Magnetic power components have a total addressable market of over $20 billion. They impact the progress in the multi-trillion dollar electronics industry because power delivery and efficiency limits the performance of future electronic systems. A key to advancing this massive market is the development of magnetic components with high power density and efficiency that can be fabricated with scalable processes at low cost.
Embodiments of the subject invention provide novel and advantageous stacked magnetic cores (e.g., for inductors and transformers) that can achieve high density with smaller lateral dimensions (or footprint), as well as methods of fabricating and using the same. A stacked magnetic core can be fabricated by depositing nanomagnetic films with control in composition and nanostructure via a continuous electroplating process. The magnetic films are interspersed with thin adhesive films (that can be insulating) in an automated roll-to-roll process to relieve both eddy currents and stresses that may otherwise be present in the magnetic films. That is, the magnetic films and adhesive films are disposed in an alternating fashion, with a first adhesive film disposed on a first magnetic film, a second magnetic film disposed on the first adhesive film, a second adhesive film (if present) disposed on the second magnetic film, and so on (for as many magnetic films and adhesive films as are present). The adhesive films can keep the magnetic films completely electrically isolated from each other, while also adhering adjacent magnetic films to each other.
In an embodiment, a method of fabricating a stacked magnetic core for an electrical component can comprise: performing an electroplating process under a magnetic field to form a first magnetic film on a plating carrier substrate; forming a first adhesive film on an upper surface of the first magnetic film; performing the electroplating process under the magnetic field to form a second magnetic film on the plating carrier substrate; forming a second adhesive film on an upper surface of the second magnetic film; removing the first magnetic film with the first adhesive film thereon from the plating carrier substrate and disposing it on an interim substrate; and removing the second magnetic film with the second adhesive film thereon from the plating carrier substrate and disposing it on the first adhesive film, forming the stacked magnetic core comprising the first magnetic film, the first adhesive film on the first magnetic film, the second magnetic film on the first adhesive film, and the second adhesive film on the second magnetic film. The first magnetic film can be electrically insulated from the second magnetic film by the first adhesive film, and a footprint of the stacked magnetic core (measured in a first plane having the upper surface of the first magnetic film) can be, for example, 10 square millimeters (mm2) or less (e.g., 5 mm2 or less, or 1 mm2 or less). The method can further comprise removing the first magnetic film, the first adhesive film, the second magnetic film, and the second adhesive film from the interim substrate. The stacked magnetic core can have an inductance density of, for example, at least 5 nanohenries per square millimeter (nH/mm2) (e.g., at least 10 nH/mm2). The stacked magnetic core can have a coercivity of, for example, no more than 1 Oersted (Oe). The first magnetic film and the second magnetic film can each be, for example, an alloy of nickel, iron, and optionally cobalt (e.g., NiFe or CoNiFe). A thickness of the first adhesive film (measured in a first direction perpendicular to the upper surface of the first magnetic film) can be in a range of, for example, from 0.01 micrometers (μm) to 10 μm (or any subrange therein; e.g., from 0.05 μm to 5 μm, or from 0.1 μm to 1 μm), and a thickness of the second adhesive film (measured in the first direction) can be in a range of, for example, from 0.01 μm to 5 μm (or any subrange therein; e.g., from 0.05 μm to 5 μm, or from 0.1 μm to 1 μm). A thickness of the first magnetic film (measured in the first direction) can be in a range of, for example, from 0.01 micrometers (μm) to 10 μm (or any subrange therein; e.g., from 0.05 μm to 5 μm, or from 0.5 μm to 2 μm), and a thickness of the second magnetic film (measured in the first direction) can be in a range of, for example, from 0.01 μm to 10 μm (or any subrange therein; e.g., from 0.05 μm to 5 μm, or from 0.5 μm to 2 μm). Each of the first adhesive film and the second adhesive film can be, for example, a polymer adhesive film or a metal-polymer composite film. The performing of the electroplating process can comprise performing the electroplating process in an electroplating bath having a rectangular frame disposed around the electroplating bath; the rectangular frame comprising a first side, a second side perpendicular to the first side, a third side parallel to the first side, and a fourth side parallel to the second side; each of the first side and the third side being longer than each of the second side and the fourth side; the first side comprising a first hard magnet; the second side comprising a first soft magnet; the third side comprising a second hard magnet; and the fourth side comprising a second soft magnet. The first hard magnet and the second hard magnet can be disposed such that they have the same North-South orientation as each other. Each of the first soft magnet and the second soft magnet can be, for example, a soft stainless steel magnet. The removing of the first magnetic film with the first adhesive film thereon from the plating carrier substrate and disposing it on an interim substrate can be performed before the performing of the electroplating process to form the second magnetic film on the plating carrier substrate. The performing of the electroplating process can comprise disposing a mask on the plating carrier substrate before forming the first magnetic film and before forming the second magnetic film, such that the respective magnetic film is formed on the plating carrier substrate only where the mask is absent, and the method can further comprise removing the mask before removing first magnetic film from the plating carrier substrate and removing the mask before removing second magnetic film from the plating carrier substrate (e.g., before disposing the first adhesive film on the first magnetic film and before disposing the second adhesive film on the second magnetic film, respectively).
In another embodiment, a method of fabricating a stacked magnetic core for an electrical component can comprise: a) forming a plurality of magnetic film-insulating film pairs, each magnetic film-adhesive film pair comprising an adhesive film disposed on an upper surface of a magnetic film, on a plating carrier substrate by performing an electroplating process under a magnetic field to form the respective magnetic film on the plating carrier substrate and then forming the respective adhesive film on an upper surface of the magnetic film; b) removing each magnetic film-adhesive film pair on its own, independently of the other magnetic film-adhesive film pairs; and c) disposing each magnetic film-adhesive film pair on an interim substrate to form the stacked magnetic core comprising a stacked structure of the plurality of magnetic film-adhesive film pairs. The magnetic film of each magnetic film-adhesive film pair can be electrically insulated from the magnetic film of each other magnetic film-adhesive film pair by its adhesive film and the adhesive film of the magnetic film-adhesive film pair with which it is in direct contact. A footprint of the stacked magnetic core (measured in a first plane having the upper surface of the first magnetic film) can be 10 mm2 or less (e.g., 5 mm2 or less, or 1 mm2 or less). Steps a), b), and c) can be performed in an automated, roll-to-roll process or in a parallel process. The method can further comprise removing the plurality of magnetic film-adhesive film pairs from the interim substrate. A thickness of the adhesive film of each magnetic film-adhesive film pair (measured in a first direction perpendicular to the first plane) can be in a range of, for example, from 0.01 μm to 5 μm (or any subrange therein; e.g., from 0.05 μm to 5 μm, or from 0.1 μm to 1 μm). A thickness of the magnetic film of each magnetic film-adhesive film pair (measured in the first direction) can be, for example, from 0.01 μm to 5 μm (or any subrange therein; e.g., from 0.05 μm to 5 μm, or from 0.5 μm to 2 μm). The performing of the electroplating process in step a) can comprise performing the electroplating process in an electroplating bath having a rectangular frame disposed around the electroplating bath; the rectangular frame comprising a first side, a second side perpendicular to the first side, a third side parallel to the first side, and a fourth side parallel to the second side; each of the first side and the third side being longer than each of the second side and the fourth side; the first side comprising a first hard magnet; the second side comprising a first soft magnet; the third side comprising a second hard magnet; the fourth side comprising a second soft magnet; and the first hard magnet and the second hard magnet being disposed such that they have a same North-South orientation as each other. Each of the first soft magnet and the second soft magnet can be, for example, a soft stainless steel magnet. The performing of the electroplating process in step a) can further comprise disposing a mask on the plating carrier substrate before forming the respective magnetic film, such that the respective magnetic film is formed on the plating carrier substrate only where the mask is absent, and removing the mask before removing the respective magnetic film from the plating carrier substrate. The stacked magnetic core can have an inductance density of, for example, at least 5 nH/mm2 (e.g., at least 10 nH/mm2). The stacked magnetic core can have a coercivity of, for example, no more than 1 Oe. The magnetic film of each magnetic film-adhesive film pair can be, for example, an alloy of nickel, iron, and optionally cobalt (e.g., NiFe or CoNiFe). The adhesive film of each magnetic film-adhesive film pair can be, for example, a polymer adhesive film or a metal-polymer composite film.
In another embodiment, a method of fabricating a stacked magnetic core for an electrical component can comprise: performing an electroplating process under a magnetic field to form a first magnetic film on a plating carrier substrate; forming a first adhesive film on an upper surface of an interim substrate; removing the first magnetic film with the first adhesive film thereon from the plating carrier substrate and disposing it on the interim substrate; performing the electroplating process under the magnetic field to form a second magnetic film on the plating carrier substrate; forming a second adhesive film on an upper surface of the interim substrate; and removing the second magnetic film with the second adhesive film thereon from the plating carrier substrate and disposing it on the second adhesive film, forming the stacked magnetic core comprising the first magnetic film, the first adhesive film on the first magnetic film, the second magnetic film on the first adhesive film, and the second adhesive film on the second magnetic film. The first magnetic film can be electrically insulated from the second magnetic film by the first adhesive film.
In another embodiment, a method of fabricating a stacked magnetic core for an electrical component can comprise: performing an electroplating process under a magnetic field to form a first magnetic film on a plating carrier substrate; removing the first magnetic film from the plating carrier substrate forming a first adhesive film on both surfaces of the magnetic film; performing the electroplating process under the magnetic field to form a second magnetic film on the plating carrier substrate; removing the second magnetic film from the plating carrier substrate; forming a second adhesive film on both surfaces of the magnetic film; and forming the stacked magnetic core comprising the first magnetic film, the first adhesive film on the first magnetic film, the second magnetic film on the first adhesive film, and the second adhesive film on the second magnetic film. The first magnetic film can be electrically insulated from the second magnetic film by the first adhesive film.
In another embodiment, the films as described herein (e.g., magnetic films) can be plated on a carrier and then released from it to form individual free-standing films. Such films can then be appropriately clamped to a frame for easier handling and then coated with an adhesive resin. These films can thus be coated in parallel, and then stacked and laminated to form a multi-layered substrate. This approach has the advantage of higher throughput as multiple free-standing magnetic films can be coated simultaneously and laminated in a single step (see also
Embodiments of the subject invention provide novel and advantageous stacked magnetic cores (e.g., for inductors and transformers) that can achieve high density with smaller lateral dimensions (or footprint), as well as methods of fabricating and using the same. A stacked magnetic core can be fabricated by depositing nanomagnetic films with control in composition and nanostructure via a continuous electroplating process. The magnetic films are interspersed with thin adhesive films (that can be insulating) in an automated roll-to-roll process to relieve both eddy currents and stresses that may otherwise be present in the magnetic films. That is, the magnetic films and adhesive films are disposed in an alternating fashion, with a first adhesive film disposed on a first magnetic film, a second magnetic film disposed on the first adhesive film, a second adhesive film (if present) disposed on the second magnetic film, and so on (for as many magnetic films and adhesive films as are present). The adhesive films can keep the magnetic films completely electrically isolated from each other, while also adhering adjacent magnetic films to each other. The adhesive films can be polymer films or metal-polymer composite films (though embodiments are not limited thereto) and can suppress the eddy currents and coupling between the magnetic films that would increase the coercivity (if the adhesive films were not present). By interspersing with thin adhesive films, a permeability of 800-1000 with a coercivity of less than 1 Oersted (Oe) can be achieved with a frequency stability of 100 megahertz (MHz) and beyond, and alternating current (AC) losses of less than 1%.
Each magnetic film can have a thickness of, for example, 0.01 μm to 10 μm (or any subrange therein; e.g., from 0.1 μm to 5 μm, or from 0.5 μm to 2 μm), and each adhesive film can have a thickness of, for example, 0.01 micrometers (μm) to 10 μm (or any subrange therein; e.g., from 0.05 μm to 5 μm, or from 0.1 μm to 1 μm). The footprint of the stacked magnetic core (i.e., the lateral area taken in a plane perpendicular to the thickness direction) can be, for example, less than 10 square millimeters (mm2) (e.g., less than 5 mm2, less than 3 mm2, less than 2 mm2, or less than 1 mm2). The, or each, magnetic film can be, for example, CoNiFe, though embodiments are not limited thereto. An electrical component (e.g., an inductor or a transformer) can comprise the stacked magnet core as its core, and such an electrical component can further comprise a coil, one or more backing layers, one or more vias, or other elements necessary to complete the electrical component. The electrical component can be configured to have a closed flux path.
The fabrication methods of embodiments of the subject invention can be based on electroplated nanocrystalline films on carrier substrates. Nanomagnetic films can be deposited with control in composition and nanostructures by a continuous electroplating process. A core having a single, monolithic magnetic film would quickly run into losses from eddy currents, process issues from stresses, and delamination and other reliability challenges. Embodiments of the subject invention address these problems by alternating the magnetic films with thin adhesive films (e.g., in an automated roll-to-roll process, or a sequential layering process from prefabricated films on carrier substrates, or coating the adhesive film directly on free-standing magnetic films) to relieve both the eddy currents and stresses. The insulating layers can suppress the eddy currents and can also prevent or inhibit coupling between the magnetic layers that would otherwise increase the coercivity of the core. By interspersing with thin insulating layers, a permeability of 800-1000 with a coercivity of less than 1 Oe can be achieved, as can a frequency stability of 100 MHz and above, and AC losses of less than 1%.
Wafer- or substrate-integrated nanomagnetic films with electroplated nanocrystalline films that have precise control in composition and nanostructures are provided, and these magnetic films are multilayered with thin insulating layers. This can be done on a substrate (e.g., on a carrier substrate), and the fabrication methods can provide scalability to adequate thickness, low eddy current losses, low stress, mechanical flexibility, and high reliability.
Magnetic properties such as coercivity, eddy current losses, and magnetostriction can degrade the performance of an electrical component unless the structure (e.g., the nanostructure) and composition meet the exchange coupling criteria. Controlling the nanostructure with electrochemical processes to achieve low coercivity and high permeability is an important goal. This is met at least in part by magnetic alloy design with dopants and the design of bath chemistry to achieve the composition. Also, magnetic annealing can be used to create anisotropic properties.
Plated films can be continuously released in an electroplating process, and the plated films can be transferred to a different substrate (i.e., a substrate different from what may be used in the electroplating process) with thin adhesive films (e.g., in a roll-to-roll process, such as an automated roll-to-roll process, or a sequential layering process from prefabricated films on carrier substrates, or coating the adhesive film directly on free-standing magnetic films). The films can be deposited onto a (relatively) thick carrier substrate via direct in situ patterning, and this can be done with or without any subsequent dry-etching or laser ablation steps to pattern the films (that is, such steps are not necessarily required but may be performed in certain embodiments). The micropatterned films can be transferred onto a different substrate (e.g., a laminate substrate) as insulating layer-isolated nanomagnetic layers. A roll-to-roll or continuous batch process, which combines plating with in situ patterning, film-transfer, and layering, can be used for high throughput and low cost. Such processes can be used in standard package integration processes to form electrical components (e.g., solenoid, spiral, and/or toroid inductors and/or transformers).
Embodiments of the subject invention provide plated magnetic composite films that achieve high Ms, high permeability, and low coercivity. Thin films (e.g., in a range of 0.1 μm to 10 μm (such as 2 μm or about 2 μm)) can be electroplated and easily released from a carrier substrate to form a layered composite structure by utilizing thin insulating layers. The films can be deposited onto a thick carrier substrate with a direct in situ patterning without the need for any subsequent lithographic steps to pattern them. These micropatterned films can be transferred onto a different substrate (e.g., a laminate substrate) as electrically-isolated nanomagnetic layers. A high throughput batch process with low cost was achieved.
An electroplating of nanostructured magnetic films (e.g., metals and/or metal alloy films) with easy release can be achieved. For example, the magnetic films can be an alloy of nickel (Ni), iron (Fe), and/or cobalt (Co) (e.g., NiFe, CoNiFe, or an alloy thereof). Plating can be performed with a stable plating bath that can yield more than 1 micrometer/minute (μm/min) of plating. The bath can be, for example, a sulphate bath (which can be utilized because of its easy scalability). In the case of the film being an alloy of Ni, Fe, and/or Co, the ratio of Fe(II)/Co(II) in the bath determines the final alloy composition. CoFe is usually activation-controlled at the cathode surface and not mass-transport-controlled. It is critical to stabilize the bath to prevent or inhibit the oxidation of Fe′ to Fe′ and prevent or inhibit the precipitation of hydroxide, and also suppress the evolution of hydrogen at the cathode. By depositing under a magnetic field, the film coercivity in the hard axis is substantially reduced while also enhancing the current-handling. A uniform unidirectional direct current (DC) magnetic field (e.g., of at least 500 Gauss, such as 1000 Gauss or about 1000 Gauss) can be applied to the working electrode (cathode). This can help create in situ magnetic orientation to achieve high field anisotropy and low coercivity in the hard axis. Both these attributes are extremely important in achieving high power density with high efficiencies. The permanent field can be created by surrounding the plating carrier with a frame (e.g., a rectangular frame) including hard magnets and soft magnets. Two parallel sides (e.g., the longer sides of the frame) can have hard magnets with the same North-South orientation as each other. The other two sides (e.g., the shorter sides of the frame), which can be parallel to each and perpendicular to the sides with the hard magnets, can comprise (or be made of) soft magnets (e.g., soft stainless steel magnets). This process is very effective in creating a uniform magnetic field around the working sample. This approach will directly pattern the films, in situ, without the need for additional plating steps.
By masking the plating carrier with a nonconducting film, plating can be prevented or inhibited in certain regions; the other regions (with no mask) are patterned in the desired designs. This is an important aspect that sidesteps several manufacturing constraints, such as patterning with laser ablation or etching (e.g., wet acid/dry plasma etching). It also can achieve layered metal-adhesive film composites without any extra process burden. Such patterning can only otherwise be achieved with screen-printing or micro-assembly of magnetic flake or particle composites, both of which constrain either the properties or the process cost. Embodiments of the subject invention can include in situ patterning.
The plated films can be released from the carrier substrate or plating carrier and transferred onto an interim carrier substrate. This is essentially based on the delamination of the films from the plating carrier and adhesion to the interim carrier. An important aspect for the success of this step is to engineer the adhesion strength between different layers. The plated film should have a poor adhesion to the plating carrier (e.g., 5 megapascals (MPa) or less). The plating stress can be controlled such that the delamination is easily accomplished. The plated films should strongly adhere to the interim carrier (e.g., with an adhesion strength of at least 30 MPa or at least 40 MPa). The adhesive films can serve two functions: provide strong adhesion to the plated film so that it can be pulled out of the plating carrier to help with the layering step on the interim carrier; and act as a strong adhesive and interlayer dielectric that electrically isolates the plated magnetic films in the final multilayered stacked magnetic core.
Referring to
Referring to
Embodiments can provide ultra-thin electrical components (e.g., inductors or transformers) with a permeability of, for example, 300-400 at a frequency of 10 megahertz (MHz). An inductance density of, for example 8 nH/mm2 or more can be achieved with, e.g., 5-10 milliohms (me) of resistance at a frequency of 1 MHz to 10 MHz. Current handling of at least 1 A/mm2 can be achieved.
Advanced composite magnetic structures with high permeability, low coercivity, high resistivity, and good frequency stability are needed to achieve the best inductor performance metrics. A key metric is L/Rdc (an inductance density of 8 nH/mm2 or more (e.g., 8-15 nH/mm2) with 10 mΩ direct current (DC) resistance at about 10 MHz). In addition, thickness scalability with innovative inductor topologies is important for handling adequate power at low cost. These attributes come at the expense of each other, and existing approaches face fundamental limitations either in terms of performance, scalability to high power, or cost. Ferrites have been widely used as the magnetic cores for low frequency applications such as transformers. Ferrites have low saturation magnetization and are not suitable for high-current handling and high-frequency applications (due to high-frequency instability above 3 MHz). The ceramic properties of ferrites provide high resistivity (>1 Ohms-meters (Ω-m)) resulting in low eddy current loss for thicker cores. However, the low field anisotropy (Hk) of ferrites limits them to low-frequency applications (about 1 MHz). Ferrites also have low saturation magnetization (about 0.5 T or less) and low permeability (μ), which lead to low inductance volumetric density and low current handling when used as the inductor cores.
Nanocrystalline magnetic ribbon alloys (e.g., Metglas® from Hitachi®) show good soft magnetic properties such as low coercivity, low core loss, and high permeability. The good soft magnetic properties are the results of absence of grain boundaries and crystal magnetic anisotropy in amorphous alloys. Another example is VITROPERM®, an iron-based nanocrystalline material with soft-magnetic properties of high saturation flux density (greater than or equal to 1.2 T), a permeability that can be adjusted in the range of from 400 to 800,000, excellent thermal stability over a wide temperature range, low core losses and low coercivity, and low or zero saturation magnetostriction. VITROPERM products are available as ribbon in thicknesses from 14 μm to 20 μm and widths from 2 millimeters (mm) to 66 mm. Because of their high eddy currents, they are more suitable for lower frequencies as transformer cores and common mode chokes. The magnetic fields of a toroid are directed within the plane of the core, and the eddy currents are orthogonal to the plane, so aligned magnetic flakes can be used as magnetic materials.
High permeability and softness (low coercivity) with frequency stability (high resistivity for low eddy currents and high ferromagnetic resonance (FMR)) requires nanostructured cobalt and iron alloys (less than 5 nanometers (nm)) with interspersed oxides (less than 5 nm) to enable exchange coupling and electrical isolation. Such “metastable” structures can be deposited with sputtering techniques, but these are unfortunately limited to lower film thickness and low-throughput processes (see also, e.g., Gardner et al., Integrated on-chip inductors using magnetic material (invited), Journal of Applied Physics, vol. 103, p. 07E927, 2008; which is hereby incorporated by reference herein in its entirety). Magnetic thin-film inductors can be integrated into integrated circuits (ICs), as depicted in
Embodiments provide plated magnetic composite films to achieve both high Ms, high permeability, and low coercivity. Nanomagnetic films can be deposited with precise control in composition and nanostructure by a continuous electroplating process. CoNiFe alloys (e.g., doped CoNiFe alloys) can achieve Ms of at least 2.0 T. Monolithic magnetic films will quickly run into losses from eddy currents, process issues from stresses, and delamination and other reliability challenges. A key innovation is to intersperse the magnetic films with thin insulating and/or adhesive layers (e.g., polymers) in an automated process to relieve both the eddy currents and stresses. The insulating layers can suppress the eddy currents and coupling between the layers that could otherwise increase the coercivity. Thin films (e.g., 2 μm or less) can be electroplated and then easily released from the carrier to form a layered composite structure by utilizing thin insulating adhesives. By interspersing with thin insulators, a permeability of 800-1000 with a coercivity of less than 1 Oersted (Oe) can be achieved with frequency stability of 100 MHz and beyond, and alternative current (AC) losses of less than 1%. The films can be deposited onto thick carriers with direct in situ patterning without the need for any subsequent lithographic steps to pattern them. That is, the fabrication method can explicitly exclude such subsequent lithographic steps. These micropatterned films can be transferred onto laminates as adhesive-isolated nanomagnetic layers. A high throughput batch process can be used for high throughput and low cost. Standard package processes can be utilized to form solenoid and/or spiral inductors after fabricating the multilayer magnetic cores. Solenoids can provide high current handling, and spiral inductors can provide low DC resistance.
A CoNiFe alloy can be used as the base magnetic material because of its high saturation magnetization (greater than 2.0 T). CoNiFe layers can be electroplated with a predetermined (e.g., designed) composition and nanostructure. Suitable additives can be used to help achieve nanostructure and lower the coercivity, while increasing the alloy resistivity. Alloying with additives allows retention of the amorphous structure with low magnetostriction. Example additives are silicon (Si), aluminum (Al), zirconium (Zr), tantalum (Ta), and hafnium (Hf), which unfortunately have reduction potential. Plating these additives is not possible under standard aqueous acid plating conditions. In this scenario, a soft magnetic material like CoNiFe is a very promising material for magnetic microsystems. Thus, embodiments can exclude any such additives, which may diminish the magnetic material's properties, and instead just use a soft magnetic material like CoNiFe.
The transitional term “comprising,” “comprises,” or “comprise” is inclusive or open-ended and does not exclude additional, unrecited elements or method steps. By contrast, the transitional phrase “consisting of” excludes any element, step, or ingredient not specified in the claim. The phrases “consisting” or “consists essentially of” indicate that the claim encompasses embodiments containing the specified materials or steps and those that do not materially affect the basic and novel characteristic(s) of the claim. Use of the term “comprising” contemplates other embodiments that “consist” or “consisting essentially of” the recited component(s).
When ranges are used herein, such as for dose ranges, combinations and subcombinations of ranges (e.g., subranges within the disclosed range), specific embodiments therein are intended to be explicitly included. When the term “about” is used herein, in conjunction with a numerical value, it is understood that the value can be in a range of 95% of the value to 105% of the value, i.e. the value can be +/−5% of the stated value. For example, “about 1 kg” means from 0.95 kg to 1.05 kg.
A greater understanding of the embodiments of the subject invention and of their many advantages may be had from the following examples, given by way of illustration. The following examples are illustrative of some of the methods, applications, embodiments, and variants of the present invention. They are, of course, not to be considered as limiting the invention. Numerous changes and modifications can be made with respect to embodiments of the invention.
A stable plating bath that can yield the desired films can be used for fabricating a multilayered inductor core. The concentration of metal ions in the bath determines the final alloy composition. It is critical to stabilize the bath to prevent or inhibit the oxidation of Fe′ to Fe′ and prevent or inhibit hydroxide precipitation and suppress the evolution of hydrogen at the cathode.
The current density can be optimized. The electroplating process was optimized such that when apply 35 milliamps per square centimeter (mA/cm2) was applied for about 4 minutes and 30 seconds, a film with a thickness of 2 μm (or about 2 μm) was achieved. The thickness of the film was confirmed using a profilometer.
Interlayered CoNiFe films and dielectric layers were fabricated to create a structure that can be placed around, above, and below the inductors.
Before going to the next step, the sample was placed in the electroplating bath and connected to the DC supply as shown in
In the third step, the predetermined desired layer was electroplated on the uncovered region. Then, the covered regions with non-conductive film were removed in the fourth step. In the fifth step, the use of a temporary carrier and adhesive was implemented to delaminate the plated films from the Ti. Ideally the temporary carrier should be as thin as possible. The plated films have a poor adhesion to the Ti, and so it is easier to delaminate and transfer onto a temporary carrier. After the sixth step of delamination, the process goes in for the sequence layering. The key in this last step is to engineer the adhesion strength between different layers. This is achieved by using an encapsulating gel (e.g., DOWSIL® encapsulating gel), which can be diluted (e.g., diluted in a ratio of 3 to 1 with Xylene).
The adhesive serves two functions, first providing strong adhesion to the electroplated CoNiFe film so that it can be easily delaminate out of the Ti and second acting as an interlayer dielectric that isolates the plated magnetic films in the final multilayered core. The adhesive is durable, fast-curing, requires no primer, and has good mechanical strength. The curing time was just 2-3 min at 100° C. In order to spread the adhesive properly, a spin coating technique or dip-coating was used. The thickness control of the adhesive was also optimized in the spin coating. A uniform spread of almost 1 μm thickness of adhesive was obtained between the two layers of film. After curing, the adhesion strength was adequate for subsequent film-transfer steps. This is how layer after layer was built using adhesive, and any number of layers can be built that is desired.
The permeability measurements were performed with a BH looper, with the sample oriented in the parallel and transverse directions. Various inductor test structures were fabricated and tested, with the results shown in the tables in
Ansys is an engineering simulation and 3D design software that was used to model various inductor topologies and magnetic properties of the composites. A 3D model for each type of inductor topology was designed keeping the area constraints within 15 mm2 (i.e., the total footprint was 15 mm2 or less). Two topologies were analyzed—spiral and solenoids.
First, a 1-turn spiral inductor with magnetic material above and below was considered. Because of the required low DC resistance, a spiral inductor with only one copper winding turn was designed, as shown in
The second structure considered included the same spiral inductor but this time by assuming a perfect wrapping of the copper trace with a magnetic composite. Initial models and simulation results, shown in
For this design, copper traces were built using the design tool in Ansys to create volumetric designs. A magnetic material was wrapped around the copper traces with the appropriate thickness and airgap to simulate a closed loop magnetic flux.
The next step was the application of material models and boundary conditions. The excitations are required to be defined and this determined the input current at which the simulation will run. A key step is to define the input currents by selecting the faces of the object into which the current will flow. ANSYS allows input of three different types of relative permeability depending on the application. There are three options available: simple permeability; anisotropic; and nonlinear. Nonlinear panel allows uploading of the BH loop of the material into the system and running of the simulation with it. However, the nonlinear option assumes that the BH loop is the same for each direction of the structure. An anisotropic permeability was worked with by uploading the BH curve in the direction in which the magnetic material was placed. For example, each copper trace from the spiral design is being covered in-plane and out-of-plane direction with magnetic material. Thus, T(1,1), T(2,2), T(3,3) in
Solenoid designs were considered in this paper because of their high current-handling capacity (see also, e.g., Mathuna et al., Power Supply on Chip for High Frequency Integrated Voltage Regulation, in IEEE APEC Charlotte, N.C., 2015; which is hereby incorporated herein by reference in its entirety). Various solenoid designs were modeled and designed to achieve high-current handling and high inductance density. The aforementioned high-permeability metal-polymer composites were used as the cores for all inductor designs. The 3D view of a designed solenoid inductor is shown in
High inductance-density inductors with magnetic cores tend to be saturated at low current. Therefore, current handling capacity become another critical parameter to judge the performance of power inductors. Saturation of the magnetic cores can be avoided by innovative inductor topologies. Different topologies have their own pros and cons. For example, solenoid inductors have low L/Itbc but high-current handling, whereas spiral inductors have high L/Itric but low-current handling. Because the topologies of inductors play an important role in determining the performance of inductors, inductor topologies can be considered as follows. Based on how the magnetic cores and copper windings are integrated, the topology can be: a) CMC (copper-magnetic-copper) inductors include magnetic cores enclosed within the copper windings (e.g., solenoid inductors); or b) MCM (magnetic-copper-magnetic) inductors include two magnetic layers sandwiching copper windings therebetween (e.g., spiral inductors). These are also shown in
Solenoids designs can be sub-categorized into: solenoids with open flux path; and solenoids with closed flux path. This is depicted in
The energy stored in power inductors as the magnetic field is given by the following equation.
where L is the inductance of the inductor and I is the current through the inductor
Equation (1) plainly shows that inductors with high inductance can store more energy. Hence, a goal is to increase the inductance of the inductor for high energy electromagnetics devices. The self-inductance of an inductor depends upon the characteristics of its construction. Optimizing these factors is crucial for the success of increasing the inductance. At first, these factors can be deduced from the fundamental equations of electromagnetism. Self-inductance of a coil is defined as the magnetic flux linkage (Nφ) divided by the current.
where φ=magnetic flux and N=number of turns. Further, magnetic flux produced in a coil is given by:
φ=BA (3)
where B=flux density and A=area. Also, the magnetic flux density within the core of a long solenoid is given by:
B=nμI (4)
where, n=number of turns per unit length and μ=permeability of the core material. Using the above formula, an approximation of inductance for any coil can be calculated.
Factors like size and length are already optimized in the electromagnetics devices and so there is very little room available here for further improvement of the inductance. Therefore, the possibility of attaining very high coefficients of self-induction reduces to just two factors: permeability of the core; and number of turns. Inductors that are designed to have a large number of copper turns result in high DC resistance and DC loss, which leads to low efficiency. Therefore, inductance per DC resistance (L/RDC) becomes an important factor in describing the efficiency of inductors. In order to increase the L/RDC value, high-permeability magnetic materials can be used as the cores of inductors (enhancing the inductance).
Without suitable designs, inductors cannot utilize the benefits brought by the magnetic materials with superior properties. Each inductor topology has its own advantages and disadvantages. Therefore, suitable inductor designs are needed to achieve high inductance, high current handling, and low DC resistance.
Another design was a four-side structure as seen in
The next step is to implement the magnetic composite. Small pieces of magnetic core were cut and placed manually around the one-turn copper trace. However, one of the main challenges remained that in simulations the inductance density requirement was achieved assuming this is a perfect closed-loop magnetic flux structure. Also, it is not as cost effective to place the side cores if it actually was manufactured on a large scale. During testing, the inductance density was not met mainly due to small separations between the core pieces, which created magnetic flux leakage.
In order to build solenoids, an 18 μm copper tape was used that could be patterned using a laser. Thin strips (0.25 mm) were cut and wrapped around the two types of magnetic composites. One of the main challenges is to meet the DC resistance, which can be optimized with via filling or photolithography for the copper windings. In addition to this, a new magnetic structure was required to surpass the inductance density and meet current handling requirements. This was performed in hand with the implementation of a magnetic paste.
The purpose of this mixture was to act as a magnetic filler to eliminate airgaps and close the magnetic loop, seal magnetic layers, and reduce power losses at high frequencies. The cross-sectional area of the flakes can reduce eddy current losses. Results on inductance performance over frequency showed enhancement in inductance. In
Inductance as a function of frequency was characterized for the inductors using a vector network analyzer (VNA). The characterization set-up is shown in
In order to measure the accurate inductance of the DUT, the inductance of the two connection wires also need to be measured by the VNA with de-embedding structures. These de-embedding structures are necessary for accurate measurements. De-embedding structures have three parts, known as open, short, and load. The function of the three de-embedding structures is to remove the effect of the parasitic from the two connecting wires and allow the VNA to solely measure the inductance from inductor.
However, when directly connected to the SMAs, the calibration process is performed with probes.
With the de-embedding structures, the inductance for inductors can be calculated after subtracting the inductance from connection wires. Without using the de-embedding structures, the measured inductance is sum of the inductance from connection wires and the inductance from inductors (Linductor=Lload−Lshort). Using this formula, Linductor is calculated for all the different structures (structure 1, structure 2, structure 3, and so on . . . ). For the comparison between different structures, the ‘X’ factor is calculated that represent how many more times the inductance is higher than its baseline (Linductor(Air core)), which is structure 1 for the solenoid topology.
where X21=number of times inductance of structure ‘2’ is higher than its baseline (Air core, structure ‘1’).
where X31=number of times inductance of structure ‘3’ is higher than its baseline (Air core, structure ‘1’).
Four different design structures were used for spirals. The use of magnetic paste was implemented in two of the spiral structures to enhance the permeability of the core. Further information can be seen in
Toroid inductors have the lowest reluctance because of the closed magnetic loop. With a few turns, enough inductance can be created with this low reluctance. However, the flux can easily saturate the magnetic core. In order to inhibit or prevent the saturation, airgaps can be introduced. In equations TR1-TR5 below, isat is saturation current; Nis number of turns; R is the reluctance, Rg is the reluctance from airgap, Rm is the reluctance from the magnetic core, and a is defined as effective reluctance enhancement factor with air-gap. Even though the reluctance is defined as Amp/(Henry×Amp) or (1/Henry), it is represented here as (α×length parameter/permeability of air). For simplicity, effective reluctance or simply reluctance is considered as just the length parameter in microns (μm) as the dimensionless permeability is considered. The reluctance can be tuned from 5 μm to hundreds of μm to achieve the target current-handling and inductance density. The saturation current is calculated as
For a toroid of core length of about 9 mm of permeability 300 and airgap of 100 the reluctance is 80 microns from Equation TR2.
After the reluctance is known, the inductance can be estimated from Equation (TR6). The inductance is estimated to be 100 nH for 3 turns, when the cross-sectional area is 2000 microns×500 microns, as shown in
Inductance enhancement (X) factor for different structures of inductors were calculated and are summarized in the table in
It should be understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application.
All patents, patent applications, provisional applications, and publications referred to or cited herein are incorporated by reference in their entirety, including all figures and tables, to the extent they are not inconsistent with the explicit teachings of this specification.
Number | Name | Date | Kind |
---|---|---|---|
4990225 | Omata | Feb 1991 | A |
20070015349 | Aublanc | Jan 2007 | A1 |
20160276269 | Peng | Sep 2016 | A1 |