WO2002087870A1

WO2002087870A1 - Electric circuit components having multiple electric characteristics and methods of making such components

Info

Publication number: WO2002087870A1
Application number: PCT/US2002/011575
Authority: WO
Inventors: Richard A. Holl; Philip L. Lichtenberger
Original assignee: Holl Technologies Company
Priority date: 2001-04-12
Filing date: 2002-04-12
Publication date: 2002-11-07
Also published as: US6830806B2; US20020148640A1

Abstract

The invention is directed to methods for producing substrates (24) for electric circuits, particularly microwave frequency circuits, and electric components (52-62) for mounting thereon employing of polymers and filler materials, the latter having specific electric characteristics, such as dielectric constant and/or resistivity. Any one substrate or component can be manufactured to have multiple body regions (241 - 24n), each of which has a different characteristic such as dielectric constant (1 -n) or resistivity (R1 - Rn).

Description

ELECTRIC CIRCUIT COMPONENTS HAVING MULTIPLE ELECTRIC CHARACTERISTICS AND METHODS OF MAKING SUCH COMPONENTS

Technical Field of the Invention

This invention relates to new methods for the manufacture from polymer/filler materials of composite substrates such as are used for electric circuits, particularly electronic circuits, and to components for such circuits, which substrates and components can have different electric constants in respective different regions thereof, and also relates to articles so . manufactured. It also relates to such methods and components with which at least one of the regions constitutes packaging for another region or regions. Background Art

An important factor in circuit design, especially at the ever higher frequencies and switching speeds that characterize modem designs, is that of the electric properties of the materials used in the manufacture of the substrates and components. For example, the dielectric properties of substrate materials used for microwave transmission media directly impact other key physical properties thereof. Three of the most widely used transmission line media at microwave frequencies, which usually are regarded as those in the range from one to one hundred gigahertz (GHz), are those known as "StriplineV'Microstrip", and "CPW", which are "planar" systems that are able to use conventional printed circuit techniques in their fabrication.

Stripline is essentially a planar development of coaxial transmission lines in which the center and outer conductors have become square or rectangular and the side walls have extended to infinity to become two spaced flat parallel ground planes with the center conductor between them, the intervening space being filled with a core of solid dielectric material. The electric field is confined to a region in the immediate vicinity of the center conductor and is uniformly distributed within the core directly between the center conductor and the ground planes, with the properties of the dielectric material playing an important role in the design and functioning of the line, as will be explained in more detail below. In icrostrip the parallel wire transmission line used at lower frequencies has been converted to a configuration with the conductors in the form of rectangular strips on opposite sides of a core layer of dielectric material, one of the strips constituting the ground plane. The electric field is also predominantly confined to the area between the conductor strip and the ground plane, but the line effectively is unshielded making it more difficult to confine all of the energy in the vicinity of the conductor strip, making the properties of the dielectric material even more important than with Stripline. In CPW the ground plane and conductor strips are on the same side of the dielectric core and the ground plane is split into two spaced parallel parts with the signal or conductor strip running in the space between them and parallel to them.

As is well known, a major difficulty with microwave circuitry is the length of the signal wavelength relative to the physical dimensions of the signal processing circuit. For, example, at 1 6GHz, which is a common operating frequency for satellite communication, the signal wavelength in free space (air or vacuum) is only 1 .875cm (0.75in), resulting in complicated phase changes between nearby points in the circuit. All the dielectric materials employed in such circuits have a dielectric constant greater than one, and the wavelength decreases as the dielectric constant increases. The three parameters which are of utmost importance in transmission line design are its characteristic impedance (Z₀), line wavelength (_g), and attenuation constant (). These parameters are very sensitive to conductor shape and conductivity, discontinuities in the surfaces on which the conductors run, as caused by surface flaws, and the nature of the dielectric.

The choice and control of dielectric constant for the core material is a major determinant of the width, length and thickness requirements for its conductors, and therefore careful selection of the substrate materials according to their electric properties, particularly their dielectric constant values, is crucial to the design, layout, and overall resultant size of such circuits. Typically, the use of higher dielectric materials results in narrower conductor widths and shorter lengths, which facilitates overall reduction in size, and this is increasingly important as the trend continues toward miniaturization of such circuits. In addition, high frequency effects such as radiation loss, dielectric loss and inductive and capacitive interactions make microwave design a rigorous challenge.

Another important problem inherent in electronic circuit design is that of impedance matching, which is one that has been addressed since the earliest days of electronic circuit design, and which usually is required for optimum performance when source and load impedances are unequal. Although many approaches are practiced, the problem has become increasingly challenging with the miniaturized electronic circuits employed a microwave frequencies. This is because prior art methods typically involve the incorporation of extra components within a circuit, requiring additional space. Most commonly the matching is accomplished by inserting matching networks into the circuit between the source and the load. Simple examples might involve the use of an inductance/capacitance circuit to match unequal source and load impedances, or to optimize the gain of an amplifier. Clearly, the addition of such components and/or networks, requiring additional space, significantly hinders miniaturization efforts.

These and other methods of providing impedance matching involve the use of circuit boards that carry a pattern of conductors onto which components are assembled. The creation of these circuit patterns becomes quite complex, and has led to the development of multi-layer boards, resulting in increased size. Furthermore, the need to select certain components of certain sizes according to their impedance matching abilities limits the overall design flexibility in creating circuits, especially if subsequently one or more components must be replaced with another of different values of electric characteristics and physical size.

In addition to careful selection of materials for the circuits and components themselves, the problem of impedance matching can be complicated by the materials employed for packaging or enclosing the circuit. A successful package design for microwave circuits must meet criteria for impedance matching, low dielectric loss at microwave frequencies, low sensitivity of the dielectric material and conductors to temperature changes, and low capacitance of interconnection to the backside of the ground plane. Therefore, selection of materials according to their electric properties for electronics packaging is yet another important facet for the performance optimization of circuits.

A major problem resides in the difficulties in choosing a dielectric material that can meet the different requirements of different respective parts of a substrate, a component, or a packaging system, and usually this has required a careful choice on the part of the designer of a single material with which an optimum balance is achieved, accepting that the properties of the material are less suitable for some aspects than would be preferred.

As far as I am aware, there is not currently a method for impedance matching that essentially requires no additional hardware, no additional space, and no constraints on component selection in the circuit designing process. Current techniques simply do not provide the needed flexibility and space-saving characteristics. Disclosure of the Invention

A principal object of the present invention to provide new methods for the manufacture of electric circuit substrates, and of components for such circuits, and of packaging therefor, with which different materials of different dielectric constants can be used successfully in different regions of the substrates, or the components, or the packaging.

Another object is to provide methods for impedance matching in electric circuit substrates, components, and their packaging that are consistent with the current need for circuit miniaturization by reducing, if not eliminating, the need for additional circuitry or additional impedance matching components. In accordance with the invention there is provided an electric circuit substrate or electric circuit component characterized in that it comprises: at least a first body region of polymer/filler composite material having a first electric characteristic; and at least a second body region of polymer/filler composite material joined to the first region to have a boundary between them, the second body region having a second electric characteristic of value different from that of the same electric characteristic of the first body region.

Also in accordance with the invention there is provided a method of manufacturing an electric circuit substrate or electric circuit component comprising a body of polymer/filler composite material characterized by: forming at least a first body region of polymer/filler composite material having a first electric characteristic; and joining to the first body region a second body region of polymer/filler composite material at at least one boundary between them by a heat and pressure operation that melts at least polymer at the boundary to effect such joining, the second body region having a second electric characteristic of value different from that of the same electric characteristic of the first body region. Description of the Drawings

The foregoing and other objects, features, and advantages of the present invention will become apparent from the following description of embodiments thereof, given by way of example, in conjunction with the accompanying diagrammatic drawings, wherein:-

Figure I illustrates a press mold apparatus used for application of heat and pressure during any of the manufacturing processes of the invention;

Figure 2 illustrates a substrate preform manufactured in the press apparatus of Figure 1 , including a depository recess or well formed therein for reception of an insert constituting a different region; Figure 3 is a cross section through the substrate preform and insert of Figure 2, taken on the line 3-3 therein;

Figure 4 is a cross section similar to Figure 3 and illustrating a completed multi-dielectric substrate;

Figure 5 is a cross section similar to Figure 4 through a completed multi-dielectric substrate preform in which the depository well and insert extend completely through the thickness thereof;

Figure 6 is a perspective view from above of a completed substrate resulting from the substrate preforms of Figures 2 through 5;

Figures 7 through 10 are perspective views respectively of other completed substrates;

Figure 1 1 is a plan view from above of a prior art circuit component with internal parts thereof shown in broken lines;

Figure 1 2 is a cross section of the circuit component of Figure 1 1 taken on the line 1 2-1 2 in Figure 1 1 ;

Figure 1 3 is a cross section of a circuit component according to the invention, the part of the component below the line 1 3b-1 3b in Figure 14 being shown in solid lines, while the part thereof above the line 1 3a-1 3a in Figure 14 is shown in broken lines;

Figure 1 4 is a cross section of the component of Figure 13 taken on the line 1 4-14 in Figure 1 3;

Figure 1 5 illustrates a resistor manufactured according to prior art methods;

Figure 1 6 illustrates the manufacture of a resistor according to the invention made by a method of the invention; and

Figure 1 7 illustrates a final step in the manufacture of the resistors of Figure 1 6.

The same references are used for similar parts throughout the drawings whenever that is possible. Modes for Carrying out the Invention and Industrial Applicability

The present invention is directed to controlling the distribution of materials of different electric constants, such as the dielectric constant and resistivity constant, in articles manufactured from such materials comprising composite polymer/filler materials. The invention is further directed to providing such articles exhibiting a plurality of different electric constants that are carefully selected and precisely confined to particular regions of the article body, and also therefore to corresponding boundary areas of the surfaces thereof on to which electric conductors can be deposited or otherwise laid thereon. One benefit of the use of such different electric constant materials is the flexibility it gives to the circuit designers who are no longer limited to the use of a single material with corresponding electric constants that usually will not be the optimum for all parts of the circuit. Another benefit is the added facility given to impedance matching in the circuit design.

While impedance matching has long been an engineering goal in the design of circuits, the problem has traditionally been addressed by adding components to a circuit for "balancing" the input and/or output impedance thereof. As discussed above, such components usually require additional space on a circuit board, which impairs a different but very important engineering goal, namely miniaturization of the circuit. The present invention provides novel approaches to the manufacture of substrates and circuit components able to provide inter alia impedance matching without necessarily requiring additional space on a circuit board. Substrates are manufactured, according to the present invention, to have a plurality of different dielectric constants in different well-defined regions of the substrate. The dielectric constants can be selected according to the kind of component that will be mounted on the respective regions or portions of the substrate. By selectively matching the components whenever possible to appropriate dielectric substrate materials, and/or by manufacturing substrates with different dielectric constant values in respective locations it becomes possible to provide at least a more closely impedance matched circuit, possibly with a reduction in the number of additional impedance matching components, and possibly with the elimination of such need. The materials used hitherto in circuit substrates and packaging materials usually cannot be fused or adequately joined after their initial fabrication to provide precise, well defined, varied dielectric constant regions. Typically, ceramics such as polished fine-grained alumina have been materials of choice for electronic substrates because of certain desirable properties known in the art, but once formed cannot subsequently successfully be pressed or fused together to form a single piece. Multiple ceramic pieces can be adhered together with a suitable bonding agent, but usually such bonds are much weaker than the sintered ceramic materials, and such adhesion results in an undesirable additional abrupt break in the characteristics, such as the dielectric constant contributed by the adhesive. Other materials, such as the green composite matrix materials commonly used in circuit boards, do not possess sufficient heat resistance to withstand heating and fusing operations for the assembly of multiple articles into a single piece with multi-dielectric zones.

As an alternative to the use of ceramics, substrates and electronic packaging are sometimes formed of metal or semiconductor filler powders embedded in a glass or polymer matrix, such as polytetrafluoroethylene ( PTFE). Unfortunately, such matrix forming polymers have excessive thermal expansion coefficients and low inherent strength. As a result, recent advances in substrate material fabrication have focused on the addition of reinforcing materials, such as woven glass fiber cloth or glass fiber mat, into the substrate body. While these materials are effective to strengthen the substrate, inherently they create an inhomogeneous structure by preventing uniform distribution of filler materials into the matrix. Additionally, the fiber containing materials have proven very difficult to machine, as conventional machining tools draw out the strengthening fibrils which then wrap around and impede the tools.

My US Patent Serial Number 09/345,81 3 , filed July 2, 1 999 entitled "Composites of Powdered Fillers and Polymer Matrix", describes materials and methods for fabricating a wide array of substrates, wherein ceramic and other powders are mixed together with a non-polar, nonfunctionalized polymer material selected from the group comprising polyarylene ether-2, polyarylene ether-3, and polyarylene ether-4, being polymers found to have adequate adhesive properties while being sufficiently thermoplastic, which requirement is explained below. Such materials are described and claimed in US Patents Serial Nos. 5,658,994 and 5,874,51 6 of Air Products and Chemicals Inc. of Allentown, Pennsylvania, USA, and are low dielectric constant materials which were developed as intermetal and interlevel dielectric films for ULSI integrated circuit fabrication. My US Patent Application Serial No. 09/973,347 , filed October 5, 2001 , discloses other highly loaded composite materials using other polymers for the matrix.

The inventions therein result from the unexpected discovery that composite materials employing a matrix polymer can be manufactured with unusually high solids loadings by choosing as a matrix material a polymer which, while not necessarily of high strength in bulk, nevertheless exhibits sufficient high inherent adhesiveness toward finely ground filler materials of the kind that can be employed in combination with such matrix materials to produce electronic substrates. Also unexpectedly, the production of useful composite materials using such matrix materials requires a complete reversal of approach from that which has previously been employed in the production of composite materials.

A major problem in the prior art processes used, and in the substrates obtained thereby, is the progressive loss of mechanical strength that results as the filler solids content is increased, and hitherto attempts to incorporate more than about 40 volume percent generally has resulted in composites which are so friable that they literally collapse to a heap of sandlike material if in testing they are stressed to the degree required in commercial practice. Moreover, it has been found difficult with prior art processes to incorporate as much as 40 volume percent solids material, since the mixtures become so viscous that uniform mixing is virtually impossible. Consequently, the approach has of necessity been to incorporate only as much filler material as will result in a substrate of adequate mechanical strength, and to accept the lower desired electric characteristics that result. It was discovered that with the methods of the prior inventions in my patent and patent application referred to above, for the successful production of composite materials, the solids content must instead be increased to values well beyond those of the conventional prior art. An acceptable minimum for these new composite materials is 60 volume percent, in that such materials are of the required minimum mechanical strength, it being found that the mechanical strength increases with increased solids content, instead of decreasing, up to the value of about 95-97 volume percent, or up to the value at which the proportion of polymer is reduced below the minimum value required to maintain adequate adhesion between the uniformly distributed filler particles.

A possible explanation for this highly unexpected result is that although the chosen polymers exhibit unusually high adhesion, especially toward oxide materials such as silica, aluminum oxide, metal powders and boron nitride, they are not particularly mechanically strong, and therefore are most effective in this new and special application if employed in the form of very thin adherent layers interposed between the filler particles, such as can only be obtained with the methods of the inventions and when the solids contents are sufficiently high. It is difficult to specify with any degree of accuracy the optimum thicknesses for the interposed polymer layers; it is known that layers of 1 -3 micrometers are very successful in giving superior adhesion with adequate strength, and a possible upper limit is 40 micrometers (0.001 in).

Such composite materials can be made by mixing together the required portion by weight, or by volume, of particles of the chosen polymer material of sufficiently small dimension, or equivalent spherical dimension, e.g. in the range 0.1 to 50 micrometers, with the corresponding portion by weight or by volume of the chosen filler material, again of sufficiently small dimension, or equivalent spherical dimension, e.g. in the range 0.1 to 50 micrometers, and subjecting the mixture to a temperature sufficient to melt the polymer material, e.g. in the range 280-400°C and to a pressure, e.g. in the range 3.5 to 1 ,380MPa (500 to 200,000 psi), preferably 70 to 1 ,380MPa (1 0,000 to 200,000psi), sufficient to disperse the melted polymer material into the interstices between the particles of filler material. In alternative processes the polymer may be added in the form of a solution thereof, provided steps are taken to remove all of the solvent once the filler and polymer materials have been uniformly mixed together.

The resultant heated and pressurized composite mixture may be formed into sheets, films, tapes, or other articles useful in circuit design. In addition to an unmatched ease of manufacture of strong substrate materials, these composite materials have highly desirable electronic properties, and of particular interest are their dielectric properties. Thus, the invention permits careful selection and control of these electric properties. The composite materials may be fabricated by using powdered filler material that is a blend of two or more individual and potentially quite dissimilar materials that could not be joined by fusing them together, and in this way, the resultant effective dielectric constant of the resultant material can be carefully tailored and controlled. The relationship between the dielectric constant of the filler material or material mixture to the effective dielectric constant of the composite material is logarithmic, and therefore the composite must be heavily loaded with the filler material of higher dielectric constant value. The resultant effective dielectric constant can be determined using what is known as the Lichtenecker equation, which is well known to those skilled in the art. Generally, such composite materials should have at least 60%, and preferably at least 80% by volume filler material. The filler material is finely powdered and can be selected from, but is not limited to, a group comprising aluminium nitride, barium titanate, barium-neodymium titanate, barium copper tungstate, lead titanate, lead magnesium niobate, lead zinc niobate, lead iron niobate, lead iron tungstate, strontium titanate, zirconium tungstate, alumina, fused quartz, boron nitride, and metal powders.

The conditions required for the formation of such substrates and electric components are applied in a heated press or mold 10, as illustrated in Figure 1 , having a fixed heated platen 1 2 supporting a mold plate 14 in a cavity 1 6, and a movable heated platen 1 8 supporting a mold plate 20. In this particular embodiment the mold plate 20 supports a mold plate extension 22 which extends into the cavity 1 6 when the mold is fully closed. Assuming that the dielectric constant is the principal electric characteristic to be considered, a first mixture of filler and polymer materials with a particular first dielectric constant Σ^ is placed in cavity 1 6 and subjected to heat and pressure by moving the heated platens 14 and 20 together, resulting in a substrate body 24 as seen in Figures 2 through 6 that is of square external shape. Owing to the presence of the extension 22, which is also of square shape but smaller than the body 24, the resultant incomplete substrate preform has a square recess 26 therein extending partly through the thickness of the body 24. However, the recess may be any shape or dimension and the sides do not need to be straight. The composite material is easily machinable, so that instead of molding it in situ using the mold plate extension 22, the recess can be formed by removing material from the substrate preform body by any one of a number of different standard methods including, but not limited to, routing, drilling, cutting by laser or tool, or etching.

The substrate preform 24 is then placed in a press mold from which the extension 22 is absent and an insert 28 of precisely the correct size and shape to fit within the recess 26 and fill it is placed therein. The insert has been formed by substantially an identical process in another press mold from a second mixture of filler and polymer materials, employing the same polymer or one that is completely compatible therewith, but with a different filler so as to obtain a particular chosen second dielectric constant Σ₂ different from the first. Alternatively the recess can be filled with precisely the correct quantity of the second filler/polymer mixture that has not yet been subjected to a forming pressing and heating operation. It will be noted that the polymer employed has the characteristic that, at temperatures at which it will provide a self-sustaining substrate preform and insert, it is still sufficiently thermoplastic to flow as required under the heat and pressure applied to the two mixtures; the temperature and pressure of the final press operation can be more intense than those used to form the preform and insert, to the extent that cross-linking and thermosetting may now take place. The combination of substrate preform 24 and insert 28 is then subjected in the press to sufficient heat and pressure that the binding matrix-forming polymer melts to form a homogenous material that will flow throughout the bodies, binding the particles of filler material into a homogenous matrix structure, as illustrated by Figures 4 and 6, having well defined regions of two distinctive dielectric constants, while at the same time solidly binding the preform and insert together.

Figure 5 shows a resultant substrate using the steps described above for the embodiment of Figures 1 -4 and 6, but in which the recess 26 extends the full depth of the substrate preform 24 and the insert 28 is of corresponding thickness so as to completely fill the recess. It is also possible to place both the first and second mixtures into a mold cavity before they have been subjected to any form of binding operation, and then subject them simultaneously to the final binding temperature and pressure. Since only the polymer melts to a fluid state the filler material tends to remain where it has been placed in the mold cavity, but there is some loss of control of the location of the boundary or boundaries between the matrixes that are formed, and also loss of control of the effective dielectric constant in undetermined portions of the regions immediately adjacent the boundary or boundaries.

Other methods according to the invention entail manufacture of each multiple electric characteristic substrate through the fusion edge to edge of a plurality of substrate preforms, usually all of the same thickness, each preform having a unique different electric constant, or any number of preforms of the same electric constant being separated by one or more other preforms of different electric constants. Since for the purpose they must be heated in a final step to a temperature at which the polymer becomes fluid they will need to be constrained within a suitable mold, such that illustrated and described. For example, as shown in Figure 7, two smaller square preforms 24, and 24₂ are butted edge to edge with one another, and are in turn butted edge to edge with a rectangular preform 24₃ to provide a square substrate having three discrete areas of dielectric constant ∑ ∑₂ and Σ₃ respectively. Boundaries 30 between the preforms are discrete, with substantially no mixing of the composite materials between the three well- defined different dielectric regions. Of course, there may be some mixing, but if the process is practiced carefully, it can be limited to an acceptable degree. If necessary the preforms can be secured in the mold cavity before they joined together by a thin layer 32 (Figure 1 ) of adhesive polymer, constituting an affixing layer. This affixing layer is as thin as possible and has a melting point and/or an evaporation point low enough for it to melt or evaporate during the operation, leaving behind only the composite materials that were affixed thereon. If any residue does remain its effect on the dielectric constants of the regions must be taken into account.

As described above, when subjected to the heat and pressure within the press mold, the matrix-forming, binding polymer material within the multiple substrate preforms becomes re-softened and the substrates become fused, but not mixed, together. Various dielectric regions can be selected and positioned within a substrate according to impedance matching requirements of a particular circuit to be mounted thereon, and substrates having multiple dielectric regions can be manufactured to have a variety of patterns. For example, as shown in Figure 8, any number (four are shown) of preforms can be of irregular shapes which are fitted closely together as with a jigsaw puzzle. Figure 9 shows the same number of preforms united to provide a disc-shaped substrate, the inner one 24τ of smallest diameter being of disc shape, while the three outer ones 24₂- 24₄ are of annular shape fitting snugly one within the other. Figure 10 illustrates the manufacture of a cylindrical rod, such as might be used as the core for an inductive component, formed by adhering together a stack of the required length of a plurality of discs 24., - 24_n of different dielectric constant Σ, - ∑_n, as required to obtain a desired characteristic that varies along the length of the rod. The possible combination and resulting pattern of multiple dielectric regions within a substrate is unlimited, and can for example form a checkerboard pattern, a spiral pattern, or any other pattern. Figures 1 1 and 1 2 show a prior art circuit element construction in which an integrated circuit (IC) chip 34 is located in a well 36 in the substrate body 24. Line conductors 38 are mounted on the upper surface of the body 24 and are connected to respective bond pads 40 on the IC chip via wire connectors 42, while ground plane conductor 44 is mounted on the lower body surface. In such an element the ground plane conductor may also constitute a heat sink and is then thicker than is required for electric transmission only. . The well and chip are closed and protected by a lid having a wall part 46a which can be a polymer or polymer/filler material, and a roof part 46b which can be of the same material as the wall part, or can be of any other suitable material, including metals or polymer filler composites in which the filler is metal or graphite. Even if the same material is used for the substrate 24 and the cover wall part 46a, so that the dielectric constant Σ, is the same, each of the line conductors 38 is subjected to a different environment along its length. Thus, the middle part (between the lines 48 in Figures 1 1 and 1 2) is sandwiched between the substrate 24 and the cover wall part 46a, and therefor needs to be treated as a stripline conductor when calculating its characteristic impedance, while the two end parts (between lines 48 and 50 in Figures 1 1 and 1 2) have one side open to air and therefore need to be treated as a microstrip conductor for the same purpose.

A basic microstrip circuit consists of a dielectric substrate such as the substrate 24 of thickness h and relative dielectric constant ∑ while the conductor line 36 has a width w and thickness t. From transmission line theory the series inductance L and shunt capacitance C per unit length of the transmission line are so related to the characteristic impedance that they change as the substrate dielectric constant changes, and must also change if the characteristic impedance is to remain constant, which is essential if the transmission line is to have uniform transmission characteristic from one end to the other. Unless great care is taken the resultant impedance mismatch can be so severe that adequate microwave transmission is impossible. The inductance decreases with increasing line width and, for example, a 50ohm line on fused silica ( ∑., = 3.88) must be twice as wide as a line on a comparably thick alumina substrate ( Σ-, = 9.9). A fundamental design problem with these transmission lines is that of finding the proper values of w for values of h (w/h ratio) and the effective dielectric constant (∑_eff ), all of which are interdependent. With this prior art example the solution adopted to maintain the characteristic impedance constant along the length of the conductor is to decrease the width, and perhaps also the thickness, of the "stripline" portion 38a of the conductor sandwiched between the substrate and the wall part 46. Such a solution requires very precise fabrication of the conductors, which are already very thin and narrow, and gives relatively limited facility for such adjustment.

A solution in accordance with the invention, as illustrated by Figures 1 3 and 1 4, enables the conductors 38 to be of uniform width and thickness along their lengths while also having sufficiently uniform effective characteristic impedance along their lengths. To this end a recess 26 of the same length as the sandwiched portion of the conductor, and registering therewith, and of slightly wider width than the conductor, is formed in the substrate body 24, as by precision molding while the substrate body is formed, or by precision routing material from the formed body. The substrate has a dielectric constant value ∑_{1 #} while the dielectric constant value of at least the component cover wall part is indicated as Σ₂, but which may be the same as Σ, . An insert 28 of precisely the size and shape to fit snugly within the recess 26, and of carefully selected and controlled dielectric constant Σ₃ is placed in the recess and permanently secured therein by a heating and pressing operation as described. Since widely varying mixtures of filler materials can be used a correspondingly wide and precisely determined range of dielectric constant values are available for accurate impedance matching by adjustment of the mixture formulation. It will be noted that with, for example, n different regions, each of the regions 24., - 24_n , and each of inserts 28., - 28_n when inserts are used, will not only have a different dielectric constant, but usually will also have a different unique resistivity R., - R_n respectively, as well as differences in other electric and physical characteristics, and all of these will need to be considered in the final circuit design. Even further variation is possible by forming the substrates and/or inserts of thin layers of different electric constants. Equivalent solutions are possible with CPW transmission lines where part of the conductor line and/or the ground plane conductors are sandwiched between two dielectric layers, instead of lying on a single layer.

The methods of the invention therefore apply to the manufacture of electric and electronic components comprising a body of material having physical and/or electric characteristics, such as resistivity (or conductance) and/or dielectric constant that must be carefully controlled and are not easily achieved using any single material to form the body. The methods are used for a variety of components including, but not limited to, resistors (for which of course the prime consideration must be the resistivity), and capacitors and inductors (for which of course the prime considerations must be respectively the dielectric constant and the magnetic permittivity). Additionally, the methods allow for the manufacture of components having any chosen impedance, resistance, capacitance, inductance, etc. without necessarily compromising the size of the component itself, so that components having different electric characteristics can be manufactured to have virtually identical physical dimensions. This can be especially important when designing the physical layout of a circuit, as the designer will be able to select a component of any desired characteristic without regard to size; if subsequently a component must be changed for one of a different value then it can be done without having to move other elements on the substrate. Moreover, prior art methods frequently produce components with leads that are not flush and are not easily accessible without first being trimmed, such as by laser, while the methods of the invention can readily be made to yield components with flush leads, eliminating the need for such trimming. Such freedom removes many sizing constraints previously known to the circuit design process.

A typical structure for a prior art resistor is shown in Figure 1 5. Conductive leads 52 are formed of highly conductive, non-corroding materials, such as gold or silver, and heat pressed onto a ceramic substrate 54. A body 56 of a highly resistive material ( a resistor ink) of prescribed dimensions and resistivity is screen printed on one surface of the substrate 54 in engagement with the leads 52. Typical resistor inks comprise a mixture of ruthenium dioxide (RuO₂) or pyrochlor together with glass powder in a suitable suspension medium. When heated the suspension component is removed, while the glass powder forms an adhesive matrix, thus producing the resistor. The substrate 54 must then be trimmed, such as by laser trimming, to expose the conductive leads, adding an additional step to the manufacturing process.

As illustrated in Figure 1 6, in a method of the invention a thin layer 32 of an appropriate affixing binding polymer, which may be that used in the formation of the substrates in view of its high adhesiveness and complete subsequent compatibility, is deposited upon heated platen 1 2 of the press 10 and is used to hold firmly in place conductive leads 52 placed thereon, also formed of any highly conductive material, such as silver or gold. A pre-formed substrate of an appropriate material, such as in the case of a resistor a mixture of ruthenium dioxide and a polyarylene ether polymer, assembled as required with any of the configurations described above, is diced to form small resistor chips 58 of size and configuration required for the resistance value of the resultant resistor when finally produced. As is illustrated the size of the chips employed can vary in size, usually in accordance with the specific requirement as to the resistance value thereof. The resistor chips are placed upon respective pairs of conductive leads 52 and pressed into the binding polymer sheet 32. The mold cavity is then back filled to the level as indicated, with any high temperature encapsulating material 60, preferably the same polymer that has been used in the production of the resistor chips so as to be completely compatible therewith. A thin metal layer 62 of, for example, platinum is then deposited on top of the back-fill material, and the assembly is subjected to heat and pressure between heated platens 1 2 and 1 8. The heat and pressure are such that the binding polymer sheet 32 melts sufficiently to be squeezed out from beneath the leads 44 and become incorporated in the backfill material, exposing the conductive leads. The result is an electric component which is characterized in that each of the regions 24τ - 24_n has as its electric characteristic a different unique resistivity R, - R_n respectively. As shown in Figure 1 7, when the resultant workpiece is removed from the press the result is a block comprising a plurality (usually a very large number) of resistor components, each with flush exposed conductive leads and no need for trimming. The block is then separated into the individual resistors along separator lines 64, preferably to be of identical exterior dimensions, despite the different sizes and resistance values of the resistor chips 58 therein.

Thus, although the invention has been described as applied to the production of resistors it is apparent that it can equally be applied to any component in which the operative element can be formed from a block of a polymer/filler material. For example, in other embodiments the resistor chips or their equivalent constitutes one region of the component, and the component can be made to have any desired impedance, including resistance and/or capacitance and/or inductance by use in the backfill material 60 of particles of chosen electric characteristic, such as resistivity, dielectric constant, etc. to produce components with any desired effective electric characteristic. In such embodiments the backfill material not only serves as an intrinsic part of the component but also as packaging material enclosing the other parts. It will be understood by those skilled in the art that any two filler materials of one different electric characteristic, such as dielectric constant, are unlikely to have any other characteristic, such as resistivity, thermal coefficient of expansion, etc. identical, and there are many other such characteristics that usually will need to be carefully considered in the final choice of the material to be used. The result is a manufacturing process capable of producing substrates for electric circuits and electric components having equal physical dimensions and variable effective characteristics. Components manufactured according to the methods of the invention can be mounted upon substrates having multiple dielectric constants, as previously described. Those skilled in the art will recognize that these novel manufacturing methods make possible an entirely new class of substrates to be used in circuit design, whereby impedance matching is inherently facilitated with the possibility of exceptional flexibility in component layout. Many modifications and variations of the invention are possible with the guidance of the above teachings. The polymer or polymers that can be employed is not limited to the polyarylene ethers specifically disclosed, but can be any non-polar polymer matenal with adequate dielectric and adhesive properties and sufficient thermoplastic capability to permit it to function also a bonding agent.

List of Reference Signs

1 0. Press

1 2. Fixed heated press platen

14. Mold plate on platen 1 2

1 6. Mold cavity

1 8. Movable heated press platen

20. Mold plate on platen 1 8

22. Mold plate extension

24. Substrate or substrate preform body

26. Recess in body 24 when provided

28. Substrate insert for recess 26

30. Boundaries between substrate preforms

32. Affixing layer of material

34. Integrated circuit (IC) chip

36. Well in substrate body 24 receiving chip 34

38. Line conductors on substrate

38a. Reduced width portion of conductor 38

40. Bonding pads on IC

42. Wire connectors between conductors 38 and pads 40

44. Ground plane conductor (plus heat sink)

46a. Wall part of component cover

46b. Roof part of component cover

48. Lines delineating mid-part of conductors 38 (Figures 1 1 and 1 2)

50. Lines delineating ends of conductors 38 (Figures 1 1 and 1 2)

52. Electric component conductive leads (Figures 1 5-1 7)

54. Ceramic substrate for prior art resistor

56. Body of heated resistive ink for prior art resistor

58. Resistor bodies

60. Backfill material (potential packaging material)

62. Conductive layer

64. Separator lines between individual components Σ. Dielectric constant of substrate materials h. Thickness of dielectric substrate 28

R. Resistivity constant of substrate materials t. Thickness of conductor 38 w. Width of conductor 38

Claims

1 . A substrate or electric component comprising a body of polymer/filler composite material characterized in that it comprises: at least a first body region (24 or 24.,) of polymer/filler composite material having a first electric characteristic; and at least a second body region (28 or 24₂) of polymer/filler composite material joined to the first region to have a boundary (30) between them, the second body region having a second electric characteristic of value different from that of the same electric characteristic of the first body region.

2. A substrate or electric component as claimed in claim 1 , characterized in that the body regions (24., - 24_n) have been joined to each other at their respective boundaries (30) by a heating and pressing operation.

3. A substrate or electric component as claimed in claim 1 or 2, characterized in that each of said body regions (24^ - 24_n) has as its electric characteristic a different dielectric constant (∑., - ∑_n respectively) .

4. A substrate or electric component as claimed in any one of claims 1 to 3, characterized in that each of said body regions (24., - 24_n) has as its electric characteristic a different resistivity (R., - R_n respectively).

5. A substrate or electric component as claimed in any one of claims 1 to 4, characterized by multiple body regions (24τ - 24_n) forming a pattern.

6. A substrate or electric component as claimed in claim 5, characterized by multiple body regions (24^ - 24_n) forming a nonuniform pattern.

7. A substrate or electric component as claimed in any one of claims 1 to 6, characterized in that at least one of the body regions (24τ - 24_n) has as its filler material a mixture of filler materials of selected electric characteristics.

8. A substrate or electric component as claimed in claim 7, characterized in that each polymer/filler material of the body regions (24τ - 24_n) comprises a mixture of filler materials of selected electric characteristics and nonpolar, nonfunctionalized polymer material, each with at least 80% by volume of filler material.

9. A substrate or electric component as claimed in any one of claims 1 to 8, characterized in that the second body region is disposed in a recess in the first body region.

10. A substrate or electric component as claimed in any one of claims 1 to 9, characterized in that: the substrate or component body (28) carries at least one transmission line conductor (38) on its surface that extends over at least two different regions thereof; that the transmission line conductor is of uniform width and thickness along its length over the said at least two different regions; and that adjustment of the transmission characteristic of the transmission line is effected by adjustment of the dielectric constant values in the different regions.

1 1 . A method of manufacturing a substrate or electric circuit component characterized by forming at least a first body region (24 or 24.,) of polymer/filler composite material having a first electric characteristic; and joining to the first body region a second body region (28 or 24₂) of poiymer/filler composite material at at least one boundary (30) between them by a heat and pressure operation that melts at least polymer at the boundary (30) to effect such joining, the second body region having a second electric characteristic of value different from that of the same electric characteristic of the first body region.

1 2. A method as claimed in claim 1 1 , characterized in that each of said body regions (24, - 24_n) has as its electric characteristic a different dielectric constant (Σ, - ∑_n respectively) .

1 3. A method as claimed in claim 1 1 or 1 2, characterized in that each of said body regions (24, - 24_n) has as its electric characteristic a different resistivity (R, - R_n respectively).

14. A method as claimed in any one of claims 1 1 to 1 3, characterized by multiple body regions (24, - 24_n) forming a pattern.

1 5. A method as claimed in claim 14, characterized by multiple body regions (24, - 24_n) forming a nonuniform pattern.

1 6. A method as claimed in any one of claims 1 1 to 1 5, characterized by the steps of forming the first region (24 or 24,) having a recess (26) therein; depositing the second body region (28 or 24₂) in the recess; and thereafter heating and pressing the body to adhere the two body regions together.

1 7. A method as claimed in any one of claims 1 1 to 1 5, characterized by the steps of forming the first region (24, or 24) having a recess (26) therein; depositing a polymer/filler composite material capable of forming the second body region (24₂ or 28) in the recess; and thereafter heating and pressing the body with sufficient pressure and at sufficient temperature to melt the polymer material and disperse the melted polymer material into the interstices between the particles of filler material and to adhere the two regions together.

1 8. A method as claimed in any one of claims 1 1 to 1 7, characterized in that at least one of the body regions (24, - 24_n) has as its filler material a mixture of filler materials of selected electric characteristics.

1 9. A method as claimed in claim 1 8, characterized in that each polymer/filler material of the body regions (24, - 24_n) comprises a mixture of filler materials of selected electric characteristics and nonpolar, nonfunctionalized polymer material, each with at least 80% by volume of filler material.

20. A method as claimed in any one of claims 1 1 to 1 9, characteriased by the steps of: applying an affixing thin layer of a polymer material (32) to a heated mold platen (1 2) within a mold cavity (1 6); placing pairs of conductive leads (52) on the thin layer at spaced intervals; placing a plurality of component bodies (58) comprising first body regions (24) on respective pairs of the conductive leads; backfilling the mold cavity (1 6) with polymer/filler mixture (60) constituting at least one second body region; applying sufficient heat and pressure to melt the polymer of the backfilled mixture and force it into the interstices between the particles of its filler material, thereby forming a workpiece containing at least one manufactured electric component; and removing the resultant workpiece from the press mold and dividing it into multiple individual electric components, each component comprising a respective component body (58) with associated conductive leads (52) .

21 . A method as claimed in claim 20, characterized in that the thus divided multiple individual electric components are all of the same external dimensions.

22. A method as claimed in any one of claims 1 1 to 21 , characterized in that the electric component is a resistor, or a capacitor, or an inductor, or a transmission line.

23. A method as claimed in any one of claims 1 1 to 22, characterized in that: at least one transmission line conductor (38) is applied to a surface of the substrate or component body (28) so as to extend over at least two different regions thereof; that the transmission line conductor is of uniform width and thickness along its length over the said at least two different regions; and that adjustment of the transmission characteristic of the transmission line has been effected by adjustment of the dielectric constant values in the different regions.