Introduction

It is well known that the design and synthesis of aromatic carboxylic acids toward rare earth compounds have received considerable attention, due to their intriguing network topologies and promising applications in fields such as catalysis, ion exchange, gas storage, sensors, luminescence, antimicrobial [1,2,3,4,5,6,7,8]. Particularly, Eu(III) and Tb(III) ions could lead to interesting and excellent luminescence emissions in visible or near-infrared regions because the narrow f → f transitions derived from 4f electrons. Benzoic acid and its derivatives have a certain rigidity and stability in the structure. Their compounds with rare earth ions not only have a variety of interesting structures but also have good luminescence properties. At the same time, nitrogen heterocyclic ligands are introduced into the compounds of rare earth carboxylic acids, which can greatly improve the luminous intensity and thermal stability of the compounds [9,10,11]. Up to now, numerous one-dimensional (1D), two-dimensional (2D) and three-dimensional (3D) rare earth compounds have been synthesized with aromatic carboxylic acids and nitrogen heterocyclic ligands [12,13,14,15]. In addition, the research on the antibacterial activity of rare earth complexes is of great significance. For example, the compounds can selectively inhibit plant pathogenic bacteria and also have good bacteriostatic activity against Candida albicans [4].

In this report, two new rare earth compounds [Ln(3-MOBA)3(terpy)(H2O)]2 (Ln = Tb (1), Eu (2); 3-MOBA = 3-methoxybenzoate, terpy = 2,2′:6′,2″-terpyridine) were synthesized. Two asymmetric dinuclear molecules and 1D chain along the x-axis direction of 1 are combined by the offset face-to-face π···π weak stacking interactions. In order to evaluate the thermal effect and thermal decomposition mechanism on the target compounds, thermogravimetric and differential scanning calorimetry coupled to infrared spectrometer (TG/DSC–FTIR) technology were performed. Additionally, the luminescent properties and antibacterial activities of the title compounds were investigated.

Experimental

Materials

LnCl3·6H2O was acquired by the reaction of Tb4O7 and Eu2O3, respectively (Ln = Tb, Eu, Beijing Lanthanide Innovation Technology Co., Ltd, 99.9%), and hydrochloric acid in aqueous solution. The other analytically pure chemicals were purchased and used without further purification.

Synthesis of [Ln(3-MOBA)3(terpy)(H2O)]2 (Ln = Tb (1), Eu (2))

Dissolve 3-MOHBA (0.6 mmol) and terpy (0.2 mmol) in ethanol (95%) and adjust the pH of the solution about 6–7 with the prepared NaOH solution (1 mol L−1). Add the mixed ligands solution to LnCl3·6H2O (0.2 mmol) aqueous solution under stirring. The precipitate was filtered off and washed with 95% ethanol and then dried. Elemental analysis: calcd. (%) for C78H68Tb2N6O20: C 54.24, H 3.97, N 4.87, Tb 18.40; found (%): C 54.04, H 3.98, N 4.83, Tb 18.52. Calcd. (%) for C78H68Eu2N6O20: C 54.68, H 4.00, N 4.91, Eu 17.74; found (%): C 54.22, H 3.97, N 4.94, Eu 17.57.

Physical measurements

Analyses for C, H, N were carried out on a Vario-EL II element analyzer. The IR spectra were measured in the range of 4000–400 cm−1 on a Bruker Tensor 27 spectrometer using KBr medium pellets. X-ray powder diffraction identification was carried out for the crystalline analyses by a D8 ADVANCE X-ray diffractometer in a scanning range of 5°–50° (2θ) with Cu Kα radiation (λ = 0.15418 Å, Bruker, Germany). The data of single-crystal X-ray diffraction were collected on a Smart-1000 diffractometer with graphite-monochromatic Mo Kα (λ = 0.71073 Å) for compounds 1 and 2 at 298(2) K. The structures were solved using the SHELXS-97 program and refined with full-matrix least squares on F2 using the SHELXL-97 program. TG/DTG-FTIR analyses were carried out with 3–6 mg compounds at a heating rate of 10 K min−1 (simulated air atmosphere) on a NETZSCH STA 449 F3 instrument coupled with Bruker Tensor 27 Fourier transform infrared spectrometer. The luminescence spectra were measured on an F-4500 Hitachi Spectrophotometer. The solid luminescence quantum yields were measured using C9920-02G Hamamatsu test system, which was constituted by integrating sphere of 10-inch diameter and connected to a CCD detector. The antibacterial activities of LnCl3·6H2O, ligands and the title compounds at 303.15 K were studied with four respective concentrations: 8 × 10−3, 1.6 × 10−2 and 3.2 × 10−2 mol L−1 in sterile DMSO. The filter diameter is 6 mm, and the sample volume of the compounds is 5 μL.

Results and discussion

Infrared spectra

The absorption bands of 3-MOHBA, terpy and compounds 1–2 are listed in Table 1. The similar IR spectra of 1 and 2 suggest that the compounds are isostructural, which is further proved by the X-ray powder diffraction. The new characteristic bands of νas(COO–) and νs(COO–) for the title compounds are observed at 1536 cm−1 and 1402, 1401 cm−1. The characteristic absorption band of νC=O (1691 cm−1) for 3-MOHBA ligand disappeared in the IR spectra of the compounds. At the same time, the νLn–O stretching vibration band of compounds is observed at 413 and 412 cm−1. All of them indicate that oxygen atoms of the carboxylate groups are coordinated to the Ln(III) ions [16]. The bands νC=N and δC–H have shifted to higher wave numbers in the compounds compared to the free terpy ligand, proving that the nitrogen atoms of terpy ligand are also coordinated to the Ln(III) ion [17].

Table 1 IR absorption bands for the ligands and compounds
Full size table

X-ray powder diffraction

The X-ray powder diffraction patterns have been measured at room temperature. The simulated curve of 1 and experimental curves of 1–2 are shown in Fig. 1. The experimental curve of 1 is almost in agreement with the simulated curve, showing that the structure of the powder of 1 is the same to the pure crystal. In addition, compounds 1–2 are isostructural due to the similar experimental curves [18].

Fig. 1
figure 1

Simulated and experimental XRD patterns of compounds (a: simulated pattern of 1; b: experimental pattern of 1; c: experimental pattern of 2)

Full size image

Crystal Structure Description of [Tb(3-MOBA)3(terpy)(H2O)]2

X-ray crystallographic data and structural refinement parameters for 1 are listed in Table 2. Compound 1 crystallizes in triclinic system, Pī space group. There are two asymmetric dinuclear molecules in unit cell. They are combined by the offset face-to-face π···π weak stacking interactions between terpy rings containing N1 and N5 of the neighboring dinuclear units, with the distance of 3.6175 Å. Each dinuclear unit in 1 consists of two Tb(III) ions, six 3-MOBA− ligands, two terpy molecules and two coordinated water molecules (Fig. 2a). Each Tb(III) ion is coordinated to nine atoms, of which five oxygen atoms are from the monodentate, bridging and bidentate chelating carboxylates, one oxygen atom is from the coordinated water molecule, and three nitrogen atoms are from terpy ligand. These nine atoms act in a distorted monocapped square antiprism (Fig. 2b) [19]. The bond lengths of Tb–O and Tb–N are listed in Table 3. The average distance of Tb1–O is 2.417(1) Å, which is shorter than that of Tb1–N bond of 2.619(5) Å. The average distance of Tb2–O is 2.414(1) Å, which is also shorter than that of Tb2–N bond of 2.601(5) Å. The 3-MOBA− ligands can be well coordinated with Tb(III) ion due to the electrostatic effects.

Table 2 Crystallographic data for compound 1
Full size table
Fig. 2
figure 2

a Crystal structure of 1 and b coordination geometry of Tb(III) ion

Full size image
Table 3 Selected bond lengths (Å) of 1
Full size table

The asymmetric units are further connected to form 1D chain along the x-axis direction through the offset face-to-face π···π weak stacking interactions between terpy rings containing N2 and N4 on neighboring as shown in Fig. 3. The distance of the terpy rings is 3.5165 Å.

Fig. 3
figure 3

1D chain structure along the x-axis

Full size image

The structure of 1 is similar to that of compound Tb(m-MOBA)3(phen)]2·2C2H5OH (m-MOBA = m-methoxybenzoate, phen = 1,10-phenanthroline) [20] and [Tb(4-EBA)3 (terpy)H2O]2 (4-EBA = 4-ethylbenzoate, terpy = 2,2′:6′,2″-terpyridine) [21]. The difference is that they both have one dinuclear molecule, which is different from 1. Therefore, the choice of carboxylate ligands and nitrogenous heterocyclic ligands plays an important role in the structure regulation of the compound.

Thermal decomposition processes of the compounds

The TG/DTG-DSC methods were used to describe thermal decomposition of 1 and 2 in air, as shown in Fig. 4a, b. The thermal analytical data are listed in Table 4. The 3D stacked plots of FTIR spectra of gaseous products for 1 and 2 are shown in Figs. 5a and 6a. The IR spectra of the evolved gases of them at different temperature are performed in Figs. 5b and 6b. Their thermal behaviors are similar, so 1 would be described in detail as the representative.

Fig. 4
figure 4

TG/DTG/DSC curves of 1 (a) and 2 (b)

Full size image
Table 4 Thermal analytical data for 1–2
Full size table
Fig. 5
figure 5

3D stacked plots of the FTIR spectra of the evolved gases (a) and the FTIR spectra of the evolved gases (b) for 1

Full size image
Fig. 6
figure 6

3D stacked plots of the FTIR spectra of the evolved gases (a) and the FTIR spectra of the evolved gases (b) for 2

Full size image

In Fig. 4a, compound 1 has three decomposition steps, which is consistent with the FTIR spectra of the evolved gases (Fig. 5). The first step mass loss of 1.98% (calcd: 2.09%) between 339.15 and 453.15 K is attributed to the release of two coordinated water molecules. However, the gas signal is too weak to detect in this temperature range (Fig. 5a). In the temperature of 453.15–629.15 K, there is obvious mass loss of 26.94% (calcd.: 27.01%) corresponding to the decomposition of terpy ligands. Meanwhile, the weak absorption peaks of CO2 are found in the range of 2404–2227 cm−1, and some weak absorption peaks: ν(C=C), 1594 cm−1, ν(C=N),1451 cm−1, ν(C–H), 2960 cm−1, ν(C=O), 1747 cm−1 are also observed in the IR spectra at 524.37 K. In the third stage, there is a mass loss of 49.48% (calcd.: 52.50%) in the temperature of 629.15–1064.15 K, which belongs to the decomposition of the 3-MOBA− ligands with the products of CO2 (2403–2268 cm−1, 668 cm−1) and H2O (3709–3535 cm−1, 1783–1342 cm−1), as shown in IR spectra at 741.53 K. Compound 1 is completely degraded into Tb4O7 with the mass loss of 78.40% (calcd.: 81.60%).

Luminescence

Solid-state luminescent spectra (at room temperature) of 1–2 are shown in Fig. 7a, b. Figure 8 shows Commission Internationale de L’Eclairage (CIE) chromaticity diagrams for two compounds.

Fig. 7
figure 7

a Emission spectra of 1 (λex = 342 nm); b emission spectra of 2 (λex = 345 nm)

Full size image
Fig. 8
figure 8

CIE chromaticity diagrams of the two compounds (1 = compound 1; 2 = compound 2)

Full size image

As expected, compound 1 exhibits green emission with the typical bands of Tb(III) ion attributed to the electronic transitions 5D4 → 7FJ (J = 6, 5, 4 and 3). The detailed CIE chromaticity coordinate values are (0.282, 0.590). The four characteristic emission bands of 1 are at 489, 543, 584 and 623 nm, respectively. The ligand-based emissions are not observed, indicating that the effective sensitization is from the ligands to the central Tb(III) ion under photoluminescence [22, 23]. The hypersensitive transition (5D4 → 7F5) at 543 nm is the most intense transition, which dominates the characteristic green fluorescence of terbium complex. Furthermore, the 5D4 → 7F5 transition is the strongest among the four bands, and the fluorescent quantum yield is 0.851 (λex = 342 nm). These indicate that the ligands are suitable for the sensitization of green fluorescence for Tb(III) ion at room temperature.

As shown in Fig. 8, the CIE chromaticity coordinate values of compound 2 are (0.643, 0.357) and “red” emission is emitted. Transitions from the excited 5D0 state to the different J (0–4) levels of the lower 7FJ state are observed in the range of 500–700 nm. The emission bands at 579 and 652 nm are very weak since their corresponding transitions 5D0 → 7F0 and 5D0 → 7F3 are forbidden in both magnetic dipole and electric dipole fields. Two lines splitting for the 5D0 → 7F1 transition at 594 nm suggest the axial symmetry exists [24, 25]. The hypersensitive transition 5D0 → 7F2 at 619 nm is the most intense transition, which dominates the whole emission spectra, leading to the characteristic red fluorescence of Eu(III) complex. The emission of Eu(III) ion often used as a sensitive probe to investigate the local environment around the ions [26]. The energy transfer from the ligands to the Eu(III) ion by intersystem crossing is efficient, which is probably attributed to the matching of energy levels between excited states of ligands and excited states of Eu(III) ion. Further, the fluorescent quantum yield is 0.441. All above prove that 3-MOBA− and terpy ligands are good chromophore to absorb energy and transfer to Eu(III) ion, and emit the characteristic fluorescence of Eu(III) ion.

Two compounds both exhibit excellent fluorescence. The quantum yield of the two complexes shows that the fluorescence property of the terbium compound is stronger than that of the europium compound. Hence, the title compounds have potential application value as luminescent material.

Antibacterial activities

The antibacterial activities of ligands and compounds for C. albicans (fungus), Escherichia coli and Staphylococcus aureus (bacteria) were determined using the filter paper disk diffusion method at 303.15 K. The ligands and compounds were, respectively, dissolved in DMSO solutions at different concentrations of 8 × 10−3, 1.6 × 10−2 and 3.2 × 10−2 mol L−1. The experimental results show that 3-MOHBA ligand has no antibacterial action on C. albicans, E. coli and S. aureus, while the two compounds and terpy ligand exerted good antibacterial activity on E. coli and S. aureus but no antibacterial action on C. albicans, which may be attributed to the selective inhibitions of the compounds and ligands on antibacterial species. As shown in Table 5, the antibacterial actions of compounds and terpy ligand are more and more remarkable with the increase in the concentration in the range of tested concentrations.

Table 5 Antibacterial activities of ligands and compounds with three different concentrations at 303.15 K
Full size table

In short, Ln(III) ions are coordinated with the donor atoms of the ligands, which make them have certain antibacterial activities on microorganism. The π-electron delocalization over the chelate ring further reduces the polarity of the central metal cation and increases the lipophilicity [27]. The bacteriostatic mechanism of the compounds is presumably that they have a good lipophilic nature arising from chelation.

Conclusions

The synthesis and characterization of two new compounds have been successfully reported. IR spectra and X-ray powder diffraction indicate that compounds 1–2 are isostructural. The crystal structure of 1 has two asymmetric dinuclear molecules in unit cell, and each Tb(III) ion is nine coordinated. The asymmetric units are further connected to form 1D chain along the x-axis direction through the offset face-to-face π···π weak stacking interactions. The thermal decomposition mechanisms of the compounds are obtained by TG/DSC-FTIR techniques. The FTIR spectra of the evolved gases show that the carboxylic ligands are completely decomposed, which is consistent with the TG analysis. What’s more, the terbium and europium compounds exhibit excellent fluorescence. The antibacterial performance study shows that two compounds have good antibacterial actions on E. coli and S. aureus.

Supporting information

The number of compound 1 CCDC 1557510 contains the supplementary crystallographic data for this paper, which can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.