US20150295154A1

US20150295154A1 - Light emitting device and manufacturing method thereof

Info

Publication number: US20150295154A1
Application number: US14/687,369
Authority: US
Inventors: Chuan-Cheng Tu; Jen-chau Wu; Yuh-Ren Shieh; Tzer-Perng Chen; Min-Hsun Hsieh
Original assignee: Epistar Corp
Current assignee: Epistar Corp
Priority date: 2005-02-03
Filing date: 2015-04-15
Publication date: 2015-10-15
Also published as: US20200227591A1; US10978615B2; US20210217926A1

Abstract

The present disclosure provides a method for forming a light-emitting apparatus, comprising providing a carrier having a plurality of first metal contacts; forming a light-emitting structure comprising a substrate, a first cladding layer on the substrate, an active layer on the first cladding layer, and a second cladding layer on the active layer; bonding the light-emitting structure to the carrier; forming a cap layer on a side of the light-emitting structure opposite to the carrier; and cutting the carrier and the cap layer to form a chip-scale LED unit.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

This application is a continuation application of U.S. patent application Ser. No. 14/082,960, filed Nov. 18, 2013, which is a continuation-in-part of U.S. patent application Ser. No. 11/674,371, filed Feb. 13, 2007, which is a continuation-in-part of U.S. patent application Ser. No. 11/249,680, filed Oct. 12, 2005, now U.S. Pat. No. 7,192,797. The disclosures of all references cited herein are incorporated by reference.

FIELD OF DISCLOSURE

The present disclosure relates to a light emitting device and a method for manufacturing the same. In addition, the present disclosure relates to a light emitting device array and a method for manufacturing the same.

BACKGROUND OF THE DISCLOSURE

For conventional light emitting device (LED) packages, a LED chip is mounted onto the sub-mount using the epoxy put thereon to form a LED element, and the process is called “Die Bonding”. Typically, the epoxy used in “Die Bonding” can be silver filled epoxy or other non-conductive resin. Then, the LED elements are assembled onto the circuit board. For a flip-chip LED, the p-type conductive layer and the n-type conductive layer are exposed on the same side to have the positive electrode and the negative electrode on the same side of the LED structure. And the LED structure with the positive electrode and the negative electrode is flipped and disposed on the solder without wire bonding. However, conventional flip-chip LEDs still require “Dicing” and “Die Bonding” for connecting and mounting the circuit board. If the electrodes of flip-chip LEDs have large contact area to be directly connected to the circuit board, a number of conventional packaging processes for LEDs can be skipped.
The operating current of a conventional LED is typically several tens to several hundreds of mAs. Therefore, the brightness of a conventional LED is not suitable for illumination purpose. When lots of LEDs are assembled into a LED lamp to improve the brightness, the volume of the LED lamp increases accordingly and results in the loss of its market competitiveness. Therefore, to improve the brightness of a single LED is a necessary approach. However, as the LED advances towards high brightness, the operating current and power of a single LED become several times to several hundred times than those that a conventional LED requires. For example, the operating current of a high brightness LED is about several hundreds of mAs to several Amps (A). As a result, the heat generated by the LED becomes an important issue. “Heat” seriously affects the performance of LEDs; for example, the thermal effect influences the wavelength of lights emitted from the LED, reduces the brightness of lights generated from the semiconductor device, and damages the LED device. Therefore, how to dissipate heat generated by the high power LED become the important issue of the LEDs.
US Applications Nos. 2004/0188696 and 2004/0203189 disclosed a LED package and the method for manufacturing the same based on the Surface Mount Technology (SMT). Each LED package includes a LED chip, and each chip is flip-chip bonded onto a frontside of the sub-mount wafer using boning bump. A plurality of arrays of openings is drilled into the electrically insulating sub-mount wafer. A metal is applied to the drilled openings to produce a plurality of via arrays. The p-type and n-type contacts of each flip-chip bonded LED electrically communicate with a solderable backside of the sub-mount wafer through a via array. A thermal conduction path is provided for thermally conducting heat from the flip-chip bonded LED chip to the solderable backside of the sub-mount wafer. Subsequent to the flip-chip bonding, the sub-mount wafer is separated to produce the surface mount LED packages.
However in US Applications Nos. 2004/0188696 and 2004/0203189, it requires drilled via array with filled metal within the sub-mount wafer and thus increases the manufacturing cost. Furthermore, it becomes complicated to flip-chip bond each chip onto the sub-mount wafer using bonding bump. Therefore, it would be beneficial if the LED packages have excellent thermal conductive paths without the provision of the sub-mount wafers.

SUMMARY OF THE DISCLOSURE

The present disclosure provides a method for forming a light-emitting apparatus, comprising: providing a carrier having a plurality of first metal contacts; forming a light-emitting structure comprising a substrate, a first cladding layer on the substrate, an active layer on the first cladding layer, and a second cladding layer on the active layer; bonding the light-emitting structure to the carrier; forming a cap layer on a side of the light-emitting structure opposite to the carrier; and cutting the carrier and the cap layer to form a chip-scale LED unit.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing aspects and many of the attendant advantages of this disclosure will become more readily appreciated as the same becomes better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein:

FIGS. 1A-1E illustrate cross-sectional views of forming a light emitting device (LED) in accordance with one embodiment of the present disclosure;

FIG. 1F illustrates LEDs in accordance with an embodiment of the present disclosure;

FIGS. 2A-2D illustrate cross-sectional views of forming a LED array in accordance with one embodiment of the present disclosure;

FIG. 2E illustrates the LED array connected to the circuit board in accordance with another embodiment of the present disclosure;

FIG. 2F illustrates a LED array having a first conductive layer with a rough surface in accordance with one embodiment of the present disclosure;

FIG. 2G illustrates a LED array package in accordance with an embodiment of the present disclosure;

FIGS. 3A-3G illustrate one embodiment of a chip-scale RGB LED unit and the manufacturing method thereof in accordance the present disclosure;

FIG. 4A is a top view of an LED array flip-bonded to a circuit board depicted in FIG. 3F;

FIG. 4B is a top view of a chip-scale RGB LED unit depicting in FIG. 3G;

FIG. 5A is a top view of an LED array flip-bonded to a circuit board depicted in accordance with another embodiment of the present disclosure;

FIG. 5B is a top view of a chip-scale LED unit in accordance with one embodiment of the present disclosure;

FIG. 5C is a cross-sectional view of a chip-scale LED unit in accordance with one embodiment of the present disclosure;

FIG. 5D is a top view of a chip-scale LED unit in accordance with another embodiment of the present disclosure;

FIG. 5E is a cross-sectional view of the chip-scale LED unit depicted in FIG. 5D;

FIG. 6A illustrates a chip-scale RGB LED unit in accordance with another embodiment of the present disclosure;

FIG. 6B illustrates a chip-scale LED unit in accordance with another embodiment of the present disclosure;

FIG. 6C illustrates a chip-scale LED unit in accordance with another embodiment of the present disclosure;

FIG. 7A illustrates a display module in accordance with one embodiment of the present disclosure;

FIG. 7B illustrates a lighting module in accordance with one embodiment of the present disclosure; and

FIG. 8 illustrates a lighting apparatus in accordance with one embodiment of the present disclosure.

DETAILED DESCRIPTION OF THE DISCLOSURE

FIGS. 1A-1E illustrate the method for forming a light emitting device (LED) according to one embodiment of the present disclosure. Referring to FIG. 1A, at first, a light emitting structure 100 is formed. The light emitting structure 100 includes a substrate 11, a first conductive layer 102 as a cladding layer, an active layer 104 disposed on the layer 102 as a light emitting layer, and a second conductive layer 106 disposed on the active layer 104 as another cladding layer. Preferably, as shown in FIG. 1A, an electrode or bonding pad 107 a is disposed on an exposed portion of the layer 102, and another electrode or bonding pad 107 b is disposed on the layer 106. The manufacture method and the material (e.g., Aluminum) of electrode or bonding pads 107 a and 107 b are well known to those skilled in the art and thus are omitted hereinafter. Furthermore, in one embodiment, the light emitting structure 100 includes a passivation layer 120 to protect the light emitting structure 100. Also, the manufacture method and the material (e.g., SiO₂) of passivation layer 120 are well known to those skilled in the art and thus omitted hereinafter.
The first conductive layer 102 and the second conductive layer 106 can be embodied as any semiconductor materials known to those skilled in the art, preferably as III-V group compound semiconductor, such as Al_xGa_yIn_1-x-yN or Al_xGa_yIn_1-x-yP, wherein 0≦x≦1, 0≦y≦1, 0≦x+y≦1, and can be doped with P/N type dopants. Light emitting layer 104 can embodied with conventional materials (e.g., Al_xGa_yIn_1-x-yN or Al_xGa_yIn_1-x-yP) and structures (e.g., Single Quantum Well, Multiple Quantum Well, and Double Heterostructure). The principles and mechanisms of the active layer 104 are known to those skilled in the art and thus omitted hereinafter. In addition, the light emitting structure 100 can be manufactured via the process of MOCVD (Metallic-Organic Chemical Vapor Deposition), molecular beam epitaxy (MBE), or hydride vapor phase epitaxy (HVPE).
Next, as shown in FIG. 1B, a first dielectric layer 122 is formed on the light emitting structure 100. Preferably, the first dielectric layer 122 is a transparent dielectric layer with the thickness D≦20 μm, and the heat generated from the light emitting structure 100 can be easily conducted. The first dielectric layer 122 can be formed with the material such as SiO₂, Si₃N₄, or the combination of them, and via the processes of E-gun or sputter.
As shown in FIG. 1C, a second dielectric layer 140 is formed on the first dielectric layer 122. The layer 140 can be formed with the material such as SiO₂, silicon nitride, polyimide, bisbenzocyclobutene, or photoresist. Preferably, the thickness of the second dielectric layer 140 is about 25 μm, and the second dielectric layer 140 is formed by using a printing technology.
As shown in FIG. 1D, the metal layer 160 is formed after the second dielectric layer 140 is formed. The metal layer 160 is disposed on the light emitting structure 100 and is electrically connected to the first conductive layer 102. A portion of the metal layer 160 is disposed on the first dielectric layer 122. Also the metal layer 162 is formed on the on the light emitting structure 100 and is electrically connected to the second conductive layer 106. A portion of the metal layer 162 is disposed on the first dielectric layer 122. Meanwhile, the first dielectric layer 122 and the second dielectric layer 140 electrically isolate the metal layer 160 from the metal layer 162. The metal layer 160 or the metal layer 162 can be embodied with materials of Au, Al, Ag, or Alloy of them. Preferably, the metal layer 160 and the metal layer 162 are formed together by using a printing technology or electroplated. After that, the manufacture process for LED 10 is completed.
In one embodiment, the dielectric layer 122 is a transparent dielectric layer. A surface of the dielectric layer 122 contacting the metal layer 160 and/or the metal layer 162 is provided for reflecting the light emitted from the light emitting structure 100. Furthermore, the metal layer 160 and/or the metal layer 162 are thermal conductive paths for the light emitting structure 100. Large contact areas A1 and A2 of the metal layer 160 and the metal layer 162 are also beneficial to the heat dissipation.
Referring to FIG. 1E, after the formation of the structure shown in FIG. 1D, the method further includes a step of removing the substrate 11 to expose the first conductive layer 102. For example, the substrate 11 can be a sapphire substrate or a GaAs substrate. When the substrate 11 is a sapphire substrate, the substrate 11 can be removed by an Excimer laser process. The Excimer laser process can be a KrF Excimer laser with an energy density of 400 mJ/cm², a wavelength of 248 nm, and a pulse width of 38 ns. As the Excimer laser radiates on the sapphire substrate at an elevated temperature, such as 60° C., the sapphire substrate is removed to expose the first conductive layer 102. Alternatively, when the substrate 11 is GaAs substrate, a solution of NH₄OH:35H₂O₂or a solution of 5H₃PO4:3H₂O₂:3H₂O can be applied to remove the GaAs substrate to expose the first conductive layer 102.
After the substrate 11 is removed, the method further includes a step of roughening the surface 102 a of the first conductive layer 102. For example, the first conductive layer 102 can be an Al_xGa_yIn_1-x-yN layer, and the surface 102 a is roughened by using an etch solution, such as a KOH solution. Alternatively, when the first conductive layer 102 is an Al_xGa_yIn_1-x-yP layer, a solution of HCl and H₃PO₄can be employed for 15 seconds to roughen the surface 102 a of the first conductive layer 102. The rough surface 102 a of the first conductive layer 102 is implemented to reduce the possibility of total reflection of light so as to increase a light extraction efficiency of the light emitting device.
Like the LED 10 in FIG. 1E, LED 10 a, 10 b, and 10 c shown in FIG. 1F, are provided with large contact areas, each of which is preferably larger than half of the section area of the LED 10. LED 10 a, 10 b, and 10 c are directly connected to the circuit board 13 by using solder 12 instead of by “Die Bonding” and “Wire Bonding”. In another embodiment, LED 10 a is provided for emitting red light, LED 10 b for emitting green light, and LED 10 c for emitting blue light. Thus LED 10 a, 10 b, and 10 c, when connected to the circuit boards 13, can be used in the image display application.
FIGS. 2A-2D illustrate the method for forming a light emitting device (LED) array according to one embodiment of the present disclosure. Referring to FIG. 2A, at first, a substrate 21 is provided. The substrate 21 can be a sapphire substrate, an GaAs substrate, or other substrates known to those skilled in the art, or the combinations of them. Next, a plurality of light emitting structures 200 a, 200 b, and 200 c are formed on the substrate 21. The materials and the manufacture processes for light emitting structures 200 a, 200 b, and 200 c can be referred to the elaboration for light emitting structure 100 illustrated by FIGS. 1A-1D. Similarly, light emitting structures 200 a, 200 b, and 200 c can be manufactured via the process of MOCVD, molecular beam epitaxy, or hydride vapor phase epitaxy.
Next, as shown in FIG. 2B, a dielectric layer 222 a is formed on the light emitting structure 200 a, a dielectric layer 222 b is formed on the light emitting structure 200 b, and a dielectric layer 222 c is formed on the light emitting structure 200 c. Preferably, like the layer 122 shown in FIG. 1B, dielectric layers 222 a, 222 b, and 222 c are transparent dielectric layers with each thickness D≦20 μm, and the heat generated from the light emitting structures 200 a, 200 b, and 200 c can be easily conducted away. The dielectric layers 222 a, 222 b, and 222 c can be formed with the material such as SiO₂, Si₃N₄, or the combination of them, and via the processes of E-gun or sputter.
Later, as shown in FIG. 2C, a dielectric layer 240 a is formed on the dielectric layer 222 a, a dielectric layer 240 b is formed on the dielectric layer 222 b, and a dielectric layer 240 c is formed on the dielectric layer 222 c. The dielectric layers 240 a, 240 b, and 240 c can be formed with the material such as SiO2, silicon nitride, polyimide, bisbenzocyclobutene, or photoresist. Preferably, like the dielectric layer 140 in FIG. 1C, the thickness of the dielectric layers 240 a, 240 b, and 240 c are about 25 μm respectively, and the dielectric layers 240 a, 240 b, and 240 c are formed by using a printing technology. In one embodiment, a dielectric layer 280 is further formed between light emitting structures 200 a, 200 b, and 200 c to electrically isolate LEDs 20 a, 20 b, and 20 c from each other (as shown in FIG. 2D). The dielectric layer 280 can be embodied with the same material (e.g., polyimide) with the layer 240 a, 240 b, or 240 c, and is formed together with them by using a printing technology. Alternatively, the dielectric layer 280 can be embodied with the different material with the layer 240 a, 240 b, or 240 c, and is formed through a different process.
As shown in FIG. 2D, metal layers 260 a, 260 b, 260 c, 262 a, 262 b, and 262 c are formed and can be embodied with materials of Au, Al, Ag, or Alloy of them. Preferably, metal layers 260 a, 260 b, 260 c, 262 a, 262 b, and 262 c are formed together by using a printing technology or electroplated. After that, the manufacture process for LED array 20 including LEDs 20 a, 20 b, and 20 c is completed.
As shown in FIG. 2E, LEDs 20 a, 20 b, and 20 c are provided with large contact areas, and are directly connected to the circuit board 23 by using solders 22. Thereafter, LED array 20 can be separated from the substrate 21 and used in the image display application. For example, after the LEDs 20 a, 20 b, and 20 c are connected to the circuit board 23 by using solder 22, the method further includes a step of removing the substrate 21. For example, the substrate 21 can be a sapphire substrate, and be removed by an Excimer laser process. The Excimer laser process can be a KrF Excimer laser with an energy density of 400 mJ/cm², a wavelength of 248 nm, and a pulse width of 38 ns. As the Excimer laser radiates on the sapphire substrate 21 at an elevated temperature, such as 60° C., the sapphire substrate 21 is removed to expose the first conductive layer 102. Alternatively, when the substrate 11 is an GaAs substrate, a solution of NH₄OH:35H₂O₂or a solution of 5H₃PO4:3H₂O₂:3H₂O can be applied to remove the GaAs substrate to expose the first conductive layer 102.
After the substrate 21 is removed, the method further includes a step of roughening the surface 102 a of the first conductive layer 102. For example, the first conductive layer 102 can be an Al_xGa_yIn_1-x-yN layer, and the surface 102 a is roughened by using an etch solution, such as a KOH solution. Alternatively, when the first conductive layer 102 is an Al_xGa_yIn_1-x-yP layer, a solution of HCl and H₃PO₄can be employed for 15 seconds to roughen the surface 102 a of the first conductive layer 102. The rough surface 102 a of the first conductive layer 102 is implemented to reduce the possibility of total reflection of light so as to increase a light extraction efficiency of the light emitting device. In another embodiment, as shown in FIG. 2G, a transparent encapsulating material 24, such as epoxy or any conventional material as appropriate, is applied to enclose the LED array 20 (including LEDs 20 a, 20 b, and 20 c) and connected to the circuit board 23 so that an LED device package 25 is formed.
FIGS. 3A-3G illustrate the method for forming a light emitting device (LED) array according to another embodiment of the present disclosure. Referring to FIG. 3A, a substrate 21 is provided. The substrate 21 is single crystalline and comprises sapphire, GaAs, GaN, or Si. Next, a first conductive layer 102 as a first cladding layer is epitaxially grown on the substrate 21, an active layer 104 comprising multiple quantum well (MQW) Structure as a light emitting layer is epitaxially grown on the first conductive layer 102, and a second conductive layer 106 as a second cladding layer is epitaxially grown on the active layer 104. Next, the first conductive layer 102, the active layer 104, and the second conductive layer 106 are etched to form a plurality of light-emitting stacks 101 separated from each other on the substrate 21, and a portion of the first conductive layer 102 of each light-emitting stack 101 is exposed. Next, a first dielectric layer 120 is formed on each light-emitting stack 101 and covers a portion of the first conductive layer 102, a portion of the second conductive layer 106, and a sidewall of the light-emitting stack 101. Next, a first electrode or bonding pad 107 a is disposed on an exposed portion of each first conductive layer 102 and electrically connected to the first conductive layer 102, and a second electrode or bonding pad 107 b is disposed on each second conductive layer 106 and electrically connected to the second conductive layer 106.
Then, as shown in FIG. 3B, a reflecting layer 221 is formed on each first dielectric layer 120, and a second dielectric layer 122 is formed on each first dielectric layer 120 and covers the reflecting layer 221. The reflecting layer 221 has a reflectivity equal to or greater than 80% for light emitted by the corresponding light-emitting stack 101. The material of the reflecting layer 221 comprises metal such as Ag, Ag alloy, Al, or Al alloy. In one embodiment, the material of the reflecting layer 221 comprises polymer mixed with inorganic particles made of metal oxide or a material having a refractive index equal to or greater than 1.8, such as an epoxy resin mixed with titanium oxide particles. The reflecting layer 221 is fully enclosed by the corresponding first dielectric layer 120 and second dielectric layer 122 such that the reflecting layer 221 is electrically insulated with the corresponding light-emitting stack 101. In another embodiment, the first dielectric layer 120 is omitted, and the reflecting layer 221 is directly formed on the second conductive layer 106 and electrically connected to the second conductive layer 106. Next, as shown in FIG. 3C, a third dielectric layer 240 is formed on each light-emitting stack 101 and the substrate 21 and exposes the corresponding first electrode or bonding pad 107 a and second electrode or boding pad 107 b. Next, a first metal layer 260 and a second metal layer 262 are formed within each third dielectric layer 240 and on a portion of the corresponding second dielectric layer 122. The first metal layer 260 and the second metal layer 262 are respectively formed on the corresponding first electrode or bonding pad 107 a and second electrode or bonding pad 107 b. The material of the first metal layer 260 and the second metal layer 262 comprise Au, Al, Ag, or the Alloy thereof. In one embodiment, the first metal layer 260 and the second metal layer 262 are formed together by using a printing process or an electroplating process.
As shown in FIG. 3D, the third dielectric layer 240 between adjacent two light-emitting stacks 101 is further patterned to form cavities in the third dielectric layer 240 and expose a portion of the substrate 21, then an opaque layer 290 is formed in the cavities. In one embodiment, the opaque layer 290 serves as an reflecting layer or a light-absorption layer for reflecting or absorbing light emitted from the corresponding light-emitting stack 101 and preventing the light emitted from neighboring light-emitting stacks 101 from mutually interfering or crosstalk. The opaque layer 290 has a transmittance less than 50% for light emitted by the corresponding light-emitting stack 101. The material of the opaque layer 290 comprises metal or comprises a polymer mixing with inorganic particles made of metal oxide or a material having a refractive index equal to or greater than 1.8, such as an epoxy resin mixing with titanium oxide particles. Therefore, the manufacture process for LED array 30 comprising a plurality of LEDs 300 is completed.
As shown in FIG. 3E, a circuit board 23 is provided. the circuit board 23 comprises a plurality of metal contacts 22 formed on the upper surface and the lower surface of the circuit board 23 and a plurality of conductive channels 22 a penetrating the circuit board 23 for connecting the metal contacts 22 on the upper surface of the circuit board 23 to the metal contacts 22 on the lower surface of the circuit board 23. The metal contacts 22 comprises metal or metal alloy. In one embodiment, the metal contacts 22 comprise solders. The circuit board 23 comprises FR-4, BT (Bismaleimide-Triazine) resin, ceramics or glass. The thickness of the circuit board 23 is between 50 and 200 microns for being sufficient to support LEDs but still compact. The LED array 30 is directly flip-bonded to the circuit board 23 by aligning the first metal layer 260 and the second metal layer 262 of each LEDs 300 to the corresponding metal contacts 22. It is noted that air gaps may be formed between the LED array 30 and the circuit board 23 other than the area of metal contacts 22. Optionally, an underfill material is filled in the gaps to enhance the bonding strength and mechanical support. Thereafter, the substrate 21 of the LED array 30 is removed after the LED array 30 is connected to the circuit board 23. In one embodiment, the substrate 21 comprises sapphire, the light-emitting stacks 101 comprises gallium nitride, and the method for removing the substrate 21 comprises using an Excimer laser to radiate at the interface of the first conductive layers 102 and the substrate 21 under an environment with an elevated temperature, such as 60° C., and then the substrate 21 is separated from the first conductive layers 102. The Excimer laser comprises a KrF Excimer laser with an energy density of 400 mJ/cm², a wavelength of 248 nm, and a pulse width of 38 ns. In another embodiment, the substrate 21 comprises GaAs, and the method for removing the substrate 21 comprises using a chemical solution with a composition of NH₄OH:35H₂O₂or a composition of 5H₃PO4:3H₂O₂:3H₂O to totally etch away the substrate 21 and then expose the first conductive layers 102, the third dielectric layers 240 a, and the opaque layers 290.
As shown in FIG. 3F, after the substrate 21 is removed, the method further includes a step of roughening the exposed surface of the first conductive layers 102. In one embodiment, the first conductive layers 102 comprises Al_xGa_yIn_1-x-yN (0≦x, y≦0), and the exposed surfaces of the first conductive layers 102 are etched by KOH solution to form a rough surface 102 a on each first conductive layer 102. In another embodiment, the first conductive layers 102 comprise Al_xGa_yIn_1-x-yP (0≦x, y≦0), and the exposed surfaces of the first conductive layers 102 are etched by HCl solution or H₃PO₄solution for 15 seconds to form a rough surface 102 a on each first conductive layer 102. The rough surface 102 a of each first conductive layer 102 can reduce the possibility of total internal reflection (TIR) of light within each LED 300 so as to increase light extraction efficiency of the device. After the roughening step, a plurality of concave regions is formed on the rough surface 102 a and substantially enclosed by the third dielectric layers 240 a. In one embodiment, in order to form a chip-scale RGB LED unit for a display, the method further includes selectively coating a first wavelength converting layer 294 on the LED 300 b, as shown in FIG. 3F, for converting light. For example, blue light having a dominant wavelength between 430 nm and 470 nm, emitted by the light-emitting stack 101 of LED 300 b is converted into a first converted light, e.g. red light having a dominant wavelength between 610 nm and 690 nm. A second wavelength converting layer 296 is further selectively coated on the LED 300 c for converting the light emitted from the light-emitting stack 101 of LED 300 c into a second converted light, e.g. a green light having a dominant wavelength between 500 nm and 570 nm. LED 300 a is not coated with any wavelength converting material for emitting the blue light directly exited from the rough surface 102 a of LED 300 a. In one embodiment, the first or second wavelength converting layers are in a form of a sheet by aggregating nano-scale quantum dots or nano-scale phosphor particles with a substantially uniform thickness and bonded to the light-emitting stack 101 by an adhesive bonding layer (not shown). In another embodiment, the first or second wavelength converting layers comprise nano-scale quantum dots or nano-scale phosphor particles having an average diameter or average feature length between 10 nm and 500 nm. The diameter or feature length of each of the nano-scale quantum dots or nano-scale phosphor particles is substantially lower than 1000 nm. The nano-scale quantum dots comprises semiconductor material, such as a group II-VI material having a composition of Zn_xCd_yMg_1-x-ySe where x and y are determined such that the group II-VI material is capable of being optically pumped to emit green or red light. The term “feature length” is defined as the maximum length between any two points of a phosphor particle or quantum dot. After that, a transparent encapsulating material 24, such as epoxy or silicone is applied to the upper surface of the LED array 32 for affixing the wavelength converting materials to the light-emitting stacks 101 and serving as optical lens for LEDs 300 a, 300 b, 300 c of the LED array 32. In another embodiment, the material of the wavelength converting layer that covers on the LEDs is the same,
FIG. 4A shows a top view of the LED array 32 flip-bonded to the circuit board 23 of FIG. 3F. The LED array 32 and the circuit board 23 are both in a wafer form with same or similar dimension. The LED array 32 comprises a plurality of RGB LED groups arranged alternately and sequentially in two dimensions, and each group comprises one LED 300 a, one LED 300 b, and one LED 300 c as enclosed by the dashed lines.
Finally, a dicing step is performed to cut the LED array 32 and the circuit board 23 simultaneously to form a plurality of chip-scale RGB LED units 35 as shown in FIG. 3G, each having one RGB LED group including an LED 300 a (blue LED) configured to emitting blue light, an LED 300 b (red LED) configured to emitting red light, and an LED 300 c (green LED) configured to emitting green light. The chip-scale RGB LED unit 35 is a package-free and surface-mounted device and is configured to be directly mounted to a printed circuited board (PCB) without further traditional packaging process after the dicing step. The transparent encapsulating material 24 commonly covers LEDs 300 a, 300 b, and 300 c without extending to sidewalls of the light-emitting stacks of LEDs 300 a, 300 b, and 300 c. In one embodiment, the dicing step is performed to cut the LED array 32 and the circuit board 23 simultaneously to form a plurality of chip-scale RGB LED units wherein each of the chip-scale RGB LED units comprises a plurality of RGB LED groups. The plurality of RGB LED groups in one chip-scale RGB LED unit is arranged as I×J array, wherein I and J are positive integers, and at least one of I and J are greater than 1. The I/J ratio is preferably approximately or equal to 1/1, 3/2, 4/3, or 16/9.
FIG. 4B shows the top view of the chip-scale RGB LED unit 35 comprising one RGB LED group in FIG. 3G. The chip-scale RGB LED unit 35 is shaped as a first rectangle with a first short side and a first lengthy side, wherein the first short side has a first width S1 and the first lengthy side has a first length S2 greater than the first width S1. Each of light-emitting stacks 101 is shaped as a second rectangle with a second short side and a second lengthy side, wherein the second short side has a second width d1 and the second lengthy side has a second length d2 greater than the second width d1. Light-emitting stacks 101 are arranged such that the second short sides of the light-emitting stacks 101 are substantially parallel to the first lengthy side of the chip-scale RGB LED unit 35 or substantially perpendicular to the first short side of the chip-scale RGB LED unit 35. In one embodiment, the chip-scale RGB LED unit 35 is used as one pixel of an indoor display panel. To realize full LED pixel in an TV display with 40 inches diagonal and 1024*768 pixel resolution, the area of each pixel has to be smaller than approximately 0.64 mm². Therefore, the area of the chip-scale RGB LED unit 35 is exemplarily smaller than 0.36 mm². The first length S2 and the first width S1 are both smaller than 0.6 mm, and the aspect ratio of the chip-scale RGB LED unit 35, i.e. S2/S1 is preferably smaller than 2/1. The distance between two metal layers 260 and 262, i.e. the first distance S3, is limited to the alignment control of the bonding process between the LED array and the circuit board in accordance with the present disclosure. The first distance S3 is equal to or greater than 25 microns to ensure the process tolerance, but smaller than 150 microns to provide sufficient contact area for electrical conduction. The distance between one edge of the chip-scale RGB LED unit 35 and one light-emitting stack 101 therein, i.e. the second distance S4, is limited to the tolerance of the dicing step. The second distance S4 is equal to or greater than 25 microns to ensure the dicing tolerance, but smaller than 60 microns to maintain the compactness. The distance between two adjacent LEDs, i.e. the third distance S5 is limited by the lithography-etching process, and is smaller than 50 microns, or preferably smaller than 25 microns for reserving more area for the light-emitting stacks. For each light-emitting stacks 101 in a chip-scale RGB LED unit 35, the second width d1 is between 20 and 150 microns and the second length d2 is between 20 and 550 microns. The ratio of the area of the chip-scale RGB LED unit 35 to the total area of the light-emitting stacks 101 is smaller than 2 or between 1.1 and 2, and preferably between 1.2 and 1.8. The area of the light-emitting stack depends on the required illumination and the pixel dimension. It is noted that the shape of the chip-scale RGB LED unit 35 can also be square with four sides being the same as the first width S1. Similarly, the shape of the light-emitting stack 101 can also be square with four sides being the same as the second width d1. In one embodiment, there are two chip-scale RGB LED units 35 included in one pixel, and one is for normal operation and the other one is for redundancy in case the one for normal operation is malfunctioned. The first width Si is preferably less than 0.3 mm in order to arrange two chip-scale RGB LED units 35 included in one pixel. By taking the advantage of the present disclosure, using LEDs as pixel elements in a flat TV can be realized, and the resolution can be further enhanced to double or quadruple of 1024*768 pixel resolution. In another embodiment, one chip-scale RGB LED unit comprises two RGB LED groups; one for normal operation and the other one for redundancy in case the one for nom al operation is malfunctioned.
FIGS. 5A-5C show one embodiment for a chip-scale LED unit in accordance with the present disclosure. The manufacturing method and structure is similar to the foregoing embodiment depicted in FIGS. 3A-3G and the related disclosure. The difference is that, before the dicing step, the LED array 34 comprises a plurality of identical LEDs 300 d as shown in FIG. 5A. Each LED 300 d is coated with an identical or different wavelength converting layer 298 for converting light emitted by the light-emitting stack 101 of the corresponding LED 300 d, e.g. blue light having a dominant wavelength between 430 nm and 470 nm, into a converted light, e.g. yellow light, green light, or red light. FIG. 5B and FIG. 5C show the top view and the cross-sectional view of the chip-scale LED unit 36 containing one singular LED after the dicing step. The dimension of the chip-scale LED unit 36 is the same or similar to that in FIG. 4B. The chip-scale RGB LED unit 36 is shaped as a first rectangle with a first short side and a first lengthy side, wherein the first lengthy side has a first length Si and the first short side has a first width S6 smaller than the first length S1. Each of light-emitting stacks 101 is shaped as a second rectangle with a second short side and a second lengthy side, wherein the second short side has a second width d1 and the second lengthy side has a second length d2 greater than the second width d1. Light-emitting stacks 101 is arranged such that the second short side of the light-emitting stack 101 is substantially parallel to the first short side of the chip-scale RGB LED unit 36 or substantially perpendicular to the first lengthy side of the chip-scale RGB LED unit 36. In one embodiment, the chip-scale RGB LED unit 36 is used as part of a pixel of an indoor display panel. The area of the chip-scale RGB LED unit 36 is exemplarily smaller than 0.12 mm². The first length S1 and the first width S6 are both smaller than 0.2 mm, and the aspect ratio of the chip-scale RGB LED unit 36, i.e. S1/S6 is preferably smaller than 2/1. The distance between two metal layers 260 and 262, i.e. the first distance S3, is limited to the alignment control of the bonding process between the LED array and the circuit board in accordance with the present disclosure. The first distance S3 is equal to or greater than 25 microns to ensure the process tolerance, but smaller than 150 microns to provide sufficient contact area for electrical conduction. The distance between one edge of the chip-scale RGB LED unit 36 and the light-emitting stack 101 therein, i.e. the second distance S4, is limited to the tolerance of the dicing step. The second distance S4 is equal to or greater than 25 microns to ensure the dicing tolerance, but smaller than 60 microns to maintain the compactness. For the light-emitting stack 101 in a chip-scale RGB LED units 36, the second width d1 is between 20 and 150 microns and the second length d2 is between 20 and 550 microns. The ratio of the area of the chip-scale RGB LED unit 35 to the total area of the light-emitting stack 101 is smaller than 2 or between 1.1 and 2, and preferably between 1.2 and 1.8. The area of the light-emitting stack depends on the required illumination and the pixel dimension. It is noted that the shape of the chip-scale RGB LED units 36 can also be square with four sides being the same as the first width S6. Similarly, the shape of the light-emitting stack 101 can also be square with four sides being the same as the second width d1. In one embodiment, there are at least three chip-scale RGB LED units 36 included in one pixel for illuminating blue, red, and green light.
FIGS. 5D-5E show another embodiment for a chip-scale LED unit in accordance with another embodiment of the present disclosure. The manufacturing method and structure is similar to the foregoing embodiment depicted in FIGS. 5A-5C and the related disclosure. The difference is that the opaque layer 290 is optionally omitted. The chip-scale LED unit 36′ is directly surface-mounted to a lighting board contained in a lighting fixture. The area of the light-emitting stack 101 depends on the required illumination and the dimension of the lighting board or the lighting fixture. The area of the light-emitting stack 101 within the chip-scale LED unit 36′ is exemplarily from 100 mil²to 200 mil²for low power application (e.g. below 0.3 watts), 201 mil²to 900 mil²for middle power application (e.g. 0.3˜0.9 watts), or greater than 900 mil²for high power application (e.g. above 0.9 watts). The dielectric layer 240 a surrounding the light-emitting stack 101 is served as a coupling lens for extracting light out of the chip-scale LED unit 36′. The ratio of the area of the chip-scale LED unit 36′ to the area of the light-emitting stack 101 is equal to or greater than 9, and preferably equal to or greater than 15 for better light extraction efficiency and light dispersion. The distance between two metal layers 260 and 262, i.e. the first distance S3′, is limited to the alignment control of the bonding process between the LED array and the circuit board in accordance with the present disclosure. The first distance S3′ is equal to or greater than 25 microns to ensure the process tolerance, but smaller than 150 microns to provide sufficient contact area for electrically conducting. It is noted that the shape of the chip-scale LED unit 36 can also be square with the length of four sides being the same as the first width S6′. Similarly, the shape of the light-emitting stack 101 can also be square with the length of four sides being the same as the second width d1′. The first width S6′ is equal to or greater than 3 times of the second width d1′, or preferably equal to or greater than 4 times of the second width d1′ for better light extraction efficiency. In one embodiment, the dielectric layer 240 a has asymmetric thickness on the sidewalls of the light-emitting stack 101 such that a first ratio of the first width S6′ to the second width d1′ is different from a second ratio of the first length Si′ to the second length d2′ to achieve an asymmetric light field under operation when viewing from the top of the chip-scale LED unit 36′. The first ratio is at least 2 times of the second ratio, or preferably 4 times of the second ratio.
FIG. 6A shows another embodiment of a chip-scale RGB LED unit in accordance with another embodiment of the present disclosure. The manufacturing method and structure of the chip-scale RGB LED unit 65 is similar to the foregoing embodiment depicted in FIGS. 3A-3G and the related disclosure. The difference is that an underfill material 680 is filled in the gaps between the LED array 32′ and the circuit board 23 to enhance the bonding strength and provide a current path between the circuit board 23 and the LEDs. The underfill material 680 comprises an anisotropic conductive film (ACF) that is capable of conducting current in a vertical path between the LED array 32′ and the circuit board 23, and isolating current in a lateral path parallel to the LED array 32′ or the circuit board 23. The underfill material 680 is coated on the circuit board 23 prior to bonding the LED array to the circuit board 23. In one embodiment, the first metal layers 260′ and the second metal layers 262′ does not contact with the metal contacts 22 of the circuit board 23. The underfill material 680 is between the metal layers and the metal contacts for conducting current between the metal layers and the metal contacts. The metal layers are patterned to form a plurality of depressions and protrusions on the surface facing the metal contacts. Therefore, the contacting surface area is increased and the bonding strength is enhanced. The plurality of depressions and protrusions is in a regular form or an irregular form, and the surface roughness Ra is between 0.5 and 5 microns. The benefit for using ACF as an underfill material is that the distance between two metal layers 260′ and 262′, i.e. the first distance S3 as shown in FIG. 4B, can be lower than 25 microns.
Similarly, FIG. 6B is a singular form of FIG. 6A. The underfill material 680 and the patterned surface of the metal layers 260′ and 262′ can be also applied to the foregoing embodiment depicted in FIGS. 5A-5C to form the structure demonstrated in FIG. 6B. The underfill material 680 is filled in the gaps between the LED 300 d and the circuit board 23 to enhance the bonding strength and provide a current path. The underfill material 680 comprises an anisotropic conductive film (ACF) that is capable of conducting current in a vertical path between the LED 300 d and the circuit board 23, and isolating current in a lateral path parallel to the LED 300 d or the circuit board 23. The underfill material 680 is coated on the circuit board 23 prior to bonding the LED array to the circuit board 23. In one embodiment, the first metal layers 260′ and the second metal layers 262′ does not contact with the metal contacts 22 of the circuit board 23. The underfill material 680 is between the metal layers 260′ and 262′ and the metal contacts 22 for conducting current between the metal layers 260′ and 262′ and the metal contacts 22. The metal layers 260′ and 262′ are patterned to form a plurality of depressions and protrusions on the surface facing the metal contacts 22. Therefore, the contact surface area is increased and the bonding strength is enhanced. The plurality of depressions and protrusions is in a regular form or an irregular form, and the surface roughness Ra is between 0.5 and 5 microns. Similarly, the underfill material 680 and the patterned surface of the metal layers 260′ and 262′ can also be applied to the foregoing embodiment of FIG. 5E to form the structure demonstrated in FIG. 6C.
FIG. 7A shows a display module 76 comprises a plurality of chip-scale RGB LED units 65 on a second circuit board 73. The configuration of the chip-scale RGB LED units 65 depends on the pixel size of the display module. For example, any two adjacent chip-scale RGB LED units 65 are either separated from each other by a distance or disposed seamlessly such that they are contacted with each other. The second circuit board comprises circuitry 72 electrically connected to LEDs of the chip-scale RGB LED units 65 for independently controlling the blue, red, and green LEDs in each chip-scale RGB LED unit 65. In one embodiment, the display module 76 contains M rows and N columns of the chip-scale RGB LED units 65 for being used in an display with X*Y pixel resolution, wherein M/N=1/1, 3/2, 4/3 or 16/9; X=a*M, Y=b*N, and a, and b are all positive integers equal to or greater than 2. The display module 76 contains over 500 units of chip-scale RGB LED units 65 in an area of one inch square. Namely, the display module 76 contains over 1500 light-emitting stacks 101 in an area of one inch square. In another embodiment, each chip-scale RGB LED unit comprises a plurality of RGB LED groups and each group comprises one blue LED, one red LED, and one green LED as depicted in the foregoing embodiments. The plurality of RGB LED groups in one chip-scale RGB LED unit is arranged as I×J array, wherein I and J are positive integers, and at least one of I and J are greater than 1. The ratio of I/J is preferably approximately or equal to 1/1, 3/2, 4/3 or 16/9. The distance between two adjacent light-emitting stacks from two adjacent RGB LED groups in one chip-scale RGB LED unit is substantially the same as the distance between two adjacent light-emitting stacks from two adjacent chip-scale RGB LED units. The display module 76 contains M rows and N columns of the chip-scale RGB LED units 65 for being used in an display with X*Y pixel resolution, wherein M/N=1/1, 3/2, 4/3 or 16/9; X=a*M*I, Y=b*N*J, and a, and b are all positive integers equal to or greater than 2. The display module 76 contains over 500 units of RGB LED groups in an area of one inch square. Namely, the display module 76 contains over 1500 light-emitting stacks 101 in an area of one inch square. Each of the LEDs in the RGB LED units and each of the RGB LED units can be independently driven by the circuitry formed on the circuit board 23 and the second circuit board 73. The material of the second circuit board 73 comprises FR-4, BT (Bismaleimide-Triazine) resin, ceramics or glass.
FIG. 7B shows a lighting module 78 comprising a plurality of chip-scale LED units 66 on a second circuit board 73. The chip-scale LED units 66 can be connected in series or parallel by the circuit of the second circuit board 73 depending on the driving voltage to be applied. In one embodiment, the lighting module 78 is installed into a lighting bulb 80 as shown in FIG. 8. The lighting bulb further comprises an optical lens 82 covering the lighting module 78, a heat sink 85 having a mounting surface where the lighting module 78 formed thereon, a frame 87 connected to heat sink, and an electrical connector 88 connected to the frame and electrically connected to lighting module.
Although specific embodiments have been illustrated and described, it will be obvious to those skilled in the art that various modifications may be made without departing from what is intended to be limited solely by the appended claims.

Claims

What is claimed is:

1. A method for forming a light-emitting apparatus, comprising:

providing a carrier having a plurality of first metal contacts;

forming a light-emitting structure comprising a substrate, a first cladding layer on the substrate, an active layer on the first cladding layer, and a second cladding layer on the active layer;

bonding the light-emitting structure to the carrier;

forming a cap layer on a side of the light-emitting structure opposite to the carrier; and

cutting the carrier and the cap layer to form a chip-scale LED unit.

2. The method of claim 1, further comprising removing the substrate before forming the cap layer.

3. The method of claim 2, wherein the step of removing the substrate is by an Excimer laser process.

4. The method of claim 1, further comprising forming a plurality of metal layers electrically connected to the light-emitting structure for bonding to the first metal contacts.

5. The method of claim 4, further comprising patterning the metal layers to form an uneven surface on the metal layers.

6. The method of claim 1, further comprising removing the substrate to expose an exposed surface of the light-emitting structure after bonding the light-emitting structure to the carrier, and comprising roughening the exposed surface of the light-emitting structure.

7. The method of claim 6, wherein after the step of roughening the exposed surface of the light-emitting structure, the exposed surface has a surface roughness Ra between 0.5 and 5 microns.

8. The method of claim 6, wherein the step of removing the substrate is by an Excimer laser process.

9. The method of claim 1, further comprising forming an underfill material substantially over the surface of the carrier before bonding the light-emitting structure to the first metal contacts.

10. The method of claim 9, wherein the underfill material is electrically conductive, and the light-emitting structure is electrically connected with the carrier through the underfill material.

11. The method of claim 1, wherein the step of forming the cap layer comprises forming a wavelength converting material on the light-emitting structure and configured to convert the light emitted from the first light-emitting structure to a first light.

12. The method of claim 11, wherein the wavelength converting material is only formed on an upper side of the light-emitting structure without extending to a sidewall of the light-emitting structure.

13. The method of claim 1, wherein the step of bonding the light-emitting structure to the carrier is by solder bonding.

14. The method of claim 1, further comprising forming a reflecting layer between the light-emitting structure and the carrier.

15. The method of claim 1, wherein the first metal contacts extend from a top surface of the carrier to a bottom surface of the carrier.

16. The method of claim 1, wherein the carrier comprises a lower surface, metal contacts on the lower surface, and a plurality of conductive channels penetrating the carrier for connecting the first metal contacts on an upper surface of the carrier to the metal contacts on the lower surface of the carrier.

17. The method of claim 1, wherein the cap layer comprises epoxy or silicone.