HK1103452B - Bistable nematic liquid crystal device - Google Patents

Description

Bistable nematic liquid crystal device

This application is a divisional application of application 200410030468.1.

The present invention relates to bistable nematic liquid crystal devices.

Liquid crystal devices typically comprise a thin layer of liquid crystal material sandwiched between cell walls. The optically transparent electrodes are mounted on the walls such that an electric field is applied between the layers and causes rearrangement of the liquid crystal molecules.

There are three known types of liquid crystal materials, nematic, cholesteric and smectic, each having a different molecular arrangement. The present invention is a device employing nematic materials.

In order to provide a display with a large number of addressable elements, the electrodes are usually manufactured as a series of row electrodes placed on one wall and a series of column electrodes placed on the other cell wall. These form an x, y matrix of addressable elements or pixels, for example, and are typically addressed using a mean square addressing method for twisted nematic devices.

The twisted nematic and phase change modes of liquid crystal devices are set in the ON state by applying a suitable voltage and in the OFF state when the applied voltage falls below a certain lower voltage value, i.e. the devices are monostable. For twisted nematic devices (e.g. 90 or 270 twist as in US 4,596,446), the number of elements that can be addressed by the mean square value is limited by the steepness of the device transmittance versus voltage curve, for which Alt and Pleschko is published in IEEE Trans ED21, 1974, page 146-. One way to improve the number of pixels is to combine a thin film transistor with each adjacent pixel; such a display is called an active matrix display. One advantage of nematic devices is the relatively low voltage requirements. They also have mechanical stability and have a wide temperature operating range. This allows the manufacture of small and portable battery powered displays.

Another method of large display addressing is to use bistable liquid crystal devices. Ferroelectric liquid crystal displays can be incorporated in bistable devices using smectic liquid crystal materials and appropriate cell wall surface alignment. Such devices are surface stabilized ferroelectric liquid crystal devices (SSFELCDs) described by the following et al: l J Yu, H Lee, C S Bak and M M Labes, Phys Rev Lett36, 7, 388 (1976); r B Meyer, Mol Crystal Liq Crystal.40, 33, (1977); NA Clark and S T Lagerwall, apple Phys Lett, 36, 11, 899 (1980). One disadvantage of ferroelectric devices is that a relatively large voltage is required to switch the material. This high voltage makes small portable, battery-driven displays expensive. Also such displays have other problems such as lack of resistance to vibration, limited temperature range and also electrically induced disadvantages such as needles.

A display having the technical advantages described above without the problems described above can be manufactured if bistable surface adhesion can be achieved with a nematic phase.

Durand et al have shown that the nematic phase can be switched between two alignment states by using chiral ions or variable electric coupling: a Charbi, R Barberi, G Durand and PMartiot-Largarde, patent application No. WO 91/11747, (1991) "Bistable chiral controlled liquid Crystal optical device", G Durand, R Barbei, M Giocondo, PMartiot-Largarde, patent application No. WO 92/00546(1991) "nematic liquid Crystal display whose surface Bitability is controlled by variable electric Effect". These summaries are as follows:

in patent application WO 91/11747 a device has the following features:

1. a liquid crystal cell fabricated with two surfaces with appropriate thickness of SiO coatings and evaporation angles allows two stable states to exist on each surface. Also the two states on one surface are designed to be at a horizontal angle that differs by 45 ° and the surface is oriented such that each of the two resulting domains is non-twisted.

2. The liquid crystal cell (6 μm thick) was filled with 5CB doped with 0.5% phenylquinine bromide and 1.8% phenyllactic acid. The former is a positively charged chiral ion with a left twist and the latter a negatively charged chiral ion with a right twist. The concentration is such that the final mixture has a long pitch so that the state is uniform in a thin cell.

3. Application of a pulse of 110V dc 40 mus causes switching between the two states. A lower threshold is observed for longer pulses, such as an 80V threshold for a 300 mus pulse.

4. The additional properly oriented polarization causes one state to appear black and the other white, with a contrast of about 20.

5. Also mentioned is a modified device that creates a short pitch mixture of chiral ions at monostable surfaces with different apex binding energies. Switching between the 180 ° twisted state and the homogeneous state was observed for pulses over 50V in a 4 μm liquid crystal cell.

In patent application WO 92/00546 a device has the following features:

the liquid crystal cell is fabricated using two surfaces with appropriate thickness of the SiO coating and evaporation angle so that two stable states exist on each surface. Also the two states on one surface are designed to differ in azimuth by 45 deg., and the surface is oriented such that each of the two resulting domains is non-twisted.

The surfaces are also oriented in such a way that the pretilt states on one surface line up with the unnumbered states on the other surface, and vice versa. When 5CB is filled, two states are shown in fig. 7B and 7C.

Application of a pulse of 14V dc for 100 mus along a 1 um cell causes switching between states. The final state depends on the sign of the pulse due to coupling with the variable electrical polarization. Switching in both directions observes the same voltage threshold.

The surface used by Durand to achieve bistable alignment is a thin layer of SiO evaporated at a precise tilt angle. However, this method has the disadvantage that any evaporation angle, deviation of the layer thickness or even any deposition parameter tends to produce a surface with a unique monostable collimation. This makes the unique oblique evaporation technique unstable, or very difficult, for large area displays.

Patent US 4,333,708 describes a multistable liquid crystal device in which the cell walls are profiled to provide a single point arrangement. Such a substrate configuration provides a multi-stable structure with director alignment, since disclination must be moved to switch in a stable structure. Switching is achieved by applying an electric field.

Another bistable nematic device is described in GB.2,286,467-A. It uses a precisely formed double grating on at least one cell wall. The double grating allows the liquid crystal molecules to adopt two different angular alignment directions when appropriate electrical signals are applied to the liquid crystal cell electrodes, such as the direct current and variable electrical polarization coupling described in patent application No. wo 92/00546. Because the director is very close in the layer plane in the two tilted states, the coupling between the director and the variable electrical component can be small, which hinders switching in certain circumstances.

According to the invention, the above-mentioned drawback can be overcome by surface treating at least one cell wall to allow nematic liquid crystal molecules to adopt one of two pretilt angles in the same azimuthal plane. The liquid crystal cell can be electrically switched between these two states so that the information display can remain after power is removed.

The same azimuth plane is explained below; the cell walls are located in the x, y plane, which means that orthogonal to the cell walls is the z axis. Two pretilt angles in the same azimuthal plane mean two different molecular positions in the same x, z plane.

The bistable nematic liquid crystal device according to the invention comprises:

two cell walls comprising a layer of liquid crystal material;

electrodes located on both walls;

surface alignment on opposing surfaces of the two cell walls to provide alignment for the liquid crystal molecules;

means for distinguishing the switching states of the liquid crystal material;

has the following characteristics:

a surface alignment grating located on at least one of the cell walls which allows the liquid crystal molecules to adopt two different pretilt angles in the same azimuthal plane;

with such an arrangement, two stable liquid crystal molecular structures can exist when appropriate electrical signals are applied to the electrodes.

The grating has a symmetrical or asymmetrical groove profile.

The grating may have an asymmetric groove profile that will produce a pretilt angle of less than 90 deg., such as 50 deg. to 90 deg.. An asymmetric profile can be defined as a surface where there is no h value:

ψ_x(h-x)＝ψ_x(h+x) ......(1)

for all values of x, ψ is a function describing the surface.

The grating may be applied to both cassette walls and may be the same or different in shape on each wall. Alternatively the grating profile may vary within each pixel area and/or within the inter-pixel gaps between the electrodes. One or both cassette walls may be coated with a surfactant such as lethicin.

The liquid crystal material may be untwisted in one or both of the stable molecular structures.

The cassette walls may be formed from a relatively thick non-flexible material, such as glass, or one or both of the cassette walls may be formed from a flexible material, such as a thin layer of glass, or a plastic flexible material, such as polyolefin or polypropylene. The plastic capsule wall may be molded on its inner surface to provide a grating. In addition, embossing may provide small mounds (e.g., 1-3 μm high and 5-50 μm or more wide) to assist in proper separation of cell wall spacings and also to block liquid crystal material flow when the liquid crystal cell is flexible. Alternatively the piers may be formed by collimating layer material.

The grating may be a photopolymer profile layer formed by a photolithographic process such as M C Hutley, diffraction grating (Academic Press, London 1982) p 95-125; and F Hom, Physics world, 33(march 1993). In addition, the double grating can be formed by pressing; m T Gale, J kanepand K Knop, J app. photo Eng, 4, 2, 41(1978), or score line formation; e G Loewenand R S Wiley, Proc SPIE, 88(1987), or by charge carrier layer transfer.

The electrodes may be formed by an arrangement of row and column electrodes and an x, y matrix of addressable elements or display pixels. Typical electrodes are 200 μm wide with a pitch of 20 μm.

In addition, the electrodes may be arranged in other display modes such as r-theta matrix or 7 or 8 grid display.

The invention will be described by way of example only with reference to the accompanying drawings.

FIG. 1 is a plan view of a matrix multiplexed liquid crystal display;

FIG. 2 is a cross-sectional portion of the display of FIG. 1;

fig. 3 shows top and side views of a mask and exposed geometric surfaces used to create the grating surface.

Fig. 4 is a cross-sectional portion of a liquid crystal director structure on the grating plane that results in a higher pretilt angle.

Fig. 5 is a cross-sectional portion of a liquid crystal director structure on the grating plane that results in a lower pretilt angle.

Fig. 6 is the energy of two pretilt structures as a function of trench depth to pitch ratio (h/w).

Figure 7 shows a cross-sectional portion of a liquid crystal cell structure which permits bistable switching between two states.

FIG. 8 shows the transmission of the liquid crystal cell and the applied signal as a function of time.

Figure 9 shows an example of a multi-path diagram for a bi-stable device.

FIG. 10 shows an alternative liquid crystal cell structure for bistable switching.

FIG. 11 shows a liquid crystal cell structure for bistable switching between untwisted and twisted states.

The display in fig. 1, 2 comprises a liquid crystal cell 1 formed by a layer 2 of a nematic or long pitch cholesteric liquid crystal material contained between glass walls 3, 4. The backing ring 5 holds the walls at a typical 1-6 μm pitch. In addition, a large number of beads of the same size are distributed within the liquid crystal to maintain accurate wall spacing. Strip-like row electrodes 6, such as SnO2 or ITO (indium tin oxide), are formed on the wall 3 and the same column electrodes 7 are formed on the other wall 4. A matrix of m x n addressable elements or pixels is formed by m rows and n columns of electrodes. Each pixel is formed by the intersection of a row and column electrode.

A row driver 8 supplies a voltage to each row electrode 6. Also a column driver 9 supplies a voltage to each column electrode 7. The control of the applied voltage is performed by a control logic circuit 10 that receives the power of the power supply 11 and the timing of the clock 12.

The liquid crystal cell 1 is flanked on both sides by polarizers 13, 13', respectively, whose polarization axes virtually intersect each other and virtually intersect the aligned director R, if any, on the adjacent walls 3, 4, described below, at 45 deg.. Further optical compensation layers 17, such as stretched polymers, may be added adjacent the liquid crystal layer 2 between the cell walls and the polarisers.

A partially reflecting mirror 16 may be arranged together with the light source 15 behind the liquid crystal cell 1. These make the display visible in reflected light and light can escape from a dim background. For the transmission means, the mirror 16 can be omitted.

Prior to installation, at least one of the cell walls 3, 4 is treated with a collimating grating to provide a bistable pretilt angle. Other surfaces may be treated with planar (i.e., zero or small angle between pretilt and collimated dipole) or perpendicular monostable surfaces, or degenerate planar surfaces (i.e., zero or small angle between pretilt and non-collimated dipole).

Finally the cell is filled with a nematic material, e.g. E7, ZLI2293 or TX2A (Merck).

An example of a method of constructing a grating surface will be described with reference to figure 3.

Example 1:

one piece of ITO coated glass formed box wall 3, 4 was cleaned with acetone and isopropanol and spiral coated with a photoresist material (Shipley 1805) at 3000rpm for 30 seconds to form a coating with a thickness of 0.55 μm. Then soft-dried at 90 ℃ for 30 minutes.

Contact exposure was then carried out on the coated crystal walls 3, 4 using a chrome mask 20 containing 0.5 μm lines 21 and 0.5 μm spaces 22 (where the overall pitch is 1 μm), as shown in fig. 3. The exposure is done with non-orthographic light, in this case at an angle of 60 °. The mask 20 is oriented such that the direction of the trenches is substantially perpendicular to the plane of incidence as shown in figure 3. Exposure of such geometric surfaces results in an asymmetric light intensity distribution and thus an asymmetric grating profile (see for example b.j.lin, opt.soc.am., 62, 976 (1972)). Coating the walls 3, 4 of the cartridge with a light intensity of 0.8mW/cm under a mercury lamp (Osram Hg/100)²Exposure time of about 40 to 180 seconds, as described below.

After exposure the coated cartridge walls 3, 4 were detached from the mask 20 and developed in Shipley MF319 for 10 seconds followed by rinsing in deionized water. This results in a box wall surface shape with an asymmetric surface mode that forms the desired grating profile. The photoresist hardened upon exposure to deep UV radiation (254nm) followed by oven drying at 160 ℃ for 45 minutes. This is done to ensure insolubility of the photoresist in the liquid crystal. Finally the grating surface is treated with a surfactant lecithin solution to create vertical boundary conditions.

Finite element analysis is employed to predict the free layer molecular (more precisely, director) configuration of nematic materials on the surface of such gratings. The results are shown in fig. 4,5 and 6, where the short lines represent the liquid crystal director throughout the thickness of the layer and the envelope of the short lines at the bottom represents the grating profile. In this case, the grating surface is described by a function:

where h is the groove depth, w is the pitch and A is the asymmetry factor. In fig. 4 and 5, a is 0.5 and h/w is 0.6. In fig. 4, the finite element mesh allows for initial director relaxation at an 80 ° tilt angle. In this case the structure reverts to a pretilt angle of 89.5 deg.. However, if the initial director tilt angle is set to 30 ° the grid reverts to a pretilt angle of 23.0 °, as shown in fig. 5. Nematic liquid crystals can thus adopt two different configurations depending on the starting conditions.

In practice nematic liquid crystal materials will revert to the one of the two configurations with the lowest overall deformation energy. Fig. 6 shows the relationship of the total energy (arbitrary units) and the ratio of the trench depth to the pitch (h/w) for the high pretilt (solid circles) and low pretilt (open circles) states. For low h/w, the high pretilt state has the lowest energy and therefore the nematic phase will adopt a high pretilt state. In contrast, for high h/w, the low pretilt state has the lowest energy and thus this state is formed. However, when h/w is 0.52, the states have the same energy and thus both can exist without reverting to the other. Bistability is therefore observed in the pretilt angle if one surface is structured as such, or close to such. Referring to the construction details above, it was found that an exposure time of 80 seconds resulted in a bistable surface. The bistability in this case is purely a surface function and is not dependent on any particular cell geometry. This means a distinction from the prior art, such as US 4333708 (1982).

A suitable liquid crystal cell structure allowing switching between bistable states is shown in figure 7, which is a pseudo-section of a device in which a layer 2 of nematic liquid crystal material having a positive dielectric anisotropy is contained between a bistable grating surface 25 and a monostable homeotropic surface 26. The latter surface 26 should be, for example, a flat photoresist surface coated with lecithin. In this device the liquid crystal molecules can exist in two stable states. In state (a) the two surfaces 25, 26 are perpendicular, whereas in (b) the grating surface 25 is in a low pretilt state resulting in a tilted structure. For many nematic materials, a tilt or bend deformation will result in a large variable electric polarization, represented by vector P in FIG. 7. A dc pulse can couple this polarization and will have good or poor structure (b) depending on its sign.

With the device in state (a), the application of a positive pulse in the vertical configuration will result in a pulsation despite the positive dielectric anisotropy. These pulsations may be sufficient to drive the system beyond the energy barrier separating the two collimation states. The system will drop to state (b) at the end of the pulse because the sign of the field couples well with the variable polarization. In state (b) with the system, a negative-sign pulse will again destroy the system. But now it will revert to state (a) when its sign does not favor the formation of a variable electrical polarization. In its vertical state, the bistable surface is tilted at an angle slightly less than 90 (e.g., 89.5). This is sufficient to control the tilt direction obtained when the liquid crystal cell is switched to state (a).

A special liquid crystal cell containing the nematic phase ZL12293(Merck) is sandwiched between a bistable grating surface and a vertically flat surface. The cell thickness was 3 μm. The transmission through the cell was measured by applying a direct current pulse at room temperature (20 ℃). The polarisers and analyser 13, 13' located on each side of the liquid crystal cell 1 intersect each other and are directed at the grating grooves at ± 45 °. In this setup, the two states (a) and (b) in fig. 7, when addressed therewith, display black and white, respectively.

Fig. 8 shows the applied voltage pulse (low trace) and the optical response (high trace) as a function of time. Each pulse has a peak of 55.0 volts and a duration of 3.3 ms. The pulse interval is 300 ms. With the first application of a positive pulse, the transmission changes from dark to light indicating that the liquid crystal cell switches from state fig. 7(a) to state (b). The second positive pulse causes a transient change in transmission since the mean square effect of the coupling of the positive dielectric anisotropy results in an immediate switching of the bulk material to state (a). In this case, however, the liquid crystal cell does not get stuck on the surface and remains in state (b). The next pulse is negative in sign and switches the liquid crystal cell from state (b) to state (a). The last second negative pulse causes the cell to be in state (a). This experiment shows that the liquid crystal cell does not change state at each pulse unless it has the correct sign. It thus proves that the system is bistable and that the final state can be reliably selected according to the sign of the applied pulse.

Switching occurs over a wide temperature interval. As the temperature increases, the voltage required for switching decreases. For example, at 30 c a voltage of 44.8V is required for bistable switching and at 50 c only a voltage of 28.8V is required. Similarly, the pulse length required for locking down for a fixed voltage decreases with temperature.

After the data is obtained, the cell is decomposed and the grating surface is characterized by AFM (atomic force microscopy). The asymmetric mode fitted by equation 2 results in an asymmetry factor of 1 μm pitch, 0.425 μm groove depth (h/w 0.425) and a 0.5. Comparing the results of fig. 6, this grating has a bistable condition with low values of h/w (0.425 compared to 0.52). Equation 2, however, does not fit AFM data exactly because the actual surface has a chamfer angle that requires the addition of higher harmonics. Other factors such as AFM tip radius also need to be considered for more accurate comparison. It can therefore be concluded that the measured surface modes are similar to the predicted conditions for bistability.

Successful switching of a single pixel allows the design of an appropriate multi-pixel approach for the selection of several adjacent pixels. Fig. 9 shows a specific example of such a method. As shown, the pixels in 4 consecutive rows R1, R2, R3, R4 in a column are to be switched. The two possible alignment states can be arbitrarily defined as ON and OFF states. Rows R1 and R4 are to be switched to the ON state, and rows R2 and R3 are in the OFF state. The + Vs strobe for three time slots followed by-Vs for three time slots (ts) is applied to each row stream. The data waveform is applied to the columns as shown and includes-Vd of 1ts with + Vd of 1ts for ON pixels and 1ts with + Vd for OFF pixels.

Now consider a particular pixel located at point a. The resultant waveform includes large positive and negative pulses that disrupt the nematic orientation and increase its energy to reach a barrier separating the two bistable surface states. In this field application condition, the liquid crystal molecules align along the electric field as in a typical monostable nematic device, as shown in FIG. 7 a. These large "reset" pulses of opposite polarity are followed by small pulses which are still large enough to dominate the final selected state of the pixel during the restoration of orientation. The electrical balance is achieved by a small pulse of opposite polarity to the switching pulse followed by two large pulses. In addition, polarity switching may be used in adjacent display addressing times.

The bistable device described above achieves a final state selection by the property of variable electrical polarization in one state. This structure must therefore contain a chamfer. In the experimental example only one surface is allowed to switch but the working device can be done in a two-surface switch. The only constraint that exists is that the low pretilt states on each surface should take different values in order to maintain a finite slope. However, even if the low pretilt state is equal, the cell can still be switched if it contains a two-frequency nematic material, i.e. a positive dielectric anisotropy at low frequencies and a negative dielectric anisotropy at high frequencies. An example of such a material is TX2A (Merck), which has a crossover frequency of 6 kHz. Fig. 9 shows a cross section of the present structure. In state (a) the nematic phase is driven to a low pretilt angle by applying a high frequency signal via the cell. The surface is then formed and the liquid crystal cell is switched to state (b). Conversely, a low frequency signal will drive the nematic phase to a high pretilt angle and the liquid crystal cell will be switched to state (a).

Example 2:

a second example of a bistable device is described below. A piece of box wall formed of ITO coated glass was cleaned with acetone and isopropanol and spiral coated with a photoresist material (Shipley 1813) at 3000rpm for 30 seconds to form a coating of 1.5 μm thickness. Then soft-dried at 90 ℃ for 30 minutes.

Contact exposure was then accomplished using a chrome mask containing 0.5 μm lines and 0.5 μm spaces (where the overall pitch was 1 μm). In this example the exposure is done with ortho-light. Exposure of such a geometric surface results in a symmetrical light intensity distribution and thus a symmetrical grating profile. Samples were tested under a mercury lamp (Osram Hg/100) with a light intensity of 0.8mW/cm²And (6) exposing.

After exposure the sample was detached from the mask 20 and developed in MF319 for 20 seconds followed by rinsing in deionised water. This gives the sample surface shape a symmetrical surface mode. The photoresist hardened by exposure to deep UV radiation (254nm) followed by oven drying at 160 ℃ for 45 minutes. This is done to ensure insolubility of the photoresist in the liquid crystal. Finally, the grating surface is treated with a chromium complex surfactant solution to create vertical boundary conditions.

A specific surface was fabricated using the 360 second exposure time method described above. AFM analysis on this grating showed it to have a 1 μm pitch and a 1.2 μm thick symmetric profile. This surface was established at a vertical plane with respect to the plane to form a liquid crystal cell having a thickness of 2.0 μm. The cell was filled with the nematic material E7(Merck) in the isotropic phase and then cooled to room temperature. Microscopic observation revealed a mixture of two bistable states, shown in fig. 7(a) and (b).

The liquid crystal cell was oriented between crossed polarizers such that the groove direction was 45 deg. to the polarizer direction. Therefore, the state (a) is a bright state and the state (b) is a dark state. Unipolar pulses of alternating sign are then applied to the liquid crystal cell. The pulse length was set to 5.4ms and the pulse interval was 1 s. When the peak voltage of the applied pulse increases to 20.3V, full switching occurs in states (a) and (b). In the same way as for the data shown in fig. 8, pairs of pulses were also applied to the liquid crystal cell. Again only the first pulse changes the system state and the second pulse only causes a non-locking transient response. In which case the optical response time can also be measured. The 10% -90% response time for switching from (a) to (b) is 8.0ms and the response time for switching from (b) to (a) is 1.2 ms. Further analysis of this cell revealed that the bistable states (a) and (b) resulted in pretilt angles of 90 ° and 0 ° at the grating face, respectively. The present sample thus demonstrates the largest possible change in pretilt angle.

The light in the structures shown in fig. 7 and 10 is optimal when the cell thickness d is given as follows:

where λ is the operating wavelength and Δ n_avThe average of the internal planar components (parallel to the liquid crystal walls) of the nematic birefringence. Δ n shown in the structure of FIG. 10_avLarger than in FIG. 7, the cell thickness is reduced and the optical switching speed is increased. However, the use of two frequencies for the nematic phase limits the choice of materials available, again leading to more complex addressing schemes, but can allow low voltage operation.

Example 3:

the bistable grating surface may also be built up opposite to the flat surface. Such a liquid crystal cell comprises a grating having the same profile as described in example 2. It is built opposite to an abrasive polymer face formed with a layer of P132 polyimide (ciba geigy). The polishing direction on the polyimide surface was set to be parallel to the grating groove direction on the grating surface. The cell gap was set to 2.5 μm and nematic phase E7 was used to fill the cell. Cooling to room temperature after filling produces two states as shown in fig. 11. This figure differs from figure 7 in that the groove direction on the bistable surface is now in the plane of the page (in the x, y plane). Thus the 90 pretilt state at the grating forms a hybrid structure as shown in (a ') and the 0 pretilt state at the grating forms a twisted structure as shown in (b'). To achieve optical contrast between the states, the cell is placed between crossed polariser 13, 13' orientations so that the grating grooves (and rubbing direction) are parallel to one polariser, but the polariser can be rotated to optimum contrast between the two switched states. Thus, state (b ') is a bright state and state (a') is a dark state. With a 5.3ms unipolar pulse, switching between (a ') and (b') occurred at a peak voltage of 56.7V. The optical response time is 110ms for switching from (a ') to (b') and 1.4ms for switching from (b ') to (a').

The bright state (b') has a 90 ° volume twist. When a general TN structure is employed, maximum transmission is obtained when N is an integer (c.h.gooch and h.a.tarry, j.phys.d: appl.phys., 81575 (1975)):

where Δ n is the nematic second refraction, d is the cell separation and λ is the operating wavelength. A bistable device with an operating wavelength of 530nm and N1 and using E7(Δ N0.22) will therefore have a cell separation of 2.1 μm.

The structure described in comparative example 2, which has the optimum thickness given by equation 3. For that example, Δ n_avA thickness of 1.2 μm is given for Δ n/2 and thus equation 3. A non-twisted bistable device will therefore always have optimal optics at thin cell spacings and will therefore switch at low voltages with short optical response times.

A cholesteric dopant (e.g., < 1% of CB 15 Merck) can be added to prevent disclination. Alternatively, these disclinations may be prevented by aligning the grooves in a direction that is not parallel to the rubbing alignment direction, e.g., about 45 °.

The grating surfaces for these devices can be fabricated using a range of techniques as listed above. The homeotropic treatment can be any surfactant that has good adhesion to the grating surface. Such treatment should result in depinning. I.e., alignment towards a particular nematic orientation does not cause rigid positional alignment of the nematic phase on the surface.

From the above analysis, it can be seen that in order to achieve bistability for a given asymmetry, the grating modes must have a certain h/w. The absolute scale of the mode is limited by other factors. If the groove depth and pitch are too large then the diffraction effect will become significant and result in a loss of device throughput. Also, if the groove depth is similar to the cell thickness, the proximity of the groove peaks of the opposing planar surfaces to the opposing planar surfaces may prevent bistable switching. If a device like that shown in fig. 10 requires two gratings, a large groove depth compared to the cell thickness will inevitably result in switching dependent on the two mode phases. This will add to the complexity of the device manufacturing process.

If the groove depth and pitch are too small, the problem still exists. For the constant h/w, the energy intensity of the overall deformation on the surface becomes larger as the pitch becomes smaller. This energy is eventually similar to the local attachment energy of a nematic phase on the surface. The structures represented in fig. 4 and 5 (assuming infinite attachment energy) will therefore no longer be obtained and the bistability inevitably lost. Typical values of h and w are about 0.5 μm and 1.0 μm in the interval of about 0.1 to 10 μm and 0.05 to 5 μm, respectively.

A small amount of dichroic dye, such as 1-5%, may be incorporated into the liquid crystal material. This may be done with or without polarizers to provide color, improve contrast, or operate as a guest-host type device, such as material D124 in E63 (Merck). The polarizer of the device (with or without dye) can be rotated to the best contrast between the two switched states of the device.

Claims (4)

1. A bistable nematic liquid crystal device comprising:

a first cell wall and a second cell wall, said first cell wall and said second cell wall enclosing a layer of liquid crystal material,

wherein the first cell wall has a bistable pre-tilt angle first surface treated with a collimating grating to provide molecules of nematic liquid crystal material with either of two pre-tilt angles in the same azimuthal plane, the second cell wall has a first surface treated to provide monostable collimation to the molecules of nematic liquid crystal material,

wherein the bistable nematic liquid crystal device provides two stable and optically distinguishable liquid crystal configurations.

2. The apparatus of claim 1, wherein the first surface of the second cartridge wall is one of planar, degenerate planar, and perpendicular surface processed.

3. The apparatus of claim 1, further comprising a plurality of beads of the same size dispersed in the layer of liquid crystal material to maintain a precise wall spacing.

4. A bistable nematic liquid crystal device providing a first stable liquid crystal configuration and a second stable liquid crystal configuration, the device having means for distinguishing between the first stable liquid crystal configuration and the second stable liquid crystal configuration, the device comprising a cell comprising two cell walls enclosing a layer of liquid crystal material, electrode structures on both walls, surface alignments on opposite surfaces of the two cell walls providing alignment for liquid crystal molecules, at least one cell wall having a surface alignment grating providing two different pre-tilt angles in the same azimuthal plane for molecules of the liquid crystal material, wherein the first stable liquid crystal configuration is a twisted molecular configuration and the second stable liquid crystal configuration is a non-twisted molecular configuration.