Patent Application
20200274234
Embodiments of the present invention relate generally to structures and methods for forming electronic structures and antenna coupled terahertz films that comprise such structures in general, and structures for harvesting energy from electromagnetic radiation. More specifically, embodiments relate to nanostructures, metamaterials, Near Field Quantum Rectifiers (NFQ Rectifiers) or, alternately, rectennas and related methods and systems for harvesting energy from, for example, infrared, near infrared and visible spectrums and capturing millimeter waves and Terahertz energy, and to films comprising such structures.
There is an immediate and great need for inexpensive renewable energy in the world right now. Ironically, there is an abundance of energy available in the form of sunlight and heat but using it to support the needs of society requires it to be converted into electrical form.
Low temperature waste heat is abundant and generally. Generally, such low temperature waste heat is found in large volume form, for instance, flue gas stacks or heated waste water. Harvesting volumes of gas or fluid requires large surface area contact with films created for this purpose. Harvesting sources of heat into usable electrical power is especially desirable at low cost. Low cost manufacturing techniques are then of importance to the proliferation of waste heat harvesting electronic films and systems.
Embodiments are directed to a system and method for making electronic components in general and NFQ Rectifiers, in particular, Embodiments use nanoimprint lithography and roll-to-roll (R2R) technology and films, such as antenna coupled terahertz films, that comprise such electronic and NFQ Rectifier structures. The technology of surfaces of paired nanoantenna and diode arrays present tremendous advantages for energy harvesting applications. In the area of waste heat recovery these systems are ideal since they can be tuned to the frequency spectra of the target source, have no moving parts, and are inexpensive to manufacture.
Embodiments described herein involve a method for fabricating electronic structures on films using nanoimprint lithography (NIL) and roll-to-roll (R2R). Developing NIL and R2R processes is expensive and time consuming. Reductions in the complexity or number of steps in a process translate to significant process development cost savings as well as reduced manufacturing cost. One such reduction, as described herein, involves an etched undercut of a key structural element in a multilevel stack.
If not made monolithically, due to the nano-scale size of the devices being manufactured, alignment is a key issue in NIL and R2R processes. Alignment of multiple structures can be assured using self aligned imprint lithography (SAIL). In the SAIL process, all device elements are assembled together in an imprint tool. A liquid polymer or monomer is applied to a substrate and the tool is pressed into the liquid. Rollers or other mechanical tools bring the substrate, liquid and imprint tool together. The liquid is cured with UV or heat and the imprint tool is separated from the cured polymer/monomer (hereinafter “polymer”).
Optimally, the NIL and R2R process of creating NFQ Rectifiers is a subtractive process. While it is possible to add layers or materials, in general the process is simplest if it is purely subtractive. In one embodiment of the invention, a substrate is coated with all the materials required to arrive at the finished component. This coated substrate is called a feedstock or a feedstock stack. For example, in making an NFQ Rectifier, a metal, at least one thin oxide, and a top metal are deposited on a substrate to create the feedstock. In an embodiment, the feedstock substrate is a substrate that can be used in roll-to-roll processes.
In an embodiment, the imprint polymer is deposited on the surface and etching of the imprinted structures and exposed feedstock layers proceeds step by step. Differential etches make it possible to selectively etch the polymer structure, metal, or oxide of the feedstock. In one embodiment, an NFQ Rectifier structure contains two metal layers separated by at least one oxide layer. The bottom metal is etched to form a left antenna leaf. The top metal is etched to form a right antenna leaf. An overlap area in the middle of the device forms a diode. In this embodiment, the simple subtraction of layers does not separate the right antenna from the lower metal below and, thus creates a short to the diode.
Undercutting the lower metal disconnects the lower metal to solve this shorting problem. The undercutting saves numerous other steps and simplifies the process significantly. In an embodiment, the undercut is performed with a wet etch of the metal and is enabled by placement of an impression, also referred to as a depression, structure in the imprint tool at the point where the undercut is desired.
Important elements for this undercut process to work are: an impression structure in the imprint tool at the region of undercut; an undercut region of narrower width than surrounding structures; and a wet etch or other isotropic etch capable of selectively removing material from beneath other permanent layers. In one embodiment of this approach, a wet etchant is used whose lateral etch rate is a function of the etchant temperature. In this manner, the lateral etch rate for the undercut is controlled by setting and maintaining a prescribed etchant temperature.
The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements. Various modifications to the described embodiments will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. Thus, the present invention is not intended to be limited to the embodiments shown, including any limitations by any dimensions included in the figures, but is to be accorded the widest scope consistent with the principles and features described herein. The drawings are not drawn to scale. Any dimensions included in the figures are merely as exemplary an exemplary embodiment for the feature they apply to, and are not intended to be limiting, to indicate scale, or to be considered relative to any other feature in the drawings.
An Antenna Coupled Terahertz film (“ACT film”) is manufactured using roll-to-roll manufacturing built around nanoimprint lithography. The ACT film comprises two subassemblies: (1) a rectenna or NFQ rectifier film and (2) a metamaterial film. In an embodiment, the metamaterial is tuned to the resonating frequency of the antenna of the rectenna. In an embodiment, a NFQ Rectifier film comprises a roll-to-roll film substrate upon which a plurality of NFQ Rectifiers is manufactured. In an embodiment, a metamaterial film comprises a roll-to-roll substrate upon which a plurality of metamaterials is manufactured. To complete manufacturing of the ACT film, the rectenna and metamaterial films are aligned to ensure the rectenna are locate over the holes in the metamaterial and then bonded together.
The metamaterial (described below) comprising the metamaterial film is tuned to the frequencies expected for energy harvesting. In this case, the metamaterial is tuned to frequencies in the Terahertz (THz) range associate with heat. More details concerning the rectenna and metamaterial can be found in U.S. patent application Ser. No. 14/745,299, filed Jun. 19, 2015, entitled, “System and Method for Converting Electromagnetic Radiation to Electrical Energy Using Metamaterials,” (the “'299 patent application”) and U.S. patent application Ser. No. 15/602,051, filed Sep. 14, 2017, entitled “Structures, System and Method for Converting Electromagnetic Radiation to Electrical Energy Using Metamaterials, Rectennas and Compensation Structures,” (the “'051 patent application”) both of which are hereby incorporated by reference herein in their entireties.
In an embodiment, manufacture of an ACT film incorporates a number of process steps as summarized below in Table 1 and in the process flow chart of
To begin manufacture of an ACT film according to an embodiment, a feedstock is created.
As shown in
(a) 1 um SiO2 isolation layer (for conductive substrate only)
(b) 150 nm Al
(c) 720 nm Si02 isolation layer
(d) 150 nm Al
(e) 30 nm Ni
(f) 2.5 nm NiOx
(g) 1 nm Al203
(h) 10 nm Cr
(i) 150 nm Al
In the description that follows, as shown for example in
Feedstocks 103 and 103a can be created in several passes. Referring to
In an embodiment, the thin layers of the feedstock are amenable to sputter deposition without discontinuities. Creation of the feedstock is referred to as step 1 in the process flow chart of
There are a variety of substrate choices for ACT film processes. The best substrate will have good dimensional stability, smooth surface, low price and be heat tolerant. Table 2 lists candidate substrate materials in current order of desirability, along with pros and cons of each:
After creation of the feedstock, such as feedstock 103 or 103a, a nanoimprint tool 101 is created by coating the feedstock with a uniform thickness of UV curable photopolymer. This can be accomplished with a reverse gravure coater or the like. In an embodiment, the coated feedstock is then imprinted using a patterned quartz roller, and UV light from within the quartz roller sets the polymer while it is in the nip between the imprint and backing roller. The imprinted feedstock then goes through a thermal or second UV curing stage to complete the cross-linking of the polymer.
In an embodiment, the final photopolymer is compatible with the wet metal etchants used in subsequent process steps. It must also be etched in an oxygen plasma and removable at the end of the process steps with an ashing or similar process. In an embodiment, the imprint step or layer height is about 0.5 microns. In an embodiment, the gravure coating is about 1 micron. Creation of nanoimprint tool 101 is referred to as step 2 in the process flow chart of
Referring to
Referring to
Antenna leaf 1302a is made of a first metal 109, M1, and antenna leaf 1302b is made from a second metal 107, M2. Diode 1304 comprises metals M1 and M2 and one or more diode oxides 108. In an embodiment, metals M1 and M2 are multilayered. For instance, metal M1 is Aluminum with a thin layer of Nickel and metal M2 is a thin layer of Chrome with Aluminum on top. Such variations are important since the requirements of a metal used in an antenna differ from that of a diode. In an embodiment, metals M1 and M2 are the same metal.
Antenna metals typically need to be highly conductive at high frequency. Thus, in an embodiment, aluminum is the primary conduction metal in both metals M1 and M2. Metals in a metal-insulator-metal (MIM) diode or metal-insulator-insulator-metal (MIIM) diode are selected for their different work functions and how they establish barriers with oxides for desired tunneling and antisymmetric diode behavior. In an embodiment, diode 1304 includes one or more diode oxides 108, for example, NiO and Al2O3. For clarity in this example, the full stack of the feedback stock Al—Ni—NiO—Al2O3-Cr—Al. A reflector metal and isolation region can also be added as shown in
Referring to
As shown in
In an embodiment, nanoimprint tool 101 is made from a polymer that can be selectively removed by etching. As described herein, manufacturing a rectenna 1300 involves alternative etches of specific features in a self-aligned imprint template.
Depression, or alternately, impression, area 102 in imprint tool 101 allows wet chemical etch access to the area near area 110, which corresponds to diode 1304, where undercut 800 (see
Referring to
Time depends on imprint process. In an embodiment, the etch time and process parameters are determined, and are influenced, by the specific polymer choice and, for this step, the thickness of the gravure coating selected in step 2. In an embodiment, the photopolymer etch time is 10 seconds or approximately 10 seconds. The de-scum etch is referred to as step 3 in the process flow chart of
Next, as illustrated in
For example, in an embodiment with a feedstock 103a, this etch removes 150 nm of top metal aluminum. (See, e.g., Al layer (i) of feedstock 103a in
In addition, in an embodiment with a feedstock 103a, an etch is performed to remove the top Cr metal (See, e.g., Cr layer (h) of feedstock 103a in
Next, in an embodiment, in a step of passivation as illustrated in
In an embodiment, in the passivation step, a CVD layer is deposited as shown. It will cover all exposed surfaces, both vertical and horizontal. In an embodiment, the thickness and composition of the CVD layer is SiN at a thickness of 0.25 um. It is helpful that there be some visual contrast to monitor subsequent process steps. This passivation step is referred to as step 6 in the process flow chart of
Next, as shown in
In embodiment using feedstock 103a, the Ni interface metal (See, e.g., layer (e) of feedstock 103a in
Next, as shown in
The lateral wet etch that determines the undercut is carefully controlled by, for example, controlling the etchant temperature. Wet etchants of Al, for example, have a lateral etch rate that is highly temperature dependent. At temperatures above about 55° C. the lateral etch rate can be as large or larger than the vertical etch rate. Careful control of the temperature in this case is key to control of the undercut. The exact etch time is determined by the thickness of the metal, the length of undercut required and the temperature of the etchant. This undercut serves not only to isolate the metal M1 and metal M2 layers but also defines the area of the active diode device 110, which corresponds to diode 1304 of rectenna 1300. Defining the dimensions of diode 1304 with undercut methods is very important since it is often advantageous to create a small diode 1304 structure. The undercut method described makes it possible to exceed the small scale limit of the imprint technology used to create even smaller structures.
Wet etch of metal 107, M1, Al in the example of feedstock 103a (See, e.g., layer (d) of feedstock 103a in
This undercut etch is referred to as step 9 in the process flow chart of
Next, as shown in
Next, as illustrated in
The step of removing the imprint layer is referred to as step 10 in the process flow chart of
Next, an etch is done through feedstock 103.
In an embodiment that uses feedstock 103a of
An RIE SF6 etch is used to etch NiO and Al2O3 layers (diode layers (f) and (g) of feedstock 103a in
The Ni layer (layer (e) of feedstock 103a in
Next, Al layer (layer (i) of feedstock 103a in
Next, as shown in
Next, as illustrated in
In an embodiment using feedstock 103a, the Cr layer is also etched, which is essentially a repeat of step 5 for layer (h) of feedstock 103a in
Next, as illustrated in
Removal of the remaining portion of nanoimprint tool 101 is referred to as step 19 in the process flow chart of
In an embodiment, the metamaterial material to be made comprises a series of patterned or unpatterned holes or posts on its surface. In the embodiment illustrated, the metamaterial comprises copper that has a patterned (periodic) series of holes on its surface. The metamaterial film subassembly comprises a plurality of such metamaterials. For manufacture of the metamaterial, the second subassembly, the following major process steps are performed:
In an embodiment, polymer or monomer 1504 is UV curable and capable of forming and curing the geometry shown in
In an embodiment, imprint pattern 1508 is characterized by a periodic placement of structures, such as structures 1508a and 1508b. In the embodiment illustrated in
In alternative embodiments, the structures do not have to have periodic displacement, have the same size, or the same shape. Instead, the structures must have a shape and placement with respect to one another such that the resulting metamaterial will be resonant at the frequency to which the rectenna to be used is tuned.
Once the plating template is completed, in step 3002, a seed metal 1602 is deposited to provide an electrically continuous path for plating to be carried out. This metal should be stable in the plating electrolyte and sufficiently conductive so that plating occurs homogeneously across the roll. In an embodiment, the seed layer material is copper that is sputtered at a rate of 2 nm/s to a thickness of 100 nm. All sections of the roll must be exposed to the copper evaporation for 50 seconds assuming this deposition rate.
In step 3003, the material comprising the metamaterial is plated with the metamaterial material to fully encapsulate the plating template structures. In an embodiment, the metamaterial is copper. In such an embodiment, where copper is the metamaterial, in step 3003, After the seed layer has been deposited, plating of the copper is carried out to fully encapsulate the plating template structures. In an embodiment, electrical continuity needs to be made to the seed layer in order to drive a current density of 20 mA/cm2. This electrical continuity can be made by direct contact to the top side of the structure towards the edges of the roll. The plating electrolyte comprises primarily copper (II) sulfate pentahydrate and sulfuric acid. The solution is very acidic with a pH of −0.25. The target plating thickness should extend past the structures. The current target thickness for wafer production of metamaterial is greater than 2× the template structure height for a total of 6 μm. With a plating deposition rate of approximately 7 nm/s the total deposition time of a single area of the roll is approximately 15 minutes. This time could be potentially decreased by a combination of increasing current density or setting the target thickness to a lower value. It is only important that the template is fully encapsulated. Therefore, the target thickness could be cut nearly in half if the plating across the roll is very homogeneous.
In step 3004, a substrate 1802 is added to the metamaterial being formed. In an embodiment, patterned side of the copper metamaterial is made face down. Making the patterned side of the copper metamaterial facedown reduces the precision tolerance of the electroplating depth and ensure a flat metamaterial surface. The metamaterial is required to have its openings facing the rectennas (when ultimately aligned and bonded to the rectenna, such as rectenna 1304). The plated structure in
In an embodiment, the material for the thermally conductive substrate 1802 can be chosen from any common metal. For instance, any of a variety of metals commonly used as roll-to-roll substrates are acceptable. In an embodiment, the bonding method for substrate 1802 uses bonds that are tolerant of temperatures up to 300° C.
After substrate 1802 is attached to the top of the plated copper, the film is submerged in a solvent that will remove imprint polymer 1504 in step 3005. Removal of imprint polymer 1504 also results in detachment of temporary substrate 1506. In an embodiment, the solvent quickly and cleanly removes the polymer/monomer. All materials in the final metamaterial structure need to be thermally conductive and low outgassing for vacuum compatibility. As shown in
In step 3006, standoff structures are built onto surface 1904 of metamaterial 1900. Before aligning and bonding the metamaterial and rectenna films, standoff pillars are built on the surface of the metamaterial to set the separation distance between the two films.
In an embodiment, there can be more than or fewer than 4 standoff structures of any shape and placement to provide the required distance between the rectenna and metamaterial films that will be aligned and bonded together.
Although manufacture of only a single rectenna and associated metamaterial device have been described, using the roll-to-roll manufacturing disclosed herein, numerous such devices will be able to be manufactured at a time using the roll-to-roll processes described above. In operation, as described in the '299 patent application and '051 patent application, in the presence of heat, electric fields are generated over the holes in the metamaterial. Rectennas that are tuned to the frequencies of those electric fields will generate an electric current when placed over them. Consequently, proper alignment is required to ensure that the rectenna film places the rectennas over the holes in the metamaterials when the rectenna film and metamaterial film are bonded to one another.
Manufacture of ACT film involves creating two subassembly films, a rectenna film comprising a plurality of rectenna 1304, and a metamaterial film comprising plurality of metamaterials 1900, as described above, aligning them, and bonding them together. The alignment of structures on these surfaces is critical to operation and must be performed to +/−250 nm precision.
There are several alignment techniques available in industry and literature. Those techniques mainly fall into three categories. In geometric imaging, two geometric marks are compared through and optical microscope. This technique is limited to the optical diffraction limit of the objective lens. However, the image resolution does not meet the alignment specification noted above. In an intensity-based detection method, the critical intensity values of diffracted beams from alignment grating marks are measured. This method can be sensitive to the alignment of the light source and the detection sensors with respect to the alignment marks. Calibrating the laser source and detection sensors position relative to the films is quite complex especially in a dynamic system. A third method, is a phase-based detection method, in which the phase of a beat signal from two diffracted beams of slightly different periods is measured. Misalignment is captured by imaging the diffracted field into a microscope-type system. Generally, the nano-scale shifts at the mask-wafer level maps to a large-scale diffraction variations that are easy to detect and process by a high resolution optical system. Depending on the type of source used and the nature of the misalignment to be captured, a diffraction scheme is selected.
Embodiments of the present invention perform alignment as a roll-to-roll process using a Moiré technique discussed below. This is an optical method involving the use of alignment marks on each film layer. At least one film layer would need to be transparent for this method to be employable. Optical CCD sensors and computer driven controlled stepper motors close the feedback loop to affect continuous alignment. As shown in
Moiré fringes are large scale interference patterns that can be produced when an opaque lined pattern with transparent gaps is overlaid on another similar pattern. For the Moiré interference pattern to appear, the two patterns are not completely identical, they must be displaced, rotated, have different, but close pitch. Moiré fringes can be detected by an optical system and a CCD. Using computer assisted codes, one can predict misalignment in the nanoscale from sub-micron scale (optical) images.
In an embodiment, alignment is performed using a combination of geometric imaging and phase-based detection described above with a Moiré pattern detection to achieve a sub-200 nm alignment between the metamaterial film and the rectenna film. In an embodiment, an alignment mark comprises four (4) sets of gratings with alternating pitches Λ1 and Λ2.
A broadband light source, such as LEDs, images identical Moiré patterns of the complimentary sets on a CCD sensor. A misalignment between the films results in a magnified phase shift between the Moiré patterns of the complimentary sets. Let the lateral grating misalignment be defined in the x and y directions as Δx and Δy, and the corresponding phase-shift in the observed Moiré patterns defined in the x and y directions as ΔX and ΔY. Then the magnification factor M, that is ΔX/Δx, or alternatively ΔY/Δy, is inversely proportional to the difference between Λ1 and Λ2 as follows,
where ΔΛ is the difference between Λ1 and Λ2. A 5% relative difference between Λ1 and Λ2 is equivalent to a magnification factor M of 20×. A substantial benefit of this approach is the detected ΔX and ΔY is independent of the relative position of the light source and the optics to the alignment marks. One important caveat is that that Moiré patterns are identical after full period shift Λ1. Hence, there is a need for an initial coarse alignment using geometric imaging methods. The initial coarse alignment is applied to the inner cross marks “+” 2210 and 2210 shown in
After coarse alignment, a CCD captures Moiré the ΔX shift in the fringes as illustrated for example, Moiré fringes 2402 associated with coarse alignment 2300 in
After fine alignment in the x-direction, the CCD captures the ΔY shift in the Moiré fringes as illustrated for example, Moiré fringes 2502 associated with a y-direction grating in the fine x-direction alignment 2501 in
An alternate method of alignment involves use of the power output signature of the rectenna film as the metamaterial and rectenna films are placed in casing structures. Power bus output is passed through an A/D converter and delivered to a computer workstation running a power and positioning optimization algorithm. Control of the relationship between the metamaterial and rectenna films is made through a motorized linear stage by the optimization software. A few degrees temperature difference is required between the two films to generate a power output signal. Initial alignment in panel case will be sufficient to begin optimization hunting algorithm. A boustrophedonic search in 1 um steps will position the plates within proximity within at most 144 steps and on average 72 steps. Fine positioning can proceed with a simple “greedy” step and adjust or other similar algorithm to bring final alignment. Once alignment is complete, edged are bonded and panel is sealed.
This application claims the benefit of the filing date of U.S. Provisional Application No. 62/808,275, filed Feb. 20, 2019, U.S. Provisional Application No. 62/816,907, filed Mar. 11, 2019, and U.S. Provisional Application No. 62/817,489, filed Mar. 12, 2019, all of which are hereby incorporated by reference herein in their entireties.
| Number | Date | Country | |
|---|---|---|---|
| 62808275 | Feb 2019 | US | |
| 62816907 | Mar 2019 | US | |
| 62817489 | Mar 2019 | US |
