Remote phosphor configuration for high luminance applications

Abstract

A micro-light-emitting diode (LED) device having a base substrate supporting one or more laser sources. The device further includes an intermediate substrate disposed on the base substrate with a curved concave region and a hole disposed through the intermediate substrate to accept the laser. A reflective layer is disposed on the curved concave region to match a shape of the curved concave region. The device further includes a top substrate disposed on the intermediate substrate, optically in-line with the intermediate substrate. The top substrate includes a micro-LED chip disposed optically in-line with the hole to convert the laser into light. The top substrate also includes a transmissive region such that the micro-LED chip converts the laser into the light and reflects the light to the reflective layer, which reflects the light through the transmissive region.

Description

Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 REMOTE PHOSPHOR CONFIGURATION FOR HIGH LUMINANCE APPLICATIONS CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of U.S. Provisional Application No.63 / 641,342 filed May 1, 2024, the specification of which is incorporated herein in its entirety by reference. FIELD OF THE INVENTION

[0002] The present invention is directed to laser-driven headlamps for automobiles with increased illumination and efficiency. BACKGROUND OF THE INVENTION

[0003] In 2010, leading car manufacturers including Mercedes-Benz®, Audi®, and Opel presented adaptive driving beam (ADB) to the United Nations Economic Commission for Europe's Working Party on Lighting and Light Signaling (GRE) in Geneva. ADB is an innovative headlamp design that aims to enhance nighttime driving safety by uniformly illuminating essential areas for the driver while deflecting beam patterns away from oncoming motorists. Shortly after this meeting, ADB was swiftly adopted and integrated into the Economic Commission of Europe (ECE) regulations. In response to ADB's success in Europe, Toyota®initiated lobbying efforts in 2013 with the U.S. government, petitioning the National Highway Traffic Safety Administration (NHTSA) to approve ADB systems for the U.S. market. By 2018, NHTSA officially announced its consideration for approving the technology. However, the momentum stalled as no new regulations were implemented to allow ADB for sale in the U.S.

[0004] The Infrastructure Bill that passed in 2022 compelled lawmakers to establish regulations concerning ADB and on February 15th, 2022, NHTSA announced that they, “issued a final rule allowing automakers to install ADB headlights to new vehicles.” With approval finally completed, numerous headlight technology advancements including laser-driven (LD) light sources are anticipated to enter the U.S. market to supply the increase in luminous demand. The prolonged lobbying to permit ADB has led to significant investment in optical research and development within the automotive industry, OEMs seek to integrate traditional lamps with complex illumination, imaging, and sensor systems.

[0005] Due to an increased emphasis on energy efficiency, light-emitting diode (LED) researchers are actively searching for methods to increase the generation of white light per unit watt. This is important for electric vehicles looking to reduce their energy budget, as a result,Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 active studies are pointing towards utilizing lasers for headlamps as they are much brighter and cost about two-thirds the amount of energy when compared to traditional sources.

[0006] A traditional LED as shown in the schematic of FIG.1A, has a phosphor layer overlaying a gallium nitride (GaN) alloy wherein the alloy generates blue light through electroluminescence. When a current is applied to the semiconductor GaN alloy a recombination of free holes and electrons occurs, and free electrons move from the n-type region to the p-type region and vice versa. The area in which the recombination of the free holes and electrons occurs is called the active region (i.e., depletion region), and is where photons are emitted. The location at which the n-type material and the p-type material meet is considered the p-n junction, this is where the term junction temperature derives its name. A remote phosphor simply implies that the phosphor layer is a distance ‘d’ away from the active layer of the LED. One immediate benefit of having a remote phosphor is that the phosphor layer absorbs less heat generated by the p-n junction and therefore can convert more of the excitation radiation without additional thermal oversaturation from the active region. In traditional LEDs, the active layer generates blue light in a Lambertian curve, for most remote applications a laser diode replaces a Lambertian emitter, and the excitation beam is assumed to be a concentrated Gaussian. Thus, there exists a present need for laser-driven headlamps for automobiles implementing phosphor illumination for white light generation. BRIEF SUMMARY OF THE INVENTION

[0007] It is an objective of the present invention to provide devices, systems, and methods that allow for laser-driven headlamps for automobiles implementing phosphor illumination for white light generation, as specified in the independent claims. Embodiments of the invention are given in the dependent claims. Embodiments of the present invention can be freely combined with each other if they are not mutually exclusive.

[0008] The present invention features a micro-light-emitting diode (LED) device. The device may comprise a base substrate comprising one or more laser sources configured to generate a laser. The device may further comprise an intermediate substrate disposed on the base substrate. The intermediate substrate may comprise a curved concave region comprising a hole disposed through the intermediate substrate such that the laser is directed through the hole. The intermediate substrate may further comprise a reflective layer disposed on the curved concave region such that the reflective layer matches a shape of the curved concave region.Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025

[0009] The device may further comprise a top substrate disposed on the intermediate substrate, optically in-line with the intermediate substrate. The top substrate may comprise a micro-LED chip disposed optically in-line with the hole, the micro-LED chip comprising an alloy configured to convert the laser into light. The top substrate may further comprise a transmissive region disposed optically in-line with the reflective layer. The micro-LED chip may be configured to convert the laser into the light and reflect the light to the reflective layer. The reflective layer may be configured to reflect the light towards the top substrate such that the light is transmitted through the transmissive region.

[0010] One of the unique and inventive technical features of the present invention is the implementation of a stackable inverse Cassegrain model into a remote phosphor micro-LED device. Without wishing to limit the invention to any theory or mechanism, it is believed that the technical feature of the present invention advantageously provides for optimized illumination and increased output of the LED device. None of the presently known prior references or work has the unique inventive technical feature of the present invention.

[0011] Furthermore, the inventive technical feature of the present invention is counterintuitive. The reason that it is counterintuitive is because it contributed to a surprising result. One of ordinary skill in the art is aware of the standard Cassegrain model and its uses for focusing light in a Cassegrain telescope. The present invention implements an inversion of the standard Cassegrain model for illumination purposes that has not yet been devised for this purpose. Surprisingly, the inverse Cassegrain model allows for efficient and optimized illumination with increased output when compared to prior illumination devices. Thus, the inventive technical feature of the present invention contributed to a surprising result.

[0012] Any feature or combination of features described herein are included within the scope of the present invention provided that the features included in any such combination are not mutually inconsistent as will be apparent from the context, this specification, and the knowledge of one of ordinary skill in the art. Additional advantages and aspects of the present invention are apparent in the following detailed description and claims. BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING(S)

[0013] The features and advantages of the present invention will become apparent from a consideration of the following detailed description presented in connection with theReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 accompanying drawings in which:

[0014] FIG. 1A shows a schematic of a traditional light-emitting diode (LED), where the active layer (i.e. GAN alloy) sits underneath a phosphor layer such that these layers touch.

[0015] FIG. 1B shows a schematic of a traditional remote LED, where the active layer and the phosphor layer are separated by a distance ‘d.’

[0016] FIG. 2 shows a schematic diagram of a stackable micro-LED based on an inverse Cassegrain model, as implemented in the present invention.

[0017] FIG. 3 shows a schematic diagram of a method for manufacturing the micro-LED device of the present invention.

[0018] FIG. 4A shows a schematic diagram of a method for manufacturing a phosphor layer of the micro-LED device of the present invention.

[0019] FIG. 4B shows a schematic diagram of a method for applying the phosphor layer to the micro-LED device of the present invention.

[0020] FIG. 5 shows a schematic diagram of the stack configuration of light sources in the micro-LED device of the present invention.

[0021] FIG. 6 shows a schematic diagram of an array configuration comprising both micro- and mini-LEDs as implemented in the device of the present invention.

[0022] FIG. 7A shows a schematic diagram of an array configuration of the micro-LED display of the device of the present invention comprising different phosphors configured to generate different wavelengths of light.

[0023] FIG. 7B shows a schematic diagram of four pixels of the array of the micro-LED display in combination with a lens system to generate a beam.

[0024] FIG. 8A shows an application of the micro-LED device of the present invention into a first lamp design.

[0025] FIG. 8B shows an application of the micro-LED device of the present invention into a second lamp design.

[0026] FIG. 8C shows an application of the micro-LED device of the present invention into a third lamp design.

[0027] FIG.9A shows a graph of YAG:Ce3+phosphor's absorbance spectra.

[0028] FIG.9B shows a graph of YAG:Ce3+phosphor's excitation spectra.

[0029] FIG.9C shows a graph of YAG:Ce3+phosphor's emission spectra.

[0030] FIG. 10 shows an illustration of a simple phosphor particle and a listing of what mayReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 occur within a simulation, a more realistic image of what the particle structure may look like, and a crystalline composition of YAG particles.

[0031] FIG. 11 shows diagrams of different scattering models, which are chosen based on particle size and wavelength of incident light.

[0032] FIGs 12A-12C shows transmission electron microscopy and scanning electron microscopy images of nano YAG particles of various sizes ranging from 50nm-100nm.

[0033] FIG.13A shows an isometric view of an edge-emitting LED (ELED).

[0034] FIG.13B shows a Vertical Cavity Surface-Emitting Laser (VCSEL).

[0035] FIGs 14A-14B show 3-dimensional views of an elliptic Gaussian (FIG. 14A) and an elliptic super Gaussian (FIG.14B) distribution with their respective expressions. Super gaussians have a curve with a flatter top and a steeper drop at the sides, causing a uniform intensity distribution along a 2D surface cut overlay, for the non-super expression P=1.

[0036] FIG. 15 shows real-color (top) and false-color intensity (bottom) distribution of the YAG phosphor in a spherical detector (IES type-A). Results are identified as being R-mode or T-mode.

[0037] FIG.16 shows laser to Phosphor matrix illustration of R mode and T mode.

[0038] FIG. 17 shows polar coordinates of intensity slice data for initial simulation results of a 450 nm laser exciting a YAG phosphor.

[0039] FIGs 18A-18B show results of varying the wavelength of the excitation beam (FIG.18A). Correlated simulated images are shown in FIG.18B, from top to bottom, results for 455 nm, 450 nm, and 445 nm. The white point value of D65 is also listed.

[0040] FIG. 19 shows a zoomed-up view of the simulated CCT values for varying wavelengths, plotted on a 1931 Colorimetry (CIE) color diagram.

[0041] FIG. 20 shows a 1931 Chromaticity Diagram, with the 450 nm source and correlating R and T plotted. A line can be generated between them, illustrating the constraints for color results by using a YAG phosphor.

[0042] FIGs 21A-21B show ray traces using phosphor particles of 1500 nm (FIG.21A) and 500 nm (FIG. 21B). Higher particle sizes mean that 2πr ≫ λ, which results in higher reflection and less forward scattering.

[0043] FIG. 22 shows longitudinal cross-sectional microstructure photographs of the sample. Larger phosphor particles are observed to sediment at the bottom.

[0044] FIG. 23 shows the optical path of a Cassegrain and inverse-Cassegrain model. (A), is a 2D cross section of a Cassegrain Telescope with a primary mirror (M1) and a secondary mirror (M2). The telescope collects light from the environment and focuses it at a point beyond M1.Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 Inverting the system (B) shows light emitting from the focal point which falls on M2 which in this case is a phosphor. Light from the phosphor is collected by M1 and is collimated out.

[0045] FIGs 24A-24B show a cross-sectional view initially theorized (FIG. 24A), of a single segmented pixel for an LED structure when an inverse-Cassegrain model (FIG.24B) is applied.

[0046] FIG. 25 shows a CPC reflector with a phosphor inside, with rays emitted from the excited surface of the phosphor being lost.

[0047] FIG. 26 shows Kyocera SLD Laser Inc®’s laser-to-phosphor (LtP) surface-mounted device (SMD) system, using the R-mode surface of the phosphor.

[0048] FIG. 27 shows LightTool®’s (LT’s) simulation set-up for a LtP system using a conic reflector.

[0049] FIG. 28A shows the initial result of the reflector system before being optimized by LT’s merit function for central intensity.

[0050] FIG.28B shows the optimized result of FIG.28A, with the change in curvature.

[0051] FIG.29A shows preoptimized results of the reflector system for k=-1.

[0052] FIG. 29B shows the optimized results of FIG.29B, showing a significant shift in average CIE color coordinates and front projection efficiency (FPE).

[0053] FIG.30 shows the deposition of a reflective layer on a phosphor chip.

[0054] FIGs 31A-31B show a comparison of metallized vs non-metallized top phosphor layer for the inverse-Cassegrain LtP system.

[0055] FIG.32A shows a y-z plane view of a laser’s angular deviation across the laser.

[0056] FIG. 32B shows a correlating schematic of the surface of the phosphor to which the laser subtends.

[0057] FIGs 33A-33C show images of a haloing effect that is reduced when a laser subtends the perimeters of the phosphor. The angular distribution of the laser increased from 1.5 to 4.5 degrees, falling towards the edge of the phosphor surface.

[0058] FIG.34A shows near-field results placed at the exit aperture.

[0059] FIG.34B shows Illuminating Engineering Society (IES) type A far-field results.

[0060] FIGs 34C-34D show color difference charts of the near-field results.

[0061] FIGs 35A-35B show far field spectral results of when the pixel’s environment is in air vs SiO2 layers. A larger spread is observed as the refractive index causes larger refraction angles.

[0062] FIG. 36A shows far field spectral results in a 180-degree sensor above the LED, with a correlating color difference chart and near field color difference chart.

[0063] FIG. 36B shows a graph of luminous intensity corresponding to the far field spectralReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 results of FIG.36A.

[0064] FIG. 37 shows OSRAM®’s Smartrix high-definition (HD) pixelated field-of-view (FOV) map for Adaptive Driving Beam (ADB) systems, given by LucidShape®’s User Guide on PixelLight®.

[0065] FIG. 38 shows simulation results (far-field) of a single μAFS chip 32x32 pixel array. Results are displayed in a logarithmic scale, false color map.

[0066] FIG. 39 shows far-field results for the simulated LtP-μLED (1W), scaled to 0.1W. Results displayed in a logarithmic scale, false color map.

[0067] FIG. 40 shows a graph of the spectral distribution of the optimized LtP LED using a YAG:Phosphor.

[0068] FIGs 41A-41B show raw data of absorbance spectra (FIG.41A) and correlating emission spectra (FIG.41B).

[0069] FIGs 42A-42B show a graph of the spectral distribution of the Sr:Eu2+phosphor, inside the IC LtP LED, post matrix optimization (FIG. 42A) with a correlating far-field spectral image (FIG. 42B). Note that at the end of the spectral distribution curve the results end at ~805 nm this is due to digitizing error, whereas the raw results don’t continue past ~805 nm.

[0070] FIG.43 shows an isometric cross section of the final device model in LightTools®.

[0071] FIG. 44A shows a schematic diagram of a first embodiment of the illumination device of the present invention.

[0072] FIG. 44B shows a schematic diagram of a second embodiment of the illumination device of the present invention.

[0073] FIG. 44C shows a flowchart of a method for fabricating the illumination device of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0074] Following is a list of elements corresponding to a particular element referred to herein:

[0075] 100 device

[0076] 110 base substrate

[0077] 112 laser sources

[0078] 120 intermediate substrate

[0079] 122 curved concave region

[0080] 124 hole

[0081] 126 reflective layerReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025

[0082] 130 top substrate

[0083] 132 micro-LED chip

[0084] 134 transmissive region

[0085] The term “phosphor” is defined herein as a solid material that emits light, or luminescence when exposed to radiation such as ultraviolet light or an electron beam.

[0086] The term “Cassegrain model” is defined herein as a lens configuration comprising a combination of a concave mirror and a convex mirror optically in-line with each other.

[0087] Referring now to FIG. 44A, the present invention features a micro-light-emitting diode (LED) device (100). In some embodiments, the device (100) may comprise a base substrate (110) comprising one or more laser sources (112) configured to generate a laser. The device (100) may further comprise an intermediate substrate (120) disposed on the base substrate (110). In some embodiments, the intermediate substrate (120) may comprise a curved concave region (122) comprising a hole (124) disposed through the intermediate substrate (120) such that the laser is directed through the hole (124). The intermediate substrate (120) may further comprise a reflective layer (126) disposed on the curved concave region (122) such that the reflective layer (126) matches a shape of the curved concave region (122).

[0088] In some embodiments, the device (100) may further comprise a top substrate (130) disposed on the intermediate substrate (120), optically in-line with the intermediate substrate (120). The top substrate (130) may comprise a micro-LED chip (132) disposed optically in-line with the hole (124), the micro-LED chip (132) comprising an alloy configured to convert the laser into light. The top substrate (130) may further comprise a transmissive region (134) disposed optically in-line with the reflective layer (126). The micro-LED chip (132) may be configured to convert the laser into the light and reflect the light to the reflective layer (126). The reflective layer (126) may be configured to reflect the light towards the top substrate (130) such that the light is transmitted through the transmissive region (134).

[0089] In some embodiments, the intermediate substrate (120) may comprise a silicate material. In some embodiments, the reflective layer (126) may comprise aluminum, gold, silver, or a combination thereof. In some embodiments, the top substrate (130) may comprise a dielectric material, an oxide material, or a combination thereof. In some embodiments, the one or more laser sources (112) may comprise a stack of laser sources. In some embodiments, the one or moreReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 laser sources (112) may comprise one or more vertical-cavity emitting lasers, one or more edge-LEDs, one or more micro-lasers, or a combination thereof. In some embodiments, the device (100) may be configured to be implemented into a lighting component for a flashlight, an automobile, a lamp, a micro-LED display, or a combination thereof. In some embodiments, the alloy may comprise a gallium nitride alloy.

[0090] Referring now to FIG. 44B, the present invention features a system (1000) for high-efficiency and high-output illumination. In some embodiments, the system (1000) may comprise an array substrate (110) comprising a plurality of laser sources, each laser source configured to generate lasers. The system (1000) may further comprise a plurality of micro-light-emitting diodes (LEDs) disposed on the array substrate such that each micro-LED (100) is configured to accept a laser generated by one or more laser sources (112) of the plurality of laser sources.

[0091] In some embodiments, each micro-LED (100) may comprise a base substrate (110) comprising one or more laser sources (112) configured to generate a laser. The device (100) may further comprise an intermediate substrate (120) disposed on the base substrate (110). In some embodiments, the intermediate substrate (120) may comprise a curved concave region (122) comprising a hole (124) disposed through the intermediate substrate (120) such that the laser is directed through the hole (124). The intermediate substrate (120) may further comprise a reflective layer (126) disposed on the curved concave region (122) such that the reflective layer (126) matches a shape of the curved concave region (122).

[0092] In some embodiments, the device (100) may further comprise a top substrate (130) disposed on the intermediate substrate (120), optically in-line with the intermediate substrate (120). The top substrate (130) may comprise a micro-LED chip (132) disposed optically in-line with the hole (124), the micro-LED chip (132) comprising an alloy configured to convert the laser into light. The top substrate (130) may further comprise a transmissive region (134) disposed optically in-line with the reflective layer (126). The micro-LED chip (132) may be configured to convert the laser into the light and reflect the light to the reflective layer (126). The reflective layer (126) may be configured to reflect the light towards the top substrate (130) such that the light is transmitted through the transmissive region (134).

[0093] In some embodiments, the system (1000) may further comprise a lens (200) disposedReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 optically in-line with one or more micro-LEDs of the plurality of micro-LEDs, configured to focus the lights from the one or more micro-LEDs into a beam. In some embodiments, one or more micro-LEDs of the plurality of micro-LEDs may be configured to produce the light at a plurality of wavelengths. In some embodiments, the intermediate substrate (120) may comprise a silicate material. In some embodiments, the reflective layer (126) may comprise aluminum, gold, silver, or a combination thereof. In some embodiments, the top substrate (130) may comprise a dielectric material, an oxide material, or a combination thereof. In some embodiments, one or more laser sources (112) of the plurality of laser sources may be arranged in a stack. In some embodiments, the plurality of laser sources may comprise vertical-cavity emitting lasers, edge-LEDs, micro-lasers, or a combination thereof. In some embodiments, the alloy may comprise a gallium nitride alloy.

[0094] Referring now to FIG. 44C, the present invention features a method for fabricating a micro-light-emitting-diode (LED) device (100). In some embodiments, the device (100) may comprise a monolithic wafer assembly. In some embodiments, the method may comprise providing an intermediate substrate (120). The method may further comprise ablating the intermediate substrate (120) to generate a micro-cavity. The method may further comprise etching the intermediate substrate (120) in the micro-cavity to form a curved concave region (122). The method may further comprise coating the curved concave region (122) with a reflective layer (126) such that the reflective layer (126) matches a shape of the curved concave region (122). The method may further comprise ablating the intermediate substrate (120) to generate a hole (124) in the curved concave region (122). The method may further comprise reconstructing the intermediate substrate (120) in a material. The method may further comprise providing a top substrate (130) comprising a transmissive region (134). The method may further comprise metalizing the top substrate (130) with a micro-LED chip (132). The method may further comprise bonding the top substrate (130) onto the intermediate substrate (120) such that the top substrate (130) is optically in-line with the intermediate substrate (120). The method may further comprise bonding the intermediate substrate (120) to a base substrate (110) comprising one or more laser sources (112) such that a laser generated by the one or more laser sources (112) is directed through the hole (124).

[0095] In some embodiments, the intermediate substrate (120) may be ablated by a pico-laser. In some embodiments, etching the intermediate substrate (120) may comprise processing the intermediate substrate (120) through a photolithography process. Etching the intermediateReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 substrate (120) may further comprise the use of potassium hydroxide (KOH). In some embodiments, the size of the micro-cavity may determine the curvature of the curved concave region (122). In some embodiments, the material that the intermediate substrate (120) is reconstructed in comprises a dielectric material, an oxide material, or a combination thereof.

[0096] In some embodiments, the base substrate (110), the intermediate substrate (120), the top substrate (130), or a combination thereof may be metallized, reconstructed into a dielectric and / or oxide substrate, or a combination thereof. In some embodiments, the top substrate (130) and the intermediate substrate (120) may be attached by dielectric bonds. In some embodiments, the micro-LED chip (132) disposed in the top substrate (130) may further comprise a reflective layer disposed on at least one side of the micro-LED chip. In some embodiments, the micro-LED chip (130) may be configured to emit infrared radiation upon excitation by the laser. In some embodiments, the micro-LED chip (130) may comprise a Yttrium Aluminum Garnet phosphor. In some embodiments, the micro-LED chip (130) may comprise a broad-band phosphor.

[0097] EXAMPLE

[0098] The following is a non-limiting example of the present invention. It is to be understood that said example is not intended to limit the present invention in any way. Equivalents or substitutes are within the scope of the present invention.

[0099] Laser-driven white light sources promise super high brightness, luminous flux and high directionality for solid-state lighting applications. In this example, the standard Cassegrain model was used, but in reverse, to develop a novel LED device package that directed and mixed all lights of the system. Utilizing LightTools®as the primary simulation software, important variables were identified and optimized inside a micro-LED configuration to show that the model device package emitted up to 360 lumens per watt, all while keeping fidelity to D65 correlated color temperature (CCT) and colorimetry (CIE) color values. Next, a ray file was generated to export results for evaluation as a light source using LucidShape®. Subsequently, the rayfile data was analyzed and integrated into an Adaptive Driving Beam (ADB) projection system, enabling detailed assessment of the LED's optical performance in comparison to an industry standard ADB light source. It was demonstrated that the prototype LED outperformed commercial LEDs in terms of luminous efficacy and projection efficiency.

[0100] Synopsy’s LightTools®were used for designing, optimizing, and analyzing solutions for illumination optics. LightTools®(LT) performs non-sequential Monte Carlo rayReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 tracing with their illumination module and has many core features, including an optimization module and an advanced physics module, the advanced physics module including phosphor particle modeling. Specifically, forward ray tracing was used, which simulated radiant flux propagation across the model (LightTools®, 2024). In forward ray tracing, rays are traced from random points within the source's volume or on its surface, shooting off in random angles based on probabilistic functions, using Monte Carlo calculations, that represent the emissive behavior of light sources. The rays hold information on wavelength and a set power (from 1 to 0) which attenuates as it passes through materials set in the simulation model. Once the ray passes through the medium, it now has a power reduced by the absorptive coefficient at the respective wavelength.

[0101] When the Monte Carlo method was applied to a phosphor, it took into consideration emission, absorption, excitation, and scattering. Each incident ray to the phosphor surface had a specific amount of power attenuated by its wavelength. The power was modified when hitting the phosphor particle, as it was either absorbed, absorbed and used for emission, or scattered. When the ray entered the phosphor material, LT calculated the average distance that a ray traveled throughout the material before it was likely to interact with a phosphor particle, determined by the mean free path (MFP).where l is the mean free path, n is the density of particles, is the scattering cross- sectional area, and is absorption cross-sectional area. Stochastic Monte Carlo calculations were done to determine the emission distribution from the excited phosphor particle.

[0102] The focus of the first simulation was to investigate several physical parameters of a phosphor when excited by a high-intensity beam (i.e., laser diode (LD)). First, important variables were identified inside Synopsys’s LightTools®. Primary variables were grouped into three categories: light source, phosphor, or matrix (i.e., phosphor encapsulant), as listed in the table below:Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025

[0103] In the context of configuring an LED for exterior automotive lighting, this investigation first focused on utilizing Yttrium Aluminum Garnet doped with Ce3+ “YAG:Ce” (Y3Al5O12:Ce3+). YAG:Ce phosphors absorbed wavelengths of blue photon energy (i.e., excitation wavelength) that were then down-converted and re-emitted by the phosphor isotopically. The phosphor particles then generated yellow light in the 500-750 nm wavelength range (i.e., emission wavelengths), typically peaking at 554 nm at room temperature. FIGs 9A-9C show the absorbance, emission, and excitation spectra, which were digitized using LT and used for simulation.

[0104] In respect to the crystalline structure of a YAG:Ce particle, when a photon interacts with a phosphor particle, several events may occur. The particle depending on size and composition, in simulation, can either: absorb, refract / transmit, reflect / scatter, and / or absorb and emit light, all of which are wavelength-dependent calculations (See FIG. 10). Absorption is understood as fundamentally a quantum process where an atom absorbs energy from incident photons, luminescence occurs when an electron from the atom is ‘promoted’ from ground state to an excited state. The excited state then decays to a metastable state (i.e., Stokes loss-- heat) before hitting the ground state, emitting a longer wavelength photon. This was understood by the collected spectra of emission, excitation, and absorption. There was an option in LT to use quantum yield replacing excitation spectra. Recall earlier that the MFP was used to calculate the probability that a ray would travel inside a phosphor medium without hitting a phosphor particle. Using MFP probability, if a ray was determined to hit a phosphor particle and had the proper absorption wavelength, LT then calculated if the incident wavelength caused an electron to jumpReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 the bandgap and therefore stimulated emission. The excitation spectrum was used instead of internal quantum yield calculations because most literature that includes emission spectra also includes a correlating absorption spectrum and therefore includes Stokes’ shift (i.e., the difference between the absorption and emission spectra' max wavelengths).

[0105] Another assumption was that the particles would undergo Mie scattering. Rays that encountered a phosphor particle had a probabilistic designation of either being absorbed or scattered, and due to the size of the particle, subjected to the wavelength of light. Mie Theory was employed. There were a few scattering theories to consider, Rayleigh, Mie and geometric. The general rule of understanding when to apply one over the other is expressed in the formula below, which assumes dielectric microspheres suspended in water. The formula shows that, Where r is the radius of the sphere (particle) and λ is the wavelength of the incident light. When 2πr ≪ λ (α ≪1), the situation is described by the Rayleigh formula. Mie scattering is described when 2πr ≈ λ (α ≈ 1), lastly when 2πr ≫ λ (α ≫1), this is what’s called Mie scattering of larger particles or geometric scattering. In this simulation, the excitation beam was 445-450 nm, and the particle size of a YAG phosphor ranged from 1 nm to 1500 nm. Based on wavelength and radius of particles, Mie scattering was assumed throughout the system or geometric scattering if particle sizes of 1000 nm or larger were used.

[0106] Inside the phosphor matrix it was assumed that the phosphor particles would act as volume scatters, meaning that statistically these particles would geometrically change the direction of any incident wavelengths based on their respective refractive index and sizes as explained above. LT allowed the definition of the volume scattering models as: phosphor, Mie, Henyey-Greenstein, Gegenbauer, User-Defined and User-Defined Dynamic Link Library (DLL). By selecting phosphor, particles were simulated that converted light. Additionally, Mie Theory was used to calculate the phosphor mean free path and the intensity distribution of unconverted rays.

[0107] Using the Mie scattering model in LT, the density of particles (mm-3) were specified, and mean free path, transmissivity, and angular scattering distributions were calculated based on Mie theory. In LT, controls were set for a phosphor mean free path based on Mie TheoryReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 and the intensity distribution of unconverted rays set to isotropic, meaning that the scattered rays equally likely exited the particle at any given angle.

[0108] Using volumetric scattering in most optical simulation models exponentially increased not only simulation time but deepening on RAM, and was also very difficult to use in the UI of the software, so it was important to realize these limitations and plan the simulation accordingly. This was mainly why every part was broken down into simpler simulation studies and models. Concerning particle size distribution, LT allowed the control of the distribution of particle sizes throughout the phosphor matrix. FIGs 12A-12C show images under transmission electron microscopy of YAG nanoparticles, the synthesis of which was done via solvothermal route. The important take away here when applying these images to LT was; for one, it was assumed that the radius of the Mie Particles was constant and that the shape was perfectly spherical. However, there was a plurality of different YAG particle synthesizing methods which would result in larger clumped shapes. Accuracy of the simulation would potentially be increased by defining a variant of Mie Particle sizes by using a gaussian size distribution, expressed in the equation below.Where is the radius of the , average statistical size and, is a small standard deviation of the set of particle sizes. The limit of the optimization merit functions considered the limit that these particles can be synthesized (as low as 5 nm) and was used as a starting limit for first optimizations.

[0109] Concerning excitation beam sources: Going back to the simulation set-up, a consideration on the type, shape, and emission of the excitation beam needed to be discussed. Semiconductor manufacturing was analyzed to choose a feasible laser system. Generally, there are two types of LEDs, edge emitting and surface emitting. As shown in FIGs 13A-13B, depending on which was chosen depended on the shape of the propagation beam. For symmetry purposes, Vertical Cavity Surface-Emitting Laser (VCSEL) was chosen, however either may ultimately be used for the function of the LED configuration. A more complete list comprises of potential laser sources is: Micro-LEDs (μLEDs), Edge-emitting lasers (EEL or ELED), Vertical cavity surface-emitting lasers (VCSELs), and Photonic Crystal Surface-emitting LasersReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 (PCSELs). ELEDs and VCSELs generated a beam pattern that was either circular or elliptical respectively, indium gallium nitride (InGaN) was the LED III-V semiconductor materials which was primarily responsible for delivering blue color in these systems. The spectral binning of these materials ranged in wavelengths of ~360-800 nm. However, in the applications listed above the binning ranged from ~0.05 nm-1.0 nm with peak wavelengths at 450 or 431 nm. Gallium(III) nitride generates a wavelength of 380nm-417nm or indium gallium nitride at 450nm.

[0110] The radiant power for the light sources was set to 1.00 Watt, this was done for simplicity, the results are typically reported in lumens therefore each lumen reported is a theoretical lumen / watt luminous efficacy (LE) result for the system. It is important to keep in mind that radiant power greatly impacts conversion via thermal and optical saturation. Thermal properties were ignored because LT, outside of spectral data manipulation, does not have a means to calculate these effects. The spectral data that was extracted was from lab test results, which were done under thermal constraints. However when the sizes of not only the phosphor but the beam were changed while keeping power constant, concerns arose that the simulated power would oversaturate the system. Understanding this, the power was kept at 1W, as it was easier to determine LE, knowing that source power would ultimately need to be scaled back linearly normalizing results to the scaled power. Lasers have different divergence angles for x / y axis, assuming projection in the z-dir, so it is common to assume that the light source may comprise a cross-section of an ellipse though are not limited to an ellipse as VCSELs propagate equally in both x / y positions. FIGs 14A-14B show the difference in LT calculations for beam patterns, taken from the user module guide for visualization of the system’s excitation beam propagation calculations, used in these simulations. The spatial gaussian values σxσy(mm) were user input values and were ultimately constrained by the size, and distance the excitation beam was from the phosphor.

[0111] Though LT can simulate coherent ray tracing, polarization and phase of the rays comprising the laser beam profile, these were not added for a few reasons. Because the coherence properties of the excitation beam were assumed to be negligible when exciting a phosphor particle for emission. Therefore, an incoherent source was assumed. Coherent source rays carried added data which compounded when simulation using hundreds of millions of rays was needed.

[0112] Due to the natural divergence of lasers, some Light to Phosphor (LtP) systems discuss using an aspheric lens in front of the light source. In these systems the distance betweenReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 the laser and phosphor was great enough so that the divergence angle created a cross-sectional of the laser surface area subtending to the phosphor surface greater than the phosphor surface area and thus needed correction. Adding aspheric lenses, especially for micro systems was costly, additionally for micro system focal lengths the natural divergence angle of the laser could be used to determine the limited distance from the phosphor needed to be to the excitation beam, therefore an additional lens was not needed. Some examples of beam divergence angles were between 5-20 degrees in the slow axis and 10 to 40 degrees in the fast axis, for VCEL’s it could be between from 1-5 degrees in both fast and slow axis. PCSELs (photo-crystal lasers) could be used to excite a phosphor matrix, and they had a less than 0.2-degree divergence angle along with a 0.05 nm band length. Ergo, there were many ranges of values to use for divergence and beam sizes.

[0113] With all that has been discussed in mind, the basic LtP system was simulated in a 360-degree detector. The goal was to observe how light in the system was generated and propagated to lay the foundation for the next simulation which looked at developing optical geometry. In the initial results, depicted in FIG. 15, the side of the phosphor the laser subtends had vastly greater luminous intensity than the opposite surface. That was because light that hit the first surface not only converted a portion of phosphor particles but reflected the unconverted excitation beam. The results of these two sides were termed R-mode and T-mode. R-mode described the surface where the excitation beam is reflected, and T-mode described the side that transmitted the remaining light (converted and unconverted) that propagates and exits the other end of the phosphor surface. A schematic representation of R and T mode is shown in FIG. 16. The phosphor was assumed to be suspended in silicone. The surface values were assumed to be 2% reflective and 98% transmissive, meaning that the reflection of the excitation beam was occurring due to the phosphor particles suspended inside the matrix.

[0114] This meant that careful consideration for the geometric design needed to focus around collecting light emitted from the surface the laser subtended. FIG.17 shows the simulated intensity distribution in polar coordinates. Skipping ahead to FIG. 26, this explained why most LtP devices model like Kyocera SLD Laser Inc®’s surface-mounted device (SMD) system, where the R-mode surface of the phosphor was the surface that faces the exit aperture of the LED.

[0115] There were two color metrics considered, correlated color temperature (CCT)14 and CIE color coordinates. These metrics are human derived, meaning that they describe, onReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 average, the perception of color based on the chromaticity of radiation rather than its spectral distribution. This was important because when creating an LED, the perceived color of the emitted radiation was heavily considered. The calculations for chromaticity were based on the spectral density of the observed light which had a matching CIE color function. The function comprised of tristimulus values, normally represented by X, Y, and Z or r(λ), g(λ), b(λ) respectively, and were used to create what was known as the 1931 x,y chromaticity diagram (seen in FIG.20).

[0116] CCT was strictly used when correlating to a white color, the spectral characteristics of blackbody radiation varied depending on its surface temperature. Consequently, a blackbody's surface temperature helped determine the color and spectral distribution of its emitted light. This had a correlating blackbody radiation curve in the central part of the CIE x,y chromaticity diagram. This metric was used to better understand the chromaticity results of the system, wherein cool white light typically had a CCT above 5000 K, indicating a higher intensity of blue tones, while warm white light, with its more orange-red hue, had a CCT below 4000 K.

[0117] Using both CIE x,y and CCT, the phosphor variables were ultimately optimized to either decrease MFP or increase it. By increasing the MFP, the likelihood of phosphor particles down-converting the excitation radiation and therefore increasing the spectral density of the yellow wavelengths tilting CCT below 4000 K towards yellow was also increased. Inversely so, changing the variables to decrease MFP caused a decrease in conversion rates and an increase in spectral density of the blue wavelengths tilting CCT above 5000 K. When using the merit function to optimize for color, the target D65 values, CCT 6508 K, and CIE x,y (0.3127, 0.3290) were assigned.

[0118] Variable excitation wavelengths were examined to observe its effect on CCT and generated lumens. The results listed in FIGs 18A-18B show that by decreasing excitation wavelength from 455 nm to 445 nm, generated lumen output was increased, as well as CCT in R-mode, and slightly in CCT in T-mode. The values of 450 nm and 445 nm yielded higher collected lumens because they aligned better to the respective spectral peaks, thus activating more particles to generate more lumens. Herein lied the optical trade off and the optimization that was conducted, optimizing parameters while keeping the fidelity of color coordinates around white (D65) while generating maximal number of lumens.Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025

[0119] The lighting industry continues to use color definitions specified in C.I.E 1931 Chromaticity Diagram, even though there have been multiple updates since. CIE defines several different white points, but one that is used in many testing facilities is the CIE Standard Illuminant D65. This white point corresponds to a CCT19 of 6,508 K and has an x / y value of 0.3127 and 0.3290 respectively. The R values were mapped on a zoomed 1931 Chromaticity Diagram to illustrate their relationship to D65. Ideally, it was preferable to achieve a white LED with CCT aligning with D65, with a maximum number of lumens generated. However, it was likely to trade color for higher luminous output, especially for devices wherein output was more important than color. As observed in FIG.19, as the CCT increased, lumens generated decreased simply due to less down-conversion. The average LE of this system for as close to white point was simulated to be 323 lumens / watt.

[0120] Displays utilized three different colored phosphors within a pixel to control color emission. The three black points represented each of the three phosphors, red, green and blue. Together they created a color triangle wherein the colors within that triangle are colors the display can replicate. For a YAG phosphor, the emission was yellow, and the excitation beam (‘source’) was blue. When plotted, the line between them represents the colors the LED can generate. Changing the variable properties previously listed moved the white point along the curve, increasing the influence of down-conversion or increasing the influence of the source beam.

[0121] In assessing the effects of larger phosphor particles within a phosphor matrix, a ray trace analysis highlighted an increase in the reflectivity of unconverted blue light as the particle size grew. Larger phosphor particles, with their greater surface area, reflected the incident blue light, reducing conversion efficiency and consequently lowering the luminous intensity of the source. This change in reflectivity shifted the CIE color coordinates towards the blue end of the spectrum, indicating a more noticeable bluish hue in the output light. The results demonstrated that larger particle sizes (where 2πr ≫ λ) tend to exhibit higher reflection rates and less forward scattering. This outcome suggested that adjustments to particle size in phosphor matrices significantly impacted light distribution and color. This of course could be modified by reducing the volume percent concentration of the phosphor particles, using larger particles while decreasing volume percent concentration to allow for increased conversion.

[0122] Due to the mechanical processing involved in phosphor production—like lapping,Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 abrasive machining, and chemical mechanical polishing (CMP) after sintering—the surface topography of phosphor samples was modeled with a Fourier function. The surface roughness affected the blue light emission when it interacted with the phosphor. LT assumed polished ‘smooth’ surface topography, meaning that the ray interacted with the surface as a flat interface. In LT, the surface roughness was changed via various methods; with a measured BSDF curve, advanced scattering models or simple scattering models. For these purposes the surface was set to Lambertian scattering.

[0123] Particles suspended within a matrix in LT are generally considered to be evenly distributed, but in practice, this assumption might not always hold true. The size distribution of phosphor particles and the time-dependent viscosity of silicone (or a similar material) are critical factors in determining the sedimentation behavior of these particles. As the material cures, viscosity changes, impacting how quickly particles settle. Stokes' law of resistance governs the rate of sedimentation, suggesting that larger or denser particles might settle faster, leading to a higher concentration at the lower part of the matrix due to gravity. If the phosphor particles accumulate near the base where the excitation laser enters, the phosphors may form a type of reflective layer that diverts the excitation beam away from particles suspended in the medium and therefore reduces luminous output of the system. To simulate, a stack of layers with different concentrations of phosphor particles was generated, though as the interest focuses on micro-structures, this was considered negligible without a lab tested comparison. The ‘feasible’ ranges from current manufacturing processes are shown in the table below.

[0124] The second simulation built on the first by placing the phosphor inside an inverse-Cassegrain model. Recall, a ‘classic’ Cassegrain reflecting telescope which comprised two mirrors, a parabolic primary mirror (M1) and a hyperbolic secondary mirror (M2), seen in FIG. 23 on the left. The hyperbolic secondary mirror reflected light collected by the primaryReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 mirror and focused it beyond M1. Applying the Cassegrainian system for illumination purposes, when it was inversed, a light source starting from the focal point illuminated the secondary mirror (i.e., phosphor). Light emitted by the phosphor was then collected by the primary mirror and collimated out, seen in FIG.23 on the right.

[0125] In the theoretical device model as shown in FIG. 24A, an inverse-Cassegrain structure was applied (FIG. 24B) to a single pixel of an LED. FIG. 24A shows a cross sectional view of an LED device (micro or mini) for a singular pixel. At the base of the structure was a light source layer like an ELED stack, a VCSEL or other previously listed. On top of that was an intermediate layer (106) and a reflective surface was placed between them (effectively acting as a primary mirror). The intermediate layer consisted of an oxide or dielectric material. Acting as the secondary mirror of a Cassegrain model, a phosphor was embedded in the top layer with a reflective coating, to direct light emitted from the phosphor to the primary mirror instead of transmitting through the phosphor. Lastly, a reflective or absorptive coating was disposed on the outer perimeter of the stacked device, which acted as a deep trench isolation (DTI) (112) and prevented optical crosstalk from pixel to pixel.

[0126] Inserting the theoretical device model into the simulation, it was first needed to define the spatial limits of the exit aperture (i.e., pixel surface). Because the interest was in developing a micro-LED, the dimensions of the pixel were limited to be less than the upper limit of a micro-LED (<100 μm). LED sizes were typically understood to fall under three categories: micro, mini, and ‘standard’. Micro-LEDs (μLEDs) have end-to-end dimensions ranging from a few microns to 100 μm. Mini-LEDs are slightly larger, with dimensions ranging from 100 μm to a few hundred microns. ‘Standard’ LEDs have an average size of approximately 1000 μm.

[0127] Upon selecting size limitations, choosing a reflector shape to use as the primary mirror was the next step. The first mirror structure that was evaluated was a CPC (compound parabolic concentrator) seen in FIG. 26. CPCs were used to concentrate the radiation from a clipped Lambertian source in a nearly etendue-preserving manner, wherein the Lambertian source spanned along the entrance aperture aimed at the exit aperture. This collected and mixed light off the R-mode surface of the phosphor. Prior results illustrated that the greatest amount of flux was generated off the R-side of the phosphor. It was postulated that because of this physical feature, most LED manufacturers for LtP systems will focus on designing their products as an SMD, pictured in FIG. 26. SMDs simply use the excited surface of the phosphor for directReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 emission. Though SMDs may be simpler to manufacture, the system itself has many optical trade-offs including lower conversion, and virtually no capacity for color mixing, or beam shaping.

[0128] The second shape considered for the reflector was a conic reflector, several conic constant values were evaluated and among them two were selected for further optimization. Optical systems benefit from being rotationally symmetric about the z-axis. One design feature in LT is the ability to generate rotationally symmetric surfaces simple or polynomial via conics and / or super conics. A conic surface is defined by a radius (r) and a conic constant (k), the conic constant refers to the type of surface which are: −1 < < 0 ellipse = 0 sphere = −1 parabola < −1 hyperbola > 0 oblate sphere the equation used to calculate the sag of a conic surface (z) is given below, where c is the surface curvature (1 / r), assuming the major axis is the Z-axis. For higher order and greater control LT offers superconic surface control (i.e., aspheric surfaces).

[0129] Four reflectors with the conic constant of -1, 0, 1, and 0.499 were evaluated, and a front projection efficiency (FPE) was calculated to determine which two to further investigate and optimize, FPE is the ratio of front collect lumens in a 30° x 30° detector to lumens collected in a full 360° detector. Due to better performance and manufacturability, 0 and -1 were selected for further study. The results for optimizing a spherical reflector as seen in FIGs 28A-28B demonstrated a slight shift in color coordinates, towards a bluer white, and an increase in FPE. The optimization was done via LT’s merit function wherein the optimized variable was the curvature of the reflector to the central intensity of the detector.Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025

[0130] A similar optimization was done for k =-1 the results of which are in FIGs 29A-29B, a parabolic reflector for a micro-LED proved to have a much higher max intensity ~2500 cd vs. 800 cd, with the respective lumens for both being 282.56 lumens. This difference led to the design decision to keep the conic constant as -1 throughout the experiment while noting that others may be used for different purposes. Metallization, to apply a reflective coating, is a common operation in semiconductor fabrication, a highly reflective layer is disposed on top a segmented phosphor chip (techniques for which can include physical vapor deposition PVD or chemical vapor deposition CVD), as seen in FIG. 30 below. These metallic layers can comprise Au, Ag, Al, or an alloy etc. Realizing that the top of the phosphor surface needed to be covered to prevent direct transmission of laser light to the observer, the effects of metalizing the phosphor with highly reflective ~98.7% aluminum coating were simulated. A metallization layer acted as a barrier for direct viewing of the beam, for beam safety. The FDA has four major hazard classification of lasers (I to IV), if the laser is classified above class I, a metallization layer to block direct transmission of the propagating beam is required.

[0131] The addition of the metalized layer was simulated and results for metalized vs non- metalized are described in FIGs 31A-31B. By simply adding the reflective coating to the top of the phosphor, total collected lumens jumped from 292.56 to 326.18, while CCT shifted substantially from 7840K to 5517K (cooler to warmer). The reasoning behind this significant change was that by adding an aluminum top layer, unconverted laser light was reflected back into the phosphor matrix, increasing the mean free path of the excitation beam. This increased the statistical likelihood that the excitation beam excited a phosphor particle which resulted in higher conversion rates (i.e., higher lumen output and yellow light phosphor emission). This simulation illustrated the optical tradeoff for LtP systems: optimizing for conversion results in an increase of total lumen output at the expense of CCT shifting in the direction of the phosphor emission wavelengths.

[0132] In terms of color uniformity, it was noted that obtaining an even color distribution is difficult due to the way each illuminating source propagates: the blue laser, a gaussian beam; the phosphor-converted-yellow, Lambertian. The uneven mixing caused a ‘haloing effect’ where the center of the white beam may have proper CCT and color coordinate but show obvious deviation towards yellow at wider angles (haloing can be observed in FIGs 31A-31B). To increase homogeneity of the blue light, one idea was to optimize the surface at which the excitation beam subtends on the phosphor. As shown in FIGs 32A-32B below, the angularReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 gaussian distribution was increased from 1.5° to 4.5° to cover a larger surface area of the phosphor. However, the beam could not reach the edge of the phosphor, since there was a chance that the beam would refract inside the matrix without being converted. This would cause blue artifacts in the image which could be viewed when the angle was at ~ 5°. Though not perfect, optimizing the angle at which the laser subtends the phosphor did create a smoother blue light distribution, as shown in FIGs 33A-33C. Lastly, these results showed when metallization was done on top of the phosphor. When metallization was performed on the sides, refraction was not an issue and there was no need for a large safety margin.

[0133] Several merit functions were written in LT to solve for the following: optimizing variables to yield highest luminous output, beam focusing, color correcting to establish average white color, editing variables to reduce color difference across projected beam. FIGs 34A-34D show a detailed result following an optimization.

[0134] Currently, all the simulations conducted were done in air. However, this is not a realistic representation since the phosphor cannot be magically suspended inside the reflector. A method for building the inverse-Cassegrain configuration may include a combination of MEMS and semiconductor manufacturing techniques. The methods may include techniques such as EPI- Wafer reconstitution process or wafer reconstruction both of which are used to form a monolithic wafer for die-to-wafer assembly. A common example of such a process is embedding materials in epoxy molding compounds (EMCs). Using this process, reconstituting singulated phosphor dies inside a substrate allowed the formation of stackable layers to build the micro-LED. Typically, an LED would need electrical interconnects to allow electricity to excite the phosphor causing it to luminesce. For an LtP system, there was no need to connect the phosphor to an IC controller and therefore only a few manufacturing process steps were needed to generate a component for an LED configuration (micro or other). Therefore taking FIG. 23 of the theorized LED configuration, take-aways from the simulation results were taken in order to construct a micro- LED that may look similar to the cross-sectional image of FIG.2.

[0135] Using LT’s immersion manager, the geometry reflected in FIG.2 was replicated to align with simulation and test several materials that the LtP system was embedded in, the results for which are found in Table 3 below.Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025

[0136] In reconstructing the phosphor in a substrate, higher scattering was observed, as well as more conversion (CCT slight shift to yellow), however a decrease in lumens and max intensity was also observed as light was absorbed by the material. Reconstructing the system in glass, for example, as illustrated in FIGs 35A-35B, showed a wider attenuation of the beam due to the higher refractive index of the reconstructed material. The higher the refractive index, the less intense the beam was, which would require a re-optimization of the curvature of the reflector.

[0137] Reconstructed stacked surfaces are beneficial for a multitude of reasons. If the phosphor sits on top of the intermediate layer housing the reflector, and the phosphor layer depletes or burns out, the top substrate may be exchanged without reconstructing the intermediate layer. Secondly, additional layers or optical films may be stacked in-between or on top to attenuate desired wavelengths, some examples may include adding layers for polarization, additional scattering, wavelength filters, and / or refractive surfaces. Additionally, stacking a meta-lens layer on top of the current configuration may assist in color uniformity while also focusing the multi-spectral beam, without increasing the color difference data across the beam.

[0138] A review of literature suggests that the most practiced method for addressing color uniformity issue is by introducing a secondary phase in the phosphor matrix, whereby the phosphor matrix is a ceramic. Prior literature has stated that when doping a phosphor ceramic with a secondary phase (i.e., TiO2, BN, Al2O3or SiO2), the microstructure of the phosphor is optimized to enable higher scattering for blue light excitation. It was reported that among secondary phases tested, TiO2proved to have the best multinational uniformity in illumination image, speckle image, illuminance curve, CCT and luminance distribution, followed by BN andReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 Al2O3. Simulated results of using various reconstructed materials are shown in the below table:

[0139] An alternative suggestion to generate better color mixing is to increase the Lambertian scattering by changing the reflectivity type of the primary mirror to Lambertian instead of assuming a specular reflective coating. The charts, in FIGs 36A-36B, show that both the near field and far field exhibit good color mixing with low dudv values. It was observed that making the reflective coating on the phosphor Lambertian had a much smaller impact on increasing color uniformity, most of the mixing comes from the primary reflector. Values for each simulation with their respective percent reflectivity are listed in a table below. It should be noted that each system’s phosphor composition could be optimized to compensate for the reduction in luminous flux, meaning that it would be possible to extract more lumens out of the system by optimizing the phosphor variables previously discussed at the cost of shifting CCT towards the emission spectra.

[0140] With the current optimized configuration, a 1 million and 50 million .ray file was exported out of LT and converted into LucidShape®(LS) binary. LS is the common design software in automotive lighting so to successfully apply the micro-LED in a larger optical system in LS would mean a potential avenue for distribution of such desired rayfiles. Additionally, LSReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 has many resources to create an ADB system which are utilized in this chapter.

[0141] There are four leading technologies for ADB systems in use, the first is a pixelated phosphor LED like OSRAM®’s μAFS (adaptive front-lighting systems), this system uses a blue LED with a pixelated phosphor coating. The second, is an LCD system composed of a LED light source, an LCD screen, and a projection lens. The third is a digital micro-mirror device (DMD) with digital light processing chips (DLP) illuminated by a high intensity light source. The fourth uses piezoelectric actuation mechanisms for its biaxial MEMS optical scanner and a laser to control how light is distributed on a phosphor plate. The IC-LtP system configured in this example was comparable to either the μAFS or the laser-scanning MEMS device, both are additive image sources meaning that a blue light source excites a yellow phosphor to generate white light, however each have comparable differences. The LCD and DMD systems are both subtractive systems, meaning that the source is already a white color, an LCD system will absorb light pixelated from the image plane whereas DMDs will reflect light. It may also be of interest that in addition to DMD systems spatial light modulators (SLMs) have been discussed as alternatives and or additions to DMDs.

[0142] Using LS, a typical ADB lens system (supplied by LS’s model gallery) was taken and evaluated and compared to OSRAM’s μAFS. Using PixelLight®(LS’ tool for creating a micro-LED array) a 32x32 pixelated light source was generated for the μAFS with size and spacing listed previously. This was surprisingly simple and fast for the software to do because the μAFS was assumed to be a Lambertian surface emitter not comprising spectral information. However, that changed when using a .rayfile, for emission information, this caused the system to lag and crash which became difficult to manage, therefore the size was reduced to a 10x10 source when evaluating the IC LtP LED.

[0143] For High Beam (HB) High Definition (HD) ADB, it was required that the FOV for X and Y respectively were 32° by 8° for a single headlamp (a Left Hand (LH) headlamp having an overlap with a Right Hand (RH) headlamp). OSRAM®’s Smartrix HD setup which uses 4 μAFS chips (per side) comprises of a two-projector unit (per car lamp), having two chips each in a single headlamp projector configuration. Upon simulating one chip in the projector system, it was found that one 32x32 chip resolved to an 8° x 8° box in the far field image plane. The results of the simulation are shown in FIG. 39. This fit precisely inside the LH requirement when 3 more chips were added as described in OSRAM®’s HD HB ADB configuration.Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025

[0144] For this same ADB projector lens model, PixelLight®was used to generate a 10x10 array with emission ascribed by the .rayfile. This configuration resulted in a projected FOV of 2° x 2° which fit into OSRAM®’s requirements for HD. The table of the simulated and calculated results shown below are for a 4mm x 4mm emission surface area where the (1W) system is the max theoretical for the LtP system, the results were also simulated with a 0.1 W laser and simulated, listed in the table below. The LtP system showed a higher projection efficiency due to the emission type. The LtP system also had a higher resolution due to its size. In an emission area of 4mm x 4mm, Osram®was able to fit 1024 controllable pixels whereas the LtP system fit 1521 pixels.

[0145] The LtP was simulated using the 10x10 configuration then normalized to a 0.1W laser source per pixel to show results. Understanding that there was more room to decrease the laser source power consumption due to heat or general electrical constraints. The far field intensity results shown in FIGs 7A-7B. Scaling the 10x10 configured results to fit into the HD FOV, it was calculated that in the x direction across the HD ADB map, 160 pixels were individually controlled as compared to OSRAM®’s μAFS 128. This resulted in the LtP- μLED having a higher angular resolution of 0.20 degrees, an increase of 0.05 degrees as compared to the simulated μAFS results. The values for the resolution of the results are shown in the table below. Aside from comparing spectral distribution and color uniformity, which were omitted due to being unable to accurately simulate spectral information of the μAFS without lab verified results. Next, the system results were projected on a 25m Illuminance map to measure lux values. For this HD ADB HB, a minimum of 32 lx per headlamp was required. For a single μAFS chip, 28.8 lx per headlamp was simulated and was found to be 115 lx and compared to OSRAMS®online results of 110 lx. This showed good correlation between the simulation and valuesReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 reported online. The results under LtP- μLED were a combination of simulated and then normalized results. The 10x10 chip was simulated to have 61.577 lx however these fit in 132x32 μAFS FOV 16 times, resulting in 985 lumens in a direct size comparison. This was also done to the simulated results at 0.1 W laser beam source. The below table shows the simulated resolution results for the μAFS and LtP-μLED.

[0146] The below table shows the simulated illuminance results for the μAFS and LtP-μLED.

[0147] IC-LtP LED systems showed potential improvements over traditional systems with higher resolution, better output efficiency, and uniform color mixing. Once more, the method in which these may be manufactured via layer stacking, offered more flexibility to customize LED chips without the need of increasing manufacturing costs. For example, the primary mirror layer could be manufactured in monolithic wafer sets and in bulk for both automotive and display LED chips. The difference would be to bond a different top and bottom layer. As the device currently stands for ADB HD systems, an array configuration of IC-LtP LED pixels was likely to be used as illustrated by FIGs 7A-7B. The sizes of the pixel’s emission surface ultimately were determined by how small a phosphor chip could be made limited by the power density of the laser emission avoiding issues caused by luminance saturation e.g., optical and thermal quenching. The spectral distribution of the IC LtP LED when comprising a broadband spectrum showed a broad curve and resulted in larger optical applications having better color rendering properties.

[0148] Using standard Cassegrain telescope design, in reverse, effective methods forReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 phosphor illumination were examined to create white light for automotive applications. Taking the inverse Cassegrain structure, the geometry was optimized, as well as phosphor matrix variables to increase lumen output while balancing with desired CCT all confined in a micro-LED package (pictured in FIG. 43). It was determined that the luminous efficacy of this configuration falls within a range of 295-360 lumens per watt with CCT ranging from 4717 K - 6042K. By reducing specular reflection of components and increasing Lambertian reflection of the mirror and phosphor surfaces, the dudv color difference values could be reduced to be uniform across the emission beam of the LED but the tradeoff was sacrificing luminous efficacy along with increasing the projection beam angle.

[0149] Understanding this model needed to be suspended in a medium for it to work, reconstruction of all the components was evaluated of the inverse-Cassegrain into stackable wafer assemblies. This allowed the generation of an IC LtP-μLED chip structure in a 10x10 array. The chip was evaluated and compared to an industry standard, like OSRAM®’s μAFS. These were both plugged into LucidShape®’s ADB projection lamp. The IC LtP-μLED chip was simulated to be 35% more efficient and had the potential to project 5 to 30 times the number of lumens per chip size. It was also calculated that the IC LtP system had an increase of 0.05 degrees allowing for more high-definition application and simulated on a 25 m illuminance map to show that the IC LtP system also had at least two times higher illuminance than the μAFS per chiplet.

[0150] As the automotive industry seeks to increase energy efficiency of electrical components, IC-LtP LEDs offer a solution to reduce energy consumption without sacrificing luminous output. In a 1:1 comparison normalizing each IC LtP pixel to emit 1 lumen would require only 0.0027 W per pixel and in a 32 x 32-pixel device package the whole chip would need 2.76 W to generate slightly above 1k lumens. This was a huge improvement over traditional devices and therefore may be utilized by other industries that require high luminance sources, like AR / VR displays. Lastly, the stickability of the wafer assembly per part-layer allowed the IC-LtP geometry to be maintained while exchanging either the reconstructed phosphor layer and / or the excitation layer to customize spectral output for a variety of applications.

[0151] Although there has been shown and described the preferred embodiment of the present invention, it will be readily apparent to those skilled in the art that modifications may be made thereto which do not exceed the scope of the appended claims. Therefore, the scope of theReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 invention is only to be limited by the following claims. In some embodiments, the figures presented in this patent application are drawn to scale, including the angles, ratios of dimensions, etc. In some embodiments, the figures are representative only and the claims are not limited by the dimensions of the figures. In some embodiments, descriptions of the inventions described herein using the phrase “comprising” includes embodiments that could be described as “consisting essentially of” or “consisting of”, and as such the written description requirement for claiming one or more embodiments of the present invention using the phrase “consisting essentially of” or “consisting of” is met.

[0152] The reference numbers recited in the below claims are solely for ease of examination of this patent application, and are exemplary, and are not intended in any way to limit the scope of the claims to the particular features having the corresponding reference numbers in the drawings.

Claims

Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 WHAT IS CLAIMED IS:

1. A micro-light-emitting diode (LED) device (100) comprising: a. one or more laser sources (112) configured to generate a laser; b. an intermediate substrate (120), comprising: i. a curved concave region (122) comprising a hole (124) disposed through the intermediate substrate (120) such that the laser is directed through the hole (124); and ii. a reflective layer (126) disposed on the curved concave region (122) such that the reflective layer (126) matches a shape of the curved concave region (122); and c. a top substrate (130) disposed on the intermediate substrate (120) comprising: i. a light source (132) disposed optically in-line with the hole (124) configured to convert the laser into light; and ii. a transmissive region (134) disposed optically in-line with the reflective layer (126); wherein the light source (132) is further configured to convert the laser into the light and reflect the light to the reflective layer (126); wherein the reflective layer (126) is configured to reflect the light towards the top substrate (130) such that the light is transmitted through the transmissive region (134).

2. The device (100) of claim 1 further comprising a base substrate (110) comprising the one or more laser sources (112), wherein the intermediate substrate (120) is disposed on the base substrate (110).

3. The device (100) of claim 1, wherein the light source (132) comprises a micro-LED chip comprising an alloy.

4. The device (100) of claim 3, wherein the alloy comprises a gallium nitride alloy.

5. The device (100) of claim 1, wherein the intermediate substrate (120) comprises a silicate material.

6. The device (100) of claim 1, wherein the reflective layer (126) comprises aluminum, gold, silver, or a combination thereof.

7. The device (100) of claim 1, wherein the top substrate (130) comprises a dielectric material, an oxide material, or a combination thereof.Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 8. The device (100) of claim 1, wherein the one or more laser sources (112) comprise one or more vertical-cavity emitting lasers, one or more edge-LEDs, one or more micro-lasers, or a combination thereof.

9. The device (100) of claim 8, wherein the one or more laser sources (112) comprise a stack of laser sources.

10. The device (100) of claim 1, wherein the device (100) is configured to be implemented into a lighting component for a flashlight, an automobile, a lamp, a micro-LED display, or a combination thereof.

11. A system (1000) for high-efficiency and high-output illumination, the system (1000) comprising: a. an array substrate (110) comprising a plurality of laser sources, each laser source configured to generate lasers; and b. a plurality of micro-light-emitting diodes (LEDs) disposed on the array substrate (110) such that each micro-LED (100) is configured to accept a laser generated by one or more laser sources (112) of the plurality of laser sources, each micro-LED (100) comprising: i. an intermediate substrate (120) comprising: A. a curved concave region (122) comprising a hole (124) disposed through the intermediate substrate (120) such that the laser is directed through the hole (124); and B. a reflective layer (126) disposed on the curved concave region (122) such that the reflective layer (126) matches a shape of the curved concave region (122); and ii. a top substrate (130) disposed on the intermediate substrate (120), optically in-line with the intermediate substrate (120), the top substrate (130) comprising: A. a micro-LED chip (132) disposed optically in-line with the hole (124), the micro-LED chip (132) comprising an alloy configured to convert the laser into light; and B. a transmissive region (134) disposed optically in-line with the reflective layer (126); wherein the micro-LED chip (132) is further configured to convert the laser into the light and reflect the light to the reflective layer (126);Reference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 wherein the reflective layer (126) is configured to reflect the light towards the top substrate (130) such that the light is transmitted through the transmissive region (134).

12. The system (1000) of claim 11 further comprising a lens (200) disposed optically in-line with one or more micro-LEDs of the plurality of micro-LEDs, configured to focus the lights from the one or more micro-LEDs into a beam.

13. The system (1000) of claim 11, wherein one or more micro-LEDs of the plurality of micro-LEDs are configured to produce the light at a plurality of wavelengths.

14. The system (1000) of claim 11, wherein the intermediate substrate (120) comprises a silicate material.

15. The system (1000) of claim 11, wherein the reflective layer (126) comprises aluminum, gold, silver, or a combination thereof.

16. The system (1000) of claim 11, wherein the top substrate (130) comprises a dielectric material, an oxide material, or a combination thereof.

17. The system (1000) of claim 11, wherein the plurality of laser sources comprise vertical-cavity emitting lasers, edge-LEDs, micro-lasers, or a combination thereof.

18. The system (1000) of claim 17, wherein the one or more laser sources (112) of the plurality of laser sources are arranged in a stack.

19. The system (1000) of claim 11, wherein the alloy comprises a gallium nitride alloy.

20. A method for fabricating a micro-light-emitting-diode (LED) device (100) comprising: a. providing an intermediate substrate (120); b. ablating the intermediate substrate (120) to generate a micro-cavity; c. etching the intermediate substrate (120) in the micro-cavity to form a curved concave region (122); d. coating the curved concave region (122) with a reflective layer (126) such that the reflective layer (126) matches a shape of the curved concave region (122); e. ablating the intermediate substrate (120) to generate a hole (124) in the curved concave region (122); f. reconstructing the intermediate substrate (120) in a material; g. providing a top substrate (130) comprising a transmissive region (134) and a micro-LED chip (130); h. bonding the top substrate (130) onto the intermediate substrate (120) such that the top substrate (130) is optically in-line with the intermediate substrate (120); andReference No.: UA24-246, ARIZ 24.26 PCT Inventor’s last name: Schur and Koshel Document Date: May 1, 2025 i. bonding the intermediate substrate (120) to a base substrate (110) comprising one or more laser sources (112) such that a laser generated by the one or more laser sources (112) is directed through the hole (124).

Images