radiation (λ = 0.71073 Å). The structures were solved by direct methods and refined by the full-matrix least-squares method based on F2 with anisotropic gthermal parameters for all non-hydrogen atoms by using the SHELXS (direct methods) and refined by SHELXL 201840 (full matrix least-squares techniques) in the Olex2 package41. The crystallographic data in CIF format were deposited at the Cambridge Crystallographic Data Centre with CCDC Nos. 2410043 and Nos. 2362879. These data are available free of charge from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB21 EZ, UK; fax: (+ 44) 1223–336-033; or e-mail: deposit@ccdc.cam.ac.uk.

Results and discussion

Syntheses and characterization

The synthetic procedures of the ligands (L1R/S and L2R/S) and their characterization ESI-TOF–MS and 1H, 19F NMR are given in the ESI (Schemes S1 − S2 and Figures S1 − S18). To obtain the (NEt4)2[Eu2(LS)4] helicates, the ligands L1S and L2S were assembled with Eu(OTf)3 in a 2:1 stoichiometric ratio with tetraethylammonium hydroxide (NEt4OH) as the base in CH3CN, respectively, as shown in Fig. 1 a and 1b (Figures S19 − S30). ESI-TOF–MS analyses affirmed the formation of the quadruple-stranded helicates of (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4]. For instance, in the negative ionization mode, both Fig. 1 c and 1 d display a cluster of peaks, which can be attributed to the charged [Eu2(LS)4]2− (LS = L1S, L2S) at m/z 1833.0378, 1889.6063, respectively. Furthermore, the 1H and 19F analyses (Figures S31 − S32) also support the formation of the quadruple-stranded helicates of (NEt4)2[Eu2(LS)4]. In contrast to the free ligands LS (Fig. 1 e and 1f.), all resonance peaks in (NEt4)2[Eu2(LS)4] show shifts and a slight increase in width, but the splitting of the peaks is still distinguishable without difficulty (Fig. 1 g and 1 h).

Fig. 1
Fig. 1
Full size image

Synthesis and characterization of complexes. Self-assembly of quadruple-stranded Eu(III) helicates by the ligands L1S (a) and L2S (b). ESI-TOF–MS spectrum of (NEt4)2[Eu2(L1S)4] (c) and (NEt4)2[Eu2(L2S)4] (d). 1 H NMR (400 MHz) spectra of L1S (e), L2S (f), (NEt4)2[Eu2(L1S)4] (g) and (NEt4)2[Eu2(L2S)4] (h) in CD3CN.

X-ray crystallography analysis provides strong evidence for the formation of quadruple-stranded helicates, and the complexes (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] were crystallized in the chiral space groups P1 (Fig. 2). The high quality, block-shaped crystals suitable for single crystal X-ray analysis were obtained by slow evaporation of the acetonitrile/1,4-dioxane solution of (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] at room temperature. In (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4], each Eu3+ ion is eight-coordinated to O atoms from four β-diketonate ligands. In each helicate, the S-BINOL spacer imparts the assembly with P-helical chirality and homochiral ΔΔ configuration of the two metal centers for (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] (Figures S33-S36 and Tables S1-S4).

Fig. 2
Fig. 2
Full size image

X-ray single crystal structures of the helicates and C-H···π interaction of (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4]. (color code for Eu: cyan, N: blue, F: green, and O: red) Hydrogen atoms on the helicates are omitted for clarity.

Optical/chiroptical properties

On the basis of X-ray single crystal structures, in combination with the NMR analyses, it is proposed that the preorganized helical chirality of the ligands effectively promotes the formation of homochiral helicates. Therefore, the helical chirality of the (NEt4)2[Eu2(LR/S)4] (LR/S = L1R/S and L2R/S) was initially examined using circular dichroism (CD) spectroscopy. As shown in Figs. 3a and 3b, (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4] exhibit mirror-image Cotton curves in the range of 250–450 nm in CH3CN, respectively, corresponding to the π − π* transition of the free ligands LR/S. These results indicate that both complexes have a high enantiomeric purity. According to the empirical relationship between the exciton coupling pattern and the absolute configuration of the metal center42,43,44,45,46,47, we observed that the relationship between the CD exciton coupling signals of (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] and the absolute configuration of the lanthanide Eu(III) ion centers in their single crystal structures is consistent with the empirical rules reported in the literature. In (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4], the S-BINOL spacer imparts the assembly with P-helical chirality and homochiral ΔΔ configuration to the two metal centers.

Fig. 3
Fig. 3
Full size image

(a) UV–vis and CD spectra of the free ligands and their helicates (NEt4)2[Eu2(L1R/S)4] and (b) (NEt4)2[Eu2(L2R/S)4]. (c) total emission and CPL spectra of the enantiomeric helicates (NEt4)2[Eu2(L1R/S)4] and (d) (NEt4)2[Eu2(L2R/S)4] in CH3CN (c = 1.0 × 10 −5 M).

Lanthanide elements have an unfilled 4f. electron orbital and a filled 5s25p6 electron orbital, where the 4f electron is located in the 5s25p6 shell. This provides the possibility for electrons to transition between different energy levels. Due to the role of electron shielding, 4f. electrons are less affected by the external field, which makes the rare earth complexes formed after the coordination of rare earth ions and ligands show relatively sharp emission bands. The PL spectra of (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4] in CH3CN are shown in Figs. 3c and 3d. The Eu(III) ions in both cases exhibit five characteristic emission bands at 579 nm, 593 nm, 613 nm, 652 nm and 701 nm, corresponding to 5D0 → 7FJ (J = 0–4) transitions, are observed. In addition, the 5D0 → 7F1 (magnetic dipole transition) and 5D0 → 7F2 (electric dipole transition) exhibit stronger emission intensity than 5D0 → 7F0,3,4. Although the total emission intensity of the 5D0 → 7F1 transition is low, its magnetic dipole character still shows the strongest circularly polarized emission. As expected, the (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4] exhibit mirror-image CPL emission at the corresponding bands observed in the PL spectrum. Glum value is an essential parameter for estimating the degree of CPL, and glum = 2(IL − IR)/(IL + IR), where IL and IR represent the left and right circularly polarized emission intensities, respectively (with − 2 ≤ glum ≤ 2). At the magnetic dipole 5D0 → 7F1 transition of 595 nm, (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] exhibit the glum value of + 1.34 and + 1.14. Notably, the glum value of (NEt4)2[Eu2(L1S)4] is close to the highest recorded glum value of 1.54 for compounds in solution. Meanwhile, the calculated intensity ratio of (5D0 → 7F2)/(5D0 → 7F1) is 10.52 for (NEt4)2[Eu2(L1S)4] and 7.14 for (NEt4)2[Eu2(L2S)4], respectively, that is, the lower the symmetry of the coordination environment in (NEt4)2[Eu2(L1S)4].

The photophysical properties of the (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4] were further characterized by steady state and transient state spectral measurements (Figures S37 − 44), with the calculated results listed in Table S5 and S6. Overall, both (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4] have high glum values and luminescent quantum yields in CH3CN, establishing them among the most comprehensively performing rare earth CPL materials. The excited-state lifetimes of the Eu(III) ion by monitoring the decay curves of the maxima emission at 613 nm were found to be 570.41 µs, 575.33 µs, 540.61 µs, and 540.74 µs for (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4] in acetonitrile, respectively (Figures S45 − 46).

On the other hand, the luminescence quantum yield is another essential parameter to evaluate the performance of the CPL materials. The triplet energy level estimated from the maximum emission band (497 and 499 nm) of the corresponding Gd(III) complex was calculated to be 20,120 and 20,040 cm−1 for (NEt4)2[Gd(L1S)4] and (NEt4)2[Gd2(L2S)4] (Figures S47-S48). Considering the relatively low energy gap ΔE (T1-5D0, 2620 and 2540 cm−1) between the triplet-state energy level of the ligands and the 5D0 energy level of the Eu3+ ion (17,500 cm−1), all are within the optimal range of energy transfer. Both (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] emit bright red light in CH3CN, with luminescence quantum yields of 37.43% and 30.30% respectively.

To investigate the luminescence mechanism of the complexes, the HOMO and LUMO distribution of the two helicates were calculated by Time-dependent density functional theory calculation (DFT). As shown in Fig. 4, the HOMO and LUMO localized on the chiral ligands, this is the source of the CPL of the helicates. In addition, the HOMO orbital is localized on the binaphthol, while the LUMO orbital locates on the β-diketone moiety. This result confirms the participation of the BINOL moiety on HOMO → LUMO electron transition.

Fig. 4
Fig. 4
Full size image

Schematic diagram showing the electronic density contours for the HOMO, and LUMO molecular orbitals of the (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4].

Origin analysis of the strong CPL

From a theoretical perspective, the glum can be calculated using the equation: glum = 4(|μ| ×|m|× cosθμ,m)/(|μ|2 +|m|2),where μ, m and θ μ,m represent the electric and magnetic transition dipole moments, and the angle between them, respectively. To achieve a high glum value, the electron transition must balance the electric and magnetic dipole moments, forming an ideal angle between them. In contrast to organic compounds, the partial f–f transitions of Ln3⁺ ions are allowed as magnetic dipole transitions, such as the 5D0 → 7F1 transition of Eu(III) ions. This typically endows chiral lanthanide complexes with exceptionally high CPL activity, often resulting in glum values that are 2–3 orders of magnitude higher than those observed in organic luminophores.

For lanthanide metals, the f–f transitions are intraconfigurational and Laporte-forbidden. To achieve strong CPL emission, the lanthanide center requires a dissymmetric ligand field that can mix the 4f. and 5 d orbitals to generate the required electric (μ) and magnetic (m) transition dipole moments. In this process, the rigidity of the complex’s conformation plays a crucial role in determining the extent of mixing at the dipole transition48. The ligands we designed are modified with two naphthalene rings based on S-BINOL. We anticipate strong C-H···π interactions in the complexes formed by these ligands, which significantly enhance the rigidity of the complex, thereby improving the glum value. For (NEt4)2[Eu2(L2S)4], the helicates modified with 1-methylnaphthalene exhibit weaker C-H···π interaction due to the presence of the methyl group49, resulting in slightly reduced rigidity compared to (NEt4)2[Eu2(L1S)4]. Since the gCPL value was tested in CH3CN solution, the solid UV–vis and 1H NMR spectra of (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] were respectively tested. It can be seen from the spectra that the solution state and crystal structure are similar (Figures S49 − S50).

To date, no universally applicable relationship has been established between CPL activity (glum) and the structures of chiral lanthanide complexes. Richardson proposed two mechanisms for explaining the origin of chiroptical activity in these complexes: static coupling and dynamic coupling mechanisms50. In this study, we focus solely on the impact of the static coupling mechanism on the CPL activity of lanthanide complexes. In the static coupling mechanism, the configurational chirality around the first ligand coordinating with the Ln3+ ion is considered. Parker et al. found a relationship between glum and the twist angle α (the angle between the top and bottom planes of the coordination polyhedron), where glum ≈ sin 4α for Eu(III) complexes51,52,53,54,55. According to this empirical relationship, the maximum glum value occurs when the twist angle is 22.5°, while a twist angle of 45° results in the disappearance of CPL activity. The helicate [Yb2(S-BTHP)4]2−, reported by our research group in 2023, typically adopts a square antiprismatic geometry (8-SAPR) coordination configuration. This helicate has an average α angle of 40.5° and exhibits a high glum value36.

The coordination configurations of the two Eu(III) metal centers in (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] both feature a fully symmetric square antiprismatic geometry. We believe this is one of the reasons these complexes exhibit high glum values. The average α angles for (NEt4)2[Eu2(L1S)4]. and (NEt4)2[Eu2(L2S)4]. are 38.603° and 38.973°, respectively (Fig. 5). These angles are very close to the 40.5° twist angle of [Yb2(S-BTHP)4]2−, suggesting that static coupling plays a limited role in contributing to the high glum values observed in these complexes.

Fig. 5
Fig. 5
Full size image

The torsion angle between the two planes of the SAPR coordination polyhedron of (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4].

Conclusions

In summary, we have successfully synthesized and characterized two pairs of quadruple-stranded helicates, (NEt4)2[Eu2(L1R/S)4] and (NEt4)2[Eu2(L2R/S)4]. Single-crystal X-ray diffraction analysis confirms the formation of homochiral quadruple-stranded dinuclear structures in both complexes. Photophysical and chiroptical investigations revealed that (NEt4)2[Eu2(L1S)4] and (NEt4)2[Eu2(L2S)4] not only exhibit intense CPL, with glum values of + 1.34 and + 1.14, respectively, but also demonstrate high luminescence quantum yields of 37.43% and 30.30%. By combining structural and chiroptical spectroscopic analyses, we propose that achieving high glum values for CPL materials in octagonal complexes requires two key factors: first, the structure must be as rigid as possible, and second, the coordination configuration should adopt a square antiprism geometry. Moreover, the relationship between glum and the twist angle α plays a crucial role in this process. This work provides valuable insights into the design of lanthanide-based CPL materials that combine both high glum values and high luminescence quantum yields.