The semiconductor are materials presenting a electric Conductivité intermediate between the metals and the Isolant S.

  • the semiconductors are paramount in electronic because they make it possible to control, by various means, at the same time the quantity of Electric current likely to cross them and the direction which this current can take.

  • In a semiconductor a Electric current is supported by two types of carriers: the electron S and the hole S.
    • the propagation via electrons is similar to that of a conducting traditional: Atome S strongly Ion ized pass their electrons in excess along the driver of an atom to another, from a zone ionized negatively with another less negatively ionized.
    • the propagation via holes is different: here, the electric charges travel from a positively ionized zone to another less positively ionized by the movement of a hole created by the absence of an electron in a quasi full electric structure.
  • the pure Silicium is a intrinsic Semi-conducteur. The properties of a semiconductor (i.e. the number of carriers, electrons or holes) can be controlled in the doping agent with impurities (other materials). A semiconductor presenting more electrons than of holes is then known as of standard NR, while a semiconductor presenting more holes than of electrons is known as of standard P.

Electronic structure of the semiconductors

Principle of the structure in bands

The properties of the semiconductors come from their electronic structure.

The states of the electrons in the matter are quantified, i.e. only certain energies and vectors of waves are accessible for them. That is explained in particular by their containment within material. In addition, there exists a relation between the vector of wave \ vec {K} = \ frac {\ vec {p}} {\ hbar} of an electron and its energy E.

For a free electron, energy is equal to its kinetic energy:

E= \ frac {p^2} {2m} = \ frac {\ hbar^2 k^2} {2m}

On the other hand, when the electron is in the middle of a crystal, a term of potential energy is to be added. This term describes the influence of the electrostatic potential of the ions of the crystal on the electron.

This influence remains weak as long as the vector of wave of the electron is without relationship with the periodicity of the crystal. On the other hand, if the vector of wave of the electron (to some extent its movement) is in phase with the crystal, it strongly will interact with him and the energy of the electron will be strongly modified about it. But this modification of energy can be as well positive as negative according to whether the electron is in advance or late of phase on the crystal. To give an image, on a side, the electron is slowed down by the ions of the crystal, other it is accelerated. There is a phenomenon of degeneration, i.e. for the same vector of wave, the electron can have different energies. The parabolic curve corresponding to the free electron E \ propto k^2 is thus found “broken” on the level of the values of k corresponding to the crystal, with the result that certain energies become inaccessible. One calls these zones of the forbidden bands (or gap) because corresponding energies cannot be reached by the electrons. To summarize, it is a phenomenon of interference between the wave of the electron and the electric field of the crystal lattice which explains the formation of the forbidden bands. The calculation of the functions of wave of the electrons in the crystal is formalized thanks to the functions of Bloch. The diagram of bands is separate horizontally in various zones called zones of Brillouin, corresponding to the frequency of periodicity of the crystal. All the zones are in general represented folded up on the first, the vectors of waves of the electrons being defined in a vector of the crystal close (similar phenomenon with the folding up of spectrum at the time of a sampling).

Elements of the type IV (C, If, Ge, Sn) have an electronic structure of type (.s ², .p ²), and can form orbital \ sigma and \ pi flexible and antiliantes. When these atoms form a crystal lattice, the energy of orbital flexible tends to decrease, whereas that of orbital the antiliantes increases (according to the interatomic distance). In parallel, the energy levels corresponding to orbital tend to be spread out around a mean level, a phenomenon due to the interaction of the orbital ones; one speaks then about bands of energy rather than of levels.

The band corresponding to the spreading out of orbital the σ antiliante is called Bande of conduction ; the band corresponding to the spreading out of orbital the π flexible is called Valence band .

As long as the energy of the band of conduction is lower or comparable with that of the valence band, of the electrons can circulate freely in the crystal: the solid is conducting. It is the case of the Magnésium (Mg). Other metals as the Cuivre (Cu) have empty states in the valence band. In this situation, the electrons of the valence band can conduct electricity while moving between these states and the material is there too good conducting.

If, on the other hand, the energy of the valence band becomes lower than that of the band of conduction (when the mesh sizes crystalline decrease), the electrons will populate all the flexible levels: the solid becomes insulating with the absolute zero. The difference in energy between the band of conduction and the valence band is called gap material.

If the gap is very important compared to thermal energy (> 200 kT), almost no electron populates the band of conduction: the material is insulating. With the room temperature kT= 26 meV (milli electronvolt), that means that with gap of 5 eV one deals with insulating good. For silicon, the gap is worth 1,12 eV and even for an intrinsic material, the band of conduction contains some electrons produced by thermal generation which are enough to ensure a minimal conduction (resistivity about 3.10^5 \ Omega.cm): the material is known as semiconductor .

Level of Fermi

The electrons are fermions, particles not being able to divide in the several same state, in accordance with the Principe of exclusion of Pauli. They thus follow a Statistique of Fermi-Dirac and the occupancy rate of the states for an energy E is written in the form:

F (E) = \ frac {1} {1 + e^ {\ frac {(E - E_f)} {kT}}}

The value E_f is called Énergie of Fermi . It is the maximum energy of the states occupied by the electrons with the Absolute zero. Its value depends on the number of electrons of valence and the density of accessible states; it is thus characteristic of material. In the semiconductors, the level corresponding to this energy (the level of Fermi) is in the gap. That has as a consequence that with 0 K, the valence band is full while the band of conduction is empty. When the temperature increases, certain electrons are able to exceed the level of Fermi and thus to reach the band of conduction.

When an electron is excited towards the band of conduction, it leaves behind him an empty state (a hole ) in the valence band, corresponding to an electron missing in one of the covalent bonds between atoms. Under the influence of an electric field, a nearby electron of valence can move in the place of the missing electron, moving this place at the same time. This hole is then able to move through material and thus to conduct electricity. The electrons and the holes are indicated under the name of charge carrier .

The holes are regarded as particles of load opposed to that of the electrons (1,602×10−19 C). In the presence of an electric field, electrons and holes move in opposite directions. The electrons are more mobile than the holes and thus conduct electricity best. This movement of conduction results from the superposition of two electric fields: the electric field applied by outside and the periodic electric field resulting from the structure of the crystal. As this last is difficult to calculate exactly, it is replaced by a total contribution, which amounts modifying the mass of the carriers; one then speaks about effective mass . In the vicinity of the minimum of the band of conduction, the effective mass is a function of the derived second of the profile of energy of the band ( parabolic approximation ).

Because of the exponential character of the distribution of Fermi-Dirac, the concentration out of carriers strongly depends on the temperature. To increase this one led to increase the number of carriers and thus increases conductivity, unlike the majority of the drivers which tend to being less conductive at high temperature. This principle is used in the thermister S.

Because of the properties of symmetry of the crystal lattice, the energy levels of the bands are not equal in all the directions: there exist privileged axes of conduction. The semiconductors where the direction corresponding to the maximum of energy of the valence band and that corresponding at least of the band of conduction coincide are said to gap direct (case, for example of AsGa); others (If) are said to gap indirect .

Doping and intrinsic semi-conduction

Formation of the forbidden bands being due to the regularity of the crystalline structure, any disturbance of this one, that it is caused by a defect in the structure of the crystal or by a chemical impurity tends to make “permeable” the gap by creating accessible states there. This property is usually used to control the properties of semiconductor materials by establishing to with it of the called quite selected atoms doping agents ( Voir also the detailed article Dopage (semiconductor) ).

Intrinsic semi-conduction

A semiconductor is known as intrinsic when it is pure. He was not doped and its electric behavior depends only on the electronic structure of material.

The carriers all are created by thermal generation, i.e. by exciting electrons in the band of conduction thanks to a rise of the temperature. Consequently, an equal number of electrons and holes are created. The level of Fermi is located, in good approximation, with the mileu of the gap. With balance and the Absolute zero, the Valence band is occupied completion by the electrons (not of holes) while the band of conduction is empty.

These semiconductors do not lead, or very little, the current, except if one brings them up to high temperature.

Doping of the type NR

The semiconductors of the type NR are called semiconductor extrinsic. The goal of a doping NR is to produce an excess of carrying electrons in the semiconductor. In order to include/understand how such a doping is carried out, let us consider the case of silicon (Si). Atoms of If four electrons of valence have, each one being related to a nearby Si atom by a Covalent bond. If an atom having five electrons of valence, as those of the group V (GOES) of the periodic table (for example, the Phosphore (P), the Arsenic (Ace) or the Antimoine (Sb)), is built-in the crystal lattice, then this atom will present four covalent bonds and a free electron. This electron, which is not an electron of connection, is only slightly related to the atom and can be easily excited towards the Bande of conduction. At the ordinary temperatures, almost all these electrons are it. Like the excitation of these electrons does not lead to the formation of holes in this kind of material, the number of electrons exceeds the number of holes by far. The electrons are carrying majority and the holes the minority carrying . And because the atoms with five electrons have an additional electron “to give”, they are called atoms donors. The materials thus formed are called semiconductor of type NR parce that they contain an excess of electrons negatively charged.

Doping of the type P

The semiconductors of the type P are also extrinsic semiconductors. The goal of a doping P is to create an excess of holes. In this case, a trivalent atom, generally an atom of Boron, are substituted for a silicon atom in the crystal lattice. Consequently, it misses an electron for one of the four covalent bonds of the adjacent silicon atoms, and the atom can accept an electron to supplement this fourth connection, thus forming a hole. When doping is sufficient, the number of holes exceeds the number of electrons by far. The holes are then carrying majority and the electrons of the minority carrying are called acceptors.

Junction P-N

A junction P-N is created by juxtaposing a semiconductor doped NR with a semiconductor doped P. If one applies a positive tension on the side of the area P, the positive majority carriers (holes) are pushed back towards the junction. In same time, the negative majority carriers on the side NR (electrons) are attracted towards the junction. Arrived at the junction, either the carriers recombine (an electron falls into a hole) by emitting a possibly visible photon (LED), or these carriers continue their race through the other semiconductor until reaching the opposite electrode: the current circulates, its intensity varies into exponential tension. If the potential difference is reversed, the majority carriers on the two sides move away from the junction, thus blocking the passage of the current on its level. This asymmetrical behavior is used in particular to rectify the alternative course.

Junction P-N is at the base of the electronics component named Diode, which allows the passage of the electric current only in one direction. In a similar way, a third area can be doped to form double junctions N-P-N or P-N-P which form the bipolar transistors. In this case, the two semiconductors in the same way standard are called the transmitting and the collecting . The semiconductor located between the transmitter and the collector is called the bases , and has a thickness about the micrometer. When one polarizes the junction transmitter-bases on line, this one is busy whereas the junction base-collector is blocked. However the base is enough fine to allow the many majority carriers injected since the transmitter (strongly doped) to cross it before having time to recombine. They are found thus in the collector, producing a current controlled by the basic tension.


Companies

The principal production companies of component in semiconductors present in France are:

Signal 10 by importance of the sales in 2006:

  1. Intel 31580 M$ (- 11% compared to 2005)
  2. Samsung 19475 (+9%)
  3. Texas Instruments 13870 (+23%)
  4. Toshiba 10030 (+11%)
  5. STMicroelectronics 9930 (+12%)
  6. TSMC 9715 (+18%)
  7. Renesas 8170 (- 1%)
  8. Hynix 7375 (+32%)
  9. NXP 6365 (+14%)
  10. Freescale 6080 (+9%)

Source: Electronics international of November 30th, 2006

See too

including Fields

Under-fields

Concepts

External bonds

  • Semiconductor Technology Information on the industry of the semiconductors.
  • Etry Animation on the industry of the semiconductors.
  • NSM-Files Physical properties of the semiconductors (If, GaAs, etc), inter alia structure of bands, mechanical properties, electric, thermal and optical.
  • Principles off Semiconductor Devices
  • Semiconductor resource launch page
  • Semiconductor glossary

Simple: Semiconductor

Random links:Common Ancolie | Champagne Charlie (film) | Katy Storie | .bt | McLeod lady