Open Access
Add to BibSonomyAbstract
This study achieved a substantial enhancement in electroluminescence by coupling localized surface plasmons in a single layer of Ag nanoparticles. Thermal evaporation was used to fabricate 20-nm Ag particles sandwiched between a gallium-doped zinc oxide film and a glass substrate to form novel window materials for use in polymer light-emitting diodes (PLEDs). The PLEDs discussed herein are single-layer devices based on a poly(9,9-di-n-octyl-2,7-fluorene) (PFO) emissive layer. In addition to low cost, this novel fabrication method can effectively prevent interruption or degradation of the charge transport properties of the active layer to meet the high performance requirements of PLEDs. Due to the surface-plasmon-enhanced emission, the electroluminescence intensity was increased by nearly 1-fold, compared to that of the same PLED without the interlayer of Ag nanoparticles.
©2011 Optical Society of America
1. Introduction
Several studies have utilized surface-plasmon-coupled emission (SPCE), excited on the interface between metallic and dielectric nanoparticles, to enhance the luminescence efficiency of polymer light-emitting diodes (PLEDs), and have reported the results [1–4]. In general, the melting point of metallic nanoparticles decreases rapidly as the particle size decreases. When metallic nanoparticles such as gold (Au) or silver (Ag) are smaller than 100 nm, their melting points are less than 400 °C. Transparent conducting oxide (TCO) films, such as indium tin oxide (ITO) and zinc oxide (ZnO), are generally formed above 400 °C to meet low resistivity requirements. Because metallic nanoparticles cannot tolerate this fabrication temperature, metallic nanoparticles must be placed between a polymer film and a TCO substrate, seriously altering the hole mobility and resulting in extremely limited electroluminescence efficiency. To solve this problem, our earlier study [5] used pulsed laser deposition (PLD) to manufacture a gallium-doped zinc oxide (GZO) thin film at a fabrication temperature below 150 °C. The resistivity of the film was on the order of 10−4 Ω-cm, achieving more than 90% visible light transmission, and the microscopic physical surface properties were comparable to those of conventional ITO films. Because this film can be fabricated at a low temperature, the metallic nanoparticle layer is no longer subjected to the TCO film process and can be placed between the GZO and glass substrate, resulting in a much weaker impact on the hole mobility of the PLEDs.
In this study, Ag nanoparticles were fabricated using thermal evaporation [6–8]. Particles that exhibited size uniformity were demonstrated according to their absorption spectra and then sandwiched between GZO film and a glass substrate. Atomic force microscopy (AFM), conducting atomic force microscopy (CAFM), and scanning surface potential microscopy (SSPM) were then employed to elucidate the impact of the interlayer of Ag nanoparticles on the surface microstructure and electrical properties of the GZO film. Finally, a polymer material was spin-coated onto the GZO film with an interlayer of Ag nanoparticles, and then deposited onto the metal electrodes to form the proposed PLED structure (Fig. 1 ). A spectral measurement system was used to observe surface plasma localization and any enhancement of the optical gain during electroluminescence. The experimental results proved that the novel PLED device is highly practical, and the low-cost fabrication makes it commercially feasible.
2. Experiments
The Ag films were fabricated via vacuum deposition from an effusive atom source (a tungsten boat containing 99.999% Ag) at normal incidence. The pressure in the diffusion-pumped chamber was 5 × 10−6 Torr during the deposition. The evaporation substrates were Fisher brand No. 1 circular glass coverslips . Prior to deposition, the glass coverslips were cleaned by sequential sonication for 20 min at room temperature in de-ionized water and acetone. The mass thickness of the Ag films was measured with a quartz crystal microbalance that was calibrated using AFM (Dimension 3100, Vecco Instrument, Santa Barbara, CA). The AFM mass thickness calibration was performed by masking half of a glass substrate and depositing Ag onto the other half. The AFM tip repeatedly scanned the boundary, and the average height was calculated. Ag film with mass thicknesses of 2 nm was deposited on glass substrates at a fixed deposition rate of 0.3 nm/s. The annealing treatments, in which a substrate heater was applied to the sample, heated the substrate at 210 °C for 30 min. Figure 2(a) shows the AFM image of the Ag nanoparticle layer and the scanned area was 1 μm × 1 μm. The average diameter and density analyzed by Fig. 2(a) were approximately 20 nm and 1400 particles per μm2, respectively. Figure 2(b) shows the absorption spectrum of the Ag nanoparticles. The surface plasmon resonance wavelength (SPRW) value determined by the wavelength corresponding to the absorption maximum was 450 nm.
The ceramic target was ZnO (99.999% purity) mixed with 3.35 wt.% Ga (99.999% purity). The ambient gas was high-purity O2 (99.999%). The GZO films were prepared using PLD at a constant O2 background pressure of 30 mTorr and a substrate temperature of 150 °C. The substrates with and without Ag nanoparticles were mounted parallel to the target surface at a distance of 5 cm. A Q-switched Nd:Yag laser (LS-2137U, LOTIS) with second harmonic operating at 355 nm and 7 ns duration was used as the light source for target ablation. The laser beam is focused using a 30 cm positive lens. The average energy is approximately 160 mJ per shot and the pulse rate is 10 Hz. Each deposition process lasts 10 min and the thickness of the films is between 100 nm and 150 nm. Ultraviolet-visible absorption spectra were obtained using a Pye Unicam PU 8800 UV/Vis spectrophotometer (Cambridge, UK). Glass microscope slides were cut into 0.9 cm × 2.6 cm rectangles to fit into the sample holder. A blank glass slide was used as a reference for each measurement. Electrical measurements were performed under ambient conditions using a commercial atomic force microscope. Rectangular Si tips (SCM-PIT7, Vecco Instruments) with a spring constant of 2.50 N/m and a resonant frequency of 75 kHz were used for the AFM measurements. The contact force was held at approximately 200 nN, as determined from a force-distance plot. The tips were pre-coated with a Cr layer and subsequently coated with a 20-nm PtIr film by ion sputtering. The CAFM and SSPM procedures employed in our studies have been described previously [9–12]. The chemical states of films are identified using X-ray photoelectron spectroscopy (XPS) using a VG Scientific ESCALAB 250 spherical sector analyzer with a monochromatic AlKα radiation source (1500 eV). The elemental composition is determined by calculating the relative peak areas at specific binding energies.
The proposed structure of the top-emitting PLED in our experiments is shown in Fig. 1. The PFO (poly[9,9-di-n-octyl-2,7-fluorene]) as the light emitting layer was spin-coated from a toluene solution to form a 50-nm-thick film on top of the GZO substrate with an interlayer of Ag nanoparticles. Finally, a 1-nm-thick lithium fluoride (LiF) film interfacial layer and an 80-nm-thick Al cathode were deposited selectively through a shadow mask to fabricate the hole array patterns with an area of 2 mm × 2 mm to define the light emitting region. A sample without Ag nanoparticles was also prepared in the same manner as a standard PLED device to quantitatively evaluate the plasmon enhancement. The current density-voltage (J-V) and luminance-voltage (L-V) characteristics of the devices were both measured using a source meter (Keithley-2400) and a luminance meter (LS-100). All current density-voltage-luminance (J-V-L) measurements were performed at room temperature and in ambient atmosphere without any protective coatings. To elucidate the gain of the quantum efficiency, electroluminescence (EL) spectra were characterized by an optical spectrum analyzer (Ando-6315).
3. Results and discussion
Figure 3 shows the absorption spectra of the GZO films with (red curve) and without (black curve) a 20-nm-diameter Ag interlayer on glass substrate. The sample must be heated to 150 °C for approximately 10 min when GZO is deposited, potentially causing structural disruption of the Ag nanoparticles on the glass substrate. As revealed by the absorption spectrum in Fig. 3, the SPRW of the Ag nanoparticles was shifted to 465 nm (red shift) after the GZO process, and the fluorescence peak of the PFO (440 nm in this case) almost coincide. Note that the transparency of the GZO film at wavelengths ranging from 400 to 800 nm remained more than 90%, even though a single layer of Ag nanoparticles was placed between the GZO film and glass substrate .
Fig. 3 Absorption spectra of a bare GZO substrate and a GZO substrate with an interlayer of 20-nm-diameter Ag nanoparticles.
Figure 4(a) shows acquired 2.5 μm × 2.5 μm AFM images of GZO substrates with (right) and without (left) an interlayer of Ag nanoparticles. The root-mean-square (RMS) roughness values of the GZO surface with and without an interlayer of Ag nanoparticles are 2.33 and 1.24 nm, respectively. Because the heat conductivity of Ag nanoparticles is higher than that of glass, in the GZO process performed by this study, the heating efficiency of a glass substrate depositing Ag nanoparticles was far higher than that of the bare glass substrate, which might easily cause the formation of large grain size on the GZO surface.
Fig. 4 (a) Topography, (b) current, and (c) contact potential difference images of a bare GZO substrate (left) and a GZO substrate with an interlayer of 20-nm-diameter Ag nanoparticles (right). (d) Histograms of contact current distribution, obtained from (b).
The current images of GZO films with (right) and without (left) an interlayer of Ag nanoparticles, with a bias of + 50 mV applied to the sample during contact-mode scanning, and corresponding to Fig. 4(a), are presented in Fig. 4(b). In the current images, both conducting and nonconducting regions are present on the surfaces, with the former characterized by its brightness. The local current-voltage (I-V) measurements show that all of the measured I-V relationships were linear when the tip was located where the current exceeded 50 nA. The contact was apparently an ohmic contact, and these surface regions are called “conducting regions”. However, in regions where the current was less than 50 nA, the I-V relationship was nonlinear. Such regions are called “nonconducting regions”, and Fowler-Nordheim electron tunneling was responsible for the current detected in these regions. The percentage of coverage of these two regions could be determined using the current distribution histograms of the current images in Fig. 4(b), which are presented in Fig. 4(d). The percentages of coverage of the conducting region of GZO with and without an interlayer of Ag nanoparticles were approximately 90.8% and 89.3%, respectively. The higher contact current of the GZO surface leads to a greater quantity or flow rate of free electrons in the region, indicating relatively stable conductivity. Figure 4(d) shows that when the GZO substrate has an interlayer of Ag nanoparticles, the stability region of conductivity in which the contact current exceeds 150 nA was raised from 67.5% to 72.7%. To explain this phenomenon, XPS measurements were performed to evaluate the Ga dopant concentration and oxygen vacancies on the GZO surface. X-ray photoelectron spectroscopy (XPS) compositional analysis demonstrated that the surface of the GZO film with an interlayer of Ag nanoparticles had no Ag atoms. However, the Ga dopant concentration determined by calculating the peak areas of the binding energy at 1115 eV decreased from 2.91 atomic% to 2.56 atomic%. Figure 5 shows the XPS spectrum of O 1s for both of the GZO substrates (a) without and (b) with an interlayer of Ag nanoparticles. The typical O 1s peak at the surface can be consistently fitted using three nearly-Gaussian curves centered at 529.96 ± 0.08, 531.37 ± 0.11, and 532.53 ± 0.06 eV. The high binding energy component located at 532.53 ± 0.06 eV has been reported as the presence of loosely bound oxygen on the surface of the GZO film that belonged to the specific species of adsorbed H2O or O2− [13,14]. The component on the low binding energy side of the O 1s spectrum at 529.96 ± 0.08 eV is attributed to O2− ions in the wurtzite structure of a hexagonal Zn2+ ion array surrounded by Zn (or the substitution of Ga) atoms with their full complement of nearest-neighbor O2− ions [15]. The medium binding energy component centered at 531.37 ± 0.11 eV is associated with O2− ions in the oxygen deficient regions within the matrix of ZnO [16]. Therefore, changes in the intensity of this component may be connected in part to variations in the concentration of oxygen vacancies. According to the analysis results, as shown in Fig. 5, Ag nanoparticles could increase oxygen vacancies on the GZO surface by approximately 45%. The free electrons provided by the increased oxygen vacancies were far more than those that were reduced, due to the decrease of Ga dopant concentration, which promoted a stability region of conductivity on the GZO surface by approximately 5.2%.
Fig. 5 XPS of O 1s for (a) a bare GZO substrate and (b) a GZO substrate with an interlayer of 20-nm-diameter Ag nanoparticles.
The contact potential difference (CPD) image was captured in “lift mode” at a lift scan height (tip-sample separation) of 30 nm with a 5-V AC voltage applied to the tip. The measurement was performed concurrently with the topography scan in tapping mode using a Cr/PtIr-coated tip. Figure 4(c) presents the CPD images of the GZO films with (right) and without (left) an interlayer of Ag nanoparticles in a 2.5 μm × 2.5 μm scanning area. The mean work function (WF) of the GZO surface with an interlayer of Ag nanoparticles was approximately 4.95 eV according to the CPD value calculation [10], which is higher than the 4.81 eV WF of the GZO film with no such interlayer. Moreover, as shown in Fig. 4(c), the maximum peak value of WF on the GZO surface decreased from 0.035 eV to 0.018 eV after inserting the Ag nanoparticles. The higher uniformity of WF on the GZO surface was caused by high heat conductivity of the Ag nanoparticles. Previously published research results [10,11] confirm that the mean WF of the GZO surface is directly proportional to its surface oxygen content. The relative peak areas of the binding energy at 528~536 eV, as shown in Fig. 5, represent the total oxygen concentration. After deducting the proportion taken up by oxygen vacancies, the total oxygen concentration of the GZO film with an interlayer of Ag nanoparticles, was approximately 58% higher than that of the bare GZO film, which was the main cause of promoting the surface mean WF. The GZO films are used as anode materials for the PLEDs and must, therefore, have a high and uniform surface WF to aid the transport of hole carriers efficiently. The results confirmed that the interlayer of Ag nanoparticles could substantially promote WF on the GZO surface and effectively ensure WF uniformity, which is helpful for increasing the luminescent efficiency of PLED components.
Figure 6(a) shows the J-V characteristics of PLEDs with (red curve) and without (black curve) an interlayer of Ag nanoparticles. As can be clearly seen, the PLED with an interlayer of Ag nanoparticles had the slight lower threshold voltage because its injection barrier for holes, taken as the difference between the highest occupied molecular orbital (HOMO) of PFO (5.8 eV) and the WF of GZO with an interlayer of Ag nanoparticles (4.95 eV), was smaller than that of the PLED with a bare GZO anode. Note that the WF of the bare GZO substrate was 4.81 eV. The decrease in threshold voltage was a reflection of improved hole injection efficiency. Figure 6(b) shows the L-V characteristics of PLEDs with (red curve) and without (black curve) an interlayer of Ag nanoparticles. Both the turn-on voltage (at 1 cd/m2) and the operation voltage (at 100 cd/m2) decreased from 6.12 V and 9.12 V to 5.91 V and 7.83 V, respectively, when using a GZO with an interlayer of Ag nanoparticle substrate as an anode. The electroluminescence photographs of PLEDs with (right) and without (left) an interlayer of Ag nanoparticles operating at 12 V bias are displayed in Fig. 6(c). We can clearly find that the brightness of the proposed PLED structure in the light emitting region is much higher than that of the standard PLED structure. In Fig. 6(d), the EL spectra normalized by the peak intensities for both samples are shown. The applied bias was set at 12 V for both samples. It can be seen that the two spectra coincide perfectly in terms of peak positions and shapes, but have obviously distinct intensities at 440 nm of wavelength. The above result strongly indicates that the origin of trap states [17–20] is not due to the presence of GZO substrate, but is due to a relatively high density of natural impurities in the polymer thin films. Thus, the emissions from both samples were due to PFO fluorescence and not simply to the localized surface plasmons of Ag nanoparticles. For electroluminescence intensity at 440 nm of wavelength, the PLED with an interlayer of Ag nanoparticles was nearly 1-fold higher than that of the PLED with a bare GZO anode, due to the surface-plasmon-enhanced emission.
Fig. 6 (a) Variation of current density with bias voltage, (b) relationship between luminance intensity and bias voltage, and (d) electroluminescence spectra normalized by the peak intensities. In all graphs, the red and black curves denote PLED with and without an interlayer of 20-nm-diameter Ag nanoparticles, respectively. (c) The electroluminescence photographs of the proposed (left) and standard (right) PLEDs.
4. Conclusion
This paper proposes a PLED structure, GZO (with an interlayer of Ag nanoparticles)/PFO/LiF/Al. Conducting atomic force microscopy (CAFM) measurement showed that the high thermal conductivity of the interlayer of Ag nanoparticles can increase the oxygen vacancies on the GZO surface, causing the high-current area (contact current > 150 nA) to increase by 5.2%. Furthermore, due to the increase of the total oxygen concentration on the GZO surface, the mean WF measured using SSPM increases from 4.81 eV to 4.95 eV, which can effectively improve hole injection efficiency for PLEDs. Through the use of Ag nanoparticles with a diameter of 20 nm embedded in a GZO layer, the electroluminescence intensity was increased substantially, owing to the strong coupling of excitons with localized surface plasmons in the Ag nanoparticle monolayer. Using chemically stable Ag nanoparticles under ambient conditions, as well as the simple fabrication process, are suitable for enhancing the emission of large-area PLEDs.
Acknowledgment
The authors gratefully acknowledge the financial support for this research from the National Science Council of Taiwan under grant No. NSC 100-2112-M-415-001-MY3.
References and links
1. T. D. Neal, K. Okamoto, and A. Scherer, “Surface plasmon enhanced emission from dye doped polymer layers,” Opt. Express 13(14), 5522–5527 (2005). [CrossRef] [PubMed]
2. M. Ishifuji, M. Mitsuishi, and T. Miyashita, “Enhanced optical second harmonic generation in hybrid polymer nanoassemblies based on coupled surface plasmon resonance of a gold nanoparticle array,” Appl. Phys. Lett. 89(1), 011903 (2006). [CrossRef]
3. D. A. Clymer and M. A. Matin, “Characterization of thin-metal anode buffers in organic devices,” Microw. Opt. Technol. Lett. 48(10), 2070–2072 (2006). [CrossRef]
4. J. Zhou, J. Yang, Y. Sun, D. Zhang, J. Shen, Q. Zhang, and K. Wang, “Effect of silver nanoparticles on photo-induced reorientation of azo groups in polymer films,” Thin Solid Films 515(18), 7242–7246 (2007). [CrossRef]
5. C.-F. Yu, S.-H. Chen, S.-J. Sun, and H. Chou, “Influence of the grain boundary barrier height on the electrical properties of Gallium doped ZnO thin films,” Appl. Surf. Sci. 257(15), 6498–6502 (2011). [CrossRef]
6. C. Douketis, T. L. Haslett, Z. Wang, M. Moskovits, and S. Iannotta, “Self-affine silver films and surface-enhanced Raman scattering: linking spectroscopy to morphology,” J. Chem. Phys. 113(24), 11315–11323 (2000). [CrossRef]
7. W. A. Weimer and M. J. Dyer, “Tunable surface plasmon resonance silver films,” Appl. Phys. Lett. 79(19), 3164–3166 (2001). [CrossRef]
8. R. Gupta, M. J. Dyer, and W. A. Weimer, “Preparation and characterization of surface plasmon resonance tunable gold and silver films,” J. Appl. Phys. 92(9), 5264–5271 (2002). [CrossRef]
9. H. N. Lin, H. L. Lin, S. S. Wang, L. S. Yu, G. Y. Perng, S. A. Chen, and S. H. Chen, “Nanoscale charge transport in an electroluminescent polymer investigated by conducting atomic force microscopy,” Appl. Phys. Lett. 81(14), 2572 (2002). [CrossRef]
10. S.-H. Chen, “Work function changes of treated indium-tin-oxide films for organic light emitting diodes investigated using scanning surface potential microscopy,” J. Appl. Phys. 97(7), 073713 (2005). [CrossRef]
11. S.-H. Chen, C.-F. Yu, Y.-S. Lin, W.-J. Xie, T.-W. Hsu, and D. P. Tsai, “Nanoscale surface electrical properties of aluminum zinc oxide thin ðlms investigated by scanning probe microscopy,” J. Appl. Phys. 104(11), 114314 (2008). [CrossRef]
12. S.-H. Chen, S.-T. Yu, Y.-Y. Liou, C.-F. Yu, C.-F. Lin, and P.-C. Kao, “Increasing the PLED luminescence efficiency by exploiting the surface plasmon resonance effect,” J. Electrochem. Soc. 158(3), J53–J57 (2011). [CrossRef]
13. S. Kawamura, J. Sakurai, M. Nakano, and M. Takagi, “Recrystallization of Si on amorphous substrates by doughnut‐shaped cw Ar laser beam,” Appl. Phys. Lett. 40(5), 394–395 (1982). [CrossRef]
14. M. N. Islam, T. B. Ghosh, K. L. Chopra, and H. N. Acharya, “XPS and X-ray diffraction studies of aluminum-doped zinc oxide transparent conducting films,” Thin Solid Films 280(1–2), 20–25 (1996). [CrossRef]
15. L. Kameswara Rao and V. Vinni, “Novel mechanism for high speed growth of transparent and conducting tin oxide thin films by spray pyrolysis,” Appl. Phys. Lett. 63(5), 608–611 (1993). [CrossRef]
16. J. C. C. Fan and J. B. Goodenough, “X-ray photoemission spectroscopy studies of Sn-doped indium-oxide films,” J. Appl. Phys. 48(8), 3524–3531 (1977). [CrossRef]
17. A. J. Campbell, D. D. C. Bradley, E. Werner, and W. Brütting, “Transient capacitance measurements of the transport and trap states distributions in a conjugated polymer,” Org. Electron. 1(1), 21–26 (2000). [CrossRef]
18. Å. Johansson and S. Stafström, “Polaron dynamics in a system of coupled conjugated polymer chains,” Phys. Rev. Lett. 86(16), 3602–3605 (2001). [CrossRef] [PubMed]
19. W. Brütting, S. Berleb, and A. G. Mückl, “Space-charge limited conduction with a field and temperature dependent mobility in Alq light-emitting devices,” Synth. Met. 122(1), 99–104 (2001). [CrossRef]
20. D. M. Basko and E. M. Conwell, “Stationary polaron motion in a polymer chain at high electric fields,” Phys. Rev. Lett. 88(5), 056401 (2002). [CrossRef] [PubMed]
