JP2006140398A - Element transfer method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an element transfer method which can easily realize element transfer while eliminating the need for bonding with resin upon transfer. <P>SOLUTION: In the element transfer method, LED elements 20 formed on the sapphire substrate 14 of an element formation wafer 13 are transferred onto a transfer substrate 11. A magnetic metal pattern 12 is formed on the transfer substrate 11 and magnetized, a magnetic material thin film (Ni thin film 23) is formed on a P bump electrode 22 formed on the sapphire substrate 14, and the LED elements 20 are held to the transfer substrate 11 by a magnetic coupling force between the magnetic metal pattern 12 on the transfer substrate 11 and the Ni thin film 23 formed on the LED element 20. Under this condition, the sapphire substrate 14 is released and the LED element 20 is transferred onto the transfer substrate 11. Then, the magnetic metal pattern 12 on the transfer substrate 11 is selectively demagnetized, the magnetic coupling force is selectively released, and the LED elements 20 at the corresponding locations are selectively taken out. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an element transfer method for transferring and arranging elements such as light emitting diodes (LEDs), and more particularly to a novel element transfer method using a magnetic coupling force.

  For example, in the manufacture of a light-emitting diode display (LED display) in which light-emitting diodes are arranged in a matrix, a method is adopted in which LED elements are individually taken out after dicing and are arranged and mounted on a circuit board. In this case, since the LED which is a light emitting element is expensive, it is considered that an LED display can be manufactured at low cost by forming as many LED elements as possible on one wafer. For example, if the size of the LED element is about tens of μm square, which is about 300 μm square in the past, the number of LED elements that can be made on a single wafer is dramatically improved, and this is rearranged. If the LED display is manufactured, it is expected that the price of the apparatus can be greatly reduced.

  In the above-described manufacturing method, it is necessary to arrange the LED elements on the circuit board corresponding to the pixel pitch, but this pixel pitch is independent of the element arrangement pitch at the time of element formation. Requires a technique for rearranging the LED elements. In particular, when a large number of LED elements are formed on a single wafer as described above, so-called enlargement transfer technology is essential.

Therefore, a technique has been proposed in which each element is formed with a high degree of integration, and after dicing, moved to a wide area while being separated by transfer or the like, to constitute a relatively large display device (for example, Patent Document 1 to Patent Document 1). (See 8).
JP 2002-118124 A JP 2002-182580 A JP 2002-185039 A JP 2002-31858 A JP 2002-314053 A JP 2002-314123 A JP 2003-77940 A JP 2003-98777 A

  In the transfer technologies described in the above-mentioned patent documents, all can be said in common that when transferring an element, the transfer substrate is attached with a resin (adhesive), and the transfer substrate is peeled off from the resin (adhesive). There is a big problem that the transfer process is complicated.

  For example, an example of a conventional element transfer method will be described. In the conventional element transfer method, an element is formed on a first transparent substrate using an epoxy resin or the like on a second transparent substrate on which a film that absorbs in the wavelength region of the laser is formed. The surface is pasted, the first transparent substrate is peeled off, and the element formed on the first transparent substrate is transferred to the second transparent substrate. After that, element isolation and electrode fabrication are performed, and a laser is selectively irradiated from the back surface of the second transparent substrate to melt the film to be absorbed in the laser wavelength range, and a plurality of elements are thinned out from the second transparent substrate. Transfer to the relay board. Further, the element transferred to the relay substrate is pressed against the embedded substrate and embedded and transferred. At this time, wiring on the back surface of the element is performed, the back surface of the element after wiring is bonded to the display substrate using a resin, and the embedded substrate is peeled off by a laser. Dry etching is performed from the peeled surface, the resin on the element surface side is removed, cueing is performed so that electrical connection can be made to the bump, and wiring is performed.

  When such a transfer process is employed, since the element is held by the resin with respect to the second transparent substrate which is a transfer substrate, it takes time and effort for peeling (separation), and the heat of the laser is used. Because it is necessary, capital investment is also required, leading to increased costs.

  Further, in the transfer process, for example, the element cannot be energized while being transferred to the second transparent substrate, and the defect of the element cannot be determined in the transfer process. Therefore, the defective portion cannot be determined until it is mounted on the mounting board (circuit board), which causes a decrease in yield.

  The present invention has been proposed for the purpose of eliminating the disadvantages of the prior art. That is, the present invention has an object to provide an element transfer method that does not require application of a resin during transfer, can easily perform element transfer, and can reduce the cost during transfer. And It is another object of the present invention to provide an element transfer method capable of determining a defective portion before transferring an element to a mounting substrate and improving yield.

  In order to achieve the above object, an element transfer method of the present invention is an element transfer method for transferring an element formed on a first substrate to a second substrate, wherein a magnetic metal pattern is formed on the second substrate. A magnetic thin film is formed on at least a part of the electrode of the element formed on the first substrate, and between the magnetic metal pattern on the second substrate and the magnetic thin film formed on the element. The element is held on a second substrate by a magnetic coupling force, and in this state, the first substrate is peeled off to transfer the element to the second substrate.

  In the element transfer method of the present invention, the element is held on the relay substrate (second substrate) using the magnetic coupling force between the magnetized magnetic metal pattern and the magnetic thin film formed on the element. Temporary fixing with resin (adhesive) is unnecessary. Therefore, there is no need for troubles such as peeling, and it is not necessary to use the heat of the laser or the like during the transfer.

  Furthermore, in the element transfer method of the present invention, it is also possible to determine element defects in the transfer process by using a magnetic metal pattern formed on the second substrate. This is defined in claim 6, characterized in that an energization test is performed in a state where the element is held on the second substrate.

  In a state where the element is held on the second substrate using the magnetic coupling force, the magnetic metal pattern formed on the second substrate is in contact with the magnetic thin film formed on one electrode of the element, Therefore, the magnetic metal pattern functions as an extraction electrode for one electrode of the element. Therefore, if energization is performed between the magnetic metal pattern and the back electrode formed on the surface opposite to the element, each element can be energized to conduct an energization test. The defective part is grasped before starting.

  According to the present invention, it is not necessary to apply a resin for transfer, and it is possible to easily perform element transfer, and it is possible to reduce the cost for transfer. Further, according to the present invention, it is possible to determine a defective portion before transferring the element to the mounting substrate, and it is possible to improve the yield.

  Hereinafter, an element transfer method to which the present invention is applied will be described in detail with reference to the drawings.

  In this embodiment, the element transfer method of the present invention will be described by taking as an example the case of manufacturing an image display device by enlarging and transferring LED elements. First, the concept of enlarging transfer will be briefly described. FIG. 1 is a diagram showing an example of enlarged transfer by selective transfer. In the selective transfer shown in FIG. 1, thinning transfer is performed in which a part of elements arranged in a matrix is thinned and transferred. Thinning transfer is performed by selectively transferring elements while facing a transfer source substrate (transfer substrate) and a transfer destination substrate (transfer target substrate), but the transfer destination transfer target substrate has a large size. As a result, it is possible to dispose all the elements on the transfer source transfer substrate separately from the transfer destination transfer target substrate.

  FIG. 1 shows an example in which the first transfer process is performed at an enlargement ratio of 3 times. When the transfer substrate 1 is used as a unit, the transfer substrate 2 has an area that is 9 times the square of 3. For this reason, a total of nine transfers are performed in order to transfer all of the elements 3 on the transfer substrate 1 which is the transfer source substrate. The elements 3 arranged in a matrix on the transfer substrate 1 are divided into 3 × 3 matrix units (in this example, 9 sets of matrix units), and the elements constituting one matrix unit in the transfer at a time 3 is transferred to the substrate 2 to be transferred, and all elements 3 are finally transferred onto the substrate 2 to be transferred by repeating this process.

  FIG. 1A schematically shows a state where the elements 3 constituting the first matrix unit among the elements 3 on the transfer substrate 1 are transferred to the transfer substrate 2. FIG. ) Schematically shows a case where the element 3 constituting the second matrix unit is transferred to the substrate 2 to be transferred. In the transfer of the element 3 constituting the second matrix unit, the alignment position of the transfer substrate 1 with respect to the transfer substrate 2 is perpendicular to the alignment position in the transfer of the element 3 constituting the first matrix unit. By repeating the same thinning transfer, the elements 3 constituting each matrix unit can be arranged separately in different regions on the substrate 2 to be transferred.

  FIG. 1C schematically shows a case where the element 3 constituting the eighth matrix unit is transferred to the substrate 2 to be transferred, and FIG. 1D shows the ninth matrix unit. The element 3 to be transferred is schematically shown on the transfer substrate 2. At the time when the elements 3 constituting the ninth matrix unit are transferred, the transfer substrate 1 has no elements 3 and all the elements 3 are separated from the transfer substrate 1 on the transfer substrate 21. It will be held in a matrix.

  Hereinafter, a specific element transfer process will be described. In the element transfer method of the present embodiment, first, as shown in FIG. 2, for example, a magnetic metal material is vapor-deposited on a transfer substrate 11 (corresponding to a second substrate) serving as a transfer relay substrate, and this is patterned. Thus, the magnetic metal pattern 12 is formed.

  The transfer substrate 11 has a size and a shape that substantially match a later-described element forming wafer, and the magnetic metal pattern 12 is formed as a belt-like pattern that matches the pitch of elements formed on the element forming wafer. ing. In this example, the element pitches are formed at two pitches of 18.75 μm and 25 μm.

  As described above, the magnetic metal pattern 12 is formed by depositing a magnetic metal material by a technique such as vapor deposition or plating, and patterning the magnetic metal material by a photolithography technique or the like. As the magnetic metal material, for example, nickel (Ni) or the like can be used. In this example, a two-layer structure of a magnetic metal layer made of nickel and an underlayer made of Ti is adopted. The element attracting magnet 12a made of Ni has a thickness of about 3 μm, and the Ti underlayer 12b has a thickness of about 1 μm.

  The magnetic metal pattern 12 needs to be preliminarily magnetized. Therefore, after Ni deposition and patterning, a magnetizing process is performed as a second process. This magnetizing step may be performed by applying a magnetic field in a predetermined direction. In this example, the magnetizing is performed so that the left side in the figure is the N pole and the right side is the S pole.

  On the other hand, the LED elements to be transferred are formed on the element forming wafer 13 as shown in FIG. For example, in the case of forming a blue or green light emitting diode, the element forming wafer 13 is formed with a gallium nitride based crystal growth layer 16 and the like on a sapphire substrate 14 (corresponding to a first substrate) via a base growth layer 15. The LED element 20 is formed by diffusing a predetermined element into the gallium nitride based crystal growth layer 16. In this example, a light emitting diode having a double hetero structure composed of an N clad layer 17, an active layer 18, and a P clad layer 19 is formed as the LED element 20. In the LED element 20 having the above-described configuration, the region between the elements 20 is filled with the insulating layer 21 so as to prevent an inadvertent short circuit or the like.

  A P bump electrode 22 is formed on the P clad layer 19 of each LED element 20, and this P bump electrode 22 functions as one electrode of the LED element 20. As shown in FIG. 3B, the P bump electrode 22 has a substantially circular planar shape, and the pitch between the center points is the formation pitch of the LED elements 20. In the case of this example, the diameter of the P bump electrode 22 is 7 μm, and as described above, the LED elements 20 are formed at two pitches of 18.75 μm and 25 μm.

  Further, in the present embodiment, a magnetic thin film, for example, a Ni thin film 23 is formed on the P bump electrode 22 corresponding to the magnetic metal pattern 12 formed on the transfer substrate 11. The Ni thin film 23 can also be formed by a technique such as vapor deposition or plating. By forming the Ni thin film 23 on the P bump electrode 22, the Ni thin film 23 is attracted to the magnetic metal pattern 12 by a magnetic coupling force. It becomes possible.

  After preparing the element forming wafer 13 and the transfer substrate 11 described above, the transfer substrate 11 is overlaid on the element forming wafer 13 as shown in FIG. At this time, the magnetic metal pattern 12 is overlaid on the Ni thin film 23 formed on the P bump electrode 22 of the element forming wafer 13 so as to face the Ni thin film 23. As a result, the Ni thin film 23 on the P bump electrode 22 and the element attracting magnet 12a constituting the magnetic metal pattern 12 on the transfer substrate 11 come into contact with each other, and the magnetism between the magnetized magnetic metal pattern 12 and the Ni thin film 23 is brought about. The P-bump electrode 22, that is, the LED element 20 is tightly fixed to the transfer substrate 11 by the mechanical coupling force.

  In the state where the LED element 20 is fixed by the magnetic coupling force using the Ni thin film 23 on the magnetic metal pattern 12 of the transfer substrate 11 as described above, the back surface side of the sapphire substrate 14 of the element forming wafer 13 as shown in FIG. The sapphire substrate 14 is peeled off by irradiating with an energy beam to cause so-called ablation (or laser ablation when laser light is used as the energy beam).

  As an energy beam used for ablation, a laser beam such as an excimer laser or a YAG laser can be used. For example, in the case of the gallium nitride-based LED element 20, the gallium nitride-based crystal growth layer 16 (underlying growth layer 15) is decomposed into metallic Ga and nitrogen at the interface with the sapphire substrate 14 and is easily peeled off from the sapphire substrate 14. The

  In addition, after the sapphire substrate 14 is peeled off by the laser ablation, metal Ga generated during ablation adheres to the surface of the gallium nitride based crystal growth layer 16. Therefore, after the laser ablation, it is preferable to remove the metal Ga deposited on the surface of the gallium nitride based crystal growth layer 16 by a technique such as wet etching. For example, if wet etching is performed using a mixed acid solution of hydrochloric acid and nitric acid, the deposited metal Ga can be easily removed.

  After the sapphire substrate 14 is peeled and removed, an element separation process for separating the LED elements 20 is performed as shown in FIG. The element separation process is a process of dividing the gallium nitride-based crystal growth layer 16 by etching, such as dry etching, and separating it into individual LED elements 20, whereby a large number of LED elements 20 are arranged on the transfer substrate 11. It becomes a state. In addition, the separated LED elements 20 are maintained on the transfer substrate 11 by magnetic coupling between the Ni thin film 23 and the magnetic metal pattern 12, respectively.

  After element isolation, a contact electrode 24 on the N clad layer 17 side is formed on the surface opposite to the previous P bump electrode 22, that is, on the gallium nitride crystal growth layer 16. The contact electrode 24 can be formed by depositing or plating a conductive material such as metal. In this example, the contact electrode 24 is formed by vapor deposition of Ti 10 nm and Pt 50 nm, and the shape thereof is a partial arc shape. Each LED element 20 emits light when energized between the contact electrode 24 and the P bump electrode 22.

  As described above, the LED elements 20 are individually separated and function as light emitting elements at this stage respectively. However, in the conventional enlarged transfer technique, the LED elements 20 are mounted on a circuit board on which a wiring is formed and a circuit is formed. Until now, it is impossible to determine the defect of the LED element 20, which is a cause of greatly reducing the yield. On the other hand, in the element transfer method of the present invention, for example, the defect of each LED element 20 can be determined at this stage, and the yield and the like can be greatly improved.

  That is, in the element transfer method of the present invention, the separated LED element 20 is caused by the magnetic coupling force between the Ni thin film 23 formed on the P bump electrode 22 and the magnetic metal pattern 12 on the transfer substrate 11. In this state, the Ni thin film 23 on the P bump electrode 22 and the magnetic metal pattern 12 are in contact with each other. Here, since both the Ni thin film 23 and the magnetic metal pattern 12 are formed of a metal material having good electrical conductivity, the P bump electrode 22 and the magnetic metal pattern 12 are in an electrically conductive state.

  Therefore, as described above, when the contact electrode 24 is formed after the sapphire substrate 14 is peeled off and energization is performed between the contact electrode 24 and the magnetic metal pattern 12, the individual LED elements 20 are energized to determine whether they are good or bad. be able to. FIGS. 7A and 7B show the state of the electrical current test. One probe P1 of the test apparatus is brought into contact with the contact electrode 24, and the other probe P2 is brought into contact with the magnetic metal pattern 12. FIG. Thus, the LED element 20 can be energized, and for example, a defect can be determined by looking at its light emission state. In addition, the probe P1 is sequentially brought into contact with the contact electrodes 24 of the LED elements 20 arranged along the magnetic metal pattern 12 while the probes P2 are in contact with the magnetic metal pattern 12 in a predetermined row, so that one row. The LED element 20 can be energized. Therefore, the magnetic metal pattern 12 with which the probe P2 is brought into contact is selected, and the probes P1 are sequentially brought into contact with the contact electrodes 24 of the LED elements 20 arranged along each magnetic metal pattern 12 as described above, whereby all the LEDs are arranged. An energization test can be performed on the element 20.

  If each LED element 20 can be determined to be defective at this point, the defective LED element 20 will not be mounted on the circuit board (display board). As a result, the yield of the product can be greatly improved, and troublesome element replacement work becomes unnecessary.

  After determining the defect of the LED element 20, the LED elements 20 arranged on the transfer substrate 11 are selectively taken out and rearranged on the display substrate (circuit board). In this rearrangement, it is first necessary to selectively remove the LED elements 20 on the transfer substrate 11 from the matrix. In the present invention, the LED elements 20 are selectively removed by magnetic demagnetization. This is done using the elimination of dynamic binding power.

  FIG. 8A and FIG. 8B show a method for selectively taking out the LED element 20. When taking out the LED elements 20, as shown in FIG. 8A, a relay substrate 26 having a silicone coating layer 25 is disposed facing each LED element 20. In this state, a jig 27 having a magnetic pole opposite to the magnetic metal pattern 12 is brought into contact with the back side of the transfer substrate 11. In the case of this example, the jig 27 is configured such that a magnet 29 is selectively formed on a glass mask 28. Further, when the magnetic metal pattern 12 is, for example, an N pole, the magnet 29 causes the S pole to face the magnetic metal pattern 12.

  As a result, the magnetic field generated by the magnetic metal pattern 12 is canceled by the magnetic field formed by the magnet 29, and the magnetic metal pattern 12 at the portion facing the magnet 29 is changed from a magnetized state to a demagnetized state. As a result, the magnetic coupling force due to the magnetic metal pattern 12 is eliminated, and the LED element 20 adsorbed on this portion is detached from the transfer substrate 11 and moved onto the relay substrate 26. Therefore, in the above-described enlarged transfer, if the magnetic coupling force is selectively eliminated by the magnet 29 corresponding to, for example, the LED elements 20 constituting the first matrix unit, the first matrix unit is constituted. Only the LED element 20 to be transferred can be transferred onto the relay substrate 26. In this example, as shown in FIG. 8B, the LED elements 20 are taken out at a pitch of 150 μm and transferred to the relay board 26.

  Note that the selective demagnetization of the magnetic metal pattern 12 is not necessarily performed by using the jig 27 (magnet 29), but can be performed by a method capable of performing selective demagnetization such as demagnetization by a magnetic head or demagnetization by an electromagnet. Any method may be used as long as it exists.

  The LED element 20 taken out on the relay substrate 26 using the selective demagnetization is then transferred by being pressed against the embedded substrate 30 as shown in FIGS. 9 (a) and 9 (b). The The embedded substrate 30 is made of, for example, a sapphire substrate, has regions partitioned by partition walls 31 corresponding to the LED elements 20, and each region is filled with a resin 32. The LED elements 20 on the relay substrate 26 are embedded in a resin 32 in a predetermined area. In this example, the LED element 20 (G) that emits green light and the LED element (B) that emits blue light are arranged adjacent to each other and are embedded in the resin 32 in different partition regions.

  After embedding the LED element 20 in the resin 32 of the embedded substrate 30, a horizontal line (cathode line) 33 corresponding to the contact electrode 24 on the back surface is formed as shown in FIGS. 10 (a) and 10 (b). Then, the horizontal line 33 and the contact electrode 24 are electrically connected. The electrical connection is performed by, for example, coating a predetermined region with a permanent resist 34 and then applying a transparent conductive ink (for example, ITO ink) 35 in a region surrounded by the permanent resist 34. 10A is a cross-sectional view seen from the 90-degree rotation direction of FIG. 9A, and FIG. 10B is a plan view seen from the bottom surface side (contact electrode 24 formation surface side). FIG.

  After the wiring (horizontal line 33 and transparent conductive ink 35) for connecting the N-side contact electrode 24 is formed, as shown in FIGS. 11A and 11B, the contact electrode 24 forming side is formed. The display substrate 36 is bonded to the substrate. The bonding of the display substrate 36 is performed by a vacuum bonding apparatus using, for example, a thermosetting adhesive 37. As the thermosetting adhesive 37, for example, an epoxy-based adhesive (trade name EPOTEK354) or the like can be used.

  Next, as shown in FIGS. 12A and 12B, the embedded substrate 30 is peeled off and removed. The embedded substrate 30 can be peeled off using laser ablation as in the case of the sapphire substrate 14 described above. That is, laser light is irradiated from the back side of the embedded substrate 30 that is a sapphire substrate. Thereby, the resin 32 is ablated, and the interface between the resin 32 and the embedded substrate 30 is peeled off. As the laser light, an excimer laser or the like can be used.

  After the embedded substrate 30 is peeled off, as shown in FIGS. 13A and 13B, the resin 32 is etched and removed to a midway position, and the P bump electrode 22 of each LED element 20 is cueed. At the same time, a lead-out via 38 for the previously formed horizontal line 33 is formed.

  Finally, as shown in FIGS. 14A and 14B, a vertical line (anode line) 39 is formed to complete the LED display. The vertical lines 39 are formed by forming a conductive film and patterning by etching. For example, Ti and Cu are formed by sputtering and patterned by wet etching to form the vertical line 39 so as to overlap the P bump electrode 22. Simultaneously with the formation of the vertical line 39, the horizontal line lead pad 40 is formed corresponding to the lead via 38 of the horizontal line 33 previously formed.

  According to the above element transfer method, when the LED element 20 is transferred from the sapphire substrate 14 which is an element formation substrate to the transfer substrate 11, it is not necessary to attach the transfer substrate 11 with a resin, and transfer is performed in a later step. The LED element 20 can be easily separated from the substrate 11. In addition, since the bonding using the resin is not performed as described above, after the LED element 20 is mounted on the mounting substrate (display substrate 36), the resin adhered to the front side of the element for bonding to the transfer substrate 11 is used. There is no need for removal, and the process can be simplified.

  Further, when the LED element 20 is selectively removed from the transfer substrate 11, a magnet having a magnetic pole opposite to the magnetic metal pattern 12, a magnetic head, or the like is brought into contact from the back side of the adsorption surface of the LED element 20 of the transfer substrate 11. By selectively demagnetizing the magnetic force attracting the element 20, it is possible to selectively take out a large number of LED elements 20 at specific positions while maintaining a constant element pitch. In addition, since the transfer process can be performed without using a laser beam or the like for this selective extraction, a significant cost reduction can be realized.

  Furthermore, since the magnetic metals (Ni thin film 26 and magnetic metal pattern 12) are in contact with each other in a state where the LED element 20 is transferred and held on the transfer substrate 11, each LED element is utilized using the magnetic metal pattern 12. 20 can be energized, and it is possible to determine a defect before transferring the LED element 20 to the mounting substrate.

It is a typical top view explaining the concept of expansion transfer. It is a schematic plan view of a transfer substrate on which a magnetic metal pattern is formed. (A) is a principal part expanded sectional view of an element formation wafer, (b) is a schematic plan view. It is a principal part expanded sectional view which shows the state which accumulated the transfer board | substrate on the element formation wafer. It is a principal part expanded sectional view which shows the peeling process of a sapphire substrate. (A) is a principal part expanded sectional view which shows an element isolation process and a contact electrode formation process, (b) is a schematic plan view. (A) is a principal part expanded sectional view which shows an electricity test process, (b) is a schematic plan view. (A) is a principal part expanded sectional view which shows the element extraction process by selective demagnetization, (b) is a schematic bottom view. (A) is a principal part expanded sectional view which shows the transcription | transfer process to an embedded substrate, (b) is a schematic bottom view. (A) is a principal part expanded sectional view which shows an N side electrical connection process, (b) is a schematic bottom view. (A) is a principal part expanded sectional view which shows the bonding process of the embedded board | substrate to a display board | substrate, (b) is a schematic bottom view. (A) is a principal part expanded sectional view which shows the peeling process of an embedded substrate, (b) is a schematic plan view. (A) is a principal part expanded sectional view which shows the cueing process of P bump electrode, (b) is a schematic plan view. (A) is a principal part expanded sectional view which shows the P side wiring formation process, (b) is a schematic plan view.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Transfer substrate, 12 Magnetic metal pattern, 13 Element formation wafer, 14 Sapphire substrate, 15 Underlayer growth layer, 16 Gallium nitride crystal growth layer, 17 N clad layer, 18 Active layer, 19 P clad layer, 20 LED element, 21 Insulating layer, 22 P bump electrode, 23 Ni thin film, 24 contact electrode, 25 silicone coating layer, 26 relay substrate, 27 jig, 28 glass mask, 29 magnet, 30 embedded substrate, 31 partition wall, 32 resin, 33 horizontal line, 34 Permanent resist, 35 Transparent conductive ink, 36 Display substrate, 37 Thermosetting adhesive, 38 Drawer via, 39 Vertical line, 40 Drawer pad

Claims (7)

In an element transfer method for transferring an element formed on a first substrate to a second substrate,
Forming a magnetic metal pattern on the second substrate and magnetizing the magnetic substrate, and forming a magnetic thin film on at least a part of the electrodes of the element formed on the first substrate;
The element is held on the second substrate by the magnetic coupling force between the magnetic metal pattern on the second substrate and the magnetic thin film formed on the element, and in this state, the first substrate is peeled to remove the element from the second substrate. An element transfer method comprising transferring to a substrate.   2. The element transfer method according to claim 1, wherein the magnetic metal pattern on the second substrate is selectively demagnetized, the magnetic coupling force is selectively released, and the element at the corresponding location is selectively taken out.   4. The element transfer method according to claim 3, wherein the elements are selectively taken out so as to maintain a constant inter-element pitch.   3. The device transfer method according to claim 2, wherein the device is selectively taken out by transferring onto a third substrate.   The demagnetization is performed by bringing a jig that generates a magnetic field in a direction to cancel the magnetization of the magnetic metal pattern of the second substrate into contact with the surface of the second substrate opposite to the surface on which the magnetic metal pattern is formed. The element transfer method according to claim 2, wherein:   2. The element transfer method according to claim 1, wherein an energization test is performed with the element held on a second substrate. 7. The element transfer method according to claim 6, wherein the energization test is performed by energizing between the magnetic metal pattern of the second substrate and the back electrode of each element.