- Email: [email protected]
Luminescence properties of the TRIMEB inclusion compound of a europium tris-β-diketonate
Available online at www.sciencedirect.com
Journal of Non-Crystalline Solids 354 (2008) 2736–2739 www.elsevier.com/locate/jnoncrysol
Luminescence properties of the TRIMEB inclusion compound of a europium tris-b-diketonate S.S. Braga a,*, A.C. Coelho a, I.S. Goncßalves a, G. Santos b, F.J. Fonseca b, A.M. Andrade b, M. Peres c, W. Simo˜es c, T. Monteiro c, L. Pereira c b
a Departamento de Quı´mica e CICECO, Universidade de Aveiro, 3810-193 Aveiro, Portugal Laborato´rio de Microeletroˆnica, Departamento de Engenharia de Sistemas Eletroˆnicos, Escola Polite´cnica da Universidade de Sa˜o Paulo, Avenida Prof. Luciano Gualberto, Travessa 3, No. 380, CEP 05508-900, Sa˜o Paulo, SP, Brazil c Departamento de Fı´sica e I3N, Universidade de Aveiro, 3810-193 Aveiro, Portugal
Available online 29 January 2008
Abstract The adduct TRIMEB:Eu(BTA)3 2H2O was prepared and primarily characterized by photoluminescence (PL), and compared with free Eu(BTA)3 2H2O. Both spectra show the Eu3+ ion emission, with subtle differences between lines for the free and encapsulated complex. The temperature dependence and chemical stability were studied, taking into account (in the latter case) the PL changes with time. The use of this new material as the emissive layer in OLEDs was tested by its successful incorporation into a device, using a conductive polymer as host. The use of the TRIMEB adduct increased the stability of the device (as compared with the free Eu complex). Ó 2008 Elsevier B.V. All rights reserved. PACS: 72.80.Le; 78.55.Kz; 78.60.Fi; 85.60.Jb Keywords: Sensors; Optical spectroscopy
1. Introduction Luminescent rare earth (RE) b-diketonate (DK) complexes have been extensively studied as the basis for light conversion molecular devices [1,2]. For practical applications in optical devices, it is desirable to immobilize these in a suitable host [2] such as cyclodextrins (CDs), which can be considered as second-sphere ligands non-covalently attached to the first-sphere ligands [3]. Previous work showed that CD inclusion can affect the photophysical behavior: a very tight encapsulation results in loss of PL efficiency [4], whereas hosts with a wider cavity such as cCD induce a more efficient ligand-to-metal ion energy transfer pathway [5]. The best results have been attained
*
Corresponding author. Tel.: +351 234370200; fax: +351 234370084. E-mail address: [email protected] (S.S. Braga).
0022-3093/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2007.09.053
using the permethylated c-CD since this host has a deeper cavity [6]. Europium DKs are suitable for incorporation into the emissive layer of organic light emitting diodes (OLEDs), yielding a stable and well defined red emission line [7–10]. The advantages of using RE is their quantum internal efficiency (100%) compared to the traditional organic small molecules and/or dyes, where the p ? p* transitions have an internal quantum efficiency of only 25% [11,12]. Although these molecules feature a very high brightness, the electroluminescence (EL) bands are quite broad, thus making them of little use for quasi-monochromatic emission with a very pure color. To overcome this problem, the uses of RE complexes like europium or terbium are a very attractive approach [9,10]. However, oxidative and electric decomposition of these materials has resulted in a limited lifetime of the OLED devices thus making them unmarketable. Cyclodextrin encapsulation is helpful in this
S.S. Braga et al. / Journal of Non-Crystalline Solids 354 (2008) 2736–2739
situation since CDs act as an insulating wire coating; their unique properties also allow them to protect the encapsulated molecule from oxidation. However, studies on conducting polymer–CD composite materials are scarce, with a report on successful insulation of polyaniline with CDs [13]. Herein, inclusion of Eu(BTA)3 2H2O (BTA = 1-benzoyl-3,3,3-trifluoroacetonate) into permethylated b-CD (TRIMEB) and this adduct’s luminescence properties are reported. The new adduct was also tested as emissive material in a red emitting OLED by mixing it with a conducting polymer of poly(9-vinylcarbazol) (PVK), using a three layer hybrid structure. This work is, to the best of our knowledge, pioneering in the field of cyclodextrinbased light emitting devices. 2. Experimental IR spectra were obtained as KBr pellets using a FTIR Mattson-7000 infrared spectrophotometer. The PL data was recorded in a Spex 1704 spectrometer between 10 K and room temperature by means of a helium liquid cryostat. The room temperature EL was obtained in an Ocean Optics CCD spectrometer. The complex Eu(BTA)3 2H2O (1) was prepared as in the literature [14]. TRIMEB was purchased from Cyclolab and used as received. Preparation of TRIMEB:Eu(BTA)3 2H2O (2): a solution of Eu(BTA)3 2H2O (1) (100 mg, 0.12 mmol) in ethanol (10 mL) was treated stepwise with solid TRIMEB (172 mg, 0.12 mmol), and the mixture was stirred at room temperature for 4 h. The solvent was then evaporated to obtain a yellow glassy product. Anal. Calcd for (C63H112O35) (C30H22F9O8Eu) 2H2O (2299.01): C, 48.58; H, 6.05; Found: C, 48.45; H, 5.74. IR(KBr/mmax/ cm 1): 3435 m, 2984 m, 2939 s, 2841 m, 1732 m, 1640 s,
2737
1615 vs, 1598 m, 1578 s, 1537 m, 1491 m, 1470 m, 1383 m, 1323 vs, 1310 s, 1296 vs, 1244 m, 1189 vs, 1142 vs, 1107 s, 1074 s, 1037 vs, 971 m, 945 m, 857 m, 811 w, 797 w, 766 m, 718 m, 703 m, 631 m, 605 vw, 580 m, 559 w, 517 m, 461 w, 431 w, 349 w, 333 w, 322 vw, 315 vw, 306 vw. For testing the material for OLEDs applications, a hybrid structure was used. The devices were fabricated onto a patterned Indium–Tin–Oxide (ITO) substrate (resistivity of 30 X/h). A three layer device structure was used, with PEDOT:PSS [poly(4-styrenesulfonate)] deposited by spin coating (expected thickness of 80 nm) and butylPBD [2-(4-biphenylyl)-5-(4-tert-butylphenyl)-1,3,4-oxadiazole. 2-(4-tert-Butylphenyl)-5-(4-biphenyl)-1,3,4-oxadiazole] deposited by thermal evaporation (20 nm tick) as hole and electron transport layers, respectively (HTL, ETL). The emissive layer (deposited by spin coating with an expected thickness of 100 nm) is a mixture of 2 (5%) in PVK. The cathode was aluminum thermally evaporated. The structures of the compounds and the device configuration are shown in Fig. 1. 3. Results and discussion Addition of solid TRIMEB to a solution of Eu(BTA)3 2H2O (1) in ethanol gave a pale yellow glassy product upon drying. The product, designated as TRIMEB:Eu(BTA)3 2H2O (2), was characterized by microanalysis (showing that the 1:1 molar ratio of TRIMEB and 1 was retained in the adduct 2), FTIR and photoluminescence spectroscopy. Crystallinity was screened by powder XRD, revealing 2 to be an amorphous material. Fig. 2 shows the photoluminescence (PL) spectrum, at 10 K, of 1 and 2. The main Eu3+ intraionic f–f emissions (rapid phosphorescence) observed are 5D0 ? 7F1,2,3,4 with the 5D0 ? 7F2 as the main band. Some differences in shape
Fig. 1. The structures of the complexes and device configuration.
2738
S.S. Braga et al. / Journal of Non-Crystalline Solids 354 (2008) 2736–2739
TRIMEB:Eu(BTA)3.2H2O
TRIMEB:Eu(BTA)3.2H2O PL Intensity (a.u.)
605
610 615 620 Wavelenght (nm)
630
5
7
7
D0-> F3
5
5
625
D0-> F4
7
D0-> F1
600
Eu(BTA)3.2H2O
PL Intensity (a.u.)
5
7
PL Intensity (a.u.)
D0-> F2
Eu(BTA)3.2H2O
525
550
575
600
625
650
675
700
725 0
Wavelength (nm)
and peak position are observed and can be clearly attributed to the local ion confinement. As an interesting observation, the PL intensity for the adduct 2 is about three times higher than that for the free complex 1. In a first observation, the second coordination sphere appears to block some possible non-radiative transitions among the ligands either by intersystem crossing, or by internal conversion for the excited levels of those ligands (followed by non-radiative decay), thus promoting the intramolecular energy transfer to the excited levels of the Eu3+. In the figure inset, a detailed spectrum in the 5D0 ? 7F2 region is shown. The adduct 2 has a broader emission line and the maximum is located near 610 nm, while non-encapsulated 1 shows a narrower line emission located near 612 nm. It must be noted that no other radiative emissions are observed indicating that the organic ligands are not emissive. The variation of this main emission with temperature is shown in Fig. 3. Though featuring similarities (in a normalized scale) bellow 120 K, the complex 1 tends to a rapid intensity decrease, in contrast to encapsulated 2. At room temperature, 2 clearly have a stronger intensity at the maximum, indicating that encapsulation lowers the complex activation energy. The complex chemical stability is of huge importance to estimate its possible behavior modifications in a practical application. In particular, we look for changes in shape, intensity and position of the radiative emissions to help predict the electroluminescence of the device. In Fig. 4, the PL spectra (at 10 K) for non-encapsulated 1 are shown in two conditions: without any environment conditions and after two days under low vacuum (about 10 1 mbar). It can be observed that the vacuum environment broadens the emission lines, changing the previously observed fine structure. Although the integrated intensity of the main emission does not change significantly, this line broadening points to conformational changes of the local coordination sphere, perhaps modifying the orbital shielding effect that
100 150 200 Temperature (K)
250
300
Fig. 3. Variation of 5D0 ? 7F2 intensity emission with temperature for both complexes used. The emission intensities are not equally scaled.
Eu(BTA)3.2H2O Eu(BTA)3.2H20 (after two days)
PL Intensity (a.u.)
Fig. 2. Photoluminescence (PL) spectra of Eu(BTA)3 2H2O (1) and TRIMEB:Eu(BTA)3 2H2O (2), at 10 K under 365 nm line excitation. Inset: detailed spectra of the 5D0 ? 7F2 emission. The emission intensities are not equally scaled.
50
550
575
600
625 650 675 Wavelength (nm)
700
725
Fig. 4. Photoluminescence spectra of Eu(BTA)3 2H2O (1) without any environment conditions and after two days under low vacuum.
allows a pure Eu3+ emission. This result is not observed for the adduct 2 that appears to be much more chemically stable. Simultaneously, its strong intensity makes it a good candidate for technological applications as the EL emission is usually some orders of magnitude lower than the PL (and the small differences in the PL spectra observed have no effect upon EL) but the device stability is always very important. Based on these findings, an attempt to make an OLED with 2 as emissive layer was made. Due to the lack of information about these nanostructures, in particular and very importantly the HOMO and LUMO levels of the BTA ligand and their modification by TRIMEB, only a comparative test can be made. The device EL is shown in Fig. 5 for a safe applied voltage of 45 V. In spite of the weak emission, the PL emission region’s reproducibility (showed in background) and good stabilization open further developments for this kind of OLEDs emissive layers. The (x, y) CIE color coordinates are (0.41, 0.34); these are a little different from those of the PL data (0.56, 0.34) due to low resolution of the EL spectrum, resulting in a noticeable line
S.S. Braga et al. / Journal of Non-Crystalline Solids 354 (2008) 2736–2739
2739
developed and tested. The main focus is the nano-encapsulation of complexes of europium with TRIMEB as an insulating ligand in the second coordination sphere. The PL data shows a clear increase in the emission intensity with only some small differences in the spectrum fine structure. On the other hand, encapsulation yields a more chemically stable complex. Upon using this encapsulated complex in a host polymeric matrix as an OLED emissive layer, we observe that, compared to the non-encapsulated complex, a more stable device is obtained, although bearing strong non-uniformity. The driving voltage is high (due to the high resistivity of the encapsulated complex) while the electrical current is very low. With these promising results, we expect to have opened up a new area of possible emissive complexes for OLEDs. References Fig. 5. Electroluminescence (EL) room temperature spectrum of an emitting device. The full line is the PL spectrum of TRIMEB:Eu(BTA)3 2H2O (2). Inset: the OLED CIE color diagram.
broadening. As TRIMEB appears to be very resistive, the driving voltage is high (near 40 V) but the electrical current is only 1–2 mA. Nevertheless, the device emitting area has a strong non-uniformity, clearly due to ineffective spin coating, that must be improved. Comparing this result with the non-encapsulated 1, we observe that, though featuring higher EL uniformity and a lower driving voltage (near 20 V), the device is less stable to the electrical field. Also, one should note that TRIMEB encapsulation has modified the local Eu3+ conformation (giving a broader emission) without significant changes in the color coordinates. This feature along with the more efficient chemical stabilization is a result of high importance for displays. Further studies are currently underway to optimize the incorporation of these nanoparticles into OLED emissive layers. In particular, the spin coating process is being optimized to obtain a higher uniformity for this layer. In parallel, the complete electrical behavior of these layers in DC and AC stage will be studied. 4. Conclusions In this work, a new method to overcome the fast chemical degradation of the emissive layer used in OLEDs was
[1] G.F. de Sa´, O.L. Malta, C. de Mello Donega´, A.M. Simas, R.L. Longo, P.A. Santa-Cruz, E.F. da Silva, Coord. Chem. Rev. 196 (2000) 165. [2] P. Lenaerts, K. Driesen, R. Van Deun, K. Binnemans, Chem. Mater. 17 (2005) 2148. [3] E. Fenyvesi, L. Szente, N.R. Russel, M. McNamara, in: J. Szejtli, T. Osa (Eds.), Comprehensive Supramolecular Chemistry, vol. 3, Pergamon, Oxford, 1996 (chapter 10). [4] S.S. Braga, R.A. Sa´ Ferreira, I.S. Goncßalves, M. Pillinger, J. Rocha, J.J.C. Teixeira-Dias, L.D. Carlos, J. Phys. Chem. B 106 (2002) 11430. [5] S.S. Braga, R.A. Sa´ Ferreira, I.S. Goncßalves, P. Ribeiro-Claro, M. Pillinger, J. Rocha, J.J.C. Teixeira-Dias, L.D. Carlos, J. Incl. Phenom. Macro. Chem. 44 (2002) 261. [6] J.A. Fernandes, S.S. Braga, M. Pillinger, R.A. Sa´ Ferreira, L.D. Carlos, A. Hazell, P. Ribeiro-Claro, I.S. Goncßalves, Polyhedron 25 (2006) 1471. [7] Y. Zheng, J. Lin, Y. Liang, Y. Yu, Y. Zhou, C. Guo, S. Wang, H. Zhang, J. Alloys Compd. 336 (2002) 114. [8] Y. Miyamoto, M. Uekawa, H. Ikeda, K. Kaifu, J. Lumin. 81 (1999) 159. [9] J. Kido, Y. Okamoto, Chem. Rev. 102 (2002) 2357. [10] H. Xin, F.Y. Li, M. Shi, Z.Q. Bian, C.H. Huang, J. Am. Chem. Soc. 125 (2003) 7166. [11] Z.Y. Xie, L.S. Hung, S.T. Lee, Appl. Phys. Lett. 79 (2001) 1048. [12] C. Adachi, M.A. Baldo, M.E. Thompson, S.R. Forrest, J. Appl. Phys. 90 (2001) 5048. [13] E. Subramanian, G. Anitha, M.K. Selvam, M.I. Ali Badusha, Bull. Mater. Sci. 28 (2005) 55. [14] M. Ismail, S.J. Lyle, J.E. Newbery, J. Inorg. Nucl. Chem. 31 (1969) 1715.
Journal of Non-Crystalline Solids 354 (2008) 2736–2739 www.elsevier.com/locate/jnoncrysol
Luminescence properties of the TRIMEB inclusion compound of a europium tris-b-diketonate S.S. Braga a,*, A.C. Coelho a, I.S. Goncßalves a, G. Santos b, F.J. Fonseca b, A.M. Andrade b, M. Peres c, W. Simo˜es c, T. Monteiro c, L. Pereira c b
a Departamento de Quı´mica e CICECO, Universidade de Aveiro, 3810-193 Aveiro, Portugal Laborato´rio de Microeletroˆnica, Departamento de Engenharia de Sistemas Eletroˆnicos, Escola Polite´cnica da Universidade de Sa˜o Paulo, Avenida Prof. Luciano Gualberto, Travessa 3, No. 380, CEP 05508-900, Sa˜o Paulo, SP, Brazil c Departamento de Fı´sica e I3N, Universidade de Aveiro, 3810-193 Aveiro, Portugal
Available online 29 January 2008
Abstract The adduct TRIMEB:Eu(BTA)3 2H2O was prepared and primarily characterized by photoluminescence (PL), and compared with free Eu(BTA)3 2H2O. Both spectra show the Eu3+ ion emission, with subtle differences between lines for the free and encapsulated complex. The temperature dependence and chemical stability were studied, taking into account (in the latter case) the PL changes with time. The use of this new material as the emissive layer in OLEDs was tested by its successful incorporation into a device, using a conductive polymer as host. The use of the TRIMEB adduct increased the stability of the device (as compared with the free Eu complex). Ó 2008 Elsevier B.V. All rights reserved. PACS: 72.80.Le; 78.55.Kz; 78.60.Fi; 85.60.Jb Keywords: Sensors; Optical spectroscopy
1. Introduction Luminescent rare earth (RE) b-diketonate (DK) complexes have been extensively studied as the basis for light conversion molecular devices [1,2]. For practical applications in optical devices, it is desirable to immobilize these in a suitable host [2] such as cyclodextrins (CDs), which can be considered as second-sphere ligands non-covalently attached to the first-sphere ligands [3]. Previous work showed that CD inclusion can affect the photophysical behavior: a very tight encapsulation results in loss of PL efficiency [4], whereas hosts with a wider cavity such as cCD induce a more efficient ligand-to-metal ion energy transfer pathway [5]. The best results have been attained
*
Corresponding author. Tel.: +351 234370200; fax: +351 234370084. E-mail address: [email protected] (S.S. Braga).
0022-3093/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2007.09.053
using the permethylated c-CD since this host has a deeper cavity [6]. Europium DKs are suitable for incorporation into the emissive layer of organic light emitting diodes (OLEDs), yielding a stable and well defined red emission line [7–10]. The advantages of using RE is their quantum internal efficiency (100%) compared to the traditional organic small molecules and/or dyes, where the p ? p* transitions have an internal quantum efficiency of only 25% [11,12]. Although these molecules feature a very high brightness, the electroluminescence (EL) bands are quite broad, thus making them of little use for quasi-monochromatic emission with a very pure color. To overcome this problem, the uses of RE complexes like europium or terbium are a very attractive approach [9,10]. However, oxidative and electric decomposition of these materials has resulted in a limited lifetime of the OLED devices thus making them unmarketable. Cyclodextrin encapsulation is helpful in this
S.S. Braga et al. / Journal of Non-Crystalline Solids 354 (2008) 2736–2739
situation since CDs act as an insulating wire coating; their unique properties also allow them to protect the encapsulated molecule from oxidation. However, studies on conducting polymer–CD composite materials are scarce, with a report on successful insulation of polyaniline with CDs [13]. Herein, inclusion of Eu(BTA)3 2H2O (BTA = 1-benzoyl-3,3,3-trifluoroacetonate) into permethylated b-CD (TRIMEB) and this adduct’s luminescence properties are reported. The new adduct was also tested as emissive material in a red emitting OLED by mixing it with a conducting polymer of poly(9-vinylcarbazol) (PVK), using a three layer hybrid structure. This work is, to the best of our knowledge, pioneering in the field of cyclodextrinbased light emitting devices. 2. Experimental IR spectra were obtained as KBr pellets using a FTIR Mattson-7000 infrared spectrophotometer. The PL data was recorded in a Spex 1704 spectrometer between 10 K and room temperature by means of a helium liquid cryostat. The room temperature EL was obtained in an Ocean Optics CCD spectrometer. The complex Eu(BTA)3 2H2O (1) was prepared as in the literature [14]. TRIMEB was purchased from Cyclolab and used as received. Preparation of TRIMEB:Eu(BTA)3 2H2O (2): a solution of Eu(BTA)3 2H2O (1) (100 mg, 0.12 mmol) in ethanol (10 mL) was treated stepwise with solid TRIMEB (172 mg, 0.12 mmol), and the mixture was stirred at room temperature for 4 h. The solvent was then evaporated to obtain a yellow glassy product. Anal. Calcd for (C63H112O35) (C30H22F9O8Eu) 2H2O (2299.01): C, 48.58; H, 6.05; Found: C, 48.45; H, 5.74. IR(KBr/mmax/ cm 1): 3435 m, 2984 m, 2939 s, 2841 m, 1732 m, 1640 s,
2737
1615 vs, 1598 m, 1578 s, 1537 m, 1491 m, 1470 m, 1383 m, 1323 vs, 1310 s, 1296 vs, 1244 m, 1189 vs, 1142 vs, 1107 s, 1074 s, 1037 vs, 971 m, 945 m, 857 m, 811 w, 797 w, 766 m, 718 m, 703 m, 631 m, 605 vw, 580 m, 559 w, 517 m, 461 w, 431 w, 349 w, 333 w, 322 vw, 315 vw, 306 vw. For testing the material for OLEDs applications, a hybrid structure was used. The devices were fabricated onto a patterned Indium–Tin–Oxide (ITO) substrate (resistivity of 30 X/h). A three layer device structure was used, with PEDOT:PSS [poly(4-styrenesulfonate)] deposited by spin coating (expected thickness of 80 nm) and butylPBD [2-(4-biphenylyl)-5-(4-tert-butylphenyl)-1,3,4-oxadiazole. 2-(4-tert-Butylphenyl)-5-(4-biphenyl)-1,3,4-oxadiazole] deposited by thermal evaporation (20 nm tick) as hole and electron transport layers, respectively (HTL, ETL). The emissive layer (deposited by spin coating with an expected thickness of 100 nm) is a mixture of 2 (5%) in PVK. The cathode was aluminum thermally evaporated. The structures of the compounds and the device configuration are shown in Fig. 1. 3. Results and discussion Addition of solid TRIMEB to a solution of Eu(BTA)3 2H2O (1) in ethanol gave a pale yellow glassy product upon drying. The product, designated as TRIMEB:Eu(BTA)3 2H2O (2), was characterized by microanalysis (showing that the 1:1 molar ratio of TRIMEB and 1 was retained in the adduct 2), FTIR and photoluminescence spectroscopy. Crystallinity was screened by powder XRD, revealing 2 to be an amorphous material. Fig. 2 shows the photoluminescence (PL) spectrum, at 10 K, of 1 and 2. The main Eu3+ intraionic f–f emissions (rapid phosphorescence) observed are 5D0 ? 7F1,2,3,4 with the 5D0 ? 7F2 as the main band. Some differences in shape
Fig. 1. The structures of the complexes and device configuration.
2738
S.S. Braga et al. / Journal of Non-Crystalline Solids 354 (2008) 2736–2739
TRIMEB:Eu(BTA)3.2H2O
TRIMEB:Eu(BTA)3.2H2O PL Intensity (a.u.)
605
610 615 620 Wavelenght (nm)
630
5
7
7
D0-> F3
5
5
625
D0-> F4
7
D0-> F1
600
Eu(BTA)3.2H2O
PL Intensity (a.u.)
5
7
PL Intensity (a.u.)
D0-> F2
Eu(BTA)3.2H2O
525
550
575
600
625
650
675
700
725 0
Wavelength (nm)
and peak position are observed and can be clearly attributed to the local ion confinement. As an interesting observation, the PL intensity for the adduct 2 is about three times higher than that for the free complex 1. In a first observation, the second coordination sphere appears to block some possible non-radiative transitions among the ligands either by intersystem crossing, or by internal conversion for the excited levels of those ligands (followed by non-radiative decay), thus promoting the intramolecular energy transfer to the excited levels of the Eu3+. In the figure inset, a detailed spectrum in the 5D0 ? 7F2 region is shown. The adduct 2 has a broader emission line and the maximum is located near 610 nm, while non-encapsulated 1 shows a narrower line emission located near 612 nm. It must be noted that no other radiative emissions are observed indicating that the organic ligands are not emissive. The variation of this main emission with temperature is shown in Fig. 3. Though featuring similarities (in a normalized scale) bellow 120 K, the complex 1 tends to a rapid intensity decrease, in contrast to encapsulated 2. At room temperature, 2 clearly have a stronger intensity at the maximum, indicating that encapsulation lowers the complex activation energy. The complex chemical stability is of huge importance to estimate its possible behavior modifications in a practical application. In particular, we look for changes in shape, intensity and position of the radiative emissions to help predict the electroluminescence of the device. In Fig. 4, the PL spectra (at 10 K) for non-encapsulated 1 are shown in two conditions: without any environment conditions and after two days under low vacuum (about 10 1 mbar). It can be observed that the vacuum environment broadens the emission lines, changing the previously observed fine structure. Although the integrated intensity of the main emission does not change significantly, this line broadening points to conformational changes of the local coordination sphere, perhaps modifying the orbital shielding effect that
100 150 200 Temperature (K)
250
300
Fig. 3. Variation of 5D0 ? 7F2 intensity emission with temperature for both complexes used. The emission intensities are not equally scaled.
Eu(BTA)3.2H2O Eu(BTA)3.2H20 (after two days)
PL Intensity (a.u.)
Fig. 2. Photoluminescence (PL) spectra of Eu(BTA)3 2H2O (1) and TRIMEB:Eu(BTA)3 2H2O (2), at 10 K under 365 nm line excitation. Inset: detailed spectra of the 5D0 ? 7F2 emission. The emission intensities are not equally scaled.
50
550
575
600
625 650 675 Wavelength (nm)
700
725
Fig. 4. Photoluminescence spectra of Eu(BTA)3 2H2O (1) without any environment conditions and after two days under low vacuum.
allows a pure Eu3+ emission. This result is not observed for the adduct 2 that appears to be much more chemically stable. Simultaneously, its strong intensity makes it a good candidate for technological applications as the EL emission is usually some orders of magnitude lower than the PL (and the small differences in the PL spectra observed have no effect upon EL) but the device stability is always very important. Based on these findings, an attempt to make an OLED with 2 as emissive layer was made. Due to the lack of information about these nanostructures, in particular and very importantly the HOMO and LUMO levels of the BTA ligand and their modification by TRIMEB, only a comparative test can be made. The device EL is shown in Fig. 5 for a safe applied voltage of 45 V. In spite of the weak emission, the PL emission region’s reproducibility (showed in background) and good stabilization open further developments for this kind of OLEDs emissive layers. The (x, y) CIE color coordinates are (0.41, 0.34); these are a little different from those of the PL data (0.56, 0.34) due to low resolution of the EL spectrum, resulting in a noticeable line
S.S. Braga et al. / Journal of Non-Crystalline Solids 354 (2008) 2736–2739
2739
developed and tested. The main focus is the nano-encapsulation of complexes of europium with TRIMEB as an insulating ligand in the second coordination sphere. The PL data shows a clear increase in the emission intensity with only some small differences in the spectrum fine structure. On the other hand, encapsulation yields a more chemically stable complex. Upon using this encapsulated complex in a host polymeric matrix as an OLED emissive layer, we observe that, compared to the non-encapsulated complex, a more stable device is obtained, although bearing strong non-uniformity. The driving voltage is high (due to the high resistivity of the encapsulated complex) while the electrical current is very low. With these promising results, we expect to have opened up a new area of possible emissive complexes for OLEDs. References Fig. 5. Electroluminescence (EL) room temperature spectrum of an emitting device. The full line is the PL spectrum of TRIMEB:Eu(BTA)3 2H2O (2). Inset: the OLED CIE color diagram.
broadening. As TRIMEB appears to be very resistive, the driving voltage is high (near 40 V) but the electrical current is only 1–2 mA. Nevertheless, the device emitting area has a strong non-uniformity, clearly due to ineffective spin coating, that must be improved. Comparing this result with the non-encapsulated 1, we observe that, though featuring higher EL uniformity and a lower driving voltage (near 20 V), the device is less stable to the electrical field. Also, one should note that TRIMEB encapsulation has modified the local Eu3+ conformation (giving a broader emission) without significant changes in the color coordinates. This feature along with the more efficient chemical stabilization is a result of high importance for displays. Further studies are currently underway to optimize the incorporation of these nanoparticles into OLED emissive layers. In particular, the spin coating process is being optimized to obtain a higher uniformity for this layer. In parallel, the complete electrical behavior of these layers in DC and AC stage will be studied. 4. Conclusions In this work, a new method to overcome the fast chemical degradation of the emissive layer used in OLEDs was
[1] G.F. de Sa´, O.L. Malta, C. de Mello Donega´, A.M. Simas, R.L. Longo, P.A. Santa-Cruz, E.F. da Silva, Coord. Chem. Rev. 196 (2000) 165. [2] P. Lenaerts, K. Driesen, R. Van Deun, K. Binnemans, Chem. Mater. 17 (2005) 2148. [3] E. Fenyvesi, L. Szente, N.R. Russel, M. McNamara, in: J. Szejtli, T. Osa (Eds.), Comprehensive Supramolecular Chemistry, vol. 3, Pergamon, Oxford, 1996 (chapter 10). [4] S.S. Braga, R.A. Sa´ Ferreira, I.S. Goncßalves, M. Pillinger, J. Rocha, J.J.C. Teixeira-Dias, L.D. Carlos, J. Phys. Chem. B 106 (2002) 11430. [5] S.S. Braga, R.A. Sa´ Ferreira, I.S. Goncßalves, P. Ribeiro-Claro, M. Pillinger, J. Rocha, J.J.C. Teixeira-Dias, L.D. Carlos, J. Incl. Phenom. Macro. Chem. 44 (2002) 261. [6] J.A. Fernandes, S.S. Braga, M. Pillinger, R.A. Sa´ Ferreira, L.D. Carlos, A. Hazell, P. Ribeiro-Claro, I.S. Goncßalves, Polyhedron 25 (2006) 1471. [7] Y. Zheng, J. Lin, Y. Liang, Y. Yu, Y. Zhou, C. Guo, S. Wang, H. Zhang, J. Alloys Compd. 336 (2002) 114. [8] Y. Miyamoto, M. Uekawa, H. Ikeda, K. Kaifu, J. Lumin. 81 (1999) 159. [9] J. Kido, Y. Okamoto, Chem. Rev. 102 (2002) 2357. [10] H. Xin, F.Y. Li, M. Shi, Z.Q. Bian, C.H. Huang, J. Am. Chem. Soc. 125 (2003) 7166. [11] Z.Y. Xie, L.S. Hung, S.T. Lee, Appl. Phys. Lett. 79 (2001) 1048. [12] C. Adachi, M.A. Baldo, M.E. Thompson, S.R. Forrest, J. Appl. Phys. 90 (2001) 5048. [13] E. Subramanian, G. Anitha, M.K. Selvam, M.I. Ali Badusha, Bull. Mater. Sci. 28 (2005) 55. [14] M. Ismail, S.J. Lyle, J.E. Newbery, J. Inorg. Nucl. Chem. 31 (1969) 1715.