KR20200105753A - Semiconductor manufacturing device and manufacturing method of semiconductor device - Google Patents

Abstract

The present invention is to provide a semiconductor manufacturing device capable of controlling a push-up unit in an optimum sequence from the viewpoint of low stress on a die or high-speed pickup. A semiconductor manufacturing device comprises: a push-up unit including a plurality of blocks in contact with a dicing tape and configured to push a die from under the dicing tape; a collet adsorbing the die; and a control unit configured to perform feedback control on a peeling model reproducing characteristics of the push-up unit so that the output of the peeling model follows target values of the peeling amount of the die from the dicing tape and the bending stress of the entire die, and to use the push-up amount which is control input to the peeling model as the push-up amount of the blocks of the push-up unit.

Description

A semiconductor manufacturing apparatus and a manufacturing method of a semiconductor device TECHNICAL FIELD [SEMICONDUCTOR MANUFACTURING DEVICE AND MANUFACTURING METHOD OF SEMICONDUCTOR DEVICE}

The present disclosure relates to a semiconductor manufacturing apparatus, and is applicable to, for example, a die bonder having a push-up (lift) unit.

In general, in a die bonder that mounts a semiconductor chip called die on the surface of, for example, a wiring board or a lead frame (hereinafter, collectively referred to as a substrate), the die is generally mounted using a suction nozzle such as a collet. An operation (operation) of carrying out bonding on a substrate, applying a pressing force, and heating the bonding material to perform bonding is repeatedly performed.

Among the die bonding processes using a semiconductor manufacturing apparatus such as a die bonder, there is a peeling process of peeling a divided die from a semiconductor wafer (hereinafter, referred to as a wafer). In the peeling step, the die is pushed up by the pushing unit from the back of the dicing tape, peeled one by one from the dicing tape held in the die supply unit, and conveyed onto the substrate using a suction nozzle such as a collet.

For example, according to Japanese Unexamined Patent Application Publication No. 2005-117019 (Patent Document 1), when the die to be peeled is pushed up among a plurality of dies affixed to the dicing tape and peeled from the dicing tape, an adsorption port (吸着A (push-up unit) is peeled off from the dicing tape with low stress from the periphery of the die by pushing up a plurality of stages of blocks in a pyramidal shape by one drive shaft of a pusher.

Japanese Patent Publication No. 2005-117019 Japanese Patent Publication No. 2017-224640

In recent years, with the advent of die stack packages and 3D-NAND (three-dimensional NAND flash), wafers (die) are becoming thinner. When the die becomes thinner, the rigidity of the die becomes very low compared to the adhesive force of the dicing tape. Therefore, for example, in order to pick up a thin die of several tens of µm or less, it is necessary to reduce the stress applied to the die (lower stress).

In the above-described push-up of a plurality of blocks by one drive shaft, since the amount of push-up of each block is limited mechanically, the push-up operation is performed at constant acceleration, constant speed, constant deceleration, and then peeling can sufficiently proceed. It is a linear sequence that waits for a certain amount of time. However, in a linear sequence, when conditions such as the type of dicing tape and die thickness are changed, the amount of the block pushed up is not necessarily optimal. In addition, in the linear sequence, there is a possibility that the sequence may not be optimal from the viewpoint of low stress on the die or high speed pickup.

An object of the present disclosure is to provide a semiconductor manufacturing apparatus capable of controlling a pushing unit in an optimal sequence from the viewpoint of low stress on a die or high speed pickup.

Other problems and novel features will become apparent from the description of the present specification and the accompanying drawings.

A brief overview of representative ones of the present disclosure is as follows.

In other words, the semiconductor manufacturing apparatus has a plurality of blocks in contact with the dicing tape, and reproduces the characteristics of a push-up unit that pushes up a die from under the dicing tape, a collet that adsorbs the die, and the push-up unit. With respect to the peeling model to be subjected to, feedback control to follow the output of the peeling model to the target value of the peeling amount of the die from the dicing tape and the bending stress of the entire die, and pushing the control input to the peeling model And a control unit configured to make the lifting amount as the lifting amount of the block of the pushing unit.

According to the semiconductor manufacturing apparatus, it is possible to control the die in an optimal sequence from the viewpoint of low stress or high-speed pickup.

1 is a diagram illustrating a configuration of a main part of a pushing unit.
Fig. 2 is a diagram explaining a pushing sequence of a pushing unit.
3 is a block diagram illustrating a feedback control system.
4 is a diagram illustrating a die peeling model.
5 is a diagram illustrating an adhesive material model.
6 is a flowchart for explaining calculation of a die peeling model.
Fig. 7 is a conceptual diagram viewed from above of a die bonder according to an embodiment.
FIG. 8 is a diagram for explaining the operation of the pickup head and the bonding head when viewed from the direction of arrow A in FIG. 7.
9 is a diagram showing an external perspective view of the die supply unit of FIG. 7.
Fig. 10 is a schematic cross-sectional view showing an essential part of the die supply section of Fig. 7.
Fig. 11 is an external perspective view of the pushing unit of Fig. 9;
12 is a top view of a part of the first unit in FIG. 11.
13 is a top view of a part of the second unit in FIG. 11.
14 is a top view of a part of the third unit in FIG. 11.
Fig. 15 is a longitudinal sectional view of the push-up unit of Fig. 11;
Fig. 16 is a longitudinal sectional view of the pushing unit of Fig. 11;
Fig. 17 is a diagram showing a configuration of a collet portion among a pushing unit and a pickup head according to the embodiment.
Fig. 18 is a flowchart for explaining a pickup operation of the die bonder of Fig. 7;
19 is a flowchart for explaining a method of manufacturing a semiconductor device according to the embodiment.
Fig. 20 is a diagram for describing a nonlinear sequence and a numerical example of a linear sequence.
Fig. 21 is a block diagram illustrating a feed forward control system for each axis.
22 is a flowchart illustrating an example of pushing up control of each block.
23 is a flowchart for explaining another example of pushing up control of each block.
24 is a block diagram illustrating a feedback control system.
Fig. 25 is a flowchart for explaining pushing up control of each block.

Hereinafter, embodiments and examples will be described with reference to the drawings. However, in the following description, the same constituent elements are denoted by the same reference numerals, and repeated explanation may be omitted. In addition, in order to make the explanation more clear, the drawings may be schematically displayed with respect to the width, thickness, shape, etc. of each part compared to the actual mode, but are only examples and do not limit the interpretation of the present invention.

First, the pushing unit will be described using Figs. Fig. 1 is a diagram showing a configuration of a main part of a pushing unit, Fig. 1(a) is a cross-sectional view taken along line A-A in Fig. 1(b), and Fig. 1(b) is a top view. Fig. 2 is a diagram showing a pushing sequence of a pushing unit, Fig. 2(a) is a diagram showing a linear sequence, and Fig. 2(b) is a diagram showing a non-linear pushing sequence of a first example 2(c) is a diagram showing a nonlinear push-up sequence of a second example.

As shown in FIG. 1, the pushing up unit (TUU) is a dome (DM) for adsorbing the dicing tape (DCT) located around the outside of the pickup object die (D), and the opening of the dome (DM). It is provided with a push-up block portion (BLK) located. The push-up block part BLK is composed of, for example, three blocks BLK1, BLK2, and BLK3. As shown in Fig. 1(b), a rectangular frame-shaped block BLK2 is located inside the rectangular frame-shaped block BLK1 in plan view, and a rectangular block BLK2 is located inside the rectangular frame-shaped block BLK2 ( BLK3) is located. 1A, the block BLK1 is pushed up above the dome DM, and the blocks BLK2 and BLK3 are pushed up above the block BLK1.

As described above, the conventional pushing up operation of the block is performed in a linear pushing up sequence in which after constant acceleration, constant speed operation, and constant deceleration, wait for a certain time until peeling proceeds sufficiently. As shown in Fig.2(a), in the linear push-up sequence, the push-up amount of the block of the push-up unit (TUU) increases in proportion to time, and after reaching the maximum value (h _gmax ) of the push-up amount Is a sequence in which the pushing operation stops and waits for the die D to be peeled off from the dicing tape DCT. Here, the time at which the maximum value of the pushing amount (h _gmax ) is reached is referred to as t _gmax . Further, after the right end of the time axis shown in Fig. 2A, the inner block is pushed up.

In the inventor's examination, the die exhibits a nonlinear behavior in which the peeling process is slowly peeled from the dicing tape at the beginning of peeling, and the peeling progress is accelerated as the peeling proceeds. The linear sequence may not be an optimal sequence from the viewpoint of low stress and high speed pickup for the die.

Therefore, in the embodiment, the push-up is performed in a nonlinear push-up sequence in which the speed during push-up is variable in accordance with the peeling of the die. In order to realize this, it is carried out by feedback control of the pushing unit. The push-up speed and the push-up amount of the block of the push-up unit can be set programmable.

Next, feedback control will be described using FIG. 3. Fig. 3(a) is a block diagram showing a general feedback control system, and Fig. 3(b) is a block diagram showing a control system of the pushing unit according to the embodiment.

As shown in Fig. 3A, a general feedback control (PID control) system is composed of a control object, a sensor, and a PID controller. The PID controller (C(s)) is the deviation signal e() between the output (control amount: y(s)) measured by the sensor from the control target (G(s)) and the target value (r(s)) to be followed. s)), an input to the control object (operation amount: u(s)) is determined by combining three operations of a proportional operation (P), an integral operation (I), and a differential operation (D).

The control amount of the pushing unit as a control target is the amount of peeling of the die D from the dicing tape DCT and the bending stress of the entire die D. These are difficult to measure with a sensor.

Thus, in the embodiment, as shown in Fig. 3B, a feedback control system is configured for the peeling model. In the embodiment, with respect to the peeling model (Gm(s)), the amount of peeling and the bending stress of the entire die are given as the target input (r(s)), and the amount of pushing is used as the control input (u(s)). It is given to each of the peeling model (Gm(s)) and the actual push-up target (G(s)). By configuring the PID controller compensator (C(s)) to follow the output (y(s)) to the target input (r(s)), the control input (u()) according to the behavior of the separation model (Gm(s)) Adjust the push-up amount, which is s)).

When the release model (Gm(s)) can sufficiently reproduce the characteristics of the actual push-up target (G(s)) by giving the amount of push-up generated above to the actual push-up target (G(s)) , As for the actual push-up target (G(s)), the peeling result designed by the feedback control system using the peeling model (Gm(s)) is obtained.

The peeling model has the following characteristics obtained from the inventor's examination.

(1) Peeling is started with a certain amount of pushing up.

(2) From the start of peeling to the progress of peeling, the peeling progress is accelerated (according to the peeling progress, the peeling progress is accelerated because the curvature increases as the die approaches the root, and the moment that becomes the restoring force w _r increases) .

(3) A dicing tape with high peel strength takes a lot of push-up, and a viscous dicing tape takes a lot of push-up time (the amount of push-up at which peeling starts by changing the parameters of the adhesive material of the dicing tape, The rate at which peeling proceeds changes).

Next, the peeling model will be described with reference to FIGS. 4 to 6. Fig. 4 is a diagram showing a die peeling model of the broken line portion of Fig. 1(a), Fig. 4(a) before pushing up, Fig. 4(b) when pushing up, Fig. 4(c) Is a diagram showing when the adhesive material breakage occurs. 5 is a diagram showing an adhesive material model. 6 is a diagram showing a calculation flow of a die peeling model.

Think of one side of the die D being pushed up as a cantilever beam fixed to the end surface of the block BLK2 being pushed up. At this time, assume that the die D is applied with an equally distributed load as shown in Fig. 4A by the adhesive material of the dicing tape DCT.

When the block BLK2 is pushed up by the push-up amount h, it is assumed that the die D is deformed as shown in Fig. 4(b) until the peeling starts.

here,

w: Adhesion force of the adhesive applied to the unit area [N]

L: Width [mm] of block BLK1

h: Amount of push-up of block (BLK2) [mm]

b: Length of one side of block BLK2 [mm]

E: Young's modulus of die [N/㎟]

δx: amount of deflection at the x position from the die end

It is called.

In addition, the adhesive material fixing the die D is modeled. 5: is a figure which shows the adhesive model (three-element model).

It is transformed over time by the damping constant (C ₂ ) by receiving the reaction force (w _r ) against the bending of the die (D). When the reaction force (w _r ) is large with respect to the spring constants (K ₁ , K ₂ ), when the displacements (ε ₁ , ε ₂ ) reach a predetermined amount of deformation, it is assumed that the adhesive material is broken. When the displacement (ε ₁ ) reaches the limit value, interfacial failure, and when the displacement (ε ₂ ) reaches the limit value, cohesive failure occurs.

When breakage occurs in the adhesive material, the die (D) has a shape as shown in Fig. 4(c), so that w is recalculated as the uniformly distributed load is applied only to the unbroken portion (non-destructive portion). .

The calculation of the die peeling model will be described using FIG. 6.

Calculating by the position of the minimum (non-destructive portion distal end) that are not destroy the x _min, hwimyang of non-destructive portion leading to _{δ xmin, E, h, L} , w, δx min using x _min, and δx _min, E, The load (w) to be updated is calculated using h, L, and x _min (step S1).

Further, the reaction force (w _r ) generated at x _min of the tip of the non-destructive portion is calculated as a moment between x _min -x _min+1 (step S2).

The deformation amount (ε ₁ ) of the adhesive material at x _min at the tip of the non-destructive part is calculated using w _r and K ₁ , and the deformation amount (ε ₁ ) is calculated using w _r , K ₁ , and C ₂ (step S3). . It is determined whether the amount of deformation has exceeded a predetermined value (whether or not the adhesive has been destroyed) (step S4).

If x _min destroys the adhesive material is encountered, thereby increasing the x width Δx _min by a unit of calculation (step S5).

Repeat until no abnormality or destruction occurs.

Then, the time is increased by the unit width Δt, and the amount of deformation of the adhesive material in x _min is calculated using w _r , ε ₁ , ε ₂ , K ₁ , and C ₂ (step S6).

It was reduced by the die (D) bending (δx _min) the amount of the adhesive material strain (ε _1, ε ₂₎ of the in the x _min, to recalculate the distribution of load (w), thus updating of the entire warp.

The calculation result (a numerical example) of the nonlinear sequence and the constant-speed rising and pushing sequence (linear sequence) will be described using FIG. 20. Fig. 20(a) is a diagram showing the push-up amount and the peeling amount in a linear sequence, and Fig. 20(b) is a diagram showing the die bending stress in the linear sequence. Fig. 20(c) is a diagram showing the push-up amount and the peeling amount of a nonlinear sequence, and Fig. 20(d) is a diagram showing the die bending stress of the nonlinear sequence.

In a linear sequence,

The push-up amount is 300㎛, the push-up speed is 1mm/sec.

It should be.

In a nonlinear sequence, for the separation model,

L=0.5mm, b=10mm, E=185000N/mm2, die thickness=20㎛, K1=20N/mm, K2=1.5N/mm, C2=0.01N/(mm/s), allowable adhesive elongation= Under the condition of 0.6mm,

C(s)=K _p +K _i /s+K _d s

K _p =[1, 0], K _i =[0, 0], K _d =[0, 0]

r(s)=[0.5, 0] (peel amount=0.5mm, die bending stress=0)

By writing. However, it is assumed that there is no error between Gm(s) and G(s).

As shown in Figs. 20A and 20B, in a linear sequence, the peeling time is 0.65 seconds and the maximum stress is 10 MPa. As shown in Figs. 20C and 20D, in the nonlinear sequence, the peeling time is 0.58 seconds and the maximum stress is 8.5 MPa, and both the peeling time and the maximum stress are more effective than the linear sequence.

Next, feedforward control for each axis using the peeling model generated by the above-described feedback control system will be described with reference to FIGS. 21 to 23. FIG. 21 is a diagram for explaining a feedforward control system for each axis, FIG. 21(a) is a block diagram for block BLK1, and FIG. 21(b) is a block diagram for block BLK2. , (C) of FIG. 21 is a block diagram of the block BLK3. 22 is a flowchart showing an example of pushing up control of each block. 23 is a flowchart showing another example of pushing up control of each block.

Each configuration and operation of FIGS. 21A, 21B, and 21C are basically the same as those of FIG. 3B, but the control input u(s) is the controller blocks BLK1, BLK2, It is input to the motor driver for BLK3) and controls the push-up operation of the blocks BLK1, BLK2, BLK3, which are the actual push-up targets (G(s)).

As shown in Fig. 21A, similarly to Fig. 3B, a peeling model Gm(s) for block BLK1 is generated. This is the generation of the peeling model of the block BLK1 in step S11 in FIGS. 22 and 23. The generated peeling model Gm(s) for block BLK1 becomes a target input r(s) for block BLK2.

As shown in Fig. 21B, similarly to Fig. 3B, a peeling model Gm(s) for block BLK2 is generated. This is the generation of the peeling model of the block BLK2 in step S21 of FIGS. 22 and 23. The generated peeling model Gm(s) for block BLK2 becomes a target input r(s) for block BLK3.

As shown in Fig. 21C, similarly to Fig. 3B, a peeling model Gm(s) for the block BLK3 is generated. This is the generation of the peeling model of the block BLK3 in step S31 in FIGS. 22 and 23.

In the flow of Fig. 22, pushing up control of each block is performed while generating a separation model of each block. That is, after generating the peeling model of the block BLK1 in step S11, in parallel with the generation of the peeling model of the block BLK2 in step S21, the pushing operation of the block BLK1 is performed (Step S12). Next, after generating the peeling model of the block BLK2 in step S21, in parallel with the generation of the peeling model of the block BLK3 in step S31, a pushing operation of the block BLK2 is performed (Step S22). Finally, after the peeling model of the block BLK3 is generated in step S31, the pushing operation of the block BLK3 is performed (step S32).

In the flow of FIG. 23, after generating the peeling model of all blocks, the pushing up control of each block is performed. That is, a peeling model of the block BLK1 is generated (step S11), a peeling model of the block BLK2 is generated (step S21), and a peeling model of the block BLK3 is generated (step S31). Thereafter, a pushing operation of the block BLK1 is performed (step S12), a pushing operation of the block BLK2 is performed (step S22), and a pushing operation of the block BLK3 is performed (step S32).

Next, the feedback control for each axis using the peeling model generated by the above-described feedback control system will be described with reference to FIGS. 24 and 25. 24 is a block diagram illustrating a feedback control system. 25 is a flowchart for explaining the pushing control of each block.

In the control system of FIG. 24, the separation model generated by the feedback control system is the same as that of FIG. 3(b), but the control input (u(s)) is input to the motor drivers for the blocks BLK1, BLK2, and BLK3, which are controllers. , Controls the push-up operation of the blocks BLK1, BLK2, and BLK3, which are the actual push-up targets G(s), and feeds back the peeling state checked by the sensor to the control input u(s).

As shown in FIG. 25, similarly to FIG. 22, the peeling model Gm(s) for block BLK1 is generated (step S11). The control input u(s) is input to the motor driver for the block BLK1 which is a controller, and controls the pushing operation of the block BLK1 which is the actual pushing target G(s) (step S12). The peeling state is checked by a sensor (step S13), and it is determined whether or not the adhesive material at the tip of the non-destructive part ( _xmin ) has been destroyed as a result of the simulation (step S14). In the case of "No", the process returns to step S11, and a peeling model Gm(s) for block BLK1 is generated again.

In the case of "Yes", the generated peeling model Gm(s) for the block BLK1 becomes the target input r(s) for the block BLK2, and the peeling model for the block BLK2 (Gm( s)) is generated (step S21). The control input u(s) is input to the motor driver for the block BLK2 which is a controller, and controls the pushing operation of the block BLK2 which is the actual pushing target G(s) (step S22). The peeling state is confirmed by the sensor (step S23), and it is determined whether or not the adhesive material at the tip of the non-destructive part ( _xmin ) has been destroyed as a result of the simulation (step S24). In the case of "No", the process returns to step S21, and a peeling model Gm(s) for block BLK2 is generated again.

In the case of "Yes", the generated peeling model Gm(s) for the block BLK2 becomes the target input r(s) for the block BLK3, and the peeling model for the block BLK3 (Gm( s)) is generated (step S31). The control input u(s) is input to the motor driver for the block BLK3 which is a controller, and controls the pushing operation of the block BLK3 which is the actual pushing target G(s) (step S32). The peeling state is checked by the sensor (step S33), and it is determined whether or not the adhesive material at the tip of the non-destructive part ( _xmin ) has been destroyed as a result of the simulation (step S34). In the case of "No", the process returns to step S13, and a peeling model Gm(s) for block BLK3 is generated again. In the case of "Yes", the push-up control ends.

The control of each axis may be performed in combination with feed forward control and feedback control.

Next, an example of a nonlinear push-up sequence will be described with reference to Figs. 2B and 2C.

As shown in Fig.2(b), in the nonlinear push-up sequence of the first example, the push-up amount of the block of the push-up unit is not necessarily proportional to time, but increases, and reaches the maximum value (h _gmax ) of the push-up amount. After doing, the pushing operation is a sequence waiting for the die to be peeled off from the dicing tape while decreasing without stopping. In the drawing, it has an inflection point that is convex upward. The t _gmax of the nonlinear push-up sequence of the first example is faster than t _gmax of the linear push-up sequence of Fig. 2A, that is, the push-up speed is high. Further, after the right end of the time axis shown in Fig. 2B, the inner block is pushed up.

As shown in Fig.2(c), in the nonlinear push-up sequence of the second example, the push-up amount of the block of the push-up unit is not necessarily proportional to time, but increases, and reaches the maximum value (h _lmax ) of the push-up amount. After doing, the pushing operation does not stop, but decreases once to reach the minimum value h _lmin , then increases, and waits for the die to peel off from the dicing tape. Here, the time to reach the maximum value of the push-up amount (h _lmax ) is called t _lmax , and the time to reach the minimum value of the push-up amount (h _lmin ) is called t _lmin . In the drawing, it has an inflection point convex upward and an inflection point convex downward. Second example of the non-linear boost sequence t _lmax is faster than t _gmax sequence Post also pushed in (a) of a second linear, that is faster push Post speed. Further, after the right end of the time axis shown in Fig. 2C, the inner block is pushed up.

<Example>

7 is a top view schematically showing a die bonder according to an embodiment. 8 is a view for explaining the operation of the pickup head and the bonding head when viewed from the arrow A direction in FIG. 7.

The die bonder 10 provides a die (D) to be mounted on a substrate (S) on which a product area (hereinafter referred to as a package area (P)) consisting of one or a plurality of final packages is printed. The die supply unit 1, the pickup unit 2, the intermediate stage unit 3, the bonding unit 4, the transfer unit 5, the substrate supply unit 6, and the substrate carrying unit 7 Wow, it has a control unit 8 that monitors and controls the operation of each unit. The Y-axis direction is the front-rear direction of the die bonder 10, and the X-axis direction is the left-right direction. The die supply unit 1 is disposed on the front side of the die bonder 10, and the bonding unit 4 is disposed on the inner side.

First, the die supply unit 1 supplies a die D to be mounted on the package area P of the substrate S. The die supply unit 1 has a wafer holding stand 12 for holding the wafer 11 and a pushing unit 13 shown by a dotted line for pushing up the die D from the wafer 11. The die supply unit 1 moves in the XY axis direction by a driving means not shown, and moves the die D to be picked up to the position of the lifting unit 13.

The pickup unit 2 lifts the pickup head 21 for picking up the die D, the Y drive unit 23 of the pickup head for moving the pickup head 21 in the Y-axis direction, and the collet 22, Each drive unit (not shown) for rotating and moving in the X-axis direction is provided. The pickup head 21 has a collet 22 (refer to FIG. 10) for adsorbing and holding the pushed die D at its tip, and picks up the die D from the die supply unit 1, and the intermediate stage It is loaded in (31). The pickup head 21 has respective driving units (not shown) for lifting, rotating, and moving the collet 22 in the X-axis direction.

The intermediate stage portion 3 has an intermediate stage 31 for temporarily loading the die D, and a stage recognition camera 32 for recognizing the die D on the intermediate stage 31.

The bonding unit 4 picks up the die D from the intermediate stage 31 and bonds it onto the package area P of the substrate S to be conveyed, or the package area P of the substrate S already ) Is bonded in the form of stacking on the bonded die. The bonding portion 4 includes a bonding head 41 including a collet 42 (see FIG. 2) that adsorbs and holds the die D at the tip, similarly to the pickup head 21, and a bonding head 41 It has a Y driver 43 for moving in the Y-axis direction, and a substrate recognition camera 44 for capturing a position recognition mark (not shown) of the package area P of the substrate S and recognizing a bonding position. .

With this configuration, the bonding head 41 corrects the pickup position and posture based on the image pickup data of the stage recognition camera 32, picks up the die D from the intermediate stage 31, and picks up the die D, and the substrate recognition camera ( The die D is bonded to the substrate based on the imaging data of 44).

The conveyance part 5 has a substrate conveyance claw 51 which grasps and conveys the board|substrate S, and the conveyance lane 52 through which the board|substrate S moves. The board|substrate S moves by driving a nut (not shown) of the board|substrate transfer claw 51 provided in the transfer lane 52 with a ball screw (not shown) provided along the transfer lane 52.

With this configuration, the substrate S moves from the substrate supply unit 6 along the transfer lane 52 to the bonding position, and after bonding, moves to the substrate unloading portion 7 to move to the substrate unloading portion 7. Transfer the substrate S.

The control unit 8 includes a memory storing a program (software) that monitors and controls the operation of each unit of the die bonder 10, and a central processing unit (CPU) that executes a program stored in the memory.

Next, the configuration of the die supply unit 1 will be described with reference to FIGS. 9 and 10. 9 is a diagram illustrating an external perspective view of the die supply unit of FIG. 7. 10 is a schematic cross-sectional view showing an essential part of the die supply section of FIG. 7.

The die supply unit 1 includes a wafer holding stand 12 that moves in the horizontal direction (XY axis direction) and a push-up unit 13 that moves in the vertical direction. The wafer holder 12 includes an expand ring 15 holding the wafer ring 14 and a dicing tape 16 held by the wafer ring 14 and adhered to a plurality of dies D. It has a support ring 17 which is positioned horizontally. The push-up unit 13 is disposed inside the support ring 17.

The die supply unit 1 lowers the expand ring 15 holding the wafer ring 14 when the die D is pushed up. As a result, the dicing tape 16 held on the wafer ring 14 is stretched, the gap between the die D is enlarged, and the die D from the lower side of the die D by the pushing unit 13 ) Is pushed up, and the pick-up property of the die D is improved. In addition, the adhesive for bonding the die to the substrate becomes a film from a liquid, and a film-like adhesive material called a die attach film (DAF) 18 is affixed between the wafer 11 and the dicing tape 16. Are doing. In the wafer 11 having the die attach film 18, dicing is performed with respect to the wafer 11 and the die attach film 18. Therefore, in the peeling process, the wafer 11 and the die attach film 18 are peeled from the dicing tape 16. In addition, in the following, the existence of the die attach film 18 is ignored, and a peeling process is demonstrated.

Next, the pushing up unit 13 will be described with reference to Figs. 11 to 16. 11 is an external perspective view of the pushing unit according to the embodiment. 12 is a top view of a part of the first unit of FIG. 11. 13 is a top view of a part of the second unit of FIG. 11. 14 is a top view of a part of the third unit in FIG. 11. 15 is a longitudinal cross-sectional view of the pushing unit of FIG. 11. 16 is a longitudinal cross-sectional view of the pushing unit of FIG. 11.

The push-up unit 13 includes a first unit 13a, a second unit 13b to which the first unit 13a is mounted, and a third unit 13c to which the second unit 13b is mounted. do. The second unit 13b and the third unit 13c are a common part regardless of the variety, and the first unit 13a is a replaceable part for each variety.

The first unit 13a includes a block portion 13a1 having blocks A1 to A4, a dome head 13a2 having a plurality of suction holes, a suction hole 13a3, and a suction hole 13a4 for dome suction. And converts the vertical motion of the concentric circular blocks B1 to B4 of the second unit 13b into the vertical motion of the four concentric rectangular blocks A1 to A4. The four blocks A1 to A4 can independently move up and down. The planar shape of the concentric rectangular blocks A1 to A4 is configured to match the shape of the die D. When the die size is large, the number of concentric rectangular blocks is composed of more than four. This is made possible by the plurality of outputs of the third unit and the concentric blocks of the second unit moving up and down independently of each other (not moving up and down). The push-up speed and push-up amount of the four blocks (A1 to A4) can be set programmable.

The second unit 13b has circular tubular blocks B1 to B6 and an outer circumferential portion 13b2, and concentrically performs the vertical motion of the output units C1 to C6 disposed on the circumference of the first unit 13a. It converts to the vertical motion of the six blocks (B1 to B6) of the image. The six blocks (B1 to B6) can be moved up and down independently. Here, since the first unit 13a has only four blocks A1 to A4, the blocks B5 and B6 are not used.

The third unit 13c has a central portion 13c0 and six peripheral portions 13c1 to 13c6. The central portion 13c0 has six output portions C1 to C6 which are arranged at equal intervals on the circumference of the upper surface and move up and down independently. The peripheral portions 13c1 to 13c6 can drive the output units C1 to C6 independently of each other. The peripheral portions 13c1 to 13c6 each have motors M1 to M6, and the central portion 13c0 is provided with plunger mechanisms P1 to P6 for converting the rotation of the motor into vertical movement by a cam or link. The plunger mechanisms P1 to P6 impart vertical movement to the output portions C1 to C6. Further, the motors M2 and M5 and the plunger mechanisms P2 and P5 are not shown. Here, since the first unit 13a has only four blocks A1 to A4, the peripheral portions 13c5 and 13c6 are not used. Therefore, the motors M5 and M6, the plunger mechanisms P5 and P6, and the outputs C5 and C6 are not used.

Next, the relationship between the pushing unit and the collet will be described using FIG. 17. Fig. 17 is a diagram showing a configuration of a collet portion among a pushing unit and a pickup head according to an embodiment.

As shown in Fig. 17, the collet portion 20 includes a collet 22, a collet holder 25 for holding the collet 22, and suction holes for adsorbing the die D ( 22v, 25v). The suction surface of the collet 22 for adsorbing the die is approximately the same size as the die D.

The first unit 13a has a dome head 13a2 around the upper surface. The dome head 13a2 has a plurality of suction holes HL and a cavity part CV, and is sucked from the suction hole 13a3, and the peripheral die Dd of the die D picked up by the collet 22 is formed into a die. It is sucked through the sinking tape 16. In Fig. 17, only one row of suction holes HL is shown around the block portion 13a1, but a plurality of rows are provided to stably hold the die Dd, which is not a pickup target. Through the gaps A1v, A2v, A3v between the blocks of the concentric rectangular blocks A1 to A4 and the cavity in the dome of the first unit 13a, the dome suction suction hole 13a4 is sucked, and the collet 22 The die D picked up is sucked through the dicing tape 16. Suction from the suction hole 13a3 and suction from the suction hole 13a4 can be performed independently.

The push-up unit 13 of this embodiment can be applied to various dies by changing the shape of the blocks of the first unit and the number of blocks. For example, when the number of blocks is 6, the die size is 20 mm square or less. Applicable to die. By increasing the number of output units of the third unit, the number of concentric blocks of the second unit, and the number of concentric rectangular blocks of the first unit, the die size is applicable to a die larger than 20 mm square.

Next, the pick-up operation by the pushing unit 13 according to the above-described configuration will be described with reference to FIG. 18. 18 is a flowchart showing a processing flow of a pickup operation.

Step PS1: The control unit 8 moves the wafer holder 12 so that the die D to be picked up is positioned directly above the push-up unit 13, and the upper surface of the third unit is placed on the back surface of the dicing tape 16. The push-up unit 13 is moved to make this contact. At this time, as shown in Fig. 19, the control unit 8 makes each block A1 to A4 of the block portion 13a1 form the same plane as the surface of the dome head 13a2, and the dome head 13a2 The dicing tape 16 is adsorbed by the adsorption holes HL of and the gaps A1v, A2v, and A3v between blocks.

Step PS2: The control unit 8 lowers the collet portion 20, positions it on the die D to be picked up, and adsorbs the die D by the suction holes 22v and 25v.

Step PS3: The control unit 8 sequentially raises the blocks of the block portion 13a1 from the outside to perform a peeling operation. Here, the control unit 8 performs feed forward control using the peeling model of the embodiment. In other words, the control unit 8 drives the plunger mechanism P4 with the motor M4, raises only the outermost block A4 by several tens of µm to several hundreds µm, then lowers it to stop. The rate of rise and fall is not constant. As a result, a push-up portion from which the dicing tape 16 protrudes is formed around the block A4, and a small space between the dicing tape 16 and the die attach film 18, that is, peeling There is a starting point. By this space, the anchor effect, that is, the stress applied to the die D is greatly reduced, and the subsequent peeling operation can be reliably performed. Subsequently, the control unit 8 drives the plunger mechanism P3 with the motor M3, and makes only the second outer block A3 rise higher than the block A4 to stop. Subsequently, the control unit 8 drives the plunger mechanism P2 with the motor M2, and makes only the third outer block A2 rise higher than the block A3 and stop. Finally, the control unit 8 drives the plunger mechanism P1 with the motor M1, and stops only the innermost block A1 by raising it higher than the block A2.

Step PS4: The control unit 8 raises the collet. In the final state of Step PS3, the contact area between the dicing tape 16 and the die D becomes an area that can be peeled off by the rise of the collet, and the die D is peeled off by the rise of the collet 22. I can.

Step PS5: The control unit 8 makes each block A1 to A4 of the block portion 13a1 form the same plane as the surface of the dome head 13a2, and the suction hole HL of the dome head 13a2, Adsorption of the dicing tape 16 due to the gaps A1v, A2v, and A3v between blocks is stopped. The control unit 8 moves the push-up unit 13 so that the upper surface of the first unit is spaced apart from the rear surface of the dicing tape 16.

The control unit 8 repeats steps PS1 to PS5 to pick up a die of the wafer 11 of good quality.

Next, a method of manufacturing a semiconductor device using a die bonder according to an embodiment will be described with reference to FIG. 19. 19 is a flowchart showing a method of manufacturing the semiconductor device of FIG. 7.

Step BS11: The wafer ring 14 holding the dicing tape 16 to which the die D divided from the wafer 11 is affixed is stored in a wafer cassette (not shown), and in the die bonder 10 Bring in. The control unit 8 supplies the wafer ring 14 to the die supply unit 1 from a wafer cassette filled with the wafer ring 14. Further, the substrate S is prepared and carried into the die bonder 10. The control unit 8 attaches the substrate S to the substrate transfer hook 51 in the substrate supply unit 6.

Step BS12: The control unit 8 peels the die D as described above, and picks up the peeled die D from the wafer 11. In this way, the die D peeled from the dicing tape 16 together with the die attach film 18 is adsorbed and held by the collet 22, and is conveyed to the next step (step BS13). And, when the collet 22 which conveyed the die D to the next process returns to the die supply part 1, the next die D is peeled from the dicing tape 16 according to the above-described procedure, and then the same The die D is peeled one by one from the dicing tape 16 according to the procedure of.

Step BS13: The control unit 8 stacks the picked-up die on the substrate S or on the already bonded die. The control unit 8 loads the die D picked up from the wafer 11 on the intermediate stage 31, picks up the die D again from the intermediate stage 31 by the bonding head 41, and is transferred. Bonding to the substrate (S).

Step BS14: The control unit 8 takes out the substrate S to which the die D is bonded from the substrate transfer claw 51 in the substrate take-out unit 7. The substrate S is taken out from the die bonder 10.

As described above, the die D is mounted on the substrate S via the die attach film 18 and is carried out from the die bonder. Then, it is electrically connected to the electrode of the substrate S through the Au wire in the wire bonding process. Subsequently, the substrate S on which the die D is mounted is carried into the die bonder, and the second die D is laminated on the die D mounted on the substrate S through the die attach film 18 After being taken out from the die bonder, it is electrically connected to the electrode of the substrate S through the Au wire in the wire bonding process. After the 2nd die D is peeled from the dicing tape 16 by the method mentioned above, it is conveyed by a pellet attaching process, and is laminated on the die D. After the above process is repeated a predetermined number of times, the substrate S is transferred to a mold process, and a plurality of dies D and Au wires are sealed with a mold resin (not shown), thereby completing a laminate package.

As described above, when assembling a laminate package in which a plurality of dies are three-dimensionally mounted on a substrate, it is required to reduce the thickness of the die to 20 μm or less in order to prevent an increase in the package thickness. On the other hand, since the thickness of the dicing tape is about 100 µm, the thickness of the dicing tape may be 4 to 5 times the thickness of the die.

If such a thin die is attempted to be peeled from the dicing tape, the deformation of the die following the deformation of the dicing tape is more likely to occur. However, in the die bonder of this embodiment, the die is picked up from the dicing tape. Damage can be reduced.

In the above, the invention made by the present inventor has been specifically described based on the embodiments and examples, but it is needless to say that the present invention is not limited to the above embodiments and examples, and various modifications are possible.

For example, although it has been described that the plurality of blocks of the first unit have a concentric quadrangular shape, they may have a concentric circular shape or a concentric elliptical shape, or may be configured by arranging rectangular blocks in parallel.

Further, in the embodiment, the pickup target die and the peripheral die are sucked/released at the same timing, but the pickup target die and the peripheral die may be sucked/released at separate timings. Thereby, more reliable peeling can be performed.

In addition, in the embodiment, the blocks of each stage are sequentially pushed up, but since each stage can be operated independently of each other, both push-up/pull-down operations may be mixed.

In addition, in the examples, an example of using a die attach film has been described, but it is not necessary to provide a preform portion for applying an adhesive to a substrate and use a die attach film.

In addition, in the embodiment, a die bonder that picks up a die from a die supply unit with a pickup head, loads it on an intermediate stage, and bonds the die mounted on the intermediate stage to a substrate with a bonding head has been described, but is not limited thereto. It is applicable to a semiconductor manufacturing apparatus that picks up a die from a die supply.

For example, it is also applicable to a die bonder that does not have an intermediate stage and a pickup head, and bonds the die of the die supply unit to the substrate with a bonding head.

In addition, it is applicable to a flip chip bonder that does not have an intermediate stage, and that picks up the die from the die supply unit and rotates the die pickup head upward to transfer the die to the bonding head and bond to the substrate with the bonding head.

In addition, it is applicable to a die sorter that does not have an intermediate stage and a bonding head, and loads a die picked up from a die supply unit by a pickup head to a tray or the like.

11: wafer
13: push up unit
16: dicing tape
22: collet
8: control unit
10: die bonder
D: Die

Claims (15)

A push-up unit having a plurality of blocks in contact with the dicing tape and pushing up a die from under the dicing tape,
A collet for adsorbing the die,
With respect to the peeling model that reproduces the characteristics of the pushing unit, feedback control is performed to follow the output of the peeling model to a target value of the peeling amount of the die from the dicing tape and the bending stress of the entire die, and the A control unit configured to make a push-up amount, which is a control input for the peeling model, as a push-up amount of the block of the push-up unit
A semiconductor manufacturing apparatus comprising a. The method of claim 1,
The peeling model includes: the sticking force of the adhesive material of the dicing tape calculated based on the push up amount of the block, the width of the block, and the Young's modulus of the die, and the adhesive material calculated based on the adhesive material model of the adhesive material. A semiconductor manufacturing apparatus including a deformation amount. The method of claim 2,
The adhesive material model includes a spring constant and a damping constant, a semiconductor manufacturing apparatus. The method of claim 3,
The peeling model,
A characteristic of starting peeling when the amount of push up is more than a predetermined value, and
The characteristic that the peeling progress is accelerated from the start of peeling to the peeling progress, and
By changing the parameters of the adhesive, the amount of push-up at which peeling starts and the speed of peeling progress are changed.
Having a semiconductor manufacturing apparatus. The method of claim 1,
The control unit is configured to raise an outer block among the plurality of blocks to reach a maximum value of the push-up amount, and then decrease the push-up amount, thereby raising the block adjacent to the inner block of the outer block, Semiconductor manufacturing equipment. The method of claim 1,
The control unit, after reaching the maximum value of the push-up amount by raising the outer block among the plurality of blocks, decreases the push-up amount to reach the minimum value of the push-up amount, and then increases the push-up amount, A semiconductor manufacturing apparatus configured to raise an inner block adjacent to the outer block. The method of claim 1,
The pushing up unit has a plurality of independent drive shafts corresponding to the plurality of blocks, and is configured to be programmable to set a pushing speed and a pushing amount of the blocks. The method of claim 1,
The semiconductor manufacturing apparatus, wherein the die further includes a die attach film between the die and the dicing tape. The method of claim 1,
A semiconductor manufacturing apparatus further comprising a pickup head to which the collet is mounted. The method of claim 9,
An intermediate stage for loading the die picked up by the pickup head;
Bonding head for bonding a die loaded on the intermediate stage onto a substrate or an already bonded die
A semiconductor manufacturing apparatus further comprising a. (a) holding the dicing tape in a semiconductor manufacturing apparatus having a plurality of blocks in contact with the dicing tape, a push-up unit for pushing up a die from under the dicing tape, and a collet for adsorbing the die The process of bringing in the supporting wafer ring
And the process of preparing
(b) step of pushing the die with the pushing unit and picking up the die with the collet
And,
In the step (b), with respect to a peeling model that reproduces the characteristics of the pushing unit, the output of the peeling model is followed by a target value of the peeling amount of the die from the dicing tape and the bending stress of the entire die. A method of manufacturing a semiconductor device, wherein feedback is controlled so as to be performed, and the die is pushed up using an amount of push-up, which is a control input to the peeling model, as a push-up amount of the block of the pushing unit. The method of claim 11,
In the step (b), after the outer block of the plurality of blocks is raised to reach a maximum value of the push-up amount, the push-up amount is reduced to raise the block adjacent to the inner block of the outer block, A method of manufacturing a semiconductor device. The method of claim 11,
In the step (b), after reaching the maximum value of the push-up amount by raising the outer block among the plurality of blocks, after reaching the minimum value of the push-up amount by reducing the push-up amount, the push-up amount is increased. The method for manufacturing a semiconductor device, wherein the block on the inner side adjacent to the outer block is raised by increasing the block. The method of claim 11,
(c) A method of manufacturing a semiconductor device, further comprising a step of bonding the die onto a substrate or an already bonded die. The method of claim 14,
The step (b) further includes a step of loading the picked up die on an intermediate stage,
The method of manufacturing a semiconductor device, wherein the step (c) further includes a step of picking up the die from the intermediate stage.

Images