HK1227015B - Ultra-high dielectric constant garnet - Google Patents

Description

Ultra-high dielectric constant garnet

Incorporation by reference to prior application

This application claims the benefit of U.S. Provisional Application No. 62/175,873, filed on June 15, 2015, entitled “Ultra-High Dielectric Constant Garnets,” and U.S. Provisional Application No. 62/343,685, filed on May 31, 2016, entitled “Ultra-High Dielectric Constant Garnets,” the entire contents of which are incorporated herein by reference.

Technical Field

The present disclosure generally relates to modified garnets having ultrahigh dielectric constants, and applications of such modified garnets.

Background Art

A variety of crystalline materials with magnetic properties have been used as components in electronic devices such as mobile phones, biomedical devices, and RFID sensors. Garnet is a crystalline material with ferromagnetic properties that is particularly useful in RF electronic components (electronics) operating in the lower frequency bands of the microwave region. Many microwave magnetic materials are derivatives of yttrium iron garnet (YIG), which is a synthetic form of garnet that is widely used in a variety of communication devices, largely due to its favorable magnetic properties such as narrow line width at its ferromagnetic resonance frequency. YIG is typically composed of yttrium, iron, and oxygen, and may be doped with one or more other rare earth metals such as lanthanides or scandium.

Summary of the Invention

Disclosed herein are embodiments of synthetic garnet materials comprising a structure including dodecahedral sites, at least some of which are occupied by bismuth, the garnet material having a dielectric constant value of at least 31.

In some embodiments, the 3dB linewidth can be less than 100. In some embodiments, the 3dB linewidth can be less than 80.

In some embodiments, the structure may include gadolinium. In some embodiments, the structure may include gadolinium at a level of up to 1.0 unit. In some embodiments, the synthetic garnet material may not include bismuthite as a second phase. In some embodiments, the structure may include at least 1.4 units of bismuth. In some embodiments, the structure may include 1.4-2.5 units of bismuth. In some embodiments, the synthetic garnet material may have a dielectric constant of at least 34.

Also disclosed herein are embodiments of synthetic garnet materials comprising a structure comprising at least 1.4 units of bismuth occupying dodecahedral sites.

In some embodiments, the synthetic garnet material may have a dielectric constant of at least 34. In some embodiments, the synthetic garnet material may have a dielectric constant of at least 36. In some embodiments, the structure may include 1.4-2.5 units of bismuth. In some embodiments, the garnet material may have a magnetization of 1900 or greater.

Also disclosed herein are embodiments of a modified synthetic garnet composition represented by the formula: BixCayGdzY3 _{-xyzFe5 - yZryO12 .} In some embodiments, 0 _< x _< 2.5, 0<y<1.0, and 0<z< _1.0 . In some embodiments, 0<x<2.5, 0<y<1.0, and 0<z<2.0. In some embodiments, the modified synthetic garnet composition may have a dielectric constant of at least 34. In some embodiments, the 3dB linewidth may be less than 80.

Also disclosed herein are embodiments of a method of making a synthetic garnet having a high dielectric constant, the method comprising providing a yttrium iron garnet structure, inserting greater than 1.4 units of bismuth into the iron garnet structure to form a modified synthetic garnet structure free of bismuthite.

In some embodiments, the modified synthetic garnet may have the following composition: _BixCayGdzY3 - _{xyzFe5 - yZryO12} , 0<x< _2.5 , 0<y<1.0, and 0 _< z< _1.0 . In some embodiments, 0<x<2.5, 0<y<1.0, and 0<z<2.0. In some embodiments, the modified _synthetic garnet may have a dielectric constant of at least 34. In some embodiments, the modified synthetic garnet may have a 3dB linewidth of less than 80.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 schematically illustrates how a material having one or more of the characteristics described herein may be designed, made, and used.

Figure 2 depicts the yttrium-based garnet lattice structure.

FIG3 illustrates an example process flow for an embodiment of making a modified synthetic garnet having one or more characteristics described herein.

FIG4 shows an example ferrite device having one or more garnet features as described herein.

5A and 5B illustrate examples of size reduction that may be implemented for ferrite devices having one or more features as described herein.

6A and 6B show example circulators/isolators having ferrite devices as described herein.

Figure 7 shows an example of a packaged circulator module.

FIG8 shows an example RF system in which one or more of the circulator/isolator devices described herein may be implemented.

FIG. 9 illustrates a process that may be implemented to produce a ceramic material having one or more features as described herein.

FIG. 10 illustrates a process that may be implemented to form a shaped object from the powder material described herein.

FIG. 11 shows an example of various stages of the process of FIG. 10 .

FIG. 12 shows a process that may be performed to sinter shaped objects such as those formed in the examples of FIGs. 10 and 11 .

FIG. 13 shows an example of various stages of the process of FIG. 12 .

14 illustrates a perspective view of a cell phone antenna base station incorporating an embodiment of the present disclosure.

FIG. 15 illustrates a housing component of a base station incorporating an embodiment of the disclosed material.

16 illustrates a cavity filter for use in a base station incorporating embodiments of the materials disclosed herein.

FIG. 17 illustrates an embodiment of a circuit board including an embodiment of the material disclosed herein.

DETAILED DESCRIPTION

Disclosed herein are embodiments of synthetic garnets (or, more generally, ferrites/ferrite garnets), methods for their manufacture, and applications of such synthetic garnets. Specifically, excess bismuth atoms can be introduced into the garnet lattice structure to increase the overall dielectric constant of the material without experiencing deleterious effects on other magnetic or electrical aspects of the garnet. In particular, bismuth-substituted ferromagnetic garnets can exhibit enhanced dielectric constants as sintered ceramics, making them particularly useful for miniaturizing isolators and circulators in commercial wireless infrastructure devices, thereby reducing the overall footprint of the devices. Additionally, the materials can maintain high magnetization, making them ideal for high-frequency applications in a range that has not previously been feasible.

Figure 1 schematically illustrates how one or more chemical elements (elements) (box 1), compounds (box 2), chemicals (box 3), and/or chemical mixtures (box 4) can be processed to produce one or more materials having one or more characteristics described herein (box 5). In some embodiments, such materials can be formed into ceramic materials (box 6) configured to include desired dielectric properties (box 7), magnetic properties (box 8), and/or advanced material properties (box 9).

In some embodiments, materials having one or more of the aforementioned properties may be implemented in applications such as radio frequency (RF) applications (block 10). Such applications may include implementing one or more features as described herein in a device 12. In some applications, such a device may further be implemented in a product 11. Examples of such devices and/or products are described herein.

synthetic garnet

Disclosed herein are methods for modifying synthetic garnet compositions, such as yttrium iron garnet (YIG), to increase the dielectric constant of the material. However, it will be understood that other synthetic garnets, such as yttrium aluminum garnet or gadolinium gallium garnet, may also be used, and the specific garnet is not limiting. Also disclosed herein are synthetic garnet materials having a high dielectric constant (and/or other advantageous properties), methods for making the same, and devices and systems incorporating such materials.

Synthetic garnets typically have a formula unit _{of A₃B₅O₁₂} , where _{A and B are trivalent metal ions. Yttrium iron garnet (YIG) is a synthetic garnet with a formula unit of Y₃Fe₅O₁₂ , which} includes yttrium (Y) in the 3+ oxidation state and iron ₍ Fe) in the 3+ oxidation state. The general crystal structure of the YIG formula unit is depicted in Figure 2. As shown in Figure 2, YIG has dodecahedral, octahedral, and tetrahedral positions. Y ions occupy the dodecahedral positions, while Fe ions occupy the octahedral and tetrahedral positions. Each YIG unit cell (cubic in crystal classification) has eight of these formula units.

The modified synthetic garnet compositions, in some embodiments, involve replacing some or all of the yttrium (Y) in yttrium iron garnet (YIG) with one or a combination of other ions so that the resulting material maintains or improves the magnetic properties desired for microwave (or other) applications, such as a high dielectric constant. Attempts have been made in the past to dope YIG with different ions to change the material properties. Some of these attempts, such as bismuth (Bi)-doped YIG, are described in D.B.Cruickshank's "Microwave Material for Wireless Applications," which is hereby fully incorporated by reference. However, in practice, the ions used as substitutes may not behave in a predictable manner due to spin canting, for example, caused by the magnetic ions themselves or by the influence of non-magnetic ions on neighboring magnetic ions in the environment, which reduces the degree alignment. Therefore, the resulting magnetic properties cannot be predicted. In addition, the amount of substitution is limited in some cases. Beyond a certain limit, the ions will not enter their preferred lattice positions and will either remain on the outside as a second phase compound or leak into another location. Additionally, ion size and crystallographic orientation preferences may compete at high substitution levels, or the substituting ion may be affected by the coordination of ions at other positions, as well as the ion size. As such, the assumption that the net magnetic behavior is the sum of the anisotropies of independent sublattices or single ions may not always hold true in predicting magnetic properties.

Considerations in selecting an effective replacement for rare earth metals in YIG for microwave magnetic applications include the density of the material, magnetic resonance linewidth, saturation magnetization, Curie temperature, dielectric constant, and optimization of the dielectric loss tangent in the resulting modified crystal structure. Magnetic resonance arises from spinning electrons, which, when excited by a suitable radio frequency (RF), will exhibit a resonance proportional to the applied magnetic field and frequency. The width of the resonance peak is typically defined in terms of the half-power point and is referred to as the magnetic resonance linewidth. It is generally advantageous for a material to have a low linewidth because a low linewidth manifests itself as low magnetic loss, which is required for all low insertion loss ferrite devices. The modified garnet compositions according to preferred embodiments of the present invention provide single or polycrystalline materials having a reduced yttrium content and still maintaining low linewidth and other desirable properties for microwave magnetic applications.

In some embodiments, yttrium-based garnets are modified by substituting bismuth (Bi ³⁺ ) for some of the yttrium (Y ³⁺ ) at the dodecahedral positions of the garnet structure, in combination with introducing one or more ions, such as divalent (+2), trivalent (+3), tetravalent (+4), pentavalent (+5), or hexavalent (+6) non-magnetic ions, into the octahedral positions of the structure to replace at least some of the iron (Fe ³⁺ ). In some embodiments, one or more high-valent non-magnetic ions, such as zirconium (Zr ⁴⁺ ) or niobium (Nb ⁵⁺ ), may be introduced into the octahedral positions.

In some embodiments, yttrium-based garnets are modified by introducing one or more high-valent ions with an oxidation state greater than 3+ into the octahedral or tetrahedral positions of the garnet structure, in combination with replacing yttrium (Y ³⁺ ) in the dodecahedral positions of the structure with calcium (Ca ²⁺ ) for charge compensation caused by the high-valent ions, thereby reducing the Y ³⁺ content. When non-trivalent ions are introduced, valence balance is maintained by balancing the non-trivalent ions with, for example, divalent calcium (Ca ²⁺ ). For example, for each 4+ ion introduced into the octahedral or tetrahedral position, one Y ³⁺ ion can be replaced with a Ca ²⁺ ion. For each 5+ ion, two Y ³⁺ ions can be replaced with Ca ²⁺ ions. For each 6+ ion, three Y ³⁺ ions can be replaced with Ca ²⁺ ions. For each 6+ ion, three Y 3+ ions can be replaced with Ca ²⁺ ions. In one embodiment, one or more high-valent ions selected ^from Zr ⁴⁺ , Sn ⁴⁺ , Ti ⁴⁺ , Nb ⁵⁺ , Ta ⁵⁺ , Sb ⁵⁺ , W ⁶⁺ , and Mo ⁶⁺ are introduced into octahedral or tetrahedral positions, and divalent calcium (Ca ²⁺ ) is used to balance the charge, which in turn reduces the Y3+ content.

In some embodiments, the yttrium-based garnet is modified by introducing one or more high-valent ions, such as vanadium (V ⁵⁺ ), into the tetrahedral positions of the garnet structure to replace Fe ³⁺ to further reduce the magnetic resonance linewidth of the resulting material. Without being bound by any theory, it is believed that the mechanism of ion substitution results in a reduced magnetization intensity at the tetrahedral positions of the lattice, which results in a higher net magnetization intensity of the garnet and, by changing the magnetocrystalline environment of the ferric ions, also reduces the anisotropy of the material and, therefore, the ferromagnetic linewidth.

In some embodiments, the following combination can effectively replace (replace) all or most of the yttrium (Y) in the microwave device garnet: high bismuth (Bi) doping, combined with calcium (Ca) valence compensation caused by vanadium (V) and/or zirconium (Zr). In addition, certain other high-valent ions can also be used in tetrahedral or octahedral positions and in order to obtain minimized magnetic resonance linewidth, a relatively high octahedral substitution level in the garnet structure is preferred. Moreover, yttrium substitution can be achieved by adding calcium to the dodecahedral positions in addition to bismuth. Doping the octahedral or tetrahedral positions with higher-valent ions (preferably greater than 3+) allows more calcium to be introduced into the dodecahedral positions to compensate for the charge, which in turn will lead to a further reduction in the yttrium content.

Modified synthetic garnet composites

Disclosed herein are modified synthetic garnets that have very high dielectric constants while also possessing high levels of magnetization, making them particularly useful for high-frequency applications. In particular, increased amounts of bismuth, along with balancing charge from other elements, can be incorporated into the crystal structure to improve the magnetoelectric properties of the garnets without degrading other magnetoelectric properties.

In some embodiments, the modified synthetic garnet composition may be defined by the following general composition: _{BixCayGdzY3 -xyzFe5 - yZryO12 ,} wherein 0≤x≤2.5, 0≤y≤1.0, and 0≤z≤1.0. In some embodiments, 0≤x≤2.5, 0≤y≤1.0, and 0≤z≤2.0. In some embodiments, 1.0< _x <2.0, 0.1<y<0.8, and 0.2<z<1.9. However, some embodiments _of the present disclosure may not be limited by the above composition.

In some embodiments, about 1.4 formula units of bismuth (Bi) may be substituted for some of the yttrium (Y) on the dodecahedral positions. In some embodiments, greater than about 1.4 formula units of bismuth (Bi) may be substituted for some of the yttrium (Y) on the dodecahedral positions. In some embodiments, about 1.4 to about 2.5 formula units of bismuth (Bi) may be substituted for some of the yttrium (Y) on the dodecahedral positions. In some embodiments, up to 3.0 formula units of bismuth (Bi) may be substituted for some of the yttrium (Y) on the dodecahedral positions. High bismuth levels that can result in advantageous properties can be achieved by including certain atoms and manufacturing methods as discussed below.

Additionally, as shown, for example, in the above formula, charge balance can be achieved by replacing some or all of the remaining yttrium (Y) with calcium (Ca) or zirconium (Zr). In some embodiments, equal amounts of Ca and Zr are added to maintain charge stability, as Ca has a formal charge of +2 and Zr has a formal charge of +4. Additionally, to balance the differential stresses on the structure caused by the inclusion of bismuth (Bi), gadolinium (Gd) or other large rare earth ions can be introduced into the dodecahedral positions of the garnet structure. For example, Gd can be added in place of Y, which can improve temperature stability. Additionally, Gd itself can increase the dielectric constant.

Table 1 below illustrates a list of different synthetic garnet compositions and their manufacturing parameters. Additionally, Table 2 discloses the corresponding properties achieved by the compositions of Table 1.

Table 1 shows the list of composition and manufacturing parameters

Table 2 shows the properties of the compositions of Table 1

As shown in the table above, using embodiments of the disclosed synthetic garnets, very high dielectric constants can be achieved. For example, in some embodiments, the dielectric constant of the composite can be greater than or equal to 31, 33, 35, 37, 39, or 40. Additionally, the 3 dB linewidth can be minimized, with some embodiments having a 3 dB linewidth of less than 100, 90, 80, 70, or 60.

Because the size of bismuth is larger than the size of the yttrium it replaces, inserting bismuth into the garnet structure can cause significant lattice distortion in the garnet structure. Typically, there is only so much bismuth that can be inserted into the garnet structure before the garnet structure decomposes, making it less useful for RF applications. For example, if too much bismuth is added to the garnet structure, the structure will reject bismuth, and a bismuth-rich phase called pyroxenite will form. When pyroxenite forms, the 3 dB linewidth of the material can increase significantly, as shown in Table 2 above, making the material difficult to use in RF applications.

Pyrrolite is a structure that is very rich in bismuth and tends to form grain boundaries. Although pyrrolite may not always be detected, due to its glass-forming or poor crystallinity, the 3 dB linewidth typically increases dramatically with pyrrolite. Therefore, for significantly high 3 dB linewidths, such as those shown in Syntheses 5 and 6 above, the presence of pyrrolite is hypothesized. Furthermore, the presence of pyrrolite is hypothesized due to abnormally high dielectric constants. Furthermore, large 3 dB linewidths may be the result of defective garnet structures with oxygen or cation vacancies.

Because it is difficult to insert too much bismuth into garnet, other atoms can be inserted to act as chemical compensators that open up the structure. For example, gadolinium (Gd) atoms can be substituted into the structure, and due to the larger size of the gadolinium atoms, a more stable garnet structure can be formed with a higher bismuth content, thereby allowing for improved properties such as the dielectric constant. Gadolinium, in particular, can have useful magnetic and radio frequency properties. For example, gadolinium is not a fast relaxer, unlike other rare earth atoms. Fast relaxers will increase the 3dB linewidth due to their stable 7f or 4f electron shells. However, gadolinium can be used without causing a significant increase in linewidth.

Other large atoms such as La, Pr, Nd, Sm, Dy, Yb and Ho can also be used instead of gadolinium. Some of these are fast relaxers and can increase the 3dB linewidth.

Table III illustrates other synthetic garnet compositions that can be formed using increased amounts of bismuth in the crystalline structure of garnet and their respective properties. In some embodiments, hafnium (Hf) and titanium (Ti) can be introduced into the octahedral positions of the lattice. In addition, rare earth ions (such as La, Ce, Pr, Nd, Sm, Eu, Dy, Tb, Ho, Er, Tm, Lu and Yb) and smaller ions (such as Mn, In, Sc, Zr, Hf, Zn and Mg) can all be introduced into the dodecahedral positions of the garnet structure. In some cases, the total charge may need to be balanced with other substitutes.

Table III: Composition and properties of synthetic garnets

As shown in Table 3, additional elements can be used in the formation of synthetic garnets. For example, hafnium (Hf), strontium (Sr), indium (In), or ytterbium can be introduced into synthetic garnets to improve properties. The change in composition can be due in part to the changing charge balance scheme. In some embodiments, the material can include some yttrium. In some embodiments, the material can be free of yttrium, such as when the material has been completely substituted.

As shown in the table above, embodiments of the synthetic garnet can achieve very high dielectric constants. For example, embodiments of the synthetic garnet can achieve dielectric constants greater than 35, greater than 36, or about 38 (or greater than about 35, greater than about 36, or greater than about 38). As a result, devices such as circulators and isolators can be approximately 5%-10% smaller in diameter compared to devices having a dielectric constant of 32. This allows for an overall smaller footprint for the devices, thereby allowing more of the devices to be located in a concentrated area.

Furthermore, embodiments of the present disclosure can have very high magnetization and high dielectric constants, which allow them to be used within certain frequency ranges. As shown above, embodiments of the synthetic garnets can have dielectric constants greater than 1600, 1700, 1800, or 1900 (or greater than about 1600, about 1700, about 1800, or about 1900), as opposed to the previously used dielectric constants of 1500. This allows devices incorporating such materials to be used within higher frequency ranges.

Preparation of modified synthetic garnet composites:

The preparation of the modified synthetic garnet material can be achieved by using known ceramic technology. A specific example of the process flow is shown in FIG3 .

3 , the process begins with weighing raw materials at step 106. The raw materials may include oxides and carbonates such as iron oxide (Fe ₂ O ₃ ), bismuth oxide (Bi ₂ O ₃ ), yttrium oxide (Y ₂ O ₃ ), calcium carbonate (CaCO ₃ ), zirconium oxide (ZrO ₂ ), gadolinium oxide (Gd ₂ O ₃ ), vanadium pentoxide (V ₂ O ₅ ), yttrium vanadate (YVO ₄ ), bismuth niobate (BiNbO ₄ ), silicon dioxide (SiO ₂ ), niobium pentoxide (Nb ₂ O ₅ ), antimony oxide (Sb ₂ O ₃ ), molybdenum oxide (MoO ₃ ), indium oxide (In ₂ O ₃ ), or combinations thereof. In some embodiments, the raw materials consist essentially of: about 35-40% by weight, more preferably about 38.61% by weight, bismuth oxide; about 10-12% by weight, more preferably about 10.62% by weight, calcium oxide; about 35-40% by weight, more preferably about 37% by weight, iron oxide; about 5-10% by weight, more preferably about 8.02% by weight, zirconium oxide; and about 4-6% by weight, more preferably about 5.65% by weight, vanadium oxide. Furthermore, organic-based materials can be used in ethanolate-based sol-gel processes and/or acrylate- or citrate-based technologies can be employed. Other methods known in the art, such as hydroxide coprecipitation, sol-gel, and laser ablation, can also be used to obtain the materials. The amount and selection of the raw materials depend on the specific formulation.

After the raw materials are weighed, they are blended in step 108 using methods consistent with the state of the art in the ceramic field, which may include aqueous blending using a mixing propeller or aqueous blending using a vibrating mill with steel or zirconium oxide media. In some embodiments, glycine nitrate or spray pyrolysis techniques may be used to blend and react the raw materials simultaneously.

The blended oxides are then dried in step 110, which can be achieved by pouring the slurry into a grid (tray) and drying in an oven, preferably between 100-400°C, or by spray drying or by other techniques known in the art.

The dried oxide blend is processed through a sieve in step 112, which homogenizes the powder and breaks up soft agglomerates that can result in dense particles after calcination.

The material is then processed by pre-sintering calcination in step 114. Preferably, the material is loaded into a container such as an alumina or cordierite saggar and heat treated in the range of about 800-1000°C. In some embodiments, heat treatment in the range of about 500-1000°C may be used. In some embodiments, heat treatment in the range of about 900-950°C may be used. In some embodiments, heat treatment in the range of about 500-700°C may be used. Preferably, the firing temperature is low, as higher firing temperatures have an adverse effect on line width.

After calcination, the material is ground in step 116, preferably using a vibratory mill, a disc mill, a jet mill, or other standard comminution technique to reduce the median particle size to a range of about 0.01-0.1 microns, although larger sizes, such as 0.5-10 microns, may also be used in some embodiments. The grinding is preferably performed in a water-based slurry, but may also be performed in ethanol or another organic-based solvent.

The material is then spray dried in step 118. During the spray drying process, organic additives such as binders and plasticizers may be added to the slurry using techniques known in the art. The material is spray dried to provide granules that are easily compressible, preferably having a size in the range of about 10 microns to 150 microns.

The spray-dried pellets are then pressed in step 120, preferably by uniaxial or isostatic pressing to achieve a pressing density as close as possible to 60% of the x-ray theoretical density. In addition, other known methods such as tape casting, tape calendering or extrusion can also be used to form the green body.

The pressed material is then processed through a calcination process in step 122. Preferably, the pressed material is placed on a setter plate made of a material that does not react easily with the garnet material, such as alumina. The setter plate is heated in a batch kiln or tunnel kiln in air or pressurized oxygen in a range of about 850°C-1000°C to obtain a dense ceramic compact. In some embodiments, a heat treatment in the range of about 500-1000°C can be used. In some embodiments, a heat treatment in the range of about 500-700°C can be used. Other known processing techniques such as induction heating, hot pressing, rapid firing, or assisted rapid firing can also be used in this step. In some embodiments, a density of >98% of theoretical density can be achieved.

The dense ceramic compact is machined in step 124 to achieve dimensions suitable for a particular application.

Devices Introducing Ultra-High Dielectric Constant Garnet

Radio frequency (RF) applications utilizing synthetic garnet compositions such as those disclosed above may include ferrite devices having relatively low magnetic resonance linewidths. RF applications may also include devices, methods, and/or systems having or involving garnet compositions having reduced or substantially zero reduced (rare) earth content. As described herein, such garnet compositions may be configured to produce relatively high dielectric constants; and such characteristics may be used to provide advantageous functionality. It will be understood that at least some of the compositions, devices, and methods described with reference to the above may be applicable to such implementations.

FIG4 shows a radio frequency (RF) device 200 having a garnet structure and chemical properties, such as those disclosed herein, and thus having a plurality of dodecahedral, octahedral, and tetrahedral structures. The device 200 may include a garnet structure (e.g., garnet structure 220) formed from such dodecahedral, octahedral, and tetrahedral structures. Various examples of how dodecahedral sites 212, octahedral sites 208, and tetrahedral sites 204 may be filled or substituted with different ions to produce one or more desirable properties for the RF device 200 are disclosed herein. Such properties may include, but are not limited to, desirable RF properties and cost-effectiveness of manufacture of ceramic materials that may be used to fabricate the RF device 200. For example, ceramic materials having a relatively high dielectric constant and having a reduced or substantially zero rare earth content are disclosed herein.

We now describe some design considerations for implementing such features. We also describe example devices and related RF performance comparisons. We also describe example applications of such devices and fabrication examples.

Bismuth Garnet:

Single crystals of materials with _the formula Bi _{(3-2x) Ca2xFe5 - xVxO12} , where x is 1.25, have been grown in the past. _4πMs values of around 600 gauss were achieved (suitable for some tunable filters and resonators in the 1-2 GHz range), with linewidths of around 1 oersted, indicating low intrinsic magnetic losses for the system. However, the Bi substitution level in the formula was only around 0.5.

Attempts to produce single-phase polycrystalline materials similar to the single-crystalline materials (having the formula _{Bi3-2xCa2xVxFe5 -xO12 ) have} been successful only in the region of x>0.96 _, which effectively limits _4πMs to less than about 700 oersteds and results in poor linewidths (greater than 100 oersteds). Small amounts of Al ⁺³ reduce the linewidth to about 75 oersteds, but increasing Al ⁺³ reduces _4πMs . In the formulas of these materials, Bi substitution is only about 0.4.

For Faraday rotation devices, the Faraday rotation can be substantially proportional to the level of Bi substitution in the garnet, thus increasing the interest in increasing substitution levels. Anisotropy is generally not a major factor for optical applications, so substitution at octahedral and tetrahedral sites can be based on maximizing the rotation. Therefore, in such applications, it may be desirable to incorporate as much Bi ⁺³ as possible into the dodecahedral sites. The maximum level of Bi ⁺³ can be influenced by the size of the dodecahedral rare earth trivalent ion.

In some cases, the level of Bi ⁺³ substitution can be affected by substitutions at other positions. Since Bi ⁺³ is non-magnetic, it can affect the Faraday rotation through its influence on tetrahedral and octahedral Fe ⁺³ ions. Since this is believed to be a spin-orbit interaction in which Bi ⁺³ alters the transition of the Fe ⁺³ pair present, changes in the anisotropy of the Fe ⁺³ ions and optical effects including large Faraday rotation can be expected. At low Bi ⁺³ substitution, the Curie temperature of Bi ⁺³ substituted YIG can also be increased.

Examples of devices with rare earth-free or rare earth-reduced garnets:

As described herein, garnets with reduced or no rare earth content can be formed, and such garnets can have desirable properties for use in devices for applications such as RF applications. In some implementations, such devices can be configured to exploit the unique properties of Bi ⁺³ ions. For example, the "lone pair" of electrons on the Bi ⁺³ ion can increase the ion polarizability and, therefore, the dielectric constant.

Furthermore, since the center frequency of a ferrite device (e.g., a garnet disk) operating in a split polarization transverse magnetic (TM) mode is proportional to 1/(ε) ^1/2 , doubling the dielectric constant (ε) can reduce the frequency by a factor of the square root of 2 (approximately 1.414). As described in greater detail herein, increasing the dielectric constant by, for example, a factor of 2 can result in a reduction in the lateral dimension (e.g., diameter) of the ferrite disk by a factor of the square root of 2. Consequently, the area of the ferrite disk can be reduced by a factor of 2. This reduction in size can be advantageous because the device's footprint on an RF circuit board can be reduced (e.g., by a factor of 2 when the dielectric constant is increased by a factor of 2). While described with respect to an example of a factor of 2 increase, similar advantages can be achieved in configurations involving a factor of more or less than a factor of 2.

Reduced size circulator/isolator having ferrite with high dielectric constant

As described herein, ferrite devices having garnets with reduced or no rare earth content can be configured to include high dielectric constant properties. Various design considerations regarding dielectric constant as applied to RF applications are now described. In some implementations, such designs utilizing garnets with high dielectric constants may or may not necessarily involve rare earth-free configurations.

The values of the dielectric constants of microwave ferrite garnets and spinels typically fall within the range of 12-18 for dense polycrystalline ceramic materials. Such garnets are typically used for above ferromagnetic resonance applications in, for example, the UHF and low microwave regions due to their low resonant linewidth. Such spinels are typically used for below resonant applications at, for example, medium to high microwave frequencies due to their higher magnetization. Most, if not substantially all, circulators or isolators using such ferrite devices are designed with a three-plate/stripline or waveguide structure.

The dielectric constant values of low linewidth garnets are typically in the range of 14-16. These materials can be based on yttrium iron garnet (YIG), which has a value of approximately 16, or substituted versions of this chemistry with aluminum, or a zirconium/vanadium combination, for example, which can reduce the value to about 14. While spinel ferrites based on lithium titanium, for example, exist with dielectric constants up to nearly 20, these generally do not have narrow linewidths and are therefore unsuitable for many RF applications. However, as described in detail above, garnets made using bismuth substituted for yttrium can have much higher dielectric constants.

In some embodiments, the increase in dielectric constant can be maintained for compositions containing bismuth, including those with other non-magnetic substitutions on either or both the octahedral and tetrahedral sites (e.g., zirconium or vanadium, respectively). The dielectric constant can be further increased by using higher polarized ions. For example, niobium or titanium can be substituted into either the octahedral or tetrahedral sites; and titanium can potentially enter both sites.

In some embodiments, the relationship between ferrite device size, dielectric constant, and operating frequency can be expressed as follows. There are different equations that can represent different transmission line representations. For example, in an above-resonance stripline configuration, the radius R of the ferrite disk can be represented as

R＝1.84/[2π(effective permeability)x(dielectric constant)] ^1/2 (1)

Where (effective permeability) = H _dc + 4M _s /H _dc , where H _dc is the magnetic field bias. Equation 1 shows that for fixed frequency and bias, the radius R is inversely proportional to the square root of the dielectric constant.

In another example, a relationship similar to Equation 1 for the ferrite disk radius R can be used for a weakly coupled quarter-wave circulator in a below-resonance stripline configuration, where a low magnetic bias field corresponds to below-resonance operation. For below-resonance waveguide configurations (e.g., in the form of a disk or rod waveguide), the lateral dimensions (e.g., radius R) and thickness d of the ferrite can affect the frequency. However, the radius R can still be expressed as

R＝λ/[2π(dielectric constant) ^1/2 ][((πR)/(2d)) ² +(1.84) ² ] ^1/2 (2)

It is similar to Equation 1 in terms of the relationship between R and dielectric constant.

The example relationship of Equation 2 is for a disk-shaped ferrite. For a triangle-shaped resonator, the same waveguide expression can be used, but in this case A (the height of the triangle) equal to 3.63 x λ/2π is used instead of the radius in the disk case.

In all of the aforementioned example cases, it can be seen that by increasing the dielectric constant (e.g., by a factor of 2), it can be expected that the size of the ferrite (e.g., a disk or triangle) can be reduced by a factor of 2, and therefore the area of the ferrite can be reduced by a factor of 2. As described with reference to Equation 2, the thickness of the ferrite can also be reduced.

In implementations where ferrite devices are used as RF devices, the size of such RF devices can also be reduced. For example, in stripline devices, the device footprint can be dominated by the area of the ferrite used. Therefore, a corresponding reduction in device size can be expected. In waveguide devices, the diameter of the ferrite used can be a limiting factor in determining size. However, any reduction in ferrite diameter can be offset by the need to maintain wavelength-related dimensions in the metal portion of the junction.

Examples of reduced-size ferrites

As described herein, ferrite size can be significantly reduced by increasing the dielectric constant associated with the garnet structure. Furthermore, as described herein, garnets with reduced yttrium and/or reduced non-Y rare earth content can be formed by appropriate bismuth substitution. In some embodiments, such garnets can include yttrium-free or rare earth-free garnets. RF devices having ferrite devices with increased dielectric constants and yttrium-free garnets are described with reference to Figures 5A-6B.

5A and 5B summarize the example ferrite size reduction described herein. As described herein and shown in FIG. 5A , ferrite device 250 can be a circular disk having a reduced diameter 2R′ and a thickness d′. The thickness may or may not be reduced. As described with reference to Equation 1, the radius R of the circular ferrite disk can be inversely proportional to the square root of the dielectric constant of the ferrite. Thus, the increased dielectric constant of ferrite device 250 is shown as resulting from its reduced diameter 2R′.

As described herein and shown in FIG5B , the ferrite device 250 can also be a triangular disk having a reduced side dimension S' and a thickness d'. The thickness may or may not be reduced. As described with reference to Equation 2, the height A of the triangular ferrite disk (which can be derived from the side dimension S) can be inversely proportional to the square root of the dielectric constant of the ferrite. Thus, the increased dielectric constant of the ferrite device 250 is shown as resulting from its reduced diameter S'.

Although described with respect to example circular and triangular ferrites, one or more features of the present disclosure may be implemented in other shaped ferrites as well.

Figures 6A and 6B show an example of a circulator 300 having a pair of ferrite disks 302, 312 disposed between a pair of cylindrical magnets 306, 316. Each of the ferrite disks 302, 312 can be a ferrite disk having one or more features described herein. Figure 6A shows an unassembled view of a portion of the example circulator 300. Figure 6B shows a side view of the example circulator 300.

In the illustrated example, first ferrite disk 302 is shown mounted to the underside of first ground plane 304. The upper side of first ground plane 304 is shown as defining a recess dimensioned to receive and retain (secure) first magnet 306. Similarly, second ferrite disk 312 is shown mounted to the upper side of second ground plane 314; and the underside of second ground plane 314 is shown as defining a recess dimensioned to receive and retain (secure) second magnet 316.

The magnets 306 and 316 arranged in the aforementioned manner can generate generally axial magnetic field lines through the ferrite disks 302 and 312. The magnetic field flux passing through the ferrite disks 302 and 312 can complete its circuit through the return path provided by 320, 318, 308, and 310 to enhance the field applied to the ferrite disks 302 and 312. In some embodiments, the return path portions 320 and 310 can be disks having a larger diameter than the diameter of the magnets 316 and 306; and the return path portions 318 and 308 can be hollow cylinders having an inner diameter that generally matches the diameter of the return path disks 320 and 310. The aforementioned portions of the return path can be formed as a single piece or can be an assembly of multiple pieces.

The example circulator device 300 may further include an inner flux conductor (also referred to herein as a center conductor) 322 disposed between the two ferrite disks 302, 312. Such an inner conductor may be configured to function as a resonator and to match the network to a port (not shown).

Various examples of novel garnet systems and devices involving the same are described herein. In some embodiments, such garnet systems may contain high levels of bismuth, which may allow for the formation of low-loss ferrite devices. Further, the rare earth content of garnets (including commercial garnets) may be reduced or eliminated through the selective addition of other elements. Such reduction or elimination of rare earth content may include, but is not limited to, yttrium. In some embodiments, the garnet systems described herein may be configured to significantly increase (e.g., double) the dielectric constant of non-Bi garnets, thereby providing the potential for a significant reduction (e.g., halving) in the printed circuit "footprint" of ferrite devices associated with conventional garnets.

In some embodiments, a ferrite-based circulator device having one or more features as described herein can be implemented as a packaged modular device. FIG7 shows an example packaged device 400 having a circulator device 300 (e.g., as shown in FIG6B ) mounted on a packaging platform 404 and enclosed by a shell structure 402. The example platform 404 is depicted as including a plurality of holes 408 sized to allow for mounting of the packaged device 400. The example packaged device 400 is shown as further including example terminals 406 a-406 c configured to facilitate electrical connection.

In some embodiments, a packaged circulator/isolator 3002 (e.g., the example of FIG. 7 ) can be implemented in a circuit board or module 3004 as shown in FIG. 17 . Such a circuit board can include a plurality of circuits configured to perform one or more radio frequency (RF)-related operations. The circuit board can also include a number of connection features configured to allow RF signals and power to be transmitted between the circuit board and components external to the circuit board.

In some embodiments, the aforementioned example circuit board may include RF circuitry associated with a front-end module of an RF device. As shown in FIG8 , such an RF device may include an antenna 512 configured to facilitate the transmission and/or reception of RF signals. Such signals may be generated and/or processed by a transceiver 514. For transmission, the transceiver 514 may generate a transmit signal that is amplified by a power amplifier (PA) and filtered (Tx filter) for transmission via the antenna 512. For reception, the signal received from the antenna 512 may be filtered (Rx filter) and amplified by a low-noise amplifier (LNA) before being passed to the transceiver 514. Within the scope of such Tx and Rx paths, a circulator and/or isolator 500 having one or more features as described herein may be implemented, for example, at the PA circuit and LNA circuit or in combination therewith.

In some embodiments, devices and circuits having one or more features as described herein may be implemented in RF applications, such as wireless base stations. Such wireless base stations may include one or more antennas 512 configured to facilitate transmission and/or reception of RF signals, such as the example described with reference to FIG8 . Such antenna(s) may be coupled to devices and circuits having one or more circulators/isolators as described herein.

As used herein, the terms "circulator" and "isolator" may be used interchangeably or independently, depending on the application as generally understood. For example, a circulator may be a passive device used in RF applications to selectively route RF signals between an antenna, a transmitter, and a receiver. If a signal is being routed between a transmitter and an antenna, the receiver should preferably be isolated. Therefore, such circulators are sometimes also referred to as isolators, and such isolation performance may represent the performance of the circulator.

RF device manufacturing

Figures 9-13 illustrate examples of how ferrite devices having one or more features as described herein may be fabricated. Figure 9 illustrates a process 20 that may be implemented to fabricate a ceramic material having one or more of the aforementioned properties. In block 21, a powder may be prepared. In block 22, a shaped object may be formed from the prepared powder. In block 23, the shaped object may be sintered. In block 24, the sintered object may be finished to produce a finished ceramic object having one or more desired properties.

In implementations where the finished ceramic object is part of a device, the device may be assembled in block 25. In implementations where the device or finished ceramic object is part of a product, the product may be assembled in block 26.

9 further shows that some or all of the steps of the example process 20 may be based on design, specifications, etc. Similarly, some or all of the steps may include or be subject to testing, quality control, etc.

In some implementations, the powder preparation step (block 21) of FIG9 can be performed using the example process described with reference to FIG14. Powders prepared in this manner can include one or more properties as described herein and/or facilitate the formation of ceramic objects having one or more properties as described herein.

In some embodiments, the powder prepared as described herein can be formed into different shapes through different forming processes. For example, FIG10 shows a process 50 that can be implemented to form a shaped object by pressing a powder material prepared as described herein. In box 52, a forming die can be filled with a desired amount of powder. In FIG11, configuration 60 shows a forming die as 61, which defines a volume 62 dimensioned to receive powder 63 and allow such powder to be pressed. In box 53, the powder in the die can be compressed to form a shaped object. Configuration 64 shows the powder in an intermediate compacted form 67 as a piston 65 is pressed (arrow 66) into the volume 62 defined by the die 61. In box 54, the pressure can be removed from the die. In box 55, the piston (65) can be removed from the die (61) to open the volume (62). Configuration 68 shows the volume (62) of the die (61) opened, thereby allowing the shaped object 69 to be removed from the die. In block 56, the shaped object (69) may be removed from the mold (61). In block 57, the shaped object may be stored for further processing.

In some embodiments, the shaped objects made as described herein can be sintered to produce the desired physical properties as ceramic devices. Figure 12 shows a process 70 that can be implemented to sinter such shaped objects. In box 71, a shaped object can be provided. In box 72, the shaped object can be introduced into a kiln. In Figure 13, a plurality of shaped objects 69 are loaded into a sintering tray 80. The example tray 80 is shown as defining a groove 83 that is dimensioned to hold the shaped object 69 on a surface 82 so that the upper edge of the tray is higher than the upper portion of the shaped object 69. Such a configuration allows the loaded trays to be stacked during the sintering process. The example tray 80 is further shown as being defined by a cutout 83 at the side wall to allow for improved circulation of hot air within the groove 83, even when the trays are stacked together. Figure 13 further shows a stack 84 of a plurality of loaded trays 80. A top cover 85 may be provided so that objects loaded in the top tray experience substantially similar sintering conditions as those in the lower tray.

In block 73, heat may be applied to the shaped object to produce a sintered object. Such application of heat may be achieved using a kiln. In block 74, the sintered object may be removed from the kiln. In Figure 13, a stack 84 of multiple loaded trays is depicted as being introduced into a kiln 87 (stage 86a). Such a stack may be moved through the kiln (stages 86b, 86c) based on a desired time and temperature profile. In stage 86d, the stack 84 is depicted as being removed from the kiln to cool.

In block 75, the sintered object may be cooled. Such cooling may be based on a desired time and temperature profile. In block 206, the cooled object may undergo one or more finishing operations. In block 207, one or more tests may be performed.

The heat treatment of various forms of powders and various forms of shaped objects is described herein as calcining, firing, annealing, and/or sintering. It will be understood that such terms may be used interchangeably in some appropriate circumstances, in a context-specific manner, or in some combination thereof.

Communication base station

Devices and circuits having one or more features as described herein may be implemented in RF applications, such as wireless base stations. Such wireless base stations may include one or more antennas configured to facilitate transmission and/or reception of RF signals. Such antenna(s) may be coupled to devices and circuits having one or more circulators/isolators as described herein.

Thus, in some embodiments, the materials disclosed above can be incorporated into various components of, for example, a communication base station used for cell phone networks and wireless communications. An example perspective view of a base station 2000 is shown in FIG14 , which includes both a cell phone tower 2002 and an electronics building 2004. Cell phone tower 2002 may include numerous antennas 2006, typically oriented in different directions to optimize service, which may be used to receive and transmit cell phone signals, while electronics building 2004 may house electronic components such as filters, amplifiers, and the like discussed below. Both antennas 2006 and the electronic components may incorporate embodiments of the disclosed ceramic materials.

FIG11 shows a schematic diagram of a base station, such as that shown in FIG14 . As shown, the base station may include an antenna 412 configured to facilitate the transmission and/or reception of RF signals. Such signals may be generated and/or processed by a transceiver 414. For transmission, the transceiver 414 may generate a transmit signal that is amplified by a power amplifier (PA) and filtered (Tx filter) for transmission via the antenna 412. For reception, the signal received from the antenna 412 may be filtered (Rx filter) and amplified by a low-noise amplifier (LNA) before being passed to the transceiver 414. Within the scope of such Tx and Rx paths, a circulator and/or isolator 400 having one or more features as described herein may be implemented, for example, in the PA circuitry and the LNA circuitry, or in conjunction therewith. The circulator and isolator may include embodiments of the materials disclosed herein. Furthermore, the antennas may include the materials disclosed herein, thereby enabling them to operate over a higher frequency range.

Figure 15 illustrates hardware 2010 that may be used in electronics building 2004 and may include the components discussed above with respect to Figure 11. For example, hardware 2010 may be a base station subsystem (BSS) that may process traffic and signaling for a mobile system.

FIG16 illustrates further details of the hardware 2010 discussed above. Specifically, FIG16 depicts a cavity filter/combiner (synthesizer) 2020 that can be incorporated into a base station. Cavity filter 2020 can include, for example, a bandpass filter such as those incorporated into embodiments of the disclosed material, and can allow the outputs of two or more transmitters based on different frequencies to be combined.

From the foregoing description, it will be appreciated that inventive garnets and methods of manufacture are disclosed. Although several components, techniques, and aspects have been described with a certain degree of particularity, it is apparent that many changes may be made to the specific designs, constructions, and methods described herein without departing from the spirit and scope of the present disclosure.

Certain features described in this disclosure with respect to separate implementations may also be implemented in combination in a single implementation. Conversely, multiple features described with respect to individual implementations may also be implemented in multiple implementations, either individually or in any suitable subcombination. Furthermore, although features may be described above as functioning in certain combinations, one or more features from a claimed combination may, in certain circumstances, be omitted from the claimed combination, and the claimed combination may be claimed as any subcombination or variations of any subcombination.

In addition, although the method may be depicted in the drawings or described in the specification in a particular order, such method does not need to be performed in the particular order shown or in a sequential order, and in order to achieve the desired result, it is not necessary to perform all methods. Other methods that are not depicted or described may be introduced into the example methods and processes. For example, one or more additional methods may be performed before, after, simultaneously with, or between any of the described methods. Further, the method may be rearranged or reordered in other implementations. Moreover, the separation (separation) of the various system components in the implementation should not be understood as requiring such separation in all implementations, and it should be understood that the described components and systems may generally be integrated together in a single product or packaged into multiple products. In addition, other implementations are within the scope of the present disclosure.

Conditional language such as "may," "could," "might," or "may" is generally intended to convey that certain embodiments include or exclude certain features, elements, and/or steps, unless specifically stated otherwise, or otherwise understood within the context of use. Thus, such conditional language is generally not intended to imply that a feature, element, and/or step is required for one or more embodiments.

Linking language such as the phrase "at least one of X, Y, and Z," unless specifically stated otherwise, is generally understood given the context in which it is used to convey that an item, term, etc. can be X, Y, or Z. Thus, such linking language is generally not intended to imply that certain embodiments require the presence of at least one of X, at least one of Y, and at least one of Z.

The degree language used herein, for example, the terms "approximately," "about," "generally (substantially, generally)," and "substantially" as used herein, means a value, amount, or characteristic that is close to the stated value, amount, or characteristic that still functions as desired or achieves the desired result. For example, the terms "approximately," "about," "generally (substantially, generally)," and "substantially" may refer to an amount that does not exceed or equal to 10%, does not exceed or equal to 5%, does not exceed or equal to 1%, does not exceed or equal to 0.1%, and does not exceed or equal to 0.01% of the deviation from the stated amount. If the stated amount is 0 (e.g., nothing, none), the ranges recited above may be specific ranges, rather than a deviation from the stated value of no more than a specific %. For example, the deviation from the stated amount does not exceed or equal to 10 wt./vol.%, does not exceed or equal to 5 wt./vol., does not exceed or equal to 1 wt./vol.%, does not exceed or equal to 0.1 wt./vol.%, and does not exceed or equal to 0.01 wt./vol.%.

Some embodiments have been described in conjunction with the accompanying drawings. The figures are drawn to scale, but such ratios should not be limiting, as scales and proportions other than those shown are contemplated and are within the scope of the disclosed invention. Distances, angles, etc. are merely illustrative and do not necessarily show an exact relationship to the actual scale and layout of the devices shown. Components may be added, removed, and/or rearranged. Further, the disclosure herein of any specific features, aspects, methods, properties, characteristics, qualities, attributes, elements (components), etc. of the various embodiments may be used in all other embodiments set forth herein. In addition, it will be appreciated that any method described herein may be practiced using any device suitable for performing the steps described.

Although many embodiments and variations thereof have been described in detail, other modifications and methods of using the same will be apparent to those skilled in the art. Therefore, it should be understood that various applications, modifications, materials, and substitutions may be made by equivalents without departing from the unique and inventive disclosure herein or the scope of the claims.

Claims (16)

1. A synthetic garnet material comprising a structure including dodecahedral positions, the structure having bismuth occupying at least some of the dodecahedral positions and the structure having gadolinium, the garnet material having a dielectric constant of at least 31, and the synthetic garnet material excluding pyrobismuthite as a second phase. 2. The synthetic garnet material of claim 1, wherein the 3dB linewidth is less than 80. 3. The synthetic garnet material of claim 1, wherein the structure comprises gadolinium at a level of up to 1.0 unit. 4. The synthetic garnet material of claim 1, wherein the structure comprises at least 1.4 units of bismuth. 5. The synthetic garnet material of claim 4, wherein the structure comprises 1.4-2.5 units of bismuth. 6. The synthetic garnet material of claim 1, wherein the synthetic garnet material has a dielectric constant of at least 34. 7. The synthetic garnet material of claim 1, wherein the garnet material has a dielectric constant of at least 36. 8. The synthetic garnet material of claim 1, wherein the garnet material has a magnetization of 1900 πM _s or higher. 9. The synthetic garnet material of claim 1, wherein the material is incorporated into a base station. 10. The synthetic garnet material of claim 1, wherein the material is incorporated into a mobile phone antenna. 11. A modified synthetic garnet compound represented by the following formula: Bi _x Ca _y Gd _z Y _3-xyz Fe _5-y Zr _y O ₁₂ , where x is between 1.0 and 2.0, y is between 0.1 and 0.8, and z is between 0.2 and 1.9, and the modified synthetic garnet compound does not include pyrobismuth as a second phase. 12. The modified synthetic garnet composition of claim 11, wherein the modified synthetic garnet composition has a dielectric constant of at least 34. 13. The modified synthetic garnet composition of claim 11, wherein the material is incorporated into a mobile phone antenna. 14. A modified synthetic garnet compound represented by the following formula: Bi _x Ca _y Gd _z Y _3-xyz Fe _5-y Zr _y O ₁₂ , wherein the modified synthetic garnet compound has a dielectric constant of at least 34, and wherein the modified synthetic garnet compound does not include pyrobismuth as a second phase. 15. A method for manufacturing synthetic garnet with a high dielectric constant, the method comprising: Provides yttrium iron garnet structure; and More than 1.4 units of bismuth are inserted into the yttrium iron garnet structure to form a modified synthetic garnet structure without pyrobismuth as a second phase, the modified synthetic garnet having a dielectric constant of at least 34. 16. The method of claim 15, wherein the modified synthetic garnet has the composition Bi _x Ca _y Gd _z Y _3-xyz Fe _5-y Zr _y O ₁₂ , 1.0 < x < 2.0, 0.1 < y < 0.8, 0.2 < z < 1.9.