WO2023115614A1 - 键合系统和键合方法 - Google Patents

键合系统和键合方法 Download PDF

Info

Publication number
WO2023115614A1
WO2023115614A1 PCT/CN2021/142073 CN2021142073W WO2023115614A1 WO 2023115614 A1 WO2023115614 A1 WO 2023115614A1 CN 2021142073 W CN2021142073 W CN 2021142073W WO 2023115614 A1 WO2023115614 A1 WO 2023115614A1
Authority
WO
WIPO (PCT)
Prior art keywords
calibration
bonding
die
sub
bonded
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/142073
Other languages
English (en)
French (fr)
Inventor
田应超
刘天建
曹瑞霞
张伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hubei Yangtze Memory Laboratories
Hubei 3D Semiconductor Integrated Innovation Center Co Ltd
Original Assignee
Hubei Yangtze Memory Laboratories
Hubei 3D Semiconductor Integrated Innovation Center Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hubei Yangtze Memory Laboratories, Hubei 3D Semiconductor Integrated Innovation Center Co Ltd filed Critical Hubei Yangtze Memory Laboratories
Priority to EP21946243.9A priority Critical patent/EP4231339B1/en
Publication of WO2023115614A1 publication Critical patent/WO2023115614A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10W—GENERIC PACKAGES, INTERCONNECTIONS, CONNECTORS OR OTHER CONSTRUCTIONAL DETAILS OF DEVICES COVERED BY CLASS H10
    • H10W72/00—Interconnections or connectors in packages
    • H10W72/01—Manufacture or treatment
    • H10W72/011—Apparatus therefor
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10W—GENERIC PACKAGES, INTERCONNECTIONS, CONNECTORS OR OTHER CONSTRUCTIONAL DETAILS OF DEVICES COVERED BY CLASS H10
    • H10W99/00—Subject matter not provided for in other groups of this subclass
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P72/00—Handling or holding of wafers, substrates or devices during manufacture or treatment thereof
    • H10P72/04—Apparatus for manufacture or treatment
    • H10P72/0446—Apparatus for mounting on conductive members, e.g. leadframes or conductors on insulating substrates
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P72/00—Handling or holding of wafers, substrates or devices during manufacture or treatment thereof
    • H10P72/50—Handling or holding of wafers, substrates or devices during manufacture or treatment thereof for positioning, orientation or alignment
    • H10P72/53—Handling or holding of wafers, substrates or devices during manufacture or treatment thereof for positioning, orientation or alignment using optical controlling means
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10W—GENERIC PACKAGES, INTERCONNECTIONS, CONNECTORS OR OTHER CONSTRUCTIONAL DETAILS OF DEVICES COVERED BY CLASS H10
    • H10W72/00—Interconnections or connectors in packages
    • H10W72/071—Connecting or disconnecting

Definitions

  • Embodiments of the present disclosure relate to but are not limited to the field of semiconductor manufacturing, and in particular, relate to a bonding system and a bonding method.
  • the three-dimensional integration of semiconductor devices can be realized by using bonding technology.
  • bonding technology By bonding two or more semiconductor structures with the same or different functions, the performance of the chip can be improved, and at the same time, the metal interconnection between the objects to be bonded can be greatly shortened, and heat generation, power consumption and delay can be reduced.
  • the bonding process can be distinguished by bonding object, including wafer to wafer (wafer to wafer) bonding, chip (or called die) to wafer (die to wafer) bonding and chip to chip (die to die) Bond.
  • a bonding system including:
  • the bonding assembly includes: a bonding head and a first optical path passing through the bonding head; wherein, the first end of the first optical path is located at the point where the bonding head picks up the first die to be bonded the pick-up surface;
  • Wafer carrying table used for carrying wafer
  • a first alignment component used to determine the offset of the pick-up surface compared to the horizontal plane when the bonding head is in the first position
  • the bonding assembly is also used to drive the bonding head to move to a second position parallel to the horizontal plane according to the offset;
  • the first alignment component is further configured to transmit a detection optical signal to the first end through the second end of the first optical path when the bonding head is at the second position; wherein, the picked-up The first die to be bonded covers the first end and reflects the detection optical signal;
  • the first alignment component is further configured to receive a reflected light signal of the detection light signal reflected by the first die to be bonded, and determine the first to-be-bonded die according to the received reflected light signal. a first deviation value between the current position of the die and the first target position;
  • the second alignment component is located on the side of the wafer carrier relatively away from the bonding component, and is used to determine the current position of the second die to be bonded on the wafer and the second target position. Deviation;
  • the wafer carrier is further configured to drive the carried wafer to move relative to the pick-up surface according to the first deviation value and the second deviation value, so that the second die to be bonded Aligning the first die to be bonded; or, the bonding assembly is further configured to drive and pick up the first die to be bonded according to the first deviation value and the second deviation value
  • the pick-up surface moves relative to the wafer carrier, so that the first die to be bonded is aligned with the second die to be bonded;
  • the bonding assembly is also used for bonding the first die to be bonded and the second die to be bonded.
  • the bonding head includes: a side surface perpendicular to the pick-up surface, a top surface parallel to the pick-up surface, and at least three calibration marks; wherein the at least three calibration marks are located on the The side surface or the top surface, the plane where the at least three calibration marks are located is parallel to the pick-up surface;
  • the first alignment component is specifically configured to determine the offset according to sub-deviations between the at least three calibration marks and preset positions.
  • the first alignment component is specifically configured to acquire a first image including a first calibration mark when the bonding head is at the first position, and determine the first alignment mark according to the first image. a first sub-deviation between the position of a calibration mark and a preset position;
  • the bonding assembly is further configured to drive the bonding head to rotate by a first angle, so that the first alignment assembly can acquire a second image including the second calibration mark;
  • the first alignment component is specifically further configured to determine a second sub-deviation between the position of the second calibration mark and the preset position according to the acquired second image;
  • the bonding assembly is further configured to drive the bonding head to rotate by a second angle, so that the first alignment assembly can acquire a third image including the third calibration mark;
  • the first alignment component is specifically further configured to determine a third sub-deviation between the position of the third calibration mark and the preset position according to the acquired third image, and according to the first A sub-deviation, the second sub-deviation and the third sub-deviation determine the offset.
  • the first alignment component includes: at least three aligners, each of which is used to obtain a sub-deviation between the calibration mark and a preset position;
  • the first alignment component is specifically configured to determine the offset according to at least three of the sub-deviations.
  • the bonding head includes: at least two alignment optical pathways penetrating through the bonding head from the sides perpendicular to the pick-up surface; wherein the at least two alignment optical pathways The plane where it is located is parallel to the pick-up surface;
  • the first alignment component is specifically configured to determine the offset according to the sub-offset between the at least two calibration light paths and the horizontal plane.
  • the first alignment component includes: a transmitting unit and a receiving unit;
  • the transmitting unit when the bonding head is at the first position, the transmitting unit is aligned with one end of the first calibration light path, and the receiving unit is aligned with the other end of the first calibration light path At least part of the area is aligned; the transmitting unit is configured to transmit a calibration optical signal to one end of the first calibration optical path; calibrating the optical signal to determine the first sub-offset;
  • the bonding assembly is also used to drive the bonding head to rotate, so that one end of the second calibration light path is aligned with the transmitting unit; wherein, the receiving unit is connected to the second calibration light path at least a partial area of the other end of the pathway is aligned;
  • the transmitting unit is further configured to transmit a calibration optical signal to one end of the second calibration optical path; the receiving unit is further configured to receive the calibration light transmitted through the second calibration optical path signal to determine a second sub-offset;
  • the first alignment component is specifically configured to determine the offset according to the first sub-offset and the second sub-offset.
  • the first alignment component includes:
  • the first transmitting unit is aligned with one end of the first calibration light path, and the first receiving unit is aligned with the first calibration light path. At least part of the other end of the path is aligned, the second emitting unit is aligned with one end of the second calibration light path, the second receiving unit is aligned with the other end of the second calibration light path at least a partial area of one end is aligned;
  • the first transmitting unit is configured to transmit a calibration optical signal to one end of the first calibration optical path
  • the first receiving unit is configured to determine the first sub-offset according to the received calibration optical signal transmitted through the first calibration optical path;
  • the second transmitting unit is configured to transmit a calibration optical signal to one end of the second calibration optical path
  • the second receiving unit is configured to determine the second sub-offset according to the received calibration optical signal transmitted through the second calibration optical path;
  • the first alignment component is specifically configured to determine the offset according to the first sub-offset and the second sub-offset.
  • a bonding method is provided, the bonding method is applied to the bonding of a first die to be bonded and a second die to be bonded located on a wafer, the bonding Combined methods include:
  • the bonding head that picks up the first die to be bonded When the bonding head that picks up the first die to be bonded is located at the first position, determine the offset of the pick-up surface where the bonding head picks up the first die to be bonded compared to the horizontal plane;
  • the wafer or the first die to be bonded move the wafer or the first die to be bonded to align the first die to be bonded with the second die to be bonded Combined die;
  • the first die to be bonded and the second die to be bonded are aligned, the first die to be bonded and the second die to be bonded are bonded.
  • the bonding head includes: a side surface perpendicular to the pick-up surface, a top surface parallel to the pick-up surface, and at least three calibration marks; wherein the at least three calibration marks are located on the side or the top surface, the plane where the at least three calibration marks are located is parallel to the pick-up surface;
  • the determining the offset of the picking surface of the bonding head compared to the horizontal plane includes:
  • the offset is determined according to the first sub-deviation, the second sub-deviation and the third sub-deviation.
  • the bonding head includes: a side surface perpendicular to the pick-up surface, a top surface parallel to the pick-up surface, and at least three calibration marks; wherein the at least three calibration marks are located on the side or the top surface, the plane where the at least three calibration marks are located is parallel to the pick-up surface;
  • the determining the offset of the picking surface of the bonding head compared to the horizontal plane includes:
  • the offset is determined based on at least three of the sub-deviations.
  • the bonding head includes: at least two alignment optical pathways penetrating through the bonding head from the sides perpendicular to the pick-up surface; wherein the at least two alignment optical pathways The plane where it is located is parallel to the pick-up surface;
  • the determining the offset of the picking surface of the bonding head compared to the horizontal plane includes:
  • the offset is determined according to the first sub-offset and the second sub-offset.
  • the bonding head includes: at least two alignment optical pathways penetrating through the bonding head from the sides perpendicular to the pick-up surface; wherein the at least two alignment optical pathways The plane where it is located is parallel to the pick-up surface;
  • the determining the offset of the picking surface of the bonding head compared to the horizontal plane includes:
  • the offset is determined according to the first sub-offset and the second sub-offset.
  • the bonding assembly can drive the bonding head to move parallel to the horizontal plane according to the offset. That is, before die-to-wafer bonding, the picked-up die can be adjusted to be parallel to the horizontal plane, which is beneficial to reduce the probability of damage due to uneven local force during die bonding, and improve the die-to-wafer bond. Combined yield.
  • the first deviation value between the current position of the first die to be bonded and the first target position can also be determined by the first alignment component, and the second alignment A component that can determine a second deviation value between the current position of the second die to be bonded and the second target position on the wafer, and adjust the wafer or the first die to be bonded according to the first deviation value and the second deviation value , the first to-be-bonded die and the second to-be-bonded die can be precisely aligned, which is beneficial to improving the bonding accuracy from the die to the wafer, thereby reducing the difficulty of bonding from the die to the wafer.
  • FIG. 1 is a schematic diagram of a wafer-to-wafer bonding process shown according to an exemplary embodiment
  • Fig. 2a is a first schematic diagram of chip-to-wafer bonding according to an exemplary embodiment
  • Fig. 2b is a second schematic diagram of chip-to-wafer bonding according to an exemplary embodiment
  • Fig. 3a is a first schematic diagram of a bonding system according to an embodiment of the present disclosure
  • Fig. 3b is a first optical path schematic diagram of a bonding system shown according to an embodiment of the present disclosure
  • Fig. 3c is a second optical path schematic diagram of a bonding system shown according to an embodiment of the present disclosure.
  • Fig. 4a is a second schematic diagram of a bonding system according to an embodiment of the present disclosure.
  • Fig. 4b is a schematic diagram of a third optical path of a bonding system according to an embodiment of the present disclosure.
  • Fig. 5a is a third schematic diagram of a bonding system according to an embodiment of the present disclosure.
  • Fig. 5b is a schematic diagram 4 of an optical path of a bonding system according to an embodiment of the present disclosure
  • FIG. 6 is a fourth schematic diagram of a bonding system according to an embodiment of the present disclosure.
  • FIG. 7 is a fifth schematic diagram of a bonding system according to an embodiment of the present disclosure.
  • FIG. 8 is a sixth schematic diagram of a bonding system according to an embodiment of the present disclosure.
  • FIG. 9 is a schematic flow diagram of a bonding method according to an embodiment of the present disclosure.
  • Fig. 10a is a process schematic diagram 1 of a bonding method shown according to the implementation of the present disclosure
  • Fig. 10b is a second process schematic diagram of a bonding method shown according to the implementation of the present disclosure.
  • Fig. 10c is a third schematic diagram of a bonding method according to the implementation of the present disclosure.
  • Fig. 10d is a fourth schematic diagram of a bonding method according to the implementation of the present disclosure.
  • FIG. 11 is a schematic diagram of a die-to-wafer structure according to an embodiment of the disclosure.
  • orientations or positional relationships indicated by the terms “upper”, “lower”, “inner”, “outer”, etc. are based on the orientation or positional relationships shown in the drawings, and are only for It is convenient to describe the present disclosure and simplify the description, but does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and thus should not be construed as limiting the present disclosure.
  • first and second are used for descriptive purposes only, and should not be understood as indicating or implying relative importance.
  • FIG. 1 is a schematic diagram of a wafer-to-wafer bonding process according to an exemplary embodiment.
  • horizontal adjustment includes adjusting the upper wafer 101 and the lower wafer 102 to be parallel to the horizontal plane
  • visual alignment includes aligning the upper wafer 101 and the lower wafer 102 in a vertical direction perpendicular to the horizontal plane.
  • the size of the wafer stage (stage) carrying the wafer is also large.
  • components can be integrated in the wafer stage to realize the levelness detection and leveling of the wafer. adjust. For example, it can be realized by measuring the height deviation of multiple positions on the surface of the wafer stage by components arranged in the wafer stage, and the method is simple and requires low precision.
  • the upper lens After adjusting the upper wafer 101 and the lower wafer 102 to be parallel to the horizontal plane, use the upper lens to identify the position of the alignment mark on the top surface of the lower wafer 102, and record it as (x 2 , y 2 ), the upper lens is located on the upper wafer
  • the side of the circle 101 is relatively far away from the lower wafer 102 .
  • the position of the alignment mark on the bottom surface of the upper wafer 101 is identified by using the lower lens, and recorded as (x 1 , y 1 ).
  • the lower lens is located on the side of the lower wafer 102 relatively away from the upper wafer 101 .
  • the theoretical alignment deviation between the alignment mark on the bottom surface of the upper wafer 101 and the alignment mark on the top surface of the lower wafer 102 is (x 1 ⁇ x 2 , y 1 ⁇ y 2 ).
  • the lower wafer 102 when using the lower lens to identify the position of the alignment mark on the bottom surface of the upper wafer 101, the lower wafer 102 needs to be moved away (for example, the lower wafer is moved from the position of 102' to the position of 102) In order to prevent the lower wafer 102 from blocking the field of view of the lower lens.
  • the upper wafer 101 when using the upper lens to identify the position of the alignment mark on the top surface of the lower wafer 102 , the upper wafer 101 needs to be moved away to prevent the upper wafer 101 from blocking the view of the upper lens.
  • Movement errors may also occur during removal of the lower or upper wafer. Specifically, when performing wafer-to-wafer bonding, since the size of the wafer is relatively large, it is only necessary to use the upper lens/lower lens to identify the alignment mark positions of the lower wafer 102/upper wafer 101 respectively, and record The movement error ( ⁇ x, ⁇ y) of the last wafer can be monitored according to the displacement deviation of the electrostatic chuck that attracts the wafer.
  • the actual alignment deviation between the alignment mark on the bottom surface of the upper wafer 101 and the alignment mark on the top surface of the lower wafer 102 is (x 1 -x 2 + ⁇ x, y 1 -y 2 + ⁇ y), according to The actual alignment deviation is corrected and aligned, and the bottom surface of the upper wafer 101 is bonded to the top surface of the lower wafer 102 .
  • Die-to-wafer bonding is not the same as wafer-to-wafer bonding.
  • the top bonding object is a wafer
  • the chip-to-wafer bonding process the top bonding object is a chip
  • the size of the chip is much smaller than that of the wafer. size. Therefore, the bonding setup for die-to-wafer bonding will be quite different from the bonding setup for wafer-to-wafer bonding. For example, incoming material receiving, level adjustment, visual alignment, down-bonding method, etc.
  • Figure 2a and Figure 2b are schematic diagrams showing a chip-to-wafer bonding according to an exemplary embodiment, referring to Figure 2a and Figure 2b, before bonding the upper chip 201 and the lower wafer 202, it is necessary Level adjustment and visual alignment. Specifically, the upper chip 201 and the lower wafer 202 are respectively adjusted to be parallel to the horizontal plane, and then the alignment mark of the upper chip 201 is aligned with the alignment mark of the die to be bonded in the lower wafer 202 .
  • chip-to-wafer bonding requires higher precision and more difficult bonding.
  • level adjustment since the size of the chip is much smaller than the size of the wafer, the size of the bonding head that picks up the chip is also much smaller than the size of the components that pick up the wafer. Integrating more components in the bonding head will increase the bond. The weight of the bonding head causes the inertia of the bonding head to increase. Therefore, the level detection and level adjustment of chip-to-wafer bonding has become a difficult problem. And when the bonding head that picks up the chip is in a non-horizontal position, parts of the chip are easily damaged due to uneven force, resulting in the failure of the chip-to-wafer bonding.
  • the upper chip 201 moves to It is also difficult for the lower lens to identify and record the displacement error caused by the alignment of the die to be bonded in the lower wafer 202 , and the bonding accuracy of the chip to the wafer is relatively low.
  • embodiments of the present disclosure provide a bonding system and a bonding method.
  • FIGS. 3b and 3c are schematic diagrams of a bonding system 300 according to an embodiment of the present disclosure.
  • FIG. 3a shows a schematic structural view of a bonding system 300
  • FIGS. 3b and 3c show a partial schematic diagram of a bonding system 300.
  • the bonding system 300 includes:
  • the bonding assembly includes: a bonding head 301 and a first optical path 302 passing through the bonding head 301; wherein, the first end 302a of the first optical path 302 is located at the point where the bonding head 301 picks up the first die to be bonded 11 Picking face 301b;
  • Wafer carrying table 303 used for carrying wafer 20;
  • the first alignment component 304 is used to determine the offset of the pick-up surface 301b compared to the horizontal plane when the bonding head 301 is located at the first position;
  • the bonding assembly is also used to drive the bonding head 301 to move to a second position parallel to the horizontal plane according to the offset;
  • the first alignment component 304 is also used to transmit a detection light signal to the first end 302a through the second end 302b of the first optical path 302 when the bonding head 301 is in the second position;
  • the integrated die 11 covers the first end 302a and reflects the detection optical signal;
  • the first alignment component 304 is also used to receive the reflected light signal of the detection light signal reflected by the first die to be bonded 11, and determine the current position of the first die to be bonded 11 according to the received reflected light signal. a first deviation value between target positions;
  • the second alignment component 305 is located on the side of the wafer stage 303 relatively away from the bonding component, and is used to determine the second deviation value between the current position of the second die 21 to be bonded and the second target position on the wafer 20 ;
  • the wafer carrier 303 is also used to drive the carried wafer 20 to move relative to the pick-up surface 301b according to the first deviation value and the second deviation value, so that the second die to be bonded 21 is aligned with the first die to be bonded core 11; or, the bonding assembly is also used to drive the pick-up surface 301b that picks up the first die 11 to be bonded to move relative to the wafer stage 303 according to the first deviation value and the second deviation value, so that the first die to be bonded The bonding die 11 is aligned with the second to-be-bonded die 21;
  • the bonding component is also used for bonding the first to-be-bonded die 11 and the second to-be-bonded die 21 .
  • the bonding assembly can be aligned in a horizontal direction (eg, x direction or y direction) and/or a vertical direction (eg, z direction) ) to adjust the bonding head 301 to the first position.
  • the first position represents the position of the bonding head 301 before performing level detection and level adjustment.
  • the first alignment component 304 can be used to detect the offset of the pick-up surface 301b compared to the horizontal plane (for example, the xoy plane), and the offset can be passed between the pick-up surface 301b and the horizontal plane. It can also be represented by the height of multiple points on the picking surface 301b compared to the same horizontal plane.
  • the bonding assembly can drive the bonding head 301 to rotate until the picking surface 301b is parallel to the horizontal plane, and adjust the bonding head 301 to the second position.
  • the second position refers to the position where the second end 302 b of the first optical path 302 can receive the detection light signal emitted by the first alignment component 304 after the levelness detection and level adjustment of the bonding head 301 .
  • the first alignment component 304 sends a detection optical signal, and the detection optical signal passes through the second end 302b of the first optical path 302 and
  • the first end 302a is transmitted to the positioning mark of the first die to be bonded 11, and forms a reflected optical signal, and the reflected optical signal is transmitted to the first alignment component through the first end 302a and the second end 302b of the first optical path 302 304.
  • first target position and the second target position have the same position in the horizontal direction and different positions in the vertical direction. That is, x 0 is the same as x 0 ', y 0 is the same as y 0 ', the coordinates in the z direction of the first target position and the second target position are different, and the different reference signs are only for the convenience of distinguishing the first target position and the second target location, and is not intended to limit the disclosure.
  • the current position (x 1 , y 1 ) of the first die 11 to be bonded, the first target position (x 0 , y 0 ), the current position (x 2 , y 2 ) of the second die 21 to be bonded ) and the second target position (x 0 ', y 0 ') are determined based on the same coordinate system.
  • the center of the wafer carrier can be used as the origin to establish a coordinate system
  • the x direction and the y direction are vertical and parallel to the plane where the wafer carrier is located
  • the z direction is perpendicular to the plane where the wafer carrier is located.
  • the positions of the first alignment component, the second alignment component, and the bonding component can be calibrated by using a standard die to ensure that the first alignment component can be aligned with the first optical path.
  • the second ends are aligned and ensure that the second alignment component can be aligned with the positioning marks on the surface of the standard die facing the wafer carrier, so as to determine the first target position and the second target position.
  • the first target position represents the position of the standard die after calibration
  • the coordinates in the horizontal direction are marked as (x 0 , y 0 )
  • the position at a preset distance from the first target position in the vertical direction is marked as the second target position
  • the horizontal coordinates of the second target position are marked as (x 0 ′, y 0 ′), where the preset distance is greater than zero.
  • first die to be bonded when the first die to be bonded is located at the first target position and the second die to be bonded is located at the second target position, it can be considered that the first die to be bonded and the second die to be bonded are alignment. That is, along the vertical direction, the projection of the first die to be bonded on the wafer carrier overlaps with the second die to be bonded.
  • the current position of the first to-be-bonded die 11 represents the actual position of the first to-be-bonded die 11 when the bonding assembly is adjusted to the second position, which may be relative to the first target position. There are deviations.
  • the current position of the second die to be bonded 21 represents that when the wafer 20 is placed on the wafer carrier and it is determined that the die to be bonded on the wafer is the second die to be bonded, the second die to be bonded
  • the actual position of the combined die 21 may be deviated relative to the second target position.
  • the first deviation value and the second deviation value are both 0, by moving the wafer vertically upward or moving the first die to be bonded vertically downward, the precise alignment bonding of the first die to be bonded and the second tube to be bonded can be realized.
  • the bonding assembly can be kept fixed (that is, the first die 11 to be bonded is kept fixed), and the wafer carrier 303 drives the wafer 20 to move horizontally according to the displacement deviation ( ⁇ x T - ⁇ x B , ⁇ y T - ⁇ y B ), So that the second die to be bonded 21 is aligned with the first die to be bonded 11 , that is, the positions of the first die to be bonded 11 and the second die to be bonded 21 in the horizontal direction are the same.
  • the direction in which the wafer stage 303 drives the wafer 20 to move can be determined according to the positive or negative value of ⁇ xT - ⁇ xB .
  • ⁇ xT - ⁇ xB is a negative value
  • the wafer stage 303 drives the wafer 20 along x
  • the axis moves in the negative direction
  • the moving distance is the absolute value of ( ⁇ x T - ⁇ x B ).
  • ⁇ x T ⁇ x B is a positive value
  • the wafer carrier 303 drives the wafer 20 to move along the positive direction of the x-axis, and the moving distance is ( ⁇ x T ⁇ x B ).
  • the direction in which the wafer stage 303 drives the wafer 20 to move can be determined according to the positive or negative value of ⁇ yT - ⁇ yB .
  • the wafer stage 303 drives the wafer 20 to move along The y-axis moves in the negative direction, and the moving distance is the absolute value of ( ⁇ y T - ⁇ y B ).
  • the wafer stage 303 drives the wafer 20 to move along the positive direction of the y-axis, and the moving distance is ( ⁇ y T ⁇ y B ).
  • the wafer carrier 303 can be kept fixed (that is, the second die to be bonded 21 is kept fixed), and the bonding assembly moves horizontally according to the displacement deviation ( ⁇ x T - ⁇ x B , ⁇ y T - ⁇ y B ), so that the first The die to be bonded 11 is aligned with the second die to be bonded 21 .
  • the moving direction of the bonded component can be determined according to the positive or negative value of ⁇ x T - ⁇ x B.
  • ⁇ x T - ⁇ x B is a negative value
  • the bonded component moves along the positive direction of the x-axis, and the moving distance is ( ⁇ x T - the absolute value of ⁇ x B ).
  • ⁇ x T - ⁇ x B is a positive value
  • the bonding assembly moves along the negative direction of the x-axis, and the moving distance is ( ⁇ x T - ⁇ x B ).
  • the moving direction of the bonded component can be determined according to the positive or negative value of ⁇ y T - ⁇ y B , for example, when ⁇ y T - ⁇ y B is negative, the bonded component moves along the positive direction of the y-axis, and the moving distance is ( ⁇ y Absolute value of T - ⁇ y B ).
  • ⁇ y T - ⁇ y B is a positive value, the bonding assembly moves along the negative direction of the y-axis, and the moving distance is ( ⁇ y T - ⁇ y B ).
  • the pick-up surface that picks up the first die to be bonded may be driven to move relative to the wafer carrying table, or the carried wafer may be driven to move relative to the pick-up surface, To align the first die to be bonded with the second die to be bonded.
  • the carried wafer is driven to move relative to the pick-up surface, so that the second die to be bonded is aligned with the first die to be bonded.
  • the size of the die and the bonding head that picks up the die is relatively small. If the bonding assembly is used to move relative to the wafer, the small-sized bonding head is more likely to fluctuate during the movement of the bonding assembly. It may be inclined relative to the horizontal plane, causing the die picked up by the bonding head to be inclined, affecting alignment and reducing bonding accuracy.
  • the bonding assembly moves relative to the wafer to align, by driving the wafer to move relative to the pick-up surface, since the size of the wafer is relatively large compared to the size of the die, during the movement of the wafer, the bond of the wafer.
  • the probability of the bonding surface fluctuating in the horizontal plane is small, that is, the second die to be bonded to be bonded to the first die to be bonded is relatively horizontal, which is beneficial to further improve the bonding accuracy from the die to the wafer.
  • the bonding head 301 can absorb the first die 11 to be bonded by means of vacuum adsorption or electrostatic adsorption.
  • the first to-be-bonded die 11 may include a bonding surface and a non-bonding surface that are arranged oppositely. When performing die-to-wafer bonding, the pick-up surface can , to pick up the first die 11 to be bonded.
  • the first die 11 to be bonded includes: a semiconductor chip, for example, a memory chip, a communication chip, an artificial intelligence chip, an LED chip, a sensor chip, or a display chip.
  • the first die to be bonded 11 includes a front and a back face oppositely arranged along the z direction, when the front of the first die to be bonded 11 is provided with a functional structure, the first die to be bonded 11
  • the front side of the first die 11 to be bonded is a bonding surface
  • the back side of the first die 11 to be bonded is a non-bonding surface.
  • first die to be bonded 11, the wafer 20 and the second die to be bonded 21 are not included in the bonding system 300, and the first die to be bonded 11, the wafer to be bonded in FIG. 20 and the second die to be bonded 21 (shown in dashed lines) are only schematic, so as to understand the positional relationship between the die to be bonded, the bonding head and the wafer when performing die-to-wafer bonding .
  • the first optical path 302 includes a first end 302a and a second end 302b.
  • the first end 302a is used for outputting a detection light signal and receiving a reflected light signal
  • the second end 302b is used for receiving a detection light signal and outputting a reflected light signal.
  • the first optical path 302 allows detection optical signals and reflected optical signals to pass through, and may be a transparent window provided in the bonding head 301 , or an optical transmission medium provided in the bonding head 301 , such as an optical fiber.
  • the first alignment component and the second alignment component can emit detection light signals, receive reflected light signals, and convert the received reflected light signals into visualized images.
  • the first alignment component can acquire a first image of the positioning mark on the first die to be bonded, and determine the first deviation value according to the first image.
  • the second alignment component can acquire a second image of the positioning mark on the second die to be bonded, and determine a second deviation value according to the second image.
  • the first alignment component and the second alignment component can transmit and receive infrared light signals (eg, far infrared light signals).
  • the wafer carrier 303 may include: a chuck (not shown in the figure), for example, an electrostatic chuck (ESC chuck), used to attract the wafer 20 .
  • a chuck for example, an electrostatic chuck (ESC chuck), used to attract the wafer 20 .
  • ESC chuck electrostatic chuck
  • the wafer stage is movable relative to the pick-up surface to drive the wafer relative to the pick-up surface.
  • the table itself used for carrying the wafer by the wafer carrier is not movable, but the wafer carrier includes a lifting pin (Lift pin) that can move relative to the pick-up surface, by driving the lift pin to move relative to the pick-up surface , which in turn can drive the wafer to move relative to the pick-up surface to realize the position adjustment of the wafer.
  • a vacuum hole may be provided on the loading needle, and the wafer may be sucked by vacuum.
  • the loading needle can move in a plane parallel to the wafer to drive the movement of the wafer to adjust the position of the wafer.
  • the electrostatic chuck includes a fixed substrate and a loading needle, and the loading needle passes through the fixed substrate along the axial direction of the electrostatic chuck, and can move along the direction perpendicular to the fixed substrate and parallel to the fixed substrate. direction to move.
  • the loading pins may contact the wafer when the electrostatic chuck is carrying the wafer.
  • Wafer 20 includes a plurality of second dies 21 to be bonded, each second die to be bonded 21 includes a substrate and a functional structure (for example, a memory array or a functional circuit) on the substrate, and two adjacent A dicing line is arranged between the second dies to be bonded.
  • a functional structure for example, a memory array or a functional circuit
  • the bonding assembly can drive the bonding head to move parallel to the horizontal plane according to the offset. That is, before die-to-wafer bonding, the picked-up die can be adjusted to be parallel to the horizontal plane, which is beneficial to reduce the probability of damage due to uneven local force during die bonding, and improve the die-to-wafer bond. Combined yield.
  • the first deviation value between the current position of the first die to be bonded and the first target position can also be determined by the first alignment component, and by setting The second alignment component can determine the second deviation value between the current position of the second die to be bonded on the wafer and the second target position, and adjust the wafer or the first die to be bonded according to the first deviation value and the second deviation value.
  • the position of the bonding die can make the first die to be bonded and the second die to be bonded be precisely aligned, which is beneficial to improving the bonding accuracy from the die to the wafer.
  • the bonding head 301 includes: a side surface 301c perpendicular to the pick-up surface 301b, a top surface 301a parallel to the pick-up surface 301b, and at least three calibration marks; wherein at least three calibration marks The marks are located on the side 301c or the top surface 301a, and the plane where at least three calibration marks are located is parallel to the pick-up surface 301b;
  • the first alignment component 304 is specifically configured to determine an offset according to sub-deviations between at least three calibration marks and preset positions.
  • Fig. 3a shows the calibration marks m 1 , m 2 and m 3 located on the side 301c, the plane formed by the calibration marks m 1 , m 2 , and m 3 is denoted as the calibration plane, and the calibration plane is parallel to the pick-up surface 301b , it can be understood that, when the bonding head 301 is at the first position, by determining the offset of the calibration plane compared to the horizontal plane, the offset of the pick-up surface 301b compared with the horizontal plane can be determined.
  • the calibration marks m 1 , m 2 , and m 3 are shown on the side 301c in this example, in other examples, the calibration marks m 1 , m 2 , and m 3 may also be located on the top surface 301a. For example, on the boundary line of the top surface 301a.
  • the number of calibration marks on the side surface 301c or on the top surface 301a is not limited to three, but can also be four, five or even more, which is not limited by the present disclosure.
  • the first alignment component 304 can respectively determine the position (x m1 , y m1 ) of the first calibration mark m 1 and the position of the second calibration mark m 2 (x m2 , y m2 ) and the position of the third calibration mark m 3 (x m3 , y m3 ), according to the height deviation between the positions of the three calibration marks and the preset position, it can be determined that the calibration plane is compared with the horizontal plane The offset, and then determine the offset of the pick-up surface 301b compared to the horizontal plane.
  • the first alignment component 304 includes a frame for determining the positions of the calibration marks m 1 , m 2 and m 3 , where the preset position represents the center position of the frame, denoted as (x c , y c ). It can be understood that when the three calibration marks m 1 , m 2 and m 3 have the same height deviation from the preset position in the y-axis direction, that is, y m1 -y c , y m2 -y c and y m3 -y c When the numerical values of are the same and the positive and negative are the same, it can be determined that the calibration plane is parallel to the horizontal plane, and then it can be determined that the pick-up plane parallel to the calibration plane is also parallel to the horizontal plane.
  • the height deviation between the position of each calibration mark and the preset position may be different.
  • the position of the third calibration mark m 3 is higher than the position of the second calibration mark m 2
  • the position of the second calibration mark m2 is higher than the position of the first calibration mark m1
  • the calibration plane is inclined compared with the horizontal plane
  • the pick-up plane parallel to the calibration plane is also inclined compared with the horizontal plane.
  • the bonding assembly can first adjust the bonding head 301 to a position parallel to the horizontal plane according to the offset, and then adjust the bonding head 301 moves along the negative direction of the z-axis to a second position where the second end 302b of the first optical path 302 can receive the detection optical signal emitted by the first alignment component 304 .
  • the bonding assembly can first adjust the bonding head 301 to a position parallel to the horizontal plane according to the offset, and then adjust the bonding head 301 to a position parallel to the horizontal plane. Move to the second position where the second end 302b of the first optical path 302 can receive the detection optical signal emitted by the first alignment component 304 along the positive direction of the z-axis.
  • the first alignment component 304 when at least three calibration marks are located on the side 301c, the first alignment component 304 is located on the side of the bonding head 301, and the second end 302b of the first optical path 302 is located on the side 301c. . In other embodiments, when the at least three alignment marks are on the top surface, the first alignment assembly is on the bond head and the second end of the first light path is on the top surface.
  • the first alignment component can be set according to the setting method of the calibration mark on the bonding head, and it is only necessary to ensure that the first alignment component can detect the calibration mark.
  • the second end of the first light path can also be set according to the setting method of the first alignment component, and it only needs to ensure that the first alignment component can be aligned with the first light path.
  • the offset of the calibration plane compared to the horizontal plane can be determined, and then the offset of the picking plane parallel to the calibration plane compared with the horizontal plane can be determined.
  • the first alignment component 304 is specifically configured to acquire the first image including the first calibration mark m1 when the bonding head 301 is at the first position, and according to the first An image determines a first sub-deviation between the position of the first calibration mark m1 and a preset position;
  • the bonding assembly is also used to drive the bonding head 301 to rotate by a first angle, so that the first alignment assembly 304 can acquire a second image including the second calibration mark m2 ;
  • the first alignment component 304 is specifically further configured to determine a second sub-deviation between the position of the second calibration mark m2 and a preset position according to the acquired second image;
  • the bonding assembly is also used to drive the bonding head 301 to rotate by a second angle, so that the first alignment assembly 304 can acquire a third image including the third calibration mark m3 ;
  • the first alignment component 304 is further configured to determine the third sub-deviation between the position of the third calibration mark m3 and the preset position according to the acquired third image, and determine the third sub-deviation according to the first sub-deviation, the second sub-deviation Bias and the third sub-bias, which determine the offset.
  • the first alignment component 304 may include an aligner 3041 located on one side of the bonding head 301, and when the bonding head 301 is in the first position, the aligner 3041 Acquire the first image including the first calibration mark m 1 , and determine the distance between the position (x m1 , y m1 ) and the preset position (x c , y c ) of the first calibration mark m 1 according to the first image
  • the first sub-deviation ⁇ y m1 y m1 ⁇ y c of .
  • the bonding assembly further includes: a fixed unit 307, a mobile unit 308, and a drive unit (not shown) connected to the mobile unit 308; the fixed unit 307 is fixedly connected to the bonding head 301 and The moving unit 308 ; the driving unit is used to drive the moving unit 308 to move the bonding head 301 relative to the wafer carrier 303 .
  • the driving unit drives the moving unit 308 to rotate clockwise by a first angle.
  • the driving unit drives the moving unit 308 to rotate clockwise by a second angle
  • the aligner 3041 acquires a third image including the third calibration mark m3 , and determines the position (x
  • the third sub-deviation ⁇ y m3 y m3 ⁇ y c between m3 , y m3 ) and the preset position (x c , y c ).
  • the offset of the pick-up plane compared to the horizontal plane is determined from ⁇ y m1 , ⁇ y m2 and ⁇ y m3 .
  • the drive unit can also drive the mobile unit to rotate counterclockwise, ensuring that the aligner rotates to the point where the second image including the second calibration mark and the third image including the third calibration mark can be obtained, which is technically skilled in the art Personnel can be selected according to actual bonding requirements, and the present disclosure is not further limited here.
  • the first angle and the second angle can be determined according to the arrangement of the calibration marks m 1 , m 2 and m 3 .
  • the shape of the bonding head can be cylindrical
  • the shape of the pick-up surface and the calibration plane can be circular
  • the first angle can be based on the center of the circle formed by the calibration mark m 1 , the calibration mark m 2 and the center of the calibration plane
  • the second angle can be determined based on the center angle formed by the calibration mark m 2 , the calibration mark m 3 and the center of the calibration plane.
  • the calibration marks m 1 , m 2 and m 3 can be arranged at equal intervals on the side surface 301c, so the first angle and the second angle are 120°.
  • the line connecting the calibration marks m1 and m3 passes through the center of the calibration plane, the line connecting the calibration mark m2 and the center of the calibration plane is perpendicular to the line connecting the calibration marks m1 and m3 , then the first angle and the second angle are 90 °.
  • the first alignment component by setting the first alignment component, the first image, the second image and the third image of the three calibration marks can be sequentially acquired, and according to the acquired first image, the second image and the third image,
  • the first sub-deviation, the second sub-deviation and the third sub-deviation can be determined in sequence, which is beneficial to accurately determine the offset of the picking surface compared to the horizontal plane.
  • the first alignment component 304 includes: at least three aligners 3041, each aligner 3041 is used to obtain a sub-deviation between a calibration mark and a preset position;
  • the first alignment component 304 is specifically configured to determine the offset according to at least three sub-deviations.
  • the first calibration mark m1 and the third calibration mark m3 are arranged side by side along the x direction
  • the first alignment component 304 includes two aligners 3041 arranged side by side along the x direction
  • the aligner 3041 on the right side of the bonding head 301 is used to obtain the first sub-deviation between the first calibration mark m 1 and the preset position
  • the aligner 3041 on the left side of the bonding head 301 is used to obtain A third sub-deviation between the third calibration mark m3 and the preset position.
  • the first alignment component 304 also includes an aligner (not shown) located on the side of the bonding head 301 facing the positive direction of the y-axis, the aligner is used to obtain the difference between the second calibration mark m2 and the preset position The second sub-deviation between .
  • the first alignment component 304 can determine the offset of the pick-up surface compared to the horizontal plane according to the first sub-deviation, the second sub-deviation and the third sub-deviation.
  • the bonding head includes three calibration marks, and the first alignment component includes three aligners.
  • the bond head can include at least four alignment marks and the first alignment assembly can include at least four aligners.
  • the setting of the number or position of the calibration marks in the bonding head is not limited here, as long as the plane on which the calibration marks are set is parallel to the pick-up surface.
  • the bonding head further includes two calibration marks arranged side by side along the y direction
  • the first alignment component further includes two aligners arranged side by side along the y direction, by The two aligners arranged side by side in the y direction can respectively determine the second sub-deviation and the fourth sub-deviation of the two calibration marks arranged side by side in the y direction
  • the first alignment component 304 , the third sub-deviation and the fourth sub-deviation can determine the offset of the picking surface compared to the horizontal plane.
  • each aligner can independently Acquiring an image of each calibration mark without driving the bondhead rotation facilitates the accuracy of the three sub-biases.
  • the three aligners can acquire the images of the three calibration marks at the same time, which can shorten the time for determining the offset, and further shorten the time for levelness detection and level adjustment before die-to-wafer bonding, Improve bonding efficiency.
  • the bonding head 301 includes: a side 301c perpendicular to the pick-up surface 301b, and at least two alignment light paths penetrating the bonding head 301 from the side 301c; wherein, at least two The plane where the calibration light path is located is parallel to the pick-up surface 301b;
  • the first alignment component 304 is specifically configured to determine an offset according to sub-offsets between at least two calibration light paths and a horizontal plane.
  • FIG. 4a shows a calibration light path 3111 that runs through the bonding head 301 from the side 301c
  • FIG. 4b shows two calibration light paths 3111 and 3112 arranged side by side along the y direction.
  • the projections of the calibration light paths 3111 and 3112 on the xoz plane coincide.
  • the number of calibration light paths penetrating the bond head 301 from the side 301c is not limited to two, and may be three or even more
  • the present disclosure is not limited here, it only needs to ensure that the calibration light path does not overlap with the first light path 302 .
  • the plane formed by the two calibration light paths 3111 and 3112 is denoted as the calibration plane, and the calibration plane is parallel to the pick-up surface 301b.
  • 3112 Determine the offset of the calibration plane compared to the horizontal plane, and then determine the offset of the picking surface 301b compared to the horizontal plane.
  • the first alignment component 304 includes: a transmitting unit 3042 and a receiving unit 3043;
  • the transmitting unit 3042 when the bonding head 301 is in the first position, the transmitting unit 3042 is aligned with one end of the first calibration optical path, and the receiving unit 3043 is aligned with at least a partial area of the other end of the first calibration optical path; the transmitting unit 3042 is configured to transmit a calibration optical signal to one end of the first calibration optical path; the receiving unit 3043 is configured to determine the first sub-offset according to the received calibration optical signal transmitted through the first calibration optical path;
  • the bonding assembly is also used to drive the bonding head 301 to rotate, so that one end of the second calibration optical path is aligned with the emitting unit 3042; wherein, the receiving unit 3043 is aligned with at least a part of the other end of the second calibration optical path allow;
  • the transmitting unit 3042 is further configured to transmit a calibration optical signal to one end of the second calibration optical path; the receiving unit 3043 is also configured to determine the second sub-offset according to the received calibration optical signal transmitted through the second calibration optical path ;
  • the first alignment component 304 is specifically configured to determine an offset according to the first sub-offset and the second sub-offset.
  • Fig. 4b shows the first calibration optical path 3111 and the second calibration optical path 3112 arranged side by side along the y direction, and the transmitting unit 3042 and the receiving unit 3043 are respectively located at the two opposite sides of the first calibration optical path 3111 along the x direction. side.
  • the transmitting unit 3042 is located on the right or left of the first calibration optical path 3111
  • the receiving unit 3043 is located on the left or right of the first calibration optical path 3111 .
  • the emitting unit 3042 When the bonding head 301 is at the first position, the emitting unit 3042 is aligned with the right end 3111a of the first calibration optical path 3111, and when the calibration optical signal sent by the emitting unit 3042 is a parallel beam, the calibration optical signal completely enters the right end 3111a, That is, the intensity of the calibration optical signal sent by the transmitting unit 3042 is basically the same as the intensity of the calibration optical signal entering the right end 3111a.
  • the substantially same can be that the intensity of the two optical signals is completely the same, or the intensity difference between the two optical signals is small , can be ignored.
  • the receiving unit 3043 is aligned with at least a partial area of the left end 3111b of the first calibration optical path 3111, and a part of the calibration optical signal transmitted through the first calibration optical path 3111 is received by the receiving unit 3043, that is, the calibration signal sent by the transmitting unit 3042.
  • the intensity of the optical signal is greater than the intensity of the calibration optical signal received by the receiving unit 3043, and the offset of the first calibration optical path 3111 relative to the horizontal plane can be calculated according to the difference between the intensity of the two optical signals.
  • the plane on which the transmitting unit 3042 and the receiving unit 3043 are arranged is parallel to the horizontal plane.
  • the calibration plane parallel to the pick-up plane 301b is also inclined compared with the horizontal plane, that is, the first calibration
  • the light path 3111 is also inclined relative to the horizontal plane. Therefore, when the right end 3111a of the transmitting unit 3042 is completely aligned, the receiving unit 3043 is aligned with a partial area of the left end 3111b. Partial area alignment does not include full alignment.
  • the receiving unit 3043 when the receiving unit 3043 is aligned with the partial area of the left end 3111b, the receiving unit 3043 can only receive a part of the calibration optical signal, and the intensity of the calibration optical signal received by the receiving unit 3043 is smaller than the intensity of the calibration optical signal sent by the transmitting unit 3042 .
  • the calibration plane parallel to the pick-up plane 301b is also parallel to the horizontal plane, that is, the first calibration optical path 3111 is also parallel to the horizontal plane. Therefore, when the right end 3111a of the transmitting unit 3042 is fully aligned, the receiving unit 3043 is also fully aligned with the left end 3111b, and the intensity of the calibration optical signal received by the receiving unit 3043 is basically the same as the intensity of the calibration optical signal sent by the transmitting unit 3042.
  • the calibration optical signal transmitted by the transmitting unit 3042 is transmitted to the receiving unit 3043 through the first calibration optical path 3111 , and the receiving unit 3043 can determine the first One offset.
  • the drive unit can pass the axis of the center of the calibration plane as the center line, and drive the bonding head 301 to rotate until the left end 3112b of the second calibration optical path 3112 is aligned with the emission unit 3042 , the receiving unit 3043 is aligned with at least a partial area of the right end 3112 a of the second calibration optical path 3112 .
  • the second calibration light path 3112 is also inclined relative to the horizontal plane. Therefore, when the left end 3112b is completely aligned with the transmitting unit 3042, the receiving unit 3043 is only aligned with a partial area of the right end 3112a.
  • the second calibration light path 3112 is also parallel to the horizontal plane. Therefore, when the left end 3112b is completely aligned with the transmitting unit 3042, the receiving unit 3043 is also completely aligned with the right end 3112a.
  • the calibration optical signal transmitted by the transmitting unit 3042 is transmitted to the receiving unit 3043 via the second calibration optical path 3112 , and the receiving unit 3043 can determine the first Two sub-offsets, the first alignment component 304 can determine the offset of the picking surface 301b compared to the horizontal plane according to the first sub-offset and the second sub-offset.
  • the first sub-offset can be determined through the first calibration optical path, and then the second sub-offset can be determined through the second calibration optical path.
  • the offset of the calibration plane for setting the two calibration light paths compared to the horizontal plane can be determined, and then the offset of the pickup plane parallel to the calibration plane compared to the horizontal plane can be determined.
  • the first alignment component 304 includes: two transmitting units 3042 and two receiving units 3043;
  • the first emitting unit 3042 is aligned with one end of the first calibration light path, and the first receiving unit 3043 is aligned with at least a partial area of the other end of the first calibration light path Alignment, the second transmitting unit 3042 is aligned with one end of the second calibration optical path, and the second receiving unit 3043 is aligned with at least a partial area of the other end of the second calibration optical path;
  • the first transmitting unit 3042 is configured to transmit a calibration optical signal to one end of the first calibration optical path
  • the first receiving unit 3043 is configured to determine the first sub-offset according to the received calibration optical signal transmitted through the first calibration optical path;
  • the second transmitting unit 3042 is configured to transmit a calibration optical signal to one end of the second calibration optical path
  • the second receiving unit 3043 is configured to determine the second sub-offset according to the received calibration optical signal transmitted through the second calibration optical path;
  • the first alignment component 304 is specifically configured to determine an offset according to the first sub-offset and the second sub-offset.
  • Fig. 4b shows two transmitting units 3042 arranged side by side along the y direction and two receiving units 3043 arranged side by side along the y direction.
  • the first transmitting unit 3042 and the first receiving unit 3043 are arranged side by side along the x direction, and are located on opposite sides of the first calibration optical path 3111 along the x direction.
  • the first transmitting unit 3042 is located on the right or left of the first calibration optical path 3111
  • the first receiving unit 3043 is located on the left or right of the first calibration optical path 3111 .
  • the second transmitting unit 3042 and the second receiving unit 3043 are arranged side by side along the x direction, and are located on opposite sides of the second calibration optical path 3112 along the x direction.
  • the second transmitting unit 3042 is located on the right or left side of the second calibration optical path 3112
  • the second receiving unit 3043 is located on the left or right side of the second calibration optical channel 3112 .
  • two calibration light paths arranged side by side along the y direction are shown, a pair of transmitting unit and receiving unit are respectively located on opposite sides of one calibration light path along the x direction.
  • the two calibration light paths can also be arranged side by side along the x direction, and a pair of transmitting units and receiving units are respectively located on opposite sides of a calibration light path along the y direction, and the two calibration light paths can also be arranged in other ways. setting, the present disclosure is not limited here.
  • the calibration optical signal emitted by the first transmitting unit 3042 is transmitted to the first receiving unit 3043 via the first calibration optical path 3111, and the first receiving unit 3043 can The strength I 1 ' of the signal determines the first sub-offset.
  • the calibration light signal emitted by the second transmitting unit 3042 is transmitted to the second receiving unit 3043 via the second calibration light path 3112 , and the second receiving unit 3043 can
  • the signal strength I 2 ′ determines the second sub-offset
  • the first alignment component 304 can determine the offset of the pick-up surface 301b compared to the horizontal plane according to the first sub-offset and the second sub-offset.
  • a pair of transmitting units and receiving units can determine the first sub-offset of the first calibration optical path, and another pair of transmitting units
  • the summing and receiving unit can determine the second sub-offset of the second calibration optical path without driving the bonding head to rotate, which is conducive to improving the accuracy of the first sub-offset and the second sub-offset.
  • two pairs of transmitting units and receiving units can simultaneously determine the first sub-offset of the first calibration light path and the second sub-offset of the second calibration light path, which can shorten the time for determining the offset. Time, thereby shortening the time for level detection and level adjustment before die-to-wafer bonding, and improving bonding efficiency.
  • the second end 302b of the first optical path 302 is disposed on the side 301c;
  • the first optical path 302 includes: a first mirror 3021, configured to reflect an optical signal transmitted between the first end 302a of the first optical path 302 and the second end 302b of the first optical path 302;
  • the first alignment component 304 includes: a second optical path 306 and an aligner 3041; wherein the aligner 3041 is used to send a detection light signal, receive a reflected light signal, and determine a first deviation value;
  • the aligner 3041 is located at the end of the second optical path 306 relatively far away from the side 301c, and the end of the second optical path 306 relatively close to the side 301c is aligned with the first end of the first optical path 302. Two ends 302b.
  • the aligner 3041 when the bonding head 301 is located at the second position, the aligner 3041 sends out a detection light signal, and the detection light signal is transmitted to the first standby through the second light path 306 and the first light path 302.
  • the positioning marks of the die 11 are bonded to form reflected optical signals, which are transmitted to the aligner 3041 through the first optical path 302 and the second optical path 306 .
  • the detection optical signal received by the two ends 302b is reflected to the first end 302a of the first optical path 302, and transmitted to the positioning mark of the first die 11 to be bonded, and is also used to connect the first end 302a of the first optical path 302 to the first end 302a of the first optical path 302
  • the received reflected optical signal is reflected to the second end 302b of the first optical path 302 and transmitted to the aligner 3041 .
  • the aligner 3041 includes: a device integrating a light emitter and a light detector.
  • the light emitter is used for sending out the detection light signal
  • the light detector is used for receiving the reflected light signal.
  • the aligner 3041 may further include an image sensor connected to the light detector, the image sensor is used to convert the received reflected light signal into a visualized image, and the first deviation value may be determined according to the image.
  • the second optical path is aligned with the first optical path, and is located between the aligner and the first optical path, which facilitates the detection of optical signals or reflected light
  • the signal transmission between the aligner and the first die to be bonded can accurately locate the current position of the first die to be bonded, which is beneficial to improving the accuracy of the first deviation value.
  • the weight of the bonding component will not be additionally increased, and the inertia of the bonding component can be made smaller.
  • the first alignment component will not affect the movement of the bonding component, which can ensure the high-frequency and high-precision movement of the bonding component in the vertical direction, and improve the key quality. While improving the bonding accuracy, the bonding efficiency can also be improved.
  • the second light path 306 when the bonding head 301 is in the second position, the second light path 306 is parallel to the pick-up surface 301 b. It can be understood that, in the embodiment of the present disclosure, the second optical path may be arranged along the horizontal direction, and the detection light signal and the reflected light signal may be transmitted along the horizontal direction in the second light path.
  • the second optical path 306' includes: a first sub-path 3061, a second sub-path 3062, and a second mirror 3063, and the first sub-path 3061 is perpendicular to the second sub-path 3062, the second mirror 3063 is located at the intersection of the first sub-circuit 3061 and the second sub-circuit 3062, and the second mirror 3063 is used to reflect the optical signal transmitted between the first sub-circuit 3061 and the second sub-circuit 3062;
  • the aligner 3041' is located at the end of the first sub-way 3061 relatively far away from the second sub-way 3062;
  • the first sub-path 3061 is perpendicular to the pick-up surface 301b
  • the second sub-path 3062 is parallel to the pick-up surface 301b
  • the end of the second sub-path 3062 relatively close to the side surface 301c is aligned with the first light beam.
  • the second end 302b of the channel 302 is aligned with the first light beam.
  • the aligner 3041' when the bonding head 301 is in the second position, the aligner 3041' sends out a detection light signal, and the detection light signal passes through the first sub-path 3061, the second sub-path 3062 and the first optical path 302 is transmitted to the positioning mark of the first die 11 to be bonded (indicated by the “cross” in FIG. A sub-path 3061 is transmitted to the aligner 3041'.
  • the second mirror 3063 is located between the first sub-circuit 3061 and the second sub-circuit 3062, and is used to reflect the detection light signal received by the first sub-circuit 3061 to the second sub-circuit 3062, and It is transmitted to the first optical path 302, and is also used to reflect the reflected optical signal received by the second sub-path 3062 to the first sub-path 3061, and transmit it to the aligner 3041'.
  • the aligner 3041' can be the same as the aligner 3041 in the above-mentioned embodiments, and the present disclosure will not repeat them here.
  • the second optical path can be set in an "L" shape, including a first sub-path in the vertical direction and a second sub-path in the horizontal direction, and the second end of the first optical path is located at When bonding the side of the head, the second sub-path is aligned with the second end of the first optical path to ensure the transmission of the detection optical signal and reflected optical signal in the first optical path and the second optical path, and accurately locate the first optical path.
  • the second optical path as an "L" shape
  • the diversity of the optical path setting methods in the first alignment component is increased.
  • those skilled in the art can reasonably
  • the light path in the first alignment component is set, the disclosure is not limited here.
  • the second light path is arranged in the horizontal direction (as shown in Figures 3a and 3b), compared to the second light path being set in an "L" shape (that is, as shown in Figures 4a and 4b), the second light path Arranging along the horizontal direction can simplify the design of the optical path, reduce the transmission loss of the detection optical signal or reflected optical signal in the second optical path, improve the positioning accuracy of the first die to be bonded, and further improve the accuracy of the first deviation value.
  • the top surface 301a includes a first region and a second region
  • the bonding assembly further includes: a fixed unit 307 and a mobile unit 308; the fixed unit 307 is fixedly connected to the first area and the mobile unit 308; the mobile unit 308 is used to move the bonding head 301 relative to the wafer carrier 303;
  • the first optical path 302' runs through the bonding head 301 in a direction perpendicular to the pick-up surface 301b; wherein, the second end 302b' of the first optical path 302' is located in the second region;
  • the first alignment component 304" the projection in the horizontal direction is located in the second area, including: the second optical path 306" and the aligner 3041"; where the aligner 3041" is used to send out the detection light signal and receive the reflection optical signal, and determine the first deviation value;
  • the second optical path 306" is perpendicular to the pick-up surface 301b and aligned with the first optical path 302';
  • the bonding head 301 includes a pick-up surface 301b oppositely arranged along the z direction, and a top surface 301a
  • the fixed unit 307 is located between the bonding head 301 and the moving unit 308, and one end of the fixed unit 307 It is fixedly connected with the top surface 301 a of the bonding head 301 , and the other end of the fixed unit 307 is fixedly connected with the moving unit 308 .
  • the first area represents the area where the top surface 301 a of the bonding head 301 is in contact with the fixing unit 307
  • the second area represents the area where the top surface 301 a of the bonding head 301 is not in contact with the fixing unit 307 . It can be understood that the projection of the fixing unit 307 in the horizontal direction is located in the first area.
  • the first optical path 302' runs through the bonding head 301, including a first end 302a' and a second end 302b'. It can be understood that the horizontal plane where the first end 302a' is located is flush with the horizontal plane where the picking surface 301b is located, and the horizontal plane where the second end 302b' is located is equal to the horizontal plane where the top surface 301a is located and is located in the second region.
  • the first alignment component 304" is located above the first optical passage 302', and the second optical passage 306" is arranged in a direction parallel to the z-axis and aligned with the first optical passage 302' , in this example, the first alignment component 304 ′′ can be fixedly connected to the mobile unit 308 .
  • the first alignment component 304 ′′ can be disposed on the top surface 301 a of the bonding head 301 , so as to ensure that the second optical path is aligned with the first optical path.
  • the first alignment component 304" can be fixed by an external bracket without being fixedly connected to the mobile unit 308, and it is enough to ensure that the second optical path is aligned with the first optical path. At this time, it is used to connect the mobile unit 308 and the fixing unit 307 of the bonding head 301 can be retractable devices, so as to prevent the first alignment component 304′′ and the external support from blocking the upward and downward movement of the bonding component.
  • the first alignment component 304" is fixed by the external bracket, which does not increase the weight of the bonded component, which can make the inertia of the bonded component small, and can ensure that the bonded component is vertically high. Frequency, high-precision movement, improve bonding accuracy, but also improve bonding efficiency.
  • the second end of the first optical path is located at the side of the bonding head.
  • the detection optical signal or the reflected optical signal can be transmitted along the vertical direction in the first optical path, and there is no need to set the first reflector in the first optical path, which simplifies the structure of the bonding assembly and is conducive to reducing the The manufacturing difficulty of small bonding components is reduced, and the manufacturing cost of bonding components is reduced.
  • the loss caused by the reflection of the detection optical signal or reflected optical signal at the position of the first mirror is reduced, which is beneficial to accurately position the first die to be bonded.
  • the current position thereby improving the accuracy of the first deviation value.
  • the bonding assembly includes: two first optical pathways, the first ends of different first optical pathways are disposed at different positions, and the second ends of different first optical pathways are disposed at different positions on the pick-up surface;
  • a first alignment assembly comprising an aligner; wherein the aligner aligns with a second end of a first optical path when the bond head is in the second position;
  • the bonding assembly further includes: a driving unit, configured to drive the two first light paths to move relative to the aligner, so that the aligner aligns with the second end of the other first light path.
  • the bonding assembly may include two first light passages 302 arranged side by side along the x direction.
  • the bonding head 301 is at the second position, the first ends 302a of the two first light passages 302 respectively expose the first Two sets of alignment marks in the bonded die 11 .
  • the first alignment component may include an aligner 3041 located on one side of the bonding head 301.
  • the aligner 3041 Aligning with the first of the first optical vias 302 to pass through the first of the first optical vias 302 locates the position of the first set of alignment marks revealed by the first of the first optical vias 302 .
  • the drive unit can be connected with the mobile unit 308 to drive the mobile unit 308 to rotate clockwise or counterclockwise relative to the aligner 3041.
  • the mobile unit 308 rotates, the fixed unit 307, the bonding head 301 and the first key to be In this way, the second first optical channel 302 can be rotated to a position aligned with the aligner 3041, so that the aligner 3041 is aligned with the second first optical channel 302, and through the second
  • the two first light passages 302 locate the positions of the second set of positioning marks exposed by the second first light passage 302 .
  • first optical passages 302 are arranged side by side along the x direction, and the second ends 302b of the two first optical passages 302 are located on the same horizontal plane.
  • the drive The unit can drive the moving unit 308 to rotate 180° clockwise or counterclockwise relative to the aligner 3041 , so that the second first optical path 302 is aligned with the aligner 3041 .
  • the first first light passage 302 rotates to the original position of the second first light passage 302 .
  • the first die to be bonded 11 also rotates synchronously, therefore, the current position of the first group of positioning marks will change after the rotation. It should be understood that the coordinate system established for locating the first target position and for locating the current position of the first group of positioning marks will also change after rotation.
  • the current position of positioning the first group of positioning marks is (x 3 , y 3 ), based on the coordinate system established after rotating clockwise or counterclockwise by 180°, positioning the first group of positioning marks
  • the current position is (-x 3 , -y 3 )
  • the first target position is (-x 0 , -y 0 ).
  • the shape of the bonding head can be cylindrical
  • the shape of the pick-up surface can be circular
  • the position of the first group of positioning marks after rotation can be determined according to the angle of rotation of the mobile unit 308, the direction of rotation, and the radius of the pick-up surface. position and the first target position after rotation.
  • the rotation angle of the moving unit relative to the aligner can be determined based on the positions of the first first light path and the second first light path.
  • the juxtaposition of the two first light paths along the x direction is only for illustration to convey the present disclosure to those skilled in the art, but the present disclosure is not limited thereto.
  • the two first light paths can also be other For example, it can be set according to the arrangement manner of multiple sets of positioning marks in the first die to be bonded.
  • one aligner can be aligned with two first optical paths respectively, and then the first tube to be bonded can be positioned through the first optical path aligned with the aligner Positioning marks on the core.
  • the bonding assembly includes: two first optical pathways, the first ends of different first optical pathways are disposed at different positions, and the second ends of different first optical pathways are disposed at different positions on the pick-up surface;
  • the first alignment component includes two aligners and two second optical paths; wherein, different aligners are located at ends of different second optical paths that are relatively far away from the bonding head;
  • the bonding assembly may include two first optical channels 302 arranged side by side along the x direction, and the first alignment assembly 304 includes two aligners 3041 and two alignment devices arranged side by side along the x direction.
  • Each aligner 3041 aligns with a first optical channel 302 through a second optical channel 306 .
  • the bonding assembly includes two first optical paths, and the first alignment assembly includes two aligners and two second optical paths.
  • the bonding assembly can include at least three first optical pathways, and the first alignment assembly can include at least three aligners and three second optical pathways.
  • the bonding assembly further includes two first optical paths arranged side by side along the y direction
  • the first alignment assembly further includes two aligners arranged side by side along the y direction and Two second optical paths arranged side by side in the y direction, so four first optical paths are arranged in the bonding assembly, four aligners and four second optical paths are arranged in the first alignment assembly, each pair The collimators are each aligned with a first optical path through a second optical path.
  • the setting of the number or position of the first optical path in the bonding assembly and the setting of the number or position of the aligner in the first bonding assembly can be adjusted according to the number or position of the positioning marks in the first die to be bonded. set, the disclosure is not further limited.
  • one aligner can align one first light path through one second light path.
  • Each aligner can position each positioning mark independently, which is beneficial to improve the positioning accuracy of the positioning mark.
  • the bonding system 300 further includes:
  • the base 309 is located on the side 301c of the bonding head 301;
  • the first bracket 310-1 is arranged parallel to the pick-up surface 301b, and is used for fixedly connecting the first aligner 3041-1 and the base 309;
  • the second bracket 310 - 2 is arranged parallel to the pick-up surface 301 b and is used for fixedly connecting the second aligner 3041 - 2 and the base 309 .
  • FIG. 7 shows a partial top view of the bonding system 300 on the xoy plane
  • the bonding assembly may include a first first optical path 302-1 and a second first optical path 302-2 arranged side by side along the y direction
  • the first alignment assembly may include a first aligner 3041-1 and a second aligner 3041-2 arranged side by side along the y direction.
  • the first aligner 3041-1 is fixedly connected to the base 309 through the first bracket 310-1
  • the second aligner 3041-2 is fixedly connected to the base 309 through the second bracket 310-2. It can be understood that , the projections of the first aligner 3041-1 and the second aligner 3041-2 on the xoz plane coincide, that is, the first aligner 3041-1 and the second aligner 3041-2 are located at the bonding The same side of the head 301.
  • the bonding system 300 further includes:
  • Two bases 309 are respectively arranged on both sides of the bonding head 301;
  • the first bracket 310-1 is arranged parallel to the pick-up surface 301b, and is used for fixedly connecting the first aligner 3041-1 and the first base 309;
  • the second bracket 310 - 2 is arranged parallel to the pick-up surface 301 b and is used for fixedly connecting the second aligner 3041 - 2 and the second base 309 .
  • FIG. 8 shows a schematic diagram of a partial structure of the bonding system 300.
  • the bonding assembly may include two first optical pathways arranged side by side along the x direction, and the first alignment assembly may include a first alignment assembly arranged side by side along the x direction.
  • the first aligner 3041-1 and the second aligner 3041-2 when the bonding head 301 is in the second position, the first aligner 3041-1 is aligned with the first optical path on the left side of the bonding head 301,
  • the second aligner 3041 - 2 is aligned with the first light path on the right side of the bond head 301 .
  • the first aligner 3041-1 is fixedly connected to the base 309 on the left side of the bonding head through the first bracket 310-1
  • the second aligner 3041-2 is connected to the base 309 on the left side of the bonding head through the second bracket 310-2.
  • the base 309 on the right side of the head is fixedly connected, and the first aligner 3041 - 1 and the second aligner 3041 - 2 are located on both sides of the bonding head 301 .
  • Fig. 9 is a schematic flowchart of a bonding method according to an embodiment of the present disclosure
  • Fig. 10a to Fig. 10d are process schematic diagrams of a bonding method according to an embodiment of the present disclosure. This bonding method is applied to the bonding of the first to-be-bonded die 11 and the second to-be-bonded die 21 located on the wafer, and the method includes the following steps:
  • S400 Move the wafer or the first die to be bonded according to the first deviation value and the second deviation value, so as to align the first die to be bonded and the second die to be bonded;
  • step S100 as shown in FIG. 10a, when the bonding head 301 picking up the first die 11 to be bonded is located at the first position, the first alignment component 304' can be used to detect that the picking surface is compared to the horizontal plane (for example , the offset of the xoy plane), which can be represented by the angle between the picking plane and the horizontal plane, and can also be expressed by the heights of multiple points on the picking plane compared to the same horizontal plane.
  • the horizontal plane for example , the offset of the xoy plane
  • step S200 after determining the offset of the pick-up plane compared to the horizontal plane, the bonding assembly can drive the bond head 301 to rotate until the pick-up plane is parallel to the horizontal plane, and adjust the bond head 301 to second position.
  • the second position indicates the position where the first optical channel 302 can receive the detection optical signal emitted by the first alignment component 304'.
  • step S300 as shown in FIG. 10c, after the first deviation value is determined, the wafer 20 is transferred to the carrier platform, and the second alignment component 305 can be used to determine the second die 21 to be bonded on the wafer 20.
  • first target position and the second target position have the same position in the horizontal direction and different positions in the vertical direction. That is, x 0 is the same as x 0 ', y 0 is the same as y 0 ', the coordinates in the z direction of the first target position and the second target position are different, and the different reference signs are only for the convenience of distinguishing the first target position and the second target location, and is not intended to limit the disclosure.
  • step S400 according to the first deviation value and the second deviation value, it is determined that the displacement deviation between the first die to be bonded 11 and the second die to be bonded 21 is ( ⁇ x T - ⁇ x B , ⁇ y T - ⁇ y B ), the bonding assembly can be kept fixed (that is, the first die 11 to be bonded is kept fixed), and the wafer 20 moves horizontally according to the displacement deviation ( ⁇ x T - ⁇ x B , ⁇ y T - ⁇ y B ), so that the second die to be bonded
  • the bonding die 21 is aligned with the first to-be-bonded die 11 , that is, the positions of the first to-be-bonded die 11 and the second to-be-bonded die 21 in the horizontal direction are the same.
  • the moving direction of the wafer 20 can be determined according to the positive or negative value of ⁇ x T - ⁇ x B.
  • ⁇ x T - ⁇ x B is a negative value
  • the wafer 20 moves along the negative direction of the x-axis, and the moving distance is ( ⁇ x T - the absolute value of ⁇ x B ).
  • ⁇ x T - ⁇ x B is a positive value
  • the wafer 20 moves along the positive direction of the x-axis, and the moving distance is ( ⁇ x T - ⁇ x B ).
  • the moving direction of the wafer 20 can be determined according to the positive or negative value of ⁇ y T - ⁇ y B , for example, when ⁇ y T - ⁇ y B is a negative value, the wafer 20 moves along the negative direction of the y-axis, and the moving distance is ( ⁇ y Absolute value of T - ⁇ y B ). When ⁇ y T ⁇ y B is a positive value, the wafer 20 moves along the positive direction of the y-axis, and the moving distance is ( ⁇ y T ⁇ y B ).
  • step S400 according to the first deviation value and the second deviation value, it is determined that the displacement deviation between the first die to be bonded 11 and the second die to be bonded 21 is ( ⁇ x T - ⁇ x B , ⁇ y T - ⁇ y B ), the wafer 20 can also be kept fixed (that is, the second die to be bonded 21 is kept fixed), and the first die to be bonded 11 moves horizontally according to the displacement deviation ( ⁇ x T - ⁇ x B , ⁇ y T - ⁇ y B ) , so that the first to-be-bonded die 11 is aligned with the second to-be-bonded die 21 .
  • the moving direction of the first die to be bonded 11 can be determined according to the positive or negative value of ⁇ x T - ⁇ x B , for example, when ⁇ x T - ⁇ x B is a negative value, the first die to be bonded 11 is positive along the x-axis direction, the moving distance is the absolute value of ( ⁇ x T - ⁇ x B ). When ⁇ x T - ⁇ x B is a positive value, the first die to be bonded 11 moves along the negative direction of the x-axis, and the moving distance is ( ⁇ x T - ⁇ x B ).
  • the moving direction of the first die to be bonded 11 can be determined according to the positive or negative value of ⁇ y T - ⁇ y B , for example, when ⁇ y T - ⁇ y B is a negative value, the first die to be bonded 11 moves along the y-axis Move in the positive direction, and the moving distance is the absolute value of ( ⁇ y T - ⁇ y B ).
  • ⁇ y T ⁇ y B is a positive value
  • the first die to be bonded 11 moves along the negative direction of the y-axis, and the moving distance is ( ⁇ y T ⁇ y B ).
  • step S400 after determining the first deviation value and the second deviation value, the wafer or the first die to be bonded 11 may be moved to align the first die to be bonded and the second die to be bonded .
  • the wafer is moved so that the second die to be bonded is aligned with the first die to be bonded.
  • step S500 as shown in FIG. 10d, after the first die to be bonded 11 and the second die to be bonded 21 are aligned, the bonding assembly moves vertically downwards to place the first die to be bonded 11 and the second die 21 to be bonded.
  • the first die to be bonded 11 is a die that exists independently of the wafer. In the manufacturing process of a semiconductor device, usually a plurality of dies are formed on a wafer.
  • the first die 11 to be bonded refers to an independent die formed after dicing the wafer carrying the plurality of dies. It can be understood that the size of the first die to be bonded is much smaller than the size of the wafer.
  • the bonding head when the bonding head is at the first position, by determining the offset of the pickup surface of the bonding head compared to the horizontal plane, the bonding head can be driven to move to the second position parallel to the horizontal plane according to the offset. Location. That is, before die-to-wafer bonding, the picked-up die can be adjusted to be parallel to the horizontal plane, which is beneficial to reduce the probability of damage due to uneven local force during die bonding, and improve the die-to-wafer bond. Combined yield.
  • the accurate alignment of the die to be bonded is beneficial to improve the bonding accuracy from the die to the wafer.
  • the bonding head includes: a side surface perpendicular to the pick-up surface, a top surface parallel to the pick-up surface, and at least three calibration marks; wherein at least three calibration marks are located on the side or top surface, at least three calibration marks The plane where the mark is located is parallel to the pick plane;
  • the offset is determined according to the first sub-deviation, the second sub-deviation and the third sub-deviation.
  • the calibration marks m 1 , m 2 and m 3 are located on the side 301c, and the plane formed by the calibration marks m 1 , m 2 , and m 3 is recorded as the calibration plane, and the calibration planes are parallel to on the pick-up surface 301b.
  • the driving unit drives the moving unit 308 to rotate clockwise by a first angle.
  • the fixing unit 307, the bonding head 301 and the first die 11 to be bonded rotate synchronously.
  • the driving unit drives the moving unit 308 to rotate clockwise by a second angle
  • the first alignment component 304' acquires a third image including the third calibration mark m3 , and determines the position of the third calibration mark m3 according to the third image
  • the third sub-deviation ⁇ y m3 y m3 ⁇ y c between the position (x m3 , y m3 ) and the preset position (x c , y c ).
  • the offset of the pick-up plane compared to the horizontal plane is determined from ⁇ y m1 , ⁇ y m2 and ⁇ y m3 .
  • the preset position represents the center position of the viewing frame used by the first alignment component 304 ′ to determine the positions of the calibration marks m 1 , m 2 and m 3 , which is denoted as (x c , y c ).
  • the driving unit can also drive the mobile unit to rotate counterclockwise, so that the aligner can be rotated to the point where the second image including the second calibration mark and the third image including the third calibration mark can be acquired.
  • the choice is made according to actual bonding requirements, and the present disclosure is not further limited here.
  • the first angle and the second angle can be determined according to the arrangement of the calibration marks m 1 , m 2 and m 3 .
  • the shape of the bonding head can be cylindrical
  • the shape of the pick-up surface and the calibration plane can be circular
  • the first angle can be based on the center of the circle formed by the calibration mark m 1 , the calibration mark m 2 and the center of the calibration plane
  • the second angle can be determined based on the center angle formed by the calibration mark m 2 , the calibration mark m 3 and the center of the calibration plane.
  • the calibration marks m 1 , m 2 and m 3 can be arranged at equal intervals on the side surface 301c, so the first angle and the second angle are 120°.
  • the line connecting the calibration marks m1 and m3 passes through the center of the calibration plane, the line connecting the calibration mark m2 and the center of the calibration plane is perpendicular to the line connecting the calibration marks m1 and m3 , then the first angle and the second angle are 90 °.
  • the calibration marks m 1 , m 2 , and m 3 are shown on the side 301c in this example, in other examples, the calibration marks m 1 , m 2 , and m 3 may also be located on the top surface 301a. For example, on the boundary line of the top surface 301a.
  • the number of calibration marks on the side surface 301c or on the top surface 301a is not limited to three, but can also be four, five or even more, which is not limited by the present disclosure.
  • the bonding head includes: a side surface perpendicular to the pick-up surface, a top surface parallel to the pick-up surface, and at least three calibration marks; wherein at least three calibration marks are located on the side or top surface, at least three calibration marks The plane where the mark is located is parallel to the pick plane;
  • An offset is determined based on at least three sub-deviations.
  • the first alignment component 304' may include three aligners 3041' and three second optical paths 306'. specifically:
  • the first aligner 3041' and the first second optical path 306' are located on the right side of the first calibration mark m1 , and are used to obtain a sub-deviation between the first calibration mark m1 and a preset position.
  • the second aligner 3041' and the second second optical path 306' are located on the side of the second calibration mark m 2 facing the positive direction of the y-axis, and are used to obtain the difference between the second calibration mark m 2 and the preset position. Sub-deviations between.
  • the third aligner 3041' and the third third optical path 306' are located on the left side of the third calibration mark m3 , and are used to obtain the sub-deviation between the third calibration mark m3 and the preset position.
  • the first alignment component 304' determines the offset of the pick-up surface 301b relative to the horizontal plane based on the three sub-deviations.
  • the bonding head includes three calibration marks, and the first alignment component includes three aligners.
  • the bond head can include at least four alignment marks and the first alignment assembly can include at least four aligners.
  • the setting of the number or position of the calibration marks in the bonding head is not limited here, as long as the plane on which the calibration marks are set is parallel to the pick-up surface.
  • the distance between the at least three calibration marks and the preset position can be obtained without driving the bonding head to rotate.
  • Sub-bias which is beneficial to improve the accuracy of the three sub-biases.
  • three sub-deviations can be obtained simultaneously, which can shorten the time for determining the offset, and further shorten the time for levelness detection and level adjustment before die-to-wafer bonding, and improve bonding efficiency.
  • the bonding head includes: at least two alignment optical pathways that are perpendicular to the side of the pick-up surface and penetrate the bonding head from the side; wherein, the planes where the at least two alignment optical pathways are located are parallel to the pick-up plane;
  • An offset is determined according to the first sub-offset and the second sub-offset.
  • two calibration light paths 3111 and 3112 penetrate the bonding head 301 from the side 301c, and the plane formed by the two calibration light paths 3111 and 3112 is marked as a calibration plane, and the calibration planes are parallel to on the pick-up surface 301b.
  • the transmitting unit 3042 transmits a calibration optical signal to the first calibration optical path 3111, and the receiving unit 3043 transmits the calibration optical signal according to the intensity I1 of the transmitted calibration optical signal and the intensity I11 of the received calibration optical signal ', determine the first suboffset.
  • the driving unit drives the bonding head 301 to rotate until the second calibration optical path 3112 is aligned with the transmitting unit 3042, and the transmitting unit 3042 transmits a calibration optical signal to the second calibration optical path 3112, and the receiving unit 3043 according to the intensity of the transmitted calibration optical signal I 2 and the intensity I 2 ′ of the received calibration optical signal determine the second sub-offset.
  • the first alignment component can determine the offset of the pick-up surface 301b compared to the horizontal plane.
  • the number of calibration light paths penetrating the bond head 301 from the side 301c is not limited to two, and may be three or even more
  • the present disclosure is not limited here, it only needs to ensure that the calibration light path does not overlap with the first light path 302 .
  • the bonding head includes: at least two alignment optical pathways that are perpendicular to the side of the pick-up surface and penetrate the bonding head from the side; wherein, the planes where the at least two alignment optical pathways are located are parallel to the pick-up plane;
  • An offset is determined according to the first sub-offset and the second sub-offset.
  • the first alignment component 304 may include two transmitting units 3042 and two receiving units 3043 . specifically:
  • the first transmitting unit 3042 and the first receiving unit 3043 are arranged side by side along the x direction and located on opposite sides of the first calibration light path 3111 along the x direction, for determining the first sub-offset
  • the second transmitting unit 3042 and the second receiving unit 3043 are arranged side by side along the x direction, and located on opposite sides of the second calibration optical path 3112 along the x direction, and are used for determining the second sub-offset.
  • the first alignment component 304 can determine the offset of the pick-up surface 301b compared to the horizontal plane.
  • FIG. 11 is a schematic diagram of a die-to-wafer structure according to an embodiment of the present disclosure.
  • the die-to-wafer structure applies the bonding system 300 in any of the above-mentioned embodiments and applies the bonding system 300 in any of the above-mentioned embodiments.
  • Made with bonding methods including:
  • the wafer 20 includes: a plurality of second dies 21 ; wherein, each first die 11 is bonded to each second die 21 .

Landscapes

  • Container, Conveyance, Adherence, Positioning, Of Wafer (AREA)
  • Wire Bonding (AREA)
  • Die Bonding (AREA)

Abstract

本公开实施例公开了一种键合系统和键合方法。键合系统包括：键合组件，包括：键合头以及贯穿键合头的第一光通路；第一对准组件，用于在键合头位于第一位置时，确定拾取面相较于水平面的偏移量；键合组件，还用于根据偏移量，驱动键合头移动至平行于水平面的第二位置；第一对准组件，还用于在键合头位于第二位置时，通过第一光通路的第二端向第一端发射检测光信号；第一对准组件，还用于接收检测光信号经第一待键合管芯反射的反射光信号，并根据接收的反射光信号确定第一待键合管芯的当前位置与第一目标位置之间的第一偏差值；第二对准组件，用于确定晶圆上第二待键合管芯的当前位置与第二目标位置的第二偏差值。

Description

键合系统和键合方法
相关申请的交叉引用
本申请基于申请号为202111594630.2、申请日为2021年12月24日、发明名称为“键合系统和键合方法”的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本公开实施例涉及但不限于半导体制造领域，尤其涉及一种键合系统和键合方法。
背景技术
在半导体制造领域，采用键合技术可实现半导体器件的三维集成。通过将两个或多个功能相同或不同的半导体结构进行键合，可以提高芯片的性能，同时也可以大幅度缩短待键合对象之间的金属互连，减小发热、功耗和延迟。
键合制程可按键合对象区分，包括晶圆到晶圆(wafer to wafer)键合、芯片(或称为管芯)到晶圆(die to wafer)键合以及芯片到芯片(die to die)键合。
发明内容
根据本公开实施例的第一方面，提供一种键合系统，包括：
键合组件，包括：键合头以及贯穿所述键合头的第一光通路；其中，所述第一光通路的第一端位于所述键合头拾取所述第一待键合管芯的拾取面；
晶圆承载台，用于承载晶圆；
第一对准组件，用于在所述键合头位于第一位置时，确定所述拾取面相较于水平面的偏移量；
所述键合组件，还用于根据所述偏移量，驱动所述键合头移动至平行于水平面的第二位置；
所述第一对准组件，还用于在所述键合头位于第二位置时，通过所述第一光通路的第二端向所述第一端发射检测光信号；其中，拾取的所述第一待键合管芯覆盖所述第一端并反射所述检测光信号；
所述第一对准组件，还用于接收所述检测光信号经所述第一待键合管芯反射的反射光信号，并根据接收的所述反射光信号确定所述第一待键合管芯的当前位置与第一目标位置之间的第一偏差值；
第二对准组件，位于所述晶圆承载台相对远离所述键合组件的一侧，用于确定所述晶圆上第二待键合管芯的当前位置与第二目标位置的第二偏差值；
所述晶圆承载台，还用于根据所述第一偏差值和所述第二偏差值，驱动承载的所述晶圆相对所述拾取面移动，以使所述第二待键合管芯对准所述第一待键合管芯；或者，所述键合组件，还用于根据所述第一偏差值和所述第二偏差值，驱动拾取所述第一待键合管芯的所述拾取面相对所述晶圆承载台移动，以使所述第一待键合管芯对准所述第二待键合管芯；
所述键合组件，还用于键合所述第一待键合管芯和所述第二待键合管芯。
在一些实施例中，所述键合头包括：垂直于所述拾取面的侧面、平行于所述拾取面的顶面、以及至少三个校准标记；其中，所述至少三个校准标记位于所述侧面或所述顶面，所述至少三个校准标记所在的平面平行于所述拾取面；
所述第一对准组件，具体用于根据所述至少三个校准标记和预设位置之间的子偏差，确定所述偏移量。
在一些实施例中，所述第一对准组件，具体用于在所述键合头位于所述第一位置时，获取包括第一个校准标记的第一图像，并根据第一图像确定第一个校准标记的位置与预设位置之间的第一子偏差；
所述键合组件，还用于驱动键合头旋转第一角度，以使所述第一对准组件能够获取到包括第二个所述校准标记的第二图像；
所述第一对准组件，具体还用于根据获取的所述第二图像，确定第二个所述校准标记的位置与所述预设位置之间的第二子偏差；
所述键合组件，还用于驱动所述键合头旋转第二角度，以使所述第一对准组件能够获取到包括第三个所述校准标记的第三图像；
所述第一对准组件，具体还用于根据获取的所述第三图像，确定第三个所述校准标记的位置与所述预设位置之间的第三子偏差，并根据所述第一子偏差、所述第二子偏差和所述第三子偏差，确定所述偏移量。
在一些实施例中，所述第一对准组件包括：至少三个对准器，每个所述对准器用于获取一个所述校准标记与预设位置之间的子偏差；
所述第一对准组件，具体用于根据至少三个所述子偏差确定所述偏移量。
在一些实施例中，所述键合头包括：垂直于所述拾取面的侧面、从所述侧面贯穿所述键合头的至少两条校准光通路；其中，所述至少两条校准光通路所在的平面平行于所述拾取面；
所述第一对准组件，具体用于根据所述至少两条校准光通路与水平面之间的子偏移，确定所述偏移量。
在一些实施例中，所述第一对准组件包括：一个发射单元和一个接收单元；
其中，在所述键合头位于所述第一位置时，所述发射单元与第一条所述校准光通路的一端对准，所述接收单元与第一条所述校准光通路的另一端的至少部分区域对准；所述发射单元用于向第一条所述校准光通路的一端发射校准光信号；所述接收单元，用于根据接收的经第一条所述校准光通路传输的校准光信号，确定第一子偏移；
所述键合组件，还用于驱动所述键合头旋转，以使第二条所述校准光通路的一端对准所述发射单元；其中，所述接收单元与第二条所述校准光通路的另一端的至少部分区域对准；
所述发射单元，还用于向所述第二条所述校准光通路的一端发射校准光信号；所述接收单元，还用于根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
所述第一对准组件，具体用于根据所述第一子偏移和所述第二子偏移，确 定所述偏移量。
在一些实施例中，所述第一对准组件包括：
两个发射单元和两个接收单元；
在所述键合头位于所述第一位置时，第一个所述发射单元与第一条所述校准光通路的一端对准，第一个所述接收单元与第一条所述校准光通路的另一端的至少部分区域对准，第二个所述发射单元与第二条所述校准光通路的一端对准，第二个所述接收单元与第二条所述校准光通路的另一端的至少部分区域对准；
第一个所述发射单元，用于向第一条所述校准光通路的一端发射校准光信号；
第一个所述接收单元，用于根据接收的经第一条所述校准光通路传输的校准光信号，确定第一子偏移；
第二个所述发射单元，用于向第二条所述校准光通路的一端发射校准光信号；
第二个所述接收单元，用于根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
所述第一对准组件，具体用于根据所述第一子偏移和所述第二子偏移，确定所述偏移量。
根据本公开实施例的第二方面，提供一种键合方法，所述键合方法应用于第一待键合管芯与位于晶圆的第二待键合管芯的键合，所述键合方法包括：
在拾取有第一待键合管芯的键合头位于第一位置时，确定所述键合头拾取所述第一待键合管芯的拾取面相较于水平面的偏移量；
根据所述偏移量，驱动所述键合头移动至平行于所述水平面的第二位置；
在所述键合头位于所述第二位置时，确定拾取的所述第一待键合管芯的当前位置与第一目标位置之间的第一偏差值，确定所述第二待键合管芯的当前位置与第二目标位置之间的第二偏差值；
根据所述第一偏差值和所述第二偏差值，移动所述晶圆或所述第一待键合管芯，以对准所述第一待键合管芯和所述第二待键合管芯；
在所述第一待键合管芯和所述第二待键合管芯对准后，键合所述第一待键合管芯和所述第二待键合管芯。
在一些实施例中，
所述键合头包括：垂直于所述拾取面的侧面、平行于所述拾取面的顶面、以及至少三个校准标记；其中，所述至少三个校准标记位于所述侧面或所述顶面，所述至少三个校准标记所在的平面平行于所述拾取面；
所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
获取包括第一个校准标记的第一图像；
根据所述第一图像，确定所述第一个校准标记的位置与预设位置之间的第一子偏差；
在获取所述第一图像后，驱动所述键合头旋转第一角度；
在所述键合头旋转第一角度后，获取包括第二个校准标记的第二图像；
根据所述第二图像，确定所述第二个校准标记的位置与所述预设位置之间的第二子偏差；
在获取所述第二图像后，驱动所述键合头旋转第二角度；
在所述键合头旋转第二角度后，获取包括第三个校准标记的第三图像；
根据所述第三图像，确定所述第三个校准标记的位置与所述预设位置之间的第三子偏差；
根据所述第一子偏差、所述第二子偏差和所述第三子偏差，确定所述偏移量。
在一些实施例中，
所述键合头包括：垂直于所述拾取面的侧面、平行于所述拾取面的顶面、以及至少三个校准标记；其中，所述至少三个校准标记位于所述侧面或所述顶面，所述至少三个校准标记所在的平面平行于所述拾取面；
所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
同时获取所述至少三个校准标记与预设位置之间的子偏差；
根据至少三个所述子偏差确定所述偏移量。
在一些实施例中，所述键合头包括：垂直于所述拾取面的侧面、从所述侧面贯穿所述键合头的至少两条校准光通路；其中，所述至少两条校准光通路所在的平面平行于所述拾取面；
所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
向第一条所述校准光通路的一端发射校准光信号；
根据接收的经第一条所述校准光通路传输的校准光信号，确定第一子偏移；
在确定所述第一子偏移后，驱动所述键合头旋转；
在所述键合头旋转之后，向第二条所述校准光通路的一端发射校准光信号；
根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
根据所述第一子偏移和所述第二子偏移，确定所述偏移量。
在一些实施例中，所述键合头包括：垂直于所述拾取面的侧面、从所述侧面贯穿所述键合头的至少两条校准光通路；其中，所述至少两条校准光通路所在的平面平行于所述拾取面；
所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
向第一条所述校准光通路的一端发射校准光信号，并向第二条所述校准光通路的一端发射校准光信号；
根据接收的经第一条所述校准光通路传输的校准光信号，确定第一子偏移；
根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
根据所述第一子偏移和所述第二子偏移，确定所述偏移量。
本公开实施例中，通过设置第一对准组件，可确定键合头的拾取面相较于水平面的偏移量，键合组件根据该偏移量可驱动键合头移动至平行于水平面。即在管芯到晶圆键合前，可将拾取的管芯调整至平行于水平面，有利于减小管芯键合时因局部受力不均而破损的概率，提高管芯到晶圆键合的良率。
在将管芯调整至平行于水平面后，还可通过第一对准组件确定第一待键合管芯的当前位置与第一目标位置之间的第一偏差值，以及通过设置第二对准组件，可确定晶圆上第二待键合管芯的当前位置与第二目标位置的第二偏差值，根据第一偏差值和第二偏差值来调整晶圆或第一待键合管芯，可使得第一待键合管芯与第二待键合管芯精确对准，有利于提高管芯到晶圆的键合精度，进而可降低管芯到晶圆的键合难度。
附图说明
为了更清楚地说明本公开具体实施方式或现有技术中的技术方案，下面将对具体实施方式或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本公开的一些实施方式，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是根据一示例性实施例示出的一种晶圆到晶圆键合过程的示意图；
图2a是根据一示例性实施例示出的一种芯片到晶圆键合的示意图一；
图2b是根据一示例性实施例示出的一种芯片到晶圆键合的示意图二；
图3a是根据本公开实施例示出的一种键合系统的示意图一；
图3b是根据本公开实施例示出的一种键合系统的光路示意图一；
图3c是根据本公开实施例示出的一种键合系统的光路示意图二；
图4a是根据本公开实施例示出的一种键合系统的示意图二；
图4b是根据本公开实施例示出的一种键合系统的光路示意图三；
图5a是根据本公开实施例示出的一种键合系统的示意图三；
图5b是根据本公开实施例示出的一种键合系统的光路示意图四；
图6是根据本公开实施例示出的一种键合系统的示意图四；
图7是根据本公开实施例示出的一种键合系统的示意图五；
图8是根据本公开实施例示出的一种键合系统的示意图六；
图9是根据本公开实施例示出的一种键合方法的流程示意图；
图10a是根据本公开实施示出的一种键合方法的过程示意图一；
图10b是根据本公开实施示出的一种键合方法的过程示意图二；
图10c是根据本公开实施示出的一种键合方法的过程示意图三；
图10d是根据本公开实施示出的一种键合方法的过程示意图四；
图11是根据本公开实施例示出的一种管芯到晶圆结构的示意图。
附图标记：
101—上晶圆；102、102’—下晶圆；201—上芯片；202—下晶圆；300—键合系统；11—第一待键合管芯/第一管芯；21—第二待键合管芯/第二管芯；20—晶圆；301—键合头；301a—顶面；301b—拾取面；301c—侧面；302、302’—第一光通路；302a、302a’—第一光通路的第一端；302b、302b’—第一光通路的第二端；3021—第一反射镜；303—晶圆承载台；304、304’、304”—第一对准组件；3041、3041’、3041”—对准器；305—第二对准组件；306、306’、306”—第二光通路；m 1—第一个校准标记；m 2—第二个校准标记；m 3—第三个校准标记；3111—第一条校准光通路；3111a—第一条校准光通路的右端；3111b—第一条校准光通路的左端；3112—第二条校准光通路；3112a—第二条校准光通路的右端；3112b—第二条校准光通路的左端；3042—发射单元；3043—接收单元；3061—第一子路；3062—第二子路；3063—第二反射镜；307—固定单元；308—移动单元；309—基座；310-1—第一支架；310-2—第二支架；3041-1—第一个对准器；3041-2—第二个对准器；302-1—第一个第一光通路；302-2—第二个第一光通路。
具体实施方式
提供下述实施例是为了更好地进一步理解本公开，并不局限于所述最佳实施方式，不对本公开的内容和保护范围构成限制，任何人在本公开的启示下或是将本公开与其他现有技术的特征进行组合而得出的任何与本公开相同或相近似的产品，均落在本公开的保护范围之内。
在本公开的描述中，需要说明的是，术语“上”、“下”、“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描 述本公开和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本公开的限制。此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性。
图1是根据一示例性实施例示出的一种晶圆到晶圆键合过程的示意图，参照图1所示，在将上晶圆101与下晶圆102键合前，需进行水平调节和视觉对位。具体地，水平调节包括将上晶圆101和下晶圆102分别调节至平行于水平面，视觉对位包括在垂直于水平面的铅垂方向上对准上晶圆101和下晶圆102。
需要指出的是，由于晶圆的尺寸较大，承载晶圆的晶圆承载台(stage)的尺寸也较大，通常可在晶圆承载台中整合部件，以实现晶圆的水平度检测及水平调节。例如，可通过设置于晶圆承载台中的部件量测晶圆承载台(stage)表面多个位置的高度偏差来实现，方式简单且精度要求较低。
在将上晶圆101和下晶圆102调节至平行于水平面后，利用上镜头识别下晶圆102顶面的对位标记位置，并记录为(x 2，y 2)，上镜头位于上晶圆101相对远离下晶圆102的一侧。利用下镜头识别上晶圆101底面的对位标记位置，并记录为(x 1，y 1)，下镜头位于下晶圆102相对远离上晶圆101的一侧。上晶圆101底面的对位标记与下晶圆102顶面的对位标记之间的理论对位偏差即为(x 1-x 2，y 1-y 2)。
需要指出的是，在利用下镜头识别上晶圆101底面的对位标记位置时，需将下晶圆102移开(例如，将下晶圆从102’所在的位置移动至102所在的位置)，以避免下晶圆102阻挡下镜头的视野。类似地，在利用上镜头识别下晶圆102顶面的对位标记位置时，需将上晶圆101移开，以避免上晶圆101阻挡上镜头的视野。
在移开下晶圆或上晶圆的过程中，还可能产生移动误差。具体地，在进行晶圆到晶圆键合时，由于晶圆的尺寸相对较大，只需利用上镜头/下镜头分别识别下晶圆102/上晶圆101的对位标记位置，并记录最后一次晶圆的移动误差(Δx，Δy)，该移动误差可根据吸附晶圆的静电吸盘的位移偏差来监控。因此， 上晶圆101底面的对位标记与下晶圆102顶面的对位标记之间的实际对位偏差即为(x 1-x 2+Δx，y 1-y 2+Δy)，根据该实际对位偏差进行校正对位，并将上晶圆101的底面与下晶圆102的顶面键合。
芯片到晶圆键合与晶圆到晶圆键合不同。具体地，在晶圆到晶圆键合的制程中，顶部键合对象为晶圆，而在芯片到晶圆键合的制程中，顶部键合对象为芯片，芯片的尺寸远小于晶圆的尺寸。因此，芯片到晶圆键合的键合装置相较于晶圆到晶圆键合的键合装置会有很大的区别。例如，来料接收、水平调节、视觉对位、向下键合方式等。
图2a和图2b是根据一示例性实施例示出的一种芯片到晶圆键合的示意图，参照图2a和图2b所示，在将上芯片201与下晶圆202键合前，需进行水平调节和视觉对位。具体地，先将上芯片201和下晶圆202分别调节至平行于水平面，再将上芯片201的对位标记与下晶圆202中待键合管芯的对位标记对位。
相较于传统的晶圆到晶圆键合，芯片到晶圆键合的精度要求更高、键合难度更大。在水平调节方面，由于芯片的尺寸远小于晶圆的尺寸，拾取芯片的键合头的尺寸也远小于拾取晶圆的部件的尺寸，在键合头中整合较多的部件，将会增加键合头的重量，导致键合头的惯性增加。因此，芯片到晶圆键合的水平度检测及水平调节成为难题。并且当拾取芯片的键合头处于非水平位置时，芯片的局部由于受力不均极易破损，导致芯片到晶圆的键合失败。
需要指出的是，由于下晶圆202以及承载下晶圆202的晶圆承载台的惯性较大，下晶圆202难以做到在竖直方向上高频率、高精度地运动。为了满足芯片到晶圆键合的高频需求，通常采用键合头竖直向下移动，以实现管芯到晶圆的键合。因此，需保证键合头足够轻巧才不至于惯性过大。
在视觉对位方面，在将上芯片201移动至与下晶圆202中待键合管芯对准位置的过程中，由于下晶圆202对于下镜头的阻挡范围较大，上芯片201移动到与下晶圆202中待键合管芯对准位置而产生的位移误差，下镜头也难以识别记录，芯片到晶圆的键合精度较低。
有鉴于此，本公开实施例提供一种键合系统和键合方法。
图3a至图3c是根据本公开实施例示出的一种键合系统300的示意图，图3a表示的是键合系统300的结构示意图，图3b和图3c表示的是键合系统300的局部示意图，参照图3a所示，键合系统300包括：
键合组件，包括：键合头301以及贯穿键合头301的第一光通路302；其中，第一光通路302的第一端302a位于键合头301拾取第一待键合管芯11的拾取面301b；
晶圆承载台303，用于承载晶圆20；
第一对准组件304，用于在键合头301位于第一位置时，确定拾取面301b相较于水平面的偏移量；
键合组件，还用于根据偏移量，驱动键合头301移动至平行于水平面的第二位置；
第一对准组件304，还用于在键合头301位于第二位置时，通过第一光通路302的第二端302b向第一端302a发射检测光信号；其中，拾取的第一待键合管芯11覆盖第一端302a并反射检测光信号；
第一对准组件304，还用于接收检测光信号经第一待键合管芯11反射的反射光信号，并根据接收的反射光信号确定第一待键合管芯11的当前位置与第一目标位置之间的第一偏差值；
第二对准组件305，位于晶圆承载台303相对远离键合组件的一侧，用于确定晶圆20上第二待键合管芯21的当前位置与第二目标位置的第二偏差值；
晶圆承载台303，还用于根据第一偏差值和第二偏差值，驱动承载的晶圆20相对拾取面301b移动，以使第二待键合管芯21对准第一待键合管芯11；或者，键合组件，还用于根据第一偏差值和第二偏差值，驱动拾取第一待键合管芯11的拾取面301b相对晶圆承载台303移动，以使第一待键合管芯11对准第二待键合管芯21；
键合组件，还用于键合第一待键合管芯11和第二待键合管芯21。
示例性地，在键合组件的键合头301拾取第一待键合管芯11后，键合组件可沿水平方向(例如，x方向或y方向)和/或竖直方向(例如z方向)移动，将键合头301调整至第一位置。这里，第一位置表示的是键合头301进行水平度检测和水平调节之前的位置。
在键合头301位于第一位置时，可利用第一对准组件304检测拾取面301b相较于水平面(例如，xoy平面)的偏移量，该偏移量可通过拾取面301b与水平面之间的夹角来表示，还可通过拾取面301b上多个点相较于同一水平面的高度来表示。
在确定拾取面301b相较于水平面的偏移量后，键合组件可驱动键合头301旋转，直至拾取面301b平行于水平面，并将键合头301调整至第二位置。这里，第二位置表示的是键合头301进行水平度检测和水平调节之后，并且第一光通路302的第二端302b能够接收到第一对准组件304发射的检测光信号的位置。
示例性地，参照图3a和图3c所示，在键合头301位于第二位置时，第一对准组件304发送检测光信号，检测光信号经由第一光通路302的第二端302b和第一端302a传输至第一待键合管芯11的定位标记，并形成反射光信号，反射光信号经由第一光通路302的第一端302a和第二端302b传输至第一对准组件304。
示例性地，第一对准组件304在接收到反射光信号后，可根据反射光信号确定第一待键合管芯11的当前位置(x 1，y 1)与第一目标位置(x 0，y 0)之间的第一偏差值(Δx T，Δy T)，其中，Δx T＝x 1-x 0，Δy T＝y 1-y 0。
示例性地，参照图3a所示，可利用第二对准组件305确定晶圆20上第二待键合管芯21的当前位置(x 2，y 2)与第二目标位置(x 0’，y 0’)之间的第二偏差值(Δx B，Δy B)，其中，Δx B＝x 2-x 0’，Δy T＝y 2-y 0’。
需要指出的是，第一目标位置和第二目标位置在水平方向上的位置相同、竖直方向上的位置不同。即x 0与x 0’相同，y 0与y 0’相同，第一目标位置和第二目标位置z方向的坐标不同，不同的附图标记仅是为了便于区分第一目标位置 和第二目标位置，而不用于限制本公开。
这里，第一待键合管芯11的当前位置(x 1，y 1)、第一目标位置(x 0，y 0)、第二待键合管芯21的当前位置(x 2，y 2)以及第二目标位置(x 0’，y 0’)基于同一坐标系确定。例如，可以晶圆承载台的中心为原点建立坐标系，x方向和y方向垂直、且平行于晶圆承载台所在的平面，z方向垂直于晶圆承载台所在的平面。
在管芯到晶圆键合前，可利用标准管芯对第一对准组件、第二对准组件以及键合组件的位置进行校正，以保证第一对准组件能够与第一光通路的第二端对准、且保证第二对准组件能够与标准管芯朝向晶圆承载台的表面的定位标记对准，进而确定第一目标位置和第二目标位置。这里，第一目标位置表示的是校正后标准管芯的位置，水平方向的坐标记为(x 0，y 0)，与第一目标位置竖直方向相距预设距离的位置记为第二目标位置，第二目标位置水平方向的坐标记为(x 0’，y 0’)，这里，预设距离大于零。可以理解的是，第一目标位置和第二目标位置虽然水平方向(x方向和y方向)上的位置相同，但在三维空间内并不是同一位置，第一目标位置和第二目标位置在竖直方向(z方向)上的位置不同。
可以理解的是，当第一待键合管芯位于第一目标位置，第二待键合管芯位于第二目标位置时，可认为第一待键合管芯和第二待键合管芯对准。即沿铅垂线方向，第一待键合管芯在晶圆承载台的投影与第二待键合管芯重叠。
需要强调的是，第一待键合管芯11的当前位置表示的是将键合组件调整至第二位置时，第一待键合管芯11的实际位置，其相对于第一目标位置可能存在偏差。第二待键合管芯21的当前位置表示的是将晶圆20置于晶圆承载台、且确定晶圆上需要键合的管芯为第二待键合管芯时，第二待键合管芯21的实际位置，其相对于第二目标位置可能存在偏差。
在第一待键合管芯11的当前位置为第一目标位置、第二待键合管芯21的当前位置为第二目标位置时，第一偏差值和第二偏差值均为0，通过垂直向上 移动晶圆或垂直向下移动第一待键合管芯，即可实现第一待键合管芯与第二待键合管的精准对位键合。
示例性地，根据第一偏差值和第二偏差值确定第一待键合管芯11和第二待键合管芯21之间的位移偏差为(Δx T-Δx B，Δy T-Δy B)，可保持键合组件固定(即保持第一待键合管芯11固定)，晶圆承载台303根据位移偏差驱动晶圆20水平移动(Δx T-Δx B，Δy T-Δy B)，以使得第二待键合管芯21对准第一待键合管芯11，即第一待键合管芯11和第二待键合管芯21水平方向上的位置相同。
这里，可根据Δx T-Δx B的正负确定晶圆承载台303驱动晶圆20移动的方向，例如，在Δx T-Δx B为负值时，晶圆承载台303驱动晶圆20沿x轴负方向移动，移动的距离为(Δx T-Δx B)的绝对值。在Δx T-Δx B为正值时，晶圆承载台303驱动晶圆20沿x轴正方向移动，移动的距离为(Δx T-Δx B)。
类似地，可根据Δy T-Δy B的正负确定晶圆承载台303驱动晶圆20移动的方向，例如，在Δy T-Δy B为负值时，晶圆承载台303驱动晶圆20沿y轴负方向移动，移动的距离为(Δy T-Δy B)的绝对值。在Δy T-Δy B为正值时，晶圆承载台303驱动晶圆20沿y轴正方向移动，移动的距离为(Δy T-Δy B)。
示例性地，根据第一偏差值和第二偏差值确定第一待键合管芯11和第二待键合管芯21之间的位移偏差为(Δx T-Δx B，Δy T-Δy B)，可保持晶圆承载台303固定(即保持第二待键合管芯21固定)，键合组件根据位移偏差水平移动(Δx T-Δx B，Δy T-Δy B)，以使得第一待键合管芯11对准第二待键合管芯21。
这里，可根据Δx T-Δx B的正负确定键合组件移动的方向，例如，在Δx T-Δx B为负值时，键合组件沿x轴正方向移动，移动的距离为(Δx T-Δx B)的绝对值。在Δx T-Δx B为正值时，键合组件沿x轴负方向移动，移动的距离为(Δx T-Δx B)。
类似地，可根据Δy T-Δy B的正负确定键合组件移动的方向，例如，在Δy T-Δy B为负值时，键合组件沿y轴正方向移动，移动的距离为(Δy T-Δy B)的绝对值。在Δy T-Δy B为正值时，键合组件沿y轴负方向移动，移动的距离为(Δy T-Δy B)。
本示例中，在确定第一偏差值和第二偏差值之后，可通过驱动拾取第一待 键合管芯的拾取面相对于晶圆承载台移动，或，驱动承载的晶圆相对拾取面移动，以将第一待键合管芯和第二待键合管芯对准。优选地，在确定第一偏差值和第二偏差值之后，驱动承载的晶圆相对拾取面移动，以使第二待键合管芯对准第一待键合管芯。
需要指出的是，管芯以及拾取管芯的键合头尺寸相对较小，若采用键合组件相对晶圆移动，在键合组件移动的过程中，小尺寸的键合头更容易发生波动，可能相对水平面倾斜，导致键合头拾取的管芯倾斜，影响对准，降低键合精度。
相较于键合组件相对晶圆移动来对准的方式，通过驱动晶圆相对拾取面移动，由于晶圆的尺寸相对管芯的尺寸较大，在晶圆移动的过程中，晶圆的键合面在水平面发生波动的几率较小，即与第一待键合管芯键合的第二待键合管芯较为水平，有利于进一步提高管芯到晶圆的键合精度。
键合头301可通过真空吸附或静电吸附的方式吸附第一待键合管芯11。第一待键合管芯11可包括相对设置的键合面和非键合面，在进行管芯到晶圆键合时，拾取面通过吸附第一待键合管芯11的非键合面，以拾取第一待键合管芯11。
第一待键合管芯11包括：半导体芯片，例如，存储芯片，通信芯片、人工智能芯片、LED芯片、传感芯片或显示芯片等。参照图3a所示，第一待键合管芯11包括沿z方向相对设置的正面和背面，当第一待键合管芯11的正面设置有功能结构时，第一待键合管芯11的正面为键合面，第一待键合管芯11的背面为非键合面。
需要强调的是，第一待键合管芯11、晶圆20以及第二待键合管芯21并未包括在键合系统300中，图3a中第一待键合管芯11、晶圆20以及第二待键合管芯21(虚线所示)仅为示意，以便于理解在进行管芯到晶圆键合时，待键合管芯与键合头以及晶圆之间的位置关系。
第一光通路302包括第一端302a和第二端302b，第一端302a用于输出检测光信号和接收反射光信号，第二端302b用于接收检测光信号和输出反射光信 号。第一光通路302允许检测光信号和反射光信号通过，可以是设置于键合头301内的透明窗口，还可以是设置于键合头301内的光传输介质，例如，光纤等。
第一对准组件和第二对准组件可以发射检测光信号，还可以接收反射光信号，还可将接收的反射光信号转化为可视化的图像。例如，第一对准组件可获取第一待键合管芯上定位标记的第一图像，根据第一图像确定第一偏差值。第二对准组件可获取第二待键合管芯上定位标记的第二图像，根据第二图像确定第二偏差值。第一对准组件和第二对准组件可以发射和接收红外光信号(例如，远红外光信号)。
晶圆承载台303可包括：卡盘(图中未示出)，例如，静电吸盘(ESC chuck)，用于吸附晶圆20。
在一些实施例中，晶圆承载台可相较于拾取面进行移动，以驱动晶圆相较于拾取面移动。
在另一些实施例中，晶圆承载台用于承载晶圆的台面本身不可移动，但晶圆承载台包括能够相对拾取面移动的装载针(Lift pin)，通过驱动该装载针相对拾取面移动，进而可驱动晶圆相对拾取面移动，实现晶圆的位置调整。装载针上可设置有真空孔，通过真空吸附晶圆。装载针可在平行于晶圆的平面内移动，以驱动晶圆的移动，从而调整晶圆的位置。
具体地，以晶圆承载台包括静电吸盘为例，静电吸盘包括固定基板和装载针，装载针沿静电吸盘的轴向穿过固定基板，且可沿垂直于固定基板的方向和平行于固定基板的方向移动。当静电吸盘承载有晶圆时，装载针可与该晶圆接触。当装载针沿垂直于固定基板的方向朝向拾取面运动时，承载的晶圆被装载针顶起进而与固定基板分离，且承载的晶圆跟随装载针进行运动。
晶圆20包括多个第二待键合管芯21，每个第二待键合管芯21包括衬底和位于衬底上的功能结构(例如，存储阵列或功能电路)，相邻的两个第二待键合管芯之间设置有切割道。
本公开实施例中，通过设置第一对准组件，可确定键合头的拾取面相较于水平面的偏移量，键合组件根据该偏移量可驱动键合头移动至平行于水平面。即在管芯到晶圆键合前，可将拾取的管芯调整至平行于水平面，有利于减小管芯键合时因局部受力不均而破损的概率，提高管芯到晶圆键合的良率。
在将管芯调整至平行于水平面的第二位置后，还可通过第一对准组件确定第一待键合管芯的当前位置与第一目标位置之间的第一偏差值，以及通过设置第二对准组件，可确定晶圆上第二待键合管芯的当前位置与第二目标位置的第二偏差值，根据第一偏差值和第二偏差值来调整晶圆或第一待键合管芯的位置，可使得第一待键合管芯与第二待键合管芯精确对准，有利于提高管芯到晶圆的键合精度。
在一些实施例中，参照图3a所示，键合头301包括：垂直于拾取面301b的侧面301c、平行于拾取面301b的顶面301a、以及至少三个校准标记；其中，至少三个校准标记位于侧面301c或顶面301a，至少三个校准标记所在的平面平行于拾取面301b；
第一对准组件304，具体用于根据至少三个校准标记和预设位置之间的子偏差，确定偏移量。
示例性地，图3a示出了位于侧面301c的校准标记m 1、m 2和m 3，校准标记m 1、m 2、m 3所构成的平面记为校准平面，校准平面平行于拾取面301b，可以理解的是，在键合头301位于第一位置时，通过确定校准平面相较于水平面的偏移量，可确定拾取面301b相较于水平面的偏移量。
虽然在本示例中仅示出了校准标记m 1、m 2和m 3位于侧面301c，但在其它示例中，校准标记m 1、m 2和m 3还可位于顶面301a。例如，位于顶面301a的边界线上。位于侧面301c或位于顶面301a的校准标记的数量也不限于三个，还可以是四个、五个甚至更多，本公开在此不作限制。
示例性地，在键合头301位于第一位置时，第一对准组件304可分别确定第一个校准标记m 1的位置(x m1，y m1)、第二个校准标记m 2的位置(x m2，y m2) 以及第三个校准标记m 3的位置(x m3，y m3)，根据三个校准标记的位置与预设位置之间的高度偏差，可确定校准平面相较于水平面的偏移量，进而确定拾取面301b相较于水平面的偏移量。
第一对准组件304包括用于确定校准标记m 1、m 2和m 3位置的取景框，这里预设位置表示的是取景框的中心位置，记为(x c，y c)。可以理解的是，当三个校准标记m 1、m 2和m 3与预设位置在y轴方向上的高度偏差相同，即y m1-y c、y m2-y c以及y m3-y c的数值大小相同且正负相同时，可确定校准平面平行于水平面，进而确定与校准平面平行的拾取面也平行于水平面。
需要强调的是，在键合头301位于第一位置时，每个校准标记的位置与预设位置之间的高度偏差可能不同。例如，当y m1-y c＜y m2-y c＜y m3-y c时，可确定沿z轴正方向，第三个校准标记m 3的位置高于第二个校准标记m 2的位置，第二个校准标记m 2的位置高于第一个校准标记m 1的位置，即校准平面相较于水平面倾斜，与校准平面平行的拾取面也相较于水平面倾斜。
示例性地，当校准标记m 1、m 2和m 3位于第二端302b下方时，键合组件可先根据偏移量将键合头301调整至平行于水平面的位置，再将键合头301沿z轴负方向移动至第一光通路302的第二端302b能够接收到第一对准组件304发射的检测光信号的第二位置。
类似地，当校准标记m 1、m 2和m 3位于第二端302b上方时，键合组件可先根据偏移量将键合头301调整至平行于水平面的位置，再将键合头301沿z轴正方向移动至第一光通路302的第二端302b能够接收到第一对准组件304发射的检测光信号的第二位置。
在一些实施例中，参照图3a所示，当至少三个校准标记位于侧面301c时，第一对准组件304位于键合头301的侧面，第一光通路302的第二端302b位于侧面301c。在其它实施例中，当至少三个校准标记位于顶面时，第一对准组件位于键合头之上，第一光通路的第二端位于顶面。
可以理解的是，第一对准组件可根据校准标记在键合头上的设置方式而设 置，保证第一对准组件可检测到校准标记即可。类似地，第一光通路的第二端也可根据第一对准组件的设置方式而设置，保证第一对准组件可与第一光通路对准即可。本领域技术人员可根据实际需求合理设置，本公开在此不做进一步限制。
本公开实施例中，通过在键合头的侧面或顶面设置至少三个校准标记，并且三个校准标记所在的平面平行于拾取面，通过三个校准标记和预设位置之间的子偏差，可确定校准平面相较于水平面的偏移量，进而确定与校准平面平行的拾取面相较于水平面的偏移量。
在一些实施例中，参照图3a所示，第一对准组件304，具体用于在键合头301位于第一位置时，获取包括第一个校准标记m 1的第一图像，并根据第一图像确定第一个校准标记m 1的位置与预设位置之间的第一子偏差；
键合组件，还用于驱动键合头301旋转第一角度，以使第一对准组件304能够获取到包括第二个校准标记m 2的第二图像；
第一对准组件304，具体还用于根据获取的第二图像，确定第二个校准标记m 2的位置与预设位置之间的第二子偏差；
键合组件，还用于驱动键合头301旋转第二角度，以使第一对准组件304能够获取到包括第三个校准标记m 3的第三图像；
第一对准组件304，具体还用于根据获取的第三图像，确定第三个校准标记m 3的位置与预设位置之间的第三子偏差，并根据第一子偏差、第二子偏差和第三子偏差，确定偏移量。
示例性地，结合图3a和图3b所示，第一对准组件304可包括一个对准器3041，位于键合头301的一侧，在键合头301位于第一位置时，对准器3041获取包括第一个校准标记m 1的第一图像，并根据第一图像确定第一个校准标记m 1的位置(x m1，y m1)与预设位置(x c，y c)之间的第一子偏差Δy m1＝y m1-y c。
示例性地，参照图3a所示，键合组件还包括：固定单元307、移动单元308以及与移动单元308连接的驱动单元(图中未示出)；固定单元307固定连接 键合头301和移动单元308；驱动单元用于驱动移动单元308相对晶圆承载台303移动键合头301。
示例性地，驱动单元驱动移动单元308顺时针旋转第一角度，在移动单元308转动时，固定单元307、键合头301以及第一待键合管芯11同步转动，对准器3041获取包括第二个校准标记m 2的第二图像，并根据第二图像确定第二个校准标记m 2的位置(x m2，y m2)与预设位置(x c，y c)之间的第二子偏差Δy m2＝y m2-y c。
然后，驱动单元驱动移动单元308顺时针旋转第二角度，对准器3041获取包括第三个校准标记m 3的第三图像，并根据第三图像确定第三个校准标记m 3的位置(x m3，y m3)与预设位置(x c，y c)之间的第三子偏差Δy m3＝y m3-y c。根据Δy m1、Δy m2和Δy m3确定拾取面相较于水平面的偏移量。
这里，驱动单元还可驱动移动单元逆时针旋转，保证对准器旋转至能够获取包括第二个校准标记的第二图像以及能够获取包括第三个校准标记的第三图像即可，本领域技术人员可以根据实际的键合需求选择，本公开在此不作进一步地限制。
第一角度和第二角度可根据校准标记m 1、m 2和m 3的设置方式而确定。在实际的应用中，键合头的形状可为圆柱形，拾取面和校准平面的形状可为圆形，第一角度可基于校准标记m 1、校准标记m 2和校准平面圆心所构成的圆心角确定，第二角度可基于校准标记校准标记m 2、校准标记m 3和校准平面圆心所构成的圆心角确定。
例如，校准标记m 1、m 2和m 3可以等间距的方式设置于侧面301c上，则第一角度和第二角度为120°。校准标记m 1和m 3的连线经过校准平面的圆心，校准标记m 2和校准平面圆心的连线垂直于校准标记m 1和m 3的连线，则第一角度和第二角度为90°。
本公开实施例中，通过设置第一对准组件，可依次获取三个校准标记的第一图像、第二图像和第三图像，并根据获取的第一图像、第二图像和第三图像， 可依次确定第一子偏差、第二子偏差和第三子偏差，有利于精确确定拾取面相较于水平面的偏移量。
在一些实施例中，第一对准组件304包括：至少三个对准器3041，每个对准器3041用于获取一个校准标记与预设位置之间的子偏差；
第一对准组件304，具体用于根据至少三个子偏差确定偏移量。
示例性地，参照图3a所示，第一个校准标记m 1和第三个校准标记m 3沿x方向并列设置，第一对准组件304包括沿x方向并列设置的两个对准器3041，位于键合头301右侧的对准器3041用于获取第一个校准标记m 1与预设位置之间的第一子偏差，位于键合头301左侧的对准器3041用于获取第三个校准标记m 3与预设位置之间的第三子偏差。第一对准组件304还包括位于键合头301朝向y轴正方向一侧的对准器(图中未示出)，该对准器用于获取第二个校准标记m 2与预设位置之间的第二子偏差。第一对准组件304根据第一子偏差、第二子偏差和第三子偏差，可确定拾取面相较于水平面的偏移量。
需要指出的是，在本示例中，示出了键合头包括三个校准标记、第一对准组件包括三个对准器的情形。在其它示例中，键合头可包括至少四个校准标记，第一对准组件可包括至少四个对准器。键合头中校准标记的数量或位置的设置本公开在此不作限制，只需保证设置校准标记的平面平行于拾取面即可。
例如，在如图3a所示的基础上，键合头还包括沿y方向并列设置的两个校准标记，第一对准组件还包括沿y方向并列设置的两个对准器，通过沿y方向并列设置的两个对准器，可分别确定沿y方向并列设置的两个校准标记的第二子偏差和第四子偏差，第一对准组件304根据第一子偏差、第二子偏差、第三子偏差和第四子偏差，可确定拾取面相较于水平面的偏移量。
相较于设置第一对准组件依次获取三个校准标记的第一图像、第二图像和第三图像，本公开实施例中，通过设置三个对准器，每个对准器可独立地获取每个校准标记的图像，无需驱动键合头旋转，有利于提高三个子偏差的精确度。
并且，本公开实施例中三个对准器可同时获取三个校准标记的图像，可缩 短确定偏移量的时间，进而缩短管芯到晶圆键合前水平度检测和水平调节的时间，提高键合效率。
在一些实施例中，参照图4a和图4b所示，键合头301包括：垂直于拾取面301b的侧面301c、从侧面301c贯穿键合头301的至少两条校准光通路；其中，至少两条校准光通路所在的平面平行于拾取面301b；
第一对准组件304，具体用于根据至少两条校准光通路与水平面之间的子偏移，确定偏移量。
示例性地，图4a示出了从侧面301c贯穿键合头301的一条校准光通路3111，图4b示出了沿y方向并列设置的两条校准光通路3111和3112，可以理解的，两条校准光通路3111和3112在xoz平面的投影重合。虽然在本示例中仅示出了两条校准光通路3111和3112，但在其它示例中，从侧面301c贯穿键合头301的校准光通路的数量不限于两条，还可以是三条甚至更多条，本公开在此不作限制，只需保证校准光通路与第一光通路302不重叠即可。
两条校准光通路3111和3112所构成的平面记为校准平面，校准平面平行于拾取面301b，可以理解的是，在键合头301位于第一位置时，可通过两条校准光通路3111和3112确定校准平面相较于水平面的偏移量，进而确定拾取面301b相较于水平面的偏移量。
在一些实施例中，参照图4a所示，第一对准组件304包括：一个发射单元3042和一个接收单元3043；
其中，在键合头301位于第一位置时，发射单元3042与第一条校准光通路的一端对准，接收单元3043与第一条校准光通路的另一端的至少部分区域对准；发射单元3042用于向第一条校准光通路的一端发射校准光信号；接收单元3043，用于根据接收的经第一条校准光通路传输的校准光信号，确定第一子偏移；
键合组件，还用于驱动键合头301旋转，以使第二条校准光通路的一端对准发射单元3042；其中，接收单元3043与第二条校准光通路的另一端的至少 部分区域对准；
发射单元3042，还用于向第二条校准光通路的一端发射校准光信号；接收单元3043，还用于根据接收的经第二条校准光通路传输的校准光信号，确定第二子偏移；
第一对准组件304，具体用于根据第一子偏移和第二子偏移，确定偏移量。
图4b示出了沿y方向并列设置的第一条校准光通路3111和第二条校准光通路3112，发射单元3042和接收单元3043分别位于第一条校准光通路3111沿x方向相对设置的两侧。例如，发射单元3042位于第一条校准光通路3111的右侧或左侧，接收单元3043位于第一条校准光通路3111的左侧或右侧。
在键合头301位于第一位置时，发射单元3042与第一条校准光通路3111的右端3111a对准，当发射单元3042发出的校准光信号为平行光束时，校准光信号完全进入右端3111a，即发射单元3042发出的校准光信号的强度与进入右端3111a的校准光信号的强度基本相同，这里，基本相同可以是两处光信号的强度完全相同，或，两处光信号的强度相差较小、可忽略不计。
接收单元3043与第一条校准光通路3111的左端3111b的至少部分区域对准，经第一条校准光通路3111传输的校准光信号的一部分被接收单元3043所接收，即发射单元3042发出的校准光信号的强度大于接收单元3043接收的校准光信号的强度，根据两处光信号的强度的差值，可计算第一条校准光通路3111相较于水平面的偏移量。
需要指出的是，设置发射单元3042和接收单元3043的平面平行于水平面，当拾取面301b相较于水平面倾斜时，与拾取面301b平行的校准平面也相较于水平面倾斜，即第一条校准光通路3111也相较于水平面倾斜。因此，当发射单元3042右端3111a完全对准后，接收单元3043与左端3111b的部分区域对准。部分区域对准不包括完全对准。可以理解的是，当接收单元3043与左端3111b的部分区域对准时，接收单元3043仅能接收一部分校准光信号，接收单元3043接收的校准光信号的强度小于发射单元3042发出的校准光信号的强度。
当拾取面301b平行于水平面时，与拾取面301b平行的校准平面也平行于水平面，即第一条校准光通路3111也平行于水平面。因此，当发射单元3042右端3111a完全对准后，接收单元3043与左端3111b也完全对准，接收单元3043接收的校准光信号的强度与发射单元3042发出的校准光信号的强度基本相同。
发射单元3042发射的校准光信号经由第一条校准光通路3111传输至接收单元3043，接收单元3043可根据发射的校准光信号的强度I 1和接收的校准光信号的强度I 1’，确定第一子偏移。
示例性地，结合图4a和图4b所示，驱动单元可以经过校准平面的圆心的轴线为中心线，驱动键合头301旋转，直至第二条校准光通路3112的左端3112b对准发射单元3042，接收单元3043与第二条校准光通路3112的右端3112a的至少部分区域对准。
类似地，当拾取面301b相较于水平面倾斜时，第二条校准光通路3112也相较于水平面倾斜。因此，当左端3112b完全对准发射单元3042后，接收单元3043仅与右端3112a的部分区域对准。
当拾取面301b平行于水平面时，第二条校准光通路3112也平行于水平面。因此，当左端3112b完全对准发射单元3042后，接收单元3043与右端3112a也完全对准。
发射单元3042发射的校准光信号经由第二条校准光通路3112传输至接收单元3043，接收单元3043可根据发射的校准光信号的强度I 2和接收的校准光信号的强度I 2’，确定第二子偏移，第一对准组件304根据第一子偏移和第二子偏移，可确定拾取面301b相较于水平面的偏移量。
可以理解的是，本公开实施例中，通过设置一个发射单元和一个接收单元，可先通过第一条校准光通路确定第一子偏移，再通过第二条校准光通路确定第二子偏移，根据第一子偏移和第二子偏移可确定设置两条校准光通路的校准平面相较于水平面的偏移量，进而确定与校准平面平行的拾取面相较于水平面的偏移量。
在一些实施例中，第一对准组件304包括：两个发射单元3042和两个接收单元3043；
在键合头301位于第一位置时，第一个发射单元3042与第一条校准光通路的一端对准，第一个接收单元3043与第一条校准光通路的另一端的至少部分区域对准，第二个发射单元3042与第二条校准光通路的一端对准，第二个接收单元3043与第二条校准光通路的另一端的至少部分区域对准；
第一个发射单元3042，用于向第一条校准光通路的一端发射校准光信号；
第一个接收单元3043，用于根据接收的经第一条校准光通路传输的校准光信号，确定第一子偏移；
第二个发射单元3042，用于向第二条校准光通路的一端发射校准光信号；
第二个接收单元3043，用于根据接收的经第二条校准光通路传输的校准光信号，确定第二子偏移；
第一对准组件304，具体用于根据第一子偏移和第二子偏移，确定偏移量。
图4b示出了沿y方向并列设置的两个发射单元3042和沿y方向并列设置的两个接收单元3043。第一个发射单元3042和第一个接收单元3043沿x方向并列设置、且位于第一条校准光通路3111沿x方向相对设置的两侧。例如，第一个发射单元3042位于第一条校准光通路3111的右侧或左侧，第一个接收单元3043位于第一条校准光通路3111的左侧或右侧。
第二个发射单元3042和第二个接收单元3043沿x方向并列设置、且位于第二条校准光通路3112沿x方向相对设置的两侧。例如，第二个发射单元3042位于第二条校准光通路3112的右侧或左侧，第二个接收单元3043位于第二条校准光通路3112的左侧或右侧。
虽然在本示例中，示出了沿y方向并列设置的两条校准光通路，一对发射单元和接收单元分别位于一条校准光通路沿x方向相对设置的两侧。但在其它示例中，两条校准光通路还可沿x方向并列设置，一对发射单元和接收单元分别位于一条校准光通路沿y方向相对设置的两侧，两条校准光通路还可以其它 方式设置，本公开在此不作限制。
第一个发射单元3042发射的校准光信号经由第一条校准光通路3111传输至第一个接收单元3043，第一个接收单元3043可根据发射的校准光信号的强度I 1和接收的校准光信号的强度I 1’，确定第一子偏移。
第二个发射单元3042发射的校准光信号经由第二条校准光通路3112传输至第二个接收单元3043，第二个接收单元3043可根据发射的校准光信号的强度I 2和接收的校准光信号的强度I 2’，确定第二子偏移，第一对准组件304根据第一子偏移和第二子偏移，可确定拾取面301b相较于水平面的偏移量。
可以理解的是，本公开实施例中，通过设置两个发射单元和两个接收单元，一对发射单元和接收单元可确定第一条校准光通路的第一子偏移，另一对发射单元和接收单元可确定第二条校准光通路的第二子偏移，无需驱动键合头旋转，有利于提高第一子偏移和第二子偏移精确度。
并且，本公开实施例中两对发射单元和接收单元可同时确定第一条校准光通的第一子偏移和第二条校准光通路的第二子偏移，可缩短确定偏移量的时间，进而缩短管芯到晶圆键合前水平度检测和水平调节的时间，提高键合效率。
在一些实施例中，结合图3a和图3c所示，第一光通路302的第二端302b设置于侧面301c；
第一光通路302包括：第一反射镜3021，用于反射在第一光通路302的第一端302a和第一光通路302的第二端302b之间传输的光信号；
第一对准组件304包括：第二光通路306以及对准器3041；其中，对准器3041用于发出检测光信号、接收反射光信号、并确定第一偏差值；
其中，在键合头301位于第二位置时，对准器3041位于第二光通路306相对远离侧面301c的一端，第二光通路306相对靠近侧面301c的一端对准第一光通路302的第二端302b。
示例性地，参照图3c所示，在键合头301位于第二位置时，对准器3041发出检测光信号，检测光信号经由第二光通路306和第一光通路302传输至第 一待键合管芯11的定位标记(图3c中“十”字型所示)，并形成反射光信号，反射光信号经由第一光通路302和第二光通路306传输至对准器3041。
示例性地，参照图3c所示，第一反射镜3021位于第一光通路302的第一端302a和第一光通路302的第二端302b之间，用于将第一光通路302的第二端302b接收的检测光信号反射至第一光通路302的第一端302a，并传输至第一待键合管芯11的定位标记，还用于将第一光通路302的第一端302a接收的反射光信号反射至第一光通路302的第二端302b，并传输至对准器3041。
对准器3041包括：集成有光发射器和光探测器的装置。光发射器用于发出检测光信号，光探测器用于接收反射光信号。进一步地，对准器3041还可包括与光探测器连接的图像传感器，图像传感器用于将接收的反射光信号转化为可视化的图像，根据该图像可确定第一偏差值。
本公开实施例中，通过设置第二光通路和对准器，第二光通路与第一光通路对准，且位于对准器与第一光通路之间，有利于检测光信号或反射光信号在对准器与第一待键合管芯之间地传输，可准确地定位第一待键合管芯的当前位置，有利于提高第一偏差值的精确度。
此外，通过将第一对准组件和键合组件分开设置，即第一对准组件设置于键合组件之外，不会额外增加键合组件的重量，可使得键合组件的惯性较小，并且，由于第一对准组件设置于键合组件之外，第一对准组件不会影响键合组件的移动，可保证键合组件在竖直方向上高频率、高精度地运动，提高键合精度的同时，还可提高键合效率。
在一些实施例中，参照图3a和图3c所示，在键合头301位于第二位置时，第二光通路306平行于拾取面301b。可以理解的是，在本公开实施例中，第二光通路可沿水平方向设置，检测光信号和反射光信号在第二光通路中可沿水平方向传输。
在一些实施例中，结合图5a和图5b所示，第二光通路306’包括：第一子路3061、第二子路3062以及第二反射镜3063，第一子路3061垂直于第二子路 3062，第二反射镜3063位于第一子路3061与第二子路3062相交处，第二反射镜3063用于反射在第一子路3061和第二子路3062之间传输的光信号；
对准器3041’，位于第一子路3061相对远离第二子路3062的端部；
其中，在键合头301位于第二位置时，第一子路3061垂直于拾取面301b，第二子路3062平行于拾取面301b，第二子路3062相对靠近侧面301c的一端对准第一光通路302的第二端302b。
示例性地，参照图5b所示，在键合头301位于第二位置时，对准器3041’发出检测光信号，检测光信号经由第一子路3061、第二子路3062和第一光通路302传输至第一待键合管芯11的定位标记(图5b中“十”字型所示)，并形成反射光信号，反射光信号经由第一光通路302、第二子路3062和第一子路3061传输至对准器3041’。
示例性地，参照图5b所示，第二反射镜3063位于第一子路3061和第二子路3062之间，用于将第一子路3061接收的检测光信号反射至第二子路3062，并传输至第一光通路302，还用于将第二子路3062接收的反射光信号反射至第一子路3061，并传输至对准器3041’。
这里，对准器3041’可与上述实施例中的对准器3041相同，本公开在此不再赘述。
可以理解的是，本公开实施例中，第二光通路可设置为“L”型，包括竖直方向的第一子路和水平方向的第二子路，在第一光通路的第二端位于键合头的侧面时，第二子路与第一光通路的第二端对准，以保证检测光信号和反射光信号在第一光通路和第二光通路中的传输，准确定位第一待键合管芯的当前位置。
此外，通过第二光通路设置为“L”型，增加了第一对准组件中光通路设置方式的多样性，在实际应用中，本领域技术人员可根据实际的键合制程需求，合理地设置第一对准组件中的光通路，本公开在此不作限制。
优选地，第二光通路沿水平方向设置(即如图3a和3b所示)，相较于第二光通路设置为“L”型(即如图4a和4b所示)，第二光通路沿水平方向设置， 可以简化光路设计，减小检测光信号或反射光信号在第二光通路中的传输损耗，提高第一待键合管芯的定位精度，进而提高第一偏差值的精度。
在一些实施例中，参照图6所示，顶面301a包括第一区域和第二区域；
键合组件还包括：固定单元307和移动单元308；固定单元307固定连接第一区域和移动单元308；移动单元308用于相对晶圆承载台303移动键合头301；
第一光通路302’，沿垂直于拾取面301b的方向贯穿键合头301；其中，第一光通路302’的第二端302b’位于第二区域；
第一对准组件304”，在水平方向的投影位于第二区域内，包括：第二光通路306”和对准器3041”；其中，对准器3041”用于发出检测光信号、接收反射光信号、并确定第一偏差值；
其中，第二光通路306”，垂直于拾取面301b且与第一光通路302’对准；对准器3041”，位于第二光通路306”相对远离键合头301的一端。
示例性地，参照图6所示，键合头301包括沿z方向相对设置的拾取面301b，和顶面301a，固定单元307位于键合头301和移动单元308之间，固定单元307的一端与键合头301的顶面301a固定连接，固定单元307的另一端与移动单元308固定连接。
这里，第一区域表示的是键合头301的顶面301a与固定单元307接触的区域，第二区域表示的是键合头301的顶面301a未与固定单元307接触的区域。可以理解的是，固定单元307在水平方向的投影位于第一区域内。
示例性地，参照图6所示，第一光通路302’贯穿键合头301，包括第一端302a’和第二端302b’。可以理解的是，第一端302a’所在的水平面与拾取面301b所在的水平面平齐，第二端302b’所在的水平面与顶面301a所在的水平面平齐、且位于第二区域内。
示例性地，参照图6所示，第一对准组件304”位于第一光通路302’上方，第二光通路306”沿平行于z轴方向设置，并与第一光通路302’对准，在本示例 中，第一对准组件304”可与移动单元308固定连接。
在一些实施例中，第一对准组件304”可设置于键合头301顶面301a，保证第二光通路与第一光通路对准即可。
在一些实施例中，第一对准组件304”可通过外部支架固定而无需与移动单元308固定连接，保证第二光通路与第一光通路对准即可，此时，用于连接移动单元308和键合头301的固定单元307可以是可伸缩装置，以避免第一对准组件304”和外部支架阻挡键合组件上、下移动。
可以理解的是，第一对准组件304”通过外部支架固定的方式，不会额外增加键合组件的重量，可使得键合组件的惯性较小，可保证键合组件在竖直方向上高频率、高精度地运动，提高键合精度的同时，还可提高键合效率。
相较于将第一光通路的第二端设置于键合头的侧面，本公开实施例中，通过设置贯穿键合头的第一光通路，第一光通路的第二端位于键合头的顶面，如此，检测光信号或反射光信号在第一光通路中可沿竖直方向传输，无需在第一光通路中设置第一反射镜，简化了键合组件的构造，有利于减小键合组件的制作难度，同时减少了键合组件的制作成本。
此外，由于第一光通路中无需额外增设第一反射镜，减少了检测光信号或反射光信号在第一反射镜位置处反射而造成的损耗，有利于精确定位第一待键合管芯的当前位置，进而提高第一偏差值的精确度。
在一些实施例中，键合组件包括：两个第一光通路，不同第一光通路的第一端的设置位置不同，不同第一光通路的第二端在拾取面的设置位置不同；
第一对准组件，包括一个对准器；其中，在键合头位于第二位置时，对准器与一个第一光通路的第二端对准；
键合组件还包括：驱动单元，用于驱动两个第一光通路相对对准器运动，以使对准器对准另一个第一光通路的第二端。
示例性地，在第一待键合管芯11包括沿x方向并列设置的两组定位标记时，每组定位标记包括一个定位标记或沿y方向并列设置的至少两个定位标记，参 照图3a所示，键合组件可包括沿x方向并列设置的两个第一光通路302，在键合头301位于第二位置时，两个第一光通路302的第一端302a分别显露第一待键合管芯11中的两组定位标记。
示例性地，结合图3a和图3c所示，第一对准组件可包括一个对准器3041，位于键合头301的一侧，在键合头301位于第二位置时，对准器3041与第一个第一光通路302对准，以通过第一个第一光通路302，定位第一个第一光通路302显露的第一组定位标记的位置。
示例性地，驱动单元可与移动单元308连接，以驱动移动单元308相对对准器3041顺时针或逆时针转动，在移动单元308转动时，固定单元307、键合头301以及第一待键合管芯11同步转动，如此，可将第二个第一光通路302转动至与对准器3041对准的位置，使得对准器3041与第二个第一光通路302对准，通过第二个第一光通路302，定位第二个第一光通路302显露的第二组定位标记的位置。
例如，两个第一光通路302沿x方向并列设置，两个第一光通路302的第二端302b位于同一水平面，在对准器3041与第一个第一光通路302对准后，驱动单元可驱动移动单元308相对对准器3041顺时针或逆时针转动180°，以使得第二个第一光通路302与对准器3041对准。此时，第一个第一光通路302转动至第二个第一光通路302的原位置处。
需要强调的是，在移动单元转动的过程中，第一待键合管芯11也同步转动，因此，第一组定位标记的当前位置在转动后会发生改变。应当理解的是，用于定位第一目标位置以及用于定位第一组定位标记的当前位置而建立的坐标系在转动后也会发生改变。
例如，基于转动前建立的坐标系，定位第一组定位标记的当前位置为(x 3，y 3)，基于顺时针或逆时针转动180°后建立的坐标系，定位第一组定位标记的当前位置为(-x 3，-y 3)，第一目标位置为(-x 0，-y 0)。
在实际应用中，键合头的形状可为圆柱形，拾取面的形状可为圆形，可根 据移动单元308转动的角度、转动的方向以及拾取面的半径确定转动后第一组定位标记的位置以及转动后的第一目标位置。
示例性地，基于顺时针或逆时针转动180°后建立的坐标系，第一组定位标记的当前位置为(-x 3，-y 3)，第二组定位标记的当前位置为(x 4，y 4)，根据第一组定位标记的当前位置和第二组定位标记的当前位置确定第一待键合管芯11的当前位置(x 1’，y 1’)，其中，x 1’＝(x 4-x 3)/2，y 1’＝(y 4-y 3)/2。第一偏差值为(Δx T’，Δy T’)，其中，Δx T’＝x 1’-(-x 0)，Δy T’＝y 1’-(-y 0)。
可以理解的是，可基于第一个第一光通路和第二个第一光通路设置的位置，确定移动单元相对对准器转动的角度。这里，两个第一光通路沿x方向并列设置仅为示意，用以向本领域技术人员传达本公开，然而本公开并不限于此，在其它示例中，两个第一光通路还可以其它的方式设置，例如，可根据第一待键合管芯中多组定位标记的设置方式而设置。
本公开实施例中，通过设置一个对准器，可实现一个对准器与两个第一光通路分别对准，进而通过与对准器对准的第一光通路定位第一待键合管芯上的定位标记。
在一些实施例中，键合组件包括：两个第一光通路，不同第一光通路的第一端的设置位置不同，不同第一光通路的第二端在拾取面的设置位置不同；
第一对准组件包括两个对准器和两条第二光通路；其中，不同的对准器位于不同的第二光通路相对远离键合头的端部；
其中，在键合头位于第二位置时，不同的第二光通路与不同的第一光通路对准。
示例性地，参照图3a所示，键合组件可包括沿x方向并列设置的两个第一光通路302，第一对准组件304包括沿x方向并列设置的两个对准器3041和两条第二光通路306，每个对准器3041各自通过一条第二光通路306与一个第一光通路302对准。
需要指出的是，在本示例中，示出了键合组件包括两个第一光通路、第一 对准组件包括两个对准器和两条第二光通路的情形。在其它示例中，键合组件可包括至少三个第一光通路，第一对准组件可包括至少三个对准器和三条第二光通路。
例如，在如图3a所示的基础上，键合组件还包括沿y方向并列设置的两个第一光通路，第一对准组件还包括沿y方向并列设置的两个对准器和沿y方向并列设置的两条第二光通路，如此，键合组件中设置有四个第一光通路，第一对准组件中设置有四个对准器和四条第二光通路，每个对准器各自通过一条第二光通路与一个第一光通路对准。
这里，键合组件中第一光通路的数量或位置的设置以及第一键合组件中对准器的数量或位置的设置，可根据第一待键合管芯中定位标记的数量或位置而设定，本公开不作进一步限制。
相较于设置一个对准器对准至少两个第一光通路，本公开实施例中，通过设置两个对准器，可实现一个对准器通过一个第二光通路对准一个第一光通路，每个对准器可独立地定位每个定位标记，有利于提高定位标记定位的精确度。
在一些实施例中，结合图3a和图7所示，键合系统300还包括：
基座309，位于键合头301的侧面301c；
第一支架310-1，平行于拾取面301b设置，用于固定连接第一个对准器3041-1和基座309；
第二支架310-2，平行于拾取面301b设置，用于固定连接第二个对准器3041-2和基座309。
图7中示出了键合系统300在xoy平面的局部俯视图，键合组件可包括沿y方向并列设置的第一个第一光通路302-1和第二个第一光通路302-2，第一对准组件可包括沿y方向并列设置的第一个对准器3041-1和第二个对准器3041-2，在键合头301位于第二位置时，第一个对准器3041-1与第一个第一光通路302-1对准，第二个对准器3041-2与第二个第一光通路302-2对准。
第一个对准器3041-1通过第一支架310-1与基座309固定连接，第二个对 准器3041-2通过第二支架310-2与基座309固定连接，可以理解的是，第一个对准器3041-1和第二个对准器3041-2在xoz平面的投影重合，即第一个对准器3041-1和第二个对准器3041-2位于键合头301的同一侧。
在一些实施例中，结合图3a和图8所示，键合系统300还包括：
两个基座309，分别设置于键合头301的两侧；
第一支架310-1，平行于拾取面301b设置，用于固定连接第一个对准器3041-1和第一个基座309；
第二支架310-2，平行于拾取面301b设置，用于固定连接第二个对准器3041-2和第二个基座309。
图8中示出了键合系统300局部结构示意图，键合组件可包括沿x方向并列设置的两个第一光通路，第一对准组件可包括沿x方向并列设置的第一个对准器3041-1和第二个对准器3041-2，在键合头301位于第二位置时，第一个对准器3041-1与键合头301左侧的第一光通路对准，第二个对准器3041-2与键合头301右侧的第一光通路对准。
第一个对准器3041-1通过第一支架310-1与位于键合头左侧的基座309固定连接，第二个对准器3041-2通过第二支架310-2与位于键合头右侧的基座309固定连接，第一个对准器3041-1和第二个对准器3041-2位于键合头301的两侧。
图9是根据本公开实施例示出的一种键合方法的流程示意图，图10a至图10d是根据本公开实施示出的一种键合方法的过程示意图。该键合方法应用于第一待键合管芯11与位于晶圆的第二待键合管芯21的键合，所述方法包括以下步骤：
S100：在拾取有第一待键合管芯的键合头位于第一位置时，确定键合头拾取第一待键合管芯的拾取面相较于水平面的偏移量；
S200：根据偏移量，驱动键合头移动至平行于水平面的第二位置；
S300：在键合头位于第二位置时，确定拾取的第一待键合管芯的当前位置 与第一目标位置之间的第一偏差值，确定第二待键合管芯的当前位置与第二目标位置之间的第二偏差值；
S400：根据第一偏差值和第二偏差值，移动晶圆或第一待键合管芯，以对准第一待键合管芯和第二待键合管芯；
S500：在第一待键合管芯和第二待键合管芯对准后，键合第一待键合管芯和第二待键合管芯。
在步骤S100中，参照图10a所示，在拾取有第一待键合管芯11的键合头301位于第一位置时，可利用第一对准组件304’检测拾取面相较于水平面(例如，xoy平面)的偏移量，该偏移量可通过拾取面与水平面之间的夹角来表示，还可通过拾取面上多个点相较于同一水平面的高度来表示。
在步骤S200中，参照图10b所示，在确定拾取面相较于水平面的偏移量后，键合组件可驱动键合头301旋转，直至拾取面平行于水平面，并将键合头301调整至第二位置。这里，第二位置表示的是第一光通路302能够接收到第一对准组件304’发射的检测光信号的位置。
在步骤S300中，参照图10b所示，可利用第一对准组件304’确定第一待键合管芯11的当前位置(x 1，y 1)与第一目标位置(x 0，y 0)之间的第一偏差值(Δx T，Δy T)，其中，Δx T＝x 1-x 0，Δy T＝y 1-y 0。
在步骤S300中，参照图10c所示，在确定第一偏差值后，将晶圆20传输至承载台上，可利用第二对准组件305确定晶圆20上第二待键合管芯21的当前位置(x 2，y 2)与第二目标位置(x 0’，y 0’)之间的第二偏差值(Δx B，Δy B)，其中，Δx B＝x 2-x 0’，Δy T＝y 2-y 0’。
需要指出的是，第一目标位置和第二目标位置在水平方向上的位置相同、竖直方向上的位置不同。即x 0与x 0’相同，y 0与y 0’相同，第一目标位置和第二目标位置z方向的坐标不同，不同的附图标记仅是为了便于区分第一目标位置和第二目标位置，而不用于限制本公开。
在步骤S400中，根据第一偏差值和第二偏差值确定第一待键合管芯11和 第二待键合管芯21之间的位移偏差为(Δx T-Δx B，Δy T-Δy B)，可保持键合组件固定(即保持第一待键合管芯11固定)，晶圆20根据位移偏差水平移动(Δx T-Δx B，Δy T-Δy B)，以使得第二待键合管芯21对准第一待键合管芯11，即第一待键合管芯11和第二待键合管芯21水平方向上的位置相同。
这里，可根据Δx T-Δx B的正负确定晶圆20移动的方向，例如，在Δx T-Δx B为负值时，晶圆20沿x轴负方向移动，移动的距离为(Δx T-Δx B)的绝对值。在Δx T-Δx B为正值时，晶圆20沿x轴正方向移动，移动的距离为(Δx T-Δx B)。
类似地，可根据Δy T-Δy B的正负确定晶圆20移动的方向，例如，在Δy T-Δy B为负值时，晶圆20沿y轴负方向移动，移动的距离为(Δy T-Δy B)的绝对值。在Δy T-Δy B为正值时，晶圆20沿y轴正方向移动，移动的距离为(Δy T-Δy B)。
在步骤S400中，根据第一偏差值和第二偏差值确定第一待键合管芯11和第二待键合管芯21之间的位移偏差为(Δx T-Δx B，Δy T-Δy B)，还可保持晶圆20固定(即保持第二待键合管芯21固定)，第一待键合管芯11根据位移偏差水平移动(Δx T-Δx B，Δy T-Δy B)，以使得第一待键合管芯11对准第二待键合管芯21。
这里，可根据Δx T-Δx B的正负确定第一待键合管芯11移动的方向，例如，在Δx T-Δx B为负值时，第一待键合管芯11沿x轴正方向移动，移动的距离为(Δx T-Δx B)的绝对值。在Δx T-Δx B为正值时，第一待键合管芯11沿x轴负方向移动，移动的距离为(Δx T-Δx B)。
类似地，可根据Δy T-Δy B的正负确定第一待键合管芯11移动的方向，例如，在Δy T-Δy B为负值时，第一待键合管芯11沿y轴正方向移动，移动的距离为(Δy T-Δy B)的绝对值。在Δy T-Δy B为正值时，第一待键合管芯11沿y轴负方向移动，移动的距离为(Δy T-Δy B)。
在步骤S400中，确定第一偏差值和第二偏差值之后，可移动晶圆或第一待键合管芯11，以将第一待键合管芯和第二待键合管芯对准。优选地，移动晶圆，以使第二待键合管芯对准第一待键合管芯。
在步骤S500中，参照图10d所示，在第一待键合管芯11和第二待键合管芯21对准后，键合组件垂直向下移动，以将第一待键合管芯11和第二待键合管芯21键合。
需要强调的是，第一待键合管芯11是独立于晶圆存在的管芯。在半导体器件的制作过程中，通常在晶圆上形成多个管芯，这里，第一待键合管芯11指的是承载有多个管芯的晶圆切割后形成的独立管芯。可以理解的是，第一待键合管芯的尺寸远小于晶圆的尺寸。
本公开实施例中，在键合头位于第一位置时，通过确定键合头的拾取面相较于水平面的偏移量，根据该偏移量可驱动键合头移动至平行于水平面的第二位置。即在管芯到晶圆键合前，可将拾取的管芯调整至平行于水平面，有利于减小管芯键合时因局部受力不均而破损的概率，提高管芯到晶圆键合的良率。
在将管芯调整至平行于水平面的第二位置后，通过确定第一待键合管芯的当前位置与第一目标位置之间的第一偏差值，以及确定晶圆上第二待键合管芯的当前位置与第二目标位置的第二偏差值，根据第一偏差值和第二偏差值来调整晶圆或第一待键合管芯，可使得第一待键合管芯与第二待键合管芯精确对准，有利于提高管芯到晶圆的键合精度。
在一些实施例中，键合头包括：垂直于拾取面的侧面、平行于拾取面的顶面、以及至少三个校准标记；其中，至少三个校准标记位于侧面或顶面，至少三个校准标记所在的平面平行于拾取面；
确定键合头的拾取面相较于水平面的偏移量，包括：
获取包括第一个校准标记的第一图像；
根据第一图像，确定第一个校准标记的位置与预设位置之间的第一子偏差；
在获取第一图像后，驱动键合头旋转第一角度；
在键合头旋转第一角度后，获取包括第二个校准标记的第二图像；
根据第二图像，确定第二个校准标记的位置与预设位置之间的第二子偏差；
在获取第二图像后，驱动键合头旋转第二角度；
在键合头旋转第二角度后，获取包括第三个校准标记的第三图像；
根据第三图像，确定第三个校准标记的位置与预设位置之间的第三子偏差；
根据第一子偏差、第二子偏差和第三子偏差，确定偏移量。
示例性地，结合图5a和图10a所示，校准标记m 1、m 2和m 3位于侧面301c上，校准标记m 1、m 2、m 3所构成的平面记为校准平面，校准平面平行于拾取面301b。
在键合头301位于第一位置时，第一对准组件304’获取包括第一个校准标记m 1的第一图像，并根据第一图像确定第一个校准标记m 1的位置(x m1，y m1)与预设位置(x c，y c)之间的第一子偏差Δy m1＝y m1-y c。
驱动单元驱动移动单元308顺时针旋转第一角度，在移动单元308转动时，固定单元307、键合头301以及第一待键合管芯11同步转动，第一对准组件304’获取包括第二个校准标记m 2的第二图像，并根据第二图像确定第二个校准标记m 2的位置(x m2，y m2)与预设位置(x c，y c)之间的第二子偏差Δy m2＝y m2-y c。
然后，驱动单元驱动移动单元308顺时针旋转第二角度，第一对准组件304’获取包括第三个校准标记m 3的第三图像，并根据第三图像确定第三个校准标记m 3的位置(x m3，y m3)与预设位置(x c，y c)之间的第三子偏差Δy m3＝y m3-y c。根据Δy m1、Δy m2和Δy m3确定拾取面相较于水平面的偏移量。
这里，预设位置表示的是第一对准组件304’用于确定校准标记m 1、m 2和m 3位置的取景框的中心位置，记为(x c，y c)。可以理解的是，当三个校准标记m 1、m 2和m 3与预设位置在y轴方向上的高度偏差相同，即y m1-y c、y m2-y c以及y m3-y c的数值大小相同且正负相同时，可确定校准平面平行于水平面，进而确定与校准平面平行的拾取面也平行于水平面。
驱动单元还可驱动移动单元逆时针旋转，保证对准器旋转至能够获取包括第二个校准标记的第二图像以及能够获取包括第三个校准标记的第三图像即可，本领域技术人员可以根据实际的键合需求选择，本公开在此不作进一步地限制。
第一角度和第二角度可根据校准标记m 1、m 2和m 3的设置方式而确定。在实际的应用中，键合头的形状可为圆柱形，拾取面和校准平面的形状可为圆形，第一角度可基于校准标记m 1、校准标记m 2和校准平面圆心所构成的圆心角确定，第二角度可基于校准标记校准标记m 2、校准标记m 3和校准平面圆心所构成的圆心角确定。
例如，校准标记m 1、m 2和m 3可以等间距的方式设置于侧面301c上，则第一角度和第二角度为120°。校准标记m 1和m 3的连线经过校准平面的圆心，校准标记m 2和校准平面圆心的连线垂直于校准标记m 1和m 3的连线，则第一角度和第二角度为90°。
虽然在本示例中仅示出了校准标记m 1、m 2和m 3位于侧面301c，但在其它示例中，校准标记m 1、m 2和m 3还可位于顶面301a。例如，位于顶面301a的边界线上。位于侧面301c或位于顶面301a的校准标记的数量也不限于三个，还可以是四个、五个甚至更多，本公开在此不作限制。
本公开实施例中，通过在键合头的侧面或顶面设置至少三个校准标记，并且三个校准标记所在的平面平行于拾取面，通过确定三个校准标记和预设位置之间的子偏差，可确定校准平面相较于水平面的平移量，进而确定与校准平面平行的拾取面相较于水平面的偏移量。
在一些实施例中，键合头包括：垂直于拾取面的侧面、平行于拾取面的顶面、以及至少三个校准标记；其中，至少三个校准标记位于侧面或顶面，至少三个校准标记所在的平面平行于拾取面；
确定键合头的拾取面相较于水平面的偏移量，包括：
同时获取至少三个校准标记与预设位置之间的子偏差；
根据至少三个子偏差确定偏移量。
示例性地，结合图5a和图10a所示，第一对准组件304’可包括三个对准器3041’和三条第二光通路306’。具体地：
第一个对准器3041’和第一条第二光通路306’位于第一个校准标记m 1的右 侧，用于获取第一个校准标记m 1与预设位置之间的子偏差。
第二个对准器3041’和第二条第二光通路306’位于第二个校准标记m 2朝向y轴正方向的一侧，用于获取第二个校准标记m 2与预设位置之间的子偏差。
第三个对准器3041’和第三条第三光通路306’位于第三个校准标记m 3的左侧，用于获取第三个校准标记m 3与预设位置之间的子偏差。
第一对准组件304’根据三个子偏差，确定拾取面301b相较于水平面的偏移量。
需要指出的是，在本示例中，示出了键合头包括三个校准标记、第一对准组件包括三个对准器的情形。在其它示例中，键合头可包括至少四个校准标记，第一对准组件可包括至少四个对准器。键合头中校准标记的数量或位置的设置本公开在此不作限制，只需保证设置校准标记的平面平行于拾取面即可。
相较于依次获取三个校准标记的第一图像、第二图像和第三图像，本公开实施例中，无需驱动键合头旋转，即可获取至少三个校准标记与预设位置之间的子偏差，有利于提高三个子偏差的精确度。
并且，本公开实施例中可同时获取三个子偏差，可缩短确定偏移量的时间，进而缩短管芯到晶圆键合前水平度检测和水平调节的时间，提高键合效率。
在一些实施例中，键合头包括：垂直于拾取面的侧面、从侧面贯穿键合头的至少两条校准光通路；其中，至少两条校准光通路所在的平面平行于拾取面；
确定键合头的拾取面相较于水平面的偏移量，包括：
向第一条校准光通路的一端发射校准光信号；
根据接收的经第一条校准光通路传输的校准光信号，确定第一子偏移；
在确定第一子偏移后，驱动键合头旋转；
在键合头旋转之后，向第二条校准光通路的一端发射校准光信号；
根据接收的经第二条校准光通路传输的校准光信号，确定第二子偏移；
根据第一子偏移和第二子偏移，确定偏移量。
示例性地，结合图4a和图4b所示，两条校准光通路3111和3112从侧面 301c贯穿键合头301，两条校准光通路3111和3112所构成的平面记为校准平面，校准平面平行于拾取面301b。
在键合头301位于第一位置时，发射单元3042向第一条校准光通路3111发射校准光信号，接收单元3043根据发射的校准光信号的强度I 1和接收的校准光信号的强度I 1’，确定第一子偏移。
驱动单元驱动键合头301旋转，直至第二条校准光通路3112对准发射单元3042，发射单元3042向第二条校准光通路3112发射校准光信号，接收单元3043根据发射的校准光信号的强度I 2和接收的校准光信号的强度I 2’，确定第二子偏移。第一对准组件根据第一子偏移和第二子偏移，可确定拾取面301b相较于水平面的偏移量。
虽然在本示例中仅示出了两条校准光通路3111和3112，但在其它示例中，从侧面301c贯穿键合头301的校准光通路的数量不限于两条，还可以是三条甚至更多条，本公开在此不作限制，只需保证校准光通路与第一光通路302不重叠即可。
在一些实施例中，键合头包括：垂直于拾取面的侧面、从侧面贯穿键合头的至少两条校准光通路；其中，至少两条校准光通路所在的平面平行于拾取面；
确定键合头的拾取面相较于水平面的偏移量，包括：
向第一条校准光通路的一端发射校准光信号，并向第二条校准光通路的一端发射校准光信号；
根据接收的经第一条校准光通路传输的校准光信号，确定第一子偏移；
根据接收的经第二条校准光通路传输的校准光信号，确定第二子偏移；
根据第一子偏移和第二子偏移，确定偏移量。
示例性地，结合图4a和图4b所示，第一对准组件304可包括两个发射单元3042和两个接收单元3043。具体地：
第一个发射单元3042和第一个接收单元3043沿x方向并列设置、且位于第一条校准光通路3111沿x方向相对设置的两侧，用于确定第一子偏移
第二个发射单元3042和第二个接收单元3043沿x方向并列设置、且位于第二条校准光通路3112沿x方向相对设置的两侧，用于确定第二子偏移。
第一对准组件304根据第一子偏移和第二子偏移，可确定拾取面301b相较于水平面的偏移量。
图11是根据本公开实施例示出的一种管芯到晶圆结构的示意图，所述管芯到晶圆结构应用上述任一实施例中的键合系统300以及应用上述任一实施例中的键合方法制作而成，包括：
多个第一管芯11；
晶圆20，包括：多个第二管芯21；其中，每个第一管芯11与每个第二管芯21键合连接。
显然，上述实施例仅仅是为清楚地说明所作的举例，而并非对实施方式的限定。对于所属领域的普通技术人员来说，在上述说明的基础上还可以做出其它不同形式的变化或变动。这里无需也无法对所有的实施方式予以穷举。而由此所引伸出的显而易见的变化或变动仍处于本公开创造的保护范围之中。

Claims (12)

  1. 一种键合系统，包括：
    键合组件，包括：键合头以及贯穿所述键合头的第一光通路；其中，所述第一光通路的第一端位于所述键合头拾取第一待键合管芯的拾取面；
    晶圆承载台，用于承载晶圆；
    第一对准组件，用于在所述键合头位于第一位置时，确定所述拾取面相较于水平面的偏移量；
    所述键合组件，还用于根据所述偏移量，驱动所述键合头移动至平行于水平面的第二位置；
    所述第一对准组件，还用于在所述键合头位于第二位置时，通过所述第一光通路的第二端向所述第一端发射检测光信号；其中，拾取的所述第一待键合管芯覆盖所述第一端并反射所述检测光信号；
    所述第一对准组件，还用于接收所述检测光信号经所述第一待键合管芯反射的反射光信号，并根据接收的所述反射光信号确定所述第一待键合管芯的当前位置与第一目标位置之间的第一偏差值；
    第二对准组件，位于所述晶圆承载台相对远离所述键合组件的一侧，用于确定所述晶圆上第二待键合管芯的当前位置与第二目标位置的第二偏差值；
    所述晶圆承载台，还用于根据所述第一偏差值和所述第二偏差值，驱动承载的所述晶圆相对所述拾取面移动，以使所述第二待键合管芯对准所述第一待键合管芯；或者，所述键合组件，还用于根据所述第一偏差值和所述第二偏差值，驱动拾取所述第一待键合管芯的所述拾取面相对所述晶圆承载台移动，以使所述第一待键合管芯对准所述第二待键合管芯；
    所述键合组件，还用于键合所述第一待键合管芯和所述第二待键合管芯。
  2. 根据权利要求1所述的键合系统，其中，
    所述键合头包括：垂直于所述拾取面的侧面、平行于所述拾取面的顶面、 以及至少三个校准标记；其中，所述至少三个校准标记位于所述侧面或所述顶面，所述至少三个校准标记所在的平面平行于所述拾取面；
    所述第一对准组件，具体用于根据所述至少三个校准标记和预设位置之间的子偏差，确定所述偏移量。
  3. 根据权利要求2所述的键合系统，其中，
    所述第一对准组件，具体用于在所述键合头位于所述第一位置时，获取包括第一个所述校准标记的第一图像，并根据所述第一图像确定第一个所述校准标记的位置与预设位置之间的第一子偏差；
    所述键合组件，还用于驱动键合头旋转第一角度，以使所述第一对准组件能够获取到包括第二个所述校准标记的第二图像；
    所述第一对准组件，具体还用于根据获取的所述第二图像，确定第二个所述校准标记的位置与所述预设位置之间的第二子偏差；
    所述键合组件，还用于驱动所述键合头旋转第二角度，以使所述第一对准组件能够获取到包括第三个所述校准标记的第三图像；
    所述第一对准组件，具体还用于根据获取的所述第三图像，确定第三个所述校准标记的位置与所述预设位置之间的第三子偏差，并根据所述第一子偏差、所述第二子偏差和所述第三子偏差，确定所述偏移量。
  4. 根据权利要求2所述的键合系统，其中，
    所述第一对准组件包括：至少三个对准器，每个所述对准器用于获取一个所述校准标记与预设位置之间的子偏差；
    所述第一对准组件，具体用于根据至少三个所述子偏差确定所述偏移量。
  5. 根据权利要求1所述的键合系统，其中，
    所述键合头包括：垂直于所述拾取面的侧面、从所述侧面贯穿所述键合头的至少两条校准光通路；其中，所述至少两条校准光通路所在的平面平行于所述拾取面；
    所述第一对准组件，具体用于根据所述至少两条校准光通路与水平面之间 的子偏移，确定所述偏移量。
  6. 根据权利要求5所述的键合系统，其中，
    所述第一对准组件包括：一个发射单元和一个接收单元；
    其中，在所述键合头位于所述第一位置时，所述发射单元与第一条所述校准光通路的一端对准，所述接收单元与第一条所述校准光通路的另一端的至少部分区域对准；所述发射单元用于向第一条所述校准光通路的一端发射校准光信号；所述接收单元，用于根据接收的经第一条所述校准光通路传输的校准光信号，确定第一子偏移；
    所述键合组件，还用于驱动所述键合头旋转，以使第二条所述校准光通路的一端对准所述发射单元；其中，所述接收单元与第二条所述校准光通路的另一端的至少部分区域对准；
    所述发射单元，还用于向所述第二条所述校准光通路的一端发射校准光信号；所述接收单元，还用于根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
    所述第一对准组件，具体用于根据所述第一子偏移和所述第二子偏移，确定所述偏移量。
  7. 根据权利要求5所述的键合系统，其中，
    所述第一对准组件包括：两个发射单元和两个接收单元；
    在所述键合头位于所述第一位置时，第一个所述发射单元与第一条所述校准光通路的一端对准，第一个所述接收单元与第一条所述校准光通路的另一端的至少部分区域对准，第二个所述发射单元与第二条所述校准光通路的一端对准，第二个所述接收单元与第二条所述校准光通路的另一端的至少部分区域对准；
    第一个所述发射单元，用于向第一条所述校准光通路的一端发射校准光信号；
    第一个所述接收单元，用于根据接收的经第一条所述校准光通路传输的校 准光信号，确定第一子偏移；
    第二个所述发射单元，用于向第二条所述校准光通路的一端发射校准光信号；
    第二个所述接收单元，用于根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
    所述第一对准组件，具体用于根据所述第一子偏移和所述第二子偏移，确定所述偏移量。
  8. 一种键合方法，所述键合方法应用于第一待键合管芯与位于晶圆的第二待键合管芯的键合，所述键合方法包括：
    在拾取有第一待键合管芯的键合头位于第一位置时，确定所述键合头拾取所述第一待键合管芯的拾取面相较于水平面的偏移量；
    根据所述偏移量，驱动所述键合头移动至平行于所述水平面的第二位置；
    在所述键合头位于所述第二位置时，确定拾取的所述第一待键合管芯的当前位置与第一目标位置之间的第一偏差值，确定所述第二待键合管芯的当前位置与第二目标位置之间的第二偏差值；
    根据所述第一偏差值和所述第二偏差值，移动所述晶圆或所述第一待键合管芯，以对准所述第一待键合管芯和所述第二待键合管芯；
    在所述第一待键合管芯和所述第二待键合管芯对准后，键合所述第一待键合管芯和所述第二待键合管芯。
  9. 根据权利要求8所述的键合方法，其中，所述键合头包括：垂直于所述拾取面的侧面、平行于所述拾取面的顶面、以及至少三个校准标记；其中，所述至少三个校准标记位于所述侧面或所述顶面，所述至少三个校准标记所在的平面平行于所述拾取面；
    所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
    获取包括第一个校准标记的第一图像；
    根据所述第一图像，确定所述第一个校准标记的位置与预设位置之间的第 一子偏差；
    在获取所述第一图像后，驱动所述键合头旋转第一角度；
    在所述键合头旋转第一角度后，获取包括第二个校准标记的第二图像；
    根据所述第二图像，确定所述第二个校准标记的位置与所述预设位置之间的第二子偏差；
    在获取所述第二图像后，驱动所述键合头旋转第二角度；
    在所述键合头旋转第二角度后，获取包括第三个校准标记的第三图像；
    根据所述第三图像，确定所述第三个校准标记的位置与所述预设位置之间的第三子偏差；
    根据所述第一子偏差、所述第二子偏差和所述第三子偏差，确定所述偏移量。
  10. 根据权利要求8所述的键合方法，其中，所述键合头包括：垂直于所述拾取面的侧面、平行于所述拾取面的顶面、以及至少三个校准标记；其中，所述至少三个校准标记位于所述侧面或所述顶面，所述至少三个校准标记所在的平面平行于所述拾取面；
    所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
    同时获取所述至少三个校准标记与预设位置之间的子偏差；
    根据至少三个所述子偏差确定所述偏移量。
  11. 根据权利要求8所述的键合方法，其中，所述键合头包括：垂直于所述拾取面的侧面、从所述侧面贯穿所述键合头的至少两条校准光通路；其中，所述至少两条校准光通路所在的平面平行于所述拾取面；
    所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
    向第一条所述校准光通路的一端发射校准光信号；
    根据接收的经第一条所述校准光通路传输的校准光信号，确定第一子偏移；
    在确定所述第一子偏移后，驱动所述键合头旋转；
    在所述键合头旋转之后，向第二条所述校准光通路的一端发射校准光信号；
    根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
    根据所述第一子偏移和所述第二子偏移，确定所述偏移量。
  12. 根据权利要求8所述的键合方法，其中，所述键合头包括：垂直于所述拾取面的侧面、从所述侧面贯穿所述键合头的至少两条校准光通路；其中，所述至少两条校准光通路所在的平面平行于所述拾取面；
    所述确定所述键合头的拾取面相较于水平面的偏移量，包括：
    向第一条所述校准光通路的一端发射校准光信号，并向第二条所述校准光通路的一端发射校准光信号；
    根据接收的经第一条所述校准光通路传输的校准光信号，确定第一子偏移；
    根据接收的经第二条所述校准光通路传输的校准光信号，确定第二子偏移；
    根据所述第一子偏移和所述第二子偏移，确定所述偏移量。
PCT/CN2021/142073 2021-12-24 2021-12-28 键合系统和键合方法 Ceased WO2023115614A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP21946243.9A EP4231339B1 (en) 2021-12-24 2021-12-28 Bonding system and bonding method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111594630.2 2021-12-24
CN202111594630.2A CN114005777B (zh) 2021-12-24 2021-12-24 键合装置和键合方法

Publications (1)

Publication Number Publication Date
WO2023115614A1 true WO2023115614A1 (zh) 2023-06-29

Family

ID=79931943

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/142073 Ceased WO2023115614A1 (zh) 2021-12-24 2021-12-28 键合系统和键合方法

Country Status (3)

Country Link
EP (1) EP4231339B1 (zh)
CN (1) CN114005777B (zh)
WO (1) WO2023115614A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2025010956A1 (zh) * 2023-07-07 2025-01-16 武汉新芯集成电路制造有限公司 一种键合装置及键合方法
WO2025010957A1 (zh) * 2023-07-07 2025-01-16 武汉新芯集成电路制造有限公司 一种键合装置及键合方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114999984B (zh) * 2022-06-23 2024-11-12 武汉新芯集成电路股份有限公司 键合装置及键合方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090141275A1 (en) * 2007-12-03 2009-06-04 Samsung Electronics Co., Ltd Alignment inspection method and alignment inspection apparatus
CN106373914A (zh) * 2016-11-10 2017-02-01 北京中电科电子装备有限公司 一种芯片键合装置
CN107665847A (zh) * 2016-07-29 2018-02-06 上海微电子装备（集团）股份有限公司 一种键合对准设备和方法
US20180102340A1 (en) * 2016-10-12 2018-04-12 Samsung Electronics Co., Ltd. Apparatus for correcting a parallelism between a bonding head and a stage, and a chip bonder including the same
CN109285803A (zh) * 2018-09-20 2019-01-29 武汉新芯集成电路制造有限公司 晶圆键合方法及装置

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000012623A (ja) * 1998-06-25 2000-01-14 Sharp Corp ボンディング装置及びボンディング方法
JP3732082B2 (ja) * 2000-09-25 2006-01-05 株式会社新川 ボンディング装置およびボンディング方法
US7727800B2 (en) * 2005-12-12 2010-06-01 Asm Assembly Automation Ltd. High precision die bonding apparatus
KR101132141B1 (ko) * 2006-03-03 2012-03-29 삼성테크윈 주식회사 다이 픽업위치 보정방법
JP5045184B2 (ja) * 2007-03-28 2012-10-10 カシオ計算機株式会社 熱圧着装置
JP2013235907A (ja) * 2012-05-08 2013-11-21 Shibuya Kogyo Co Ltd ボンディング装置
CN103367208B (zh) * 2013-07-02 2015-10-28 华中科技大学 一种用于高密度芯片的倒装键合平台
CN109451763B (zh) * 2018-05-16 2019-11-08 长江存储科技有限责任公司 用于晶圆键合对准补偿的方法和系统
DE102018115144A1 (de) * 2018-06-23 2019-12-24 Besi Switzerland Ag Stellantrieb für einen Bondkopf
CN110660723B (zh) * 2018-06-29 2022-05-10 上海微电子装备（集团）股份有限公司 一种机械手、键合腔体、晶圆键合系统及键合方法
CN111312611B (zh) * 2018-12-11 2022-09-30 武汉新芯集成电路制造有限公司 一种晶圆键合方法、控制单元和系统
CN111446191A (zh) * 2020-05-15 2020-07-24 长江存储科技有限责任公司 晶圆键合设备及检测方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090141275A1 (en) * 2007-12-03 2009-06-04 Samsung Electronics Co., Ltd Alignment inspection method and alignment inspection apparatus
CN107665847A (zh) * 2016-07-29 2018-02-06 上海微电子装备（集团）股份有限公司 一种键合对准设备和方法
US20180102340A1 (en) * 2016-10-12 2018-04-12 Samsung Electronics Co., Ltd. Apparatus for correcting a parallelism between a bonding head and a stage, and a chip bonder including the same
CN106373914A (zh) * 2016-11-10 2017-02-01 北京中电科电子装备有限公司 一种芯片键合装置
CN109285803A (zh) * 2018-09-20 2019-01-29 武汉新芯集成电路制造有限公司 晶圆键合方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4231339A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2025010956A1 (zh) * 2023-07-07 2025-01-16 武汉新芯集成电路制造有限公司 一种键合装置及键合方法
WO2025010957A1 (zh) * 2023-07-07 2025-01-16 武汉新芯集成电路制造有限公司 一种键合装置及键合方法

Also Published As

Publication number Publication date
CN114005777A (zh) 2022-02-01
EP4231339B1 (en) 2025-07-16
EP4231339A1 (en) 2023-08-23
CN114005777B (zh) 2022-03-29
EP4231339A8 (en) 2024-03-27
EP4231339A4 (en) 2024-03-06

Similar Documents

Publication Publication Date Title
WO2023115614A1 (zh) 键合系统和键合方法
CN113990790B (zh) 键合系统和键合方法
WO2023115610A1 (zh) 键合系统和键合补偿方法
KR102364002B1 (ko) 고정밀 본드 헤드 위치 결정 방법 및 장치
KR102132094B1 (ko) 전자 부품 실장 장치 및 전자 부품 실장 방법
KR101051498B1 (ko) 전자 부품 위치 맞춤 방법 및 그 장치
CN115799146B (zh) 一种光学对位系统
CN116092957B (zh) 一种键合系统和键合方法
CN107134419B (zh) 倒装芯片键合装置及其键合方法
US6566669B2 (en) Semiconductor parts and semiconductor mounting apparatus
JP7585001B2 (ja) ウェハ接合装置
JP2020020990A (ja) 接合装置および接合方法
CN111326442A (zh) 基板接合装置
US12610780B2 (en) Bonding systems using a predetermined tilt profile
WO2023115615A1 (zh) 键合系统和键合方法
JP5813330B2 (ja) コレット位置検出方法及び装置
CN117420338A (zh) 一种探针吸取方法及装置
CN117420337A (zh) 一种探针校正方法及装置
CN117976599B (zh) 一种高精度晶圆键合设备
KR102804682B1 (ko) 하이브리드본딩 정렬방법
CN223624930U (zh) 一种键合对准组件及键合装置
CN117092775A (zh) 用于制造相机模块的透镜模块相对于图像传感器的被动对准
CN121646302A (zh) 一种可视化集成力控反馈的工作头、芯片贴装及测试方法
TW202230577A (zh) 基板接合裝置及基板接合方法
JP2005236270A (ja) 発光素子の実装方法

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021946243

Country of ref document: EP

Effective date: 20221226

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21946243

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWG Wipo information: grant in national office

Ref document number: 2021946243

Country of ref document: EP