WO2004081997A1 - 半導体装置製造装置および半導体装置の製造方法 - Google Patents
半導体装置製造装置および半導体装置の製造方法 Download PDFInfo
- Publication number
- WO2004081997A1 WO2004081997A1 PCT/JP2004/003332 JP2004003332W WO2004081997A1 WO 2004081997 A1 WO2004081997 A1 WO 2004081997A1 JP 2004003332 W JP2004003332 W JP 2004003332W WO 2004081997 A1 WO2004081997 A1 WO 2004081997A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- chip
- chips
- wafer
- semiconductor device
- joining
- 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
Links
Classifications
-
- 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
- H10P72/06—Apparatus for monitoring, sorting, marking, testing or measuring
- H10P72/0612—Production flow monitoring, e.g. for increasing throughput
Definitions
- the present invention relates to an apparatus for manufacturing a three-dimensionally mounted semiconductor device in which a plurality of chips are stacked, and to a manufacturing apparatus capable of flexibly responding to the properties of individual chips.
- CMO S-FET complementary MOS field-effect transistor
- LSI processing capability through miniaturization centered on lithography.
- wiring delays inside the LSI become apparent, and it is necessary to reduce the chip size as much as possible to reduce the RC delay, which is the product of resistance and capacitance. Has occurred.
- the spread of mobile phones and other devices has led to a demand for multifunctional and miniaturized LSIs.
- three-dimensional mounting LSI which vertically stacks two-dimensional LSI chips, is becoming widespread.
- chip formation by three-dimensional stacking there are several forms of chip formation by three-dimensional stacking.
- the first method is to select the non-defective LSI chip, mount it on a thin package, and stack multiple packages. This method is disclosed, for example, in Japanese Patent Publication No. Hei 10-1233481 (applicant: Samsung Electronics).
- the second method is to directly stack multiple non-defective chips vertically. This method is disclosed in, for example, Japanese Patent Publication No. 2002-1000072 (applicant: Sharp).
- the third method is to stack the wafer itself and then divide it into chips. This one The law is disclosed, for example, in Japanese Patent Laid-Open Publication No. 11-2016 (Applicant: Japan Science and Technology Corporation).
- the yield can be reduced by stacking non-defective chips.However, in addition to the cost increase due to the package cost, additional connection wiring between the chip and the package is required to achieve the LSI. There is a limit to the speeding up of the processing capacity.
- the third method is stacking for each wafer, so the handling is relatively excellent, but the yield in the process before chip division does not reach 100%, and the total number of wafers when the yield of 3D SI is reached Degrades in proportion to the power of N. Therefore, at present it is difficult to apply it to the actual LSI manufacturing process.
- the most likely form of LSI manufacturing is the second method of stacking chips, in two ways.
- the first method is to wire the pads (wire bond) for each chip after stacking the chips. In this case, although very effective for inexpensive chips and memories, delay occurs because the final wiring of the chip is wire wiring.
- the second method is an improvement of this method, in which a direct connection is made using bumps on the chip.
- the present invention does not require human intervention on wafers and chips and has a high throughput. It is an object to propose an apparatus for manufacturing a three-dimensionally mounted semiconductor device.
- Another object of the present invention is to provide a method capable of manufacturing a three-dimensional mounting semiconductor device at low cost.
- the present invention relates to a step of joining electrode portions formed on a chip on which a circuit pattern is formed or on an assembly of the chips, and formed on each of the chips conducting with the circuit pattern.
- a bonding device for bonding the electrode portions of each chip of the chip or the assembly of the chips, and an execution device for performing each of the other steps.
- Transport means for transporting the chip or the aggregate of the chips, and the chip or the aggregate of the chips including a step using the bonding step according to the chip or the aggregate of the chips.
- Control means for determining a step for stacking the chips, wherein the control means obtains information on the chip or the assembly of chips, The ⁇ corresponding to the process that sends to said conveying means and said execution means.
- the execution means may include at least cutting means for cutting the chip assembly, and the control means may provide cutting information to the cutting means.
- the information on the chip or the aggregate of the chips may be used as the chip design information or the chip inspection information.
- a selection means for selecting the chip based on the design information or the inspection information may be provided.
- control means may also serve as a control means for controlling the joining by the joining means.
- the present invention also relates to a semiconductor device manufacturing apparatus for processing chips formed on a wafer, wherein the plurality of execution units are arranged at least in a plurality of different locations and perform processing performed on the wafer or the chip.
- a transport unit disposed between the execution units, for transporting the wafer or the chip, a control unit provided in each of the execution units, and a network for communicatively connecting the control units with a common protocol.
- the execution unit may be configured such that a common transport unit can be attached.
- the present invention provides a method of manufacturing a semiconductor device, comprising: a step of joining electrode portions formed on a chip on which a circuit pattern is formed or an aggregate of the chips and formed on each of the chips to be electrically connected to the circuit pattern.
- the present invention relates to a method of manufacturing a semiconductor device obtained by stacking chips on which circuit patterns are formed, wherein the step of patterning the stacked chips on the same wafer includes the steps of: And separating the divided chips for each circuit pattern formed on the chip; and laminating the chips separated for each circuit pattern in a predetermined order. is there.
- FIG. 1 is a flowchart showing one representative step of the manufacturing method according to the embodiment of the present invention, and shows a manufacturing process of a laminated chip.
- FIG. 2 is a diagram showing the arrangement of various chips in the same exposure area on a wafer.
- FIG. 3 is a diagram showing a schematic configuration of the semiconductor device manufacturing apparatus according to the embodiment of the present invention.
- FIG. 4 is a view showing a manufacturing process of the laminated chip according to the embodiment of the present invention, and is a flowchart showing a part of a process of manufacturing the laminated chip after laminating the wafer unit.
- FIG. 5 is a view showing a manufacturing process of the laminated chip according to the embodiment of the present invention, and is a flow chart showing a part of a process of producing the laminated chip after laminating the chip in units of a chip class.
- the embodiment of the present invention exemplifies a three-dimensionally mounted semiconductor device in which a through electrode is formed.
- the through-electrode is an electrode formed so as to penetrate the wafer on which the circuit is formed up and down, and has a substantially shorter wiring length than a case where each chip is bonded by wire bonding, thereby improving signal delay. Therefore, this through-electrode method is set as a promising method.
- FIG. 1 is a flowchart showing a procedure for manufacturing a three-dimensionally mounted semiconductor device (laminated chip) according to the present embodiment.
- FIG. 3 is a block diagram showing a layout of the semiconductor device manufacturing apparatus of the present embodiment.
- the semiconductor device manufacturing apparatus 1 has a plurality of execution units that individually perform each process in the manufacture of a laminated chip.
- the execution units are through via processing module 11, through via sidewall insulating film formation module 12, electrode metal burying processing module 13, support body formation processing module 14, thinning processing module 15, micro bump formation module 16 ⁇ Individualization processing module 17 ⁇ Taping processing module 18 ⁇ Washing module 19 ⁇ H Expanding processing module 20, chip selection module 21, individual chip library processing module 22, chip library module 23, high-precision alignment pressure module 24, resin filler device 25, and chip stacking processing module 26 Have been.
- a vacuum inspection system module 30 and an atmospheric inspection system module 31 for inspecting a result processed by each module or device are provided.
- a resist coating / developing device 40 and an exposure device 41 for forming a through electrode are provided as an execution unit.
- the three-dimensional mounted semiconductor chip manufactured by the semiconductor device manufacturing apparatus 1 is carried out of the semiconductor device manufacturing apparatus 1 by a 3D chip loader 51.
- the semiconductor device manufacturing apparatus 1 is provided with a transfer device that transfers wafers or chips and connects between modules and between devices. Have been. These transfer devices are appropriately selected and arranged from existing transfer devices such as a robot arm using a vacuum chuck or a Verneuil chuck.
- Each execution unit is equipped with a processor (PC) that controls these units, and each processor (PC) and the processor of the transport device (not shown) are connected to a network using a common protocol. . -, The quality information of the chips obtained in the KGD sorting process (S2) described later, the design information of each chip formed on the wafer (EDA information described later), and the process in which the corresponding chip is being processed. Information on each chip, such as information, is provided to the processor (PC) of each execution unit via the network.
- PC processor
- the control of each execution unit is designed by EDA so that the execution unit can accurately perform the processing for 3D mounting on each chip. This can be done based on the information.
- the manufacturing process is determined by the processor (PC) provided in the high-precision alignment pressurizing module 24, which is one of the execution units.
- Step S 1 In the flow chart of FIG. 1 showing the manufacturing procedure of the laminated chip, S4 to S14 The process is performed in the semiconductor device manufacturing apparatus 1 of the present embodiment shown in FIG. (Step S 1)
- a wafer pre-process is performed by an apparatus different from the semiconductor device manufacturing apparatus 1 used in the present embodiment, and the pre-processed wafer is input.
- the pre-process in this embodiment is a process in which a wafer is formed through the same manufacturing process as a method for forming a circuit or element of a normal semiconductor device on a wafer. The circuit pattern of the chip is formed.
- a circuit pattern is usually formed on a wafer by a projection exposure apparatus.
- the masks required to use this projection exposure apparatus are very expensive. Since a plurality of different chips are required in a three-dimensionally mounted semiconductor device, it is preferable to design a mask so that these plural types of chips can be projected and exposed with a small number of masks. Note that, in the present embodiment, description will be made below on the assumption that one semiconductor device is manufactured by stacking four types of chips.
- FIG. 2 is a schematic configuration diagram showing a part of the wafer, on which patterns used for the three-dimensionally mounted semiconductor device according to the present embodiment are formed.
- the hatched portion in FIG. 2 is a portion corresponding to a chip, and the illustrated range is a region corresponding to an exposure field on the wafer.
- Chips are regularly formed on the entire surface of the wafer in an arrangement as shown in FIG. In a wafer having a chip arrangement as shown in Fig. 2, for example, if the exposure field is a chip of 4 O mm x 25 mm, the four types of chips A, B, C, and D exist in this field. Design the mask.
- the four types of chips A, B, C, and D automatically increase as described later.
- the alignment and the lamination are performed with high accuracy, and one three-dimensional laminated chip is completed. Therefore, a three-dimensional mounting semiconductor device (laminated chip) can be made with one mask, so that the number of masks can be reduced and the cost for masks can be reduced as compared with a method in which only the same type of chip is made on one wafer. It is possible to reduce.
- EDA Electronic Des According to (nation)
- the design is performed on the assumption that each chip is stacked. This is designed so that the connection between each chip or IP block is optimized (automatic routing optimization). Then, the design information of the wiring pattern designed by the EDA is supplied to a semiconductor device manufacturing apparatus 1 according to the present embodiment, which will be described later, in association with a chip formed on the wafer.
- KGD known Good Die: quality assurance equivalent to packaged products
- S2 chip coordinates
- the sorting process of KGD etc. is left up to the manufacturer to carry out the same testing as KTD (Known Test Die: probing test that is the same as the package. Not covered by warranty), PD (Probed Die: Different levels of selection criteria may be adopted, such as those specified and not covered by the guarantee.
- This KGD sorting step (S2) may be taken into the semiconductor device manufacturing apparatus 1 described later.
- the wafer is cleaned in the initial cleaning step (S3) shown in FIG. Since an existing wafer cleaning apparatus is used in this cleaning step, a detailed description is omitted here.
- the washed wafer is supplied to the semiconductor device manufacturing apparatus 1 shown in FIG.
- the wafer that has completed the initial cleaning step (S 3) is temporarily stored in the loader 50.
- the wafer stored in the header 50 enters a through via processing step (S4).
- a through via processing step S4
- a resist is applied to the wafer by the resist coating and developing device 40, and then a mask corresponding to the through-via pattern to be formed is selected by the exposure device 41 based on the wafer information provided by the EDA. Is done.
- the through-via pattern is projected and exposed on the wafer by using the selected mask.
- the resist is developed by the resist coating / developing device 40 and the wafer is transferred to the through via processing module 11. Between each of these devices and modules, there is disposed a known transfer device disclosed in Japanese Patent Application Laid-Open No. 9-148384, which mechanically holds a wafer.
- vias having a predetermined depth are formed in the provided wafer.
- a via having a predetermined depth is formed by using reactive ion etching or the like.
- the PC that controls the penetrating peer processing module 11 can transmit the reactive ion etching processing conditions on the network in association with the processed wafer.
- the wafer in which the via is formed is transferred to the through-via-side-wall-side-insulation-film forming module 12, and the side-wall-side-insulation-film forming step (S 5) is performed.
- a transfer device that mechanically holds the wafer is used as the transfer device.
- an oxide film such as SiO 2 is formed on the via sidewall of the wafer by a plasma CVD film forming method or the like.
- the PC that controls the through-via-sidewall-insulating-film forming module 12 is also connected to the network I and the workpiece, so that the film forming conditions for each processed wafer can be transmitted on the network. I have. Therefore, when a chip is determined to be defective, it is easier to obtain clues to clarify the cause of the defect.
- the wafer having the oxide film formed in the via is transferred to the electrode metal burying module 13, and the pier having the insulating film formed is filled with a copper material or the like.
- a barrier layer such as titanium nitride
- a seed layer such as copper
- the copper is grown by an electrode method or the like, so that the inside of the via is formed.
- the copper material is filled.
- each of the through via processing module 11, the through via side wall insulating film forming module 12, and the electrode metal burying processing module 13 is connected to the vacuum inspection module 30 by a transfer device (not shown).
- a transfer device not shown.
- whether a via is formed at a predetermined depth and a predetermined shape in each process Inspect whether the film or copper material is formed without defects.
- the 30 is provided with an X-ray image processing device, an infrared image processing device, and the like, and inspects the inside of the via as appropriate to detect a defective chip. Then, information on the defective chip is provided to the processor (PC) of the module or device of each of the other execution units via the network.
- PC processor
- the wafer on which the through electrodes are formed is transferred to a support forming step (S7) by the transfer device.
- a UV curable resin is applied to the wafer, a substrate such as quartz is attached to the surface (surface) of the wafer on which the circuit pattern is formed, and UV curing is performed to transfer the wafer to the substrate as the support. Glue.
- the wafer to which the support is adhered is transferred to a slicing process (S8).
- the wafer is transferred to the thinning processing module 15 and thinned to a thickness of about 50 micrometers or less. Also, by thinning the wafer, the other end surface of the through electrode is exposed, so that the through electrode can be exposed on both surfaces of the wafer.
- the back surface of the wafer is ground by a grinder, and the surface is smoothed and flattened by a CMP (Chemical Mechanical Olisliing) device. The flatness is set according to the joining method.
- microbump forming process step (S9) for forming a microbump on the back surface of the wafer is performed.
- microbump patterning is performed by a photolithography method as in the case of forming a through via, and a bump such as gold is formed using a known film forming method.
- the wafer with the micro-bumps formed on the back surface is then moved to a singulation process (S10).
- the wafer is cut into chips by a singulation module 17 having a dicing saw.
- a very thin wafer Since the wafer is diced while being attached to the support, it is possible to reduce the problem of chipping, which is likely to occur for a thinned chip.
- dicing is performed so that the support is not completely separated.
- the chips diced and singulated are transferred to a taping module 18 while being fixed to a support, and subjected to a taping step (S11).
- a resin resin tape for transfer is adhered to a surface of the chip opposite to a surface to which the support is adhered, in a state where the singulated chips are fixed to the support.
- the resin tape has an area such that all the chips fixed to the support can be bonded with the same tape, and the chips fixed to the same support are fixed to the same resin tape. Then, in the taping module 18, the support is peeled off from the chip bonded to the resin tape.
- the singulating processing module 17 and the taping processing module 18 are each a vacuum chuck. It is connected to the atmospheric inspection module 31 by a transfer device having Atmospheric inspection system module 31 is equipped with an optical microscope and an image measurement device.a Use these to observe the state of wafers and chips processed by each module, and to check for the presence of defects in each process. Find out.
- the steps of the micro-bump formed by the micro-bump forming module 16 can also be examined by the atmospheric inspection module 31. The information obtained from the inspection result is transmitted to other modules and devices via a network. Provided through.
- each chip adhered to the resin tape is transported to the cleaning module 19, and after being cleaned, is transferred to the individual chip library processing step (S12).
- the resin tape is stretched along the chip arranging direction by the tape expansion module 20 to facilitate handling of individual chips.
- the chip selection module 21 has a part determined to be defective or defective in the previous process based on the information obtained through the network. Separate the chips from non-defective chips and pick up only non-defective chips, or pick up chips that have a part determined to be defective or defective. In this way, only non-defective chips are processed in the subsequent steps.
- the chips are classified according to chip types (A, B, C, D) among the good chips.
- the number of chips of each type may not be the same by removing chips having a portion determined to be defective or defective from the subsequent process.
- individual chips The library processing module 22 supplies the missing chips from the chips manufactured on different wafers in advance, and matches the number of non-defective chips for each chip type. Thereafter, the non-defective chips selected in the KGD selection process (S 2) and the non-defective chips of the four chip block data relating to the four types of chips A, B, C, and D shown in FIG.
- the chip library module 23 stores, for example, every four chip blocks or arranges four types of chips A, B, C, and D in order, and associates the arrangement order with the chip type. Attach. Then, the array information and the chip type information are provided to the execution unit that performs the subsequent processing through the network.
- the chips selected by the chip library module 23 are transferred to the high-precision alignment pressurizing module 24 by a transfer device having a Bernoulli chuck or the like, and the chip stacking process (S13) is performed. I do.
- the chips are sequentially aligned and stacked so that the bumps and electrodes are at the same position according to the type of chip.
- the high-precision alignment pressurizing module 24 obtains the positions of the electrodes and the bumps of the microphone opening from the image, for example, as in an existing flip-chip bonder, and connects the electrodes and the micro-bumps to be joined together. It is possible to use an arrangement that performs alignment so as to be at the same position, and to which a high-precision alignment mechanism as used in an exposure apparatus is added.
- Methods of bump bonding include metal bonding by heating and pressing or ultrasonic waves, or N There are non-metallic connections such as CP (Non Conductive Paste), NCF (Non Conductive Film), ACP (Anisotro pic Conductive Paste), ACF (Anisotropic Conductive Film).
- CP Non Conductive Paste
- NCF Non Conductive Film
- ACP Anisotro pic Conductive Paste
- ACF Anisotropic Conductive Film
- the resin-sealed chip is transferred to a three-dimensional mounting semiconductor device (3D-Chip) slot 51, and the stacked chip is output from the device (S14).
- the resin fill-in device 25 is not limited to a device in which a resin is filled between chips after lamination, and is also used in the case of non-metal connection by NCP or ACF.
- a thermosetting epoxy resin is applied to the stacked surfaces of the chips, and the high-precision alignment pressurizing module 24 applies heat and pressure during the stacking, so that the bumps are joined together.
- the device 1 is a device that can cope only by adjusting the control of the transport device irrespective of the manufacturing method because the respective execution units are connected by the transport device. For example, even if the joining method changes, it can be handled.
- each module and each apparatus are connected by a transfer device and are connected to each other by a network
- the manufacturing process distributed on the network is transferred. It can be received by the control unit of the device.
- the transfer device By controlling the transfer device according to the type of chip and the manufacturing process, it becomes possible to give an appropriate processing to each chip.
- a defective chip occurs at a fixed position on the wafer or if the probability of occurrence of a defective chip per wafer is extremely low, it is possible to form the same chip on one wafer and stack it for each wafer. preferable. By doing so, And the production time can be reduced.
- the design information by EDA is acquired from the network to detect what kind of chips are formed on each wafer, and after the respective wafers are stacked one on another, If it is determined that a laminated chip can be manufactured even after cutting for each chip, when the steps from S1 to S6 in FIG. 1 are completed, the subsequent processing steps are switched to the steps shown in FIG. You may do it.
- each module, device, and transport means are controlled to perform the following steps. First, the wafer is processed in the same manner from S1 to S6 shown in FIG. 1, and after the embedded electrodes are formed, the steps shown in FIG. 4 are performed.
- the space between the bonded wafers is filled with resin by the resin fill-in device 25.
- a method may be used in which the resin is coated in advance by a spray coating or the like as a function included in the resin filling device 25.
- the micro-bumps are transported to the vacuum inspection module 30 to check the bonding of the micro-bumps. If the bonding is confirmed, the bonded wafers are separated. Conveyed to the thinning module 15 Then, at least one of the bonded wafers is sliced by the thinning module 15 (S43), and when the embedded electrode is exposed from the back surface of one of the wafers, the wafer is transferred to the microphone opening bump forming module 16. A microbump is formed on the exposed embedded electrode (S41).
- the new wafer is processed up to step S6 in Fig. 1. Then, the new wafer is transferred to the microbump forming module 16 to form microbumps on the circuit pattern forming surface. Next, the stacked body of the wafer having the embedded electrodes exposed from the back surface and the new wafer are respectively washed, and after the washing, the wafers are stacked and pressure-bonded by the high-precision alignment pressure module 24 (S42). ). Thereafter, the back surface of the new wafer is thinned (S43). However, at the time of the last lamination, either the back surface of a new wafer or the wafer that has been bonded first and has no exposed electrodes may be thinned. At that time, it is sufficient to make the flakes thin enough to expose the embedded electrodes.
- the wafer stack is transported to the singulation module 17 and singulated to obtain a stacked chip.
- the three-dimensional stacked chip can be output by transferring it to the 3D chip loader 51 (S45).
- the number of lamination processing steps and the number of individualization processing steps are reduced, so that the number of manufacturing steps can be reduced and the manufacturing cost can be reduced. Even if no support is provided, one of the wafers is always kept in a thick state. Therefore, the slicing process can be performed without providing a support.
- the manufacturing process using each process shown in Fig. 5 It is preferable to manufacture a three-dimensionally mounted semiconductor device by a method.
- the steps up to the micro-bump processing step (S 9) shown in FIG. 1 are performed in the same manner as the above-described manufacturing method, and a series of steps shown in FIG. 5 are executed following the microphone opening bump processing step (S 9). I do. That is, without performing the processes after the singulation process (S 10) in FIG. 1 and proceeding to the wafer dividing process (S 51) in FIG. 5, the process is performed to form an aggregate of several chips from the wafer. The wafer is divided into chip classes.
- chip cluster when a chip is cut out from a wafer into a continuous state (hereinafter referred to as “chip cluster”), the outer shape of the chip cluster is However, it is preferable that the chip clusters stacked on each other be cut out so as to be the same.
- chip clusters are cut out from a wafer, it is preferable that the chip clusters stacked on each other are cut out so that defective chips have the same positional relationship. In such a case, defective chips are stacked at the time of stacking, so that it is possible to collectively discard only the stacked chips composed of the defective chips.
- This wafer dividing step is performed by the singulation module 17.
- the division is performed by the PC that controls the singulation processing module 17.
- the division method is determined for each wafer based on the KGD information acquired from the network and the information acquired from the vacuum inspection module 30 and the atmospheric inspection module 31.
- the high-precision alignment pressurizing module 24 is not provided for the actual installation. Therefore, on the premise that lamination is performed, it may be preferable to control the high-precision alignment by using a PC that controls the pressurization module 24 for the items related to the process that is reliably performed by performing lamination. . By doing so, there is no need to make many programs that are not used in each processing module or device resident, so that there is an advantage that it is easy to secure the resources of the PC to be controlled.
- the obtained chip class is transferred to a taping processing step (S52).
- This taping step (S52) is the same as the taping step (S11) shown in FIG. 1.
- the divided wafers are directly taped and the support is removed from the chip cluster.
- the taped chip class is removed by the tape expansion module 20. At this time, the tape is stretched so that the chip cluster can be easily removed from the tape. With the tape stretched, the chip cluster is removed, and each chip cluster is stored in the chip library module 23 (S53). At this time, each chip cluster is integrated in a chip library so that the chip clusters to be stacked can be identified. After that, the chip clusters are stacked by the high-precision alignment pressure module 24, and the micro bumps formed in the chip class are joined together. Then, the resin is introduced during the chip class by the resin fill-in device 25, and the chip cluster laminating step (S54) is completed.
- the chip class stacked on the singulation module 17 is transported again and cut into individual chips.
- the laminated chip thus separated is conveyed to a 3D—chip loader 51, and a three-dimensionally mounted chip is output.
- the number of times of stacking the chips is smaller than in the manufacturing method shown in FIG.
- the defective chips it becomes easier to remove the defective stacked chips.
- non-defective chips and defective chips can be prevented from being stacked, and the probability that even non-defective chips are removed together with defective chips is reduced.
- each processing module or apparatus is connected to each other by a network and a transporting means, and each processing module, apparatus, and transporting means are determined via the network according to the determined manufacturing process. Controllable. As a result, there is an advantage that an appropriate manufacturing method can be adopted according to the state of the chips formed on the wafer and the location of defective chips.
- the determination of the manufacturing method is preferably performed by the processor (P C) that controls the high-precision alignment pressurizing module 24 as described above.
- the high-precision alignment pressure module 24 is a device for stacking chip assemblies including chips and wafers, and has no possibility of being used for other purposes. On the other hand, other modules and devices may be used for other purposes. Therefore, the process required for stacking a chip or a set of chips, such as a wafer and a chip cluster, is selected by the processor (PC) of the high-precision alignment pressurizing module 24, so that the high-precision alignment pressurizing module 24 is selected. Processor (PC) can be used effectively.
- the semiconductor device manufacturing apparatus 1 has the following features. example W
- each processing module including each processing module other than those shown in the figure, is connected to a central control unit called a main champer, which has a vacuum inspection system and an air inspection system.
- the vacuum inspection system is equipped with a scanning electron microscope (SEM), an optical microscope, and an X-ray / IR image measurement device.
- the atmospheric inspection system is equipped with an optical microscope and an image measurement device.
- the inspection equipment used in this inspection system is a three-dimensional shape measurement device that inspects the shape of bumps, an X-ray or infrared inspection device that can inspect through-electrodes in a nondestructive manner, or inspects the surface condition. Optical microscopes and electron microscopes.
- the alignment accuracy between the electrodes is as low as 0,5 ⁇ (3 ⁇ ), and room-temperature bonding can be applied to the stacking method between chips as wiring becomes finer. In this case, it is necessary to add a device for surface activation to the high-precision alignment pressurizing module 24.
- a rewiring layer for rewiring the electrodes may be incorporated in the semiconductor device manufacturing apparatus 1 of the present embodiment so that the respective chips may be directly bonded to each other.
- the electrodes of each chip may be made conductive through the electrodes.
- Each processing module or device is a distributed processing system with individual processors (PCs).
- PCs processors
- a “central control system” in which all the modules are centrally managed by a single host computer is also effective, but here a distributed processing type is adopted as a system that can flexibly respond to various user requests.
- information exchange between each module / device and unit can be handled by performing common protocol communication.
- each module, device, and unit can recognize each other and have flexibility in the process that can be selected.
- a common transfer device can be attached, and it is easy to add modules, devices, and units, and to change the layout. Therefore, it is possible to easily add modules, devices, and units according to the process to be adopted.
- the inspection equipment for the vacuum inspection system and air inspection system installed in the main chamber described above are also connected on the same network, and the inspection results obtained here are fed back to EDA design data. By doing so, it is also expected to be able to design circuit layouts that are less prone to defects during manufacturing.
- wafers are sequentially sent, and processing is performed for each module, so that processing can be performed under high-throughput and optimal conditions without human intervention.
- This equipment can easily cope with stacking at the wafer level by replacing the unit transport system, etc.
- the chip library was used to select chips, but the expanded wafer was used. It goes without saying that it can be mounted directly on the die bonder from the ring.
- the present invention does not relate to optimization of individual processes, but realizes high throughput and high yield by consolidating each process manufacturing apparatus and sometimes introducing an inspection process. .
- optimal wafer processing proceeds for each module, apparatus, and unit, and good chips are sorted by the KGD sorting process and stacked from EDA design data. Since chips are automatically recognized and the optimal manufacturing method is selected as appropriate, 3D can be automatically generated with extremely high efficiency.
- processors (PCs) of each execution unit can communicate with each other using a common protocol.
- PCs processors
- a common transfer device can be attached, it is easy to add modules, devices, and units, and to change the layout. Therefore, expansion of each module, device, and unit can be easily performed according to the process to be adopted.
- the semiconductor manufacturing apparatus of the present invention can be used for three-dimensional mounting of a semiconductor chip. Further, the method for manufacturing a semiconductor device of the present invention can be used for mounting a semiconductor chip.
Landscapes
- Wire Bonding (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2003069296A JP2006049334A (ja) | 2003-03-14 | 2003-03-14 | 半導体装置製造装置及び半導体装置の製造方法 |
| JP2003-069296 | 2003-03-14 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2004081997A1 true WO2004081997A1 (ja) | 2004-09-23 |
Family
ID=32984613
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2004/003332 Ceased WO2004081997A1 (ja) | 2003-03-14 | 2004-03-12 | 半導体装置製造装置および半導体装置の製造方法 |
Country Status (3)
| Country | Link |
|---|---|
| JP (1) | JP2006049334A (ja) |
| TW (1) | TW200425352A (ja) |
| WO (1) | WO2004081997A1 (ja) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5212410B2 (ja) * | 2010-03-19 | 2013-06-19 | 富士通株式会社 | 半導体部品製造システム、制御装置、およびコンピュータプログラム |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06140459A (ja) * | 1992-10-27 | 1994-05-20 | Olympus Optical Co Ltd | 半導体製造装置 |
| JPH10256109A (ja) * | 1997-03-11 | 1998-09-25 | Canon Inc | 半導体製造装置及びデバイス製造方法 |
| JPH11265841A (ja) * | 1998-12-18 | 1999-09-28 | Nikon Corp | デバイス製造支援装置、デバイス製造システム、及びデバイス製造方法 |
| JP2000228489A (ja) * | 1999-02-08 | 2000-08-15 | Rohm Co Ltd | チップオンチップの半導体チップおよび半導体装置 |
| JP2001210783A (ja) * | 2000-01-27 | 2001-08-03 | Seiko Epson Corp | 半導体装置の製造方法 |
-
2003
- 2003-03-14 JP JP2003069296A patent/JP2006049334A/ja active Pending
-
2004
- 2004-03-12 WO PCT/JP2004/003332 patent/WO2004081997A1/ja not_active Ceased
- 2004-03-12 TW TW093106571A patent/TW200425352A/zh unknown
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06140459A (ja) * | 1992-10-27 | 1994-05-20 | Olympus Optical Co Ltd | 半導体製造装置 |
| JPH10256109A (ja) * | 1997-03-11 | 1998-09-25 | Canon Inc | 半導体製造装置及びデバイス製造方法 |
| JPH11265841A (ja) * | 1998-12-18 | 1999-09-28 | Nikon Corp | デバイス製造支援装置、デバイス製造システム、及びデバイス製造方法 |
| JP2000228489A (ja) * | 1999-02-08 | 2000-08-15 | Rohm Co Ltd | チップオンチップの半導体チップおよび半導体装置 |
| JP2001210783A (ja) * | 2000-01-27 | 2001-08-03 | Seiko Epson Corp | 半導体装置の製造方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2006049334A (ja) | 2006-02-16 |
| TW200425352A (en) | 2004-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8994163B2 (en) | Semiconductor assemblies, stacked semiconductor devices, and methods of manufacturing semiconductor assemblies and stacked semiconductor devices | |
| TW202044507A (zh) | 整合扇出型裝置、三維積體電路系統及其製作方法 | |
| US8021981B2 (en) | Redistribution layers for microfeature workpieces, and associated systems and methods | |
| EP4418314A2 (en) | Die processing | |
| TWI727261B (zh) | 製造堆疊封裝式半導體封裝的方法 | |
| US20250299314A1 (en) | Automatic optical inspection system and method | |
| CN116635996A (zh) | 芯片堆叠结构及其制作方法、芯片封装结构、电子设备 | |
| Lau | Critical issues of TSV and 3D IC integration | |
| TW202406047A (zh) | 半導體裝置封裝及其形成方法 | |
| CN111933533A (zh) | 半导体封装结构及其制造方法 | |
| KR100886706B1 (ko) | 적층 패키지 및 그의 제조 방법 | |
| CN221613889U (zh) | 半导体装置封装 | |
| CN221201166U (zh) | 半导体元件封装 | |
| JP2005026346A (ja) | 半導体チップの積層方法 | |
| JP2006049334A (ja) | 半導体装置製造装置及び半導体装置の製造方法 | |
| EP4553896A2 (en) | Method for manufacturing integrated device packages | |
| TWI909176B (zh) | 半導體裝置封裝及其製造方法 | |
| US20250391828A1 (en) | Method for Producing a Semiconductor Component Assembly | |
| CN116207083A (zh) | 异质芯片封装方法及封装结构 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
| AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
| NENP | Non-entry into the national phase |
Ref country code: JP |
|
| WWW | Wipo information: withdrawn in national office |
Country of ref document: JP |
|
| 122 | Ep: pct application non-entry in european phase |