Mount structure

Abstract

A mount structure having a joining capable of withstanding development of cracks generated by thermal stress due to repeated temperature changes in a mount structure having the joining of a large area is formed by joining a ceramic substrate electrode of a ceramic substrate and a metal substrate electrode of a metal substrate by a laminate, in which the laminate is formed by stacking a first interface layer, a first solder joining portion, a second interface layer, a first buffer material electrode, a buffer material, a second buffer material electrode, a third interface layer, a second solder joining portion and a fourth interface layer in this order from the ceramic substrate electrode toward the metal substrate electrode, a thickness of the laminate is 30 m or more and 100 m or less, a difference between a thickness of the first solder joining portion and a thickness of the second solder joining portion is within 25%, and differences in elastic moduli and in linear expansion coefficients between the first solder joining portion and the buffer material are respectively within 62%.

Claims

1. A mount structure comprising a ceramic substrate electrode comprising a ceramic substrate, and a metal substrate electrode comprising a metal substrate, wherein the ceramic substrate electrode and the metal substrate electrode are joined by a laminate, wherein the laminate is formed by stacking a first interface layer, a first solder joining portion, a second interface layer, a first buffer material electrode, a buffer material, a second buffer material electrode, a third interface layer, a second solder joining portion and a fourth interface layer in this order from the ceramic substrate electrode toward the metal substrate electrode, wherein a thickness of the laminate is 30 m or more and 100 m or less, wherein a difference between a thickness of the first solder joining portion and a thickness of the second solder joining portion is within 25%, wherein a difference in elastic moduli between the first solder joining portion and the buffer material is within 10%, and wherein a difference in linear expansion coefficients between the first solder joining portion and the buffer material is within 1%.

2. The mount structure according to claim 1, wherein the first solder joining portion and the second solder joining portion are formed of Au-20Sn.

3. The mount structure according to claim 1, wherein the buffer material is formed of Cu.

4. The mount structure according to claim 1, wherein the ceramic substrate is formed of AlN.

5. The mount structure according to claim 1, wherein the metal substrate is formed of Cu.

6. The mount structure according to claim 1, wherein the ceramic substrate electrode is a two-layer plating layer.

7. The mount structure according to claim 1, wherein the metal substrate electrode is a two-layer plating layer.

8. The mount structure according to claim 1, wherein the first buffer material electrode and the second buffer material electrode are plating layers.

9. The mount structure according to claim 1, wherein the second solder joining portion is a plating layer.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a cross-sectional view showing a first solder joining portion to a second solder joining portion in parallel with respect to a thickness direction in a mount structure according to Embodiment 1.

(2) FIG. 2 is a cross-sectional view showing a first solder to a second solder in parallel with respect to a thickness direction before the mount structure according to Embodiment 1 is joined.

DESCRIPTION OF EMBODIMENTS

(3) A mount structure according to an aspect of the present disclosure is formed by joining a ceramic substrate electrode of a ceramic substrate and a metal substrate electrode of a metal substrate by a laminate, in which the laminate is formed by stacking a first interface layer, a first solder joining portion, a second interface layer, a first buffer material electrode, a buffer material, a second buffer material electrode, a third interface layer, a second solder joining portion and a fourth interface layer in this order from the ceramic substrate electrode toward the metal substrate electrode, a thickness of the laminate is 30 m or more and 100 m or less, a difference between a thickness of the first solder joining portion and a thickness of the second solder joining portion is within 25%, and a difference in elastic moduli between the first solder joining portion and the buffer material and a difference in linear expansion coefficients between the first solder joining portion and the buffer material are respectively within 62%.

(4) In the mount structure according to another aspect of the present disclosure, the difference in elastic moduli between the first solder joining portion and the buffer material and a difference in linear expansion coefficients between the first solder joining portion and the buffer material may be respectively within 20%.

(5) In the mount structure according to further another aspect of the present disclosure, the first solder joining portion and the second solder joining portion may be formed of Au-20Sn.

(6) In the mount structure according to further another aspect of the present disclosure, the buffer material may be formed of any of Cu, Al and Ag.

(7) In the mount structure according to further another aspect of the present disclosure, the ceramic substrate may be formed of AlN.

(8) In the mount structure according to further another aspect of the present disclosure, the metal substrate may be formed of Cu.

(9) Hereinafter, amount structure according to embodiments of the present disclosure will be explained with reference to the attached drawings. The same numerals are added to members which are substantially the same in the drawings.

Embodiment 1

(10) FIG. 1 is a cross-sectional view showing a first solder joining portion 104 to a second solder joining portion 110 in parallel with respect to a thickness direction in a mount structure 114 according to Embodiment 1.

(11) The mount structure 114 according to Embodiment 1 is a mount structure in which a ceramic substrate electrode 102 of a ceramic substrate 101 and a metal substrate electrode 112 of a metal substrate 113 are joined by a laminate 117. The laminate 117 is formed by stacking a first interface layer 103-1, the first solder joining portion 104, a second interface layer 103-2, a first buffer material electrode 106, a buffer material 107, a second buffer material electrode 108, a third interface layer 103-3, the second solder joining portion 110 and a fourth interface layer 103-4 in this order from the ceramic substrate electrode 102 toward the metal substrate electrode 112. A thickness of the laminate 117 is 30 m or more and 100 m or less. Moreover, a difference between a thickness of the first solder joining portion 104 and a thickness of the second solder joining portion 110 is within 25%. Furthermore, a difference in elastic moduli between the first solder joining portion 104 and the buffer material 107 and a difference in linear expansion coefficients between the first solder joining portion 104 and the buffer material 107 are respectively within 62%.

(12) The first and second solder joining portions 104 and 110 are formed of, for example, Au-20Sn (an AuSn alloy with a composition of 20 wt % Sn), in which an elastic modulus is 68 GPa and a linear expansion coefficient is 16 ppm/K.

(13) The first and second solder joining portions 104 and 110 are formed to have approximately the same thickness.

(14) When the ceramic substrate electrode 102 and the metal substrate electrode 112 are joined by the laminate 117, thicknesses of the first and second solder joining portions 104 and 110 are equivalent, wetting forces (wettabilities) are equivalent in the ceramic substrate electrode 102 and the metal substrate electrode 112. Accordingly, the wetting force in the horizontal direction is not biased, and it is possible to suppress inclination of the ceramic substrate 101 which is a problem when mounting an LED device.

(15) When repeated temperature changes occur by operation of a device such as an LED, thermal stress is generated due to a difference in linear expansion coefficients between the ceramic substrate 101 and the metal substrate 113. The generated thermal stress is respectively applied to the first and second solder joining portions 104, 110 and the buffer material 107, which causes distortion. The thermal stress generated at the time of distortion is affected by respective elastic moduli and linear expansion coefficients. When respective elastic moduli and linear expansion coefficients differ, respective distortion behaviors also differ, therefore, stress concentration occurs.

(16) Au-20Sn is a strong solder material and can withstand large thermal stress in the joining of a large area, however, it is difficult to suppress development of cracks when stress concentration occurs.

(17) On the other hand, elastic moduli and linear expansion coefficients of the first and second solder joining portions 104, 110 and the buffer material 107 are close to one another in the mount structure according to Embodiment 1. Accordingly, it is possible to suppress stress concentration to thereby allow the mount structure to withstand large thermal stress in the case where the temperature change is large.

(18) Consequently, a mount structure capable of withstanding development of cracks even in the joining of a large area can be obtained by the mount structure according to Embodiment 1.

(19) In the case where the mount structure is used for devices such as an LED, the thickness of devices is increased when the thickness of the mount structure is large. The thickness between the first solder joining portion 104 to the second solder joining portion 110 is preferably within a range of 30 to 100 m.

(20) Members forming the mount structure will be explained below.

(21) <Ceramic Substrate>

(22) The ceramic substrate 101 may be, for example, an AlN substrate. Specifically, an AlN chip of 10 mm square or 15 mm square can be used.

(23) <Ceramic Substrate Electrode>

(24) The ceramic substrate electrode 102 is provided on the ceramic substrate 101, which may be, for example, a plating layer. The ceramic substrate electrode 102 may also be a two-layer plating layer including a Ni-plating as a lower layer and an Au-plating as an upper layer.

(25) <Metal Substrate>

(26) The metal substrate 113 may be, for example, a Cu plate.

(27) <Metal Substrate Electrode>

(28) The metal substrate electrode 112 is provided on the metal substrate 113, which may be, for example, a plating layer. The metal substrate electrode 112 may also be a two-layer plating layer including a Ni-plating as a lower layer and an Au-plating as an upper layer.

(29) <Laminate>

(30) The metal structure 114 is formed by joining the ceramic substrate electrode 102 of the ceramic substrate 101 and the metal substrate electrode 112 of the metal substrate 113 by the laminate 117. The laminate 117 is formed by stacking the first interface layer 103-1, the first solder joining portion 104, the second interface layer 103-2, the first buffer material electrode 106, the buffer material 107, the second buffer material electrode 108, the third interface layer 103-3, the second solder joining portion 110 and the fourth interface layer 103-4 in this order from the ceramic substrate electrode 102 toward the metal substrate electrode 112. A thickness of the laminate 117 is 30 m or more and 100 m or less. Respective layers included in the laminate 117 will be explained below.

(31) <First Solder Joining Portion>

(32) The first solder joining portion 104 may be a plating layer. A material for the layer may be, for example, Au-20Sn (an AuSn alloy with a composition of 20 wt % Sn) plating. The material is not limited to Au-20Sn. The thickness of the first solder joining portion 104 can be 1 m or more and 10 m or less.

(33) The first interface layer 103-1 is formed as an interface layer between the ceramic substrate electrode 102 and the first solder joining portion 104. For example, in an interface between the Au plating as the upper layer of the ceramic substrate electrode 102 and the Au-20Sn plating of the first solder joining portion 104, the first interface layer 103-1 which is richer in Au than Au-20Sn is formed.

(34) <Buffer Material>

(35) As the buffer material 107, for example, Cu, Ag and Al can be used. Moreover, Cu foil, Ag foil and Al foil may be used. The Cu foil may have, for example, an elastic modulus of 75 GPa, a linear expansion coefficient of 16.2 ppm/K and a thickness of 35 m. The first buffer material electrode 106 and the second buffer material electrode 108 may be plating layers, for example, Ni plating, which are applied to both surfaces of the Cu foil as the buffer material 107. A difference in elastic moduli between the first solder joining portion 104 and the buffer material 107 can be within 62%. Furthermore, a difference in linear expansion coefficients between the first solder joining portion 104 and the buffer material 107 can be within 62%.

(36) The second interface layer 103-2 is formed as an interface layer between the first solder joining portion 104 and the first buffer material electrode 106. For example, in an interface between the Au-20Sn plating of the first solder joining portion 104 and the Ni plating of the first buffer material electrode 106, the second interface layer 103-2 containing three components of AuSnNi is formed.

(37) <Second Solder Joining Portion>

(38) The second solder joining portion 110 may be a plating layer. A material may be, for example, Au-20Sn (an AuSn alloy with a composition of 20 wt % Sn). The material is not limited to Au-20Sn. A thickness of the second solder joining portion 110 can be 1 m or more and 10 m or less. A difference between the thickness of the first solder joining portion 104 and the thickness of the second solder joining portion 110 can be within 25%.

(39) The third interface layer 103-3 is formed as an interface layer between the second buffer material electrode 108 and the second solder joining portion 110. For example, in an interface between the Ni plating of the second buffer material electrode 108 and the Au-20Sn plating of the second solder joining portion 110, the third interface layer 103-3 containing three components of AuSnNi is formed.

(40) The fourth interface layer 103-4 is formed as an interface layer between the second solder joining portion 110 and the metal substrate electrode 112.

(41) (Manufacturing Method of Mount Structure)

(42) The mount structure according to Embodiment 1 will be fabricated by the following procedure.

(43) FIG. 2 is a cross-sectional view showing a first solder 115 to a second solder 116 in parallel with respect to a thickness direction before the mount structure according to Embodiment 1 is joined.

(44) (1) The ceramic substrate 101 and the metal substrate 113 are prepared. The ceramic substrate electrode 102 is provided on one side of the ceramic substrate 101. The metal substrate electrode 112 is provided on one side of the metal substrate 113.

(45) (2) The buffer material 107 is prepared. The first buffer material electrode 106 and the second buffer material electrode 108 are provided on both surfaces of the buffer material 107, and further, the first and second solders 115 and 116 are respectively provided on the surfaces thereof. The first and second solders 115 and 116 are fabricated by, for example, plating, a material of which is Au-205n plating and a thickness of which is 1 m or more and 10 m or less. The Cu foil to which Ni-plating is applied is prepared as the first and second buffer material electrodes 106 and 108 under these layers. The Cu foil portion will be the buffer material 107 after joining. Thicknesses of the first and second buffer material electrodes 106 and 108 are equivalent to each other and, for example, 1 m or more and 3 m or less. The Cu foil having an elastic modulus of 75 GPa, a linear expansion coefficient of 16.2 ppm/K and a thickness of 35 m is used in Embodiment 1.
(3) Next, the ceramic substrate 101 and the buffer material 107 are installed so that the ceramic substrate electrode 102 faces the surface of the first solder 115. Then, the buffer material 107 and the metal substrate 113 are installed so that the metal substrate electrode 112 faces the second solder 116.
(4) Subsequently, these components are heated to 310 C. which is higher than 280 C. as a melting point of Au-20Sn of the first and second solders 115 and 116 in the above state to melt the first and second solders 115 and 116. Accordingly, the ceramic substrate electrode 102 and the first buffer material electrode 106 are joined through the first interface layer 103-1 and the second interface layer 103-2 on respective surfaces thereof. Also, the metal substrate electrode 112 on the metal substrate 113 and the second buffer material electrode 108 are joined through the third interface layer 103-3 and the fourth interface layer 103-4 on respective surfaces thereof. The mount structure 114 is formed by cooling the entire structure.

EXAMPLES AND COMPARATIVE EXAMPLES

(46) Mount structures fabricated and evaluated in Examples 1 to 3 and Comparative Examples 1 to 4 are shown in Table 1. In Examples 1 to 3 and Comparative Examples 1 to 4, the thickness of Au-20Sn was changed and effects on the thicknesses of the first and second solder joining portions 104 and 110 were evaluated.

(47) In Examples 1 to 3 and Comparative Examples 1 to 4, mount structures having large areas of 10 mm square and 15 mm square to join were fabricated and verified. Here, the ceramic substrate 101 is an AlN chip with 10 mm square or 15 mm square. The ceramic substrate electrode 102 has Au plating on the outermost layer and Ni-plating under the outermost layer on a back surface of the ceramic substrate. The metal substrate 113 is a Cu plate, which has the metal substrate electrode 112 having Au plating on the outermost layer and Ni-plating under the outermost layer on a back surface thereof.

(48) <Check of Inclination State>

(49) Initial states of inclination in the fabricated mount structures were checked by using a measuring microscope. States in which inclination is 10 m or less is determined as good, and states in which inclination is larger than the above is determined as poor.

(50) <Temperature Cycle Test>

(51) After the initial states were checked, repeated temperature cycle tests were performed. Liquid-tank type temperature cycle tests were performed at 40 C. to 125 C. for 1000 cycles while holding 5 minutes respectively, and observation of cracks was performed by a scanning acoustic tomograph (SAT) after completing the tests. Observation results were also shown in Table 1, and whether the joining after development of cracks was secured or not was determined by the following criteria. That is, an area ratio of cracks was or less of the entire area in both sizes of 10 mm square and 15 mm square in the ceramic substrates 101 was determined as excellent, the area ratio of cracks was higher than of the entire area in any of sizes of 10 mm square and 15 mm square and or less was determined as good and the area ratio of cracks was higher than of the entire area was determined as poor.

(52) TABLE-US-00001 TABLE 1 Thickness of Thickness of first solder second solder joining potion joining portion Development (m) (m) inclination of cracks Example 1-1 1 1 good excellent Example 1-2 5 5 good excellent Example 1-3 10 10 good excellent Comparative 1 5 poor excellent Example 1-1 Comparative 1 10 poor excellent Example 1-2 Comparative 5 1 poor excellent Example 1-3 Comparative 10 1 poor excellent Example 1-4

(53) Referring to Table 1, initial inclinations are small and determined as good in all Examples 1-1 to 1-3. However, initial inclinations are large and determined as poor in Comparative Examples 1-1 to 1-4. Furthermore, wetting spreading states of solder at the time of melting are not uniform and inclinations are large in Comparative Examples 1-1 to 1-4.

(54) Development of cracks is small and determined as excellent in all Examples 1-1 to 1-3 as well as in Comparative Examples 1-1 to 1-4.

(55) Accordingly, when comparing Examples 1-1 to 1-3 with Comparative Examples 1-1 to 1-4, it can be found that the thickness of the first solder joining portion 104 and the thickness of the second solder joining portion 110 are preferably equivalent as a preferable condition for producing effects of the mount structure according to the present disclosure.

Examples 2-1 to 2-8, Comparative Examples 2-1 to 2-2

(56) In Examples 2-1 to 2-8 and Comparative Examples 2-1 to 2-2, evaluations were performed by changing the material of the buffer material 107 for evaluating effects of physical properties of the buffer material 107.

(57) Evaluation levels in Examples 2-1 to 2-8 and Comparative Examples 2-1 to 2-2 are shown in Table 2. Thicknesses of the first and second solder joining portions 104 and 110 are respectively 5 m and the fabrication method and the evaluation method of the mount structures are the same as those in Example 1.

(58) TABLE-US-00002 TABLE 2 Buffer material Linear Elastic expansion Thickness modulus coefficient Crack Material (m) (GPa) (ppm/K) Inclination development Example 2-1 Cu 35 75 16.2 good excellent Example 2-2 Cu 70 75 16.2 good excellent Example 2-3 Cu 35 110 16.2 good good Example 2-4 Cu 70 110 16.2 good good Example 2-5 Ag 35 76 19.3 good excellent Example 2-6 Ag 70 76 19.3 good excellent Example 2-7 Al 35 68.3 23.7 good good Example 2-8 Al 70 68.3 23.7 good good Comparative Ni 35 207 15 good poor Example 2-1 Comparative Ni 70 207 15 good poor Example 2-2

(59) Referring to Table 2, initial inclinations are small and determined as good in all Examples 2-1 to 2-3 as well as Comparative Examples 2-1 and 2-2. Accordingly, it can be found that the material of the buffer material 107 does not affect the initial inclination.

(60) The development of cracks after performing repeated temperature cycle tests is small in Examples 2-1 to 2-8 and determined as excellent or good. In particular, the development of crack is extremely small in Examples 2-1, 2-2 using Cu with low elasticity and in Examples 2-5, 2-6 using Ag as the buffer material 107, which are determined as excellent.

(61) On the other hand, the development of cracks is relatively large in Comparative Examples 2-1 and 2-2 using Ni as the buffer material 107 and determined as poor.

(62) Cu foil used as the buffer material 107 in Examples 2-1 and 2-2 has a low elastic modulus, and a difference as a comparison with 68 GPa as an elastic modulus of Au-20Sn is 10%. The linear expansion coefficients are extremely similar and a difference is 1%. Differences in elastic moduli and linear expansion coefficients between Ag foil used as the buffer material 107 in Examples 2-5 and 2-6 and Au-20Sn are respectively 3% and 20%.

(63) This is because stress concentration does not occur and the joining of the Au-20Sn solder with high strength can be maintained as the elastic modulus and the linear expansion coefficient of the buffer material 107 are close to those of the first and second solder joining portions 104 and 110.

(64) Differences in elastic moduli and the linear expansion coefficients between materials used as the buffer material 107 in other examples and comparative examples and the Au-20Sn solder used as the first and second solder joining portions 104 and 110 are as follows. Differences in elastic moduli and linear expansion coefficients between the Cu foil used as the buffer material 107 in Examples 2-3 and 2-4 and the Au-20Sn solder used as the first and second solder joining portions 104 and 110 are respectively 63% and 3%. Differences in elastic moduli and linear expansion coefficients between the Al foil used as the buffer material 107 in Examples 2-7 and 2-8 and the Au-20Sn solder used as the first and second solder joining portions 104 and 110 are respectively lower than 1% and 48%. Differences in elastic moduli and the linear expansion coefficients between the Ni foil used as the buffer material 107 in Comparative Examples 2-1 and 2-2 and the Au-20Sn solder used as the first and second solder joining portions 104 and 110 are respectively 204% and 6%.

(65) According to the results of the examples and the comparative examples, the difference in elastic moduli and the linear expansion coefficients between the first and second solder joining portions 104, 110 and the buffer material 107 are preferably within 62% as a preferable condition for producing effects of the mount structure according to the present disclosure. As a particularly preferable condition, it is preferable that the difference in elastic moduli and linear expansion coefficients between the first and second solder joining portions 104, 110 and the buffer material 107 are within 20%.

(66) The present disclosure includes suitable combinations of arbitrary embodiments and/or examples in the above various embodiments and/or examples, and advantages possessed by respective embodiments and/or examples can be obtained.

(67) The mount structure according to the present disclosure has a joining which can withstand development of cracks generated by thermal stress due to repeated temperature changes. Accordingly, it is possible to provide a mount structure without inclination and excellent in the heat dissipation property for the joining of a large area in devices generating large heat such as an LED or power devices, which may contribute to improvement of performance in devices.