Introduction

The study of lanthanide complexes combining with aromatic carboxylic acids has attracted extensive attention, according to their diverse coordination types and fascinating properties [1–4]. At 2014, Sun et al. [5] reported that the rare earth complexes show potential antifungal properties, exerting various and selective inhibitions against the plant pathogenic fungi. The lanthanide complexes also had good bacteriostatic activity against Candida albicans, which has been reported by Zong at 2015 [6]. In addition, a wide variety of applications including gas separation and storage [7], nonlinear optics, electroluminescent materials [8], luminescent bioprobes [9], and catalysis [10] have been utilized as functional materials. However, the direct excitation of lanthanide metal centers is limited by small absorption cross sections, low molar absorption coefficients, and the Laporte forbidden f–f transitions [9, 11]. So the suitable ligands are used to stimulate luminescence of lanthanide ions via the effective energy transformation, which is called “antenna effect” [12], but the triplet states energies of organic chromophores should be high or close to lanthanide ions [13].

Aromatic carboxylic acids are usually chosen as the organic ligands since they can blind to the lanthanide cations in various modes: monodentate, chelating, bridging, and bridging–cheating to form supramolecular structures [14]. Terpy has been shown to be an excellent antenna unit for lanthanide luminescence and also performed the good coordination ability as the second ligand [15].

Therefore, we choose 2,4-dichlorobenzoic acid (2,4-DClHBA) as the main ligands, and 2,2′:6′,2″-terpyridine (terpy) as auxiliary ligands to validly sensitize the lanthanide metal ions to assemble two novel hybrid compounds [Ln(2,4-DClBA)3(terpy)(H2O)]·H2O (Ln = Dy(1), Er(2); 2,4-DClBA = 2,4-dichlorobenzoate; terpy = 2,2′:6′,2″-terpyridine). The structures of the title complexes are all different from our previous reports [16–19], and they are all mononuclear molecules, which makes them to form the multidimensional structure. The thermal decomposition processes were studied by the TG/DSC-FTIR technology. The heat capacities of the complexes were studied, which is crucial for designing chemical processes as well as for the progress of thermodynamic theories. Luminescence spectrum for complex 1 and the average activation energy E values of the first decomposition stage for title complexes were also studied.

Experimental

Materials and methods

LnCl3·6H2O was acquired by the reaction of Ln2O3 (Ln = Dy, Er, Beijing Lanthanide Innovation Technology Co., Ltd, 99.9 %) and hydrochloric acid aqueous solution in the condition of water bath heating at a constant temperature of 353.15 K, followed by evaporation of the liquid. The other analytically pure chemicals were purchased and used without further purification.

Equipment and conditions of the experiment

Analyses for C, H, N in the corresponding complexes were carried out via Vario-EL II element analyzer, and the percentage of the lanthanide cations were obtained by EDTA titrimetric analysis. The data of single-crystal X-ray diffraction were collected on a Smart-1000 diffractometer with graphite–monochromatic Mo Kα (λ = 0.71073 Å) at the temperature of 298(2) K. The structures were solved using the SHELXS-97 program (direct methods) and refined by the full-matrix least squares on F 2 using the SHELXL-97 program. Powder X-ray diffraction identification was carried out by a Bruker D8 ADVANCE X-ray diffractometer in a scanning range of 5°–50° (2θ) with graphite–monochromatic Cu Kα (λ = 1.54178 Å) at 298 (2) K. TG/DSC-FTIR analyses were performed with a heating rate of 10 K min−1 using a NETZSCH STA 449 F3 instrument with a Bruker TENSOR27 Fourier transform infrared spectrometer. The luminescence spectra were measured on an F-4600 Hitachi Spectrophotometer. Heat capacities of the prepared complexes were carried out on a NETZSCH DSC 200 F3 (nitrogen atmosphere) in the temperature of 263.15–337.15 K using an indirect measurement method. Non-isothermal kinetics of the first decomposition stage for title complexes were determined with the NETZSCH STA 449 F3 thermal analyzer under simulated air atmospheres by the linear heating rate of 10, 12, 15, 20 K min−1.

Synthesis

Dissolve 2,4-DClHBA (0.6 mmol) and terpy (0.2 mmol) in ethanol (95 %) and adjust the solution at the pH of 5–7 with the prepared NaOH solution (1 mol L−1). Add the mixed ligands solution to LnCl3·6H2O (0.2 mmol) aqueous solution under stirring and deposit it for 12 h. The powders of target complexes were obtained by filtration, and the colorless, pink platy single crystals were acquired through the method of solvent extraction at room temperature. Element analysis: calcd (%) for C36H24Cl6N3O8Dy: C, 43.16; H, 2.41; N, 4.19; Dy, 16.22. Found: C, 42.84; H, 2.43; N, 4.10; Dy, 16.15. calcd (%) for C36H24Cl6N3O8Er: C, 42.96; H, 2.40; N, 4.17; Er, 16.62. Found: C, 42.53; H, 2.36; N, 4.04; Er, 16.48.

Results and discussion

Crystal structure

Single-crystal X-ray diffraction reveals that complexes 1–2 are isomorphous, and crystallize in the triclinic space group Pī (Table 1). As a consequence, the crystal structure of complex 1 will be described here in detail.

Table 1 Crystallographic data for complexes 1–2
Full size table

Complex 1 is mononuclear, which is different from the most synthesized binuclear complexes [20, 21]. The mononuclear unit contains one Dy3+ cation, three 2,4-DClBA acid ligands, one tridentate terpy ligand, a coordinated water molecule, and an uncoordinated water molecule (Fig. 1a). Each Dy3+ cation is coordinated with nine atoms. There are five oxygen atoms from 2,4-DClBA acid ligands, which coordinate to Dy3+ by bidentate chelating, monodentate coordination modes. One oxygen atom comes from the coordinated water, and three nitrogen atoms supported by the tridentate terpy (Fig. 1b). The Dy–O bonds range from 2.290(5) to 2.632(8) Å and average Dy–N bond distances for the terpy molecule are 2.544 Å, which is obviously to found that the distance of Ln–N is reducing almost with the decrease in lanthanide radius, which is also according with the process of thermal decomposition (Table 2).

Fig. 1
figure 1

a Crystal structure of complex 1, all H atoms omitted for clarity. b Coordination geometry of Dy3+ cation

Full size image
Table 2 Selected bond lengths (Å) of the title complexes
Full size table

As shown in Fig. 2a, the mononuclear molecular structure of complex 1 is assembled into a supramolecular 1D chain by two independent and moderate Cl–π interactions [22] along the z-axis direction at the distance 3.368 and 3.560 Å. The chains are further packed into a 2D sheet in the xz plane (Fig. 2b) through two kinds of different hydrogen bonding interactions (Table 3). One is between carboxylic oxygen from monodentate 2,4-DClBA acid ligands and C–H of the adjacent terpy ligands unit at a distance of 2.820 Å along x-axis, and anther C–H···O hydrogen bonds [23] exists between uncoordinated water molecules and C–H of the neighboring terpy unit along the line of angle from z-axis and negative x-axis at the distance of 2.530 Å. There exist O–H···C hydrogen bonds between carbon from 2,4-DClBA ligands unit and O–H of the adjacent uncoordinated water unit at the distance of 2.890 Å to form 3D supramolecular network (Figure 2c).

Fig. 2
figure 2

a 1D chain structure along the z-axis. b 2D sheets connected by hydrogen bond interactions in the xz plane. c 3D structure interconnected by hydrogen bonds (uncoordinated water units have been omitted for clarity)

Full size image
Table 3 Selected hydrogen bond lengths (Å) and angles (°) for complex 1
Full size table

Previously, the binuclear molecular complex [Dy(2,4-DClBA)3phen]2 has been reported [20]. Each Dy3+ cation is surrounded by eight atoms, which is described as a distorted square antiprism. The structure of the mononuclear [Dy(2,4-DClBA)3(terpy)(H2O)]·H2O is totally different from the complex [Dy(2,4-DClBA)3phen]2, and the water molecule has coordinated with the Dy3+ at the molecular complex [Dy(2,4-DClBA)3(terpy)(H2O)]·H2O structure. As shown in Table 2, the average distance of the Dy–N for complex 1 is 2.544 Å and is shorter than the complex [Dy(2,4-DClBA)3phen]2 with the distance of 2.547 Å, indicating that the terpy has a stronger blinding affinity for the Ln3+ ions than the phen ligand.

Powder X-ray diffraction

The powder X-ray diffraction (PXRD) patterns [24, 25] of complexes 1–2 and two ligands have been carried out at room temperature. As shown in Fig. 3, the curves of the title complexes are different from the two ligands, indicating the two new complexes have been produced. The curves are almost similar with each other, indicating they are isostructural. As shown in Fig. 4, the diffraction peaks are almost in agreement with simulated data, showing that the structures of the powder complexes 1–2 are similar with the pure crystal.

Fig. 3
figure 3

a Experimental PXRD spectra of the complexes 1–2 and two ligands (a experimental complex 1; b experimental complex 2; c 2,4-DClHBA; d terpy). b Simulated and experimental PXRD spectra of the complexes 1–2

Full size image
Fig. 4
figure 4

TG/DTG/DSC curves (a) and stacked plots of the FTIR spectra of the evolved gases (b) for complex 1

Full size image

Thermal decomposition processes

The thermal decomposition processes of complexes 1–2 were determined from 298.15 to 1173.15 K at a heating rate of 10 K min−1 under a simulated air atmosphere. The data of thermal decomposition of the title complexes are listed in Table 4, and the IR spectra of the evolved gases at different temperature are shown in Figs. 4 and 5. Thermal decomposition process of complex 1 is different from complex 2, according to the TG curves. Complex 1 and 2 would be described here in detail.

Table 4 Thermal decomposition data of the title complexes
Full size table
Fig. 5
figure 5

TG/DTG/DSC curves (a) and stacked plots of the FTIR spectra of the evolved gases (b) for complex 2

Full size image

There are at least five thermal decomposition processes for complex 1 according to the DTG curve, which was shown in Fig. 4a, and it is consistent with the FTIR spectra of the evolved gases (Fig. 4b). The first stage between 373.15 and 402.15 K belongs to the release of one uncoordinated water molecule with a mass loss of 1.77 % (Calcd: 1.80 %), and at the same time, there is a small endothermic peak (T p = 389.45 K; ΔH = 84.15 kJ mol−1). As shown in Fig. 6a, the IR spectra at 393.15 K appear weak absorption peak at 3560–3875 and 1376–1853 cm−1, indicating the production of water at first stage. The second stage at temperature of 402.15–514.15 K accounts for the loss of one coordinated water and part terpy with the mass loss of 8.78 % and absorbs little quality of heat (T p = 494.45 K; ΔH = 19.15 kJ mol−1). In the IR spectra at 494.05 K, we can find the weak absorption peak of water at 3571–3879, 1372–1847 cm−1 and the peak of CO2 at 2367–2319 cm−1. In the third step, there is a mass loss of 16.29 % at the temperature of 514.15–684.15 K, belonging to the decomposition of the rest terpy, and one small endothermic peak (T p = 527.85 K; ΔH = 24.93 kJ mol−1) on the DSC curve can be observed. The thermal decomposition processes also generate CO2 (2365–2317 cm−1), H2O (3571–3879 cm−1) and appear the characteristic peaks of ν C=C, νC=N at 1597, 1489 cm−1, according to the IR spectra. In the fourth and fifth stages, the 2,4-DClBA occurs to decompose, which shows in the TG/DTG-DSC curve at the temperature of 684.15–793.15 and 793.15–1045.15 K with the mass loss of 39.85 and 14.58 %. Different from the former stages, the two stages appear one big exothermic peak at DSC curve (T p = 819.35 K; ΔH = −2830.17 kJ mol−1), and the complex 1 is completely degraded into Dy2O3 with a mass loss of 81.27 % (Calcd: 81.40 %). In Fig. 6a, there is the same absorption at temperature of 771.15 and 527.85 K, except that the absorption peak of CO2 is much stronger. But at 819.35 K, the peak of H2O disappears and only emerges the CO2 absorption at 2364–2316,679 cm−1.

Fig. 6
figure 6

FTIR spectra of the evolved gases of complex 1 (a) and complex 2 (b) at different temperature

Full size image

For complex 2, the thermal decomposition processes divide into four steps, according to the DTG curve (Fig. 5a). One uncoordinated water decomposes between 374.15 and 401.15 K in the first step with a mass loss of 1.80 % (Calcd. 1.79 %) and appears a small endothermic peak at the DSC curve (T p = 386.45 K; ΔH = 78.99 kJ mol−1). There is no IR absorption in the first stage and can be explained that the signal of gas is too weak to be detected. In the second step, the terpy ligand was decomposed at 401.15–696.15 K with a mass loss of 25.24 % (Calcd 25.17 %), which absorbs the heat of 8.63 kJ mol−1. In the IR spectra (Fig. 6b) at 605.01 K, there are the characteristic bands of CO2 (2367–2315 cm−1) and ν C=C, ν C=N and γ =C–H (1596, 1487 and 795 cm−1), assigning to the decomposition product. The third and fourth step could be seen one big process, which is the decomposition of 2,4-DClBA with the mass loss of 54.87 % (Calcd 54.37 %). According to the DSC, there is a big exothermic peak, indicating the variation of the energy of the process. In the IR spectra, we can find the absorptions of CO2 (2368–2314, 675 cm−1), ν C=C, ν C=N and γ =C–H (1596, 1487 and 795 cm−1) at 772.34 K, and the stronger absorption of CO2 (2368–2314, 675 cm−1), the peak of CO (2183, 2114 cm−1) at 814.92 K. At last, the complex 2 is completely degraded into Ho2O3 with a mass loss of 81.91 % (Calcd 81.33 %).

Kinetics of the first decomposition stage

The non-isothermal multiple scan method has been widely used to obtain the reliable value of the average activation energy E, in which the kinetic model function is not involved. Therefore, the average activation energy E [26] of the first decomposition for complexes 1–2 was been calculated via the integral isoconversional nonlinear (NL-INT) method [27, 28]. As shown in Fig. 7, it is obviously to observe that the activation energy E of the title complexes varies with conversion rate (α), respectively, indicating that the first thermal decomposition stage with losing H2O for complexes 1–2 is complex actions [29].

Fig. 7
figure 7

Relationship between E and α for the first decomposition stage of complexes 1–2

Full size image

Heat capacities of the complexes

The molar heat capacities of the complexes 1–2 were measured via DSC from the temperature of 263.15–347.15 K at a heating rate of 10 K min−1 under the nitrogen atmosphere. The average molar heat capacities of title complexes are listed in Table S1 and plotted in Fig. 8. As shown in Fig. 8 the molar heat capacities of complexes gradually increase with the elevation of temperature. The heat capacities of the title complexes were fitted to the following polynomial equations by the method of least squares [19, 30, 31] at reduced temperature (x). The equations are given in following:

Fig. 8
figure 8

Relationship of the molar heat capacities varying with temperature of complexes 1–2

Full size image
$$\begin{aligned} & {\text{Complex}}\;1 \, \left[ {{\text{Dy}}\left( {2,4 - {\text{DClBA}}} \right)_{3} \left( {\text{terpy}} \right)\left( {{\text{H}}_{2} {\text{O}}} \right)} \right]\cdot{\text{H}}_{2} {\text{O}} \\ & T = 268.15\,{\text{K}} - \, 338.15\,{\text{K}} \\ & C_{{{\text{p}},{\text{m}}}} /{\text{J}}\,{\text{mol}}^{ - 1} \,{\text{K}}^{ - 1} = 862.39412 + 76.06434x + - 1.4464x^{2} \\ & \quad - 2.97542x^{3} + 8.84492x^{4} + 6.2375x^{5} - 6.80803x^{6} \\ & R^{2} = 0.99999\;\;{\text{ SD}} = 0.15707 \\ & {\text{Complex}}\;2 \, \left[ {{\text{Er}}\left( {2,4 - {\text{DClBA}}} \right)_{3} \left( {\text{terpy}} \right)\left( {{\text{H}}_{2} {\text{O}}} \right)} \right]\cdot{\text{H}}_{2} {\text{O}} \\ & T = 268.15\,{\text{K}} - \, 338.15\,{\text{K}} \\ & C_{{{\text{p}},{\text{m}}}} /{\text{J}}\,{\text{mol}}^{ - 1} \,{\text{K}}^{ - 1} = 845.74344 + 70.20721x + 0.61469x^{2} \\ & \quad + 0.49654x^{3} + 12.288x^{4} + 7.55106x^{5} - 7.3677x^{6} \\ & R^{2} = 0.99999\;\;{\text{SD}} = 0.12196 \\ \end{aligned}$$

The reduced temperature (x) is calculated from the formula x = [T − (T max + T min)/2]/[(T max − T min)/2] (T max = the highest temperature, T min = the lowest temperature). R 2 is the correlation coefficient, and SD is the standard deviation. The smoothed heat capacities, enthalpy, entropy, and Gibbs free energy relative to the reference temperature 298.15 K of the title complexes with an interval of 3 K which is shown in Table S2-S3 were calculated according to the fitted polynomial equations and the following thermodynamic equations:

$$\begin{aligned} H_{\text{T}} - H_{298.15} & = \int_{298.15}^{T} {C_{\text{P,m}} } {\text{d}}T \\ S_{\text{T}} - S_{298.15} & = \int_{298.15}^{T} {C_{\text{P,m}} } T^{ - 1} {\text{d}}T \\ G_{\text{T}} - G_{298.15} & = \int_{298.15}^{T} {C_{\text{P,m}} } {\text{d}}T - T\int_{298.15}^{T} {C_{\text{P,m}} } T^{ - 1} {\text{d}}T \\ \end{aligned}$$

Luminescent behavior

The solid-state luminescent behavior of complex 1 has been investigated in detail at room temperature. The emission spectrum of complex 1 was obtained with excitation wavelength at 279 nm, and the emission spectrum of Dy3+ cation in the range of 450–620 nm shows two different characteristic peaks (Fig. 9) at 483, 580 nm, respectively, which are associated with the 4f–4f transitions of 4F9/2 → 6H15/2, 4F9/2 → 4H13/2. The hypersensitive transition (4F9/2 → 4H13/2) at 580 nm dominates the main emission spectra, resulting in the characteristic yellow luminescence of arrowhead complex. This indicates that organic chromophore terpy with triplet states energies is in the appropriate for sensitizing the Dy3+ cation.

Fig. 9
figure 9

Excitation (a) and emission (b) spectra of complex 1

Full size image

Conclusions

In summary, the synthesis and characterization of the title complexes have been successfully reported. These complexes are isomorphous with nine coordinated atoms, and the crystal structures of them have performed the mononuclear, which is totally different from the previous complexes. Complexes 1–2 show the 1D, 2D, 3D supramolecular structures. According to the study of TG/DSC-FTIR, thermal decomposition mechanisms of the title complexes are obtained and the uncoordinated water all lost firstly. The average molar heat capacity value of the complexes gradually increased with the elevation of temperature, and their heat capacities fitted to a polynomial equation. The derived thermodynamic functions (H T − H298.15), (S T − S298.15), and (G T − G298.15) of the complexes relative to the standard reference temperature 298.15 K were also obtained. Complex 1 shows the characteristic emission of Dy3+ with strong yellow emission.

Supporting information

The numbers of the three complexes (CCDC 1459059 (1), 1459061 (2)) contain 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. Experimental molar heat capacities of complexes 1–2 have been performed in Table S1, and smoothed molar heat capacities and thermodynamic functions of complexes 1–2 were shown in Table S2-S3, respectively.