JP2018041989A - Semiconductor device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a semiconductor device capable of preventing warpage of a substrate.SOLUTION: A semiconductor device 100 comprises: a substrate 8 having a first principal surface and a second principal surface facing an opposite side to the first principal surface; a first nonvolatile semiconductor memory 10 and a second nonvolatile semiconductor memory 10 which are mounted on the first principal surface of the substrate 8; and an adhesive part 31 which is filled in a gap between the first nonvolatile semiconductor memory 10 and the second nonvolatile semiconductor memory 10 and generates force in directions in which the first nonvolatile semiconductor memory 10 and the second nonvolatile semiconductor memory 10 are drawn while exposing surfaces of the first nonvolatile semiconductor memory 10 and the second nonvolatile semiconductor memory 10.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a semiconductor device.

  Conventionally, a semiconductor device is used in which a nonvolatile semiconductor memory element such as a NAND flash memory is mounted on a substrate on which a connector is formed. In addition to the nonvolatile semiconductor memory element, the semiconductor device includes a volatile semiconductor memory element and a controller that controls the nonvolatile semiconductor memory element and the volatile semiconductor element.

  In such a semiconductor device, the shape and size of the substrate may be restricted in accordance with the usage environment, standards, and the like. For example, a substrate that has a rectangular shape in plan view may be used. In addition, due to recent demands for miniaturization of semiconductor devices, substrates tend to be thinner. When such a thin rectangular substrate is used, it is required to suppress warping of the substrate.

JP 2010-79445 A

  An object of this invention is to provide the semiconductor device which can suppress the curvature of a board | substrate.

  According to one aspect of the present invention, a semiconductor device includes a first main surface, a substrate having a second main surface facing away from the first main surface, and a first main surface of the substrate. A first non-volatile semiconductor memory and a second non-volatile semiconductor memory. The semiconductor device is filled in a gap between the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory while exposing the surfaces of the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory, An adhesive portion that generates a force in a direction in which the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory are attracted to each other.

FIG. 1 is a block diagram illustrating a configuration example of the semiconductor device according to the first embodiment. FIG. 2 is a plan view showing a schematic configuration of the semiconductor device. FIG. 3 is a side view of the semiconductor device. FIG. 4 is a diagram showing a layer structure of the substrate. FIG. 5 is a diagram showing the wiring density of each layer of the substrate. FIG. 6 is a diagram showing a wiring pattern formed on the back layer (eighth layer) of the substrate. FIG. 7 is a diagram showing the wiring density of each layer of a substrate as a comparative example. FIG. 8 is a diagram for explaining the line width and interval of the wiring pattern formed on the back layer (eighth layer) of the substrate. FIG. 9 is a diagram illustrating an adhesive portion filled in a gap of the NAND memory. FIG. 10 is a diagram showing slits formed in the seventh layer of the substrate. FIG. 11 is a diagram illustrating a layer configuration of a substrate included in the semiconductor device according to the second embodiment. FIG. 12 is an external perspective view of a holding member used in the method for transporting a semiconductor device according to the third embodiment. FIG. 13 is a cross-sectional view showing a state where the holding member shown in FIG. 12 is housed in a box. FIG. 14 is a front view of a holding member according to a modification of the third embodiment. FIG. 15 is a diagram illustrating a state in which the movable portion of the holding member illustrated in FIG. 14 is opened.

  Hereinafter, a semiconductor device according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited to these embodiments.

(First embodiment)
FIG. 1 is a block diagram illustrating a configuration example of the semiconductor device according to the first embodiment. The semiconductor device 100 is connected to a host device (hereinafter abbreviated as a host) 1 such as a personal computer or a CPU core via a memory connection interface such as a SATA interface (ATA I / F) 2 and functions as an external memory of the host 1. To do. Examples of the host 1 include a CPU of a personal computer, a CPU of an imaging apparatus such as a still camera and a video camera. Further, the semiconductor device 100 can transmit and receive data to and from the debugging device 200 via the communication interface 3 such as an RS232C interface (RS232C I / F).

  A semiconductor device 100 includes a NAND flash memory (hereinafter abbreviated as “NAND memory”) 10 as a nonvolatile semiconductor memory element, a drive control circuit 4 as a controller, and a volatile semiconductor capable of a higher-speed storage operation than the NAND memory 10. It includes a DRAM 20, which is a storage element, a power supply circuit 5, a status display LED 6, and a temperature sensor 7 for detecting the temperature inside the drive. The temperature sensor 7 measures the temperature of the NAND memory 10 directly or indirectly, for example. The drive control circuit 4 limits the writing of information to the NAND memory 10 when the measurement result by the temperature sensor 7 is equal to or higher than a certain temperature, and suppresses further temperature rise.

  The power supply circuit 5 generates a plurality of different internal DC power supply voltages from an external DC power supply supplied from a power supply circuit on the host 1 side, and supplies these internal DC power supply voltages to each circuit in the semiconductor device 100. Further, the power supply circuit 5 detects the rise of the external power supply, generates a power-on reset signal, and supplies it to the drive control circuit 4.

  FIG. 2 is a plan view showing a schematic configuration of the semiconductor device 100. FIG. 3 is a side view of the semiconductor device 100. The power supply circuit 5, DRAM 20, drive control circuit 4, and NAND memory 10 are mounted on a substrate 8 on which a wiring pattern is formed. The substrate 8 has a substantially rectangular shape in plan view. A connector 9 that is connected to the host 1 and functions as the above-described SATA interface 2 and communication interface 3 is provided on one short side of the substrate 8 having a substantially rectangular shape. The connector 9 functions as a power input unit that supplies power supplied from the host 1 to the power circuit 5. The connector 9 is, for example, an LIF connector. The connector 9 is formed with a slit 9a at a position shifted from the center position along the short direction of the substrate 8 so as to be fitted with a protrusion (not shown) provided on the host 1 side. It has become. This can prevent the semiconductor device 100 from being attached upside down.

  The substrate 8 has a multilayer structure formed by overlapping synthetic resins, and has, for example, an eight-layer structure. Note that the number of layers of the substrate 8 is not limited to eight. FIG. 4 is a diagram showing a layer structure of the substrate 8. On the substrate 8, wiring patterns are formed as wiring layers 8b in various shapes on the surface or inner layer of each layer (insulating film 8a) made of synthetic resin. The wiring pattern is made of, for example, copper. The power supply circuit 5, DRAM 20, drive control circuit 4, and NAND memory 10 mounted on the substrate 8 are electrically connected to each other through a wiring pattern formed on the substrate 8. Further, the front surface (first layer side) and the back surface (eighth layer side) of the substrate are covered with a solder resist 8c as a protective film.

  FIG. 5 is a diagram showing the wiring density of each layer of the substrate 8. Here, the first to fourth layers formed on the surface layer side from the center line 30 (see also FIG. 4) of the layer structure of the substrate 8 are referred to as upper layers, and are formed on the back layer side from the center line 30. The formed fifth to eighth layers are called lower layers.

  As shown in FIG. 5, the wiring layer 8b formed in each layer of the substrate 8 functions as a signal layer for transmitting / receiving signals, and a plane layer serving as a ground or power supply line. Then, the wiring density of the wiring pattern formed in each layer, that is, the ratio of the wiring layer to the surface area of the substrate 8 is configured as shown in FIG.

  In the present embodiment, the eighth layer functioning as the ground is not a plane layer but a net-like wiring layer, so that the wiring density is suppressed to 30 to 60%. Here, the wiring density in the entire upper layer of the substrate 8 is about 60%. Therefore, by forming the wiring pattern with the wiring density of the eighth layer being about 30%, the wiring density of the entire lower layer can be made about 60%, and the wiring density of the entire upper layer and the wiring density of the entire lower layer can be reduced. Can be approximately equal. Note that the wiring density of the eighth layer may be adjusted to be approximately equal to the wiring density of the entire upper layer by adjusting in the range of about 30 to 60%.

  FIG. 6 is a diagram showing a wiring pattern formed on the back surface layer (eighth layer) of the substrate 8. As shown in FIG. 6, a wiring pattern is formed in a net pattern on the back surface layer (eighth layer) of the substrate 8. In this way, the eighth layer of the substrate 8 is a net-like wiring layer, so that the wiring density is kept lower than that formed as a plane layer.

  The wiring layer formed on the back layer is also required to have a function as a shield layer that reduces the influence of noise leaking from the semiconductor device 100 and affecting other devices. FIG. 8 is a diagram for explaining the line width and interval of the wiring pattern formed on the back layer (eighth layer) of the substrate. As shown in FIG. 8, a net-like wiring having a line width L of 0.3 mm and a line interval S of 0.9 mm is formed on the eighth layer of the substrate 8. In the net-like wiring formed in this way, the opening width W is 0.9 × √2 = 1.27 mm.

For example, the shielding effect against high frequency noise such as 3 GHz SATA fundamental wave is as follows. First, the half wavelength (λ / 2) of the second harmonic of the SATA fundamental wave is calculated from C = f × λ × √ε. Here, C is the speed of light, which is 3.0 × 10 ⁸ m / s. f is the frequency of the second harmonic, which is 6.0 × 10 ⁹ Hz. ε is a relative dielectric constant, which is 4.6.

  According to the above conditions, λ is 23.3 mm, and ½ wavelength (λ / 2) is 11.7 mm. That is, the ½ wavelength (λ / 2) is about 10 times the opening width W (1.27 mm). Further, since λ / 20 = 1.2 mm and substantially equal to the opening width W, the shielding effect is about −20 dB.

  FIG. 9 is a diagram illustrating an adhesive portion filled in a gap of the NAND memory 10. As shown in FIG. 9, a gap between the NAND memory 10 and the substrate 8 is filled with an adhesive portion 31 made of a synthetic resin material, thereby bonding the NAND memory 10 and the substrate 8 together. Further, a part of the bonding portion 31 protrudes from the gap between the NAND memory 10 and the substrate 8. The protruding portion is filled in a gap between the NAND memories 10 arranged along the longitudinal direction of the substrate 8. Therefore, the bonding portion 31 bonds the NAND memories 10 to each other on the side surface. The bonding portion 31 protrudes to the extent that the height of the NAND memory 10 is not exceeded, and the surface of the NAND memory 10 is exposed. In FIG. 9, the bonding portion 31 is filled up to the middle of the height of the NAND memory 10. However, the bonding portion 31 may be lower than this, and if the bonding portion 31 is in contact with adjacent NAND memories 10. Good. Of course, the bonding portion 31 may be filled between the NAND memories 10 so as to be higher than the height shown in FIG.

  FIG. 10 is a view showing slits formed in the seventh layer of the substrate 8. In FIG. 10, the state which looked at the board | substrate 8 from the back surface layer side is shown, and the 8th layer is abbreviate | omitted and shown. Further, the NAND memory 10 mounted on the surface layer side is indicated by a broken line. A plane layer is formed as a wiring layer on the seventh layer of the substrate 8. As shown in FIG. 10, the seventh layer of the substrate 8 is provided with a slit 32 (a portion where the wiring layer is not formed) in a part of the seventh layer as a plane layer while forming a wiring pattern over almost the entire area of the seventh layer. ing. The slit 32 is provided in a portion of the wiring pattern formed over substantially the entire area of the seventh layer that faces the gap of the NAND memory 10.

  FIG. 7 is a diagram showing the wiring density of each layer of a substrate as a comparative example. As shown in the comparative example of FIG. 7, in the conventional substrate, the wiring density is about 90% by using the eighth layer as a plane layer. Therefore, the lower layer wiring density is about 75%, and the difference from the upper layer wiring density (about 60%) is large. Because the wiring density is different, the ratio of the insulating film 8a (synthetic resin) and the wiring portion (copper) in the entire upper layer of the substrate 8 is different from the ratio of the synthetic resin and copper in the entire lower layer of the substrate 8. Become. Thereby, the thermal expansion coefficient also differs between the upper layer and the lower layer of the substrate 8. Due to this difference in thermal expansion coefficient, a warp that tends to have a convex shape on the surface layer side along the longitudinal direction of the substrate 8 (a convex shape upward in FIG. 3) easily occurs along with the temperature change of the substrate 8. . Such a temperature change is likely to occur during the manufacturing process of the semiconductor device 100. Further, due to the recent demand for miniaturization of semiconductor devices, the substrate 8 tends to be thinned, and such warpage is likely to occur.

  On the other hand, in the present embodiment, the wiring density of the eighth layer is adjusted in a range of about 30 to 60%, and the wiring density of the entire upper layer is made substantially equal to the wiring density of the entire lower layer. Are also approximately equal. Therefore, it is possible to prevent the substrate 8 from warping. In addition, since the wiring density is adjusted in the eighth layer farthest from the center line 30 (see also FIG. 4), a larger moment for suppressing warpage can be generated.

  In addition, since the wiring density is adjusted in the eighth layer of the substrate 8, the wiring design becomes easier and the cost can be reduced compared with the case where the wiring density is adjusted in a layer having a limited wiring layout such as the signal layer. Can be suppressed.

  Further, since the bonding portion 31 is filled in the gap between the adjacent NAND memories 10, a force for pulling the NAND memories 10 as indicated by the arrow X is generated by the bonding force of the bonding portion 31. Since the force that pulls the NAND memories 10 toward each other is a force that opposes the force that warps the substrate 8 so as to have a convex shape, the occurrence of warpage of the substrate 8 can be suppressed.

  Further, among the wiring patterns formed over substantially the entire seventh layer of the substrate 8, the wiring patterns are weakened at the slits 32 because the wiring patterns are provided in the portions facing the gaps of the NAND memory 10. Therefore, the force that opposes the force (see also the arrow X in FIG. 9) generated by filling the bonding portion 31 in the gap between the NAND memories 10 is weakened, and the warpage of the substrate 8 is more effectively suppressed. be able to.

  In the present embodiment, in order to adjust the wiring density of the entire lower layer of the substrate 8, the eighth wiring layer is a net-like wiring layer. However, the present invention is not limited to this. For example, a wiring layer is formed on a line. May be. Further, the wiring density of the lower layer as a whole may be adjusted by adjusting the wiring density of the lower layers other than the eighth layer, that is, the wiring layers from the fifth layer to the seventh layer. Of course, the wiring density of all the layers from the fifth layer to the eighth layer may be adjusted to adjust the wiring density of the entire lower layer.

  The layer in which the slit 32 is formed is not limited to the seventh layer. In the lower layer, slits may be formed in layers other than the seventh layer, that is, in the fifth to sixth and eighth layers.

(Second Embodiment)
FIG. 11 is a diagram illustrating a layer configuration of a substrate included in the semiconductor device according to the second embodiment. In the present embodiment, the outermost layer is provided as the ninth layer outside the eighth layer of the substrate 8. The entire outermost layer is covered with copper foil to form a shield layer. Thus, by covering the entire outermost layer with the copper foil, noise leakage from the semiconductor device can be more reliably prevented. In addition, it is good also as a shield layer by covering the whole region of a layer inside the 9th layer with copper foil.

(Third embodiment)
FIG. 12 is an external perspective view of a holding member used in the method for transporting a semiconductor device according to the third embodiment. FIG. 13 is a cross-sectional view showing a state where the holding member shown in FIG. 12 is housed in a box. In the present embodiment, the semiconductor device 100 is packed and transported by the holding member 50. The holding member 50 suppresses the warp of the substrate 8 due to a change with time.

  The holding member 50 includes a sandwiching part 51 and a connecting part 52. Two holding portions 51 are provided for one holding member 50. The sandwiching portion 51 sandwiches and holds a portion along the longitudinal direction of the substrate 8. In order to hold the substrate 8 from both sides, two holding portions 51 are provided for one holding member 50. The sandwiching portion 51 is formed in a U-shaped cross section, and sandwiches a portion along the longitudinal direction of the substrate 8 in the gap. The clamping part 51 suppresses the warp of the substrate 8 against a force that causes a warp along the longitudinal direction of the substrate 8 with a change with time. Therefore, the sandwiching portion 51 is formed with a strength that can counteract the force to warp the substrate 8.

  Further, in order to suppress the warpage of the substrate 8, it is preferable that the holding portion 51 is in close contact with the substrate 8 while the substrate 8 is held. For example, the gap formed in the sandwiching portion 51 may be formed to be slightly narrower than the thickness of the substrate 8, and the substrate 8 may be inserted into the sandwiching portion 51 while expanding the gap. Further, it may be formed so as to have a width substantially equal to or slightly wider than the substrate 8, and the substrate 8 can be easily inserted into the gap.

  The connecting part 52 connects the two clamping parts 51. Thereby, the holding member 50 can be integrated. As shown in FIG. 13, the connecting portion 52 is a buffer material that maintains a distance between the semiconductor devices 100 when the plurality of semiconductor devices 100 are accommodated in a box and relaxes an impact applied to the semiconductor device 100 during transportation. Also works.

  In addition, the holding | maintenance part 53 is formed in the clamping part 51, respectively. The space | interval holding | maintenance part 53 is formed so that it may extend in the opposite side to the side in which the connection part 52 was provided with respect to the clamping part 51. FIG. As shown in FIG. 13, the gap holding unit 53 holds a gap between the semiconductor devices 100 when a plurality of semiconductor devices 100 are housed in a box, and cushions the shock applied to the semiconductor device 100 during transportation. Function as.

  In the present embodiment, the sandwiching unit 51 is described as sandwiching the substrate 8. However, for example, an electronic component (not shown) such as a resistor or a capacitor, the NAND memory 10, or the like is mounted on the substrate 8. . Therefore, when an electronic component or the like is mounted on the peripheral portion of the substrate 8, the holding portion 51 may be formed with a width that allows the substrate 8 and the electronic component or the like to be sandwiched together.

  FIG. 14 is a front view of a holding member 50 according to a modification of the third embodiment. In this modification, the clamping part 51 is configured to include a fixed part 51a and a movable part 51b. The fixed portion 51a and the movable portion 51b are rotatably connected at a portion corresponding to the bottom of the gap formed in the holding portion 51, and can open and close the movable portion 51b.

  A closing portion 55 is formed in each of the movable portions 51b. As shown in FIG. 14, the closing portions 55 are hooked to each other when the movable portion 51 b is closed, and keep the movable portion 51 b closed. Further, since the movable portion 51b is in a closed state, the width of the gap formed in the holding portion 51 is kept constant.

  FIG. 15 is a view showing a state in which the movable portion 51b of the holding member 50 shown in FIG. 14 is opened. As shown in FIG. 15, the gap of the clamping part 51 can be expanded by opening the movable part 51b. If the semiconductor device 100 is placed on the fixed portion 51a and the movable portion 51b is closed while the gap of the holding portion 51 is widened, the semiconductor device 100 is inserted into the holding portion 51 while pushing the gap wide. The semiconductor device 100 can be easily held by the holding member 50.

  1 host, 2 SATA interface (ATA / IF), 3 communication interface, 4 drive control circuit (controller), 5 power supply circuit, 7 temperature sensor, 8 substrate, 8a insulating film, 8b wiring layer, 9 connector, 9a slit, 10 NAND memory (NAND flash memory, non-volatile semiconductor memory element), 20 DRAM (volatile semiconductor memory element), 30 center line, 31 adhesive part, 32 slit, 50 holding member, 51 clamping part, 52 connecting part, 53 interval Holding unit, 55 closing unit, 100 semiconductor device, 200 debugging device.

Claims (5)

A substrate having a first main surface and a second main surface facing away from the first main surface;
A first nonvolatile semiconductor memory and a second nonvolatile semiconductor memory mounted on the first main surface of the substrate;
The gap between the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory is filled while exposing the surfaces of the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory. An adhesive portion that generates force in a direction in which the nonvolatile semiconductor memory of 1 and the second nonvolatile semiconductor memory are attracted;
A semiconductor device comprising: The substrate includes a wiring layer provided on the first main surface, a wiring layer provided on the second main surface, a plurality of wiring layers formed as inner layers, and a plurality of wiring layers provided between these wiring layers, respectively. A gap between the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory in the wiring layer formed on the second main surface side of the center line of the layer structure of the substrate. The semiconductor device according to claim 1, wherein a slit is formed in at least a part of a portion facing the substrate. The substrate has a substantially rectangular shape in plan view,
A connector provided on a side along the short direction of the substrate for connecting to an external device, and the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory electrically connected to the connector and controlling the first nonvolatile semiconductor memory And a controller mounted on the first main surface of the substrate,
3. The semiconductor device according to claim 1, wherein the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory are provided on a side opposite to the connector when viewed from the position of the controller in a plan view. .   An internal voltage is generated based on a power source mounted on the substrate and supplied from the external device via the connector, and the generated internal voltage is used as the first nonvolatile semiconductor memory and the second nonvolatile memory. The semiconductor device according to claim 3, further comprising a power supply circuit configured to supply the semiconductor memory.   The bonding portion is also filled in a gap between the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory and the substrate, and the first nonvolatile semiconductor memory and the second nonvolatile semiconductor memory The semiconductor device according to claim 1, wherein the surface is exposed only in the gap.

Images