Nucleotide-based green synthesis of lanthanide coordination polymers for tunable white-light emission

2.1 Materials and reagents

The following chemical reagents were purchased from J&K Scientific Co. Ltd (Beijing, China): TbCl₃·6H₂O (99.9%), EuCl₃·6H₂O (99.9%), GdCl₃·6H₂O (99.9%), adenosine-5-monophosphate disodium salt (AMP, 98%), tyrosine, and ciprofloxacin. Ultrapure water (18 MΩ) was used for the preparation of all aqueous solutions. Unless otherwise stated, all chemicals used were of analytical reagent grade and were used as received without further purification.

2.2 Synthesis of AMP/Tb_0.1Eu_0.9Gd_99.0-CIP(4)

The aqueous solutions of 2 mL of LnCl₃·6H₂O (Tb³⁺ = 0.01 mM, Eu²⁺ = 0.09 mM, Gd³⁺ = 9.9 mM), 2 mL of AMP (10 mM), and 2 mL of CIP (2.5 mM) were mixed together and incubated for 20 min. The white precipitate was formed immediately. After stirring for 4 h at room temperature, the white precipitate was collected by centrifugation at 13,000 rpm for 5 min, and then washed with ultrapure water for several times to remove unreacted reactants. Finally, the obtained precipitate of AMP/Tb_0.1Eu_0.9Gd_99.0-CIP(4) was dried in an oven at 60°C. For comparison, the single Tb/AMP-CIP(1), Eu/AMP-CIP(2), and Gd/AMP-CIP(3) were also prepared by a similar procedure using EuCl₃·6H₂O (10 mM), TbCl₃·6H₂O (10 mM), and GdCl₃·6H₂O (10 mM) as the precursors, respectively.

2.3 Apparatus and characterization

The morphology of LMOCP was examined by scanning electron microscope (SEM) (JSM-6490LV, Japan). The elemental analysis was performed on an energy-dispersive X-ray spectrometer (EDX, X-Max Oxford, UK) and inductively coupled plasma atomic emission spectrometer (ICP-AES, USA). A D8 advance diffractometer (Bruker, Germany) was used for the collection of diffraction data which were useful for phase determination. The fluorescence spectra were recorded on an Agilent CaryEclipse fluorescence spectrophotometer (USA) with a xenon lamp as an excitation source. The fluorescence lifetime was measured by F-7000 FL spectrophotometer. The detection solution was placed in a microscale quartz cuvette. UV-visible absorption spectra were recorded on an Agilent Cary 60 UV-visible spectrophotometer (USA). Nicolet FTIR IS 10 spectrometer (USA) was employed to record the Fourier transform infrared (FTIR) spectra. The Commission International de L’Eclairage (CIE) color coordinates were calculated based on the international CIE standards.

2.4 Quantum yield of AMP/Tb_0.1Eu_0.9Gd_99.0-CIP

The relative quantum yields of AMP/Tb_0.1Eu_0.9Gd_99.0 were determined using tyrosine as a standard. The values of quantum yield were calculated by the following equation:

where Φ is the quantum yield, F is the fluorescence integral intensity, A is the absorbance intensity, and n is the refractive index of the solvent. The subscripts s and x are the reference fluorophore of known quantum yield and the sample, respectively.

3.1 Synthesis and characterization

AMP/Tb_0.1Eu_0.9Gd_99.0-CIP was prepared by the self-assembling of Ln³⁺ ions, AMP, and CIP in aqueous solution. SEM images showed that AMP/Tb_0.1Eu_0.9Gd_99.0-CIP was the same network-exterior as the single AMP/Ln-CIP {Ln = Tb(1) or Eu(2) or Gd(3)} (Figure 1). This result indicates that different kinds of lanthanides ions involved in the same framework have no influence on its appearance. No diffraction peak appeared in the images of X-ray diffraction (XRD) spectra (Figure 2), proving that these LMOCPs were all amorphous. The energy-dispersive X-ray (EDX) confirmed that Ln³⁺, AMP, and CIP were all involved in the formation of LMOCP (Figure S1). With same lanthanide ratios in LMOCP, we also compared the fluorescence of the mixtures of LMOCP 1, 2, and 3 with that of LMOCP 4. The results indicated that the mixtures of LMOCP 1, 2, and 3 do not emit pure white light (Figure S2). Therefore, to obtain white-emitting material, current synthesis strategy is perfect. Furthermore, the metal compositions analyzed by ICP-AES (mole ratio, Tb:Eu:Gd = 0.1:0.9:99.0) are matched with the initial compositions of the Ln³⁺ ions. However, the element analysis of C, H, N, O, and P is not accordant with the initial appending amount of AMP and CIP because(1) during washing and centrifugation, excess AMP and/or CIP was eliminated, which is unproportional with the original ratio and (2) the sensitivity of different elemental analysis instrument to each element is not completely identical (Table S1). The UV-Vis absorption spectra of AMP and CIP and AMP/Tb_0.1Eu_0.9Gd_99.0-CIP are also shown in Figure S3; it is observed that AMP displays a characteristic absorption peak at 260 nm and free CIP exhibits two maximum absorption peaks at 270 and 323 nm, respectively. In the absorption peaks of LMOCP 4, these peaks shifted to 268 and 320 nm, respectively. These changes confirmed that both AMP and CIP had coordinated with Ln³⁺. Furthermore, the peaks that correspond to the absorption spectrum of LMOCP 4 are stronger, indicating an efficient energy transfer and luminescence sensitization between CIP and Ln³⁺.

Figure 1

SEM images of (1) AMP/Tb-CIP, (2) AMP/Eu-CIP, (3) AMP/Gd-CIP, and (4) AMP/Tb_0.1Eu_0.9Gd_99.0-CIP.

Figure 2

XRD analysis of (1) AMP/Tb-CIP, (2) AMP/Eu-CIP, (3) AMP/Gd-CIP, and (4) AMP/Tb_0.1Eu_0.9Gd_99.0-CIP.

To confirm the chemical coordination between ligands and Ln³⁺, FTIR was conducted (Figure S4). The characteristic absorption peaks corresponding to C2–N1 stretching vibrations, NH₂ scissoring vibrations, and P–O stretching of AMP were observed at 1,576, 1,654, and 978 cm⁻¹, respectively. These peaks shifted to 1,566, 1,643, and 996 cm⁻¹ in the spectra of LMOCP 4, suggesting both nucleobase moieties and phosphate groups of AMP are involved in the coordination to Ln³⁺ [36]. Furthermore, when compared with pure CIP, the disappearing of the band at 1,708 cm⁻¹ (ν_COOH) and the shifting of the peak from 1,627 cm⁻¹ (ν_C═O) to 1,579 cm⁻¹ reflects the coordination between Ln³⁺ and CIP via carboxylate O and carbonyl groups [37,38,39].

3.2 Photoluminescence properties

The solutions and solid-state photoluminescent properties of LMOCP [1,2,3,4] were investigated (Figure 3). In the fluorescence spectrum of AMP/Tb-CIP(1), the peaks at 488, 545, 584, and 619 nm are the characteristic emissions of the Tb³⁺ ions from the ⁵D₄ → ⁷F_J (J = 3–6) transitions, respectively. For AMP/Eu-CIP, typical peaks at 545, 589, 616, and 696 nm were also observed [40,41], which were associated with the 4f–4 f transitions of ⁵D₀ exited state to the low lying ⁷F_J (J = 1–4) of Eu³⁺. LMOCP 1 and 2 display strong green and red emission at room temperature, respectively. Furthermore, the characteristic emission of CIP ligand at 412 nm disappeared in spectra of LMOCP 1 and 2, indicating the efficient energy transfer from CIP to Tb³⁺ and Eu³⁺ (Figure S5 and Figure 3a, b). However, AMP/Gd-CIP(3) does not display the typical emission of Gd³⁺ at 340 nm but a single broad blue emission peak at 412 nm originates from the π–π* transition of free CIP (Figure S4 and Figure 3c). We contribute this result to the energy mismatch between the absorption band of Gd³⁺ and the emission of CIP, leading the characteristic 4f–4f transition of Gd³⁺ at 340 nm is invisible.

Figure 3

Excitation and emission spectra of AMP/Tb-CIP(a), AMP/Eu-CIP (b), and AMP/Gd-CIP (c). Inset is the corresponding images of their solutions and solid powder under a common UV lamp.

To obtain a pure white light emission, the molar ratio of Tb³⁺, Eu³⁺, and Gd³⁺ ions doped into LMOCP 4 was optimized. According to three primary color theory, green, red, and blue light are all needed for obtaining white emission. However, in LMOCP of 1 and 2, the blue emission of CIP ligand is completely suppressed by the strong red and green luminescence of the Eu³⁺ and Tb³⁺ ions, an excess amount of nonluminescent Gd³⁺ ion is required to dilute these two Ln³⁺ ions in the solid state. With an increase in the concentration of Gd³⁺, the emission intensities of blue light components gradually raised, when the mole ratio of Tb:Eu:Gd was adjusted to 0.1:0.9:99.0 (Figure 4), a white emission with CIE chromaticity coordinates of (0.336, 0.327) which is close to those of international pure white light (0.333, 0.333) was obtained (Figure 5). In addition, the emission intensity of the white light is reaction time-dependent. We checked its fluorescence peak variation at 617 nm with the reaction time. As shown in Figure S6, when the reaction time is 4 h, the luminescence intensity at 617 nm of AMP/Tb_0.1Eu_0.9Gd_99.0-CIP reached the maximum. Furthermore, LMOCP 4 exhibits a high relative quantum yield of 36.5% that are comparable with that of other MOCPs of 6.5% [7], 15.3% [8], 8% [9], 7.93% [26], and 50% [10]. Additionally, the obtained LMOCP 4 also possesses long luminescence lifetime of 4.359 ms (Figure S7). These distinct photophysical properties make LMOCP 4 a kind of preeminent white light materials.

Figure 4

Emission spectra of the doped LMOCP excited at 273 nm. (a–h) (Tb, Eu, Gd%): (a) (1.01, 22.52, 76.47), (b) (2.34, 3.91, 93.75), (c) (1.015, 2.075, 96.91), (d) (1.515, 1.275, 97.21), (e) (0.015, 1.075, 98.91), (f) (0.1, 0.5, 99.4), (g) (0.15, 0.85, 99.0), and (h) (0.1, 0.9, 99.0).

Figure 5

Emission spectra of LMOCP4 with the excitation wavelength of 273 nm (4) (left) and its corresponding CIE chromaticity diagram (right). Inset is the photograph of the solution and a solid powder sample of LMOCP 4 under the UV lamp.

3.3 Thermogravimetry (TGA)

Finally, we investigated the thermal stability of the LMOCP 4 by TGA techniques. As can be seen from Figure S8, even exceeding to 350°C, the LMOCP 4 only lost 12% of its total weight. The result indicates that the obtained white materials LMOCP 4 based on three primary color theory possess high thermal stability.