This invention was made with Government support under Contract No. F49620-84-C-0096 awarded by ^' he Department of the Air Force. The Government has certain rights in this invention.

RELATED APPLICATION

This invention is related to Serial No. 879,719 , filed JUNE 27. 1986 by Bernard Soffer, "Optical Intensi¬ ty-to-Position Mapping Apparatus and Method" , and assigned to Hughes Aircraft Company, the assignee of the present invention.

BACKGROUND OF THE INVENTION Field of the Invention This invention relates to the characterization and processing of optical beams, and more particularly to a system and method for deflecting light to convert the spatial intensity spectrum of a light beam to a positional mapping, or to implement a switching mechanism.

Description of the Related Art

The distribution of optical intensities across a single light beam or a series of separate light beams may form the basis for optical computing and logic processing by first converting the distribution of light intensities to a positional mapping, and then operating upon the signal at each different position to perform the desired

computing or processing. The cross-section of a light beam 2 can be envisioned as an aggregation of a large number of pixel locations 4, as illustrated in FIG. 1. Because the pattern of light intensity at each different pixel location can be imaged or mapped elsewhere in the system, this pattern may be employed as an information coding system. For various computing and logic processing purposes it is desirable to group the various pixels by their respective optical intensity levels, rather than by their spatial positions within the beam. In other words, it is desirable to be able to convert the spatial intensi¬ ty pattern of beam 2 to a positional mapping 6, in which all pixels having a common optical intensity are mapped to a common position on a distance spectrum. In the example of FIG. 1 it is assumed, for a simple special case, that the optical intensity levels are grouped into a limited number of discrete levels, rather than extending over a continuum of intensities. Thus, the mapping process may yield a number of discrete spikes 8 which are separated by positional distance, the height of each spike varying with the number of pixels having an optical intensity corre¬ sponding to the position of the spike. This representa¬ tion is called a histogram. Where separate input light beams are presented, such as in discrete optical fibers, it may likewise be desirable to switch the separate beams among various output locations in accordance with their intensities.

A system which accomplishes the desired intensity-to- position mapping is described in U.S. Patent No. 4,351,589 to Pierre H. Chavel et al. , assigned to Hughes Aircraft Company, the assignee of the present invention. The patent discloses the use of liquid crystals to produce variable gratings, which diffract incoming light by vary¬ ing amounts depending upon the grating period. The opti- cal intensities at the different locations in an input

light beam control the grating period at corresponding locations in the liquid crystal media. The variable gratings were employed to convert the spatial intensity distribution within the input beam to a positional mapping of intensities, from which the desired computing and logic functions could be accomplished. Applications of this variable grating mode (VGM) device are disclosed in an article by B. H. Soffer et al., "Optical Computing With

Variable Grating Mode Liquid Crystal Devices", 1980 Inter- national Optical Computing Conference, SPIE Vol. 232, pages 128-136.

While it facilitates optical data and image process¬ ing and optical logic and computing, the variable grating device disclosed in the patent in its present state of development has somewhat slow response times, and works only at low temporal frequencies.

A paper has been published disclosing a single pris¬ matic wedge of liquid crystal, across which a variable voltage is applied to deflect a light beam, M. A. Muriel and J. A. Martin-Pereda, "Digital Light Beam Deflector with Liquid Crystals", 1980 European Conference on Optical Systems and Applications (Utrecht) , SPIE Vol. 236, pages 386-388. The amount of deflection which a light ray undergoes after crossing the wedge is said to depend upon the applied voltage. However, the light was subjected to a large amount of scattering in transit through the wedge, and the single cell disclosed was not suitable for use in intensity-to-positional mapping. The response time is also slow. A related technological issue involved an attempt to control the focal length of a lens by means of an applied voltage. This was described in an article by Susumu Sato, "Liquid-Crystal Lens-Cells - with Variable Focal Length" , Japanese Journal of Applied Physics, Vol. 18, No. 9, September 1979, pages 1679-1684. In this approach liquid

crystal cells shaped like a plano-convex lens or a plano¬ concave lens were prepared. Electric or magnetic fields were applied across the lens-cell to vary its focal length. The authors encountered excessive light scatter- ing and slow response time due to the thickness of the liquid crystal cell.

SUMMARY OF THE INVENTION

In view of the above problems associated with the prior art, the object of the present invention is to provide a light deflection system and method capable of mapping the spatial optical intensities within a single light beam or a series of light beams onto a positional array, or for switching among light beams, with a faster response time than available in the prior art, without excessive light scattering, and with an improved temporal frequency response.

These objects are accomplished in .the present inven¬ tion by providing an array of voltage gradient modules, with each module comprising a layer of electro-optic material such as liquid crystals and a means for estab¬ lishing a variable spatial voltage gradient and electric field across the material in a direction transverse to the input light. The voltage gradients and fields of respec- tive modules are varied in accordance with the optical intensities at corresponding locations in the input light. The modules operate primarily by a diffractive process to produce optical outputs which vary in direction in accor¬ dance with the modules' voltage gradient and field varia- tions, thereby mapping the modules' input optical intensi¬ ties onto the directions of their optical outputs. A lens in the output path focuses the directionally mapped module outputs to a positional mapping of the input light inten¬ sities in the focal plane of the lens. The processed

input light intensity distribution is reconstructed in the image plane of the lens.

The voltage gradients and fields are created by the interaction between an array of interdigitated electrodes on one side of the electro-optic layer, and a counter electrode on the other side of the layer which may be either a planar electrode or a second array of interdigi¬ tated electrodes. Voltages are applied to the electrodes on either side of the electro-optic layer to establish the desired spatial voltage gradients through the layer.

The invention may be implemented in a variety of specific applications. These include a reflective liquid crystal light valve, a transmissive liquid crystal light valve, and an optical switch adapted to switch a matrix of input light beams among a plurality of output locations. The invention is capable of accomplishing all of the non-linear optical functions described in Patent No. 4,351,589, mentioned above, but with a higher speed and better resolution. The invention is particularly suited for optical information processing applications.

These and other features and advantages of the inven¬ tion will be apparent to those skilled in the art from the following detailed description of preferred embodiments, taken together with the accompanying drawings, in which:

DESCRIPTION OF THE DRAWINGS

FIG. 1 is an illustrative diagram of the intensity- to-position mapping, described above, achieved by the invention; FIG. 2 is a side elevation view of a lens and ray diagram illustrating one of the principles of the inven¬ tion, that angles and positions can be mapped to each other;

FIG. 3 is a side elevation view illustrating the voltage gradient modules employed in the invention;

1 FIG. 4a is a simplified sectional view illustrating one electrode arrangement for establishing the voltage gradients;

FIG. 4b is a graph indicating the voltage gradients 5 and fields established by the embodiment of FIG. 4a;

FIG. 5 is a simplified sectional view of another electrode arrangement;

FIG. 6 is a sectional view of an embodiment in which input light is transmitted through a liquid crystal light 10 valve to achieve positional mapping;

FIG. 7 is a sectional view of another embodiment in which a reflective liquid crystal light valve utilizes separate input and readout beams to achieve positional mapping; 15 FIG. 8 is a fragmentary plan view of an electrode section;

FIGs. 9 and 10 are graphs of typical output light distributions for two different input levels; and

FIG. 11 is a perspective view of an optical switch 20 constructed in accordance with the present invention.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

The present invention makes use of a novel light deflection mechanism to achieve a positional mapping of a

25 spatial optical intensity profile with subsequent recon¬ struction of the profile, with an intermediate directional mapping. To illustrate the transformation of the interme¬ diate directional mapping to the final positional mapping, two light beams 10 and 12 are shown directed upon a lens

3.0 14 in FIG. 2. Each beam is collimated and occupies a substantial cross-sectional area, as indicated by the multiple ray lines for each beam. So long as the rays of each beam are parallel, they are focused by lens 14 essen¬ tially to points on the lens focal plane 16. The focal

35 position of each beam on the focal plane will depend upon

its direction relative to the optical axis of the lens. Beam 10, which is parallel to the lens axis, will be focused to a point along that axis. This will apply to all of the rays in beam 10, including those which are far removed from the lens axis. Since all of the rays in beam 12 are parallel, they will likewise be focused to a single point on the lens focal plane, even though the rays them¬ selves may be distributed over a substantial portion of the lens. The distance of the beam's focus point from the lens axis will vary with the angle between the beam and axis.

One embodiment of the present invention utilizes the phenomena just described to achieve a mapping of spatial beam intensities to position in a two-step process. First, the beam is processed so that a series of small output beams are produced, with the position of each output beam corresponding to a particular location in the input beam, and the directions of the output beams corre¬ sponding to the optical intensities at their respective locations in the input beam. All of the output beams for common-intensity locations of the input beam are thus parallel, and the difference in direction among different output beams will increase as the difference between the corresponding input beam intensities increase. In the second step, each set of parallel output beams is focused to respective common points in the lens focal plane, thereby producing a positional distribution corresponding to the optical intensities of the input beam.

The mechanism used to control the deflections of the output beams may be conceptualized as a pattern of sepa¬ rate voltage gradient "modules" . A series of such modules which may be used for positional mapping is shown in FIG. 3, which plots voltage against distance. Each module has an associated spatial voltage gradient transverse to the beam and an associated phase gradient, which deflects its

SUBSTITUTE SHEET

corresponding portion of the beam by an amount which varies with the gradient magnitude. The physical appara¬ tus used to establish the gradients is described in a later section. Three beam pixels 18, 20 and 22 are illustrated as being directed towards modules 24, 26 and 28, respective¬ ly. Module 24 has an intermediate voltage gradient, indicated by line 30 which slopes down from an intermedi¬ ate voltage level along a unit distance transverse to the input beam pixel 18. The voltage gradient produces an electric field which has a component El that is also transverse to pixel 18, and has an intermediate magnitude. When the voltage gradient is established through an electro-optic material, a phase gradient is established which will deflect the input light pixel 18 by an angle A which varies in positive proportion to the gradient magni¬ tude. When an array of small modules are employed, they will operate' in a manner analogous to a blazed grating, deflecting the incoming light into a Fourier array of discrete deflection angles, with the intensity of the deflected light peaking at angle A and generally progres¬ sively diminishing therefrom.

Module 26 illustrates a lower voltage gradient, with a proportionately lower electric field E2 transverse to the input pixel 20. This will produce a smaller deflec¬ tion angle B. Module 28, on the other hand, has a larger voltage gradient transverse to its input pixel 22. This produces a larger electric field E3 transverse to the input light, and a correspondingly larger deflection angle C.

A physical set-up for establishing the spatial volt¬ age gradients is illustrated in FIG. 4a. An array of interdigitated electrodes 30 are positioned across the gradient area from a counter electrode 32, which in this embodiment is a planar electrode. The interdigitated

electrodes 30 are flat and preferably elongated into the page. Alternate electrodes 34 are connected to one side of a voltage source 36, while the intervening electrodes 38 are connected in common with the counter electrode 32 to the opposite side of the voltage source, which may be at ground potential. Each of the interdigitated elec¬ trodes is laterally spaced from its adjacent electrodes. A resistor sheet 40 may be provided in contact with each of the interdigitated electrodes to linearize the voltage gradients therebetween.

The voltage gradients produced by the electrode array is illustrated in FIG. 4b, which is vertically aligned with FIG. 4a. The solid lines 42 sloping down and to the right indicate voltage gradients having an associated electric field component E directed to the right; the solid lines 44 sloping down and to the left indicate voltage gradients having an associated electric field component E directed towards the left. ^' The solid voltage gradient lines result from the use of resistive sheet 40; voltage gradients indicated by dashed curved lines result when no resistive sheet is used.

An alternate embodiment in which interdigitated electrodes 46 and 48 are employed on both sides of the voltage gradient area is illustrated in FIG. 5. In this embodiment alternate electrodes in the upper array 46 are connected to two different voltage potentials, respective¬ ly, and similar connections are made for the alternate electrodes in the lower array 48. The directions of the transverse electric field components between the two electrode arrays are indicated by arrows.

The application of the invention to a transmissive liquid crystal light valve (LCLV) is shown in FIG. 6. An interdigitated electrode array 50 is formed on a trans¬ parent glass plate (not shown) on one side of a layer of nematic liquid crystals 52, while a planar electrode 54 is

formed on another glass plate (not shown) on the other side of the liquid crystal layer. A layer of photoconduc- tive material 56 is disposed between planar electrode 54 and liquid crystal layer 52. The two sets of electrodes form a plurality of small, replicated voltage gradient modules for deflecting incoming light. A voltage differ¬ ential is established between the interdigitated elec¬ trodes 50 and planar electrode 54, such as by applying a voltage V to every other one of the interdigitated elec- trodes 50, and grounding the intervening interdigitated electrodes and the planar counter electrode 54. With a liquid crystal medium between the electrodes, the voltage between the two electrodes would preferably be AC driven at about 10 kHz. Since the photoconductive layer 56 has a very high impedance in the absence of applied light, most of the voltage between the electrodes will normally be established across the photoconductor 56. In this embodi¬ ment both electrodes should be transparent.

An input light beam 57 is shown directed onto the underside of the planar electrode 54 and photoconductive layer 56. As used herein, the term "light" is intended in a broad sense to include any electromagnetic radiation capable of being diffracted by the electro-optic modules described herein, rather than just the visible portion of the spectrum. Although in practice it might be continu¬ ous, beam 57 is illustrated for simplicity as constituting four separate rays. One ray has a relatively low optical intensity designated 1, two rays have an intermediate optical intensity designated 2, and the last ray has a relatively high optical intensity designated 3.

As is well known, a photoconductor will become par¬ tially conductive in the presence of light, with the degree of conductance varying with the intensity of the imposed light. Also, most photoconductors are not subject to significant light spreading, and can be designed with a

high transverse electrical impedance. Accordingly, the conductivity of the photoconductive layer in the path of each light ray will vary in a positive fashion with the intensity of that light ray. Specifically, the conductiv- ity of the photoconductive layer in the vicinity of the low intensity ray will be relatively low, its conductivity in the vicinity of the two intermediate intensity rays will be of an intermediate level, and the photoconductor conductivity in the vicinity of the high intensity ray will be relatively high. A voltage divider is established between each module of the liquid crystal and the underly¬ ing portion of the photoconductive layer, and the portion of the voltage dividers attributable to the liquid crys¬ tals will vary in accordance with the conductivity of the underlying photoconductor material. Thus, the module in line with the highest intensity ray will support the greatest voltage differential through the liquid crystal layer (and therefore the greatest electric field) , while the module in line with the lowest intensity ray will support the lowest ventage differential.

A pattern of voltage gradients transverse to the input light rays will thus be established through the liquid crystal layer in a fashion similar to FIG. 4b, but the magnitude of each voltage gradient will vary in accor- dance with the voltage divider effect between the liquid crystal and underlying photoconductive layer at that location. Thus, as the input light rays are transmitted through the photoconductive layer and their respective modules in the liquid crystal, the highest intensity ray will undergo the greatest deflection as it exits the light valve, while the lowest intensity ray will experience the lowest deflection as it exits its respective portion of the light valve. The two rays of equal intermediate intensity will each be deflected by the same intermediate angle, and will thus exit their respective modules as

parallel rays. This controlled deflection implements the first step of the mapping process described previously, that of mapping the spatial distribution of light inten¬ sities onto the direction of output rays. A lens 58 is positioned over the light valve so as to focus light emerging therefrom onto its back focal plane 60, which is centered on the lens axis 62. Because it was deflected the least upon exiting the light valve, the light ray corresponding to the lowest optical intensity will be focused onto the focal plane at point 64, nearest to the lens axis (for a practical device it will actually be focused as. a series of points corresponding to a Fourier spectrum, as described further below, but the peak will be located at point 64) . The ray corresponding to the highest input intensity experiences the greatest deflection, and accordingly is focused in a similar manner onto point 66 of the focal plane at a much greater dis¬ tance from the lens axis. The two rays of intermediate intensity exit the light valve as parallel rays, and accordingly are focused to a common intermediate point 68 on the focal plane between points 64 and 66. Thus, the optical intensities at separate pixel locations in the original input beam are mapped onto a positional array in which increasing distance from the lens axis corresponds to pixels of increasing optical intensity, regardless of the positions of the pixels in the input beam. The origi¬ nal input light intensity distribution is reconstructed as a phase image in the back image plane of lens 58.

While a liquid crystal layer 52 is used in the pre- ferred embodiment, in principle any electro-optic material could be used. An -electro-optic material is defined as one whose refractive index varies with the magnitude of an applied electric field. Nematic liquid crystal presently in common use is preferred because, although its response is somewhat slower than many other materials, it exhibits

the largest change in refractive index for relativel small changes in electric field. In the future "quantu well" or ferroelectric liquid crystal devices may offer the potential for faster response than is presently avail- able with conventional nematic liquid crystals.

A reflective light valve is illustrated in FIG. 7 which is somewhat analogous to a conventional reflective liquid crystal light valve, such as that shown in U.S. Patent No. 3,824,002 to Terry D. Beard and assigned to Hughes Aircraft Company, the assignee of the present invention. Elements of this device which are common with the transmissive device of FIG. 6 are indicated by the same reference numerals. Specifically, the upper inter¬ digitated electrode array 50, the lower planar electrode 54, the layer of nematic liquid crystal material 52 and photoconductive layer 56 are common to both devices. In addition, a light blocking ^' layer 70 is interposed immedi¬ ately above the photoconductive layer 56, and a reflecting means 72 such as a dielectric or metal matrix mirror is provided between the light blocking layer and the liquid crystal layer 52. An input light beam is applied to the underside of the structure, as in the transmissive embodi¬ ment. While the input light again provides a variable voltage divider effect with respect to the photoconductive and liquid crystal layers, the input light is blocked by layer 70 from reaching the liquid crystal. Instead, a readout light beam 74 is directed from above the structure through the transparent interdigitated electrodes 50 and liquid crystal to the underlying mirror 72, and are reflected back through the liquid crystal to emerge as output rays. The readout rays undergo two deflections, once upon entering and again upon exiting the liquid crystal layer.

The voltage gradients and electric fields transverse to the readout rays are controlled by the' intensity of the

input beam on the photoconductive layer immediately underlying each readout ray. Accordingly, the deflection experienced by each readout ray will vary in accordance with the immediately underlying input beam intensity, as in the transmissive embodiment. However, since each ray of the readout beam undergoes two deflections, the angular spread between the output rays will be greater than in the transmissive embodiment, resulting in a more easily resolved pattern at the focal plane of the lens (not

10 shown) in FIG. 7. In both FIGs. 6 and 7 only the output rays which are deflected to the right are illustrated. -Voltage gradients are electric fields oriented to produce deflections in the opposite direction will also be established in the liquid crystal between each of the - illustrated rays. These voltage gradients will produce a positional mapping to the left of lens axis 62.

The maximum deflection, Θ _max , from one transit of the liquid crystal is given in radians by the expression:

Θ _max = Δn d 20 ^max a where Δn is equal to the difference in refractive index between the extraordinary and ordinary directions (the birefringence) of the liquid crystal, d is the thickness of the liquid crystal cell, a is the electrode gap width.

2 ₅ A typical value for Δn is 0.22. The resolution and deflection angle produced by the device will increase with the thickness of the liquid crystal layer, but on the other hand a thicker liquid crystal will significantly slow down the device's response time. A suitable compromise _Q for many applications will be a liquid crystal thickness in the order of 5-15 microns.

5

SUBSTITUTESHEET

FIG. 8 illustrates a plan view of a portion of the interdigitated electrode array, consisting of elongated electrodes 76, 78, 80 and 82. The electrodes are each long enough to encompass a plurality of pixels. For good quality images it is desirable to have a large number of electrodes. For a typical light valve resolution of 20 microns, the electrode pairs will also have a periodicity of 20 microns. Thus, the width of each electrode as well as the gap between adjacent electrodes will typically be on the order of 5 microns.

In practice, with a large number of relatively narrow electrodes, the present invention will produce deflections primarily through a diffractive process analogous to a blazed grating. Blazed gratings are physically structured with tilted surfaces, that is, gradients, and are there¬ fore not adjustable, whereas the gradients of the present invention are voltage gradients.

The pattern produced at the focal plane by a light valve such as that shown in FIGs. 6 or 7 is illustrated in FIGs. 9 and 10, with FIG. 9 showing a smaller deflection corresponding to a lower input light intensity. The two graphs have normalized horizontal and vertical scales. The horizontal scale corresponds to distance along the focal plane, and also to the input optical intensity and to the Fourier spatial frequency. The vertical scale indicates the number of pixels mapping onto a particular portion of the horizontal scale. All along the horizontal scale a series of minor diffraction orders 86 will be produced, due to the periodicity of the electrodes and voltage gradient modules. These discrete orders corre¬ spond to the Fourier components of the electrodes. At a location corresponding to the intensity of the input pixel, a series of elevated spikes will be mapped to form a distinct envelope 88. The width of the envelope, or in other words the number of spikes included within the

envelope, varies inversely with the width of the modules formed by each electrode pair (pixel width) . As the pixel's optical intensity increases or decreases, the location of the various diffractive orders will remain fixed in Fourier space, but their heights will change so that the envelope 88 moves a corresponding distance to the right or left, respectively. FIG. 10 illustrates the mapping for a pixel intensity twice that of FIG. 9. The number of resolvable deflection angles is given by the product of the maximum deflection given above and the pixel width, and is approximately 18 for the above para¬ meter values.

Once the intensity-to-position mapping of the input beam has been achieved, any desired portion of the mapped image can be fully or partially blocked or modified. One or more intensities of interest can be simply blocked or modified by operating upon its position in the focal plane. Thus, the present invention is adaptable to all of the optical computing and logic processing that can be achieved with the aforementioned Patent No. 4,351,589, the contents of which are incorporated herein by reference. In addition, the present invention is capable of operating significantly faster than the referenced patent, and therefore over a greater temporal frequency range. The response time can be lowered to less than 10 msec, with a high output uniformity and resolution exceeding 25 line pairs per mm.

The application of the invention to a fiber optic switch is illustrated in FIG. 11. The light from indivi- dual input fibers 90 can be coupled into any one of a plurality of output fibers 92. The switching mechanism consists of a layer of electro-optic material 94, prefer¬ ably liquid crystals, with each input fiber directed at a target area 95 on the liquid crystal cell. Pairs of electrodes (or a plurality of electrodes) 96 are

positioned along one face of the liquid crystal cell on opposite sides of each target area 95. Voltage differen¬ tials are applied across each electrode pair, establishing electric fields in the liquid crystal transverse to the direction of light emitted from the input fibers. This produces a deflection of the input light which is a func¬ tion of the electric field for each fiber. The resulting deflected output light beams 98 are focused by a lens 100 onto corresponding output fibers 92. The electric field produced by each electrode pair can be controlled from an external source, and the beams from each input fiber can be switched independently of the other input beams. Many such input fibers could be coupled via the optical deflec¬ tor to a large number of output fibers. A particular input fiber could also be coupled to multiple output fibers simultaneously by spreading the light from the input fiber transverse to the electrodes before it is input to the liquid crystal deflector.

Several embodiments of the invention have thus been shown and described. However, it should be understood that numerous variations and alternate embodiments will occur to those skilled in the art. Accordingly, it is intended that the invention be limited only in terms of the appended claims.