The valence electrons in molecules like carbon dioxide (CO2) can be visualized using aLewis electron dot diagram. Incovalent bonds, electrons shared between two atoms are counted toward the octet of both atoms. In carbon dioxide each oxygen shares four electrons with the central carbon, two (shown in red) from the oxygen itself and two (shown in black) from the carbon. All four of these electrons are counted in both the carbon octet and the oxygen octet, so that both atoms are considered to obey the octet rule.
An isolated chlorine atom (Cl) has two and eight electrons in itsfirst and second electron shells, located near the nucleus. However, it has only seven electrons in the third andoutermost electron shell. One additional electron would completely fill the outer electron shell with eight electrons, a situation the octet rule commends. Indeed, adding an electron to the produce thechloride ion (Cl−)releases 3.62 eV of energy.[1] Conversely, another surplus electron cannot fit in the same shell, instead beginning the fourth electron shell around the nucleus. Thus the octet rule proscribes formation of a hypothetical Cl2−ion, and indeed the latter has only been observed as aplasma under extreme conditions.
A sodium atom (Na) has a single electron in its outermost electron shell, the first and second shells again being full with two and eight electrons respectively. The octet rule favors removal of this outermost electron to form the Na+ ion, whichhas the exact same electron configuration as Cl−. Indeed, sodium is observed to transfer one electron to chlorine during the formation of sodium chloride, such that the resulting lattice is best considered as a periodic array of Na+ and Cl− ions.
To remove the outermost Na electron and return to an "octet-approved" staterequires a small amount of energy: 5.14 eV.[2] This energy is provided from the 3.62 eV released during chloride formation, and theelectrostatic attraction between positively-charged Na+ and negatively-charged Cl− ions, which releases a 8.12 eVlattice energy.[3] By contrast, any further electrons removed from Na would reside in the deeper second electron shell, and produce an octet-violating Na2+ ion. Consequently, the second ionization energy required for the next removal is much larger – 47.28 eV[4] – and the corresponding ion is only observed under extreme conditions.
In 1864, the English chemistJohn Newlands classified the sixty-two known elements into eight groups, based on their physical properties.[5][6][7][8]
In the late 19th century, it was known that coordination compounds (formerly called "molecular compounds") were formed by the combination of atoms or molecules in such a manner that the valencies of the atoms involved apparently became satisfied. In 1893,Alfred Werner showed that the number of atoms or groups associated with a central atom (the "coordination number") is often 4 or 6; other coordination numbers up to a maximum of 8 were known, but less frequent.[9] In 1904,Richard Abegg was one of the first to extend the concept ofcoordination number to a concept ofvalence in which he distinguished atoms as electron donors or acceptors, leading to positive and negative valence states that greatly resemble the modern concept ofoxidation states. Abegg noted that the difference between the maximum positive and negativevalences of anelement under his model is frequently eight.[10] In 1916,Gilbert N. Lewis referred to this insight asAbegg's rule and used it to help formulate hiscubical atom model and the "rule of eight", which began to distinguish betweenvalence andvalence electrons.[11] In 1919,Irving Langmuir refined these concepts further and renamed them the "cubical octet atom" and "octet theory".[12] The "octet theory" evolved into what is now known as the "octet rule".
Walther Kossel[13] andGilbert N. Lewis saw that noble gases did not have the tendency of taking part in chemical reactions under ordinary conditions. On the basis of this observation, they concluded thatatoms ofnoble gases are stable and on the basis of this conclusion they proposed a theory ofvalency known as "electronic theory of valency" in 1916:[14]
During the formation of a chemical bond, atoms combine together by gaining, losing or sharing electrons in such a way that they acquire nearest noble gas configuration.
The quantum theory of the atom explains the eight electrons as aclosed shell with an s2p6 electron configuration. A closed-shell configuration is one in which low-lying energy levels are full and higher energy levels are empty. For example, theneon atom ground state has a fulln = 2 shell (2s22p6) and an emptyn = 3 shell. According to the octet rule, the atoms immediately before and after neon in the periodic table (i.e. C, N, O, F, Na, Mg and Al), tend to attain a similar configuration by gaining, losing, or sharing electrons.
Theargon atom has an analogous 3s23p6 configuration. There is also an empty 3d level, but it is at considerably higher energy than 3s and 3p (unlike in the hydrogen atom), so that 3s23p6 is still considered a closed shell for chemical purposes. The atoms immediately before and after argon tend to attain this configuration in compounds. There are, however, somehypervalent molecules in which the 3d level may play a part in the bonding, although this is controversial (see below).
Forhelium there is no 1p level according to the quantum theory, so that 1s2 is a closed shell with no p electrons. The atoms before and after helium (H and Li) follow a duet rule and tend to have the same 1s2 configuration as helium.
In thetrigonal planar coordination geometry, onep orbital points out of the bonding plane, and can onlyoverlap with nearby atomic orbitals in aπ bond. If thatp orbital would be empty in an isolated atom, it may be filled through an intramoleculardative bond, as withaminoboranes. However, in some cases (e.g.boron trichloride and variousboranes,triphenylmethanium), no nearby filled orbital can profitably overlap with the emptyp orbital. In such cases, the orbital remains empty, and the compound obeys a "sextet rule". Likewise, linear compounds, such asdimethylzinc, have twop orbitals perpendicular to the bonding axis, and may obey a "quartet rule".[15] In either case, the empty unshielded orbitals tend to attract adducts.
Radicals satisfy the octet rule in onespin orientation, with four spin-up electrons in the valence shell, and almost satisfy it in the opposite spin orientation. Thus, for example, themethyl radical (CH3), which has an unpaired electron in anon-bonding orbital on the carbon atom and no electron of opposite spin in the same orbital. Another example is the radicalchlorine monoxide (ClO•) which is involved inozone depletion.
Stable radicals tend to adopt states in which the unpaired electron candelocalize through resonance. In such cases, the octet rule can be restored through the formalism of a1- or 3-electron bond.
Species such ascarbenes can be interpreted two different ways, depending on their spin state.Triplet carbenes are best thought of as two radicals localized on the same atom, and obey the octet rule in those radicals' shared spin-up orientation.Singlet carbenes tend to adopt a planar configuration, and are best thought of as obeying the planar sextet rule.
Main-group elements in the third and later rows of the periodic table can form hypercoordinate orhypervalent molecules in which the central main-group atom is bonded to more than four other atoms, such asphosphorus pentafluoride, PF5, andsulfur hexafluoride, SF6. For example, in PF5, if it is supposed that there are five truecovalent bonds in which five distinct electron pairs are shared, then the phosphorus would be surrounded by 10 valence electrons in violation of the octet rule. In the early days of quantum mechanics,Pauling proposed that third-row atoms can form five bonds by using one s, three p and one d orbitals, or six bonds by using one s, three p and two d orbitals.[16] To form five bonds, the one s, three p and one d orbitals combine to form five sp3dhybrid orbitals which each share an electron pair with a halogen atom, for a total of 10 shared electrons, two more than the octet rule predicts. Similarly to form six bonds, the six sp3d2 hybrid orbitals form six bonds with 12 shared electrons.[17] In this model the availability of empty d orbitals is used to explain the fact that third-row atoms such as phosphorus and sulfur can form more than four covalent bonds, whereas second-row atoms such as nitrogen and oxygen are strictly limited by the octet rule.[18]
5 resonance structures of phosphorus pentafluoride
However other models describe the bonding using only s and p orbitals in agreement with the octet rule. Avalence bond description of PF5 usesresonance between different PF4+ F− structures, so that each F is bonded by a covalent bond in four structures and an ionic bond in one structure. Each resonance structure has eight valence electrons on P.[19] Amolecular orbital theory description considers thehighest occupied molecular orbital to be a non-bonding orbital localized on the five fluorine atoms, in addition to four occupied bonding orbitals, so again there are only eight valence electrons on the phosphorus.[citation needed] The validity of the octet rule for hypervalent molecules is further supported byab initio molecular orbital calculations, which show that the contribution of d functions to the bonding orbitals is small.[20][21]
Nevertheless, for historical reasons, structures implying more than eight electrons around elements like P, S, Se, or I are still common in textbooks and research articles. In spite of the unimportance of d shell expansion in chemical bonding, this practice allows structures to be shown without using a large number of formal charges or using partial bonds and is recommended by the IUPAC as a convenient formalism in preference to depictions that better reflect the bonding. On the other hand, showing more than eight electrons around Be, B, C, N, O, or F (or more than two around H, He, or Li) is considered an error by most authorities. In particular, instead of pentavalent N, tetravalent N+ is written (e. g. not H−O−N(=O)=O but H−O−N+(=O)−O−).
The octet rule is only applicable tomain-group elements. Other elements follow otherelectron counting rules as theirvalence electron configurations are different from main-group elements. These other rules are shown below:
Theduet rule orduplet rule of the first shell applies to H, He and Li—the noble gashelium has two electrons in its outer shell, which is very stable. (Since there is no 1p subshell, 1s is followed immediately by 2s, and thus shell 1 can only have at most 2 valence electrons).Hydrogen only needs one additional electron to attain this stable configuration, whilelithium needs to lose one.
Fortransition metals, molecules tend to obey the18-electron rule which corresponds to the utilization of valenced,s andp orbitals to form bonding and non-bonding orbitals. However, unlike the octet rule for main-group elements, transition metals do not strictly obey the 18-electron rule and the valence electron count can vary between 12 and 18.[22][23][24][25]
^Housecroft, Catherine E.; Sharpe, Alan G. (2005).Inorganic Chemistry (2nd ed.). Pearson Education Limited. p. 883.ISBN0130-39913-2. Per source, the enthalpy change for
Cl +e− → Cl-
is -349 kJ/mol. Unit conversion performed usingWolfram|Alpha database, 13 April 2025.
^Housecroft & Sharpe 2005, p. 880. Source gives ionization energy of +495.8 kJ/mol. Unit conversion performed usingWolfram|Alpha database, 13 April 2025.
^Housecroft & Sharpe 2005, p. 880. Source gives ionization energy of +4562 kJ/mol. Unit conversion performed usingWolfram|Alpha database, 13 April 2025.
English translation:Werner, Alfred; Kauffman, G.B., eds. (1968).Classics in Coordination Chemistry, Part I: The selected papers of Alfred Werner. New York City, New York, USA: Dover Publications. pp. 5–88.
^Albright, T. A.; Burdett, Jeremy K.; Whangbo, Myung-Hwan (1985).Orbital Interactions in Chemistry. Wiley. pp. 298–299.ISBN0-471-87393-4.LCCN84-15310.
^L. PaulingThe Nature of the Chemical Bond (3rd ed., Oxford University Press 1960) p.63. In this source Pauling considers as examples PCl5 and thePF6− ion.ISBN0-8014-0333-2
^R.H. Petrucci, W.S. Harwood and F.G. Herring, General Chemistry (8th ed., Prentice-Hall 2002) p.408 and p.445ISBN0-13-014329-4
^Douglas B.E., McDaniel D.H. and Alexander J.J.Concepts and Models of Inorganic Chemistry (2nd ed., John Wiley 1983) pp.45-47ISBN0-471-21984-3
^Magnusson, E., J.Am.Chem.Soc. (1990), v.112, p.7940-51Hypercoordinate Molecules of Second-Row Elements: d Functions or d Orbitals?
^Frenking, Gernot; Shaik, Sason, eds. (May 2014). "Chapter 7: Chemical bonding in Transition Metal Compounds".The Chemical Bond: Chemical Bonding Across the Periodic Table. Wiley -VCH.ISBN978-3-527-33315-8.
^Bayse, Craig; Hall, Michael (1999). "Prediction of the Geometries of Simple Transition Metal Polyhydride Complexes by Symmetry Analysis".J. Am. Chem. Soc.121 (6):1348–1358.doi:10.1021/ja981965+.
^King, R.B. (2000). "Structure and bonding in homoleptic transition metal hydride anions".Coordination Chemistry Reviews.200–202:813–829.doi:10.1016/S0010-8545(00)00263-0.