JP3547687B2 - Method for manufacturing display device - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a liquid crystal display or similar display. The present invention particularly relates to an active matrix display device, a display method thereof, and a manufacturing method thereof. One of the objects of the present invention is a display of a black-and-white display, which does not require a high-speed operation or a high-speed operation such as a gray scale display, but relates to a display that requires visibility and low cost. In particular, a display having such a function is used for a read-only display device such as various information displays.
[0002]
[Prior art]
With the recent miniaturization and power saving of various OA devices, display devices are being replaced by flat panel displays (FPDs) such as liquid crystal displays (LCDs) and plasma displays from conventional cathode ray tubes (CRTs). . In particular, LCDs have been used in portable devices because of their low power consumption.
[0003]
However, LCDs still have many problems to solve. At present, an LCD that is frequently used is called a simple matrix type LCD, and may be called an STN LCD after taking the name of a liquid crystal material. STNLCDs are easy to manufacture, have low cost, and are widely used.
[0004]
However, STN as a liquid crystal material has a problem that the response speed, which is an inherent characteristic of the material, is extremely slow, and when an object moving at high speed is displayed, the object cannot follow the object and cannot be displayed.
[0005]
Further, from the operation method, the time during which one pixel is lit in one frame (usually 10 to 30 msec) is from several tens μsec to several msec. This is inversely proportional to the number of rows of the matrix. In a matrix of 200 rows, one frame is 30 msec and only about 150 μsec is lit. For this reason, there is a drawback that the screen has low contrast and is very difficult to see when the screen is viewed obliquely. Furthermore, if there is a very bright or dark portion on a part of the screen, a phenomenon (crosstalk) that affects the surrounding area occurs.
[0006]
On the other hand, in recent years, LCDs having a method in which each pixel has an active element to switch the pixel by using the active element have been proposed and marketed. These are collectively called active matrix LCDs, but are called TFTLCDs or MIMLCDs depending on the type of active element. The TFT is a thin film transistor, and the MIM is a diode having a metal / insulator / metal structure.
[0007]
In these LCDs, the lighting time of the pixel during one frame is almost equal to one frame, so that the contrast is high and the viewing angle is wide. However, due to technical problems, its production yield is low, its cost and selling price are high, and at present, it is practically used only for high-end computer displays.
[0008]
At present, demand for LCDs is mainly used for portable computers, but a wider range of applications is expected in the future. For example, there is a use as an information display such as a cordless telephone, a display attached to a mobile telephone, or a portable electronic dictionary. In such a case, visibility and low cost are required, and power saving is also required. However, conventional LCDs have not been satisfactory in that respect.
[0009]
For example, although the cost of the STNLCD is low, it is difficult to see from the above problems. In addition, there are two main types of TFTLCD: TFTLCD using TFT using amorphous silicon (hereinafter a-Si TFTLCD) and TFTLCD using TFT using polysilicon (hereinafter polysilicon TFTLCD). Although the former and the latter have no problem in the visibility of the image, the cost cannot be matched with the STNLCD.
[0010]
In particular, when an a-Si TFT LCD is used as a small-sized read-only display, one of the factors that raises the cost is that a driver circuit cannot be built in. Therefore, a driver IC must be connected by a TAB method or the like. Costs will be a major part of the cost increase.
[0011]
FIG. 3 shows the relationship between the number of pixels (number of dots) of the LCD and the cost. This relationship is conceptual and semi-quantitative. In a simple matrix system such as STNLCD, the matrix itself is relatively easy to manufacture, and most of the cost of a small matrix is occupied by the driver IC. That is, the number of driver ICs is proportional to the number of terminals in the matrix, while the number of dots is proportional to the square of the number of terminals. Consequently, the price of the driver IC is proportional to the square root of the number of dots. The cost is dominated by the price of The state is shown in a simple matrix: TAB in the figure.
[0012]
In the a-Si TFT LCD, the production of the matrix is complicated, the yield of the matrix itself is low, and the overall shift is higher than that of the simple matrix. The state is shown in a-Si TFT: TAB in the figure. In the a-Si TFT LCD, the factor occupying the price is different between the small matrix and the large matrix. In the case of a small-scale matrix, the cost of the driver IC occupies a large part of the cost as in the case of the STNLCD. On the other hand, in a large-scale matrix, the cost due to a decrease in the yield of the matrix is a major factor.
[0013]
In a polysilicon TFT LCD, the driver IC can be manufactured simultaneously with the matrix formation by using polysilicon, so that there is no need to mount the IC, and thus the driver IC does not contribute to the cost. Particularly, the mounting of the driver IC has a technical problem, and the purpose of thinking of miniaturization is originally not suitable. Therefore, the polysilicon TFT LCD is also characterized in that it can be miniaturized. However, it is more difficult to fabricate the matrix of the polysilicon TFTLCD than the a-Si TFTLCD, and the cost increases significantly as the number of dots increases. However, since there is no cost factor of the driver IC in the small-scale matrix, as shown in the figure for the complete polysilicon TFT, the cost is competitive with the a-Si TFT LCD.
[0014]
Now, even a-SiTFTLCD, if constituting the drivers a-Si as shown in the dotted line (complete a-SiTFT), can STNLCD and conflict. However, this was not possible with the conventional TFTLCD method. That is, for example, when a relatively small matrix of 160 × 100 is considered, the frame frequency is 30 Hz in a normal operation, and therefore, a signal of 480 kHz is input particularly to the data line driver. However, an a-Si TFT cannot follow such a high-speed operation. The same is true for compound semiconductors such as cadmium selenium (CdSe) based semiconductors. Although these semiconductor materials are not actively used as active devices, they have toxicity and resource problems, but their low response speed is also a serious problem.
[0015]
To avoid this difficulty, the frame frequency may be reduced. In particular, when it is not necessary to display a moving image, it seems that there is no problem in reducing the frame frequency. However, due to the current technical problems of the TFTLCD, the state of frame scanning is visible and the screen is extremely low. It becomes hard to see.
[0016]
FIG. 2 shows a pixel circuit of a TFTLCD using a TN liquid crystal as a conventional liquid crystal material and an operation example thereof. A gate electrode of the TFT is connected to a selection line (also called a gate line), a drain is connected to a data line (also called a drain line), and a source is connected to a pixel electrode. The opposing electrode of the pixel electrode is usually kept at a constant voltage as a common electrode. Generally, it is grounded.
[0017]
As shown in FIG. 2B, a pulse is periodically applied to the selection line, and pixel information is applied to the data line as a voltage signal. The pulse cycle of the selection line is one frame cycle in a normal operation, and is typically 10 to 30 msec. The width of the pulse is approximately equal to or less than the period divided by the number of rows of the matrix. For example, it is 100 to 300 μsec for a relatively small 100-row matrix used for an information display or the like.
[0018]
The signal of the data line is in a voltage state when the pixel is turned on, and is in a non-voltage state when the pixel is turned off. The polarity of the voltage state is periodically changed. This is because if a direct current is applied to the TN liquid crystal material for a long time, the TN liquid crystal material is decomposed due to electrolysis. This operation is called alternating current.
[0019]
Now, the source side of the signal applied TFT of such a signal is shown in V _1. First, the TFT is turned on by application of the pulse of the selection line, and the voltage of the source rises to become the same as the voltage of the drain. However, at the same time as the pulse is cut off, the voltage has an effect of ΔV due to the parasitic capacitance between the gate electrode of the TFT and the source region. Thereafter, the TFT is turned off, so that the pixel electrode is in an electrically floating state, and the voltage gradually decreases due to the leak current of the TFT.
[0020]
Next, a pulse is again applied to the selection line, and when the TFT is turned on, the voltage of the source approaches the negative drain voltage this time. Thereafter, the pulse is cut off, and the voltage shifts negatively by ΔV due to the influence of the parasitic capacitance, and the voltage is attenuated by the leak current. Because when a pulse of the last selection line is applied drain voltage is 0, the release electric charge stored in the pixel electrode, V ₁ is zero.
[0021]
If the frame frequency is reduced, such voltage fluctuations will be visible at the frame frequency. The lowering of the frame frequency is limited to 10 Hz.
[0022]
Another solution is to fabricate only the driver IC using polysilicon. However, in order not to limit the type of the glass substrate, annealing at a high temperature, which is usually performed, cannot be performed. The advanced technology must be adopted. However, laser annealing is a method that has not yet established its technology and is inferior in mass productivity.
[0023]
[Problems to be solved by the invention]
The present invention is to provide a new active matrix system capable of competing with the simple matrix system in terms of cost and realizing the same image quality as the active matrix system in a display device which does not particularly need to display a moving image, and a display device thereof. It is.
[0024]
In particular, the present invention eliminates the need for a driver IC by simultaneously forming a pixel matrix and a peripheral driver circuit using an a-Si TFT or a TFT that can be manufactured at a relatively low temperature, thereby improving the yield and reducing the cost. Is what you do.
[0025]
[Means for Solving the Problems]
As described above, lowering the frame frequency is particularly necessary for a display that does not need to display a moving image when using a TFT using a low mobility semiconductor such as an a-Si TFT or a CdSe-based semiconductor in a driver circuit. This is an important method. However, the reduction in frame frequency must not cause visible image degradation such as flicker.
[0026]
By the way, according to the conventional idea, the meaning of the frame frequency overlaps with the meaning of the frequency of AC conversion and the meaning of the frequency of rewriting. Even if the two were separated, the rewriting frequency was naturally higher than the frequency of the alternating current. Here, pay attention to the word rewriting used in the text. In the text, rewriting does not only mean a change in the display content, but also means that a signal is externally injected or that there is a chance even if the display content is the same. Therefore, in the conventional TFTLCD, the pixels that were lit in one frame can be maintained in the lit state in the next frame. For this reason, a pulse is applied to the selection line and a signal is sent to the data line at the same time. , Is rewritten.
[0027]
As described above, the frame frequency cannot be set to 10 Hz or less due to a visual problem. In the present invention, the above-mentioned problem is solved by clearly distinguishing between AC and rewriting and controlling both independently. It will be clear that if these elements are separated, it is the alternating frequency that affects vision, not the rewriting frequency. For example, in the segment type LCD, the operation of rewriting is substantially different from the operation of AC conversion. In fact, the rewriting frequency of the calculator's LCD is extremely slow. However, the frequency of the alternating current is about 30 Hz. The reason why the display on the LCD flickers due to battery exhaustion or the like is due to a decrease in the AC frequency, not because the rewrite cycle has dropped.
[0028]
Also in the present invention, the frequency of the alternating current must be 10 Hz or more in order to prevent flicker. However, the rewriting frequency needs to be 1 Hz or less.
[0029]
For example, if the rewriting is 1 Hz, the signal sent to the driver of the data line of the LCD of 160 × 100 dots is 16 kHz, which is 1/30 of the conventional value, which is a speed that can be sufficiently driven by the a-Si TFT.
[0030]
Now, in order to achieve such an object, a conventional TFT LCD system is extremely inappropriate. This is because, in the conventional TFT LCD, one TFT plays two roles of selecting a pixel and supplying a voltage to the pixel. Therefore, in the present invention, these two roles are assigned to the respective active elements separately. Here, an element that selects a pixel is a first element, and an element that receives an output of the first element and supplies a voltage to the pixel is a second element.
[0031]
These elements are composed of active elements such as TFTs and various diodes, or passive elements such as resistors and capacitors. At the time of manufacturing these, a-Si or one manufactured under conditions equivalent thereto is desired, and adoption of a high-temperature process at 600 ° C. or higher is avoided.
[0032]
Most simply, as shown in FIG. 1A, two TFTs are a first element (Tr ₁ ) and a second element (Tr ₂ ), respectively. In the present invention, it is necessary to provide a voltage supply line which is not included in the conventional TFTLCD method in the sense that AC is performed regardless of pixel rewriting. For the connection between the wires, to the data lines one of the drain or the source of the Tr ₁ as shown in FIG, connecting the gate electrode to the selection line, the other one thereof is connected to the gate electrode of Tr _2. Further, also the source of the Tr ₂ to one voltage supply line of the drain, the other of the source and the drain is connected to the pixel electrode.
[0033]
The operation of this example will be described below with reference to FIG. Here, for the sake of simplicity, it is assumed that the rewriting is performed once while the exchange is performed twice. Of course, the case where the rewriting is performed once while the AC conversion is performed ten times, or the case where the rewriting is performed once while the AC conversion is performed 30 times can be similarly expanded.
[0034]
In this example, it is assumed that the pixel which has been turned off at first is turned on and then turned off again at the next rewriting. The select line V _G, a pulse as in the prior art are regularly applied. On the other hand, necessary signals are also applied to the data lines. The signal applied to the data line has two values, positive and negative, or a voltage state and a non-voltage state. Here, it is assumed that both Tr ₁ and Tr ₂ are NMOS. Further, the potential of the counter electrode of the pixel is set to 0.
[0035]
When a pulse is first applied to the selection line, the signal on the data line is positive, so the potential V ₁ on the source side of Tr ₁ becomes a positive value, and the voltage increases as in the case of the conventional TFTLCD. Then, it falls by the end of the pulse, and thereafter discharges spontaneously. The time required for this discharge is determined by the OFF resistance of Tr ₁ and the capacitance C between the gate electrode of Tr ₂ and the channel. For example, in an a-Si TFT, the OFF resistance is about 10 ¹³ Ω, and the C is about 10 ⁻¹³ F, so that the attenuation constant is about 1 second. That is, the voltage is reduced to about 40% after one second. This time can be extended by making C larger.
[0036]
On the other hand, a signal synchronized with the pulse of the selection line is sent to the voltage supply line, and an AC pulse is sent to this voltage supply line as shown in FIG. Here, for each selection pulse, the signal polarity of the voltage supply line changes twice between positive and negative. Of course, the polarity may be changed more for one selection pulse.
[0037]
Since the gate electrode of Tr ₂ already takes a positive voltage, Tr ₂ is turned ON, the voltage of the voltage supply line is directly applied to the pixel electrode, the voltage V ₂ of the pixel electrode, FIG. 1 (B First, as shown in ()), it takes a negative value, and then takes a positive value as the voltage of the voltage supply line is inverted. It can be said that this is a feature of the present invention. By such a two-stage operation, a voltage substantially equal to the voltage of the voltage supply line is supplied to the pixel, and this is achieved by natural discharge as in the prior art. It does not decrease. Therefore, black and white is clearly discriminated.
[0038]
Next, a pulse is again applied to the selection line. Since this time, the voltage of the data line is 0, the charge stored in C is discharged, V ₁ is zero. As a result, Tr _{2 is} also turned off, and the supply of the voltage to the pixel is stopped.
[0039]
In the related art, since the cycle of the AC conversion is the same as or longer than the cycle of the rewriting, a pulse as shown by a dotted line must be applied to the selection line. However, with the present invention, the pulse is not needed and the operating signal is halved.
[0040]
Considering further the effect of the present invention, if the rewriting is performed once a second, for example, by a method similar to that of FIG. 1, this is 1/30 of the conventional speed. This means that the pulse applied to the selection line and the signal on the data line can be 30 times longer. For example, in the case of a pulse of a selection line, conventionally, it was about 100 μsec for a matrix of 200 rows, but in the present invention, it can be set to 30 msec, 3 msec. This means that the required voltage can be charged and supplied in a reliable manner even if the operation of the TFT is slow. Conventionally, the response of the a-Si TFT was a short time in which the operation of the TFT was difficult, so that variations in the characteristics of each TFT resulted in pixels that were sufficiently charged and pixels that were not sufficiently charged, leading to deterioration in image quality.
[0041]
In the present invention, a semi-analog voltage is not applied to the pixel by the operation of the two-stage TFT, but such a feature further reduces TFT defects and contributes to an improvement in yield.
[0042]
In this description, it is desirable to use an a-Si TFT as the TFT. An NMOS a-Si TFT may be used for both, but an enhancement type TFT may be used for Tr ₁ and a depletion type TFT may be used for Tr ₂ . When an a-Si TFT is used, a PMOS is not suitable for the purpose because its operation speed is extremely slow. However, in a silicon semiconductor in an intermediate state between amorphous silicon and polysilicon, the mobility of holes is considerably large, so that a PMOS can be used. In that case, the peripheral circuit can also be CMOS.
[0043]
FIG. 4 shows an example of the overall configuration of the apparatus of the present invention. The number of dots of this LCD is, for example, 320 × 480 (half the screen of a normal laptop computer). However, the screen is roughly divided into four parts, up, down, left, and right, and each is driven by a driver (401) for four selection lines and voltage supply lines arranged beside the LCD matrix (406). Further, each of the four divided screens is further divided into halves, and driven by data line drivers (402) provided above and below. Each driver is connected from a wire bonding terminal (403) to an external circuit by a wiring (405) connected by a wire bonding method.
[0044]
For example, paying attention to the lower left screen, the pixel located here is 19200, which is one eighth of the whole. If the operation of rewriting only once per second is performed, the frequency of the signal transmitted from the wiring 405 to the data line driver 402 is 19.2 kHz. Also, as for the signals sent to the selection line and the voltage supply line, it is necessary that a signal of 30 Hz is sent to the voltage supply line at least for one row, and the number of rows is 120 half of 240 rows (the other 120 rows). Is handled by the driver on the opposite side), so a 3.6 kHz signal is sent. In each case, the frequency is extremely low, and there is almost no problem even if the driver is constituted by an a-Si TFT.
[0045]
Further, when the driver circuits are formed simultaneously with the matrix as described above, the reduction in yield due to the driver circuits can be almost ignored.
The capacitance C is especially problematic between the gate electrode and the channel of the Tr ₂ in the present invention. As mentioned earlier, in order to maintain the potential of _{V 1,} OFF resistance and C Tr ₁ becomes its parameters. The OFF resistance of the TFT can be changed to some extent by changing the thickness and width of the channel. However, it is difficult to achieve a high resistance of 10 ¹³ Ω or more. Meanwhile, C is is determined by the size of the gate electrode of Tr _2. For example, if the thickness of the insulating film is 100 nm for a gate electrode of 10 × 100 μm ² , C is 10 ⁻¹³ to 10 ⁻¹² F. If silicon nitride having a high dielectric constant is used as the insulating film, C increases.
[0046]
Use of an area of 10 × 100 μm for the gate electrode of Tr ₂ leads to a decrease in aperture ratio, which is not desirable. In fact, it is not advisable to devote much more area to the TFT. Therefore, to resolve this conflict, the voltage supply line, the overlapping source electrode and wiring Tr ₁ may. In this case, a large capacity can be obtained without lowering the aperture ratio. In this case, one of the methods is to use a material having a large dielectric constant for the interlayer insulator.
[0047]
Since a large C is connected to Tr ₁ as described above, there may be a person who is concerned about a decrease in the ON / OFF operation speed of Tr ₁ . However, in the present invention, the signal of each data line also lasts much longer than the pulse of the selection line, for example, 30 times longer. On the other hand, in the conventional TFTLCD, the capacitance of the pixel electrode as a load was about 10 ⁻¹³ F. In the case of the present invention, a load capacity that is as large as that of the related art or about one order of magnitude is required. However, since the response speed is reduced to one tenth or less, there is no problem at all, and there is more margin than in the related art. May be able to respond and operate.
[0048]
According to the present invention, when rewriting (including maintaining) the display, there is a method of performing the rewriting every suitable number of lines in accordance with the timing of the AC conversion. For example, a method as shown in FIG. For example, let's say you have a 100 row matrix. It is assumed that the same signal is applied to the voltage supply lines of five rows, that is, the first row, the 21st row, the 41st row, the 61st row, and the 81st row. Similarly, it is assumed that five rows, ie, the second row, the 22nd row, the 42nd row, the 62nd row, and the 82nd row, and the other rows also form a set, and operate in synchronization with each other.
[0049]
At the time of the first AC conversion (FIG. 5A), it is assumed that the pixels in the first to twentieth rows are rewritten. At this time, a pulse is applied to the selection line and a positive voltage is applied to the voltage supply line to the pixels in the first row. On the other hand, a voltage is applied to the voltage supply line but no pulse is applied to the selection line for the other pixels that move in synchronization with the 21st row and the other first rows. Therefore, at this time, rewriting is performed only on the first row among the rows operating as a set of five. The same applies to the other sets. In this case, only the first to twentieth rows are rewritten.
[0050]
Next, it is assumed that a negative voltage is applied to the voltage supply line simultaneously with a pulse to the selection line in the 21st row. However, at this time, no pulse is applied to the selection line in the first row and the other rows that operate synchronously. A voltage is applied to the voltage supply line in the same manner as in the 21st row. The same applies to other sets of rows, as shown in FIG. 5B, where only the 21st to 40th rows are rewritten.
[0051]
Thereafter, the same operation is repeated. In FIG. 5C, lines 41 to 60 are rewritten. At this time, a positive voltage is applied to the voltage supply line. In FIG. 5D, the data is rewritten from the 61st line to the 80th line. At this time, a negative voltage is applied to the voltage supply line. In FIG. 5E, the 81st to 100th rows are rewritten. At this time, a positive voltage is applied to the voltage supply line.
[0052]
In this way, in FIG. 5F, the first to twentieth rows are rewritten again. At this time, the voltage applied to the voltage supply line is negative. Each pixel has been rewritten once between FIGS. 5A to 5E, but the voltage of the pixel has changed five times, such as positive, negative, positive, negative, and positive. This is exactly what should be the feature of the present invention. That is, the rewriting cycle is longer than the alternating cycle. In particular, in the present invention, the load on the driver circuit is significantly reduced by setting the ratio of the period to 30 times or more.
[0053]
Now, in the present invention, the power for driving the LCD can be reduced. In a conventional TFTLCD or STNLCD, the frequency of a signal output to each data line is (number of rows × 30) Hz. However, in the present invention, if rewriting is performed only once per second, for example, the frequency is (the number of rows × 1) Hz.
[0054]
On the other hand, in the conventional LCD, the frequency of the signal output to each selection line is 30 Hz, whereas in the present invention, it is 1 Hz. However, in the present invention, since a signal of 30 Hz is output to the voltage supply line, this point is almost equal to the conventional one.
[0055]
As a result, power consumption can be reduced by reducing the number of data line signals. Further, in the conventional STNLCD, since the operation was performed in the dynamic mode, it was necessary to illuminate the screen with a backlight in order to make the screen easily viewable. However, in the present invention, since the operation was performed in the static mode, the backlight was operated. Good visibility can be obtained without it.
[0056]
In order to implement the present invention, a known thin film semiconductor manufacturing technique may be used. The details thereof will not be described one by one, but examples will be shown and described below.
[0057]
【Example】
FIG. 6 shows an example of a pixel driving circuit for implementing the present invention and a manufacturing method thereof. This shows a state when the pixel circuit is viewed from above. The circuit of this embodiment has an inverted stagger type double TFT with triple metal wiring. Such a circuit may be manufactured as follows.
[0058]
First, a selection line (which becomes a gate electrode and a wiring of the Tr ₁ ) 601 made of a metal material such as aluminum is patterned on an appropriate substrate (mask 1). At this time, if a metal oxide film with good insulating properties is formed on the surface of the selection line by an anodic oxidation method or the like, the probability of occurrence of a defect in a later process is reduced. Then, a first insulating layer functioning as a gate insulating film and an interlayer insulating film is formed. Next, an amorphous silicon or polysilicon film is formed by a CVD method or the like, and is patterned (mask 2). Next, using the mask 1, an etching stopper such as a silicon nitride film is formed so as to overlap the selection line. Alternatively, light may be irradiated from the back surface of the substrate to pattern the etching stopper in a self-aligned manner so as to overlap the selection line.
[0059]
Next, a semiconductor film doped with impurities is formed and patterned (mask 3). Thus, the semiconductor region 602 of the first TFT is manufactured. FIG. 6A shows this state.
[0060]
Next, the data line 603 is formed of a metal material. The data line is formed so as to be connected to one of the source and the drain of the first TFT (mask 4). At the same time, a wiring 604 extending from an electrode connected to the other of the source and the drain of the first TFT is formed using the same material. At this time, the reason why the metal wiring 604 presents such a complicated calculation is to make it overlap with the voltage supply line later. FIG. 6B shows this state.
[0061]
Further, as in the case of manufacturing the first TFT, a second insulating film (to be a gate insulating film of the second TFT) is formed, and the activated semiconductor film of the second TFT is patterned (mask). 5) Next, using the mask 4, an etching stopper is formed, and an impurity-doped semiconductor film is formed and patterned (mask 6). Thus, the semiconductor region 605 of the second TFT is formed. Further, a voltage supply line 606 is formed of a metal material (mask 7), and a contact is formed with one of the source and the drain of the second TFT. Thus, a circuit as shown in FIG. 6C is obtained. Finally, as shown in FIG. 6D, the transparent conductive film 607 is patterned (mask 8) to complete the circuit.

[0062]
The above steps require a total of eight masks, and require ten mask processes. In order to actively reduce the mask process, it is desirable to introduce a self-alignment process. In order to form a TFT without using an etching stopper, first, an impurity semiconductor to be a source / drain region may be formed by patterning, and thereafter, an activated semiconductor film may be formed.
[0063]
In this circuit, the voltage supply line and the gate electrode wiring of the second TFT are designed to intentionally overlap. This means that the capacity of both of them (corresponding to C in FIG. 1A) is increased, the retention time of the electric charge accumulated in the gate electrode of the second TFT is lengthened, and the number of times of rewriting is reduced. Because it was intended.
[0064]
【The invention's effect】
According to the present invention, it is possible to provide an LCD which is equivalent to an active matrix type such as a TFT LCD in viewability and which can be competitive in price with the STNLCD type.
[0065]
An object of the present invention is to provide an LCD used for a display device that does not need to display a moving image. For example, as an accessory of an electric device, a device used for displaying a method of operating the device or an operation state of the device is used. Traditionally, such applications have been very limited and the market has been small. Conventionally, a segment type LCD or SETN LCD has been used as a read-only display.
[0066]
However, the segment system has a limited display capacity. In addition, in STNLCD, it was necessary to mount a driver IC. At present, the TAB method is generally used as a technique for mounting such an IC, but it becomes technically difficult to adopt the TAB method as the pixels become smaller. Generally, when one side of a pixel is less than 100 μm, the TAB method cannot be used.
[0067]
In the present invention, such a problem does not occur because the driver IC is also integrally formed. However, in the conventional a-Si TFT LCD, it was difficult to configure the driver IC with the a-Si TFT due to the difficulty in the operation method. The present invention has successfully solved this point.
[0068]
According to the present invention, a completely new use of a read-only LCD is expected. For example, the present invention does not require an external IC, and thus can be extremely miniaturized. Therefore, it can be used for a card type display device. For example, it can be used for a card type pager, a display device for various credit cards, and the like. Although such uses are expected, they cannot be put to practical use because there are no suitable display devices and LCDs. At present, the market scale for such purposes is small, but it is expected that there will be enormous potential demand, and it is expected to grow into a large market.
[0069]
In the present invention, it is desirable to use a material manufactured at a low temperature of 600 ° C. or lower as a material of the TFT. Although the a-Si TFT is taken up in the embodiment, there is no particular problem even if it is a compound semiconductor such as CdS or CdSe.
[Brief description of the drawings]
FIG. 1 shows a circuit example of a pixel of a TFTLCD of the present invention and an operation example thereof.
FIG. 2 shows a circuit example of a pixel of a conventional TFTLCD and an operation example thereof.
FIG. 3 schematically shows the relationship between the number of pixels and the cost of various LCDs.
FIG. 4 shows a configuration example of a panel of the TFTLCD of the present invention.
FIG. 5 shows an example of a display method of the TFTLCD of the present invention.
FIG. 6 shows an example of a circuit of a pixel of a TFTLCD of the present invention and an example of a manufacturing method thereof.
[Explanation of symbols]
401 selection line / voltage supply line driver circuit 402 data line driver circuit 403 bonding pad 405 bonding wire 406 matrix area

Claims (3)

Drive having a first element and a pixel formed on and a substrate and a second element formed of amorphous silicon, the pixels and formed on the substrate simultaneously, the element formed of amorphous silicon A method for manufacturing a display device having a circuit ,
Forming a gate electrode of a first element connected to a selection line on the substrate;
Forming a gate insulating film of the first element on the gate electrode of the first element ;
Forming an amorphous silicon film of the first element on the gate insulating film of the first element ;
Forming a silicon nitride film on the amorphous silicon film of the first element ;
Irradiating light from the back surface of the substrate to pattern the silicon nitride film,
By patterning to form an impurity-doped semiconductor film patterned the silicon nitride film, forming a source and a drain of the first element,
Forming a data line connected to one of the source or the drain of the first element, and simultaneously forming a metal wiring connected to the other of the source or the drain of the first element;
Forming a gate insulating film of a second element on the metal wiring;
Forming an amorphous silicon film of the second element on the gate insulating film of the second element;
Forming an etching stopper on the amorphous silicon film of the second element;
Forming a source and drain of the second element by forming and patterning a semiconductor film doped with impurities on the etching stopper;
Forming a voltage supply line connected to one of a source and a drain of the second element;
Forming a pixel electrode connected to the other of the source and the drain of the second element;
The method for manufacturing a display device, wherein the metal wiring is a gate electrode wiring of a second element and overlaps with the voltage supply line . Drive having a first element and a pixel formed on and a substrate and a second element formed of amorphous silicon, the pixels and formed on the substrate simultaneously, the element formed of amorphous silicon A method for manufacturing a display device having a circuit ,
Forming a gate electrode of a first element connected to a selection line on the substrate;
Forming a gate insulating film of the first element on the gate electrode of the first element ;
Forming an amorphous silicon film of the first element on the gate insulating film of the first element ;
Forming a film serving as an etching stopper on the amorphous silicon film of the first element ;
By irradiating light from the back surface of the substrate, the film serving as the etching stopper is patterned to form an etching stopper of the first element ,
Forming a source and drain of the first element by forming and patterning a semiconductor film doped with impurities on the etching stopper of the first element ;
Forming a data line connected to one of the source or the drain of the first element, and simultaneously forming a metal wiring connected to the other of the source or the drain of the first element;
Forming a gate insulating film of a second element on the metal wiring;
Forming an amorphous silicon film of the second element on the gate insulating film of the second element;
Forming an etching stopper for the second element on the amorphous silicon film of the second element;
Forming a source and drain of the second element by forming and patterning a semiconductor film doped with impurities on the etching stopper of the second element;
Forming a voltage supply line connected to one of a source and a drain of the second element;
Forming a pixel electrode connected to the other of the source and the drain of the second element;
The method for manufacturing a display device, wherein the metal wiring is a gate electrode wiring of a second element and overlaps with the voltage supply line . It said first element and said second element, a method for manufacturing a display device according to claim 1 or claim 2, characterized in that an inverse stagger type thin film transistor.

Images