Materials having two surfaces with different coefficients of thermal expansion

Abstract

A body comprising at least two components having one or more different properties and a method of producing the same are disclosed. One of the body components is in the form of particles with optional adhesive interlayers. A second of the components has a surface locally melted in a predetermined pattern and only to a predetermined depth by scanning an electron beam there across to incorporate the particles and form a metal composite film. Thereby, a predetermined volumetric concentration of the incorporated particles varies continuously from the locally melted surface so as to provide two surfaces in the body having different coefficients of thermal expansion.

Claims

1. A metal composite film, comprising a first component and a second component, wherein the first component is comprised of particles, the second component is a metal or metal alloy body, and the particles are distributed on a surface of the metal or metal alloy body and also incorporated within the metal or metal alloy body to a predetermined depth under the surface of the metal or metal alloy body such that a predetermined volumetric concentration of the incorporated particles varies continuously throughout the film from the surface of the body toward an opposite surface of the body so that the film has a tuned gradient of coefficient of thermal expansion values, wherein an electron beam produced by a superconducting linear electron accelerator is moved relative to the surface to locally melt the surface and the predetermined depth under the surface of the metal or metal alloy body to receive the particles.

2. The body of claim 1, wherein the second of the components has a thermal conductivity lower than that of the particles and a higher coefficient of thermal expansion than that of the particles.

3. The body of claim 1, wherein the body is operatively associated with a heat generating device such that the second of the components beyond the predetermined depth is a heat sink.

4. The body of claim 1, wherein the second of the components is comprised of copper or copper alloy, and the incorporated particles are constituted of diamond to form the metal composite film as a copper diamond film.

5. The body of claim 1, wherein the first of the components includes adhesive interlayers that are one of chromium and titanium.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) These and other objects, advantages and features of this invention will be apparent from the following specification thereof taken in conjunction with the accompanying drawings in which:

(2) FIG. 1 is a general schematic of a conventional Ebeam-LINAC used to carry out materials processing;

(3) FIG. 2 is a conventional and representative copper heat spreader mounted on finned heat sink;

(4) FIG. 3 is a computational plot of the electron trajectories in a copper diamond substrate;

(5) FIG. 4 is a computational plot of the energy deposition profile in a copper diamond substrate;

(6) FIG. 5 is an example of the computational prediction of the three-dimensional thermal distribution in a copper-diamond heat spreader showing the melting of copper in desired process region;

(7) FIG. 6 is an example of the computational prediction of the temperature profile for the process region of a copper diamond substrate; and

(8) FIG. 7 is a table of example ranges of parameters suitable for processing copper diamond composite for both superconducting and non-superconducting beams;

DETAILED DESCRIPTION OF THE DRAWINGS

(9) The properties of a high thermal conductivity and tunable-CTE heat spreader can be achieved using a copper-diamond film. This results from the fact that copper has a high thermal conductivity (401 W/(m.Math.K)) and a relatively high CTE (1710.sup.6/ C.), while diamond has an extremely high thermal conductivity (2200 W/(m.Math.K)) and a relatively low CTE (110.sup.6/ C.).

(10) A copper-diamond film has a very high thermal conductivity and a CTE between that of copper and diamond. The exact thermal conduction is a direct function of the relative quantities of copper and diamond used. Most pure metals and other materials used in thermal control applications, like silicon, have a CTE somewhere in this range. The diamond content can be varied in the film spatially to create a gradient of CTE values. However, up until now, it was not possible to reliably produce such a film using known techniques. The approach can also be used to create other metal-diamond films, such as aluminum-diamond films. It is of course possible to use this localized thermal processing technique with any number and type of materials. Copper-diamond is the example used here due to the extremely high cost of producing materials of this type using conventional approaches and therefore as a means to illustrate the benefits of the this new approach.

(11) We have found that the copper-diamond film can now be produced using electron beam processing in a way that was previously impractical or unachievable. The desired tuned heat spreader can be formed using copper and diamond particles that are put in a cast of a particular shape, and then sintered using an electron beam.

(12) In order for the necessary gradient to be preserved during processing, we have also found that the copper should be melted locally instead of melting the entire piece at once in order to minimize the flow of diamond particles within the copper. The electric field effects of the electron beam will enhance wetting and adhesion of the copper to the diamond particles. Alternatively, adhesive interlayers such as chromium, titanium or other materials can be use to first coat the diamond particles prior to their incorporation into the copper matrix. Our novel method for designing, simulating, and producing a copper diamond composite is demonstrated using the following example.

(13) The first step in the process is to outline all of the design parameters. The properties of the electron beam are: accelerating voltage, scan width, beam angle, and working distance.

(14) These parameters, along with the target material properties and simulation parameters, such as the number of electrons trajectories, are used in a Monte Carlo statistical simulation to predict the most probable trajectories of electrons in the same to be treated and determine backscattering, generation of X-rays, and energy deposition.

(15) FIG. 3 shows an exemplary simulation of trajectories from a 5 MeV electron beam in copper-diamond substrate target. The simulation consists of 200,000 electron trajectories, only a portion of which are shown. The thin lines show trajectories of electrons that terminate within the target, while the solid lines that originate and terminate at the top edge show the trajectories of electrons that are backscattered. This simulation indicates that the approximate penetration depth of the electron beam is 3 mm.

(16) The energy deposition profile for this example simulation is shown in FIG. 4. The innermost contour represents a region that contains 10% of the total energy deposited by the beam. The outermost contour represents a region that contains 95% of the total energy deposited by the beam.

(17) A Monte Carlo simulation of the trajectories can be used along with a thermal analysis to develop a profile of the energy deposited in the solid as a function of position. This profile is essentially a snapshot in time for an electron beam. Coupling these results with the beam current and the scan rate provides an accurate description of the total amount of energy deposited in the sample over time for a given process.

(18) As is well known in the art, the thermal analysis of the effect of the E-Beam on the sample can be used to generate plots such as the 3D temperature profile shown in FIG. 5 where higher temperatures are shown both by darker regions and height to provide a graphical representation of temperature gradients. The plot shows the thermal profile for a copper diamond sample with the darkest regions at the plateau representing the region that has melted.

(19) Our method of linking Monte Carlo simulations to thermal models provides a valuable tool for determining the requirements of a specific process. Without linking the two, manual drawing of the geometry would be required, applying heat generation terms for each element, and prescribing boundary conditions for each design case.

(20) The following describes an analysis for the production of a 1.2 mm thick copper heat spreader using electron beam processing. The nominal thickness of 1.2 mm was chosen because it is a typical value for current heat spreader technology.

(21) Using the methods described above, the temperature profile of the process region was obtained and is shown in FIG. 6. The z axis at the left edge of the figure represents the centerline of the beam and is an axis of symmetry, while the r axis represents the radial coordinate. The origin is located at the surface where the beam penetrates the material.

(22) The parameters selected for this process are listed below:

(23) TABLE-US-00001 Substrate Material: Copper-Diamond Accelerating voltage: 5.0 MeV Beam current: 1 mA Beam radius: 0.1 mm Exposure time. 16 msec Energy deposited: 4922 keV Power consumption: 4922 W Processed depth: 1.2 mm Processing rate: 20 mm.sup.2/sec

(24) The accelerating voltage of 5 MeV was chosen to match the desired process depth of 1.2 mm. Even though the beam penetrates approximately 3 mm, most of the energy is concentrated near the surface. Therefore, the top 1.2 mm of material will melt regardless of the total depth of the substrate.

(25) The beam radius of 0.1 mm was chosen because it resulted in a process region of approximately 0.5 mm by 0.5 mm, which is appropriate for this size substrate.

(26) The beam current and exposure time were chosen to effectively melt the copper in the process region with the least amount of power. The beam current for this process is 1 mA and the exposure time is 16 msec. The exposure time indicates the amount of time that the beam radiates in a given process region, in this case a 0.5 mm radius. Both of these values are within the capabilities of current electron beam technology.

(27) The energy deposited in the material by the electron beam is 4922 keV, indicating an efficiency of 98.4% for the 5 MeV beam. The process consumes approximately 4922 W of power and the processing rate is 20 square millimeters of surface area of the copper substrate per second. This indicates that a typical heat spreader could be processed in a few minutes at a very high efficiency.

(28) The above described materials processing can be performed by either a normal conducting or superconducting electron beam. The beam parameters depend on the processing rate, intensity and the type of electron beam. Example ranges of parameters suitable for processing copper diamond composite are shown in FIG. 7. Both superconducting and normal conducting electron beams can be modeled and optimized for materials processing applications using a predictive simulation algorithm as above described. The optimal type of electron beam for each application is determined by the trade-offs between capital expenses, operating expenses and throughput.

(29) While we have shown and described a currently preferred embodiment in accordance with our invention, it should be understood that the same is susceptible of other changes and modifications as will now be apparent to one skilled in the art without departing from the spirit of our invention. Therefore, we do not intend to be limited to the details shown and described herein but intend to cover all such changes and modifications as may fall within the scope of the appended claims.