Molecular Construction and Photophysical Properties of Luminescent

14168

J. Phys. Chem. C 2008, 112, 14168–14178

Molecular Construction and Photophysical Properties of Luminescent Covalently Bonded Lanthanide Hybrid Materials Obtained by Grafting Organic Ligands Containing 1,2,4-Triazole on Silica by Mercapto Modification J. L. Liu and B. Yan* Department of Chemistry, Tongji UniVersity, Siping Road 1239, Shanghai 200092, People’s Republic of China ReceiVed: December 21, 2007; ReVised Manuscript ReceiVed: June 27, 2008

This study focuses on the syntheses of a series of organic-inorganic hybrid materials in which the triazole heterocyclic organic components were grafted into the silica backbone via covalent bonds through a sol-gel process. The organic parts 3-alkyl-4-amino-5-ylsulfanyl-1,2,4-triazole (O1 and O2) were first prepared by the reaction of thiocarbohydrazide with acetic acid or propionic acid, respectively, and then functionalized with trialkoxysilyl groups, and the as-obtained silylated monomers (P1-P4) were used as the siloxane network precursors to coordinate to Eu3+ or Tb3+ and further introduced into silica matrixes by Si-O bonds after hydrolysis and polycondensation processes. The preparation of the hybrid materials (LnM1-LnM4, Ln ) Eu and Tb) including covalent grafting is described, as well as their structures and the photophysical properties. All of the materials are totally amorphous and no phase separation happened. The spectroscopic date revealed that all these hybrids can show the characteristic luminescence of Eu3+ or Tb3+ ions for the energy transfer process takes place successfully between the organic parts and the RE ions. Modifications by different trialkoxysilyl groups (3-(triethoxysilyl)propyl isocyanate or 3-chloropropyl trimethoxysilane) lead to the different coordination structures and thus influence the absorption efficiency or the ability of the organic ligands to transfer the absorbed energy to Ln3+ ions and consequently changed the luminescence lifetimes and the quantum yield of the emission. Introduction Organic-inorganic hybrid materials have been attracting a great deal of attention because of their extraordinary properties as they combine the mutual advantages of both organic and inorganic networks.1,2 The very large range of choices for the organic and inorganic components offers the possibility of obtaining materials with attractive performances such as mechanical, thermal, and other physical and chemical properties.3 Besides, most of the hybrid materials can be prepared by the mild synthetic method named sol-gel approach, which is based on hydrolysis/polycondensation reactions of metal alkoxides. This method exhibits its unique characteristics such as convenience, low temperature, and versatility.4-8 Moreover, the microstructure, the external shape, or the degree of combination between the two phases can be further controlled by modifying the sol-gel processing conditions. Rare earth ions (noted as RE) have been well-known as important components in luminescent materials for their sharp and intense emission bands based on f-f electronic transitions and a wide range of lifetimes suitable for various applications, which are due to the fact that the emitting excited state and the ground state have the same fn electronic configuration and that the f orbitals are shielded from the environment by the outer s and p electrons.9 Organic chelates are well-known to be the efficient sensitizers for the luminescence of lanthanide ions (Eu3+ and Tb3+ in particular): complex formation between lanthanide ions and certain organic ligands has the double beneficial effect of both protecting metal ions from vibrational coupling and increasing light absorption cross-section by * To whom the correspondence should be addressed. E-mail: byan@ tongji.edu.cn. Fax: +86-21-65982287. Phone: +86-21-65984663.

“antenna effects”. In such cases, the organic ligand absorbs exciting radiation and transfers excitation energy to the lanthanide emitter. Entrapping of rare earth complexes with β-diketones, aromatic carboxylic acids, and heterocyclic ligands in sol-gel-derived host structures has been described in some recent studies.10-15 Typically, these materials were obtained by doping silica gels with organometallic complexes (class hybrid materials16). However, this conventional doping method seems unable to solve the problem of clustering of emitting centers because only weak interactions (such as hydrogen bonding, van der Waals forces, or weak static effects) exist between organic and inorganic moieties. In addition, inhomogeneous dispersion of two phases and leaching of the photoactive molecules frequently occur in this sort of hybrid materials for which the concentration of complexes is also prohibited. As a consequence, a few studies in terms of the covalently bonded hybrids (class hybrid materials16) have appeared and the as-derived molecularbased materials exhibit monophasic appearance even at a high concentration of rare earth complexes;17-22 besides, the reinforcement of thermal and mechanical resistances has been clearly established. The key procedure to construct molecular-based materials is to design a functional bridge molecule (ligand) by the grafting reaction, which can have the double function of coordinating to rare earth ions and allowing the sol-gel process to constitute the covalent Si-O network.23,24 Some previous research has concentrated on the modification of pyridine-dicarboxylic acid or their derivatives: Zhang et al. focus on the modification of heterocyclic ligands like 1,10-phenanthroline and bipyridyl,16,25 Carlos and co-workers have prepared hybrids that contain OCH2CH2 (polyethylene glycol, PEG) repeat units grafted onto a siliceous backbone by urea (-NHC(dO)NH-), or urethane

10.1021/jp712018n CCC: $40.75  2008 American Chemical Society Published on Web 08/19/2008

Luminescent Covalently Bonded Lanthanide Hybrid Materials

J. Phys. Chem. C, Vol. 112, No. 36, 2008 14169

Figure 1. Scheme of the synthesis process of thiocarbohydrazide S3, organic components O1 and O2, and the silylated precursors P1-P4 (A) and the predicted structure of the hybrids (B).

(-NHC(dO)O-) bridges.26,27 Our research team presently did extensive work and we have successfully realized four paths to construct rare earth hybrid systems with chemical bonds. The first is to modify the amino groups of aminobenzoic acids using 3-(triethoxysily)propyl isocyanate (abbreviated as TESPIC) through a hydrogen transfer addition reaction between the amino group of acids and the internal ester group (isocyanate) of TESPIC.28,29 The second path is to modify the carboxylate group of aromatic carboxylic acids through the amidation reaction between the carbonyl groups of acids and the amino group of amino-silane cross-linking reagents.30,31 The third is to modify the hydroxyl groups of hydroxyl compounds by using ester groups of TESPIC.32 The fourth is to modify the sulfonic groups of an acylamine derivation to graft with 3-aminopropyl trimethoxysilane.33 After the modification, we assemble the above modified bridge ligands with rare earth ions and inorganic precursors as tetraethoxysilane (TEOS) to construct hybrid systems. Moreover, according to the molecular fragment principle to assemble the ternary rare earth complexes, we also achieved the cooperative design of rare earth hybrid systems with crossing reagent as structural ligand and photoactive sensitizer as functional ligand.34 Triazole compounds have been used extensively as insecticides, anticancer medicines, and so on. 1,2,4-triazole and its

derivatives show strong coordination ability for many metal ions, and their applications are strengthened by being modified and coordinated with metal ions. In this paper, we use 3-alkyl-4amino-5-ylsulfanyl-1,2,4-triazole (O1 and O2) compounds as the organic parts to graft with the silica network. The organic compounds O1 and O2 possess the triazole heterocyclic organic functions which can absorb exciting radiation and transfer excitation energy to the lanthanide emitter to sensitize luminescence of lanthanide ions, which are called the “antenna effects”. Besides, the mercapto groups of the organic compounds O1 and O2 possess the reactive hydrogen atoms which can be expected to realize hydrogen transfer reaction with the silane cross-linking reagent 3-(triethoxysilyl)propyl isocyanate and the obtained silylated monomers can be used as siloxane network precursors to be introduced into silica matrixes by Si-O bonds after hydrolysis and polycondensation processes. The mercapto modification method is a new path to construct rare earth covalently bonded hybrid systems. Because the luminescence properties of the corresponding lanthanide ions have been largerly improved as the energy transfer process takes place successfully between the organic parts and the RE ions, these kinds of materials are expected to realize potential applications in nonlinear optics, light emitting devices, and functional membranes. Moreover, the strengthening of the thermal stability

14170 J. Phys. Chem. C, Vol. 112, No. 36, 2008

Liu and Yan

Figure 3. Ultraviolet absorption spectra of the organic components O1 and O2 and the precursors P1-P4.

spectrophotometer. 1H NMR and 13C NMR spectra were recorded on a Bruker AVANCE-500 spectrometer with tetramethylsilane (TMS) as internal reference. Ultraviolet absorption spectra of these samples (5 × 10-4 mol L-1 chloroform solution) were recorded with an Agilent 8453 spectrophotometer. Melting points were measured on a XT4-100XA apparatus and were uncorrected. Scanning electronic microscope (SEM) images were obtained with a Philips XL-30. The X-ray diffraction (XRD) measurements were carried out on powdered samples via a BRUKER D8 diffractometer (40 mA /40 kV), using monochromated Cu KR1 radiation (λ)1.54 Å) over the 2θ range of 5° to 70°. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) were performed on a NETZSCH STA 449C with a heating rate of 10 deg/min under a nitrogen atmosphere (flow rate: 40 mL/min). Reflectivity spectra were recorded on a Bws003 spectrometer equipped with a diffuse reflectance accessory. Fluorescence excitation and emission spectra were obtained on a Perkin-Elmer LS-55 spectrophotometer. Luminescent lifetimes were recorded on an Edinburgh FLS 920 phosphorimeter, using a 450 W xenon lamp as the excitation source (pulse width, 3 µs). Materials. Starting materials were purchased from China National Medicines Group. Tetraethoxysilane (TEOS) was distilled and stored under a N2 atmosphere. The solvents were purified according to literature procedures.35 Other starting reagents were used as received. Europium and terbium nitrates were obtained from their corresponding oxides in concentrated nitric acid. Synthesis. The organic compounds 3-alkyl-4-amino-5-ylsulfanyl-1,2,4-triazole (O1 and O2) and the silylated precursors (P1-P4) were prepared according to the procedure depicted in Figure 1A. The hybrid materials (LnM1-LnM4) were prepared by the hybridization of the precursors and TEOS through hydrolysis and polycondensation processes.

Figure 2. Infrared spectra of the precursors P1-P4 (A) and the hybrid materials M1-M4, LnM1-LnM4 (B) in the 4000-400 cm-1 range.

and mechanical stability of these hybrid materials should also arise considerable usages as tunable solid state lasers or chemical/biomedical sensors. Experimental Section Physical Measurements. FT-IR spectra (KBr) were measured within the 4000-400 cm-1 region on a Nicolet model 5SXC

TABLE 1: Assignments of the Main Infrared Absorption Bands (cm-1) for Thiocarbohydrazide S3, Organic Components O1 and O2, and Their Corresponding Silylated Precursors P1-P4 compds S3 O1 O2 P1 P2 P3 P4

ν(S-H)

ν(CH2)

ν(CONH)

2638 2641 2972, 2975, 2972, 2973,

2928 2928 2928 2929

1737 1735

ν(N-H) 3271, 3269, 3261, 3302, 3297, 3281, 3278,

3202 3177 3143 3192 3190 3149 3149

ν(CdN)

ν(Si-C)

1621 1617 1636 1635 1626 1619

1198 1196 1192 1192

δ(Si-O)

1105, 1103, 1105, 1105,

1087, 1080, 1087, 1093,

δ(C-S-C)

462 464 473 459

672 690 692 685



TABLE 2: ν(CdO) and ν(CdN) Stretching Vibration Frequencies (cm-1) for Free-Ln Materials M1-M4 and Coordinated Hybrids LnM1-LnM4 ν(CdO) ν(CdN)

M1

EuM1

TbM1

M2

EuM2

TbM2

M3

EuM3

TbM3

M4

EuM4

TbM4

1680 1632

1654 1621

1655 1619

1675 1630

1645 1615

1648 1623

1624

1610

1604

1620

1608

1609

TABLE 3: The Absorption Edges of the Reflectivity Spectra for Free-Ln Materials and Their Complexes M1 M2 M3 M4

free-Ln materials

Eu3+ complex

Tb3+ complex

228-285 228-286 228-281 228-284

229-297 228-294 229-297 228-297

229-295 228-296 229-297 228-296

Synthesis of Thiocarbohydrazide (S3). Hydrazine hydrate (S2; 40 mL, 85 %) was dissolved in 120 mL of water, and then 12 mL of CS2 (S1) was added dropwise, the reaction mixture was kept under room temperature for 1 h and then heated to 90 °C for an additional 10 h. After cooling, the precipitate was filtered off, washed with water, and dried. The crude product was purified by recrystallization form water and finally obtained as white needles, yield 12.04 g (72 %). mp 172-173 °C. Synthesis of 3-Alkyl-4-amino-5-ylsulfanyl-1,2,4-triazole (O1 and O2). S3 (3.18 g, 0.03 mol) was added to 10 mLof acetic acid and then the resulting mixture was heated under reflux for

Figure 4. The X-ray diffraction (XRD) graphs of the organic components O1 and O2 (A) and the hybrids TbM1, EuM2, EuM3, and TbM4 (B).

about 10 h. The solution was evaporated in vacuo and the remaining residue washed with water. The solid obtained was collected and recrystallized from water to give the crystals of O1, yield 2.60 g (68 %). Mp 199-201 °C. 1H NMR (DMSO 500 MHz) δ 2.25 (s, 3H, CH3), 5.56 (s, 2H, NH2), 13.34 (s, 1H, SH). 3-Ethylic-4-amino-5-ylsulfanyl-1,2,4-triazole (O2) was prepared by the same method except that acetic acid was replaced by propionic acid and given a yield of 2.59 g (60 %). Mp 153-155 °C. 1H NMR (DMSO 500 MHz) δ 1.05 (t, 3H, CH3), 2.46 (q, 2H, CH2), 5.50 (s, 2H, NH2), 13.44 (s, 1H, SH). Synthesis of the Precursor P1. To a warm solution of O1 (2 mmol) in 10 mL of pyridine was added 3-(triethoxysilyl)propyl isocyanate (2 mmol) dissolved in 10 mL of pyridine dropwise with stirring, then the mixture was warmed at 80 °C for approximately 12 h under argon in a covered flask. The solvent was removed in vacuum, and the residue was washed with 20 mL of hexane three times when aclear yellow oil was obtained in a 95 % yield. 1H NMR (CDC13, 500 MHz) δ 0.69 (t, 2H, CH2Si), 1.23 (t, 9H, CH3CH2), 1.68 (m, 2H, CH2CH2CH2), 2.28 (s, 3H, CH3C), 3.33 (m, 2H, NHCH2), 3.80 (q, 6H, SiOCH2), 5.02 (s, 2H, NH2), 7.53 (t, 1H, NH). 13C NMR (CDC13, 100 MHz) δ 7.8 (CH2Si), 12.9 (CH3C), 18.4 (CH3CH2O), 25.4 (CH2CH2CH2), 45.5 (NCH2CH2), 58.4 (CH3CH2O), 148 (SCdN), 159 (CH2CdN), 167 (CdO). The other precursor P2 was synthesized by the same manner from the reaction between 3-(triethoxysilyl)propyl isocyanate with O2 and was also obtained as a clear yellow oil in a 92 % yield. 1H NMR (CDC13, 500 MHz) δ 0.72 (t, 2H, CH2Si), 1.22 (t, 9H, CH3CH2O), 1.24 (t, 3H, CH3CH2C), 1.78 (m, 2H, CH2CH2CH2), 2.81 (q, 2H, CH3CH2C), 3.40 (t, 2H, NHCH2), 3.83 (q, 6H, SiOCH2), 4.89 (s, 2H, NH2), 7.45 (t, 1H, NH). 13C NMR (CDC13, 100 MHz) δ 7.7 (CH2Si), 14.3 (CH3CH2C), 18.4 (CH3CH2O), 22.7 (CH3CH2C), 25.5 (CH2CH2CH2), 45.5 (NCH2CH2), 58.3 (CH3CH2O), 148 (SCdN), 159 (CH2CdN), 167 (CdO). Synthesis of the Precursor P3. O1 (2 mmol) was dissolved in 10 mL of anhydrous acetone. A solution of 3-chloropropyl trimethoxysilane (2 mmol) in 10 mL of acetone was added dropwise, then the resulting mixture was refluxed overnight under argon in a covered flask followed by an addition of 3 drops of triethylamine. Filtration of the precipitated triethylamine hydrochloride followed by evaporation of acetone and excess triethylamine in vacuo led to a residue. The residue was washed with 20 mL of hexane three times when a clear yellow oil was obtained in a 95 % yield. 1H NMR (CDC13, 500 MHz) δ 0.58 (t, 2H, CH2Si), 1.68 (m, 2H, CH2CH2CH2), 2.30 (s, 3H, CH3C), 3.05 (t, 2H, SCH2), 3.58 (s, 9H, OCH3), 5.60 (s, 2H, NH2). 13C NMR (CDC13, 100 MHz) δ 8.7 (CH2Si), 12.3 (CH3C), 16.5 (CH2CH2CH2), 39.5 (SCH2CH2), 56.3 (CH3O), 148(SCdN), 159 (CH2CdN). The other precursor P4 was synthesized by the same manner and obtained as a clear oil in a 94 % yield. 1H NMR (CDC13, 500 MHz) δ 0.64 (t, 2H, CH2Si), 1.22 (t, 3H, CH3CH2), 1.78 (m, 2H, CH2CH2CH2), 2.54 (q, 2H, CH3CH2C), 2.95 (t, 2H, SCH2), 3.54 (s, 9H, OCH3), 5.62 (s, 2H, NH2). 13C NMR (CDC13, 100 MHz) δ 8.7(CH2Si), 14.3(CH3CH2C), 16.5 (CH2CH2CH2), 22.5 (CH3CH2C), 39.5 (SCH2CH2), 56.3 (CH3O), 148 (SCdN), 159 (CH2CdN).


Liu and Yan

Figure 5. The DSC and TGA traces of TbM2 hybrid material.

Synthesis of the Sol-Gel DeriWed Hybrid Materials Containing Rare Earth (LnM1-LnM4, Ln ) Eu or Tb). A typical procedure for the preparation of the hybrid materials was as follows. The precursor (P1-P4) was dissolved in DMF, and TEOS and H2O were added with stirring, then one drop of diluted hydrochloric acid was added to promote hydrolysis. A stoichiometric amount of Ln(NO3) · 6H2O was added to the final stirring mixture. The mole ratio of Ln(NO3) · 6H2O/Pn (n ) 1-4)/TEOS/H2O was 1:3:6:24. At the beginning of the reaction, the individual hydrolysis of the silylated precursors and TEOS are predominant. After hydrolysis, an appropriate amount of hexamethylene-tetramine was added to adjust the pH value to 6-7 and then the polycondensation reactions between hydroxyl groups of both the silylated precursors and TEOS take place. The resulting mixture was agitated magnetically to achieve a single phase, and thermal treatment was performed at 60 °C in a covered Teflon beaker for about 6 days until the sample solidified. The obtained materials were washed with ethanol and dried at 70 °C for 2 days. The final molecular hybrid materials were collected as monolithic bulks and were ground into powdered material for the photophysical studies. Free-Ln materials M1-M4 were also prepared by the same manner without the addition of Ln(NO3) · 6H2O. Results and Discussion Chemical Characterizations. As detailed in the Experimental Section, 1H and 13C spectra relative to the organic materials (O1 and O2) and the precursors (P1-P4) are in full agreement with the proposed structure. The 1H NMR chemical shifts relative to SH bonds observed in O1 and O2 have disappeared in the corresponding silylated precursors while the shifts attributed to NH2 are still reserved, which indicates the hydrogen transfer reaction takes place between the trialkoxysilyl groups with OH but not the NH2 groups. The peaks observed for carbonyl groups in 13C NMR spectra attributed to -CONHgroups are observed in P1 and P2, which can further demonstrate the grafting reaction between the organic compounds with TESPIC. Integration of the 1H NMR signals corresponding to methoxy and ethoxy groups shows that no hydrolysis of the precursors occurred during the grafting reaction.

FTIR experiments were performed on all the organic compounds and the silylated precursors. Figure 2A shows the FTIR spectra of P1-P4 and assignments of the main infrared absorption bands of S3, O1, O2, and P1-P4 are shown in Table 1. The apparent characteristic absorption bands of NH2 groups at 3302-3149 cm-1 can be observed in both the organic compounds O1 and O2 and the corresponding silylated precursors P1-P4, while the peaks at around 2640 cm-1 originating from the absorption of SH groups disappeared, which means the grafting reaction corresponds to the mercapto modification and not the amide groups. Besides, the presence of a series of strong bands at around 2975-2928 cm-1 due to the vibrations of methylene -(CH2)3- indicates that 3-(triethoxysilyl)propyl isocyanate or 3-chloropropyl trimethoxysilane has been successfully grafted onto organic compounds (O1 and O2). For the precursors P1 and P2, the absorption at about 1735 cm-1 corresponds to the formation of carbonyl group, meanwhile, no absorption band characteristic of the ν(NdCdO) vibration was detected at 2277 cm-1, which is a further proof of the completion of reaction. As for P1 and P2, the occurrence of the P3 and P4 synthesis reactions was evidenced by the increase of ν(C-S-C) at 685 cm-1, while, the vanishing of the ν(S-H) at 2638 cm-1and the ν(C-Cl) at 800 cm-1 can further provide the completion of the grafting reactions. Besides, the stretching vibration of Si-C located at 1192 cm-1 exists in all of the four precursors, and the stretching vibration of Si-O at 1107 and 1085 cm-1 together with the bending vibration at 460 cm-1 indicate the absorption of the siloxane bonds. The ultraviolet absorption spectra of the organic materials (O1 and O2) and the silylated precursors (P1-P4) were exhibited in Figure 3. From the spectra, it is observed that there was a red shift (255 nm f 265 nm) between the organic materials and the precursors P1 and P2, because the π f π* transition of the triazole groups was influenced by the conjugating effect of the CdO groups after the modification reactions. So far as P3 and P4 were concerned, there was nearly no shift (255 nm f 257 nm). This is due to the fact that there was no double bond in 3-chloropropyl trimethoxysilane so that the conjugating systems of the trizole groups were nearly unchanged after grafting to the 3-chloropropyl trimethoxysilane.



Figure 6. SEM images for molecular-based hybrid materials with organic and inorganic networks: (A) EuM1, (B) TbM1, (C) EuM2, (D) TbM2, (E) EuM3, (F) TbM3, (G) EuM4, and (H) TbM4.


Liu and Yan

Figure 7. The excitation and emission spectra of the europium hybrid materials (A) EuM1, (B) EuM2, (C) EuM3, and (D) EuM4.

The amorphous hybrid materials (LnM1-LnM4) and the corresponding free-Ln materials (M1-M4) were also characterized by infrared spectroscopy. The FTIR spectra of these hybrids are shown in Figure 2B. The formation of siloxane bonds can be indicated by the broad absorption located at 1120-1050 cm-1 (ν(Si-O-Si)). Besides, the ν(Si-C) vibration located in the 1192-1200 cm-1 wavelength range can be observed in all of the materials, which is consistent with the fact that no (Si-C) bond cleavage occurs during the hydrolysis/condensation reactions.22 However, the ν(O-H) vibration at 3740 and 3400 cm-1 can also be observed, which means the existence of residual silanol groups. And the ν(Si-OH) stretching vibration at 956 cm-1 is a further evidence of the incompleteness of condensation reactions. Complexation of Ln3+ can be clearly shown by infrared spectroscopy. Table 2 summarizes the ν(CdO) and ν(CdN) frequencies observed for free-Ln materials (M1-M4) and coordinated hybrid complexes (LnM1-LnM4). In LnM1 and LnM2, the ν(CdO) vibrations are shifted to lower frequencies (∆ν ) 20-40 cm-1) compared with the free-Ln materials M1 and M2, which can indicate the complexation of the RE ions with the oxygen atom.22 Exchanging the RE ions (replacing Eu3+ by Tb3+) and changing the organic ligand (changing M1 by M2) seems to have no influence on the CdO bond strength. The ν(CdN) vibrations originating from the triazole heterocyclic parts in all of the four coordinated hybrid complexes are also shifted to lower frequencies (∆ν ) 10-30 cm-1) compared with their corresponding free-Ln materials. This may be ascribe to

the complexation of the RE ions with the nitrogen atom of the triazole system thus decreasing the electron density on triazole groups. Diffuse reflectance experiments were performed on powdered materials for both free-Ln materials (M1-M4) and the Eu3+ or Tb3+ hybrid complexes (LnM1-LnM4). The absorption edges extracted from the spectra are shown in Table 3. Reflectivity spectra exhibit for all samples a broad absorption band in the near-UV range (200-300 nm). These absorption bands correspond to transition from the ground state of the organic ligands to the first excited state (S0 f S1). It is more specifically attributed to π f π* transition of the 1,2,4-triazole group. For free-Ln materials and the Eu3+ or Tb3+ hybrid complexes, a red shift is observed upon complexation that may be due to the complexation of RE ions with the N atom of the 1,2,4-triazole group, since the singlet level is sensitive to the ligand structure, and the influence of 1,2,4-triazole group is more sensitive. The complexation of RE ions enhanced the delocalization in the 1,2,4-triazole group and then the shift to longer wavelength is observed. In consideration of the infrared spectroscopy and the diffuse reflectivity spectra of the materials mentioned above, and other references previously reported,22,36 the resulting predicted LnMn (n ) 1-4) complexe structures are presented in Figure 1B. However, the presence of molecular H2O and OH groups in the complexes cannot be excluded. The diffractograms of the hybrid materials reproduced in Figure 4B reveal that all the materials are totally amorphous.



Figure 8. The excitation and emission spectra of the terbium hybrid materials (A) TbM1, (B) TbM2, (C) TbM3, and (D) TbM4.

TABLE 4: The Luminescence Efficiencies and Lifetimes for the Hybrid Materials hybrids

EuM1

υ01 (cm-1)a υ02 (cm-1)a A0J (s-1) Aexp (s-1) Arad (s-1) Anrad (s-1) τ (µS)b η (%)

16978 16313 50,164,13 4950 227 4723 202 4.59

a

TbM1

550

EuM2 16978 16313 50,143 5405 193 5212 185 3.57

TbM2

745

EuM3 16978 16313 50,88 8403 138 8265 119 1.64

TbM3

368

EuM4 16978 16313 50,87 7752 137 7615 129 1.77

TbM4

338

The energies of the 5D0 f 7FJ transitions (υ02). b For 5D0 f 7F2 transition of Eu3+ or 5D4 f 7F5 transition of Tb3+.

The broad peak centered around 23.91° in the XRD patterns may be ascribed to the coherent diffusion of the siliceous backbone of these hybrids.37,38 The structural unit distance, calculated by using the Bragg law, is approximately 3.83 Å, which could be attributed to a periodic distance due to the organic moiety. The small narrow peaks in the XRD figure can be due to the incompleteness of hydrolysis-condensation reactions between the excessive TEOS molecule. To enhance the sol-gel reaction, the molar ratio of silylated precursors with TEOS is 1:2, and excessive TEOS can also take place in hydrolysis-condensation reactions among themselves. Subsequently, there may exist some simple Si-O component that can readily show a better crystalline state in the hybrid systems and thus exhibit narrow peaks in the XRD patterns. Otherwise, the hydrolysis-condensation reactions of silylated precursors with TEOS can also cause the formation of some simple Si-O

component. However, the intension of these narrow peaks is very weak, which indicates that the content of these simple Si-O components is small. By comparison of the materials and the organic compound patterns (Figure 4A), it is evident that none of the materials contains measurable amounts of phases corresponding to the pure organic compounds. Moreover, no free salt is detected in any of the diffraction patterns. This is an initial indication for the formation of the true covalent-bonded hybrid materials. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) were performed on all of the amorphous materials. Figure 5 shows the DSC and TGA traces of TbM2 as an example. From the TGA curve we can see that, from 40 to 100 °C, there was a large mass decrease (8.98 %) in heat flow around 71.5 °C, and it was an endothermic process that can be observed from the DSC curve, which is due to the loss

14176 J. Phys. Chem. C, Vol. 112, No. 36, 2008 of the adsorbed and the coordinated H2O molecules. From 100 to 200 °C, the speed of the mass loss was slowed and only 6.0 % was discovered. This may be due to the evaporation of residuary DMF. The mass loss from 200 to 800 °C may be due to the decomposition of the materials which can be explained by the thermal instability of the organic parts. The main mass decrease (14.04 %) was located around a temperature of 247.1 °C. Generally, the thermal stability of the organic-inorganic hybrid is higher than that of the pure organic complexes because of the molecular level hybridization between the rigid silica matrix and the organic complexes.10,12,13 Unfortunately, in the present study, the thermal stability of this kind of hybrid was still lower, whereas compared with the organic compound O2 whose melting point is only 153 °C, the thermal stability of the organic-inorganic materials was still enhanced. This kind of material is stable under 180 °C, which can be satisfied with current applications in many areas. The scanning electron micrographs for the hybrid materials demonstrate that a homogeneous system with the size of a micrometer order of magnitude was obtained. Figure 6 shows the selected micrographs for EuM1 (A), TbM1 (B), EuM2 (C), TbM2 (D), EuM3 (E), TbM3 (F), EuM4 (G), and TbM4 (H). The quite uniform frameworks on the faces for all of the hybrid materials suggest that a self-assembly process might occur during the polymerization reaction and a complicated huge molecular system was obtained containing a functional bridge ligand with strong covalent bonds between the inorganic and organic phases. Compared to hybrid materials with doped lanthanide complexes which generally experiencing phase separation phenomena, the two phases in these molecular-based hybrids with chemical covalent bonds (Si-O) can exhibit their distinct properties together.24,39-42 There are many granules on the surfaces of the hybrids EuM1 and TbM1 which can be seen from the micrographs, which may be due to the construction of the three-dimensional Si-O-Si polymeric network structure in the hydrolysis and copolycondensation processes of silica. The granules in EuM1 and TbM1 have similar sizes, with diameters about 50-100 nm, indicating that the construction of the hybrids is mainly based on the polymeric networks of the inorganic and the organic parts through the Si-O-Si bonds, and the RE ions have nearly no influence on the sizes of the hybrids because the sizes of RE ions are too small when compared with that of the organicinorganic network. However, from the micrographs of EuM2 and TbM2 we can see that the EuM2 has a larger size than TbM2. This may be due to the enlarged extent of polycondensation reaction in EuM2, since the polycondensation reaction can be influenced by the temperature, pH value, time of the aging procedure, and so on during the sol-gel process. As for LnM3 and LnM4, the LnM1 and LnM2 materials possess a more regular microstructure and micromorphology. This may be due to the difference of the precursor molecules. Compared with the precursors P3 and P4, P1 and P2 have more coordination atoms and longer chains, which may readily provide orientation and induction ability.41 From the discussion mentioned above we can see that the self-assemble process can be influenced not only by the different sol-gel processing conditions but also by the different organic compounds in the hybrid materials. However, in this study, exchanging the RE ion (replacing Eu3+ by Tb3+) seems to have little influence on the microstructure. Luminescence Properties. Some organic functions, such as β-diketones, aromatic carboxylic acids, and heterocyclic ligands, are already well-known to be good chelating groups to sensitize luminescence of lanthanide ions. The mechanism usually

Liu and Yan described for sensitized emission in rare earth chelates proceeds through the following steps: (a) energy absorption via a ground singlet-excited singlet transition. (b) radiationless intersystem crossing from the excited singlet to the triplet states, (c) energy transfer from the ligands to lanthanide ions, and (d) lanthanide ions emission from the excited states.43-45 The excitation and emission spectra of the resulting hybrid materials are shown in Figures 7 and 8. The excitation spectra were obtained by monitoring the emission of Eu3+ or Tb3+ at 614 or 545 nm. For Eu3+ complexes, the excitation spectra were dominated by a peak centered at 392 nm. As a result, the emission lines of the hybrid materials were assigned to the 5D0 f 7FJ transitions located at 590, 614, 650, and 700 nm, for J ) 1, 2, 3, and 4, respectively. The 5D0 f 7F2 emission around 614 nm is the most predominant transition, which agrees with the amorphous characters of the hybrid materials. From Figure7, only minor differences can be seen in the spectral repartition of various organic compounds with respect to the position of emission bands which indicate that the effective energy transfer takes place between the organic ligands and the chelated Eu3+ ions. For Tb3+ complexes, a broad band centered at around 310 nm was observed in the excitation spectra and as a result, the emission lines were assigned to the 5D4 f 7FJ transitions located at 490, 544, 587, and 622 nm, for J ) 6, 5, 4, and 3, respectively. The most striking green fluorescence (5D4 f 7F5) was observed due to the fact that this emission is the most intense one. Accordingly, we may expect that through this efficient pathway, leaching of the photoactive molecules can be avoided: a higher concentration of metal ions is reached and clustering of the emitting centers may be prevented. The typical decay curves of the Eu and Tb hybrid materials were measured and they can be described as a single exponential (ln(S(t)/S0) ) -k1t ) -t/τ), indicating that all Eu3+ and Tb3+ ions occupy the same average coordination environment. As shown in Table 4, it appears as a general trend that the Eu3+ lifetimes in hybrid materials are lower than those in the corresponding Tb3+ materials, which may be ascribed to the energy match between the ligands triplet state and the resonant emissive energy level of the central Eu3+ (17 250 cm-1) not as proper as for Tb3+ (20 500 cm-1). For the different organic ligands, the lifetimes of M1 and M2 complexes with lanthanide ions are much higher than those of M3 and M4 complexes, this may be due to the coordination modality of the ligands with lanthanide ions. In LnM1 and LnM2, the O atom of the CdO group and the N atom of the 1,2,4-triazole group participate in the coordination with lanthanide ions and thus a more stable ring containing six atoms is formed, while in LnM3 and LnM4, only the N atom of the 1,2,4-triazole group participates in the coordination and this coordination effect is much weaker than that in LnM1 and LnM2, and thus led to a lower energy transfer efficiency, for both the energy match and the coordination behavior can be directly related to the capability of the ligands to absorb and transfer energy to the lanthanide ions and then influence the luminescence lifetimes. Besides, the resulting lifetimes for all samples are on a micron order of magnitude, and are slightly lower than that of some organic complexes, being ascribed to a possible quenching by OH or silanol groups. This effect, however, remains limited, confirming that the ligands used are efficiently shielding the lanthanide ions from their surroundings. On the basis of the emission spectra and lifetimes of the 5D0 emitting level, we selectively determined the emission quantum efficiencies of the 5D0 excited state of europium ion for Eu3+ hybrids. Assuming that only nonradiative and radiative processes



are essentially involved in the depopulation of the 5D0 state, the quantum efficiency of the luminescence step, η, can be defined as how well the radiative processes compete with nonradiative processes.46-54

due to the formation of the Si-O-Si network during the hydrolysis and polycondensation reactions between the precursors and TEOS. From further investigation of the luminescence properties, we can prove that an intramolecular energy transfer process took place between the chromophore part of these molecular-based hybrids and Eu3+ or Tb3+ ions thus all these hybrids can show the characteristic luminescence of corresponding Ln3+ ions. A few differences are also observed in the luminescence properties such as the quantum yield of the emission and the luminescence lifetimes showing that the coordination structures of the organic parts with Ln3+ affect the optical characteristics. As the synthesis process can be easily applied to other organic ligands and to different alkoxysilanes, we may expect to obtain stable and efficient hybrid materials in optical or electronic areas for which the desired properties can be tailored by an appropriate choice of the precursors. However, the influence of the coordination structure and the silica network on the luminescence properties in such materials needs further fundamental investigation.

η)

Ar Ar + Anr

(1)

where Ar and Anr are radiative and nonradiative transition rates, respectively. Nonradiative processes influence the experimental luminescence lifetime by the equation:46-54

τexp ) (Ar + Anr)-1

(2)

So quantum efficiency can be calculated from the radiative transition rate constant and experimental luminescence lifetime from the following equation:46-54

η ) Arτexp

(3)

where Ar can be obtained by summing over the radiative rates A0J for each 5D0 f 7FJ transition of Eu3+:46-54

Ar ) ∑A0J ) A00 + A01 + A02 + A03 + A04

(4)

Since 5D0 f 7F1 belongs to the isolated magnetic dipole transition, it is practically independent of the chemical environments around the Eu3+ ion, and thus can be considered as an internal reference for the whole spectrum; the experimental coefficients of spontaneous emission, A0J, can be calculated according to the equation:46-54

A0J ) A01(I0J/I01)(υ01/υ0J)

(5)

Here A0J is the experimental coefficient of spontaneous emissions, A01 being Einstein’s coefficient of spontaneous emission between the 5D0 and 7F1 energy levels. In vacuum, the value of A01 can be determined to be 50 s-1 approximately (A01) n3A01(vacuum)).47,52 I is the emission intensity and can be taken as the integrated intensity of the 5D0 f 7FJ emission bands.53,54 υ0J refers to the energy barycenter and can be determined from the emission bands of Eu3+’s 5D0 f 7FJ emission transitions. Here the emission intensity, I, taken as integrated intensity S of the 5D0 f 7F0-4 emission curves, can be defined as below:

Ii-j ) pωi-jAi-jNi ≈ Si-j

(6)

where i and j are the initial (5D0) and final levels (7F0-4), respectively, ωi-j is the transition energy, Ai-j is the Einstein’s coefficient of spontaneous emission, and Ni is the population of the 5D0 emitting level. On the basis of the above discussion, it can be seen the value η mainly depends on the values of two quanta: one is lifetime and the other is I02/I01 (red/orange ratio). As shown in Table 4, the quantum efficiencies of the four kinds of europium complex hybrid materials can be determined in the following order: EuM1 > EuM2 > EuM4 > EuM3, which are in agreement with the order of lifetimes. So the different compositions of the hybrid materials may have an influence on the luminescent lifetimes and quantum efficiencies. If the lifetimes and red/orange ratios are large, the quantum efficiency must be high. Conclusions In summary, compared with the conventional way of doping silica gels with organometallic complexes, we have successfully designed a novel way to construct molecular-based hybrid materials and a series of organic-inorganic materials are prepared by the sol-gel process. These materials display the connection of inorganic and organic parts on a molecular level

Acknowledgment. This work was supported by the National Natural Science Foundation of China (20671072). References and Notes (1) Suratwala, T.; Gardlund, Z.; Davidson, K.; Uhlmann, D. R. Chem. Mater. 1998, 10, 190. (2) Molina, C.; Dahmouche, K.; Santilli, C. V.; Craievich, A. F.; Ribeiro, S. J. L. Chem. Mater. 2001, 13, 2818. (3) Schneider, M.; Muller, K. Chem. Mater. 2000, 12, 352. (4) Matthews, L. R.; Knobbe, E. T. Chem. Mater. 1993, 5, 1697. (5) Casalboni, M.; Senesi, R.; Prospositio, P. Appl. Phys. Lett. 1997, 30, 2969. (6) Lebeau, B.; Fowler, C. E.; Hall, S. R. J. Mater. Chem. 1999, 9, 2279. (7) Innocenzi, P.; Kozuka, H.; Yoko, T. J. J. Phys. Chem. B. 1997, 101, 2285. (8) Mstsui, K.; Momose, F. Chem. Mater. 1997, 9, 2588. (9) Sabbatini, N.; Guar digli, M.; Lehn, J. M. Coord. Chem. ReV. 1993, 123, 201. (10) Yan, B.; Zhang, H. J.; Ni, J. Z. Mater. Sci. Eng. 1998, B52, 123. (11) Serra, O. A.; Nassar, E. I.; Zapparolli, G.; Rosa, I. L. V. J. Alloys Compd. 1994, 207/208, 454. (12) Zhang, Y.; Wang, M.; Xu, J. J. Mater. Sci. Eng. 1997, B47, 23. (13) Serra, O. A.; Nassar, E. J.; Rosa, I. L. V. J. Lumin. 1997, 72-74, 263. (14) Czarnobaj, K.; Elbanowski, M.; Hnatejko, Z.; Klonkowski, A. M.; Lis, M.; Pietraszkiewicz, M. Spectrochim. Acta, Part A 1998, 54, 2183. (15) Bredol, M.; Kynast, U.; Boldhaus, M.; Lau, C. Ber. Bunsen-Ges. Phys. Chem. 1998, 102 (11), 1557. (16) Li, H. R.; Lin, J.; Zhang, H. J.; Fu, L. S. Chem. Mater. 2002, 14, 3651. (17) Dong, D. W.; Jiang, S. C.; Men, Y. F.; Ji, X. L.; Jiang, B. Z. AdV. Mater. 2000, 12, 646. (18) Li, H. R.; Fu, L. S.; Zhang, H. J. Thin Solid Films 2002, 416, 197. (19) Li, H. R.; Lin, J.; Fu, L.S.; Guo, J. F.; Meng, Q. G.; Liu, F. Y.; Zhang, H. J. Microporous Mesoporous Mater. 2002, 55, 103. (20) Liu, F. Y.; Fu, L. S.; Wang, J.; Liu, Z.; Li, H. R.; Zhang, H. J. Thin Solid Films 2002, 419, 178. (21) Binnemans, K.; Lenaerts, P.; Driesen, K.; Gorller-Walrand, C. J. Mater. Chem. 2004, 14, 291. (22) Franville, A. C.; Zambon, D.; Mahiou, R. Chem. Mater. 2000, 12, 428. (23) Wang, Q. M.; Yan, B. Inorg. Chem. Commun. 2004, 7, 1124. (24) Wang, Q. M.; Yan, B. Appl. Organomet. Chem. 2005, 19, 952. (25) Li, H. R.; Lin, J.; Zhang, H. J.; Fu, L. S. Chem. Commun. 2001, 12, 12. (26) Carlos, L. D.; De Zea Bermudez, V.; Ferreira, R. A. S. J. NonCryst. Solids 1999, 247, 203. (27) Bermudez, V. D.; Carlos, L. D.; Duarte, M. C.; Silva, M. M.; Silva, C. J.; Smith, M. J.; Assuncao, M.; Alca´cer, L. J. Alloys Compd. 1998, 21, 275. (28) Wang, Q. M.; Yan, B. J. Mater. Chem. 2004, 14, 2450. (29) Wang, Q. M.; Yan, B. Inorg. Chem. Commum. 2004, 7, 747. (30) Wang, Q. M.; Yan, B. J. Mater. Res. 2005, 20, 592. (31) Wang, Q. M.; Yan, B. Cryst. Growth Des. 2005, 5, 497. (32) Wang, Q. M.; Yan, B. J. Organomet. Chem. 2006, 691, 540.

14178 J. Phys. Chem. C, Vol. 112, No. 36, 2008 (33) Lu, H. F.; Yan, B. J. Non-Cryst. Solids 2006, 352, 5331. (34) Yan, B.; Zhao, L. M. Mater. Lett. 2005, 59, 795. (35) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals;, Pergamon Press: Oxford, UK, 1980. (36) Franville, A. C.; Zambon, D.; Mahiou, R.; Chou, S.; Troin, Y.; Cousseins, J. C. J. Alloys Compd. 1998, 275-277, 831. (37) Carlors, L. D.; Bermudez, V. D.; Ferreira, R. A. S.; Marques, L.; Assuncao, M. Chem. Mater. 1999, 11, 581. (38) Goncalves, M. C.; Bermudez, V. D.; Ferreira, R. A. S.; Carlos, L. D.; Ostrovskii, D. J.; Rocha, Chem. Mater 2004, 16, 2530. (39) Wang, Q. M.; Yan, B. J. Photochem. Photobiol. A 2005, 175, 159. (40) Wang, Q. M.; Yan, B. J. Photochem. Photobiol. A 2006, 178, 70. (41) Yan, B.; Wang, F. F. J. Organomet. Chem. 2007, 692, 2395. (42) Yan, B.; Qiao, X. F. J. Phys. Chem. B 2007, 111, 12362. (43) Crosby, G. A.; Whan, R. E.; Alire, R. E. J. Chem. Phys. 1961, 34 (3), 743. (44) Malta, O. L. J. Lumin. 1997, 71, 229. (45) Malta, O. L. Spectrochim. Acta, Part A 1998, 54, 1593. (46) Malta, O. L.; Couto dos Santos, M. A.; Thompson, L. C.; Ito, N. K. J. Lumin. 1996, 69, 77.

Liu and Yan (47) Malta, O. L.; Brito, H. F.; Menezes, J. F. S.; Silva, F. R. G. E.; Alves, S.; Farias, F. S.; Andrade, A. V. M. J. Lumin. 1997, 75, 255. (48) Carlos, L. D.; Messaddeq, Y.; Brito, H. F.; Ferreira, R. A. S.; Bermudez, V. D.; Ribeiro, S. J. L. AdV. Mater. 2000, 12, 594. (49) Sa’ Ferreira, R. A.; Carlos, L. D.; Goncau¨lves, R. R.; Ribeiro, S. J. L.; Bermudez, V. D. Chem. Mater. 2001, 13, 2991. (50) Soares-Santos, P. C. R.; Nogueira, H. I. S.; Felix, V.; Drew, M. G. B.; Ferreira, R. A.; Carlos, L. D.; Trindade, T. Chem. Mater. 2003, 15, 100. (51) Teotonio, E. E. S.; Espinola, J. G. P.; Brito, H. F.; Malta, O. L.; Oliveria, S. F.; de Foria, D. L. A.; Izumi, C. M. S. Polyhedron 2002, 21, 1837. (52) Ribeiro, S. J. L.; Dahmouche, K.; Ribeiro, C. A.; Santilli, C. V.; Pulcinelli, S. H. J. J. Sol-Gel Sci. Technol. 1998, 13, 427. (53) Werts, M. H. V.; Jukes, R. T. F.; Verhoeven, J. W. Phys. Chem. Chem. Phys. 2002, 4, 1542. (54) Peng, C. Y.; Zhang, H. J.; Yu, J. B.; Meng, Q. G.; Fu, L. S.; Li, H. R.; Sun, L. N.; Guo, X. M. J. Phys. Chem. B 2005, 109, 15278.

JP712018N

Molecular Construction and Photophysical Properties of Luminescent

Recommend Documents