- Email: [email protected]
Surface plasmon-enhanced localized electric field in organic light-emitting diodes by incorporating silver nanoclusters
Applied Surface Science 295 (2014) 266–269
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Surface plasmon-enhanced localized electric field in organic light-emitting diodes by incorporating silver nanoclusters Ying-Chung Chen a , Chia-Yuan Gao a , Kan-Lin Chen b , Chien-Jung Huang c,∗ a b c
Department of Electrical Engineering, National Sun Yat-sen University, Kaohsiung, Taiwan Department of Electronic Engineering, Fortune Institute of Technology, Kaohsiung, Taiwan Department of Applied Physics, National University of Kaohsiung, Kaohsiung, Taiwan
a r t i c l e
i n f o
Article history: Received 7 October 2013 Received in revised form 28 November 2013 Accepted 7 January 2014 Available online 15 January 2014 Keywords: Silver nanoclusters Localized electric field Electron injection
a b s t r a c t
The influence of silver nanoclusters (SNCs) on the performance of organic light-emitting diodes is investigated in this study. The SNCs are introduced between the electron-injection layer and cathode alumina by means of thermal evaporation, resulting that different absorption peaks of SNCs were formed. A higher luminance and electron-injection ability are obtained when the mean cluster size is 34 nm. The surfaceenhanced Raman scattering spectroscopy reveals that the localized electric field around the SNCs is enhanced, resulting in an increase in electron injection from cathode electrode. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Organic light-emitting diode (OLED) has displayed significant potentiality in the application of lighting and display, which is mainly because of its low power consumption. So far, the efficiency of device is not optimal, so many researchers are still trying to improve the efficiency of device by different means, e.g., doping phosphorescence material into the emitting layer [1,2], blocking hole with hole-blocking layer [3–6], or improving the hole/electron balance by increasing electron injection [7–9], which is to improve the efficiency of device. The efficiency of OLED can be improved, but these methods increased the complexity of device structure which increased the fabrication time and cost of device. Recently, it has drawn a lot of attention that the surface plasmon resonance effect (SPRE) of metal nanocluster is applied to OLED which is mainly because of surface plasmon-enhanced luminescence of OLED being enhanced. In recent reports on both organic and inorganic light-emitting diodes, the coupling process between surface plasmons and radiated light is rapider than spontaneous recombination of excitons [10,11]. Therefore, the exciton lifetime in surface plasmon resonance (SPR) devices is considerably reduced. Furthermore, spontaneous emission rate of exciton is inversely proportional to the exciton lifetime [12]. In other words, emission efficiency is increased with the subtraction in the
∗ Corresponding author. Tel.: +886 7 5919475; fax: +886 7 5919357. E-mail address: [email protected] (C.-J. Huang). 0169-4332/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2014.01.031
exciton lifetime. Besides, a thin silver layer was incorporated as an interlayer between the electron-transport layer and the electrode which can improve the efficiency of OLEDs [13]. However, the cause of improving the efficiency has not yet been studied in detail. In this article, the silver nanoclusters (SNCs) were introduced between electron-injection layer (EIL) and cathode, and it was also observed that how SPR wavelength of different mean cluster sizes of SNCs influenced for the enhancement of device efficiency. In addition, a different viewpoint in surface plasmon-enhanced luminescence by a theory of surface-enhanced Raman scattering (SERS) is presented. 2. Experimental The schematic structure of OLEDs is shown in Fig. 1. Indium tin oxide (ITO) coated on glass with a sheet resistance of 10 /sq was used as the starting substrate. The substrate was cleaned with acetone, methanol and deionized water, and then dried with nitrogen gas. After cleaning process, the substrates were loaded into a thermal evaporator. Afterward, N,N -bis-(1naphthyl)-N,N-dipheny1,1 - bipheny1 -4-diamine (NPB; 35 nm), tris-(8-hydroxyquinoline) aluminum (Alq3 ; 40 nm), 4, 7-dipheny11, 10-phenanthroline (BPhen; 10 nm), lithium fluoride (LiF; 0.5 nm), SNCs (× nm) and aluminum (Al; 100 nm) were deposited. NPB and BPhen were used as the hole and electron transporting layers, respectively. Alq3 was used as emitting layer. LiF, Al and ITO were used as EIL, cathode and anode, respectively. To obtain the absorption peak of SNCs being the closest to the photoluminescence (PL) peak of Alq3 , the mean cluster size of the SNCs was
Y.-C. Chen et al. / Applied Surface Science 295 (2014) 266–269
changed from 15 to 43 nm. Furthermore, the different mean cluster sizes of SNCs were deposited on top of glass substrate in order to measure SERS. During the deposition, the base pressure of the chamber was maintained as low as 2.4 × 10−6 Torr. The active area of the device was 6 × 6 mm2 . The deposition rates of all organic materials were maintained at 0.01 nm/s, except for the Al that deposition rate was 0.5 nm/s. In addition, the SNCs were fabricated by thermal evaporation of silver slug. The deposition rate of SNCs is 0.01 nm/s. Besides, the different mean cluster sizes of SNCs are obtained by different deposition time. The deposition time for the mean particle sizes of 15, 24, 34, 43 nm is 100, 200, 300, and 400 s, respectively. The thickness of the layers is controlled by using a quartz–crystal monitor. The current density–voltage (J–V) characteristics and luminance–voltage (L–V) of the devices were measured by using a Keithley 2400 (Keithley instruments Inc, USA) and a PR-655 (Photo Research Inc, USA), respectively. The SERS and absorption spectrum were measured by using a BWII RAMaker
267
Fig. 1. Schematic device structure of the OLED with SNCs.
Fig. 2. The SEM images of SNCs at (a) 15 nm, (b) 24 nm, (c) 34 nm, and (d) 43 nm. (e) Normalized absorption spectra of the SNCs with different mean cluster sizes and PL spectrum of the Alq3 .
268
Y.-C. Chen et al. / Applied Surface Science 295 (2014) 266–269
1.4
SNCs (15 nm) SNCs (34 nm) SNCs (54 nm)
(a)
Normalized absorption
1.2
444,
1.0
525, 577 (nm)
0.8 0.6 0.4 0.2 0.0 350
400
450
500
550
600
650
700
Wavaelength (nm) 7000
(b)
Intensity (counts/15s)
6000
SNCs (15 nm) SNCs (34 nm) SNCs (54 nm)
5000 4000 3000 2000 0
500
1000
1500
2000
Raman Shift (cm-1)
Fig. 3. (a) Current density–voltage curves and (b) Luminance–voltage curves of OLEDs. Inset of (a) shows the efficiency–voltage characteristics of OLEDs.
(Protrustech, Taiwan) and a U-3900 spectrophotometer (Hitachi, Japan), respectively. The photoluminescence (PL) spectrum and scanning electron microscope (SEM) images were measured by using an IK series (Kimmon Koha, Japan) and a JSM-6700F (JEOL, Japan). All measurements were made at room temperatures under air ambient. 3. Results and discussion Fig. 2(a–d) show the SEM images of SNCs with different mean cluster sizes on glass substrate. The mean cluster sizes of SNCs in Fig. 2(a–d) were calculated and found to be 15, 24, 34, and 43 nm, respectively. It is well known that the absorption peak depends on SNCs size, shape, and spacing between SNCs. The absorption spectra of SNCs with different mean cluster sizes and the PL spectrum of Alq3 are shown in Fig. 2(e). It can be clearly seen that the absorption peaks were red-shift with increase in mean cluster size of SNCs. Similarly, when the mean cluster size of metal NCs is increased, the absorption peaks were red-shift, which is consistent with our previous study [14]. Besides, it is observed that the PL peak of Alq3 is about 523 nm. When the mean cluster size of SNCs is 34 nm, the absorption peak of SNCs is about 525 nm that is closest to the PL peak of Alq3 . Consequently, as is expected, the best performance of device can be obtained when the mean cluster size of SNCs is 34 nm. The resulting current density–voltage curves are plotted in Fig. 3(a). The current density of the device with SNCs is remarkably increased because the localized electric field around the SNCs is enhanced, which results in an increase in electron injection from cathode electrode. In addition, the spacing between the emissive layer and SNCs is necessary because the spacing can avoid the nonradiative quenching of exciton [15]. The current efficiency–voltage curves are plotted in the inset of Fig. 3(a). The current efficiency of the device with SNCs at 34 nm is quite high compared with that of the device without SNCs. Generally, the mobility of electrons in
Fig. 4. (a) Normalized absorption spectra of the SNCs with different mean cluster sizes. (b) SERS spectrum at excitation laser’s wavelength (532 nm) for SNCs with different mean cluster sizes.
the electron transporting layer is a few orders of magnitude lower than that of holes in the hole transporting layer [16,17], resulting in the imbalance between holes and electrons within the emitting layer. Therefore, the current efficiency of the devices with SNCs are higher than that of the device without SNCs because of enhancement in electron injection from the cathode to emitting layer, resulting in better balance for the ratio of hole/electron injection. Fig. 3(b) shows the L–V characteristic of the device with different sizes of SNCs and without SNCs. The luminance of devices with SNCs is higher than that of the device without SNCs due to its surface plasmon-enhanced spontaneous emission rate. It is well known that the SPRE was induced by interaction between the surface charge of metal and the electromagnetic field of the incident light [18]. On the other hand, luminance depends on the current density and current efficiency. A high current density and high current efficiency will benefit the high luminance of device. Consequently, the maximum luminance of 14070 cd/m2 is obtained when the SNCs is 34 nm. Furthermore, the result is consistent with what we previously extrapolated. That is to say, when the SNCs were introduced between the LiF and Al cathode, the SNCs can generate the SPRE, which can not only enhance the spontaneous emission rate but also improve the balance for the ratio of hole/electron injection. However, the enhanced electron injection is attributed to SPRE which induces the enhanced localized electric field around the SNCs. Fig. 4(a) shows the absorption spectra of the SNCs with different mean cluster sizes. It is observed that the peak wavelength is 525 nm for SNCs of 34 nm, which is closest to the excitation laser’s wavelength (532 nm) of SERS. In addition, when the size of the SNCs becomes bigger, it was found that the absorbance peak obviously shifts from the short wavelength to the long wavelength, and the full-width at half-maximum (FWHM) is from narrow into wide. The widening and red shift for the larger SNCs are because of an increasing contribution of higher-order plasmon modes for the larger SNCs [19]. To further confirm the relationship between the electric field and the SNCs, the SNCs were deposited on top of glass substrate to
Y.-C. Chen et al. / Applied Surface Science 295 (2014) 266–269
measure SERS. The SERS data obtained from different mean cluster sizes of SNCs are shown in Fig. 4(b). The spectral feature characteristics of SNCs were about 1260, 1348, and 1555 cm−1 . As anticipated, Raman spectrum in SNCs of 34 nm shows the highest intensity due to the fact that the peak wavelength in SNCs of 34 nm is closest to the excitation laser’s wavelength of SERS, resulting in large electric field around the SNCs. The Raman intensity is proportional to the square of the incident field. The relationship between the Raman intensity and the incident field can be expressed as follows [20]:
IRS ∝
∂˛ E0 [cos (ωo − ωv ) t + cos (ωo + ωv ) t] ∂Q
2
(1)
Where IRS is the Raman intensity, ˛ is the molecular polarizability, Q is the nuclear displacement, E0 is the amplitude of the incident field, and ωo and ωv are the angular frequency of the incident field and the molecule vibrating with angular frequency, respectively. Thus, the formula can clearly illustrate that increasing the field amplitude of the excitation light can enhance the Raman intensity. Furthermore, the SPRE of metal NCs results in a localized electric filed around the SNCs [21]. It was suggested that the high SERS is obtained, which is relationship with the large electric field produced by SPR on SNCs. The mechanism of SERS with SPR was divided into five parts [20]: (i) the illumination photon at exc and incident angle ωexc is coupled to a metal surface and produces surface plasmon polaritons (SPPs) for excitation. (ii) The SPPs polarize the molecules and produce large dipole moments, which cause enhancement of the excitation source. (iii) The molecular polarizations change, provided that the molecules change vibrational state. (iv) The molecular polarizations change couples energy into SPPs at scattered frequency. (v) Last, the outgoing SPPs decay into radiation at scattered angle RS and radiation at ωRS , rising the emission enhancement. Therefore, the Raman intensity can be enhanced by SPRE. The Raman spectra have the highest intensity when SNCs is 34 nm, which is closest to the excitation laser’s wavelength of SERS. It generates larger localized electric field around the SNCs. Furthermore, the excitation laser’s wavelength of SERS is as good as PL peak of Alq3 . Therefore, it is confirmed that the electron injection is enhanced from the cathode to emitting layer when SNCs are introduced between the LiF and Al cathode because of SPRE. SPRE caused the enhanced localized electric field around the SNCs. 4. Conclusion In summary, the SNCs are introduced between the EIL and Al cathode, and its best performance is obtained when the SNCs is
269
34 nm because the absorption peak of SNCs is closest to the PL peak of Alq3 . Furthermore, SPRE causes the enhanced localized electric field around the SNCs. Therefore, SPRE is the crucial factor to contribute the electron injection being enhanced from the cathode to emitting layer when SNCs are introduced between the LiF and Al cathode. Consequently, the SNCs play dual important roles: (i) It enhanced the spontaneous recombination of excitons due to the subtraction in the exciton lifetime. (ii) It enhanced the electron injection because of SPRE, which induced the enhanced localized electric field around the SNCs. Acknowledgments This work was partially supported by the National Science Council of Taiwan (NSCT) under Contract no. NSC-101-2221-E-390-021. References [1] Y. Zhang, M. Slootsky, S.R. Forrest, Appl. Phys. Lett. 99 (2011) 223303–2233033. [2] J. Lee, J.-I. Lee, J.Y. Lee, H.Y. Chu, Appl. Phys. Lett. 95 (2009) 253304–2533043. [3] H.S. Woo, R. Czerw, S. Webster, D.L. Carroll, J. Ballato, A.E. Strevens, D. O’Brien, W.J. Blau, Appl. Phys. Lett. 77 (2000) 1393–1395. [4] A. Haldi, B. Domercq, B. Kippelen, R.D. Hreha, J.-Y. Cho, S.R. Marder, Appl. Phys. Lett. 92 (2008) 253502–2535023. [5] S.O. Jeon, K.S. Yook, C.W. Joo, J.Y. Lee, Appl. Phys. Lett. 94 (2009) 013301–0133013. [6] Y. Xhao, L. Duan, D. Zhang, L. Hou, J. Qiao, L. Wang, Y. Qiu, Appl. Phys. Lett. 100 (2012) 083304–0833044. [7] K. Kim, K. Hong, I. Lee, S. Kim, J.-L. Lee, Appl. Phys. Lett. 101 (2012) 141102–1411024. [8] T. Earmme, S.A. Jenekhe, Appl. Phys. Lett. 102 (2013) 233305–2333054. [9] Y.W. Park, J.H. Choi, T.H. Park, E.H. Song, H. Kim, H.J. Lee, S.J. Shin, B.-K. Ju, W.J. Song, Appl. Phys. Lett. 100 (2012) 013312–0133124. [10] A. Kumar, R. Srivastava, D.S. Mehta, M.N. Kamalasanan, Org. Electron. 13 (2012) 1750–1755. [11] A. Fujiki, T. Uemura, N. Zettsu, M. Akai-Kasaya, A. Saito, Y. Kuwahara, Appl. Phys. Lett. 96 (2010) 043307–0433073. [12] A. Kumar, R. Srivastave, P. Tyagi, D.S. Mehta, M.N. Kamalasanan, Org. Electron. 13 (2012) 159–165. [13] K.S. Yook, S.O. Jeon, C.W. Joo, J.Y. Lee, Appl. Phys. Lett. 93 (2008) 013301–0133013. [14] C.J. Huang, P.H. Chiu, Y.H. Wang, K.L. Chen, J.J. Linn, C.F. Yang, J. Electrochem. Soc. 153 (2006) D193–D198. [15] A. Kumar, P. Tyagi, R. Srivastava, D.S. Mehta, Appl. Phys. Lett. 102 (2013) 203304. [16] T. Matsushima, M. Takamori, Y. Miyashita, Y. Honma, T. Tanaka, H. Aihara, H. Murata, Org. Electron. 11 (2010) 16–22. [17] S. Naka, H. Okada, H. Onnagawa, Appl. Phys. Lett. 76 (2000) 197–199. [18] W.L. Barnes, A. Dereux, T.W. Ebbesen, Nature 424 (2003) 824–830. [19] K.G. Stamplecoskie, J.C. Scaiano, J. Phys. Chem. C 115 (2011) 1403–1409. [20] C.S.S.R. Kumar, Raman Spectroscopy for Nanomaterials Characterization, Springer, Berlin, 2012. [21] D. Liu, M. Fina, L. Ren, S.S. Mao, Appl. Phys. A 96 (2009) 353–356.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Surface plasmon-enhanced localized electric field in organic light-emitting diodes by incorporating silver nanoclusters Ying-Chung Chen a , Chia-Yuan Gao a , Kan-Lin Chen b , Chien-Jung Huang c,∗ a b c
Department of Electrical Engineering, National Sun Yat-sen University, Kaohsiung, Taiwan Department of Electronic Engineering, Fortune Institute of Technology, Kaohsiung, Taiwan Department of Applied Physics, National University of Kaohsiung, Kaohsiung, Taiwan
a r t i c l e
i n f o
Article history: Received 7 October 2013 Received in revised form 28 November 2013 Accepted 7 January 2014 Available online 15 January 2014 Keywords: Silver nanoclusters Localized electric field Electron injection
a b s t r a c t
The influence of silver nanoclusters (SNCs) on the performance of organic light-emitting diodes is investigated in this study. The SNCs are introduced between the electron-injection layer and cathode alumina by means of thermal evaporation, resulting that different absorption peaks of SNCs were formed. A higher luminance and electron-injection ability are obtained when the mean cluster size is 34 nm. The surfaceenhanced Raman scattering spectroscopy reveals that the localized electric field around the SNCs is enhanced, resulting in an increase in electron injection from cathode electrode. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Organic light-emitting diode (OLED) has displayed significant potentiality in the application of lighting and display, which is mainly because of its low power consumption. So far, the efficiency of device is not optimal, so many researchers are still trying to improve the efficiency of device by different means, e.g., doping phosphorescence material into the emitting layer [1,2], blocking hole with hole-blocking layer [3–6], or improving the hole/electron balance by increasing electron injection [7–9], which is to improve the efficiency of device. The efficiency of OLED can be improved, but these methods increased the complexity of device structure which increased the fabrication time and cost of device. Recently, it has drawn a lot of attention that the surface plasmon resonance effect (SPRE) of metal nanocluster is applied to OLED which is mainly because of surface plasmon-enhanced luminescence of OLED being enhanced. In recent reports on both organic and inorganic light-emitting diodes, the coupling process between surface plasmons and radiated light is rapider than spontaneous recombination of excitons [10,11]. Therefore, the exciton lifetime in surface plasmon resonance (SPR) devices is considerably reduced. Furthermore, spontaneous emission rate of exciton is inversely proportional to the exciton lifetime [12]. In other words, emission efficiency is increased with the subtraction in the
∗ Corresponding author. Tel.: +886 7 5919475; fax: +886 7 5919357. E-mail address: [email protected] (C.-J. Huang). 0169-4332/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apsusc.2014.01.031
exciton lifetime. Besides, a thin silver layer was incorporated as an interlayer between the electron-transport layer and the electrode which can improve the efficiency of OLEDs [13]. However, the cause of improving the efficiency has not yet been studied in detail. In this article, the silver nanoclusters (SNCs) were introduced between electron-injection layer (EIL) and cathode, and it was also observed that how SPR wavelength of different mean cluster sizes of SNCs influenced for the enhancement of device efficiency. In addition, a different viewpoint in surface plasmon-enhanced luminescence by a theory of surface-enhanced Raman scattering (SERS) is presented. 2. Experimental The schematic structure of OLEDs is shown in Fig. 1. Indium tin oxide (ITO) coated on glass with a sheet resistance of 10 /sq was used as the starting substrate. The substrate was cleaned with acetone, methanol and deionized water, and then dried with nitrogen gas. After cleaning process, the substrates were loaded into a thermal evaporator. Afterward, N,N -bis-(1naphthyl)-N,N-dipheny1,1 - bipheny1 -4-diamine (NPB; 35 nm), tris-(8-hydroxyquinoline) aluminum (Alq3 ; 40 nm), 4, 7-dipheny11, 10-phenanthroline (BPhen; 10 nm), lithium fluoride (LiF; 0.5 nm), SNCs (× nm) and aluminum (Al; 100 nm) were deposited. NPB and BPhen were used as the hole and electron transporting layers, respectively. Alq3 was used as emitting layer. LiF, Al and ITO were used as EIL, cathode and anode, respectively. To obtain the absorption peak of SNCs being the closest to the photoluminescence (PL) peak of Alq3 , the mean cluster size of the SNCs was
Y.-C. Chen et al. / Applied Surface Science 295 (2014) 266–269
changed from 15 to 43 nm. Furthermore, the different mean cluster sizes of SNCs were deposited on top of glass substrate in order to measure SERS. During the deposition, the base pressure of the chamber was maintained as low as 2.4 × 10−6 Torr. The active area of the device was 6 × 6 mm2 . The deposition rates of all organic materials were maintained at 0.01 nm/s, except for the Al that deposition rate was 0.5 nm/s. In addition, the SNCs were fabricated by thermal evaporation of silver slug. The deposition rate of SNCs is 0.01 nm/s. Besides, the different mean cluster sizes of SNCs are obtained by different deposition time. The deposition time for the mean particle sizes of 15, 24, 34, 43 nm is 100, 200, 300, and 400 s, respectively. The thickness of the layers is controlled by using a quartz–crystal monitor. The current density–voltage (J–V) characteristics and luminance–voltage (L–V) of the devices were measured by using a Keithley 2400 (Keithley instruments Inc, USA) and a PR-655 (Photo Research Inc, USA), respectively. The SERS and absorption spectrum were measured by using a BWII RAMaker
267
Fig. 1. Schematic device structure of the OLED with SNCs.
Fig. 2. The SEM images of SNCs at (a) 15 nm, (b) 24 nm, (c) 34 nm, and (d) 43 nm. (e) Normalized absorption spectra of the SNCs with different mean cluster sizes and PL spectrum of the Alq3 .
268
Y.-C. Chen et al. / Applied Surface Science 295 (2014) 266–269
1.4
SNCs (15 nm) SNCs (34 nm) SNCs (54 nm)
(a)
Normalized absorption
1.2
444,
1.0
525, 577 (nm)
0.8 0.6 0.4 0.2 0.0 350
400
450
500
550
600
650
700
Wavaelength (nm) 7000
(b)
Intensity (counts/15s)
6000
SNCs (15 nm) SNCs (34 nm) SNCs (54 nm)
5000 4000 3000 2000 0
500
1000
1500
2000
Raman Shift (cm-1)
Fig. 3. (a) Current density–voltage curves and (b) Luminance–voltage curves of OLEDs. Inset of (a) shows the efficiency–voltage characteristics of OLEDs.
(Protrustech, Taiwan) and a U-3900 spectrophotometer (Hitachi, Japan), respectively. The photoluminescence (PL) spectrum and scanning electron microscope (SEM) images were measured by using an IK series (Kimmon Koha, Japan) and a JSM-6700F (JEOL, Japan). All measurements were made at room temperatures under air ambient. 3. Results and discussion Fig. 2(a–d) show the SEM images of SNCs with different mean cluster sizes on glass substrate. The mean cluster sizes of SNCs in Fig. 2(a–d) were calculated and found to be 15, 24, 34, and 43 nm, respectively. It is well known that the absorption peak depends on SNCs size, shape, and spacing between SNCs. The absorption spectra of SNCs with different mean cluster sizes and the PL spectrum of Alq3 are shown in Fig. 2(e). It can be clearly seen that the absorption peaks were red-shift with increase in mean cluster size of SNCs. Similarly, when the mean cluster size of metal NCs is increased, the absorption peaks were red-shift, which is consistent with our previous study [14]. Besides, it is observed that the PL peak of Alq3 is about 523 nm. When the mean cluster size of SNCs is 34 nm, the absorption peak of SNCs is about 525 nm that is closest to the PL peak of Alq3 . Consequently, as is expected, the best performance of device can be obtained when the mean cluster size of SNCs is 34 nm. The resulting current density–voltage curves are plotted in Fig. 3(a). The current density of the device with SNCs is remarkably increased because the localized electric field around the SNCs is enhanced, which results in an increase in electron injection from cathode electrode. In addition, the spacing between the emissive layer and SNCs is necessary because the spacing can avoid the nonradiative quenching of exciton [15]. The current efficiency–voltage curves are plotted in the inset of Fig. 3(a). The current efficiency of the device with SNCs at 34 nm is quite high compared with that of the device without SNCs. Generally, the mobility of electrons in
Fig. 4. (a) Normalized absorption spectra of the SNCs with different mean cluster sizes. (b) SERS spectrum at excitation laser’s wavelength (532 nm) for SNCs with different mean cluster sizes.
the electron transporting layer is a few orders of magnitude lower than that of holes in the hole transporting layer [16,17], resulting in the imbalance between holes and electrons within the emitting layer. Therefore, the current efficiency of the devices with SNCs are higher than that of the device without SNCs because of enhancement in electron injection from the cathode to emitting layer, resulting in better balance for the ratio of hole/electron injection. Fig. 3(b) shows the L–V characteristic of the device with different sizes of SNCs and without SNCs. The luminance of devices with SNCs is higher than that of the device without SNCs due to its surface plasmon-enhanced spontaneous emission rate. It is well known that the SPRE was induced by interaction between the surface charge of metal and the electromagnetic field of the incident light [18]. On the other hand, luminance depends on the current density and current efficiency. A high current density and high current efficiency will benefit the high luminance of device. Consequently, the maximum luminance of 14070 cd/m2 is obtained when the SNCs is 34 nm. Furthermore, the result is consistent with what we previously extrapolated. That is to say, when the SNCs were introduced between the LiF and Al cathode, the SNCs can generate the SPRE, which can not only enhance the spontaneous emission rate but also improve the balance for the ratio of hole/electron injection. However, the enhanced electron injection is attributed to SPRE which induces the enhanced localized electric field around the SNCs. Fig. 4(a) shows the absorption spectra of the SNCs with different mean cluster sizes. It is observed that the peak wavelength is 525 nm for SNCs of 34 nm, which is closest to the excitation laser’s wavelength (532 nm) of SERS. In addition, when the size of the SNCs becomes bigger, it was found that the absorbance peak obviously shifts from the short wavelength to the long wavelength, and the full-width at half-maximum (FWHM) is from narrow into wide. The widening and red shift for the larger SNCs are because of an increasing contribution of higher-order plasmon modes for the larger SNCs [19]. To further confirm the relationship between the electric field and the SNCs, the SNCs were deposited on top of glass substrate to
Y.-C. Chen et al. / Applied Surface Science 295 (2014) 266–269
measure SERS. The SERS data obtained from different mean cluster sizes of SNCs are shown in Fig. 4(b). The spectral feature characteristics of SNCs were about 1260, 1348, and 1555 cm−1 . As anticipated, Raman spectrum in SNCs of 34 nm shows the highest intensity due to the fact that the peak wavelength in SNCs of 34 nm is closest to the excitation laser’s wavelength of SERS, resulting in large electric field around the SNCs. The Raman intensity is proportional to the square of the incident field. The relationship between the Raman intensity and the incident field can be expressed as follows [20]:
IRS ∝
∂˛ E0 [cos (ωo − ωv ) t + cos (ωo + ωv ) t] ∂Q
2
(1)
Where IRS is the Raman intensity, ˛ is the molecular polarizability, Q is the nuclear displacement, E0 is the amplitude of the incident field, and ωo and ωv are the angular frequency of the incident field and the molecule vibrating with angular frequency, respectively. Thus, the formula can clearly illustrate that increasing the field amplitude of the excitation light can enhance the Raman intensity. Furthermore, the SPRE of metal NCs results in a localized electric filed around the SNCs [21]. It was suggested that the high SERS is obtained, which is relationship with the large electric field produced by SPR on SNCs. The mechanism of SERS with SPR was divided into five parts [20]: (i) the illumination photon at exc and incident angle ωexc is coupled to a metal surface and produces surface plasmon polaritons (SPPs) for excitation. (ii) The SPPs polarize the molecules and produce large dipole moments, which cause enhancement of the excitation source. (iii) The molecular polarizations change, provided that the molecules change vibrational state. (iv) The molecular polarizations change couples energy into SPPs at scattered frequency. (v) Last, the outgoing SPPs decay into radiation at scattered angle RS and radiation at ωRS , rising the emission enhancement. Therefore, the Raman intensity can be enhanced by SPRE. The Raman spectra have the highest intensity when SNCs is 34 nm, which is closest to the excitation laser’s wavelength of SERS. It generates larger localized electric field around the SNCs. Furthermore, the excitation laser’s wavelength of SERS is as good as PL peak of Alq3 . Therefore, it is confirmed that the electron injection is enhanced from the cathode to emitting layer when SNCs are introduced between the LiF and Al cathode because of SPRE. SPRE caused the enhanced localized electric field around the SNCs. 4. Conclusion In summary, the SNCs are introduced between the EIL and Al cathode, and its best performance is obtained when the SNCs is
269
34 nm because the absorption peak of SNCs is closest to the PL peak of Alq3 . Furthermore, SPRE causes the enhanced localized electric field around the SNCs. Therefore, SPRE is the crucial factor to contribute the electron injection being enhanced from the cathode to emitting layer when SNCs are introduced between the LiF and Al cathode. Consequently, the SNCs play dual important roles: (i) It enhanced the spontaneous recombination of excitons due to the subtraction in the exciton lifetime. (ii) It enhanced the electron injection because of SPRE, which induced the enhanced localized electric field around the SNCs. Acknowledgments This work was partially supported by the National Science Council of Taiwan (NSCT) under Contract no. NSC-101-2221-E-390-021. References [1] Y. Zhang, M. Slootsky, S.R. Forrest, Appl. Phys. Lett. 99 (2011) 223303–2233033. [2] J. Lee, J.-I. Lee, J.Y. Lee, H.Y. Chu, Appl. Phys. Lett. 95 (2009) 253304–2533043. [3] H.S. Woo, R. Czerw, S. Webster, D.L. Carroll, J. Ballato, A.E. Strevens, D. O’Brien, W.J. Blau, Appl. Phys. Lett. 77 (2000) 1393–1395. [4] A. Haldi, B. Domercq, B. Kippelen, R.D. Hreha, J.-Y. Cho, S.R. Marder, Appl. Phys. Lett. 92 (2008) 253502–2535023. [5] S.O. Jeon, K.S. Yook, C.W. Joo, J.Y. Lee, Appl. Phys. Lett. 94 (2009) 013301–0133013. [6] Y. Xhao, L. Duan, D. Zhang, L. Hou, J. Qiao, L. Wang, Y. Qiu, Appl. Phys. Lett. 100 (2012) 083304–0833044. [7] K. Kim, K. Hong, I. Lee, S. Kim, J.-L. Lee, Appl. Phys. Lett. 101 (2012) 141102–1411024. [8] T. Earmme, S.A. Jenekhe, Appl. Phys. Lett. 102 (2013) 233305–2333054. [9] Y.W. Park, J.H. Choi, T.H. Park, E.H. Song, H. Kim, H.J. Lee, S.J. Shin, B.-K. Ju, W.J. Song, Appl. Phys. Lett. 100 (2012) 013312–0133124. [10] A. Kumar, R. Srivastava, D.S. Mehta, M.N. Kamalasanan, Org. Electron. 13 (2012) 1750–1755. [11] A. Fujiki, T. Uemura, N. Zettsu, M. Akai-Kasaya, A. Saito, Y. Kuwahara, Appl. Phys. Lett. 96 (2010) 043307–0433073. [12] A. Kumar, R. Srivastave, P. Tyagi, D.S. Mehta, M.N. Kamalasanan, Org. Electron. 13 (2012) 159–165. [13] K.S. Yook, S.O. Jeon, C.W. Joo, J.Y. Lee, Appl. Phys. Lett. 93 (2008) 013301–0133013. [14] C.J. Huang, P.H. Chiu, Y.H. Wang, K.L. Chen, J.J. Linn, C.F. Yang, J. Electrochem. Soc. 153 (2006) D193–D198. [15] A. Kumar, P. Tyagi, R. Srivastava, D.S. Mehta, Appl. Phys. Lett. 102 (2013) 203304. [16] T. Matsushima, M. Takamori, Y. Miyashita, Y. Honma, T. Tanaka, H. Aihara, H. Murata, Org. Electron. 11 (2010) 16–22. [17] S. Naka, H. Okada, H. Onnagawa, Appl. Phys. Lett. 76 (2000) 197–199. [18] W.L. Barnes, A. Dereux, T.W. Ebbesen, Nature 424 (2003) 824–830. [19] K.G. Stamplecoskie, J.C. Scaiano, J. Phys. Chem. C 115 (2011) 1403–1409. [20] C.S.S.R. Kumar, Raman Spectroscopy for Nanomaterials Characterization, Springer, Berlin, 2012. [21] D. Liu, M. Fina, L. Ren, S.S. Mao, Appl. Phys. A 96 (2009) 353–356.