In hydraulics, I think perfect sealing is achievable - it is not science fiction to consider a material whose pores are smaller than a water molecule (or is it?), which would yield zero flow rate.

However I can't apprehend a material which would reflect electrons - given that the number of electrons in an atom if fixed for it to be stable, and otherwise it can re-emit then in any direction, including through the material.

So it seems it's impossible to get perfect insulation. To prove it's right or wrong, a recap of how insulation works would be nice.

I hope this is the right stackexchange.

  • 1
    \$\begingroup\$ Probably a better question for physics really... \$\endgroup\$
    – Matt Young
    Jan 25, 2015 at 15:03
  • \$\begingroup\$ No, there's no "perfect" insulator, and there's no "perfect" seal, either. To understand what "perfect" means in this context, you need to understand something about quantum mechanics, and this question would be better asked on physics.se \$\endgroup\$
    – Dave Tweed
    Jan 25, 2015 at 15:25
  • \$\begingroup\$ If you think of voltage as the equivalent of pressure, then you'll understand that there's no perfect seal because given a high enough pressure, any seal will break. The same thing goes for voltage. \$\endgroup\$
    – horta
    Jan 25, 2015 at 16:07

3 Answers 3


Insulation \$\rho\$, breakdown voltage \$U_{br}\$ and dielectric permittivity \$\varepsilon\$

Insulation is the ability of a material to block the flow of current through it, so that means that this material has a high resistivity. It is measured in \$\Omega m\$(Ohm-meter).

Breakdown on the other hand is a process in insulators, it can be quantified in terms of a breakdown voltage (for specific frequency, temperature etc.). That means that if you pass a specific voltage the material loses its insulating ability and fails catastrophically(melting, arcing, etc.). You can also measure it as \$E_{br}\$, or breakdown field it is measured in \$\frac{kV}{mm}\$(kilovolt per millimeter), air has \$3\frac{kV}{mm}\$ for example.

The dielectric permittivity is yet another property of insulating materials. The relative permittivity \$\varepsilon_r\$ tells you how many times the electric field will be lower in the material in comparison with vacuum. Say if you had a capacitor with vacuum as dielectric, and another one with some material with a specific \$\varepsilon_r\$, the electric field in the (with everything else the same) in the second capacitor will be \$\varepsilon_r\$ times smaller than in the first one.


Similar to superconductors, there are superinsulators. Superconductors will be superconducting only for some specific circumstances. That means specific temperature, magnetic field and current density. You can't put infinite current through it, as at some point it will lose its superconducting ability.

For superinsulators you have a similar story, they won't let any current pass until you keep the parameters (voltage, temperature) within specific boundaries. They will therefore lose their superinsulating property at one point.


Because of the discussion on Majenkos answer I'll have to talk about this, too.

How can vacuum not be a perfect insulator? It has no current carrying ability as it has no particles that could conduct, so there can be no current flow. Therefore the conductivity should be zero, isn't it? Well there are always very small amounts of particles that will create some conductivity, and there can be breakdown processes in the vacuum.

Look at high voltage switchgear. There are vacuum switches which go (as far as I know) up to 160kV. That means that they can handle the separation of two contacts that have this kind of voltage between them. During the switching operation an electric arc is created. This arc is mainly fed by the let's say evaporated contact materials. So the vacuum will conduct only because of foreign particles.

There always have to be two electrodes at a voltage difference for conduction, if vacuum is in between them we just need a big enough voltage difference so that particles start to move from one electrode to the other creating a conducting path. So eventually everything will conduct because of the electrode material.


Image of a high voltage switch, source for image: Mitsubishi Electric


No, there is no such thing as perfect insulation. All insulation is merely a very very high resistance with a high breakdown voltage.

Give it enough voltage and the insulation will break down and start to conduct. Below that voltage it can still conduct, but only very very very tiny amounts. Insulation is typically measured in GΩ or even TΩ for the good stuff.

  • 2
    \$\begingroup\$ Remember vacuum tubes? Just heat the cathode metal a bit, and the electrons will be free to move. \$\endgroup\$
    – Turbo J
    Jan 25, 2015 at 15:25
  • 2
    \$\begingroup\$ You don't think an electric field can move electrons across a vacuum? \$\endgroup\$
    – Dave Tweed
    Jan 25, 2015 at 15:29
  • \$\begingroup\$ I'm thinking a vacuum many millions of lightyears across. Generating the energy to make the electron move across that vacuum would be impractical at best. \$\endgroup\$
    – Majenko
    Jan 25, 2015 at 15:35
  • 1
    \$\begingroup\$ @Majenko: All that's necessary is to get the electron started and then it'll go on its merry way forever until something interferes with it. Or, at least, that's what Newton's First Law of Motion points out. \$\endgroup\$
    – EM Fields
    Jan 25, 2015 at 21:53
  • 2
    \$\begingroup\$ I don't know how observant you all aren't but 6 hours or so ago I completely deleted that erroneous statement. And yet you're all still going on about it. Look I admit it - I was wrong (something you'll never be able to force out of Olin), now just drop it will you? \$\endgroup\$
    – Majenko
    Jan 25, 2015 at 21:54

If there is no perfect insulator, theres no perfect capacitor. All capacitors will throw away so much energy over time, and there's seemingly nothing you can do about it.

  • 1
    \$\begingroup\$ Welcome to EE.SE but answers are supposed to answer the question at the top of the page. This one seems to have gone off on a tangent and does not address the question. You can always edit to improve your answer before (or after) it gets downvotes. \$\endgroup\$
    – Transistor
    Feb 17, 2021 at 10:39
  • \$\begingroup\$ There is an enormous difference between "perfect" and "good enough", and good engineers focus on "good enough". Your second sentence really doesn't make any sense. \$\endgroup\$ Feb 17, 2021 at 12:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.