What distinguishes a metal from a nonmetal in terms of atomic structure?

[live4physics]

What´s the main difference between a metal and nonmetal in terms of atomic structure of elements ?

Thanks.

 

[fatra2]

Hi there,

From what I remember from my solid state physics, the difference lies in the valence eletron.

To be a more precise, in a conductor (metal), the energy gap to a zero potential of the valence electron is close to null. Otherwise, the greater the energy gap, the better insulation the matter will be.

Hope this helps. Cheers

 

[granpa]

the electrons in a metal form a degenerate gas (so i am told). I don't know if this has any connection with 'degenerate matter' or not.

in conductors resistance increases with increasing temperature.
in semiconductors resistance decreases with increasing temperature.

 

[fatra2]

Hi granpa,

You seem to mix up different theories. The conductivity matter is definetily dependant on its temperature. But this conductivity is a macroscopic measure that gives an idea on the ease of electrons to move into this matter. Where temperature plays a role is in the energy gap between bound and free electrons.

You are almost right, "free" electrons in a metal are often link to a gas of electrons, mainly because of its macroscopic behaviour. Once again, physics is trying to model movement of many, many, many electrons in a conductor with simple macroscopic equations, in order to obtain approximate data.

 

[granpa]

metals are conductors. electrons are free to move
nonmetals are dielectrics. electrons are bound to the nuclei.

conductors reflect all of the light from their surfaces. dielectrics relfect some of the light from their surfaces (like from the surface of water)

you might want to study same electrostatics.

 

[fatra2]

Hi there,

To make it clear (and so you should go back to your solid state physics book), electrons are not completely free in a metal. In some cases, we model them to be free, but they still "belong" to an atom. The only difference is that, the electrons need very little energy to go off from the valence band.

 

[ZapperZ]

metals are conductors. electrons are free to move
nonmetals are dielectrics. electrons are bound to the nuclei.

conductors reflect all of the light from their surfaces. dielectrics relfect some of the light from their surfaces (like from the surface of water)

you might want to study same electrostatics.

Hi there,

To make it clear (and so you should go back to your solid state physics book), electrons are not completely free in a metal. In some cases, we model them to be free, but they still "belong" to an atom. The only difference is that, the electrons need very little energy to go off from the valence band.

fatra2 is correct. While the simplistic description of "free" electrons is sufficient in many cases, it isn't accurate in many others.

In most cases, when we talk about "conductors" and "insulators/semiconductors", we are talking about band solids, which means that the standard band structure is what defines what they are. In fact, for these materials, if you have a band dispersion that crosses the Fermi energy, that's a sufficient criteria to call something a "metal" or a conductor. This is similar to what fantra2 has said, there is practically no gap between the occupied and unoccupied states, and thus, electrons requires almost no energy to be mobile.

Reflectivity is a VERY weak criteria to use to distinguish between a metal and an insulator. I have a UV light (248 nm wavelength), and I use a dielectric "mirror" to reflect it, and it looks TRANSPARENT in the visible range.

Zz.

 

[granpa]

the op might find this interesting:
http://en.wikipedia.org/wiki/Plasmon

 

[genneth]

See here for a discussion:
http://nanoscale.blogspot.com/2009/01/what-does-it-mean-for-material-to-be.html
http://nanoscale.blogspot.com/2008/12/more-about-insulators.html

 

[genneth]

If I had to nail down what counts as a metal, it would be the existence of free charge carriers at zero temperature. Or equivalently, the existence of finite current states arbitrarily close to zero energy.

A common way for this to occur is if there are charge-carrying excitations at arbitrarily low energy --- these excitations can be collective or otherwise complex.

The links above point out several ways in which a material can fail to have this property, and some other properties that all follow from the above summary. However, the line is not absolute --- there's definite grey areas no matter how you try to define it.

On a slightly off-topic point, this is the kind of question where there is no one right answer. Condensed matter is about the simple answers --- but here, there's not really a *simple* yet completely correct answer. It entirely depends on the OP's background and purpose. For a school course some words about valence electrons, or even rough pictures of bands would be sufficient. For an advanced undergraduate course I would expect a discussion of Bloch wavefunctions, band dispersion and Fermi surfaces. Beyond, I expect people to know better than to ask.

 

[ZapperZ]

If I had to nail down what counts as a metal, it would be the existence of free charge carriers at zero temperature. Or equivalently, the existence of finite current states arbitrarily close to zero energy.

A common way for this to occur is if there are charge-carrying excitations at arbitrarily low energy --- these excitations can be collective or otherwise complex.

The links above point out several ways in which a material can fail to have this property, and some other properties that all follow from the above summary. However, the line is not absolute --- there's definite grey areas no matter how you try to define it.

On a slightly off-topic point, this is the kind of question where there is no one right answer. Condensed matter is about the simple answers --- but here, there's not really a *simple* yet completely correct answer. It entirely depends on the OP's background and purpose. For a school course some words about valence electrons, or even rough pictures of bands would be sufficient. For an advanced undergraduate course I would expect a discussion of Bloch wavefunctions, band dispersion and Fermi surfaces. Beyond, I expect people to know better than to ask.

That is why I made specific reference about "band solids", which is what we commonly associate the definition for "metals" and "semiconductors/insulators". I certainly would not want to go into more exotic solids such as Mott insulator and charge transfer metals. From the way the OP asked the question, I only expect the most rudimentary answer based on the simple band solids description.

One of the things I used to investigate in strongly correlated electron system is the metal-insulator transition, and specifically to distinguish between the Mott-Hubbard system versus the Brinkmann-Rice metals. We characterize "metallic" and "semiconducting" behavior predominantly based on the slope of the resistivity curves versus temperate. A positive slope we characterize as "metallic" behavior, while a negative slope signifies a semiconducting behavior. In some material, such as the 2D Ba-doped cobaltates, both behavior exists, where one transition to another as one changes the temperature, meaning that the material went from metallic to semiconducting, and vice versa (see, for example, Valla et al, Nature 417, 627 (2002)). In both "phases", there ARE "conducting charges", since the conductivity isn't zero. But it is how the properties behave as you change temperature that distinguish it as being metallic or semiconducting.

So certainly the physics here can get VERY complicated if we want to consider all the possible systems. However, I believe that for THIS particular question in this thread, I would guess that the band solids explanation is sufficient.

Zz.