JPH07104311A - Liquid crystal display device - Google Patents

Abstract

PURPOSE:To provide the liquid crystal display device for which a redundancy structure is adopted and with which a high yield is obtainable and a difference in display characteristics between normal pixels and pixels after repair is made small. CONSTITUTION:A first thin-film transistor(TFT) 101 has a cut part 111 which is electrically disconnectable from a pixel electrode 105. A second TFT 102 is provided with a juncture 109 which is electrically connectable to the pixel electrode 105 in a source electrode route. The second TFT 102 decreases a difference in the capacitance between a scanning line and the pixel electrode in the state that the first TFT 101 is electrically connected to the pixel electrode 105 and the capacitance between the scanning line and the pixel electrode in the state that the first TFT 101 is electrically disconnected from the pixel electrode 105 and in turn the second TFT 102 is electrically connected to the pixel electrode 105 by using the parasitic capacitance of the juncture 109.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a liquid crystal display device.

[0002]

2. Description of the Related Art In recent years, a new display device which replaces the cathode ray tube display device has been actively developed. Among them, the liquid crystal display device is applied to various fields because it is thin and can operate at low power.

Among liquid crystal display devices, active matrix type display devices having excellent display characteristics are expected.
In particular, those using thin film transistors (hereinafter abbreviated as TFT) as switching elements are expanding the market in the field of small televisions. Recently, products such as 10 to 20-inch TVs and projection TVs have been developed with the aim of increasing their size and definition. ,
FIG. 9 shows a conventional liquid crystal display device using a TFT array.
A plan configuration diagram of pixels is shown.

In the figure, 101 is a TFT, 103 is a scanning line, 104 is an auxiliary capacitance line, 105 is a pixel electrode, and 106 is a signal line. In this figure,
Although the scanning line 103 is shared with the gate electrode of the TFT 101, the gate electrode may be drawn from the scanning line 103. The signal line 106 is connected to the drain electrode 107 of the TFT 101, and the intersection with the scanning line 103 is insulated by an insulating film. The source electrode 108 of the TFT 101 is connected to the pixel electrode 105. The auxiliary capacitance line 104 is insulated from the pixel electrode 105 by an insulating film.

When a signal voltage and a scanning voltage are applied to the TFT array thus constructed, the individual TFTs 101 become conductive, and a voltage corresponding to the signal voltage is applied to the pixel electrode 105. When the scanning voltage is not applied, the individual TFTs 101 are turned off and the voltage applied to the pixel electrode 105 is held.

As described above, such a liquid crystal display device is becoming larger and finer. As a result, an increase in the number of pixels or an increase in pixel density is caused, resulting in an increase in the occurrence rate of pixel defects and a significant reduction in manufacturing yield, which is a major problem.

As a method of solving this problem, a method of correcting a pixel defect using a high energy beam such as a laser has been proposed. One of them is a method of cutting and correcting a short-circuit portion that mainly occurs during pattern formation.
In this method, for example, as shown in FIG.
The short-circuited portion 401 between the signal line 106 and the pixel electrode 105, which is caused by the defective pattern formation of No. 06, is irradiated with the beam 402 passing through the aperture to cut and separate the short-circuited portion.

The other is a method of electrically connecting the upper and lower electrodes by irradiating the structure shown in FIG. 11 with a laser beam. That is, when the laser beam 501 is irradiated from the back surface of the substrate 505, first, the lower electrode 502 absorbs the energy of the laser beam and is rapidly heated, and is liquefied or vaporized to expand its volume. As a result, the insulating film 5
03 or the upper electrode 504 is pierced. At this time, the liquid phase of the lower electrode 502 adheres to the periphery of the hole generated by the laser beam irradiation, and serves to make electrical contact with the upper electrode 504. As a result, the upper electrode 504
And the lower electrode 502 are electrically connected. Note that FIG.
Although the laser beam is irradiated from the back surface of the substrate 505 in No. 1, the electrical connection can be made by the same process even when the laser beam is irradiated from the front surface of the substrate 505. Further, it can be applied to a structure in which a passivation film is formed on the upper electrode 504.

Recently, it has been considered to use the above structure and the redundant structure together. That is, two TFTs are provided for each pixel. The first TFT is connected to the signal line, the scanning line and the pixel electrode. The second TFT is provided as a spare and is not normally connected to the signal line. Therefore, the second TFT does not substantially participate in the signal writing to the pixel electrode. If the first TFT does not operate normally for some reason, the laser beam is used to disconnect and separate the first TFT from the circuit. Next, the second TFT is connected to the signal line using a laser beam. As a result, instead of the first TFT, the second TFT
Can be operated and defective pixels can be relieved.

As another example of the redundant structure, a method in which TFTs are connected in parallel and both TFTs operate in a normal state can be considered. However, in this case, although it is effective for the defective ON characteristics of the TFT, it is not effective for the defective OFF characteristics. Therefore, it is ideal to provide a preliminary structure to completely repair the pixel.

In the TFT array, it is known that the pixel potential shifts due to switching noise when the TFT is turned off. If the shift amount of the pixel potential is dV _p , dV _p has a great influence on display characteristics. Therefore, it is necessary to make dV _p uniform within the panel. This is the same even when the redundant structure is adopted, and it is necessary to suppress the difference in dV _p between the normal pixel and the repaired pixel to a sufficiently small value. However, in the conventional liquid crystal display device, the reality is that there is no liquid crystal display device having a redundant structure and a sufficiently small difference in dV _p .

[0012]

SUMMARY OF THE INVENTION Therefore, the present invention provides a liquid crystal display device which employs a redundant structure, can obtain a high yield, and can reduce the difference in display characteristics between normal pixels and repaired pixels. It is an object.

[0013]

In order to achieve the above object, the present invention provides at least a thin film transistor and a pixel electrode in an intersection region between two adjacent scanning lines and two adjacent signal lines. In a liquid crystal display device including a pixel including a plurality of thin film transistors per pixel, at least one of the plurality of thin film transistors is provided with a cutting portion that can be electrically separated from the pixel electrode, and another thin film transistor is provided. The source side electrode path is provided with a connection portion that can be electrically connected to the pixel electrode. Then, using the parasitic capacitance determined by the arrangement of the source side electrode path of the thin film transistor including the connection portion,
The scanning line-pixel electrode capacitance under the state where the thin film transistor having the cutting portion is electrically connected to the pixel electrode, and the thin film transistor having the cutting portion is electrically disconnected from the pixel electrode. In addition, the difference between the scanning line / pixel electrode capacitance when the thin film transistor having the connection portion is electrically connected to the pixel electrode is reduced.

[0014]

With the adoption of the redundant structure, not only the yield is increased, but also the difference in the capacitance between the scanning line and the pixel electrode between the normal pixel and the repaired pixel can be significantly reduced, so that the display characteristics can be improved. It will be possible.

[0015]

Embodiments will be described below with reference to the drawings. FIG. 1 shows a liquid crystal display device 1 according to an embodiment of the present invention.
A plan configuration diagram of pixels is shown.

In the figure, 101 indicates a first TFT, and 102
Indicates a second TFT, 103 indicates a scanning line, and 104
Is a storage capacitor line, 105 is a pixel electrode, and 106
Indicates a signal line. The signal line 106 is the first TFT1
01 and the drain electrode 107 of the second TFT 102, and the intersection with the scanning line 103 is insulated by an insulating film. The auxiliary capacitance line 104 is the pixel electrode 10
5 is insulated by an insulating film.

The source electrode 108 of the second TFT 102
Is connected to the pixel electrode 1 via the connecting portion 109 provided on the way.
05 is connected. The connection portion 109 is provided with a connection portion for melting the electrode material and electrically connecting the source electrode 108 to the pixel electrode 105 when irradiated with the laser beam 110.

On the other hand, the source electrode 1 of the first TFT 101
08 is connected to the pixel electrode 105 via a cutting portion 111 provided on the way. The cutting portion 111 melts and cuts when the laser beam 112 is applied to the first TF.
A cut portion is provided to electrically separate T101 from the pixel electrode 105.

In this example, the normal pixel and the repaired pixel are scanned using the parasitic capacitance existing in the connection portion 109, that is, the parasitic capacitance existing between the source electrode 108 of the second TFT 102 and the pixel electrode 105. The difference in capacitance between the line and pixel electrode is reduced.

Here, a specific example will be described. A Mo-Ta alloy is formed into a film with a thickness of 250 nm on the washed glass substrate,
The scanning line 103, the auxiliary capacitance line 104, and the lower electrode of the connection portion 109 were patterned.

Next, a gate insulating film (SiO) is formed to a thickness of 350 n.
m, SiN thickness 50nm, a-Si film thickness 50nm, etching topper (SiN) thickness 200nm
Pattern the etching stopper, then source,
After forming an n ⁺ a-Si film having a thickness of 50 nm, which is an ohmic contact layer in the drain region and doped with impurities such as phosphorus, the a-Si layer was patterned into an island shape. Furthermore, I
TO was formed into a film having a thickness of 100 nm to form the pixel electrode 105.

Next, the first insulating film SiO 2 on the terminal portion of the gate electrode was removed by etching. After that, Mo
Is deposited to a thickness of 100 nm and Al is deposited to a thickness of 400 nm.
6, the drain electrode 107 and the source electrode 108 were formed, and the n ⁺ a-Si was removed by etching to electrically separate the drain electrode 107 and the source electrode 108 to form an active matrix substrate. Finally, a SiN film having a thickness of 150 nm was formed as a passivation film and patterned.

The active matrix substrate thus obtained has two TFs per pixel as shown in FIG.
T is provided. The first TFT 101 is the signal line 10
6 and the pixel electrode 105, but the second TF
T102 is in a standby state. That is, the second TFT 102 is electrically separated from the pixel electrode 105 via the connection portion 109.

If the first TFT 101 is defective,
The laser beam 112 is irradiated through the cut portion 111 through the aperture to cut the source electrode 108, and the first TFT 101 is cut.
Are electrically separated from the pixel electrode 105. Cutting point 11
Since the electrode film scatters from No. 1, it is preferable to make the cut portion 111 thin so that the rate of occurrence of wire breakage does not increase so that cutting can be performed with as few irradiation times as possible. In this example, the width of the cut portion was set to about 5 μm, and good cutting was possible.

Next, a laser beam 110 is applied to the connecting portion 109.
Of the second TFT 102 by irradiating and electrically connecting
Is connected to the pixel electrode 105. As a result, the TFT performing the switching operation is switched from the first TFT 101 to the spare second TFT 102. First TFT 10
Since the probability that both the first TFT 102 and the second TFT 102 are defective is extremely low, this method can repair almost 100% of defective pixels related to the TFT.

By the way, in the TFT array, the pixel potential is shifted by the switching noise when the TFT is turned off, as described above. As a reference example in FIG.
An equivalent circuit for one pixel in the TFT array shown in FIG. 9 is shown. Voltage shift amount dV _{p at} point S in the figure
Is the shift amount of the gate voltage when the TFT 701 is off.
V _g , parasitic capacitance 70 between the gate and source of the TFT 701
When 2 is C _gs , the liquid crystal capacitance 703 is Cl _c , and the auxiliary capacitance 704 is C _s , dV _p = C _gs / (C _gs + Cl _c + C _s ) × dV _g .

Since dV _p has a great influence on the display characteristics, it is necessary to make dV _p uniform within the panel. An equivalent circuit for pixel potential shift in the pixel structure shown in FIG.
As shown in FIGS. 3 (a) and 3 (b).

The capacitance between points G and S in the figure is the capacitance between the scanning line and the pixel electrode. In a normal pixel, as shown in FIG. 3A, the C _gs 801 of the first TFT 101 and the second TFT 1
The total scanning line-pixel electrode capacitance is represented by a combination of C _gs 802 of _No. 02 and the parasitic capacitance 803 of the portion separated by the insulating film of the connection portion 109.

On the other hand, the cut portion 11 of the first TFT 101
When 1 is cut and separated and the connection portion 109 of the second TFT 102 is connected, an equivalent circuit relating to the pixel potential shift is shown in FIG.
It becomes as shown in (b). In this case, the second TFT1
The C _gs 802 of 02 becomes the capacitance between the entire scanning line and the pixel electrode.

In the normal pixel, the first and second TFs are
When T101 and T102 are conductive, the potentials at points F and S in the figure are equal to the potential of the signal line 106. Therefore, it is equivalent to the parasitic capacitance 803 of the connection unit 109 being substantially absent. In other words, the entire scan line at this time
The inter-pixel electrode capacitance is C _gs 801 of the first TFT 101. In the normal pixel, the first and second TFTs 10
When points 1 and 102 are in the non-conducting state, the potentials at points F and S in the figure are not equal, so the overall capacitance between the scanning lines and pixel electrodes is C _gs 801 of the first TFT and the second TFT 10.
C _gs 802 of 2 and the parasitic capacitance 803 of the connecting portion 109 are combined capacitances.

FIG. 4 shows C _gs in FIGS. 3 (a) and 3 (b).
The behavior of and, that is, the gate-source voltage dependence is shown. In the figure, the characteristic curve 901 shows the state after repair shown in FIG. 3 (b), and the characteristic curve 902 shows the state in FIG. 3 (a).
The state of the normal pixel shown in FIG.

When V _gs is increased, C _gs above the threshold value
Will increase. After the TFT becomes conductive, as can be seen from the above explanation, the entire scanning line of normal pixels and repaired pixels.
The capacitance between the pixel electrodes is almost equal. On the other hand, when the TFT is in a non-conducting state, the entire scanning line-pixel electrode capacitance between the normal pixel and the repaired pixel is the parasitic capacitance 803 of the connecting portion 109.
It depends on

Therefore, the parasitic capacitance 80 of the connecting portion 109 is
The C _gs characteristic can be adjusted by changing 3.
C _gs in the non-conduction state of the second TFT 102 is C _gs0ff
2, the parasitic capacitance of the connection portion 109 is C _x , the first TFT 10
1 in a conducting state, when the difference between the non-conductive state C _gs and _{dC gs, 1 / dC gs =} 1 / C x + 1 / C gs0ff Under 2 relations, the voltage of C _gs as shown in FIG. 5 Dependency can be eliminated. As a result, the drain voltage dependence of the voltage shift amount can be eliminated. Further increasing the C _x, larger such conditions than C _gs of C _gs nonconductive conductive state can also be realized.

FIG. 6 shows a plan view of one pixel of a liquid crystal display device according to another embodiment of the present invention. In addition,
In this figure, the same parts as those in FIG. 1 are designated by the same reference numerals.
In this embodiment, the source electrode 10 of the second TFT 102
A part 113 of 8 extends in parallel with the scanning line 103 and is connected to the source electrode 108 of the first TFT 101. Then, the connecting portion 109 is interposed on the way.

That is, in this example, the connection portion 109 is provided with a parasitic capacitance and the portion 113 and the scanning line 103 are also provided with a parasitic capacitance, and the normal pixel and the repaired pixel are scanned by these parasitic capacitances. It suppresses the capacitance fluctuation between the line and the pixel electrode.

A specific example will be described below. A Mo-Ta alloy having a thickness of 250 nm was formed on the washed glass substrate, and the scanning line 103, the auxiliary capacitance line 104, and the lower electrode of the connecting portion 109 were patterned. Next, gate insulating film (SiO)
Is 350 nm thick, SiN is 50 nm thick, and a-Si film is 50 nm thick.
After forming a 200 nm thick etching stopper (SiN) continuously, the etching stopper was patterned. After forming an n ⁺ a-Si film with a thickness of 50 nm doped with impurities such as phosphorus, which is an ohmic contact layer in the source and drain regions, the a-Si layer was patterned into an island shape. Further, ITO was deposited to a thickness of 100 nm to form the pixel electrode 105.

Then, the first insulating film SiO 2 on the terminal portion of the gate electrode was removed by etching. Then M
o is formed to a thickness of 100 nm, Al is formed to a thickness of 400 nm, and the signal line 10
6, the drain electrode 107 and the source electrode 108 were formed, and n ⁺ a-Si was removed by etching using a metal as a mask to electrically separate the drain electrode 107 and the source electrode 108 to form an active matrix substrate. Finally, a SiN film having a thickness of 150 nm was formed as a passivation film and patterned.

The active matrix substrate thus obtained is provided with two TFTs per pixel as in the case shown in FIG. Although the first TFT 101 is connected to the signal line 106 and the pixel electrode 105,
The TFT 102 is in a standby state and is separated from the pixel electrode 105 via the connection portion 109.

The source electrode 108 of the second TFT 102
Has a portion 113 along the scanning line 103, and this portion 113 has a parasitic capacitance between the scanning line 103 and the source electrode 108 of the second TFT 102.

If the first TFT 101 is defective,
The laser beam 112 is irradiated through the cut portion 111 through the aperture to cut the source electrode 108, and the first TFT 101 is cut.
Are electrically separated from the pixel electrode 105.

Since the electrode film scatters from the cutting points 111, it is preferable that the cutting points 111 be thin so that the rate of occurrence of wire breakage does not increase so that the cutting can be performed with as few irradiation times as possible. In this example, the width of the cutting point 111 is
As a result of setting it to about 5 μm, good cutting became possible.

Next, a laser beam 110 is applied to the connecting portion 109.
Of the second TFT 102 by irradiating and electrically connecting
Is connected to the pixel electrode 105. As a result, the TFTs that perform the switching operation are changed from the first TFT 101 to the second TFT.
Switched to FT102. Since the probability that both the first TFT 101 and the second TFT 102 are defective is extremely low, the number of defective pixels related to the TFT is almost 1 by this method.
Can be repaired by 00%.

Here, the shift of the pixel potential is examined by an equivalent circuit as in the example of FIG. An equivalent circuit for pixel potential shift in the pixel structure shown in FIG. 6 is as shown in FIGS. 7 (a) and 7 (b).

In the figure, the capacitance between the points G and S is the capacitance between the scanning line and the pixel electrode. In the normal pixel, as shown in FIG. 7A, the C _gs 801 of the first TFT 101 and the second TF
C _gs 802 of T102 and parasitic capacitance 803 of connecting portion 109
And the parasitic capacitance 11 of the portion 113 provided adjacent to the scanning line 103.
The total capacitance between the scanning lines and the pixel electrodes is represented in the form of a combination of 01 and 01.

On the other hand, the cut portion 11 of the first TFT 101
When 1 is cut and separated and the connection portion 109 of the second TFT 102 is connected, an equivalent circuit relating to the pixel potential shift is shown in FIG.
It becomes as shown in (b). In this case, the second TFT1
The combined capacitance of the C _gs 802 of _No. 02 and the inter-wiring parasitic capacitance 1101 becomes the entire scanning line-pixel electrode capacitance.

In the normal pixel, the first and second TFs are
When T101 and T102 are in the conductive state, the potentials at points F and S in FIG. 7A are equal to the potential of the signal line 106.
Therefore, the parasitic capacitance 803 of the connecting portion 109 is substantially the same as that of the parasitic capacitance 803, and the C _gs 80 of the second TFT 102 is 80.
2. It is not necessary to consider the inter-wiring parasitic capacitance 1101.

In the normal pixel, the first and second TFs are
When T101 and 102 are in the non-conducting state, Fig. 7 (a)
Since the potentials of the points F and S are not equal, the total capacitance between the scanning lines and the pixel electrodes is C _gs 801 of the first TFT, C _gs 802 of the second TFT 102, and the parasitic capacitance of the connection portion 109. 803 and the inter-wiring parasitic capacitance 1101 are combined capacitances.

After the repair, the inter-wiring parasitic capacitance 1101 is changed to C _gs 802 of the second TFT 102 as shown in FIG. 7B.
Connected in parallel with. For this reason, the entire scanning line-pixel electrode capacitance after repair is larger than that in the case where there is no inter-wiring parasitic capacitance. Therefore, the inter-wiring parasitic capacitance 11
By appropriately selecting the size of 01, the difference in C _gs between the normal pixel and the repaired pixel can be reduced.

The behavior of C _gs will be further described with reference to FIG. A characteristic curve 1203 shows the C _gs characteristic of the normal pixel having the structure of FIG. 1, and a characteristic curve 1204 shows the C _gs characteristic of the pixel after repair. As described above, the C _gs when the two TFTs are in the conductive state does not differ from that of the normal pixel after repair, but when the two TFTs are in the non-conductive state, the normal pixel is affected by the connection portion 109. C _{gs of} is larger than that after repair. The C _gs characteristic of a normal pixel in this example is 1201
become that way.

In the case of this example, two T
When the FT is in the non-conducting state, the inter-wiring parasitic capacitance has the second T
Since it is connected in parallel with the FT 102, the characteristic curve 1203
C _gs is slightly larger than that of. The C _gs characteristic of the pixel after repair is shown by 1202. This has a form in which a parasitic capacitance 1101 between wirings is added to the characteristic curve 1204.

Therefore, the C _gs characteristic of the normal pixel 1201
By selecting the inter-wiring parasitic capacitance 1101 so that the difference between the repaired C _gs characteristic 1202 and the repaired C _gs characteristic 1202 becomes small, the difference in the display characteristic can be suppressed.

Specific examples are shown below. The conductive C _gs (hereinafter, referred to as C _gson ) of the TFT is 0.03 pF, and the non-conductive C _gs (hereinafter, _referred to as C _gsoff ) is 0.02 pF. Also,
The parasitic capacitance of the connecting portion 109 is 0.006 pF.

In the configuration of FIG. 1, the normal pixel has C _gson = 0.03.
pF, Cg _gsoff = 0.0247pF, C after repair
_gson = 0.03 pF and C _gsoff = 0.02 pF. C _gsoff
There is a difference of about 0.005 pF, and a pixel potential shift occurs.
The maximum difference in brightness will be about 6%.

Next, in the structure of this example, the inter-wiring parasitic capacitance is
_Assuming 0.002 pF, the normal pixel has C _gson = 0.03 pF, C
_gsoff = 0.0248pF, C _gson = 0.032 after repair
pF, C _gsoff = 0.022 pF. This results in a maximum of about
The brightness difference can be suppressed to about 3%. Since the pixel potential shift is reduced to suppress the brightness difference, the DC voltage component applied to the mixed liquid can also be reduced, and as a result,
Display reliability can also be improved.

As for the inter-wiring parasitic capacitance, a parasitic capacitance of about 0.003 pF can be obtained by setting the space between the wirings to about 4 to 5 μm. Further, it is appropriate that the inter-wiring parasitic capacitance is about one half of the parasitic capacitance of the connecting portion. Further, although two TFTs are provided per pixel in each of the above embodiments, the number of TFTs is not limited to this.

[0056]

As described above, according to the present invention,
A redundant structure in which a plurality of thin film transistors is arranged per pixel is adopted, and the difference in the capacitance between the scanning line and the pixel electrode between the normal pixel and the repaired pixel is reduced by the structure of the source side electrode path. Without complicating
The difference in display characteristics between normal pixels and repaired pixels can be reduced.

[Brief description of drawings]

FIG. 1 is a plan configuration diagram of one pixel in a liquid crystal display device according to an embodiment of the present invention.

FIG. 2 is an equivalent circuit diagram when one thin film transistor is incorporated for one pixel.

3A is an equivalent circuit diagram of one normal pixel in the liquid crystal display device according to the embodiment, and FIG. 3B is an equivalent circuit diagram of one pixel after repair.

FIG. 4 is a diagram showing the gate-source voltage dependence of the capacitance between the scanning line and the pixel electrode in the example.

FIG. 5 is a diagram showing the gate-source voltage dependence of the scanning line-pixel electrode capacitance when the parasitic capacitance is changed.

FIG. 6 is a plan configuration diagram of one pixel in a liquid crystal display device according to another embodiment of the present invention.

FIG. 7A is an equivalent circuit diagram of one normal pixel in the liquid crystal display device according to the embodiment, and FIG. 7B is an equivalent circuit diagram of one pixel after repair.

FIG. 8 is a diagram showing the gate-source voltage dependence of the capacitance between scanning lines and pixel electrodes in the same embodiment as compared with the example of FIG.

FIG. 9 is a plan configuration diagram of one pixel of a liquid crystal display device in which one thin film transistor is incorporated for each pixel.

FIG. 10 is a diagram for explaining an example of a repair method.

FIG. 11 is a diagram for explaining another example of the repair method.

[Explanation of symbols]

101 ... First thin film transistor 102 ... Second
Thin film transistor 103 ... Scan line 104 ... Auxiliary capacitance line 105 ... Pixel electrode 106 ... Signal line 107 ... Drain electrode 108 ... Source electrode 109 ... Connection part 111 ... Disconnection part

Claims (1)

[Claims] 1. A liquid crystal display device comprising a pixel including at least a thin film transistor and a pixel electrode in an intersection region between two adjacent scan lines and two adjacent signal lines, wherein the thin film transistor is one pixel. A plurality of thin film transistors are provided for each pixel, at least one of the plurality of thin film transistors is provided with a cutting portion that can be electrically separated from the pixel electrode, and another thin film transistor is electrically connected to the pixel electrode through a source electrode path. The thin film transistor including the connection portion capable of connecting the scanning line and the pixel electrode capacitance and the thin film transistor including the disconnection portion under the state of being electrically connected to the pixel electrode. Instead of electrically disconnecting the thin film transistor having the cutting portion from the pixel electrode, the thin film transistor having the connecting portion is used for the pixel. A liquid crystal display device characterized by comprising reducing parasitic capacitance which is determined the difference between the scanning line-pixel electrode capacitance under a state of being electrically connected to a pole in the arrangement of the source electrode path.