High Dielectric polymer compounds & additives for RF applications 2

[maniachalengineer]

I am trying to formulate silicone compounds that ultimately will have higher dielectric (e) properties for radio and antenna applications. Many formulas and "recipes" exist for liquid (water/salt/sugar) solutions with high dielectric but these are quite problematic for storage, mold, leakage, etc. Thus I'd prefer to have a "solid" high dielectric polymer or silicone based solution. I've been able to add moderate amounts of high dielectric (e= approx 90) powders such as barium titanate to conventional polymers (e=approx 3) resulting in composite materials with medium dielectric constant (e=approx 10). The complex dielectric properties are measured with an Agilent vector network analyzer and dielectric probe.
As one might suspect, the upper and lower limits of this "mixing" approach are defined by the percentage of the high dielectric "guest" material (barium titanate) incorporated into the low dielectric "host" material (ex silicone. In other words the effective dielectric constant of the resultant compound can never go lower than silicone or higher than barium titanate.
I have located a few public research papers online that show mixing low percentages (5%-10%) of various carbon powders and fibers into polymer with very high dielectric constant of the resultant compound. Unfortunately my experiments with commercially available carbon powders and fibers have not produced similar results. I suspect that I'm doing something wrong or using the wrong types of carbon additives. I'm assuming that the polymer "host" is unimportant and simply performs the role of a "carrier" for the dielectric modifying additives.
Has anybody encountered any experimental results or formulations for high dielectric solid/polymer materials? I'm targeting dielectric constants from 20 to 100 and I'd like to tailor the dielectric conductivity, permittivity, and loss to specific values for various radio frequencies...
Any ideas? Suggestions?
Thanks in advance!

 

[Pud]

It's been a long time, but IIRC, around 2.5 is classed as "high" for an unfilled polymer. A cast polystyrene was (used to be?) used for antenna microwave lenses as it had a "high" dielectric!

Does any material have a dielectric in the 20 - 100 range?

Cheers

Harry

ps: I could be writing rubbish of course!

http://www.tynevalleyplastics.co.uk

 

[Demon3]

Some comments about your question. When you add filler to a polymer nearly all properties change linearly with the volume % of filler added. The effect is not linear if you plot versus weight % of filler.

As you add filler, the viscosity inreases with the volume % filler until at around 60 volume % the viscosity gets so high you can't process the polymer anymore.

The density of barium titanate is around 6gcm-3 whereas polymers are around 1.0 gcm-3. This means that adding say 20 weight % filler is only 4 or 5 volume %, i.e. not much and not likely to help the dielectric constant much.

So, my advise is to add 60-85 weight % of barium titanate. Make sure it's surface treated with a good dispersant such as dodecyl trimethoxy silane, or look up what works well to disperse that filler.

The funny results you are seeing with carbon black are probably due to the percolation threshold. I.e. the electrical properties of composites are not linear with the volume % filler. The don't chabnge much as you add filler until the percolation point where there are enough particles for them to join up and make a continuous network that conducts electricity. You can find lots on information on that, probably at Wikipeia as a starting point. You can find excellent information at the Cabot website, they make carbon black. You will see that the type of carbon black has a huge effect on its electrical properties.


Chris DeArmitt

 

[maniachalengineer]

Reply to "Demon3";

Your comments are right on target from what I have learned so far from papers and experiments- the percolation effects are paramount insofar as the dielectric constant of mixtures. Furthermore the various formulas for mixing of "host and guest" compounds are highly non-linear and do not encompass or predict the effects of conductive compounds like carbon.

Unfortunately I've also found that various carbon types are largely unpredictable in this application when trying to use the commonly available properties- in fact the conductivity of carbon powders and fibers is frequently listed with a "disclaimer" about using the conductivity for any type of compounding since it is typically measured after being compressed into a block...

I appreciate your feedback, I will try to post any noteworthy progress notesd as I move forward on this task.

 

[Demon3]

Hello,

Glad to hear that I contributed something that makes some sense. I had some issues recently and downloaded the techical brochures from Cabot which plot how the percolation threshold moves as you change the type of carbon black. Ability to form structure, surface area and degree of surface oxidation all play a role. Also, I would recommend calling them. Their experts really know their stuff. Ask to speak to Dan Callahan (plastics) or John Foster (coatings). Because the conductivity depends on the degree of dispersion, the mixing has a tremendous influence. Getting that good and consistent is half the battle.


Chris DeArmitt

 

[maniachalengineer]

Reply to "Demon3" and general update on carbon based mixtures, excellent source for carbon properties, etc.

Thanks for the lead pointing me to Cabot, I am checking it out. I have also really benefitted from looking at the Asbury carbon website (see weblink below) for excellent technical info on carbon and it's myriad properties. Seems like carbon is the most versatile "simple" structure that exists and it certainly gets complicated as it takes on various properties during treatment- ranging from pencil lead to diamond to superconductors, to ??? Here is a good, fairly complete and detailed reading source for carbon information:

http://www.asbury.com/Table/Literature/Technical-Info/

Russ

 

[unclesyd]

On things that I found out when using carbon as filler for some of our products.
Though we didn't use a silicone carrier we did use a very viscous carrier in our process. We made two materials, one was just for coloring and the other was to make a polymer electrically conductive. One method I used to help with and control the distribution was reverse the normal mixing sequence. That is add all the powder to the liquid pool instead of incremental powder additions to a liquid. We used Bradbury mixers as well as rod and ball mills depending on the dispersion.

There are some interesting papers around on preparing and adding pigments to paint bases.

 

[unclesyd]

While looking up another problem I woke up some neurons. Have you looked at Boron Nitride powder as a filler. Just before I left work I was working with BN on an entirely different polymer problem. Incorporated in a polymer it has some very unique properties.
Though not conductive it is known as the White Graphite.

http://bnpowder.com/

 

[maniachalengineer]

Regarding "Unclesyd" comments and about mixing in general;

We have observed that mixing and effective dispersion are very important- the measurements are strongly "polarized" and affected by any non-homogenous (non-isotropic) distribution of the carbon powders and fibers in the silicone. We also have some difficulty with "wetting" the mixes- especially since the silicone "base" is initially somewhat high viscosity (approx 5000-10000 Centipoise)and the carbon powder has EXTREMELY high surface area (some activated carbons have 250 sq meters of surface area per gram)- this makes for a non-mixable "paste" in short order. A good overall description of the mixture is "brownie mix" from kitchen experience- lots of fine powder into a thick viscous base- it takes a while to wet the powder and start truly mixing as a "semi-liquid" suspension.

We have employed "thinners and diluents" in the mixing process as well as avoiding the high surface area carbon powders, these help considerably. We also have tried high energy ultrasonic mixing (sonicators)and mechanical kitchen style mixers (in addition to manual stick mixing) but several aspects of the mixed compound are still problematic...

We also use a vacuum degassing step to remove air and other vapors from the mixture followed by curing in a low temp oven for several hours- but it is apparent that some settling and "stratification" is still taking place, possibly other unwanted (non isotropic) arrangements in the cured matrix. Does anybody know of a truly "random mixing" process or machine for something like this? How can you mix these thick mixes randomly during curing without introducing more air or other problems?

Who knows? Maybe we'll also make better brownies eventually!

Thanks in advance for your thoughts and suggestions!

 

[unclesyd]

Additional things you could look at.

Making a master batch of carbon then adding more silicone.

Doing the reverse mixing at master batch levels of carbon.

Using an small amount of additive like BN or other dielectric material with the carbon prior to incorporating it into the silicone.

Don't as me why but heat the carbon.

Saturate the carbon prior to mixing. If my recall is working I used benzoyl alcohol, as it was compatible, with my process.

 

[Demon3]

Just a note on boron nitride. It's available in two forms. In fact it's directly analagous to carbon. There is a graphite (sheet) form that is called white graphite as it's soft and lubricious.

Then there is a diamond-like form that is extremely hard.

Check Wikipedia for more info.

Make sure you don't get the cubic, diamond-like type by mistake and ruin your equipment due to the extreme wear it will cause.

You can get the hexagonal, soft form from Momentive I think and Saint Gobain for sure.


Chris DeArmitt

 

[Compositepro]

I most applications where a high dielectric constant is desired it is also desired to have a high dielectric strength. Conductive fillers increase apparent dielectric constant of materials because of the way it it defined and measured. The effect of the filler is the same as moving your measurement plates together but but using the larger gap to calculate your number. It is generally not a useful effect.

In any case, three roll mills are the best way to get very high dispersion of powders into viscous materials. On the other hand carbon black can be too highly dispersed as well as poorly dispersed (as the Cabot literature will explain).

 

[unclesyd]

The link that I posted is to ZYP Coatings. The majority of their materials are of the hexagonal persuasion.

 

[maniachalengineer]

All;

Thanks a bunch for the continued suggestions and ideas!

The Boron Nitride ("white graphite") is a "dead end" for this application because it cannot raise the dielectric by either being conductive or by inherently having a high dielectric itself...

I'll look into 3 roll mills- I'm not familiar with these. I also don't have all the Cabot literature yet- but according to "Compositepro" and "Demon3" I definitely need it! I will renew my efforts on this goal.

It seems that the effects of dispersed carbon in a plastic (or presumably silicone) carrier are highly unpredictable so far- I've tried for weeks to duplicate results in published papers that used percentages (by weight) less than 10% to achieve remarkably high dielectric constant (DC)- without raising my measured dielectric appreciably. I have needed to add 50% (or more) to get any appreciable increase in DC.

"Compositepro" also made a good comment about the "effective dielectric" by adding conductive materials into the compound- it does raise the DC while jeopardizing or lowering the dielectric strength- but my application can utilize this quite nicely. So just about anything that raises DC in this "solid compound" is good for me!

Thanks again everybody!

Russ

 

[unclesyd]

Have you looked at shear mixers made by people like Ross?

This type mixer is used to make some pretty heavy mixes like inks and tints. Ross had a broad section of mixers along with some good testing programs.

http://www.highshearmixers.com/

Find out who makes the black silicone rubber for sealing glass cook tops and go from there.

 

[unclesyd]

Here is a good article on mixing that I learned the hard way many years ago.

Checkout: "The New Rules of Mixing"

http://www.highshearmixers.com/articles.html

I'm trying to interpret my notes from a long time ago when we first started working mixing carbon and polymers.
I don't have copy of my final reports and can't understand why my notes that were so clear and informative are now offering a challenge to even read.

 

[Demon3]

There are other fillers apart from carbon black that have electrical conductivity and are far easier to disperse than carbon black. Indium tin oxide is one.


Chris DeArmitt

 

[unclesyd]

Have you looked at the various zinc ferrite's as I recall some have fairly high dielectric values. They might make good diluent for high value materials.
If I'm looking at my notes correctly we had no trouble incorporating them with several different polymers.

Also you might want to contact Lubrizol about your mixing and dispersion problems.

http://www.lubrizol.com/products/

 

[unclesyd]

After you posted about modifying the dielectric materials I recalled that I had a little information about this Nano Nickel material. We were looking at this material and similar materials to make a cloth that would act as EMF shielding.

If this type material is of any use I also information on silver coated particles. We used a similar material for anti-microbial properties.

http://conductivecomposites.com/index.html