Chapter 5 Current-Driven AMOLED with Fully Integrated Drivers
5.5 Design verification
The complete operation of this integrated current driving system was verified after depositing OLED materials onto a substrate fabricated by poly-Si TFT technology. The photograph of a 2.2 inch TFT substrate with QCIF+ resolution (176 x RGB x 220) is shown in Fig. 5-21.
Scan driver 1
Scan driver 2
Logic circuit 176 x RGB x 220
current-driven pixel array
Voltage latch DCC
SI current memory
6-bits IREF
Fig. 5-21. Photograph of glass substrate with integrated data and scan drivers and current-driven pixel array.
A high work function ITO electrode was deposited onto the substrate as an anode contact for OLED. Since available states of OLED materials are less than 5 eV, a thin layer of copper phthalocyanine (CuPc) was used to facilitate the injection of holes by presenting a low energy barrier. Next, a layer of naphtha-phenyl-benzidene (NPB)
was deposited to form a hole transporting layer (HTL). An emitting layer followed, often consisting of tris-(8-hydroxyquiniline) aluminum (Alq3) doped with a fluorescent dye such as Coumarin 540 for generating R. G. and B three primary colors.
For patterning the emitting layer, a shadows mask was required to assist in depositing the R, G, B materials onto the corresponding pixel areas. An additional layer of Alq3
serving as the electron transporting layer (ETL) was deposited in sequence. Finally, low-work-function cathode metal completed the thin film stack as an electron injector.
0 10 20 30 40 50 60
300 400 500 600 700 800
0.0
Fig. 5-22. Optical performance of fabricated current-driven AM-OLED. (a) Display luminance as a function of gray scales. (b) Display spectral response. (c) Display luminance versus viewing angle. (d) Display contrast ratio as a function of viewing angle.
To verify the optical performance, the AM-OLED was configured in the QCIF+
resolution and 60 Hz frame rate. The pattern is an all-white flat field. The luminance
of the OLED, recorded by a GmbH Conoscope providing direct readout, varies linearly with input digital signals as seen in Fig. 5-22(a). The peak brightness of 440 cd/m2 can be achieved under operation voltage less than 15 volts. The spectral response shown in Fig. 5-22(b), reveals a color gamut that is adequate for general display applications. The luminance and contrast ratio versus the viewing angle are shown in Figs. 5-22(c) and 5-22(d). Exactly as a self-emission device, the AM-OLED display possesses a remarkably wide viewing angle as well as a superior contrast ratio of higher than 2500 in dark environment.
Fig. 5-23. A color photograph taken directly from the current-driven AM-OLED display panel.
Operation of the entire current-driven AM-OLED has been verified through successful performance in an actual display application as shown in Fig. 5-23. Besides, pixel to pixel luminance variation is almost invisible in this display so that the proposed integrated current-driven circuits including pixel and drivers are capable of achieving an extremely uniform luminance. Specifications of this AM-OLED are
summarized in Table 5-4.
Table 5- 4. AM-OLED display specifications.
Display size 2.2 inch diagonal
Resolution QCIF+, 176xRGBx220
Pixel pitch 198µm x 198µm
Gray scale 6 bits
Input data Digital RGB signals Peak luminance > 400 cd/m2
CIE coordinate R (0.591, 0.364) G (0.412, 0.542) B (0.204, 0.272) Contrast ratio > 2500:1 in dark
5.6 Summary
Design and implementation of current driven AM-OLED with fully integrated data drivers are described and the measurement results of such the panel fabricated with poly-Si TFT technology are presented. In the integrated current type driving system, the SI memory operates with only one control signal and has a capability to improve the output current accuracy by suppressing the influence of charge injection.
Moreover, fast response time is achieved by the capacitance of small size without sacrificing the output accuracy. The features, such as fast response time and high accuracy of output current, favor the applications of the proposed SI memory in driver circuits of current driving active-matrix OLED panels. The digital-to-current converter with the new-designed SI memory is capable of generating desired analog current signals for current-driven pixel circuit regardless of the operation temperature and device aging. The expensive current-type driver ICs are no longer needed due to the fully integrated data drivers which can also provide a digital interface to increase
the design flexibility. The measured display performance shows that the current-driven AM-OLED is superior in high uniformity of display luminance to meet the requirements of high performance display applications.
Reference
1. M. Kimura, H. Maeda, Y. MAtsueda, S. Miyashita, T. Shimoda, S. W. B. Tam, P. Migliorato, J. H. Burroughes, C. R. Towns, and R. H. Friend, “Low-Temperature Poly-Si TFT driven Light-Emitting Polymer Displays and Digital Gray Scale for Unniformity,” in Proc. IDW’99, 1999, pp. 171-174.
2. D. Pribhat and F. Palis, “Matrix addressing for organic electroluminescent displays”, Thin Solid Films, vol. 383, pp. 25, 2001.
3. R. M. A. Dawson, Z. Shen, D. A. Furst, S. Connor, J. Hsu, M. G. Kane, R. G. Stewart, A.
Ipri, C. N. King, P. J. Green, R. T. Flegal, S. Pearson, W. A. Barrow, E. Dickley, K. Ping, C. W.
Tang, S. Van Slyke, F. Chen, J. Shi, J. C. Sturm, and M. H. Lu, “Design of an improved pixel for a polysilicon active-matrix organic LED display,” in Symp. Dig. 1998 SID, 1998, pp.
11-14.
4. J. C. Goh, H. J. Chung, and C. H. Han, “A New Poly-Si TFT Pixel Circuit Scheme for an Active-Matrix Organic Light Emitting Diode Displays,” in Symp. Dig. 2001 IDW, 2001, pp.
312-322.
5. Y. W. Kim, S. R. Lee and O. K. Kwon, "A New Pixel Structure for AMOLED Panel with 6-bit Gray Scale," in Symp. Dig. 2002 EuroDisplay, 2002, pp. 613-617.
6. Y. He, R. Hattori, and J. Kanicki, “Current-Source a–Si:H Thin-Film Transistor Circuit for Active-Matrix Organic Light-Emitting Displays,” in IEEE Electron Device Lett., vol. 21, no.
12, Dec 2000, pp. 590-592.
7. Y. He, R. Hattori, and J. Kanicki, “Improved A-Si:H TFT pixel electrode circuits for active-matrix organic light emitting displays,” in IEEE Trans. Electron Devices, vol. 48, no. 7, Jul, 2001, pp. 1322-1325.
8. D. Vallancourt and S. J. Daubert, “Applications of current-copier circuits,” in Analog IC Design: The current mode approach. C. Toumazou, F. Lidgey, and D. J. Haigh, Eds. London:
Petegrinus, 1990, p. 515-534.
9. T. Fiez, G. Liang, and D. J. Allstot, “Switched-current design issues,” in IEEE J. Solid-State Circuits, Vol. 25, no. 3, 1990, p. 699-705.
10 . R. E. Suares, P. R. Gary, and D. A. Hodges, “All-MOS charge redistribution analog-to-digital conversions techniques-Part II,” in IEEE J. Solid-State Circuits, vol. SC-10, 1975, p379-385.
11. M. S. Shur, H. C. Slade, T. Ytterdal, L. Wang, Z. Xu, K. Aflatooni, Y. Byun, Y. Chen, M.
Froggatt, A. Krishnan, P. Mei, H. Meiling, B.-H. Min, A. Nathan, S. Sherman, M. Stewart, and S. Theiss, “Modeling and Scaling of a-Si:H and Poly-Si Thin Film Transistors,” Mat. Res.
Soc. Symp. Proc., vol. 467, 1997, pp. 831-842.
12. M. S. Shur, H. C. Slade, M. D. Jacunski, A. A. Owusu, and T. Ytterdal, “SPICE Models for Amorphous Silicon and Polysilicon Thin Film Transistors,” J. Electrochem. Soc., vol. 144, no. 8, 1997, pp. 2833-2839.
Chapter 6
Current-Scaling Pixel Electrode Circuit for AM-OLED
6.1 Introduction
To modulate the OLED current, two approaches have been often used. In the first approach, a voltage signal is used to directly control the driving current of two-TFT pixel electrode circuit. Unfortunately, in this pixel configuration, non-negligible TFT characteristic variations (threshold voltage and field-effect mobility shifts) due to the manufacturing process variation and to the device aging can result in non-uniform luminance over the display area. Current driving schemes with four-TFT pixel electrode circuit has been demonstrated in previous chapter as another approach to drive AM-OLED, whereby the current signal provided by external driver modulates directly the pixel electrode circuits. The four-TFT circuits can not only provide a continuous excitation to OLED, but at the same time can also compensate for the TFT threshold voltage variation.
Although the current driving scheme improves the display luminance uniformity, a large timing delay can be observed at a low data current that is due to combination of a high OLED efficiency and charging of a large interconnect parasitic capacitances. For example, a current of 70 nA is sufficient to achieve luminance of 100 cd/m2 when an OLED with efficiency of 20 cd/A or higher is used. However, for such small current, an interconnect parasitic capacitance of about 10 pF needs more than 150 µsec to build up a sufficient voltage level. This charging time is much larger than 30 µsec, that is needed for a display with VGA (640 x RGB x 480) resolution
operated at 60 Hz. To reduce the programming time delay, the pixel electrode circuits based on an adjustable TFTs geometric ratio with the current scaling function have been proposed [1][2]. One example of such circuit is current mirror type pixel electrode circuit, as shown in Fig. 6-1(a), where a high data-to-OLED-current ratio can only be achieved for a large geometric ratio of T4 to T3, yet, significantly limit the pixel electrode aperture ratio. A possible solution to this problem is top emission OLED structure in which a nearly entire pixel area can be used as light-emitting region [3]. In general, the pixel aperture ratio should not be influenced by the size of TFT and the complexity of pixel electrode circuit. This is especially true for high resolution displays.
(a) (b)
Fig. 6-1. Schematic diagrams of current driven pixel circuits with (a) conventional current mirror and (b) cascade structure of storage capacitors.
In this dissertation, we present an improved current driven pixel circuit based on poly-Si TFT technology with the current scaling function. A cascade structure of storage capacitors is proposed here to achieve a high data-to-OLED-current ratio without increasing TFTs size in comparison to conventional current mirror pixel circuit, shown in Fig. 6-1(a). The proposed pixel electrode circuit can also compensate
for poly-Si TFT threshold voltage variation so that uniform display luminance can be expected. First we describe the structure and discuss the operation principles of pixel electrode circuit. The parameters used for circuit simulation are discussed in Section III. The simulation results and circuit performance are discussed in Section IV, with conclusion presented in Section V.
6.2 Proposed pixel electrode circuit
The proposed current driven pixel electrode circuit consists of three switching TFTs (T1, T2, T4), one driving TFT (T3) and two storage capacitors (CST1, CST2) connected between a scan line and ground with a cascade structure, as shown in Fig.
6-1(b). The operation of the circuit is controlled by four external terminals: V
SCAN, VCTRL, IDATA and ground. The signals of VSCAN, VCTRL, and IDATA are supplied by external drivers while the cathode of OLED is grounded. It should be noticed that to simplify the circuit analysis, one node of CST1 connected to the ground is adopted. In practice, the ground electrode of CST1 can be connected to the Vdd line in order to reduce number of lines. The operation of this pixel electrode circuit is described as follow:During the ON-state, the scan line signal VSCAN turns on the switching transistors T1 and T2. During this period, a data current signal IDATA passes through T1 and T3 to OLED, shown as the solid line in Fig. 6-1(b), and sets the voltage at the T3 drain electrode (nodes A). At the same time, the voltage at the T3 gate electrode (node B) is set by IDATA passing through T2 (dash line). The control signal VCTRL turns T4 off to ensure that no current flows through T4. Consequently, the ideal OLED current in ON-state should be equivalent to IDATA. Since the T3 drain and gate electrodes are at the same potential, T3 will operate in the deep saturation region, e.g., VDS > VGS-VTH
(threshold voltage) and the VA and VB voltages at both nodes are determined automatically according to Eq. 6-1.
( )
2 length of TFT(T3), respectively. If T3 threshold voltage changes and if this change is not higher than VSCAN amplitude, the T3 gate voltage, VB-ON, will be adjusted accordingly to ensure the identical IDATA in ON-state. Therefore, VB-ON is always adjusted to keep IDATA at about the same value regardless of poly-Si TFT threshold voltage. The VB-ON will also be stored in both CST1 and CST2 and the voltage across CST2 is VSCAN-VB-ON.When the pixel changes from ON- to OFF-state, VSCAN turns off T1 and T2 and VCTRL simultaneously turns on T4. Because CST2 is connected between the scan line and the node B to form a cascade structure with CST1, VSCAN change from high to ground state will reduce VB-ON to VB-OFF due to the feed-through effect of the capacitors. VB-OFF can be derived from the charge conversation theory, and is given by
Eq. 6-2, in which ∆V
SCAN and COV-T2 are an amplitude of VSCAN (=VSCAN-ON – VSCAN-OFF) and the overlap capacitance of T2, respectively.2 will turn on T3 during this period. Since the overdrive voltage of T4 (=VCTRL-VA-VTH) is lower than Vdd-VA, the T4 is working in saturation region. In order to ensure that the VA is similar to Vdd and the T3 is operating in the deep saturation region, the width of T4 should be large enough to reduce the turn-on resistance of T4. A current smaller than IDATA , shown as the dash line in Fig. 6-1(b), will be generated by VB-OFF
and will pass through T4 and T3 to OLED. Consequently, the OLED current in OFF-state, IOLED-OFF, will be smaller than IDATA.
Since the T3 gate voltage decreases from VB-ON to VB-OFF, the OLED driving current is scale-down from ON- to OFF-state by the storage capacitor cascade structure. The quantity of voltage drop, shown as ∆VSCAN·CST2/(CST1+CST2) in
2
will increase with increasing ∆VSCAN and CST2 values and will lead to a smaller IOLED-OFF. In other words, the scale-down ratio, RSCALE=IOLED-ON/IOLED-OFF, is related to both the size of CST2 and to ∆VSCAN. Therefore it is expected that a larger CST2 will result in larger RSCALE. Consequently, when a very large data current IDATA is used to charge the pixel electrode and to shorten the pixel programming time, at the same time, a smaller driving current IOLED-OFF can be achieved for lower gray scales.
6.3 Parameter extraction and pixel electrode circuit design
Synopsis H-SPICE simulation tool with the Rensselaer Polytechnic Institute (RPI) Troy, NY, poly-Si TFT model [4][5] was used to evaluate the proposed pixel electrode circuit. The poly-Si TFT parameters developed within our group were used in this simulation [6][7]. The transfer characteristics (ID-VGS, drain current versus gate-to-source voltage) of poly-Si TFT are shown in Fig. 6-2 and its transconductance is given in the insert. To simulate the behavior of OLED the conventional semiconductor diode model, with the parameters extracted for organic polymer light-emitting device (PLED) fabricated in our laboratory, was used. The opto-properties of PLED are shown in Fig. 6-2 and were described elsewhere [8]. In the pixel design, a CST1 with the fixed size of 2.5 pF was used and CST2 size was
varied from 210 to 625 fF to achieve different CST2/CST1 ratios. Since T2 works as a switch in this circuit, its size can be smaller in comparison with other TFTs. Based on our own experience, we conclude that a high-performance poly-Si TFT with µFE
higher than 1.5 cm2/V-sec is essential for future poly-Si TFT pixel electrode circuit.
-5.0 -2.5 0.0 2.5 5.0 7.5 10.0
Fig. 6-2. (a) Transfer characteristics of poly-Si TFT. The transconductance versus gate
voltage is shown in insert. (b) An example of measured PLED current density and
brightness variation with supplied voltages.
The poly-Si TFT with lower µFE will need a higher driving voltage and larger geometric size to achieve an adequate OLED driving current level. Then, increased display power consumption and reduced pixel aperture ratio, when light is emitted through the substrate, are expected. In addition, it is expected that a higher performance TFT will have better electrical stability over the time. The poly-Si TFTs and OLED parameters used for this pixel electrode circuit simulation are given in
Table 6-1. The parameters used in pixel circuit simulation..
Table 6-1. The parameters used in pixel circuit simulation.
Device parameters for TFT W/L (T1, T3, T4) (µm) 150/6
W/L (T2) (µm) 50/6
VTH (V) 2
µFE (cm2/V-sec) 1.9
COV (nF/m) 0.2
IOFF (pA) 0.1
CST1 (pF) 2.5
CST2 (fF) 210~625
Device parameters for OLED
n 31
RS (Ω) 20
IS (A) 8 x 10-5
COLED (pF) 3
Supplied signals
VSCAN (V) 0~35
VCTRL (V) 0~35
Vdd (V) 35
IDATA (µA) 0~5
Times (mSec)
tON 0.33
tOFF 33
6.4 Simulation results and discussion 6.4.1 Current scaling ratio
Since IOLED-ON (=IDATA) is larger than IOLED-OFF by a factor of RSCALE, the average OLED current (IAVG) for the pixel electrode circuit must be properly defined:
OFF
Where tON and tOFF denote the select and deselect periods during the frame time, respectively.
From this equation, an accurate IAVG can be calculated for various combinations of IOLED-OFF and RSCALE to satisfy the display requirements for different gray scales. In order to display low gray scales, not only a low IOLED-OFF but also a high IOLED-ON is needed to control both a low display luminance and a fast programming speed at the same time. Combination of a low IOLED-OFF and a large RSCALE can be used to satisfy such this display requirement. For higher gray scales, a high IOLED-ON is not needed since a high IOLED-OFF can be achieved. Therefore, a combination of a large IOLED-OFF
and a low RSCALE is appropriate to display high gray scales.
Since the scale-down ratio (RSCALE=IOLED-ON/IOLED-OFF) will affect the performance of the proposed pixel electrode circuit, it is important to evaluate its evolution with the IDATA (=IOLED-ON) and CST2/CST1. The variation of RSCALE as a function of IDATA is shown in Fig. 6-3(a), where we can conclude that when CST2/CST1=1/12, RSCALE decreases from 210 to 1.5 as IDATA increases from 0.1 to 10 µA. In this specific case, since VB-ON at a high gray scale is larger than that at a low gray scale, it is expected that a large I will pass through T3. And a fixed voltage
drop induced by ∆VSCAN·CST2/(CST1+CST2) is relatively small in comparison to VB-ON, hence data current drop is expected to be small.
0.1 1 10
Fig. 6-3. Variation of the scale-down ratio versus (a) data current and (b) ratio of storage capacitances.
In the other words, a fixed voltage drop can dramatically affect VB-ON at low gray scales where VB-ON is small. Therefore, a desirable high RSCALE at low gray scales and
OFF
a low RSCALE at high gray scales can be achieved by proposed pixel electrode circuit.
The variation of RSCALE with the CST2/CST1 is as shown in Fig. 6-3 (b); which was derived from Fig. 6-3 (a). It should be mentioned that in Eq. 6-2, a large CST2/CST1
ratio can induce a large VB offset between pixel ON- and OFF-state. Consequently, VB
decrease will result in the scale-down of the data current and in a high RSCALE. The simulation results show that when IDATA is fixed, RSCALE increases when CST2
increases from 210 to 625 fF, corresponding to an increase of CST2/CST1 from 1/12 to 1/4. Fig. 6-3(b) also demonstrates that when a smaller IDATA is used, a higher RSCALE
can be achieved with the constant CST2/CST1.
The current-scaling function is performed so that the large programming current can be reduced to an appropriate value when the pixel operates from the ON- to the OFF-state. In ON-state, the IOLED-ON (=IDATA) are identical in not only the conventional but also the proposed pixel electrode circuits because the external driver directly controls the current, as shown in Fig. 6-4(a). When pixels work in OFF-state, the proposed pixel circuit reveals superior current-scaling ability in comparison with the conventional current-driven pixel electrode circuit which just ideally keeps the IOLED-OFF equivalent to IOLED-ON, as shown in Fig. 6-4(b). It should be noticed that the IOLED-OFF versus IDATA of the conventional pixel circuit changes from linear to curved behavior due to the charge injection phenomenon. This charge injection can occur when the gate voltage is removed, and when the charge carriers in the T2 channel are released and redistributed into the drain and source electrodes. The carrier redistribution will alter the voltages at both nodes. Therefore, the charge injection from T2 causes the IOLED-OFF slightly deviating from IOLED-ON.
0.1 1 10
IOLED-ON of both the proposed and conventional pixel circuts are identical to I
DATA.
(a)
Fig. 6-4. Variation of the I
OLED-ON, I
OLED-OFFand I
AVGduring one frame period versus I
DATA(=I
OLED-ON) at various C
ST2/C
ST1ratio.
From
Fig. 6-4(b), it is obvious that the large C
ST2/CST1 results in significant decrease of the IOLED-OFF. Moreover, since the OFF-state period is much longer than ON-state, the small IOLED-OFF in OFF-state can further reduce the IAVG even if the IOLED-ON is large. Using Eq. 6-3, the plots of IAVG versus IDATA (=IOLED-ON) in one frame period (tON + tOFF) with CST2/CST1 ratios as a parameter are shown in Fig. 6-4(c).For example, the proposed pixel electrode circuit can generate IAVG ranging from 1 nA
to 5 µA with IDATA ranging from 0.1 to 10 µA. By contrast, the IAVG of conventional pixel electrode circuit is almost equal to IDATA. In other words, a very small IAVG can only be achieved by the IDATA having a similar magnitude. From Fig. 6-4, it is evident that IDATA larger than IAVG can be used to program the proposed pixel circuit in ON-state without increasing the poly-Si TFTs geometric size. Hence using an additional CST2 to form a cascade capacitors structure, a large RSCALE can be achieved and a high IDATA can be used to accelerate the pixel circuit programming in ON-state.
0.1 1 10
10-3 10-2 10-1 100 101
Proposed pixel CST2/CST1=1/8
Current-mirror pixel (with scaling function) T4/T3=4/1
I AVG (µA)
IDATA (µA) tON=0.33 mSec, tOFF=33 mSec
Conventional current-driven pixel (without scaling function)
Fig. 6-5. Comparison of I
AVGas a function of I
DATAamong conventional current-driven, current-mirror, and proposed pixels.
To demonstrate the proposed pixel electrode circuit outstanding current scaling function in comparison with both the conventional current-driven and current-mirror
To demonstrate the proposed pixel electrode circuit outstanding current scaling function in comparison with both the conventional current-driven and current-mirror