Lifting method

The scanning reduction projection optical system addresses high-speed and high-precision transfer challenges by using a lens array zoom homogenizer and telecentric projection lens to achieve efficient and accurate retransfer and defect removal in micro-element manufacturing, overcoming scanner limitations and reducing costs.

JP2026090347APending Publication Date: 2026-06-02SHIN ETSU CHEMICAL CO LTD

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
SHIN ETSU CHEMICAL CO LTD
Filing Date
2026-02-04
Publication Date
2026-06-02

Smart Images

  • Figure 2026090347000001_ABST
    Figure 2026090347000001_ABST
Patent Text Reader

Abstract

This device provides a low-cost scanning reduction projection small optical system that can scan a small irradiation area with a uniform and stable energy distribution over a wide area with high precision and high speed, while compensating for the lack of accuracy of scanners, without using large-aperture, expensive fθ lenses or telecentric reduction projection lenses. It also provides a lift device for mounting or retransferring the system, and describes its implementation method. [Solution] A scanning reduction projection optical system is used to image a transverse multimode pulsed laser beam onto a donor substrate in the size of a minute area via a lens array type zoom homogenizer, array mask 10, scanning mirror 4, photomask 6, and telecentric projection lens 8, thereby lifting a minute element on the donor substrate onto an opposing receptor substrate with high positional accuracy. The implementation process involves the following steps: inspection for acquiring positional information, lift area division, irradiation position selection, transfer, and stage movement.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a scanning reduction projection optical system and a laser processing apparatus using the same.

Background Art

[0002] As a technique used in the manufacturing process of displays and the like, there is a technique of transferring micro-elements such as micro LEDs arranged in a matrix on a donor substrate to a receptor substrate. In addition, there are transfer techniques to a receptor substrate for various functional films and material films such as conductive and adhesive films, and organic EL films applied on the donor substrate. For example, there are various techniques such as LIFT (Laser Induced Forward Transfer) technique, stamp technique, roll transfer technique, etc. However, it is not easy to achieve both high-speed processing and high positional accuracy required in the process for any of these techniques, and transfer leakage, positional deviation, etc. may occur even at high speed.

[0003] Therefore, the applicant has developed a high-speed and high-precision lift device (Patent Document 1). However, there may be defects reaching 1% or more in the micro-elements and the like on the donor substrate and the receptor substrate that are transferred and mounted by the transfer device using the above-described stamp technique or roll transfer technique used in the lift device. Therefore, it is required to have both high precision and high speed for a technique for re-transferring micro-elements and the like to such defective portions and a device using the same.

[0004] In addition, such a technique can be used not only for re-transfer from a donor substrate to a receptor substrate, but also for removing irradiated objects such as defective micro-elements and unnecessary material portions located on the substrate. In addition to the mounting process by transfer of micro-elements and the like, it is also expected to be used in a defective removal process.

[0005] On the other hand, unlike Patent Document 1, there is a high-precision transfer technology using a step-and-repeat method. For example, in a technology that re-transfers micro-elements to areas without elements after defective micro-elements have been removed from the substrate to be corrected, an excimer laser beam is shaped into a laser beam with a uniform energy distribution using a beam homogenizer made of a lens array, and this is projected at a reduced size onto the micro-elements located on the donor substrate (for re-transfer) using a photomask and a reduction projection lens, thereby lifting the micro-elements onto the substrate to be corrected. This technology makes it possible to accurately lift (re-transfer) the micro-elements on the donor substrate onto the substrate to be corrected using a stage with high positional accuracy, but since it requires a processing time of about 1 to 2 seconds per micro-element, it is not practical as a re-transfer device used in the manufacturing process of displays and the like, where a large number of defective areas of up to 1% may occur, considering its production efficiency.

[0006] While it is well known that a technique combining a galvanometer scanner and an fθ lens to scan and irradiate an object with laser light at high speed is a technology that enables high-speed processing, this technique can remove defective micro-elements from the substrate to be repaired at high speed, as long as the positional accuracy of the scanner allows it. However, subsequent re-transfer, which requires high positional accuracy, is difficult due to the limitations of this accuracy.

[0007] Therefore, in order to resolve the positional accuracy problem with this scanner, it is possible to construct a high-speed, high-positional-accuracy scanning reduction projection optical system that reduces dependence on the scanning accuracy of the scanner by selectively irradiating the scanned laser light towards apertures arranged on a photomask via an fθ lens or the like, and projecting this reduced light onto a predetermined irradiated object located on the substrate. Patent Document 2 shows an example of an optical system that compensates for positional misalignment of the irradiated area on a donor substrate caused by low scanning accuracy by scanning the beam of an Nd:YAG laser with a galvanometer mirror and projecting it reducedly via an fθ lens and a photomask.

[0008] However, in order to increase the size of each substrate and achieve even faster processing, it is necessary to enlarge the irradiation area on the donor substrate and reduce the laser light scanned onto many irradiated objects in a short time. In other words, it is necessary to enlarge the scannable irradiation area on the photomask. In this case, it is necessary to enlarge the aperture of the fθ lens and reduction projection lens used in Patent Document 2, and designing it to be telecentric would be costly.

[0009] Furthermore, in order to accommodate the miniaturization and high density of irradiated objects such as micro-elements, it is necessary to compensate for the accuracy limits of the scanner and to have a laser beam with a stable and uniform intensity distribution and an extremely small irradiation area size on the donor substrate so as not to interfere with adjacent irradiated objects. In addition, there is a desire for the realization of lift devices, retransfer devices, and defect removal devices equipped with such optical systems, as replacements for the devices described in Patent Document 1. [Prior art documents] [Patent Documents]

[0010] [Patent Document 1] Japanese Patent Publication No. 2020-4478 [Patent Document 2] Japanese Patent Publication No. 2006-41500 [Overview of the project] [Problems that the invention aims to solve]

[0011] Therefore, the objective is to provide a low-cost scanning reduction projection optical system that can scan a small irradiation area with a uniform and stable energy distribution over a wide area with high precision and high speed, while compensating for the lack of precision of scanners, without using large-aperture, expensive fθ lenses or telecentric reduction projection lenses, as well as a lift device for implementation or retransfer that incorporates this system, and furthermore, to provide a method for implementing these systems. [Means for solving the problem]

[0012] The first invention is a scanning reduction projection optical system that can be used in a laser processing apparatus that utilizes to irradiate an object to be irradiated, such as a plurality of mini-LEDs or micro-LEDs arranged on a substrate, or a material film or functional film attached to a substrate by coating or printing, etc., with multimode pulsed laser light to induce a reaction directly on the object to be irradiated or through a substance between the substrate and the object to be irradiated, and comprises a lens array type zoom homogenizer, a scanning mirror scanned by a drive axis control device of one or more axes, a photomask, and a projection lens system in which at least the image side is telecentric, and the photomask has a plurality of apertures of a predetermined shape to be reduced and projected, arranged at a predetermined pitch.

[0013] Furthermore, this homogenizer includes a first lens array, a second lens array, and a condenser lens, and is a zoom homogenizer that constitutes an infinity-correcting optical system with the second lens array and the condenser lens, and images an illumination area of ​​a predetermined size covering one or more adjacent aperture groups on the photomask onto the photomask, and in particular compensates for fluctuations in the position and size of the illumination area and the energy intensity distribution within the illumination area.

[0014] "Compensating for fluctuations" refers to a state in which, for example, fluctuations in the oscillation state are known to cause fluctuations in the beam pointing stability, beam size, and even the intensity distribution of the beam cross-section of the laser light. This zoom homogenizer avoids these effects within the irradiation area imaged on the photomask. As a result, an image of a minute area with an extremely uniform energy distribution is obtained on the donor substrate onto which the mask is projected.

[0015] The predetermined size is such that the irradiation area does not extend to any other apertures adjacent to the aperture group. In other words, considering the irradiation position accuracy on the photomask due to the scanning accuracy of the scanning mirror, if the laser light were to pass through its adjacent aperture and irradiate an object on the substrate (that is not intended to be irradiated), the boundary (outer edge) of the energy distribution above the threshold that would induce a reaction would not extend to any aperture adjacent to the aperture group contained within that irradiation area. For example, in Figure 1, the irradiation area (DP) enclosed by the dashed line exemplifies the maximum acceptable size of the irradiation area covering the aperture group consisting of four apertures (61), and the small irradiation area enclosed by the dotted line exemplifies the minimum size of the irradiation area covering a single aperture. In the figure, each example of an irradiation area is shown twice to represent the deviation due to the positional accuracy of the scanner.

[0016] For example, as shown in Figure 1, when apertures (squares) are arranged in a matrix on a photomask (6), the side length (DP) of a predetermined size of the irradiation area (square) that irradiates n × n (n ≥ 1) of these apertures at once is within the range of the following equation, where Ma is the side length of each aperture, Pi is the pitch, and St is the scanning position accuracy on the photomask by the scanner. Pi×(n-1)+Ma+St ≦ DP < Pi×(n+1)-Ma-St If the predetermined size exceeds this range, laser light with energy exceeding the aforementioned threshold that passes through an unintended adjacent aperture (or part thereof) may be projected onto an unintended irradiated object at a reduced size, potentially inducing a reaction. The shape of the aperture is not square, and the range of the predetermined size (DP) in the simultaneous irradiation of n × m objects is a design consideration.

[0017] When using a photomask with a matrix of apertures for simultaneous irradiation, the pitch (Pi) of the apertures on the photomask is fixed at c times the pitch of the irradiated object to be lifted onto the receptor substrate, assuming the reduction ratio of the reduction projection lens is 1 / c.

[0018] Depending on the specifications of the reduction projection lens, various designs are available for the number of apertures, ranging from a single row arrangement as shown in Figure 2 to a matrix arrangement as shown in Figure 1 above. These also depend on the scannable range of the scanning mirror.

[0019] The lens elements constituting the first and second lens arrays are not limited to fly-eye type, but may also be cylindrical or spherical. Therefore, each lens array may be composed of a combination of orthogonal lens elements. Furthermore, there may be a zoom homogenizer with an additional third lens array.

[0020] The second invention is a scanning reduction projection optical system in which, in the first invention, the projection lens system comprises a field lens disposed between the condenser lens and the photomask, and a reduction projection lens that is at least telecentric on the image side.

[0021] Here, the specifications of the field lens are determined based on the specifications of the zoom homogenizer, condenser lens (3), and telecentric lens, and in this invention, its preferred position is directly in front of the photomask (6) as shown in Figure 3. The focal length of this field lens (5) is designed with a curvature that allows the laser light from the zoom homogenizer (a number of focusing points corresponding to the number of lens elements in the second lens array (2)) to pass through the aperture (7) placed at the entrance pupil position of the image-side telecentric reduction projection lens (8).

[0022] The third invention is a scanning reduction projection optical system in which the scanning mirror (4) in the first or second invention is composed of a two-axis galvanometer scanner. This allows scanning of the laser beam irradiation area toward multiple rows of apertures arranged on the photomask (6). The control device for the scanning mirror can vary, including a dedicated controller or a combination of a dedicated board and a PC. In either case, it can also be used to control the oscillation timing of the pulsed laser beam.

[0023] The fourth invention is a scanning reduction projection optical system using a zoom homogenizer in which an array mask composed of an opening group having a size smaller than the size of each lens element constituting the lens array is arranged facing each of the lens elements, immediately before the first lens array or between the first lens array and the second lens array. Thereby, on the donor substrate on which the opening shape on the photomask is reduced and projected, virtual images due to stray light between the lens elements are removed, and an extremely small imaging with a stable and highly uniform energy distribution can be obtained. Further, regardless of the lens element shape, a minute image of an arbitrary shape combined with the opening shape or arrangement of the photomask can be obtained.

[0024] When the lens element is cylindrical, an array mask of elongated openings may be used in combination. The position of the array mask is a design matter to be determined while checking the irradiation area on the photomask with a beam profiler or the like within a range located immediately before or after the first lens array or between the first lens array and the second lens array. Note that the number of lens elements and the number of the opening group of the array mask do not have to match. For example, the NA of the optical system can be adjusted by reducing the number of openings in the outer peripheral portion of the array mask.

[0025] As an example of the predetermined size, the element size of the fly-eye type first lens array, or the size of the opening of the array mask arranged immediately before the first lens array is dA, the focal length of the first lens array is f1, the focal length of the second lens array is f2, the distance between these first and second lens arrays is a (here a = f2), and the focal length of the condenser lens is fC, the focal length of the field lens constituting the projection lens system is fF, and the lens interval between these is b. Then, the predetermined size (DP) of the irradiation area imaged on the photomask is expressed by the following formula.

[0026]

Equation

[0027] However, this configuration is not ideal due to the effects of aberrations in the first lens array and diffraction caused by the array mask. Therefore, these problems can be avoided by moving the first lens array towards the light source and placing the array mask near the light source-side focal point of the second lens array. Furthermore, the utilization efficiency of the pulsed laser light cut out by the array mask can also be improved.

[0028] The fifth invention is a scanning reduction projection optical system in which the array mask in the fourth invention has multiple types of aperture groups arranged within the plane of the substrate, which enable switching between and using aperture groups of different sizes, shapes, or numbers of apertures.

[0029] If the number of apertures in the array mask is less than the number of lens elements in the first lens array, it becomes an array mask with the function of shaping illumination. Figure 4 shows an example of an array mask in which a group of apertures for shaping illumination and a number of opposing groups of apertures equal to the number of lens elements in the lens array are arranged on a single substrate.

[0030] The sixth invention is a scanning reduction projection optical system in which, in the fourth or fifth invention, the array mask is mounted on a mount that includes a θ-axis that allows for minute rotational adjustment around the optical axis.

[0031] Highly uniform imaging is achieved by superimposing the light emitted from each lens element of the second lens array onto the photomask using a condenser lens. Here, positional misalignment in the plane perpendicular to the optical axis in the relative positional relationship between the array mask and the second lens array does not affect uniformity. However, if there is a misalignment in the rotational direction (θ) around the optical axis, the outline of the formed image becomes blurred, resulting in multiple images. This is shown in Figure 5. (Here, an array mask with a circular aperture shape is used.) The effect of such rotational misalignment becomes larger as the image size decreases.

[0032] The seventh invention is a scanning reduction projection optical system in which the various optical elements in any of the first to sixth inventions correspond to the oscillation wavelength of an excimer laser.

[0033] The eighth invention is a laser processing apparatus that utilizes the induction of a reaction by irradiating an object to be irradiated located on a substrate with multimode pulsed laser light, wherein the multimode pulsed laser light emitted from the laser device is projected onto a substrate held on a stage having at least X-axis and Y-axis drive axes by a scanning reduction projection optical system described in any one of the first to sixth inventions.

[0034] The objects to be irradiated located on the substrate here can vary, including the aforementioned defective micro-elements and unwanted portions of functional films on circuit boards. Furthermore, the "reactions" induced here include, but are not limited to, mechanical, optical, electrical, magnetic, and thermal reactions.

[0035] The ninth invention is a laser processing apparatus for mounting, retransferring, or both, in which the substrate is a donor substrate on which the object to be irradiated is located, and pulsed laser light is irradiated toward the object to be irradiated from the back surface of the donor substrate to selectively peel or separate the object to be irradiated and lift it onto a receptor substrate facing the donor substrate. More specifically, it is a lifting apparatus.

[0036] Furthermore, the stage is a donor stage that holds the donor substrate with its back surface facing the incident side of the pulsed laser light, and it also has a receptor stage that holds the receptor substrate and has an X axis, a Y axis, a vertical Z axis, and a θ axis that rotates in the XY plane, and the scanning reduction projection optical system and this donor stage are installed on a first base plate, the receptor stage is installed on a second base plate or a base base plate, and furthermore, the first base plate and the second base plate are each independently installed on a base base plate.

[0037] Here, selectively peeling or separating the irradiated object means, in the case of a micro-element, selectively peeling the micro-element itself from the donor substrate, and, in the case of a functional film printed or coated on a donor substrate, selectively peeling or separating the functional film, etc., in the portion corresponding to the imaging position and size of the laser beam projected at a reduced size through an aperture on the photomask. Note that this peeling or separation includes cases in which a so-called ablation process is not involved.

[0038] Furthermore, since all stages perform a step-and-repeat operation, meaning that all stages are stationary when the laser beam is irradiated, the rolling guide system, which has physical contact with the guide and is stable, is preferable to the air bearing system, even considering cost.

[0039] On the other hand, the first surface plate, the second surface plate, and the foundation surface plate on which these stages are installed must be made of highly rigid materials such as steel, stone, or ceramic. As for stone, granite is a suitable material to use.

[0040] With the above configuration, the positional accuracy and stability of the irradiated object lifted onto the receptor substrate are determined by the relative positional relationship between the mask, reduction projection lens, and donor substrate, which is suppressed by the scanning reduction projection optical system and the installation structure of each substrate stage that eliminates vibration, as well as the alignment positional accuracy of the irradiated object on the donor substrate.

[0041] The tenth invention is a lift device in which, in the ninth invention, the control device for the scanning mirror includes the functions of controlling the scanning mirror that scans the optical axis of the pulsed laser toward an aperture on the photomask selected based on previously acquired positional information of the object to be irradiated on the donor substrate and information of the planned lift position on the receptor substrate, and controlling the irradiation of the pulsed laser light.

[0042] In particular, when using a lift device as a retransfer device, the possibility of defective areas with no elements or defective elements mounted on the donor substrate for retransfer cannot be ruled out. Therefore, based on information about the defective areas on both the receptor substrate to be lifted and the donor substrate to be lifted, the scanning mirror is controlled so that the optical axis is scanned to the aperture on the photomask facing the selected target object, so that the laser light from the scanning reduction projection optical system is irradiated only to the target object on the donor substrate to be lifted.

[0043] The eleventh invention is a lift device in which, in the ninth or tenth invention, the donor stage is capable of holding two or more donor substrates and can be switched between and used.

[0044] For example, a pulsed laser beam of a predetermined size, reduced in size by the scanning reduction projection optical system according to the present invention, is irradiated onto a first donor substrate coated with a conductive paste film, and a portion of the conductive paste film corresponding to that size is lifted (paste printed) onto the opposing receptor substrate. Next, the first donor substrate is switched to a second donor substrate (carrier substrate) by moving the donor stage, and a micro-element (device) on the carrier substrate is lifted to the same position on the receptor substrate and fixed via the conductive paste film.

[0045] The twelfth invention is a lift device in which the donor stage is suspended from the lower surface of the first base plate, as in the inventions ninth to eleventh.

[0046] The order in which the axes constituting the donor stage are installed is a design consideration, but preferably, they are installed from the bottom surface of the horizontally installed first base plate in the order of the X-axis, then the Y-axis, and if the θ-axis is included, it is suspended below it. The axis configuration of the receptor stage is also a design consideration.

[0047] The 13th invention is a laser processing apparatus according to the 8th invention, or a lift apparatus according to any of the 9th to 12th inventions, characterized in that the laser apparatus is an excimer laser apparatus.

[0048] The 14th invention is a lifting method for mounting or retransferring an object to be irradiated from a donor substrate onto an opposing receptor substrate, using a lifting device according to the present invention equipped with a scanning reduction projection optical system as described in any of the 9th to 13th inventions, comprising: an inspection step of acquiring in advance "position information D", which is the position information of the object to be irradiated on the donor substrate, and "position information R", which is the planned lifting position of the object to be irradiated onto the receptor substrate; a division step of dividing the area on the donor substrate into "divided areas D" of a predetermined size; and a selection step of selecting the position of the object to be irradiated within the divided area D to be lifted based on the position information D and position information R. This lifting method includes a transfer step in which a laser beam is emitted from a photomask (through a reduction projection lens) that passes through an aperture on the photomask opposite the position of the selected object to be irradiated, and irradiates the donor substrate with the selected object to be irradiated, thereby lifting the selected object to be irradiated in the divided area D toward a "divided area R" which is conveniently defined as an area on the receptor substrate opposite to it; and a moving step in which, after the transfer step, the donor substrate and the receptor substrate are moved to the next lift area. Thereafter, the transfer step and the moving step are repeated to mount or retransfer the selected object to be irradiated on the donor substrate onto the receptor substrate, covering the entire area of ​​the receptor substrate that is to be lifted.

[0049] In the inspection process, there are various methods for acquiring positional information of the irradiated objects being mounted, but it is a design matter to obtain the centroid coordinates of each object by processing images of all or two or more individual irradiated objects by sampling. The determination of the origin position of these coordinates is also a design matter. Positional information D and / or positional information R may be inspected using an independent inspection device separate from this lift device, and the results may be acquired by the control device of this lift device via communication means, or they may be calculated from design values ​​based on the measurement results during the positioning (alignment) of the donor substrate and receptor substrate in this lift device. It is desirable to perform this inspection process before the splitting process, although this depends on the total cycle time of mounting or retransfer.

[0050] The position of the selected irradiated object within divided area D is, when this lift method is used for mounting, the position of all irradiated objects on the donor substrate that are opposite the planned lift position in the opposing divided area R (excluding defective areas), and when used for retransfer, the position of the irradiated object on the donor substrate that is opposite the defective area (element-free area) on the receptor substrate (excluding defective areas (element-free areas)).

[0051] The transfer process includes both cases: one in which the scanning mirror is not stopped, and pulsed laser light is emitted in synchronization with the time the optical axis is scanned to the position of the selected object to be irradiated; and another in which scanning and stopping are repeated.

[0052] Furthermore, position information D may include not only the position coordinates of micro-elements or irradiated objects that are properly mounted on the donor substrate and can be properly peeled or separated, but also the position coordinates of irradiated objects identified as defective or missing (as abnormal position information). The same applies to position information R when this lift method is used for retransfer.

[0053] The maximum size of the divided area D depends on the reduction projection lens that constitutes the scanning reduction projection optical system mounted on this lift device. In particular, the numerical aperture and magnification of telecentric reduction projection lenses are limited in consideration of their manufacturing cost, and depending on these specifications, the area on the photomask, and consequently on the donor substrate, that can be lifted in a single scan is limited.

[0054] In this lifting method, regardless of whether the lifting device is used for mounting or re-transferring, the above-mentioned divided area D is set in the area of ​​the donor substrate, and the donor substrate and receptor substrate are moved by the step-and-repeat operation of each stage that holds them, and are lifted accurately while avoiding vibration when these stages stop.

[0055] The 15th invention is a lift method in which, in the 14th invention, the designed mounting pitch of the object to be irradiated on the donor substrate is 1 times, 1 / 2 times, 1 / 3 times, etc., an integer fraction of 1 or more, when the lift method is used for mounting, or when it is used for retransfer, when it is 1 times, 1 / 2 times, 1 / 3 times, etc., the designed mounting pitch of the object to be irradiated on the already mounted receptor substrate.

[0056] Furthermore, if the planned lift positions on the receptor substrate are arranged in an X x Y matrix, this also includes cases where the mounting pitch on the opposing donor substrate of the same size is 1 / n times in column X and 1 / m times in column Y (where n and m are distinct integers greater than or equal to 1).

[0057] In the 16th invention, first, it is assumed that there is an error between the substrates (taking into account the difference in mounting density on the donor substrate) between the actual mounting pitch of the object to be irradiated on the donor substrate, calculated from the position information D acquired in the inspection process, and the mounting pitch on the receptor substrate, calculated from the position information R. When this lift method is used for mounting, this mounting pitch on the receptor substrate is calculated from the planned lift position (design position) of the object to be irradiated, and when it is used for retransfer, it is the actual mounting pitch of the object to be irradiated that has already been mounted.

[0058] Even in donor boards where no defective or missing elements are present, the actual mounting pitch of the irradiated objects may have an error (δPi) compared to the designed mounting pitch, in light of the donor board manufacturing process. Furthermore, even if this error does not tend to vary from place to place on the same board, it is conceivable that there may be differences (errors) between manufacturing lots of donor boards, and even between donor boards themselves. In this case, the error δPi accumulates according to the number of irradiated objects contained within the divided area D.

[0059] Therefore, the 16th invention is a lifting method in which the amount of movement of each substrate in the moving process of the 15th invention is set to an amount of movement that cancels out this "cumulative error amount".

[0060] In the 16th invention, when attempting to lift an object to be irradiated within a divided area D using a donor substrate with an error (δPi) of a certain magnitude or greater, the cumulative error amount in that area may exceed the acceptable range in terms of lift position accuracy, depending on the position of the object to be irradiated within that area (the number of objects to be irradiated from the reference position). The discrepancy between the planned lift position on the receptor substrate and the position of the object to be irradiated on the donor substrate is a problem, but the discrepancy between the planned imaging position on the donor substrate where the laser light passing through the aperture on the photomask is expected to form and the position of the object to be irradiated is particularly problematic when the mounting density of objects to be irradiated on the donor substrate is high and the pitch is narrow.

[0061] Therefore, in the 17th invention, the distance between adjacent objects to be irradiated on the donor substrate is set as the upper limit, and the range of acceptable cumulative error is determined considering the difference between the irradiation size of the laser beam directed towards the objects on the donor substrate and the size of the objects to be irradiated, as well as the effect of their positional displacement on the lift position accuracy. If the cumulative error within the division area D exceeds this acceptable range at the position where it is maximum (for example, the lower right end when the upper left end is used as the reference point), the size of the division area D in the division process of the 16th invention is further reduced to a size that allows the cumulative error at that position to fit within the acceptable range, resulting in a "modified division area D".

[0062] Then, the selected irradiated object within the corrected division area D is mounted or re-transferred to the "corrected division area R" on the opposing receptor substrate, which is also conveniently defined as being the same size. Subsequently, a moving process is performed in which the donor substrate and the receptor substrate are moved by a stage in order to lift them to the next corrected division area R. At this time, the amount of movement of each stage is adjusted to cancel out the accumulated error amount.

[0063] Furthermore, when determining the allowable range for this cumulative error, it is advisable to assume, or simulate in advance, that even if the cumulative error does not exceed the aforementioned maximum value within a single divided area D, the maximum value may be exceeded as a result of repeating the movement process on the entire receptor substrate according to the design pitch on the donor substrate (or on the receptor substrate), and then determine this allowable range (size of the corrected divided area D).

[0064] The 18th invention is a lifting method in which, using position information D, position information R, the size of the divided area D, and the allowable range as parameters, a simulation program determines the size of the modified divided area D, the combination of movement amounts of each stage, and the execution order of each process so as to minimize the time required for mounting or retransferring the entire receptor substrate. [Effects of the Invention]

[0065] Without using large-diameter fθ lenses or telecentric projection lenses, this system compensates for the insufficient scanning accuracy of scanning mirrors, rapidly scanning a minute and stable irradiation area with a uniform and unchanging energy distribution toward apertures arranged on a photomask, and projecting it onto the irradiated object with high uniformity and high precision. This enables the low-cost realization of a defect removal device, or a lift device for mounting or re-transferring, equipped with this system. [Brief explanation of the drawing]

[0066] [Figure 1] A conceptual diagram illustrating the appearance of a photomask with apertures arranged in a matrix. [Figure 2] A conceptual diagram illustrating the appearance of a photomask with openings arranged in a row. [Figure 3] A conceptual diagram illustrating the arrangement of the constituent elements of a scanning reduction projection optical system. (Without array mask) [Figure 4] A schematic diagram illustrating the appearance of an array mask with multiple aperture groups arranged in a sequence. [Figure 5]Beam profiler image of the irradiation area (θ-axis offset) on the photomask. [Figure 6A] A schematic (3D) diagram of a lift device equipped with a scanning reduction projection optical system. [Figure 6B] A schematic diagram of a lift device equipped with a scanning reduction projection optical system. [Figure 7] Conceptual diagram of the element configuration of a zoom homogenizer. [Figure 8A] Conceptual diagram of an array mask. [Figure 8B] A photograph of an array mask. [Figure 9] A conceptual diagram showing the arrangement of each optical element from the zoom homogenizer to the donor substrate. [Figure 10A] A beam profile image showing the imaging state on the photomask when a rectangular array mask is inserted. [Figure 10B] Beam profile image showing the imaging state on the photomask when a circular array mask is inserted. [Figure 11] Conceptual diagram of the photomask used in Example 1. [Figure 12] Beam profile image of laser light projected onto a donor substrate at a reduced size. [Figure 13] A conceptual diagram showing the area where micro-elements are mounted across the entire effective area of ​​the donor substrate. [Figure 14] A conceptual diagram showing how a 6-inch donor substrate is divided into 27 separate areas. [Figure 15] A conceptual diagram illustrating the relationship between the position on the donor substrate and the cumulative error amount. [Figure 16] A conceptual diagram illustrating the overlap between fault location information D and fault location information R. [Figure 17] A conceptual diagram illustrating how each opposing substrate is lifted by facing its respective divided area. [Figure 18] A conceptual diagram illustrating how each non-opposing substrate division area is lifted by facing it. [Modes for carrying out the invention]

[0067] The embodiments for carrying out the present invention will be described below with specific examples and diagrams. For convenience, the aforementioned conceptual diagrams and other figures may also be used in the following description. [Examples]

[0068] This embodiment 1 describes a lift device in which a total of 49,950 micro-LED elements (30 × 60 [μm] (X axis × Y axis)) are mounted in a matrix on a receptor substrate of the same size, with 222 × 225 elements arranged without defects on a 6-inch donor substrate. The required lift position accuracy for these approximately 50,000 micro-elements mounted on the receptor substrate is ±2 [μm], and the pitch in each axial direction is 450 [μm]. On the donor substrate, the micro-elements are arranged without defects (no missing elements) at a pitch of 1 / 2 times the mounting pitch of the planned lift positions on the receptor substrate, and the total number of these elements is approximately 200,000. The distance (spacing) between adjacent micro-elements is X: 195 [μm], Y: 165 [μm]. In this embodiment, for simplicity, the size of the receptor substrate was set to be the same as the donor substrate, and the arrangement pitch of the micro-elements was set to be the same in both the X and Y axes; however, these are all design considerations.

[0069] First, an example of the external appearance of the lift device according to the present invention is shown in Figure 6A. The configuration shown in this external view is compatible with receptor substrates of 55 inches or larger. A conceptual diagram of the arrangement of the main components is shown in Figure 6B. Note that in Figure 6B, the laser device, various control devices, and mounts for various optical elements are omitted, but the X, Y, and Z axis directions are shown in the figure. The first surface plate (G11, G12) and the second surface plate (G2) are all made of granite. The base surface plate (G) is made of highly rigid iron.

[0070] Furthermore, it is desirable to have a rotation adjustment mechanism between each base plate and between each base plate and each stage for fine-tuning the installation angle (orthogonal / parallel). Specifically, the rotation adjustment mechanism described in Patent Document 1 above is preferred. In addition, it is desirable that the high-magnification camera that monitors the position of each substrate is installed in a location different from the vibration system, such as the stage that holds the substrate.

[0071] The laser device used in this embodiment 1 is an excimer laser with an oscillation wavelength of 248 nm. The spatial distribution of the emitted laser light is approximately 8 × 24 mm, and the beam divergence angle is 1 × 3 mrad. Both are expressed in (length × width) notation, and the values ​​are in FWHM. Excimer lasers have various specifications, including differences in output power, repetition frequency, beam size, and beam divergence angle, as well as the emitted laser light being vertically elongated (the vertical and horizontal orientations reversed). However, by adding, omitting, or modifying the optical system, many excimer lasers can be used in this embodiment. Also, depending on its size, the laser device may be installed on a surface plate different from the foundation on which the stages of the lift device are generally installed.

[0072] The light emitted from the excimer laser enters the telescope optical system and propagates to the zoom homogenizer beyond it. Here, as shown in Figure 6B, the zoom homogenizer is positioned on the first base plate (G11) so that its optical axis is aligned with the X-axis. The laser light immediately before entering the zoom homogenizer is adjusted by the telescope optical system to be approximately parallel, and enters it along the X-axis with approximately the same size regardless of the position of the zoom homogenizer. In this embodiment, the size is approximately 25 × 25 [mm] (Z × Y).

[0073] In this embodiment, each lens array (1, 2) constituting the zoom homogenizer is formed by combining two uniaxial cylindrical lens arrays at a right angle in the YZ plane perpendicular to the optical axis, as shown in the conceptual diagram in Figure 7. The laser light is incident on the first lens array (1) of the initial stage, focuses, passes through the array mask (10) placed near the light source side focal position of the subsequent second lens array (2), and propagates in the order of the second lens array (2) and the condenser lens (3). In this embodiment 1, an array mask was used in which 0.75 mm square apertures were arranged in a matrix. A conceptual diagram of this is shown in Figure 8A. In this conceptual diagram, the positional relationship between the array mask (10) and its apertures (101), which are positioned directly in front of the fly-eye type lens array (1), is shown in order to represent its relationship with the lens array. The appearance of the array mask used in this embodiment is shown in Figure 8B.

[0074] Figure 9 shows a conceptual diagram of the arrangement of each optical element from the zoom homogenizer to the donor substrate. The detailed arrangement is a design matter. The laser light emitted from the zoom homogenizer is scanned by a two-axis scanning mirror (4) and its control device, propagates to the field lens (5), and forms an image on the photomask (6).

[0075] The imaging plane on the photomask (6) is the image plane of the infinity-correcting optical system, which is composed of the second lens array (2) and the condenser lens (3), with the aperture (101) of the array mask (10) as the object plane. The laser beam is scanned by a scanning mirror (4) and propagated toward a selected aperture (61) on the photomask (6), where it forms an image in an irradiation area of ​​a predetermined size. Figure 10A is the beam profile image. This predetermined size is the boundary (outer edge) of the energy distribution that has an energy level above the threshold at which an unintended irradiation target would be induced if irradiated through an adjacent aperture on an unselected photomask. In this embodiment, this is approximately 1 [mm] (FWHM). Figure 10B shows the beam profile on the same photomask (6) when an array mask (10) with a circular aperture shape is used.

[0076] In this embodiment 1, the photomask (6) is made of a synthetic quartz plate on which a pattern has been drawn (applied) by chrome plating. A conceptual diagram is shown in Figure 11. The laser light passes through the white window portion (61) which is not chrome plated and is blocked by the colored portion (62) which is chrome plated. The size of the aperture is 60 × 100 [μm], and 74 of these apertures are arranged in the X-axis direction at a pitch of 600 [μm], and 25 apertures are arranged in the Y-axis direction at the same pitch, for a total of 1850 apertures. The surface to which chrome plating is applied is the laser light emission side, while the incident side is coated with an anti-reflective film for 248 [nm]. Furthermore, aluminum vapor deposition or dielectric multilayer films can be used instead of chrome plating.

[0077] The photomask (6) and array mask (10) are each fixed to a dedicated mount (not shown), which has a total of six-axis adjustment mechanism: the W-axis, U-axis, V-axis which move in the X-axis, Y-axis, and Z-axis directions respectively, the R-axis (θ-axis) which is a rotation axis in the YZ plane (around the optical axis), the TV-axis which adjusts the tilt relative to the V-axis, and the TU-axis which adjusts the tilt relative to the U-axis.

[0078] Furthermore, the laser beam (optical axis) passing through the zoom homogenizer is scanned at high speed across the apertures on the photomask by the scanning mirror (4), and the excimer laser device pulses in synchronization with the timing when its optical axis is scanned to the position of each aperture.

[0079] The laser light that has passed through the photomask pattern propagates through the aperture (7) to the image-side telecentric projection lens (8) with a reduction ratio of 3 / 4, and is then projected onto the donor substrate (91) from its back surface to the position where the micro-elements with a size of 30 × 60 [μm] (X × Y) on the front (bottom) surface are mounted. Figure 12 is a beam profile image of the projected laser light. The micro-elements on the donor substrate that are irradiated by the scanning laser light are successively lifted in a matrix-like manner onto the opposing receptor substrates at a pitch of 450 [μm] and mounted.

[0080] The number of lifted 30 × 60 [μm] irradiated objects corresponds to the number of 74 × 25 apertures on the photomask mentioned above, and the lift range is approximately 33 × 11 [mm].

[0081] Furthermore, the reduction projection lens (8) is image-side telecentric, and in addition to adjustment by a Z-axis driven stage that holds it, it is also possible to add a Z-axis adjustment function for the donor substrate (Z-axis stage (Zd)) to support imaging onto the light absorption layer. However, it is necessary to consider the decrease in lift position accuracy due to the increased load on the donor stage.

[0082] Furthermore, when adjusting the imaging position at the interface between the donor substrate surface and the light-absorbing layer, real-time monitoring using a confocal beam profiler (BP) with an imaging plane that is conjugate to the mask surface and the reduction projection lens is effective. In this embodiment 1, the spatial intensity distribution of the laser light projected at reduction at the interface between the donor substrate surface and the micro-element is monitored in real time and with high resolution.

[0083] The laser beam is scanned across all apertures on the photomask (6), and the laser beam is sequentially irradiated onto the micro-elements arranged on the donor substrate, whose imaging position has been adjusted. As a result, 1850 micro-elements, corresponding to the number of numerical apertures on the photomask, are mounted to the same number of planned lift positions on the receptor substrate. These 1850 lift areas correspond to one area when the receptor substrate is divided into 3 sections in the X-axis direction (A-C) and 9 sections in the Y-axis direction (1-9), for a total of 27 sections (A1-A9, B1-B9, C1-C9). The size of this area is approximately 33 × 11 [mm], determined by the scanning angle of the scanning mirror and the aperture diameter of the reduction projection lens. After the lifting of one area is complete, the donor substrate and receptor substrate are moved to the next lift area. For example, a movement of 33.3 [mm] in the X-axis direction and 11.25 [mm] in the Y-axis direction. Subsequently, another 1850 lifts are performed, and this process is repeated across the entire area where lifts are planned, completing the mounting of 49950 micro-elements from the donor board to the receptor board. As mentioned above, the micro-elements are mounted on the donor board without defects at a high density of half the pitch of the areas where lifts are planned to be placed on the receptor board. Therefore, one donor board can be used to mount elements onto four receptor boards.

[0084] Furthermore, depending on the relationship between the imaging size of each aperture on the donor substrate and its energy density, if the predetermined size (DP) of the laser beam irradiation area on the photomask can be set to a size that allows all four apertures to be irradiated simultaneously, as shown by the dashed line in Figure 1, the time required for implementation can be reduced to approximately 1 / 4.

[0085] Furthermore, as shown in Figure 13, when using a donor substrate that is wafer-shaped and has micro-elements mounted across its entire effective area, it is possible to lift micro-elements located in the corners by setting a lift area (shown by the shaded area in the figure) along the mounting area of ​​the micro-elements and scanning the scanning range of the scanning mirror accordingly.

[0086] The above is a specific example of a lift device and mounting method for mounting a micro-element on a donor substrate onto an opposing receptor substrate. [Examples]

[0087] In this second embodiment, assuming there are no transfer defects, an example of correction by re-transferring to defective areas corresponding to approximately 1% of the 6-inch receptor substrate to be corrected (hereinafter simply referred to as the "receptor substrate" in this second embodiment) is shown using the correction lift device according to the present invention. This substrate consists of 495 × 495 = 245,025 40 [μm] square microelements mounted in a matrix with a pitch of 200 [μm] between adjacent elements. Similarly to the receptor substrate, it is assumed that approximately 1% of the microelements used for re-transferring (correction) are distributed on the 6-inch donor substrate for correction (hereinafter simply referred to as the "donor substrate" in this embodiment). In this second embodiment, the donor substrate has microelements arranged at a pitch of 50 [μm], which is 1 / 4 times the design mounting pitch of the microelements mounted on the receptor substrate, and the total number of microelements is over 3.9 million. Therefore, the spacing between adjacent microelements is 10 [μm]. Furthermore, the general configuration of the scanning reduction projection optical system mounted on the lift device of this embodiment 2 and the structure of the device are the same as in embodiment 1, however, the specifications of each optical element, their placement, and the beam profile shape determined by them are design considerations.

[0088] In this embodiment 2, on the photomask (6), as shown in Figure 1, 200 [μm] square apertures (Ma), which are slightly larger than the shape determined by the magnification (1 / 4x) of the reduction projection lens (8) and the size of the micro-element (40 [μm] square), are arranged in a matrix of 165 x 55 apertures with an 800 [μm] pitch (Pi). This arrangement corresponds to the arrangement of micro-elements mounted on the receptor substrate. The size of the irradiation area imaged onto this photomask by the scanning reduction projection optical system is approximately 1 [mm] square, the same as in this embodiment 1, and is a size that allows irradiation without interference with adjacent apertures on the photomask. The laser light passing through this aperture is imaged towards the micro-elements used for retransfer (correction) on the donor substrate (9) via the image-side telecentric reduction projection lens (8).

[0089] The excimer laser light, which has the irradiation area size and is pulsed in synchronization with the operation of the scanning mirror (4), is irradiated onto a 200 [μm] square aperture on the photomask (6). The laser light that has passed through this aperture passes through an image-side telecentric projection lens (8) with a reduction magnification of 1 / 4 and is irradiated from the back surface of the donor substrate toward the micro-elements arranged thereon, without interfering with adjacent micro-elements at 10 [μm] intervals. In this embodiment, the irradiated pulsed laser light forms an image on the surface (bottom surface) of the donor substrate at a 50 [μm] square area, which is slightly larger than the size of the micro-element, inducing a reaction and lifting the micro-element at that position toward an element-free area on the receptor substrate.

[0090] The following shows specific lifting methods for re-transfer (correction) for each step. (1) Inspection process The positional information of the micro-elements mounted on the donor substrate and the receptor substrate is obtained from both the design position information and the actual mounting position obtained from image processing. The specific coordinates are determined by the centroid position obtained from the shape of the micro-element, and the coordinate origin is determined by referring to the position of the orientation flat on the substrate. Here, the position information on the donor substrate is referred to as "position information D," and the position information on the receptor substrate is referred to as "position information R."

[0091] Location information D includes micro-elements that are deemed defective and should not be used for retransmission, as well as areas where no elements are installed at all. The position coordinates of these are calculated from adjacent micro-elements and obtained as "defective location information D". The same applies to "defective location information R" in location information R.

[0092] Furthermore, in this embodiment, the design pitch of the micro-elements arranged on the donor substrate is 1 / 4 the design pitch of the micro-elements mounted on the receptor substrate, and the actual pitch has errors between the donor substrates and even between the receptor substrates. Therefore, this error is calculated from the position information D and R. In this embodiment, for simplicity, the position information R is taken as the design pitch, and the value of the error δPi between the pitch of the donor substrate and that pitch is +0.0075 [μm].

[0093] (2)Dividing process The area on the 6-inch donor substrate is divided into 27 divided areas ("divided areas D"), similar to Example 1. As shown in Figure 14, each area is 33 × 11 [mm], and for convenience, each divided area is shown in the figure as A1~A9, B1~B9, and C1~C9. The receptor substrate is similarly divided into 27 "divided areas R", and the lift is performed between opposing divided areas.

[0094] The size of this divided area D is a design consideration, similar to Example 1, determined by the effective aperture diameter of the reduction projection lens, the size of the photomask limited by other specifications, and the reduction projection magnification.

[0095] Here, the acceptable range for the cumulative error obtained by multiplying the aforementioned error δPi by the number of 660(-1) objects in the direction of the long axis (X axis) contained within the divided area D was set to ±5 [μm]. As shown in Figure 15, this tolerance range is arbitrarily set based on the limit that the laser beam passing through the aperture (61) on the photomask (6) can illuminate the entire surface of the micro-element (solid line) at the rightmost position where the cumulative error is maximum, with a 50 [μm] square (dashed line in the figure) being the image size formed on a 40 [μm] square micro-element on the donor substrate. The 40 [μm] square of the double-dotted line in the figure represents the mounting position of the micro-element on the donor substrate in the design, and is located in the center relative to the image size of the dashed line. In this embodiment, since the cumulative error amount of any micro-element mounted within the division area D is within the tolerance range (659 × 0.0075 ≈ 4.94) relative to its design position there, it is not necessary to set a reduced "corrected division area D" in this division process.

[0096] (3) Selection process For a divided area R of the same size opposite to divided area D, the micro-elements arranged within divided area D are selectively lifted one-to-one towards the areas without elements (approximately 91 defective location information R), which represent about 1% of the 9075 micro-elements implemented in divided area R, by selecting the micro-elements located opposite to these areas.

[0097] However, of the 660 × 220 = 145,200 micro-elements implemented within the divided area D, approximately 1%, or 1,452, are defective elements, and these (defect location information D) may overlap with the defect location information R. Figure 16 illustrates this. This image shows the receptor substrate being observed through the donor substrate. A portion of the arrangement of micro-elements on the donor substrate is shown. In this figure, the area near the upper left enclosed by the dashed line within the divided area A1 is magnified, illustrating the arrangement of micro-elements on the donor substrate located within this area.

[0098] Here, for convenience, the locations of areas where defective elements have been removed from the receptor substrate by a prior removal process are shown as white rectangles (Mr), the locations of micro-elements that are properly mounted on the receptor substrate are shown as black rectangles (Er), and the locations of micro-elements on the donor substrate, which is densely arranged 16 times more densely than the receptor substrate, are shown as gray rectangles (Ed). (Ed elements that overlap with Er or Mr are omitted from the illustration.) As mentioned above, the pitch of micro-elements Ed on the donor substrate is 50 [μm], and the pitch of micro-elements Er on the receptor substrate is 200 [μm].

[0099] First, we will explain the case where the first receptor board is modified using a donor board that is being used for modification for the first time. (3-1) The defect location information D and defect location information R are compared on a board-by-board basis to pre-confirm locations where the element-free area (Mr) and the defective element location (Md) overlap. The presence or absence of overlap to be confirmed here refers to the overlap on a board-by-board basis (all 27 areas) at the intersection of the β column group aligned in the X-axis direction and the α column group aligned in the Y-axis direction of the donor board shown in the figure (245,025 locations / board in total). Note that the other α', α'', α''', β', β'', and β''' column groups represent the arrangement positions of other minute elements mounted on the donor board at 16 times the density. (In the figure, only the 3 columns from the left of the α column group are shown with arrows, and only the 2 columns from the top of the β column group are shown with arrows. The other "'" column groups are similarly shown with arrows in a limited manner.)

[0100] If there are no duplicates, the selection process involves selecting the position (Ed) of a micro-element on the donor substrate that corresponds to the element-free area (Mr). If duplicates are found, the process will be described later.

[0101] (3-2) On the other hand, in the case of correction using a donor substrate that has been used for correction and a second or subsequent receptor substrate (first or second), the position information of all the defective elements (Md) that originally existed on the donor substrate, plus the position information of the used substrate where minute elements are missing due to past retransfer, is compared with the distribution position information of element-free areas on the receptor substrate (Mr) to check for any overlaps. If there are no overlaps, the position of the minute element on the donor substrate corresponding to the element-free area (Mr) is selected as the selection step.

[0102] (4) Transfer process In the cases described in (3-1) or (3-2) above, after the position of the micro-element on the donor substrate to be used for retransfer is selected, the modification of the first or subsequent receptor substrates is started from the aforementioned divided area A1. The scanning mirror (4) scans the optical axis of the laser beam along the row of β on the donor substrate via the photomask (6) and reduction projection lens (8). Within this area, at the selected position, the micro-element mounted at the selected position on the donor substrate is lifted toward the opposing element-free area (Mr) on the receptor substrate by the excimer laser beam emitted at the timing when the optical axis is scanned.

[0103] (5) Moving process After the correction of area A1 of the receptor board to area A1 by the micro-element at the selected position within area A1 of the donor board is completed, the same correction is performed on the next area A2. The area movement is performed in any order (for example, A1~A9 → B1~B9 → C1~C9) by moving the stage that holds each board, and corrections are performed on all areas. The amount of stage movement when moving each divided area is set to offset the aforementioned cumulative error amount (approximately +4.95 [μm]). After the correction of all divided areas that require correction is completed, this receptor board is replaced with the next receptor board to be corrected.

[0104] On the other hand, if duplication is found during the substrate-level matching in (3-1) or (3-2) above, the column group to be matched is changed from α to α' column group, or from β to β' column group, and the combination of the changed column groups (the intersection of the changed column groups) is checked for duplication at the substrate level in the same way as in (3-1) or (3-2) above. In this case, there are 15 possible combinations of column groups. If a combination of column groups without duplication is found at the substrate level, the position (Ed) of the micro-element on the donor substrate corresponding to the element-less area (Mr) is selected based on that combination. Before the transfer process, the substrate is moved according to the combination of column groups. Figure 17 shows the position of the substrates during re-transfer between each divided area (A1).

[0105] Furthermore, in the substrate-level matching described in (3-1) or (3-2) above, if no combination of column groups without overlap is found in any of the 15 possible combinations, the matching is performed at a narrower, opposing divided area level rather than a broader substrate level. For example, the matching is limited to area A1 of the receptor substrate and area A1 of the donor substrate opposite it. If no overlap is found at the intersection of the α and β column groups, the process proceeds directly to the transfer step. If there is an overlap, other combinations of column groups without overlap (e.g., α' column group and β'' column group) are searched for, at least within the divided area (A1) of each opposing substrate. There are up to 16 such combinations. After the correction of divided area A1 of the receptor substrate by divided area A1 of the donor substrate is completed using the found combination of column groups without overlap, the same matching process is performed for the next area A2, and area A2 of the receptor substrate is corrected using the combination of column groups without overlap. From this point onward, the process is repeated: donor A3 to receptor A3, donor A4 to receptor A4, and so on. Note that if, for example, there is overlap only between divided areas B5, and no overlap exists on a board-by-board basis outside of this area, then area-specific matching before and after this point is unnecessary.

[0106] In the aforementioned matching of opposing divided areas (A1 and A1, B1 and B1, etc.), if no combination of non-overlapping column groups is found, the matching of areas on the donor substrate is then expanded from matching only with opposing divided areas to matching with the other 26 divided areas on the donor substrate, and non-overlapping column group combinations are searched for on an area-by-area basis. Figure 18 illustrates this process. Here, for example, in the combination of divided area C5 on the donor substrate and divided area A3 on the receptor substrate, re-transfer is performed based on the intersection position of the α''' group and the β''' group. However, it should be noted that non-overlapping column group combinations must be searched between the divided areas, taking into account the defective position information (Md) that reflects the missing positions of the minute elements used for correction up to the point in time when the divided area to be matched was changed. In addition, it is necessary to consider (calculate) whether a reasonable cycle time can be achieved by comprehensively taking into account the time required for frequent movement of column groups and stage movement for moving to the opposing divided area. Based on the calculation results, the appropriate selection of the matching range for duplicates and the determination of when to replace the donor board are made.

[0107] Furthermore, as long as direct irradiation of the receptor substrate with laser light does not have adverse effects, the transfer process can be performed regardless of whether or not there is duplication of defective location information. For example, areas without elements that were not re-transferred due to duplication can be left to subsequent corrections using different donor substrates. Moreover, it is possible to combine substrate-by-substrate verification and division-area verification in a timely manner based on cycle time simulations. In addition, it is possible to determine the combination of verification methods and their ranges that have the shortest simulated cycle time based on the position information of each substrate acquired in advance during the inspection process.

[0108] As in this embodiment, when retransferring for each divided area R, that is, when scanning a scanning mirror with a scanning speed of approximately 30 units / second for each numerical aperture (165 × 55 units) on the photomask and retransferring, assuming there is no overlap of defective positions between substrates, the time required to correct (approximately 90 locations) per divided area is about 3 seconds. Even considering the time required to do this for 27 divided areas, as well as the time required to move each substrate to move between divided areas, the time required from when the substrate is set until the correction is completed is roughly 90 seconds.

[0109] Compared to the approximately 2450-second cycle time when using the aforementioned conventional device, which requires approximately 1 second to re-transfer a single micro-element, this embodiment enables the correction of defective elements up to 1% at an overwhelmingly faster speed. Furthermore, even considering the calculation time required for verification (duplicate confirmation) and the stage movement time required for the column group movement and divided area movement to avoid duplication, the overwhelming advantage in cycle time in this embodiment can be maintained.

[0110] Furthermore, when randomly matching divided areas, not limited to opposing divided areas, or when considering the cumulative error amount, it is advisable to optimize each process by using a simulation program with position information D and R acquired in the inspection process, the size of the divided area D, and the allowable range of the cumulative error amount as parameters, to determine the size of the corrected divided area D, the combination of movement amounts for each stage, and the execution order of each process so as to minimize the time required for mounting or retransferring the entire receptor substrate. [Industrial applicability]

[0111] It can be used in part of the manufacturing process for microLED displays. [Explanation of symbols]

[0112] 1. First lens array 10 Array Masks 101 Opening (on array mask) 2. Second lens array 3. Condenser lens 4 Scanning mirror 5 Field Lenses 6 Photomasks 61. Aperture (on photomask) 62 Shielding portion of the photomask 7. Aperture (Entrance Pupil) 8. Reduction projection lens (image-side telecentric) 91 Donor board 92 Receptor boards BP Beam Profiler CCD High Magnification Camera Laser beam irradiation area (predetermined size) on a DP photomask ED position of micro-elements on donor substrate Position of micro-elements on Er receptor substrate G, G11, G12, G2 (Foundation, 1st (Column), 1st (Beam), 2nd) Surface Plate La laser light Ma Aperture size on photomask Location of defective elements on the Md donor board Element-free areas on the Mr Receptor board Pitch of apertures arranged on the Pi photomask Scanner accuracy on St photomask Xd, Xr (for donor and receptor substrates) X-axis stage Yd, Yr (for donor and receptor substrates) Y-axis stage Zd, Zr (for donor and receptor substrates) Z-axis stage α, β (along the Y and X axes) rows of micro-elements arranged on the donor substrate. δPi Pitch error of elements mounted on the substrate θd, θr (for donor and receptor substrates) θ-axis rotation stage

Claims

1. A method for lifting an object to be irradiated from a donor substrate onto an opposing receptor substrate, An inspection step to acquire position information D of the object to be irradiated on the donor substrate, and position information R which is the planned lift position of the object to be irradiated on the receptor substrate, A division step of dividing the region on the donor substrate into a predetermined size division area D, A selection step of selecting the position of the object to be irradiated to be lifted within the divided area D based on the position information D and the position information R, A transfer step in which the selected object to be irradiated is lifted from the divided area D to the opposing divided area R by laser light irradiated through an aperture on a photomask that is opposite to the position of the selected object to be irradiated, After the transfer step, a transfer step is performed to move the donor substrate and / or the receptor substrate, A lifting method, including...

2. The lifting method according to claim 1, wherein the transfer step and the transfer step are repeated a plurality of times.

3. The lift method according to claim 1, wherein the design mounting pitch of the irradiated object on the donor substrate is 1 / 1 or more integer fractions of the design mounting pitch of the irradiated object mounted on the receptor substrate.

4. When there is an error between the actual mounting pitch of the object to be irradiated, calculated from the position information D, and the design mounting pitch R, calculated from the position information R, The lift method according to claim 3, wherein the amount of movement of each substrate in the moving step is an amount of movement that cancels out an accumulated error amount corresponding to the number of objects to be irradiated contained within the divided area D.

5. If the cumulative error amount exceeds an arbitrary tolerance range with the distance between adjacent irradiated objects on the donor substrate as the upper limit, The division process is a division process that reduces the size of the division area D and makes it a modified division area D. The lift method according to claim 4, wherein the amount of movement of each substrate in the moving step is the amount of movement that cancels out the amount of error accumulated within the correction division area D.

6. The lift method according to claim 5, wherein the size of the modified divided area D, the combination of movement amounts for each stage, and the execution order of each process are determined by a simulation program using the position information D, the position information R, the size of the divided area D, and the tolerance range as parameters.