JP2009194357A - Semiconductor device and method of manufacturing the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a semiconductor device having a structure capable of easily bonding a conductive ball to a pad whose main constituent is aluminum, and a method of manufacturing the same. <P>SOLUTION: The semiconductor device includes a pad 12 whose main constituent is aluminum and which is for electrically connecting a semiconductor element formed in a semiconductor substrate 11 to the outside, a sintered conductive layer 13 formed on the pad 12 whose main constituent is aluminum, and a conductive ball 14 welded on the sintered conductive layer 13. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a semiconductor device and a manufacturing method thereof.

Conventionally, after a semiconductor chip is die-bonded to a lead frame, the semiconductor element and the lead frame are electrically connected by wire bonding.
In recent years, in response to high integration, high functionality, and miniaturization of semiconductor devices, flip chip connection using bumps, which is superior in terms of high-speed signal processing over wire bonding, has been performed.

  As a method of forming a bump on a bonding pad of a semiconductor chip, using a resist having an opening at a position corresponding to the bonding pad as a mask, solder is plated on the bonding pad and reflowed to form a ball-shaped solder bump. A method is known in which a solder paste containing granular solder is applied on a bonding pad by a printing method and reflowed to form ball-shaped solder bumps.

  However, when forming solder bumps on the aluminum pads of the semiconductor chip, there is a problem that it is difficult to directly bond the aluminum pads and solder balls because aluminum has poor wettability with solder.

  On the other hand, a method is known in which metal solder bumps are formed through a metal film called UBM (Under Bump Metal) in order to improve the adhesion between the connection terminals and the metal balls (see, for example, Patent Document 1). .)

  In the solder bump forming method disclosed in Patent Document 1, an electroless nickel alloy plating film and an electroless gold plating film are formed in this order on the surface of a conductor, and solder balls not containing lead are deposited thereon.

However, the bump forming method disclosed in Patent Document 1 has a problem that the UBM is formed by electroless plating, so that the process becomes complicated, and high productivity and cost reduction are difficult.
JP 2002-118135 A

  The present invention provides a semiconductor device having a structure in which a conductive ball can be easily bonded to a pad containing aluminum as a main component, and a method for manufacturing the same.

  A semiconductor device according to one embodiment of the present invention is formed over a pad containing aluminum as a main component for electrically connecting a semiconductor element formed over a semiconductor substrate to the outside, and the pad containing the aluminum as a main component. It comprises a sintered conductive layer and a conductive ball welded on the sintered conductive layer.

  A method for manufacturing a semiconductor device of one embodiment of the present invention includes a step of forming a pad mainly composed of aluminum for electrically connecting the semiconductor element to the outside on a semiconductor substrate on which the semiconductor element is formed; Using a first mask having a first opening at a position corresponding to a pad containing aluminum as a main component, and selectively applying a conductive paste on the pad containing aluminum as a main component; Using a second mask having a second opening at a position corresponding to a pad as a main component, and selectively placing a conductive ball having a smaller size than the second opening on the conductive paste; Heat-treating the conductive paste to form a sintered conductive layer, melting the conductive balls and welding them onto the sintered conductive layer, and using the aluminum as a main component. Through the sintered Yuishirube conductive layer on the pad is characterized by comprising a step of bonding the conductive balls.

  According to the present invention, a semiconductor device having a structure in which a conductive ball can be easily bonded to a pad containing aluminum as a main component and a method for manufacturing the same are obtained.

  Embodiments of the present invention are described below with reference to the drawings.

  A semiconductor device according to Embodiment 1 of the present invention will be described with reference to FIG. 1A and 1B are diagrams showing a main part of a semiconductor device according to Embodiment 1 of the present invention. FIG. 1A is a cross-sectional view thereof, and FIG. 1B is an enlarged cross-sectional view of the main part of FIG. is there.

  As shown in FIG. 1A, a semiconductor device 10 of this embodiment includes a pad mainly composed of aluminum for electrically connecting a semiconductor element (not shown) formed on a semiconductor substrate 11 to the outside. 12, a sintered conductive layer 13 formed on a pad 12 containing aluminum as a main component, a solder ball (conductive ball) 14 welded on the sintered conductive layer 13, and a pad containing aluminum as a main component 12 is provided with a protective film 15 that protects 12, for example, a polyimide film.

  As shown in FIG. 1B, the pad 12 mainly composed of aluminum reacts with the metal contained in the sintered conductive layer 13 between the pad 12 mainly composed of aluminum and the sintered conductive layer 13. The first alloy layer 16 is formed, and the second alloy layer 17 in which the metal contained in the sintered conductive layer 13 reacts with the solder ball 14 is formed between the sintered conductive layer 13 and the solder ball 14. Yes.

Here, the pad 12 composed mainly of aluminum contains a small amount of impurities such as silicon (Si) or copper (Cu) for the purpose of improving the conductivity and strength of the pure aluminum pad and film. Means an aluminum pad.
Hereinafter, the pad 12 mainly composed of aluminum is simply referred to as an aluminum pad 12.

The sintered conductive layer 13 is a conductive cured layer formed by thermally curing a conductive paste containing tin (Sn) and silver (Ag) as main components and added with a small amount of copper (Cu).
Here, a small amount of copper (Cu) is added to increase the elasticity of the sintered conductive layer 13. Moreover, that the main component is tin (Sn) and silver (Ag) means that the total component of Sn and Ag is 50% wt or more.

The solder ball 14 is an alloy mainly composed of at least two selected from tin (Sn), antimony (Sb), copper (Cu), gold (Au), and silver (Ag). Contains 1% to 1% by weight of phosphorus (P).
The solder ball 14 has a size of about 100 to 300 μmφ, and is so-called lead (Pb) free solder.

Since a small amount of phosphorus (P) has a property of being oxidized and sublimated to phosphorus pentoxide (P ₂ O ₅ ) when the solder ball 14 is melted, the solder ball 14 is prevented from being oxidized. In addition, it is added to activate.

  Next, a method for manufacturing the semiconductor device 10 will be described with reference to FIGS. FIG. 2 is a flowchart showing the manufacturing process of the semiconductor device 10, and FIGS. 3 and 4 are cross-sectional views showing the main parts of the manufacturing process of the semiconductor device 10.

  As shown in FIG. 2, a semiconductor element, for example, a discrete semiconductor element is formed on a semiconductor wafer 11 by a known method, and an aluminum pad 12 for electrically connecting the semiconductor element to the outside is formed.

Next, as shown in FIG. 3A, the surface of the aluminum pad 12 is subjected to plasma treatment using argon (Ar) or oxygen (O ₂ ) to contaminate the surface (attached organic matter, surface oxide film). It is removed and the surface of the aluminum pad 12 is activated.
For example, the surface of the aluminum pad 12 is etched by about 10 to 30 nm using an RIE (Reactive Ion Etching) apparatus.

Next, as shown in FIG. 3B, a conductive paste 22 is selectively applied onto the aluminum pad 12 using a first mask 21 having a first opening 20 at a position corresponding to the aluminum pad 12. To do.
For example, the conductive paste 22 is imprinted in the first opening 20 by the roller 23 using the screen on which the resist film having the first opening 20 is formed as a mask by the screen printing method, and the conductive paste is formed on the aluminum pad 12. 22 is applied so that the thickness is about several tens of microns.

Next, as shown in FIG. 3C, on the conductive paste 22 applied on the aluminum pad 12, the second mask 25 having the second opening 24 at a position corresponding to the aluminum pad 12 is used. The solder ball 14 having a size smaller than that of the second opening 24 is selectively placed.
For example, a stainless steel second mask 25 having a second opening 24 is disposed in close proximity via a spacer 26 on the aluminum pad 12 of the semiconductor substrate 11, and the solder ball 14 is moved to the second mask while being vibrated. One solder ball 14 is thrown into one second opening 24 while being uniformly dispersed on the second opening 24.

  Here, it is desirable to inspect the arrangement state of the solder balls 14 by image processing after sweeping out the excessive solder balls 14 and removing the second mask 25.

Next, as shown in FIG. 4A, high temperature reflow (320 to 550 ° C. × 60 to 300 sec) is performed, and the solder balls 14 are joined to the aluminum pads 12 via the sintered conductive layer 13.
By heating the heater 27, the conductive paste 22 is thermally cured to form the sintered conductive layer 13 that is a conductive cured layer on the aluminum pad 12, and the solder balls 14 are melted to form on the sintered conductive layer 13. Solder balls 14 are welded.
FIG. 4C is a SEM (Secondary Electron Microscopy) photograph showing the solder ball 14 bonded to the aluminum pad 12 through the sintered conductive layer 13.

  At this time, between the aluminum pad 12 and the sintered conductive layer 13, a first alloy layer 16 in which Al, Sn, and Ag react by solid diffusion is formed, and the sintered conductive layer 13 and the solder balls 14 In the meantime, a second alloy layer 17 in which Sn, Ag and solder are diffused and reacted is formed.

  Next, as shown in FIG. 5, using the knife edge 28, the solder ball 14 welded to the sintered conductive layer 13 is pressed from the side, and the load when the solder ball 14 peels from the sintered conductive layer 13 is determined. When the shear strength of the solder ball 14 was measured, the value was inferior to that obtained when a metal plating film was used as a conventional UBM.

  The shear strength of the solder ball 14 depends on the high temperature reflow conditions. In general, the shear strength tends to increase as the reflow temperature increases, but it is easy to obtain a value of about 0.7 to 5 N (0.07 to 0.51 kgf).

  A longitudinal section from the solder ball 14 to the aluminum pad 12 is exposed, and a longitudinal profile of a metal element such as Al, Sn, Ag, Sb, Cu, Au is measured by XMA (X-ray Micro Analyzer). The presence of the first alloy layer 16 and the second alloy layer 17 was confirmed. For example, the first alloy layer 16 was presumed to be mainly interdiffusion of Al and low melting point Sn.

  From this, it is presumed that due to the presence of the first alloy layer 16 and the second alloy layer 17, the aluminum pad 12 and the solder ball 14 are substantially integrated, and the above-described shear strength is obtained. The thicknesses of the first alloy layer 16 and the second alloy layer 17 vary depending on the high temperature reflow conditions.

Next, a masking tape is affixed to the surface of the semiconductor wafer 11 on the solder ball 14 side, back surface grinding is performed until a predetermined thickness is achieved, and polishing grinding (Ra: 0.02 μm or less) is performed.
Next, so-called stealth dicing is performed on the semiconductor wafer 11 by condensing and irradiating a laser beam from the back surface side and introducing a modification mark into the dicing line.
Next, the semiconductor wafer 11 is supported by a heat-resistant tape, and a back electrode is formed on the back surface of the semiconductor wafer 11 by sputtering or vapor deposition.

Next, the semiconductor wafer 11 is transferred to a highly extensible tape, the tape is stretched, and the semiconductor wafer 11 is separated into semiconductor chips by the stretching force.
As a result, a semiconductor chip is obtained in which the solder balls 14 are bonded onto the aluminum pad 12 via the sintered conductive layer 13 formed by thermally curing the conductive paste.
Next, the semiconductor device 10 is obtained by flip-chip bonding a semiconductor chip having the solder balls 14 to the substrate and molding with a resin.

  As described above, in this embodiment, the conductive paste is applied on the aluminum pad 12 whose surface is activated by the plasma treatment, and the solder balls 14 are placed on the conductive paste, and then high-temperature reflow is performed. Yes.

As a result, the solder balls 14 can be joined via the sintered conductive layer 13 formed by thermally curing the conductive paste on the aluminum pad 12.
Therefore, the semiconductor device 10 having a structure capable of easily joining the solder ball 14 to the aluminum pad 12 and the manufacturing method thereof are obtained.

  Here, a case where a small amount of phosphorus (P) is added to the solder ball 14 has been described. However, when high-temperature reflow is performed in a non-oxidizing atmosphere, phosphorus (P) may not be added. .

  A semiconductor device according to Embodiment 2 of the present invention will be described with reference to FIG. 6A and 6B are diagrams showing the main part of the semiconductor device according to the second embodiment of the present invention. FIG. 6A is a cross-sectional view thereof, FIG. 6B is an enlarged cross-sectional view of the main part of FIG. FIG. 6C is a cross-sectional view showing a state where the semiconductor device is mounted on the substrate.

  In the present embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, description thereof will be omitted, and different portions will be described. This example differs from Example 1 in that the sintered conductive layer is made as thick as a solder ball.

As shown in FIG. 6A, the semiconductor device 40 of this example includes a sintered conductive layer 41 having a thickness L <b> 1 formed on the aluminum pad 12.
As shown in FIG. 6B, the first alloy layer 16 in which the aluminum pad 12 and the metal contained in the sintered conductive layer 41 are reacted is formed between the aluminum pad 12 and the sintered conductive layer 41. Yes.

As shown in FIG. 6C, the semiconductor device 40 is mounted on the substrate 42 by flip chip connection.
In the semiconductor device 40, the upper surface 41 a of the sintered conductive layer 41 is bonded to a conductive pad 43 formed on an insulating substrate 42, and is electrically connected to the outside through a wiring 44 extending from the conductive pad 43. Is done.

  Therefore, the thickness L1 of the sintered conductive layer 41 is sufficient if it is sufficient to hold the semiconductor substrate 11 on the substrate 42 without hindrance, that is, about 100 to 300 μm as the solder ball 14 shown in FIG. It is.

The sintered conductive layer 41 is a conductive cured layer formed by thermally curing a conductive paste containing tin (Sn) and silver (Ag) as main components and added with a small amount of copper (Cu).
As is well known, the conductive paste is composed of a solder component and a flux component. In order to obtain the height L1 by thick coating, the conductive paste has 10-30 wt% Sn and 90-70 wt% Ag as the solder components. In addition, it is desirable that the Sn particle size is 5 to 30 μm and the Ag particle size is about 1 to 4 μm.

  As is well known, the flux component is a mixture of a rosin, an activator, a thixotropic agent, an additive, a solvent, and the like. A solder component is 80 to 90 wt%, and a flux component of about 20 to 10 wt% is appropriate.

  As will be described later, when the conductive paste is applied onto the aluminum pad 12, the conductive paste can be separated from the mask, that is, the mask can be easily removed. This is to ensure transfer.

  As a result, the conductive paste can be applied to a thickness similar to that of the solder ball 14 and about 100 to 300 μm. As a result of various tests, it was possible to apply to a height of about 500 μm.

  FIG. 7 is a flowchart showing the manufacturing process of the semiconductor device, and FIG. 8 is a cross-sectional view showing the main part of the manufacturing process of the semiconductor device.

  As shown in FIG. 7, in this embodiment, since the sintered conductive layer 41 having a thickness similar to that of a solder ball can be directly bonded to the conductive pad 43 of the substrate 42, a solder ball transfer step is not necessary. Other steps are performed in the same manner.

  That is, a semiconductor element, for example, a discrete semiconductor element is formed on the semiconductor wafer 11 by a known method, and an aluminum pad 12 for electrically connecting the semiconductor element to the outside is formed.

Next, as shown in FIG. 8A, the surface of the aluminum pad 12 is subjected to plasma treatment using argon (Ar) or oxygen (O ₂ ) to contaminate the surface (attached organic matter, surface oxide film). It is removed and the surface of the aluminum pad 12 is activated.

Next, as shown in FIG. 8B, a conductive paste is formed on the aluminum pad 12 using a mask 51 having an opening 50 at a position corresponding to the aluminum pad 12 and a thickness L2 slightly larger than L1. 52 is applied selectively.
For example, the conductive paste 52 is imprinted on the aluminum pad 12 by imprinting the conductive paste 52 into the opening 50 by the roller 23 using the screen on which the resist film having the opening 50 is formed as a mask 51 by screen printing.

  The thickness L2 of the mask 51 is calculated from the height L1 of the sintered conductive layer 41 by assuming that the solvent in the flux component volatilizes when the conductive paste 52 is thermally cured and the conductive paste 52 shrinks. It is desirable to keep it as large as possible.

  Since the conductive paste 52 is well separated from the mask 51, that is, has good detachability, when the mask 51 is removed, the conductive paste 52 is reliably transferred to the semiconductor substrate 11 and applied to the thickness L <b> 2.

  Next, as shown in FIG.8 (c), high temperature reflow (320-550 degreeC x 60-300 sec) is performed. By heating the heater 27, the conductive paste 52 is thermally cured to form a sintered conductive layer 41 having a thickness L 1 that is a conductive cured layer on the aluminum pad 12. Thereby, the semiconductor device 40 shown in FIG. 6A is obtained.

Next, after the upper surface 41 a of the sintered conductive layer 41 is brought into contact with the conductive pad 43, for example, a solder bump, the solder is melted by heating to weld the conductive pad 43 to the sintered conductive layer 41.
Thereby, the semiconductor device 40 is mounted on the insulating substrate 42 by flip chip connection.

As described above, the semiconductor device 40 of this embodiment includes the sintered conductive layer 41 having a thickness similar to that of the solder ball 14.
Thereby, since the semiconductor device 40 can be directly mounted on the substrate 42, there is an advantage that the conductive ball 14 becomes unnecessary, and the material can be reduced and the assembly process can be simplified.

FIG. 1A is a cross-sectional view of a main part of a semiconductor device according to a first embodiment of the present invention, and FIG. 1B is an enlarged cross-sectional view of the main part of FIG. 3 is a flowchart showing manufacturing steps of the semiconductor device according to the first embodiment of the present invention. Sectional drawing which shows the principal part of the manufacturing process of the semiconductor device which concerns on Example 1 of this invention. Sectional drawing which shows the principal part of the manufacturing process of the semiconductor device which concerns on Example 1 of this invention. Sectional drawing which shows the principal part of the manufacturing process of the semiconductor device which concerns on Example 1 of this invention. FIG. 6A is a cross-sectional view of the semiconductor device according to the second embodiment of the present invention, FIG. 6B is an enlarged cross-sectional view of the main portion of FIG. c) A sectional view showing a state where a semiconductor device is mounted on a substrate. 9 is a flowchart showing manufacturing steps of a semiconductor device according to Embodiment 2 of the present invention. Sectional drawing which shows the principal part of the manufacturing process of the semiconductor device which concerns on Example 2 of this invention.

Explanation of symbols

10, 40 Semiconductor device 11 Semiconductor substrate 12 Aluminum pad (pad mainly composed of aluminum)
13, 41 Sintered conductive layer 14 Solder ball (conductive ball)
DESCRIPTION OF SYMBOLS 15 Protective film 16 1st alloy layer 17 2nd alloy layer 20 1st opening 21 1st masks 22 and 52 Conductive paste 23 Roller 24 2nd opening 25 2nd mask 26 Spacer 27 Heater 28 Knife edge 42 Substrate 43 Conductivity Pad 44 Wiring 50 Opening 51 Mask

Claims (5)

A pad mainly composed of aluminum for electrically connecting a semiconductor element formed on the semiconductor substrate to the outside;
A sintered conductive layer formed on the aluminum-based pad;
Conductive balls welded on the sintered conductive layer;
A semiconductor device comprising:   A first alloy layer in which the aluminum and a metal contained in the sintered conductive layer are reacted is formed between the aluminum-based pad and the sintered conductive layer, and the sintered conductive layer and the sintered conductive layer The semiconductor device according to claim 1, wherein a second alloy layer in which a metal contained in the sintered conductive layer and the conductive ball are reacted is formed between the conductive balls.   2. The semiconductor device according to claim 1, wherein the sintered conductive layer is a conductive cured layer formed by thermally curing a conductive paste containing Sn and Ag as main components and added with Cu. .   The semiconductor device according to claim 1, wherein the conductive ball includes at least two selected from Sn, Sb, Cu, Au, and Ag as main components. Forming a pad mainly composed of aluminum for electrically connecting the semiconductor element to the outside on a semiconductor substrate on which the semiconductor element is formed;
Selectively applying a conductive paste on the aluminum-based pad using a first mask having a first opening at a position corresponding to the aluminum-based pad;
A conductive ball having a smaller size than the second opening is selectively placed on the conductive paste using a second mask having a second opening at a position corresponding to the pad containing aluminum as a main component. And a process of
Heat treatment is performed, the conductive paste is thermally cured to form a sintered conductive layer, the conductive balls are melted and welded onto the sintered conductive layer, and the baking is applied to the pad mainly composed of aluminum. Bonding the conductive balls through the conductive layer;
A method for manufacturing a semiconductor device, comprising:

Images