Dipolar Electro-optic Compositions And Method Of Preparation

Abstract

Methods and apparatus incorporating suspensions of assymetric minute dipolar particles for the control of electromagnetic radiation. The physical electrical and optical properties of the dipolar particles and their suspending medium are specified in conjunction with apparatus employing them.

Parent Case Text

This application is a continuation-in-part of an application for patent entitled, Dipolar Electro-Optic Structures and Method, filed June 29, 1964, Ser. No. 378,836, now U.S. Pat. No. 3,512,876, in the name of Alvin M. Marks.

Description

BACKGROUND OF THE INVENTION

It has previously been suggested to employ a suspension of orientable dipolar particles as a light-controlling element, and to orient the particles in such a suspension by the application of an external electric or magnetic force field. Devices of this general type that have so far been proposed, however, have had little use because of a number of important deficiencies. One of such prior art faults was the tendency of the oriented particles to coagulate or clump together, rather than remain uniformly dispersed. Another shortcoming was that the optical properties of the devices, either in the oriented or disoriented condition, were of a low order. Thus, when such a suspension was switched from maximum transmittance to minimum transmittance, or maximum reflectance to minimum reflectance, the obtainable ratios of these transmittances, or reflectances, were too small. Moreover, clear suspension of dipolar particles, free from light scatter, were not available. Furthermore, the response of such a system to an applied electric or magnetic force field tended to be slow. Orientation and disorientation control techniques were lacking. Consequently, prior art devices were not suitable for incorporation into most electro-optical systems. In general the underlying physical laws governing electrodichroic systems were not at all well understood, and the physical parameters of such systems were relatively unknown.

DEFINITIONS

Electrodichroic systems as used herein means, dipolar suspensions which exhibit changes in optical properties upon the application of electric or magnetic fields.

Optical density is defined as the negative logarithm to the base 10 of the light transmittance of an optical element. Thus, if the element is completely transparent, it transmits 100 percent of the incident light, the transmittance is 1.00, and the optical density is -log.sub.10 1.0, or 0.

Similarly, if the element transmits 10 percent of the incident light, the transmittance is 0.10 and the optical density is:

-log.sub.10 10/100 = -log.sub.10 0.10 = log.sub.10 = 1

Similarly, if an element transmits 1 percent of the incident light, the transmittance is 0.01, and the optical density is 2. In the same way an element that transmits 0.1 percent of the incident light corresponds to a transmittance of 0.001, and an optical density of 3, etc.

It is useful to employ the following convention: The plane of a dipole layer is taken as the XY plane, X generally being considered the horizontal and Y the vertical axis. The direction of incident light normal to the plane of the cell is taken as the Z axis. The X, Y and Z axes are all mutually perpendicular. The electric field may be applied along the X, Y or Z axes, and the subscripts x, y and z indicate the electric field direction. The subscript r indicates no voltage with the dipole orientation randomized.

A suspension of rod-shaped particles in a fluid which exhibits change in optical properties upon the application of an electric field is defined as a dipolar material having electrodichroic properties.

The electrodichroic ratio of a dipolar material is defined as the ratio of the optical density in the opaque condition for dipoles in random orientation, to the optical density in the transparent condition for dipoles partially or completely oriented parallel to the electric field direction.

The parallel Electrodichroic ratio refers to the electrodichroic ratio with the electric field applied parallel to the light path; that is: q.sub.rz = D.sub.r /D.sub.z.

The normal electrodichroic ratio refers to the electrodichroic ratio with the electric field applied normal to the light path; that is: q.sub.rx = D.sub.r /D.sub.x.

The electrodichroic response is defined as the change in the electrodichroic ratio with respect to the change in applied electric field intensity E; that is: .sigma. = (dq/.DELTA.E).

The electrodichroic sensitivity is defined as the change of electrodichroic ratio with respect to the change in electric field intensity, per unit of mass in the unit area. Thus the electrodichroic sensitivity in the electrodichroic response per unit mass in a unit area of a dipole suspension layer; that is: S = 1/M (.DELTA.q/.DELTA.E) = .alpha./M. Subscripts define parallel and normal cases.

Relaxation means the disorientation in the absence of aligning field of previously aligned dipolar particles.

For most effective performance, an electro-optical shutter should be characterized by an electrodichroic ratio of preferably 10 or more.

An electrodichroic ratio of 15, therefore, signifies that the optical density of the shutter in the opaque condition is 15 times that of optical density of the same shutter in the transparent condition.

As a specific example, a shutter capable of transmitting 60 percent of the incident light in the transparent state, and only 0.1 percent of the incident light when the opaque state, would have the following optical densities:

Transparent: D.sub.t = log.sub.10 (100/60) = 0.22

Opaque: D.sub.o = log.sub.10 (100/0.1) = 3

The electrodichroic ratio of such a shutter, then would be:

D.sub.o /D.sub.t = 3/0.22 = 13.6

An object of the present invention, is to provide improved dipole particle suspensions, and methods and apparatus for electrically controlling light and other electromagnetic radiation.

Another object is to provide light-controlling compositions whose optical properties can be varied electrically without the use of mechanical moving parts.

Still another object is to provide light-controlling compositions and devices as aforesaid, characterized by improved electro-optical characteristics; greater electrodichroic ratios, greater electrodichroic sensitivity and smaller alignment.

A further object is to provide an electro-optical layer having an electrodichroic ratio in excess of 10.

Another object still is to provide a light controlling panel having electrical means for decreasing relaxation time.

Another object is to provide a light controlling panel having electrical means for causing dipole orientation or relaxation, selectively confined to a particular area.

Yet another object is to provide a thin electro-optical light control panel or shutter of large area suitable for use as an electrically controlled variable density window, visor, optical element, or ophthalmic lens.

A feature of the present invention is the use as a light-controlling medium of a suspension of dipole particles having optimum optical and electrical properties resulting from novel relationships established amongst the physical dimensions, resistivity and concentration of the dipole particles and the viscosity of the suspending fluid.

Still another feature is the utilization of the "antenna" effect influencing the optical properties of dipolar particles, as hereinafter more fully described.

Another feature of the present invention is the control of alignment rise time to maximum transmittance of a suspension of dipolar particles by correlation of dipole particle dimensions, concentration of the dipolar particles, the viscosity of the suspending fluid, and the application of pulsed electric fields of high intensity.

Another feature is the use of an electric or magnetic field to orient or disorient the dipole particles in such a suspension.

Another feature is the use of current-carrying shielding means for confining an orienting electric field.

Still another feature is the use of transparent conductive films which serve as shielding means for establishing different orientations of the dipolar particles to change the transmittance or reflectance of the device.

Another feature of the invention is the use of the "curtain effect", and transparent conductive shielding electrodes to reorient a dipolar suspension.

A further feature of this invention is a novel electrooptical iris or curtain diaphragm without mechanical moving parts.

Other objects, advantages, and novel features of the present invention will become apparent from the following more complete description and claims.

In one form, the present invention contemplates a light-controlling device employing a suspension of particles hereinafter referred to as dipoles or dipole particles, said particles having at least one dimension large relative to at least one other dimension. The suspended particles are orientable in response to an applied electric, magnetic or mechanical shear field. The application of a non-constant force field to said suspension enables maximum alignment to be attained without coagulation of the particles.

In another form, this invention contemplates an electro-optical light control device having a cell containing a suspension of dipole particles in a transparent medium, capable of interacting with electromagnetic radiation, said cell having spaced transparent walls and being provided with spaced, transparent electrically conductive films generally parallel with the transparent walls. This embodiment also has a pair of electrodes at opposite edges of the cell, near the edges of the transparent walls, and insulated from the conductive films. Such a cell is made transparent by orienting the dipole particles in the suspension with their long dimensions normal to the transparent walls. Orientation is achieved by imposing an electrical potential between the transparent conductive films. The cell is rendered opaque by starting to orient the long dimensions of the dipole particles parallel with the transparent walls, by imposing an electrical potential in a direction parallel to the walls between the electrodes at the edges of the cell, but stopping the orienting influence while the particles are in an intermediate, random phase.

In this phase of operation, the field between the two edge electrodes is confined within the cell by simultaneously passing an electric current through each of the conductive films. Passage of such a current effectively prevents the lines of force from short-circuiting through the conductive films and thus by-passing the interior of the cell where the dipole particles are located.

In still another form, this invention contemplates an electro-opticallight control device comprising in combination a first cell and a second cell, each of which is enclosed in part by generally parallel, spaced, transparent walls, both of said cells being located in the space between a pair of generally parallel spaced conductive loops. The first cell has a first pair of electrodes located at opposite edges of the cell, and a second pair of electrodes, angularly spaced from the first pair of electrodes by approximately 90.degree. measured in a plane parallel with the transparent walls. The shutter is rendered transparent by imposing electrical fields between the conductive layers, thus orienting the particles in both cells normal to the transparent walls. To render the cell opaque, each of the two pairs of electrodes is connected to a source of electrical potential difference thereby creating a first electrical field between the first pair of electrodes in the first cell, and a second electrical field between the second pair of electrodes in the second cell, the first and second electrical fields being at right angles. The effect of the two fields is to orient the dipole particles in the first cell in a first direction parallel to the transparent walls, and the dipoles in the second wall in a second direction parallel to the transparent walls, these first and second directions being perpendicular. Since the dipole particles, when aligned normal to the light path, act like polarizing elements, the cross-orientation effectively blocks all but a very small portion of the light.

BRIEF DESCRIPTION OF THE FIGURES

The invention consists in the construction, combination and arrangement of parts and of operating steps as hereinafter more fully described and claimed, and as illustrated in the drawings, in which like parts appearing in more than one view are given the same reference numeral throughout, and in which:

FIG. 1 is a fragmentary view on an enlarged scale of an electrically responsive light-controlling structure made in accordance with the present invention showing disoriented dipole particles in a reflecting or light absorbing state,

FIG. 2 is a view similar to FIG. 1, showing the dipole particles in aligned orientation, with the long dimension of the particle normal to the plane of the structure, in a transmittive state,

FIG. 3 is a fragmentary view similar to FIGS. 1 and 2, showing a protective coating between the conductive coating and the dipole suspension,

FIG. 4 is a cross-sectional view showing a structure similar to that shown in FIGS. 1 and 2, provided with an electromagnet to effect orientation,

FIG. 5 is a perspective view of another embodiment of the invention, showing a comparatively bulky high voltage switching device utilizing a single plane dipolar suspension and unshielded electrostatic fields for controlling the orientation of a dipolar particle suspension,

FIG. 6 is a fragmentary diagrammatic detail of a portion of the embodiment of FIG. 5, on a larger scale,

FIG. 7 is a view similar to FIG. 6, showing another stage in the operation of the device in FIG. 5,

FIG. 8 is a schematic diagram of an electrical circuit used to apply potential to the electrodes of the device in FIG. 5,

FIG. 9 is a fragmentary view, on a greatly enlarged scale, of a single dipole in an elementary volume of suspending fluid,

FIG. 10 is a fragmentary perspective view, similar to FIG. 1, of another embodiment of the invention, namely a reflective-absorptive panel,

FIG. 11 is a partially cut away perspective view, partially schematic, of another embodiment of the invention, in the nature of an electro-optical iris diaphragm,

FIG. 12 is a perspective view of the same electrooptic iris diaphragm as in FIG. 11 but with the device in another stage of operation,

FIG. 13 is a view similar to FIG. 12, showing the device of FIG. 11 at still another stage of operation,

FIG. 14 shows a dipole curtain shutter in cross section, used in producing a "curtain effect" according to another embodiment of the invention,

FIG. 15 is another view similar to FIG. 14 showing an intermediate stage of the "curtain effect",

FIG. 16 is a curtain shutter as shown in FIG. 14 but at a stage in which the dipole particles are all oriented in the X-direction,

FIG. 17 shows a front view of the device in FIG. 14 at an intermediate stage of the curtain effect,

FIG. 18 is a front view of the device of FIG. 14 in another mode of operation,

FIG. 19 graphically shows the percent transmittance versus time during a change in dipole particle orientation from the Z-direction as shown in FIG. 2 through random orientation to the orientation in the X-direction, and also shows the corresponding E.sub.z and E.sub.x alternating electrical pulses applied first along the Z axis, and then along the X axis to achieve this result,

FIG. 20 shows a graph of percent transmittance versus time during orientation and disorientation when pulses are on and off respectively;

FIG. 21 shows a transmittance-time curve corresponding on the application of a series of pulses timed to give an increasing degree of orientation with intervening periods of partial relaxation between successive pulses,

FIG. 22 is a graphic representation of the voltage-time characteristic of a reversing D.C. pulsed current used in certain embodiments of the invention,

FIG. 23 shows the relative power absorbed or reradiated versus wavelength for thick and thin half-wave dipoles,

FIG. 24 shows a polar graph of response versus angle to ray path of the dipole antenna,

FIG. 25 shows a graph of response versus angle to the polarization direction for a dipole antenna,

FIG. 26 shows a conventional half-wave dipole with a central electrical load,

FIG. 27 shows a half-wave dipole with a distributed electrical load,

FIG. 28 illustrates diagrammatically the effective cross section of a dipole antenna,

FIG. 29 shows a graph of the random-normal electrodichroic ratio versus the electric field intensity, steady state 60 cycle A.C. for an herapathite suspension,

FIG. 30 shows a random-parallel electrodichroic ratio versus electric field intensity, steady state 60 Hz for herapathite suspension of two different concentrations.

FIG. 31 shows the transmittance versus time for an herapathite dipole suspension having various electric field intensities applied parallel to the light path as a D.C. pulse,

FIG. 32 shows the random-parallel electrodichroic ratio versus the electric field intensity from the data of FIG. 31,

FIG. 33 shows the peak transmittance versus electric field intensity from the data of FIG. 31, for various initial transmittances for the dipole layer in the random state,

FIG. 34 shows a plot of the inverse rise time versus the electric field intensity from the data of FIG. 31,

FIG. 35 shows the transmittance versus time for the various given electric field intensities plotted from an empirical equation which closely fits the experimental data shown in FIGS. 31-34 inclusive.

FIG. 36 shows the random-parallel electrodichroic ratio versus electrical field intensity steady state 60 cycle A.C. for an aluminum flake suspension,

FIG. 37 is the same as FIG. 36 except that the parallel electrodicroic ratio is plotted on a log scale versus electric field intensity on a linear scale,

FIG. 38 is an exploded perspective view of a two layer dipole suspension, current shielded transparent electrode type of shutter in the transparent condition,

FIG. 39 is an exploded perspective view showing the device of FIG. 38 in the opaque condition, using current shielded transparent electrodes,

FIG. 40 is an assembled cross section of the device of FIGS. 38 and 39 in the opaque condition, taken along line 40--40 of FIG. 41,

FIG. 41 is a front view of the device, corresponding to the cross-sectional view of FIG. 40,

FIG. 42 is an exploded perspective view of a single dipole suspension layer shutter of the present invention, in a condition to polarize transmitted light, utilizing current-shielded transparent electrodes,

FIG. 43 is another exploded perspective view showing the device of FIG. 42 in the transparent condition,

FIG. 44 is a perspective view of a shutter according to another embodiment of the invention, in the opaque condition without shielding, utilizing electrostatic fields in air,

FIG. 45 shows a schematic diagram of the pulse circuit for actuating the dipole cell,

FIG. 46 shows the wiring diagram for an intermediate pulse amplifier,

FIG. 47 shows a high voltage pulse amplifier,

FIG. 48 shows various cross sectional areas for half-wave dipole,

FIG. 49 shows the cross section versus the ratio of radiation resistance to absorption resistance,

FIG. 50 shows the maximum percent transmittance versus electrodichroic ratio for constant minimum transmittance of 0.01 percent, 0.1 percent and 1.0 percent,

FIG. 51 shows the optical density versus (1) Alignment Time (2 KV.sub.rms 3 CM) (2) Relaxation Time, for a Herapathite Suspension,

FIG. 52 shows the optical density versus electric field intensity for herapathite suspension for Z orientation,

FIG. 53 shows the parallel electrodichroic ratio versus electric field intensity for a herapathite suspension,

FIG. 54 shows the parallel electrodichroic ratio versus frequency at various electric field intensities for a herapathite suspension,

FIG. 55 shows the parallel electrodichroic ratio versus electric field intensity in KV.sub.rms /CM for various frequencies,

FIG. 56 shows the transmittance versus wavelength for a herapathite suspension in aligned (open) and random (closed) orientation,

FIG. 57 shows the relative electrodichroic ratio versus wavelength,

FIG. 58 shows the relative transmittance versus angle for herapathite dipole layer for:

1. Nonpolarized

2. Polarizer parallel

3. Polarizer crossed

Dipole layer with dipoles oriented parallel to Z axis

FIG. 59 shows the transmittance for polarized light versus wavelength, for various electric field intensities;

A. polarizer Parallel

B. polarizer Crossed

For chromium dipoles of average particle size 7700A .times. 700A oriented in the plane of the suspension, parallel to the plane of polarization or at right angles thereto;

FIG. 60 shows the percent polarization versus wavelength for a chromium metal dipole suspension (computed from data of FIG. 59.);

FIG. 61 shows the transmittance versus wavelength for electric field parallel to light path, for various electric field intensities for chromium metal dipoles oriented in Z direction,

FIG. 62 shows the parallel electrodichroic ratio versus electric field intensity at 2300 nm for a chromium dipole suspension,

FIG. 63 shows the parallel electrodichroic ratio versus wavelength for chromium metal dipole suspension oriented in Z direction, at various applied electric field intensities in Kv/cm,

FIG. 64 shows the cross section of a typical cell for Z orientation,

FIG. 65 shows the front view of the cell shown in FIG. 64,

FIG. 66 shows the cross section of a typical cell for X orientation,

FIG. 67 shows the front view of the cell shown in FIG. 66,

FIG. 68 shows a graph of relative transmission versus time across a cell containing a chromium dipole suspension for an applied voltage pulse applied in the Z direction, on a time scale of 0 to 5 milliseconds,

FIG. 69 shows the same graph as FIG. 68 on a time scale of 0-100 milliseconds.

Light-controlling devices according to this invention are useful in varying embodiments, as camera shutters, variable iris diaphragms, variable density windows for control of room lighting, visors for automobiles, opthalmic and optical elements, 2-dimensional and 3-dimensional displays, radiation absorption and reflection control elements for buildings, spacecraft, decorative elements, signalling devices, and in a variety of other ways which their novel characteristics will readily suggest to those skilled in the art.

The dipole particles useful in the present invention are characterized in that they have at least one dimension large relative to at least one other dimension -- that is to say, they are in the form of flakes, needles or the like. The dipole particles should have at least one dimension equal to one-half of the wavelength of the radiation to be controlled, (normally, visible light, but in some cases infrared, ultraviolet, microwave, or other portions of the electromagnetic spectrum) and at least one other dimension substantially smaller than one-half of said wavelength. The magnitude of the third dimension -- that is, whether the particle is a needle or a flake -- depends on the requirements of the specific embodiment of the invention, as more fully discussed below.

For purposes of brevity, the term "light" is used throughout the present specification and claims in a general sense and is intended to encompass not only visible light but also infrared and ultraviolet "light", as well as microwave radiation in the neighboring portions of the electromagnetic spectrum.

In addition to the dimensional and resistance requirements for strong interaction with electromagnetic radiation herein disclosed, the electrical or magnetic properties of the dipolar particles, i.e., in the electrical case the conductivity or the dielectric constant, must be such as to facilitate orientation in an electric or magnetic field.

The suspending medium is a fluid, non-reactive with the dipole particles, or is a substance capable of being converted to a fluid, at a temperature sufficiently low to avoid any adverse effect on the dipole particles.

It is not in all cases necessary that the suspending medium be in a liquid state. Providing the applied torque is sufficiently strong to orient the dipole particles against a certain amount of plastic resistance of the suspending medium, it is sufficient if the suspending medium is in a deformable plastic or thixotropic state. The term "fluid" as used herein should, therefore, be understood to encompass such a plastic condition. For most applications of the present invention, the suspending medium is present as a liquid during alignment or disorientation of the dipole particles.

The suspending medium may be thermoplastic which is fluid about a given temperature, for example 40.degree. to 60.degree.C. The dipole layer is oriented by an electric field, while the suspending medium is fluid. Then the orientation may be fixed by causing the suspending medium to solidify.

The dipole particles must also be of such a nature that they are capable of being oriented by an applied electric, magnetic or in certain cases a mechanical shear force field.

Some particles have an inherent dipole moment by reason of their internal structure, in which the effective center of positive charge in the molecule or crystal is spaced from the center of negative charge. Such an inherent dipolar character, if present, is effective to some degree in augmenting the tendency of the particles to orient themselves in an applied force field. Inherent dipolarity is, however, neither essential nor a major factor in determining the effectiveness of the dipole particles.

The major factor in effecting orientation of the dipolar particles in an applied field is induced dipolarity, which may arise because of a difference between the dielectric constant of the insulating dipole particles and that of the surrounding medium. Alternatively, the dipolarity may arise because the dipolar particle is a semi-conductor, or a conductor permitting opposing charges to accumulate at the long opposite ends of the dipole particle. Ordinarily, an insulating dipole particle has a larger dielectric constant than the medium, and tends to concentrate the lines of force within itself. In so doing, it acquires an induced dipolarity, the end of the particle nearer the positive electrode acquiring an induced negative charge, and vice versa. Once the induced dipolarity has arisen, the particle tends to orient itself by swinging so that the end having the induced positive charge points directly toward the negative electrode, and vice versa.

Similar considerations apply when the orienting field is a magnetic field, except that the induced dipolarity is magnetic, rather than electrostatic, in character.

In the unusual case where the particles have a smaller dielectric constant than the surrounding medium, the same general principles apply, but in such a case, the lines of force tend to concentrate in the medium rather than in the particles, and the medium then tends to push the particles into alignment in the process of shortening the lines of force.

Suitable dipole particles according to the present invention, therefore, include such materials as herapathite crystals, which are particularly advantageous because of their optical properties, as well as other materials which, by virtue of their shape, dielectric constant or conductivity characteristics tend to concentrate the lines of force of a magnetic or electrostatic field within themselves. Needle-shaped particles of a ferro-magnetic substance such as iron will orient themselves when subjected to the influence of a magnetic force field. Similarly, needles of any electrically conductive substance tend to align themselves parallel to the lines of force of an applied electrostatic field.

When reference appears herein to "dipolar particles", or "dipoles", it is therefore intended to include conductive asymmetric particles and insulating asymmetric particles having a substantial difference in dielectric constant from the medium in which they are immersed. All such particles are capable of acquiring induced dipolarity, and reference to the particles as "dipoles" is not intended to limit them to particles characterized by inherent or permanent dipolarity.

To illustrate the above considerations with reference to some specific examples, dipole particles may be insulating providing the difference in dielectric constant or index of refraction between the particle and the liquid in which it is suspended is substantial. An example of this is a lead carbonate which forms minute hexagonal flakes having an index of refraction of approximately 2.25 and which may be immersed in a fluid having a relative index of refraction of approximately 1.5. The electrostatic lines of force tend to concentrate in the vicinity of the material having the higher index of refraction or higher dielectric constant and thus produce a torque causing alignment of the particles.

Herapathite forms flat needles having a length to thickness of approximately 25 to 1 and having a much higher index of refraction than the suspending fluid. Moreover, the particles are in themselves minute polarizing elements tending to polarize light passing therethrough by virtue of their molecular structure.

Graphite flakes are minute hexagonal crystals having a very high conductivity in the plane of the hexagon and a very low conductivity across the plane. They are thus similar to metallic flakes insofar as their conductivity is concerned since electric charges are free to flow across the plane of the hexagon.

Because the conductivity is anisotropic, it is very low normal to the plane of the hexagon, and hence relatively thick particles of graphite may be effectively oriented.

Still another and preferred class of materials are those comprising metals in which the electric charge is free to move. These metals in fact show the quickest alignments in the smallest fields. Moreover, they are suited for the practice of the "antenna effect" which is more fully described hereinafter.

THE DIPOLE PARTICLE

In FIG. 9 there is shown a single dipole particle 50 of length L and thickness L/a, its cross section being shown as square for simplicity. The particle is shown aligned along the Z axis but the dipole particle 50 can be tilted through an angle .phi. as shown. The dipole particle is shown in an elementary cubic volume of fluid 51. This Figure is useful in connection with the mathematical-physics discussion given in a subsequent section.

For example, the dipole particle length is L = .lambda./2n where n is the index of refraction of the fluid. Usually n is approximately 1.5 so that the dipole length is in almost all cases then .lambda./3. The thickness of the dipole L/a depends on whether the dipole is to be reflecting or absorbing and upon the resistivity of the metal from which the dipole is formed.

The length to width ratio "a" also controls the resonant response of the dipole to radiation, in effect determining the wavelength range to which the dipoles are tuned to absorb or reflect. All these factors will be more fully described hereinafter.

The dipoles may be oriented by electrical or magnetic fields as described herein.

As an example of the alignment of non-metallic needle-like crystals, we may take the FIG. 35 which shows the transmittance versus time due to the alignment of a suspension of dipolar herapathite needles, for various electric field intensities applied as a high voltage D.C. pulse as shown in FIG. 45.

The curves shown in FIG. 29 and 30 are based on data obtained for suspensions of herapathite dipoles using 60 cycle A.C. FIG. 29 shows that the normal electrodichroic ratio increases linearly with the field strength at first, increasing more slowly as the field strength is increased. An empirical exponential equation shows with fair agreement. FIG. 30 shows that in the range 1<q.sub.rz <7 the parallel electrodichroic ratio varies directly with the electric field intensity, and as the square of the concentration of herapathite dipoles in the suspension.

The graph of FIG. 36 demonstrates that for thin aluminum flake dipoles the parallel electrodichroic ratio increases more rapidly as the electric field strength increases and follows an empiric equation which is an increasing exponential function. This is further demonstrated in FIG. 37 in which the data of FIG. 35 is plotted on a semi-log linear scale obtaining a straight line.

In FIG. 36 it can be seen that ultrathin flakes produced by the floatation method in which only particles from the upper layer of the suspending fluid are used, show a marked increase in the electrodichroic ratio at much lower electric field intensity, without coagulation.

In the FIG. 36 above referred to, the end points of the curve or the last experimental observation represents the voltage at which coagulation or agglomeration occurred. The test data shown in FIGS. 29 and 30, were made with a steady applied A.C. voltage. When the voltage exceeded the values indicated the agglomeration occurred.

The force field referred to herein is preferably electrostatic for most embodiments of the present invention, but it may also be magnetic, and the latter is preferred in certain cases. The field is also described as "non-constant", by which is meant that it is non-constant with respect to time. It may, therefore, be continuous alternating voltage, or a pulsed voltage, the pulses being either direct or alternating. A steady direct current, however, is intended to be excluded by the term "non-constant". The non-constant field is required for dipolar suspending fluids containing ions, as with herapathite fluid. In certain applications a constant D.C. voltage will be useful to provide a momentary light pulse. In other cases, where the dipolar suspending fluid is substantially free from ions and is a good insulator a steady D.C. field may be employed.

CONDUCTIVE FILMS

In the practice of the present inventions, suitable transparent conductive coatings are required which are known to the art. One such material is a stannic oxide film on glass or plastic such as is sold by the Liberty Mirror Company under the designation EL-SX; by Pittsburgh Plate Glass Company under the designation NESA. These transparent conductive films have a transmittance of between 70 and 90 percent.

ELECTRO-OPTICAL LIGHT CONTROLLING PANEL

Referring now to the drawings and more particularly to FIGS. 1 and 2, 52 indicates a transparent sheet of glass, plastic or the like. A second sheet of the transparent material 53, also made of glass, plastic or other fluid-impervious material is spaced from the first sheet 52. A fluid-tight gasket 54 is disposed between the sheets 52 and 53, adjacent the edges thereof, to form a small, thin sheet-like area 55 between said sheets. The surfaces of sheets 52 and 53 which define the sheet-like area 55 are covered with an electrically-conductive transparent coating or electrode 56 hereinafter more fully described. The insulating gasket 54 may extend beyond the transparent sheets 52, 53 so as to form a longer electrical air path, and thus prevent arcing between the electrodes 57 at the edge of the sheets. The conductive coatings 56 are connected to suitable metallic strips or bus bars 57 which are disposed along the edges of sheets 52 and 53. Electrical leads 58 are in turn connected to the bus bars 57 and lead to a suitable source of electrical potential (not shown).

The thin sheet-like space 55 between sheets 52 and 53 is filled with a fluid 51 in which there is carried a suspension of dipole particles 50.

When the dipole particles 50 are free to move about in the suspending fluid within the sheet-like space 55 they are subject to Brownian movement and become randomized as shown in FIG. 1. The dipole particles within the sheet-like area 55 may be highly-reflective, or strongly absorptive, flat flake-like or needle like particles. In FIG. 1, light ray 59, in the direction indicated by arrow is shown reflected by the particles emerging from the structure as the reflected beam 60.

When an electric field is imposed across the conductive coatings 56 by the application of an electrical potential difference to leads 58, the dipole particles 50 become aligned with their long dimension parallel to the electric field and normal to the surfaces of sheets 52 and 53, as shown in FIG. 2. Since the thickness of dipole particles 50 is small compared to their length, the light 59 is able to pass between them and reach the second sheet 53. The second sheet 53 also being transparent, the light then passes unimpeded out of the cell as transmitted beam 61. For purposes other than one presently under consideration namely, the electro-optical shutter - it is also within the purview of the invention to make sheet 53 of a non-transparent reflective or absorptive material, so that when the dipole particles 50 are oriented as described, the light passes through the suspension and is reflected, absorbed, or partially reflected and partially absorbed upon striking second sheet 53.

When the electric field is decreased or zero, the dipole particles 50 again become randomized by Brownian movement, with the result that many of them assume positions in which their long dimension is at an angle to the incident light ray 59. Because the dipole particles have then random angular positions the incident light is reflected back in a more or less diffused pattern.

It will be apparent that the optical characteristics of the assembly may be varied from highly reflective to highly absorptive and also may be employed to change from highly reflective or absorptive to light transmitting. Whether the particles, in random array, reflect or absorb light depends on their optical and electrical properties, particularly their dimensions and electrical resistivity, as more fully explained below.

In FIG. 3, the conductive coatings 56 are covered by a transparent protective layer 62 which is disposed upon the coatings 56 on the faces thereof nearest the dipole suspension. The protective layer 62 is necessary in certain cases where the dipole suspension may be chemically reactive with the conductive coating. Protective layer 62 may be, for example, a layer of transparent silicon monoxide, tantalum oxide, or magnesium fluoride.

While the flat sheet-like area 55 has been shown with substantial thickness in the drawings, it is to be understood that the thickness is exaggerated for the purpose of clarity, and in actual practice sheets 52 and 53 may be spaced apart, for example, a distance of from 0.01 to 1.0 millimeter. As a result of a small spacing between the sheets, a substantially complete alignment of the dipoles within the sheet-like area 55, may be obtained using voltages as small as 10 to 500 volts.

For a rapid response, the electric field intensity employed should be as large as practicable, short of the point at which the suspending medium, the dipoles, or other components are broken down. In most embodiments of the invention, voltages of the order 1 to 100 kv/cm are preferred. The voltage required to attain these electric field intensities depends on the thickness of the dipole suspension layer required to obtain the required transmittance range, as set forth hereinafter. The use of larger electric field intensities gives a proportionately shorter response time in switching from a random to an oriented state, vice versa, or from one oriented state to another, an advantage in certain embodiments. Also, it has been discovered, in accordance with the present invention, that troublesome coagulation and clumping of the dipole particles, experienced in prior art devices, can be overcome by using one or more pulses of high intensity and short duration, or a non-constant field, in the form of one or more pulses having a suitable peak electric field intensity, time duration and repetition rate.

The effect which tends to cause coagulation is believed to be explained as follows: When the electric field is applied and the dipoles become oriented, each particle assumes an induced polarity (which reinforces its own inherent polarity, if any). When any two dipoles are aligned in approximately end-to-end relationship, the two ends which are close together are of opposite polarity. They, therefore, attract each other, resulting in longitudinal migration and coagulation. This effect may be avoided by pulsing the field so that it lasts only long enough to effect the desired orientation, and is discontinued before any migration can take place.

The speed of orientation, and consequently the required dureation of the pulse, depends on the electric field intensity of the pulse, the dimensions and electrical characteristics of the dipole particles, and the viscosity of the suspending medium.

A very small mass per unit area of dipoles is effective to control the light, because of their large effective cross section for the capture of incident radiant energy (as discussed below under the heading "Antenna Effect"). Avoidance of the coagulation or clumping effect by the use of a non-constant force field enables greater concentrations of dipole particles to be used in the suspension.

An advantage in using greater concentrations is that the layer thickness and operating voltage is decreased and the dipole suspensions are more responsive to the orienting field. The more rapid response occurs because the particles are closer together at greater concentrations, and the induced dipoles on each pulse exert an increased mutual attraction, much as two compass needles would attract each other when they are brought closer together. Uncontrolled, this increased attraction may result in coagulation, but using a pulsed D.C. or a pulsed alternating field, leads to the desirable result that each dipole particle helps to align its neighbors, producing a rapid response to the orienting field. Concentrations of dipoles which give rise to high viscosities, however, should be avoided where rapid response is desired, because they slow down the response by viscous drag.

After the initial voltage pulse is applied, and the field is off, the particles start to disorient at a rate dependent on temperature (which is normally almost constant), viscosity, and particle size. Pulses of lower voltage, and of such repetition rate as to make the time between pulses considerably less than the disalignment time constant, will keep the particles in substantial alignment without coagulation.

The use of a pulsed field, rather than a steady D.C field, has still another advantage. Despite all precautions, stray ions may be present in the system, either originally as impurities, or produced by breakdown of the suspending medium and/or the dipole particles. When a steady D.C. field is employed, such ions tend to migrate toward the electrodes (i.e. transparent conductors 56). The positive ions migrate toward and collect at the transparent wall in the vicinity of the negative electrode, and the negative ions collect at the transparent wall in the vicinity of the positive electrode, thus shielding the applied electric field and partially offsetting or neutralizing the electric field applied across the dipole suspension layer. When a pulsed field is employed, this migration is avoided, and such shielding does not take place. A simple, intermittent unidirectional D.C. pulse of high intensity and short duration is effective to momentarily orient the particles without causing coagulation, and is also largely effective to avoid migration of ions. Migration of ions is better avoided, however, by using a reversing D.C. pulse, in which alternate pulses are opposite in direction, as illustrated in FIG. 22. Still more effective, and the preferred type of pulse, is a pulsed A.C., in which each pulse is of sufficient duration to include several cycles of the A.C. alternation, as illustrated for example in FIG. 20, which are of sufficiently high frequency to prevent substantial ion separation.

When it is desired to return the particles to their original random state, it is usually sufficient merely to discontinue application of the orienting field. The particles then quickly return to the random condition by the action of Brownian movement. Where more rapid randomization is required, the dipole particles 50 may be randomized by applying a second electrostatic field in a different direction, or a rotating field. The disorientation in certain cases may be alternatively accomplished by applying a viscous drag in the plane of the sheet by a relative linear or rotary motion of the two sheets 52, 53, or a mechanical vibrator may be employed to agitate the cell.

In one form of the invention illustrated in FIGS. 1 and 2, the electro-optical shutter is "opened", that is to say the dipoles are oriented normal to the transparent sheets 52 and 53, by an electrostatic field applied between electrically-conductive transparent coatings 56, which then serve as electrodes. The orientation may if desired be produced by a magnetic field instead of an electric field, provided the dipoles are ferromagnetic, or diamagnetic or paramagnetic material relative to the suspending fluid. Magnetic orientation can be achieved, for example, as shown in FIG. 4, by positioning the cell containing the suspension between the poles 63, 64 of an electromagnet 65. When the magnet is energized by closing switch 66, the ferromagnetic dipoles 68 are oriented. The use of a magnetic field for orienting the dipoles, however, is not preferred in most cases because it requires more cumbersome equipment and greater power input. Magnetic orientation is useful, nevertheless, in connection with some of the embodiments of the invention hereinafter described.

In FIG. 5 there is shown a cell containing a dipole suspension layer 67 between discs of transparent sheets such as glass 52 and 53 separated at the rim by a gasket ring 54. Electrodes 69 and 70 are located on the Y axis and electrodes 71 and 72 are located on the X axis, at the rim as shown in FIG. 6. An electrostatic field extending in the Y direction is obtained by applying a voltage between electrode 69 and 70, or an electrostatic field in the direction of the X axis is obtained by applying a voltage between electrodes 71 and 72. An electric field along the Z axis may be obtained by applying a voltage between rings 75 and 76. The electrode rings 75 and 76 must be separated a sufficient distance in air so that the fields may be effectively applied along the X or Y axis without being diverted toward the rings 75 and 76.

As a result of suitably spacing the electrodes, electric fields may be applied alternatively along the X, Y or the Z axes. However, the construction shown in FIG. 5 is relatively bulky and the large spacing between electrodes 75 and 76 necessitates the application of very large voltages to obtain substantial alignment of the dipole suspension layer in the Z direction..

To bring the ring electrodes 75 and 76 into close prosimity to the dipole suspension and obtain a uniform electric field between these electrodes requires that the metal rings 75, 76 be replaced by transparent conductors as shown in FIG. 1. However, when this is done, a voltage, for example, applied to the electrodes 71, 72 on the X axis, tends to terminate on the transparent conductors and does not pass across along the X axis to align the body of the dipole suspension in the X direction. This condition, shown in FIG. 14, occurs particuarly when the spacing between the transparent electrodes is small.

However, when the transparent electrodes have a critical spacing of 2 to 10 times the suspension layer thickness, the "curtain effect" occurs as hereinafter described in connection with FIGS. 14 to 18 inclusive. The employment of the curtain effect enables effective alignments to be obtained in the X, Y or Z directions with compact spacing of the electrodes. Consequently utilizing the curtain effect shutter of FIG. 14 smaller voltages along the Z axis produce substantial alignments of dipole particles in the suspending fluid compared to the relatively large voltages required to similarly align the dipole particles in the shutter shown in FIG. 5.

Another method of obtaining the alignment in the X, Y or Z directions with a compact shutter operating at relatively small voltages, utilizes the technique of current shielding in which an electric current is passed along transparent conductors parallel to an aligning X or Y electric field, while no current is passed through the transparent electrodes when a voltage is applied between them in the Z direction. A compact shutter operating at relatively small voltages is achieved by this current shielding technique as more fully described in connection with FIGS. 38-43 inclusive. FIGS. 38-41 show two layer suspensions in which the current shielding and cross polarization is utilized to achieve an opaque state; and in which the transparent state is achieved by aligning the dipolar particles in both layers along the Z direction by electric fields applied along the Z direction, with no current flow along the transparent conductors.

In FIGS. 42 and 43 a single layer suspension cell is shown with FIG. 43 showing the alignment in the Z direction achieved by applying the electric field between the transparent electrodes without current flow along the electrodes, while FIG. 42 shows an alignment in the Y direction accompanied by current flow in the transparent electrodes in the Y direction.

It will be understood in connection with FIG. 42 that the passage of the dipole particle alignment from Z to the Y direction may be interrupted while the dipole particles are at the random alignment stage. The phenomena herein employed is shown graphically in FIG. 19. Random alignment of the dipolar particles occurs during the time in which the particles pass from the alignment in the Z direction, to the random state, and then into an alignment in the Y direction. To accomplish this the pulse of electric field intensity applied along the Y direction, or the X direction, may be stopped at a critical time t.sub.2 at which the particles have assumed a random position as indicated by a minimum transmittance through the dipole suspension layer.

ROTATOR

A dipolar device having applications for example for 3D television and movies is a polarized filter in which the plane of polarization is rotated electrically through 90.degree.; herein termed an XY rotator.

Various devices illustrated herein may be employed for this purpose: Thus, in FIGS. 6 and 7 the plane of polarization may be shifted from the Y direction to the X direction; and then applying the electric field as in FIG. 7 between electrodes 71 and 72, by first applying the electric field as in FIG. 6 between 69 and 70.

By modifying the electric circuit and the cycling of cells 50 and 51 as in FIG. 38; first switch cell 50 to the Y direction keeping 51 in the Z direction; then switch cell 50 to the Z direction and switch cell 51 to the X direction. This structure can also act as a shutter by switching as shown in FIG. 39; that is, by cross polarizing cells 50 and 51, Y and X respectively.

Another method (not shown) may incorporate electrodes embedded in the X direction on one sheet and in the Y direction on the other sheet.

Various other embodiments of this invention may be made.

An electro-optical device of the type shown in FIGS. 1 and 2 is highly satisfactory for many purposes such as camera shutters, control of room lighting by means of windows equipped with electro-optical "shades", etc. For other uses, such as environmental radiation control panels for buidling walls, roofs and space vehicles, data display screens, and the like, a suspension is required which may be switched from a light-transmitting state to a reflective state. Such a suspension is described hereinafter.

REFLECTING-TRANSMITTING PANEL

A reflecting-transmitting shutter is shown in FIGS. 5, 6 and 7. In a cell 78 is a suspension of flake-shaped particles 77. Electrodes 69 and 70 are provided at the top and bottom edges of the cell to impress a first electrostatic field across the cell 78 along the Y axis and the electrodes 71 and 72 are provided to impress a second electrostatic field along the X axis. A third pair of electrodes 75 and 76 for example, in the form of rings surrounding the field of view on either side of the cell 78 (as shown in FIG. 5), are provided for the purpose of impressing a third electrostatic field along the Z axis normal to the plane of the suspension layer and generally parallel to the incident light path. Alternatively, alignment in the Z direction can be obtained by a compact disposition of the electrodes as explained hereinafter.

A compact disposition of transparent electrodes to switch from the Z to the X or Y axes or vice versa may be achieved as hereinafter described in connection with current shielded fields along the X or Y axes, as explained in connection with FIGS. 38-40 inclusive.

In the operation of the light reflecting-transmitting shutter, the first two pairs of electrodes 69, 70, 71 and 72 are used to orient the flakes 77 parallel to the transparent sheets 52, 53 of the shutter. A suitable voltage is applied between the two side electrodes 71 and 72. As a result of the applied field, each of the random flakes represented by flakes 77, tends to move under the influence of a couple turning on its axis, so as to bring the electric forces on the flake into opposition and alignment, so that no further turning of the flake will result. Next, the voltage across side electrodes 71 and 72 is reduced to zero and an equivalent voltage is applied across top and bottom electrodes 69 and 70, respectively, resulting in a rotation of the applied field through 90.degree.. This causes a couple to act on each flake 77, turning it about its axis, which is perpendicular to a second axis. The result of the two-fold rotation is that the flake is aligned in a plane parallel with the transparent faces 52 and 53 of the cell. Since all flakes of the suspension are so oriented, the net result is a substantially specular reflection of light incident upon the cell with corresponding high opacity.

The alternation of the applied field between top and bottom electrodes 69 and 70 on the one hand, and side electrodes 71 and 72 on the other hand, is readily accomplished by using a suitable single-phase A.C. input (for example at 10 KHz and 1-3 Kv/cm). To insure that the voltages across the cell are balanced, a bridge circuit such as that shown in FIG. 8 may be employed. Terminals 81 and 82 are connected to electrodes 69 and 70, respectively, and terminals 79 and 80 are connected to electrodes 71 and 72 respectively. For example, when terminal 81 is at +E/2 volts and terminal 82 is at -E/2 volts (E being the total applied voltage), then terminals 79 and 80, which are connected to the other pair of electrodes 71 and 72, are at zero potential.

In the circuit as illustrated in FIG. 8, it will be noted that there are provided a first pair of power-supply leads 83, 84 connected to electrodes 69 and 70, respectively; a second paid of power-supply leads 85 and 86, connected respectively to electrodes 71 and 72, and a set of four balancing resistors 73, 83.sub.a, 84.sub.a, 85.sub.a, and 86.sub.a.

When it is desired to switch the shutter from a reflective to a transparent condition, the power to electrodes 69, 70, 71, and 72 is shut off, and ring electrodes 75 and 76 are energized. This results in a field substantially normal to the transparent sheets 52 and 53 of the cell, which causes the particles to be aligned parallel with the light path, making the cell transparent in the manner similar to that shown in FIG. 2.

SINGLE LAYER DIPOLAR ORIENTATION AND ITS EFFECT ON LIGHT

Light passing through a single layer dipolar suspension is affected by the orientation of the long dimension of the dipolar particles. Using the device shown in FIG. 5, with a dipole particle suspension comprising an electrically conductive needle shaped metal or an herapathite crystal, and with the dipolar particles oriented normal to the light path in the plane of the cell, unpolarized transmitted light is polarized. The electric vector of the transmitted polarized light is normal to the direction of the alignment of the dipolar particles. The electric vector in the direction of dipole alignment of the incident unpolarized light is absorbed because it induced a motion of charges in dipole particles. In the absence of an electric field, the dipole particles are randomly oriented and the dipolar suspension layer may be substantially opaqued to light but does not polarize the light.

In similar manner, when the particles are oriented in the direction of the light by an electrical field along the Z axis, applied for example between the ring 75 and 76 in FIG. 5, the cell is highly transmitting to ordinary non-polarized light which is transmitted without polarization.

FORCED RANDOMIZATION OF DIPOLES

In most of the embodiments of the invention thus far discussed, the dipoles will revert to random orientation within a fraction of a second after the orienting field is withdrawn, and no additional steps are necessary to achieve rapid opaquing of the shutter. Where it is desired to hasten the randomization process, it may be speeded up by switching from a Z to a Y orientation and interrupting the reorientation at the intermediate opaque condition or in various other ways. Randomization may be hastened for example, by mechanically vibrating the cell, by sliding or rotating one face of the cell relative to the other, or by application of a rotating electric or magnetic field. The rotating magnetic field may be generated by a conventional rotating field assembly such as is used in a single phase A.C. motor, the dipole chamber being disposed within the "cage" of field coils where the armature of such a motor is housed.

In switching from a Z orientation to an X or Y orientation it will be observed that the transmittance goes through a minimum as shown in FIG. 19. At a position intermediate the Z orientation and the X or Y orientation, the cell becomes quite opaque. This phase is ascribed to an intermediate condition, in which the dipoles are oriented in all directions at 45.degree. to the plane of the suspension, and behave in effect as if they were randomly oriented, and probably are quite close to the random state.

FIG. 19 shows a graph of transmittance versus time for an area whose particules have been initially oriented in the Z direction with a transmittance of 50-80 percent by a pulse E.sub.z of time duration t.sub.1. At the time t.sub.1 a pulse E.sub.x is applied, and in a time duration (t.sub.2 - t.sub.1) the particles have started to revolve into the X direction, but however, have only just passed into the random state, and for this reason the transmittance curve is dropped to a minimum at time t.sub.2, corresponding to the random state.

The E.sub.x pulse may be discontinued at this point and the particles will remain in the random state. However, if the E.sub.x pulse is continued for an additional time duration (t.sub.3 - t.sub.2) the particles now reorient themselves in the X direction polarizing transmitted light, with the transmittance increasing to between 20 and 50 percent.

Less time is taken for the particles to pass from one given orientation to the random state than from the given orientation to another at 90.degree. thereto. Moreover, the random state has the minimum transmittance for a single layer dipole suspension.

DIPOLE ROTATIONAL INERTIA EFFECT

In a high-speed switching of the electro-optical shutter by means of pulsed high intensity electric fields, the transmittance of the suspension may reach a maximum and then decrease, particularly in suspending fluids of small viscosity, (less than 10mp). Under these circumstances the dipole particles acquire an appreciable angular velocity, and by reason of inertia tend to shoot past the parallel position of maximum transparency, provided however that the fluid viscosity is sufficiently small so that the particles will rotate through a considerable angle before stopping. This effect may be overcome by compensation, for example, by shortening the duration of the pulse so that it has fallen to zero before the particles are fully aligned and letting them "coast" into alignment.

The inertia effect may also be utilized to provide a shutter which automatically transmits a light pulse of predetermined duration and shuts itself off. This is accomplished by applying a high-voltage pulse sufficient to impart a predetermined angular momentum to the particles, and allowing them to coast to and through the transmitting position to a position of extinction. The duration of the light pulse may be controlled by the momentum imparted by the intensity duration of the applied voltage and viscosity.

ORIENTATION BY REPEATED VOLTAGE PULSES

FIG. 20 is a graph showing a transmittance versus time for light passing through a cell containing a dipole layer subject to a voltage pulse, for example an A.C. voltage pulse 122 having an amplitude E.sub.z and time duration t.sub.1, resulting in the transmittance vs. time curve 123 shown. The rise time on this curve depends upon the particle dimension and concentration, the viscosity of the suspending fluid, and the electric field intensity. For the most rapid alignment and disalignment generally it is preferred to use fluids of relatively low viscosity in the range of 2 to 20 millipoise and electric field intensities just below the breakdown strength of the fluid, which is usually of the order of 100-300 kv/cm. Dipolar particles having a length of the order of 0.18 microns, 25 to 1 length to thickness ratio, and concentrations, and preferably as high as possible in the range 0.01 to 10 percent are also employed for this purpose. Minimum alignment times are about 100 nanoseconds. When the pulse terminates and there is no field, particles start to disalign by Brownian motion as indicated at 124 on the curve. When they are partially disaligned, a shorter pulse may be applied to cause the particles to realign following the transmittance time graph shown at 125.

FIG. 22 shows the application of a series of A.C. voltage pulses of amplitude E.sub.z in which the time duration t' is so short that only partial alignment is obtained during each pulse as shown by the segments of the transmittance time graph, FIG. 21, at 126, 127, and 128. Corresponding A.C. voltage pulses 129, 130 and 131 respectively are applied with a repetition rate of 1/t" pulses per second. These pulses cause a maximum alignment to be achieved bit by bit. Moreover, once the alignment has been attained, the alignment is maintained by the application of pulses of even shorter duration or lower amplitude.

A maximum peak voltage applied for a small time duration enables the particles to be quickly oriented into parallel position in the field. When the pulse discontinues the relaxation time is relatively long, as shown in FIG. 20, so that considerable time may intervene before particles are substantially oriented away from the parallel position. Thereafter, a pulse of relatively shorter duration or lesser voltage is sufficient to reestablish alignment.

If A.C. or D.C. electric fields are applied continuously they first orient the particles into parallel alignment; then, since the neighboring particles have absolute induced charges on their adjacent faces, the closest particles are drawn into contact, causing coagulation of the dipole suspension, rendering it inoperable. To avoid this coagulation, peak voltages below electrical breakdown of the shortest duration time, sufficient to cause the particles to orient, may be applied to establish maximum orientation in the least time. Thereafter, alignment may be maintained indefinitely and coagulation avoided by the application of a voltage pulse of suitable peak intensity and duration just sufficient to maintain alignment and counterbalance the effects of Brownian disorientation.

Pulses which may be employed may be D.C. pulses, or voltage pulses alternating in polarity of the type shown in FIG. 21, in which the pulse is an envelope for an alternating field of greater frequency as shown in FIG. 22. For example, in FIG. 22 the pulse length t' is 10 microseconds, and the time t" is 100 microseconds, then the alternating frequency of the electric field may be 10 megacycles; which would provide 100 alternations in the pulse. All the pulses may be of the same voltage. Alternatively, the initial pulse may be of a high voltage to align the particles quickly, and the subsequent pulses may be of smaller voltage or shorter time duration.

DIPOLE REFLECTIVE ABSORPTIVE PANEL

In another embodiment, this invention is useful in the form of a panel which becomes reflective on applying a voltage and absorptive when the voltage is removed.

A reflective-absorptive panel shown in FIG. 10 in fragmentary fashion comprises a cell enclosed in part by parallel glass plates, 52, 53. Plate 52 is provided with a transparent conductive coating 56, and plate 51 is coated with a conductive coating 88 which is a mirror such as a metallic coating. The cell is filled with a suspension of dipole particles 50, for example, herapathite particles, suspended in a suitable medium 51. The medium 51 may be, for example, a plasticizer such as dibutyl sebacate or the like. When a voltage is applied between coating 56 and coating 88, the particles 50 are aligned normal to glass plates 52, 53 and parallel with the direction of an incident light ray 59, so that light which enters through transparent plate 52 (and coating 56) is transmitted through the suspension, reflected from the mirror surface of coating 88, and transmitted back out through the suspension as a specular reflection 60. When the applied voltage is zero, the dipole particles disalign and the panel becomes absorptive (dark). Thus by applying a voltage from zero to the maximum aligning voltage, the panel may be regulated from highly absorptive (black) to reflective (silvery) or any condition between, as partially absorptive and partially reflective.

DIPOLE IRIS DIAPHRAGM

An iris, electrically controllable to any given diameter is shown in FIGS. 11 to 13. Referring to FIG. 11 there is shown a pair of point electrodes 89, 90 along the same axis in the Z direction in the center of the field, one near each face of the cell 91. When an electrical potential is applied between the point electrodes 89, 90, no effect is observed until the potential reaches a threshold value which depends on the chacteristics of the cell 91 and of the suspension. When the threshold voltage is reached, a small transparent spot 92 appears between the electrodes 89, 90. The diameter of the transparent spot 92 may be reversibly increased or decreased by increasing or decreasing the separation between the electrodes or by increasing or decreasing the applied voltage or both.

By the way of example, a transparent circular spot surrounded by a very dark area 93 approximately 8mm in diameter can be made to appear in a herapathite suspension by applying a 60 Hz A.C. potential difference and 5,000 volts or about 10KHz and 500 volts across electrodes 89, 90 spaced 3mm apart.

As shown in FIG. 12, the apparatus comprises a cell 91 enclosed in part by transparent plates 52, 53 filled with a suspension of dipole particles 50 in fluid 51 and equipped with two pairs of electrodes 89, 90. X-orienting electrodes 95, 96 are located at opposite edges of the cell 91 and are optional, depending on the service in which the apparatus is to be employed. Z-orienting electrodes 89, 90 are centrally located at or near the outer surfaces of transparent plates 52 and 53 respectively.

If the dipole suspension is initially in the random condition, the entire field of view is opaque. When a suitable voltage is applied between electrodes 89 and 90, a transparent spot 92 appears between them, the size of which depends on the electric potential difference and the spacing between the tips of electrodes 89 and 90. The dipoles 50 within the transparent spot are Z oriented by the applied field, while the dipoles outside the spot are outside the force field and remain randomly oriented. For this mode of operation, X orienting electrodes 95, 96 are not needed and may be omitted altogether.

The line of demarcation between the transparent spot 92 and the dark randomized area 93 surrounding it, is quite sharp. There is a critical initial electric field intensity required to start the formation of a small central transparent spot. The rim of the transparent circular area shows an abrupt change from transparent to opaque, (see FIG. 13).

In applying a field between point electrodes 89, 90 along the Z axis, the field intensity is greatest along the axis and then decreases as the distance radius from Z axis increases. The sharp line of demarcation between the transparent to the opaque areas appears to occur at that radius from the Z axis at which the field intensity falls below the critical field intensity required to cause the Z alignment.

The aligned dipoles also produce a counterfield which tends to offset the applied field. The counterfield of the aligned dipoles thus modifies the applied field.

When a step D.C. electric field is applied, the Z alignment within the dipole iris may be momentarily established, but then disappears due to the formation of the shielding field produced by ion migration to the outer surfaces of the dipole layer. A light pulse thus occurs. However, the dipole iris is permanently maintained by the application of an A.C. electric field. While 60 Hz A.C. for example, is very satisfactory, a greater frequency such as 5-10 KHz is preferred because of much smaller operating voltage.

A constant D.C. electric field may be employed when the axial electrodes 89, 90 are in contact with the dipole suspension layer. This continuously drains off ions which may be present in the suspending fluid and prevents the establishment of an ionic shielding field. A dipole iris of controllable diameter requires that the electrodes 89 and 90 be sufficiently far apart to establish an electric field of suitable intensity and diameter.

If the dipoles in the suspension layer are oriented in the X direction by a voltage applied between electrodes 95, 96 as shown in FIG. 12, the cell is partially transparent, transmitting and polarizing about 45 percent of the incident light. The layer has light-polarizing properties of about 45 percent of the incident light. The layer has light-polarizing properties, because of the X orientation of the dipoles.

If, when the cell is in the condition described and illustrated in FIG. 12, the voltage to the X oriented electrodes 95, 96 is cut off, the dipoles will retain their X orientation for a small time interval before appreciable randomization takes place. This time interval depends on the characteristics of the suspension, particularly on the viscosity of the suspending fluid and the dipole particle dimensions.

If, after the voltage to X orienting electrodes 95, 96 is discontinued, a voltage is immediately thereafter applied across Z orienting electrodes 89 and 90 a transparent spot 92 will again appear in the center of the cell, between electrodes 89 and 90. This phase is illustrated in FIG. 13, showing a transparent spot 92 which transmits light along the Z axis without polarization leaving the area 97 outside of the spot 92 partially transparent and polarizing, instead of opaque as in FIG. 11. This result is obtained because the particles unaffected by the field between Z orienting electrodes 89 and 90 still retain their initial X orienting in the area 97. The condition shown in FIG. 13 can be made to persist using a fluid of medium or high viscosity; or permanently if the fluid is solidified.

The appearance of the cell 91 in this mode of operation is also characterized by an opaque ring 93 defining the borderline between the fully transparent interior of the spot 92 and the partially transparent, polarizing area 97 outside the spot. Inside the spot 92, the dipole particles 50 are Z oriented. Outside the spot, the particles are X oriented. In the dark band there exists an opaque "pseudo-random" orientation, characteristic of particles in transition from X or Y to Z orientation, or vice versa.

An apparatus of this type provides a field of view comprising a fully-illuminated non-polarizing central pupil surrounded by a partially transparent, polarizing general field, and a dark line of demarcation between the two in the form of a dark ring 93 separating the central pupil 92 from the surrounding field. An element having these characteristics is a useful component for optical range finders, gunsights, navigational instruments and the like.

DIPOLE "CURTAIN SHUTTER"

FIGS. 14-16 inclusive show cross-sectional views, and 17 and 18 front views of a dipole curtain shutter. A pair of X or Y orienting electrodes in the form of parallel bus bars 110,111 are placed at the opposite edges of a rectangular cell 100 containing a dipole suspension layer between spaced transparent electrodes 106,109.

The dipole curtain effect may be applied to what is herein termed a dipole curtain shutter, the operation of which will be understood in connection with FIGS. 14-18.

FIG. 14 shows the cell initially in a random, opaque condition. When a voltage is applied between the bus bar electrodes and gradually increased after the threshold value is reached, the dipoles 50 closest to the bus bars become oriented in the X-Y plane, producing a narrow, transparent (and polarizing) strip near each of the electrodes. The oriented dipoles nearest the bus bar electrodes then act as secondary electrodes (see FIGS. 15-18), and these, in turn, cause orientation of particles still farther away. The effect produced is that of a shrinking opaque "curtain", which leaves behind it a transparent strip of a width which is a function of time, applied voltage difference, particle size and fluid viscosity.

Referring to FIG. 14, there is shown a dipole cell 100 containing a suspension of dipoles 50 between transparent members 102 and 103. The transparent member 102, for example, may comprise two transparent glass or plastic sheets 104 and 105 laminated together with a transparent conductive film 106 therebetween. In a similar manner, transparent member 103 may comprise transparent glass or plastic sheets 107 and 108 with a transparent conducting film 109 laminated therebetween.

Spaced X orienting electrodes 110 and 111 are supported respectively within insulating transparent or opaque blocks 98 and 99. Blocks 98 and 99 serve to insulate the electrodes 110 and 111. In addition, the transparent glass sheets 105 and 108 serve to insulate the transparent conducting films 106 and 109 from each other and from the electrodes 110 and 111. In FIG. 14 the field configuration is shown when a voltage is applied to the electrodes 110 and 111. The electrostatic field lines 113 and 114 issuing respectively from electrodes 110 and 111 have only a relatively short range of effect upon the dipole suspension in the immediate vicinity of the electrodes. In effect the field is shorted by the transparent conductive films 106 and 109. The result is that the central region between the transparent members 102, 103 generally indicated as 116 is effectively field free. The dipole particles are in random direction and hence opaque to light.

If the suspension was initially in the disoriented state, an alignment of the dipoles will initially occur in the vicinity of the electrodes 110 and 111. However, as soon as the dipoles closest to the electrodes 110 and 111 are aligned, the electrostatic field is effectively moved toward the central area 116 as shown in FIG. 15. In fact, the aligned particles tend to act as a pathway for the electrostatic lines of force which then travel along the aligned particles leaving a smaller central strip 116a still in the random condition. This process continues, the dipole particles 50 aligning in succession like a series of falling dominoes, so that the field free central region 116a becomes smaller and smaller and the planes at which alignment is occuring continue to approach each other until the condition shown in FIG. 16 prevails in which all of the dipole particles 50 between the electrodes 110, 111 are aligned in the X direction.

FIG. 17 is a front view of the condition shown in FIG. 15 showing two horizontally X aligned areas 112 and 115, and a central disaligned area or band 116. The areas 112 and 115 generally have a transmittance of between 15 percent and 45 percent and strongly polarize light. The central band 116 forms a black absorbing strip having a transmittance of the order of 0.01 to 1 percent.

Assuming no voltage initially, and the particles in a random state, the application of the voltage along the X axis causes the black absorbing band 116 to become progressively narrower until it disappears, the entire cell then being uniformly oriented and transparent to light polarized with the H vector parallel to the orientation direction of the dipole particles.

FIG. 18 shows another mode of operation.

As a first step the cell shown in FIG. 14 may be rendered initially transparent by applying an electric field along the Z axis which aligns the dipoles normal to the face of the cell; and then as a second step FIG. 18, an electric field is applied along the X axis. When these steps are taken, the dipole particles in the strips 112 and 115 in the vicinity of the X orienting electrodes 110 and 111 attempt to realign themselves in the X direction but first pass through a random state of very low transmittance as shown in strips 117 and 118.

As this process continues, the central area 116b in FIG. 18, which was initially transparent, becomes narrower and narrower. On both sides of the transparent strip 116b there are dark strips 117 and 118 which are in that random phase through which the initially Z oriented particles pass to become X oriented particles.

As the process continues further, the transparent band 116b becomes narrower and narrower, finally merging into a single central dark band. As the strip areas 117 and 118 come together and merge, transparent area 116b disappears; thereafter, all of the randomly directed particles in the dark central band produced by merging of bands 116 and 117 disappear, and the entire cell then contains dipole particles which are X oriented.

When the sequence just described is complete, all of the dipole particles are oriented in the X direction and the cell is partially transparent and polarizing. From this condition the cell may be switched as desired to either an opaque or a transmitting non-polarizing condition, by application of a suitable voltage pulse between Z orienting electrodes via the transparent conductive films 106, 109.

If it is desired to switch the cell from the X orientation to an opaque condition, the electric field in the Z direction is applied between the transparent electrodes 106, 109 in a short pulse. The short pulse swings the dipole particles part of the way from the X toward the Z orientation, but is discontinued before the orientation is complete, leaving the dipoles in an intermediate random condition. The cell is then opaque, and is either absorptive or diffusely reflective, depending upon the characteristics of the particular dipole suspension used.

If it is desired to switch from partially transparent X orientation to a fully transparent condition, it is necessary only to apply the electric field between the Z orienting electrodes, via the transparent conducting films 106, 109, in a longer pulse of sufficient duration to allow the oriented action in the Z direction, and the cell to become fully transparent. FIG. 19 graphically illustrates the operation of this cell.

A dipole curtain shutter of the type just described is useful in a variety of ways, among which may be mentioned exposure control and masking in photographic processes, as a light control element for displays, etc.

COMPACT LIQUID DIPOLE SHUTTER

FIGS. 38 and 39 show exploded views of still another embodiment of the present invention, comprising two cells, 119 and 120, each containing a suspension of dipole particles 50 in a transparent medium 51. Each of the cells is enclosed by suitable enclosure means including transparent walls of glass or the like which are omitted in this view for purposes of clarity.

Each of the two cells 119, 120, is located between a pair of electrically conductive, transparent films. The films 106 and 109 are located on either side of cell 119 and films 106a and 109a are located on either side of cell 120. The conductive films may be enclosed in and protected by walls of glass or the like in the manner set forth above in connection with FIGS. 14, 15 and 16. Where appropriate, a single sheet of glass, transparent plastic or the like may serve as one of the enclosing walls of one of the cells and simultaneously as one of the enclosing and protecting walls for a neighboring conductive film. In certain cases the conductive film forms a layer on the glass surface in direct contact with the dipole suspension layer.

FIG. 38 shows the cells 119, 120 in the transparent condition. This condition is brought about by orientation of dipole particles 50 in a "Z" direction, normal to the faces of cells 119 and 120, and hence parallel with the direction of an incident light ray 121. In order to effect this orientation, a voltage is imposed between conductive films 106, 109, and 106a, 109a, in both cells. The voltage between the films 106 and 109 orients the dipole particles in the cells 119, 120. The voltage is applied to the conductive films by suitable leads electrically connected to bus bars 101a, 101b, 101c, 101d or the like in the form of metallic strips along the edges of the conducting films 106, 109, 106a, 109a. Strip 101b along the top of film 109, is connected as shown to one side of a voltage source, which may be an A.C. or D.C. generator or the like (not shown), and a charge is thereby imparted to the whole surface of conductive film 109. Strip 101a, along the bottom of film 106, is connected to the other side of the same voltage source and film 106 thereby acquires an opposite charge. The opposiing charges of the two films set up an electrical field in the Z direction indicated by the arrows, and this field is effective to orient dipole particles 50 parallel with the arrows, thus making the cell transparent.

In similar manner, strip 101c on the right edge of film 106a, is connected to the one end of the voltage source and strip 101d on the left edge of film 109a is connected to the other, thus generating a similar electric field through cell 120, and orienting the dipole particles therein also in the Z direction.

When it is desired to switch the shutter to an opaque condition, the electrical connections are switched to the arrangement, shown in FIG. 39. With the apparatus in this condition, strip 101b, on the top edge of cell 119, is connected to one side of the voltage source, and strip 101a, on the bottom edge of cell 119 is connected (through ground) to the other, thus setting up an orienting electric field within the cell in the vertical or Y direction.

If conductive films 106 and 109 were widely spaced as illustrated in FIGS. 38 and 39, the electric field applied across the dipole layer alone would be sufficient to orient dipole particles 59 in the vertical or Y direction, as shown. FIGS. 38 and 39, however, are exploded views, and for the sake of compactness films 106 and 109 should be located in close proximity to the surfaces of the dipole suspension layer. When so located, the conductive films 106 and 109 whould tend to deflect the field by "short-circuiting" the lines of force, thereby causing the field to bypass the interior of the cell, leaving no field throughout most of the cell area to orient dipole particles 50 along the X or Y axes. To overcome this effect, strip 101b along the top of film 109, is also connected to the one side of the voltage source and strip 101a, along the bottom thereof, to the other. The conductive films have an optimum resistivity per square which limits current and ohmic heating to a low value, yet provides a current shield sufficient to provide a uniform voltage gradient in the X or Y directions. This shield causes a direct or alternating current to flow through film 109 in or against the direction indicated by the arrows. Similarly, strip 101b, on the top of film 109, is connected to the one end of the voltage source and strip 101a, on the bottom thereof, to the other side of the voltage source, thus causing current flow in film 109. The effect of the current flow in films 106 and 109 is to prevent their functioning as conductive paths for the electric field between strips 101a and 101b. Consequently, the electric field gradient is produced parallel to the X or Y axes, which is effective to orient the particles as indicated in FIG. 39.

In similar fashion, an electric field is generated in cell 120 by connecting strip 101c, on the right edge thereof to one side of the voltage source and strip 101d, on the left edge, to the other. Short-circuiting of the electric field is avoided and an electric field gradient is established in the cell parallel to the X or Y axes by connecting strip 101c to the one end of the voltage source and strip 101d to the other, thus causing current flow through films 106a and 109a, respectively, in or against the direction indicated by the arrows.

With the cell in the opaque condition, as illustrated in FIG. 39, the dipole particles in cell 120 are oriented in the horizontal or X direction, and the dipole particles in cell 119 are oriented in the Y direction. The opaquing effect is the same as excluding light by crossed polarizers.

For the type of cell illustrated ijn FIGS. 38 and 39, best advantage can be taken of the "crossed" condition of the two cells in the opaque state, by using a light polarizing type of dipole particle, such as a suspension of herapathite crystals in a transparent, inert, non-conductive fluid, as hereinafter set forth in Example A. Alternatively, the suspension may comprise metallic dipoles in a fluid, as hereinafter described.

The conductive films 106, 109 and 106a and 109a may be any suitable transparent electrically conductive film as previously described herein.

The cell illustrated in exploded form in FIGS. 38 and 39 is shown in assembled form in FIGS. 40 and 41. In FIGS. 40 and 41, some of the electrical connections illustrated in FIGS. 38 and 39 have been omitted, and the transparent plates of glass, transparent plastic or the like, forming part of the shutter assembly, have been included. Thus, in FIG. 40, the cell is seen to comprise a first glass plate 133, and a second glass plate 134, with conductive film 106 laminated between them. Second plate 134 serves also as one of the containing walls of cell 119. The opposite facing wall of cell 119 is a third glass plate 135. Conductive films 109 is laminated between plate 135 and a fourth plate 136. Similarly, conductive film 106a is laminated between plate 136 and a fifth plate 137, which also serves as one of the containing walls for cell 120. The opposite facing wall of cell 120 is a sixth plate 138. Conductive film 109a is laminated between plate 138 and a seventh plate 139. It will be understood that the relative dimensions are distorted in FIG. 40 for purposes of clarity, and the entire assembly may be, and preferably is, quite thin -- for example having an overall thickness of 6mm, while the width as viewed from the front (as in FIG. 39) may be 60mm or more. In FIG. 40 the dipoles in layers 140 and 141 are shown oriented the same as in the exploded view of FIG. 39, for the opaque state.

Another embodiment of the compact liquid dipole shutter according to the present invention is illustrated in exploded perspective in FIGS. 42 and 43. In this design, only one cell is used. In the transparent state shown in FIG. 43, all of the dipole particles 50 are oriented normal to the faces of cell 142, in the Z direction, in essentially the same manner as illustrated for the two cells 119 and 120 of FIGS. 38. Thus to render the shutter transparent, metal strip 143 at the top of conductive film 109 is connected to the one side of a voltage source, so that film 109 acquires a charge. Strip 144, at the bottom of conductive film 106, is connected to the other side of the voltage source, and film 106 thus acquires an opposite charge. The two oppositely charged films create an electrical field effective to orient the particles in the Z orientation, as shown.

When it is desired to switch the shutter of FIG. 43 to the opaque condition, the electrical connections are switched to the arrangement shown in FIG. 42. When so connected, bus bar 101 at the top of cell 142 is connected to one side of the voltage source, and bus bar 101a at the bottom thereof, connected to the opposite side, setting up an electric field effective to orient dipole particles 50 in the vertical or Y direction. Simultaneously, diversion of the electric field due to the presence of conductive films 106 and 109 is avoided by connecting strips 143 and 145 to bus bar 101 and strips 144 and 146 to the bus bar 101a, thereby causing a current to flow in each of the conductive films 106, and 109. In connection with FIG. 42 it is preferred to induce the opaque random state by a voltage pulse of suitable amplitude and duration, as explained in connection with FIG. 19.

In this type of shutter, in order to achieve a large electrodichroic ratio, it is important to provide dipole particles having suitable electrodichroic characteristics. One type of dipolar particle which may be employed in a cell such as shown in FIG. 43 is a metal flake, for example a suspension of minute flakes of aluminum or the like. Such flakes may be prepared by chopping or milling a thin, aluminum layer, preferably while the same is carried on a suitable carrier film such as a soluble polymer film or the like, and then dissolving away the carrier film and concentrating the desired size fraction by centrifugation.

When a suspension of flakes is oriented by a field parallel to the path of incident light rays, as shown in FIG. 43, the particles are aligned edgewise to the incident light, and the suspension is transparent.

When, on the other hand, the particles are oriented by a field parallel to the Y axis, as in FIG. 42, the major axes of the particles are vertically oriented, while the minor axes are randomly arranged in the horizontal plane. The effect might be likened to a roomful of panels each hung by a single string from the ceiling, and the passage of light is effectively prevented.

DIPOLE SHUTTER WITH SPACED ELEMENTS

The dipole shutters just described, and illustrated in FIGS. 38 to 43 inclusive, have the advantage in that they may be compact. However, a small current must be drawn which does not constitute a disadvantage in most applications. Where a very small power drain is a primary factor and compactness is a secondary factor, the embodiment of the invention illustrated in FIG. 44 may be utilized.

FIG. 44 shows a shutter comprising a pair of spaced conductive loops or rings 75 and 76, which may be of any conductive material, for example copper. In the space between the rings are two cells 147 and 148.

Cell 147 is enclosed by transparent glass walls 149 and 150, spaced by annular gasket 151. In the interior of the cell, within the central space bounded by walls 149 and 150 and gasket 151, is a suspension of dipole particles 50 in a transparent, non-conductive fluid medium. The suspension may advantageously be a suspension of herapathite dipoles as described in Example A.

Cell 148 is of construction similar to cell 147, and is bounded by transparent glass walls 152 and 153 and gasket 154. Cell 148 contains a dipole particle suspension similar to that contained in cell 147.

Electrodes 155 and 156 are provided at opposite side edges of cell 147, and electrodes 157 and 158 are provided at the top and bottom, respectively, of cell 148.

The electrical system comprises a source of electrical potential (not shown), having one side indicated at 159 and the other as ground 160.

Various switches or relays or a single multiple-throw switch, are provided as indicated at A,B at various points in the figure.

When the shutter is to be made transparent all switches are thrown to the "B" position indicated in the drawing. This causes ring 76 to be connected to one side of the voltage source and ring 75 to the other side. The electrodes 155, 156, 157 and 158 are disconnected from the voltage source in this phase of operation. The result of the connection of rings 75 and 76 as just indicated is to impart one charge to ring 76 and an opposite charge to ring 75. This arrangement sets up an electrostatic force field through the cells in the direction indicated by the dashed lines 161a, generally parallel with the path of an incident light ray indicated at 161. The force field orients the long direction of the dipole particles in both cells in the same direction as the light ray and renders the shutter transparent.

When it is desired to make the shutter opaque, the switches are thrown to the position in which they are actually illustrated (the "A" position in each case). This position disconnects rings 75 and 76 from the voltage source 159, and connects electrodes 156 and 157 to the one side of the voltage source. The result is to set up a vertically directed electric field in cell 148, and a horizontally directed field in cell 147, and the field in each case being parallel with the transparent walls of the cell. The long directions of the dipole particles are thereby cross-oriented, those in cell 148 being vertically oriented and those in cell 147 horizontally oriented. The incident light is thereby effectively blocked. In one embodiment of the present invention the dipole particles employed are minute elongated crystals of herapathite. The electrodichroic ratio of such a shutter is particularly high, ratios of 12 and greater being obtainable.

In order to prevent the conductive rings 75 and 76 from "short-circuiting" the electric field lines and thus diverting the field away from the interiors of the cells, each ring should be separated from the cell nearest it by a suitable air gap, or some transparent insulating material (indicated as distance d in FIG. 44). Also, to prevent shorting of the field between electrodes 156 and 157, or between electrodes 155 and 158, there should be provided an air gap or other transparent insulating means between cell 147 and 148, also indicated in the drawing as an air gap of width d. The magnitude of d, or the insulating value of other insulating means employed, depends primarily on the voltage of the field, and the distance between electrodes 156 and 157, or 155 and 159, respectively and whether air or other medium is employed. For a typical shutter having an aperture, or cell diameter of 3cm, and using a dipole suspension having a resistivity of 30 megohm-cm, and a maximum voltage just under breakdown in the range of about 30kv/cm, a spacing of 3cm between each ring and its neighboring cell gives satisfactory results. The spacing between cells should ordinarily be approximately the same as the spacing between the cells and the rings.

It may be noted that current flow is not required either in rendering the shutter transparent or opaque. Although high voltages are required, the only power loss results from the usually small leakage current in a well designed shutter, so that the power demand is very small notwithstanding the voltages involved.

PULSE CIRCUITS

FIG. 45 shows an electronic circuit employed for the application of D.C. pulses of a large voltage across a dipole cell.

In FIG. 45 a switch 161 operates a conventional pulser 162. This pulser produces a positive 50 volt rectangular D.C. pulse 163 of controlled duration which has a rise time of about 1 microsecond.

The intermediate pulse amplifier 164 is to increase the amplitude of the voltage pulse 163 from +50 volts to +200 volt pulse 165. FIG. 46 shows the intermediate pulse circuit amplifier 164, with the values of its components.

The 200 volt pulse 165 applied to the control grid of the high voltage pulse amplifier circuit 166 results in an output negative 4,000 volt pulse 167 having a pulse duration which is controlled by the pulser 162. FIG. 47 shows the circuit and component values of the high voltage pulse amplifier 160. The high voltage pulse 167 is applied to the transparent electrodes 168 and 169 of the dipole cell 170 which is shown schematically in FIG. 45. Between the transparent conductive layers 168 and 169, a dipole suspension layer 171 is provided. The construction of the cell may be generally similar to that shown in FIG. 1.

In the tests made and illustrated herein the graphs of FIGS. 31 through 35 inclusive, a single rectangular D.C. pulse was applied and the effect observed upon a light beam from a suitable light source. The light beam passed through the dipole cell 170 to a photocell (not shown) which was amplified by a logarithmic amplifier and viewed on a storage oscilloscope, which displays a single transient.

The storage oscilloscope was triggered by the voltage pulse applied to the dipole cell. A curve of transmittance versus time was displayed and photographs obtained from which the graphs of FIGS. 31-35 were made.

Other well known circuits may be employed. In place of the pulser 162, there may be employed a frequency-pulse generator which can put out any of the voltage pulses shown in FIGS. 19 to 22, inclusive, which may be suitably applied to the X,Y or Z electrodes to obtain the required optical response.

POLARIZERS

A polarizing medium results if the fluid layer shown in FIG. 12 is solidified (as by cooling if the fluid is a thermoplastic or a glass). For example, the dipoles may be metal needles, such as platinum, and the medium a low melting point low viscosity glass, such as "solder glass".

The dipole particles utilized in polarizers according to this invention differ from those of prior art polarizers such as the Polaroid "J" polarizers which were an oriented herapathite suspension in cellulose acetate butyrate. The dipoles of the present invention are controlled in size and shape to close tolerances, whereas those of the prior art were of random size and shape. Consequently, polarizers produced in accordance with this invention have a greater percent transmission and a greater percent polarization with little perceptible light scatter. Light scatter was a particularly serious disadvantage of prior art polarizers and was a result of the process of manuacture in which larger particles were produced in situ.

PREPARATION OF DIPOLE SUSPENSIONS

The state of the prior art was unsatisfactory in relation to the preparation of asymmetric particle suspensions having electro-optical properties suitable for utility for the present invention. The mathematical section hereinafter sets forth the ranges of physical variables of dipolar particle suspensions which provides electro-optical properties suitable for the purposes of this invention. As a result of the specifications provided by this mathematical analysis, certain novel compositions and methods of preparation of dipole particle suspensions suitable for practice of this invention were discovered, examples of which follow:

PREPARATION OF SUBMICRON HERAPATHITE CRYSTALS

To produce submicron herapathite crystals in high concentration in a low viscosity suspending fluid, which form an optically clear, non-scattering dipole particle suspension of suitable electrodichroic ratio and sensitivity, the reacting solutions should be:

1. miscible

2. near maximum concentration

3. at low viscosity

4. at low temperature

5. rapidly mixed in reacting proportions

6. violently agitated

An example follows:

EXAMPLE A

No. 1 Parts by Weight Iodine 20 Normal propanol 80 Total 100

The iodine is dissolved in the normal propanol by heating and shaking.

No. 2 Quinine Bisulphate 32.5 Methanol 67.5 Total 100

For complete solution warm with agitation in a hot water bath to about 70.degree..

No. 3 Nitrocellulose, 5-6 second type RS (solids) 12.5 Isopropyl Alcohol 5.5 Isopropyl Acetate 16.0 Toluol 16.0 Methanol 50.0 Total 100.0

Solutions Nos. 2 and 3 are then heated to 70.degree.C and used to prepare No. 4.

No. 4 Material % Solution %Solids No. 2 Quinine Bisulphate 32.5 12.5 4.06 No. 3 Nitrocellulose 12.5 60.6 7.55 Methanol 13.0 Butyl Acetate 14.0 Total 100.0 11.61

This solution is then warmed to 70.degree.C and pressure filtered at the same temperature to remove any small undissolved crystal which would act as nuclei for crystallization.

Solutions Nos. 1 and 4 are then mixed in proportion and rapidly mixed in a container cooled by an acetone dry-ice bath. The result is:

No. 5 Before Reaction After Reaction % Solids Solids % Solids No. 1 9 pts. Iodine 20.0 1.8 (Quinine Bisulphate) 4.06 3.7 5.5 IQS 44.4 No. 4 91 pts. (Nitrocellulose ) 7.55 6.87 N/C 55.6 Total 100 5.5 12.37 100.0

While Solution No. 5 is being prepared, akyl epoxy stearate (Celluflex-23) a high boiling solvent also known as a "plasticizer" is cooled in an ice bath to 0.degree.C, and added in the following proportions to make a paste containing the submicron herapathite particles in suspension:

No. 6 Paste % Pts. Material Solids Solids Solution (Iodoquinine 4.24 Sus- 13. No. 5 77 Sulphate) pended Cellu- Nitrocellulose 5.30 Sol- 16.3 flex-23 23 Celluflex-23 23.0 ution 70.7 Total 100 32.54 100.0

No. 6 is then mixed with a mechanical stirrer for about 10 minutes to insure complete reaction and homogenity. After this, to remove the volatile solvents, the suspension No. 6 is placed in a rotating evacuator for about two hours and a paste is then obtained which is substantially free from solvents except the plasticizer and which has a resistivity of at least 30 megohm-cm.

The analysis of the paste resulting from No. 6 after the volatiles have been removed is:

No. 7 % by Wt. Iodoquinine Sulphate 13.0 Nitrocellulose 16.3 Celluflex 23 70.7 Total 100.0

As a diluent for the paste there is then prepared:

No. 8 Xylol 80 parts Butyl Acetate 20 parts Total 100.0 No. 9 No. 7 50 parts No. 8 50 parts Total 100 parts

A solids analysis of No. 9 is as follows:

Solids % Solids Iodoquinine Sulphate 6.5 44.3 Nitrocellulose 8.15 55.7 Total 14.65 100.0

% Solids Total -- 14.65%

% IQS in Suspension--6.5%

No. 9 may be used directly or be centrifuged to obtain a supernatent liquid for use in an electrodichroic system.

A herapathite suspension prepared in this manner is characterized by elongated submicron crystals of herapathite, which remain in suspension without settling and which is suitable for use as a dipole particle suspension in the practice of this invention.

Chemically herapathite is quinine trisulphate dihydroiodide tetraiodide hexahydrate, the chemical name for 4C.sub.20 H.sub.24 O.sub.2 N.sub.2 .sup.. 3H.sub.2 SO.sub.4 .sup.. 2HI .sup.. I.sub.4 .sup.. 6H.sub.2 O. The molecular weight 2,464.

Stoichiometrically herapathite contains approximately 25.8 percent of iodine which is approximately a ratio of iodine to quinine bisulphate of 1/3.

However, I have found that the proportions can be varied from one-half through one-fourth. This is apparently due to herapathite being a molecular compound or a mixed crystal in which the proportion of the components may vary.

Moreover, the HI in the compound is present in the proportion of two moles of quinine to one of HI. The heating of the iodine solution No. 1 usually suffices to provide sufficient HI as set forth in the above example. The presence of HI in stoichiometric quantities is required to form a stable crystalline compound. An additional quantity of HI may be added to achieve the molar ratio set forth.

Generally I have found the composition of Example A to be satisfactory, and this composition has been used in most of the tests.

FLAKE DIPOLE SUSPENSIONS

To prepare metallic flakes for use as dipole particles, a layer of metal is deposited, for example, by known vacuum deposition techniques, on a film of plastic or other convenient substrate, and the substrate is subsequently dissolved, thus causing the metal film to be suspended as a flake in the solvent. The suspended film is then chopped to flakes of the desired size.

EXAMPLE B

ALUMINUM FLAKE SUSPENSIONS

Aluminum flake suspensions were prepared in silicone oils utilizing irregularly shaped flake-like aluminum particles of a diameter between 1 to 17 microns and a thickness about 0.01 to 0.1 microns. Two different suspensions were prepared:

Material Density Viscosity Parts of Weight Aluminum Flake 5 Silicone Oil 0.65 10 Silicone Oil 3.0 50 Silicone Oil 1000.0 35 Total 100

Viscosity: -- 23 cs

Concentration: 0.05 gms Al per gm suspension

The aluminum is put into suspension by shaking the mixture at room temperature.

EXAMPLE C

THIN ALUMINUM FLAKE SUSPENSIONS

A new method of preparing ultrathin aluminum flake suspensions has been developed. Aluminum flakes 1-17 microns in diameter, and 0.01 to 1.0 micron in thickness are used as the starting point. A suspension is prepared by adding 48 grams of the aluminum flake material to 300 cubic centimeters of di-iso-octyl adipate. This mixture is then shaken and poured into a 500 cubic centimeter graduated cylinder and allowed to settle. Most of the aluminum flakes then settle to the bottom of the graduate. However, a small portion of the flakes remain suspended in a thin layer at the top of the graduate. This top layer then comprises ultrathin aluminum flakes, approximately 0.01 micron in thickness. The layer is then poured off and utilized for testing.

Thus, by means of this flotation method, the thinner flakes are separated from the thicker flakes. These thin flakes may be further separated and concentrated by centrifuging.

With the low viscosity, thin flake suspension, a large electrodichroic ratio and large sensitivity is obtained.

EXAMPLE D

ULTRATHIN ALUMINUM FLAKE SUSPENSIONS

Another way to make thin flakes of aluminum or the like is to coat a thin rubber or other stretchable sheet with a film of aluminum by exposing it to aluminum vapor, until a film of 50-200 A thickness has been built up. This sheet is then stretched to break up the surface into flakes of aluminum. The underlying sheet is next dissolved in order to place the flakes in suspension. Finally, the large flakes are eliminated and the small flakes in the desired size range are concentrated, by centrifugation. This technique can also be employed using polyvinyl alchol or polyvinyl chloride sheets by heating the sheets after the coating step, to facilitate their being stretched.

The resulting suspension is suited for use in those embodiments of the invention which require a suspension of dipoles in the form of flakes - for example the Reflective-Absorptive panels discussed earlier herein.

NEEDLE-LIKE METAL PARTICLES FROM STRETCHED POLYMERS

A method for the production of suspension metal rod dipole particles is: dissolve a metal salt in a matrix of polyvinyl alcohol, cast the solution as a polyvinyl alcohol film, soften and stretch the film in known manner, reduce the metal salt to the metal by exposure of the film to a reducing liquid or gas, and then dissolve the polyvinyl alcohol in water.

NEEDLE-LIKE METAL DIPOLES FROM "WHISKERS"

Needle-shaped metallic dipoles may be formed from a metal such as gold, platinum, paladium, chromium, tin or the like, which are known to grow submicron-diameter crystal whiskers under appropriate conditions, for example, from the vapor phase. These crystal whiskers may then be incorporated into fluid to form a dipole suspension. Such needles, if classified to a uniform length, may be made sharply selective as to the wavelengths of light affected by them. This property results from their large length-to-thickness ratio and resistivity, as explained below. Such materials constitute a new class of pigments different in effectiveness and mode of operation from conventional pigments.

METHOD OF VAPOR PHASE GROWTH

In the vapor phase, factors controlling the growth of needle-like whisker dipoles are partial pressure, temperature of the metal vapor, temperature and nature of the deposition surface and the time of growth. The growth usually occurs best under vacuum, or inert gas such as helium or nitrogen, but in some cases as with gold whiskers can be grown in air. Two gold sheets separated by a few millimeters and by a few degrees temperature difference, held in air at a temperature such as to generate an appreciable gold partial vapor pressure, will cause gold whisker crystals to grow normal to the surface of the cooler gold sheet. The dimensions of the whiskers are such as to fall within the size ranges herein specified. On cooling, the whiskers may be incorporated in a plastic film formed by coating the surface of the gold sheet, encompassing the whiskers. Upon drying, the film may be stripped away and dissolved leaving the gold dipoles in suspension in the fluid. This process may be performed continuously using an endless belt of a material such as stainless steel, which is initially provided with active sites for initiation of whisker growth.

METHOD OF LINEAR COOLING OR EUTECTIC MELT

Another important method is indirectional cooling of a eutectic melt to precipitate oriented fibres. Examples are chromium fibres (whiskers) grown in a copper matrix; and aluminum nickelide - Al.sub.3 NI grown in an aluminum matrix. The matrix is then dissolved in acid leaving behind the metal fibres which are washed in a solvent, suspending in a fluid, chopped into suitable lengths in a high speed blender, and centrifugally assorted into various lengths. The chromium metal suspension reporter later herein was prepared in this manner.

Submicron whiskers may be obtained by using small diameter melted rods, such as 0.1 to 1cm minimum and large temperature gradients over a similar axial distance.

FLAT CRYSTALS

Flakes made from crystalline material such as lead carbonate (pearlescence) may be grown to any desired size by methods well known to the art. These flakes have an index of refraction of about 2.4, and when placed in a fluid having an index of refraction of about 1.5, are readily aligned by an electric field, and in the equivalent of about 15-20 layers almost totally reflect visible ultraviolet and near infrared radiation, when disoriented or oriented in the plane of the cell wall or sheet; while being almost completely transparent when aligned normal to the sheet surface.

Zinc vapor will deposit submicron flat crystals on a substrate, which can be dissolved away as above described, to yield a metal flake suspension having a dipolar characteristic.

Graphite forms flat hexagonal flakes which, when suspensed in an oil of small viscosity, show dipolar characteristics.

ANTENNA EFFECT

The submicron dipole particles of this invention may be described as behaving like minute dipole antennae, exhibiting many of the physical properties associated with the macroscopic dipole antennae in a manner similar to those used for transmission and reception of microwaves for television, radar signals and the like. The submicron dipoles differ from these large antennae primarily in that they are "tuned" to very much shorter wavelengths namely those in the visible and neighboring portions of the electromagnetic spectrum. Despite the difference in dimensions, certain dipole particles of this invention behave toward light rays in the manner very similar to that in which large dipole antennae behave toward microwaves.

To review briefly the theory underlying this concept, light is an electromagnetic wave having three functional attributes, which are (1) amplitude or intensity, (2) wavelength or color, and (3) polarization or the vibration direction at right angles to the direction of propagation of the ray. Both television waves and light waves are electromagnetic waves, and share the same fundamental properties.

A half-wave dipole antenna, of the type used for television reception, is responsive to all three attributes, and absorbs and reradiates energy in a manner dependent on all three, depending on its length, thickness, resistivity and angular orientation with respect to the incident wave. In the same way a half-wave dipole tuned to visible light is capable of controlling all three attributes of light by varying its length, thickness, resistivity and angular orientation.

The electric power absorbed from the radiation by the half-wave dipole depends upon two orientation angles of the dipole. The first angle, .phi., is that between the length of the dipole and the direction of polarization of the signal. The direction of polarization of an electromagnetic wave is herein defined as the vibration plane of the electric vector of the wave. The second angle .theta., is the angle between the long axis of the dipole and the ray direction.

FIG. 24 shows, for a half-wave dipole antenna, a polar graph of absorbed or reflected radiant power versus signal direction .phi..

In FIG. 25 the radiation ray path is normal to the plane of the diagram, and there is shown the angle between the dipole length and the polarization direction .theta. versus the power absorbed or reflected by the dipole.

The radiation interacting with the dipole depends upon two angles; the angle between the length of the dipole and the ray path, and the angle between the length of the dipole and the direction of the electric vector of polarization of the ray.

Maximum absorbed or reflected radiant power results when the antenna is aligned parallel to the polarized electric vector of the radiation and at right angles to the signal path (.theta. = 0 and .phi. = 90.degree.). The antenna absorbs or reflects no power when it is placed at right angles to the polarized electric vector of the radiation, or arranged parallel to the ray path.

When adjusted for a maximum absorption or reflection of radiant power, a half-wave or .lambda./2 antenna is then said to become resonant to the particular wavelength .lambda..

The power absorbed by the dipole from the radiant energy may be reradiated, or absorbed and dissipated as heat, depending on the length and width and the electrical resistance of the half-wave dipole antenna.

If power is to be absorbed from the dipole antenna and utilized in an outside electric circuit, as for example in a television set, a matched or characteristic resistance of about 73 ohms must be inserted at the center of the half-wave dipole antenna, as shown in FIG. 26.

An antenna may be made of such material, thickness and length as to achieve almost complete power absorption, or almost complete reflection.

In FIG. 27 there is also shown a half-wave (.lambda./2) antenna in which the central resistor is replaced by a single rod 172 having a distributed resistance of approximately 80 ohms, which results in total absorption of radiation in a wavelength range .DELTA..lambda. centered about the wavelength .lambda..

Now, if instead of a half-wave antenna with a central resistor or an equivalent distributed resistance, a half-wave antenna of low resistance is employed, then the half-wave dipole antenna becomes relfective for the full wavelength. The radiant power may be said to be absorbed by the half-wave dipole and then reradiated in all directions, with the intensity direction pattern shown in FIG. 24.

FIG. 28 illustrates a very important property of the half-wave dipole antenna, to the "effective cross section". The half-wave dipole antenna shown has a thickness of (1/25) its length. Its length is .lambda./2 and its thickness .lambda./50. The physical cross section of this half-wave dipole at right angles to the light ray is: (.lambda./2) (.lambda./50) = .lambda. .sup.2 /100. However, it is known that the effective cross section of a half-wave dipole antenna is much larger. The cross section from which the half-wave dipole appears to absorb power from a polarized wave with the electric vector parallel to the length of the dipole, is approximately .lambda..sup.2 /8, or in this example 12.5 times.

Dipole antennas have been employed for the electromagnetic spectrum all the way from long wave radio down through the television range into the microwave and millimeter wave spectrum.

To date, however, no practical method has been suggested for making controlled use of dipole antennae in the visible or adjacent portions of the spectrum.

According to the present invention, visible-light dipoles are readily prepared. Methods and devices for readily putting them to controlled, practical use are described.

Because their effective cross section is much greater than the physical cross section, the dipolar particles may be very sparsely distributed in space. The dipolar particles maybe sufficiently far apart from each other so as to have no physical interreaction. Each dipolar particle may act independently of the other.

From the known resistivity of metals the ideal length to width ratio of absorbing or reflecting dipole particles of various materials have been computed.

FIG. 12 shows a cell in the XY plane in which the dipole particles 50 are aligned in the X direction. Light transmitted along the Z axis into the surface emerges from the other side plane polarized with the electric vector E.sub.y in the ZY plane. Reflected light, if any, is plane polarized with the electric vector E.sub.x parallel to the ZX plane. Reflected light is polarized and scattered.

A mathematical study applying the principles of the electromagnetic radiation-antenna theory to dipolar conducting particles was made. In accordance with this theory, the flow of incident radiation power into the antenna is defined by a space cross section A.sub.s.

Consider a dipole antenna of a length .lambda./2 for receiving electromagnetic radiation tuned to absorb a maximum of incident radiation of wavelength .lambda.. For maximum absorption this antenna has a distributed resistance R.sub.a equal to its radiation resistance R.sub.b. This antenna intercepts incident radiation-power from an equivalent area approximately (.lambda..sup.2 /4), for dipoles whose long direction is in the direction of electric vector of the incident polarized light. Half this radiation power is absorbed and converted to heat, and the other half is re-radiated as scattered light.

The space cross section A.sub.s varies with .beta., defined as R.sub.a /R.sub.b. The equivalent cross section of a half-wave dipole antenna is greater than its actual physical dimensions. This important effect applies to suspensions of dipoles in a transparent medium.

The ratio "a" of antenna length to diameter determines the width of the wavelength band that will interact with the antenna. The spectral interaction bands of a conducting half-wave dipole particle become narrower as the length to diameter ratios increase. The interaction is for a wavelength band centered about wavelength .lambda..

If the antenna is relatively thick; i.e., the ratio of length to thickness is about 5 to 15, the antenna is capable of absorbing a broad band of incident energy.

If the antenna is relatively thin; i.e., the ratio of length to thickness is large, say 30 or more, the antenna is tuned to accept a narrow band of frequencies.

For most dipole devices, such as windows or shutters, a wide band response is usually required, and there may be used a range of lengths from about 1000A to 2500A, having a length to width ratio a = (L/d) of between 10 and 25.

Metal of lower resistivity results in a greater ratio "a" for the dipole antenna rod, and a narrower absorption band for radiation absorbed or reflected.

For rod shapped conducting dipole particles in random orientation, the mass concentration to substantially attenuate transmitted light is less than that otherwise required using plates of the same size, by a factor of l/a. The volume of the dipole is L.sup.3 /a.sup. 2, while the volume of a plate is L.sup.3 /a. (l/a) equals 2 to 10 percent for 50> a> 10.

The physical cross sections of rod and plate are respectively in the ratio L.sup.2 /a and L.sup.2 ; hence, the rod shape has a smaller physical cross section than the plate by a factor of l/a.

When the dipoles particles in a suspension are arranged with their length normal to the light rays and parallel to the electric vector of the light, they absorb light from an area of approximately 10 to 100 times their physical cross section, or, 5 to 50 times their physical cross section for ordinary light incident on randomly directed dipoles.

However, when these dipole particles are oriented by the applied electric field, with their length parallel to the light rays, the physical cross section presented to the rays is decreased by a factor of "a" or from about 5 to 100 times. As the dipoles are oriented, the space cross section and absorbance diminish faster than the physical cross section of the dipole particles. When the dipole rods are randomly oriented in a suspending layer, the absorbance is almost complete.

Thus, for a dipole layer containing long rod shaped dipole particles operating in the random-parallel, the absorption in the open (parallel) state is negligible, while absorption in the closed state is almost complete. The light scattered from the rod shaped particles is negligible because the particle diameter is less than (.lambda./30).

When the suspended particles are metal conductors, electromagnetic radiation antenna theory as used in microwave technology, is applied to the suspended dipole particles which are considered as a plurality of antenna elements. By selecting the length, thickness, resistivity, and angular position of these dipole particles, the transmittance, absorption, reflectance, peak wavelength and polarization of the light interacting with such a dipole suspension can be controlled.

In considering a half-wave dipole suspended in a fluid medium the wavelength of light in the suspending medium must be used. This wavelength is inversely proportional to the index of refraction of the medium. Thus, .lambda./2 or a half-wave dipole for light radiation in vacuum or air has a length of .mu./2n in a suspending medium with an index of refraction of n. For example, a half-wave dipole for light of wavelength 5600 A as measured in air has a length of 2800 A, but has a length of 1867A in a suspending medium with an index of refraction of 1.5.

The dipole suspensions disclosed and described herein comprise suspensions of herapathite crystals and metal crystals rods.

Chemically, herapathite is iodoquinine-sulphate which forms long blade shaped hexagonal crystals, having a length to width ratio of about 25 and a thickness of about 1/10 the width, and which strongly polarize transmitted light.

Herapathite contains parallel polyiodide chains of various lengths within the crystal structure held in a dielectric crystalline cage, or cathrate crystal structure. Electron transfer occurs along the polyiodide chains, which act as conductive dipoles rigidly mounted in parallel arrays within a dielectric matrix. Light transmitted through a herapathite crystal has the greatest polarization in the visible, over a wide range.

Herapathite polyiodide chains react to light as though they were groups of metallic dipoles held in parallel array separated by an insulating structure.

Because the dipoles in herapathite are in the form of rigid parallel arrays held within a bulky dielectric crystal matrix, the idealized theory presented for isolated metal dipoles is not exact, and an empirical approach is employed.

Table of symbols

the cgs unit system is used throughout, except as noted.

A = real cross-sectional area of the dipole antenna = Ld.

A.sub.s = A.sub.a + A.sub.b = the equivalent area of space from which the dipole antenna absorbs and/or reflects incident radiation.

A.sub.a = cross section for absorbed power of antenna

A.sub.b = cross section for re-radiated power of antenna

a = ratio of dipole length to width = L/d

a = [1n (2 L/d) - 0.80]

C.degree. = concentration of dipole particles in suspending fluid in proportion by mass.

D = optical density, or log.sub.10 [100/percent transmittance]

D = minimum optical density obtained by applying an electric field E (where E< E, and D <D).

D = minimum optical density, corresponding to peak transmittance T, corresponding to that peak electric field intensity E required to achieve maximum alignment of the dipoles.

d = diameter of dipole considered as a square rod.

d.sub.1 = thickness of a layer of the dipole suspension

d.sub.p = mean distance between dipole particles centers

E = electric field intensity

E' = electric field intensity which just causes coagulation

E.sub.r = electric field intensity which just balances relaxation due to Brownian motion

f = frequency of electric field

g = A.sub.s /A = ratio of space cross section to physical cross section of dipole

l = Current induced by radiation field in dipole antennae

k = Boltzmann's constant = 1.38 .times. 10.sup..sup.-23 J .degree.K.sup..sup.-1 (mks -.degree.K)

K = randomizing constant = .pi./18k = 1.26 .times. 10.sup.22 .degree.K J.sup..sup.-1 (mks - .degree.K)

L = length of dipole particle

M = mass of dipole particles per unit area of layer of a suspension of thickness d.sub.1 ; M=Cd.sub.1

M.sub.1 = mass of liquid volume per unit area of dipole suspension of thickness d.sub.1

m.sub.p = mass of dipole particle

N = number of dipole particles per unit area of suspension

n = index of refraction of the suspending fluid

p = number of dipoles aligning per unit time in a unit volume

q = electrodichroic ratio = D.sub.r /D, corresponding to E.

R.sub.a = resistance of dipole particle (absorption)

R.sub.b = radiation resistance of the dipole antenna (reflection)

S = sensitivity, defined as (1/M) (.DELTA.q/.DELTA.E) = .sigma./M

T = transmittance

V.sub.p = volume of one dipole particle of square cross section = Ld.sup.2

V.sub.1 = volume of fluid occupied by one dipole particle (.alpha.L)3 (.alpha.L)3

Greek Symbols

.delta..sub.1 = density of the fluid in which the dipole is suspended

.delta..sub.p = density of the dipole particle

.epsilon..sub.o = dielectric constant for free space

.epsilon. = extinction factor for dipoles

.alpha. = d.sub.p /L

.beta. = r.sub.a /R.sub.b

.gamma. = A.sub.s /.lambda..sup.2 = space across section factor

.sigma. = .DELTA.q/.DELTA.E, electrodichroic response; or the change in electrodichroic ratio with respect to the change in the electric field intensity

.eta. = viscosity of the suspending liquid

.theta. = absolute temperature in .degree.K

.lambda. = wavelength of incident radiation in vacuum

.rho. = resistivity of the material comprising the dipole particle

.tau..sub.B = relaxation time factor; the characteristic time for an aligned suspension to randomize due to Brownian motion, in which time the optical density increases by (D.sub.r - D)/e

.tau. = Alignment time factor; a characteristic time for a random suspension to orient due to the applied electric field intensity E; in which time the optical density decreases by (D.sub.r - D)/e

Subscripts

a = for absorbed radiation

b = for re-radiated radiation

r = random or most opaque state

x = in the x direction or normal to the light ray

z = in the z direction or parallel to the light ray

rx = in which the orientation of the dipole changes from random to normal to the light ray

rz = in which the orientation of the dipole changes from random to parallel to the light ray

ms = milliseconds = 10.sup..sup.-3 sec

.mu.s = microseconds = 10.sup..sup.-6 sec

A = angstrom = 10.sup..sup.-10 m

mp = millipoise = 10.sup..sup.-3 poise (cgs viscosity)

Superscripts

= maximum value

= a value, below maximum

THEORY OF ISOLATED DIPOLES

Antenna Cross Section

A dipole antenna absorbs, reflects and transmits electromagnetic radiation over an area of space termed its cross section.

FIG. 48 shows a half-wave antenna with its long axis parallel to the Y axis. For Cases I and II incident light is directed along the Z axis, and is polarized with the E vector parallel to the Y axis.

FIG. 49 shows the reflection-scattering cross section A.sub.b and the absorption cross section A.sub.a as a function of the ratio .beta. = R.sub.a /R.sub.b, or the ratio of the ohmic self-resistance to its radiation resistance of the antenna. As .beta. increases, the reradiation cross section A.sub.b becomes small compared to the absorption cross section A.sub.a.

Case I - The radiation resistance of the antenna is equal to its ohmic resistance; R.sub.a = R.sub.b. The re-radiated power is equal to the absorbed power. The absorbed power is a maximum. The effective cross section for the absorbed power is the area ABCD; or A.sub.a = .lambda..sup.2 /8. The cross section for the power re-radiated as scattered light by the antenna, is also area ABCD: A.sub.b = .lambda..sup.2 /8 = 1/2 (.lambda./2).sup.2 For a half wave dipole antenna in a medium of index of refraction n, L = .lambda./2n and A.sub.a = A.sub.b = 1/2 L.sup.2

The total cross section for the absorbed power and the re-radiated scattered power is twice the area ABCD; or the area EFGH; In a random suspension the re-radiated power from a single dipole is absorbed by other dipoles. In this case, substantially all the power is absorbed, hence:

A.sub.s = A.sub.a + A.sub.b = L.sup.2 (.gamma. = 1) (1)

case II - The antenna resistance R.sub.a is very small; the power incident upon the antenna cross section is all reradiated, or reflection-scattered; and the area is IJKL:

A.sub.s = 0 + A.sub.b = .lambda..sup.2 /2 = 2L.sup.2 (.gamma. = 2). (2)

cases III and IV are the same as Cases I and II respectively, except that (a) the light is ordinary light, and (b) the dipoles have a random orientation in space. The absorption factor .gamma. is decreased by a factor of 1/2 by (a), and another 1/2 by (b) so that:

Case III - Same as Case I, except for (a) and (b)

A.sub.r = 1/2 .times. 1/2 .times. L.sup.2 = (1/4) L.sup.2 (.gamma..sub.r = 1/4) (3)

Case IV - Same as Case II except for (a) and (b)

A.sub.r = 1/2 .times. 1/2 .times. 2L.sup.2 = (1/2)L.sup.2 (.gamma..sub.r = 1/2) (4)

Generalizing in terms of a cross section factor y, the space cross sectional area A.sub.s for a half-wave dipole of length L = .lambda./2n suspended in a medium with index of refraction n, is:

A.sub.s = .gamma.L.sup.2 (5)

for a random dipole suspension and ordinary light A.sub.s = A.sub.r and .gamma. = .gamma..sub.r ; and for an aligned suspension with the dipoles aligned parallel to the Z axis A.sub.s = A.sub.z ; .gamma. = .gamma..sub.z, and L is decreased by the length to width ratio a = L/d. Assuming a square cross section for the antenna, it's physical cross section A = Ld when normal to the ray; that is:

A = L.sup.2 /a (normal); or L.sup.2 /2a (random) (6)

In Case III, for ordinary light, and with the dipoles at random directions, the ratio g of the space cross section A.sub.s for which a dipole antenna is absorbing and/or reflecting ordinary light, to the actual physical cross section of the dipole antennae A, is found from (3) and (6):

g = (A.sub.r /A) = (1/4)L.sup.2 / (L.sup.2 /2a) = a/2 (7)

As example, with a thick antenna a = 4 and g = 2; and with a thin antenna, a = 100 and g = 50.

This discussion does not imply that there is no power transmitted through the cross section. For a suspension of a large number of dipole antennae, the absorption cross section may be replaced by the absorption factor in accordance with Beer's law and the scattering reflection factor. The absorption cross section is variable and depends upon dipole orientation, which is a function of the applied electric field intensity. As a consequence, in a dipolar medium, the absorption, reflection, scattering and transmittance is a function of the electric field intensity, time and other variables. These relationships are derived and discussed in this paper.

The cross section .gamma. is evaluated empirically for a suspension of rod-shaped conductive particles at random or aligned by the electric field. In the case of a rod-shaped conductive particle aligned parallel to the light path, where the diameter of the rod is less than about .lambda./30, the light scattering or re-radiated component is negligible. The signal pickup by an antenna aligned parallel to the light path is a minimum.

For large particles the absorption cross section is twice the physical cross section, but this does not exactly apply to very small particles. Explicit mathematical physics relationships for aggregates of conductive rodlike submicron particle suspensions is not known. The analysis herein empirically combines mathematical-physics theory with experimental results to characterize the observed properties of a rod-shaped dipolar suspension in an electrical field.

Radiation Resistance

The radiation resistance R.sub.b corresponds to the reradiated power 1.sup.2 R.sub.b.sup.2 from the dipole antennae.

The radiation resistance R.sub.b of a short dipole with uniform current is:

R.sub.b = 80.pi. .sup.2 (L/.lambda.).sup.2 (8)

for a .lambda./2 dipole, thus:

R.sub.b = 20.pi..sup.2 = 197 ohms (9)

For a .lambda./2 dipole with a center resistance, and with a sinusoidal current distribution, for peak absorbed power:

R.sub.b = 73 ohms, and for .epsilon. = 1 R.sub.a = 73 ohms

The radiation resistance of an isolated dipole with distributed resistance in a medium having an index of refraction n; and the radiation resistance of a dipole in random infinite arrays in which interaction occurs, are not explicitly known.

The distributed ohmic resistance of the dipole can be calculated from its resistivity .rho., cross sectional area d.sup.2 and length:

R.sub.a = .rho. L/d.sup.2 (10)

THEORY OF DIPOLE SUSPENSIONS

Relationships of the Physical Variables

The relationships between the physical variables are found from the space cross section of a rod-shaped dipole particle.

Let the dipoles be uniformly spread on a plane of unit area in a suspending medium with their cross sections contiguous, to substantially absorb or reflect incident light; whereupon the number of dipole particles per unit area are:

N = 1/A.sub.r = 1/.gamma..sub.r (.lambda./2n).sup.2 = 4n.sup.2 /.gamma..sub.r .lambda..sup.2 (11)

Each dipole particle occupies a cubic volume V.sub.1 in the medium, and a cube of volume V.sub.1 has a side .alpha.L; from which:

V.sub.1 = (.alpha.L).sup.3 = (.alpha..lambda./2n).sup.3 (12)

Total volume per unit area of the medium in which the dipoles are suspended is:

V = N V.sub.1 = 1.sup.2 . d.sub.1 (13)

The layer thickness d.sub.1 is from (11), (12) and (13):

d.sub.1 = .alpha..sup.3 .lambda./2 .gamma..sub.r n (14)

The mass of a half wave dipole is:

m.sub.p = .delta..sub.p L d.sup.2 = .delta..sub.p .lambda..sup.3 /8a.sup.2 n.sup.3 (15)

The mass of dipoles per unit area is:

M = m.sub.p N = .delta..sub.p .lambda. /2n.delta..sub.r a.sup.2 (16)

The mass concentration C, or mass of dipoles per unit mass of the suepending medium, may be obtained by the mass of the dipole and the mass of the medium in which a single dipole is suspended, for .alpha.>> 1:

C = .delta..sub.p V.sub.p /.delta..sub.1 V.sub.1 = (.delta..sub.p L.sup.3 /a.sup.2) /.delta..sub.1 .alpha..sup.3 L.sup.3 = (.delta..sub.p /.delta..sub.1)/a.sup.2 .delta..sup.3 (17)

EXAMPLE 1

Given: .alpha. = 5; .lambda. = 5 .times. 10.sup..sup.+5 cm; .gamma..sub.r = 1/4; n = 1.5

Find: Thickness of dipole layer d.sub.1

Answer:

From (14): d.sub.1 = 5.sup.3 .times. 5 .times. 10.sup..sup.-5 /(3/4) = 0.083 cm

EXAMPLE 2

Given: .delta..sub.p = 10; .lambda. = 5 .times. 10.sup..sup.-5 cm; .gamma..sub.r = 1/4; a = 10; n = 1.5

Find: Mass per unit area of dipoles M, for a = 10, and a = 26

Answer:

From (16): the mass per unit area of dipoles is:

for a = 10:

M = 10 .times. 5 .times. 10.sup..sup.-5 /2 .times. 1.5 .times. (1/4) 10.sup.2

m = 6.67 .times. 10.sup..sup.-6 gms/cm.sup.2,

For a = 26.sub.2 :

M .congruent. 1 .mu.g/cm.sup.2

EXAMPLE 3

Given: A silver dipole .delta..sub.p - 10.5 is suspended in a fluid of density .delta..sub.1 = 1. The silver dipole has a length to width ratio a = 26, and the interparticle spacing to particle length ratio is .alpha. = 5.

Find: Concentration C

Answer:

From (17):

C = (10.5/1) /26.sup.2 .times. 5.sup.3

c = 1.24 .times. 10.sup..sup.-4 gm silver/gm of fluid.

The Electrodichroic Ratio

The random-parallel electrodichroic ratio is defined for a dipole suspension layer which has a minimum (closed) transmittance when the dipole particles have a random orientation, and a maximum (open) transmittance when the dipole particles are partially or completely oriented parallel to the light path normal to the plane of a dipole suspension layer by an electric or other force field. For zero electric field the dipole particles are oriented at random. As the applied electric field intensity is increased from zero, the dipole particles become more completely aligned parallel to the field.

Beer's law, in which M = Cd.sub.1, may be utilized for the closed state:

T.sub.r = e .sup..sup.-.sup..epsilon. .sup.M (18)

beer's law may be utilized for the open state:

T.sub.r = e.sup.-.sup..epsilon. .sup.M

Using the definition of the optical density, equations (18) and (19) are now expressed in terms of the corresponding optical densities:

D.sub.r = log.sub.10 (1/T.sub.r) = (log.sub.10 .epsilon.) .epsilon..sub.r M = 0.434 .epsilon..sub.r M (20) D.sub.z = log.sub.10 (1/T.sub.z) = (log.sub.10 .epsilon.) .epsilon..sub .z M = 0.434 .epsilon..sub .z M (21)

by definition, and from (20) and (21), the random-parallel electrodichroic ratio is:

q.sub.rz = D.sub.r /D.sub.z = .epsilon..sub.r /.epsilon..sub.z (22)

In a similar manner, other electrodichroic ratios have been defined.

The electrodichroic ratio q.sub.rz is measure of the effectiveness of the dipole layer as a light control medium, and is particularly useful because it is parameter which is independent of the dipole concentration and layer thickness.

The significance of the electrodichroic ratio will be clear from the following discussion:

EXAMPLE 4

An electro-optic dipole shutter which has an electric field along the Z axis, transmits 63 percent of the incident light in the maximum transparent state, and 0.1 percent of the incident light in the random opaque state. The corresponding optical densities are:

Transparent: D = log.sub.10 (100/63) = 0.2

Opaque: D.sub.r = log.sub.10 (100/0.1) = 3

The parallel electrodichroic ratio is:

q.sub.rz = D.sub.r /D.sub.z = 3/0.2 = 15.0

A q.sub.rz = 15 is highly satisfactory for most applications, such as electrodichroic photographic shutters, and variable transmittance windows.

FIG. 50 shows maximum percent transmission versus the electrodichroic ratio for the minimum percent transmission: 0.01, 0.1 and 1. The corresponding maximum percent transmission may be read off directly.

For particles aligned parallel to the Z axis the physical cross section is (L/a).sup.2. However, for a>3, data from the literature for conducting spheres of radius r, shows that: the scattering cross section A.sub.b is negligibly small; only the absorption cross section A.sub.a need be considered; and this relationship holds:

A.sub.a = .pi. r.sup.2 . 2 .pi.r/.lambda. (23)

The term .pi.r.sup.2 is the physical cross section and the term (2.pi. r/.lambda.) is the factor .gamma. for the case of the sphere. For an equivalent square dipole rod r = L/2a and .lambda. = 2nL. On this basis, for a dipole rod aligned parallel to the Z axis:

.gamma..sub.z = .pi./2na .congruent. 1/a (24)

Hence A.sub.z = .gamma..sub.z (1/a).sup.2 .congruent. (1/a.sup.3) L.sup.2 (25)

the random-parallel electrodichroic ratio may be expressed in terms of the absorption factors according to (22) and in terms of the absorption cross sections, as follows:

q.sub.rz = .epsilon..sub.r /.epsilon..sub.z = A.sub.r /A.sub.z (26)

From (25) and (26):

q.sub.rz = .gamma..sub.r L.sup.2 / (1/a.sup.3)L.sup.2 .congruent. .gamma..sub.r a.sup.3 (27)

This expression may be used to empirically determine the cross section .gamma..sub.r from measurements of the electrodichroic ratio and the particle dimensions.

EXAMPLE 5

Given: A herapathite dipole suspension for which

q.sub.rz = 15, and a = 5

Find: .gamma..sub.r

Answer:

From (27):

.gamma..sub.r .congruent. q.sub.rz /a.sup.3 = 15/125 .congruent. 1/8

EXAMPLE 6

Given: q.sub.rz = 100 and T.sub.z = 0.96, then D.sub.z = 0.04 and D.sub.r = 100 .times. 0.04 or T.sub.r = 0.01 percent. This dipole suspension will change from practically transparent to practically opaque. For this data and .gamma..sub.r = 1/10, according to (27), a = 10.

EXAMPLE 7

Given: an electrodichroic ratio of 6 at (1/e) (D -D.sub.r) for an applied electric field intensity of E.sub.z = 10.sup.3 volts/cm

Find: The electrodichroic response

Answer:

.sigma..sub.rz = (.DELTA.q/.DELTA.E) = (6-1)/10.sup.3

.sigma..sub.rz = 5 .times. 10.sup..sup.-3 .DELTA. q (volts/cm).sup..sup.-1

EXAMPLE 8

Given: In example (7 ) a dipole mass of

M = 1.0 .times. 10.sup..sup.-6 gm/cm.sup.2

Find: the electrodichroic sensitivity S.sub.rz

S.sub.rz = .sigma..sub.rz /M = (5 .times. 10.sup..sup.-3 /1 .times. 10.sup..sup.-6)

S.sub.rz = 5000.DELTA.q cm.sup.3 /volt gm.

Dimension Ratios and the Resistivity of Materials

If .rho. is the resistivity of a conductor, then the resistance R.sub.a of a ;conductor of length L and cross-sectional area A is found from (10); into which substitute the length of a half-wave dipole in a suspensing medium of index of refraction n, L = .lambda./2n.

Then, for a dipole rod having a square section:

R.sub.a = .rho. (.lambda./2n) / (.lambda./2na).sup.2 = 2n .rho. a.sup.2 /.lambda. (28)

Solving (27) for a:

a = .sqroot.R.sub.a .lambda. /2n.rho. (29)

Known resistivities .rho. for metals at 20.degree.C are: silver 1.6 .times. 10.sup..sup.-6 ohm cm; gold 2.4 .times. 10.sup..sup.-6 ohm cm; chromium 2.6 .times. 10.sup..sup.-6 ohm cm.

EXAMPLE 9

An ideal absorbing dipole is assumed to have a distributed resistance R.sub.a of about 80 ohms, and an ideal reflecting dipole is assumed to have a distributed resistance R.sub.a of 8 ohms.

Given: Values of 1.5 for n and 5 .times. 10.sup..sup.-5 cm for .lambda., and the resistivities .rho. for silver and chromium given above.

Find: the length to width ratio "a" for an absorbing and a reflecting half-wave dipole.

Answer: Substituting these values, it is found for these metals. From equation (29) the length to width ratio "a" was 20 and 30 for an absorbing dipole; and 6 and 10 for a reflecting half-wave dipole.

Relaxation Time Factor

After the dipoles are aligned by the applied electric field, the electric field is suddenly turned off. Brownian molecular impacts produce torques which cause the dipoles to turn to random directions. From the literature in a related field an equation was derived for .tau. .sub.B the relaxation time factor:

.tau..sub.B = (.pi./18k) (.eta.L.sup.3 /T{ [ln (21/d)] - 0.80 }) (30)

Setting K = .pi./18k; and a.sub.o = ln(21/d) - 0.80:

.tau..sub.B = K (.eta./T) (L.sup.3 /a.sub.o) (31)

For example for (L/d) = 10, a.sub.o = 2.20; (1/d) = 20, a.sub.o = 2.89; and (L/d) = 30, a.sub.o = 3.29.

Evaluating (31) by expressing L in A, .tau..sub.B in .mu.sec, and in millipoise or .eta..sub.mp and T in degrees Kelvin:

.tau..sub.B.sub..mu.s = 1.26 .times. 10.sup..sup.-6.sub..sub..eta.mp LA.sup.3 /T.sub.o a.sub.o (32)

Since the optimum length L is approximately .lambda./3, for 5650A, L = 1800A. For .eta. = 10 millipoise, this results in a relaxation time factor of .tau..sub.B = 127 microseconds.

By way of example, the viscosities for various liquids at 30.degree. C in millipoise are: water, 10; hexane 2.9; Toluene, 5.8; Dioctyl adipate 129.; Glycerine 6240.

Using Equation (32) it is shown that a relaxation factor of .tau..sub.B = 6.mu.s is obtained with a low viscosity fluid such as Hexane (2.9) millipoise and a dipole length L = 1000A.

A Herapathite dipole suspension in a fluid with a viscosity of 100 millipoise and a dipole length of 2000A has a relaxation time of about three milliseconds.

A dipole suspension in a fluid with a viscosity of 1000 millipoise and a dipole length L = 4000A has a relaxation time of about 0.24 seconds.

Thus, the relaxation time may be controller over a wide range.

Alignment and Relaxation vs. Frequency

A dipole suspension disorients by Brownian motion from the aligned state to the random state with a relaxation time .tau..sub.B. An applied alternating voltage has a critical frequency f.sub.c, in which a half period is equal to the Brownian relaxation time .tau..sub.B ; that is:

f.sub.c = 1/2 .tau..sub.B (33)

if the frequency of the applied alternating voltage field is equal to or a little less than f.sub.c ; then the orientation of dipole rods in the suspension will oscillate between partial alignment and randomization; causing an electro-optic modulation of the dipole suspension. Better alignment is obtained at greater frequencies, f, for which half period 1/2 f is much smaller than .tau..sub.B. In this case less relaxation occurs during field reversals. Hence, to assure the substantial maintenance of alignment, and no apparent modulation of transmitted light, the frequency should greatly exceed the critical frequency f.sub.c ; that is:

f>>f.sub.c (34) EXAMPLE 10 The relaxation factor is .tau..sub.B = 1 millisecond, then from (34) f>>1/2 .times. 10.sup..sup.-3 ; that is, f>> 500 hz; or f = 5000 Hz.

In the case of dipole suspensions having a greater fluid viscosity, or having longer dipoles, .tau..sub.B is greater and a smaller frequency can be used without apparent modulation of transmitted light. When the relaxation factor .tau..sub.B is smaller, the critican frequency increases.

EXAMPLE 11

For .eta. = 1000 millipoise, and L = 3000A, the Brownian relaxation time .tau..sub.B = 0.1 sec; and the critical frequency f.sub.c >> 5 hz; hence a 60 cycle frequency will suffice to align this dipole suspension without apparent modulation of the transmitted light.

FIG. 54 shows for an herapathite suspension, an experimental plot of the parallel electrodichroic ratio q.sub.rz versus frequency f, for various electric field intensities E.sub.z. In general, for constant E.sub.z, the q.sub.rz increases rapidly as the frequency increases from 0.5 KHz to 3KHz, less rapidly from 3KHz to 10KHz, and substantially reaches an assymptotic value above 30KHz.

The ions in the dipolar suspension have a mobility expressed in cm/sec per volt/cm, or cm.sup.2 /volt-sec. The ion mobility in the dipole cell was calculated as follows:

The ions are assumed to travel during the half-cycle a distance equal to the length of a dipole 0.5 .times. 10.sup..sup.-4 cms in a field of 1.5 Kv/cm. The length of the half-wave of light is taken as 0.5 .times. 10.sup..sup.-4 cm. At a frequency of 10kc, the half-cycle time is 0.5 .times. 10.sup..sup.-4 seconds. The rms ion velocity is then 0.5 .times. 10.sup..sup.-4 /0.5 .times. 10.sup..sup.-4 or 1 cm/sec at 1500 volt/cm. Hence the mobility of the ion in the dipole suspension must be 1/1500 = 6.7 .times. 10.sup..sup.-4 cm.sup.2 /volt-sec. Experimentally, ion mobilities reported in the literature were found to be: 6.9 .times. 10.sup..sup.-4 for Cl, 18.1 for OH.sup.- , and 32.0 for H.sup.+ in cm.sup.2 /volt sec.

For a herapathite suspension the ion is probably I.sup.-, which should have a mobility approximating that of Cl.sup.-. The calculations of ion mobility in the dipole suspension are in good agreement with the experimental values of ion mobility given in the literature.

Effect of Ions on Dipole Alignment

In a strong electric field at lower frequencies, the ions migrate a greater distance in the dipole layer, periodically concentrating at the ends of the dipole particles, and at the surfaces of the dipole layer. These charge concentrations counteract the applied electric field, and the dipole alignment is decreased. However, as the frequency increases, the ions oscillate only a short distance about a main position, the ionic shielding effect is diminished or eliminated, and the applied electric field more effectively produces alignment of the dipoles. A low voltage high frequency field will align as well as a high voltage low frequency field.

When ions are present, a step DC voltage momentarily partly aligns the dipole particles. An ionic shielding layer is soon set up near the induced charges at the end of each dipole, counteracting the applied field. The dipole particles then start to disalign under randomizing molecular impacts due to Brownian motion. For the step DC voltage to cause a substantial initial alignment, the rise time of the applied step voltage must be less than 10 microseconds.

With an AC electric field having a frequency in excess of 1 KHz the dipoles align during each half-cycle. Greater alignment is achieved with a smaller electric field as the frequency is increased, substantially reaching an assymptotic value of frequencies greater than 30 KHz. At frequencies greater than 10KHz, the disalignment due to Brownian motion during field reversals is usually small compared to the large alignment effect occuring during each half-cycle.

With DC or low frequency AC electric fields, positive and negative ions may migrate to opposite surfaces of the dipole layer, or to the ends of the induced dipoles, decreasing or cancelling the aligning electric field within the dipole layer. This field-neutralizing effect depends upon the presence of ions in the dipole suspension.

With non-ionic or slightly ionic suspensions such as a DC or low frequency AC field may be used and the transparent conductive films may be in direct contact with the suspension layer surfaces to pick up a small current and prevent charge build up.

Where the dipole layer contains a substantial concentration of ions, direct contact with the transparent conductors may cause their electrolysis and destruction. Consequently, a herapathite suspension which contains a concentration of ions, requires a protective transparent insulating layer over the transparent conductive layer, (see FIG. 3).

With an electric field having a frequency of 5 KHz or more ion migration and separation of oppositely charged ions is diminished and the field-neutralizing effect substantially eliminated.

Using a herapathite dipole suspension, for frequencies up to a few hundred Hz, the transmittance increase of the cell is small at 1-3.5 Kv/cm but as the frequency increases in the 1 to 30 KHz range, the transmittance increases substantially and above 30 KHz there is little further increase in transmittance.

Herapathite dipole suspensions are particularly sensitive to electrolytic destruction yet, when placed between a dipole cell in which the transparent conducting electrodes are covered with thin transparent protective transparent layer; these suspensions are stable.

ELECTRIC ALIGNMENT TIME FACTOR

A gated alternating pulse having an rms electric field intensity E is applied to a random dipole suspension. The electric alignment time factor .tau. is a function of the applied electric field intensity E.

The Brownian motion impacts result in randomizing-torques which disorient the dipoles. The average randomizing-torque corresponds to an rms electric field intensity E.sub.r. Before alignment of the dipoles can occur, the torque on the dipoles due to the applied electric field intensity E must exceed the randomizing torque E.sub.r.

The alignment time factor .tau. due to the applied electric field E is proportional to the relaxation time factor .tau..sub.B and to the ratio of these torques.

.tau.=.tau..sub.B [E.sub.r / (E - E.sub.r) ] = .tau..sub.B {1/ [(E/E.sub.r) - 1 ]} (35)

For E >>E.sub.r Equation (35) becomes

E .tau. = E.sub.r .tau..sub.B (36)

combining equations (31) and (36)

.tau. = (K.sub..eta.L.sup.3 /Ta.sub.o) (E.sub.r /E) (37)

equation (36) may be used to evaluate E.sub.r from measurements of .tau., .tau..sub.B and E.

For an herapathite suspension, the ratio (.tau..sub.B /.tau.) usually measures 20 to 32, for an electric intensity of about 3 Kv/cm at 10 KHz.

EXAMPLE 12

Given: An herapathite suspension layer in a cell had a relaxation time .tau..sub.B = 8 ms and an alignment time .tau. = 0.25 ms at an applied electric field of 3.2 Kv/cm rms at 30 KHz.

Find: E.sub.r

Answer: r

From (36):

E.sub.r = E(.tau./.tau..sub.B) = 3.2 (0.25/8)

E.sub.r .congruent. 0.1 Kv/cm; or 100 volts/cm.

It may seem strange that E.sub.r is so large, compared to the small Brownian motion force, usually associated with molecular impact having energies in the fractional voltage range. However, the torques are exerted by an applied electric field on a dipole rod only about 2 .times. 10.sup.-.sup.5 cm long. For 100 volts/cm, this corresponds to an aligning potential difference V = EL = 100 .times. 2 .times. 10.sup.-.sup.5 = 2 .times. 10.sup.-.sup.3 volts applied to the dipole rod to balance the disaligning effect due to Brownian motion.

A rapid alignment time (say 25 s) is obtained by applying a pulse comprising large electric field (say 30 kv/cm) (less than breakdown, about 400 kv/cm) for a short time (say 25 s). After this alignment the electric field intensity is decreased to about 2.0 kv/cm to maintain alignment without coagulation.

ALIGNMENT OF DIPOLES IN AN ELECTRIC FIELD

Insight into the alignment process may be had from the following consideration. An applied DC or AC electric field induces opposite charges on each end of the dipole. The opposite induced electrical charges on the neighboring ends of the particles produces torques tending to align the particles in the direction of the electric field. It might be thought that when the electric field is applied, all the randomly directed dipolar particles start to align simultaneously at a rate determined only by their initial direction and other factors which are the same for all dipoles of the same size and shape. However, while alignment occurs in this manner, a greater factor is that the ends of adjacent induced dipole particles in closest proximity exert attractive forces on each other, producing greater torques, and more rapid alignment. The most closely proximate particles are aligned most quickly, while other dipoles of the suspension remain almost randomly directed. Another group of the most closely proximate particles then align. The process continues until all the dipoles are aligned. The alignment principle may now be set forth:

Groups of the most closely proximate particles align first, leaving the remaining particles more or less random in direction. Let p be the proportion of dipole particles aligned parallel to the electric field at time t. Then the proportion of dipole particles aligning per unit of time (dp/dt), varies directly as (1-p) the proportion of remaining randomized dipole particles:

(dp/dt) = (1/.tau.) (1-p) (38)

where .tau. is the electric alignment time factor; which is a constant under given conditions. Integrating (38) and evaluating for p = O, when t = 0, there is obtained:

p = 1-e.sup.-.sup.t/.sup..tau. (39)

The absorption factor of a layer of suspension of which a proportion p of dipole rods is aligned and a proportion of the dipole rods (1-p) is in the random state, is:

.epsilon..sub.z p +.epsilon..sub.r (1-p) = .epsilon..sub.r + (.epsilon..sub.r -.epsilon..sub.z)p (40)

Use of the combined absorption factor (40) in Beer's law determines the transmittance: ##SPC1##

Characteristic Electro-Optic Equations

FIG. 51 shows optical density vs. time for a voltage pulse Ez applied at time t = 0. Curve 1 shows the optical density exponentially decreasing to time t.sub.1, reaching the assymptotic value D. At the time t.sub.1 the voltage is cut off, and the dipole particles relax, the optical density then increases exponentially along the curve 2. For a strong electric field the electric alignment time factor.tau.is very small compared to the Brownian relaxation time factor .tau..sub.B.

In FIG. 51 Curve 1 is empirically expressed by the following formula:

D.sub.z = D.sub.z + (D.sub.r - D.sub.z) e.sup..sup.-t/.sup..tau. (47)

Equation (47) may be put in the form of Equation (45) which was theoretically derived in Section on alignment of dipoles in an electric field.

By the relationship between optical density D and the absorption factor .epsilon. :

D.sub.r = 0.434 M.sub..epsilon. (48)

d.sub.z = 0.434 M.sub..epsilon.

D.sub.z = 0.434 M.sub..epsilon.

Substituting the relations (48) into (47) and using the definition of optical density equation (45) follows as before.

As a result of the applied step voltage, at a sufficiently long time t.sub.1 >> .tau..sub.B, the optical density is presumed to have reached the minimum D.sub.z. The applied voltage is suddenly turned off at time t.sub.1. The dipole rod suspension then starts to randomize, represented by Curve 2 of FIG. 51, and by the following empirical expression:

D = D.sub.r - (D.sub.r - D.sub.z) e.sup..sup.-t/.sup..tau. (49)

FIG. 52 shows a plot of optical density D.sub.z vs. applied electric field intensity E.sub.z. There is no change in the optical density until the electric field intensity exceeds a threshold value E.sub.r. However, because of the statistical spread in the molecular impact velocities the step due to E.sub.r is not abrupt, but curved. E.sub.r may be determined by extending curve 3 until it intercepts the line D.sub.r, as shown at 4. As E.sub.z increases the optical density decreases exponentially to a minimum optical density D. The practical operating voltage range is below E'.sub.z ; above which coagulation of the dipole suspension occurs.

The curve 3 in FIG. 52 is represented by the following equation:

D.sub.z = D +(D.sub.r - D) e.sup..sup.-(E .sup.- .sup.E .sup.)/E (50)

for E >>E.sub.r (50) becomes:

D.sub.z = D + (D.sub.r - D) e.sup..sup.-E/E (51)

the curve 5 of electrodichroic ratio as a function of electric field intensity is shown in FIG. 53. The equation of curve 5 is found from (51) the definition of the electrodichroic ratio:

q.sub.rz = (D.sub.r /D.sub.z) = q.sub.rz / [1 + (q.sub.rz - 1) e.sup..sup.-E/E ] (52)

the electrodichroic response .sigma..sub.rz = .DELTA.q.sub.rz /.DELTA.E is obtained by differentiating 52, and evaluated at E.fwdarw.0 (actually E.fwdarw.E.sub.r) and for q .sub.rz >> 1. There is thus obtained:

.sigma..sub.rz .fwdarw. (1/E.sub.r) (53)

The electrodichroic response has a limiting value, given by (53), which for a particular herapathite suspension measured about 1 .DELTA. q (kv/cm).sup..sup.-1. In this case the limiting electrodichroic response .sigma..sub.rz = (q.sub.rz - 1)/3 - 10, solving this q.sub.rz maximum .congruent. 31.

EXPERIMENTAL

The figures are the result of many tests using:

1. a herapathite suspension

2. a chromium rod metal suspension

The transmittance and optical density measurements were taken in the visible range using an RCA 931 photomultiplier tube, which has a peak response at 510 nm. Optical density was obtained directly using a logarithmic amplifier. Dynamic measurements were shown on a storage scope. The measurements were taken on electro-optic cells having glass windows constructed in accordance with this invention as shown in FIGS. 64 - 67 for z parallel orientation, (b) as shown in FIGS. 2 and 4 for x normal orientation.

The cells shown in section in FIGS. 66 and 65 are numbered to correspond to FIGS. 1 and 3. The conductive coatings 56 expand to about 2 mm from the edge of this glass plate, surrounded by a conductive 62' bus bar which is connected to the leads 58. The circumferential bus bar 62' results in decreasing the resistance of the conductive coatings in the circuit by a factor of 12 relative to that of a single edge bus bar. This is important since the more transparent coatings have a greater resistance which may be decreased by an order of magnitude in this manner.

In FIG. 67, the strips 200 and 201 were solid conductors with edges separated by a distance d" = 5 mm. However, any other value may be used. Alternatively, invisible lines inscribed in the lamination may be used as described in my copending application Ser. No. 551,113 -filed June 8, 1966.

Herapathite Suspension

The herapathite suspension contained blade shaped particles of L .congruent. 3000A, d .congruent. 150 A and a blade thickness of 50 A. The suspension was adjusted to a D.sub.r .congruent. 3 for d.sub.1 .congruent. 0.075cm.

The electrodichroic response is obtained by the slope .sigma..sub.rz = .DELTA. q.sub.rz /.DELTA.E = 11.DELTA.q(kv/cm).sup..sup.-1. The mass unit area of herapathite dipoles was M = 410 .times. 10.sup..sup.-6 gms/cm.sup.2, and the sensitivity was S.sub.rz = .sigma..sub.rz /M = 27000.DELTA.q(kv gm/cm.sup.3).sup..sup.-1. Based on the percent iodine in herapathite, the iodine conducting rods are 31 percent of the total herapathite mass. The active M is 127 .times. 10.sup..sup.-6 gms/cm.sup.3. On this basis S.sub.rz = 87000.DELTA.q (kv gm/cm.sup.3).

FIG. 54 shows the relative parallel electrodichroic ratio q.sub.rz vs. frequency f for various electric field intensities E.sub.z in kv(rms)/cm.

A maximum electrodichroic ratio is reached assumptotically at about 3 KHz for all applied electric field intensities.

FIG. 55 shows the relative parallel electrodichroic ratio q.sub.rz vs. the electric field intensity in kv(rms)/cm; for 1, 10, and 100 KHz. This shows that about 10 KHz, and about 3 kv(rms)/cm., the alignment, and q.sub.rz has reached a maximum value.

FIG. 56 shows present transmittance versus wavelength for parallel (open) and random (closed) states.

FIG.57 shows relative electrodichroic ratio q.sub.rz versus wavelength, plotted from data of FIG. 56. The peakq.sub.rz occurs at about 575 nm.

FIG. 58 shows wedge angle effect of transmittance versus angle, the dipoles being aligned in the Z direction for:

1. ordinary light

2. polarized light parallel

3. polarized light crossed

Permanent wedge-angle filters may be prepared with a film containing a dipole suspension having a fixed Z orientaion and laminated to a linear polarized film. The dipole layer is is oriented by an electric field, while the suspending medium is fluid. Then the orientation is fixed by causing the suspending medium to solidify.

Metal Dipole Suspension

The results on a chromium metal dipole suspension are shown in FIGS. 59-63 inclusive.

The chromium metal dipole particles were producd in a limited sample, of an average length of about 7700 A and a diameter from 400-600 A. There was a considerable spread of sizes, and many very small irregularly shaped particles; so that, this suspension was not refined or optimized. Nevertheless these results substantiate the antenna theory as applied to dipole suspensions. The results show (1) a strong polarizing effect (FIGS. 59 and 60); and (2) a large parallel electrodichroic ratio q.sub.rz = 28, with a peak at .lambda. = 3L = 3 .times. 0.77 = 2.3.mu.. According to the analysis herein, the dipole particles of length 1880A will have a peak 565nm.

MANUFACTURE OF SUBMICRON FLAKES

An aluminum film is coated on to a two mil thickness mylar. The coating thickness may be varied and preferably has a thickness to provide a transparency from 1 percent to 3 percent.

AL FILMS

THICKNESS VS. % TRANSMISSION THICKNESS TRANSMISSION A % 50 50 100 18.5 150 3. 200 1

the coated mylar is cut into one inch squares and 100 grams of this are placed in a jar with 1000 grams of acetone. The jar is shaken in a paint shaker for about two hours and cooled as needed to prevent excessive pressure from the acetone vapor. The suspension of flakes is decanted from the mylar. The process is repeteed as many times as needed to get the required concentration of flake; for example, five or ten times. 8 grams of 12 1/2 percent nitrocellulose solutiln and 2 grams of plasticizer are added; for example, Drapex which is an epoxy stearate. The fluid suspension which is a dark blue color, is placed in Ehrlenmyer flask. The Ehrlenmyer flask is placed on a water bath to evaporate all the acetone. A black paste remains containing a flake suspension. The paste is then placed on an evaporator until the resistivity is in excess of 30 meghom-cm. The paste is then diluted with a suitable solvent, for example, the same used in herapathite suspension to provide a fluid having an optical density of about three in a 1mm thick cell.

MANUFACTURE OF DIPOLE NEEDLES

Metal powder having particles which are about one micron in diameter is placed in a pyrex test tube and subjected to an elevated temperature for a period of many hours. As an exmaple, gold or platinum powder is heated for 30 hours at 400.degree. C to grow on them submicron whiskers of about 2000 A long and 100 A in diameter. One gram of the micron powder with the whiskers is then placed in 10 grams of acetone and the process repeated as above with the flakes; however, in this case after the agitation the metal needles remain in the suspension. The suspenstion stands until the one micron metal particles settle leaving the submicron needles in suspension. Centrifugation may be used instead of settling. The same procedure as with the flakes is then followed and the paste formed in this case contains the required submicron needles. ##SPC2##

Table 1 summarizes the results of these investigations presenting the electro-optical and response characteristics for random-parallel alignment, showing values for:

a. theoretical

b. goal

c. herapathite dipoles

d. metal dipoles-chromium

The values given in column (a) follow from the theory. Numerous examples are presented herein.

Under (b), goal represents what appears to be attainable results on further development, with no major breakthrough shorter dipoles and lower viscosity fluids will lead to faster response times for electric alignment, and Brownian relaxation.

The herapathite suspension is most fully developed. Its characteristics are shown in column (c). The chromium metal dipole suspension characteristics are shown in column (d). The chromium metal dipole suspension may be optimized.

Comparison of the data presented herein for herapathite dipole suspensions, and metal dipole suspensions shows:

1. The parallel electrodichroic ratio q.sub.rz of the metal dipole suspension at 2300 nm was 28 compared to about 15 for the herapathite suspension at 575.

2. The parallel electrodichroic response o.sub.rz of the metal suspension was 3.2 times that of the herapathite suspension.

3. The threshold electric field intensity E.sub.r of the herapathite suspension was 0.1 - 0.2 kv/cm; while that of the metal suspension was probably only about 0.03 kv/cm.

The much greater electrodichroic response and much smaller threshold observed for the metal dipole suspension relative to that of the herapathite dipole suspension may be ascribed to the greater conductivity of the metal. This results in a greater dipole torque for the metal in a given electric field intensity. This also explains the much smaller threshold electric field intensities for the metal dipole suspension compared to that for the herapathite dipole suspension, since a smaller electric field intensity will then be effective in the former case to turn the dipole against the randomizing effects of molecular impact.

The metal dipoles are cylindrical rods, while the herapathite dipoles are blade-shaped. Molecular impact on a blade will cause it to rotate about its long axis, as well as about its center; while molecular impact on a cylindrical rod would be more effective to rotate it about its center. In alignment by an electric field or in relaxation by molecular impact, the rod will have less resistance than a blade and will turn faster. The shape and the increased conductivity of the metal dipole particle compared to the herapathite dipole particle, explains the much smaller threshold electric field intensity required to align the metal dipoles.

4. For a dipole layer to be substantially opaque with the dipoles in random orienation, theory predicts a dipole mass/unit area M of about 1 .mu.g/cm.sup.2. (example 2). The herapathite dipole layer required 410 .mu.g/cm.sup.2 total, or about 127 .mu.g/cm.sup.2 calculated as iodine conducting rods only; while the metal dipole rods 0.77.mu. long required only 33.mu.g/cm.sup.2.

The herapathite dipole crystal comprises a bulky insulating clathrate cage for the conducting iodine rods. This extra bulk slows alignment and relaxation, increases the mass per unit area; and decreases response and sensitivity. If the metal rods were only 0.19.mu. long compared to 0.77.mu.(1/4 the length), then the peak electrodichroic ratio would be at 3 .times. 0.19 or 570 nm; and the mass/unit area required would be (1/4).sup.3 = 1/64 or approximately 0.5.mu.g/cm.sup.2, which is of the order of that predicted.

5. The parallel electrodichroic sensitivity S.sub.rz =94 .sub.rz /M expressed in .DELTA.q (cm.sup.3 /gm kv) of the herapathite dipole suspension was 2.7 .times. 10.sup.4 ; while that of the metal dipole suspension was 10.sup.6 ; theoretically for M = 1 .mu.g/cm.sup.2, and for .sigma..sub.rz = 33, S.sub.rz = 33 .times. 10.sup.6.

6. The relaxation time .tau..sub.B compared to the rise time .tau. at 10 kv/cm, had a ratio of .tau./.tau..sub.B of about 30 for the herapathite to about 8 for the metal suspension. The rise and relaxation times were much faster for the metal suspension, despite the greater density of the metal dipoles 7.2 gms/cm.sup.3 compared to about 2 gms/cm.sup.3 for the herapathite dipoles; and lengths of 0.77.mu. and 0.5.mu. respectively. The equations (31) and (32) may possibly have to be modified by the inclusion of the density .delta..sub.p in the numerator on the right hand side; although this is not suggested in the literature.

By decreasing the length of the metal dipole to about 0.19.mu. (1/4) 0.77.mu. the rise times and relaxations times should decrease by a factor of (1/4).sup.3 = 1/64 or to about 30.mu.s and 240.mu.s respectively.

The fluid viscosity in all these tests was about the same, 180 millipoise. By decreasing the fluid viscosity to about 18 millipoise, these times should be further decreased to about 3.mu.s and 24.mu.s respectively.

7. FIGS. 67 and 68 show that the voltage of AC pulse, applied across the Z cell, decreases as the metal dipole rods are aligned by the electric field, which causes an increase in the cell capacitance. The alignment of the metal dipole rods causes an increase in the apparent dielectric constant.

This effect may be useful as a capacitance control device, which has many applications in electronic circuits -- timing control, modulation, amplification, etc.

8. The metal dipoles operate at lower operating voltages -- for the present example about 10 percent the voltage of the herapathite suspension for the same electrodichroic ratio 10. Thus for a herapathite cell operating at 200 volts rms, the same cell would operate at 20 volts rms/cm for the metal suspension.

9. The transmittance -- time -- electric field equations are extraordinary; an exponential power raised to an exponential power, having the unique characteristics of a rapid tremendous change in transmittance from substantially opaque to substantially transparent; for a relatively small voltage applied parallel to the light path. Thus large area electro-optic windows, display panels, shutters and other devices become feasible.

10. The metal dipole suspensions are chemically more stable than herapathite suspensions.

11. The metal suspensions demonstrate the predicted antennae effects.

12. The metal dipoles are generally superior to the herapathite suspensions .

Claims

I claim:

1. An electrodichroic composition of matter comprising a suspension of conductive particles in a non-reactive transparent fluid suspending medium, said particles having lengths in the rage of 0.1 -1 .mu. and each of said particles having length to thickness ratios of at least 5.

2. An electrodichroic composition of matter according to claim 1 in which the transparent fluid suspending medium is a thermoplastic medium.

3. An electrodichroic composition of matter according to claim 1 in which the transparent fluid suspending medium is thixotropic.

4. An electrodichroic composition of matter according to claim 1 in which the transparent fluid suspending medium has a resistivity of at least 30 megohm-cm.

5. An electrodichroic composition of matter according to claim 1, in which the conductive particles are selected from the class consisting of aluminum, aluminum nickelide, antimony, cadmium, chromium, copper, gold, indium, lead, palladium, platinum, silver tantalum, thalium, titanium herapathite and tungsten.

6. An electrodichroic composition of matter according to claim 1 in which the particles are needle-like.

7. A pigment comprising an electrodichroic composition of matter according to claim 6, in which the dipoles have maximum lengths in the rnage of 1000 to 3000 A.

8. An electrodichroic composition of matter according to claim 1 in which the particls have a long dimension of the order of .lambda./2n, and at least one other dimension not exceeding .lambda./10n, where .lambda. is the wavelength of light in the visual range and n is the index of refraction of said transprent suspending medium.

9. An electrodichroic composition according to claim 5 in which the particles are flakes having long dimensions in the range of 1000 to 10,000 A, and said flakes having thicknesses of 50 to 250A.

10. The method of preparing a dipolar crystalline suspension defined by claim 1 which comprises forming from solution a suspension of needle-like crystals in a solvent media, saod solution being in at least a saturated condition, adding a miscible non-solvent of higher boiling point to said solution, and thereafter removing the solvent component.

11. The method of preparing a dipolar crystalline suspension defined by claim 1 which comprises forming from solution a suspension of needle-like crystals in a solvent media containing a polymer, said solution being in at least a saturated condition, adding a miscible non-solvent of higher boiling point to said solution and thereafter removing the solvent component.

12. The method of preparing a dipolar crystalline suspension defined by claim 1 which comprises forming from solution a suspension of needle-like crystals in a solvent media containing a polymer, said solution being in at least a saturated condition, adding a miscible non-solvent of higher boiling point to said solution and thereafter removing the solvent component, said non-solvent being of a sufficient quantity to form a paste upon removal of the solvent.

13. The method of preparing a dipolar crystalline suspension defined by claim 1 which comprises forming from solution a suspension of needle-like crystals from a solvent media containing a polymer, said solution being in at least a saturated condition, adding a miscible non-solvent of higher boiling point to said solution and thereafter removing the solvent component by vacuum, said non-solvent being of a sufficient quantity to form a paste upon removal of the solvent, and thereafter adding a low viscosity miscible non-solvent to dilute the paste to form a low viscosity dipolar suspension.

14. A method according to claim 13 in which the needle-like crystals are Herapathite crystals.

15. An electrodichroic composition of matter of a high viscosity concentrate including a polymer 1000 parts, plasticizer 1000-5000 parts, dipolar particles 1-1000 parts suspended therein, and said dipolar particles having lengths in the range of 0.1-1.mu..

16. An electrodichroic composition of matter including a nonionic fluid containing 1-10 percent of a high viscosity concentrate according to claim 15, with which the polymer and plasticizer is compatible, and dissolved therein, and in which total composition the dipoles are suspended.

17. An electrodichroic composition of matter comprising a suspension of conductive particles in a non-reactive transparent fluid suspending medium, said particles having lengths in the range of 0.1-0.2 .mu. and each of said particles having length to thickness ratios of at least 5.