JP2007311828A - Semiconductor device, electronic component, circuit substrate, and electronic instrument - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a semiconductor device in which the package size is near its chip size, and which can effectively absorb thermal stresses separately from stress absorbing layer, and its manufacturing method, a circuit substrate, and an electronic instrument thereof. <P>SOLUTION: The semiconductor device has the package size near its chip size, and can effectively absorb the thermal stress separately from the stress absorbing layer. The semiconductor device 150 comprises a semiconductor chip with an electrode 158; a resin layer 152 as a stress-relaxing layer arranged on the semiconductor chip; an interconnection 154 formed from the electrode 158 to over the resin layer 152; and a solder ball 157 formed on the interconnection 154 above the resin layer 152. The resin layer 152 is formed so that it may include a recess 152a on front surface, and the interconnection 154 is formed so that it may pass on the recess 152a. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a semiconductor device, a manufacturing method thereof, a circuit board, and an electronic device, and more particularly to a semiconductor device having a package size close to a chip size, a manufacturing method thereof, a circuit substrate, and an electronic device.

  In pursuit of high-density mounting of semiconductor devices, bare chip mounting is ideal. However, bare chips are difficult to guarantee quality and handle. Therefore, a CSP (chip scale package) having a package close to the chip size has been developed.

  Among the CSP-type semiconductor devices developed in various forms, as one form, a patterned flexible substrate is provided on the active surface side of the semiconductor chip, and a plurality of external electrodes are formed on the flexible substrate. There is something that has been. It is also known to absorb heat stress by injecting resin between the active surface of a semiconductor chip and a flexible substrate.

However, if the resin alone does not sufficiently absorb heat stress, another means is required.
JP-A-8-250549 JP-A-8-203906 JP-A-1-196856 JP-A 64-1257 JP-A 63-229839 JP-A-6-69211 JP-A-4-10429 JP-A-3-20041 Japanese Utility Model Publication No. 4-74427 JP-A-2-109358

  SUMMARY OF THE INVENTION The present invention solves the above-described problems, and an object of the present invention is to provide a semiconductor device capable of effectively absorbing thermal stress separately from the stress absorption layer, and its package size is close to the chip size. It is to provide a manufacturing method, a circuit board, and an electronic device.

A method of manufacturing a semiconductor device according to the present invention includes a step of preparing a wafer on which an electrode is formed,
Providing a first stress relaxation layer on the wafer while avoiding at least a portion of the electrode;
Forming a first conductive portion from the electrode over the first stress relaxation layer;
Forming an external electrode connected to the first conductive portion above the first stress relaxation layer;
Cutting the wafer into individual pieces;
Have
A structure for increasing stress relaxation is formed in at least one of the step of providing the first stress relaxation layer and the step of forming the first conductive portion.

According to the present invention, since the conductive portion and the external electrode are formed on the stress relaxation layer, a substrate such as a film patterned by providing the external electrode in advance becomes unnecessary.

  In addition, since the conductive portion that connects the electrode and the external electrode can be freely formed according to the design, the arrangement of the external electrode can be determined regardless of the arrangement of the electrodes. Therefore, various semiconductor devices having different positions of the external electrodes can be easily manufactured without changing the circuit design of the elements formed on the wafer.

  Furthermore, according to the present invention, after the stress relaxation layer, the conductive portion and the external electrode are formed on the wafer, the wafer is cut to obtain individual semiconductor devices. Therefore, since the stress relaxation layer, the conductive portion, and the external electrode can be simultaneously formed for many semiconductor devices, the manufacturing process can be simplified.

  As a structure for increasing the stress relaxation, a recess may be formed on the surface of the first stress relaxation layer, and the first conductive portion may be formed so as to pass over the recess.

  By doing so, the conductive portion is formed by bending in a direction intersecting the surface of the stress relaxation layer, so that the stress can be absorbed by changing the bent state, and disconnection is prevented.

  As a structure for increasing the stress relaxation, in the step of forming the first conductive portion, the first conductive portion may be bent in the planar direction on the first stress relaxation layer.

  You may include the process of filling an elastic body on the said 1st conduction | electrical_connection part located in the said hollow part. Stress is further absorbed by this elastic body.

  A step of providing a second stress relaxation layer and a second conduction part connected to the first conduction part on the first stress relaxation layer in which the first conduction part is formed may be included. .

  By doing so, the stress relaxation layer is formed in a plurality of stages, and it becomes easier to disperse the stress.

  At least one of the first conductive portion and the second conductive portion may be formed in a planar shape having a planar extension larger than the thickness.

  By doing so, the signal is transmitted in the vicinity of the planar ground potential, so that an ideal transmission path is obtained.

On the first stress relaxation layer on which the first conductive portion is formed, a second stress relaxation layer and a second conductive portion are provided,
A third stress relaxation layer and a third conduction portion are provided on the second stress relaxation layer on which the second conduction portion is formed,
The second conductive portion may be formed in a linear shape, and the first and third conductive portions may be formed in a planar shape so as to have a larger planar extension than the second conductive portion.

  By doing so, the second conductive portion formed in a linear shape is sandwiched between the pair of planar conductive portions, so that the periphery is covered with the ground potential wiring. If it does so, the structure similar to a coaxial cable will be obtained and the signal which passes the 2nd conduction | electrical_connection part will become difficult to receive the influence of noise.

  A pair of wirings having a ground potential may be formed in parallel so as to sandwich the first conductive portion.

By doing so, the first conductive portion formed in a linear shape is sandwiched between the pair of wirings, so that the periphery is covered with the ground potential wiring. If it does so, the structure similar to a coaxial cable will be obtained and a signal will become difficult to receive to the influence of noise.

A semiconductor device according to the present invention includes a semiconductor chip having an electrode,
A first stress relaxation layer provided on the semiconductor chip so as to avoid at least a part of the electrode;
A first conductive portion formed from the electrode to the first stress relaxation layer;
An external electrode formed on the first conductive portion located above the first stress relaxation layer;
Have
The first stress relaxation layer is formed so as to have a depression on the surface, and the first conduction part is formed over the depression.

  By doing so, the conductive portion is formed by bending in a direction intersecting the surface of the stress relaxation layer, so that the stress can be absorbed by changing the bent state, and disconnection is prevented.

  An elastic body may be provided on the first conductive portion located in the recess so as to fill the recess.

  The first conductive portion may be formed by bending on the first stress relaxation layer.

  The first conductive part may be formed in a loose shape.

  You may have a 2nd conduction | electrical_connection part connected to a 2nd stress relaxation layer and a said 1st conduction | electrical_connection part on the said 1st stress relaxation layer in which the said 1st conduction | electrical_connection part was formed.

  By doing so, the stress relaxation layer is formed in a plurality of stages, and it becomes easier to disperse the stress.

  One of the two conducting parts including the first conducting part and the second conducting part is formed in a linear shape, and the other is formed in a planar shape having a planar extension wider than the linear conducting part. May be.

  The planar conductive portion may be a ground potential, and a signal may be input to the linear conductive portion.

A second stress relaxation layer and a second conduction portion provided on the first stress relaxation layer in which the first conduction portion is formed;
A third stress relaxation layer and a third conduction portion provided on the second stress relaxation layer in which the second conduction portion is formed;
Have
The second conductive portion may be formed in a linear shape, and the first and third conductive portions may be formed in a planar shape so as to have a larger planar extension than the second conductive portion.

  By doing so, the second conductive portion formed in a linear shape is sandwiched between the pair of planar conductive portions, so that the periphery is covered with the ground potential wiring. If it does so, the structure similar to a coaxial cable will be obtained and the signal which passes the 2nd conduction | electrical_connection part will become difficult to receive the influence of noise.

  You may have a pair of wiring formed in parallel so that the said 1st conduction | electrical_connection part may be pinched | interposed and used as a grounding potential.

  By doing so, the first conductive portion formed in a linear shape is sandwiched between the pair of wirings, so that the periphery is covered with the ground potential wiring. If it does so, the structure similar to a coaxial cable will be obtained and a signal will become difficult to receive to the influence of noise.

  You may have a protective film on the opposite side to the surface which has the said electrode of the said semiconductor chip.

  You may have a heat radiator on the opposite side to the surface which has the said electrode of the said semiconductor chip.

  The semiconductor device is mounted on a circuit board according to the present invention.

  The electronic device according to the present invention has this circuit board.

  Prior to describing the preferred embodiment of the present invention, the technology which is the premise of the present invention will be described.

(First prerequisite technology)
FIG. 5 is a plan view showing a semiconductor device as a premise of the present invention. This semiconductor device is classified as a so-called CSP. A wiring 3 is formed from the electrode 12 of the semiconductor chip 1 toward the center of the active surface 1 a, and an external electrode 5 is provided on each wiring 3. Since all the external electrodes 5 are provided on the stress relaxation layer 7, it is possible to alleviate stress when mounted on a circuit board (not shown). Further, a solder resist layer 8 is formed on the external electrode 5 as a protective film.

  As shown in the figure, the external electrode 5 is provided not on the electrode 12 of the semiconductor chip 1 but in an active region of the semiconductor chip 1 (region where active elements are formed). The external electrode 5 can be provided in the active region by providing the stress relaxation layer 7 in the active region and further disposing (withdrawing) the wiring 3 in the active region. Therefore, when the external electrode 5 is disposed, an active region, that is, a region as a fixed surface can be provided, and the degree of freedom of the setting position of the external electrode 5 is greatly increased.

Then, by bending the wiring 3 on the stress relaxation layer 7, the external electrodes 5 are provided so as to be arranged in a lattice pattern. In addition, at the junction between the electrode 12 and the wiring 3, the size of the electrode 12 and the size of the wiring 3 shown in the figure are
Wiring 3 <electrode 12
But
Electrode 12 ≦ Wiring 3
It is preferable that In particular,
Electrode 12 <Wiring 3
In this case, not only the resistance value of the wiring 3 is decreased, but also the strength is increased, so that disconnection is prevented.

  1A to 4C are views for explaining a method of manufacturing a semiconductor device according to the first prerequisite technology, and correspond to a cross section taken along line II in FIG.

  First, the electrode 12 and other elements are formed on the wafer 10 by a known technique. In this example, the electrode 12 is made of aluminum. As another example of the electrode 12, an aluminum alloy material (for example, aluminum silicon or aluminum silicon copper) may be used.

Further, a passivation film (not shown) made of an oxide film or the like is formed on the surface of the wafer 10 in order to prevent chemical changes. The passivation film is formed not only avoiding the electrode 12 but also avoiding a scribe line where dicing is performed. By not forming a passivation film on the scribe line, generation of dust during dicing can be avoided, and further, generation of cracks in the passivation film can be prevented.

  As shown in FIG. 1A, a photosensitive polyimide resin is applied to the wafer 10 having the electrodes 12 (for example, by “spin coating method”) to form a resin layer 14. The resin layer 14 is preferably formed in a thickness of 1 to 100 μm, more preferably about 10 μm. In the spin coating method, since a large amount of polyimide resin is wasted, a device that discharges the polyimide resin in a belt shape by a pump may be used. As such an apparatus, for example, there is an FAS ultra-precise discharge type coating system manufactured by FAS (see US Pat. No. 4,696,885).

As shown in FIG. 1B, a contact hole 14 a for the electrode 12 is formed in the resin layer 14. Specifically, the contact hole 14a is formed in the resin layer 14 by removing the polyimide resin from the vicinity of the electrode 12 by exposure, development, and baking. In the figure, there is no region where the resin layer 14 overlaps the electrode 12 when the contact hole 14a is formed. Although the resin layer 14 is not left on the electrode 12 at all, there is an advantage that an electrical contact with a metal such as a wiring provided in the subsequent process becomes good, but such a structure has to be necessarily provided. Do not mean. In other words, even if the resin layer 14 is in the vicinity of the outer periphery of the electrode 12, if the hole is formed so that a part of the electrode 12 is exposed, the object is sufficiently achieved. In this case, since the number of bends in the wiring layer is reduced, it is possible to prevent a decrease in wiring reliability due to disconnection or the like. Here, the contact hole 14a is tapered. Accordingly, the resin layer 14 is inclined at the end where the contact hole 14a is formed. Such a shape is formed by setting exposure and development conditions. Furthermore, if the electrode 12 is subjected to O ₂ plasma treatment, the polyimide resin can be completely removed even if a little polyimide resin remains on the electrode 12. The resin layer 14 thus formed becomes a stress relaxation layer in a semiconductor device as a finished product.

In this example, a photosensitive polyimide resin is used as the resin, but a non-photosensitive resin may be used. For example, a silicone-modified polyimide resin, an epoxy resin, a silicone-modified epoxy resin, or the like that has a low Young's modulus when solidified (1 × 10 ¹⁰ Pa or less) and can perform a stress relaxation function may be used.

  As shown in FIG. 1C, a chrome (Cr) layer 16 is formed on the entire surface of the wafer 10 by sputtering. The chrome (Cr) layer 16 is formed from the electrode 12 to the resin layer 14. Here, the material of the chrome (Cr) layer 16 was selected because of its good adhesion to the polyimide constituting the resin layer 14. Alternatively, considering crack resistance, aluminum alloy such as aluminum, aluminum silicon, aluminum copper, or copper alloy, or a metal having a spreadability (extension property) such as copper (Cu) or gold may be used. Alternatively, if titanium having excellent moisture resistance is selected, disconnection due to corrosion can be prevented. Titanium is preferable from the viewpoint of adhesion to polyimide, and titanium tungsten may be used.

Considering the adhesion with the chrome (Cr) layer 16, it is preferable to roughen the surface of the resin layer 14 made of polyimide or the like. For example, the surface of the resin layer 14 can be roughened by performing dry treatment exposed to plasma (O ₂ , CF ₄ ) or wet treatment with acid or alkali.

  Further, since the end of the resin layer 14 is inclined in the contact hole 14a, the chrome (Cr) layer 16 is similarly inclined in this region. The chrome (Cr) layer 16 becomes the wiring 3 (see FIG. 5) in the semiconductor device as a finished product, and also becomes a diffusion prevention layer for the polyimide resin when the layer is subsequently formed during the manufacturing. The diffusion preventing layer is not limited to chrome (Cr), and all the above-described wiring materials are effective.

  As shown in FIG. 1D, a photoresist is coated on the chrome (Cr) layer 16 to form a resist layer 18.

  As shown in FIG. 1E, a part of the resist layer 18 is removed by exposure, development, and baking. The remaining resist layer 18 is formed from the electrode 12 toward the center of the resin layer 14. Specifically, the remaining resist layer 18 is such that the resist layer 18 on one electrode 12 and the resist layer 18 on the other electrode 12 do not continue on the resin layer 14 (each in an independent state). It has become.

  Then, leaving only the region covered with the resist layer 18 shown in FIG. 1E (that is, using the resist layer 18 as a mask), the chrome (Cr) layer 16 is etched, and the resist layer 18 is peeled off. As described above, the metal thin film forming technique in the wafer process is applied to these pre-processes. The chrome (Cr) layer 16 thus etched is as shown in FIG. 2A.

  In FIG. 2A, the chrome (Cr) layer 16 is formed from the electrode 12 to the resin layer 14. Specifically, the chrome (Cr) layer 16 is not continuous between one electrode 12 and another electrode 12. That is, the chrome (Cr) layer 16 is formed so that wiring corresponding to each electrode 12 can be configured.

  As shown in FIG. 2B, a copper (Cu) layer 20 is formed on the uppermost layer including at least the chrome (Cr) layer 16 by sputtering. The copper (Cu) layer 20 serves as a base layer for forming external electrodes. Alternatively, a nickel (Ni) layer may be formed instead of the copper (Cu) layer 20.

  A resist layer 22 is formed on the copper (Cu) layer 20 as shown in FIG. 2C, and a part of the resist layer 22 is removed by exposure, development and baking as shown in FIG. 2D. Then, the region to be removed is at least part of the resist layer 22 located above the resin layer 14 and above the chrome (Cr) layer 16.

  As shown in FIG. 2E, a pedestal 24 is formed in a region where the resist layer 22 has been partially removed. The base 24 is formed by copper (Cu) plating, and a solder ball is formed thereon. Accordingly, the pedestal 24 is formed on the copper (Cu) layer 20 and is electrically connected to the electrode 12 through the copper (Cu) layer 20 and the chrome (Cr) layer 16.

  As shown in FIG. 3A, on the base 24, a solder 26 that becomes a solder ball as the external electrode 5 (see FIG. 5) is formed in a thick layer. Here, the thickness is determined by the amount of solder corresponding to the ball diameter required at the time of subsequent solder ball formation. The layer of the solder 26 is formed by electrolytic plating or printing.

  As shown in FIG. 3B, the resist layer 22 shown in FIG. 3A is stripped, and the copper (Cu) layer 20 is etched. Then, the base 24 becomes a mask, and the copper (Cu) layer 20 remains only under the base 24 (see FIG. 3C). Then, the solder 26 on the pedestal 24 is formed into a hemispherical ball or more by wet back to form a solder ball (see FIG. 3D).

A solder ball as the external electrode 5 (see FIG. 5) is formed by the above process. Subsequently, a process for preventing the oxidation of the chromium (Cr) layer 16 and the like, improving the moisture resistance of the completed semiconductor device, and achieving the objectives such as mechanical protection of the surface is shown in FIGS. 4A and 4B. This is done as shown in

  As shown in FIG. 4A, a photosensitive solder resist layer 28 is formed on the entire surface of the wafer 10 by coating. Then, exposure, development, and baking are performed to remove a portion of the solder resist layer 28 that covers the solder 26 and a region in the vicinity thereof. Thus, the remaining solder resist layer 28 serves as an antioxidant film, as a protective film when finally becoming a semiconductor device, and further as a protective film for the purpose of improving moisture resistance. Then, the electrical characteristics are inspected, and if necessary, the product number, manufacturer name, etc. are printed.

  Subsequently, dicing is performed to cut into individual semiconductor devices as shown in FIG. 4C. Here, the position where dicing is performed is a position where the resin layer 14 is avoided, as is apparent from a comparison between FIGS. 4B and 4C. Therefore, since dicing is performed only on the wafer 10, a problem in cutting a plurality of layers made of materials having different properties can be avoided. The dicing process is performed by a conventional method.

  According to the semiconductor device thus formed, the resin layer 14 becomes the stress relaxation layer 7 (see FIG. 5), so that the thermal expansion coefficient between the circuit board (not shown) and the semiconductor chip 1 (see FIG. 5). The stress due to the difference is relaxed.

  According to the semiconductor device manufacturing method described above, almost all the steps are completed in the wafer process. In other words, the process of forming the external terminals connected to the mounting substrate can be performed within the wafer process, and the conventional packaging process, that is, the individual semiconductor chips are handled and the inner lead bonding is performed on each individual semiconductor chip. It is not necessary to perform a process, an external terminal formation process, or the like. Further, when the stress relaxation layer is formed, a substrate such as a patterned film becomes unnecessary. For these reasons, a low-cost and high-quality semiconductor device can be obtained.

  In this example, the resin as the stress relaxation layer is a photosensitive polyimide resin, but other than that, a non-photosensitive resin may be used. In this example, two or more wiring layers may be provided. If the layers are stacked, the layer thickness generally increases and the wiring resistance can be lowered. In particular, when one of the wirings is made of chrome (Cr), copper (Cu) or gold has a lower electrical resistance than chrome (Cr), so that the wiring resistance can be lowered by combining them. Alternatively, a titanium layer may be formed on the stress relaxation layer, and a nickel layer or a layer made of platinum and gold may be formed on the titanium layer. Alternatively, two layers of platinum and gold may be used as the wiring.

(Second prerequisite technology)
6A to 7C are diagrams illustrating a method for manufacturing a semiconductor device according to the second prerequisite technology. The present technology differs from the first prerequisite technology in the steps after FIG. 3A, and the steps up to FIG. 2E are the same as the first prerequisite technology. Therefore, the wafer 110, the electrode 112, the resin layer 114, the chrome (Cr) layer 116, the copper (Cu) layer 120, the resist layer 122, and the pedestal 124 shown in FIG. 6A are the same as the wafer 10, the electrode 12, and the resin layer shown in FIG. 14, the chrome (Cr) layer 16, the copper (Cu) layer 20, the resist layer 22 and the pedestal 124, and the manufacturing method is the same as that shown in FIGS.

  In the present technology, as shown in FIG. 6A, thin solder 126 is plated on the pedestal 124, and the resist layer 122 is peeled off as shown in FIG. 6B. Further, using the thin solder 126 as a resist, the copper (Cu) layer 120 is etched as shown in FIG. 6C.

  Subsequently, as shown in FIG. 7A, a solder resist layer 128 is formed on the entire surface of the wafer 110, and as shown in FIG. 7B, the solder resist layer 128 in the region of the pedestal 124 is removed by exposure, development, and baking.

  Then, as shown in FIG. 7C, a thick solder 129 thicker than the thin solder 126 is plated on the base 124 where the thin solder 126 remains. This is done by electroless plating. The thick solder 129 is then formed into a hemispherical or more ball shape by wet back as in the state shown in FIG. Thus, the thick solder 129 becomes a solder ball as the external electrode 5 (see FIG. 5). Subsequent steps are the same as those in the first prerequisite technology described above.

  Also with this technique, almost all steps can be performed in the wafer process. In the present technology, the thick solder 129 is formed by electroless plating. Therefore, the pedestal 124 can be omitted, and the thick solder 129 can be formed directly on the copper (Cu) layer 120.

(Third prerequisite technology)
8A to 9D are diagrams for explaining a method for manufacturing a semiconductor device according to the third prerequisite technology.

  The wafer 30, the electrode 32, the resin layer 34, the chrome (Cr) layer 36, the copper (Cu) layer 40, and the resist layer 42 shown in FIG. 8A are the same as the wafer 10, the electrode 12, the resin layer 14, and the chrome (Cr ) Layer 16, copper (Cu) layer 20, and resist layer 22, and the manufacturing method is the same as that shown in FIGS.

  Then, a part of the resist layer 42 shown in FIG. 8A is removed by exposure, development and baking. Specifically, as shown in FIG. 8B, the resist layer 42 at other positions is removed, leaving only the resist layer 42 located above the chrome (Cr) layer 36 to be a wiring.

  Subsequently, the copper (Cu) layer 40 is etched to peel the resist layer 42, leaving the copper (Cu) layer 40 only on the chrome (Cr) layer 36, as shown in FIG. 8C. Thus, a wiring having a two-layer structure of the chrome (Cr) layer 36 and the copper (Cu) layer 40 is formed.

  Next, as shown in FIG. 8D, a photosensitive solder resist is applied to form a solder resist layer 44.

  As shown in FIG. 9A, contact holes 44 a are formed in the solder resist layer 44. The contact hole 44a is formed above the resin layer 34 and on the copper (Cu) layer 40 that is the surface layer of the wiring having a two-layer structure. The contact hole 44a is formed by exposure, development, and baking. Or you may print a soldering resist, providing a hole in a predetermined position so that the contact hole 44a may be formed.

  Subsequently, the solder cream 46 is printed in the contact hole 44a so as to form a raised shape (see FIG. 9B). The solder cream 46 becomes a solder ball by wet back as shown in FIG. 9C. Then, dicing is performed to obtain individual semiconductor devices shown in FIG. 9D.

  In the present technology, the solder ball pedestal is omitted, and solder cream printing is applied, so that the formation of the solder balls is facilitated and the manufacturing process is reduced.

  Moreover, the wiring of the semiconductor device to be manufactured has two layers of chromium (Cr) and copper (Cu). Here, chrome (Cr) has good adhesion to the resin layer 34 made of polyimide resin, and copper (Cu) has good crack resistance. Since the crack resistance is good, it is possible to prevent disconnection of the wiring or damage to the electrode 32 or the active element. Alternatively, the wiring may be constituted by two layers of copper (Cu) and gold, two layers of chromium and gold, or three layers of chromium, copper (Cu) and gold.

  In this technique, an example without a pedestal is given, but it goes without saying that a pedestal may be provided.

(Fourth prerequisite technology)
FIG. 10 is a diagram for explaining a method of manufacturing a semiconductor device according to the fourth prerequisite technology.

  The wafer 130, the electrode 132, the resin layer 134, the chrome (Cr) layer 136, the copper (Cu) layer 140, and the solder resist layer 144 shown in the same figure are the same as the wafer 30, the electrode 32, the resin layer 34, and the chrome (shown in FIG. 9A). The method is the same as that of the (Cr) layer 36, the copper (Cu) layer 40, and the solder resist layer 44, and the manufacturing method is the same as that shown in FIGS.

  In the present technology, instead of using the solder cream 46 in FIG. 9B, a solder ball 148 is mounted on the contact hole 144 a formed in the solder resist layer 144 by applying a flux 146. Thereafter, wetback, inspection, marking, and dicing processes are performed.

  According to the present technology, a pre-formed solder ball 148 is mounted and used as the external electrode 5 (see FIG. 5). Further, the pedestals 24 and 124 can be omitted as compared with the first and second prerequisite technologies. Furthermore, the wiring 3 (see FIG. 5) has a two-layer structure of a chrome (Cr) layer 136 and a copper (Cu) layer 140.

  In this technique, an example without a pedestal is given, but it goes without saying that a pedestal may be provided.

(Fifth prerequisite technology)
11A to 12C are diagrams illustrating a method for manufacturing a semiconductor device according to the fifth prerequisite technology.

  First, as shown in FIG. 11A, a glass plate 54 is bonded to a wafer 50 having electrodes 52. A hole 54 a corresponding to the electrode 52 of the wafer 50 is formed in the glass plate 54, and an adhesive 56 is applied.

  The thermal expansion coefficient of the glass plate 54 is a value between the thermal expansion coefficient of the wafer 54 serving as a semiconductor chip and the thermal expansion coefficient of the circuit board on which the semiconductor device is mounted. From this, the value of the thermal expansion coefficient changes in the order of the semiconductor chip obtained by dicing the wafer 54, the glass plate 54, and the circuit board (not shown) on which the semiconductor device is mounted. The difference between the two becomes smaller and the thermal stress becomes smaller. That is, the glass plate 54 becomes a stress relaxation layer. A ceramic plate may be used instead of the glass plate 54 as long as it has a similar thermal expansion coefficient.

When the glass plate 54 is bonded to the wafer 50, the adhesive 56 that has entered the hole 54 is removed by O ₂ plasma treatment, as shown in FIG. 11B.

Next, as shown in FIG. 11C, an aluminum layer 58 is formed on the entire surface of the wafer 50 and on the glass plate 54 by sputtering. Thereafter, if a film is formed on the surface of the hole 54, it is possible to protect aluminum which is relatively easily broken. Next, a resist layer 59 is formed as shown in FIG. 12A, and as shown in FIG. 12B, a part of the resist layer 59 is removed by exposure, development, and baking treatment. The resist layer 59 to be removed is preferably at a position other than the wiring pattern forming portion.

  In FIG. 12B, the resist layer 59 is left from above the electrode 52 to above the glass plate 54. Further, there is an interruption so that the upper part of one electrode 52 and the upper part of the other electrode 52 are not continuous.

  When the aluminum layer 58 is etched, as shown in FIG. 12C, the aluminum layer 58 remains in a region to be a wiring. That is, the aluminum layer 58 is formed as a wiring from the electrode 52 to the glass plate 54. In addition, an aluminum layer 58 is formed so that the electrodes 52 are not connected to each other and become wiring for each electrode 52. Alternatively, if it is necessary to make the plurality of electrodes 52 conductive, an aluminum layer 58 serving as a wiring may be formed correspondingly. In addition to the aluminum layer 58, any of the materials selected in the first prerequisite technology can be applied as the wiring.

  Since the wiring from the electrode 52 is formed by the above steps, solder balls are formed on the aluminum layer 58 as the wiring, and the wafer 50 is cut into individual semiconductor devices. These steps can be performed in the same manner as in the first prerequisite technique.

  According to this technique, although the glass plate 54 has the hole 54a, the formation of the hole 54a is easy. Therefore, it is not necessary to pattern the glass plate 54 so as to form bumps and wirings in advance. In addition, the metal thin film forming technique in the wafer process is applied to the formation process of the aluminum layer 58 and the like to be the wiring, and almost all processes are completed by the wafer process.

  Note that another stress absorbing layer such as a polyimide resin may be further provided on the glass plate 54 in the same manner as in the first prerequisite technology. In this case, since the stress absorption layer is newly provided, the thermal expansion coefficient of the glass plate 54 may be equal to that of silicon.

(Sixth prerequisite technology)
13A to 13D are views for explaining a method of manufacturing a semiconductor device according to the sixth prerequisite technology. In this technique, a polyimide plate was selected as the stress relaxation layer. This is because polyimide is a member having a low Young's modulus and suitable as a stress relaxation layer. In addition, for example, a composite plate such as a plastic plate or a glass epoxy type may be used. In this case, it is preferable to use the same material as the mounting substrate because there is no difference in thermal expansion coefficient. In particular, since there are many plastic substrates as mounting substrates today, it is effective to use a plastic plate for the stress relaxation layer.

  First, as shown in FIG. 13A, a polyimide plate 64 is bonded to a wafer 60 having an electrode 62 so as to be shown in FIG. 13B. The polyimide plate 64 is pre-coated with an adhesive 66.

  Next, as shown in FIG. 13C, a contact hole 64a is formed in the region corresponding to the electrode 62 using an excimer laser or the like, and as shown in FIG. 13D, an aluminum layer 68 is formed by sputtering. In addition to the aluminum layer 68, any of the materials selected in the first prerequisite technique can be applied.

  Thus, since the state is the same as that in FIG. 11C, the semiconductor device can be manufactured by performing the steps in FIG. 12A and thereafter.

  According to the present technology, since the polyimide plate 64 in which no holes are formed is used, a patterned substrate is not necessary. Other effects are the same as those of the first to fifth prerequisite technologies.

  As another technique, the stress relaxation layer may be preliminarily machined with a drill or the like to provide a hole, and thereafter, an arrangement process such as bonding to the wafer may be performed. In addition to machining, holes can be provided by chemical etching or dry etching. In addition, when forming a hole using chemical etching or dry etching, it may be performed on the wafer or in a previous preliminary process.

(First embodiment)
The present invention has been made to further improve the above-described technique, and a preferred embodiment of the present invention will be described below with reference to the drawings.

  14A to 14D are views showing a first embodiment according to the present invention.

In the semiconductor device 150 shown in FIG. 14A, a resin layer 152 made of polyimide is intermittently formed. The resin layer 152 becomes a stress relaxation layer. As the stress relaxation layer, a photosensitive polyimide resin is preferable, but a resin having no photosensitivity may be used. For example, a silicone-modified polyimide resin, an epoxy resin, a silicone-modified epoxy resin, or the like that has a low Young's modulus when solidified (1 × 10 ¹⁰ Pa or less) and can perform a stress relaxation function may be used.

  Further, the resin layer 152 is formed with a tapered recess 152a. And since the wiring 154 is formed along the surface shape of this hollow part 152a, the wiring 154 is bent in cross-sectional shape. A solder ball 157 is formed on the wiring 154. As described above, the wiring 154 is arranged on the resin layer 152 as a stress relaxation layer, and moreover, the wiring 154 is easily bent and expanded as compared with the case where the wiring 154 is simply arranged flat. Therefore, when the semiconductor device 150 is mounted on the circuit board, it becomes easy to absorb the stress due to the difference in thermal expansion coefficient. From the portion where the wiring 154 is displaced (bent portion or the like) to the solder ball 157, it is preferable to select a material having a higher elastic deformation rate as the resin layer 152. This applies to the following embodiments.

  Furthermore, as shown in FIG. 14A, it is preferable to provide an elastic body 156 above the depression 152a, specifically, in a wiring region formed in a depression corresponding to the depression 152a. The elastic body 156 may be formed of the material used for the resin layer 152 as a stress relaxation layer. The elastic body 156 can further absorb the stress that causes the wiring 154 to expand and contract. For example, a photoresist that forms the outermost layer (protective layer) may function as the elastic body 156. Further, the elastic bodies 156 may be provided individually corresponding to the individual depressions 152a.

  Thus, disconnection of the wiring 154 is prevented, or damage to the electrode 158 and the like through the wiring 154 due to stress is prevented. The electrode 158 and the wiring 154 are covered and protected by an outermost layer (protective layer) 155.

  Next, in the semiconductor device 160 illustrated in FIG. 14B, the second resin layer 166 and the second resin layer 166 are formed on the first resin layer 162 of the first wiring 164 formed from the electrode 169 to the first resin layer 162. Two wirings 168 are formed. The first wiring 164 is connected to the electrode 169, the second wiring 168 is connected to the first wiring 164, and a solder ball 167 is formed on the second wiring 168. In this way, if a plurality of resin layers and wirings are formed, the degree of freedom in wiring design increases. Note that the electrode 169 and the wirings 164 and 168 are covered and protected by an outermost layer (protective layer) 165.

In addition, an elongated wiring having an area that can be ignored can be formed in a planar shape having a planar extension (width or size). Moreover, when the resin layer becomes a plurality of layers, the stress is easily dispersed. Further, if the wiring formed in a planar shape is set to a GND (ground) potential or a power supply voltage potential, impedance control becomes easy and the high frequency characteristics become very excellent.

  Next, a semiconductor device 170 shown in FIG. 14C is a combination of the semiconductor devices 150 and 160 described above. That is, the first wiring 174 is formed on the first resin layer 172, and the second resin layer 176 is formed on the first wiring 174 so as to have a depression 176 a. The second wiring 178 formed on the second resin layer 176 is bent in the cross-sectional shape. Note that a solder ball 177 is formed on the second wiring 178. Further, the electrode 179 and the wirings 174 and 178 are covered and protected by an outermost layer (protective layer) 175. According to the present embodiment, the effect of combining the semiconductor devices 150 and 160 can be achieved.

  Next, in the semiconductor device 180 shown in FIG. 14D, a wiring 184 is formed on the stress relaxation layer 187 formed in the region indicated by the broken line so as to be bent in a planar shape from the electrode 182, and the wiring 184 is soldered. A bump 186 such as a ball is formed. Also in this embodiment, although the direction is different from that of the semiconductor device 150 (see FIG. 14A), the wiring 184 is bent, and therefore, the ability to absorb stress is excellent.

  Note that the wiring 184 that is bent into the planar shape shown in FIG. 14D may be bent three-dimensionally as shown in FIGS. 14A to 14C. By doing so, the effect of preventing disconnection is further enhanced. However, the stress relaxation layer 187 must be present under the wiring 184. The electrode 182 and the wiring 184 are covered and protected by an outermost layer (protective layer) not shown.

(Second Embodiment)
Next, the semiconductor device 190 shown in FIG. 15 is characterized by a wiring 200 that connects the aluminum pad 192 and a solder ball 196 provided on the stress relaxation layer 194. As the wiring 200, any of the wiring materials selected in the first prerequisite technology or the like may be used. The wiring 200 has a loose portion 200a. As shown in FIG. 14D, the loose portion 200a is in a state where the inside of the wiring is a cavity (slit), and a plurality of loose portions 200a are continuously formed with the normal wiring interposed therebetween. The loose portion 200a is more excellent in stress absorption than the bent wiring 184. By having the loose portion 200a, the wiring 200 is not cracked on the semiconductor chip, and the aluminum pad 192 and other active elements are not damaged, and the reliability of the semiconductor device is improved. Further, since the loose portion 200a is provided in one wiring, a fine space for the stress absorbing structure is sufficient. As a result, the degree of freedom in design can be improved while maintaining the miniaturization of the semiconductor device without departing from the category of CSP. In the present embodiment, the loose portion 200a is an example with respect to the planar direction, but it may be provided in the thickness direction.

  In the above-described embodiment or base technology, solder has been described as an example of the external electrode. However, there is no problem even if other well-known connection materials such as gold bumps are used. The external electrode is an active region of the semiconductor chip, and can be formed anywhere except on the electrode.

(Third embodiment)
FIGS. 16-20 is a figure which shows 3rd Embodiment based on this invention. FIG. 16 is a view showing a cross section of the semiconductor device according to the present embodiment. The semiconductor device 300 has a multi-layer (four-layer) structure on a semiconductor chip 302 and the surface is protected by a solder resist 350. Also in this embodiment, the materials and manufacturing methods described in the other embodiments and the prerequisite technology can be applied.

  17A and 17B are diagrams showing the first layer. Specifically, FIG. 17B is a plan view, and FIG. 17A is a sectional view taken along line VII-VII in FIG. 17B. The semiconductor chip 302 is formed with electrodes 304 to which signals are input or output. In the vicinity of the electrode 304, a stress relaxation layer 310 having an inclined end surface is formed. The stress relaxation layer 310 is an insulator, and specifically, a polyimide resin is preferable. A signal wiring 312 is formed from the electrode 304 to the stress relaxation layer 310. As shown in FIG. 17B, the signal wiring 312 has an island-shaped connection portion 312 a at the end opposite to the electrode 304. In addition, a GND plane 316 is formed so as to surround the connecting portion 312a and not to contact the connecting portion 312a. The GND plane 316 is connected to a ground electrode (not shown) of the semiconductor chip 302.

  18A and 18B are diagrams showing the second layer. Specifically, FIG. 18B is a plan view, and FIG. 18A is a cross-sectional view taken along line VIII-VIII in FIG. 18B. As shown in these drawings, the stress relaxation layer 320 is formed on the first layer described above. However, the stress relaxation layer 320 is formed so as to avoid the central portion of the connection portion 312a of the signal wiring 312 of the first layer. A signal wiring 322 is formed from the connection portion 312 a of the first layer to the stress relaxation layer 320 of the second layer. The signal wiring 322 includes a connection portion 322a connected to the connection portion 312a and another connection portion 322b. In addition, a signal wiring 324 that is not electrically connected to the signal wiring 322 is formed in the stress relaxation layer 320. The signal wiring 324 includes connection portions 324a and 324b. Further, other wirings 324 and 325 are formed in the stress relaxation layer 320, but the description is omitted because they are not linearly related to the present invention. Further, a GND plane 326 is formed so as to surround the signal wirings 322 and 324 and the wirings 324 and 325 so as not to contact each other. The GND plane 326 is connected to a ground electrode (not shown) of the semiconductor chip 302 via the first-layer GND plane 316.

  19A and 19B are diagrams showing the third layer. Specifically, FIG. 19B is a plan view, and FIG. 19A is a cross-sectional view taken along the line IX-IX in FIG. 19B. As shown in these drawings, the stress relaxation layer 330 is formed on the second layer described above. However, the stress relaxation layer 330 is formed to avoid the central portion of the connection portion 322b of the signal wiring 322 of the second layer. A signal wiring 332 is formed from the connection portion 322 b of the second layer to the stress relaxation layer 330. The signal wiring 332 includes a connection portion 332a connected to the second-layer connection portion 322b and another connection portion 332b. Further, a signal wiring 334 that is not electrically connected to the signal wiring 332 is formed in the stress relaxation layer 330. The signal wiring 334 has connection portions 334a and 334b. In addition, a GND plane 336 is formed so as to surround the signal wiring 332 and the signal wiring 334 and not to contact each other. The GND plane 336 is connected to a ground electrode (not shown) of the semiconductor chip 302 via the first-layer GND plane 316 and the second-layer GND plane 326.

  20A and 20B are diagrams showing the fourth layer. Specifically, FIG. 20B is a plan view, and FIG. 20A is a cross-sectional view taken along line XX of FIG. 20B. As shown in these drawings, a stress relaxation layer 340 is formed on the third layer described above. However, the stress relaxation layer 340 is formed so as to avoid the central portion of the connection portion 334b of the signal wiring 334 of the third layer. A connection portion 342 is formed on the connection portion 334b of the third layer, a base 344 made of copper (Cu) is formed on the connection portion 342, and a solder ball 348 is formed on the base 344. ing. Solder balls 348 serve as external electrodes. In addition, a GND plane 346 is formed so as to surround the connection portion 342 and not to contact the connection portion 342. The GND plane 346 is connected to a ground electrode (not shown) of the semiconductor chip 302 via the first-layer GND plane 316, the second-layer GND plane 326, and the third-layer GND plane 336.

Next, the conduction state in this embodiment will be described. The electrode 304 formed on the semiconductor chip 302 is connected to the first layer signal wiring 312, and the signal wiring 312 is connected to the second layer signal wiring 322. The signal wiring 322 is connected to the third-layer signal wiring 332 through the connection portion 322b, and the signal wiring 332 is connected to the second-layer signal wiring 324 through the connection portion 332b. The signal wiring 324 is connected to the third-layer signal wiring 334 through the connection portion 324b. A solder ball 348 is formed on the connection portion 334 b of the signal wiring 334 via the connection portion 342 and the base 344.

  In this way, the electrode 304 at an arbitrary position of the semiconductor chip to which signals are input or output and the solder ball 348 as an external electrode formed at an arbitrary position on the semiconductor chip are connected.

  Of course, the external electrodes may be arranged in a matrix as mentioned in other embodiments or base technologies.

  Further, the GND planes 316, 326, 336 and 346 of the first to fourth layers are all at the same ground potential.

  Therefore, according to the present embodiment, the wiring between the electrode 304 and the solder ball 348 is surrounded by the conductor having the ground potential via the insulator. That is, since the inner conductor is surrounded by the outer conductor having the ground potential via the insulator, the structure is the same as that of the coaxial cable. As a result, the signal is less susceptible to noise and an ideal transmission path can be obtained. For example, a semiconductor device as a CPU can operate at a high speed exceeding 1 GHz.

  In order to reduce the layer formation cost, any of the GND planes 316 and 346 formed in the first layer or the fourth layer may be omitted.

(Other embodiments)
The present invention is not limited to the above embodiment, and various modifications can be made. For example, in the above embodiment, the present invention is applied to a semiconductor device, but the present invention can be applied to various electronic components for surface mounting regardless of whether they are active components or passive components.

  FIG. 21 is a diagram showing an example in which the present invention is applied to an electronic component for surface mounting. The electronic component 400 shown in the figure is provided with electrodes 404 on both sides of the chip portion 402, and is, for example, a resistor, a capacitor, a coil, an oscillator, a filter, a temperature sensor, a thermistor, a varistor, a volume, or a fuse. Similar to the above-described embodiment, a wiring 408 is formed on the electrode 404 via a stress relaxation layer 406. A bump 410 is formed on the wiring 408.

  FIG. 22 is also a diagram showing an example in which the present invention is applied to an electronic component for surface mounting. The electrode 424 of the electronic component 420 is formed on the mounting side surface of the chip portion 422, and the stress relaxation layer 426. A wiring 428 is formed via the. Bumps 430 are formed on the wiring 428.

  In addition, since the manufacturing method of these electronic components 400 and 420 is the same as that of the said embodiment or base technology, description is abbreviate | omitted. In addition, the effect obtained by forming the stress relaxation layers 406 and 426 is the same as that of the above-described embodiment or the base technology.

Next, FIG. 23 is a diagram showing an example in which a protective layer is formed on a semiconductor device to which the present invention is applied. A semiconductor device 440 illustrated in FIG. 4 is obtained by forming a protective layer 442 on the semiconductor device illustrated in FIG. 4C, and a description of the semiconductor device 440 is omitted because it is the same as the semiconductor device illustrated in FIG. 4C except for the protective layer 442.

  In the semiconductor device 440, the protective layer 442 is formed on the surface opposite to the mounting side, that is, the back surface. By doing so, it is possible to prevent the back surface from being scratched.

  Furthermore, it is possible to prevent damage to the semiconductor chip itself due to cracks originating from scratches on the back surface.

  The protective layer 442 is preferably formed on the back surface of the wafer before being cut into individual semiconductor devices 440. Thus, the protective layer 442 can be formed on the plurality of semiconductor devices 440 at the same time. Specifically, it is preferable to form the protective layer 442 on the wafer after all the metal thin film forming steps are completed. By carrying out like this, a metal thin film formation process can be performed smoothly.

  The protective layer 442 is preferably made of a material that can withstand high temperatures in the reflow process of the semiconductor device 440. Specifically, it is preferable to withstand the melting temperature of the solder. The protective layer 442 is formed by applying a potting resin. Alternatively, the protective layer 442 may be formed by attaching a sticky or adhesive sheet. This sheet may be organic or inorganic.

  In this way, since the surface of the semiconductor device is covered with a substance other than silicon, the marking property is improved, for example.

  Next, FIG. 24 is a diagram showing an example in which a radiator is attached to a semiconductor device to which the present invention is applied. A semiconductor device 450 shown in the figure is obtained by attaching a heat radiator 452 to the semiconductor device shown in FIG. 4C. Except for the heat radiator 452, the semiconductor device 450 is the same as the semiconductor device shown in FIG.

  The heat radiator 452 is attached to the surface opposite to the mounting side, that is, the back surface of the semiconductor device 450 via a heat conductive adhesive 454. By doing so, heat dissipation is improved. The radiator 452 has a large number of fins 456 and is often formed of copper, copper alloy, aluminum nitride, or the like. In this example, the case with fins is taken as an example. However, even if a simple plate-like radiator (heat radiating plate) having no fins is attached, a corresponding heat radiation effect can be obtained. In this case, since it is a simple plate-like attachment, handling is easy and cost can be reduced.

  In the embodiment or the base technology, solder bumps or gold bumps are provided in advance on the semiconductor device side as external terminals, but as another example, a pedestal such as copper is used as it is without using solder bumps or gold bumps on the semiconductor device side. It may be used as an external terminal. In this case, it is necessary to provide solder in advance at the junction (land) of the mounting substrate (motherboard) on which the semiconductor device is mounted before mounting the semiconductor device.

  Moreover, it is preferable that the polyimide resin used in the said embodiment is black. By using the black polyimide resin as the stress relaxation layer, it is possible to avoid malfunction when the semiconductor chip receives light, and to improve light resistance and improve the reliability of the semiconductor device.

  FIG. 25 shows a circuit board 1000 on which an electronic component 1100 such as a semiconductor device manufactured by the method according to the above-described embodiment is mounted. As an electronic device including the circuit board 1000, a notebook personal computer 1200 is shown in FIG.

1A to 1E are views for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 2A to 2E are views for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 3A to 3D are views for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 4A to 4C are diagrams for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. FIG. 5 is a plan view showing a semiconductor device as a premise of the present invention. 6A to 6C are diagrams for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 7A to 7C are diagrams for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 8A to 8D are views for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 9A to 9D are views for explaining a method for manufacturing a semiconductor device which is a premise of the present invention. FIG. 10 is a diagram for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 11A to 11C are diagrams for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 12A to 12C are diagrams for explaining a semiconductor device manufacturing method which is a premise of the present invention. 13A to 13D are views for explaining a method of manufacturing a semiconductor device which is a premise of the present invention. 14A to 14D are diagrams showing the semiconductor device according to the first embodiment of the present invention. FIG. 15 is a diagram illustrating the semiconductor device according to the second embodiment. FIG. 16 is a diagram illustrating a semiconductor device according to the third embodiment. 17A and 17B are views for explaining a method for manufacturing a semiconductor device according to the third embodiment. 18A and 18B are views for explaining a method of manufacturing a semiconductor device according to the third embodiment. 19A and 19B are views for explaining a method of manufacturing a semiconductor device according to the third embodiment. 20A and 20B are views for explaining a method of manufacturing a semiconductor device according to the third embodiment. FIG. 21 is a diagram showing an example in which the present invention is applied to an electronic component for surface mounting. FIG. 22 is a diagram showing an example in which the present invention is applied to an electronic component for surface mounting. FIG. 23 is a diagram showing an example in which a protective layer is formed on a semiconductor device to which the present invention is applied. FIG. 24 is a diagram showing an example in which a radiator is attached to a semiconductor device to which the present invention is applied. FIG. 25 is a diagram showing a circuit board on which electronic parts manufactured by applying the method according to the present invention are mounted. FIG. 26 is a diagram showing an electronic apparatus including a circuit board on which an electronic component manufactured by applying the method according to the present invention is mounted.

Explanation of symbols

DESCRIPTION OF SYMBOLS 150 ... Semiconductor device, 152 ... Resin layer, 154 ... Wiring, 156 ... Elastic body, 157 ... Solder ball, 158 ... Electrode, 160 ... Semiconductor device, 164 ... First wiring, 167 ... Solder ball, 168 ... Second Wiring, 169 ... Electrode, 170 ... Semiconductor device, 174 ... First wiring, 176 ... Second resin layer, 177 ... Solder ball, 178 ... Second wiring, 180 ... Semiconductor device, 182 ... Electrode, 184 ... Wiring 186: Bump, 187 ... Stress relaxation layer, 190 ... Semiconductor device, 192 ... Aluminum pad, 194 ... Stress relaxation layer, 196 ... Solder ball, 200 ... Wiring, 300 ... Semiconductor device, 302 ... Semiconductor chip, 304 ... Electrode, 310 ... Stress relaxation layer, 312 ... Signal wiring,
316 ... Plane, 320 ... Stress relaxation layer, 322 ... Signal wiring, 324 ... Signal wiring, 326 ... Plane, 330 ... Stress relaxation layer, 332 ... Signal wiring, 334 ... Signal wiring, 336 ... Plane, 340 ... Stress relaxation layer, 342 ... connection part 344 ... pedestal,
346: Plane, 348 ... Solder ball, 350 ... Solder resist, 400 ... Electronic component, 402 ... Chip part, 406 ... Stress relaxation layer, 408 ... Wiring, 410 ... Bump, 420 ... Electronic component, 422 ... Chip part, 424 ... Electrode, 426 ... Stress relaxation layer, 428 ... Wiring, 430 ... Bump, 440 ... Semiconductor device, 442 ... Protective layer, 450 ... Semiconductor device, 452 ... Radiator, 454 ... Thermally conductive adhesive, 456 ... Fin

Claims (21)

Preparing a wafer on which electrodes are formed;
Providing a first stress relaxation layer on the wafer while avoiding at least a portion of the electrode;
Forming a first conductive portion from the electrode over the first stress relaxation layer;
Forming an external electrode connected to the first conductive portion above the first stress relaxation layer;
Cutting the wafer into individual pieces;
Have
A method of manufacturing a semiconductor device, wherein a structure for increasing stress relaxation is formed in at least one of a step of providing the first stress relaxation layer and a step of forming the first conductive portion. In the manufacturing method of the semiconductor device according to claim 1,
As a structure for increasing the stress relaxation, a method of manufacturing a semiconductor device, wherein a recess is formed on a surface of the first stress relaxation layer, and the first conductive portion is formed so as to pass over the recess. In the manufacturing method of the semiconductor device according to claim 1,
A method of manufacturing a semiconductor device, wherein, as the structure for increasing the stress relaxation, in the step of forming the first conductive portion, the first conductive portion is bent in a planar direction on the first stress relaxation layer. The method of manufacturing a semiconductor device according to claim 2.
A method for manufacturing a semiconductor device, comprising a step of filling an elastic body on the first conductive portion located in the recess. In the manufacturing method of the semiconductor device in any one of Claims 1-4,
A semiconductor device including a step of providing a second stress relaxation layer and a second conduction portion connected to the first conduction portion on the first stress relaxation layer in which the first conduction portion is formed. Manufacturing method. In the manufacturing method of the semiconductor device according to claim 5,
A method of manufacturing a semiconductor device, wherein at least one of the first conductive portion and the second conductive portion is formed in a planar shape having a planar extension larger than a thickness. In the manufacturing method of the semiconductor device in any one of Claims 1-4,
On the first stress relaxation layer on which the first conductive portion is formed, a second stress relaxation layer and a second conductive portion are provided,
A third stress relaxation layer and a third conduction portion are provided on the second stress relaxation layer on which the second conduction portion is formed,
A method of manufacturing a semiconductor device, wherein the second conductive portion is formed in a linear shape, and the first and third conductive portions are formed in a planar shape so as to have a larger planar extension than the second conductive portion. . In the manufacturing method of the semiconductor device in any one of Claims 1-4,
A method of manufacturing a semiconductor device, wherein a pair of wirings having a ground potential are formed in parallel so as to sandwich the first conductive portion. A semiconductor chip having electrodes;
A first stress relaxation layer provided on the semiconductor chip so as to avoid at least a part of the electrode;
A first conductive portion formed from the electrode to the first stress relaxation layer;
An external electrode formed on the first conductive portion located above the first stress relaxation layer, and
The first stress relaxation layer is formed so as to have a depression on the surface, and the first conductive part is formed over the depression. The semiconductor device according to claim 9.
A semiconductor device in which an elastic body is provided so as to fill the inside of the recess on the first conducting portion located in the recess. The semiconductor device according to claim 9.
The first conductive portion is a semiconductor device formed by bending on the first stress relaxation layer. The semiconductor device according to claim 11.
The first conductive portion is a semiconductor device formed in a loose shape. The semiconductor device according to claim 9.
A semiconductor device having a second stress relaxation layer and a second conduction portion connected to the first conduction portion on the first stress relaxation layer in which the first conduction portion is formed. The semiconductor device according to claim 13.
One of the two conducting parts including the first conducting part and the second conducting part is formed in a linear shape, and the other is formed in a planar shape having a planar extension wider than the linear conducting part. Semiconductor device. The semiconductor device according to claim 14.
The semiconductor device in which the planar conductive portion is set to a ground potential, and a signal is input to the linear conductive portion. The semiconductor device according to claim 9.
A second stress relaxation layer and a second conduction portion provided on the first stress relaxation layer in which the first conduction portion is formed;
A third stress relaxation layer and a third conduction portion provided on the second stress relaxation layer in which the second conduction portion is formed;
Have
The second conductive portion is formed in a linear shape, and the first and third conductive portions are formed in a planar shape so as to have a larger planar extension than the second conductive portion. The semiconductor device according to claim 9.
A semiconductor device having a pair of wirings that are formed in parallel so as to sandwich the first conductive portion and are set to a ground potential. The semiconductor device according to claim 9.
A semiconductor device having a protective film on a side surface opposite to a surface having the electrode of the semiconductor chip. The semiconductor device according to claim 9.
The semiconductor device which has a heat radiator on the opposite side to the surface which has the said electrode of the said semiconductor chip.   A circuit board on which the semiconductor device according to claim 9 is mounted.   An electronic device having the circuit board according to claim 20.

Images