Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Electrically non-conductive thermally conductive material

  1. Jul 5, 2006 #1
    Hi friends

    Is there any material which is electrically non-conductive but thermally highly conductive.

  2. jcsd
  3. Jul 6, 2006 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member


    Really, to get the best help here, you should describe your application and constraints.

    Extra tip : Go to matweb.com -> property search. You can then search for materials with thermal conductivity > X and electrical resisivity > Y.
    Last edited: Jul 6, 2006
  4. Jul 6, 2006 #3
    Actually I am trying to develop thermal control for a solar panel battery pack. When the batteries are extracting power they dissipate much heat and become very hot. We cant use a conductive material for wiping the heat.So please recommend some material which is a good heat conductor but electrical insulator.
  5. Jul 7, 2006 #4


    User Avatar
    Science Advisor
    Gold Member

    These are only qualitative quantities, high conductivity compared to what? You should decide what your requirements are before you go searching for a material.

    You could try coating something like aluminum with a non-conductive coating. Just a thought.

    BTW, why CANT you use a heatsink or something on the batteries? It's a fairly common practice I think.
  6. Jul 7, 2006 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The standard heatsinking trick is to use a thin layer of insulating material (eg: mica sheet) as an electrical spacer between the two high conductivity materials.

    But without a better description of a system (picture, specs, dimensions, requirements), it would be hard to help much more. It essentially comes down to doing the calculations and deciding what works for you.

    You want the electrical resistance of your heatsink arrangement to be large compared to resistance-to-ground of the primary electrical path. How much larger? What level of current loss can you tolerate? Then you must have numbers for the power dissipation and a target value for the temperature difference you are willing to tolerate. This will set a lower bound on the thermal conductance you need. With these two numbers you then work on the geometry to see if you can achieve your target with real materials.
  7. Jan 30, 2008 #6
    Did you find any such material? I am looking for a similar one...


  8. Jan 30, 2008 #7
    BACT DLC Diamond coating? May be what you're looking for. I think a Boron Nitride powder may have a thermal conductivity a bit too low for what's you're looking for. Check out www.bactusa.com.

    Check out Boron Nitride powders still, though, may be of some use.
  9. Jan 31, 2008 #8


    User Avatar
    Science Advisor
    Gold Member

    Boron Nitride, Sapphire, or Alumina are both useful in some limited applications. The fact is that you need to decide exactly what kind of voltage you're insulating against, and how much heat you need to conduct.

    I have heard that Boron Nitride is difficult to get in commercially pure grades, and is strongest as an electrical insulator in very thin sheets.
    Last edited: Jan 31, 2008
  10. Jan 31, 2008 #9


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Can you use a liquid? I have to believe various oils could be used. There's one used commonly in tranformers - now I forget the name. I'm sure someone can comment.
  11. Jan 31, 2008 #10
    also, select graphite materials can be very thermally conductive, and can just be coated with a thin paralyne coating or something to give electrical insulation. this could give you something a lot lighter weight and with better corrosion resistance than some of the metal alternatives. Of course if money is no object, then yeah diamond is where it is at.
  12. Feb 8, 2008 #11
    AlN seems to be good solution...Did any body work AlN?
    I may get some samples next week. I'll try them and let you people also know...
  13. Nov 24, 2008 #12
    This looks to be a common problem.

    I too am looking for a thermally conductive electrical insulator for a 12v system. Space is a serious issue and one of the components on the PCB (aluminum) is not properly insulated from the board (and the manufacturer has no plans to rectify this any time soon). I'm hoping to use the PCB's housing as its heat sink (the housing is also the part's cathode) while using thin plastic as a thermal break/insulator to protect sensitive areas of the PCB's componentry. At some point on the back of the PCB there has to be a thermal connection between heatsink and PCB, I need to know what I can place there that wont short out the electronics but allow proper heat flow.

    The dimentions are seriously small, I need something of a maximum 1 mm thickness, and the load on the PCB can get up to 45C when driven at high amperage. the surface area is 1 sq.cm

    Any advice/solutions would be greatly appreciated, Thanks
  14. Nov 26, 2008 #13
    Hi Tasty,

    the thickness range of thermal conductive materials is 0.5mm to 13mm,i have delt with this issue before.and the thermal pads are placed between PCB and heatsink which is also available for your question.
  15. Nov 26, 2008 #14
    Thermal conductive materials (of course Electrically insulator) provides an effective thermal interface between heat sink and electronic devices where uneven surface topography,air gaps and rough surface textures are present.
  16. Dec 18, 2008 #15


    User Avatar

  17. Dec 18, 2008 #16
    aluminum nitride its applications have been developed mainly for military aeronautics and transport fields,I think it takes the higher cost to be used in high vacuum applications, am i right?
  18. Dec 18, 2008 #17


    User Avatar

    cost was not an issue for me (this was for basic research, not commercial application) so I admit I didn't pay attention to how much it cost; also I used it because for high vacuum we want minimal outgassing and AlN was sufficient for that criteria along with its electrical and thermal properties. But yes in general components that are high vacuum compatible tend to be quite expensive. (e.g. comparing the cost of a ultrahigh vacuum valve to a low vacuum valve, the cost can be 100 times more)
  19. Jan 18, 2009 #18
    Beryllium oxide meets your requirements. I used it successfully with silicone grease coupling to remove heat from the collectors of rf power transistors in telemetry transmitters (1961 - 1962). Caution: Beryllium oxide powder is poisonous (similar to lead oxides ). Any cutting or grinding of the material must be done with suitable precautions.
  20. Jan 18, 2009 #19
    What is the fabric stuff they use to insulate power transistors these days? Mica was used in the past.

    We used to be very wary of BeO as used in Klystrons. Carcinogenic.
  21. Jan 19, 2009 #20


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Perhaps your thinking of this?
    http://www.bergquistcompany.com/tm_sil_pad_list.cfm [Broken]
    Last edited by a moderator: May 3, 2017
  22. Apr 17, 2009 #21
    mica would do, for thin layers one can use GE varnish.
  23. Mar 18, 2010 #22
  24. Jun 14, 2010 #23
    Mica, Beryllium Oxide (toxic).
  25. Jul 19, 2011 #24
    Hi, I am currently trying to develop a rig on a major budget. I need to give electrical insulation between a thermal heating element and a heat sink as cheaply as possible.

    The problem is this rig is going to be operating at 750C temp in a steam/nitrogen environment for up to 1000 hour periods.

    Any ideas?
  26. Aug 1, 2011 #25
    BeO is nearly abandoned because of toxicity. Al2O3, AlN, (BN) are common. Mica is older.

    Do you have a minimum thickness, like a breakdown voltage or a maximum capacitance? If not, a thin foil is the solution.

    Also: good contact is more important than the material itself. Use special grease, or at least some oil, but no dry contact. Some elastomer films provide the good contact naturally.

    All this is well-known from power electronics engineering. Look in catalogues from Berquist and the like.


    Breakdown voltage or maximum capacitance impose a limit to the heat conductivity, once you've selected the best possible material. If you have to go beyond, a solid won't fit. Blow some gas, or use a heat pipe (with an insulating wall, sure).


    Just because someone wrote "price is not an issue": diamond isn't the ultimate material. If you remove 13C and synthesize diamond from pure 12C, you improve thermal conductivity a lot, even at room temperature.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook