Lanthanide Coordination Polymers as Luminescent Sensors for the Selective and Recyclable Detection of Acetone

Abstract

Three new isostructural lanthanide coordination polymers {[Ln(L)₂·2H₂O]·Cl·4H₂O}, Ln = La (LaL 1), Tb (TbL 2), Eu (EuL 3), L = 4-carboxy-1-(4-carboxybenzyl)pyridinium, have been synthesized under hydrothermal conditions and characterized by single crystal X-ray diffraction, IR, TG, PXRD, and luminescence. The solid-state luminescence properties of those Ln-CPs were investigated, realizing the zwitterionic ligand (L) is an excellent antenna chromophore for sensitizing both Tb³⁺ and Eu³⁺ ions. We utilized TbL 2 as a representative chemosensor to consider the potential luminescence sensing properties in different solvent suspension, which has the potential to serve as the first case of a luminescent Ln-CP material based on the zwitterionic type of organic ligand for selective and recyclable sensing of acetone in methanol solution.

1. Introduction

Over the last decade, lanthanide-based metal–organic coordination polymers (Ln-CPs), as novel luminescent functional materials, have attracted a good deal of attention. They have attracted this attention not only because of their diverse architectures, but also because of their potential luminescent sensor applications for detecting cations [1,2], anions [3,4], small molecules [5], gases and vapors [6], biomolecules [7], temperature [8,9], and so on [10,11]. Compared with other luminescent sensors, Ln-CPs based luminescent sensing materials have been a rapidly developing area due to their outstanding merits such as high surface area, easily designed crystal structures, stable frameworks, and permanent porosity, as well as exposed active sites [12,13,14]. Recently, the zwitterionic type of organic ligands have drawn our attention [15,16], which simultaneously bear positive and negative charges with a certain separated distance on the coordination skeleton endowing special physical properties, for example, improved adsorption selectivity of gases or vapors [17]. However, to date, the Ln-CPs materials reported are usually based on the common organic ligands, while the study of Ln-CPs based on the zwitterionic types of organic ligands have rarely been reported [18]. Therefore, further systematic investigation on the Ln-CPs based on the zwitterionic types of organic ligands is absolutely necessary.

Acetone is a highly volatile organic solvent, which could stimulate cornea or metabolism disorder and have a toxic effect on the human body [19]. Considering the extensive application and potential harm of acetone, it is urgent to develop new luminescent sensors to detect it. To date, only a few CPs have been reported for the sensing of acetone [20,21,22,23], most of which were constructed by d¹⁰ transition metal ions (Zn and Cd) and π-conjugated ligands, whereas coordination polymer materials based on lanthanide ions have been rarely published [24]. The advantages of those lanthanide materials over the transition metal luminescent materials include the sharp and strong emissions, large stokes shifts, high quantum yields, and long luminescence lifetimes, which originate from f–f transitions of lanthanide ions [25]. Taking into account these kinds of sensor materials are still defective, the synthesis and development of new materials based on lanthanide CPs for fluorescent high selective detection of acetones is still a major challenge.

As an effective combination of the above mentioned aspects, we selected the zwitterionic types of organic ligand 4-carboxy-1-(4-carboxybenzyl)pyridinium chloride (H₂LCl) as ligand to construct three isostructrual lanthanide coordination polymers {[Ln(L)₂·2H₂O]·Cl·4H₂O}, Ln = La (LaL 1), Tb (TbL 2), Eu (EuL 3). Those materials have the potential to serve as the first example of a luminescent Ln-CP material based on the zwitterionic type of organic ligand for selective and recyclable sensing of acetone in methanol solution.

2. Results

2.1. Crystal Structures

2.1.1. Descriptions of the Crystal Structures of {[La(L)₂·2H₂O]·Cl·4H₂O} (1)

The X-ray single-crystal diffraction studies reveal that compound LaL 1 crystallizes in the Pī space group of the triclinic system. As shown in Figure 1a, each La³⁺ center is eight-coordinated by six oxygen atoms (O1, O5, O8A, O4C, O2B, and O3D) from different L ligands and two oxygen atoms (O9 and O10) from two coordinated water molecules. The asymmetric unit of LaL 1 contains one crystallographically independent La³⁺, two deprotonated L ligands, two coordinated water molecules, one free Cl⁻ ion, and four guest water molecules. There are two kinds of coordination modes for the two carboxylate groups of L ligand; one adopts bidentate bridging coordination fashion μ₂-η¹η¹ and the other one shows monodentate bridging coordination mode μ₁-η¹. The two adjacent La³⁺ ions are connected together through four carboxylate groups of four L ligands in bidentate bridging coordination mode to form a dinuclear unit (Figure 1), then neighboring dinuclear subunits are bridged by these four L ligands in opposite direction to form 1D beaded chain. When adjacent, these chains are further connected by other 1D beaded chains, building by L ligands in unidentate bridging coordination mode, to form a 2D network (Figure 1). The angle between these two chains is 78.71°. The 3D supramolecular architecture of compound LaL 1 is obtained from the interlayer π···π stacking interactions between neighboring phenyl rings and benzene rings in adjacent layers (centroid−centroid: 3.534(2) Å, 3.888(3) Å and 3.623(2) Å). (Figure 1)

2.1.2. PXRD Analysis

To determine whether the crystal structures are truly representative of the bulk materials tested in property studies, powder X-ray diffraction (PXRD) experiments were carried out for compounds 1–3. The PXRD experimental and as-simulated patterns of compounds 1–3 are shown in the Figure 2. Failing to obtain crystals suitable for single-crystal crystallography, we were unable to determine the structures of TbL 2 and EuL 3, but the PXRD patterns of TbL 2 and EuL 3 are in good agreement with that of LaL 1, with only minor shifts in peak positions, indicating that TbL 2 and EuL 3 are isomorphous with LaL 1, and indicating that the crystal structures are truly representative of the bulk crystal products. The differences in intensity may be owing to the preferred orientation of the crystal samples.

2.2. Solid-State Luminescence of 1–3

The solid-state luminescence properties and the excitation spectrum of the H₂LCl ligand and compounds 1–3 were recorded at room temperature as shown in Figure 3 and Figure S1, respectively. As shown in Figure 3, complex LaL 1 presents a broad emission band centered at 409 nm (λ_ex = 332 nm), which is assigned to the typical π*–π transition of ligands (Figure 3). The blue shift compared to the ligand emission is presumably due to a conformational change in the ligand upon binding with the metal ions [20,21,22,23]. Upon excitation at 332 nm, complex TbL 2 showed four characteristic emission bands of Tb³⁺ ion, centered at 489, 544, 582, and 619 nm (Figure 3), and these emissions are attributed to the f–f electronic transitions ⁵D₄→⁷F_J (J = 6, 5, 4 and 3), respectively. Among these transitions, ⁵D₄→⁷F₅ (544 nm) green emission is a dominating intensity. While being excited at 332 nm, complex EuL 3 displays emissions at 590, 613, and 653 nm, which are attributed to the ⁵D₀→⁷F_J (J = 1, 2, and 3) transitions of the Eu³⁺ ion, with a maximum at ⁵D₀→⁷F₂ transition (613 nm) (Figure 3). Notably, there was no emission band of the ligand observed in the emission spectra of the compounds, indicating efficient energy transfer from the ligand to the Tb³⁺ ion or Eu³⁺ ion. These results clearly indicate that the L ligand is an excellent antenna chromophore for sensitizing both Tb³⁺ and Eu³⁺ ions.

2.3. Organic Small Molecule Sensing

The high luminescence intensities of those CPs prompt us to utilize TbL 2 as a representative to consider the potential luminescence sensing properties in different solvent suspension. Firstly, the finely ground sample of TbL 2 (10 mg) is immersed in different common organic solvents (5 mL), treated by ultrasonication for 60 min, and then aged for 24 h to generate stable suspensions before the fluorescence study. The solvents used are methanol, trichloromethane, tetrahydrofuran, dichloromethane, cyclohexane, n-hexane, acetonitrile (CH₃CN), N,N-dimethylacetamide (DMA), N,N-dimethylformamide (DMF), and acetone. Interestingly, the most interesting feature is that its luminescent emission spectrum, monitored at 613 nm (⁵D₄→⁷F₃), is largely dependent on the solvent molecules, particularly in the case of acetone, which exhibit the most significant quenching effects (Figure 4). Such solvent-dependent luminescence properties are of interest for the sensing of acetone solvent molecules. To explain the reason for the quenching effect, the absorption spectra of methanol and acetone are investigated, which reveals that acetone has a wide absorption range from 308 to 360 nm, while methanol exhibits no absorption (Figure S2). The absorbing band of acetone overlays part of the absorption band of sensor TbL 2, which may lead to energy transfer occuring between the sensor TbL 2 and the acetone molecules. Due to the intermolecular solute-solvent interactions between TbL 2 and acetone, the energy absorbed by TbL 2 is transferred to acetone molecules, resulting in a decrease in the luminescent intensity, which is similar to the previously reported CP-based sensors [22,23].

For better understanding of the sensing ability of TbL 2 for acetone, the sensing properties of TbL 2 for acetone were also investigated by recording the emissive spectra of the 2-methanol suspension with the gradual addition of acetone. As shown in Figure 4, the luminescence intensity monitored at 613 nm decreases clearly with the increasing acetone concentration. Moreover, the quenching effect can be rationalized for the low concentrations by the Stern–Volmer equation: I_o/I = 1 + K_sv × [M], where I_o and I are the luminescence intensities of ethanol suspensions of complex TbL 2 before and after the addition of acetone, respectively; K_sv is the quenching constant, and [M] is the molar concentration of acetone [20,21,22,23,24]. On the basis of the quenching experimental data, the linear correlation coefficient (R²) in the K_sv curve of TbL 2 with addition of acetone is 0.9877 (Figure 4), suggesting that the quenching effect of TNP on the fluorescence of TbL 2 fits well the Stern–Volmer model. The K_sv is 2.0263 × 10⁴ M⁻¹ at low concentrations of acetone, which is comparable to those of previously reported CP-based sensors [20,21,22,23,24].

Additionally, we also found that TbL 2 can be regenerated and reused for five numbers of cycles by centrifuging the dispersed crystals in methanol after sensing and washing several times with methanol (Figure 5). The quenching efficiencies of every cycle are basically unchanged through monitoring the emission spectra of TbL 2 dispersed in the presence of 200 μM acetone in methanol. Furthermore, the PXRD patterns of the initial sample and recovered by centrifuging of the dispersed crystals in methanol after sensing and washing several times with methanol indicate the high stability of this compound (Figure S3). The result reveals that TbL 2 could be applied as a fluorescence sensor for acetone with high selectivity and recyclability.

3. Experimental Section

CCDC 1498357 (1) contains the supplementary crystallographic data for this paper. This data can be obtained free of charge from the Cambridge Crystallographic Data Center via the internet at http://www.ccdc.cam.ac.uk/data_request/cif, or by e-mailing [email protected].

3.1. Material and Methods

All of the reagents and solvents used in this paper were purchased from Adamas-beta Co., Ltd. (Shanghai, China) and used as received without further purification, unless otherwise indicated. The NMR spectra were recorded on a Bruker DRX400 (¹H: 400 MH_Z, ¹³C: 100 MH_Z), chemical shifts (δ) are expressed in ppm, and J values are given in Hz, and deuterated DMSO was used as solvent. IR spectra were recorded on a FT-IR Thermo Nicolet Avatar 360 using KBr pellet. The phase purity of the samples was investigated by powder X-ray diffraction (PXRD) measurements carried out on a Bruker D8-Advance diffractometer equipped with CuKα radiation (λ = 1.5406 Å) at a scan speed of 1°/min. Thermal stability studies were carried out on a NETSCHZ STA-449C thermoanalyzer with a heating rate of 10 °C/min under a nitrogen atmosphere. All fluorescence measurements were performed on an Edinburgh Instrument F920 spectrometer. C, H, and N were determined on an Elementar Vario III EL elemental analyzer.

3.2. Synthetic Procedures

3.2.1. Synthesis of 4-Carboxy-1-(4-Carboxybenzyl)Pyridinium Chloride (H₂LCl)

A mixture of isonicotinic acid (1.231 g, 10 mmol) and methyl 4-(bromomethyl)benzoate (2.291 g, 10 mmol) in CH₃CN (50 mL) was refluxed for 8 h. After the mixture was cooled down to room temperature, the resulting precipitate was filtered to provide a white solid which was dissolved and refluxed in 100 mL 2 N HCl aqueous solution for 2 h. After the solution cooled down to ambient temperature, the white precipitate formed were collected by filtration and washed with ether (20 mL × 3) to afford H₂LCl (2.712 g, 92%) which was identical to data in the literature [15,26]. ¹H NMR (400 Hz, DMSO-d6): δ = 9.46 (d, 2H, ArH, J = 6.4 Hz), 8.52 (d, 2H, ArH, J= 6.4 Hz), 7.99 (d, 2H, ArH, J = 8.4 Hz), 7.67 (d, 2H, ArH, J = 8.4 Hz), 6.12 (s, 2H, CH₂); ¹³C NMR (100 Hz, DMSO-d6): δ = 167.2, 163.8, 146.9, 146.5, 139.0, 132.1, 130.45, 129.5, 128.2, 63.2. (Figure S4).

3.2.2. Synthesis of {[La(L)₂·2H₂O]·Cl·4H₂O} (LaL 1)

LaCl₃·6H₂O (0.3 mmol), H₂LCl (0.1 mmol), NaOH (0.2 mmol) and H₂O (10 mL) were added to a 15 mL Telfon-lined stainless steel autoclave, and the solution was heated at 150 °C for 24 h, then cooled down to room temperature at a rate of 2 °C/h. Colorless block crystals of the product were collected by filtration and washed with water several times, then dried in air. The obtained yield based on the H₂LCl was 22.5%. Main IR (KBr, cm⁻¹): 3382, 3113, 3051, 1621, 1540, 1397, 1172, 1131 (Figure S5). The TG curve for LaL 1 shows a gradual weight loss 13.8% between 90 and 280 °C, which can be ascribed to the removal of two lattice water molecules and four free water molecules (Figure S6). The phase purity of the product was confirmed by PXRD experiments (Figure 2). Anal. Calcd for C₂₈H₃₂ClLaN₂O₈ (698.9337): C, 48.12; H, 4.61; N, 4.01. Found: C, 48.05; H, 4.69; N, 3.98.

3.2.3. Synthesis of {[Tb(L)₂·2H₂O]·Cl·4H₂O} (TbL 2)

All attempts to get single crystals of TbL 2 by different methods were in vain. The procedure was similar to that of complex LaL 1 except that TbCl₃·6H₂O was used instead of LaCl₃·6H₂O. Yield: 19.2% (based on the ligand). The phase purity of the product was confirmed by PXRD experiments (Figure 2). Main IR (KBr, cm⁻¹): 3395, 3113, 3047, 1625, 1544, 1397, 1176, 1131. (Figure S5). Anal. Calcd for C₂₈H₃₂ClN₂O₈Tb (718.9437): C, 46.78; H, 4.49; N, 3.90. Found: C, 46.71; H, 4.53; N, 3.87.

3.2.4. Synthesis of {[Eu(L)₂·2H₂O]·Cl·4H₂O} (EuL 3)

All attempts to get single crystals of EuL 3 by different methods were in vain. The procedure was similar to that of complex LaL 1 except that EuCl₃·6H₂O was used instead of LaCl₃·6H₂O. Colorless precipitates were isolated in a yield of 13.5% (based on the ligand). The phase purity of the product was confirmed by PXRD experiments (Figure 5). Main IR (KBr, cm⁻¹): 3391, 3109, 3051, 1621, 1548, 1397, 1176, 1131. (Figure S5). Anal. Calcd for C₂₈H₃₂ClEuN₂O₈ (711.9837): C, 47.24; H, 4.53; N, 3.93. Found: C, 47.19; H, 4.58; N, 3.91.

3.3. Crystallography

Crystallographic data were collected at 296 K on a Bruker Smart AXS CCD diffractometer with graphite-monochromated Mo Kα radiation (λ = 0.71073 Å) using ω-scan technique. Cell parameters were retrieved using SMART software and refined with SAINT on all observed reflections. Absorption corrections were applied with the program SADABS. Structure LaL 1 was solved by direct methods using SHELXS-97 [27] and refined on F2 by full-matrix least-squares procedures with SHELXL-97 [28]. All nonhydrogen atoms were located in different Fourier syntheses and finally refined with anisotropic displacement parameters. Hydrogen atoms attached to the organic moieties were either located from the difference Fourier map or fixed stereochemically. The final chemical formula was estimated and combined with the TGA results. Details of the crystallographic data collection and refinement parameters are summarized in

Table 1. Crystal data and structure refinement for 1.

Compound	1
Chemical formula	C28H32N2O14ClLa
Formula mass	794.92
Crystal system	Triclinic
a/Å	11.3001(7)
b/Å	11.7346(7)
c/Å	14.6280(9)
α/°	100.2340(10)
β/°	93.7830(10)
γ/°	109.1450(10)
Unit cell volume/Å3	1787.09(19)
Space group	Pī
Z	2
DX/Mg m−3	1.343
μ/mm−1	1.318
Reflections with I > 2σ(I)	7215
Independent reflections	8114
F(000)	720
Rint	0.0245
GOF on F2	1.094
R1, wR2 [I > 2σ(I)]	0.0327, 0.0955
R1, wR2 [all data]	0.0375, 0.1000
Residuals/e Å−3	0.793, −0.638
CCDC number	1498357

. Main bond lengths and angles are presented in

Table 2. Elected bond lengths (Å) and angles (°) for compound 1.

Bond lengths (Å) for 1
La(1)-O(5)	2.448(3)	O(2)-C(6)	1.254(4)
La(1)-O(8)A	2.490(3)	O(3)-C(14)	1.246(4)
La(1)-O(1)	2.504(3)	O(4)-C(14)	1.260(4)
La(1)-O(2)B	2.508(3)	O(5)-C(20)	1.237(5)
La(1)-O(4)C	2.512(2)	O(6)-C(20)	1.252(6)
La(1)-O(3)D	2.562(2)	O(7)-C(28)	1.247(5)
La(1)-O(10)	2.569(3)	O(8)-C(28)	1.269(5)
La(1)-O(9)	2.605(3)	N(1)-C(4)	1.356(5)
La(1)-O(4)D	2.959(2)	N(1)-C(3)	1.380(5)
O(1)-C(6)	1.244(4)	N(1)-C(7)	1.502(4)
Bond angles (°) for 1
O(5)-La(1)-O(8)A	91.37(11)	O(3)D-La(1)-O(10)	68.16(9)
O(5)-La(1)-O(1)	145.38(9)	O(5)-La(1)-O(9)	71.04(10)
O(8)A-La(1)-O(1)	79.47(10)	O(8)A-La(1)-O(9)	69.88(10)
O(5)-La(1)-O(2)B	79.90(9)	O(1)-La(1)-O(9)	133.44(9)
O(8)A-La(1)-O(2)B	139.19(10)	O(2)B-La(1)-O(9)	69.58(10)
O(1)-La(1)-O(2)B	127.76(9)	O(4)C-La(1)-O(9)	75.85(9)
O(5)-La(1)-O(4)C	144.80(9)	O(3)D-La(1)-O(9)	138.14(10)
O(8)A-La(1)-O(4)C	88.20(9)	O(10)-La(1)-O(9)	127.61(10)
O(1)-La(1)-O(4)C	68.85(8)	O(5)-La(1)-O(4)D	119.81(9)
O(2)B-La(1)-O(4)C	77.69(9)	O(8)A-La(1)-O(4)D	146.12(9)
O(5)-La(1)-O(3)D	79.70(9)	O(1)-La(1)-O(4)D	67.36(8)
O(8)A-La(1)-O(3)D	141.40(9)	O(2)B-La(1)-O(4)D	65.49(8)
O(1)-La(1)-O(3)D	86.90(9)	O(4)C-La(1)-O(4)D	73.97(8)
O(2)B-La(1)-O(3)D	76.46(9)	O(3)D-La(1)-O(4)D	46.25(7)
O(4)C-La(1)-O(3)D	120.19(8)	O(10)-La(1)-O(4)D	101.20(8)
O(5)-La(1)-O(10)	74.02(10)	O(9)-La(1)-O(4)D	129.85(9)
O(8)A-La(1)-O(10)	73.27(9)	C(6)-O(1)-La(1)	139.4(2)
O(1)-La(1)-O(10)	71.36(9)	C(6)-O(2)-La(1)B	136.0(2)
O(2)B-La(1)-O(10)	138.94(10)	C(4)-N(1)-C(3)	120.5(3)
O(4)C-La(1)-O(10)	138.48(9)	C(4)-N(1)-C(7)	119.4(3)
O(3)D-La(1)-O(10)	68.16(9)	C(3)-N(1)-C(7)	120.1(3)

.

4. Conclusions

In summary, we have designed three isostructural lanthanide coordination polymers (Ln-CPs) to explore chemosensors. Firstly, the solid-state luminescence properties of those Ln-CPs were investigated, realizing the zwitterionic ligand (L) is an excellent antenna chromophore for sensitizing both Tb³⁺ and Eu³⁺ ions. In addition, TbL 2 as a representative chemosensor, material could not only serve as a chemosensor for acetone but also can be regenerated and reused for at least five cycles. Owing to a few luminescent CP, sensors for detecting acetone have been realized and reported [20,21,22,23], to the best of our knowledge, those materials have the potential to serve as the first case of a luminescent Ln-CP material based on the zwitterionic type of organic ligand for selective and recyclable sensing of acetone in methanol solution. Further studies are under way in our lab.