Lanthanide nitrato complexes bridged by the bis-tridentate ligand 2,3,5,6-tetra(2-pyridyl)pyrazine: Syntheses, crystal structures, Hirshfeld surface analyses, luminescence properties, DFT calculations, and magnetic behavior

Abstract

Six dinuclear lanthanide(III) nitrato complexes [Ln(NO₃)₃(H₂O)]₂(μ-tppz) (where tppz = 2,3,5,6-tetra(2-pyridyl) pyrazine and Ln(III) = Nd (1), Sm (2), Eu (3), Gd (4), Tb (5), and Dy (6)) with bis-tridentate N-heterocyclic 2,3,5,6-tetra(2-pyridyl)pyrazine as bridging ligand have been solvothermally synthesized and characterized via elemental analysis, infrared spectroscopy, thermogravimetric analysis, single-crystal X-ray diffraction, and powder X-ray diffraction. The 3-D Hirshfeld surface and 2-D fingerprint plots show that the main interactions in 1–6 are the O⋯H/H⋯O intermolecular interactions with relative contributions of about 62%. Although the poor lanthanide(III)-centered luminescence properties clearly point to the efficiency of nonradiative quenching processes (presence of water molecules in the coordination sphere of the lanthanide(III) ions), the ligand tppz is better suited to sensitize the lanthanide(III)’s emissions of Eu^III and Nd^III than Sm^III, Tb^III, and Dy^III. Finally, the magnetic data of Dy^III comple×6 reveals antiferromagnetic coupling between Dy^III ions.

1. Introduction

On account of their excellent luminescent properties such as sharp line-like and high color-pure emissions, long lifetimes, and large Stokes/anti-Stoke shifts, lanthanide(III) complexes with luminescence arising from f–f transitions possesses diverse applications in the fields of lighting, telecommunications, road marking, laser materials, sensors for analytes, upconversion materials, bioimaging, immunoassays, medical diagnosis and therapy, luminescent lanthanide-based Single-Molecule Magnets (SMM), and so on [1–9]. To sensitize the Laporte-forbidden f–f transitions, various organic chromophores are selected to coordinate to the lanthanide(III) ions by utilizing the “antenna effect” [1,5,8,10]. Among them, multi-dentate N-heterocyclic ligands have been shown excellent antenna-ligands to sensitize the luminescence of lanthanide (III) complexes, due to their fascinating electrochemical, photophysical and photochemical properties stemming from the conjugated aromatic cores [11–14]. The π-conjugated N-heterocyclic ligand 2,3,5,6-tetra (2-pyridyl) pyrazine (tppz) possessing two trans-terpyridine coordination sites is one of the most fascinating molecules [15–21]. Although the tppz ligand was first synthesized by Goodwin and co-workers in 1959 [22], little work on tppz-containing complexes appeared until 1993 [23]. Since then, numerous studies on the synthesis and application of the tppz-based complexes have been reported [23–31]. Although most of them focused on complexes with transition metals, there has been a few cases with lanthanide(III) ions [15,17,32–35]. In particular, [Cp*₂Yb]₂(μ-tppz) bridged by tppz was reported by Morris and John in 2003 and 2006 to explore electron transfer molecular wire complexes [32,33]. [(Cp*₂Ln)₂(μ-tppz)]⁺ and [(Cp*₂Ln)₂(μ-tppz)]⁻ (Ln(III) = Gd, Tb, and Dy) were reported by Long in 2014 to demonstrate how tppz can be used to synthesize a series of dinuclear lanthanide(III) complexes for which the redox-active ligand tppz existed in the monoanionic and tri-anionic forms [34]. Fukuzumi used tppz as a probe molecule and a host ligand for metal ions, enabling “OFF-OFF-ON” switch ability for metal ion-fluorescence sensors and metal ion-promoted electron transfer reactions [35]. More recently, tppz was used to construct highly luminescent mononuclear lanthanide(III) complexes [15]. Owing the capability of tppz to possess different modes of coordination towards transition metal ions and lanthanide(III) ions, the challenge work, consisting of construction of d-f hetero-dinuclear luminescent complexes with tppz as π-conjugated bridge, is of great interest to us. To fulfill this goal, a series of lanthanide(III) tppz-containing complexes were firstly studied.

Herein, a series of dinuclear lanthanide(III) nitrato complexes, [Ln (NO₃)₃(H₂O)]₂(μ-tppz) (Ln(III) = Nd (1), Sm (2), Eu (3), Gd (4), Tb (5), and Dy (6)), bridged by the bis-tridentate N-heterocyclic ligand 2,3,5,6-tetra(2-pyridyl) pyrazine ligand, are reported. Complexes 1–6 were characterized by elemental analysis, infrared spectroscopy, thermogravimetric analysis, single-crystal X-ray diffraction, and powder X-ray diffraction. Of special importance, we investigated the near-infrared (1–2) and visible luminescent (2–6) properties for these solid-state compounds, whereas the Hirshfeld surface analysis was used to explore the interactions within the crystal structures of 1–6 for a deeper understanding of the chemical binding in these luminescent lanthanide (III) compounds. Furthermore, the HOMO–LUMO energy gaps of the structures of tppz, Eu^III complex 3, Gd^III complex 4, and Tb^III complex 5 were revealed by employing the density functional theory (DFT) calculations. Finally, the magnetic behavior of the Dy^III comple×6 was also studied.

2. Experimental

2.1. Materials and methods

Samarium nitrate hexahydrate (99.9%) and gadolinium nitrate hexahydrate (99.9%) were purchased from Amethyst Chemicals, whereas neodymium nitrate hexahydrate (99.9%), europium nitrate hexahydrate (99.9%), and terbium nitrate hexahydrate (99.9%) were purchased from Strem Chemicals. Dysprosium nitrate hexahydrate (99.9%) was purchased from Beijing HWRK Chem Ltd and 2,3,5,6-tetra (2-pyridyl)pyrazine (tppz) was synthesized according to the literature method [22].

[Nd(NO₃)₃(H₂O)]₂(μ-tppz) (1). 2,3,5,6-tetra(2-pyridyl)pyrazine (19.4 mg, 0.05 mmol) in 6 mL dichloromethane was added dropwise to a 3 mL acetonitrile solution containing Nd(NO₃)₃·6H₂O (43.8 mg, 0.1 mmol). The mixture was added to a 25 mL teflon-lined stainless auto-clave and heated at 120 °C for 2 days. Pale yellow crystals of 1 suitable for X-ray crystallographic measurements were collected by filtration, washed twice with dichloromethane, and then dried in air. Yield: 50.6 mg, 93.3% based on Nd(NO₃)₃·6H₂O. Anal. Calcd. for C₂₄H₂₀N₁₂O₂₀Nd₂: C, 26.55; H, 1.84; N, 15.49%; Found: C, 26.44; H, 1.85; N, 15.48%. IR (KBr pellets, cm⁻¹): 3350, 3095, 1634, 1598, 1503, 1385, 1281, 1157, 1024, 811, 785, 739.

[Sm(NO₃)₃(H₂O)]₂(μ-tppz) (2). For the synthesis of 2 the same procedure as that for 1 was employed, using the 2,3,5,6-tetra(2-pyridyl)pyrazine (19.4 mg, 0.05 mmol) and Sm(NO₃)₃·6H₂O (44.4 mg, 0.1 mmol). Yield 50.9 mg, 92.8% based on Sm(NO₃)₃·6H₂O. Calcd. for C₂₄H₂₀N₁₂O₂₀Sm₂: C, 26.25; H, 1.82; N, 15.31%; Found: C, 26.35; H, 1.83; N,15.37%. IR (KBr pellets, cm⁻¹): 3350, 3093, 1634, 1598, 1503, 1385, 1283, 1157, 1024, 812, 785, 738.

[Eu(NO₃)₃(H₂O)]₂(μ-tppz) (3). For the synthesis of 3 the same procedure as that for 1 was employed, using the 2,3,5,6-tetra(2-pyridyl) pyrazine (19.4 mg, 0.05 mmol) and Eu(NO₃)₃·6H₂O (44.6 mg, 0.1 mmol). Yield 50.1 mg, 91.1% based on Eu(NO₃)₃·6H₂O. Calcd. for C₂₄H₂₀N₁₂O₂₀Eu₂: C, 26.17; H, 1.82; N, 15.27%; Found: C, 25.95; H, 1.83; N,15.10%. IR (KBr pellets, cm⁻¹): 3356, 3095, 1635, 1598, 1505, 1386, 1286, 1158, 1025, 811, 785, 739.

[Gd(NO₃)₃(H₂O)]₂(μ-tppz) (4). For the synthesis of 4 the same procedure as that for 1 was employed, using the 2,3,5,6-tetra(2-pyridyl)pyrazine (19.4 mg, 0.05 mmol) and Gd(NO₃)₃·6H₂O (45.1 mg, 0.1 mmol). Yield 50.8 mg, 91.4% based on Gd(NO₃)₃·6H₂O. Calcd. for C₂₄H₂₀N₁₂O₂₀Gd₂: C, 25.92; H, 1.80; N, 15.12%; Found: C, 26.05; H, 1.81; N,15.08%. IR (KBr pellets, cm⁻¹): 3354, 3095, 1635, 1598, 1505, 1386, 1287, 1158, 1025, 812, 785, 741.

[Tb(NO₃)₃(H₂O)]₂(μ-tppz) (5). For the synthesis of 5 the same procedure as that for 1 was employed, using the 2,3,5,6-tetra(2-pyridyl)pyrazine (19.4 mg, 0.05 mmol) and Tb(NO₃)₃·6H₂O (45.3 mg, 0.1 mmol). Yield 51.1 mg, 91.7% based on Tb(NO₃)₃·6H₂O. Calcd. for C₂₄H₂₀N₁₂O₂₀Tb₂: C, 25.84; H, 1.79; N, 15.08%; Found: C, 25.76; H, 1.80; N, 15.05%. IR (KBr pellets, cm⁻¹): 3361, 3095, 1634, 1598, 1506, 1388, 1288, 1157, 1026, 811, 784, 742.

[Dy(NO₃)₃(H₂O)]₂(μ-tppz) (6). For the synthesis of 6 the same procedure as that for 1 was employed, using the 2,3,5,6-tetra(2-pyridyl)pyrazine (19.4 mg, 0.05 mmol) and Dy(NO₃)₃·6H₂O (45.6 mg, 0.1 mmol). Yield 51.3 mg, 91.5% based on Dy(NO₃)₃·6H₂O. Calcd. for C₂₄H₂₀N₁₂O₂₀Dy₂: C, 25.68; H, 1.78; N, 14.98%; Found: C, 25.59; H, 1.79; N, 14.95%. IR (KBr pellets, cm⁻¹): 3377, 3097, 1634, 1599, 1506, 1388, 1291, 1158, 1027, 811, 785, 745.

2.2. X-ray diffraction analysis

The diffraction data was collected at 293(2) K with an Agilent G8910A CCD diffractometer with graphite monochromated Mo-Kα radiation (λ = 0.71073 Å), using the ω-θ scan mode in the range 3.34° ≤ θ ≤ 25.01°. The structures were solved by direct methods and refined by full matrix least-squares on F². An empirical absorption correction was applied with the program SADABS. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms were added geometrically. Calculations and graphics were performed with SHELXL [36,37]. The crystallographic data and the refinement of structure parameters for 1–6 are summarized in Table S1. Selected bond lengths and angles for 1–6 are given in Table S2.

2.3. Other measurements

Elemental analyses (C, H, N) were performed with a vario MICRO cube elemental analyzer. FT–IR spectra were recorded from KBr pellets with a Bio-Rad FTS-7 spectrophotometer in the range of 4000–400 cm⁻¹. Thermogravimetric analyses (TGA) were run on a SDT-Q600 comprehensive thermal analyzer under N₂ atmosphere with a flow rate of 100 mL min⁻¹, a heating rate of 10 °C·min⁻¹, and a temperature range of 30–900 °C. Powder X-ray diffraction (PXRD) data were obtained on a PANalytical X’Pert³ power diffractometer with a PIXcel detector. The X-ray generator was operated at 40 kV and 40 mA. The Cu Kα line at 1.54056 Å was used as the radiation source. Samples were scanned from 5° to 50° (2θ) in stage sizes of 0.02626° with a scanning speed of 0.1347°/s. Hirshfeld surface analysis was performed using the CrystalExplorer program [38,39]. Solid-state luminescence properties of tppz and complexes 1–6 were recorded in the solid state at room temperature on an Edinburgh FLS920 lifetime and steady-state spectrometer. The absolute quantum yield of Eu^III complex 3 was obtained by following the standard procedures using the same Edinburgh FLS920 instrument, which is equipped with an integrating sphere.

2.4. Computational details

Geometries were optimized at the B3LYP level of theory [40]. The geometries obtained from X-ray diffraction data were used as input structures for the geometry optimizations, which were performed under the freezing of no-hydrogen atoms. The SDD basis set with large-core quasi-relativistic Pseudopotential ECP52MWB was used for Eu atoms, ECP53MWB for Gd atoms, and ECP54MWB for Tb atoms [41–43], whereas the 6–31G(d) basis set was employed for all other atoms [44, 45]. The spin state of all the optimized neutral structures was considered as a singlet. Frequencies were analytically computed at the same level of theory to obtain the gas phase energies, and to confirm whether the structures are minima (no imaginary frequency). All of the calculations were performed using the Gaussian 16 program package, and the structures and molecule orbitals (MOs) were generated by GaussView 6.0. And the Cartesian coordinates of the ligand tppz and the complexes 3–5 used in the DFT calculation are provided in Table S8.

3. Results and discussion

3.1. Crystal structures

Complexes 1–6 are isostructural and crystallize in the monoclinic space group P2₁/n. Therefore, Eu^III complex 3 was selected as a representative of the six compounds to discuss in detail their X-ray structures. The molecular structure of 3 is shown in Fig. 1 (a). Each ten-coordinated Eu^III center is surrounded by three nitrogen atoms located on one side of the bis-tridentate ligand tppz (N2 comes from the pyrazine, whereas N1 and N3 originate from the pyridine), six oxygen atoms (O1 and O3, O4 and O6, O7 and O9) of three bidentate nitrates, and one oxygen atom (O10) from the coordinated water. The analysis of the coordination geometries around the Ln^III center using SHAPE 2.1 reveals that the LnN₃O₇ lanthanide cores adopt a Sphenocorona J87 geometry (Fig. 1b) [46–48]. The different figures of merits calculated for 1–6 are listed in Table S3.

Fig. 1.

(a) Molecular structure of [Eu(NO₃)₃(H₂O)]₂(μ-tppz) (3) showing 50% thermal ellipsoids (hydrogen atoms are not shown for clarity, symmertry code a: 1-x, 1-y, 2-z); (b) Sphenocorona J87 geometry of EuN₃O₇ in 3; (c) and (d) Perspective view of the molecular structure of 3 showing the pyrazine and pyridyl rings numbering, symmertry code a: 1-x,1-y, 2-z.

In order to describe the structural characteristics of the coordinated tppz ligand, the label of the pyrazine and pyridyl rings was done in accordance with the reported notation found in the literature [19]. The pyrazine ring is labeled as E and the four pyridyl rings are labeled as A, B, C and D corresponding to the rings on the 2, 3, 5, and 6 positions of the pyrazine, respectively (Fig. 1c and d). There are fourteen conformations of tppz, which can be divided into five families based on the relative positions of the four pyridyl nitrogen atoms. According to the torsion angles (N2C7C8N1 = 33.577°, N2C5C6N3 = −31.577°, N2aC7aC8aN1a = −33.577°, and N2aC5aC6aN3a = 31.577°), the conformation of tppz in 1–6 could be designated as 5XXXX. The number 5 indicates that the nitrogen atoms in the A and D rings are “up” and the nitrogen atoms in the B and C rings are “down”. The letter “X” following the number 5 indicates an “exo” conformation, where the pyridyl nitrogen is pointed towards the pyrazine nitrogens (|Npz–C–C–Npy| < 90°) [19]. The dihedral angles between the ring planes are listed in Table 1. The dihedral angles between the A-C rings and B-D rings are 0.311° and 0.689°, respectively. This suggests that the two rings are roughly coplanar. On the other hand, the dihedral angles between the A-B rings and A-D rings are 49.104° and 49.304°, which are consistent with the reported value of 50° [19]. Finally, it must be mentioned that the pyridyl rings A and D twist from coplanarity with the pyrazine E by 32.734° and 36.908°, respectively.

Table 1.

Dihedral angles of the pyridyl rings (A-D) with respect to the pyrazine ring (E) in 3.

Rings	Angle (°)	Rings	Angle (°)
A-E	32.734	A-B	49.104
A-D	49.304	D-E	36.908
A-C	0.311	B-D	0.689

The Eu–N and Eu–O bond distances in the 2.583–2.609 Å and 2.391–2.604 Å ranges, are comparable to those reported for lanthanide complexes (2.556–2.610 and 2.309–2.620 Å), respectively [49–51]. The average Ln-O and Ln-N bond lengths (Table S2 of ESI) and the distances between the Ln^III⋯Ln^III centers (Table 2) decrease gradually from 1 to 6, which are consistent with the effect of the lanthanide contraction [52, 53]. The cross-bridge Ln^III⋯Ln^III distances (7.694–7.871 Å) are within those reported for the [Cp*₂Yb]₂(tppz) (7.57 Å) [33] and [(Cp*₂Dy)₂(μ-tppz)](BPh₄) (7.71 Å) complexes [34], but longer than the ones observed for the transition metal tppz complexes (6.497–7.523 Å) [27,30,54–59].

Table 2.

Ln^III⋯Ln^III distances in complexes 1–6.

LnIII⋯LnIII	Distances (Å)	LnIII⋯LnIII	Distances (Å)
NdIII⋯dIII	7.871	GdIII⋯GdIII	7.759
SmIII⋯SmIII	7.815	TbIII⋯TbIII	7.718
EuIII⋯EuIII	7.753	DyIII⋯DyIII	7.694

As shown in Fig. 2, the dinuclear lanthanide units show a parallel arrangement to each other through layered π-π type interactions originated from the two coordinated pyridyl rings (perpendicular distance of 3.469 Å, centroid-to-centroid distance of 4.303 Å, and offset angle of 60.53°). Furthermore, they are linked together by the hydrogen bonds O10–H10A⋯O3, O10–H10A⋯O2, and O10–H10B⋯O2, which involve the coordinated water molecule (O10) and two coordinated nitrate oxygen atoms (O2 and O3) and, thus, giving rise to a two-dimensional frame network (Table S4 of ESI).

Fig. 2.

(a) 2D structure of 3 interconnected by hydrogen bonds and π-π stacking interactions; (b) A view of the hydrogen bonds and π-π stacking interactions (yellow dotted lines) linking the double Ln^III units.

3.2. IR spectroscopy

The IR spectral data of the tppz ligand (Fig. S1) and complexes 1–6 (Fig. S2) are listed in Table S5. The broad absorption bands around 3400 cm⁻¹ shown in the spectra are assigned to ʋ(O–H), indicating the presence of water molecules. The aromatic C–H stretching vibrations at 3093 – 3097 cm⁻¹ observed in the complexes shift towards higher wave numbers with respect to the free ligand (3049 cm⁻¹). After coordination to the lanthanide centers the ʋ(C=C) and ʋ(C=N) stretching vibrations of the free ligand at 1585 and 1564 cm⁻¹ are blue-shifted to 1634 and 1598 cm⁻¹ for 1–6, respectively [27,54]. In addition, the ʋ(C–C) and ʋ (C–N) of the free ligand at 1391, 1125, and 1035 cm⁻¹, and the aromatic C–H deformation vibrations of adjacent hydrogen from the benzene ring at around 806, 784 and 753 cm⁻¹ are somewhat shifted upon coordination to the lanthanide ions. These observations are consistent with those reported in the literature and suggesting the coordination of the tppz ligand to the lanthanide ions [31,57]. The vibration of nitrate ν_as (NO₃) appears at 1503–1506 cm⁻¹ (ν₄) and 1281–1291 cm⁻¹ (ν₁). The differences of 215–222 cm⁻¹ between ν₄ and ν₁ suggest a bidentate chelating mode of the nitrato ligand [60]. All of the spectroscopic observations have been confirmed by the X-ray structures of 1–6.

3.3. Thermogravimetric analysis and PXRD

The thermal properties from crystalline samples of 1–6 were investigated. They all show a similar thermal decomposition behavior. The TGA/DTG curves of 1–6 all show a two-step weight loss pattern, as seen in Fig. S3 and from the data summarized in Table S6. Thus, the Nd^III complex 1 was selected as a representative of the six compounds to discuss in detail their thermogravimetric analyses. The first decomposition step, which is attributed to the loss of the coordinated water molecule, occurs in the temperature range of 187–256 °C with a weight loss of 3.37 % (calcd. 3.32 %). The second step from 322 to 616 °C is assigned to the removal of the tppz ligand and nitrates with a weight loss of 69.66 % (calcd. 70.18 %). As a result, the final residue is attributed to Ln₂O₃, as suggested by the weight loss calculations.The powder XRD patterns of 1–6 were measured at room temperature. It must be noted that the simulated X-ray powder curves obtained from the single-crystal diffraction data and the experimental curves from bulk samples were also calculated and compared (Fig. S4). The simulated intensity and diffraction angles of the main diffraction peaks of 1–6 are in agreement with the experimental ones, which confirmed the phase purity of the complexes [52,61].

3.4. Hirshfeld surface analysis

To get a deeper insight into the intermolecular interactions within the crystal structures, CrystalExplorer was used to analyze the crystal structures of 1–6 [62,63]. The 3D Hirshfeld surface and the associated two-dimensional (2D) fingerprint plots of 3 (as an example since the six compounds showed similar patterns) are plotted in Fig. 3. The Hirshfeld surface is mapped with d_norm (the normalized contact distance), which is normalized from d_e (the nearest external distance), d_i (the nearest internal distance), and the van der Waals (vdW) radii of the two atoms to the surface. The fingerprint plots are derived from the 3D Hirshfeld surface and provide a visual summary of the frequency of each combination of d_e and d_i across the surface of a molecule [39,52]. Table S7 summarizes the relative contributions of various intermolecular interactions to the Hirshfeld surfaces in the 1–6 crystal structures.

Fig. 3.

(a) Hirshfeld surface mapped with d_norm, (b) shape index, (c) curvedness, and (d–f) fingerprint plots for Eu^III complex 3.

In the d_norm Hirshfeld surface, the regions of shortest intermolecular contacts are visualized by red circle areas [64]. The main partial contributions of the different intermolecular contacts are H⋯O and O⋯H contacts in 3 with a relative contribution of 62.3% (Fig. 3a). In fact, the H⋯O/O⋯H contacts vary from 57.5% in 1–62.4% in 5 (Table S7). Thus, the Hirshfeld surface analysis for the X-ray structure of the six complexes reveals that the crystal packing is primarily determined by H⋯O and O⋯H intermolecular contacts.⁶² The red triangles in the shape index plot represent concave regions, which result from the carbon atoms of the π-stacked molecule above it. On the other hand, the blue triangles in the curvedness plot refer to convex regions because of the ring carbon atoms of the molecule inside the surface. The outline of alternating red and blue triangles is an indicator of π⋯π stacking interactions [65]. The π⋯π stacking information conveyed by the shape index (Fig. 3b) and the curvedness plot (Fig. 3c) are consistent with the crystal structure analyses.

3.5. UV–Vis absorption and photoluminescence

Ambient-temperature UV–Vis absorption spectra (Fig. S5) of tppz ligand and complexes 1–6 were recorded in DMF solution (1 × 10⁻⁶ mol/L). The spectrum of tppz displays two absorption bands positioned at around 278 and 310 nm, which can be ascribed to the n→π* or π→π* transitions centered on the π-conjugated unit, respectively. Upon complexation to the Ln^III center, the bands observed in the UV–Vis absorption spectrum of tppz are slightly blue-shifted to about 276 and 308 nm. Similar results were obtained for all complexes.

The solid-state luminescent properties of tppz and 1–6 were investigated at 77 and 298 K. Under excitation at 342 nm, the emission spectrum of tppz exhibits one broad maximum emission centered at 385 nm at room temperature, which originates from the ligand-centered (LC) singlet excited state energy level ¹ππ* (Fig. S6). Knowing that tppz is known to form two polymorphs (monoclinic P2₁/n and tetragonal I4₁/a), the observed solid-state luminescent features are consistent with those reported for the monoclinic polymorph. At 77 K, the emission band maxima at about 392 and 541 nm are assigned to the singlet and triplet excited state energy levels ¹ππ* and ³ππ*, respectively (Fig. S7). On the other hand, the time-resolved emission spectrum recorded with a time delay of 1.0 ms only shows the triplet excited state energy level ³ππ* with a maximum around 542 nm.

Upon complexation to the nonluminescent Gd^III center, the maximum of the emission band of the ¹ππ* is red-shifted to about 476 nm with a much weaker and broad emission at 627 nm tentatively ascribed to the ligand-to-ligand charge transfer transitions (LLCT) [54, 55,66] (Fig. S8, room temperature). At 77 K, the steady-state emission spectrum (Fig. S9, left) showed that the LLCT band at around 633 nm becomes stronger than the ¹ππ* band at 474 nm.^{52, 64} Since the nonluminescent Gd^III ion is commonly used to determine the position of the excited triplet state energy level of coordinated ligands, the time-resolved luminescence spectrum of 4, measured at 77 K with a time delay of 1 ms (Fig. S9, right), showed a more structure emission band with two maxima at about 538 and 571 nm, and thus corresponding to a triplet energy level located at about 538 nm (or 18,587 cm⁻¹). It is worth noting that this value is very close to the one observed for the triplet energy level of 4′-phenyl-2,2′:6′,2″-terpyridine (18,500 cm⁻¹) [52]. Knowing that the corresponding singlet energy level is located at about 474 nm (21,097 cm⁻¹), the energy difference ΔE(¹ππ*−³ππ*) is determined to be 2510 cm⁻¹, suggesting the presence of an inefficient intersystem crossing (ISC) process. This is in agreement with the fact that one expects that an efficient ISC process takes place when ΔE (¹ππ*−³ππ*) is greater than 5000 cm⁻¹ [2,67].

Upon excitation of the ligand-centered absorption band, the emission spectra of 1, 2, 3, and 5 are dominated by narrow bands corresponding to the typical Nd^III, Sm^III, Eu^III, and Tb^III lanthanide-centered transitions, respectively. The Nd^III complex 1 shows typical NIR emissions upon excitation at 449 nm. As shown in Fig. 4a, 1 displays the strong characteristic emission band of the Nd^III ion centered at 1054 nm and assigned to the ⁴F_3/2 → ⁴I_11/2 transition. It can also be seen the two splitting peaks of the ⁴F_3/2 → ⁴I_9/2 (881 and 901 nm) and ⁴F_3/2 → ⁴I_13/2 (1321 and 1333 nm) transitions. The energy of the ligand triplet state (18,587 cm⁻¹), which is much higher than the corresponding emitting levels of Nd^III (11,460 cm⁻¹), thus facilitating their population through ligand-to-metal energy transfer [68]. Indeed, the observation of the characteristic emission bands of the Nd^III confirms that the tppz moieties can act as effective sensors for the luminescence of Nd^III ions [5,67].

Fig. 4.

(a–d) Solid-state excitation (λ_em = 1056, 494, 616 and 541 nm) and emission (λ_ex = 449, 375, 439 and 337 nm) and emission spectra of 1, 2, 3 and 5 at room temperature; (e) CIE chromaticity coordinates.

As shown in Fig. 4b, Sm^III complex 2 exhibits weak characteristic emission bands of the Sm^III ion upon excitation at 375 nm at room temperature. The bands observed at 561, 599, 642, and 705 nm correspond to ⁴G_5/2 → ⁶H_5/2 (forbidden), ⁴G_5/2 → ⁶H_7/2 (magnetic dipole), ⁴G_5/2 → ⁶H_9/2 (electric dipole), and ⁴G_5/2 → ⁶H_11/2 transitions, respectively [69]. The observation of the ligand-centered emission around 450–470 nm in the emission spectrum of the Sm^III complex at 77 K suggests an inefficient ISC, thus leading to unfavorable conditions to observe significantly the characteristic Sm^III-centered emission (Fig. S10) [70].

The luminescent bands of Eu^III in complex 3 (Fig. 4c) appear at 579, 592, 616, 648 and 689 nm, and are assigned to the ⁵D₀ → ⁷F_J (J = 0, 1, 2, 3, and 4) transitions of the Eu^III center with the CIE value of (0.6654, 0.3344) (Fig. 4e), respectively. The hypersensitive electric dipole transition ⁵D₀ → ⁷F₂ at 616 nm is the strongest emission band in the luminescence spectrum of 3 measured at 298 K and upon excitation of 449 nm. These observations suggest that the Eu^III ion occupied a site of low symmetry without an inversion center. Furthermore, the appearance of the symmetry-forbidden emission ⁵D₀ → ⁷F₀ at 579 nm indicates that the Eu^III cation possesses a non-centrosymmetric environment [52]. The absolute quantum yield of 3 is 1.31%, which is much less than the reported value (16.8%) of the terpyridine analogue complex [Eu (NO₃)₃(ptpy)(H₂O)] (where ptpy is 4′-phenyl-2,2′:6′,2′′-terpyridine) [52]. In addition, the ⁵D₀→⁷F₂ transition was used to measure the luminescence lifetime. The decay curve of 3 was fitted using a mono-exponential function, which gave a lifetime of 0.035 ms that is much shorter than the one observed for [Eu(NO₃)₃(ptpy)(H₂O)] (0.545 ms) [52]. It can be concluded from the low absolute quantum yield and much shorter luminescence lifetime observed for complex 3 that the tppz ligand does not have efficient sensitization abilities towards the Eu^III cation unlike its analog ptpy ligand. However, it is noteworthy that the ligand-centered emissions are not observed in the steady-state and time-resolved emission spectra of 3 (Fig. S11) at 77 K, indicating efficient energy transfer processes taking place [71]. Additionally, the efficiency of the luminescence sensitization by the ligand, Φ_sen, amounted to 85.6% (see the Supporting Information for a detailed determination of this value).

Upon excitation of the ligand-centered absorption band at room temperature, the Tb^III complex 5 displays the emission bands of ⁵D₄ → ⁷F₆ (488 nm), ⁵D₄ → ⁷F₅ (541 nm), ⁵D₄ ⁷F₄ (581 nm), ⁵D₄ → ⁷F₃ (622 nm), ⁵D₄ → ⁷F₂ (647 nm), ⁵D₄ → ⁷F₁ (668 nm), and ⁵D₄ → ⁷F₀ (680 nm) of the Tb^III cation (Fig. 4d), which are responsible for the characteristic green color of Tb^III with the CIE value of (0.2985,0.5620) (Fig. 4e) [72]. The observation of the emission of the excited singlet state energy level ¹ππ* in the luminescence spectra of 5 at room temperature and 77 K confirms that there is an inefficient ISC process taking place (Fig. S11).

At room temperature, the Dy^III complex 6 only exhibits one broad emission centered at 483 nm (Fig. S12), which is assigned to the ligand-centered (LC) ¹ππ* transition. At 77 K, this emission band appears more structured with a maximum around 463 nm (Fig. S12). Although the emission spectrum does not show the characteristic emission bands of the Dy^III ion, the red-shift of the emission band of the ¹ππ* state in the spectrum of 6 compared to its position for the free ligand confirms the coordination of the tppz ligand to the Dy^III cation (a difference of about 94 nm). Like for the Sm^III ion, it must be concluded that the tppz ligand is not a good sensor for the luminescence of the Dy^III cation.

Normally, it necessitates the energy of the lowest excited triplet state level of the sensitizer to be nearly equal or above the resonance level. The energy-level match between the triplet state of the antenna (aka coordinated ligand) and the resonance level of Ln^III ions is a vital factor that defines the luminescence efficiency of Ln^III complexes. Evidently, the effectiveness of the energy transfer from the antenna moiety to the Ln^III cation is somewhat proportional to the extent of overlap between the phosphorescence spectrum of the ligand and the absorption spectrum of the Ln^III ion. An effective ligand to Ln^III ion energy transfer is apparent from the quenching of the emission of the coordinated ligands [2]. The determined triplet energy level of the ligand (18,587 cm⁻¹) is 1316 and 762 cm⁻¹ higher than the excited state of Eu^III (17,271 cm⁻¹) and Sm^III (17825 cm⁻¹), but less than the excited states of Tb^III (20,400 cm⁻¹), and Dy^III (21,000 cm⁻¹). As a result, one expects that the energy transfer from tppz to the Sm^III, Tb^III, and Dy^III centers is not favorable. It is worth noting that the 0-phonon of the ³ππ* state of the complexed ligand in the Gd^III-containing compound 4 at 18,587 cm⁻¹ is better suited to sensitize the Eu^III (17,271 cm⁻¹) ions than the Sm^III (17,825 cm⁻¹), Tb^III (20,400 cm⁻¹) and Dy^III (21,000 cm⁻¹) cations. It must be pointed out that it was not possible to obtain a measurable lifetime and quantum yield for the Tb^III complex 5. This is also in line with the fact that the Eu^III complex 3 has a low quantum yield (1.31%) and short lifetime (0.035 ms). These poor luminescence properties may be explained by the presence of O⋯H/H⋯O intermolecular interactions [73], as evidenced by the 3-D Hirshfeld surface and 2-D fingerprint plots. The O⋯H/H⋯O intermolecular interactions are the main’s interactions in 1–6 with a contribution of 62% as determined for 3. As a result, the presence of O–H oscillators from water molecules in the coordination sphere of the 1–6 compounds effectively quench the characteristic luminescence of the lanthanide(III) ions.^{50, 66, 68}

3.6. DFT calculations

Density functional theory (DFT) calculations were conducted to unravel the HOMO–LUMO energy gaps of the structures of tppz and complexes 3, 4 and 5. The HOMO–LUMO energy gaps are listed in Table 3. As shown in Fig. 5, the energy gaps between HOMO and LUMO for tppz, 3, 4 and 5 are 4.80, 3.38, 3.30, and 3.37 eV, respectively. It is worth mentioning that the HOMO and LUMO profiles are all located on the pyrazine and pyridyl groups for the tppz ligand. After coordination with the lanthanide(III) center in 3, 4 and 5, the HOMO profiles are more located on one of the nitrates, whereas the LUMOs are located on the tppz ligand. As a result, the energy gaps between the HOMOs and LUMOs decrease by 1.42–1.50 eV (11,453–12,098 cm⁻¹), which corroborate the formation of the complexes 3, 4 and 5. The structure calculated for the complexes 3, 4 and 5 presents average Eu–N, Gd–N and Tb–N distances of 2.601, 2.596 and 2.580 Å, respectively, which compare well with the average experimental values obtained from X-ray diffraction measurements (Eu–N: 2.601, Gd–N: 2.597 and Tb–N: 2.581 Å).

Table 3.

Theoretically energy values for tppz and complexes 3–5.

Compd.	Energy
	LUMO (eV)	HOMO (eV)	ΔE (eV)	ΔE (cm−1)
Tppz	−1.77	−6.57	4.80	38,714
3-Eu	−3.96	−7.34	3.38	27,261
4-Gd	−3.98	−7.28	3.30	26,616
5-Tb	−3.95	−7.32	3.37	27,181

Fig. 5.

HOMOs, LUMOs and HOMO-LUMO gaps of tppz, complexes 3, 4 and 5.

3.7. Magnetic properties

The temperature dependence of the magnetic susceptibilities of the Dy^III comple×6 was recorded from 2 to 300 K at 1000 Oe, as depicted in Fig. 6. It shows that the χ_m value decreases with T in a monotonous form (going 0.09 and 9.46 cm³ mol⁻¹ from 300 to 2 K). The χ_mT values (where χ_m being the molar magnetic susceptibility) of 6 at 300 K is 28.01 cm³ K mol⁻¹, which is slightly lower than the theoretical value of 28.34 cm³ K mol⁻¹ for two magnetic uncoupled Dy^III ions (⁶H_15/2, g = 4/3 and χ_mT = 14.17 cm³ K mol⁻¹). The χ_mT value decreases gradually and reaches a minimum of 18.95 cm³ K mol⁻¹ at 2 K. The decrease of χ_mT profiles may arise from the progressive excited depopulation of stark sublevel and/or non-negligible intra- or intermolecular antiferromagnetic interactions between the Dy^III ions [74,75]. From Fig. 6 inset, it can be seen that the temperature dependence of the reciprocal susceptibility (${\chi }_{\text{m}}^{-1}$) above 50 K follows the Curie–Weiss law [χ_m = C/(T − θ)] with a Weiss constant of θ = − 9.87 K and Curie constant of C = 28.99 mol cm⁻³. Both the overall profile of χ_mT-T curve and the negative θ value might suggest antiferromagnetic interactions between Dy^III ions in the system [76].

Fig. 6.

Plots of χ_mT-T, χ_m-T and Curie-Weiss fitting for the ${\chi }_{\text{m}}^{-1}-\text{T}$ data of 6.

4. Conclusions

The characterizations, Hirshfeld surface analyses and solid-state luminescence properties of six dinuclear bis-tridentate N-heterocyclic ligand bridging lanthanide complexes [Ln(NO₃)₃(H₂O)]₂(μ-tppz) (Ln (III) = Nd (1), Sm (2), Eu (3), Gd (4), Tb (5), and Dy (6)) were reported. Complexes 1 and 3 exhibit characteristic emissions of the central Nd^III and Eu^III ions under the UV excitation, which suggest that the tppz ligand could be a good organic chromophore to absorb energy and transfer it to these lanthanide(III) ions. Unlike for the Nd^III and Eu^III ions, tppz is not a good sensor for the luminescence of the Tb^III, Sm^III, and Dy^III cations (the emission spectra of 2, 5 and 6 mainly displayed the ligand-centered (LC) ¹ππ* emissions). Finally, the DFT calculations were conducted to unravel the HOMO–LUMO energy gaps of the structures of tppz, 3, 4, and 5. After coordination of tppz to the lanthanide(III) center, it showed that the HOMO–LUMO energy gaps decreased to 1.42–1.50 eV (11,453–12,098 cm⁻¹). This observed decrease was a confirmation of the formation of the complexes studied. Additionally, the magnetic data of 6 confirmed the antiferromagnetic coupling phenomenon between the Dy^III ions. It is worth noting that the ligand tppz is better suited to sensitize Eu^III and Nd^III, than the other lanthanide(III) ions investigated (Sm^III, Tb^III, and Dy^III). Additionally, the overall poor luminescence properties of all lanthanide(III)-containing compounds are most likely the consequence of the presence of nonradiative quenching processes such as the presence of water molecules in the coordination sphere of the lanthanide(III) ions (important contribution of O⋯H/H⋯O intermolecular interactions observed for the Eu^III complex 3). To remedy that, ongoing work involves the formation of related lanthanide(III) complexes by using co-ligands to act as additional sensitizers to replace the OH quenchers in order to improve the luminescent properties of the tppz-bridging systems. Thus, the study of such hetero-binuclear complexes with this bis-tridentate ligand is currently under investigation.