JPH02213452A - Low thermal expansion and high heat conductivity alloy - Google Patents

Abstract

PURPOSE:To produce the alloy having a low thermal expansion characteristic, high heat conductivity and excellent workability by mixing the powders of the metals respectively having specified average thermal expansion coefficient and heat conductivity and binding the mixture into a specified state by sintering. CONSTITUTION:The powder of one or >=2 kinds among the metals and alloys having <=6X10<-6>/ deg.C average thermal expansion coefficient from 20 to 150 deg.C and <=60W/mK heat conductivity (1W/mK=0.24cal/sec.cm.deg) and the powder of one or >=2 kinds among the metals and alloys having >=11X10<-6>/ deg.C average thermal expansion coefficient from 20 to 150 deg.C and a 150W/mK heat conductivity are mixed. The mixture is then bound by sintering to obtain an alloy in which the respective powders have a fine alloyed mixed structure only at their interface and the powders of the metals and alloys having >=150W/mK heat conductivity have a continuous layer. By this method, the low thermal expansion and high heat conductivity alloy having 6-11X10<-6>/ deg.C average thermal expansion coefficient from 20 to 150 deg.C and 60-150W/mK heat conductivity is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an alloy that has both low thermal expansion properties and high thermal conductivity properties and is used in ultra-high integrated circuits and the like.

[Conventional technology]

There has been a need for new materials with new functions that did not previously exist, mainly in the electronics field. For example, materials with both low thermal expansion and high thermal conductivity are required for housings and packaging materials for ultra-highly integrated circuits.

However, as can be seen from FIG. 1 showing the coefficient of thermal expansion and thermal conductivity, there are materials with properties corresponding to the shaded area in the figure that satisfy both the low thermal expansion and high thermal conductivity properties that the present invention aims to obtain. I can't find it.

W and Mo are used as single metals, and W and Mo are used as alloys.
Only alloys based on this show values close to the required properties.

On the other hand, among compounds, ceramic materials such as SiC-based and AIN-based materials have close to the required characteristics, and some of them are currently in practical use.

[Problem to be solved by the invention]

However, the above-mentioned W, MO, and alloys based on these have extremely high material costs, are difficult to plastic work, weld, and join, and have a large specific gravity that makes them difficult to handle.
There are problems that make it difficult to use as an industrial material.

In addition, in the case of metallic materials, attempts are usually made to obtain desired properties by melting and alloying multiple metal elements, but in terms of physical properties,
A that generally forms a total solid solution.

When 82 types of metals are alloyed, the thermal expansion coefficient changes linearly depending on the composite side, but the thermal conductivity is lower than the composite side, and it is impossible to satisfy these two types of physical properties at the same time by alloying. The reality is that there is no practical alloy that meets the above-mentioned required properties based on the conventional concept.

On the other hand, ceramic materials such as SiC and AIN have problems in processability, reliability of mechanical properties, cost, etc., and are not necessarily satisfactory materials.

An object of the present invention is to provide an alloy that is inexpensive, has excellent workability, and satisfies both low thermal expansion and high thermal conductivity as an industrial material for semiconductor devices that require high reliability.

[Means to solve the problem]

The present invention differs in concept from the past, and provides an interface in which one or more metal powders or alloy powders selected from each of a group with a low coefficient of thermal expansion and a group with a high thermal conductivity are combined. An alloy that is alloy-bonded only at the bonding interface, with virtually no diffusion of alloying elements other than the bonding interface.
This is a new functional alloy that combines low thermal expansion and high thermal conductivity, which have never existed before, by combining the low thermal expansion and high thermal conductivity characteristics of each.

The alloy of the present invention is a metal material that has much superior industrial properties in terms of not only processability but also excellent mechanical properties and economic efficiency compared to heavy alloys and ceramic materials such as MOLW alone or based on them. It is.

That is, the first invention has an average coefficient of thermal expansion from 20°C to 150°C of 6 to 11 x 10'/'C1 and a thermal conductivity of 60 to 150 W/mK (1W/mK = 0.
24 cal/sec-an-deg), the second invention is a low thermal expansion and high thermal conductivity alloy, characterized by:
The average coefficient of thermal expansion from 20℃ to 150℃ is 6XIO-
Powder of one or more metals and alloys with a thermal conductivity of 60 W/a+ or less at 6/°C or less and 20°C
The average coefficient of thermal expansion from to 150℃ is 11 x 10'
/''C or higher and a thermal conductivity of 150 W/wK or higher and one or more powders of metals and alloys are mixed and then combined by sintering to form an alloy only at the interface of each powder. It has a fine mixed structure that is bonded,
Furthermore, the low thermal expansion and high thermal conductivity alloy according to the first invention has a continuous layer of powder of one or more of the metals and alloys having a thermal conductivity of 150 W/aK or more,
Furthermore, a third invention is the low thermal expansion and high thermal conductivity alloy according to the second invention, in which the metal having a thermal conductivity of 150 W/mK or more is Cu.

In the present invention, a mixed layer low thermal expansion alloy can be used as the low thermal expansion powder. For example, an alloy called Invar has an average thermal expansion coefficient of 6X10=/'C or less from 20°C to 150°C. , Ni 34-3
Fe-Ni alloy consisting of 8% and the remainder substantially Fe, (2) (1) Part of the Ni of Fe-Ni alloy is M
n < 2, 0%, C < 0.5%, Or < 2%
, Ti < 2.5%, Cu < 4%, (3) Fe-N1-Goo alloy consisting of Ni 29-40%, Co (12%, the balance substantially Fe) ( 4) Part of the Ni of the Fe-N1-Go alloy in item (3) is M n < 2, 0%, C < 0.5%, Cr <
(5) Coo-Fe-Cr consisting of 34-38% Fe, 8-11% Cr, and the remainder substantially Go. Alloy, (6) (5) Fe-Cr-G
%, and alloys substituted with one or more of Cu<4%.

Furthermore, it is effective to use inexpensive Cu or the like as a high thermal conductivity alloy. For example, the thermal conductivity is 200 to 400 W/mK
(7) The purity of Cu is 99% with electrolytic copper, deoxidized copper, and oxygen-free copper.
or more pure Cu, (8) Cu with Cd<2%, Si<1%, Zr<1
%, Cu alloys to which one or more types of AI<10% are added, etc.

The low thermal expansion and high thermal conductivity alloy of the present invention can be realized by alloy-bonding these low thermal expansion powders and high thermal conductivity powders only at the interfaces of the respective powders. In other words, the two types of functions of low thermal expansion and high thermal conductivity are achieved by alloying and bulking only at the interface of each powder within a range where the respective physical properties do not substantially change.

In order to effectively combine the above two types of functions, it is necessary that the low thermal expansion powder has a metal structure in which the interparticle gaps are filled with a highly thermally conductive material. Furthermore, it is desirable that the residual porosity is 5% or less.

Here, the above-mentioned range in which the physical property values of each powder do not substantially change means that two different types of powder are alloyed only at the bonding interface, and other than the bonding interface is in the other particle. This refers to the fact that one metal element does not diffuse. More specifically, it is desirable that the amount of diffusion of metal elements from different particles within 10 μm from the interface between the particles is within 5% by weight. In particular, the high thermal conductivity materials mentioned above are C
F contained in the low thermal expansion material in the Cu phase
Experiments have confirmed that when alloying elements such as e, Ni, and Go are diffused into a solid solution, the thermal conductivity decreases rapidly.

In this way, diffusion of metal elements between powders results in a significant decrease in both desired properties, but it is effective to perform alloy bonding only at the bonding interface of each powder in order to ensure the mechanical properties of the composite material. It is.

To achieve alloy bonding only on the bonding surfaces of the above-mentioned powders, 1. For example, hot press, hot isostatic pressing (HIP)
However, the sintering conditions are determined depending on the type and amount of powder to be mixed, as well as the powder particle size and shape.

Although it is effective to form an oxide film or a nitride film on the powder surface and then sinter it as a method for preventing the diffusion of elements between particles, this is not always necessary.

By the way, in order to obtain a composite material that simultaneously satisfies both the properties of the first invention of the present application, it is necessary to select starting materials that take into account the reduction in both properties due to alloy joining. Therefore, in the second invention of the present application, powders satisfying each characteristic are selected as starting materials, and the powders are mixed and then sintered.

〔Example〕

Examples will be described in detail below.

Example 1 First, an example of the infiltration method used to produce a comparative alloy will be described.

A water atomized powder of a so-called super invar alloy consisting of 32.3% Ni, 5.5% Go, and the balance Fe and unavoidable impurities by weight was prepared. - 3 and 4 tons of this powder with 100 mesh and average particle size of 52 μl, respectively.
After compacting with the molding pressure of
A test piece of M) was cut out.

The porosity of each test piece was 48% for the former and 27% for the latter. Same cross-sectional shape as this test piece (30W x 45L)
The electrolytic Cu powder was molded. At this time, the amount of Cu powder used is 1 of the weight ratio of Cu required to fill the voids with Cu.
．． It was set to 1 times. C on the top of the super invar alloy molded body
U molded bodies are loaded and heated at 1130°C in a dry H1 atmosphere.
Infiltration was performed for 40 iin. It was confirmed that 99.2% or more of the weight of Cu added after cooling was infiltrated into the super invar alloy compact. Density, thermal expansion, thermal conductivity, and tensile test pieces were cut out from each infiltrated test piece, and the physical properties and mechanical properties of each item were measured. The results will be described below together with the results of subsequent examples.

Next, the HIP method for manufacturing the alloy of the present invention will be described.

An Ar gas atomized powder of a low thermal expansion alloy consisting of 32.1% Ni, 5.6% Go, the remainder Fe and unavoidable impurities was prepared by weight, and classified using -20 mesh.

The shape of the powder is substantially spherical, with an average particle size of 270μ
It was m. To this powder, 27% and 38% of spherical Cu powder was added by Ar gas atomization with an average particle size of 70μI, respectively.
%, 48%, and 58% by weight, and after canning, HIP treatment was performed at 800° C., 1000 atm, and 2 hours. Various test pieces were cut out from each compacted body, and a comparison test with the above-mentioned Cu molten material was conducted. The results are shown in Table 1.

The evaluation items are density ratio, average coefficient of thermal expansion from 20°C to 150°C, thermal conductivity (using the flash method, the surface of a flaky sample is irradiated with a xenon lamp to instantaneously heat it, and the rate of temperature rise on the back side of the sample is measured). calculation) and alloy diffusion amount. Here, the amount of alloy diffusion is the result of measuring the weight ratio of Cu and Fe at a point where each powder enters the inside of the powder by 10 μm from the bonding interface between the low thermal expansion alloy powder and the Cu powder. That is, low thermal expansion alloy powder was analyzed for Cu, and Cu powder was analyzed for Fe.

The results of Example 1 are shown in Table 1, and the results of plotting the average coefficient of thermal expansion and thermal conductivity against the Cu content are shown in FIG. In a comparison between the infiltration method and the HIP method, better results were obtained with the HIP method for each Cu content. That is, the HIP method provides a smaller average coefficient of thermal expansion and a larger value of thermal conductivity at each Cu content level. This tendency is particularly noticeable in the value of thermal conductivity; in the infiltration method, there is almost no increase in thermal conductivity as the amount of Cu increases.
The expected composite properties were not obtained.

On the other hand, in the HIP method, the average coefficient of thermal expansion changes on the composite side (proportional to the amount of each addition), and the thermal conductivity is lower than the predicted value from the composite side, but maintains high thermal conductivity. It is possible to do so.

Sample No. 048 has an average coefficient of thermal expansion of 9. lXlO4
/°C, and has a thermal conductivity of 83 W/@, physical properties that cannot be obtained with conventional metal or ceramic materials.

In the infiltration method, 7% or more of Ni is dissolved in Cu and 21% or more of Cu is dissolved in the super-invar alloy phase, which causes a decrease in thermal conductivity and an increase in thermal expansion. HI
In the F' method, C that has entered 10 μm from the interface of the Cu powder
Ni in u is 3.5% or less, and Cu at a position 10 μm from the interface in the super invar powder is also 1.2% or less, so interdiffusion of alloying elements is extremely low.

As can be seen from Table 1, the alloy produced by the infiltration method has sufficient properties because the density ratio is insufficient, the highly thermally conductive Cu layer is a discontinuous layer, and the amount of diffusion between mutual particles is large. is not obtained.

In addition, when the metallographic structure of the HIP-sintered alloys of sample No. G27 was observed, it was confirmed that Cu was a discontinuous layer. This is because the blending ratio of Super Invar powder was higher than that of Cu powder, so Cu, which is a highly thermally conductive material, could not sufficiently fill the gaps between Super Invar particles, and desired physical properties could not be obtained.

Example 2 Weight ratio: Ni 32.8%, Co 5.5%, Ti 1.
A spherical powder with a low coefficient of thermal expansion was prepared by gas atomization using N with an average particle diameter of 270 ILII, consisting of 2% Fe and unavoidable impurities. The results of ESCA analysis of the surface of the as-atomized powder show that there are 5
It was confirmed that TIN powder of 00A (Angstrom) or less was produced. The same C as in Example 1 was added to the same powder.
U powder was mixed at a weight ratio of 27%, 38%, 48%, and 58%, respectively, and subjected to HIP treatment under the same conditions as in Example 1.

The evaluation results are shown in Table 2. Comparing the results in Table 2 where TiN is formed on the surface of Super Invar powder and the results in Table 1 where TiN is not formed for alloys obtained at the same blending ratio, it is found that TiN is formed on the surface of Super Invar powder. It was observed that the thermal conductivity of the alloy was increased by about 20%. This is the T formed on the surface of Super Invar powder.
This is thought to be due to the effect of the iN layer in preventing diffusion of metal elements between powders.

In addition, when the metallographic structure of the HIP-sintered sample No. H27 was observed in the same manner as the No. G27 alloy of Example 1, it was confirmed that Cu was a discontinuous layer, and the desired physical property values were obtained. I couldn't.

Example 3 Four types of materials of 30sφX300mmQ obtained in Example 2 were cold swaged to 10mmφX2700ae.
Processing was performed up to R. A tensile test was conducted on this material in conjunction with the evaluation items in Tables 1 and 2.

The results are shown in Table 3. From 30mm diameter to 10mm by cold working
By reducing the pressure to mmφ, the microscopically detected pores disappear and the density ratio becomes 100% of the true density, the average coefficient of thermal expansion decreases slightly and the thermal conductivity decreases accordingly. It increased by 0%.

Mechanical properties as cold worked are 32 kg/s' tensile strength
Above, the elongation value showed 6% or more.

It should be noted that even though the HIP-sintered alloy sample No. C27 was subjected to cold working, it was confirmed that Cu was still a discontinuous layer, and desired physical property values could not be obtained.

〔Effect of the invention〕

According to the present invention, it simultaneously satisfies both properties of an average thermal expansion coefficient of 6 to 11 x 10''/'C1 from 20°C to 150°C and a thermal conductivity of 60 to 150 W/mK, and can be processed at low cost. It is possible to provide a metal material with excellent properties and is industrially useful.

[Brief explanation of the drawing]

Figure 1 shows the relationship between the coefficient of thermal expansion and thermal conductivity of existing materials (the shaded area is the metal composite material of the present invention), and Figure 2 shows the relationship between the coefficient of thermal expansion and thermal conductivity of existing materials. It is a figure which plotted against Cu content. 2 Shinka 5150'Cl21. n 3 Anti-KWI Foreign Rebellion (X
10/”c)

Claims (1)

[Claims] 1. The average coefficient of thermal expansion from 20°C to 150°C is 6 to 1.
1×10^-^6/℃ and thermal conductivity of 60 to 150W
/mK (1W/mK=0.24cal/sec・cm・
An alloy with low thermal expansion and high thermal conductivity, characterized in that: deg). 2 Average coefficient of thermal expansion from 20℃ to 150℃ is 6×1
A powder of one or more metals and alloys having a thermal conductivity of 60 W/mK or less at a temperature of 0^-^6/°C or less,
Average thermal expansion coefficient from 20℃ to 150℃ is 11×10
After mixing powders of one or more metals and alloys with a thermal conductivity of 150 W/mK or more at a temperature of 6/°C or higher, the powders are combined by sintering to form an interface between the respective powders. 2. The powder has a fine mixed structure in which the alloy is bonded only in the alloy, and further has a continuous layer of powder of one or more of metals and alloys having a thermal conductivity of 150 W/mK or more. Low thermal expansion, high thermal conductivity alloy. 3. The low thermal expansion and high thermal conductivity alloy according to claim 2, wherein the metal having a thermal conductivity of 150 W/mK or more is Cu.