OLED manufacturing for large area lighting applications
M.Eritt a ,⁎,C.May a ,K.Leo a ,M.Toerker a ,C.Radehaus b
a Fraunhofer-Institut für Photonische Mikrosysteme (IPMS)/Maria-Reiche-Straße 2,D-01109Dresden,Germany b
Technische Universität Chemnitz,09107Chemnitz,Germany
a b s t r a c t
a r t i c l e i n f o Available online 24October 2009Keywords:OLED
Indium –tin oxide (ITO)
Vacuum thermal evaporation
We present first results of a new developed large area manufacturing system for organic light-emitting diodes (OLED)dedicated to lighting and signage applications.The system combines high throughput with flexibility in production at highly-precise deposition conditions.After introducing the sy
stem with its modules,first results for organic and metal layer deposition properties are shown.Next orange/red p –i –n type OLED samples are prepared on large ITO substrates and the characterization by luminance and current measurements are presented.The devices achieve very high ef ficiencies up to 31cd/A on large area substrates which are comparable to devices on smaller substrates.
©2009Elsevier B.V.All rights reserved.
1.Introduction
Organic light-emitting diodes (OLEDs)have made great progress since the first presentation of thin film devices based on small molecule organic materials by Tang and VanSlyke in 1987[1].The OLED performance and stability have been rapidly increased over the last years [2].Also,the size of the devices is growing and several applications like displays [3]and general lighting [4]are produced.The developments for lighting applications will reach high ef ficiencies and low cost production to replace fluorescent and phosphorescent lamps [5].Because of very thin active layers (several 10–100nm),the low material amount used for the production of OLED results in cheap and lightweight products.
Stable deposition conditions are necessary for accurate thin film depositions of organic semiconduct
ors.Layer thickness homogeneity better than ±5%allows a uniform distribution of optical and electrical OLED properties over the whole substrate area.
Mass production of high ef ficient OLED on larger area is still a challenge.To show the scalability of OLED process technology for larger substrates and high throughput,a new deposition system was speci fied by IPMS and provided by Sunic Systems,Inc.It is designed for OLED signage and lighting application and doesn't need high accuracy shadow masking as used for RGB display manufacturing.Two more advantages should be mentioned:long up time of several days of production without any maintenance and a low tact time down to 3min to reach a high throughput of substrates.
The general process flow for the production of an OLED can be subdivided in different steps.The first step is the anode layer deposition on top of the bare glass typically done by sputtering of a
thin conductive oxide (TCO)and structuring by a photolithography process.The requirements for TCO-coatings for OLED applications can be summarized by high transmission (>90%),low sheet resistance of 1–50Ω/sq,high work function (~5.0eV)and low roughness below 1nm (RMS).Additionally a passivation layer is deposited optionally and structured to prevent a shorting be
tween anode and cathode wire crossings.Several cleaning steps are applied to remove residuals from the anode surface.Before the vacuum process starts,the adsorbed water will be removed by an annealing step at inert conditions.Directly connected,the plasma pre-treatment process and the OLED stack deposition are performed and closed with the transfer to the encapsulation system.The encapsulation is done by covering the active area with an additional glass to protect the sensitive OLED structure from atmospheric conditions.But a more and more interesting process is a thin film encapsulation (TFE)[6]by the deposition of a barrier layer to produce thinner devices.
This work focuses on the investigation of the ITO anode layer and presents the results of vacuum processes to deposit organic and metal layers for OLED devices with the new manufacturing tool (Fig.1).The achieved OLED device characteristics on large substrates are compa-rable to diodes produced under laboratory conditions.
2.Experimental
The manufacturing system is based on different clusters including process modules for pre-treatment,organic layer deposition,metal deposition and automatic transfer to an encapsulation system.A substrate with the size of 370×470mm 2(or so-called “GEN2”-size)is used and the transfer i
s done by robot handlers fully automated.After loading,the substrates are annealed in an oven chamber to reduce the adsorbed water.Afterwards the pre-cleaning module removes organic residuals from the substrate surface by combined argon/oxygen
Thin Solid Films 518(2010)3042–3045
⁎Corresponding author.
E-mail itt@ipms.fraunhofer.de (M.
Eritt).
0040-6090/$–see front matter ©2009Elsevier B.V.All rights reserved.doi:
10.1016/j.tsf.2009.09.188
depositionContents lists available at ScienceDirect
Thin Solid Films
j o u r n a l h o m e p a g e :w w w.e l s ev i e r.c o m /l o c a t e /ts f
plasma that is generated by an ICP (ion coupled plasma)source under vacuum conditions.The plasma process also increases the work function of indium –tin oxide (ITO)[7].
The cluster system contains five process modules for the organic and two for the metal layer depositions.These modules are equipped with point sources optimized for high layer homogeneity.Their base pressure is below 10−7mbar to generate high purity layers.A p –i –n OLED device structure [3]consists of a p-doped hole transport layer (p-HTL),an electron blocking layer (EBL),an emitter doped emission layer (EML),a hole blocking layer (HBL),a n-doped electron transpo
rt layer (n-ETL)and an aluminum cathode.The layer thicknesses of EBL,EML,and HBL range between 10and 20nm.The HTL and ETL are co-evaporated layers containing a host as matrix material and an electrical dopant.The emitter layer bases on a phosphorescent orange/red dye embedded in the host material (see Ref.[8]).The p –i –n OLED technology was pro-vided by Novaled AG [9].
The system is designed to reach a tact time of 3min which implies an output of one substrate every 3min.It is able to use rigid glass substrates or flexible sheets.The substrates used in this work are float glass coated with indium –tin oxide (ITO)and structured by photolithography and wet etching processes (substrates supplied by Thin Film Devices)without passivation and metallization layers.The cleaning of the structured ITO substrates is done by surfactants and DI-water before the processing.
The film uniformity data of the organic layers are measured by a VASE system (variable angle spectroscopic ellipsometry,J.A.Woollam Co.,Inc.)to determine optical data including thickness information at different points on the substrate [10].Morphology is investigated by non contact mode atom force microscopy (AFM)(Nanoscope D3100,Veeco).The luminance and current vs.voltage characteristics are carried out using a DMS 401system (supplied by Autronic-Melchers).3.Results
3.1.Properties of the thin films
The analysis of the ITO surface is done by AFM and shows a grain structure with smooth surface.The measured R q (RMS)value is only 0.6nm and the maximum peak (R max )value represents 12.0nm.An AFM micrograph is shown in Fig.2.The measured low roughness values are suf ficient for the deposition of thin organic layers.The sheet resistance is measured by a value of ~35Ω/sq.
To proof the deposition homogeneity of organic films,the material tris(8-hydroxy-quinoline)aluminum (Alq3)is evaporated at two different deposition rates (Fig.3).Typically rates for the host materials are 2.0Å/s and for the dopants 0.5Å/s.Measurement points are set diagonal on the substrate area to see the largest deviation [11].
Excellent thickness homogeneity is determined for both,low and high deposition rates.The variation for the host material is at 1.7%and for the dopant at 2.9%.The set point for the thickness is 100nm in both cases but a slight drift of +1.5%and −2.2%respectively,is observed.Together with the homogeneity results it is suf ficient for high-precision and stable rate OLED deposition processes.
Aluminum is used as cathode material for the OLEDs.It is evaporated from crucibles at two different r
ates.The morphology of the surface is investigated by AFM measurements (Fig.4).The sample preparation is done on cleaned bare glass by an evaporation of 100nm thick aluminum.The roughness and maximum peak values for the aluminum coating at the two rates are nearly independent from the rate and summarized in Table 1.The achieved low roughness data of the aluminum coatings are suf ficient for the cathode deposition on smooth organic layers to realize the second electrode of an OLED.
3.2.OLED devices
After the investigation of the single deposition processes shown above,materials for a p –i –n type OLED are installed.The deposition rates are constant at 2Å/s for the host materials and respectively lower for the dopant materials.For the aluminum layer the deposition rate is adjusted to 12.5Å/s.These values for the deposition rates are identical to the previous experiments with single layers.The
layer
Fig.2.AFM micrograph of the electrode ITO film on
glass.
Fig.3.Layer homogeneity of Alq3layer deposition for host and dopant source by VASE
ellipsometry.
Fig.1.View of large area OLED deposition tool.
3043
M.Eritt et al./Thin Solid Films 518(2010)3042–3045
depositions are structured by thin metal shadow masks to form 1×1cm2OLED devices.
The current–density–voltage-characteristic J(V)and luminance–voltage-characteristic L(V)are shown in Fig.5for device#1.The low voltage of2.5V for a luminance of100cd/m2and3.25V for1000cd/m2 is promising for signage applications and comparable to other published results[8].Also the reverse current at−4V is below10−4mA/cm2and the curves showing good diode behaviors.In Fig.6the values for current efficiency at a luminance of1000cd/m2are plotted against the position of the devices on the substrate.The device positions are equal to the layer homogeneity positions above.The connection of the individual devices is not equal to each other therefore only the current efficiency can be presented and shown in Fig.6.The J(V)and L(V)characteristics of the single device#1are shown in Fig.5.
4.Discussion
Considering the good homogeneities,the variation for the current efficiencies is higher than expected.The efficiency of the OLED depends mostly on doping ratio and thickness of each organic layer.To understand the different efficiencies achieved for the p–i–n devices distributed at the substrate,further investigations are necessary. Especially for the emitter dopant material at lower deposition rate (below0.5Å/s)the thickness distribution has to be determined.One open point is the influence of the thickness distribution of the deposition sources shown in Fig.3to the device efficienci
es.The observed drop down of the layer thickness to the middle of the substrate could affect the efficiency of the devices by changing the optical out coupling of the thinfilm structure[12].To improve the up scaling process to larger area, the characteristics of the OLED devices need better understanding.
5.Conclusions
We have shown thefirst results realized with a new OLED manufacturing system.The single organic and metal layers were deposited at low and high deposition rates to prove the homogeneity with respect to roughness information.Homogeneity measurements provided good results for Alq3below1.7%(2.0Å/s)and2.9%(0.5Å/s) that are better than the predicted5%.The roughness of the aluminum was not influenced significantly by the deposition at two different rates.A thick aluminum cathode layer of several hundred nanometers can be evaporated in a short tact time without any influence to the underlying organic layers.The indium–tin oxide used as anode contact had also low roughness below1nm.We presented orange/ red p–i–n OLED devices showing good efficiency data between19and
Table1
Morphology data for aluminum at different rates(AFM).
Deposition rate(Å/s)R q(RMS)(nm)R max(nm)
12.5 1.520.6
25.0 2.0
19.5
Fig.6.Current efficiency values at different positions on the
substrate.
Fig.4.AFM micrographs of the aluminum cathode at different rates12.5(a)and25.0Å/s(b).
3044M.Eritt et al./Thin Solid Films518(2010)3042–3045
31cd/A and also low driving voltages.A picture of the test substrate with all 21lighting devices can be seen in Fig 7.
The next step will be the fabrication of a white emission OLED stack,the upsizing of the OLED areas to create devices for signage and lighting applications.In the future organic solar cells (OSC)processes [13]can be transferred from smaller lab systems to the described large area pilot line.
Acknowledgements
The research is funded within the framework for technology promotion by means of the European Fund for Regional Development (EFRE)as well as by means of the Free State of Saxony.Parts of the work leading to these results have received funding from the European Community's Seventh Framework Programme under grant agreement no.FP7-224122(OLED100.eu).
References
[1]C.W.Tang,S.A.VanSlyke,Appl.Phys.Lett.51(1987)913.
[2]K.Walzer,B.Maennig,M.Pfeiffer,K.Leo,Chem.Rev.107(2007)1233.
[3]M.Pfeiffer,S.Forrest,K.Leo,M.E.Thompson,Adv.Mater.14(22)(2002)1633.[4]Z.H.Kafa fi,CRC Press,Bellingham,2005.
[5]F.So,J.Kido,P.Burrows,MRS Bull.33(7)(2008)683.[6]S.-H.K.Park,et al.,ETRI J.27(5)(2005)545.
[7]Y.Hashimoto,Y.Osato,M.Tanaka,M.Hamagaki,T.Sakakibara,Jpn.J.Appl.Phys.41(part 1,no.4A)(2002)2249b.
[8]
R.Meerheim,S.Scholz,S.Olthof,G.Schwartz,S.Reineke,K.Walzer,K.Leo,J.Appl.Phys.104(014510)(2008).
[9]J.Birnstock,M.Hofmann,S.Murano,M.Vehse,J.Blochwitz-Nimoth,Q.Huang,G.He,M.Pfeiffer,K.Leo,SID Symposium Digest,vol.36,2005,p.40.
[10]J.A.Woollam,B.Johs,C.Herzinger,J.Hil fiker,R.Synowicki,C.Bungay,SPIE Pro-ceedings,vol.CR72,1999,p.3.[11]E.Lee,Vacuum 83(2009)848.
[12]B.C.Krummacher,et al.,Org.Electron.10(2009)478.[13]
C.W.Tang,Appl.Phys.Lett.48(183)
(1986).
Fig.7.Substrate with 21orange/red devices in series connection.
3045
M.Eritt et al./Thin Solid Films 518(2010)3042–3045
版权声明:本站内容均来自互联网,仅供演示用,请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198,我们将在24小时内删除。
发表评论