Semiconductor device and method of manufacturing the same

Abstract

A semiconductor device includes an interconnect member, a first semiconductor chip, a second semiconductor chip, a resin layer, an inorganic insulating layer, and a through electrode. The first semiconductor chip is mounted in a face-down manner on the interconnect member. The resin layer covers the side surface of the first semiconductor chip. This inorganic insulating layer is in contact with the back surface of the first semiconductor chip, and directly covers the back surface. Also, the inorganic insulating layer extends over the resin layer. The through electrode penetrates the inorganic insulating layer and the semiconductor substrate of the first semiconductor chip. The second semiconductor chip is mounted in a face-down manner on the inorganic insulating layer that covers the back surface of the first semiconductor chip in the uppermost layer.

Description

[0001] This application is based on Japanese patent application No. 2005-349794, the content of which is incorporated hereinto by reference. BACKGROUND [0002] 1. Technical Field [0003] The present invention relates to a semiconductor device and a method of manufacturing the same. [0004] 2. Related Art [0005] As a conventional method of manufacturing a semiconductor device, there is a method disclosed in Japanese Laid-Open patent publication No. 2002-343904 (patent document 1), for example. This document discloses a method of manufacturing a multiple-chip type semiconductor device in which a plurality of semiconductor chips are stacked. [0006] According to the manufacturing method disclosed in the patent document 1, first, a printed substrate is prepared in which an electrode post is formed on one surface, and a connection electrode is formed on the other surface. Next, a semiconductor chip is mounted in a face-down manner on the surface of the printed substrate on which the electrod...

AI Technical Summary

Benefits of technology

[0018] In this manufacturing method, electrical connection between the first semiconductor chip and another semiconductor chip is established by the through electrode. This reduces the path length of the signals transmitted between the two chips as compared with the case in which the electrical connection between these two chips is established by an interconnect that detours through the outside of the chips. For this reason, a semiconductor device suitable for high-speed operation can be obtained.

[0019] Further, an inorganic insulating layer is formed that directly covers the back surface of the first semiconductor chip. This inorganic insulating layer prevents the device region (the region where semiconductor elements such as transistors are to be formed) of the semiconductor chip from being contaminated with metals. For this reason, the problem of metal contamination can be prevented from occurring even if the semiconductor chip is thinned, unlike the case in which the back surface of the semiconductor chip is stripped bare.

[0022] In this semiconductor device, electrical connection between the first semiconductor chip and another semiconductor chip is established by the through electrode. This reduces the path length of the signals transmitted between the two chips as compared with the case in which the electrical connection between these two chips is established by an interconnect that detours through the outside of the chips. For this reason, a semiconductor device suitable for high-speed operation can be realized.

[0023] Further, an inorganic insulating layer is provided that directly covers the back surface of the first semiconductor chip. This inorganic insulating layer prevents the device region of the semiconductor chip from being contaminated with metals. For this reason, the problem of metal contamination can be prevented from occurring even if the semiconductor chip is thinned, unlike the case in which the back surface of the semiconductor chip is stripped bare.

Problems solved by technology

However, according to the manufacturing methods disclosed in the patent documents 1, 2, it is difficult to obtain a semiconductor device suitable for high-speed operation.

When the interconnect length is large, there will be a delay in the signals, so that it is difficult to apply this manufacturing method to a semiconductor device that requires a high-speed operation.

For example, when considering the connection to a DDR2 (Double Data Rate 2) memory, even an interconnect length of about 2 to 3 mm could be a problem.

Moreover, the manufacturing methods disclosed in the patent documents 1-4 are all unsuitable for manufacturing a semiconductor device on which a thin type semiconductor chip is mounted.

The reason lies in that, with regard to the patent document 1, since the back surface of the semiconductor chip is stripped bare, a problem of metal contamination will occur when the thickness of the chip is reduced.

Such metal contamination will give adverse effects on the semiconductor device.

When the chip is extremely thin (for example, less than 50 μm), the chip will be damaged even with a little force, so that it is difficult to handle a single chip.

Also, due to the stress between the silicon and the insulating layer, the chip will be greatly warped when a free surface is given to the chip, thereby making the handling all the more difficult.

Images