Thursday, May 8th, 2008
 |
Posted by Amanda Whittemore
|
Have you seen the statement on Indium’s product data sheet for liquid metals that says:
“Care should be taken in reheating the alloy in the original packaging provided. Temperatures should not exceed 65.6ºC”?
Recently there has been some confusion regarding this statement, so I’d like to make a clarification the current and potential liquid metal users out there. This statement does not mean that these alloys cannot be used above 65C. The reason for the statement is to address the fact that the gallium alloys increase in corrosiveness above this temperature and the bottles that the alloys are supplied in are not rated to withstand that corrosiveness.
It is suggested that you avoid heating the gallium alloys above 65.6 ºC while they are in their original packaging, but once they are dispensed on your corrosion-resistant substrate, they should not cause an issue.
Posted by Amanda Whittemore at 7:54 AM (4 days ago)
Tuesday, May 6th, 2008
|
Posted by Amanda Whittemore
|
Continuing with the topic of thermal interface material options, a popular material choice is a clad metal thermal interface preform including multiple layers of various metals, or a metal preform clad with an adhesive synthetic material. For specific applications where an interface is sandwiched between two drastically different substrates, clad metal preforms are great. Using clad preforms containing multiple metals, it is possible to clad a stiff material on one side which prevents deforming and a soft, conductive material on the other.
While there are applications that these materials cater to, it is important to realize the full impact of using such a TIM material.
Bob Jarrett recently wrote a summary of the resistance impact of adding an adhesive layer to an existing metal interface material.
Bob wrote:
The adhesive on the interface will act as another interface layer, increasing the thermal resistance, additively. The typical acrylic polymer adhesive has a thermal conductivity of ~0.1 W/m-K. If this adhesive is applied as a typical ½ mil (13 μm) film, the resistance increase is:
R = x/k
= (13/10000) cm / (0.1/100) W/cm-K
= 0.13 cm²-°C/W.
This is added to the 0.06-0.08 cm²-°C/W for the Heat-Spring a total resistance of about 0.2 cm²-°C/W. The thin layer of adhesive more than doubles the resistance.
This posting is not meant to deter you from using cladded interface materials, just to help you realize the potential effects of using such a material. Every layer added to the interface also adds some amount of resistance.
Posted by Amanda Whittemore at 15:00 PM (6 days ago)
Monday, May 5th, 2008
|
Posted by Amanda Whittemore
|
Bob Jarrett gave an excellent explanation of the thermal conductivity of metal alloys and their relationship with electrical conductivity. It is as follows:
The high thermal conductivity of metals is a result of energy transfer by free (or valence) electrons. These mobile electrons conduct electricity and thermal energy. For metals, the ratio of the thermal conductivity to the electrical conductivity is the Lorenz constant or the Weidemann-Franz ratio (L = κ/σT ~ 2.45E-8 WΩ/K²). When metals are mixed to form an alloy, the energy states of the free electrons are modified as the different elements have differing affinities for the wandering electrons.
The electrical and thermal conductivity of the mixture often deviates (negatively) from a linear rule of mixing. For example, Sn-Cu forms two intermetallic compounds (Cu3Sn and Cu6Sn5). The conductivity relationship still holds, but both the thermal and electrical conductivity are less than either constituent. Similarly in the Sn-Ag system, the conductivity of the 96.5Sn-3.5Ag eutectic composition is lower than either element.
Posted by Amanda Whittemore at 10:53 AM (May 5th, 2008)
Friday, May 2nd, 2008
|
Posted by Amanda Whittemore
|
Calculating the thermal conductivity of an alloy is not done by taking an average (or weighted average) of its elemental constituents. When two metals are mixed, their properties dramatically change, often in ways which would be difficult to speculate.
For example, the thermal conductivity of silver metal is approximately 429W•m−1•K−1 (according to Wikipedia) and tin metal is 73W•m−1•K−1. Both of these values are very high. Surprisingly, the conductivity of Indalloy #121 (96.5Sn 3.5Ag) is a considerably lower 33W•m−1•K−1.
This may not seem logical, but is correct. The thermal conductivity of an alloy is impacted more by the chemical bonding structure between the inclusive elements than by the measured conductivity each element exhibits alone. The bonding orientation of the metals affect the rate which heat can pass through the material.
Thermal conductivity of common metal alloys can be found in Indium’s Solder and Alloy Directory.
Posted by Amanda Whittemore at 10:33 AM (May 2nd, 2008)
Thursday, May 1st, 2008
|
Posted by Amanda Whittemore
|
The typical extreme device needs an advanced thermal interface material to keep the interface cool while the device generates extreme amounts of heat. As discussed in my previous posting, the interface material chosen for extreme conditions often melts at a very high temperature so that it remains solid and stable during high temperature operation.
Other types of extreme devices have an opposite issue, a low reflow temperature requirement. A significant consideration when choosing a solder TIM is whether the entire package will withstand soldering temperatures. As indium supplies TIM materials with high temperatures, we have also studied and understand low temperature metals.
At the very lowest temperatures, there are metal alloys which are liquid at room temperature. These alloys contain indium, gallium, and tin. These materials can be deposited and remain molten during the entire life of the product. They do not require any elevated temperatures for their application.
For those thermal designers who are not ready to take the liquid metal leap, there are also alloys available for use in thermal interfaces which melt at approximately 60C. These alloys will wet a substrate, forming an interface with very low resistance.
Unfortunately, without a flux, these low temperature alloys are unlikely to form any intermetallic with the heat sink or spreader. If the interface material alone is to provide mechanical stability, an intermetallic is needed. The flux is used to remove oxides from the substrates, allowing for intermatallic formation. Most fluxes do not activate until temperatures or 125C. There are some fluxes available which activate as low as 100C however. For more information on these, please read Jim Hisert’s solder blog on this topic.
Posted by Amanda Whittemore at 11:19 AM (May 1st, 2008)
View All Entries
