A Pixel Circuit for Compensating Electrical Characteristics Variation and OLED Degradation

Abstract

In recent years, the active-matrix organic light-emitting diode (AMOLED) displays have been greatly required. A voltage compensation pixel circuit based on an amorphous indium gallium zinc oxide thin-film transistor is presented for AMOLED displays. The circuit is composed of five transistors–two capacitors (5T2C) in combination with an OLED. In the circuit, the threshold voltages of both the transistor and the OLED are extracted simultaneously in the threshold voltage extraction stage, and the mobility-related discharge voltage is generated in the data input stage. The circuit not only can compensate the electrical characteristics variation, i.e., the threshold voltage variation and mobility variation, but also can compensate the OLED degradation. Furthermore, the circuit can prevent the OLED flicker, and can achieve the wide data voltage range. The circuit simulation results show that the OLED current error rates (CERs) are lower than 3.89% when the transistor’s threshold voltage variation is ±0.5V, lower than 3.49% when the mobility variation is ±30%.

1. Introduction

OLEDs (organic light-emitting diodes) have gained widespread attention for their advantages such as low power consumption, high contrast, fast response time, thinner, and more foldable characteristics. According to the different driving methods, OLED driving technology can be divided into PMOLEDs (passive-matrix OLEDs) and AMOLED (active-matrix OLEDs). PMOLED has the advantages of a simple structure and low cost. However, PMOLED requires a larger driving voltage, and its power consumption is significantly higher than the AMOLED. As shown in Figure 1, AMOLEDs use independent thin-film transistors to control each pixel, so that each pixel can be continuously and independently driven and lit. Therefore, AMOLED is suitable for large and high-resolution displays, and has high application prospects for the displays [1,2,3,4].

In AMOLED pixel circuits, the oxide thin-film transistor (TFT) has great advantages, such as high carrier mobility, high light transmittance, good uniformity, and low off current, so it is widely applied to drive OLED, especially for large size AMOLED [5,6,7,8,9].

On the one hand, the TFT’s electrical characteristics variation, i.e., the threshold voltage variation and mobility variation would lead to the OLED current change. On the other hand, the OLED degradation would also lead to the OLED current change. The OLED current change brings out the uniformity of displays [10,11]. Therefore, in AMOLED displays, the circuit is needed to compensate the TFT’s electrical characteristics’ variation and the OLED degradation. Generally, the compensation circuit is divided into the current compensation circuit and the voltage compensation circuit.

The current compensation circuit can successfully compensate the TFT’s electrical characteristics’ variation [12,13]. However, the compensation speed is relatively slow at low gray level. This problem could be solved by the voltage compensation circuit [14,15,16].

Therefore, lots of valuable voltage compensation pixel circuits have been reported [17,18,19,20,21,22,23,24,25,26,27,28,29,30].

In [17,18,19,20,21], the circuits can compensate the threshold voltage variation successfully. In [22,23,24,25], the circuits can compensate both threshold voltage and mobility variations successfully. In [26,27,28], the circuits can compensate both threshold voltage variation and OLED degradation successfully. To obtain higher uniformity of displays, it would be better if the circuits can compensate the above three items.

In [29], the circuit can compensate the above three items successfully, but the circuit cannot prevent the OLED flicker. In [30], the circuit not only can compensate the above three items, but also can prevent the OLED flicker successfully. However, the data voltage must be less than the OLED threshold voltage. Thus, the data voltage range is limited.

In this paper, a voltage compensation pixel circuit is proposed. In the threshold voltage extraction stage, the circuit can extract the threshold voltages of both the TFT and the OLED simultaneously. In the data input stage, the circuit can generate the mobility-related discharge voltage. The simulation results show that the circuit can compensate three items: the threshold voltage variation, the mobility variation, and the OLED degradation, can prevent the image flicker, and can achieve the wide data voltage range.

2. Materials and Methods

The circuit structure and the driving schematic diagram are shown in Figure 2. As shown in Figure 2a, the circuit consists of one driving TFT (T2), four switching TFTs (T1, T3, T4, T5), and two capacitors (C1, C2).

As shown in Figure 2b, the driving schematic diagram contains four stages: (1) the initialization stage, (2) the threshold voltage extraction stage, (3) the data input stage, and (4) the emission stage.

The working principle of the circuit is described as follows.

2.1. Initialization Stage

In the initialization stage, as shown in Figure 2b, SCAN1, SCAN3, and SCAN4 are high. SCAN5 is low. Therefore, T1, T3, and T4 are turned on. T5 is turned off. The schematic of the circuit in this stage is shown in Figure 3a.

Because T1 and T4 are turned on, the voltage of node B is charged to VDD.

Because T3 is turned on, no current flows through the OLED. Therefore, the OLED flicker is prevented.

2.2. Threshold Voltage Extraction Stage

In the threshold voltage extraction stage, as shown in Figure 2b, SCAN1, SCAN3, and SCAN4 go to low. SCAN5 remains low. Therefore, T1, T3, T4, and T5 are turned off. The schematic of the circuit in this stage is shown in Figure 3b.

Because T1 is turned off, no current flows through the OLED. Therefore, the OLED flicker is prevented.

The voltage of node B is gradually discharged until T2 is turned off. The voltage of node B goes to

$$V_B=V_{\mathit{TH}\_T2}+V_{\mathit{TH}\_OLED}.$$

(1)

Consequently, the threshold voltages of both T2 and OLED are extracted simultaneously in this stage.

2.3. Data Input Stage

In the data input stage, as shown in Figure 2b, SCAN1, SCAN3 and SCAN4 remain low; SCAN5 goes to high. Therefore, T1, T3, and T4 are turned off; T5 is turned on. The schematic of the circuit in this stage is shown in Figure 3c.

Because T1 is turned off, no current flows through the OLED. Therefore, the OLED flicker is prevented.

In [29], to prevent OLED flicker, the data voltage range is limited: it is much less than the OLED threshold voltage. In this paper, the above limitation is avoided.

At the beginning and the end of this stage, t is defined as t₀ and t₀ + T, respectively. They are indicated in Figure 2b.

At the time t₀, the data voltage (V_DATA) is input to the circuit; V_C and V_B are expressed as

$$V_C\left(t=t_0\right)=V_{DATA}+V_{TH\_OLED},$$

(2)

$$V_B\left(t=t_0\right)=V_{T{H}_{T2}}+V_{TH\_OLED}.$$

(3)

After the time T, V_C remains unchanged; V_B discharges through C1, C2, and T2. At the time t₀ + T, V_C and V_B are expressed as

$$V_C\left(t=t_0\right)=V_{DATA}+V_{TH\_OLED},$$

(4)

$$V_B\left(t=t_0\right)=V_{TH\_T2}+V_{TH\_OLED}-\left(V_{DATA}+V_{TH\_OLED}\right)\times \frac{C2}{C1+C2}-\Delta V\mu .$$

(5)

where $\Delta V\mu$ is the discharged voltage related to the mobility of T2.

The expression of $\Delta V\mu$ is derived as follows. When DATA is input to the circuit, T2 keeps the diode-connected structure. Therefore, the mobility-related discharge voltage $\Delta V\mu$ is stored in C1. By the law of discharge conservation, we have [30]

$$\left(C1+C2\right)\frac{d{V}_{GS\_T2}}{dt}=\frac{1}{2}\mu C_{OX}\frac{W}{L}{\left(V_{GS\_T2}-V_{TH\_T2}\right)}^2.$$

(6)

where $\mu$ is the mobility of T2, C_OX is the gate oxide capacitance per unit area, and $\frac{W}{L}$ is the width–length ratio of T2.

Integrating (6), we have

$${{\displaystyle \int }}_{V_{GS\_T2}\left(t=t_0\right)}^{V_{GS\_T2}\left(t=t_0+T\right)}\frac{1}{{\left(V_{GS\_T2}-V_{TH\_T2}\right)}^2}d{V}_{GS\_T2}={{\displaystyle \int }}_{t=t_0}^{t=t_0+T}\frac{\mu C_{OX}\frac{W}{L}}{2\left(C_1+C_2\right)}dt.$$

(7)

where

$$V_{GS\_T2}\left(t=t_0\right)=V_B\left(t=t_0\right)-V_C\left(t=t_0\right),$$

(8)

$$V_{GS\_T2}\left(t=t_0+T\right)=V_B\left(t=t_0+T\right)-V_C\left(t=t_0+T\right).$$

(9)

Substituting (2)–(5) to (9), we obtain

$$\Delta V\mu =\left(V_{DATA}+V_{TH\_OLED}\right)\times \frac{C2}{C1+C2}-\frac{1}{\frac{\mu C_{OX}\frac{W}{L}}{2\left(C_1+C_1\right)}T+\frac{1}{\left(V_{DATA}+V_{TH\_OLED}\right)\times \frac{C2}{C1+C2}}}.$$

(10)

Consequently, the mobility-related discharge voltage $\Delta V\mu$ is generated in this stage.

2.4. Emission Stage

In the emission stage, as shown in Figure 2b, SCAN1 goes to high, SCAN3 and SCAN4 remain low, and SCAN5 goes to low. Therefore, T1 is turned on, and T3, T4, and T5 are turned off.

The schematic of the circuit in this stage is shown in Figure 3d.

The driving TFT (T2) operates in the saturation region; the OLED current is expressed as follows:

$$I_{OLED}=\frac{1}{2}{\mu C}_{\mathit{OX}}\frac{W}{L}{\left(V_{\mathit{GS}\_T2}-V_{\mathit{TH}\_T2}\right)}^2.$$

(11)

Substituting (9) to (11), we obtain

$$I_{OLED}=\frac{1}{2}\mu C_{OX}\frac{W}{L}{(\left(V_{DATA}+V_{TH\_OLED}\right)\times \frac{C2}{C1+C2}-V_{DATA}-\Delta V\mu )}^2.$$

(12)

Substituting (10) to (12), we obtain

$$I_{OLED}=\frac{1}{2}\mu C_{OX}\frac{W}{L}{(\frac{1}{\frac{\mu C_{OX}\frac{W}{L}}{2\left(C_1+C_1\right)}T+\frac{1}{\left(V_{DATA}+V_{TH\_OLED}\right)\times \frac{C2}{C1+C2}}}-V_{DATA})}^2.$$

(13)

From (12), it is found that the OLED current is independent of the threshold voltage V_{TH_T2}. That is, when V_{TH_T2} varies, I_OLED remains stable. Therefore, the circuit can compensate the threshold voltage variation.

From (10), it is found that when the mobility $\mu$ increases, $\Delta V\mu$ will increase, and vice versa. Consequently, in (12), when the mobility varies, I_OLED remains stable. Therefore, the circuit can compensate the mobility variation. This point also can be explained by (13). In (13), when $\mu$ increases, both $\frac{1}{2}\mu C_{OX}\frac{W}{L}$ and $\frac{\mu C_{OX}\frac{W}{L}}{2\left(C_1+C_1\right)}T$ will increase; thus, I_OLED remains stable, and vice versa.

From (13), it is found that the OLED current is positively correlated with V_{TH_OLED}. Therefore, the circuit can compensate the OLED degradation [29,30,31].

3. Results and Discussions

In the circuit simulation, to evaluate the compensation performance, the SPICE model (level = 35) is used for the oxide TFTs. The TFTs’ threshold voltage and mobility are 1.5 V and 50 cm²/V, respectively. The TFTs’ threshold voltage variation and mobility variation are ±0.5 V and ±30%, respectively [9,14,28]. The OLED model is equivalent to a TFT and a C_OLED in parallel [22,27,32]. The oxide TFT and the OLED models are verified by the experimental data [8,33], which are shown in Figure 4.

The values of the design parameters are shown in

Table 1. Design parameters of the circuit.

Parameters (Unit)	Value	Parameters (Unit)	Value
WT2 (µm)	3	VDD (V)	5
LT2 (µm)	3	C2 (pF)	0.2
WT1,T3,T4,T5 (µm)	22	COLED (pF)	0.2
LT1,T3,T4,T5 (µm)	3	SCAN1,3,4,5 (V)	−5~5
C1 (pF)	0.4	DATA (V)	VDATA~0

; the range of the values is reasonable, which is consistent with the previous pixel circuit applications [11,19,20,27,32,34].

Figure 5a shows the transient waveforms of V_B, i.e., the gate voltage of T2, at V_DATA = −4 V. It is found that when ΔV_{TH_T2} = ±0.5 V, ΔV_B approximates ±0.5 V in the threshold voltage extraction stage, i.e., V_B senses the threshold voltage variation successfully. Figure 5b shows the transient waveforms of the OLED current I_OLED. It is found that I_OLED = 0 except for the emission stage, i.e., the OLED flicker is prevented. In the emission stage, when ΔV_{TH_T2} = −0.5, 0, and +0.5 V, the transient waveforms of I_OLED = 94.11, 90.59, and 87.06 n A, respectively. The current error rates (CERs) are 3.74% and 0, 3.89%, respectively. Thus, the circuit compensates ΔV_{TH_T2} successfully.

The CER for ΔV_{TH_T2} is defined as

$$\frac{I_{OLED}\left({\Delta V}_{\mathit{TH}\_T2}=0\right)-I_{OLED}\left({\Delta V}_{\mathit{TH}\_T2}=\pm 0.5 \mathrm{V}\right)}{I_{OLED}\left({\Delta V}_{\mathit{TH}\_T2}=0\right)}\times 100\%.$$

(14)

Figure 5c shows the transient waveforms of CER vary when ΔV_{TH_T2} is +0.5 V and −0.5 V, respectively. It is found that the CERs are less than ±9.59% within the whole data range. Thus, the threshold voltage variation ΔV_{TH_T2} is compensated successfully.

Figure 6a shows the transient waveforms of V_B at V_DATA = −4V. It is found that when Δu= ±30%, the variation of V_B is similar to Δu, i.e., V_B senses the mobility variation successfully. Figure 6b shows the transient waveforms of the OLED current I_OLED. It is found that I_OLED = 0 except for the emission stage, i.e., the OLED flicker is prevented. In the emission stage, when Δu = −30, 0, and +30%, the transient waveforms of I_OLED = 88.06, 90.59, and 93.87 n A, respectively. The current error rates are 2.81% and 0, 3.49%, respectively. Thus, the circuit compensates Δu successfully.

The CER for Δu is defined as

$$\frac{I_{OLED}\left(\Delta \mathrm{u}=0\right)-I_{OLED}\left(\Delta \mathrm{u}=\pm 30\%\right)}{I_{OLED}\left( \Delta \mathrm{u}=0\right)}\times 100\%.$$

(15)

Figure 6c shows the transient waveforms of CER varies when Δu is +30% and −30%, respectively. It is found that the CERs are less than ±9.28% within the whole data range. Thus, the mobility variation Δu is compensated successfully.

In this paper, the circuit can compensate the OLED degradation. It is explained as follows. For the long time operation, the OLED luminance degrades while V_{TH_OLED} increases [29,30,31]. Therefore, I_OLED (13) increases. The increase in I_OLED brings about the increase in the OLED luminance. Thus, the OLED luminance degradation is compensated.

In Figure 5a,b and Figure 6a,b, the time of the third stage, i.e., the data input stage, is set to 3.8 us. It is suitable for the 8K4K ultrahigh definition (7680 × 4320, UHD) for high-performance display [11,35,36].

As shown in

Table 2. Comparison between this paper and the previous publications.

Publications	Structure	VTH Compensation	µ Compensation	OLED Degradation	Prevent Image Flicker
2015 [22]	5T2C	√	√	-	√
2015 [27]	5T2C	√	-	√	√
2015 [28]	4T1C	√	-	√	√
2016 [30]	4T2C	√	√	√	√
2017 [26]	4T1C	√	-	√	√
2018 [23]	6T2C	√	√	-	√
2018 [29]	5T2C	√	√	√	-
2020 [19]	6T1C	√	-	-	√
2020 [20]	9T2C	√	-	-	√
2022 [17]	6T2C	√	-	-	√
This paper	5T2C	√	√	√	√

, the valuable publications are compared with this paper. In [29], the circuit can compensate the threshold voltage variation, the mobility variation, and the OLED degradation successfully, but the circuit cannot prevent the OLED flicker. In [30] and this paper, the circuits not only can compensate the above three items successfully, but also can prevent the OLED flicker.

However, in [30], V_DATA must be less than V_{TH_OLED}, that is, the range of V_DATA is limited. In this paper, the range of V_{DATA is} not limited by V_{TH_OLED}. Therefore, the circuit achieves the wide data voltage range.

Figure 7 shows the layout structure of the circuit. SCAN1, SCAN3, SCAN4, and SCAN5 are transverse lines, which are set to 4 um. These transverse lines are shared by the pixels of the same row. VDD, GND, and DATA are the vertical lines, which are set to 6 um. VDD and GND are shared by the entire display panel. DATA is shared by the pixels of the same column. The total layout area is 180 um × 110 um. The proposed layout achieves an aperture ratio of 39.14%.

4. Conclusions

The pixel circuit is presented for improving AMOLED displays uniformity. In the threshold voltage extraction stage, the threshold voltages of the driving TFT and the OLED are extracted. In the data input stage, the discharge voltage related to mobility is generated. Consequently, the circuit not only compensates the threshold voltage variation, the mobility variation, and the OLED degradation, but also prevents the image flicker and achieves the wide data voltage range.