JPH07209628A - Liquid crystal display element and its driving method - Google Patents

Abstract

PURPOSE:To obtain a high contrast and to enable long-period stable operation by providing a frequency range wherein at least one of three characteristic values of the dielectric tensor of chiral smectic C layer liquid crystal is larger than the remaining two components. CONSTITUTION:A couple of substrates where striped electrodes are formed while one of them is transparent are arranged opposite each other so that the electrodes are approximately orthogonal each other, and ferroelectric liquid crystal which has a chiral smectic C phase is sandwiched between the substrates to form the liquid crystal element, this element has the frequency range wherein at least one of epsilon3 and epsilon2 is larger than the remaining two components, where epsiloni (i=1, 2, and 3) is one of three characteristic values of the dielectric tensor of the chiral smectic C phase liquid crystal. The dielectric constant of liquid crystal molecules in the long-axis direction is denoted as epsilon3, the component in the tangential direction of a circle between the two components having a short-axis direction perpendicular to epsilon3 is denoted as epsilon2, and a component orthogal to epsilon2 is denoted as epsilon1. As for chevron structure, dielectric response to the effective voltage of an AC pulse in a frequency range wherein the response of a self-polarized component can be ignored is classified by the large/small relation of the three components.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a matrix driving liquid crystal display device using a ferroelectric liquid crystal and a driving method thereof.

[0002]

2. Description of the Related Art Liquid crystal displays
play (hereinafter referred to as LCD) is widely used as a device for displaying figures and characters on a flat surface for display or decoration of measuring instruments such as pocketable calculators and testers, and for POP because it is lightweight and thin. Recently, a thin film transistor (THIN FILM TRANSISTER: hereinafter referred to as TFT) has been used, and it has come to be used as a large-capacity terminal display for a television, a personal computer, or a workstation that displays moving images in full color.

The origin of the shutter property in these is a so-called Twisted nematic type (hereinafter referred to as TN type), which is a known technique (Kobayashi,
Okano ed. "Liquid Crystal", 1985, Baifukan). In order to display a high-quality and large-capacity image with the TN type, it is necessary to form active elements such as TFTs on the substrate by the number of pixels. Since it is difficult to stably form a TFT element on a substrate with a high yield, the yield of manufacturing a panel with a size of about 14 inches cannot be improved.
CD manufacturing has become quite difficult. In terms of characteristics, since the TFT element is provided, there is also a problem that the aperture ratio of the screen is reduced and the image quality becomes dark.

In the technological development of LCDs, the technical possibility of displaying a clear static and moving image in a matrix without using a TFT is continuously pursued, and a super twisted nematic type (hereinafter referred to as STN type) is called. Ferroelectric liquid crystal display (FERRO-ELECTRIC LIQ
UID CRYSTAL DISPLAY: hereinafter referred to as FLCD) is expected.

In principle of operation, the STN type is similar to the TN type. By devising the twist of the liquid crystal layer, a steeper threshold characteristic is provided. Even with this, the number of scanning lines is 200 to 4 which is equivalent to that of TFT-LCD.
The limit is about 00 lines, and the display capacity and the image quality cannot exceed the TFT. The latter is a ferroelectric liquid crystal (FLC: FERRO-EL).
ECTRIC LIQUID CRYSTAL) CHIRALSMECTIC C PHASE (hereinafter referred to as SmC ^* ) is used, and there is a memory effect (bistability) in the liquid crystal layer itself. Therefore, in principle, the number of scanning lines is limited. A large-capacity high-quality display is possible. Regarding FLCs and electro-optical elements using them, see, for example, Fukuda and Takezoe, “Structure and Physical Properties of Ferroelectric Liquid Crystal” (Corona Co., Ltd. 1990).
Year) is detailed. Since there is no element for each pixel like a TFT, a high aperture ratio and a bright image quality are expected.

However, the SmC ^* phase has a structure and an operating mechanism which are completely different from those of the TN type, and there are still many technical problems to be overcome for practical use, and orientation control is still present. , Gap control, impact resistant panel manufacturing, development of fast response materials. For more information on these, see the book (Next-generation LCD display and liquid crystal materials, supervised by Fukuda, CMC, published in 1992). Another important problem is that even if these are solved once, the possibility of displaying a high-contrast image stably for a long period of time is not guaranteed. Particularly, in the matrix driving method based on the voltage modulation borrowed from the TN type, it is difficult to display a stable image with high contrast due to the incomplete memory state due to the layer structure and the insufficient threshold characteristic.

In order to improve this point, the applicant of the present invention has a patent application that if a pulse width modulation drive is applied to an FLC having a certain dielectric property, a high quality image can be displayed due to the dielectric effect. No. 4-186715, Japanese Patent Application No. 4-21
This is indicated by No. 9080.

[0008]

However, when the present applicant diligently studied the above-mentioned driving method, there were the following problems. That is, the output waveform on the scanning signal side requires a 5-level voltage output, which complicates the drive circuit. In this regard, the present invention provides a means to configure all drive output waveforms in three levels. Another problem is not only the liquid crystal specified in the above application, but also FL having a wider range of dielectric properties.
Since it was found that the method can be applied to C liquid crystal, it is necessary to correct this point.

[0009]

That is, according to the present invention, a pair of substrates, at least one of which is provided with stripe-shaped electrodes and are transparent, are opposed to each other so that the electrodes are substantially orthogonal to each other, and a chiral smectic is provided between the substrates. In a liquid crystal display device in which a ferroelectric liquid crystal exhibiting a C phase is sandwiched, the three eigenvalues of the dielectric tensor of the chiral smectic C phase liquid crystal are expressed by ε.
When i (i = 1, 2, 3), the liquid crystal display element is characterized in that at least one of ε _{3 and} ε ₂ has a frequency range larger than the remaining two components. A pulse width modulation drive in which an AC pulse having a pulse width shorter than a pulse width used for forming a state at the time of selection is applied at the time of selection, and the AC pulse has a constant width and a polarity is continuously inverted. And, the frequency is in a frequency region in which at least one of ε _{3 and} ε ₂ is larger than the remaining two components.

[0010]

In order to improve the image quality of the FLCD, it is desirable to use the matrix driving method for inducing the change of the layer structure so that the static layer structure peculiar to FLC is linked to the higher image quality. Therefore, first, the layer structure of the SmC ^* phase, the spontaneous polarization of the liquid crystal molecules that regulate the electrical response, and the special property (frequency dependence) of the dielectric tensor will be described.

An ideal bookshelf structure of SmC ^* layers is shown in FIG. The FLC molecule (100) has a spontaneous polarization P (503) in a direction substantially perpendicular to the molecular axis, and the applied electric field E
(505) is coupled with P, and the torque (P · E) acts so that P points in the direction of E. This acts so that the molecular axis moves on the cone from S1 to S2 (or S2 to S1) in a collective manner. The response of this spontaneous polarization component is F
The first mode of LC response, which is the origin of switching (called DC response or pulse response). The force required to cause the reversal of spontaneous polarization varies from liquid crystal to liquid crystal, but generally depends on the product of pulse width and voltage. If the apex angle θ (504) representing the amount of rotation of the molecular axis is approximately 45 degrees in the horizontal direction, it is possible to maximize the contrast change due to birefringence under crossed Nicols, and the FLC layer is shuttered. That is why it can be used for.

Bookshelf structure S1 and S2 are 1 μm
If the thickness is about m, a memory effect is produced by the effect of the interface, and even if the electric field E is blocked, it can stay at each position (memory effect). This is the greatest feature of the SmC ^* layer of FLC, and suggests that the image quality may not change even when matrix display is performed. However,
Actually, in most cases, the apex angle θ is 45 degrees, that is, it is not stabilized at the position where the contrast is maximum, but is stabilized at a position greatly deviated.

[0013] The cause of this is,
Physical Review Letters, 2658 pages, 5
9 (1987)).
mC ^*The layers are different from the ideal structure shown in FIG.
The "<<" shaped shelves shown in Fig. 1 (a) with the direction inclined by α
This is probably because it has a Bron structure. Center of cone
The line (101) is not parallel to the substrate surface (102) and
It is tilted by a degree α. With this layer structure, the liquid crystal molecules are
It tries to be almost parallel to the plate surface, so after all, it is a cone
It is quiet at (S1 ') and (S2') which are deviated from the above S1 and S2.
Stop. At this position, the shadow on the substrate surface is more than the apex angle 2θ.
It will be significantly smaller. Cool from SmA phase to Sm
C^*There are some exceptions when it comes to competing, but it is generally unavoidable
It is believed that the chevron structure of FIG.
It

On the other hand, for the driving of the FLC which responds to the direct current, the voltage modulation driving is used from the analogy of the TN type driving. In this case, the pulse width as shown in FIG. 4C is the same, and the voltage is the voltage V (30
The voltage waveform (403) that is 1/3 to 1/4 (bias voltage) of 0) is added when the device is not selected. Since spontaneous polarization responds to this voltage, the liquid crystal molecules oscillate in the vicinity of S1 'or S2', and the light transmittance vibrates (405), further lowering the contrast. In the worst case, state reversal occurs and it is difficult to write an image with a stable image flickering. If no voltage is applied during non-selection, the response is as shown in FIGS.

From the above description, in order to increase the stability and the contrast by the chevron structure, it is necessary to hold the liquid crystal molecules in S1 or S2 and constrain them so that they do not respond with the bias voltage. . As a means for simultaneously solving the two problems of relaxation and fluctuation, it is conceivable to utilize the dielectric response which is the second response mode of liquid crystal molecules. The dielectric response of liquid crystal molecules that can be approximated to a spheroid is determined by the magnitude of the eigenvalue of the dielectric tensor T = (εxy) as shown in FIG. In conclusion, it means that the liquid crystal molecular axis is changed and fixed so that the maximum component of the absolute value of the eigenvalue is aligned in the direction of the applied electric field as much as possible. The complexity of the problem is that this magnitude relationship changes with frequency. In addition, it is desirable to move the molecular axis by a dielectric effect, but even if the molecular axis does not move depending on the magnitude of the effect, the fluctuation due to spontaneous polarization is suppressed (stabilized), so that a desirable result is obtained. provide.

As shown in FIG. 2, the dielectric constant in the major axis direction of the liquid crystal molecule is ε ₃ (= parallel to ε), and the component in the tangential direction of the circle is ε ₂ out of the two components in the minor axis direction orthogonal to this. The component orthogonal to it is represented by ε ₁ (= ε vertical). With the chevron structure,
The dielectric response to the effective voltage of the AC pulse in the frequency domain, where the response of the spontaneous polarization component can be ignored, can be classified by the magnitude relationship of the three components.

(1) ε ₃ > ε ₂ and ε ₃ > ε ₁

The dielectric anisotropy Δε (= ε ₃ −ε ₁ ) is positive and the dielectric response is similar to that of a nematic liquid crystal in which the dielectric constant in the major axis direction is dominant. That is, when a sufficient effective value is given, the molecular axis stands on the substrate. In this case, Japanese Patent Application No. 4-186715 and Japanese Patent Application No. 4-21
9080 for further details.

(2) ε ₁ > ε ₂ and ε ₁ > ε ₃

In this case, the dielectric anisotropy Δε is negative,
As long as the chevron structure is retained, the molecular axis is parallel to the substrate and no improvement in contrast is expected. In the bookshelf structure, stabilization (suppression of oscillation) due to dielectric action is expected.

(3) ε ₂ > ε ₃ and ε ₂ > ε ₁

In this case, since the component in the minor axis direction works, the molecular axis tends to tilt with respect to the substrate as in (1).
However, the effect is affected by the magnitude relationship between ε ₃ and ε ₁ . If ε ₃ > ε ₁ , the effect of (1) is synergized and the rise is more immediate. In the opposite case, the effect of (2) is superimposed and the effect is weakened.

In the chevron structure, moving the molecular axis so as to rise from the substrate directly leads to improvement in contrast. (1) and (3) correspond to this, and claim 1 of the present invention is this case. The present invention selects S1 'or S when selected.
The molecules brought to either side of 2'are subjected to S pulse by applying an AC pulse in which the dielectric action is dominant in the non-selection time.
1'or S2 ', or preferably S1
Alternatively, the purpose is to move and restrain in the direction of S2. For that purpose, the pulse width at the time of non-selection needs to be narrower than the pulse width at the time of selection, and the pulse width modulation drive in which the response of the spontaneous polarization component is suppressed is required. In the voltage modulation drive of the conventional method, the pulse width is the same as the write pulse width, and the response of the spontaneous polarization component is always present. Therefore, the dielectric response is blocked, and the effect of the present invention is expected. Can not.

The extent of the frequency range of the AC pulse is determined by a measurement example of the magnitude relationship of the three eigenvalues, for example, J. Jones et al., Ferroelectrics, 1
Volume 21, pages 91, 1991 and F Gouda et al.,
Physical Review A, Volume 46, Pages 951, 19
1992). Determining the eigenvalue requires three independent experiments for each frequency. Experimental data is described in these, but it is necessary to configure the matrix drive waveform so as to make use of it and to exert a dielectric torque, and that it is 1/2 of the write pulse width as shown in the examples described later. No mention is made of what is possible with ~ 1/5 AC pulses.

Since pulse width modulation driving is extremely desirable by the mechanism described above, the matrix driving waveform according to the present invention will be described below with reference to the drawings.

Although not conscious of pulse width modulation driving, an example of applying a high frequency pulse to a non-selected electrode based on voltage modulation driving so that a high frequency is applied at the time of non-selection is the Prod- ing of Society of Information Display (Sato et al., Vol. 28, p. 189, 1987). In this case, since the high frequency pulse is superimposed on the low voltage data signal, the response of the spontaneous polarization component cannot be ignored, and it is necessary to increase the frequency and voltage, which is not desirable. An example of a high-frequency voltage fluctuation-free drive waveform that improves this point is disclosed in the National Technical Report (Wakida et al., Vol. 33, No. 1, pages 44 to 50). Although this waveform is effective, it is necessary to prevent inversion to another state within the selective writing time, high voltage and high frequency are required, and the number of output voltage levels is 5.

The present inventors have constructed two types of waveforms that have improved these points. If the pulse width required for state formation is τ, a type that requires 4τ for selection (claim 3, FIG. 7 to FIG. 1).
0) and 2τ which is a half of this (claim 4 and FIGS. 11 to 16). Vd and Vs are amplitudes including signs. In these figures, a typical unit waveform is described only when Vd = Vs and the ratio of the pulse width of the non-selection signal to τ is 2: 1 to 5: 1. A pulse ratio higher than this can be easily inferred because it can be easily inferred, but according to the experiments by the present inventor, good results were obtained with a pulse ratio of 5: 1 or less. The preferred ratio is 3: 1 to 4: 1. A too high pulse ratio is disadvantageous because of high energy consumption.

A common feature of these unit waveform groups is that the number of output voltage levels is 3 (± Vs,
0 and ± Vd, 0), which means that pulses of the same width are continuously applied while changing the polarity when not selected.

The waveforms shown in FIGS. 7 to 10 are for writing one state (ON) on the select electrode first and then writing the other state, and thus need not be particularly described. After the selection signal (700) is applied to the pixel of interest, pulse-like non-selection signal trains (701, 702) corresponding to the number of scanning lines are applied. Even if the output voltage level is 3, waveforms other than those shown are possible. As an example, in FIG.
Seed. In FIG. 7, the shaded portion (703) of the selection signal and the shaded portion (704) of the non-selection signal that follows it are integrated into the off state.

FIG. 10 shows a pulse width ratio of 5: 1, but after the ON state (1000) is formed, there are three pulses (3τ / 5, 101 in total) which may be inverted within the selected time.
0) Applied. Since at least τ is required to invert the state, even this does not cause any inversion. The number of this pulse can be reduced to two, but if this is done, there is a relation that the number of high voltage pulses (1020) increases by one set. There is no difference in the display in either case.

From FIG. 11 onward, in order to reduce the scanning time by half,
A preliminary signal is applied to the electrode to be selected next to preliminarily align the states to one state. When n = 3, two types were described. To the pixel of interest, the selection preliminary signal (1100 or 1100 ′) is first applied, then the selection signal (1110 or 1120) is applied, and then the non-selection signal (1130, 1140) is applied. As a result, the state is reset (temporarily turned off) by the auxiliary signal (1150) before the new state is written.
This means that the scanning time is halved.

Further, in order to completely invert the reset state (ON state), a pulse having a width τ, which is the shortest possible pulse width, is usually applied. On the contrary, when it is necessary to stay in the reset state, it is necessary that the total sum of pulse widths that may cause inversion is limited to be shorter than τ. In the waveform of the present invention, τ / 2 at n = 2 and n = 3
For 2τ / 3 or τ / 3, for n = 4 τ / 2, for n = 5 2
τ / 5 or 3τ / 5. It is preferable that this total sum is small in order to ensure writing of two states, but it is impossible to make it zero. When n is large, there is a relation that the number of pairs of high-voltage pulses increases at the time of non-selection as described above when the total sum is reduced.

In this case, the amplitude of the same polarity is large (= V
The d + Vs) pulse may be repeated twice with a pulse of opposite polarity interposed. This could be a DC component and induce oscillation, but in the range of this waveform, the reset state was reliably maintained and oscillation did not occur at all. It goes without saying that the amplitudes Vd and Vs of the selection signal and the data signal can be independently changed within a writable range, if necessary.

[0034]

EXAMPLES In the present invention, the eigenvalue of the dielectric tensor in the frequency range of the AC pulse should be clarified, but as already pointed out, the experiment is difficult and the measurement with high accuracy is difficult. However, if the target liquid crystal has ε _i defined in claim 1, the liquid crystal molecules should surely rise from the substrate surface. The magnitude of the eigenvalue is unknown, but it is not difficult to guess that they have the same effect.

Example 1 The electrode parts were patterned by a photolithography method by a conventional method so that the electrode parts would face each other when the glass substrates with ITO transparent electrodes were opposed to each other. Thereafter, a spacer member for adjusting the cell gap (photoresist MP1400: manufactured by Shipley Co., Ltd.) was provided outside the electrode region by photolithography so that the step difference was about 2.4 μm. Polyimide film (trade name HL1100:
Rubbing treatment was carried out after forming (Hitachi Chemical Co., Ltd.) according to a conventional method. Approximately 170 with these facing each other and crimped
Post-baking was performed in an oven at 0 ° C. for 1 hour to obtain a cell having a gap of 2.2 μm. After this, Δε is positive (frequency 1
CS1014 with Δε = 0.2) at KHz to 30 KHz
FLC (trade name, manufactured by Chisso Corporation) was introduced into the cell at about 100 ° C., and cooled to obtain an oriented SmC ^* phase. The pretilt angle of this system was estimated to be almost 0 degree.

A waveform for on / off writing having a pulse width ratio of 4: 1 in FIG. 6A was synthesized by an arbitrary waveform generator and applied to this cell, and changes in the amount of light transmission synchronized with it were examined.
(For waveforms other than 4: 1, the pulse width that can cause inversion can be 3τ / 5 or less, so it is sufficient to show that writing is possible in this case. Is included in the figure, but it is omitted in the figure). Two different states may be formed by the two write signals (202 and 203) and may be held separately by the bias unit 204. Write pulse voltage Vpp (= 2Vs, Vd = Vs
) And the pulse width τ are changed to examine the writeability, and Table 1 shows the result. ○ is a combination that can be completely written, and × is an impossible combination.

[0037]

[Table 1]

In the case of ◯, the results are as shown in FIG. 6 (b), and when the voltage was increased, good results without vibration were obtained at a position close to the maximum contrast. × is the case of separation, but there was shaking. In the voltage modulation drive of 4: 1 bias (FIG. 4C), the oscillation caused by the bias portion 403 was large, and the center of oscillation was also inward, which was clearly inferior. This difference is due to the effect of the bias signal train, and it is clear that in pulse width modulation, switching occurs at the S1 and S2 positions in FIG. 1 due to the dielectric torque, and is constrained at this position. In voltage modulation, the spontaneous polarization P is in a frequency range in which it can respond, and this restraining effect cannot be expected at all, and there is only the effect that the voltage is lowered and the amplitude is reduced.

The pulse width at which the dielectric torque started to take effect was from 80 μs. In order to completely suppress and restrain the relaxation, the length should be shorter than 60 μs. The voltage is about 4
0 Vpp or more was required. Vpp = 50V, τ = 24
At 0 μs (bias pulse width 60 μs), extremely good results were obtained. Writing was possible even with 3: 1, 5: 1, but the pulse width ratios of 4: 1 and 5: 1 were characteristically superior. However, it was difficult to distinguish between them. When the pulse width of the bias was set shorter than 25 μs, linear cracks were observed to occur, so it was not preferable to shorten it too much.

Next, the same liquid crystal-polyimide system having a cell gap of 1.6 μm was obtained by the same procedure. In this case, the dielectric torque was developed under almost the same conditions. The difference is that the pulse width τ can be shortened to 120 μs when the amplitude Vpp is the same. In this case, writing was possible even if the pulse width ratio was reduced to 2: 1.

From the observation of the extinction position, Vpp
It was found that the tilt angle 2θ gradually increased toward the maximum value as the voltage was increased from 40V to 60V. It was thought that the molecule was moving on the cone.

<Example 2> FL represented by the following chemical formula
C was synthesized.

[0043]

[Chemical 1]

This liquid crystal is described in the literature (F. Gouda et al., Physical Review A, Vol. 46, pages 951, 1992).
According to (Year), it is known that ε ₂ is larger than the remaining two components up to about 1 MHz. A cell with a gap of 2.0 μm was prepared by the same procedure as in Example 1, and this liquid crystal was sealed to obtain a device. At room temperature (25
Degree), the width of the AC pulse in which the dielectric effect is dominant is 5
It was 0 μs or less and the voltage was 50 V or more. Also in this case, writing was possible with τ = 120 μs, Vs = Vd = 25 V, and a pulse width ratio of 4: 1. I was able to write at other ratios.

In the first embodiment, it is doubtful whether ε ₃ is the maximum in the frequency range (≈10 kHz) where the dielectric effect is dominant. Therefore, the crystal rotation method proposed by Nakano et al. (Nakano et al., Japan Journal of Upright Physics, 1
No. 9, page 2013, 1980) was extended to a chevron structure, and it was investigated whether liquid crystal molecules rise from the substrate. The light transmittance of the chevron structure was obtained as a function of the tilt angle from the numerical calculation using the theoretical formula of the above-mentioned document, and it was estimated how the transmittance changes as the tilt angle increases. When this was compared with the measured light transmittance of the device to the effective value of the AC pulse, it did not completely match quantitatively, but qualitatively, when the voltage was increased, the liquid crystal molecules became It was confirmed that it stood up from a horizontal direction. The same applies to Example 2. In addition, it is obvious that the present invention can be applied to a tilted bookshelf structure (FIG. 1B).

[0046]

As described above, the present invention makes use of a ferroelectric liquid crystal having a specific dielectric property, not destroying its natural state, the chevron structure or the inclined bookshelf structure. It has become clear that matrix drive can be performed in the form.

The driving method of the present invention can realize a high contrast and stable operation for a long period of time, and provides unprecedented practicality as a driving method of a ferroelectric liquid crystal element.

[0048]

[Brief description of drawings]

FIG. 1 is a perspective view showing a layer structure of a liquid crystal of SmC ^* phase.

FIG. 2 is an explanatory diagram of directions of three eigenvalues of a dielectric tensor of a cone and liquid crystal molecules constrained by the cone.

FIG. 3A is a waveform diagram showing an ideal pulse modulation drive waveform, and FIG. 3B is a waveform diagram showing a temporal change in light transmittance of an FLCD in synchronization with it.

FIG. 4A is a waveform diagram showing an applied waveform of a write pulse, FIG. 4B is a waveform diagram showing a time change of light transmittance of an FLCD synchronized therewith, and FIG. 4C is a voltage modulation drive. A waveform diagram showing an applied waveform including an actual bias portion to be applied, and (d) is a waveform diagram showing a time change of the light transmittance of the FLCD synchronized with it.

FIG. 5 is a perspective view showing a bookshelf structure which is an ideal state of an alignment structure of liquid crystals of SmC ^* phase.

FIG. 6A is a waveform diagram showing an example of a drive waveform of the present invention actually applied to a pixel with a pulse width ratio of 4: 1, and FIG. 6B is a waveform diagram of an FLCD synchronized therewith. It is a wave form diagram which shows the time change of light transmittance.

FIG. 7 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming a matrix driving waveform having a pulse width ratio of 3: 1 according to the present invention.

FIG. 8 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming a matrix driving waveform having a pulse width ratio of 4: 1 according to the present invention.

FIG. 9 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming a matrix driving waveform having a pulse width ratio of 5: 1 according to the present invention.

FIG. 10 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming a matrix driving waveform having a pulse width ratio of 2: 1 according to the present invention.

FIG. 11: Applying a preliminary signal to the next selected electrode,
FIG. 6 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming the matrix driving waveform of the present invention in which one of the states is aligned in advance and which has a pulse width ratio of 2: 1.

FIG. 12: Applying a preliminary signal to the next selected electrode,
FIG. 3 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming a matrix driving waveform of the present invention with a pulse width ratio of 3: 1, which is arranged in one state in advance.

FIG. 13: Applying a preliminary signal to the next selected electrode,
FIG. 3 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming a matrix driving waveform of the present invention with a pulse width ratio of 3: 1, which is arranged in one state in advance.

FIG. 14: Applying a preliminary signal to the next selected electrode,
FIG. 3 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming the matrix driving waveform of the present invention, which has been arranged in one state in advance and has a pulse width ratio of 4: 1.

FIG. 15: Applying a preliminary signal to the next selected electrode,
FIG. 3 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming the matrix driving waveform of the present invention, which has been arranged in one state in advance and has a pulse width ratio of 4: 1.

FIG. 16: Applying a preliminary signal to the next selected electrode,
FIG. 6 is a waveform diagram showing a data signal, a selection signal, and a non-selection signal for forming the matrix driving waveform of the present invention with a pulse width ratio of 5: 1, in which the states are made uniform in advance.

[Explanation of symbols]

101 ... Center line of cone 102 ... Substrate surface 200 ... Pulse portion necessary for aligning states to one side by a part of selection preliminary signal and selection signal 201 ... Off-write unipolar pulse portion 202 which may cause state inversion ... write monopolar pulse 203 ... write monopolar pulse 204 ... non-selection signal (bias signal) 300 ... write bipolar pulse 302 ... position and amount exhibiting maximum contrast 400 ... applied signal train 401 of write pulse only ... write Bipolar pulse 402 ... Writing bipolar pulse 403 ... Bias signal sequence during voltage modulation 405 ... Change of light transmission amount synchronized with 403 Diagram 501 ... Stable position of molecular axis 502 ... Stable position of molecular axis 503 ... Spontaneous polarization Direction 504 ... Apex angle of cone 505 ... Applied electric field in two directions 700, 700 '... Synthesized writing Only signals 701, 702 ... Synthesized non-selection signal 706 ... Pulse for state inversion 1100, 1100 '... Synthesized selection preliminary signal 1110 ... Synthesized write signal (ON) 1120 ... Synthesized write signal (OFF ) 1130 ... 1140 Combined non-selection signal 1150 ... Preliminary signal pulse portion that contributes to state reset

Claims (4)

[Claims] 1. A pair of substrates, at least one of which is provided with stripe-shaped electrodes and are transparent, are arranged so as to oppose each other so that the electrodes are substantially orthogonal to each other, and a ferroelectric having a chiral smectic C phase between the substrates. In a liquid crystal display device in which a liquid crystal is sandwiched, when the three eigenvalues of the dielectric tensor of the chiral smectic C layer liquid crystal are εi (i = 1, 2, 3), at least one of ε _{3 and} ε ₂ is the remaining 2 A liquid crystal display device having a frequency region larger than the component. 2. A pulse width modulation drive in which an AC pulse having a pulse width shorter than a pulse width used for forming a selected state is applied during non-selection, wherein the AC pulse has a constant width and the polarity is continuously inverted. _2. The liquid crystal display device according to claim 1, wherein the alternating current pulse is in a frequency range in which at least one of ε _{3 and} ε ₂ is larger than the remaining two components. Driving method. 3. A pulse width used to form a state when selected is τ
Then, the two data signals have pulse widths of τ / n (n
= 1, 2, 3, 4, 5, ..., The same applies to the following), and the pulse width of the selection signal and the non-selection signal, which are scanning signals, is τ / 3. The method for driving a liquid crystal display element according to claim 2, wherein the number of output voltage levels of the selection signal and the data signal is 3, which is composed of a combination of pulses of n and an amplitude of ± Vs. 4. When the pulse width used for forming one state at the time of selection is τ, two data signals have pulse widths of τ /
n, composed of a combination of pulses with an amplitude of ± Vd,
The selection signal, the selection preliminary signal and the non-selection signal which are scanning signals are composed of a combination of pulses having a pulse width of τ / n and an amplitude of ± Vs, and the number of output voltage levels of the selection signal and the data signal is three. 3. The liquid crystal display element according to claim 2, wherein the selection preliminary signal is applied to the scan electrode to be selected next, and the total scanning time is shorter than that of the method according to claim 3. Driving method.