US3245011A - Ferroelectric signal-translating device having voltage-controlled signal delay - Google Patents

Description

G. RUPPRECHT Filed May 31. 1961 FERROELECTRIC SIGNAL-TRANSLATING DEVICE HAVING VOLTAGE-CONTROLLED SIGNAL DELAY April 5, 1966 BIAS and CONTROL VOLTAGE SOURCE //Vl E/V7'0/? GEORG RUPPRECHT POM/mm ATTORNEY LOAD SOU IfQC E O MICROWAVE POWER FIG. 3

United States Patent 3 245 011 FERROELECTRIC SlGl lAll -TRANSLATING DEVICE HAVHNG VGLTAGE-QONTROLLED SIGNAL DE- LAY Georg Rupprecht, West Newton, Mass, assignor to Raytheon Company, Lexington, Mass, a corporation of Delaware Filed May 31, 1961, Ser. No. 113,927 7 Claims. (Cl. 33331) This invention relates generally to signal-translating de vices using ferroelectric materials and, in particular, to signal translating devices using ferroelectric materials having the perovskite crystal structure.

Ferroelectric materials, such as barium and strontium titanate, have dielectric constants which vary with the strength of an applied field. In the usual device using ferroelectric material, an applied voltage is used to pro duce an electric field in the ferroelectric material. Varying this applied voltage varies the dielectric constant of the ferroelectric material and therefore the velocity, phase, or wavelength of a signal propagating through the ferroelectric. The larger the relative variation of the dielectric constant, the larger the variation in velocity, phase, or wavelength. It is, therefore, important to know under what conditions such devices should be operated for maximum relative nonlinean'ty, that is, for maximum change in the relative value of the dielectric constant for a given change in the applied electric field. The relative nonlinearity may be represented by the ratio:

where As is the incremental change in the value of the dielectric constant e for a change in the strength of the electric field from EAE to E-I-AE, and e is the value of the value of the dielectric constant about which the change is made at a temperature T and electric field strength E and at a direction of that electric field with respect to the crystallographic axes of the material indi cated by the Miller indices, 11, k, and I. It is also important to know under what conditions such devices should be operated for minimum loss in the material. The total loss will be a combination of other circuit losses and loading as well as the loss in the material, but the material losses will set a lower limit on the extent to which the total loss can be reduced. However, little of a detailed nature has been revealed on the relationship of relative nonlinearity and loss to the significant parameters of temperature, bias electric field strength, and crystal orientation. Briefly, this invention provides a signal-translating device comprising signaltranslating means including fer-' roelectric material and means for operating the material in a region in which the ratio of relative nonlinearity to loss tangent is a maximum. An embodiment of such a device is described having two conductors with ferroelectric material positioned between the conductors and improved impedance matching, and offering other advantages not found in the prior art.

A feature of this invention is that operating conditions are given so that ferroelectric material may be operated substantially in a region of temperature, electric field strength, and crystal orientation in which the ratio of relative nonlinearity to loss tangent of the material is a maximum.

Another desirable feature of this invention is that the uniformity of the electric field distribution within the ferroelectric material is improved. This uniformity means that spurious modes with phase velocities diiferent from the desired mode are not likely to be propagated in the material. The disadvantage of these modes is that they tend to cut down the power transmitted as well as producing undesirable resonances in the device.

Another desirable feature of this invention is the relative ease with which brittle crystalline materials may be machined into a shape suitable for the type of transmission lin used. By using a planar transmission line a shape having surfaces which follow natural crystallographic planes is favored. Such a shape is easier to machine than other shapes.

Another desirable feature of this invention is an improvement in the bandwidth characteristic of the device obtained by employing an improved impedance match. In one embodiment of the device this is done by using a body of material in which the dielectric constant increases continuously from a value close to that of air at one end to a region of maximum dielectric constant (substantially pure ferroelectric) in the center, and then decreases smoothly to a value close to that of air at the other end. In another embodiment an approximation to a continuous impedance match is made by using several dielectric materials with different dielectric constants.

The device will be better understood by reference to the following description taken in conjunction with the appended drawings:

FIG. 1 is a cross-sectional view of an embodiment of the device;

FIG. 2 is a cross-sectional view taken along line 22 of FIG. 1; and

FIG. 3 is a cross-sectional view taken along line 3-3 of FIG. 1.

With reference now to FIGS. 1, 2, and 3, the device is shown generally at 10. A conducting strip 11 is supported on an insulating sheet shown generally at 12. The sheet 12 insulates the conducting strip 11 from a conducting base plane 13. The strip 11 and base plane 13 with insulating sheet 12 form a microstrip transmission line. The conducting strip 11 and the conducting base plane 13 are electrically connected to coaxial cables 14 and 15. Coaxial cable 14 transmits a microwave signal from a source of microwave power 16 and a coaxial cable 15 transmits the phase-shifted or modulated signal to load 17. Coaxial cable 14 is connected to the conducting strip 11 and the conducting base plane 13 by means of a type N coaxial connector 18. The inner conductor 19 of the coaxial cable 14 is connected to the conducting strip 11 by means of a Wire 20 which passes through a hole in the insulating sheet 12. The outer conductor 19a is connected to the conducting base plane 13. Coaxial cable 15 is connected in a similar manner by a connector 21.

The insulating sheet 12 is composed of a sheet 22 of dielectric material, a sheet 23 of a dielectric material hav ing a higher dielectric constant than that of sheet 22, two strips 24 and 25 of a dielectric material having a still higher dielectric constant and a body 26 of a ferroelectric material. The ferroelectric material has a higher dielectric constant than sheets 22, 23, or strips 24 and 25. The insulating sheet 12 has a uniform thickness throughout. The dielectric materials 22, 23, 24, and 25 serve as impedance matching elements. The conducting strip 11 is tapered in that part which is in contact with sheet 22 in order to further improve the impedance match.

A cable 27 and connector 28 are used to impress a bias and control voltage from a source 29 across the body 26 of ferroelectric material so that an electric field of variable magnitude and frequency may be applied to the body 26. The cable 27 is connected to the conducting base plane 13 and a conducting strip 31) in a manner similar to that for coaxial cable 15. The conducting strip 30 is composed of four sections, 31, 32, 33, and 34, which make up an example of an impedance transformer. Section 31 is a narrow section, one-quarter of a wavelength long, where the wavelength is that of the signal entering 9 a through connector 15. Section 32 is a wider section of the same length. Section 33 is the same as section 31. Section 34 connects to connector 28. This impedance transformer prevents leakage of power from microwave power source 16 into conducting strip 30.

Two D.C. blocks 35 and 36 are provided in conducting strip 11 in order to isolate the bias and voltage control source 29 from the source of microwave power 16 and the load 17. The DC blocks are made by cutting conducting strip 11 and overlapping the cut pieces onequarter of a wavelength and separating these pieces by sheets of insulation 37 and 38.

The base plane 13, the insulating sheet 12 and the conducting strips 11 and 30 with DC. blocks 35 and 36 are sealed in a protective container 39 of which base plane 13 forms one side.

The container 39 rests in a container 4! which has space for a liquid coolant 41. The conducting strip 11 is curved to allow the device to be easily inserted in the coolant 41.

In operation the voltage source 29 applies an electric field between the conducting strip 11 and the conducting base plane 13. This electric field appears across the body 26 of ferroelectric material. The magnitude of the bias voltage is that which produces an electric field giving substantially the highest ratio of relative nonlinearity to loss tangent as explained below. A signal from the source of microwave power enters the device at connector 18 and is propagated through the insulating sheet 12 and out connector 21 to a load 17.

When the device is to be used as a phase shifter a control voltage, which may be direct current or alternating current and may be in the form of a pulse, from voltage source 29 applies an electric field to the body 26 of ferroelectric material, which changes the total electric field across the material. This change in electric field changes the dielectric constant of the material'from that obtaining with only the electric field produced by the bias voltage. The change in dielectric constant modifies the phase of the signal. The other materials which make up insulating sheet 12, namely the dielectric sheets 22 and 23 and the strips 24 and 25, do not show a change in dielectric constant with a change in electric field and, therefore, have no eifect on the phase change.

In order to modulate the signal from the source of microwave power a control voltage varying in accordance with the desired modulating pattern is used.

An improved impedance match can also be obtained by using-an elongate body of material in which the dielectric constant changes continuously through a maximum region from one surface of the body to an oppositely disposed surface. For example, the material could be a ceramic mixture of strontium titanate and alumina or strontium titanate, titanium oxide and polyester, or other appropriate combinations, the proportion of the various ingredients differing throughout the body. Any ceramic or crystal growing process in which the concentration of ingredients may be continuously varied could be used. A ceramic, for example, would be built up at first of nearly pure ferroelectric. Then, the ferroelectric would be diluted with increasing percentages of a plurality of low-dielectric constant materials. The result would be a body in which the active section (ferroelectric) merges smoothly in a continuous impedance matching section. In the embodiment of FIG. 1 the sheets 22 and 23 might both be a low dielectric constant plastic, to provide insulation and support, and the ferroelectric sections 26 and dielectric sections 24 and 25 would be a body of material as described above.

As pointed out above, a device of the type described relies on a change in dielectric constant with applied voltage. Investigation by this inventor has determined that single crystals of ferroelectric material having the perovskite crystal structure such as the titanates of barium and strontium, have a dielectric constant, above the Curie temperature, whose real part can be substantially defined by the equation:

T T 2m 0 2 1+ 0 TT. E where e is the real part of the dielectric constant at a temperature T, an applied electric field strength E, and at a direction of that electric field with respect to the crystallographic axes of the material indicated by the Miller indices h, k, and l, C is the Curie constant of the material, T is the Curie temperature, A is the anisotropic nonlinearity constant of the material, and E is the applied electric field strength. A is essentially frequency and temperature independent, but has a marked dependence upon the direction of the applied electric field with respect to the crystallographic axes. The Curie temperature is a constant for a particular material and is defined as the intercept at 1/ 6 0 of a linear extrapolation of a Curie-Weiss plot of 1/ 6 versus T.

The prototype of the perovskite'crystal strucure is the mineral perovskite which is chemically known as calcium titanate. This mineral has a cubic crystal structure in which eight doubly ionized calcium ions occupy the eight cube corners, six doubly ionized oxygen ions occupy the center of the six faces of the cube, and a quadruply ionized titanium ion occupies the center of the cube. Barium titanate has substantially the same structure above the Curie temperature, the calcium ions being replaced by barium ions. Therefore, it is said to have the perovskite structure. Ferroelectric crystals having the perovskite structure are, among others, strontium titanate, in which strontium ions replace the calcium ions of the prototype crystal and potassium tantalite in which potassium ions replace the calcium ions and a tantalum ion replaces the titanium ion of the prototype crystal.

' The subscripts h, k, and l of e and A of Equation (1) refer to. the Miller indices of the crystallographic axis which is parallel to the direction of the applied electric field. In a cubic crystal such as perovskite, a direction perpendicular to any face of the cube is one characterized by h-=1, i=0, I 0, and written as Thus, in strontium titanate A is the value of the constant in the [100] direction which is a direction perpendicular to any plane of a unit cube of the crystal containing four strontium ions and. one oxygen ion, i.e. a face plane of the unit cube. The direction, h: l, k=l, [:0, is a direction perpendicular to a plane passing through four strontium ions, two oxygen ions, and the titanium ion of the unit cube. Other directions are given by other values of h, k, and l.

Polycrystalline ferroelectric material, including mixtures of barium titanate and strontium titanate in ceramic gum, follow Equation 1, with a suitable averaging of Microwave losses in ferroelectric material having the perovskite crystal structure have also-been investigated. The loss tangent has been found to consist of a field-inde pendent part and a field-dependent part, where the field is the applied electric field across the ferroelectric. The

field-dependent loss tangent for single crystal strontium titanate can be substantially defined by the equation:

frequency of the microwave energy, B is an anisotropic constant, C 1s the Curie constant, T is the Curie temperatan 5 ture,-T is the temperature, and E is the applied electric field strength. This constant is substantially temperature, frequency and field independent, the subscripts h, k, and l denoting the Miller indices of the crystallographic axis which is parallel to the direction of the applied electric field. Since the real part of the dielectric constant e decreases with increasing temperature T, the field-dependent losses also decrease with increasing temperature.

The field-independent losses, on the other hand, have a temperature and frequency variation defined substantially by the equation:

tan WW (3) where the symbols are as defined above.

It should be noted that Equation 4 depends on both temperature and the magnitude of the electric field as does the relative nonlinearity.

Polycrystalline material exhibits generally the same behavior as described in Equation 4 with a suitable change of constants. B must be suitably averaged over all crystal directions.

The temperature at which a device employing ferroelectric material should be operated will be that at which the ratio of nonlinearity to loss is a maximum.

Any ferroelectric material or combination of ferroelectric materials such as a mixture of barium titanate and strontium titanate may be used in the device. A good choice is a single crystal strontium titanate which has high relative nonlinearity and low loss. The Curie constant C of this material is approximately 825x K., T is approximately 37 K. and the values of the anisotropic nonlinearity constant A for three crystalloraphic directions are approximately:

l00=1.15 10-15 no=0.96 X 10-13 26% u1=0.69 10-1 2&9};

The value of the anisotropic constant B in Equation 2 above is approximately 4.8 l0 m? sec. /K./volt in the [100] direction. The values of the constants on and B in Equation 3 are:

The evaporation temperature of liquid nitrogen under atmospheric pressure is a value which makes the ratio of relative nonlinearity to loss tangent substantially a maximum and is one that is easily obtained and controlled. However, use of other ferroelectrics and in particular mixtures of barium titanate and strontium titanate permit operation at room temperature.

From a study of the values of the anisotropic nonlinearity constant above, it can be seen that the crystal should be aligned so that its [100] direction is parallel to the electric field.

For materials other than strontium titanate, the equations given above will sufl'ice to define optimum operating conditions to those skilled in the art.

When strontium titanate is selected as the ferroelectric material and three dielectric materials having dilferent dielectric constants are used as impedance matching, the dielectric sheet 22 may be made of rexolite or Teflon impregnated fiberglass, the dielectric piece 23 of alumina 6 and the dielectric pieces 24 and 25 of titanium oxide. The ratio of the impedances in the matching section is then:

Rexolite to alumina 0.543 Alumina to titanium oxide 0.274 Titanium oxide to strontium titanate 0.246

If only rexolite and alumina were used, the ratio of the impedances would be:

Rexolite to alumina 0.543 Alumina to strontium titanate 0.065

from which the improvement in impedance matching can be seen.

The insulation 23 and 24 for DC. blocks may be Mylar and the container 26 may be Styrofoam.

Although there have been described what are considered to be preferred embodiments of the present invention, various adaptations and modifications thereof may be made without departing from the spirit and scope of the invention as defined in the appended claims. In particular, the features of the invention residing in obtaining a maximum ratio of relative nonlinearity to loss tangent are not limited to phase shifters and modulators.

In achieving the advantages associated with the improved impedance matching and uniformity of field, as well as the two plane-parallel construction, it is not necessary to operate above the Curie temperature or in the region of maximum ratio of relative nonlinearity to loss tangent.

What is claimed is:

1. In combination, a transmission line circuit, means for providing high frequency energy to said transmission line circuit, a ferroelectric material having a perovskite crystal structure forming part of said transmission line circuit, means for providing impedance matching between the line circuit at said ferroelectric material and at other portions of said transmission line circuit, means for applying said energy from said transmission line circuit to said ferroelectric material through said means for providing impedance matching, and means for initiating phase shifting of said energy comprising means for applying a substantially uniform electric field to said ferroelectric material.

2. In combination, a transmission line circuit, means for providing high frequency energy to said transmission line circuit, a ferroelectric material having a perovskite crystal structure forming part of said transmission line circuit, means for providing impedance matching between the live circuit at said ferroelectric material and at other portions of said transmission line circuit, means for applying said energy from said transmission line circuit to said ferroelectric material through said means for providing impedance matching, means for initiating phase shifting of said energy comprising means for applying a substantially uniform electric field to said ferroelectric material with said field being aligned substantially parallel to the indices of the crystallographic axis of said material and means for varying the magnitude and frequency of said electric field to vary the dielectric constant value of said ferroelectric material.

3. The combination as claimed in claim 2 wherein said impedance matching means comprise a plurality of dielectric materials having different dielectric constants disposed in contiguous relationship with said ferroelectric material.

4. The combination according to claim 2 wherein said ferroelectric material and impedance matching means comprise at least three dielectric materials having different dielectric constants.

5. The combination according to claim 2 wherein the ferroelectric material has the highest dielectric constant value and the materials disposed on either side thereof are of a dielectric material having a progressively lower dielectric constant extending in the direction toward the terminal ends of the transmission line circuit.

6. A signal-translating device comprising two plane parallel conductors having an intermediate layer of dielectric material therebetween to define a transmission line circuit, rneans for coupling high frequency microwave energy to said transmission line circuit, said intermediatetlayer comprising at least three dielectric materials having different dielectric constants including a ferroelectric material having a perovskite crystal structure and the maximum dielectric constant value and adjoining materials extending in the direction toward the high frequency coupling means having continuously decreasing dielectric constant values, means for initiating phase shif ing of said microwave energy comprising means for applying a substantially uniform electric field to said ferroelectric material oriented parallel to the [100] indices of the orystallographic'axis of said material, and means for varying the magnitude and frequency of said electric field to vary the dielectric constant of said ferroelectric material.

7. A signal translating device according to claim 6 wherein opposing faces of said ferroelectric mate-rial are in substantial contact with said conductors.v

References Cited by the Examiner UNITED STATES PATENTS 2,555,959 5/1951 Curtis 332-30 2,825,876 4/195 Levine et al. 33384 2,838,736 6/1958 Poster 333-73 2,906,970 9/1959 Mason 33372 2,916,615 12/1959 L-undberg 333- 31 2,922,125 1/1960 Suhl 333-81 2,944,167 I 7/1960 Mature 30'7--88.5 2,969,512 1/1961 Jaffe 333-72 3,001,151 9/1961 Morris 33381 3,005,168 10/1961 Fye 333-84 3,050,643 8/1962 Connell 307- 885 HERMAN KARL SAALBACH, Prz'mm-y Examiner.

Claims (1)

1. IN COMBINATION, A TRANSMISSION LINE CIRCUIT, MEANS FOR PROVIDING HIGH FREQUENCY ENERGY TO SAID TRANSMISSION LINE CIRCUIT, A FERROELECTRIC MATERIAL HAVING A PERVOSKITE CRYSTAL STRUCTURE FORMING PART OF SAID TRANSMISSION LINE CIRCUIT, MEANS FOR PROVIDING IMPEDANCE MATCHING BETWEEN THE LINE CIRCUIT AT SAID FERROELECTRIC MATERIAL AND AT OTHER PORTIONS OF SAID TRANSMISSION LINE CIRCUIT, MEANS FOR APPLYING SAID ENERGY FROM SAID TRANSMISSION LINE CIR-

Images