Photoluminescence and Hirshfeld surface analysis of the binuclear mixed-ligand complex [Eu 2 (o -MBA) 6 (phen) 2 ]

1 Introduction

Lanthanide coordination compounds, particularly Eu(III)-based systems, have attracted considerable attention due to their fundamental and applied importance (Motson et al., 2004; Reisfeld, 2015; Blasse and Grabmaier, 1994; Thor et al., 2024). Although Eu(III) complexes have been studied for several decades, interest in these materials remains high, driven by a number of factors. Lanthanide coordination compounds consist of Ln(III) metal ions coordinated by organic ligands, and the wide variety of available ligands allows the synthesis of structurally diverse materials (Reisfeld, 2015; Thor et al., 2024; Tanner, 2010). In particular, the use of new ligands or novel combinations of ligands enables the design of coordination compounds with tailored physicochemical properties and expanded application potential. Equally important is the possibility of fine-tuning the properties of Ln(III) complexes through rational ligand selection and purpose-oriented molecular design (Mortensen and Sorensen, 2023; Bünzli and Eliseeva, 2010; Brown et al., 1990; Gawryszewska et al., 2005; Kariaka et al., 2023; Félix et al., 2024). Special attention has been devoted to binuclear mixed-ligand complexes, which offer additional flexibility for tuning the coordination environments of the two metal centres (Li et al., 2020; Bünzli, 2015; Culeac et al., 2022; Culeac et al., 2024; Demchenko, 2015).

Eu(III) complexes are especially attractive for photoluminescence spectroscopy and practical applications due to their narrow emission bands, long excited-state lifetimes, and relatively simple emission band splitting patterns (Bünzli, 1987; Hasegawa et al., 2022). Literature studies highlight the significant potential of Eu(III) coordination compounds and emphasize the need for continued research aimed at developing new luminescent materials for emerging applications, as, for example for quantum information systems (Riedel et al., 2022).

To date, relatively few studies have addressed the coordination compound [Eu₂(o-MBA)₆(phen)₂]. Previous reports have focused primarily on its synthesis, structural features, and thermal behavior, including TGA/DTA analyses (Kalinovskaya et al., 2021; Kalinovskaya and Zadorozhnaya, 2022; Chen Z.M. et al., 2009). This dinuclear Eu(III) complex contains o-methylbenzoate (o-MBA) and 1,10-phenanthroline (phen) as chelating ligands. To the best of our knowledge, a comprehensive peer-reviewed study addressing both the structural and luminescent properties of this specific complex has only recently been published (Culeac et al., 2024).

The combination of mixed ligands (phen and o-MBA) provides a more efficient sensitization pathway compared to single-ligand systems or mononuclear species. The relatively large energy gap ΔE(3ππ∗−⁵D₀) ≈ 4854 cm^-1 (Kalinovskaya and Zadorozhnaya, 2022) significantly suppresses nonradiative deactivation, thereby enhancing the sensitization efficiency. In the case of [Eu₂(o-MBA)₆(phen)₂] complex two Eu(III) centers are bridged and coordinated by six o-MBA ligands and two phen ligands. The presence of multiple o-MBA ligands enables both bridging and terminal coordination, resulting in a stable binuclear Eu(III) architecture. This structural arrangement increases rigidity and lowers symmetry, factors that strongly influence emission intensity and spectral line splitting.

Single-crystal X-ray diffraction analysis confirmed that each Eu(III) ion is coordinated by five o-MBA ligands and one phen ligand, forming a Chinese lantern–type (paddlewheel) structure. The rigid coordination environment surrounding the Eu³⁺ ions minimizes vibrational energy losses and nonradiative decay pathways. Thermal analysis further demonstrates good thermal stability, with a decomposition temperature of approximately 273 °C (Culeac et al., 2024). The aim of the present work is to extend the previous characterization of the [Eu₂(o-MBA)₆(phen)₂] complex, including low-temperature photoluminescence studies and estimation of partial ligand field parameters.

2 Materials and methods

The coordination complex [Eu₂(o-MBA)₆(phen)₂] was synthesized by chemical method and was characterized by single-crystal XRD and PL spectroscopy. All chemicals and reagents were purchased from commercial sources (Sigma-Aldrich, St. Louis, MO, USA) and used without further purification: europium(III) chloride hexahydrate (99.9%), 1,10-phenanthroline monohydrate (≥99%), o-methylbenzoic (o-toluic) acid (99%), and sodium hydroxide (≥98%). A solution of o-methylbenzoic acid (0.50 g, 3.6 mmol) in 15 mL of warm 96% ethanol was mixed with 1,10-phenanthroline monohydrate (0.24 g, 1.2 mmol) and 3.6 mL of 1 M NaOH solution. The resulting mixture was slowly added, under continuous stirring, to an aqueous solution of EuCl₃·6H₂O (0.45 g, 1.2 mmol) dissolved in 5 mL of distilled water. The reaction mixture was stirred for 1 h at room temperature. The precipitated product was collected by filtration, washed successively with ethanol and diethyl ether, and dried in air to afford a white powder with a slight pink tint. The isolated yield was 0.75 g (45.7%).

Analytical calculations for compound C₇₂H₅₈N₄O₁₂Eu₂ (%): C, 58.63; H, 3.97; N, 3.80; Eu, 20.62. Found (%): C, 58.97; H, 4.11; N, 3.72; Eu, 20.81. The europium content was determined from the Eu₂O₃ residue obtained after combustion of the sample in oxygen during carbon and hydrogen analysis.

The final air-dried product appears as a white powder with a faint pink tint under daylight conditions (Figure 1a). Pink single crystals suitable for single-crystal X-ray diffraction were obtained by recrystallization from a DMF/toluene (1:1, v/v) mixture at room temperature through slow solvent evaporation over approximately 4–5 weeks.

Photoluminescence spectra (PL) were recorded for powder samples (crystals ground into powder) for different temperatures in the range 15–300 K. PL temperature studies were carried out with a Leybold RDK 10–3202 closed-cycle refrigerator system. The temperature of the sample was controlled by a thermocouple with an accuracy of 0.02 K. High-resolution PL emission spectra (2 cm⁻¹) were registered with a double grating spectrophotometer DFS-52 (LOMO, St Petersburg, Russia) and a Hamamatsu photomultiplier module H8259-01 in a photon counting mode. PL emission spectra were registered with a Thorlabs L375P70MLD Laser Diode (375 nm, 70 mW) or Collimated Laser Diode Module CPS405 (405 nm, 4.5 mW) as excitation sources. Low-resolution excitation spectra were registered with an MDR-23 single grating monochromator (LOMO) as an excitation source and a double grating spectrometer DFS-52 for collecting the PL emission. A halogen lamp OSRAM 64623 HLX 12 V 100 W was used as a light excitation source. Fluorescence lifetimes were measured using a 3^rd harmonics (355 nm) of pulsed Q-switched Nd:YAG laser, while emission was detected by a photomultiplier tube and processed with a boxcar integrator. The emission spectra were corrected for the instrument spectral sensitivity. Under UV irradiation the [Eu₂(o-MBA)₆(phen)₂] complex exhibits a characteristic bright-red luminescence with high color purity (Figure 1). The absolute photoluminescence quantum yield (PLQY) was determined using an integrating sphere technique. The measurements were carried out with a G8 integrating sphere (GMP General Microtechnology & Photonics, Switzerland).

3 Hirshfeld surface analysis

The coordination complex [Eu₂(o-MBA)₆(phen)₂] (CSD - reference code MOZCUD) crystallizes in triclinic space group P-1 with two crystallographically independent but identical complexes in the asymmetric unit (Z′ = 2). Both complexes have Ci molecular low-symmetry. The unit cell parameters are a = 13.1368(5) Å, b = 14.3488(5) Å, c = 17.4196(5) Å, α = 83.193(3)°, β = 83.754(3)°, γ = 72.940(3)°, V = 3107.5(2) Å3. The structures of both crystallographically independent binuclear complexes reveal that the coordination environment of the Eu(III) centers represents a distorted tricapped trigonal prism, as illustrated in Figure 2. The detailed structural data can be found in the previously published work of Culeac et al. (2024).

The chemical identity of the recrystallized crystals was confirmed by comparing of experimentally obtained powder X-ray diffraction (PXRD) data with theoretical simulations based on single-crystal X-ray diffraction data. The two crystallographically independent binuclear complexes exhibit identical Eu³⁺ coordination environments, with average Eu–O bond distances of 2.460 Å (Eu1) and 2.468 Å (Eu2), and average Eu–N distances of 2.645 Å and 2.616 Å, respectively. Their geometrical parameters coincide within the accuracy of the data. The intramolecular Eu···Eu separations represent 4.0087(4) and 4.0371(4) Å (Culeac et al., 2024).

The fingerprint plot profiles of the two molecules are virtually superimposable, with no more than a 0.5 percentage point difference in any individual contact contribution (H···H 57.2% vs. 56.8%, C···H/H···C 21.2% vs. 21.7%, O···H/H···O 7.3% vs. 7.3%, C···C 3.4% vs. 3.1%, N···H/H···N 0.7% vs. 0.6%), confirming the close similarity of their intermolecular interaction landscapes. For these reasons, the Hirshfeld surface and fingerprint analysis are represented by only one of them.

To examine the intermolecular interaction distribution in the title crystal structure, Hirshfeld surface (HS) analysis was performed using CrystalExplorer 21 software (Spackman et al., 2021; Spackman and Jayatilaka, 2009; Spackman and McKinnon, 2002). The Hirshfeld surface is a three-dimensional surface that is defined based on the electron density distribution of a molecule within a crystal. It is constructed using spherical atomic electron densities, resulting in a surface that is unique to a given crystal structure.

The Hirshfeld surface corresponds to regions where the ratio of the promolecule to procrystal electron densities is 0.5. Intermolecular interactions are visualized using the normalized contact distance, d _*norm*, which depends on the internal (d _*i*) and external (d _*e*) distances to the surface and the respective van der Waals radii of the atoms (Spackman and Jayatilaka, 2009; Mckinnon et al., 2007). Since the two independent molecules in the asymmetric unit exhibit very similar intermolecular interaction patterns, only one molecule was selected for detailed Hirshfeld surface and fingerprint analysis. Based on the distances to the nearest atoms inside (d _*i*) and outside (d _*e*) the surface, the normalized contact distance (d _*norm*) is calculated. Hirshfeld surfaces mapped using d _*norm* values of - 0.2892 to 1.5557 are shown in Figure 3.

Mapping the Hirshfeld surfaces over shape index and curvedness provides insight into molecular packing and π–π stacking interactions. Regions with low curvedness and flat surfaces indicate π-stacking, as evidenced by the green patches on the surface in Figure 3c. On the shape index map, π–π interactions appear as red and blue triangle pairs in characteristic locations: blue bumps correspond to convex regions (i.e. inward-facing ring carbons) and red cavities correspond to concave regions (i.e. external π-stacked rings) (see Figure 3b). Furthermore, the dark blue edges on the curvedness surface reflect structural corrugation, which is consistent with the continuous π–π stacking that is observed in the crystal structure (Culeac et al., 2024).

Two-dimensional fingerprint diagrams obtained via HS analysis provide in-detail information about the intermolecular interaction types and significance within a crystal structure. The HS analysis reveals that H···H interactions dominate the surface, contributing 57.2% of the total Hirshfeld area (Figure 4). This suggests the predominance of van der Waals contacts involving hydrogen atoms. The next most significant contribution comes from C···H/H···C contacts, accounting for 21.2%, which indicates the presence of weak but directional interactions, such as C–H···π interactions, often associated with molecular packing and structural stabilization. O···H/H···O interactions contribute 7.3%, suggesting a moderate role in crystal stabilization. C···C interactions (3.4%) indicate limited π–π stacking between aromatic rings. Less significant contributions come from N···H/H···N interactions, representing only 0.7%, indicating their minimal role in the overall landscape of intermolecular interactions (Figure 4).

4 Photoluminescence spectra

The photoluminescence spectra of [Eu₂(o-MBA)₆(phen)₂] complex were registered for powder samples at different temperature in the range 15–300 K. The excitation spectrum (300 K), detected at 614 nm, assigned to the ⁵D₀→⁷F₂ transition of Eu³⁺, shows a broad band in the 300–390 nm range with a maximum at ∼ 360 nm (Figure 5). Several narrow lines in the 384–590 nm range can be assigned to intra-4f⁶ transitions from the ⁷F_0,1 levels to the ⁵D_0,1,2,3 and ⁵L₆, higher excited states. The sharp features at 395, 464, and 535.6 nm correspond to the Eu³⁺ transitions ⁷F₀→⁵L₆, ⁷F₀→⁵D₂, and ⁷F₁→⁵D₁, respectively (Culeac et al., 2024; Da Rosa et al., 2021; Gawryszewska et al., 2000; Puntus et al., 2010). The dominant broad band at ∼ 360 nm, with a shoulder near 335nm, can be assigned to ligand-centred π–π* transitions (Biju et al., 2006; Ruan et al., 2017; Mautner et al., 2021).

The emission spectrum of the [Eu₂(o-MBA)₆(phen)₂] powder sample obtained at room temperature using 375 nm laser excitation is represented in Figure 6. The luminescence profile is characterized by a series of narrow lines arising from the Eu^{3+ 5}D₀→⁷F_*J* (J = 0–4) transitions, with barycenters at 580.09, 593.29, 617.76, 651.29, and 694.08 nm. No emission from the short-lived ⁵D₁ level or other higher excited states is observed, indicating efficient nonradiative relaxation to the metastable ⁵D₀ state. The absence of any broad ligand-centered emission band further confirms efficient energy transfer from the ligands to the resonant Eu³⁺ level.

The emission spectra of [Eu₂(o-MBA)₆(phen)₂] powder sample were obtained for cryogenic-to-room different temperatures under 405 nm excitation, and for temperature range 20–300 K are represented in Figure 7. The complex exhibits a typical Eu(III) photoluminescence pattern with well-resolved splitting of the ⁵D₀→⁷F_*J* (J = 0–4) transitions.

The observed asymmetry of the band ⁵D₀→⁷F₀ (Figure 8) and its fwhm ≈ 0.42 nm (12.6 cm^-1) can be attributed to the presence of two crystallographically independent binuclear complexes. Within each binuclear unit, the two Eu³⁺ ions are related by an inversion center and are therefore crystallographically equivalent. However, the two independent complexes exhibit slightly different coordination environments around Eu³⁺ (Culeac et al., 2024). Indeed, deconvolution of the ⁵D₀→⁷F₀ emission band (Figure 8, inset) reveals two closely spaced components separated by approximately 5.9 cm⁻¹, indicating the presence of two spectroscopically nearly equivalent Eu³⁺ sites. (Detailed peak-fitting information can be find in Supplementary Figure S1).

The magnetic-dipole (MD) transition ⁵D₀→⁷F₁ at 15 K displays 3 almost equal spaced components, with approximately the same integrated intensity, and a very weak minor component with negligible contribution to the integrated intensity (Figure 9a). (More details of peak fitting can be find in Supplementary Figure S2). This is consistent with a binuclear, low-symmetry structure and with crystallographic data, indicating two non-equivalent, yet nearly identical Eu³⁺ coordination sites. The hypersensitive electric-dipole transition ⁵D₀→⁷F₂ dominates the spectrum, with a barycenter at ∼ 617 nm and with a contribution of ∼80% of the total integrated PL intensity. The dominant electric dipole transition ⁵D₀→⁷F₂ can be deconvoluted into at least 6 components (Figure 9b; Supplementary Figure S3). The asymmetry ratio calculated for room temperature R = I(⁵D₀→⁷F₂)/I(⁵D₀→⁷F₁) = 7.78 correlates with the strong red emission of the compound. The pronounced dominance of the ⁵D₀→⁷F₂ transition points to a highly polarizable ligand environment around the Eu³⁺ ions (Binnemans, 2015; Tanner, 2010; Tanner, 2013; Marques et al., 2016). The integrated intensity of the band ⁵D₀→⁷F₄ is comparable with integrated intensity of the band ⁵D₀→⁷F₁, while the weakest emission band ⁵D₀→⁷F₃ appears as almost unnoticeable.

The fluorescence lifetime was determined using a Q-switched Nd:YAG laser operated at 355 nm (3^rd harmonic generation). Figure 10 illustrates the PL time decay characteristic measured for the transition ⁵D₀→⁷F₂ monitored at 614 nm at room temperature. The measured PL decay of the ⁵D₀→⁷F₂ transition can be approximated by a single exponential function, and the experimentally lifetime (τ _*obs*) obtained from the plot (Figure 10; Supplementary Figure S4) represents τ _obs ≈ 1.26 ± 0.01 ms.

A number of photophysical parameters of the Eu³⁺ complex was estimated from the PL spectral and time decay characteristic based on the Judd–Ofelt (JO) approach (Judd, 1962; Ofelt, 1962). These parameters comprise the spontaneous emission probabilities A 0 − J, branching ratios β 0 J, Judd–Ofelt intensity parameters Ω 2 and Ω 4, total radiative transition probability A r a d, radiative (τ r a d) and experimental lifetime (τ o b s), non-radiative decay rate A n r a d, and internal quantum efficiency (η) (Table 1; Supplementary Table S1).

TABLE 1. JO intensity and radiative parameters (300 K) of the [Eu₂(o-MBA)₆(phen)₂] complex.

The Judd–Ofelt intensity parameters Ω 2 and Ω 4, as well as the radiative and efficiency parameters A r a d , A n r a d , and η, were derived from the experimental emission spectra and lifetime data using the standard equations described in the literature (Cantelar et al., 2020; De Santana et al., 2018; De Sá et al., 2000; Binnemans, 2015) (Table 1). The transition-resolved radiative rates A r a d J, total radiative rate A r a d, and branching ratios β 0 J were also calculated using the LUMPAC software (Filho et al., 2014; Malta et al., 1996). Detailed transition-resolved Judd–Ofelt analysis, including integrated emission intensities, barycenters, branching ratios, and individual radiative transition probabilities, is provided in the Supplementary Table S1.

The radiative transition probability A 0 − J of the Eu³⁺ ion, can be written in terms (Equation 1) of the area I 0 − J under the corresponding PL emission curve, and the reference radiative transition probability of the MD transition A 0 − 1 (De Santana et al., 2018; Cantelar et al., 2020; Yang et al., 2013):where λ 0 − J represents the wavelength barycenter of the Eu³⁺ transition ⁵D₀→⁷F_J (J = 0–4), I 0 − J is the integrated intensity of the emission band ⁵D₀→⁷F_J (the area under the corresponding emission curve). Because the magnetic dipole transition ⁵D₀→⁷F₁ has no electric dipole contribution, and is practically independent of the Eu³⁺ ions environment, it is generally accepted as a reference. The total radiative transition probability A r a d, can be written in terms (Equation 2) of the sum of radiative transition probability from a higher energy level ⁵D₀ to all the lower energy level A 0 − J (De Santana et al., 2018):

A 0 − J = I 0 − J I 0 − 1 λ 0 − J λ 0 − 1 A 0 − 1 (1)

A r a d = ∑ A 0 − J (2)

In order to calculate the experimental Judd–Ofelt intensity parameters Ω λ for Eu³⁺ from the integrated emission intensities I 0 − J and the MD transition ⁵D₀→⁷F₁ as an internal reference, the standard relation (Equation 3) was used (Cantelar et al., 2020; De Santana et al., 2018; De Sá et al., 2000; Binnemans, 2015):where υ ₁ is the average wavenumber of the transition ⁵D₀→⁷F₁, υ _j is the average wavenumber of the transition ⁵D₀→⁷F_J, e is the electron charge, h is the Planck constant, n is the refractive index. The dipole line strength of the magnetic-dipole ⁵D₀→⁷F₁ transition was taken as D m d = 9.6 × 10⁻⁴² esu² cm² (Bünzli and Eliseeva, 2010; Binnemans, 2015). The refractive index was taken as n ≈ 1.50, which is widely adopted for solid Eu³⁺ coordination compounds when the experimental refractive index is not available. For the average index of refraction n ≈ 1.5 the transition probability of the reference magnetic dipole transition, is known as A 0 − 1 ≈ 50 s^-1 (De Sá et al., 2022; Malta et al., 1997; Molina et al., 2005). The squared reduced matrix elements values F 7 J ‖ U λ D 5 0 2 were taken from (Carnall et al., 1978; Binnemans, 2015) and are equal to 0.0032 and 0.0023 for λ = 2 and λ = 4, respectively. The branching ratios β 0 J for the emission transitions are obtained from the relation β 0 J = A 0 J / A r a d. The experimental PL lifetime τ o b s, determined from the PL decay characteristic (Figure 10), is related to radiative A r a d and nonradiative A n r a d transitions probabilities as follows (Equation 4) (Biju et al., 2006):

Ω λ = D m d υ 1 3 e 2 υ J 3 F 7 J ‖ U λ D 5 0 2 9 n 3 n n 2 + 2 2 I 0 − J I 0 − 1 (3)

1 τ o b s = A r a d + A n r a d (4)

The intrinsic quantum yield η is determined as follows (Equation 5):

η = A r a d A r a d + A n r a d (5)

The absolute photoluminescence quantum yield (PLQY) was determined under 375 nm laser excitation using an integrating sphere, according to a generally accepted procedure described in the literature. The [Eu₂(o-MBA)₆(phen)₂] powder sample exhibited an absolute photoluminescence quantum yield of 33.7% ± 8%.

As is known, the JO intensity parameters are sensitive to asymmetry and covalency of the ligand environment (local distortion, bonding character). Specifically, Ω 2 is determined by the covalence between Eu³⁺ and the ligand anions and by the structural changes in the vicinity of the Eu³⁺ ion, while the parameter Ω 4 is associated with microscopic parameters such as the matrix rigidity and viscosity as well as the dielectric constant (Binnemans, 2015; Ferhi et al., 2014). The numerical values of JO intensity parameters vary significantly among glasses, oxides, and molecular complexes. In general, higher Ω 2 values are associated with increased metal–ligand covalency and lower site symmetry around the Eu³⁺ ion (Chen and Liu, 2005; Manju Gopinath et al., 2020). In the [Eu₂(o-MBA)₆(phen)₂] complex both the intensity parameters Ω 2 and Ω 4, as well as the radiative parameters, are sufficiently high and comparable to those reported for other Eu³⁺ compounds (Cantelar et al., 2020; De Santana et al., 2018). For instance, a highly polarizable chemical environment around the Eu³⁺ ion may result in a large value of Ω 2, as reported for the [Eu(DBM)₃(PTSO)₂] complex, Ω 2 = 61.9 × 10⁻²⁰ cm² (Niyama et al., 2005).

The binuclear complex [Eu₂(o-MBA)₆(phen)₂] exhibits moderate thermal quenching over the temperature range 80–300 K (Figure 7, inset). Actually, below room temperature the Eu³⁺ compounds generally show average luminescence stability (Zhang et al., 2017; Gálico et al., 2017; Thor et al., 2022; Culeac et al., 2022), but it can undergo moderate thermal quenching in certain hosts through mechanisms that become active even at these relatively low temperatures. The quenching in this temperature range is typically moderate because the large energy gap of ΔE(3ππ∗−⁵D₀) usually require significant thermal energy to overcome. Commonly, the PL quenching mechanisms in Eu³⁺ compounds may be related to multiphonon relaxation, thermally activated energy transfer or electron transfer processes (Zhang et al., 2017; Thor et al., 2022; Tsaryuk et al., 2023).

The temperature dependence of the integrated emission intensity was analyzed using an Arrhenius-type thermal quenching model (Gwak et al., 2014; Berry et al., 1996). To characterize the quenching behavior, the monotonic region of the intensity dependence (80–300 K) was fitted using the relation (Equation 6):where I 0 is the initial emission intensity, I(T) is the emission intensity at temperature T, A is a host-dependent constant, k B is the Boltzmann constant, and E a is the activation energy. For selected range 80–300 K the Arrhenius plot ln ⁡ I 0 / I − 1 vs. 1 / k B T (Figure 11) exhibits an approximately linear behavior. An apparent activation energy of E a = 254 ± 35 cm^-1 was obtained, with a coefficient of determination R 2 = 0.91 (Supplementary Figure S4). The moderate R 2 value suggests that the thermal quenching process is not fully described by a single-channel Arrhenius mechanism.

I T = I 0 / 1 + A e x p − E a k B T (6)

The evaluated value of the activation energy is much smaller than the estimated energy gap between the ligand triplet state and the Eu^{3+ 5}D₀ level (≈4854 cm^-1) (Kalinovskaya and Zadorozhnaya, 2022), suggesting that direct thermal population of the ligand triplet state is unlikely to be the dominant quenching pathway. However, the activation energy obtained from the temperature dependence of the emission intensity alone does not permit a definitive identification of the quenching mechanism. Possible contributions may include low-energy vibrationally assisted non-radiative processes or other thermally activated relaxation channels reported for Eu³⁺ (Shi and Mudring, 2022; Pazik et al., 2012). The question of thermal quenching mechanism of [Eu₂(o-MBA)₆(phen)₂] complex remains to be further investigated and temperature-dependent lifetime and quantum-yield measurements would be required to establish the dominant thermal quenching mechanism.

The color purity was calculated using the standard relation (Equation 7) (Hasabeldaim et al., 2023):

C o l o r p u r i t y = x − x s 2 + y − y s 2 x d − x s 2 + y d − y s 2 × 100 % (7)

Here, (x, y) are the chromaticity coordinates of the sample, (x _s, y _s) are the coordinates of the reference white point, and (x _d, y _d) are the chromaticity coordinates corresponding to the dominant wavelength on the CIE 1931 spectral locus. Following common practice in photoluminescence studies, Illuminant E (equal-energy white light) was selected as the reference white point because it provides a neutral, device-independent reference for evaluating the chromaticity of luminescent materials. Accordingly, the white-point coordinates (x _s, y _s) = (0.333, 0.333) were used in the calculations. The dominant wavelength was determined geometrically from the CIE 1931 chromaticity diagram by extending a straight line from the Illuminant E point through the sample chromaticity coordinates until it intersected the spectral locus. The resulting intersection defines the dominant wavelength coordinates (x _d, y _d) = (0.677, 0.322), which were used in the color-purity calculation.

The CIE chromaticity coordinates were obtained from the measured photoluminescence emission spectrum, and the color-purity values were calculated using the standard CIE methodology implemented in the SciApps color-calculation tool, following the procedure described by Hasabeldaim et al. (2023).

Figure 12 illustrates the CIE (Commission Internationale de Eclairage) chromaticity coordinates of [Eu₂(o-MBA)₆(phen)₂] complex estimated from the photoluminescence emission spectra (300 K), and these data confirm high color purity 97.07%. The CIE coordinates for powder sample (x = 0.667, y = 0.333) fall in pure red region and are close to the standard red color values established by the European Broadcasting Union (EBU, x = 0.64, y = 0.33).

Splitting of the ⁵D₀→⁷F₁ transition can be used to estimate the partial ligand field parameters of the complex (Binnemans and Gӧrller-Walrand, 1995; Gӧrller-Walrand and Binnemans, 1996; Binnemans, 2015). The three Stark components of this manifold (Figure 9a) and Supplementary Figure S2 are located at 37.64, 46.34, and 55.21 meV (respectively 303.6, 373.8 and 445.3 cm^-1) relative to the ground level ⁷F₀, and are distributed nearly symmetrically around their barycenter (≈46.4 meV). The apparent shoulder contribution at 56.65 meV is small, therefore it is treated effectively as a perturbation to the dominant peak. Although the peaks are broadened somewhat even at 15 K, they are still easily resolved with a separation of about 71 cm^-1. Indeed, for truly low-symmetry species, the three components are equally spaced and tend to have the same intensity (Bünzli and Eliseeva, 2010). For Eu(III) ion in low symmetry (like Ci), the ligand field splits the degenerate ⁷F₁ level into 3 non-degenerate Stark components, associated with M _*J* = 0, ±1.

A preliminary estimate of the second-rank crystal-field parameters was obtained from the Stark splitting of only the ⁷F₁ multiplet (Culeac et al., 2025). Since this treatment relies on a simplified effective Hamiltonian restricted to the J = 1 manifold and includes only the lowest-rank terms, the resulting parameters should be regarded as approximate indicators of the local anisotropy rather than a complete description of the crystal field.

A more reliable description of the Stark level scheme was obtained by fitting the experimental Stark energies using the crystal-field parameters (CFPs) within the SPECTRE program (Boothroyd, 1990), employing the Wybourne tesseral formalism (Wybourne, 1965; Rudowicz and Qin, 2004; Ghosh et al., 2023). It should be noted that the CFPs depend on the choice of the coordinate system, and in the absence of symmetry constraints (Cᵢ site symmetry), the orientation of the axes is not uniquely defined. Consequently, the obtained CFPs represent an effective parameter set, corresponding to an arbitrary axis system defined by the fitting procedure. Direct comparison with CFPs reported in other studies requires consideration of the coordinate system used. In the SPECTRE program the output parameters appear as B(2,0) and B(2,2), which correspond to the conventional crystal-field parameters B 20 and B 22, respectively. For simplicity, the latter notation is used in the following text. The experimental Stark energies derived from the ⁵D₀→⁷F₁ and ⁵D₀→⁷F₂ emission bands (Supplementary Figures S2, S3) were used as input for the fitting procedure.

The best restricted fit was obtained by varying the dominant second-rank parameters while keeping higher-rank terms fixed at reasonable starting values. The resulting parameters were B 20 = − 271 ± 119 cm^-1 and B 22 = 128.6 ± 45.2 cm^-1, with a goodness-of-fit parameter χ² = 6.09. The calculated energies reproduce most of the experimentally observed Stark levels within approximately 0–2 cm^-1, although the lowest component of the ⁷F₁ multiplet remains somewhat less accurately reproduced. The relatively large magnitude of the B 22 parameter indicates a substantial non-axial component of the crystal field, which is characteristic of Eu³⁺ ions located in strongly distorted coordination environments (Ghosh et al., 2023; Thomsen et al., 2022; Lavín et al., 2001).

This interpretation is consistent with reported structural features of the complex [Eu₂(o-MBA)₆(phen)₂] (Culeac et al., 2024), where each Eu³⁺ ion is nine-coordinated by seven oxygen atoms from carboxylate groups and two nitrogen atoms from the phenanthroline ligand. The coordination polyhedron is markedly irregular due to the combination of bridging carboxylate groups and chelating phenanthroline ligands, leading to a highly anisotropic ligand field around the metal center. Using the values derived from the SPECTRE fit, the rhombicity ratio defined in the Wybourne notation as ρ = | B 22/ B 20 | is approximately 0.47. Such a value indicates a substantial non-axial component of the crystal field, implying that the rhombic distortion is comparable in magnitude to the axial contribution (Rudowicz and Qin, 2004; Thomsen et al., 2022).

Such a coordination geometry leads to a strongly anisotropic ligand field and consequently to significant non-axial crystal-field contributions. The agreement between the Stark splitting pattern, the crystal-field analysis, and the structural features of the coordination polyhedron therefore confirms that the Eu³⁺ ion occupies a markedly low-symmetry site in the complex, illustrating the well-known sensitivity of Eu³⁺ emission spectroscopy to the local symmetry of the metal center (Binnemans, 2015; Tanner, 2010). The spectroscopic results support the structural picture derived from X-ray diffraction, confirming that the Eu³⁺ ions experience a strongly distorted coordination environment. This behavior is typical for Eu³⁺ ions in low-symmetry coordination compounds, where the Stark structure of the ⁷F_*J* multiplets provides a sensitive probe of the local symmetry of the metal site.

5 Conclusion

The coordination compound [Eu₂(o-MBA)₆(phen)₂] was characterized by single-crystal X-ray diffraction and photoluminescence spectroscopy over the temperature range 15–300 K. Low-temperature emission measurements resolved the Stark structure of the ⁵D₀→⁷F_J transitions, confirming that the Eu³⁺ ions occupy markedly low-symmetry coordination environments consistent with the crystallographic model.

The [Eu₂(o-MBA)₆(phen)₂] complex exhibits moderate thermal quenching of the emission intensity in the 80–300 K temperature range. The photoluminescence results are in good agreement with the single-crystal X-ray structure, which indicates that both Eu³⁺ ions occupy nearly identical nine-coordinate environments defined by seven oxygen atoms and two nitrogen atoms. The estimated second-rank crystal-field parameters, B 20 ≈ −271 cm^-1 and B 22 ≈ 127 cm^-1, are consistent with a low-symmetry (Cᵢ) coordination environment in the triclinic P-1 crystal structure. The rhombicity ratio, ρ = | B 22/ B 20 | ≈ 0.47, indicates a substantial non-axial component of the crystal field, demonstrating that the rhombic distortion is comparable in magnitude to the axial contribution.

The complex exhibits intense red emission dominated by the hypersensitive ⁵D₀→⁷F₂ transition, high color purity (≈97%), and moderate thermal stability, with an apparent activation energy for thermal quenching of approximately E a ≈ 254 cm^-1. The Judd–Ofelt intensity parameters, radiative characteristics, intrinsic quantum efficiency, and absolute photoluminescence quantum yield (PLQY ≈33.7%) were determined, providing a comprehensive evaluation of the luminescence performance of the compound.

Hirshfeld surface analysis, reported here for the first time for the [Eu₂(o-MBA)₆(phen)₂] complex, revealed the role of intermolecular interactions in shaping the local crystal-field environment and the photoluminescence properties. Overall, the combined structural and spectroscopic investigations provide new insights into the relationships between local coordination geometry, intermolecular interactions, and the photoluminescence behavior of the [Eu₂(o-MBA)₆(phen)₂] complex.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. ChatGPT Plus was used for English improvement and for guiding through the SPECTRE User Manual.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fnano.2026.1875127/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Arun Kumar, Marche Polytechnic University, Italy

Reviewed by: Virender., Deenbandhu Chhotu Ram University of Science and Technology, India

Duan Gao, Liaoning Police Academy, China