Holography– a method of recording and subsequent reconstruction of the spatial structure of light waves, based on the phenomena of interference and diffraction of coherent light beams.

The photo plastic on which this information is recorded is called hologram.

It is not the optical image of the object that is recorded on the hologram, but the interference pattern that appears when a light wave scattered by the object (subject wave) is superimposed and a reference (or reference) wave coherent with it.

The main areas of application of holography:

Recording and storing information, incl. and visual (optical holographic memory);

Optical information processing and object recognition system;

Holographic interferometry.

Build a diagram, consider the recording processholograms.

In this process, a complex interference pattern is recorded and recorded on a photographic material (for example, photographic film), which is created by the superposition (interaction) of two light waves - the basic (reference) monochromatic wave and the secondary wave reflected or scattered by the object. The hologram is recorded according to the scheme shown in Fig. 1.

A monochromatic coherent laser beam is expanded by a collimator and further divided into two beams by a splitter. One (reference) beam is reflected from the mirror and sent directly to the photographic film. Another (object) beam is directed by the corresponding mirror to the object, reflected from it and perceived (recorded) by photographic film. It is this (reflected, scattered) beam that carries a variety of visual information about the volumetric (three-dimensional) parameters and characteristics (size, surface, contour, irregularities, transparency) of the object. Such a beam essentially creates a three-dimensional image of an object that a person can see and observe directly (with natural vision).

Light waves from the reference and scattered object beams create an interference pattern on the surface of the photographic film, consisting of many spots, the shape and intensity of which depend on the amplitude and phase of the incident and interacting light waves. The photographic film is exposed and then developed according to standard recipes. The resulting (developed) film is a hologram that preserves the interference pattern of the recorded object. The hologram has the appearance of a foggy negative, in which the details of the object are clearly not visible.

Build a diagram, consider the recovery (reproduction) processholograms.

Restoring a three-dimensional image of an object from its hologram (developed photographic film) is carried out according to the scheme presented in Fig. 2.

The hologram is illuminated by one reference beam, and the original conditions, the previous relative orientation of the reference beam and the photographic film, are preserved. If the specified conditions of laser illumination of the hologram are met, two images appear due to diffraction of light. It should be taken into account that earlier, in the process of the initial formation of the object’s hologram, a certain diffraction pattern with closely spaced interference fringes arose, the exact appearance of which is determined by the three-dimensional structure of the object. When this diffraction pattern is re-illuminated according to the scheme (Fig. 2), the diffracted light will have the parameters and characteristics specified by the original holographic shooting object.

One of the two images obtained when reproducing a hologram is virtual (Fig. 2), since a lens is required to observe it. However, the natural lens of the human eye is sufficient for this and the observer can see a virtual (but undistorted and three-dimensional) image of the object by viewing it directly through the hologram.

The second (actual, real) image is formed in a different direction of the laser beam passing through the hologram. This image can be projected onto a screen and observed without an intermediate lens. Part of the reproducing beam passes through the hologram without diffraction, without changing direction. This undiffracted beam does not have any noticeable practical value.

The considered schemes for recording (Fig. 1) and playback (Fig. 2) of a hologram, proposed by E. Leith and J. Upatnieks, belong to the category of optimal (technically advanced). These designs use off-axis geometry in which the reference and object rays strike the photographic film at an angle to each other. Therefore, when reproducing a hologram, the real and virtual images appear on opposite sides of the reference beam, which greatly facilitates separate observation of the images.

Holography- a set of technologies for accurately recording, reproducing and reshaping the wave fields of optical electromagnetic radiation, a special photographic method in which, using a laser, images of three-dimensional objects are recorded and then reconstructed, highly similar to real ones.

This method was proposed in 1947 by Dennis Gabor, who also coined the term hologram and received the Nobel Prize in Physics in 1971 “for the invention and development of the holographic principle.”

History of holography

The first hologram was received in 1947 (long before the invention of lasers) by Dennis Gabor during experiments to increase the resolution of the electron microscope. He also coined the word “holography” itself, with which he emphasized the complete recording of the optical properties of an object. Unfortunately, his holograms were of poor quality. It is impossible to obtain a high-quality hologram without a coherent light source.

Features of the scheme:

After creation in 1960 year of red ruby ​​(wavelength 694 nm, operates in pulsed mode) and helium-neon (wavelength 633 nm, operates continuously) lasers, holography began to develop intensively.

In 1962 year, a classic scheme for recording holograms by Emmett Leight and Juris Upatnieks from the Michigan Institute of Technology (Leith-Upatnieks holograms) was created, in which transmission holograms are recorded (when restoring a hologram, light is passed through a photographic plate, although in practice some of the light is reflected from it and also creates an image , visible from the opposite side).

Leith-Upatnieks scheme

In this recording scheme, the laser beam is divided by a special device, a divider (in the simplest case, any piece of glass can act as a divider) into two. After this, the rays are expanded using lenses and directed using mirrors to the object and recording medium (for example, a photographic plate). Both waves (object and reference) fall on the plate from one side. With this recording scheme, a transmission hologram is formed, which requires a light source with the same wavelength at which the recording was made, ideally a laser, to be restored.

In 1967 The first holographic portrait was recorded with a ruby ​​laser.

As a result of long work in 1968 year, Yuri Nikolaevich Denisyuk received high-quality (until that time the lack of necessary photographic materials prevented obtaining high quality) holograms that restored the image by reflecting white light. To do this, he developed his own hologram recording scheme. This scheme is called the Denisyuk scheme, and the holograms obtained with its help are called Denisyuk holograms.

Features of the scheme:

• observing images in white light;
• insensitivity to vibrations of the “object-RS” element;
• high resolution recording medium.

In 1977 Lloyd Cross created the so-called multiplex hologram. It is fundamentally different from all other holograms in that it consists of many (from tens to hundreds) individual flat views, visible from different angles. Such a hologram, naturally, does not contain complete information about the object; in addition, it, as a rule, does not have vertical parallax (that is, you cannot look at the object from above and below), but the dimensions of the recorded object are not limited by the laser coherence length (which rarely exceeds several meters, and most often only a few tens of centimeters) and the size of a photographic plate.

Moreover, you can create a multiplex hologram of an object that does not exist at all, for example, by drawing a fictitious object from many different angles. Multiplex holography is superior in quality to all other methods of creating three-dimensional images based on individual angles (for example, lens rasters), but it is still far from traditional holography methods in terms of realism.

In 1986 Abraham Secke put forward the idea of ​​creating a source of coherent radiation in the near-surface region of a material by irradiating it with X-rays. Since spatial resolution in holography depends on the size of the source of coherent radiation and its distance from the object, it turned out to be possible to reconstruct the atoms surrounding the emitter in real space.

Unlike optical holography, in all electron holography schemes proposed to date, the restoration of an object’s image is carried out using numerical methods on a computer.

In 1988 Barton proposed such a method for reconstructing a three-dimensional image, based on the use of Fourier-like integrals, and demonstrated its effectiveness using the example of a theoretically calculated hologram for a cluster of a known structure. The first reconstruction of a three-dimensional image of atoms in real space from experimental data was carried out for the Cu(001) surface by Harp in 1990.

Physical principles

Basic law of holography

If a photosensitive material on which a pattern of interference of several light waves is recorded is placed in the position in which it was during the recording process and illuminated again with some of these waves, then the rest will be restored. This feature is explained by the fact that not only the intensity is recorded on the hologram, as on a regular photographic plate, but also the phase of the light emanating from the object. It is information about the phase of the wave that is necessary for the formation of three-dimensional space during reconstruction, rather than the two-dimensional one provided by ordinary photography. Thus, holography is based on wavefront reconstruction.

The holographic process consists of two stages - recording and restoration.

• The wave from the object interferes with the “reference” wave, and the resulting pattern is recorded.
• The second stage is the formation of a new wave front and obtaining an image of the original object.

Recording information about the phase of a wave coming from an object can only be done with a light source with stable phase characteristics. Ideal for this purpose is laser- coherent light source of high intensity and high monochromaticity.

Superposition principle

Everyday experience shows that the illuminance produced by two or more ordinary incoherent light sources is the simple sum of the illuminances produced by each of them separately. This phenomenon is called superposition principle.

Huygens also wrote in his Treatise: “One of the most wonderful properties of light is that when it comes from different directions, its rays produce an effect, passing through one another without any interference.” The reason for this is that each source, consisting of many atoms and molecules, simultaneously emits a huge number of waves that are out of phase. The phase difference changes quickly and randomly, and, despite the fact that interference occurs between some waves, the interference patterns change with such frequency that the eye does not have time to notice the changes in illumination. Therefore, the intensity of the resulting oscillation is perceived as the sum of the components of the original oscillations, and the radiation of the source is "White light, i.e. not monochromatic, but consisting of different wavelengths. For the same reason, this light is unpolarized, but natural, that is, it does not have a predominant plane of vibration.

Coherent oscillations

Under special conditions, the principle of superposition is not observed. This is observed when the phase difference between light waves remains constant for a long enough time to be observed. The waves seem to “sound in time.” Such oscillations are called coherent.

The main feature of coherence is the possibility of interference. This means that when two waves meet, they interact, forming a new wave. As a result of this interaction, the resulting intensity will differ from the sum of the intensities of the individual oscillations - depending on the phase difference, either a darker or lighter field is formed, or instead of a uniform field, alternating bands of different intensities are formed, interference fringes.

Monochromatic waves are always coherent, However, light filters, often called monochromatic, in reality never produce strictly monochromatic radiation, but only narrow the spectral range and, of course, do not transform ordinary radiation into coherent radiation.

Obtaining coherent radiation

Previously, only one way of producing coherent radiation was known - using a special device - interferometer. The radiation from a conventional light source was divided into two beams, coherent with each other. These beams could interfere. Another method is now known, using stimulated radiation. Lasers are based on this principle.

Diffraction in holography

The main physical phenomenon on which holography is based is diffraction- deviation from its original direction of light passing near the edges of opaque bodies or through narrow slits. If not one, but several slits are applied to the screen, then an interference pattern appears, consisting of a series of alternating light and dark stripes, brighter and narrower than with a single slit. In the middle there is the brightest band of “zero order”, on both sides of it there are bands of gradually decreasing intensity of the first, second and other orders. As the number of slits on the screen increases, the stripes become narrower and brighter. A screen with a large number of thin parallel slits, the number of which is often increased to 10,000, is called a diffraction grating.

The grating, which is a hologram, is characterized primarily by the fact that diffraction occurs not at the slit, but at the circle. The diffraction pattern from a round opaque object is a bright central circle surrounded by gradually weakening rings. If, instead of an opaque disk, a disk with rings surrounding it is placed in the path of the wave, the circle in the image will become brighter and the stripes will become paler. If the transparency from a dark to a light area does not change abruptly, but gradually, according to a sinusoidal law, then such a grating forms only stripes of the zero and first orders, and interference in the form of stripes of higher orders does not appear. This property is very important when recording a hologram. If the transition from a dark ring to a light one is carried out strictly according to a sinusoidal law, then the rings in the image will disappear and the image will be a small bright circle, almost a dot. Thus, a circular sinusoidal grating will form from a parallel beam of rays (a plane wave) the same image as a collecting lens.

This lattice, called zone lattice(Soret plate, Fresnel plate), sometimes used instead of a lens. For example, it is used in glasses, replacing heavy high-refraction spectacle lenses. Obtaining zone gratings is possible in various ways, both mechanical and optical, interference. The use of these gratings, obtained by interference, forms the basis of holography.

Hologram recording

To record a hologram of a complex non-self-luminous object, it is illuminated with laser radiation. A coherent reference wave is directed onto the same plate onto which the scattered light reflected by the object falls. This wave is separated from the laser radiation using mirrors.

The light reflected by each point of the object interferes with the reference wave and forms a hologram of that point. Since any object is a collection of light-scattering points, many elementary holograms are superimposed on the photographic plate - points that together give a complex interference picture of the object.

The developed hologram is placed in the place where it was during recording, and the laser is turned on. Just as when restoring a hologram of a point, when the hologram is illuminated by a beam of light from a laser involved in recording, the light waves emanating from the object during recording are restored. Where the object was located during recording, a virtual image is visible. The real image associated with it is formed on the other side of the hologram, on the observer’s side. It is usually invisible, but unlike the imaginary one, it can be obtained on the screen.

Yu. N. Denisyuk (1962) developed a method in which three-dimensional media are used instead of a thin-layer emulsion to register a hologram. In such a thick hologram, standing waves arise, which significantly expanded the capabilities of the method. A three-dimensional diffraction grating, in addition to the previously described properties of a hologram, has a number of important features. The most interesting is the possibility of image restoration using a conventional continuous spectrum source - an incandescent lamp, the sun and other emitters. In addition, in a three-dimensional hologram there are no zero-order waves and a real image, and therefore interference is reduced.

The hologram recording scheme is shown in Figure 1. Denisyuk recorded a hologram in a three-dimensional environment, thus combining Gabor's idea with Lippmann's color photography. Then the sections of the hologram with maximum light transmission will correspond to those sections of the front of the object wave in which its phase coincides with the phase of the reference wave. Therefore, when the hologram is subsequently illuminated by a reference wave, the same distribution of amplitude and phase is formed in its plane as was the case of the object wave, which ensures restoration...

55. Holography. Scheme for recording and restoring holograms. Recording holograms on thick-layer emulsions. Application of holograms

Holography (from the Greek holos - whole, complete and grapho I write) a method of recording and reconstructing a wave field, based on recording the interference pattern, which is formed by a wave reflected by an object illuminated by a light source S (object wave), and a coherent wave coming directly from the source (reference wave). The recorded interference pattern is called hologram . The hologram recording scheme is shown in Figure 1.

The foundations of holography were laid in 1948 by physicist D. Gabor (Great Britain). Wanting to improve the electron microscope, Gabor proposed recording information not only about the amplitudes, but also about the phases of electronic waves by superimposing a coherent reference wave on the object wave. However, due to the lack of powerful sources of coherent light, he was unable to obtain high-quality holographic images. Holography experienced its rebirth in 1962–1963, when American physicists E. Leith and J. Upatnieks used a laser as a light source and developed a scheme with an inclined reference beam, and Yu.N. Denisyuk recorded a hologram in a three-dimensional environment, thus combining Gabor's idea with Lippmann's color photography. By 1965 1966 The theoretical and experimental foundations of holography were created. In subsequent years, the development of holography proceeded mainly along the path of improving its applications.

Let the interference structure formed by the reference and object waves be recorded by positive photographic material. Then the sections of the hologram with maximum light transmission will correspond to those sections of the front of the object wave in which its phase coincides with the phase of the reference wave. These areas will be more transparent, the greater the intensity of the object wave. Therefore, when the hologram is subsequently illuminated by a reference wave, the same distribution of amplitude and phase is formed in its plane as was the case of the object wave, which ensures the restoration of the latter.

Recovery object wave, the hologram is illuminated by a source creating a copy supporting waves. As a result of light diffraction on the interference structure of the hologram in a diffraction beam first order a copy of the object wave is restored, forming undistorted virtual imageobject, located in the place where the object was during holography. If the hologram is two-dimensional, the conjugate wave is simultaneously reconstructedminus of the first order, forming distorted real imagesubject (Figure 2).

The angles at which diffraction beams of zero and first orders propagate are determined by the angles of incidence of the object and reference waves on the photographic plate. In the Gabor scheme, the reference wave source and the object were located on the hologram axis ( axial diagram ). In this case, all three waves propagated behind the hologram in the same direction, creating mutual interference. In the Leith and Upatnieks scheme, such interference was eliminated by tilting the reference wave ( off-axis scheme).

Interference structurecan be recorded by a photosensitive material in one of the following ways:

1. in the form of variations in light transmittance or reflection. Such holograms, when reconstructing the wavefront, modulate the amplitude of the illuminating wave and are called amplitude;
2. in the form of variations in refractive index or thickness (relief). Such holograms, when reconstructing the wavefront, modulate the phase of the illuminating wave and are therefore called phase.

Often phase and amplitude modulation are performed simultaneously. For example, a conventional photographic plate records the interference structure in the form of variations in blackening, refractive index and relief. After bleaching the hologram, only phase modulation remains.

Interference structure recorded on a photographic plate usually lasts for a long time, that isthe recording process is separated from the recovery process (stationary holograms). However, there are photosensitive media (some dyes, crystals, metal vapors) that almost instantly respond with phase or amplitude characteristics to illumination. In this case, the hologram exists during the influence of the object and reference waves on the medium, and the restoration of the wave front is carried out simultaneously with the recording, as a result of the interaction of the reference and object waves with the interference structure formed by them (dynamic holograms). On principles dynamic holographypermanent and random access memory systems, laser radiation correctors, image intensifiers, laser radiation control devices, and wavefront inversion systems can be created.

If the thickness of the photosensitive layer is significantly greater than the distance between adjacent surfaces of the interference maxima, then the hologram should be considered as volumetric . If the interference structure is recorded on the surface of the layer, or if the layer thickness is comparable to the distance d between adjacent elements of the structure, then holograms are called flat. Criterion for the transition from two-dimensional holograms to three-dimensional ones: .

Volume hologramsare three-dimensional structures in which the surfaces of nodes and antinodes are recorded as variations in the refractive index or reflectance of the medium. The surfaces of nodes and antinodes are directed along the bisector of the angle that constitutes the object and reference beams. Such multilayer structures, when illuminated by a reference wave, act like three-dimensional diffraction gratings. Light specularly reflected from the layers restores the object wave.

Beams reflected from different layers reinforce each other if they are in phase, that is, the path difference between them is equal to (LippmannBragg condition). The condition is automatically satisfied only for the wavelength in the light of which the hologram was recorded. This determines the selectivity of the hologram with respect to the wavelength of the source, in the light of which the wavefront is restored. It becomes possible to restore an image using a continuous spectrum source (Sun, incandescent lamp). If the exposure was carried out with light containing several spectral lines (red, blue, green), then for each wavelength its own three-dimensional interference structure is formed. The corresponding wavelengths will be separated from the continuous spectrum when the hologram is illuminated, which will lead to the restoration of not only the structure of the wave, but also its spectral composition, that is, obtaining a color image. Three-dimensional holograms simultaneously form only one image (imaginary or real) and do not produce zero-order waves.

Properties of holograms.

A) The main property of holograms, which distinguishes it from a photograph, is that in the photograph only the distribution of the amplitude of the object wave incident on it is recorded, while in the hologram, in addition, the distribution of the phase of the object wave relative to the phase of the reference wave is also recorded. Information about the amplitude of the object wave is recorded on the hologram in the form of the contrast of the interference relief, and information about the phase in the form of the shape and frequency of the interference fringes. As a result, the hologram, when illuminated by a reference wave, restores a copy of the object wave.

B) The properties of a hologram, usually recorded on negative photographic material, remain the same as in the case of positive recording: light areas of the object correspond to light areas of the reconstructed image, and dark areas correspond to dark areas. This is easy to understand, taking into account that information about the amplitude of the object wave is contained in the contrast of the interference structure, the distribution of which on the hologram does not change when replacing a positive process with a negative one. With such a replacement, it only shifts to the phase of the restored object wave. This is not noticeable by visual observation, but sometimes appears in holographic interferometry.

IN) If, when recording a hologram, light from each point of an object hits the entire surface of the hologram, each small section of the latter is capable of reconstructing the entire image of the object. However, a smaller section of the hologram will reconstruct a smaller section of the wave front carrying information about the object. If this area is very small, the quality of the reconstructed image will deteriorate.

In the case of focused image holograms, each point of the object sends light to its corresponding small area of ​​the hologram. Therefore, a fragment of such a hologram restores only the corresponding section of the object.

G) The total brightness range transmitted by a photographic plate, as a rule, does not exceed one or two orders of magnitude, while real objects often have significantly larger brightness differences. A hologram with focusing properties uses all the light falling on its entire surface to construct the brightest areas of the image, and it is capable of conveying gradations of brightness up to five and six orders of magnitude.

D) If, when reconstructing the wave front, the hologram is illuminated with a reference source located relative to the hologram in the same way as during its exposure, then the reconstructed virtual image coincides in shape and position with the object itself. When the position of the reconstruction source changes, when its wavelength or the orientation of the hologram and its size changes, the correspondence is violated. As a rule, such changes are accompanied by aberrations in the reconstructed image.

E) The minimum distance between two adjacent points of an object that can still be seen separately when observing an image of an object using a hologram is calledhologram resolution. It grows with increasing size of the hologram. Angular resolution of round (diameter D ) of a hologram is determined by the formula: . Angular resolution of a square hologram with a square side equal to L , is determined by the formula: .

In most holographic schemes, the maximum size of the hologram is limited by the resolution of the recording photographic material. This is due to the fact that an increase in the size of the hologram is associated with an increase in the angle between the object and reference beams and the spatial frequency. An exception is the scheme of lensless Fourier holography, in which it does not increase with increasing size of the hologram.

AND) The brightness of the reconstructed image is determineddiffraction efficiency, which is defined as the ratio of the luminous flux in the reconstructed wave to the luminous flux incident on the hologram during reconstruction. It is determined by the type of hologram, the conditions of its recording, as well as the properties of the recording material.

The maximum achievable diffraction efficiency of holograms is:

For two-dimensional transmitting holograms

amplitude 6.25%,

phase 33.9 5;

For two-dimensional reflective6.25 and 100%, respectively;

For three-dimensional transmittingholograms 3.7 and 100%;

for three-dimensional reflective 7.2 and 100%.

Applications of Holography. When restoring holograms, a complete illusion of the existence of an object is created, indistinguishable from the original. This property of holograms is used in lecture demonstrations, when creating three-dimensional copies of works of art, and holographic portraits. Three-dimensional holographic images are used to study moving particles, raindrops or fog, and tracks of nuclear particles in bubble and spark chambers.

Using holographic devices, various wave transformations are carried out, including wavefront reversal in order to eliminate optical aberrations. One of the first applications of holography was related to the study of mechanical stress. Holography is used to store and process information. This ensures high recording density and recording reliability.

The three-dimensionality of the image makes the creation of holographic cinema and television promising. The main difficulty in this case is the creation of huge holograms that could be viewed simultaneously by a large number of viewers. In addition, the hologram must be dynamic. To create holographic television, it is necessary to overcome the difficulty caused by the need to expand the frequency band by several orders of magnitude in order to transmit three-dimensional moving images.

A hologram can be produced not only by the optical method, but also designed on a computer (digital hologram). Machine holograms are used to obtain three-dimensional images of objects that do not yet exist. Machine holograms of complex optical surfaces are used as standards for interference testing of product surfaces.

Acoustic holography is also known, which can be combined with methods for visualizing acoustic fields.

Additional material

When the reference and object waves meet in space, a system of standing waves is formed. The maxima of the amplitude of standing waves correspond to zones in which the interfering waves are in the same phase, and the minima correspond to zones in which the interfering waves are in antiphase. For a point reference source O 1 and a point object O 2 the surfaces of maxima and minima represent a system of hyperboloids of revolution. The spatial frequency of the interference structure (the reciprocal of its period) is determined by the angle at which light rays emanating from the reference source and emanating from the object converge at a given point: , where is the wavelength. Planes tangent to the surface of nodes and antinodes at each point in space bisect the angle. In the Gabor scheme, the reference source and object are located on the hologram axis, the angle is close to zero and the spatial frequency is minimal. Axial holograms are also called single beam , since one beam of light is used, one part of which is scattered by an object and forms an object wave, and the other part, which passes through the object without distortion, is a reference wave.

In the Leith and Upatnieks scheme, a coherent inclined reference beam is formed separately (double beam hologram). For double-beam holograms, the spatial frequency is higher than for single-beam holograms. Therefore, recording double-beam holograms requires photographic materials with a higher spatial resolution.

If the reference and object beams fall on the photosensitive layer from different sides (~ 180 0 ), then it is maximum and close to 2/ (holograms in colliding beams). Interference maxima are located along the surface of the material in its thickness. This scheme was first proposed by Denisyuk. Since when such a hologram is illuminated by a reference beam, the reconstructed object wave propagates towards the illuminating beam, such holograms are sometimes called reflective.

Types of holograms. The structure of the hologram depends on the method of generating the object and reference waves and on the method of recording the interference pattern. Depending on the relative position of the object and the plate, as well as on the presence of optical elements between them, the relationship between the amplitude-phase distributions of the object wave in the planes of the hologram and the object is different. If the object lies in the plane of the hologram or is focused on it, then the amplitude-phase distribution on the hologram will be the same as in the plane of the object (focused image hologram; Figure 3).

When the object is far enough from the plate, or at the focus of the lens L, then each point of the object sends a parallel beam of light to the plate. In this case, the connection between the amplitude-phase distributions of the object wave in the hologram plane and in the object plane is given by the Fourier transform (the complex amplitude of the object wave on the plate is the so-called Fourier image of the object). The hologram in this case is calledFraunhofer hologram(Figure 4).

If the complex amplitudes of the object and reference waves are Fourier images of the object and reference source, then the hologram is calledFourier hologram. When recording a Fourier hologram, the object and the reference source are usually located in the focal plane of the lens (Figure 5).

In the case of a lensless Fourier hologram, the reference source is located in the plane of the object (Figure 6). In this case, the front of the reference wave and the fronts of elementary waves scattered by individual points of the object have the same curvature. As a result, the structure and properties of the hologram are almost the same as those of the Fourier transform hologram.

Fresnel hologramsare formed when each point of an object sends a spherical wave to the plate. As the distance between the object and the plate increases, Fresnel holograms turn into Fraunhofer holograms, and as this distance decreases, into focused image holograms.

S

Real Image

Virtual image

Figure 6 Scheme of lensless recording of a Fourier hologram

Hologram

Figure 5 Fourier hologram recording scheme

Reference source

Support beam

L

Support beam

Figure 4 Fraunhofer hologram recording diagram

Figure 3 Scheme of recording a focused image hologram

Figure 1 Schematic of hologram recording

Figure 2 Recovery scheme

holographic image of an object

Support beam

Hologram

Which coincide with a very high degree of accuracy, a standing electromagnetic wave arises. When a hologram is recorded, two waves are added in a certain region of space: one of them comes directly from the source (reference wave), and the other is reflected from the recording object (object wave). In the region of a standing electromagnetic wave (or other recording material) is placed, as a result, a complex pattern of darkening bands appears on this plate, which correspond to the distribution of electromagnetic energy (pattern) in this region of space. If now this plate is illuminated by a wave close to the reference one, then it will convert this wave into a wave close to the object one. Thus, we will see (with varying degrees of accuracy) the same light that would be reflected from the recording object.

Sources of light

When recording a hologram, it is extremely important that the lengths (frequencies) of the object and reference waves coincide with each other with maximum accuracy and do not change during the entire recording time (otherwise a clear picture will not be recorded on the plate). This can only be achieved if two conditions are met:

1. both waves were initially emitted by the same source
2. this source emits a wave with a very stable wavelength (radiation)

The only light source that satisfies the second condition well is . Before the invention of lasers, holography practically did not develop. Today, holography places one of the most stringent requirements on laser coherence.

Most often, coherence is usually characterized by the coherence length - that difference in the optical paths of two waves, at which the clarity of the interference pattern drops by half compared to the interference pattern produced by waves that have traveled the same distance from the source. For various lasers, the coherence length can range from several millimeters (high-power lasers intended for welding, cutting and other applications that do not require this parameter) to tens of meters (special, so-called single-frequency lasers for applications that require coherence).

History of holography

The first hologram was obtained in the year (long before the invention of lasers) during experiments to increase resolution. He also coined the word “holography” itself, with which he emphasized the complete recording of the optical properties of an object. Unfortunately, his holograms were of poor quality. It is impossible to obtain a high-quality hologram without a coherent light source.

Leith-Upatnieks notation scheme

In this recording scheme, the laser beam is divided by a special device, a divider (in the simplest case, any piece of glass can act as a divider) into two. After this, the rays are expanded using lenses and directed to the object and the plate using mirrors. Both waves (object and reference) fall on the plate from one side. With this recording scheme, a transmitting hologram is formed, which requires for its reconstruction a source emitting light in a very small range of wavelengths (monochrome radiation), ideally - .

Denisyuk's recording scheme

In this scheme, the laser beam is expanded and directed towards. Part of the beam passing through it illuminates the object. Light reflected from an object forms an object wave. As can be seen, the object and reference waves fall on the plate from different sides. In this scheme, a reflective hologram is recorded, which independently cuts out a narrow section (sections) from the continuous spectrum and reflects only this. Thanks to this, the hologram image is visible in ordinary white light or a lamp (see the picture at the beginning of the article). Initially, the hologram cuts out the wavelength with which it was recorded (however, during processing and when storing the hologram, it can change its thickness, and the wavelength also changes), which makes it possible to record three holograms of one object on one plate, and with lasers, ultimately obtaining one a color hologram that is almost impossible to distinguish from the object itself.

This scheme is characterized by extreme simplicity and in the case of application (having extremely small dimensions and producing a diverging beam without the use of ) it is reduced to only one laser and some base on which the laser, plate and object are fixed. It is precisely these schemes that are used when recording amateur holograms.

Photo materials

Holography is extremely demanding on the resolution of photographic materials. The distance between the two maxima of the pattern is of the same order of magnitude as the laser wavelength, the latter most often being 633 (helium-neon) or 532 (second harmonic laser) nanometers. Thus, this value is on the order of 0.0005 mm. To obtain a clear image of the interference pattern, photographic plates with from 3000 (Leit-Upatnieks) to 5000 (Denisyuk) lines per millimeter were required.

The main photographic material for recording holograms is special photographic plates based on traditional silver bromide. Thanks to special additives and a special development mechanism, it was possible to achieve a resolution of more than 5000 lines per millimeter, but this comes at the cost of extremely low sensitivity of the plate and a narrow spectral range (precisely matched to the laser radiation). The sensitivity of the plates is so low that they can be exposed to direct sunlight for a few seconds without the risk of flare.

In addition, photographic plates based on bichromated gelatin are sometimes used, which have even greater resolution and allow recording very bright holograms (up to 90% of the incident light is converted into an image), but they are even less sensitive, and they are sensitive only in the short wavelength region (blue and, to a lesser extent, the green portions of the spectrum).

At the moment, there is only one industrial (except for a number of small) production of photographic plates for holography in the world - the Russian Slavich Company.

Some recording schemes make it possible to write on plates with lower resolution, even on ordinary photographic films with a resolution of about 100 lines per millimeter, but these schemes have a lot of limitations and do not provide high image quality.

Amateur holography

As already written above, Denisyuk’s scheme, when using a laser diode as a source of coherent light, turns out to be extremely simple, which made it possible to record such holograms at home without the use of special equipment.

To record a hologram, it is enough to create a certain frame on which a laser, a photographic plate (usually PFG-03M) and a recording object will be fixedly mounted. The only serious requirement imposed on the design is minimal vibration. The installation should be installed on vibration-damping supports; a few minutes before and during the exposure, you must not touch the installation (usually the exposure is measured by opening and closing the laser beam with a screen that is not mechanically connected to the installation; in the simplest case, you can simply hold it in your hand).

Amateur holography uses cheap and readily available semiconductor lasers:

1. laser pointers
2. laser modules
3. separate laser diodes

Laser pointers are the easiest to use and affordable source of coherent light. You can buy them for little money almost anywhere. After unscrewing or sawing off the lens that focuses the beam, the pointer begins to shine like a flashlight (except that its spot is elongated in one direction), allowing it to illuminate the photographic plate and the scene located behind it. You just need to secure the button in some way (for example, with a clothespin) in the on state. The disadvantages of pointers include their unpredictable quality and the need to constantly buy new batteries.

A more advanced source is a laser module, whose focusing lens again needs to be unscrewed or sawed off. Unlike a pointer, the module is not powered by batteries inside it, but by an external source, which can be a stabilized 3V power supply. Such a power supply, like the laser module itself, is usually sold in radio parts stores for relatively little money. The absence of low batteries contributes to stable operation. As a rule, laser modules are made better than pointers, but their coherence is also unpredictable.

Finally, laser individual diodes are the most difficult light sources to operate. Unlike modules and pointers, they do not have a built-in power supply, so you will have to assemble one or buy one (the latter is very expensive). The fact is that laser diodes, as a rule, use a non-standard supply voltage, for example 1.8V, 2.7V, etc. In addition, what is more important for them is not the supply voltage, but the current. The simplest power supply consists of a milliammeter, a variable resistor and a standard 3-5V stabilized power supply. In addition, the laser diode is not capable of cooling itself; it must be installed on a radiator. The thermal power of diodes used for amateur holography does not exceed hundreds of milliwatts, so a minimally sized radiator is sufficient; however, the larger the radiator, the more stable the temperature, and coherence directly depends on temperature stability.

As already written above, the coherence of pointers and modules is completely unpredictable, because this parameter is not important for their normal use. It is quite possible that you will have to buy several modules/pointers before you come across an instance with high coherence. You can understand that coherence is insufficient from the recorded hologram: if it has characteristic stripes that move when it rotates, then the laser generates several wavelengths and its coherence is low.

In the case of laser diodes the situation is noticeably better. First, if the diode exhibits a poor emission spectrum (i.e., low coherence) in its normal operating mode, then by slightly lowering or increasing the current through it, you can try to obtain a good spectrum. Secondly, some diodes are manufactured by the manufacturer taking into account the requirements of high coherence. These are lasers with a single longitudinal mode (Single longitudinal mode) or single-frequency lasers. Their coherence length significantly exceeds a meter, which greatly exceeds the needs of amateur holography. Moreover, the price of such lasers starts from several tens of dollars, which is quite affordable for most amateurs. In particular, such laser diodes are produced by Opnext together with Hitachi.

Red semiconductor lasers with a wavelength of 650 nm are most widely used in a wide variety of applications. These same lasers are most widely used in amateur holography. They are distinguished by their low price, fairly high power, and the sensitivity of the eye (and the PFG-03M photographic plates used to record Denisyuk holograms) to this wavelength is quite high. Less common in holography are lasers with wavelengths of 655-665 nm. The sensitivity of the photographic plate (and the eye) to this range is noticeably (about 2 times) less than to 650 nm, but such lasers have many times more power at a similar price. 635nm lasers are even less common. Their spectrum is extremely close to the spectrum of the red He-Ne laser (633 nm), for which photographic plates are sharpened, which ensures maximum sensitivity (the sensitivity of the eye is also significantly, twice as high as at 650 nm). However, these lasers have a high price, low efficiency and rarely have high power. In addition, the polarization of these lasers is perpendicular to the polarization of longer wavelength lasers, but this is neither an advantage nor a disadvantage, it just needs to be taken into account when installing the laser to ensure minimal reflection of light from the glass of the photographic plate.

Links

• Holography - Virtual Gallery- the largest website dedicated to holography in the CIS

3. Optical information processing. Introduction to Optoelectronics

3. Optical information processing

Human activity is constantly associated with the need to compare, analyze and summarize information of a very different nature. The most important trend towards improving the methods and means used for this is the continuous increase in the volume and speed of data processing. Electronics has achieved significant success in this regard, the development of which led, in particular, to the creation of computers based on integrated circuits. Solving increasingly complex problems requires a further increase in the number of electronic components, memory units, and an increase in the number of operations performed.

The use of optical methods makes it possible to radically improve the defining characteristics of data processing systems. Indeed, an increase in the carrier frequency due to the use of electromagnetic oscillations in the optical range) leads to a colossal increase in the information capacity of the signal transmission and processing channel. The short wavelength of light allows the use of signal modulation not only in time, but also in spatial coordinates, i.e., parallel processing and storage of huge amounts of information (for example, over 1 10 6 channels or more). In this case, the influence of both cross-talk and external interference is relatively easily eliminated.

Optical information processing presupposes the presence of fundamentally new elements and means: high-speed light modulators (single-channel and two-coordinate), optical beam deflection devices (deflectors), storage devices with adequate information capacity and speed, multi-element photodetectors, information display devices, etc.

Optical methods make it possible to process and record information in both analog and digital (binary) form. It must be borne in mind that in the first case, a linear dependence of the optical characteristics of the devices on the value of the control signal is desirable; in the second, on the contrary, it is better if the device has threshold properties. Digital methods, characterized by significantly greater accuracy, less sensitivity to the effects of distortion and external interference, ease of recording and signal conversion, require a wider frequency band. However, this is precisely what optical methods easily provide, so processing and recording information in digital form in optical devices is widespread. Any analog signal, as is known, can be represented in digital form, going to pulse code modulation(ICM).

3.1. Optical radiation modulators

As in radio engineering, modulation consists of introducing information into a light wave by changing one of its characteristics over time - amplitude, phase, frequency, as well as polarization. Photodetectors used in optoelectronics are usually sensitive only to light intensity, so in practice modulation of the phase, frequency or polarization of light is usually converted to amplitude.

If optical radiation is transformed in the necessary way during its generation in the source itself, modulation is called internal (direct). In the case of LEDs or semiconductor lasers, modulation of the radiation intensity can be achieved by changing the exciting current. This is a simple and convenient method used in practice. However, very often there is a need to modulate radiation that has already left the source (external modulation). Optical modulators can operate at higher frequencies than those achievable with internal modulation. Of course, one cannot rely on the use of moving curtains, screens, mirrors, prisms, disks with holes or other mechanical devices whose speed does not exceed ~1·10 4 Hz. Optical radiation modulators in information processing and transmission systems operate on the basis of various physical processes that occur when light passes through a modulating medium under the influence of external factors.

3.1.1. Operating principles of optical modulators

For light modulation, the well-studied electro-optical Kerr effect(1875), which consists in the appearance of optical anisotropy under the influence of an external electric field in an isotropic substance. To observe the effect (Fig. 3.1, a), a transparent dielectric substance is placed between the plates of a flat capacitor, to which a voltage is applied U, creating a sufficiently strong electric field in the modulating environment of the MS E. A Kerr cell is placed between crossed polarizers P and analyzer A. At U=0, the light intensity at the output of the device is also zero, however, when a voltage is applied, the modulating medium becomes optically similar to a birefringent crystal with an optical axis parallel to the direction of the electric field. Therefore, after passing through a Kerr cell, a light wave splits into two linearly polarized components. One of them is polarized so that its electric vector is oriented perpendicular to the external field E(ordinary wave), and the other - in parallel E(not an ordinary wave). To ensure maximum modulation depth, it is necessary that the main plane of the polarizer P composed with vector E corner 45°. Ordinary and extraordinary waves have different refractive indices ( P about and P f) and therefore propagate in the environment at different speeds. After passing through the Kerr cell, the light becomes elliptically polarized and passes through the analyzer to a greater or lesser extent.

Theory and experience show that the difference P about and P e is proportional E 2 (hence the name used - quadratic Kerr effect):

Where k K- coefficient independent of E. Phase difference between ordinary and extraordinary rays after passing the path l in a modulating environment is

, (3.2)

Where B =kTo/λ - so-called Kerr constant.

The quadratic Kerr effect is explained by the optical anisotropy of the molecules of the modulating medium, i.e., the difference in their ability to be polarized by the electric field of a light wave in different directions. In the absence of an external electric field E anisotropic molecules are randomly oriented and the substance as a whole is isotropic. If molecules have their own electric dipole moment, then a sufficiently strong electric field causes their preferential orientation and the substance becomes macroscopically anisotropic.

In substances consisting of molecules that do not have their own dipole moment, an external electric field can induce it, and due to the anisotropy of the molecules, the dipole moment does not necessarily coincide with the direction E. A pair of forces arises that force the molecules to orient themselves in a certain way relative to E. In accordance with the above, they distinguish orientational And Kerr polarization effects. The order of magnitude time of orientational relaxation of dipole molecules is 10 -9 s. This means that at modulation frequencies greater than 10 8 -10 9 Hz, the orientational Kerr effect practically does not appear and only the polarization effect remains effective, the speed of which is limited by the time of 10 -12 -10 -13 s.

Electro-optical phenomena are observed not only in isotropic substances, but also in crystals with natural optical anisotropy. So that double refraction does not appear when E= 0, a uniaxial crystal is cut so that faces are formed perpendicular to its optical axis, and light is directed along it. The control electric field is created in a direction perpendicular to the direction of light propagation, i.e., the same as in a Kerr cell (Fig. 3.1, a). A modulating device is also possible, in which the electric field is directed parallel to the propagation of light. To do this, transparent electrodes are applied to the corresponding faces of the anisotropic crystal (Fig. 3.1,6). According to Fig. 3.1 use the terms - longitudinal And transverse electro-optical effects. The change in birefringence of an anisotropic crystal placed in an electric field is called Pockels effect- named after the physicist who discovered it (1894). Unlike the Kerr effect, the difference n 0 And n e in the Pockels effect is proportional to the first power E:

, (3.3)

Where k n is the electro-optical coefficient, different from k K in formula (3.1) both in value and in dimension. How

and the Kerr effect and Pockels effect are characterized by low inertia, which allows light to be modulated to frequencies of ~1·10 13 Hz. It should, however, be borne in mind that the upper limit of the modulation frequency is most often determined not by processes in the substance, but by the capacitance of the device and turns out to be several orders of magnitude lower.

The operation of optical radiation modulators can be based on magneto-optical effects, in particular on the effect first studied by Cotton and Mouton (1907). This effect is similar to the electro-optical Kerr effect (Fig. 3.2, A): the modulating medium is placed between the crossed Polaroid and the analyzer, the direction of the magnetic field is perpendicular to the light beam, the main planes of the polarizers are 45° with the direction of the magnetic field. Cotton effect-Moutona observed in a macroscopically isotropic substance consisting of molecules or aggregates of molecules that have a constant magnetic moment, but are randomly oriented. An external magnetic field, interacting with the magnetic moments of molecules, orders their orientation, as a result of which the substance becomes anisotropic, acquiring the properties of a birefringent crystal. As in the case of the Kerr effect, under the influence of a magnetic field the light beam is divided into two beams - ordinary and extraordinary - and, having passed through the modulating medium, becomes elliptically polarized due to the difference P 0 and P e, and this difference is proportional to the square of the tension N magnetic field:

, (3.4)

Where k KM - Cotton's coefficient- Moutona(sometimes it means the value k km/λ).

Can also be used in optical modulators Faraday effect(1845), which consists in rotating the plane of polarization of light propagating in a medium along the magnetic field (Fig. 3.2, b). The effect is explained by the fact that in a magnetized substance the refractive indices for circularly right- and left-handed polarized light differ n + And P - . Plane-polarized light is the sum of left- and right-hand polarized components. After passing the modulating medium, a path difference arises between them, as a result of which the plane of polarization rotates by an angle φ, proportional to the length l paths of light in matter and the first degree H:

where ρ - Verdet constant, named after the researcher who studied in detail the magnetic rotation of the plane of polarization of light.

The action of an optical modulator can be based on a number of other effects discussed in subsequent sections of this chapter: acousto-optical, inverse piezoelectric, as a result of changes in the optical absorption coefficient, the ability of a material to scatter light, etc.

3.1.2. Characteristics and parameters of optical modulators

Regardless of the operating principle, optical modulators are characterized by a number of parameters: signal modulation depth, optical losses, transparency band, modulated frequency band, specific power consumption, control voltage, etc.

If Ф min denotes the intensity of light passing through the modulator in the absence of a control signal (in complete darkness), and Ф m ах - when it is supplied (in complete brightening), then depth(degree) modulation defined as

The modulation depth is often also understood as the ratio Ф m ах to Ф min, which is usually expressed in decibels:

If Ф min ≈ 0, the modulator can be used as optical shutter(light valve), i.e. a device that turns the light on and off.

Optical loss a modulator or shutter is characterized by the ratio of the light intensity Ф 0 in the absence of the device to its value Ф max with full clearing of the modulator and is also expressed in decibels:

(3.8)

Transparency strip determines the spectral range of radiation passing through the modulator without noticeable attenuation.

Under bandwidthΔ f modulator refers to the range of modulation frequencies in which it can operate. Usually Δ f is defined as the difference between the upper f in and bottom f n frequencies and, since f in >> f n, then Δ f ≈ f V. The optical shutter is also characterized response time (speed), which is close in order of magnitude to / f in 1 .

Energy is expended for modulation, and the greater the Δ f. Therefore, as a characteristic of the modulator, a parameter is introduced, determined power consumption per unit modulation frequency band(usually expressed in milliwatts per megahertz).

In the case of crossed polarizers at the input and output of the modulator, the amplitude of the transmitted light wave is proportional to sinφ, where φ is the angle of rotation of the plane of polarization caused by the application of voltage U, a output light intensity

, (3.9)

Where Uλ/2- the so-called half wave voltage, equal to this U, at which the maximum light transmission of the device is achieved, i.e. when the phase of the output light changes by π.

Electro-optical modulators have become widespread. Effective materials for such devices are: lithium niobate LiNbO 3 with a transparency range of 0.4 - 4.5 microns, lithium tantalate LiTaO 3 (0.4 - 5 microns), barium and bismuth titanates (BaTiO 3 and Bi 4 Ti 3 O 12 ), potassium niobate and tantalate (KNbO 3 and KTaO 3), as well as KTa x Nb 1- x O 3 (KTN) (0.5–4.5 µm). Such “classical” electro-optical materials are also used as potassium dihydrogen phosphate KH 2 PO 4 (abbreviated designation KDP) and its deuterated modification KD 2 PO 4 (DKDP) with a transparency range of 0.3 - 1.2 μm, ammonium dihydrogen phosphate NH 4 H 2 PO 4 (ADP), ammonium dihydroarsenide NH 4 H 2 AsO 4 (ADA) and many other materials.

Magneto-optical modulators use ferromagnetic materials, in particular ferrites, which combine ferromagnetic and semiconductor (dielectric) properties and are complex oxides of iron and some other elements. Some of their varieties are widely used to cover tapes of tape recorders and video recorders. Many types of ferrites can be used, in particular yttrium iron garnet Y 3 Fe 5 O 12, yttrium aluminum garnet Y 3 A1 5 O 12 (YAG), other materials (Bi x Y 1- x Fe 5 O 12, Y 2 BiFe 3, 8 Ga 1,2 O 12), transparent in the red and near-infrared regions of the spectrum.

Optical modulators can also use many other effects, described in later sections of the chapter.

3.2. Optical deflectors

3.2.1. Electro-optical deflectors

Common elements in optical information processing systems are devices for changing the spatial position of a light beam - the so-called deflectors(from Latin deflectio - I reject). There are deflectors with a discrete set of positions of the deflected beam, as well as those intended for its continuous scanning - scanners.

As already noted, the ordinary and extraordinary rays emerging from a birefringent crystal are linearly polarized in mutually perpendicular planes. If light polarized in the polarization plane of an ordinary beam is directed onto a plane-parallel plate cut from such a crystal at an angle to its optical axis, the extraordinary beam will be absent at the output of the crystal, and the ordinary beam will pass through the crystal without changing its spatial position. If the plane of polarization of the beam incident on the plate is rotated by 90°, only an extraordinary beam will pass through the crystal, which will no longer be a continuation of the primary one, but will shift in parallel relative to it. In other words, using a polarizer, one of two spatially separated beams emerging from the crystal can be isolated. In deflectors, the orientation of the plane of polarization of the primary beam is changed not by rotating the polarizer, but by using an electro-optical cell, which, when passed through in the absence of a control voltage, U the polarization of light does not change, but when U, equal to half-wave U λ/2, the polarization plane rotates 90°, which is required for the deflector to operate. The beam displacement depends on the material from which the birefringent plate is cut and its thickness, i.e. it cannot be controlled electrically. So that the beam at the output of the deflector can have many positions, the light is passed through a sequence of pairs “electrically controlled polarization modulator - birefringent plate” (Fig. 3.3). To obtain the same step in a discrete sequence of positions of the light beam at the output of the deflector, it is necessary that the thickness of the birefringent crystals arranged in cascade one after another differ by a factor of two.

For definiteness, let us assume that the main section of all crystals (the plane passing through the direction of the light beam and the optical axis of the crystal) coincides with the plane of the drawing. Let us direct a linearly polarized beam to the deflector, so that the plane of the electric vector in the light wave is perpendicular to the main cross section of the crystals. If voltage is not applied to all polarization modulators, the plane of polarization of the beam does not change, it does not deviate from the original direction of propagation and at the output of the device will be in the position 1. Let us now apply voltage to the third polarization modulator U 3 , equal to half-wave, i.e., rotating the plane of polarization of light by 90°. This corresponds to the plane of polarization of the extraordinary beam in the third birefringent plate. In this case, the beam will deviate, leaving the plate in direction 2. So that the beam at the output of the modulator takes the position 3, you need to apply a half-wave voltage to the second stage of the modulator, the thickness of the birefringent plate in which is twice as large as in the third, and to prevent the beam from being deflected by the third stage, you need to apply voltage Uλ/2 both on the second and third cascades. In order for the output light beam to hit the point 4, half-wave voltage needs to be applied only to the second stage, etc. (Table 3.1).

To expand the beam deflection range by half (with the same step) into the device shown in Fig. 3.3, you need to introduce a cascade with a birefringent plate twice as thick as the first cascade. Further expansion of the beam deflection range requires the introduction of cascades with even thicker plates.

With help m- you can get 2 cascade deflectors T discrete positions of the light beam at the output. To obtain a total beam position number of, for example, 256, an 8-stage deflector is required. To obtain a beam deflection along two coordinates, birefringent crystals are introduced into the deflector, the main sections of which are mutually perpendicular (in this case, 10 4 -10 5 resolved positions of the beam at the output are quite achievable with a switching time of 10 -6 -10 -7 s). It is obvious that it is not at all necessary that the thickness of the birefringent plates decrease in the direction of light propagation of the second ray. If it is not the same as in Fig. 3.3 (in reverse order or alternately), only the switching of control voltages will change.

Table 3.1.Switching control voltages three-stage deflector shown in Fig. 3.3.

Voltage			Position beam
U 1	U 2	U 3

One of the main parameters of the deflector is resolution, which for the device in question is determined by the material and thickness of the birefringent plates l, as well as their orientation relative to the optical axis of the crystal. Obviously, the deviation of the extraordinary ray h equals l tanψ, where ψ is the angle of deflection of the beam in the plate (Fig. 3.3).

The same materials can be used in deflectors as in electro-optical modulators: KDP, ADP, DKDP, LiNbO 3, BaTiO 3, etc. The mineral calcite CaCO 3 (56% CaO + 44% CO 2 with impurities) or its especially transparent variety - Iceland spar, which has high birefringence. At a wavelength of 0.63 μm, the angle ψ for a KDP crystal, for example, is equal to ~1.5°, for calcite - about 6°. It should be emphasized that the beam deflection at the output of the deflector of the type considered does not depend on the voltage on the polarization modulator. If you make it unequal U λ / 2, the position of the extraordinary ray will not change, but only its intensity will decrease. In addition, an ordinary beam will appear at the output of the deflector, the intensity of which will increase as the U compared with U λ / 2. This allows the deflector to be used as a modulator.

Continuous deviation (scanning) The beam can be obtained using a prism made of electro-optical material (for example, KTN, KDP, BaTiO 3) with metal electrodes deposited on its end faces, to which a control voltage is applied U. The angle θ at which the beam leaves the prism depends on the refractive index of the prism material, and therefore on U. The resolution of the scanner is defined as the ratio of the maximum angle change Δθ to the beam divergence δθ. The Δθ / δθ value for a prism electro-optical scanner can reach ~1 10 2 .

3.2.2. Application of the acousto-optic effect in deflectors and for other radiation transformations

The operation of acousto-optic devices is based on the interaction of simultaneously propagating optical and sound waves in a substance. Back at the beginning of the 19th century. T. Seebeck and D. Brewster discovered a change in the refractive index of light under the influence of elastic mechanical stress P substances, which leads to artificial optical anisotropy, manifested in birefringence and dichroism. This is the so-called elastic-optical effect (photoelasticity, acousto-optical effect), explained by the deformation of the electronic shells of atoms and molecules, the orientation of anisotropic molecules, etc. Under the influence of mechanical stresses introduced by a sound wave, alternating stripes with different refractive indices appear in the substance, propagating with sound frequency v sound If a light beam with transverse dimensions comparable to the sound wavelength is also directed at the substance λ sound = v sound / v sound, where v sound is the speed of sound, the path of the light beam will periodically bend. This phenomenon is of little interest for optical information processing simply because of its low frequency. However, with increasing frequency v sound (during the transition to ultrasound), as predicted by L. Brillouin back in 1922, light experiences alternating bands with different P diffraction, similar to the diffraction of x-rays on atomic planes in a crystal.

To observe the acousto-optical effect (Fig. 3.4), a sound wave in the crystal is excited using an acoustoelectric transducer, which is a piezoelectric plate attached to the crystal or a thin film deposited on its surface (LiNbO 3, CdS, ZnO). Application to AC voltage converter U causes mechanical vibrations of the plate (film) and can excite sound waves in the crystal in a wide range of frequencies up to tens of gigahertz (1 GHz = 1 10 9 Hz), going into an acoustic absorber at the opposite end of the crystal (for example, epoxy resin with filler, bismuth alloy with indium, etc.).

There are two possible schemes for the diffraction of light by sound waves.

A coherent optical beam can be sent normal to the direction of propagation of the sound wave (Raman diffraction- Nata), and then at the output the light wave is divided into a series of beams symmetrically diverging at angles θ T to the incident beam

Where T= 0, ±1, ±2, ..., λ-wavelength of light. Condition T= 0 corresponds to the zero diffraction order, t= ± 1 - first order, etc. Thus, the energy of the incident beam is distributed among many beams. The ratio of the intensity of diffracted beams depends on the frequency and intensity of the sound, the length of the path traveled by the light in the zone of action of the sound wave (interaction lengthsL), For Raman-Nath diffraction to occur, the following condition must be satisfied: λ L/λ sv 2<<1 . With the opposite inequality, another type of diffraction is observed when light falls on the crystal not perpendicular to the direction of sound propagation (Bragg diffraction).

If the angle between the direction of the incident light beam and the normal to the surface of the crystal θ B (Fig. 3.4) satisfies a condition similar to the Bragg-Wulf condition for x-ray scattering

all the energy of the light beam is concentrated practically in a beam corresponding to the first order of diffraction.

The ratio between the intensities of the diffracted beam and the beam emerging from the crystal parallel to the incident one depends on the interaction length and the amplitude of the sound wave. In order for a significant part of the incident light flux (50-90%) to be diffracted at a sound intensity of 1 W/cm 2, the interaction length for various substances should be in the order of magnitude 0.1 - 10 cm. The sound frequency v 3 B usually exceeds 1 10 9 Hz, which corresponds to the so-called hypersound (ultrasound called elastic waves in the range from 15-20 kHz to 1 10 9 Hz, and hypersound - from 10 9 to 10 12 -10 13 Hz).

According to (3.11) with sin θ B = 1 (backscattering of light), the equality 2 must be satisfied λ sv = λ , which corresponds to a certain limiting sound frequency for a given λ . The value depends not only on the wavelength of the light, but also on the material of the acousto-optic cell, since the speed of sound propagation is different in different materials. In the visible region of the spectrum, the value ranges from several gigahertz to several tens of gigahertz.

Bragg diffraction has been successfully used in high-speed optical deflectors. An undoubted advantage of such deflectors compared to electro-optical ones is the ability to change the beam deflection angle by changing the sound frequency. However, in accordance with (3.11), in this case it is necessary to consistently change the angle of incidence of the light beam so that the Bragg condition is satisfied each time. This can be achieved by changing the direction of propagation of the incident beam or sound wave as needed, which significantly complicates the operation of the deflector. However, condition (3.11) for various angles is satisfied when using a diverging, rather than plane, sound wave (this is shown schematically in Fig. 3.4). Such a wave can be considered as a set of plane waves directed within a certain angular interval. For a given sound frequency, Bragg diffraction will be observed on that component of the sound wave for which condition (3.11) is satisfied. Obviously, the greater the divergence of the sound wave, the larger the angular interval over which the light beam can be deflected, changing the frequency of the sound. In this case, however, the length of the acousto-optical interaction decreases and to obtain the same intensity of the diffracted beam it is necessary to increase the intensity of the sound wave.

If an anisotropic crystal is used as the working medium of an acousto-optic cell, the picture of the occurring phenomena becomes more complicated compared to that considered, the diffraction conditions become dependent on the mutual orientation of the direction of sound propagation and the optical axis of the crystal, the position of the plane of polarization of light, etc. In this case, however, it may the range of changes in the frequency of sound in which the Bragg condition is satisfied will be noticeably expanded, and therefore the interval of the angular position of the diffracted beam at the same power consumption will be increased.

Using acousto-optical devices, it is possible to carry out not only one-coordinate, but also two-coordinate deflection of a light beam. In this case, deflectors with mutually perpendicular scanning planes can be combined in one acousto-optical cell. The number of distinguishable positions of the light beam (resolution) of an acousto-optical deflector can be 10 3 -10 4, and scanning can be carried out not only along a set of fixed directions, but also during continuous scanning, which is achieved by a stepwise or smooth change in the frequency of acoustic oscillations.

The performance of an acousto-optic deflector is determined by the time during which the sound wave passes through the active zone of the crystal, i.e., it is limited by the relatively low speed of sound. However, the switching time of the light beam can be less than 1 10 -6 s.

Acousto-optic deflectors can use many materials that weakly absorb sound vibrations and are transparent in the corresponding region of the optical spectrum: fused quartz, chalcogenide and other glasses of various chemical compositions (for example, As 2 S 3), tellurium dioxide TeO 2 (paratellurite), lead molybdate PbMoO 4 (wulfinite), as well as KDP, DKDP, LiNbO 3 crystals, etc. The fraction of energy of the deflected beam relative to the incident energy (deflection efficiency) of acoustoelectric deflectors is usually close to 50-70%.

A deflector based on the acousto-optical effect can be used as a modulator (it is easier to achieve 100% modulation if you use a diffracted beam rather than a transmitted one), as well as to perform other light wave transformations.

If broadband (rather than monochromatic) radiation is directed at a crystal with a sound wave introduced into it, then light of predominantly one wavelength will be deflected by an angle of 2θ B. This makes it possible to isolate a narrow spectral range of radiation from the incident beam. By changing the frequency of sound, the wavelength of diffracted light can be changed over a wide range spanning visible, ultraviolet and infrared radiation. This is the basis for the operation of high-speed tunable acousto-optical filters. The spectral half-width of such filters is 0.01 - 1 nm.

Since light in an acousto-optic cell is diffracted by a sound wave, i.e., by a “moving grating,” a shift in the frequency of light occurs due to Doppler effect. For light incident on the crystal in the direction of sound propagation, and for light propagating in the opposite direction (in quantum mechanical language, this corresponds to the processes of emission or absorption of a phonon), the frequency of light becomes equal, respectively v - v 3 B and v + v 3B. This phenomenon can be used in practice to shift the frequency of light up or down by an amount v 3 B, which can also be changed.

3.3. Optical transparency

Optical transparency(OT) is a flat device, the optical parameters of which (transparency, scattering, refractive index, polarization) under the influence of a control signal change from point to point over its area, i.e. the light beam passing through such a device or reflected from it, appears to be spatially modulated. Spatial modulation of light for optical transparency does not exclude, in addition, temporal modulation of signals. A transparency that allows both possibilities is called dynamic or space-time light modulator(PVMS). Using high-speed PVMS, it is possible to perform parallel processing of large amounts of information (images, pictures) in real time, which is difficult to achieve in electronic devices and systems. It is obvious that PVMS can be used not only for conversion, but also for parallel input of information arrays, as well as for its output and display, including in visual form. Finally, if the properties of the materials and the principle of operation of the transparency allow the “optical relief” to be preserved for some time, PVMS can be used as a high-capacity memory device.

Various physical effects are used to modulate the signal in OT. Modulation can be carried out either by applying electrical voltage to different parts of the transparency (EUT - electrically controlled banner) or by projecting an optical image onto it (OUT- optically controlled transparency). Devices are also possible in which the OT serves as a target in a cathode ray tube, and its parameters are controlled using a scanned electron beam. However, such devices (such as Titus, Eidofor and their modifications) as “non-solid-state”, requiring evacuation, high accelerating and control voltages, will not be considered below.

Most of the parameters introduced for optical modulators are also applicable to transparency. The most important and characteristic parameters specifically for banners are resolution, defined by the number of distinguishable lines per unit length (usually expressed in lines per millimeter), and energy sensitivity to control signal(joules per square centimeters). The ratio Ф m ах / Ф min) of the intensities of radiation passing through the transparency at maximum clearing and darkening is called optical contrast.

3.3.1. Electrically controlled banners

When creating a transparency, it is natural to strive to obtain the greatest possible spatial resolution, and if it exceeds ~ 10 lines/mm, which is quite realistic, then with a transparency area of ​​several square centimeters, individual electrical connection of each element using a separate conductor becomes almost impossible. Therefore, the EUT uses the so-called X- Y-addressing(two-dimensional, matrix, multiplex). In this case, parallel conductive transparent stripes (tires) are applied to a thin layer of modulating medium on both sides, so that on opposite surfaces they are oriented mutually perpendicular (Fig. 3.5, A). An electric field is created in the desired place on the transparency by applying it to the corresponding X- Y-tires control voltage, which causes a local change in the optical properties of the modulating medium at the point of their intersection. To carry out optical modulation over the entire surface of the transparency, the electrical signal must “run” through all points of intersection of the buses (with the number of lines and columns of 100x100, there are already 1·10 4 such points!). To do this, use a different sequence of addressing the control voltage to the EUT elements. It can be applied one by one to all elements (element-by-element addressing), simultaneously to all elements of the whole line with the desired distribution among the elements within a given line (line-by-line addressing), the same for columns, etc. However, in all cases, the control voltage is not applied simultaneously to all elements of the transparency, i.e., parallel processing of information in real time, strictly speaking, is excluded. Nevertheless, the EUT is the most important element of information processing systems simply because it ensures the conversion of electrical signals into optical ones, without which it is impossible to combine electronic and optical devices (multi-element photosensitive matrices play a similar role in converting optical signals into electrical ones).

The characteristics and parameters of the EUT are primarily determined by the material used as a modulating medium.

Many electro-optical crystals discussed in 3.1.2 are suitable for high-speed EUTs. In most cases, rotation of the plane of polarization of light under the influence of an applied electrical voltage is used as a modulating effect. To convert polarization modulation into amplitude, an EUT of this type is placed between a crossed polarizer and analyzer. Under the influence of voltage, the transparency brightens in the appropriate place. The widespread use of traditional electro-optical crystals in EUTs is hampered by high control voltage (more than 1 10 3 V).

Electrically controlled PVMS can be created based on ferroelectric ceramics- a mixture of lead zirconate and titanate pressed at high temperatures with the addition of lanthanum (PbZrO 3 + PbTiO 3 + La, abbreviated PLZT, in Russian spelling - TSTL). Depending on the ratio of the components and the sintering mode, CTSL ceramic plates are obtained that have certain electro-optical properties. The transparency of the plates with a thickness of about 0.1 mm in the visible region exceeds 90%, the linear dimensions are several centimeters, and the operating voltages are in the range of 100-200 V.

The use of CTSL ceramics in optical transparency is based on the orientation (reorientation) of the polarization vector under the influence of external voltage domains- regions of spontaneous polarization, having optical properties similar to uniaxial crystals and randomly oriented in the initial state of the EUT. As a result of the preferential orientation of the domains, birefringence is induced. If the ceramics are relatively coarse-grained (4-5 microns), its scattering properties change under the influence of an external electric field. In the latter case, the light passing through the transparency is modulated in amplitude without the use of crossed polaroids. At temperatures above the Curie point, the ferroelectric phase of CTSL ceramics is replaced by the paraelectric phase. In accordance with this, the EUT has either long-term memory or high speed (up to 10 -7 - 10 -9 s). Such an effective device, in addition, has a fairly low cost.

The most sensitive to control signals and economical are EUTs based on liquid crystals(LC) - complex organic substances that have the properties of a liquid (fluidity) and at the same time a crystal (anisotropy of properties, including optical ones). Liquid crystal state (mesophase) exists only in a certain temperature range. Beyond its limits, the liquid crystal turns into an isotropic liquid at high temperatures, and into a solid phase at low temperatures.

LC molecules have an elongated, cigar-shaped shape (they represent a kind of uniaxial “crystal”) and therefore tend to mutual parallel packing, and ultimately to anisotropy of the LC layer. Depending on the nature of the arrangement of molecules, several types of FAs are distinguished: nematic, smectic, cholesteric. In EUT, an LC is placed in a narrow space (3-30 μm) between two transparent substrates. Mutually perpendicular transparent electrode busbars are created on the inner surfaces of the substrates. These same surfaces are polished (rubbed) with translational (rather than rotational) movement of the substrate relative to the grinding material, or a thin SiO 2 film is sprayed onto them at an angle. This treatment leads to the fact that the LC molecules are oriented parallel to the plane of the substrate and, in addition, in one direction. For light directed perpendicular to the substrates, such a layer has maximum birefringence. If a voltage exceeding a certain threshold is applied to the cell, the LC molecules rotate parallel to the acting electric field and the LC layer no longer causes birefringence. At V=0 a large difference Δ is achieved P = P e - P 0 = =0.2÷0.4, which provides the maximum modulation depth even with an LC layer thickness of approximately 1 μm.

Using the orienting action of the substrates, rotating them relative to each other, the LC molecules can be twisted so that their long axes in the layers adjacent to one and the other substrate will be mutually perpendicular. Such a structure becomes optically active and rotates the plane of polarization by 90°. Under the influence of a voltage applied to the cell, the molecules rotate parallel to the field and the “twisted” state of the LC disappears. This is the so-called twist effect(from the English twist-twist), widely used in indicators of watches, microcalculators, etc.

To obtain amplitude modulation in an LC using birefringence or the twist effect, it is necessary to use two film polaroids. Direct amplitude modulation of light using an LCD is also possible. To do this, an insignificant addition of a dye can be introduced into the LC, the orientation of the molecules of which depends on the orientation of the surrounding LC molecules. Since the absorption of light by a dye depends on the orientation of the long axis of its molecules, by controlling the orientation of the LC molecules, it is possible to change the optical absorption of the device (effect guest-master). It is possible to introduce into the LC not coloring additives, but doping impurities that lead to ionic conductivity of the substance. Then, when an external voltage is applied, the flow of current causes a vortex, turbulent movement in the LC and the LC layer, which is transparent in the initial state, becomes cloudy gray (dynamic scattering effect or electrohydrodynamic effect).

The operating voltages of EUTs operating on various effects in the liquid crystal vary from several tens to several volts, and at fairly low flowing currents (for example, 1-3 μA/cm2). Liquid crystal devices, which are also characterized by high manufacturability and low cost, have a significant drawback - relatively low performance. The electro-optical response time for some LCs can be in the microsecond range, but the time it takes for the molecules to return to their original state is at least one to two orders of magnitude longer, so LC devices typically respond in the millisecond range. The switching time of the EUT decreases as the thickness of the LC layer decreases and the control voltage increases. By optimizing these parameters and using other techniques (both technological and power mode), the switching frequency of the device can be increased to 1 10 3, and sometimes exceed 1·10 4 Hz. This is still not enough to solve many problems of optical information processing, although it is quite acceptable, for example, for indicator instruments.

Another disadvantage of LC devices is associated with the limited temperature range of existence of the mesophase, which is several tens of kelvins (for example, from -(10-20) to +(40-50) ° C).

High performance, as well as an almost unlimited service life, can be achieved by using magneto-optical effects in ferrites in EUT. Difficulties in the use of ferrites, in particular garnet ferrites and orthoferrites (differing in the composition of rare earth elements and crystal structure), are associated with significant absorption of light in the visible region of the spectrum. Nevertheless, optical transmission acceptable for practice (-10% in the red region of the spectrum) is achieved, for example, in transparency based on orthoferrites YFeO 3, YFeGaO 3, ferrite garnets YGaScFeO, YGdGaFeO, Y 2 BiFeGaO 12, etc. Local

a magnetic field that causes a change in the optical properties of ferrite can be created using the so-called current loop(Fig. 3.5, b). After a short-term connection of current (exceeding a certain threshold value), the created magnetization of the section of the banner (shaded in the figure) can persist for an extremely long time. When reading information, Faraday rotation of the plane of polarization of light passing through the ferrite layer is used, as well as when it is reflected (magneto-optical Kerr effect), Magneto-optical EUTs with a magnetization reversal time of ~ 1 × 10 -8 s, an information capacity of at least 100x100 elements and a practically unlimited resource are quite possible. Some difficulties with the use of magneto-optical transparency are associated with switching fairly large control currents (~1A).

3.3.2. Operating principles of optically controlled banners

In the most common case, the OUT is (Fig. 3.6, A) a thin plate of electro-optical material MS with a layer of photoconductor FP and two continuous transparent electrodes deposited on it 3 (for example, layers of tin oxides, indium, indium-tin ITO, transparent films of platinum, gold, etc.), to which the voltage is connected U. Such a multilayer structure is placed between a polarizer P and analyzer A and a parallel beam of light Ф 0 is directed at it, for which the photoconductive layer is insensitive and transparent (reading light wave). A translucent mirror is installed between the OUT and the analyzer 4, with the help of which a light control wave F control is projected onto the photoconductive layer (through the MS), creating the desired picture and having a spectral composition corresponding, in contrast to F 0, to the sensitivity of the photoconducting layer. In the absence of control light, the resistance of the FP is high and almost all of the applied voltage drops across it. Under the influence of F control, the FP resistance decreases and the voltage is redistributed between the FP and the electro-optical layer, locally changing its optical parameter, for example, causing birefringence (longitudinal Pockels effect). If in the initial state the polarizer P and analyzer A are crossed, then in the absence of F control, the reading light F 0 will not reach the output of the entire device. In places illuminated by the control wave, the light Ф 0 will be modulated in phase or polarization and the device will, to a lesser or greater extent, become transparent to the light beam Ф 0 . The device under consideration makes it possible to perform a number of transformations: the picture created by the F control wave has a different spectral composition; unpolarized control light can be transformed into coherent light (using a laser as a source Ф 0); the beam intensity F can exceed the intensity F control, i.e., light amplification is realized; if the main planes of the polarizer and analyzer in the initial state are not perpendicular, but parallel, then the image created by the F control wave will be converted into a negative one, etc.

Typically, a photoconductor is a weakly conductive material, and the charge relief created as a result of projecting an image onto it persists for some time (the F control wave is therefore also called recording). If necessary, the recorded information can be erased by short-term uniform illumination of a suitable spectral composition. Thus, an ODT with a photoconductive layer with a long relaxation time can be used both as a two-dimensional random access memory device and for processing rapidly changing information (if the write-read-erase cycle time is short). It should also be noted that recording can be made not only by projecting an image, but also by scanning a focused and intensity-modulated beam.

It is possible to operate the OUT not only for transmission, but also for reflection (Fig. 3.6, b). At the same time, the structure of the OUT itself has been changed: the FGT and MS layers are separated by an opaque dielectric mirror 5. Reading light Ф 0 passing through a polarizer P and reflected from a rotating translucent mirror 4, is sent to the OUT, and then, passing through the electro-optical material, is reflected from the mirror separation layer 5, again passes through the MS and is sent to the analyzer A. The control (recording) light beam F control is directed to the OUT from the opposite side of the separation layer. Otherwise, the transparency in question works the same way as in the transmission scheme. Since the dielectric mirror is opaque, the input and output of the device are optically isolated; There is freedom in choosing the spectral composition of the reading light. Another advantage of the ODT operation scheme for reflection is that due to the double passage of the reading beam, the depth of its modulation also doubles.

3.3.3. Various types of optically controlled banners

The variety of problems that can be solved using ODTs and the optimization of their parameters for each specific case have led to the search for various designs, materials used for the photosensitive and modulating layers, the involvement of various mechanisms leading to light modulation, etc.

In optical transparency type Phototitus(Fototitus) amorphous selenium is used as a photoconductor, and a KDP or DKDP crystal is used as a modulating material. The banner is placed in a vacuum chamber and its temperature is reduced to approximately -50°C (usually using a thermoelectric refrigerator). Cooling reduces the operating voltage of the device to 100-200V, and the information storage time increases to 1 hour compared to ~0.2 s at room temperatures, i.e., we can assume that within a few minutes there is no noticeable decrease in the contrast of the recorded image. Recording is done by exposing the image in the ultraviolet or blue region of the spectrum, reading - in the red (for example, with helium-cadmium and helium-neon lasers). The erasure of the charge relief, and therefore the spatial distribution of birefringence, is carried out by uniform illumination from an additional source, for example, a pulsed xenon lamp. The duration of recording and erasing information in the Phototitus device is quite short and amounts to ~1·10 -4 s. With a DKDP crystal thickness of about 100 microns, the spatial resolution of the transparency is no worse than 20 lines/mm. The widespread use of the device in modern information processing systems is hampered by the need for vacuuming and cooling.

Interesting aspects are the use of CTSL ceramics in optically controlled PVMS. In this case (Fig. 3.6, A) A photoconductive layer, for example polyvinylcarbazole (PVC), is applied to a plane-parallel ceramic plate, and then transparent electrodes are applied to both outer surfaces. The structure is glued to plexiglass, which is slightly bent, as a result of which mechanical stress is created in the ceramics. This leads to the fact that the electrical domains in the ceramic plate, the direction of which was chaotic in the initial state, are oriented along the direction of mechanical stress, in pairs and antiparallel, so that the resulting polarization of the ceramic plate is zero. Electrical voltage is applied to transparent electrodes U, however, somewhat less than what is required to reorient the domains along the direction of the external electric field. If, without removing U, the structure is illuminated, the resistance of the photoconductive layer will decrease, most of the voltage will be applied to the ceramic plate, which will lead to the orientation of the electrical domains along the direction of the electric field. Thus, when an image is projected onto a transparency in illuminated areas, birefringence disappears. Reading of the recorded information can be done using a polarizer and analyzer, erasing by illuminating the entire plate with U =0.

In addition to birefringence, the scattering effect is used in OUTs based on coarse-grained CTSL ceramics. In this case, mechanical stress is not created in the plate. Under the influence of external electrical voltage and uniform illumination, the plate is polarized. The polarity of the external source is then reversed, but the voltage is set lower so that repolarization does not occur. If a control light beam is now directed at the OUT, then in the illuminated areas the domains will be misoriented, which will lead to local scattering of light. To erase the recorded information, the transparency is uniformly illuminated with the polarizing voltage turned on, as a result of which the domains are oriented parallel to the field and the plate becomes transparent.

Finally, in OUT based on fine-grained CTSL ceramics it is possible to use reverse piezoelectric effect-change in the geometric dimensions of the body under the influence of an external electric field. In a transparency of this type, one of the electrodes is a mirror-reflective layer (Fig. 3.7, a). First, the plate is uniformly illuminated from the side of the photoconductive layer, and the voltage necessary to polarize the ceramic is applied between the transparent and opaque electrodes. The polarity of the source is then reversed, simultaneously reducing the voltage to a level insufficient to reorient the electrical domains in the dark. If an image is projected onto the OUT, then in illuminated areas the resistance of the photoconductive layer will become small, resulting in a reorientation of the domains. This will cause local mechanical stresses and a geometric relief will appear on the reflective layer, reproducing the recorded image (Fig. 3.7, 6). In this case, the curvature of the OUT surface usually does not exceed several tenths of a micrometer. This difference in the path of the light beam, however, is sufficient to read the image.

The described PVMSs based on induced birefringence, controlled scattering, and geometric relief are called respectively Ferpic(Ferpic-Ferroelectric Picture), Kerampic(Ceramic-Ceramic Picture), Ferikon(Fericon - Ferroelectric leonoscope).

An optically controlled transparency can be built on a material that has both photosensitive and electro-optical properties. The so-called PROM device(PROM - Pockels Readont Optical Modulator) is designed as follows. On a bismuth silicate plate Bi 12 SiO 20 (it is possible to use Bi 12 GeO 20, ZnS, ZnSe, CdS, CdSe, ZnO and other materials capable of maintaining the polarization state for a long time) with a thickness of about 100 μm, thin dielectric layers (~ 3 μm) are applied on both sides ), and on top of them are transparent layers of platinum. There is no special photoconductor layer in the PROM device, since the photoconductivity of the electro-optical material is used. The sensitivity of Bi 12 Si0 20, in particular, falls in the spectral region of 0.4-0.5 μm, and at λ≥0.5 μm it drops sharply. The structure is connected to a constant voltage source (1000-2000 V) and illuminated with a xenon lamp flash. Electrons generated by light in Bi 12 SiO 20 move to the interface with the dielectric layer, are localized there at energetically deep centers and polarize the plate (no through current flows due to the presence of dielectric layers). The movement of electrons continues until the polarization charge compensates (screens) the external field. When the electrodes are short-circuited in the crystal, due to polarization, an electric field arises in the direction opposite to the external one.

If an image is projected onto the PROM structure in the blue-blue region (0.4-0.5 µm), the polarization field will disappear in bright areas, and remain unchanged in unlit areas. Image reading is carried out with linearly polarized red light (X> 0.6 μm), which does not cause changes in the crystal, but is spatially modulated in phase. Phase modulation is converted into amplitude modulation if, as usual, the structure is placed between a crossed polarizer and analyzer.

If an external voltage source of the same polarity is connected to the structure, in previously illuminated areas the polarization field will be compensated by the external one, and in unlit areas the electric field will act. As a result, when read, the positive image will turn into a negative one. Information is erased by uniform illumination in the blue-blue region of the spectrum at U= 0. The recording-reading time is ~1·10 -3 s, the memory can be stored for 1-2 hours, the spatial resolution of the transparency is several tens or hundreds of lines per millimeter. The disadvantages of the PROM device are high supply voltage, low image changing frequency (no more than 1 kHz).

A peculiar modification of the PROM is the PRIZ device (from the words “image converter”). Its difference is that a wafer of electro-optical semiconductor material (for example, bismuth silicate or germanate) is cut not parallel (as in the PROM device), but perpendicular to the optical axis, i.e., so that the external electric field applied to the structure does not cause modulation reading light. However, when the transparency is illuminated unevenly, as a result of the migration of current carriers generated by light, a transverse field component arises, which leads to a change in the refractive index due to the electro-optical effect. Identification of areas with the maximum illumination gradient turns out to be very useful in image processing, in particular in object recognition. In fact, with the help of the PRIZ device, spatial differentiation of the image, and without the use of a special optical processor.

In another modification of the considered OUT, the electrode layers are applied directly to the plate of the electro-optical crystal. In this case, the polarization transverse field that appears after the start of image exposure gradually decreases due to the passage of current (for bismuth silicate with a characteristic time of about 1 s). The device thus makes it possible to highlight changing details in the image, i.e., to produce temporal differentiation of the image.

Of independent interest are liquid crystal(LCD) optically controlled banners. Both the LCD FP structure and those with an opaque dielectric mirror between the layers are used. The undoubted advantage of such banners, like LCD-based EUTs, are low operating voltages, simple and cheap manufacturing technology; The disadvantage is significant inertia (~1·10 -2 s). Since LCs are high-resistance materials, for electrical matching it is also necessary to use high-resistance semiconductors (ZnS, ZnSe, CdS, Se, etc.) as photoconductors. The use of low-resistance photoconductors (in particular, silicon) in combination with LC (as well as other electro-optical materials) is possible in devices with photosensitive MIS structures.

OUT can use not only electro-optical effects, but also thermo-optical (thermal) method of recording information, based on changes in the properties of the liquid crystal during its phase transition under the influence of heating. In this type of OUT, a thin LC film is placed between ITO electrodes, which are opaque in the infrared region of the spectrum. If a laser beam is directed at such a structure, the radiation energy will be absorbed in the electrode layer and cause local heating of the LC. In an initially transparent LC layer, heating and then rapid cooling will lead to a “frozen” disorder of the structure, intensely scattering light. The recording can be erased by heating and then cooling the cell in an electric field created by a voltage applied to the electrodes.

An optically controlled transparency can be built on the basis of a material in which, at a certain temperature, a transition from a metallic state to a semiconductor occurs. Vanadium oxides, in particular, have such threshold properties, and among them the most suitable is vanadium dioxide VO 2 with a phase transition temperature of ~70° C. The manufacture of a banner involves applying a layer of VO 2 0.1-0.2 μm thick on a substrate made of glass, quartz, sital or other suitable material. A scanning laser beam is directed onto the VO 2 layer or an image of such intensity is projected that in illuminated areas, as a result of light absorption, the vanadium oxide layer heats up and passes from a semiconductor state to a metallic one. After the image is exposed, the transparency returns to its original state. To read information, you can use a change in either the absorption coefficient or the refractive index. The energy sensitivity of the transparency is not very low (1·10 -2 J/cm 2), the spatial resolution is several thousand lines per millimeter, the recording time can be increased to ~1·10 -8 s. The abbreviation used is OUT of the type in question - FTIROS(phase transformation reversible light reflector).

Other thermal-action OUTs are also possible, in particular using thermoplastics-plastics that can soften when heated and retain their shape after cooling (for example, polystyrene, polyvinyl chloride, etc.). A layer of photoconductor (usually polyvinylcarbazole) is applied to a glass plate with a conductive transparent layer of tin dioxide or metal, and a layer of thermoplastic is applied on top of it. Next, the surface of the thermoplastic is charged using a corona discharge, resulting in a potential difference between the surface of the transparency and the conductive electrode. When an optical image is projected onto the structure, the resistance of the photoconductor in illuminated areas decreases and the electric field in different places of the thermoplastic turns out to be different. If a current pulse is passed through an SnO 2 electrode, the thermoplastic layer will briefly heat up (to the softening point) and in places of a strong electric field the film will shrink, which will be fixed for a long time after the device has cooled. As a result, a surface relief is formed that repeats the recorded picture, and the reading light will be phase modulated. The image is erased by heating the film in the dark. OUTs are possible that use the photosensitivity of the thermoplastic itself (photothermoplastics), and then the need for a separate photoconductor layer disappears. The energy sensitivity of a thermoplastic device is high and comparable to the sensitivity of a photoemulsion; the spatial resolution is 1000-4000 lines/mm.

Most of the considered OUTs can operate in a mode where the intensity of both the recording and read light changes along the coordinates as smoothly as desired. To process digital information in the form of binary images, they use matrix OUT. This type of transparency includes many regularly spaced “photodetector-electro-optical material” cells, operating almost independently of each other and designed to perform operations on one bit of information. The design of a reflective-type matrix OUT is illustrated in Fig. 3.8, A. In contrast to the previously discussed structures, the light-insulating layer between the photoconductor 2 and the modulating medium 3 made in the form of a matrix of mirror-reflecting metal pads 5, separated by a resistive layer 4, opaque and non-photosensitive. External electrodes 1 made in the form of a metal mask with windows located in alignment with reflective areas on the optical separating layer. This ensures independent operation of the transparency cells and high recording reliability. As in the diagram shown in Fig. 3.1,6, For image reading, translucent mirrors, polarizers, etc. are used. KDP, ADP, LiNbO 3 crystals, etc. can be used as electro-optical material in matrix OUTs.

The disadvantage of such devices is their relatively low performance. To increase it, the electro-optical layer is applied not on top of a continuous resistive layer, but on a transparent substrate created on it (Fig. 3.8, b) integrated matrix of photosensitive silicon circuits with the necessary amplification elements (transistors) 6 . The operating speed of such photosensitive cells can be 10 -6 - 10 -7 s.

To provide optical memory, the electro-optical material does not necessarily have to have hysteresis properties. For example, ferroelectric ceramics are suitable for this, but at temperatures above the Curie point. The RAM of such a transparency (muna Latria, as it is called) is provided by an electronic circuit of photosensitive cells. Its duration is determined by the time of charge leakage through the reverse biased silicon r-p-transition (usually up to 1·10 -2 s), which in some cases is quite sufficient for optical information processing systems.

3.4. Optical memory

The advantages of optical modulators, deflectors, controlled banners and other elements of optical information processing systems cannot be fully realized without adequate optical memory devices with high capacity, recording density and speed, short search (sampling) time, high durability and reliability of information storage.

As noted above, an optical transparency, in which the modulating medium, even for a short time, can retain an optical parameter different from the equilibrium state after the cessation of external influence, is essentially a RAM device (for many cases, a storage duration of only 10 -8 -10 -9 s is sufficient ). Below, in addition to storage devices (storage) of this type, long-term (permanent, archival) memory devices will be considered.

The creation of optical memory is dictated by the fact that magnetic memory used in electronic devices faces serious difficulties due to the increasing demands placed on information processing systems. In addition to a dramatic increase in recording density and speed, a significant reduction in size, weight and cost, optical memory devices allow parallel recording and retrieval of two-dimensional information arrays. However, optical memory uses both recording methods - both parallel and sequential. Although optical memories allow recording information directly in analog form, devices with recording in digital binary form will also be considered below, which provides greater accuracy, noise immunity, and versatility of recording.

3.4.1. Permanent optical memory with a sequential way of writing and reading information

A simplified block diagram of recording sequential type information using a scanning laser beam is shown in Fig. 3.9. To ensure high recording density, they try to focus the laser radiation into a spot of the smallest possible size (due to diffraction, these dimensions cannot be less than the radiation wavelength and are usually close to 1 µm). The beam, modulated in the required manner, is directed through the lens onto the storage medium, and its geometric position is set by an optical two-coordinate deflector. In the simplest case, silver halide emulsions deposited on a transparent substrate are used as such a medium. Photographic emulsions, which, of course, provide permanent (irreversible) memory, have high resolution (thousands of lines per millimeter) and high energy sensitivity of 10 -4 - 10 -6 J/cm (for various types of photographic emulsion). After development and fixation, the image is projected using a reading lens onto a radiation detector, for example onto a photodetector array. The light source in this case is the scanning beam of the same laser (the modulator is open when reading).

The search for media for optical memory with an optimal combination of sensitivity, resolution and other characteristics has led to the use of many other materials, in particular photoresistors, in addition to photoemulsion. All these materials require processing using liquids, and quite long-term, at best, a few seconds (for some resistors, “dry” heat treatment is possible at a temperature of 150 - 200 o C).

Bitwise recording of information can be carried out by burning (melting) using a focused laser beam through holes about 1 µm in size in thin (~0.05 µm) layers of Pt, Bi, Rh, As, Cr and other substances deposited on a transparent, for example polyester, basis. The advantage of such a recording, which can be read by the same laser, but with a lower beam intensity so as not to damage the recording, is a high signal-to-noise ratio, high reliability and a long service life. Another method of recording in the form of a coded sequence of pulses is to create micro depressions or spots (pits) on a polyvinyl chloride or polymethacrylate plate with a layer of tellurium deposited on its surface (20 - 40 μm), as a fusible material that strongly absorbs infrared radiation.

Finally, microbumps can be formed in the metal layer. In this case, refractory materials (Ti, Pt) are used, and a well-evaporated material is used as a dielectric sublayer. Under the action of a laser beam, the metal film is not burned out or melted, and as a result of evaporation of the sublayer, a bulge is formed in the appropriate place. The film with the recorded information is covered with a layer of transparent material, which is intended primarily to protect the information carrier from damage and guarantees a long service life. If the protective layer is relatively thick (as is usually the case), foreign particles, scratches and other microdefects on its surface are out of focus of the reading object and, therefore, slightly distort the signal.

The memory film structure can be reinforced or deposited on a rotating disk of glass, quartz, glass-ceramic or polymer. Information is recorded on tracks with a pitch of 1.5 - 2 microns, which, with a disk diameter of 30 cm, allows recording more than 1·10 10 bits of information. This capacity is enough to encode a 20-30 minute color television program, or several tens of thousands of pages of typewritten text, which is comparable to the information in the Great Soviet Encyclopedia.

Difficulties in using optical disks are associated with the need for precise alignment of the laser head and the storage medium. Reliable reading is almost impossible without a special servo system that ensures accurate tracking and following of the scanning beam along the information track. Obviously, in order for the marks on the disk not to be “smeared” during recording due to its rotation, the laser radiation pulses must be quite short (~1·10 -8 s). The photodetector used for reading must have high speed (10 -8 - 10 -9 s).

A comparison of magnetic and optical memory indicates the undoubted advantages of the latter. Optical memory is distinguished by high quality recording and playback with a much longer service life (there is no mechanical contact between the reading device and the storage medium), high recording density, long shelf life (tens of years instead of 1 g with magnetic recording) and much lower cost. The disadvantage of the considered optical memory devices is that they are written only once; making copies is, of course, possible. To replicate a recording from a primary optical disk (without a protective coating), a metal original is produced using electroplating methods, and plastic copies are pressed from it in the required quantity. A highly reflective film (aluminum) is applied to the secondary disks on the recording side, and a transparent protective layer is applied on top of it. Optical discs of small diameter (11.5 - 12 cm) used for high-quality sound reproduction are called CDs. In a similar way, it is also possible to replicate discs for video playback.

3.4.2. Optical RAM

Devices random access memory, unlike those discussed above, must be reversible, i.e., after a short-term erasing effect, they must be ready to record new information. The properties of the medium used should not change with a large number of write-erase cycles and allow information to be written and erased in the shortest possible time. In random access optical memory, many physical effects are used, in particular, the previously discussed devices Phototitus, PROM are used, as well as photoconductor structures - LC, photoconductor - ferroelectric CTSL ceramics and many others.

Optical memory devices that use recording on photochromic materials- substances whose absorption changes reversibly under the influence of optical radiation directly, i.e., without any manifestation. Among the large number of photochromic materials, polymers, silicate glasses, and alkali halide crystals (KS1, NaF, CaF, etc.) are quite widely used. During the photochromic process, a substance, absorbing light quanta, passes from the initial state to a photoinduced one, characterized by a change in optical transmittance in a different spectral region. To write and read information, therefore, radiation with different wavelengths is required (for example, 0.2 - 0.4 μm when writing and 0.4 - 0.7 μm when reading). The reverse transition to the initial state occurs spontaneously, but can be noticeably accelerated by the action of light absorbed in the photoinduced state, so when reading, the light energy must be several orders of magnitude higher than when writing.

The storage time of recorded information varies for different materials: from 1·10 -6 s to several years. Photochromic materials are characterized by short recording times (~1·10 -8 s) and high resolution (~3000 lines/mm). Recording can be done in different planes of the photochromic material, and the transition from one plane to another is carried out by changing the focal length of the recording and reading lenses. Despite some loss of optical contrast, it is possible to use multiple layers for recording, which leads to enormous volumetric recording density.

Memory devices based on magneto-optical effects use layers of ferromagnetic materials with high coercive force, capable of maintaining magnetization for a long time after the external magnetic field is turned off. In a thin layer of such material, under the influence of radiation from a laser beam, local heating occurs and, if the temperature exceeds the Curie point, the magnetization vector changes abruptly. The rotation of the plane of polarization of the reading light passing through the layer (Faraday effect) turns out to be different in pre-illuminated and unilluminated areas. Reading can also be carried out by reflected light, using the already mentioned magneto-optical Kerr effect.

To erase information recorded by a ferromagnetic layer, it is heated with a light pulse or some other method in the presence of a magnetic field, as a result of which its original magnetic state is restored. Although magneto-optical effects are used when reading information in the cases under consideration, this recording method is also commonly called thermomagnetic. Among the suitable materials for thermomagnetic recording, manganese bismuth MnBi has been well studied, having a Curie temperature of approximately 360 o C, fairly good resolution (10 3 lines/mm), short recording time (~ 1·10 -8 s), long storage life of recorded information, as well as a job resource. MnA1Ge, MnCuBi alloys, lanthanide oxides (for example, EuO, etc.), bismuth-containing garnets, as well as amorphous Tb 1- films are used as storage material in magneto-optical disks. X Fe x and compounds based on them (with the addition of cobalt, chromium, cadmium, gadolinium, etc.).

Films Tb 1- X Fe x are ferrimagnetic, i.e. the magnetic moments of terbium and iron atoms are oriented antiparallel, and in a certain range X anisotropy appears in the film with an axis perpendicular to the film plane. Recording, reading and erasing information is done in almost the same way as in the case of a memory device based on MnBi. The advantage of amorphous films Tb 1- X Fe x consists in the absence of scattering effects at grain boundaries, in contrast to polycrystalline MnBi or other similar materials. Curie temperature Tb 1- X Fe x depending on X varies within 40 - 140 o C, resolution - more than 1·10 4 lines/mm, recording - erasing cycle time - about 1·10 -8 s. The information capacity of magneto-optical disks with a diameter of 30 cm is 10 9 - 10 10 bits.

Recording in chalcogenide glasses containing sulfur, tellurium, arsenic and other elements (for example, As - Se, Sb - S, As - Sb - S, As - Bi - S, Ge - S, Te - Ge) is based on local heating by a laser beam - As, etc.). However, the memory mechanism in this case is different. When the devitrification temperature is exceeded, but below the melting point, a phase transition occurs from the amorphous state of the material to the crystalline state, as a result of which the refractive index of light changes, which is used when reading information. The transition of the film to the amorphous state (erasure) is carried out by heating to the melting temperature and subsequent rapid cooling. Recording on such films, as with thermomagnetic recording, lasts for a long time, the energy sensitivity is approximately the same, the resolution exceeds 1·10 lines/mm, however, the optical transmittance of glasses can reach ~~80% (1·10 -3 for MnBi). Amorphous TeO films obtained by vacuum evaporation are also used for reverse recording. X(X=1.1÷1.2). Under the action of a laser beam, a photothermal transition occurs, as a result of which the optical transmittance and reflection of the film noticeably changes. Optical discs operating on this principle allow multiple re-recordings (for example, of music programs) up to 1·10 -6 times.

The operation of high-speed multi-channel banners with reverse memory can be based on an element proposed in the early 80s and called transphasor. This device uses the optical nonlinearity of the material, which manifests itself in a change in the refractive index with increasing intensity of the incident light. In a transphasor, a light beam is directed onto a plane parallel plate of a nonlinear crystal, forming a Fabry-Perot interferometer, the role of mirrors in which can be played either by natural (polished) crystal faces or by thin translucent metal films deposited on them. The thickness of the plate is chosen such that at low light intensities, when the crystal can be considered linear, the phase difference of the rays repeatedly reflected from the mirror faces is equal to an odd number π and the beam intensity at the output is low (Ф out = 0). This condition is violated in the region of high light fluxes (quite achievable using lasers), when the value P, which means the optical path length begins to increase. This causes an increase in the intensity of light inside the resonator, which, in turn, leads to an even greater increase P etc. The device jumps into a state with transmission close to unity.

In practice, two laser beams are directed at the transphasor. One of them has a constant intensity F post corresponding to low transmission, but close to the threshold state. A small illumination with another beam (F control) switches the transphasor to a state with maximum F output. Due to the F post, such a state can be maintained for as long as desired, and when the F post is turned off, the crystal jumps into its original state, i.e., it will no longer miss the second ray F control. So the transphasor is optically bistable element, which can be considered as an optical analogue of an electronic transistor.

A transphasor switches much faster than a transistor. Indeed, the speed of the transphasor is limited by the time it takes to establish the light field inside the resonator, and it is of the order of magnitude equal to hn/s, i.e., with the thickness of the plate h=10 µm is ~1·10 -13 s. In any case, the operation of the transphasor in the picosecond range (10 -12 s) is quite realistic. Its transverse dimensions are limited by the cross section of the laser beam, i.e. the transphasor can be as miniature as a transistor. When using, for example, indium antimony or gallium monoselenide as a material for the transphasor, the switching energy is only 1·10 -15 J with a constant prethreshold illumination power of ~10 mW. Difficulties in the implementation of devices based on transphasors are associated with the fact that the materials used for this require cooling.

Other effects and materials can be used in RAM devices (see d 3.3).

3.4.3. Principles of holographic recording of information

Holographic memory is based on recording an interference pattern formed by the addition of a light wave reflected from or transmitted through an object (object wave) and a coherent wave coming directly from the light source (reference wave). If the recorded picture (hologram) is then illuminated by the same reference source, located relative to it in exactly the same way as during recording, then as a result of the interaction of the reference wave with the hologram in space, a wave is formed that restores the image of the object, matching it in shape and spatial position (a mandatory requirement for the light fluxes used is their coherence).

It is important that a hologram, unlike a photograph, records not only the distribution of amplitudes, but also the distribution of phases of the object wave relative to the reference one. Information about the phase relationship between the object and reference waves is reflected by the pattern and frequency of the fringes of the interference pattern, and information about the amplitude is reflected by its contrast. Using a hologram, the amplitude-phase distribution of the wave field is thus restored, i.e., a copy of the object wave is created, and not just the light-contrast characteristic of the object, as with conventional photography. This explains the extremely high information capacity of the holographic method of recording information.

Since during recording, light from each point of the object falls on the entire surface of the hologram, each small section of it is able to restore the image of the object, although with a lower signal-to-noise ratio and with a loss of resolution of small details. Therefore, the quality of hologram recording is slightly affected by various defects - spots, specks of dust, scratches, etc. This ensures high reliability and noise immunity of holographic recording. A quantitative characteristic that reflects the ability of a hologram to transform a reference wave into a reconstructed image is called diffraction efficiency and is defined as the ratio of the power of the light flux in the reconstructed image to the power of the light flux in the reconstruction wave.

Holograms are often recorded on photographic plates, and different sections of the photographic plate can be used to record different holograms. The minimum sizes of these areas are theoretically determined by diffraction phenomena, but in practice the recording density turns out to be noticeably lower.

The optical design of holographic recording (Fig. 3.10) usually includes a beam splitter (for example, a translucent mirror), which is installed in the path of the laser beam that illuminates the recorded object and forms an object wave. Using deflecting devices (deflectors, mirrors, etc.), the reference wave is directed to the desired area of ​​the photographic plate (as is the object wave). If the object of recording is an optical transparency, then in each such area, usually not exceeding 1 - 2 mm 2, not one bit of information is recorded, but a whole image (information page with a capacity of 1·104 - 1·10 5 bits). One and the same section of recording material can contain several superimposed holograms that do not affect each other, if the angle of incidence of the reference beam is changed each time during recording. Of course, when reading, its direction must change accordingly to be the same as when writing. However, it must be kept in mind that an increase in the number of superimposed holograms leads to a decrease in diffraction efficiency.

Until now it was assumed that the thickness of the recording medium is much less than the period of the interference pattern ( two-dimensional holograms). In the opposite case, the hologram is not a flat pattern of interference fringes, but a volumetric structure that repeats the spatial pattern of interference between the object and reference waves. Three-dimensional method of recording holograms as the most general one was proposed and justified in 1962 by Yu.N. Denisyuk. When reconstructing an image, a volumetric hologram acts like a three-dimensional diffraction grating. Reflection of light from interference layers (Bragg) occurs only when a condition similar to (3.11) is met: , where d-distance between adjacent layers; θ B- the angle between the incident light and the plane of the layers.

Thus, a three-dimensional hologram has spectral selectivity (selectivity), i.e., sources with a continuous spectrum (for example, an incandescent lamp, the Sun) can be used to restore the image. In this case, the hologram will “select” radiation of the wavelength that was used during recording (two-dimensional holograms do not have spectral selectivity and the reconstructed image will be blurred). The property of three-dimensional holograms to reproduce the spectral composition of the recording radiation makes it possible to significantly increase the information capacity by recording multiple images in the same area of ​​the recording medium, each time using radiation with a different wavelength. The desired image can be read independently, for which it must be reconstructed using radiation of the appropriate wavelength. Another advantage of a 3D hologram is that it reconstructs only one image. A two-dimensional hologram transforms a reference wave into both an object wave and a so-called conjugate wave, which creates a false image, which can make it difficult to read the information.

Holographic recording can be done in both digital and analogue forms; used in both permanent and reverse memory devices, including real-time data processing systems.

The development of holographic recording methods has led to the use of many materials suitable for this purpose. At the same time, the most important requirement for them is high resolution. For special silver-halide photographic emulsions, it reaches 3000 - 5000 lines/mm (in the red region of the spectrum). Some loss of resolution, but a gain in diffraction efficiency, can be obtained by using dichromated gelatin and photoresistors of various types to record holograms. A hologram fixed on a photographic emulsion, due to blackening, modulates the light flux in amplitude, but at the same time its phase modulation occurs, since the thickness and refractive index of the emulsion simultaneously change. A hologram obtained on a transparent material modulates light only in phase. In accordance with this, they distinguish phase and amplitude holograms. In the first case, the diffraction efficiency of the hologram can approach 100%, in the second, it is usually several percent (the photographic emulsion with the recorded hologram is therefore bleached).

For holograms that can be rewritten many times, many of the materials used in other optical recording methods are used. To obtain phase holograms immediately after exposure, photothermoplastics are used, which provide high diffraction efficiency, as well as other reversible materials: photochromic, magneto-optical, chalcogenide glasses, etc.

For three-dimensional recording of holograms, a reoxane-polymer material with the addition of a sensitizer dye and anthracene is used. Writing in reoxane is based on the photoinduced oxidation reaction of anthracene, resulting in a change in the refractive index with virtually no decrease in optical transmittance. In this case, the depth of hologram recording can reach several millimeters.

3.5. Digital and analog conversions in the optical path

3.5.1. Perform basic logical operations

The devices considered make it possible to implement a variety of calculations and transformations of information in both analog and digital forms. The analog form of processing is attractive because all kinds of sensors, receivers and means of displaying physical quantities operate in the mode of continuous change of input and output signals, while the digital form, as already noted, is characterized by higher accuracy, reliability and noise immunity, since it is based on the identification of easily distinguishable states.

Let us first consider how elementary logical operations are performed by optical methods. We will depict (Fig. 3.11) an optical element that transmits light in the presence of a control signal X, unshaded rectangle ( T), and the element that transmits light in the absence of a control signal is a shaded rectangle ( T). In Fig. 3.11 arrows indicate control signals ( X, X 1 , X 2) and a controlled beam (optical “power”). The optical signal at the output is indicated at.

In the case when the optical beam passes sequentially through the controlled elements T, is being implemented logical multiplication operation (at= X 1 ^ X 2, I). This can be easily seen using the example of a device with two inputs (Fig. 3.11, A). Light will not pass through the device ( at= 0) as in the absence of both control signals ( X 1 =0, X 2 =0), and when a signal is applied to only one of the elements ( X 1 =l, X 2 =0 or X 1 =0, X 2 =1); light enters the output ( at= 1), only if control signals are applied to both one and the other inputs ( X 1 =l, X 2 =1).

To perform a logical summation operation (y= X 1X 2, OR) controlled optical elements T“connected” in parallel (Fig. 3.11, b). In this case, for light to reach the output ( at= 1) it is enough that the control signal is applied to at least one of the elements ( X 1 =0, X 2 =1 or X 1 =l, X 2 =0). Of course y=l and when a control signal is applied to both elements ( X 1 =l, X 2 =1).

Inversion operations (at= , NOT performed using one element that transmits light in the absence of a control signal, i.e., using an element (Fig. 3.11, c). It is easy to verify that if two elements are connected in parallel, the AND-NOT operation is implemented (Schaeffer stroke, at = X 1 X 2), and the OR operation is NOT (Pierce arrow, y = X 1 ↓X 2) - when they are connected in series (Fig. 3.11, g, d). If light passes sequentially through the elements T and , the prohibition operation is performed at=X 1 ←X 2 (Fig. 3.11, e). The meaning of this transformation is that in the absence of a prohibiting signal ( X 2 =0) light passes through the device at X 1 =1 and does not pass when X 1 =0. When a prohibiting signal is given ( X 2 = 1) light does not reach the output at any value X 1 .

In Fig. 3.11, g, h demonstrated how optical methods can be used to perform equivalence operations (at = X 1 ~ X 2) and disparity (at = X 1 X 2). In this case, the control signal X 1, like X 2, hits two optical elements at once T And . It can be seen from the figure that in the equivalence circuit, light reaches the output when the states of the input signals coincide, i.e., as when X 1 =0, X 2 =0, and when X 1 =1, X 2 =1. If a control signal is applied to one of the inputs and not to the other, there is no light at the output. In a disparity circuit, on the contrary, the output light enters under different conditions X 1 and X 2 (X 1 =1, X 2 =0 or X 1 =0, X 2 =1) and does not fall under the same X 1 and X 2 (X 1 =1, X 2 =1 or X 1 =0, X 2 =0). Combining elements T and , other transformations can be performed.

3.5.2. Conversions over digital and analog paintings

As optical elements T and matrix optical transparency cells, controlled either optically or electrically, can be used. It is important that with the help of banners it is possible to process information simultaneously through many channels in parallel, i.e. transformation over the paintings. One of the possible schemes designed for this optical processor(calculator) is shown in Fig. 3.12.

A collimated beam of light F control is directed to the OUT, which operates on reflection and records the result of calculations. The OUT reading circuit (using the F 0 beam) includes a crossed polarizer P and analyzer A, so that in the absence of a control optical signal F control at the input, the light intensity at the output (F out) is also zero. This banner must have memory and, in addition, allow, by changing the diet, U transform a positive image into a negative one and vice versa (see § 3.3). A transparent transparency is placed in the path of the beam F control T, designed to form the necessary pictures and project them onto the OUT.

Let's write on the OUT what is specified by the banner T image, and then, without relieving tension U, another image. When reading, those places of the OUT that were hit by the control signal when projecting the first, second, or both images will turn out to be light. Obviously, this is how it is produced addition operation images (paintings). Picture multiplication operation can be carried out if on the path of the beam F there is a control behind the transparency T(or in front of it) place another banner T". If with banners T And T" specify the desired pictures, then when they are simultaneously recorded on the OUT, the light will only hit those places; against which they are transparent as T, so and T", which is what is required for the multiplication operation.

If after recording one image under voltage U record another image under voltage - U (using one banner T), then mutual erasure of signals will occur in those places of the OUT that were hit by light when projecting both one and the other image. When reading (U=0) the output signal will be present only in those places where light was present in one image and not in the other, or vice versa. This transformation corresponds to picture subtraction operation.

By changing the recording sequence and the power supply mode of the OUT when writing and reading, you can perform many other transformations on images. If, for example, from a picture containing many elements, we subtract the same one, but differing in the absence or presence of some new details, then in the resulting image both of them will be represented in light places on a black background. It is often more convenient to observe such “extra” or “missing” details against the background of a weak, low-contrast image of the original picture, which is quite simple to achieve by introducing, when recording one of the pictures, uniform illumination of the entire OUT of suitable intensity, i.e. introducing the so-called optical shift. This technique can also be used to eliminate the background, if there is one in any picture (for example, a veil in a photograph). To do this, the desired picture is recorded on the OUT at voltage U, and then under tension - U record uniform illumination. At a certain exposure of the illumination, the background in the resulting picture will disappear (of course, if the background and illumination are strictly uniform).

Using a processor, the circuit of which is shown in Fig. 3.12, can be produced spatial differentiation of images. To do this, you need to record the original image on the OUT, and then subtract from it the image of the same picture, but slightly shifted or defocused. In this case, the resulting picture on a black background will depict not the objects themselves, but their contours. By introducing an optical shift, you can simultaneously observe a low-contrast image of the original picture, the details of which are bordered by light lines. The benefit of such a transformation becomes obvious if one performs spatial differentiation not of a digital picture, but of a halftone image. In this case, at the output of the OUT, black places will remain black, white places will also become black, and only those places that correspond to the largest illumination gradient will appear. With the help of this analog conversion, Consequently, hard-to-see small details can be highlighted in the imaged object.

The use of other operations on images also has important practical aspects. The multiplication operation, for example, can be effectively used to reduce the impact of clutter on the image. For this purpose on banners T And T" (Fig. 3.12) two images of the same object are formed. If these images contain random noise (uncorrelated noise), then, by multiplying images, i.e. passing a beam of light sequentially through the transparency T And T" and recording the resulting picture on the OUT, at its output an image is obtained with an increase in the signal-to-noise ratio compared to images on T And T". Some interference will still pass through to the output, but only those whose spatial position on the banners T And T" will coincide by chance.

3.5.3. Transformations in coherent beams

The use of coherent radiation expands and enriches the possibilities of optical information processing.

In Fig. Figure 3.13 shows a simplified diagram explaining the operation of an optical processor using spatial filtering. To the plane S BX (input) a parallel beam of coherent radiation collimated from a point source is directed. The design includes two spherical converging lenses with a focal length F, located at a distance of F and 3F from the input plane. If in plane S BX place an optical transparency (for example, EUT), forming any picture F in, then in accordance with the laws of ray optics in the plane Sout (output or correlation) the picture will be reproduced, inverted with respect to F input. In plane S f, which is called frequency or filtration plane, a distribution of amplitude and phase of the light field will be formed, proportional to the spectrum of spatial frequencies of the pattern F in (this will be done Fourier transform function Ф in). Any transparency placed in a plane S f essentially performs the function of a spatial filter. A screen with a rectangular hole, for example, is a two-dimensional spatial low-pass filter, an opaque rectangle is a two-dimensional high-pass filter, a narrow slit is a one-dimensional spatial low-pass filter, etc. It is attractive to use an EUT as a spatial filter, which allows filtering that varies in time.

Opportunities that are practically inaccessible to computers are opened by the use of holographic methods in optical information processing systems. Introduction to Plane S f of holograms allows you to analyze the spatial spectrum of the pattern formed in the input plane S BX, in particular, to solve such an important applied problem as pattern recognition.

The selection of the desired object includes the preliminary production of the so-called matched filter and subsequent identification of the object in the array of information arriving at the input plane of the processor. Suppose that on a page of text it is necessary to identify and determine the coordinates of some character, for example, numbers or letters. To fabricate a matched filter in the input plane S BX place a banner with the image of this sign. In plane Sf a light wave corresponding to its spatial spectrum will be formed. If using a beam splitter and mirrors on a plane Sf simultaneously send a reference wave coherent with the wave illuminating the input plane S BX(Vander Lugt scheme), then in the plane S f an interference pattern is formed, which is Fourier image hologram object placed in a plane S VH. The recorded hologram represents a matched filter of the spatial frequencies of this object. This hologram contains amplitude and phase information about the sign that is depicted at the input, and it can be used to recognize that sign. To do this, the hologram filter is left in the plane S f, the reference wave is removed, and in the plane S BX place a page with text, illuminating it with the same light source. The holographic method of pattern recognition is based on the fact that if a hologram is restored by the radiation of the object that was used during its registration, i.e., its light field is used as a reading wave, then the image of the point source used will be restored. In the exit plane S OUT, therefore, in places corresponding to the image of a given sign, images of the light source, i.e., bright spots, will appear. Light coming from other signs in the plane S BX, will not restore images of a point source and there is no light in the corresponding places of the output plane.

It is obvious that using the described method it is possible to identify arbitrarily complex characters, whole words, phrases, drawings, for example, fingerprints, an image of a section of terrain, etc. Using, again, a reverse medium with a sufficiently high speed to record holograms, it is possible to conduct processing a whole array of information in real time. Unique applications of optical processors of the type under consideration open up if we use the possibility of synthesizing spatial-frequency filters using a computer.