# Looking for a special dielectric

Discussion in 'General Electronics' started by Lord Snooty, Feb 1, 2004.

1. ### Lord SnootyGuest

I'm looking for a dielectric material that can be used in a capacitor subject
to AC, which should have the following characteristics:
1. Should be able to be used in high power and high frequency situations
2. Should have as high a permittivity as possible
3. Exhibits the highest possible polarising velocity of its ions when subject
to AC electric fields.

#3 is the unusual one. It refers to the small displacements that occur in a
dielectric when polarised. I want the charges on the polarised molecules to
move as fast as possible for a given applied voltage. Maximum velocity will
correspond, for sinusoidal AC, to the zero-volts crossing point, since maximum
acceleration occurs at peak volts - and velocity and acceleration are phased
90 degrees apart for a sinewave.

I am wondering whether a liquid dielectric would be the best bet in respect of
#3. How does one calculate the value of the polarising velocity in any case,
for the kind of dielectrics commonly used in HV capacitors?

2. ### Don LancasterGuest

Distilled water?

--
Many thanks,

Don Lancaster
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
voice: (928)428-4073 email: fax 847-574-1462

Please visit my GURU's LAIR web site at http://www.tinaja.com

3. ### Paul Hovnanian P.E.Guest

Just some thoughts: Maximizing permittivity and charge velocity might be
working at odds when selecting an optimal material. For polar materials
(where there is an inherent charge distribution across each molecule),
the relative permittivity is proportional to the dipole moment of charge
in each molecule. This is proportional to the charges and their physical
separation in each molecule. The higher the permittivity, the more
charge (and mass) at greater separation, which also results in greater
physical moments of inertia and slower response to align with an
electric field. Nonpolar materials polarize only due to charge
displacement within a molecule rather than physical realignment of the
molecule. These would be faster, but might not have as high a charge
separation for a given field.
Possibly. At least molecules in liquids aren't locked into a crystal
structure and are free to
align with a field.

You might want to look at curves of permittivity vs frequency. The less
the permittivity drops off at higher frequencies, the faster the charge
dipoles can move.

4. ### Roy McCammonGuest

I'm not sure "polarising velocity" is a well defined concept.
Did you mean maybe smallest time constant (settles to equilibrium
position in the least amount of time)?

5. ### Lord SnootyGuest

Nope, I actually meant that, when subjected to an AC electric field within
which the ionic dipoles will oscillate, the maximum velocity during one cycle
of this oscillation will be at the "equilibrium (non-polarised) point. Same
with any harmonic motion of course. I am therefore talking about the actually
velocity of the charges in the dielectric.

-Andrew

6. ### Lord SnootyGuest

Good points. If the ionic dipole moment were solely due to the charge
separation (i.e. all materials had equal charges/masses), then one could make
the simple connection
big separation->big polarisation->big permittivity. One can also say that big
separation->big velocity, since, for two materials identical except for their
charge separation distance, in identical electric fields, it's the case that
(because local charge screening effects are equal) the maximum charge velocity
in a cycle will be highest for the longest path, because they have a larger
distance over which to accelerate, and hence for the largest separation.
However, variable charge is, as you rightly point out, the fly in the
ointment.
Thanks - I think intuitively that this is probably a very useful criterion. I
sure am not going to get such detailed data from the capacitor manufacturers.

-Andrew

7. ### Joseph.D.WarnerGuest

Look at various perovskites. A common one is Ba(x)Sr(1-x)TiO3. By
100 K to above room temperature. It has been used in microwaves to at
least 30 GHz and probably can be higher. Dielectric constants can vary
from 500 to over 5000 depending on composition and growth methods.
Breakdown fields are typically >5 E+5 V/cm. This varies too according to
processing.
Have no idea for this.
If you're using liquid you wil still need to make it very thin or you're
time constant will be determined by the reactance of the circuit and
not by the material.

8. ### Uncle AlGuest

stupid should be silent - or declared open season. Think of it as
evolution in action.

9. ### Lord SnootyGuest

Why don't you go take a flying hoorah at a rolling doughnut, you grumpy old
fart?
You contribute nothing except bile. Drink some. After all, it's yours.

I suspect you're not getting enough.

-Andrew

10. ### tweakGuest

And your ignorant? We all lack knowledge in some area crank.
That's the whole point of this NG.
If your going to flame at least make a point. Berating for the sake of
it is the sign of a weak mind.
And if you want natural selection dialed into the equation I doubt
that many of you brainiac types would last long either.

11. ### Michael VarneyGuest

That is "you're", you ignorant dolt.

Hey stupid, noting that we humans lack sharp teeth, claws, sheer strength,
and speed, how is it do you think we survived to become the dominant species
on the planet?

12. ### Robert C MonsenGuest

Must have been our winning personalities!