CN111630646A

CN111630646A - Solder joint

Info

Publication number: CN111630646A
Application number: CN201980003606.1A
Authority: CN
Inventors: 日野英治; 泽渡広信
Original assignee: JX Nippon Mining and Metals Corp
Current assignee: JX Nippon Mining and Metals Corp
Priority date: 2018-12-28
Filing date: 2019-08-07
Publication date: 2020-09-04
Also published as: JPWO2020136979A1; WO2020136979A1; TW202039241A; TWI717833B

Abstract

The present invention provides a solder joint portion having excellent durability in a high temperature region by a joint portion which joins a UBM to a solder alloy and comprises, in order from the UBM side, a Ni layer, a NiSn alloy layer, (Cu, Ni, Pd) alpha Sn alloy layer, a BiSn alloy layer, and a Bi layer continuing to the solder alloy side.

Description

Solder joint

Technical Field

The present invention relates to a solder joint.

Background

For environmental considerations, it is recommended to use solder alloys that do not contain lead. Depending on the composition of the solder alloy, the temperature range suitable for use as a solder also changes.

Power devices are used as power conversion elements in a wide range of fields such as hybrid vehicles and power transmission and transformation. Although the conventional devices using Si wafers are adaptable, in recent years, attention has been focused on SiC, GaN, and the like having a larger band gap than Si in the field where high withstand voltage, large current application, and high-speed operation are required.

The operating temperature of the conventional power module reaches about 170 ℃, but it is considered that SiC, GaN, and the like of the next generation may be in a temperature range of 200 ℃ or more and 200 ℃. Accordingly, heat resistance and heat dissipation are required for each material used for a module on which these wafers are mounted.

As a bonding portion having such characteristics, for example, a bonding portion using Sn-3.0Ag-0.5Cu solder (3.0 mass% Ag, 0.5 mass% Cu, and the balance Sn) is considered preferable from the viewpoint of Pb-free. However, since the next-generation module may have an operating temperature exceeding 200 ℃, it is required to have heat resistance higher than that of a bonding portion using a Sn-3.0Ag-0.5Cu solder having a melting point of about 220 ℃. Specifically, the joint portion is required to be resistant to a high temperature of 250 ℃ or higher, from the viewpoint of cooling of the radiator and tolerance of the temperature around the engine. Alternatively, it is considered that the operating temperature of the next-generation module can be adjusted by using a bonding portion of Pb solder (Pb-5Sn) which is not good from the viewpoint of environmental regulations.

In recent years, a joint using a metal powder paste has been attracting attention as a joint of a next-generation module. Due to the small size of the metal powder, the surface energy is high and sintering starts at temperatures well below the melting point of the metal. Unlike solder, once sintered, it is not remelted until the temperature is raised to a temperature close to the melting point of the metal. A joint using Ag powder paste is under development by taking advantage of such characteristics (patent document 1).

The Pb-5Sn solder has a sufficient function as a bonding material for a next-generation power module, but is preferably not used from the viewpoint of future environmental regulations because it contains lead. The Ag powder paste can provide sufficient bonding strength and heat resistance to the bonding portion depending on the conditions, but has a problem of material price.

Background of the invention

Patent document

Patent document 1: international publication No. 2011/155055.

Disclosure of Invention

[ problems to be solved by the invention ]

Solder joints having excellent durability even in a high temperature region required for bonding materials of next-generation power modules, for example, a temperature region exceeding 250 ℃.

Accordingly, an object of the present invention is to provide a solder joint portion having excellent durability in a high temperature region.

[ means for solving problems ]

Research and development of solder joints having excellent characteristics in a high temperature region have been carried out with attention paid to the melting point of a solder alloy. However, the present inventors have made further intensive studies and developments, and as a result, have found that the high-temperature characteristics of the solder joint portion are not so much dependent on the structure of the joint portion formed by solder joint, and have specified the structure, thereby completing the present invention.

Therefore, the present invention includes the following (1).

(1)

A bonding portion for bonding a UBM to a solder alloy, comprising the following layers in order from the UBM side:

a Ni layer continued from the UBM side,

A NiSn alloy layer,

A (Cu, Ni, Pd) alpha Sn alloy layer,

BiSn alloy layer, and

a Bi layer continuing toward the solder alloy side.

[ Effect of the invention ]

According to the present invention, a solder joint having excellent durability in a high temperature region can be obtained.

Drawings

Fig. 1 is an EPMA (electron probe microanalyzer) composite map image after the reflow soldering process of the joined portion of example 1.

Fig. 2 is an EPMA composite map image after being held at a temperature of 250 ℃ for 1000 hours after the reflow process of the joint portion of example 1.

Fig. 3 is an EPMA composite map image after reflow processing of the joint portion of comparative example 1.

Fig. 4 is an EPMA composite map image after being held at a temperature of 250 ℃ for 1000 hours after the reflow process of the joint portion of comparative example 1.

Fig. 5 is an STM (scanning electron microscope) image after reflow processing of the joint portion of example 1.

Fig. 6 is a graph in which the atomic concentration (mol%) obtained by analysis along the analysis line (55) of fig. 5 is taken as the vertical axis and the distance from the analysis starting point of the analysis line is taken as the horizontal axis.

Fig. 7 is an STM image after reflow processing of the joint portion of comparative example 1.

Fig. 8 is a graph in which the atomic concentration (mol%) obtained by analysis along the analysis line (75) of fig. 7 is on the vertical axis and the distance from the analysis starting point of the analysis line is on the horizontal axis.

Description of the symbols

11: ni layer

12: NiSn alloy layer

13: (Cu, Ni, Pd) alpha Sn layer

14: layer of Bi

21: ni layer

22: NiSn alloy layer

23: (Cu, Ni, Pd) alpha Sn layer

24: layer of Bi

31: ni layer (Bi invasion free)

32: (Cu, Ni, Pd) alpha Sn layer

32': NiSn alloy layer

33: free (Cu, Ni, Pd) alpha Sn phase

34: layer of Bi

35: invasion of Bi

41: ni layer (Bi invasion free)

42: ni layer (with Bi intrusion)

43': NiSn alloy phase (not continuous)

43: (Cu, Ni, Pd) alpha Sn layer

44: free (Cu, Ni, Pd) alpha Sn phase

45: layer of Bi

46: ni phase

51: ni layer

52: p-rich region

53: (Cu, Ni, Pd) alpha Sn layer

54: layer of Bi

55: analysis line

71: ni layer

72: p-rich region

73: (Cu, Ni, Pd) alpha Sn layer

74: layer of Bi

75: and (6) analyzing the line.

Detailed Description

The present invention will be described in detail below with reference to embodiments. The present invention is not limited to the specific embodiments listed below.

[ Joint joining UBM to solder alloy ]

The joint part of the present invention is a joint part for joining a UBM and a solder alloy, and comprises, in order from the UBM side, a Ni layer, a NiSn alloy layer, (Cu, Ni, Pd) alpha Sn alloy layer, a BiSn alloy layer, and a Bi layer continuing to the solder alloy side.

By such a layer structure, a solder joint having excellent durability in a high temperature region can be obtained.

[UBM]

An Under Bump Metal (UBM) is a Metal layer interposed on an electrode (for example, an Al electrode) on a wafer (for example, an Si wafer) to form a solder Bump. As the UBM, a Ni layer or a Ni alloy layer is generally formed, and a Pd layer and/or an Au layer or the like is formed thereon. In a preferred embodiment of the present invention, the UBM is, for example, a metal layer formed on an Al electrode on a Si wafer and composed of a Ni layer, a Pd layer formed thereon, and an Au layer formed thereon.

[ solder alloy ]

In a preferred embodiment, the solder alloy is a lead-free solder alloy, and examples thereof include alloys having a Bi-Cu-Sn composition as disclosed in examples.

[ UBM side of the joint ]

Since the bonding portion of the present invention is a bonding portion where the UBM and the solder alloy are bonded, the UBM loses the original structure of the metal layer due to the bonding, and only the Ni layer which is the main layer of the UBM remains as a detectable layer. The Ni layer from the UBM extends to the joint portion, and is a Ni layer extending from the UBM side in the vicinity of the joint portion. The UBM loses its original structure by joining, but in the present specification, the side where the UBM exists on both joined sides is referred to as the UBM side. In this specification, the Ni layer continuing from the UBM side is sometimes referred to as only the Ni layer. The Ni layer is a layer extending from the UBM side, but in order to clarify the structure of the joint portion, this description will be made as a part of the layer constituting the joint portion.

[ solder alloy side of joint ]

The joint part of the present invention is a joint part for joining the UBM and the solder alloy, so that the solder alloy side of the joint part extends toward the solder alloy. In a preferred embodiment, the solder alloy is an alloy containing Bi as a main component, and the solder alloy side of the joint portion is a Bi layer extending toward the solder alloy side. The solder alloy has a phase different from that of the original solder alloy due to bonding, but in the present specification, the side where the solder alloy exists on both sides of bonding is referred to as the solder alloy side. In this specification, the Bi layer continuing to the solder alloy side is sometimes referred to as a Bi layer only. The Bi layer is a layer extending toward the solder alloy side, but in order to clarify the structure of the joint portion, the Bi layer is described as a part of the layer constituting the joint portion in the present specification.

[ NiSn alloy layer ]

In a preferred embodiment, the NiSn alloy layer contains Ni, Sn and P. In a preferred embodiment, in the NiSn alloy layer, the Ni content at the boundary with the Ni layer is greater than the Ni content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer. In a preferred embodiment, in the NiSn alloy layer, the Sn content at the boundary with the Ni layer is smaller than the Sn content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer.

In a preferred embodiment, the content of Sn at the boundary with the Ni layer in the NiSn alloy layer is 0.4 mol% or less, preferably 0.35 mol% or less.

In a preferred embodiment, in the NiSn alloy layer, the P content at the boundary with the Ni layer is greater than the P content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer.

In a preferred embodiment, the content of P in the boundary between the NiSn alloy layer and the (Cu, Ni, Pd) α Sn alloy layer is 0.5 mol% or less, preferably 0.3 mol% or less.

In a preferred embodiment, the Ni content in the NiSn alloy layer at each distance from the boundary with the Ni layer is in the range of 21 mol% to 83 mol%, preferably in the range of 22 mol% to 80 mol%. The Ni content at each distance from the boundary with the Ni layer in the above range means the following meaning: the Ni content was measured at each measurement point by setting 1 or more measurement points at a distance from the boundary with the Ni layer, all within the above range; or both may be predicted to be within the above ranges. The predictable condition is within the above range, and means that the condition is within the above range by extrapolation from measurement values at measurement points close to each other, for example. The number of measurement points is not particularly limited, and from the viewpoint of actual measurement operation, for example, about 1 to 20 or about 1 to 5 measurement points may be provided. In the following description, the description of "each distance" is used as the same meaning as described above.

In a preferred embodiment, the content of Sn in the NiSn alloy layer at each distance from the boundary with the Ni layer is in the range of 0.2 mol% to 48 mol%, preferably in the range of 0.25 mol% to 45 mol%. The Sn content at each distance from the boundary with the Ni layer is in the above range, and is the same as that described above with respect to the Ni content.

In a preferred embodiment, the content of P in the NiSn alloy layer at each distance from the boundary with the Ni layer is in the range of 0.1 mol% to 10 mol%, preferably in the range of 0.2 mol% to 9 mol%. The P content at each distance from the boundary with the Ni layer is in the above range, and the same meaning as described above for the Ni content is used.

In a preferred embodiment, the content of Bi at the boundary with the (Cu, Ni, Pd) α Sn alloy layer in the NiSn alloy layer is 2 mol% or less, preferably 1 mol% or less.

In a preferred embodiment, the content of Bi in the NiSn alloy layer at each distance from the boundary with the Ni layer is in the range of 0.2 mol% to 2 mol%, preferably in the range of 0.3 mol% to 1.5 mol%. The Bi content at each distance from the boundary with the Ni layer is in the above range, and is the same as that described above with respect to the Ni content. As described above, in a preferred embodiment, the content of Bi in the NiSn alloy layer is extremely reduced, that is, the intrusion of Bi is effectively prevented.

In a preferred embodiment, the thickness of the NiSn alloy layer is in the range of, for example, 0.03 to 0.1[ mu ] m, preferably 0.04 to 0.1[ mu ] m, preferably 0.05 to 0.1[ mu ] m, preferably 0.06 to 0.1[ mu ] m.

In a preferred embodiment, in the NiSn alloy layer, the Cu content at the boundary with the Ni layer is less than the Cu content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer. In a preferred embodiment, the content of Cu in the NiSn alloy layer at the boundary with the Ni layer is 4 mol% or less, preferably 3 mol% or less.

In a preferred embodiment, the NiSn alloy layer further contains Pd. In a preferred embodiment, in the NiSn alloy layer, the Pd content at the boundary with the Ni layer is less than the Pd content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer. In a preferred embodiment, the content of Pd at the boundary with the Ni layer in the NiSn alloy layer is 3 mol% or less, preferably 2 mol% or less.

[ Ni layer ]

In a preferred embodiment, the Ni layer is a layer from UBM as described above. In a preferred embodiment, the Ni content of the Ni layer at each distance of 0.2[ μm ] or less from the boundary with the NiSn alloy layer is 83 mol% or more, preferably 85 mol% or more. In a preferred embodiment, the Ni layer may contain an elemental composition from the UBM. In a preferred embodiment, the Ni layer may contain P, for example.

In a preferred embodiment, the Ni layer has a Bi content of 0.2 mol% or less, preferably 0.1 mol% or less, at distances of 0.2[ μm ] or less from the boundary with the NiSn alloy layer. The content of Bi at each distance from the boundary with the NiSn alloy layer is the same as the content of Ni at each distance from the boundary with the Ni layer in the NiSn alloy layer, as described above, except that the content of Bi at each distance from the boundary with the Ni layer is the same as the content of Bi at each distance from the boundary with the Ni layer. As such, in a preferred embodiment, the content of Bi in the Ni layer is extremely reduced, i.e., Bi intrusion is effectively prevented.

[ (Cu, Ni, Pd) alpha Sn alloy layer ]

The (Cu, Ni, Pd) α Sn alloy layer is a Sn alloy layer in which Cu, Ni, Pd are dissolved. The (Cu, Ni, Pd) α Sn alloy layer is considered to be formed of the UBM and the constituent elements of the solder alloy. In a preferred embodiment, the (Cu, Ni, Pd) α Sn alloy layer contains Cu, Ni, Pd, and Sn.

In a preferred embodiment, in the (Cu, Ni, Pd) α Sn alloy layer, the Cu content at each distance from the boundary with the NiSn alloy layer is in the range of 10 mol% to 22 mol%, preferably in the range of 12 mol% to 20 mol%.

In a preferred embodiment, in the (Cu, Ni, Pd) α Sn alloy layer, the Ni content at each distance from the boundary with the NiSn alloy layer is in the range of 13 mol% to 21 mol%, preferably in the range of 14 mol% to 20 mol%.

In a preferred embodiment, in the (Cu, Ni, Pd) α Sn alloy layer, the Pd content at each distance from the boundary with the NiSn alloy layer is in the range of 5 mol% to 19 mol%, preferably in the range of 6 mol% to 18 mol%.

In a preferred embodiment, in the (Cu, Ni, Pd) α Sn alloy layer, the Sn content at each distance from the boundary with the NiSn alloy layer is in the range of 44 mol% to 55 mol%, preferably in the range of 45 mol% to 54 mol%.

In a preferred embodiment, the (Cu, Ni, Pd) α Sn alloy layer further contains Bi. In a preferred embodiment, the content of Bi at the boundary with the NiSn alloy layer in the (Cu, Ni, Pd) α Sn alloy layer is 2 mol% or less, preferably 1 mol% or less. In a preferred embodiment, in the (Cu, Ni, Pd) α Sn alloy layer, when the Bi content is measured from the distance from the boundary with the NiSn alloy layer, there is a distance that the Bi content is in the range of 0.1 mol% to 24 mol% (preferably in the range of 0.2 mol% to 23 mol%, or in the range of 1 mol% to 24 mol%, or in the range of 2 mol% to 23 mol%). That is, in a preferred embodiment, the (Cu, Ni, Pd) α Sn alloy layer has a peak of the Bi content at a position distant from the boundary with the NiSn alloy layer, which the inventors believe means that the NiSn alloy layer prevents Bi from entering from the Bi layer side. That is, in a preferred embodiment, the peak of the Bi content existing at a position of the (Cu, Ni, Pd) α Sn alloy layer away from the boundary with the NiSn alloy layer is the Bi content in the above range and is larger than the Bi content at the boundary with the NiSn alloy layer.

In a preferred embodiment, the content of Bi in the (Cu, Ni, Pd) α Sn alloy layer at the boundary with the BiSn alloy layer is 0.5 mol% or less, preferably 0.4 mol% or less.

[ BiSn alloy layer ]

In a preferred embodiment, the BiSn alloy layer contains Bi and Sn. In a preferred embodiment, the Bi content in the BiSn alloy layer at the boundary with the (Cu, Ni, Pd) α Sn alloy layer is 0.5 mol% or less, preferably 0.4 mol% or less. In a preferred embodiment, the content of Sn in the boundary between the BiSn alloy layer and the Bi layer is 2 mol% or less, and preferably 1 mol% or less.

In a preferred embodiment, the BiSn alloy layer further contains Cu, Ni, and Pd.

In a preferred embodiment, in the BiSn alloy layer, the Cu content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer is greater than the Cu content at the boundary with the Bi layer. In a preferred embodiment, the Cu content in the BiSn alloy layer at the boundary with the Bi layer is 0.5 mol% or less, preferably 0.3 mol% or less.

In a preferred embodiment, in the BiSn alloy layer, the Ni content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer is greater than the Ni content at the boundary with the Bi layer. In a preferred embodiment, the content of Ni at the boundary with the Bi layer in the BiSn alloy layer is 1 mol% or less, preferably 0.5 mol% or less.

In a preferred embodiment, in the BiSn alloy layer, the Pd content at the boundary with the (Cu, Ni, Pd) α Sn alloy layer is greater than the Pd content at the boundary with the Bi layer. In a preferred embodiment, the content of Pd in the boundary with the Bi layer in the BiSn alloy layer is 0.6 mol% or less, preferably 0.3 mol% or less.

[ Bi layer ]

In a preferred embodiment, the Bi layer is a Bi layer that continues toward the solder alloy side as described above, with Bi from the solder alloy as a main component. In a preferred embodiment, the Bi content of the Bi layer at each distance of 0.2[ μm ] or less from the boundary with the BiSn alloy layer is 97 mol% or more, preferably 98 mol% or more. In a preferred embodiment, the Bi layer may contain elemental constituents from the solder alloy.

In a preferred embodiment, the Bi layer does not have a phase of (Cu, Ni, Pd) α Sn alloy that is free from the (Cu, Ni, Pd) α Sn alloy layer, the phase of (Cu, Ni, Pd) α Sn alloy that is free from the (Cu, Ni, Pd) α Sn alloy layer, specifically, a phase described as a phase in an image of a comparative example in the examples described later²]When the image near the bonding portion is observed in the above visual field and the Bi layer does not have the phase of the (Cu, Ni, Pd) α Sn alloy that is free from the (Cu, Ni, Pd) α Sn alloy layer, it is considered that the condition that the Bi layer does not have the phase of the (Cu, Ni, Pd) α Sn alloy that is free from the (Cu, Ni, Pd) α Sn alloy layer is satisfied.

[ element concentration ]

The concentration (% by mole) of each element can be measured by the means disclosed in examples described later, and specifically, it can be measured using STM (manufactured by JEOL, device name: JEM-2100F). The element concentration in the present invention means the element concentration when the total of Ni, Sn, Bi, Cu, Pd, and P is 100 mol%.

[ joining Strength (shear Strength) ]

In a preferred embodiment, the joint of the present invention has a joint strength (shear strength) after being held at a high temperature of 250 ℃ for 1000 hours of, for example, 40MPa or more, preferably 42MPa or more. That is, the joint portion has excellent durability in a high temperature region. The bonding strength can be measured by the procedure described in the examples below. The reason why the joined portion of the present invention has excellent durability in a high temperature region is not clear, but the present inventors have found that, in the present invention, by making the joined portion have the above-described layer structure, particularly, by having a specific NiSn alloy layer, the penetration of Bi from the Bi layer side toward the Ni layer side can be prevented, and excellent characteristics can be obtained, based on a comparison with comparative examples to be described later.

[ preferred embodiment ]

In a preferred embodiment, the present invention includes the following items (1).

(1)

a Ni layer continued from the UBM side,

A NiSn alloy layer,

A (Cu, Ni, Pd) alpha Sn alloy layer,

BiSn alloy layer, and

a Bi layer continuing toward the solder alloy side.

(2)

The joint according to (1), wherein the NiSn alloy layer contains Ni, Sn and P,

in the NiSn alloy layer, the Ni content at the boundary with the Ni layer is larger than that at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer,

in the NiSn alloy layer, the Sn content at the boundary with the Ni layer is less than the Sn content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer,

in the NiSn alloy layer, the Sn content at the boundary with the Ni layer is less than 0.4 mol%,

in the NiSn alloy layer, the P content at the boundary with the Ni layer is larger than the P content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer,

in the NiSn alloy layer, the P content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer is less than or equal to 0.5 mol%.

(3)

The joint according to any one of (1) to (2), wherein the NiSn alloy layer contains Ni, Sn and P,

in the NiSn alloy layer, the Ni content at each distance from the boundary with the Ni layer is in the range of 21 mol% to 83 mol%,

in the NiSn alloy layer, the Sn content at each distance from the boundary with the Ni layer is in the range of 0.2 mol% to 48 mol%,

in the NiSn alloy layer, the P content at each distance from the boundary with the Ni layer is in the range of 0.1 mol% to 10 mol%.

(4)

The joint according to any one of (1) to (3), wherein a Bi content at a boundary with the (Cu, Ni, Pd) α Sn alloy layer in the NiSn alloy layer is 2 mol% or less.

(5)

The joint according to any one of (1) to (4), wherein the NiSn alloy layer has a Bi content in a range of 0.2 mol% to 2 mol% at each distance from a boundary with the Ni layer.

(6)

The joint according to any one of (1) to (5), wherein the NiSn alloy layer has a thickness in the range of 0.05 to 0.1[ μm ].

(7)

The joint according to any one of (2) to (6), wherein in the NiSn alloy layer, a Cu content at a boundary with the Ni layer is smaller than a Cu content at a boundary with the (Cu, Ni, Pd) α Sn alloy layer,

in the NiSn alloy layer, the Cu content at the boundary with the Ni layer is 4 mol% or less.

(8)

The bonding part according to any one of (2) to (7), wherein the NiSn alloy layer further contains Pd,

in the NiSn alloy layer, the Pd content at the boundary with the Ni layer is less than the Pd content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer,

in the NiSn alloy layer, the Pd content at the boundary with the Ni layer is 3 mol% or less.

(9)

The joint according to any one of (1) to (8), wherein the Ni layer has a Bi content of 0.2 mol% or less at distances of 0.2[ μm ] or less from a boundary with the NiSn alloy layer.

(10)

The bonding part according to any one of (1) to (9), wherein the (Cu, Ni, Pd) α Sn alloy layer contains Cu, Ni, Pd and Sn,

in the (Cu, Ni, Pd) alpha Sn alloy layer, the Cu content at each distance from the boundary with the NiSn alloy layer is in the range of 10 mol% to 22 mol%,

in the (Cu, Ni, Pd) alpha Sn alloy layer, the Ni content at each distance from the boundary with the NiSn alloy layer is in the range of 13 mol% to 21 mol%,

in the (Cu, Ni, Pd) alpha Sn alloy layer, the Pd content at each distance from the boundary with the NiSn alloy layer is in the range of 5 mol% to 19 mol%,

in the (Cu, Ni, Pd) α Sn alloy layer, the Sn content at each distance from the boundary with the NiSn alloy layer is in the range of 44 mol% to 55 mol%.

(11)

The bonding part according to (10), wherein the (Cu, Ni, Pd) α Sn alloy layer further contains Bi,

in the (Cu, Ni, Pd) alpha Sn alloy layer, the Bi content at the boundary with the NiSn alloy layer is 2 mol% or less,

in the (Cu, Ni, Pd) alpha Sn alloy layer, when the Bi content is measured from the distance from the boundary with the NiSn alloy layer, there is a distance that the Bi content is in the range of 0.1 mol% to 24 mol%,

in the (Cu, Ni, Pd) α Sn alloy layer, the Bi content at the boundary with the BiSn alloy layer is 0.5 mol% or less.

(12)

The bonded part according to any of (1) to (11), wherein the BiSn alloy layer contains Bi and Sn,

the Bi content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer in the BiSn alloy layer is less than 0.5 mol%,

in the BiSn alloy layer, the Sn content at the boundary with the Bi layer is 2 mol% or less.

(13)

The bonding part according to (12), wherein the BiSn alloy layer further contains Cu, Ni, and Pd,

in the BiSn alloy layer, the Cu content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer is larger than the Cu content at the boundary with the Bi layer,

the Cu content at the boundary with the Bi layer in the BiSn alloy layer is 0.5 mol% or less,

in the BiSn alloy layer, the Ni content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer is larger than that at the boundary with the Bi layer,

the Ni content at the boundary with the Bi layer in the BiSn alloy layer is 1 mol% or less,

in the BiSn alloy layer, the Pd content at the boundary with the (Cu, Ni, Pd) alpha Sn alloy layer is larger than that at the boundary with the Bi layer,

in the BiSn alloy layer, the Pd content at the boundary with the Bi layer is 0.6 mol% or less.

(14)

The bonding portion according to any one of (1) to (13), wherein the Bi layer does not have a phase of the (Cu, Ni, Pd) α Sn alloy that is free from the (Cu, Ni, Pd) α Sn alloy layer.

(15)

The joint according to any one of (1) to (14), wherein the UBM is formed by sequentially laminating nickel, palladium and gold on the electrode.

(16)

The joint according to any one of (1) to (15), which has a joint strength of 40MPa or more after being held at a high temperature of 250 ℃ for 1000 hours.

(17)

An electronic component having the joint portion according to any one of (1) to (16).

(18)

A power device having the joint according to any one of (1) to (16).

(19)

A printed circuit board having the joint portion according to any one of (1) to (16).

In a preferred embodiment, the present invention includes an electronic component, a power device, a printed circuit board, an LED, a flexible circuit material, and a heat dissipating material, each of which has the above-described joint portion.

[ examples ]

The present invention will be described in detail below with reference to examples. The present invention is not limited to the following examples.

[ example 1]

[ formation of UBM ]

An a1 side (thickness 3 μm) was prepared on one side of an Si wafer by sputtering, and a polyimide film was formed by coating, followed by exposure and development to form a pad (land) having an opening of 300 μm in diameter on the polyimide film.

Further, an Ni layer (thickness: 2.5 μm), a Pd layer (thickness: 0.05 μm), and an Au layer (thickness: 0.02 μm) were formed in this order on the pad portion by electroless plating, thereby providing an UBM (under Bump Metal). Furthermore, since the electroless Ni plating solution contains a reducing agent for hypophosphite ions, P is co-deposited on the Ni layer.

[ solder powder ]

The composition of the solder powder used in example 1 by ICP analysis is shown in table 1. 300 μm Φ was used as the solder powder.

[ reflow treatment ]

Flux was applied to the UBM, and solder powder of 300 μm was mounted thereon, and reflow treatment was performed to heat-bond the flux. The conditions of the reflow process were as follows:

the temperature was raised from room temperature to 110 ℃ at a rate of 1.4 ℃/sec. Next, the temperature was raised from 110 ℃ to 140 ℃ at a rate of 0.7 ℃/sec. The temperature is raised from 140 ℃ to 200 ℃ at a temperature rise rate of 0.8 ℃/second. Next, the temperature was raised from 200 ℃ to 290 ℃ at a temperature raising rate of 4.5 ℃/sec. Next, the temperature of 290 ℃ was maintained for 40 seconds. Next, the sample was initially cooled from 290 ℃ to room temperature at a cooling rate of 7 ℃/sec. The operation was carried out in a nitrogen atmosphere.

[ EPMA analysis, STM analysis ]

Thereafter, the sample was sealed in a resin and the cross section was polished. The clean and smooth surface of the sample was analyzed by EPMA (manufactured by JEOL, device name: JXA-8500F). The thickness of the solder joint after the reflow process is about 1 μm or less, and therefore, it is measured by a high-precision STM (manufactured by JEOL, device name: JEM-2100F). The image accuracy of the EPMA measurement is inferior to that of STM, but the image processing is performed so that the size of 1 pixel is 0.08. mu.m. For further quantification, the same sample was processed to be thin and then measured by STM.

The joint section of example 1 was measured by EPMA on the joint section immediately after reflow treatment, and an image of the composite map thus prepared is shown in fig. 1.

The joint of example 1 was subjected to EPMA measurement and composite mapping in the same manner as for the cross section of the joint after being held at 250 ℃ for 1000 hours after the reflow treatment. The image of this composite map is shown in fig. 2.

[ measurement of shear Strength ]

The joint of example 1 was subjected to the above-described reflow treatment, and then held at a temperature of 250 ℃ for 1000 hours in an atmospheric environment, and then the shear strength was measured as follows. The results are shown in table 1.

The bond strength was measured according to MIL STD-883G. A tool mounted on a load sensor is lowered to a substrate surface, the apparatus detects the substrate surface and stops the lowering, the tool is raised to a predetermined height from the detected substrate surface, and the load when the joint is broken by the tool pressing is measured. These results are summarized in table 1.

< measurement Condition >

The device comprises the following steps: product of dage corporation, dage series 4000

The method comprises the following steps: die shear strength test

Testing speed: 100 μm/sec

And (3) testing height: 20.0 μm

Tool movement amount: 0.9mm

Comparative example 1

As comparative example 1, a joint portion was formed by the same procedure as in example 1 except that a solder powder having a different composition from that of example 1 was used and a reflow process under different conditions from those of example 1 was performed, and EPMA measurement, STM measurement, and shear strength measurement were performed on the joint portion. The ICP analysis values and the shear strength measurement results of the composition of the solder powder of comparative example 1 are shown in table 1. The reflow conditions of comparative example 1 were as follows:

the temperature was raised from room temperature to 150 ℃ at a rate of 0.9 ℃/sec. Next, the temperature was raised from 150 ℃ to 290 ℃ at a temperature raising rate of 4.5 ℃/sec. Next, the temperature of 290 ℃ was maintained for 40 seconds. Next, the sample was initially cooled from 290 ℃ to room temperature at a cooling rate of 8 ℃/sec. The operation was carried out in a nitrogen atmosphere.

The joint section of comparative example 1 immediately after the reflow process was measured by EPMA, and an image of the composite map thus prepared is shown in fig. 3.

For the joint of comparative example 1, a composite map was created by measuring the cross section of the joint after the solder reflow treatment and holding the joint at a temperature of 250 ℃ for 1000 hours by the same EPMA. The image of this composite map is shown in fig. 4.

[ Table 1]

[ STM image and line analysis ]

Fig. 5 shows an STM image of a cross section of the joint portion immediately after the reflow process in the joint portion of example 1. Fig. 6 shows a graph obtained by converting the concentration (mol%) of each element along the analysis line shown in the STM image of the junction of example 1. The element concentration (mol%) is 100% of the total of 6 elements in the graph, i.e., Ni, Sn, Bi, Cu, Pd, and P.

Fig. 7 shows an STM image of a cross section of the joint portion immediately after the reflow process in the joint portion of comparative example 1. Fig. 8 shows a graph obtained by converting the concentration (mol%) of each element along the analysis line shown in the STM image of the junction of comparative example 1.

[ evaluation ]

Fig. 1 is an EPMA composite map image after reflow processing of a joint portion in example 1.

In fig. 1, a layer considered to be a (Cu, Ni, Pd) α Sn layer (13) was observed in the vicinity of the interface between the Ni layer (11) and the Bi layer (14), and a NiSn alloy layer (having a thickness of about 0.1 μm) (12) was present directly below the (Cu, Ni, Pd) α Sn layer (13). In the NiSn alloy layer (12), Bi intrusion is not observed at all in the image of the colorized composite mapping. The thickness of the (Cu, Ni, Pd) α Sn layer (13) is relatively uniform, and no (Cu, Ni, Pd) α Sn phase released from the (Cu, Ni, Pd) α Sn layer (13) is observed in the Bi layer (14). In fig. 1, the intrusion of Bi into the Ni layer (11) is not observed at all in the image of the colorized composite map. The black area occupying more than the lower half of the image in fig. 1 is an Al layer (3 μm) which becomes the base of the Ni layer (11) and an Si base material which becomes the base of the Al layer.

In fig. 2, a layer considered to be a (Cu, Ni, Pd) α Sn layer (23) was observed in the vicinity of the interface between the Ni layer (21) and the Bi layer (24), and a NiSn alloy layer (about 0.2 μm in thickness) (22) was present directly below the (Cu, Ni, Pd) α Sn layer (23). In the NiSn alloy layer (22), Bi penetration is not observed at all in the image of the colorized composite mapping. In addition, although the thickness of the (Cu, Ni, Pd) α Sn layer (23) is increased as compared with fig. 1, the entire layer remains in close contact with the NiSn alloy layer (22). In the Bi layer (24), no (Cu, Ni, Pd) α Sn phase released from the (Cu, Ni, Pd) α Sn layer (23) is observed. In the Bi layer (24), the Ni phase considered to be derived from the Ni layer is not observed at all in the colored composite-mapped image. In fig. 2, the intrusion of Bi into the Ni layer (21) was not observed at all in the image of the colorized composite map.

In fig. 3, a layer considered to be a (Cu, Ni, Pd) α Sn layer (32) is observed in the vicinity of the interface between the Ni layer (31) and the Bi layer (34), and a layer considered to be a NiSn alloy layer (32') is present directly below the (Cu, Ni, Pd) α Sn layer (32) (thickness about 0.1 μm), but in the image of the colorized composite map, a slight amount of Bi intrusion (35) is observed to the extent that it is not observed in the gradation level in the NiSn (32') alloy layer. The thickness of the (Cu, Ni, Pd) alpha Sn layer (32) is not uniform, and a (Cu, Ni, Pd) alpha Sn phase (33) released from the (Cu, Ni, Pd) alpha Sn layer (32) is observed in the Bi layer (34). In fig. 3, the intrusion of Bi into the Ni layer (31) is not observed in the image of the colorized composite map. The black area occupying more than the lower half of the image in fig. 3 is an Al layer (3 μm) which becomes the base of the Ni layer (31) and an Si base material which becomes the base of the Al layer.

In fig. 4, the Ni layer is present as in fig. 3, but in the image of the colorized composite map, the Ni layer becomes the Ni layer (42) in which Bi intrusion is observed almost in the entire thickness (thickness of about 1.8 μm), and the Ni layer (41) in which Bi intrusion is not observed is present only very thinly (thickness of about 0.4 μm). A layer considered to be a (Cu, Ni, Pd) alpha Sn layer (43) is observed in the vicinity of the interface between the Ni layer (42) and the Bi layer (45), and a layer considered to be a NiSn alloy layer is not observed in an image of a colorized composite map directly below the (Cu, Ni, Pd) alpha Sn layer (43), but lumps of the NiSn alloy phase (43') are not observed continuously. The (Cu, Ni, Pd) α Sn layer (43) has a very irregular shape, and a plurality of free (Cu, Ni, Pd) α Sn phases (44) are observed in the Bi layer (45). Furthermore, a plurality of Ni phases (46) that are thought to originate from the Ni layer are observed in the Bi layer (45).

Fig. 5 is an STM image after reflow processing of the joint portion of example 1.

In fig. 5, a layer considered to be a (Cu, Ni, Pd) α Sn layer (53) is observed in the vicinity of the interface between the Ni layer (51) and the Bi layer (54). An area (52) with elongated holes as seen in the up-down direction is observed. As will be described later, this region (52) is a P-rich region having a relatively large P concentration. The thickness of the P-rich region is about 0.13 μm. FIG. 6 is a graph showing the results of analysis along the analysis line (55) and measurement of the atomic concentration at each analysis point.

Fig. 6 is a graph in which the atomic concentration (mol%) obtained by analysis along the analysis line (55) of fig. 5 is taken as the vertical axis and the distance from the analysis starting point of the analysis line is taken as the horizontal axis. The left side of the graph is the Ni layer side, and the right side of the graph is the Bi layer side. The starting point of the horizontal axis is located in the Ni layer. The Ni concentration gradually decreased at a distance of 0.15 μm, and the Sn concentration started to increase rapidly from a nearly nonexistent state in the vicinity of 0.22 μm. The NiSn alloy layer is believed to start from this position. At this point, the P concentration is relatively higher than before and after the point, and the P concentration is decreased with the increase in distance, and almost disappears in the vicinity of 0.26. mu.m. In parallel with the decrease in the P concentration, the Ni concentration decreases and the Sn concentration increases. It is considered that the NiSn alloy layer ends at the position where the P concentration disappears, and the (Cu, Ni, Pd) α Sn layer starts newly. In the initial position of the (Cu, Ni, Pd) α Sn layer, the Pd concentration is relatively higher than before and after the layer. At the boundary between the NiSn alloy layer and the (Cu, Ni, Pd) α Sn layer, the Sn concentration was estimated to be 48 mol%, the Ni concentration was estimated to be 21 mol%, and the value of α was estimated to be 0.8. When the distance from the starting position of the (Cu, Ni, Pd) α Sn layer is increased, a region having a relatively high Bi content appears temporarily in the vicinity of 0.30 μm, but almost no Bi is contained before or after the region. Thereafter, the (Cu, Ni, Pd) α Sn layer continues for a period of increasing distance. The alpha value is estimated to be 0.8-0.9, and the average value is 0.83. Thereafter, when the distance increases, the Bi concentration starts to increase sharply from the almost nonexistent state in the vicinity of 0.48 μm. At the position where the Bi concentration starts to rise, the (Cu, Ni, Pd) α Sn layer ends and the BiSn alloy layer starts. In the BiSn alloy layer, Cu, Ni and Pd are present in addition to Bi and Sn. The concentrations of Cu, Ni, and Pd in the BiSn alloy layer all decreased with increasing distance, and disappeared in the vicinity of 0.56 μm. It is considered that the BiSn alloy layer ends and the Bi layer starts at the position where the concentrations of Cu, Ni, and Pd disappear.

In fig. 7, a layer considered to be a (Cu, Ni, Pd) α Sn layer (73) is observed in the vicinity of the interface between the Ni layer (71) and the Bi layer (74). In comparison with fig. 5, the P-rich region (72) having the elongated holes in the vertical direction is slightly irregular and uneven in shape, and the atomic concentration at each analysis point is measured by analyzing along the observed analysis line (75) to obtain a graph shown in fig. 8.

Fig. 8 is a graph in which the atomic concentration (mol%) obtained by analysis along the analysis line (75) of fig. 7 is on the vertical axis and the distance from the analysis starting point of the analysis line is on the horizontal axis. The left side of the graph is the Ni layer side, and the right side of the graph is the Bi layer side. In the graph of fig. 8, it is observed that Bi penetrates to a position corresponding to the boundary between the NiSn alloy layer and the (Cu, Ni, Pd) α Sn layer in the graph of fig. 6. In the graph of fig. 8, it is observed that Bi passes through a position corresponding to the NiSn alloy layer in the graph of fig. 6, and penetrates to a position corresponding to the Ni layer. It is predicted that the Sn concentration is 18 mol%, the Ni concentration is 23 mol%, and the value of α is 2.0 at the position where the P concentration disappears, i.e., at the boundary between the NiSn alloy layer and the (Cu, Ni, Pd) α Sn layer. The alpha of the (Cu, Ni, Pd) alpha Sn layer is predicted to be 0.7-1.2, and the average value is 0.85.

[ industrial applicability ]

The invention provides a solder joint part which does not contain lead and has excellent durability in a high-temperature area. The present invention is industrially useful.

Claims

1. A bonding portion for bonding a UBM to a solder alloy, comprising the following layers in order from the UBM side:

a Ni layer continued from the UBM side,

A NiSn alloy layer,

A (Cu, Ni, Pd) alpha Sn alloy layer,

BiSn alloy layer, and

a Bi layer continuing toward the solder alloy side.

2. The joint according to claim 1, wherein the NiSn alloy layer contains Ni, Sn and P,

3. The joint according to claim 1 or 2, wherein the NiSn alloy layer contains Ni, Sn and P,

4. The joint according to any of claims 1 to 3, wherein a Bi content at a boundary with the (Cu, Ni, Pd) α Sn alloy layer in the NiSn alloy layer is 2 mol% or less.

5. The joint according to any of claims 1 to 4, wherein the Bi content at each distance from the boundary with the Ni layer in the NiSn alloy layer is in the range of 0.2 mol% to 2 mol%.

6. The joint according to any of claims 1 to 5, wherein a thickness of the NiSn alloy layer is in a range of 0.03 to 0.1[ μm ].

7. The joint according to any of claims 2 to 6, wherein in the NiSn alloy layer, a Cu content at a boundary with the Ni layer is smaller than a Cu content at a boundary with the (Cu, Ni, Pd) alpha Sn alloy layer,

8. The joint portion according to any of claims 2 to 7, wherein the NiSn alloy layer further contains Pd,

9. The joint according to any of claims 1 to 8, wherein the Bi content of the Ni layer at each distance within 0.2[ μm ] from the boundary with the NiSn alloy layer is 0.2 mol% or less.

10. The joint portion according to any of claims 1 to 9, wherein the (Cu, Ni, Pd) α Sn alloy layer contains Cu, Ni, Pd and Sn,

11. The joint according to claim 10, wherein the (Cu, Ni, Pd) α Sn alloy layer further contains Bi,

12. The joint according to any of claims 1 to 11, wherein the BiSn alloy layer contains Bi and Sn,

13. The bonding part according to claim 12, wherein the BiSn alloy layer further contains Cu, Ni, and Pd,

14. The joint portion according to any of claims 1 to 13, wherein the Bi layer has no phase of the (Cu, Ni, Pd) α Sn alloy that is dissociated from the (Cu, Ni, Pd) α Sn alloy layer.

15. The joint portion according to any of claims 1 to 14, wherein the UBM is formed by sequentially depositing nickel, palladium, and gold on the electrode.

16. The joint according to any one of claims 1 to 15, having a joint strength of 40MPa or more after being held at a high temperature of 250 ℃ for 1000 hours.

17. An electronic part having the joint portion as set forth in any one of claims 1 to 16.

18. A power device having the joint of any one of claims 1 to 16.

19. A printed circuit board having the joint portion of any one of claims 1 to 16.