Valence Electron

A valence electron is one of the outermost electrons of an atom, the ones that sit in the highest occupied shell and therefore take part in chemical bonding, while the inner core electrons stay tightly bound to the nucleus and are almost never disturbed. An atom's chemistry, and much of its electrical behavior when packed into a solid, is set by how many valence electrons it has and how willing it is to share, lend or borrow them. For the elements used in transistors the count is the single most useful number on the periodic table, because it decides whether an atom acts as the crystal's backbone or as a deliberate impurity.

Silicon has four valence electrons, which is the property that makes it a semiconductor rather than a metal or an insulator. With atomic number 14, silicon has the electron configuration 1s2 2s2 2p6 3s2 3p2, so ten electrons form a stable neon-like core and the remaining four, two in the 3s and two in the 3p subshell, are valence electrons. Germanium, directly below silicon in the same column of the periodic table, also has four, and so does carbon in diamond. Elements in this column sit at the midpoint of the octet: they would need to gain or lose four electrons to reach a closed shell, which is too costly, so they prefer to share. That is why the group forms strong directional bonds in a solid instead of releasing free electrons like sodium or accepting them like chlorine.

The idea of valence electrons grew from G. N. Lewis's 1916 picture of the shared electron pair. In a paper that year and a book in 1923, Gilbert Newton Lewis proposed that atoms bond by sharing pairs of electrons so that each attains a stable group of eight, the octet; Irving Langmuir popularized and extended the idea around 1919. The octet rule is a heuristic and has many exceptions, especially for heavier elements and for hydrogen, which settles for two, but it explains the covalent bonding of silicon exactly: each atom contributes one electron to each of four bonds, and each shared pair counts toward both neighbors. Quantum mechanics later replaced the cubical atoms of Lewis with orbitals, describing the silicon bond as mixed sp3 hybrids pointing to the corners of a tetrahedron, but the bookkeeping of four per atom is unchanged.

In a crystal, valence electrons are the raw material for conduction, but only a few of them are ever free to carry it. In a perfect silicon crystal at absolute zero, every valence electron is locked into a bond, so the crystal conducts nothing. Supplying roughly 1.12 electron-volts, from heat or a photon, can tear one electron out of its bond; the electron wanders off and leaves behind a vacancy that behaves like a positive charge. At room temperature only about one bond in every twenty trillion or so is broken at any moment, which is why pure silicon is a poor conductor. The same reasoning applies to compound semiconductors, where the average count is still four: gallium contributes three and arsenic five.

Doping works by changing the number of valence electrons at a few chosen sites. Substituting phosphorus or arsenic, which have five, for a silicon atom leaves one electron without a bond partner, and it is released into conduction with little energy; substituting boron, which has three, leaves a bond one electron short, which captures an electron from a neighbor and so creates a mobile hole. Concentrations of from parts per billion to parts per thousand are enough to change conductivity by many orders of magnitude, and it is this controllability, unavailable in metals, that makes the transistor possible. Metals behave differently: in copper the single outer 4s electron per atom becomes a delocalized sea, so nearly every atom contributes a carrier, while in silicon the valence electrons are bound and only the freed minority conduct.

Misconceptions are common, and the first is that valence electrons are simply the electrons able to move. In silicon they are mostly immobile, held in bonds, and conduct only after excitation. Counting valence electrons for transition metals and for atoms beyond the second row is also less tidy than the octet picture suggests, since inner d subshells can participate. And a Lewis diagram is bookkeeping, not a photograph; the electrons are delocalized probability clouds, and the bond is a region of enhanced electron density between nuclei rather than two dots on a line. In today's chips the same logic still organizes the choices of materials, from the boron and phosphorus used to dope silicon to the gallium and nitrogen in power electronics.

Read valence electron as the outermost electron count that sets how an atom bonds, and therefore whether a crystal holds its charge still, releases it, or leaves a hole, the single number behind both the structure of silicon and the art of doping it.