- Email: [email protected]
Ultra-bright alternating current organic electroluminescence
Organic Electronics 13 (2012) 1589–1593
Contents lists available at SciVerse ScienceDirect
Organic Electronics journal homepage: www.elsevier.com/locate/orgel
Ultra-bright alternating current organic electroluminescence Ajay Perumal ⇑, Björn Lüssem, Karl Leo Institut für Angewandte Photophysik,1 Technische Universität Dresden, D-01062 Dresden, Germany
a r t i c l e
i n f o
Article history: Received 23 December 2011 Received in revised form 2 April 2012 Accepted 18 April 2012 Available online 17 May 2012 Keywords: Organic small molecules Molecular doping Charge transport Electroluminescence Alternating current
a b s t r a c t We report on an alternating current (AC) field induced organic electroluminescence (EL) device with internal charge carrier generation and recombination luminance of over 5000 cd m2 under AC drive without charge carrier injection from external electrodes. The ultra-bright AC-EL is attributed to an optical optimization performed on the devices via numerical optical simulations based on an optical thin film model as well as an increase in the number of charge carriers achieved via the concept of molecular doping within the device. The luminance levels achieved are highest reported so far in literature for AC organic light emitting devices. Ó 2012 Elsevier B.V. All rights reserved.
The general lighting and signage market needs alternative and efficient large area illumination panels. As all the buildings and houses are currently wired for AC power (110/220 V and 50/60 Hz), it is advantageous to operate the lighting and illumination modules directly from AC power lines. Several efforts have been proposed in the past for AC-EL devices [1–8]. None of the above mentioned approaches were successful in achieving high brightness levels required for lighting applications. Recently we have reported on high brightness AC field induced organic electroluminescence with controlled internal charge carrier generation within an organic inorganic hybrid device [9,10]. With an organic light emitting layer embedded between insulating dielectric layers, we reached a luminance of over 1500 cd m2. Our approach is encouraging to work towards potential lighting panels which can operate directly from the AC power lines without the need for intermediate power converter or the expensive back end electronics. In this report, we demonstrate ultra-bright AC-EL achieved as a result of an optical optimization performed
1
http://www.iapp.de.
⇑ Corresponding author. E-mail addresses: [email protected] (A. Perumal), [email protected] (K. Leo). 1566-1199/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.orgel.2012.04.024
on the previously reported device [10] via numerical optical simulations based on an optical thin film model [11]. Furthermore, we enhance number of charge carriers within the device achieved by increasing the thickness of molecular doped charge transport layers. The optical optimization of the device is accomplished by varying and adjusting the thickness of the molecular doped charge transport layers so that the emission zone is placed at the optical cavity maxima of the electromagnetic (EM) field. By increasing the thickness of the molecular doped charge transport layers the coupling of the emitting dipole to the evanescent waves between the metallic electrode and the organic dielectric media is reduced. It is known that when an emitting dipole is placed close to the metallic electrode, this coupling becomes prominent and is strongly dependent on the distance between the emitting molecule and the metallic electrode [12]. Also by increasing the thickness of doped charge transport layers it is possible to enhance the number of free charge carriers available within the device. By doing so we are able to achieve very high luminance of over 5000 cd m2 at a low threshold voltage of 10 V for the onset of luminance. The luminance levels achieved are highest reported so far in literature for either organic or inorganic AC light emitting devices. The optical model [11] used for optical optimization of the device treats the radiative molecules as point dipoles
1590
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
which are approximated as damped harmonic oscillators within the classical limit, the model does not take into account quantum physical effects. Transfer matrix formalism is used to calculate the electromagnetic (EM) field profiles of the planar device. The numerical model calculates the emission of the device within a given solid angle X, in a direction forming an angle h between the surface normal to the device and the line of sight. The luminous flux/ power per unit area of the device is calculated by integrating the luminous intensity per unit area I(h) with respect to the given solid angle in the forward hemisphere,
LF ¼
Z
IðhÞ dX X
LF ¼ 2p
Z
ð1Þ
p 2
IðhÞ sinðhÞ dh
ð2Þ
0
where h is the viewing angle between the surface normal to the device and the line of sight. The device structures are shown in Fig. 1. The devices are prepared on a pre-coated and structured glass substrate with indium tin oxide (ITO) film (90 nm). The hafnium oxide (HfO2) insulating layers are coated via planar radio frequency (RF) magnetron sputtering with 100 W RF power. The organic and metal layers are deposited via thermal vapor deposition process. The devices are processed in a cluster tool under ultra high-vacuum (UHV) conditions. The device area is 6.7 mm2. Device A consists of an indium tin oxide (ITO)/HfO2 (60 nm)/20 nm p-doped hole transporting layer (HTL)/organic light emission layer (EML) (20 nm)/ 20 nm n-doped electron transporting layer (ETL)/HfO2 (60 nm)/aluminum (Al) (100 nm). The EML consists of unipolar/ambipolar host in this case alpha-NPD (N,N0 -di(naphthalen-1-yl)-N,N0 -diphenyl-benzidine) doped with 10 wt.% of orange phosphorescent dye dopant Ir(MDQ)2(acac) (bis(2-methyldibenzo-[f,h]quinoxaline)(acetylace tonate) iridium(III)). The device also consists of 10 nm thin layers of alpha-NPD and BPhen (4,7-diphenyl-1,10-phenanthroline) layers, which confine the charge carriers to the EML by acting as electron and hole blocking layers. Device B is similar to device A except 90 nm p-doped and 165 nm ndoped charge transport layers. MeO-TPD (N,N,N0 ,N0 tetrakis(4-methoxyphenyl)-benzidine) doped with F4TCNQ (2,3,5,6-tetrafluoro-7,7,8,8-tetracyanoquinodimethane) acts as p-doped HTL while BPhen doped with cesium (Cs) Fig. 2. (a) Simulated far field outcoupled luminous flux has a strong influence on the the doped HTL and ETL layer thickness variations. (b) ACEL spectra of devices A and B along with the PL emission of the emitter molecule Ir(MDQ)2(acac). (c) The emission affinity (the EM field profile of the device without emitter in place) as a function of wavelength.
Fig. 1. Device structures.
acts as n-doped ETL. The charge transport layers act as charge transport cum charge generation layers. The AC current–voltage–luminance (I–V–L) characteristics of the devices are measured using an automated set up consisting of arbitrary signal generator coupled to bipolar power amplifier, an AC power meter (ZES Zimmer GmbH) and a luminance meter (KonicaMinolta CS100A). The angular-dependent AC-electroluminescence (EL) spectra are measured with a custom built spectrogoniometer consisting
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
of rotational sample holder. The AC-EL is coupled to calibrated USB-2000 mini spectrometer (Ocean Optics) via optical fiber for recording the spectra. We compare the performance of a non optimized device (device A) with an optically optimized device (device B). The molecularly doped HTL and ETL layer thickness variation has a strong influence on the luminous flux as shown in Fig. 2(a). For device A with 20 nm molecularly doped HTL and ETL thickness, the luminous flux is not optimal while for device B with 90 nm HTL and 165 nm ETL thickness the luminous flux is optimal and shows the maximum value. The AC-EL spectral emission of the devices A and B are shown in Fig. 2(b). As reference, the photoluminescence (PL) emission spectrum of the iridium based phosphorescent emitter Ir(MDQ)2(acac) is also shown. Although the AC-EL spectra for both devices peak at 610 nm, the spectral shape for the non optimized device A is slightly different. Especially at the full width half maximum position, it spreads beyond the PL spectra of the emitter molecule, which is usually observed for devices with non optimized cavity. For the optically optimized device the AC-EL spectra are in accordance with the PL spectra of the emitter molecule. The altered AC-EL spectra for device A is due to a weak cavity effect as the emitter molecule is not placed at the EM field maxima of the cavity as evident from the plot of emission affinity (quantity which describes for which wavelength the cavity emission is maximized) shown in
1591
Fig. 2(c). For the optimized device the emitter molecule is placed at the EM field maxima of the cavity and emission is in accordance with the PL emission of the emitter molecule. Fig. 3(a) shows the luminance voltage plot for device A and device B. The optically optimized device B shows very high luminance levels close to 5000 cd m2 at 31 V and 10 kHz (2500 cd m2 at 23 V and 10 kHz). For the non optimized device, we achieve maximum luminance of 1581 cd m2 at 23 V and 10 kHz. Although the molecularly doped ETL and HTL thickness have huge difference for both the devices, the threshold voltage for the onset of luminance for the device B remains approximately the same or even lower. This clearly demonstrates the addition of molecularly doped charge transport layers do not alter the electrical property of the device due to their high conductivity. The threshold voltage for the onset of luminance would shift to higher voltage for the device B in comparison to device A, if the molecularly doped layers had a large resistance. The DC injection currents for both the devices are of same magnitude and DC luminance levels are also of same order for both the devices (Supplementary information, Fig. S1). With the intention not to change the electrical properties of the device, we have chosen the same thickness of insulating layers for both the devices. For a capacitive device, the applied AC voltage lags behind the current flowing through the device by 90°. Fig. 3(b) is a plot of phase angle (between the AC voltage
Fig. 3. (a) Luminance–voltage plot (L–V) at 10 kHz frequency for the devices A and B. (b) Phase variation between the applied AC voltage and the current with respect to the AC voltage at a frequency of 10 kHz. (c) Luminance–frequency plot (L–F) at a fixed AC voltage (20 V for devices A and B). (d) The plot of AC current and impedance variation with respect to the AC frequency. Solid line in the plot corresponds to device A and dashed line corresponds to device B.
1592
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
Fig. 4. (a) The plot of the experimental (solid line) and the simulated (dash dotted line) angular AC EL power spectra for the device B. Both the simulated and the experimental power spectra are normalized with respect to normal emission of the device (0° emission). (b) The luminance intensity and intern the luminance corrected to angular emission. (c) The luminous efficacy of the device plotted as a function of AC voltage at a frequency of 10 kHz. (d) The lifetime of the device plotted as a function of operation time. The device is operated at a constant AC voltage of 20.80 V and current of 6.85 mA (Supplementary information, Fig. S2) during the lifetime measurement.
and the current) versus AC voltage applied to the device. The phase angle is close to 90° at low AC voltages and drops at the onset of luminance, which is correlated to the steep increase in luminance as seen in Fig. 3(a) and (b). The drop in phase can be seen as an indication for the resistive contribution of the organic layers, which is in accordance with the previously proposed [9] device operation model. The applied AC voltage divides across the device depending on the capacitances of the top and bottom insulating dielectric layers (Cit and Cib), and the organic layer (C0). Above the threshold voltage for the onset of luminance, due to the charge carrier generation and movement of the charge carriers the organic layer becomes conductive and the capacitive contribution from the organic layer (C0) drops and the capacitance contribution of insulating layers dominates. The luminance saturation also corresponds to the phase saturation for higher voltages. The luminance and phase saturation indicates there are limited number of charge carriers within the device. Fig. 3(c) shows the luminance-frequency variation for devices A and B. The increase of luminance observed as a function of frequency is related to impedance or capacitive reactance of the device. The capacitive reactance of the device is inversely proportional to frequency. As the frequency increases, the capacitive reactance decreases and the current flowing through the device increases as shown in Fig. 3(d). Hence, more charge carriers are available for
recombination and this results in more luminance output. The AC impedance for device A (solid line) is higher in comparison to device B (dashed line) as shown in Fig. 3(d). Although the molecularly doped HTL and ETL thickness is much higher for device B, the impedance is lower in comparison to device A. This clearly indicates the thicker doped layer is more conductive and has more free charge carriers resulting in the flow of higher currents in the device B. This leads to higher luminance in device B in comparison to device A, which is in agreement with the L–V curve of the devices. The numerical optical model as well predicts the angle dependent power spectra emitted per unit solid angle by the device. Fig. 4(a) shows the comparison of the experimental (solid line) and simulated (dash dotted line) angle dependent emission spectra for the optically optimized device B. There is good agreement between the experimental and simulated angular emission patterns clearly indicating the validity of the numerical optical simulation model. The spectral positions have a minor blue shift (10 nm) even for angles as high as 70° resulting in stable color coordinates with respect to viewing angle. The normalized luminous intensity for different viewing angles shows a superLambertian behavior of the device emission as shown in Fig. 4(b). The luminous efficacy (ratio of luminous flux to power input) for device B is plotted as a function of the AC voltage is shown in Fig. 4(c). We achieve a maximum
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
luminous efficacy of close to 1 lm W1 (luminous efficacy values are corrected to non-Lambertian emission). Finally, the stability of the device is crucial and Fig. 4(d) shows the lifetime plot of the device B over the operation time. The lifetime is defined as the time interval where in the set initial luminance drops to 50% of its initial value, when the device is operated at a constant AC voltage and current. The device shows an adequate stability with life time of 100 h with initial luminance set at 500 cd m2. The device is operated at a constant AC voltage of 20.80 V and current of 6.85 mA (Supplementary information, Fig. S2) during the lifetime measurement. The initial increase observed in relative luminance with operation time could be due to better charge carrier transport during the initial hours of operation. In conclusion, we demonstrate very bright AC organic electroluminescence of well over 5000 cd m2 with a luminous efficacy close to 1 lm W1. We also demonstrate utilization of numerical optical simulation model to optically optimize the device and thereby enhancing the luminance output and the device performance significantly. The high luminance values achieved for the device combined with the luminous efficacy and adequate device stability is encouraging to work towards potential future lighting panels which can operate directly from AC power lines. Acknowledgment The authors thank Dr. Mauro Furno for the optical simulation tool SIM-OLED.
1593
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.orgel.2012.04.024. References [1] X.L. Xu, X.H. Chen, Y.B. Hou, Z. Xu, X.H. Yang, S.G. Yin, Z.J. Wang, X.R. Xu, S.P. Lau, B.K. Tay, Chem. Phys. Lett. 325 (2000) 420. [2] S.Y. Yang, L. Qian, F. Teng, Z. Xu, X.R. Xu, J. Appl. Phys. 97 (2005) 126101. [3] T. Tsutsui, S.B. Lee, K. Fujita, Appl. Phys. Lett. 85 (2004) 2382. [4] S.B. Lee, K. Fujita, T. Tsutsui, Jpn. J. Appl. Phys. 44 (2005) 6607. [5] D. Adachi, S. Hasui, T. Toyama, H. Okamoto, Appl. Phys. Lett. 77 (2000) 1301. [6] P.D. Rack, P.H. Holloway, Mater. Sci. Eng. R 21 (1998) 171. [7] J.C. Hitt, J.P. Bender, J.F. Wager, CRC Crit. Rev. Solid State Mater. Sci. 25 (2000) 29. [8] V. Wood, J.E. Halpert, M.J. Panzer, M.G. Bawendi, V. Bulovic, Nano Lett. 9 (2009) 2367. [9] A. Perumal, M. Fröbel, S. Gorantla, T. Gemming, B. Lüssem, J. Eckert, K. Leo, Adv. Funct. Mater. 22 (2012) 210. [10] A. Perumal, B. Lüssem, K. Leo, Appl. Phys. Lett. 100 (2012) 103307. [11] M. Furno, R. Meerheim, M. Thomschke, S. Hofmann, B. Lüssem, K. Leo, Proc. SPIE 7617 (2010) 761716. [12] H. Becker, S.E. Burns, R.H. Friend, Phys. Rev. B 56 (1997) 1893.
Contents lists available at SciVerse ScienceDirect
Organic Electronics journal homepage: www.elsevier.com/locate/orgel
Ultra-bright alternating current organic electroluminescence Ajay Perumal ⇑, Björn Lüssem, Karl Leo Institut für Angewandte Photophysik,1 Technische Universität Dresden, D-01062 Dresden, Germany
a r t i c l e
i n f o
Article history: Received 23 December 2011 Received in revised form 2 April 2012 Accepted 18 April 2012 Available online 17 May 2012 Keywords: Organic small molecules Molecular doping Charge transport Electroluminescence Alternating current
a b s t r a c t We report on an alternating current (AC) field induced organic electroluminescence (EL) device with internal charge carrier generation and recombination luminance of over 5000 cd m2 under AC drive without charge carrier injection from external electrodes. The ultra-bright AC-EL is attributed to an optical optimization performed on the devices via numerical optical simulations based on an optical thin film model as well as an increase in the number of charge carriers achieved via the concept of molecular doping within the device. The luminance levels achieved are highest reported so far in literature for AC organic light emitting devices. Ó 2012 Elsevier B.V. All rights reserved.
The general lighting and signage market needs alternative and efficient large area illumination panels. As all the buildings and houses are currently wired for AC power (110/220 V and 50/60 Hz), it is advantageous to operate the lighting and illumination modules directly from AC power lines. Several efforts have been proposed in the past for AC-EL devices [1–8]. None of the above mentioned approaches were successful in achieving high brightness levels required for lighting applications. Recently we have reported on high brightness AC field induced organic electroluminescence with controlled internal charge carrier generation within an organic inorganic hybrid device [9,10]. With an organic light emitting layer embedded between insulating dielectric layers, we reached a luminance of over 1500 cd m2. Our approach is encouraging to work towards potential lighting panels which can operate directly from the AC power lines without the need for intermediate power converter or the expensive back end electronics. In this report, we demonstrate ultra-bright AC-EL achieved as a result of an optical optimization performed
1
http://www.iapp.de.
⇑ Corresponding author. E-mail addresses: [email protected] (A. Perumal), [email protected] (K. Leo). 1566-1199/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.orgel.2012.04.024
on the previously reported device [10] via numerical optical simulations based on an optical thin film model [11]. Furthermore, we enhance number of charge carriers within the device achieved by increasing the thickness of molecular doped charge transport layers. The optical optimization of the device is accomplished by varying and adjusting the thickness of the molecular doped charge transport layers so that the emission zone is placed at the optical cavity maxima of the electromagnetic (EM) field. By increasing the thickness of the molecular doped charge transport layers the coupling of the emitting dipole to the evanescent waves between the metallic electrode and the organic dielectric media is reduced. It is known that when an emitting dipole is placed close to the metallic electrode, this coupling becomes prominent and is strongly dependent on the distance between the emitting molecule and the metallic electrode [12]. Also by increasing the thickness of doped charge transport layers it is possible to enhance the number of free charge carriers available within the device. By doing so we are able to achieve very high luminance of over 5000 cd m2 at a low threshold voltage of 10 V for the onset of luminance. The luminance levels achieved are highest reported so far in literature for either organic or inorganic AC light emitting devices. The optical model [11] used for optical optimization of the device treats the radiative molecules as point dipoles
1590
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
which are approximated as damped harmonic oscillators within the classical limit, the model does not take into account quantum physical effects. Transfer matrix formalism is used to calculate the electromagnetic (EM) field profiles of the planar device. The numerical model calculates the emission of the device within a given solid angle X, in a direction forming an angle h between the surface normal to the device and the line of sight. The luminous flux/ power per unit area of the device is calculated by integrating the luminous intensity per unit area I(h) with respect to the given solid angle in the forward hemisphere,
LF ¼
Z
IðhÞ dX X
LF ¼ 2p
Z
ð1Þ
p 2
IðhÞ sinðhÞ dh
ð2Þ
0
where h is the viewing angle between the surface normal to the device and the line of sight. The device structures are shown in Fig. 1. The devices are prepared on a pre-coated and structured glass substrate with indium tin oxide (ITO) film (90 nm). The hafnium oxide (HfO2) insulating layers are coated via planar radio frequency (RF) magnetron sputtering with 100 W RF power. The organic and metal layers are deposited via thermal vapor deposition process. The devices are processed in a cluster tool under ultra high-vacuum (UHV) conditions. The device area is 6.7 mm2. Device A consists of an indium tin oxide (ITO)/HfO2 (60 nm)/20 nm p-doped hole transporting layer (HTL)/organic light emission layer (EML) (20 nm)/ 20 nm n-doped electron transporting layer (ETL)/HfO2 (60 nm)/aluminum (Al) (100 nm). The EML consists of unipolar/ambipolar host in this case alpha-NPD (N,N0 -di(naphthalen-1-yl)-N,N0 -diphenyl-benzidine) doped with 10 wt.% of orange phosphorescent dye dopant Ir(MDQ)2(acac) (bis(2-methyldibenzo-[f,h]quinoxaline)(acetylace tonate) iridium(III)). The device also consists of 10 nm thin layers of alpha-NPD and BPhen (4,7-diphenyl-1,10-phenanthroline) layers, which confine the charge carriers to the EML by acting as electron and hole blocking layers. Device B is similar to device A except 90 nm p-doped and 165 nm ndoped charge transport layers. MeO-TPD (N,N,N0 ,N0 tetrakis(4-methoxyphenyl)-benzidine) doped with F4TCNQ (2,3,5,6-tetrafluoro-7,7,8,8-tetracyanoquinodimethane) acts as p-doped HTL while BPhen doped with cesium (Cs) Fig. 2. (a) Simulated far field outcoupled luminous flux has a strong influence on the the doped HTL and ETL layer thickness variations. (b) ACEL spectra of devices A and B along with the PL emission of the emitter molecule Ir(MDQ)2(acac). (c) The emission affinity (the EM field profile of the device without emitter in place) as a function of wavelength.
Fig. 1. Device structures.
acts as n-doped ETL. The charge transport layers act as charge transport cum charge generation layers. The AC current–voltage–luminance (I–V–L) characteristics of the devices are measured using an automated set up consisting of arbitrary signal generator coupled to bipolar power amplifier, an AC power meter (ZES Zimmer GmbH) and a luminance meter (KonicaMinolta CS100A). The angular-dependent AC-electroluminescence (EL) spectra are measured with a custom built spectrogoniometer consisting
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
of rotational sample holder. The AC-EL is coupled to calibrated USB-2000 mini spectrometer (Ocean Optics) via optical fiber for recording the spectra. We compare the performance of a non optimized device (device A) with an optically optimized device (device B). The molecularly doped HTL and ETL layer thickness variation has a strong influence on the luminous flux as shown in Fig. 2(a). For device A with 20 nm molecularly doped HTL and ETL thickness, the luminous flux is not optimal while for device B with 90 nm HTL and 165 nm ETL thickness the luminous flux is optimal and shows the maximum value. The AC-EL spectral emission of the devices A and B are shown in Fig. 2(b). As reference, the photoluminescence (PL) emission spectrum of the iridium based phosphorescent emitter Ir(MDQ)2(acac) is also shown. Although the AC-EL spectra for both devices peak at 610 nm, the spectral shape for the non optimized device A is slightly different. Especially at the full width half maximum position, it spreads beyond the PL spectra of the emitter molecule, which is usually observed for devices with non optimized cavity. For the optically optimized device the AC-EL spectra are in accordance with the PL spectra of the emitter molecule. The altered AC-EL spectra for device A is due to a weak cavity effect as the emitter molecule is not placed at the EM field maxima of the cavity as evident from the plot of emission affinity (quantity which describes for which wavelength the cavity emission is maximized) shown in
1591
Fig. 2(c). For the optimized device the emitter molecule is placed at the EM field maxima of the cavity and emission is in accordance with the PL emission of the emitter molecule. Fig. 3(a) shows the luminance voltage plot for device A and device B. The optically optimized device B shows very high luminance levels close to 5000 cd m2 at 31 V and 10 kHz (2500 cd m2 at 23 V and 10 kHz). For the non optimized device, we achieve maximum luminance of 1581 cd m2 at 23 V and 10 kHz. Although the molecularly doped ETL and HTL thickness have huge difference for both the devices, the threshold voltage for the onset of luminance for the device B remains approximately the same or even lower. This clearly demonstrates the addition of molecularly doped charge transport layers do not alter the electrical property of the device due to their high conductivity. The threshold voltage for the onset of luminance would shift to higher voltage for the device B in comparison to device A, if the molecularly doped layers had a large resistance. The DC injection currents for both the devices are of same magnitude and DC luminance levels are also of same order for both the devices (Supplementary information, Fig. S1). With the intention not to change the electrical properties of the device, we have chosen the same thickness of insulating layers for both the devices. For a capacitive device, the applied AC voltage lags behind the current flowing through the device by 90°. Fig. 3(b) is a plot of phase angle (between the AC voltage
Fig. 3. (a) Luminance–voltage plot (L–V) at 10 kHz frequency for the devices A and B. (b) Phase variation between the applied AC voltage and the current with respect to the AC voltage at a frequency of 10 kHz. (c) Luminance–frequency plot (L–F) at a fixed AC voltage (20 V for devices A and B). (d) The plot of AC current and impedance variation with respect to the AC frequency. Solid line in the plot corresponds to device A and dashed line corresponds to device B.
1592
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
Fig. 4. (a) The plot of the experimental (solid line) and the simulated (dash dotted line) angular AC EL power spectra for the device B. Both the simulated and the experimental power spectra are normalized with respect to normal emission of the device (0° emission). (b) The luminance intensity and intern the luminance corrected to angular emission. (c) The luminous efficacy of the device plotted as a function of AC voltage at a frequency of 10 kHz. (d) The lifetime of the device plotted as a function of operation time. The device is operated at a constant AC voltage of 20.80 V and current of 6.85 mA (Supplementary information, Fig. S2) during the lifetime measurement.
and the current) versus AC voltage applied to the device. The phase angle is close to 90° at low AC voltages and drops at the onset of luminance, which is correlated to the steep increase in luminance as seen in Fig. 3(a) and (b). The drop in phase can be seen as an indication for the resistive contribution of the organic layers, which is in accordance with the previously proposed [9] device operation model. The applied AC voltage divides across the device depending on the capacitances of the top and bottom insulating dielectric layers (Cit and Cib), and the organic layer (C0). Above the threshold voltage for the onset of luminance, due to the charge carrier generation and movement of the charge carriers the organic layer becomes conductive and the capacitive contribution from the organic layer (C0) drops and the capacitance contribution of insulating layers dominates. The luminance saturation also corresponds to the phase saturation for higher voltages. The luminance and phase saturation indicates there are limited number of charge carriers within the device. Fig. 3(c) shows the luminance-frequency variation for devices A and B. The increase of luminance observed as a function of frequency is related to impedance or capacitive reactance of the device. The capacitive reactance of the device is inversely proportional to frequency. As the frequency increases, the capacitive reactance decreases and the current flowing through the device increases as shown in Fig. 3(d). Hence, more charge carriers are available for
recombination and this results in more luminance output. The AC impedance for device A (solid line) is higher in comparison to device B (dashed line) as shown in Fig. 3(d). Although the molecularly doped HTL and ETL thickness is much higher for device B, the impedance is lower in comparison to device A. This clearly indicates the thicker doped layer is more conductive and has more free charge carriers resulting in the flow of higher currents in the device B. This leads to higher luminance in device B in comparison to device A, which is in agreement with the L–V curve of the devices. The numerical optical model as well predicts the angle dependent power spectra emitted per unit solid angle by the device. Fig. 4(a) shows the comparison of the experimental (solid line) and simulated (dash dotted line) angle dependent emission spectra for the optically optimized device B. There is good agreement between the experimental and simulated angular emission patterns clearly indicating the validity of the numerical optical simulation model. The spectral positions have a minor blue shift (10 nm) even for angles as high as 70° resulting in stable color coordinates with respect to viewing angle. The normalized luminous intensity for different viewing angles shows a superLambertian behavior of the device emission as shown in Fig. 4(b). The luminous efficacy (ratio of luminous flux to power input) for device B is plotted as a function of the AC voltage is shown in Fig. 4(c). We achieve a maximum
A. Perumal et al. / Organic Electronics 13 (2012) 1589–1593
luminous efficacy of close to 1 lm W1 (luminous efficacy values are corrected to non-Lambertian emission). Finally, the stability of the device is crucial and Fig. 4(d) shows the lifetime plot of the device B over the operation time. The lifetime is defined as the time interval where in the set initial luminance drops to 50% of its initial value, when the device is operated at a constant AC voltage and current. The device shows an adequate stability with life time of 100 h with initial luminance set at 500 cd m2. The device is operated at a constant AC voltage of 20.80 V and current of 6.85 mA (Supplementary information, Fig. S2) during the lifetime measurement. The initial increase observed in relative luminance with operation time could be due to better charge carrier transport during the initial hours of operation. In conclusion, we demonstrate very bright AC organic electroluminescence of well over 5000 cd m2 with a luminous efficacy close to 1 lm W1. We also demonstrate utilization of numerical optical simulation model to optically optimize the device and thereby enhancing the luminance output and the device performance significantly. The high luminance values achieved for the device combined with the luminous efficacy and adequate device stability is encouraging to work towards potential future lighting panels which can operate directly from AC power lines. Acknowledgment The authors thank Dr. Mauro Furno for the optical simulation tool SIM-OLED.
1593
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.orgel.2012.04.024. References [1] X.L. Xu, X.H. Chen, Y.B. Hou, Z. Xu, X.H. Yang, S.G. Yin, Z.J. Wang, X.R. Xu, S.P. Lau, B.K. Tay, Chem. Phys. Lett. 325 (2000) 420. [2] S.Y. Yang, L. Qian, F. Teng, Z. Xu, X.R. Xu, J. Appl. Phys. 97 (2005) 126101. [3] T. Tsutsui, S.B. Lee, K. Fujita, Appl. Phys. Lett. 85 (2004) 2382. [4] S.B. Lee, K. Fujita, T. Tsutsui, Jpn. J. Appl. Phys. 44 (2005) 6607. [5] D. Adachi, S. Hasui, T. Toyama, H. Okamoto, Appl. Phys. Lett. 77 (2000) 1301. [6] P.D. Rack, P.H. Holloway, Mater. Sci. Eng. R 21 (1998) 171. [7] J.C. Hitt, J.P. Bender, J.F. Wager, CRC Crit. Rev. Solid State Mater. Sci. 25 (2000) 29. [8] V. Wood, J.E. Halpert, M.J. Panzer, M.G. Bawendi, V. Bulovic, Nano Lett. 9 (2009) 2367. [9] A. Perumal, M. Fröbel, S. Gorantla, T. Gemming, B. Lüssem, J. Eckert, K. Leo, Adv. Funct. Mater. 22 (2012) 210. [10] A. Perumal, B. Lüssem, K. Leo, Appl. Phys. Lett. 100 (2012) 103307. [11] M. Furno, R. Meerheim, M. Thomschke, S. Hofmann, B. Lüssem, K. Leo, Proc. SPIE 7617 (2010) 761716. [12] H. Becker, S.E. Burns, R.H. Friend, Phys. Rev. B 56 (1997) 1893.