From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Physics (Rated B-class, High-importance)
WikiProject icon This article is within the scope of WikiProject Physics, a collaborative effort to improve the coverage of Physics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.

Complex Permittivity, Gauss Law and Ampere's Law[edit]

Something is not clear about the definition of the complex permittivity in harmonic analysis: it seems to be related to the conductivity in Ampere's Law, so to include ohmic conduction. What is not clear is whether the complex conductivity here discussed would also appear in Gauss Law or not: could you deepen this topic?

Vacuum polarisation[edit]

Is the intrinsic reason for the non zero value of epsilon0 that the vacuum can in fact be polarised by an alternating electric field? Or not? Any thoughts?--Light current 04:40, 13 December 2005 (UTC)

Nah - the fact that certain books still term the constant "Permittivity of free space" is anachronistic really. Epsilon nought is really just an expression of the strength of interactions in freespace like the gravitational constant, and certain sources now term it the electric constant, for instance the NIST reference website to avoid such confusion. --Neo 11:58, 13 December 2005 (UTC)

'Light Current''s question is mine too. I am totally satisfied by NeilTarrant's answer. But how was/is the "permittivity of free space" determined? Another question is whether the constant indicate the ratio of electric capacitance per unit length of free space as it's unit may impart? Thanks

'and thereby reduce the field inside the material.' should be 'and thereby reduce the total electric field inside the material.' See page 155 Jackson Electrodynamics TheDeuce1123

Old talk (section title added)[edit]

I re-did the last 2 changes because someone deleted the entire permettivity page.


Might consider splitting up the page into complex perm vs. the basic definition of perm.

I don't think multiple pages are needed (the current length is not too great), but subsections would be a good idea. —Steven G. Johnson 03:05, Dec 10, 2004 (UTC)

I liked the page how it was a few minutes ago. I liked being able to find the definition of permittivity right away.


Quantum Mechanical section is in need of clarity (and I think correction - but I'm no expert). Nowhere is there any mention that this can be measured (not modeled), how it can be measured, and the historical development of the concept.

Yes, alot of things are missing here. I am organizing the article for start... any comments? Karol 17:21, May 24, 2005 (UTC)


OK, I generally cleaned up, reorganized and added things. I removed the attention tag, but if someone thinks it's still needed then please put it up again. IMHO the quantum-mechanical and measurement sections need the most attention of improvement currently. Cheers! Karol 11:58, May 25, 2005 (UTC)

Might want to add this?[edit]


Also known as the vacuum permittivity, this is the amount that a vacuum allows electric current to flow through it.

ε0 = 8.8542 * 10 -12 C/(N*m2) = 1/(1.1294*1011*V*m)

If you have two charged plates in a vacuum 1 mm apart with 112940 volts or more between them, an electric arc will jump between the plates.

I don't want to step on any toes, but as a beginner no statement on the current page has a comparable simplicity, so I found it very helpful. Buhsra: no dont merge it because it ie easy to understand now.

Thanks for trying to help, but you mustn't believe everything you read on the Web. The statement you quote is either a hoax or is written by someone with no knowledge of physics. Permittivity is not "the amount that [something] allows electric current to flow through it". The units of ε are F·m-1 or C·V-1·m-1, which are not equivalent to those above. The "112940 volts" thing is just pure nonsense. --Heron 11:47, 23 August 2005 (UTC)
Is this editor maybe confusing field emission that depends on work function of the matl?--Light current 12:54, 3 August 2006 (UTC)

Correction required for consistency?[edit]

Under the discussion of complex permitivity there appear to be two conflicting statements...

"In the equation above, is the imaginary part of the permittivity. The real part of the permittivity, , is related to the fraction of the energy absorbed by the medium."

"At a given frequency, the imaginary part of leads to absorption loss if it is positive (in the above sign convention) and gain if it is negative. More generally, the imaginary parts of the eigenvalues of the anisotropic dielectric tensor should be considered."

I believe that the second statement is correct and the imaginary part is what leads to absorption or gain. That is definitely true of the complex portion of refractive index (a related quantity). Perhaps the first statement needs to be corrected in order to jibe with this and ensure consistency? -- Rob, Sept 2 2005

You are right, the first statement is false, and it might have been me who wrote both statements, the first with a mistake, although I don't remember if it was me in fact (I rewrote most of the article in May this year). Should be dispersed instead of absorbed. Karol 17:40, September 2, 2005 (UTC)

Unit for permitivity of free space.[edit]

Shouldn't the unit for be ?PoorLeno 19:53, September 4, 2005 (UTC)

That is identical to farads/metre, as given in the article, since farads = coulombs^2 newton^-1 metre^-1. -- DrBob 23:10, 4 September 2005 (UTC)
Yeah, I got it this morning. Was a bit baffled by the is all. PoorLeno 16:21, September 5, 2005 (UTC)

Even though these units are the same, the former should be used for the sake of simplicity. -- (talk) 15:16, 28 July 2010 (UTC)

Table with DC values[edit]

I moved the table of dielectric constant for different materials to the more relevant page. Karol 09:25, 29 November 2005 (UTC)

Permittivity of free space[edit]

"permittivity of free space" is a commonly used term that redirects here (as it should), so I'm putting that term in bold where it is mentioned (in the vacuum permittivity section). Xezlec 22:03, 31 December 2005 (UTC)

The text of the separate article "E0" can be merged with this section. This would reduce confusion between the use of E0 to denote the ground state energy eigenvalue of a potential (among other uses) and the use of in electromagnetism.

This bit- "The permittivity of a linear isotropic homogeneous(LIH) material is usually given relative to that of vacuum, as a relative permittivity \varepsilon_{r} (also called dielectric constant). The actual permittivity is then calculated by multiplying the relative...susceptibility of the material." Should be moved to the section below. It doesn't belong in the vacuum permittivity part.Nathan

Reduces electric field[edit]

How can increased permittivity reduce an externally applied electric field? Yet this is what the article says! It can reduce D, but D is not applied is it? It is E that is applied--Light current 22:56, 2 January 2006 (UTC)

Well, reasonable question. Let me know if this helps. Increased permittivity increases D if we hold E constant, or reduces E if we hold D constant. You can "apply" whatever you want, if by "apply" you mean hold constant, which I think you do.
For instance, if the material in question is a sheet between two oppositely charged plates (with no charge ever added or removed from the plates), then D is held constant (equal to the plate charge divided by the area of the sheet, no matter what), and E depends on the permittivity of the material. But, if instead of holding the plate charge constant, we hold constant the voltage between the plates (perhaps with a 1.5-volt battery), then E is held constant (equal to the voltage divided by the distance between the plates), and D depends on the permittivity of the material.
If this is a point of confusion maybe someone should attempt to clarify this on the page! Anyone? Xezlec 23:35, 2 January 2006 (UTC)
Whoops, gotta add one more comment. In the real world, of course, we don't always hold either E or D constant in every situation. Rather, they both are usually somewhat free to change. In this case we might expect both D to increase a little and E to decrease a little when permittivity is increased. Thus it can be reasonably stated that increased permittivity has a general tendency to increase D and decrease E. Xezlec 00:01, 3 January 2006 (UTC)
Your explanation is crystal clear. And it helps show why I find "applied field" so unclear. I really wish it would be replaced in all these articles by symbols (D or E) and operational descriptions such as you've given. (talk) 22:43, 6 April 2014 (UTC)

Yes I agree with your analysis, so it should be made clear on the page to avoid confusing the unwary! I think E should be the independent variable and D the dependent variable in this description. (as in fact the equation shows).--Light current 00:35, 3 January 2006 (UTC)

Ok, I changed the offending sentence as per your suggestion. I also think the whole idea that permittivity relates to a material's ability to "store charge" is a little iffy. A dielectric in a capacitor doesn't really store the charge; the conductor does. Permittivity relates to a material's ability to transmit (or "permit") electric field. I'll worry about that later. Xezlec 00:50, 3 January 2006 (UTC)

Well I think increased permittivity permits more electric flux for a given applied filed, does it not? --Light current 00:57, 3 January 2006 (UTC)

If you put a dielectric object (with ε > 1) into an externally applied electric field, the electric field within the object is reduced. If you put a charged particle in a dielectric medium, its electric field is reduced by a factor of the dielectric constant. This is because the bound charges in the dielectric orient so as to partially cancel or "screen" the field. And few people use the term "electric flux density" for the D field, as far as I can tell.
In any case, this article is somewhat confused; when you introduce a dielectric material into a system, the electric field doesn't simply stay the same while D gets multiplied by ε. In a homogeneous dielectric medium, for electrostatics, it is the converse: D remains fixed while E gets reduced by a factor of ε for a given charge distribution. I rewrote the offending paragraph, but it's clear that the article as a whole is a mishmash that needs a lot of work.
—Steven G. Johnson 02:11, 3 January 2006 (UTC)

THe E field within the object is reduced , yes. But the externally applied E field is NOT reduced. I think electric flux density is an easier term to understand, which is why i changed it.--Light current 02:21, 3 January 2006 (UTC)

It doesn't matter what term is easier to understand. It matters what term people use. Pfalstad 06:05, 4 January 2006 (UTC)

This is a chicken and egg situation and I feel you are splitting hairs here.--Light current 02:21, 3 January 2006 (UTC)

The total field is reduced compared to the applied field; I think you're the one who is splitting hairs in order to defend a misleading formulation. This is the fundamental thing that one has to understand about dielectric media: a larger dielectric constant reflects the tendency of charges in the material to move so as to oppose applied fields. And using the phrase "electric flux density", which hardly anyone even uses (try googling for "electric flux density" vs. "electric field"), doesn't clarify matters. —Steven G. Johnson 02:26, 3 January 2006 (UTC)
I could argue that the charges move so as to oppose field within the material and strengthen field outside it (what you are perhaps calling "applied field", a confusing term with which I take strong exception), in fact, hmmm, I think I will argue that. ;) Xezlec 21:37, 3 January 2006 (UTC)

When you say total field, do you mean the algebraic sum of the applied field and the field in the dielectric? If so I might agree with you, but I thought E was the external applied filed.-- is it not?--Light current 03:32, 3 January 2006 (UTC)

What is an "externally applied field"? I don't think the article was at any point trying to imply anything about whatever was "outside" the material in question. I think we are assuming that the entire universe as far as we care is filled with the material whose permittivity is under discussion (maybe this should be mentioned). Boundaries between materials are far, far, far more complex and merit discussion in an entirely different article. What I think we are taking about is a material with some properties, and some kind of conditions imposed within it artificially, and some other conditions result from this situation. Or at least that's what we should be talking about since it's the simplest case. Xezlec 21:29, 3 January 2006 (UTC)
The concept of an external applied field is very standard in electrostatics, especially when considering the effect of a field on a finite object, in which case it can be taken as the boundary condition that E tends to a constant at infinite distance. Or, in this context, you could take it as the field due to an arrangement of free charges, not including the bound charge. In circuits, one has the concept of the "driving voltage", which is not the same as the total voltage due to internal resistance, and so on. In any case, for a given static arrangement of free charges, if you multiply the permittivity everywhere by some constant, the electric fields are decreased by that constant.
While I do find that a perfectly valid definition of the term, I don't see that definition anywhere in the article, or anywhere on Wikipedia for that matter, and if I were new to the subject, that definition surely would not have spontaneously occurred to me. Furthermore, I don't consider it a simple thing. I think what you are describing is a pretty complex idea, not at all intuitive to anyone who does not have a very strong math/physics background already (i.e. someone who certainly already knows what permittivity is). You are talking about defining permittivity by talking about a concept which is the limit of a quantity as distance approaches infinity! Why not define it in a literally correct (though perhaps un-elegant) way that everyone can understand, and then explain it the more "fundamental" way later, after the reader knows what you're talking about? My brain hurts already, and I'm supposedly pretty good with this stuff, by engineering standards. Xezlec 23:32, 7 January 2006 (UTC)
Please note that I just made some changes to the part about the relationship between E and D in the explanation section. This is the very subject I've made it my goal to clarify on Wikipedia. The "flux of a charge" is meaningless as far as I know, and since E and D have very specific, well-understood meanings, I don't think this needs to be ambiguous. Xezlec 22:00, 3 January 2006 (UTC)
Unfortunately, your revised discussion is still unclear. The relationship D=εE is between the total displacement and electric fields, and does not depend only on the "applied" field. In most contexts (a capacitor attached to a fixed-voltage battery is an exception), the field changes in response to the polarization of the material. —Steven G. Johnson 22:13, 3 January 2006 (UTC)
You're talking to the wrong guy. I'm not sure what edit you're referring to, but the part I changed didn't say anything about that. I just changed a few words along the lines of "D is the amount of flux of the charge" to "D relates to the charge densities involved" and stuff like that. I didn't add (or remove) anything about "applied" versus "total" fields, I don't think. If I did, it was probably by accident. I always think in "total" fields, myself. They feel like they have a more solid and concrete meaning to me, a fact I should probably quit harping on about here! Xezlec 23:32, 7 January 2006 (UTC)

D is the independent variable because both the free charges and the polarised charges have to be considered in any region of space (D= epsilonE+P). So if D and P are constant, (which is usually the case) increases in epsilon must result in reduction of E. However I dont see how P can remain constant when E changes (unless its cancelled by the change in epsilon). It would appear the (asst)professor is correct!--Light current 03:30, 4 January 2006 (UTC)

---The wording is so confusing at the beginning of the article. Instead of saying It's a measure of how much electric field is "generated" per unit charge (but backwards, because less is "generated" for more permittivity) why not say it's a measure of how much charge is required to generate a given electric field? That way there's no mind game where everything is backwards, and its measure is no longer inversely proportional to its quantity. 11:57 30 October 2015 (US central) — Preceding unsigned comment added by (talk) 16:58, 30 October 2015 (UTC)


As for terminology: 'electric flux densily' gives about 2M google hits whilst 'electric displacement field ' gives about 2.3M. Hardly a big difference!--Light current 03:37, 3 January 2006 (UTC)

I was always taught "flux density" in school and didn't hear "displacement field" until much later. I find them both equally clear since I know what flux is, and I know what displacement current is, and they are very closely related ideas in my mind. I doubt it matters which term we use, especially since they link to the same page. If I had to choose sides (but why bother, this is so nitpicky) I would pick flux density because it's less likely to be confused with electric field. Xezlec 21:29, 3 January 2006 (UTC)
You forgot to put quotes in your Google search. If you search for the specific phrase "electric flux density" you get 17,000 hits. If you search for "displacement field" you get about 190,000 hits. If you search for "electric displacement" you get 50,000. "electric displacement field", which hardly anyone uses because it is too verbose, gets 530. —Steven G. Johnson 02:28, 4 January 2006 (UTC)

You are correct. however I still think 'flux density' will be understandable by more people and is less likely to be confused with electric field. --Light current 02:35, 4 January 2006 (UTC)

I fail to see the reasoning behind your assertion, especially confronted with the factor of 10 difference for Google. The only reason for the term "flux density" is in analogy with the B field which was historically called the "magnetic flux density", but these days most people refer to both B and H as a "magnetic field" and have forgotten this 19th-century accident. "Displacement field" is also the term used in major textbooks such as Jackson, and is mnemonic with "D". Anyway, this discussion belongs in Talk:electric displacement field. —Steven G. Johnson 03:00, 4 January 2006 (UTC)
I agree with light current and xezlec in that a better term to use would be electric flux density. First, I’m hesitant to differ to google as to what is the best term to use for describing a physics phenomenon. As to the textbook usage, Jackson (along with Griffiths) mainly use the term electric displacement, however, David Cheng’s, Sadiku’s, Ulaby’s, and Balanis’s books all use the term “electric flux density.” While non consensus may be reached, my small sample has shown a greater use of the second term. Second, examination of the SI units of coulombs/m^2 would intuitively lead to a “density” type term. If I heard the term “current,” I would naturally think of coulombs/second. Third, a better symmetry results with Maxwell’s equations when:
E, electric field intensity, V/m
H, magnetic field intensity, A/m
D, electric flux density, C/m^2
B, magnetic flux density, Wb/m^2
Mak17f 20:04, 25 February 2006 (UTC)

Well its not a big deal I suppose!--Light current 03:03, 4 January 2006 (UTC)


There seems to be a lot of unfortunate duplication going on. Compare permittivity, polarization density, electric susceptibility, dielectric constant, and refractive index. I'm not sure how to fix it, however. :-( —Steven G. Johnson 07:25, 4 January 2006 (UTC)

Could all be pages linked from a hub page called dielectrics or properties of dielectrics--Light current 19:36, 4 January 2006 (UTC)
Good idea, but I would prefer a navigation sidebar. —Steven G. Johnson 23:49, 4 January 2006 (UTC)
Whatever turns you on! Actually thats a good idea we should use on a lot more pages!--Light current 23:54, 4 January 2006 (UTC)
Please see the Electronics pages for an example of a series list solution to the organisation and avoidance of duplication.--Light current 16:23, 6 March 2006 (UTC)

Split Infinitive[edit]

"partially to cancel"? Yuck! See split infinitive. Pfalstad 20:10, 15 January 2006 (UTC)

Ive changed it to the other alternative--Light current 20:14, 15 January 2006 (UTC)

Merge proposal[edit]

Oppose: I think there's enough distinction betweeen the two topics to have separate pages--Light current 21:51, 25 February 2006 (UTC)

What do you feel is the main distinction? The difference of 1 between susceptibility and relative permittivity? The content at the bottom of the susceptibility page could be inserted into any of the "electric fields in materials" type pages I mentioned below. The definition section is repeated on several pages. Mak17f 22:04, 25 February 2006 (UTC)

Support: As Steven Johnson mentioned above, there is a lot of overlap in several articles that deal with fields within dielectrics: dielectric constant, permittivity, electric flux density, polarization density, displacement current, dielectrics, electric susceptibility and refractive index. One massive article may not be appropriate, but some consolidation is probably in order. Having the electric susceptibility article absorbed into this article is probably a first step. Mak17f 21:55, 25 February 2006 (UTC)

Oppose: One could even argue that susceptibility is a 'more fundamental' (as in, stands in closer relation to particle physics) property of matter than relative permitivity is. susceptibility should definitely have it's own page I think. Tjeerd130.89.18.106 13:21, 6 March 2006 (UTC)

Ok, then the argument should be made on the susceptibility page as to how it closely relates to particle physics. I still don't see how it is more than only a difference in name and a value of 1 from relative permittivity. Of the topics I and Steven listed above, a case could be made for each of them of how they are very subtly different. However, I think the goal should be to best present the whole topic. Your brief point with the comparison to particle physics should certainly be made, but is it necessary for a reader to visit another page to read it? Mak17f 15:57, 6 March 2006 (UTC)
Also, can you find a text where Polarization density, permittivity, and electric susceptibility are presented as functionally seperate topics? Mak17f 16:18, 6 March 2006 (UTC)

There are also the articles Electric constant and Magnetic constant. I've changed the redirect of Permeability of free space to Magnetic constant, but now that I see the discussion here, it seems that the merging supporters might not like it. But as it is, if the articles are not merged, than the redirects (and what links there) should point to the more specific article. —Yoshigev 10:32, 3 April 2006 (UTC)

What a mess! This calls for massive merging. (I am in favour of dielectric constant.) /Pieter Kuiper 19:31, 6 August 2007 (UTC)
I tend to see dielectric constant in my own field as well. One note of caution: some people use dielectric constant to mean only the static (zero frequency) dielectric constant, and others use it for the (frequency dependent) relative permittivity. (And of course, those of us who are theorists and use sensible units don't distinguish between permittivity and relative permittivity. ;-) Anyway, one thing we can all agree on is that it is a mess. —Steven G. Johnson 16:48, 7 August 2007 (UTC)
I will try to have a look at what the preferred terminology is according to IUPAP, when I get around to it. After that one should probably take this up at Portal:Physics. /Pieter Kuiper 17:11, 7 August 2007 (UTC)
I checked the IUPAP Red Book, and they use 'permittivity', 'relative permittivity' and 'permittivity tensor'. The red book is 20 years old now. My guess is that if IUPAP would make a new recommendation based on a description of current usage, that 'dielectric constant' might be chosen.
So this did not help much :( /Pieter Kuiper 08:50, 8 August 2007 (UTC)

With the chemists of IUPAC, I now found a reference article about nomenclature:

It would be best if wikipedia conformed. I will start with an article about the "electric constant", and changing existing redirects so that they conform to the guidelines. After that there remains the problem of merging. I also started diskussing this at Portal_talk:Physics#Reference_for_nomenclature /Pieter Kuiper 14:59, 8 August 2007 (UTC)

I do research in classical electromagnetism, and I've never seen "electric constant" (which your reference is using as a synonym for vacuum permittivity) as far as I can recall, so this terminology is far from universal. Please use "vacuum permittivity", which everyone (even the chemists) understands, for . For the reason I mentioned above, "dielectric constant" (while common) is ambiguous because it historically referred only to the static (zero-frequency) relative permittivity. My suggestion would be to use terms that are unambiguous and universally accepted: vacuum permittivity, permittivity, relative permittivity, and static relative permittivity. The articles can then note the common synonyms: electric constant for vacuum permittivity, and dielectric constant (or dielectric tensor in the case of anisotropic media) for the relative permittivity (or relative permittivity tensor), and dielectric constant as the classic term for the static relative permittivity. (People also talk about the "dielectric function" for media that are inhomogeneous.) —Steven G. Johnson 16:13, 8 August 2007 (UTC)
Electric constant may not be the most common term, but it is a recommendation, also in physics. NIST is conforming, see electric constant. So should we, the only way out of the quagmire. /Pieter Kuiper 17:22, 8 August 2007 (UTC)
See Talk:Vacuum permittivity: just because one NIST page uses "electric constant", doesn't mean that it is the "recommended term" — you can easily find (with Google) other NIST pages that use "vacuum permittivity". Thus, all we have is evidence that both terms are common. —Steven G. Johnson 18:12, 8 August 2007 (UTC)

Please take further discussion of this particular issue (what to call ) to Talk:Vacuum permittivity so that we don't have to carry on the same discussion in multiple places. Note that this is separate from what to call the relative permittivity, etc. —Steven G. Johnson 18:16, 8 August 2007 (UTC)

Merge for now: This material is too important to be left as an ignorable stub. It should be brought into permittivity, connected to Kramers-Kronig relation, Green-Kubo relations, Green's function (many-body theory), linear response function, Density functional theory and to optical absorption and expanded to give examples for various materials, e.g. a semiconductor, a metal , and an insulator. It should be related to methods of computation; see Silva I and Silva II. Eventually this topic will become big enough to go back to a separate article, but that awaits some authors. Brews ohare (talk) 17:18, 2 April 2008 (UTC)

I redirected dielectric function to Permittivity#Complex permittivity. Probably someone will think that this was too bold, and revert. That is fine, I won't mind. /Pieter Kuiper (talk) 21:27, 2 April 2008 (UTC)
I added more of the other article and threw in a bunch of references that show what kind of stuff should go here; but have not summarized it for the article. Brews ohare (talk) 01:02, 3 April 2008 (UTC)

Classification of materials[edit]

I was a bit surprised by the line:

A perfect dielectric is a material that exhibits a displacement current only, therefore it stores and returns electrical energy as if it were an ideal battery.

Does anyone object if I replace battery with capacitor? --Arthur

Why capacitor? Karol 16:43, 25 April 2006 (UTC)

Because it is more accurate and relevant than a perfect "secondary" battery. However, a battery is an electrochemical system with a different (chemical) energy storage mechanism than the (electrostatic) energy storage system being discussed. I've made the change. Bert 13:30, 27 May 2006 (UTC)

I have some concern over the passage,

Materials can be classified according to their permittivity. Those with a permittivity that has a negative real part are considered to be metals, in which no propagating electromagnetic waves exist. Those with a positive real part are dielectrics.

Metals are associated with a very high conductivity, giving rise to a highly lossy medium, not with the sign of the real part of the permittivity. A negative real part would be associated more with a resonating plasma or left-handed material in my opinion.

This also brings me to another point about the sign of the imaginary part of the complex permittivity. Using the Poynting theorem, the specifications for a lossy source-free medium is that both and are Hermitian, negative definite. Thus, the imaginary part of our permittivity tensor must be positive for a lossy medium. A negative sign would be associated with an active medium. As such, the sign should be corrected in the complex permittivity tensor and throughout the preceding section such that we have positive imaginary parts. For reference look at Chapter 1.1.5 of Weng Chew's "Waves and Fields in Inhomogeneous Media." Chapter 7.5.B of 3rd Edition of Jackson's text also states, "Since a positive imaginary part to represents dissipation of energy from the electromagnetic wave into the medium, ..." Sorry to be a little long-winded on this but I found it annoying that the sign error was reintroduced. --Born2bwire 22:46, 5 June 2006 (UTC)

OK you seem to know what you are talking about! Would you like to fix it for us? I think it would be good to include the ref you quoted also! Thanks 8-)--Light current 22:59, 5 June 2006 (UTC)
Yes please.. The part about metals having a negative real part seems obviously wrong. Pfalstad 23:10, 5 June 2006 (UTC)

Done, though it would be best if someone could adjust the formatting for me. The loss tangent relations, , are inexplicably large compared to the body of text they are located in. --Born2bwire 23:48, 5 June 2006 (UTC)

Looks OK in my browser! And thanks 8-) --Light current 00:08, 6 June 2006 (UTC)


Do we really need this unit to be quoted?--Light current 01:23, 23 May 2006 (UTC)

Link to the statement on permeability of free space[edit]

There is a link at the bottom of the article which takes the victim^h^h^h^h^h^hreader to a statement which basically begs the original question "What is the significance of the premiability of free space?". It is not an uncommon response to the question, but I believe it is neither helpful nor substantive. The fact that units can be scaled to eliminate the constant coefficients does not demonstrate that the constants has no physical significance. By appropriately selecting units, one can also eliminate the constance c from the equations of relativity. That does not, however, change the fact that the parameter of time for which it is a scaling factor stands on a different footing from the other three dimensions. Indeed, by eliminating c from the expression of relativity, we lose an explicit indicator of that uniqueness of the time dimension.

Dielectric absorption phenomena[edit]

Dielectric absorption can also refer to the development of a voltage across the terminals of a previously discharged capacitor. I've added a short discussion about it in this section. I don't know it this is the right section to mention it, but it should be covered somewhere either in capacitors, dielectrics, or permittivity... Bert 13:09, 14 June 2006 (UTC)

Making light move "backward"[edit]

See Scientific American this year, either the July or August article, which describes how the wave peaks can be decoupled from the energy envelope and made to move in the opposite directions. Requires a material with the right permittivity and has major consequences for imaging and data storage.

The negative-refractive index stuff you are referring to requires a material that has both negative ε and μ, which was first proposed by Veselago in 1968. The recent developments center on composite "metamaterials" that achieve this property via an array of tiny (subwavelength) structures, but their practical impact is still being debated. Opposite phase and group velocity, however, is possible with "ordinary" materials and has been known for decades (at least), for example in hollow metallic waveguides with an inhomogeneous dielectric core (e.g. Clarricoats and Waldron, J. Electron. Contr. 8, p. 455, 1960). —Steven G. Johnson 17:31, 14 August 2006 (UTC)
Main thing to note that this is partly a consequence of how we describe geometry. When we talk about the cross product of vectors, we always choose the right-hand orientation for the resulting vector. But this orientation is arbitrary, we could have chosen to use our left-hand. So what happens in a double-negative/metamaterial/left-handed material is that the wave now follows the left-handed convention. So the progression of phase, as Steven noted, is reversed and the refractive angles between media refract in the opposite direction. One of the resulting effects is that you can have a planar lens that focuses a point source back. Hence why press releases generally say that the light travels "backwards." A negative permittivity is not a new phenomenon though, this also happens around the resonance frequencies of plasmas, like in the ionosphere. Personally, I do not like a lot of the language used in these articles where the say that the light travels backwards or that they can stop it all together. This does not happen but they cannot find a way to actually explain the subtle difference why to laymen. Either way, I believe the Wikipedia already has separate articles pertaining to left-handed materials.--Born2bwire 03:41, 24 August 2006 (UTC)
People in glass houses shouldn't throw stones? Negative permittivity alone (e.g. in metals or plasmas) leads to evanescent waves only, not backwards waves. The combination of both negative ε and μ, which leads to opposite phase and group velocities, was achieved experimentally only very recently as I alluded to. And group velocities approaching zero (in a context where group velocity is a meaningful concept, unlike the "superluminal" stuff from a few years ago) does indeed occur both theoretically and experimentally. —Steven G. Johnson 04:09, 24 August 2006 (UTC)
I can't understand the claim of Bornb2wire, which appears to say that that negative refraction is due to a convention about the handedness of the cross product. It seems that if you perform an orientation-reversing transformation of space, you won't transform a positively refracted ray to a negatively refracted one. They are that different. (talk) 22:53, 6 April 2014 (UTC)


Permittivity measurements are experimentally difficult to do accurately. There are many ways to do it. Split-Post resonators, like that invented by Mike Janezic at NIST, and the calibration comparison technique pioneered by Dylan Williams and Roger Marks. I work in this field, but as I have said I am a newbie so I would prefer if someone would help me write this section. —Nathan

Epsilon Naught[edit]

Why does "Epsilon naught" redirect here? First of all, this page never mentions anything about Epsilon Naught. If somehow Epsilon Naught does relate to this topic (I have no knowledge of it myself) will someone mention it, even briefly in the article on how it relates to give a reason for its redirect here?

And furthermore, I would believe that Epsilon naught should redirect to Epsilon nought, because "naught" is a common misspelling of "nought" (I must admit, that spelling error is what brought me here in the first place.). 22:36, 16 July 2007 (UTC)

Quantum Mechanical Interpretation:

1rst Point :

Is water blue because it absorbs more red light ? Or because it scatters more blue light which can only be seen in large amounts of water like lakes ?

Also how much of a lake blue color is due to simple reflexion of the blue sky above ? The same for the reflexion of an ice-block and the ocean blue color ?

So, I'm not convinced at all that water is blue, specially when by just looking at it you can see that it's colorless. Sure our eyes sensitivity can't be compared to a UV-VIS spectrometer and that's why we would need a very large amount of water to see a color. I think this part of the text really needs more data, like the spectra of water, some relative absorptions, and the relative scattering effect on lab conditions, and not on a lake under the blue sky ;-)

2nd Point :

Our eyes are not damaged by light because they have water ? That should really be removed ! The eyes have water because water was abundant during evolution and also is a transparent colorless liquid at body temperature, as any other liquid with same caracteristics will do, which allows one to see under the VIS spectrum.

And yes even near UV-light can definitely damage the eyes, a typical example are the crystalline, lenses made of 'almost dead' cells that do not regenerate, which don't have nucleus but are just filled with a water + proteins solution. The light changes the proteins conformations over time (even will all that water around) and when we are old enough we have to change them by some gel-like artificial lenses. So water does not protect the eyes from light !


Phase velocity?[edit]

The permittivity \varepsilon and magnetic permeability μ of a medium together determine the phase velocity v of electromagnetic radiation through that medium

Phase velocity, or signal velocity? Phase velocity is not a physically meaningful quantity. It can go to infinity if it wants to. -- (talk) 06:44, 8 January 2008 (UTC)


I've read through this article and haven't been able to glean whether a conductor has a high permitivity!? While I only have an A-level in physic I should think it'd be even worse for a complete layman! (talk) 03:00, 6 April 2008 (UTC)

So this is a couple years late, but hopefully this will be helpful for anyone who wonders the same thing. At low frequencies it is not very useful to talk about permittivity of a metal/conductor because the response is thoroughly dominated by the conductivity. Using the common complex permittivity definition, you get an epsilon=something-j*(something huge). The sign in the middle depends on whether you're using physics or engineering notation, it's just determined by energy conservation. The real part (commonly called the permittivity) doesn't really matter because the imaginary part is so large. At much higher frequencies (around THz or so) you get close to a resonance frequency of many metals. In this case it DOES make sense to talk about the permittivity, since it is dominating the response. The most interesting ranges are where the permittivity is negative, which directly leads to plasmonics.
I guess the take-home (or tl;dr version) is that since permittivity depends on frequency (in the time-harmonic case) or all previous times (in the instantaneous case), it's hard to discuss the permittivity of metals without specifying what frequency range you're referring to. Hey, at high enough frequencies, everything has a permittivity equal to that of free space (the charged particles can no longer move fast enough to respond to a field). Hope this helps. (talk) 23:12, 28 September 2010 (UTC)

Just a pun, or useful material[edit]

The recent research into the use of titanate compounds for capacitors makes me wonder how long before the market sees a "High permittivity paint" ? This could be a very useful thing to have around for constructing enormous capacitors capable of enabling off-hours generation of electrical energy, as it would make construction of them much cheaper than the current microfabrication techniques require - even if the resulting capacitor ended up being several times larger. Zaphraud (talk) 14:21, 9 September 2008 (UTC)


as a hardly technical point why is an ellipsis claimed to indicate a terminating decimal? This is just the opposite of the usual convention. —Preceding unsigned comment added by (talk) 21:24, 4 February 2009 (UTC) On further thought it properly indicates continuation not termination of the irrational number. In any case the expression is an approximate equality, so the ellipsis is optional and surely the cumbersome explination of its meaning can go. —Preceding unsigned comment added by (talk) 17:00, 5 February 2009 (UTC)

Permittivity is a measured quantity[edit]

There was a serious error in this article in that the section that deals with the numerical value of ε uses the equation c^2 = 1/(εμ), with c already being a defined value, in order to justify the value of ε. The problem here is that since both c and μ are defined quantities in SI units, we cannot use them to determine a measured quantity. When ε is measured using a capacitor circuit, we obtain a value which can then be substituted into this equation to yield a value that is very close to the speed of light. However, we cannot work in reverse and use the defined speed of light to determine the measured value of ε. David Tombe (talk) 11:50, 13 August 2009 (UTC)

You cannot measure the permittivity of vacuum, because the meter is defined in terms of the speed of light which comes from the permittivity. (e.g. how do you measure the distance between the plates of your capacitor in a way that independent of the permittivity of vacuum? You can't.) You can only try to measure the relative permittivity of different materials/conditions.
Please direct further discussion to Talk:Vacuum permittivity rather than propagating your misunderstanding of SI units to multiple pages simultaneously. — Steven G. Johnson (talk) 00:07, 14 August 2009 (UTC)

Steven, I'll do as you say and continue this at Talk:Vacuum permittivity. David Tombe (talk) 11:38, 14 August 2009 (UTC)

Unsourced article[edit]

This article lack source particularly article or book (talk) 14:14, 18 December 2009 (UTC)


Near the end of the section called "Quantum-mechanical interpretation" there is a reference to the Lorentz model, but it directs to a mathematical page, which does not really seem like an appropriate reference. Can this be improved? —Preceding unsigned comment added by (talk) 14:13, 9 September 2010 (UTC)

Dielectric Spectrum diagram[edit]

The diagram of the dielectric spectrum needs to be changed - 'depolar' ought to be 'dipolar'. —Preceding unsigned comment added by (talk) 20:08, 10 October 2010 (UTC)

Yes check.svg Done (It was not difficult: SVG is just a program in ascii format, one can search and replace the text string, and reupload the file.) /Pieter Kuiper (talk) 10:05, 11 October 2010 (UTC)

Removed erroneous generalization about sunlight and water-filled organs[edit]

Removed: "This is why sunlight does not damage water-containing organs such as the eye.[1]"

The preceding statement was referring to the reduced absorption of UV, which is a feature of the clear, watery, "vitreous humor" of the eye. This actually allows UV to penetrate to the retina and cause damage. Ryan Westafer (talk) 17:22, 15 December 2010 (UTC)


  1. ^ Braun, Charles L.; Smirnov, Sergei N. (1993). "Why is water blue?". Journal of Chemical Education. 70 (8): 612. doi:10.1021/ed070p612. 

Plots of Metal Permittivities[edit]

Hi all. As part of my academic work, I have produced the wavelength dependent permittivity and refractive index for quite a lot of metals produced using different models (Drude, Lorentz-Drude, Brendel-Bormann). I thought it would be useful to include somewhere here. Any suggestions on the best place for the data, and the appropriate format to plot it in (HTML5, static images, table, etc.)? Drnathanfurious (talk) 16:24, 17 September 2012 (UTC)

P(omega) integral over the frequency domain[edit]

Is , when integrated over the frequency domain, equal to the polarization current density?siNkarma86—Expert Sectioneer of Wikipedia
86 = 19+9+14 + karma = 19+9+14 + talk
09:45, 2 January 2013 (UTC)

Error in Formula[edit]

In the section on lossy medium, the second equation (for epsilon^hat) has a typo. The term "-i" should be "+i".

Elee1l5 (talk) 20:58, 17 February 2013 (UTC)

Bad Notation[edit]

The notation used for epsilon^hat in the section on lossy medium is not good, since it is inconsistent with the notation used earlier in the article (when there is no conductivity). If epsilon^hat denotes the complex conductivity due to dipolar loss, then the complex effective permittivity (which accounts for conductivity and dipolar loss) would be

epsilon^effective = epsilon^hat + i (sigma / omega).

I recommend using the notation epsilon^effective instead of epsilon^hat to denote effective permittivity. The total current is then

J^total = -i omega epsilon^effective E.

Elee1l5 (talk) 20:59, 17 February 2013 (UTC)

Ambiguous in intro[edit]

I propose that this be struck from the intro, or reworded to eliminate ambiguity:

"Thus, permittivity relates to a material's ability to transmit (or "permit") an electric field."

As a new learner, this threw me off. 'Permitting' an electric field to 'transmit' to me means not interfering with it, which is the exact opposite of what it means. It does still 'relate to' that, but inversely. Maybe that's what it implies. Regardless, 'relates to' can mean either thing, and there's no reason to be ambiguous when it can confuse.

I am not confident to make this change, given the uncertainty it caused me. Please someone do, if I'm not wrong. (talk) 04:59, 12 September 2013 (UTC)


The permittivity of a medium describes how much electric field (more correctly, flux) is 'generated' per unit charge in that medium: unit change of what? requirement: need to make notes Face-smile.svg thanks ~"aGastya" ✉ let’s talk about it :) 15:19, 19 February 2015 (UTC)

It says charge, not change. SpinningSpark 17:54, 21 February 2015 (UTC)