作者简介:曾文进(1981—),男,南京邮电大学讲师、博士.研究方向:有机光电器件.E-mail:iamwjzeng@njupt.edu.cn
中文责编:方 圆; 英文责编:木 南
1)南京邮电大学材料学院,江苏南京 210023; 2)华南理工大学发光材料与器件教育部重点实验室,广东广州 510641
1)School of Materials Science and Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210023, Jiangsu Province, P.R.China2)State Key Lab of Luminescent Materials and Devices, South China University of Technology, Guangzhou 510640, Guangdong Province, P.R.China
chemical physics; polymer light-emitting diodes; Ag paste; blade-coating; printed cathode; interface resistance
DOI: 10.3724/SP.J.1249.2016.01018
研究银粉含量和印刷阴极型聚合物发光二极管(polymer light-emitting diode,PLED)之间的构-效关系.实验比较两种不同银粉含量的导电银胶,通过刮涂法制备PLED的阴极.两种银胶的胶体基底相同,区别在于银粉颗粒的含量不同.实验研究银粉的分布状态与器件性能之间的关系.结果表明,银胶中的银粉含量越高,器件的性能越好,主要体现在驱动电压更低、电流密度更大和量子效率更高.偏光显微镜图片显示,提高银胶中银粉的含量,可以优化银粉在印刷阴极/电子传输层之间的分布.通过银粉覆盖率的数据模拟也证明了这一点.为确定银粉覆盖率的提高能够优化器件效率,在器件中通过蒸镀添加薄银层.结果表明,由于薄银层的插入,器件的驱动电压随之下降,器件性能也得到优化.因此,在印刷型的PLED器件中,提高银胶中银粉的含量可以有效减低载流子的注入势垒,达到器件优化的效果.
The structure-activity relationship of the cathode-printed polymer light-emitting diodes(PLEDs)is investigated. Two kinds of Ag pastes on the same paste resin but with different content of Ag particles were applied to prepare the cathodes of PLEDs by the blade-coating. The relationship between the distribution of Ag particles and the performance of PLEDs was investigated. The results indicate that the paste with a higher silver content exhibites better performances, including a lower driving voltage, higher current density and quantum efficiency. In-situ polarized microscopic images reveal that a higher silver content in the paste could lead to a better distribution of Ag at the interface of the cathode and the electron-transporting layer(ETL), which can also be proved by the simulation of the coverage percentage. A thin layer of Ag was inserted by evaporation between the ETL and Ag-paste cathode, which is regarded as equivalent to the increase of Ag coverage at the interface. As expected, the driving voltage of the devices was reduced and the performance improved after the thin layer of thermally-deposited Ag was inserted. Therefore, large Ag contents at the interface benefits the performance of PLED due to the low injection barriers.
The polymer light-emitting diode(PLED)has attracted tremendous attention due to its superiorities in solution processability, low cost, richness of display colors and its potential applications in large-area display panels and solid state light source, etc. Moreover, PLED can be fabricated by the technique of full-printing[1- 4]. Until now, the series of novel electroluminescent polymers, which are suitable for solution-processing, have been developed since electronic luminescence was reported from devices made of conjugated polymers[5-9].
The cathode of PLED can also be printed from the metal pastes, such as silver, copper or gold pastes. However it should be noted that high efficiency required the balanced injection of charge carriers from both the electrodes(anode and cathode)before we applied the metal pastes on the cathode as mentioned above[10]. In addition, metals with low work function, such as Ca, Ba and Mg, are not suitable for printing due to their high chemical activities. Currently, Ag paste is considered as the most promising material for the cathode printing in full-printed PLED based on the following three reasons: ① Silver paste can be achieved with comparable conductivity to evaporated metal after curing at room or moderate temperature, unlike CNTs or graphene which need super high temperature to achieve considerable conductivity. ② Ag paste possesses strong adhesive strength which leads to its wide application in the field of electronic circuits. ③ Ag paste is applicable to most printing techniques such as blade-coating, inkjet printing, screen printing, etc[11].
Due to the high work function of silver, generally an electron-transporting layer(ETL)is needed to match the energy levels of the polymer layer and the cathode. Amino-/ammonium-functionalized polyfluorene were synthesized ETL materials in the full-printed PLED by Cao and other groups[12-20].
However, we still notice that the performance of PLED with printed cathode needs further improvement, which mainly reflects its higher driving voltage and lower current density. It is necessary to investigate the injection barriers of the silver particles at the interface between the polymer layer and the Ag-paste cathode. And the injection barriers of the silver particles may be greatly affected by the distribution of silver particles in the paste, which arises from the particles size, the silver content and contact resistance at the interface of the polymer/silver paste.
In this study, PLED was fabricated with the cathode made from two kinds of Ag pastes, based on the same resin base but the different Ag contents. It was found that the Ag contents significantly influence the coverage of Ag particles at the polymer-cathode interface. The coverage of Ag particles is related to the interfacial resistance of the cathode. With the simulation of the polarized microscopic images by scientific image processing software, we can analyze the relationship between the distribution of Ag particles and the performance of the device.
Indium tin oxide(ITO)glass with a surface resistance of ca. 25 Ω/sq was purchased from China South Glass Co. Ltd. Poly[2-(4-(3',7'-dimethyloctyloxy)-phenyl)-p-phenylenevinylene](P-PPV)and poly[9, 9-bis(3'-(N, N-dimethylamino)propyl)-2,7-fluorene-alt-2,7-(9,9-dioctylfluorene)](PFN)were synthesized as reported elsewhere[21-22]. Poly(3,4-ethylenedioxythiophene)-polystyrenesulfonic acid(PEDOT:PSS)(Baytron P 4083)was purchased from Bayer company and used without further purification. The conducting Ag pastes with a viscosity of ca. 18 Pa·s and a conductivity higher than 3 × 103 s·cm-1 were prepared in laboratory. Two kinds of Ag pastes with different Ag content were applied in this study. Paste KD-1 has a mass fraction of 75% for the Ag content, while paste KD-2 has a higher mass fraction of 95%.
ITO glass was cleaned ultrasonically by a solvent bath of acetone, detergent, deionized water and isopropyl alcohol in sequence. Surface treatment by O2 plasma was performed on ITO surface to remove the organic residue and improve the work function as well. The layers of PEDOT:PSS, P-PPV solution(5.5 mg/mL in p-xylene)and PFN solution(4 mg/mL in methanol)were formed on the clean ITO glass in sequence by spin-coating. The optimum thickness for each layer was found to be 40 nm of PEDOT:PSS, 80 nm of P-PPV and 20 nm of PFN. For the devices with the cathode made of Ag paste, the Ag paste was patterned on top of the PFN layer by the method of blade coating in a glove box under an inert atmosphere. The shape and thickness of the Ag paste were controlled by a plastic mask. For the control device using Ag as a cathode, 150 nm Ag was thermal evaporated at a rate of 0.2 nm/s under high vacuum below 3 × 10-4 Pa, with the metal thickness controlled by a calibrated crystal oscillator. The architecture of electron-only device was ITO/Sn(30 nm)/ P-PPV(80 nm)/PFN(20 nm)/cathode. The 30 nm layer of Sn was thermally evaporated under high vacuum below 3 × 10-4 Pa. The subsequent deposition of P-PPV, PFN and the cathode was same as that of the standard devices.
The thickness of the polymer thin films was determined by a surface profiler(Tencor Alpha-Step 500). Current density-luminance-voltage(J-L-V)characteristic curves were collected on a semiconductor testing system consisting of a Keithley 236 source-meter and a calibrated silicon photodiode. The external quantum efficiency(QE)was calculated by measuring the light output in a calibrated integrated sphere(IS-080, Labsphere). The polarized microscopic images were collected on the polarized microscope(Nikon Eclips E600, Tokyo, Japan). The coverage of Ag particles was calculated using an image-processing software(Image J, a widely-used software to calculate the area of irregular shape).
The chemical structures of the polymers P-PPV and PFN are shown in Fig.1(a)and(b)respectively, in which the P-PPV acts as the emission layer(EML)and PFN as the electron-transporting layer(ETL). The amino group of PFN can induce dipoles under the applied electric filed, thus benefits the electron injection from the PLED cathode. As known, Ag is a noble metal with high work function, which is unfavorable for electron injection from the cathode. Therefore, the thin layer of PFN plays a very important role in the realization of using Ag metal as the cathode. Fig.1(c)indicates the device architecture has the optimum configuration of ITO/PEDOT:PSS(40 nm)/P-PPV(80 nm)/PFN(20 nm)/Ag paste, which had been verified in previous study[18].
Fig.1 Chemical structures of P-PPV, PFN, and the device architecture with the configuration of ITO/PEDOT:PSS(40 nm)/P-PPV(80 nm)/PFN(20 nm)/Ag paste
In our previous study[15-16], it has been revealed that the thickness of PPV and PFN can affect device performance. The thickness of each film layer indicated in the structure has been optimized. For comparison, devices with evaporated Ag as cathodes were also fabricated in the same configuration as the control. For simplicity, devices with cathodes made of the above-mentioned materials were referred as device 1, device 2 and device 3, corresponding to evaporated Ag, Ag paste KD-1 and KD-2, respectively.
It is expectable that there are differences between the electric properties of PLED with evaporated Ag and Ag paste as the cathode. Fig.2(a)and(b)respectively demonstrate the J-L-V characteristic curves and QE-J curves of P-PPV devices with different cathode materials. It can be seen that devices 1 and 3 have similar on-voltage, lower than that of device 2. However, when the luminance at a specified voltage is taken into account, device 1 possesses a higher value than device 2 and device 3. As a result, indicated in Fig.2(b), device 1 demonstrates the highest efficiency. More details are summarized in table 1. At a specific current density of 10 mA/cm2, device 1 achieves a luminance of 880 cd/m2 at 7.9 V, with a QE of 3.8%, a QEmax of 4.2% and a maximum luminance efficiency(LEmax)of 11.0 cd/A. For devices with two kinds of Ag paste as a cathode, such as device 2 and device 3, the latter has a much better performance, with a luminance of 740 cd/m2 at 7.8 V, a QEmax of 3.0% and a LEmax of 7.8 cd/A.
Fig.2 Electronic properties of P-PPV devices with two different Ag pastes(KD-1 and KD-2)as device cathodes. Performance of control device with evaporated Ag is also shown for comparison.
Despite the obvious results that device 1 has attained the best performance among the three types of devices, it would be more interesting to make clear the reason why device 3 possesses a superior performance over device 2 merely arising from the different Ag contents in the paste. It can be taken for granted that pastes of different Ag content have different resistance, which can influence device performance. However, to further understand the mechanism, silver distribution of Ag particles should be investigated, especially at the interface between the cathode and the polymer layer. Therefore, polarized microscopy was applied in-situ to observe the distribution of the Ag particles within the actual device. The polarized images of the two pastes are presented in Fig.3(a)and(b), in which the brighter spots represent the Ag particles directly contacting the polymer layer, while the darker background is the opaque paste resin. By comparing the two images, it is clear that Ag particles have a richer distribution in device 3. As known, Ag particles are the effective components making an Ag paste conductive. The enrichment of Ag particles can facilitate better formation of an ohmic contact[23-24] between the cathode and the polymer layer and would thus effectively benefit the transporting and injection of electrons. To quantitatively evaluate the enrichment of Ag particles, the value of Ag particles coverage was calculated using simulation software. The simulated images are demonstrated in Fig.3(c)and(d), which correspond to the polarized images in(a)and(b). The coverage of Ag particles in Fig.3(a)was calculated to be 27.6%, less than 65.0% in Fig.3(b). Direct contact of silver particles with the polymer layer can reduce the internal resistance of the device since the resin of the silver paste is usually isolated[25]. Therefore, it is reasonable to speculate that the lower coverage of Ag particles in paste KD-1 led to its poorer device performance.
Fig.3 In-situ polarized microscopic images collected from the ITO side to compare the distribution of Ag particles between the two Ag pastes
Electron-only devices were prepared with tin(Sn)as the anode. The injection barrier height of different cathode configurations can be calculated according to the Fowler-Nordheim(FN)tunneling mechanism. In the FN model, the injection current density(J)is related to the magnitude of electric field(F)by the following equation[26-27],
J=AF2exp(-K/F)
where
A=(q3)/(8πhφ)
K=8π((2m)1/2φ3/2)/(3qh)
Here φ is the barrier height, m the effective mass of electrons in the active materials, q is the charge of an electron, and h is Planck's constant. The barrier heights were calculated to be 0.40, 0.51 and 0.44 eV, respectively, for the cathode configurations of Ag, Ag paste KD-1 and Ag paste KD-2, as indicated in Fig.4. The results indicate that higher coverage of Ag particles can lower the injection carrier of the electrons injecting from the cathode.
图4 不同电极材料的P-PPV器件的阴极注入势垒计算曲线To further verify the speculation, a thin layer of Ag in a different thickness was thermally evaporated on top of the polymer layer before the coating of the Ag paste KD-1. With the thickness of the evaporated Ag increasing, the Ag particles can finally form a complete layer of Ag which can fully cover the polymer surface. The evaporated thin Ag film and the following coated Ag paste together constituted the device cathode. Therefore, the thermal evaporation of the inserted Ag film can be regarded as an analogous process in which the coverage of Ag particles in the paste can increase gradually with the growth of the evaporated Ag film. The J-V curves and L-V curves are shown in Fig.5(a)and(b).
Fig.5 Electronic properties of P-PPV devices with a thin layer of evaporated Ag in different thickness inserted between polymer layer and Ag paste cathode KD-1
It can be seen that the electric behavior of the paste device, with the thickness of the evaporated Ag layer increasing from 0 to 30 nm, became more similar to that of the evaporated control one. And its performance also improves gradually towards that of the control device. Therefore, it is proved again that a low coverage of Ag particles in the paste is unfavorable to device performance. It is crucial to develop efficient methods to increase the coverage of Ag particles to achieve high-efficiency PLED with the cathode made of Ag paste.
In summary, we have undertaken an investigation on the relationship between the silver content in the paste and the device performance of the cathode-printed PLED. Our experiment results reveal that higher silver content in the paste facilitates better distribution of Ag at the interface of the cathode and the electron-transporting layer. According to the simulation results, higer coverage of Ag particles at the interface is favorable to the device performance, therefore it provides guidance to the further improvement of the printed PLED with Ag paste as cathode. As a result, the maximum QE of 3.0% are achieved for the devices with the cathode of higher silver contents.
深圳大学学报理工版
JOURNAL OF SHENZHEN UNIVERSITY SCIENCE AND ENGINEERING
(1984年创刊 双月刊)
主 管 深圳大学
主 办 深圳大学
编辑出版 深圳大学学报理工版编辑部
主 编 阮双琛
国内发行 深圳市邮电局
国外发行 中国国际图书贸易集团有限公司(北京399信箱)
地 址 北京东黄城根北街16号
邮 编 100717
电 话 0755-26732266
0755-26538306
Email journal@szu.edu.cn
标准刊号 ISSN 1000-2618
CN 44-1401/N