litz skin effect and topological insulators

[Jamie M]

Hi,

I was reading about topological insulators which can super-conduct
electrons on their edges, I was thinking about this in relation to
the high frequency skin effect. At a high enough frequency, the
skin depth should be 1 atom deep on the conductor, and then the electron
may have to travel on the "edge" only of the conductor, so
there could be zero losses from resistance and supercondivity.

The losses would be 100% from emission from the vibrating AC
electrons. At an infinite frequency the vibration amplitude of
the electrons would approach zero, and radiation losses would
approach zero. Does this make sense that losses from skin effect
could start to decrease or am I thinking about it completely backwards?!
Would it be possible to approach a high enough
frequency where the radiation losses could be recovered?

cheers,
Jaime

 

[Jamie M]

Pretty skinny premise, man.

They have carbon films one atom thick that stop smaller atom gasses
like Helium. A whole new technology will come out of this.

Hi,

If you use two conductors separated by a thin insulator layer
(nanometers) and run high frequency AC 180 degrees out of phase in
them, will the radiated losses cancel out? If so then at high enough
frequency maybe that could be a superconductor.

cheers,
Jamie

 

[RobertMacy]

Hi,

I was reading about topological insulators which can super-conduct
electrons on their edges, I was thinking about this in relation to
the high frequency skin effect. At a high enough frequency, the
skin depth should be 1 atom deep on the conductor, and then the electron
may have to travel on the "edge" only of the conductor, so
there could be zero losses from resistance and supercondivity.

The losses would be 100% from emission from the vibrating AC
electrons. At an infinite frequency the vibration amplitude of
the electrons would approach zero, and radiation losses would
approach zero. Does this make sense that losses from skin effect
could start to decrease or am I thinking about it completely backwards?!
Would it be possible to approach a high enough
frequency where the radiation losses could be recovered?

cheers,
Jaime

No. The skin effect you describe requires the current down inside to force
the current to the outside and the current down inside will eat power.

To test your idea, use free simulation tool, femm 4.2, and try it out,
you'll better understand where skin effect comes from. And learn a lot
about magnetics.

 

[[email protected]]

Not 'pluralize'! Possessive form! Doh!

The possessive form of boss is boss'. When the last letter is an 's'
or a 'z', whether it's plural or not, the possessive form has a
trailing apostrophe and no additional 's'.

 

[Jamie M]

No. The skin effect you describe requires the current down inside to
force the current to the outside and the current down inside will eat
power.

To test your idea, use free simulation tool, femm 4.2, and try it out,
you'll better understand where skin effect comes from. And learn a lot
about magnetics.

Hi,

Thanks I checked wiki: http://en.wikipedia.org/wiki/Skin_effect

The current down inside that forces the current to the outside, is the
AC induced eddy currents, and these eddy currents also move towards the
skin at higher frequency, so at some high enough frequency even the
eddy currents approach the outer layer of atoms on the conductor. Not
sure what would happen in this case, but probably the AC eddy currents
would create a plasma discharge or something off the surface?!

cheers,
Jamie

 

[RobertMacy]

Hi,

Thanks I checked wiki: http://en.wikipedia.org/wiki/Skin_effect

The current down inside that forces the current to the outside, is the
AC induced eddy currents, and these eddy currents also move towards the
skin at higher frequency, so at some high enough frequency even the
eddy currents approach the outer layer of atoms on the conductor. Not
sure what would happen in this case, but probably the AC eddy currents
would create a plasma discharge or something off the surface?!

cheers,
Jamie

I thought about that just after posting. Should check using FEA and a
model with a VERY thin surface layer of almost infinite conductivity and
see what happens. Something must not work here, else a single layer of
superconductivity would ALWAYS steal all the carriers so is not an easy
thing to even understand.

 

[[email protected]]

I thought about that just after posting. Should check using FEA and a
model with a VERY thin surface layer of almost infinite conductivity and
see what happens. Something must not work here, else a single layer of
superconductivity would ALWAYS steal all the carriers so is not an easy
thing to even understand.

Current does flow on the surfaces in superconductors, even at DC.
The current layer has finite thickness, however, it's called the
London penetration depth.

The current tries to distribute itself across the cross-section
of the wire such that the stored energy is minimized. Each current
filament can lower its energy by moving away from other filaments.
One would then expect that all the filaments would move as far away
from each other as they can, i.e. to the surface of the conductor.

Anyway, there is only a finite density of superconducting charge
carriers available in a material. The smaller fraction of the
cross-sectional area the current filaments occupy, the smaller
number of carriers must carry the current, and the faster each
carrier has to move. Because the carriers have mass, there is
kinetic energy stored in their motion. This kinetic energy storage
looks to the driving electric circuit exactly like inductance,
hence it's called "kinetic inductance". However, the energy is stored
as kinetic energy, not in the magnetic field. Therefore the kinetic
inductance does not couple magnetically to other nearby inductors,
either.

So, there is a tendency of the current filaments to move farther
from each other, and thus reduce the stored magnetic energy. But
by doing so they induce a larger and larger kinetic energy storage
when the current gets packed to the surfaces. The two effects
balance when the current distribution reaches (in the case of an
infinite slab geometry IIRC) and exponentially decaying shape,
with decay length equaling the London penetration depth. For our
Nb thin films at LHe temperature this is roughly 90 nm.

Regards,
Mikko

 

[[email protected]]

Depends on whom you ask. Strunk & White says always to add "'s"
regardless. Chicago Manual of Style says to add "'s" if the noun is
singular, and just "'" if it's plural. The one constant in all this
confusion is RWWATP.(*)

Cheers

Phil Hobbs

(*) Real Writers Write Around The Problem.

If there are standards on every side of the argument, it sounds like a
"real writer" can do it any way sheit pleases. "That's the nice thing
about standards..."

 

[Jamie M]

Current does flow on the surfaces in superconductors, even at DC.
The current layer has finite thickness, however, it's called the
London penetration depth.

The current tries to distribute itself across the cross-section
of the wire such that the stored energy is minimized. Each current
filament can lower its energy by moving away from other filaments.
One would then expect that all the filaments would move as far away
from each other as they can, i.e. to the surface of the conductor.

Anyway, there is only a finite density of superconducting charge
carriers available in a material. The smaller fraction of the
cross-sectional area the current filaments occupy, the smaller
number of carriers must carry the current, and the faster each
carrier has to move. Because the carriers have mass, there is
kinetic energy stored in their motion. This kinetic energy storage
looks to the driving electric circuit exactly like inductance,
hence it's called "kinetic inductance". However, the energy is stored
as kinetic energy, not in the magnetic field. Therefore the kinetic
inductance does not couple magnetically to other nearby inductors,
either.

So, there is a tendency of the current filaments to move farther
from each other, and thus reduce the stored magnetic energy. But
by doing so they induce a larger and larger kinetic energy storage
when the current gets packed to the surfaces. The two effects
balance when the current distribution reaches (in the case of an
infinite slab geometry IIRC) and exponentially decaying shape,
with decay length equaling the London penetration depth. For our
Nb thin films at LHe temperature this is roughly 90 nm.

Regards,
Mikko

Hi,

I think a microwave waveguide is acting kind of like a topological
insulator too, the microwaves reflect off the surface and don't
penetrate, but still there is some interaction with the surface
atoms that is near superconducting type.

cheers,
Jamie