Hybrid bonding equipment and hybrid bonding method based on visual alignment

Abstract

The invention discloses a hybrid bonding device and a hybrid bonding method based on visual alignment, and belongs to the field of bonding processing equipment.The hybrid bonding device comprises a bonding assembly, a first displacement mechanism, a second displacement mechanism, a first visual assembly, an optical calibration assembly, a second visual assembly and a bearing table; according to the technical scheme of the utility model, the position coordinates of the calibration point, the chip to be bonded and the mark point on the target chip can be accurately obtained through the corresponding arrangement of the optical calibration assembly, so that the compensation position information between the two chips can be accurately obtained, and the visual alignment of the two chips before hybrid bonding can be realized. The hybrid bonding equipment is compact in structure and convenient to use, bonding of the to-be-bonded chip and the target chip on the wafer can be accurately completed, visual alignment between the to-be-bonded chip and the target chip is accurately achieved before hybrid bonding, the alignment precision during hybrid bonding of the chips is effectively guaranteed, errors in the bonding process are reduced, and the bonding efficiency is improved. And the quality of a chip product is improved.

Description

Technical Field

[0001] The invention belongs to the field of bonding processing equipment, and in particular relates to a hybrid bonding equipment and a hybrid bonding method based on visual alignment. Background Art

[0002] Hybrid bonding technology is a new type of semiconductor packaging technology that can integrate homogeneous and heterogeneous materials to achieve vertical stacking, thereby improving chip performance. Compared with traditional thermal compression bonding technology, hybrid bonding technology does not require solder bumps, can achieve higher density interconnection, and reduce chip power consumption.

[0003] In general, the hybrid bonding process can be divided into three process stages: surface treatment, alignment bonding, and post-processing. Among them, the surface treatment process uses plasma or acid and alkali to clean and activate the surface of the chip and wafer to enhance adhesion. Alignment bonding is to complete the high-precision alignment of the chip and the wafer to ensure that the pads correspond one by one, and then directly bond through van der Waals forces, the forces between molecules and atoms under a certain pressure. The post-processing process is to strengthen the bonding through annealing, and the annealing temperature is usually between 200 and 300 ° C. In the entire process flow, alignment accuracy is a key indicator that affects the electrical connection and reliability of the chip. Once there is a problem with the alignment accuracy, it will greatly affect the yield and cause huge losses. Therefore, how to accurately and effectively realize the alignment operation in the bonding process has become the focus of researchers, and it is also the first problem to be solved when wanting to improve the efficiency and yield of hybrid bonding processing. Summary of the invention

[0004] In response to one or more of the above-mentioned defects or improvement needs of the prior art, the present invention provides a hybrid bonding device and a hybrid bonding method based on visual alignment, which can accurately realize visual alignment during the hybrid bonding operation, ensure the alignment accuracy of the two objects before hybrid bonding, and improve the quality of hybrid bonding.

[0005] To achieve the above-mentioned object, one aspect of the present invention provides a hybrid bonding device based on vision alignment, which includes a bonding component, a first displacement mechanism, a second displacement mechanism, a first vision component, an optical calibration component, a second vision component and a carrier; The carrying platform is used to carry the target chip, and is spaced apart from the second visual component in the first direction; The bonding assembly includes a bonding head connected to a pressure motor, which is used to take the chip to be bonded and complete the mixed bonding of the chip to be bonded and the target chip; The first displacement mechanism includes a horizontal displacement component, and the second displacement mechanism includes a vertical lifting component and a plane adjustment component; the first visual component and the vertical lifting component are respectively assembled and connected with the horizontal displacement component, so that the horizontal displacement component can drive the bonding component and the first visual component to synchronously displace in the first direction; the bonding component is assembled and connected with the vertical lifting component through the plane adjustment component, and the bonding component can be lifted and lowered by the vertical lifting component, and the bonding component can be adjusted in the XYθ direction relative to the first visual component in the plane by the plane adjustment component; The first visual component is arranged on one side of the bonding component, and includes a camera module and a semi-transmissive and semi-reflective mirror and a reflective mirror arranged in sequence on the optical path of the camera module, so that the shooting optical path of the camera module can obtain a first detection light and a second detection light that are parallel to each other and vertically downward after the action of two lenses; the second visual component includes a camera module for generating a third detection light that is vertically upward; The optical calibration component is arranged on a side of the bonding component close to the first visual component, and includes a black film layer and a light-transmitting layer arranged in sequence in a first direction, and a calibration point is arranged at the bottom of the light-transmitting layer; the position information of the calibration point and the first marking point on the chip to be bonded is respectively obtained through the second visual component, and the position information of the calibration point and the second marking point on the target chip is respectively obtained through the first visual component, and the compensation position information between the first marking point and the second marking point is obtained by coordinate transformation, and finally the plane adjustment component adjusts the chip to be bonded and the target chip to be aligned according to the compensation position information to complete the visual alignment before hybrid bonding.

[0006] As a further improvement of the present invention, the bottom of the optical calibration component protrudes from the bottom of the bonding head, and the protruding distance of the optical calibration component is equal to the thickness of the chip to be bonded, so that the first marking point and the calibration point are located in the same horizontal plane.

[0007] As a further improvement of the present invention, the bottom of the optical calibration component is flush with the bottom of the bonding head, and the depth of field range of the camera module in the second visual component is not less than the thickness of the chip to be bonded.

[0008] As a further improvement of the present invention, the thickness of the black film layer is equal to that of the light-transmitting layer; and / or The optical calibration component is movably connected to the side wall surface of the bonding head via a displacement adjustment mechanism, and the displacement adjustment mechanism is used to adjust the vertical position of the optical calibration component relative to the bonding head; and / or The thickness of the optical calibration component in the first direction is 0.5 to 3 times the opening width of the first visual component lens barrel.

[0009] As a further improvement of the present invention, the optical calibration component has its two ends flush with the two side end surfaces of the bonding head in the second direction, and the second direction is a horizontal direction perpendicular to the first direction.

[0010] As a further improvement of the present invention, both the first visual component and the second visual component are two spaced apart in the second direction; and the calibration points are two spaced apart in the second direction.

[0011] As a further improvement of the present invention, the bonding head is a suction cup with negative pressure adsorption capability, and a plurality of adsorption holes are arranged at intervals on the bottom surface thereof.

[0012] Another aspect of the present invention provides a hybrid bonding method based on vision alignment, which uses the hybrid bonding device based on vision alignment to complete the bonding between the chip to be bonded and the target chip, and includes the following process: (1) Controlling the bonding head to pick up the chip to be bonded, and loading the target chip onto the carrier, so that the first marking point on the chip to be bonded faces downward and the second marking point on the target chip faces upward; (2) controlling the first displacement mechanism to operate, driving the bonding component and the first vision component to move synchronously, so that the bonding component moves to above the second vision component; (3) controlling the second visual component to operate, so that the second visual component respectively obtains the position coordinates of the calibration point and the first marking point in the second coordinate system; (4) Controlling the first displacement mechanism to drive the first vision component and the bonding component to move synchronously to the top of the target chip; controlling the second displacement mechanism to drive the bonding component to translate relative to the first vision component, and marking the black film layer on the optical path of the reflector, and marking the second marking point on the target chip on the optical path of the semi-transmitting and semi-reflecting mirror; (5) controlling the first visual component to obtain the position coordinates of the second marking point in the first coordinate system; (6) Controlling the second displacement mechanism to continue to drive the bonding component to translate relative to the first vision component, calibrating the black film layer to the optical path of the semi-transmitting and semi-reflecting mirror, and calibrating the calibration point to the optical path of the reflector; continuing to control the first vision component to work, and obtaining the position coordinates of the calibration point in the first coordinate system; (7) performing coordinate conversion between the second coordinate system and the first coordinate system, obtaining a compensation position difference between the first marking point and the second marking point, and controlling the second displacement mechanism to perform a compensation movement according to the compensation position difference, so that the chip to be bonded is vertically aligned with the target chip; (8) Control the pressure motor in the bonding assembly to press the bonding head down so that the chip to be bonded fits tightly with the aligned target chip until the bonding process is completed.

[0013] As a further improvement of the present invention, in process (1), after the bonding head absorbs the chip to be bonded, the position of the optical calibration component is calibrated such that the black film layer is aligned with the second detection light emitted by the second reflector, and the first detection light emitted by the semi-transmissive and semi-reflective mirror passes over the optical calibration component and is emitted downward; and / or In process (1), the bottom of the optical calibration component is aligned with the bottom of the chip to be bonded, and the vertical distance between the calibration point and the target chip is calibrated to be equal to the spacing between the two detection lights in the first visual component.

[0014] As a further improvement of the present invention, in process (7), the coordinate transformation between the two coordinate systems is completed by the following transformation matrix formula:

[0015] In the formula, X and Y describe the origin of the second coordinate system O name2 Position coordinates in the first coordinate system; It is defined as the rotation matrix of the second coordinate system relative to the first coordinate system, which is used to represent the rotation angle of the X-axis and the Y-axis; R 11 , R 12 , R 21 , R 22 The unknown quantity is solved by calibrating the coordinates of the vector in the two coordinate systems.

[0016] The above-mentioned improved technical features can be combined with each other as long as they do not conflict with each other.

[0017] In general, compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects: (1) The hybrid bonding device based on vision alignment of the present invention comprises a bonding component, a first displacement mechanism, a second displacement mechanism, a first vision component, an optical calibration component, a second vision component and a carrier. Through the mutual cooperation of the two vision components and the two displacement mechanisms, and the corresponding setting of the optical calibration component, the position coordinates of the calibration points, the chip to be bonded and the marking points on the target chip can be accurately obtained, so as to accurately obtain the compensation position information between the two chips, thereby realizing accurate alignment of the two chips before hybrid bonding, ensuring the alignment accuracy of the chips before bonding, avoiding too low a yield rate due to equipment control errors, and reducing the hybrid bonding cost of the chips.

[0018] (2) The hybrid bonding device based on vision alignment of the present invention, by preferably setting the optical calibration component and the relative position between the optical calibration component and the first vision component, ensures the accuracy of obtaining the calibration point and the second marking point position based on the first vision component, and the accuracy of obtaining the calibration point and the first marking point position based on the second vision component, thereby avoiding the alignment error caused by the focus difference of the camera, and further improving the accuracy and reliability of the vision alignment.

[0019] (3) The hybrid bonding device based on visual alignment of the present invention further improves the accuracy of visual alignment between the chip to be bonded and the target chip, thereby improving the bonding quality of the chip, by preferably setting two sets of first visual components and two sets of second visual components, and setting two calibration points on the optical calibration component. By obtaining the corresponding positions of the two sets of first marking points and the two sets of second marking points, the accuracy of visual alignment between the chip to be bonded and the target chip is further improved.

[0020] (4) The hybrid bonding method based on visual alignment of the present invention is completed based on a combined hybrid bonding device. The steps are simple and the control is convenient. By acquiring the corresponding positions of the calibration point and the two marking points and converting the coordinates of each point, the chip to be bonded and the target chip can be accurately aligned before bonding, thereby avoiding the bonding error introduced by the inaccurate alignment of the two chips, improving the quality of chip bonding, and reducing the scrap rate in the chip bonding process.

[0021] (5) The hybrid bonding equipment based on visual alignment in the present invention has a compact structure and is easy to use. It can accurately complete the bonding of the chip to be bonded with the target chip on the wafer, and accurately realize the visual alignment between the chip to be bonded and the target chip before hybrid bonding, effectively ensuring the alignment accuracy during chip hybrid bonding, reducing equipment errors in the bonding process, improving the quality and yield of chip products, and having good economic value. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a schematic structural diagram of a hybrid bonding device based on visual alignment in an embodiment of the present invention; Figure 2 is a schematic diagram of positioning using a second visual component in an embodiment of the present invention; Figure 3 is a schematic diagram of performing a positioning operation using a first visual component in an embodiment of the present invention; Figure 4is a schematic diagram of performing secondary positioning using a first visual component in an embodiment of the present invention; Figure 5 is a structural side view of a hybrid bonding device based on visual alignment in an embodiment of the present invention; In all the drawings, the same reference numerals represent the same technical features, specifically: 1. Bonding assembly; 2. First displacement mechanism; 3. Second displacement mechanism; 4. First visual assembly; 5. Optical calibration assembly; 6. Second visual assembly; 7. Chip to be bonded; 8. Target chip; 9. Wafer; 10. Carrier; 101. bonding head; 102. pressure sensor; 103. pressure motor; 201, translation motor; 202, translation guide rail; 203, fixing plate; 301, lifting motor; 302, lifting guide rail; 401, camera module; 402, first reflector; 403, semi-transmissive and semi-reflective mirror; 404, second reflector; 501, black film layer; 502, light-transmitting layer; 503, calibration point; 701, first marking point; 801, second marking point. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0025] In the description of the present invention, it should be understood that, unless otherwise expressly specified and limited, the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like to indicate orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0026] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0027] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0028] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0029] Example: See also Figure 1~Figure 5 In a preferred embodiment of the present invention, a hybrid bonding device based on visual alignment includes a bonding component 1, a first visual component 4, an optical calibration component 5, a second visual component 6 and a carrier 10.

[0030] The bonding assembly 1 is used to pick up the chip 7 to be bonded and perform the bonding process; the carrier 10 is used to carry the target chip 8 , specifically, to carry the wafer 9 carrying a plurality of target chips 8 .

[0031] At the same time, the optical calibration component 5 in the preferred embodiment is arranged on one side of the bonding component 1, and is used to cooperate with the first visual component 4 and the second visual component 6 to complete the visual alignment between the first marking point 701 on the chip to be bonded 7 and the second marking point 801 on the target chip 8, thereby ensuring the accurate alignment between the chip to be bonded 7 and the target chip 8 when the subsequent bonding component 1 performs mixed bonding, thereby avoiding the occurrence of defective products due to the existence of alignment errors.

[0032] More specifically, the first visual component 4 in the preferred embodiment is arranged on one side of the bonding component 1, and the detection light output from the lens of the first visual component 4 is two parallel light beams emitted downward, that is, Figure 1 The negative direction of the Z axis is shown in .

[0033] Correspondingly, the second visual component 6 is arranged on one side of the carrier 10 along the first direction, that is, Figure 1 The third detection light outputted from the lens of the second visual component 6 is outputted upward, that is, Figure 1 The positive direction of the Z axis is shown in .

[0034] Furthermore, the optical calibration component 5 in the preferred embodiment includes a black film layer 501 and a light-transmitting layer 502 stacked in the X-axis direction, and a calibration point 503 is arranged at the bottom of the light-transmitting layer 502. The calibration point 503 is imaged in the first visual component 4 and the second visual component 6 to achieve visual alignment between the optical calibration component 5 and the two visual components.

[0035] In order to accurately realize the visual alignment process and the hybrid bonding process, a first displacement mechanism 2 and a second displacement mechanism 3 are also provided in the preferred embodiment. Among them, the first displacement mechanism 2 is an X-axis direction displacement mechanism, which includes a horizontal displacement component, which is used to drive the bonding component 1 and the first visual component 4 to perform horizontal displacement in the first direction at the same time, so as to realize the alignment of the bonding component 1 with the second visual component 6 and the carrier 10. At the same time, the second displacement mechanism is set corresponding to the bonding component 1, including a vertical lifting component and a plane adjustment component, which is used to drive the bonding component 1 to lift and lower and perform XYθ adjustment on the plane relative to the first visual component 4, so as to ensure that the chip to be bonded 7 is accurately aligned with the target chip 8.

[0036] In actual setting, marking points are respectively set on the bottom surface of the chip to be bonded 7 and the top surface of the target chip 8 after the bonding component 1 picks up the material, namely the aforementioned first marking point 701 and the second marking point 801. Then, the second visual component 6 obtains the position coordinates of the calibration point 503 and the first marking point 701, and then the first visual component 4 obtains the position coordinates of the calibration point 503 and the second marking point 801. By using the position of the calibration point 503 as the conversion reference, the position conversion of the first marking point 701 and the second marking point 801 in the same three-dimensional coordinate system is performed, and finally the compensation position information between the first marking point 701 and the second marking point 801 is obtained, and accordingly, the position of the chip to be bonded 7 is adjusted to align with the corresponding target chip 8 before hybrid bonding.

[0037] Specifically, the bonding assembly 1 in the preferred embodiment includes a bonding head 101, the top of which is connected to the output shaft of a pressure motor 103, and a pressure sensor 102 is provided corresponding to the connection between the two. The pressure sensor 102 can accurately obtain the pressure applied by the bonding head 101 during bonding operations.

[0038] In more detail, in order to accurately pick up the chip 7 to be bonded, the bonding head 101 is preferably configured as a suction cup with negative pressure adsorption capability, and a plurality of adsorption holes are arranged at intervals on the bottom surface thereof.

[0039] Furthermore, the first displacement mechanism 2 in the preferred embodiment includes a translation guide rail 202 extending along the first direction, and a translation motor 201 and a fixing plate 203 are provided corresponding to the translation guide rail 202. The translation motor 201 is assembled and connected with the translation guide rail 202, and can be reciprocated in the first direction relative to the translation guide rail 202 during operation. At the same time, the fixing plate 203 is fixedly connected to the translation motor 201, and can be synchronously displaced with the translation motor 201 when the translation motor 201 is translated.

[0040] Correspondingly, the first visual component 4 in the preferred embodiment is mounted on the fixed plate 203 , and the bonding component 1 is disposed on the fixed plate 203 via the second displacement mechanism 3 .

[0041] The second displacement mechanism 3 in the preferred embodiment includes a lifting rail 302 extending vertically and a lifting motor 301 mounted on the lifting rail 302. The vertical displacement of the bonding head 101 can be accurately achieved through the drive of the lifting motor 301 and the guidance of the lifting rail 302.

[0042] Meanwhile, the second displacement mechanism 3 also includes a plane adjustment component (not shown in the figure) arranged corresponding to the bonding component 1 and the first visual component 4 , which is used to realize the first direction displacement of the bonding component 1 relative to the first visual component 4 .

[0043] Exemplarily, in actual settings, the aforementioned plane adjustment component is preferably arranged between the lifting motor 301 and the pressure motor 103, and is further preferably an XYθ displacement adjustment module, which can perform displacement adjustment within the XY axis plane and θ angle adjustment around the plane Z axis as shown in the figure, thereby completing the application of compensation position information.

[0044] By combining the first displacement mechanism 2 and the second displacement mechanism 3, the first visual component 4 can be synchronously displaced following the bonding component 1 during the visual alignment process, ensuring that the relative position between the first visual component 4 and the bonding head 101 can be maintained at a smaller distance, thereby reducing the displacement adjustment amount of the bonding component 1 relative to the first visual component 4 during the visual alignment process and the compensation position adjustment process.

[0045] Furthermore, in the preferred embodiment, the first visual component 4 includes a camera module 401 and a corresponding visual alignment light path. The camera module 401 in the preferred embodiment includes a camera, a lens and a light source, and is capable of emitting detection light into the visual alignment light path. At the same time, a semi-transmitting and semi-reflecting mirror 403 and a second reflector 404 are sequentially arranged in the visual alignment light path. The two mirrors are spaced apart in the first direction, so that the detection light emitted by the camera module 401 can obtain a downward first detection light beam and a transmitted light beam directed toward the second reflector 404 after passing through the semi-transmitting and semi-reflecting mirror 403. After the transmitted light is reflected by the second reflector 404, a second detection light is obtained which is spaced apart from the first detection light by a certain distance and is parallel to the first detection light.

[0046] In actual setting, the interval between the first detection light and the second detection light ( Figure 4 W3 in the figure corresponds to the size of the optical calibration component 5 in the first direction (the X-axis direction in the figure), ensuring that the two detection lights can be simultaneously irradiated onto the optical calibration component 5 under test requirements.

[0047] At the same time, in the preferred embodiment, a first reflector 402 is further preferably provided between the camera module 401 and the semi-transmissive and semi-reflective mirror 403 for reflecting the detection light beam from the camera module 401 to the semi-transmissive and semi-reflective mirror 403, so as to facilitate the arrangement of the camera module 401.

[0048] In more detail, the second reflector 404 in the preferred embodiment is disposed on a side of the semi-transmissive and semi-reflective mirror 403 close to the bonding assembly 1 , that is, the second detection light at this time is located between the first detection light and the bonding head 101 .

[0049] Furthermore, the optical calibration component 5 in the preferred embodiment includes a black film layer 501 and a light-transmitting layer 502 stacked in a first direction, and the center distance between the two layers is further preferably equal to the distance between the first detection light and the second detection light.

[0050] At the same time, a calibration point 503 is arranged on the bottom surface of the light-transmitting layer 502, so that the corresponding detection light beam (the second detection light in the preferred embodiment) can pass through the light-transmitting layer 502 and complete the shooting and imaging of the calibration point 503, thereby obtaining the position information of the calibration point 503 in the corresponding coordinate system.

[0051] In more detail, the optical calibration component 5 in the preferred embodiment is arranged on the side of the bonding head 101 close to the first visual component 4, and the bottom of the optical calibration component 5 protrudes from the bottom surface of the bonding head 101, and the protruding distance is further preferably equal to the thickness of the chip to be bonded 7, so that after the bonding head 101 completes the material collection of the chip to be bonded 7, the calibration point 503 and the first marking point 701 are on the same horizontal plane.

[0052] Based on the above-mentioned setting of the optical calibration component 5, it is also convenient for the second visual component 6 to quickly and accurately complete the positioning and shooting of the calibration point 503 and the first marking point 701 when performing visual positioning.

[0053] For the second visual component 6 in the preferred embodiment, it is preferably a camera component, including a camera, a lens and a light source, and its configuration can be further preferably the same as the camera module 401.

[0054] Of course, according to the actual setting requirements, the optical calibration component 5 can also be preferably set to a form where the bottom is flush with the bottom of the bonding head 101. In this case, the second visual component 6 preferably selects a camera component with a depth of field range not less than the thickness of the chip to be bonded 7. For this method, the installation difficulty of the optical calibration component 5 can be reduced to a certain extent, and the impact of the optical calibration component 5 on other chips other than the target chip 8 during subsequent hybrid bonding can be reduced.

[0055] For example, in a more specific preferred embodiment, the optical calibration component 5 and the side wall surface of the bonding head 101 are preferably movably connected via a displacement adjustment mechanism, and the vertical position of the optical calibration component 5 relative to the bonding head 101 can be adjusted as needed.

[0056] During the actual alignment, after the chip 7 to be bonded is taken out, the position of the bonding component 1 in the first direction is preferably adjusted to the following position: The optical calibration component 5 does not block the first detection light reflected by the semi-transmissive and semi-reflective mirror 403 , and the second detection light output after being reflected by the second reflector 404 is emitted toward the black film layer 501 .

[0057] At the same time, the vertical distance H1 between the chip to be bonded 7 and the target chip 8 is adjusted to be equal to the distance W3 between the semi-transmissive and semi-reflective mirror 403 and the second reflector 404. This arrangement ensures that the optical paths of the first detection light for photographing the second marking point 801 and the second detection light for photographing the calibration point 503 are equal.

[0058] By setting the optical calibration component 5 so that the bottom is flush with the chip to be bonded 7, and setting the distance between the chip to be bonded 7 and the target chip 8 to be equal to the spacing between the two detection light beams of the first visual component 4 after being emitted, the alignment focus of each detection light beam at the desired shooting point is consistent, thereby reducing the visual accuracy error caused by inconsistent focus.

[0059] More preferably, in actual setting, the thickness W2 of the optical calibration component 5 in the first direction is preferably 0.5 to 3 times the width W1 of the lens barrel opening of the first visual component 4 .

[0060] Exemplarily, in the preferred embodiment, W1 is 1-10 mm, and W2 is 0.5-30 mm.

[0061] Further preferably, the thickness of the black film layer 501 and the light-transmitting layer 502 in the optical calibration component 5 are preferably equal, and the thickness of each is preferably 0.25 mm to 15 mm.

[0062] For example, in actual configuration, the length dimension of the optical calibration component 5 in the second direction is preferably equal to the dimension of the bonding head 101 in the second direction, and the two ends of the optical calibration component 5 along the second direction are respectively flush with the two end surfaces of the bonding head 101, such as Figure 5 It should be noted that the second direction is a horizontal direction perpendicular to the first direction, that is, Figure 1 The Y-axis direction shown in .

[0063] In actual operation, in a preferred embodiment, a first coordinate system {name1} is established based on the bonding component 1 and the first visual component 4 , and a second coordinate system {name2} is established based on the second visual component 6 .

[0064] Based on this, set The matrix is ​​used to transform the vector described in the second coordinate system {name2} to the description in the first coordinate system {name1}. In the two-dimensional coordinate system, Preferably, it is a 3×3 matrix.

[0065] The transformation matrix formula is as follows:

[0066] Where X and Y describe the origin of the {name2} coordinate system O name2 Position coordinates in the {name1} coordinate system; It is defined as the rotation matrix of the {name2} coordinate system relative to the {name1} coordinate system, which is used to represent the rotation angles of the X-axis and the Y-axis and describes the posture of the {name2} coordinate system relative to the {name1} coordinate system; R 11 , R 12 , R 21 , R 22 is an unknown quantity, which can be obtained by calibrating the vector P 1. P 2 Solve the coordinates in the {name1} and {name2} coordinate systems.

[0067] Through the design of the above-mentioned conversion matrix formula, after the two visual components complete the acquisition of the position information of each point, the conversion of the coordinate positions in the two coordinate systems can be accurately completed, and then the compensation position information between the second marking point 801 and the first marking point 701 can be accurately obtained, thereby completing the visual alignment of the chip 7 to be bonded relative to the target chip 8.

[0068] For example, if a vector is known P The description in the second coordinate system {name2} is name2 P , at the same time, the known transformation matrix , then the vector P Description in the first coordinate system {name1}: name1 P = name2 P For the two visual components, the position coordinates of the calibration point 503 and the corresponding marking point in the two coordinate systems can be accurately obtained, and then a vector based on the bright spot can be obtained in each of the two coordinate systems; in the design based on the aforementioned transformation matrix formula, the position compensation information between the first marking point 701 and the second marking point 801 can be accurately obtained to complete the alignment between the chip to be bonded 7 and the target chip 8.

[0069] After the alignment is completed, the bonding head 101 is controlled to press down. When the pressure reaches 1-5N, the vacuum is broken to tightly bond the two chips together, thus completing the direct bonding process.

[0070] In a preferred embodiment, after the wafer 9 and the target chip 8 are loaded on the carrier 10, the setting position of the carrier 10 remains fixed, and the relative position between the second visual component 6 and the carrier 10 is fixed. However, in actual setting, it is also preferred to set a displacement mechanism for the carrier 10 to achieve the adjustment of the position of the carrier 10 in the entire device and the relative position between the carrier 10 and the second visual component 6.

[0071] In addition, it is also preferred to provide a displacement mechanism for the second visual component 6, which will not be elaborated here.

[0072] As another aspect of the present invention, based on the aforementioned hybrid bonding device, a hybrid bonding method based on visual alignment is further proposed, which preferably includes the following process: (1) Controlling the bonding head 101 to correspondingly absorb the chip to be bonded 7 so that the first marking point 701 on the chip to be bonded 7 faces downward; accordingly, loading a wafer 9 having a plurality of target chips 8 onto the carrier 10; When performing process (1), the optical calibration component 5 is preferably calibrated to the following position: the black film layer 501 is aligned with the second detection light emitted by the second reflector 404, and the first detection light emitted by the semi-transmissive semi-reflective mirror 403 passes over the optical calibration component 5 and is emitted downward; At the same time, it is preferred to align the bottom of the optical calibration component 5 with the bottom of the chip 7 to be bonded, and calibrate the vertical distance between the calibration point 503 and the target chip 8 to be equal to the spacing between the two detection beams in the first visual component 4 .

[0073] (2) Control the first displacement mechanism 2 to drive the bonding component 1 to move above the second visual component 6, such as Figure 2 As shown in ; at this time, the first visual component 4 moves synchronously with the bonding component 1; (3) Controlling the second visual component 6 to work, so that the second visual component 6 respectively obtains the position coordinates of the calibration point 503 and the first marking point 701 in the second coordinate system; (4) Controlling the first displacement mechanism 2 to work, driving the first visual component 4 to move to above the target chip 8; at this time, the bonding component 1 moves synchronously with the first visual component 4; (5) Control the second displacement mechanism 3 to drive the bonding component 1 to translate relative to the first vision component 4, and mark the black film layer 501 on the optical path of the second reflector 404, and mark the second marking point 801 on the target chip 8 on the optical path of the semi-transmissive and semi-reflective mirror 403, control the first vision component 4 to operate, and obtain the position coordinates of the second marking point 801 in the first coordinate system; The above alignment process (5) can be completed in process (1), or can be performed after completing process (4), which will not be described in detail here.

[0074] (6) Control the second displacement mechanism 3 to continue to drive the bonding component 1 to translate relative to the first visual component 4 (i.e., along the Figure 4 The black film layer 501 is calibrated to the optical path of the semi-transmissive and semi-reflective mirror 403, and the calibration point 503 is calibrated to the optical path of the second reflective mirror 404; thereafter, the camera module 401 continues to take pictures to obtain the position coordinates of the calibration point 503 in the first coordinate system; (7) performing coordinate conversion between the second coordinate system and the first coordinate system, converting the positional difference between the first marking point 701 and the second marking point 801 according to the positional coordinates of the first marking point 701, the calibration point 503 and the second marking point 801, and controlling the second displacement mechanism 3 to perform compensatory motion according to the positional difference, so that the first marking point 701 moves to the position directly above the second marking point 801, thereby completing the alignment between the chip to be bonded 7 and the target chip 8; In a preferred embodiment, the coordinate transformation between the two coordinate systems is accomplished by the aforementioned transformation matrix formula, which will not be described in detail here.

[0075] (8) Control the pressure motor 103 in the bonding assembly 1 to work and press the bonding head 101 downward so that the chip to be bonded 7 and the aligned target chip 8 are tightly fitted together until the bonding process is completed.

[0076] In another specific preferred embodiment, the marking points 503 at the bottom of the light-transmitting layer 502 are two respectively arranged at the two ends of the Y-axis direction, and the first marking point 701 on the chip to be bonded 7 and the second marking point 801 on the target chip 8 are also two in the Y-axis direction. Accordingly, two first visual components 4 (i.e., Figure 5 The visual component 4a and the visual component 4b are provided, and the second visual component 6 is also provided as two sets in the Y-axis direction.

[0077] At this time, in the above process (3), the two second visual components 6 respectively obtain the position coordinates of the two calibration points 503 at the bottom of the light-transmitting layer 502 and the two first marking points 701 of the chip to be bonded 7 close to one end of the optical calibration component 5 in the second coordinate system.

[0078] At the same time, in the above processes (5) and (6), the two first visual components 4 respectively obtain the position coordinates of the two second marking points 801 at one end of the target chip 8 in the X-axis direction and the two calibration points 503 at the bottom of the light-transmitting layer 502 in the first coordinate system.

[0079] Thereafter, in the above process (7), based on the multiple position coordinates obtained in the first coordinate system and the second coordinate system, the compensation position difference between the two first marking points 701 and the two second marking points 801 is converted, and then the position of the chip to be bonded 7 is adjusted according to the compensation position difference, so that the chip to be bonded 7 and the target chip 8 are accurately aligned in the Z-axis direction.

[0080] Finally, the hybrid bonding between the chip to be bonded 7 and the target chip 8 is completed by the lifting and lowering movement of the bonding component 1 in the Z-axis direction.

[0081] The hybrid bonding equipment based on visual alignment in the present invention has a compact structure and is easy to use. It can accurately complete the bonding of the chip to be bonded with the target chip on the wafer, and accurately realize the visual alignment between the chip to be bonded and the target chip before hybrid bonding, effectively ensuring the alignment accuracy during chip hybrid bonding, reducing equipment errors in the bonding process, improving the quality and yield rate of chip products, and having good economic value.

[0082] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A hybrid bonding device based on visual alignment, characterized in that: It includes a bonding component, a first displacement mechanism, a second displacement mechanism, a first vision component, an optical calibration component, a second vision component and a bearing platform; The carrying platform is used to carry the target chip, and is spaced apart from the second visual component in the first direction; The bonding assembly includes a bonding head connected to a pressure motor, which is used to take the chip to be bonded and complete the mixed bonding of the chip to be bonded and the target chip; The first displacement mechanism includes a horizontal displacement component, and the second displacement mechanism includes a vertical lifting component and a plane adjustment component; the first visual component and the vertical lifting component are respectively assembled and connected with the horizontal displacement component, so that the horizontal displacement component can drive the bonding component and the first visual component to synchronously displace in the first direction; the bonding component is assembled and connected with the vertical lifting component through the plane adjustment component, and the bonding component can be lifted and lowered by the vertical lifting component, and the bonding component can be adjusted in the XYθ direction relative to the first visual component in the plane by the plane adjustment component; The first visual component is arranged on one side of the bonding component, and includes a camera module and a semi-transmissive and semi-reflective mirror and a reflective mirror arranged in sequence on the optical path of the camera module, so that the shooting optical path of the camera module can obtain a first detection light and a second detection light that are parallel to each other and vertically downward after the action of two lenses; the second visual component includes a camera module for generating a third detection light that is vertically upward; The optical calibration component is arranged on a side of the bonding component close to the first visual component, and includes a black film layer and a light-transmitting layer arranged in sequence in a first direction, and a calibration point is arranged at the bottom of the light-transmitting layer; the position information of the calibration point and the first marking point on the chip to be bonded is respectively obtained through the second visual component, and the position information of the calibration point and the second marking point on the target chip is respectively obtained through the first visual component, and the compensation position information between the first marking point and the second marking point is obtained by coordinate transformation, and finally the plane adjustment component adjusts the chip to be bonded and the target chip to be aligned according to the compensation position information to complete the visual alignment before hybrid bonding.

2. The hybrid bonding device based on vision alignment according to claim 1, characterized in that: The bottom of the optical calibration component protrudes from the bottom of the bonding head, and the protruding distance of the optical calibration component is equal to the thickness of the chip to be bonded, so that the first marking point and the calibration point are located in the same horizontal plane.

3. The hybrid bonding device based on vision alignment according to claim 1, characterized in that: The bottom of the optical calibration component is flush with the bottom of the bonding head, and the depth of field range of the camera module in the second visual component is not less than the thickness of the chip to be bonded.

4. The hybrid bonding device based on vision alignment according to any one of claims 1 to 3, characterized in that: The thickness of the black film layer is equal to that of the light-transmitting layer; and / or The optical calibration component is movably connected to the side wall surface of the bonding head via a displacement adjustment mechanism, and the displacement adjustment mechanism is used to adjust the vertical position of the optical calibration component relative to the bonding head; and / or The thickness of the optical calibration component in the first direction is 0.5 to 3 times the opening width of the first visual component lens barrel.

5. The hybrid bonding device based on vision alignment according to claim 4, characterized in that: The optical calibration component has two ends aligned with the two side end surfaces of the bonding head in the second direction, and the second direction is a horizontal direction perpendicular to the first direction.

6. The hybrid bonding device based on vision alignment according to any one of claim 5, characterized in that: The first visual component and the second visual component are both two spaced apart in the second direction; and the calibration points are two spaced apart in the second direction.

7. The hybrid bonding device based on vision alignment according to any one of claims 1 to 3, 5 and 6, characterized in that: The bonding head is a suction cup with negative pressure adsorption capability, and a plurality of adsorption holes are arranged at intervals on the bottom surface of the suction cup.

8. A hybrid bonding method based on vision alignment, which uses the hybrid bonding device based on vision alignment described in any one of claims 1 to 7 to complete the bonding between the chip to be bonded and the target chip, characterized in that: The process includes the following: (1) Controlling the bonding head to pick up the chip to be bonded, and loading the target chip onto the carrier, so that the first marking point on the chip to be bonded faces downward and the second marking point on the target chip faces upward; (2) controlling the first displacement mechanism to operate, driving the bonding component and the first vision component to move synchronously, so that the bonding component moves to above the second vision component; (3) controlling the second visual component to operate, so that the second visual component respectively obtains the position coordinates of the calibration point and the first marking point in the second coordinate system; (4) Controlling the first displacement mechanism to drive the first vision component and the bonding component to move synchronously to the top of the target chip; controlling the second displacement mechanism to drive the bonding component to translate relative to the first vision component, and marking the black film layer on the optical path of the reflector, and marking the second marking point on the target chip on the optical path of the semi-transmitting and semi-reflecting mirror; (5) controlling the first visual component to obtain the position coordinates of the second marking point in the first coordinate system; (6) Controlling the second displacement mechanism to continue to drive the bonding component to translate relative to the first vision component, calibrating the black film layer to the optical path of the semi-transmitting and semi-reflecting mirror, and calibrating the calibration point to the optical path of the reflector; continuing to control the first vision component to work, and obtaining the position coordinates of the calibration point in the first coordinate system; (7) performing coordinate conversion between the second coordinate system and the first coordinate system, obtaining a compensation position difference between the first marking point and the second marking point, and controlling the second displacement mechanism to perform a compensation movement according to the compensation position difference, so that the chip to be bonded is vertically aligned with the target chip; (8) Control the pressure motor in the bonding assembly to press the bonding head down so that the chip to be bonded fits tightly with the aligned target chip until the bonding process is completed.

9. The hybrid bonding method based on vision alignment according to claim 8, characterized in that: In process (1), after the bonding head absorbs the chip to be bonded, the position of the optical calibration component is calibrated such that the black film layer is aligned with the second detection light emitted by the second reflector, and the first detection light emitted by the semi-transmissive and semi-reflective mirror passes over the optical calibration component and is emitted downward; and / or In process (1), the bottom of the optical calibration component is aligned with the bottom of the chip to be bonded, and the vertical distance between the calibration point and the target chip is calibrated to be equal to the spacing between the two detection lights in the first visual component.

10. The hybrid bonding device based on vision alignment according to claim 8 or 9, characterized in that: In process (7), the coordinate transformation between the two coordinate systems is completed by the following transformation matrix formula: In the formula, X and Y describe the origin of the second coordinate system O name2 Position coordinates in the first coordinate system; It is defined as the rotation matrix of the second coordinate system relative to the first coordinate system, which is used to represent the rotation angle of the X-axis and the Y-axis; R 11 , R 12 , R 21 , R 22 The unknown quantity is solved by calibrating the coordinates of the vector in the two coordinate systems.

Images