Covalent bonds form primarily between nonmetals, the elements clustered in the upper-right portion of the periodic table that tend to share electrons rather than surrender or seize them outright. Carbon, oxygen, nitrogen, hydrogen, sulfur, and phosphorus are the most prolific covalent bonders, and they account for the vast majority of molecules you encounter in daily life, from water to DNA to the plastic in your phone case. But the real answer is broader and stranger than “nonmetals only,” because metals, metalloids, and even supposedly inert noble gases can all form covalent bonds under the right conditions.
The Six Core Nonmetals That Drive Most Covalent Chemistry
The elements carbon, hydrogen, nitrogen, oxygen, phosphorus, and sulfur, sometimes grouped under the shorthand CHNOPS, are the workhorses of covalent bonding. These elements sit near the top of the periodic table and need to gain only one to four electrons to fill their outer electron shells. As the biochemist George Wald pointed out decades ago, this property is what allowed them to form stable bonds, including multiple bonds like double and triple bonds, and ultimately to become the molecular backbone of life on Earth.1BioSystems. Short non-coded peptides interacting with cofactors facilitated the integration of early chemical networks The fact that these elements can share electrons in so many different arrangements explains why organic chemistry alone contains millions of known compounds while most of the periodic table contributes far fewer.
The idea that a covalent bond is fundamentally a shared pair of electrons dates to 1916, when Gilbert Newton Lewis proposed the model that still underpins how chemists think about bonding. Lewis recognized that two atoms could each contribute one electron to a shared pair, and that shared pair would count toward filling the outer shell of both atoms simultaneously.2Journal of the Chemical Society. The modern theory of valency That basic picture has been refined enormously by quantum mechanics, but the core insight holds: covalent bonding is about sharing.
Why Sharing Instead of Transferring
Whether two atoms share electrons (covalent bond) or one hands them over entirely (ionic bond) depends largely on how strongly each atom attracts electrons. Chemists call this property electronegativity, and it varies across the periodic table. Fluorine, sitting at the top right, has the highest electronegativity of any element. Cesium and francium, down at the bottom left, have the lowest.
When two atoms with similar electronegativities come together, neither can yank electrons away from the other, so they share. Two chlorine atoms bonding to form Clâ‚‚ is a clean example: identical atoms, identical pull, perfectly equal sharing. When two atoms with very different electronegativities meet, the electron transfer is lopsided enough to count as ionic. Sodium and chlorine forming table salt is the textbook case: sodium essentially gives up an electron, and chlorine takes it.
Most real bonds fall somewhere in between. The practical consequence is that covalent bonding is not an all-or-nothing category. It is the dominant mode whenever the bonding partners are both nonmetals with roughly similar electron-attracting strength, and it fades gradually into ionic character as the electronegativity gap widens.
Polar Covalent Bonds and Unequal Sharing
When two different nonmetals form a covalent bond, the sharing is almost never perfectly equal. In a water molecule, for instance, oxygen pulls the shared electrons closer than hydrogen does, giving the oxygen end of each bond a slight negative charge and the hydrogen end a slight positive charge. This unequal sharing is called a polar covalent bond, and it is responsible for many of water’s unusual properties, from its high boiling point to its ability to dissolve salts.
Quantum mechanical studies of polar bonds reveal that the distribution of electron density in the bonding region is shaped by competing forces. Polarization effects tend to push electron density toward the less polarizable atom, while quantum interference effects push it back toward the space between the two nuclei, ultimately producing the observed charge distribution and the molecule’s dipole moment.3ACS Publications (J. Chem. Theory Comput.). Description of Polar Chemical Bonds from the Quantum Mechanical Interference Perspective For practical purposes, the bigger the electronegativity difference between two nonmetals, the more polar their covalent bond will be.
A common rule of thumb puts the dividing line between polar covalent and ionic at an electronegativity difference of about 1.7 to 2.0, but that cutoff is genuinely fuzzy. Some bonds that “should” be ionic by that rule behave with significant covalent character, and vice versa. The spectrum between pure covalent and pure ionic is continuous, not a switch.
Metalloids and the In-Between Elements
Metalloids, the elements straddling the zigzag line between metals and nonmetals on the periodic table, are enthusiastic covalent bonders in their own right. Silicon, the second most abundant element in Earth’s crust, forms a network of covalent bonds in quartz (silicon dioxide) and in pure silicon crystals used for semiconductor chips. Germanium does something similar. These elements share electrons with neighbors in extended three-dimensional networks, producing solids that are extremely hard and have high melting points.
Boron is a particularly fascinating case. It has only three electrons available for bonding but four bonding positions, which makes it perpetually electron-deficient. To cope, boron forms unusual bonding arrangements in which three atoms share just two electrons rather than the standard two-atom, two-electron bond. Experimental and theoretical studies of boron clusters have shown that this electron deficiency leads to planar or quasi-planar structures, and at larger sizes boron can even form atomically thin sheets analogous to carbon’s graphene.4Nature Reviews Chemistry. From planar boron clusters to borophenes and metalloborophenes
Analysis of these electron-deficient systems shows that the unpaired electron density is not confined to the nuclear positions as it is in conventional bonds. Instead, it spreads out into the bonding regions to compensate for the shortage of electrons, producing bonding patterns that look nothing like the neat Lewis dot structures you see in introductory courses.5PubMed. Topology of the Effectively Paired and Unpaired Electron Densities for Complex Bonding Patterns: The Three-Center Two-Electron Bonding Case Boron’s weirdness is a reminder that the rules governing covalent bonding are more flexible than any simple classification scheme can capture.
When Metals Form Covalent Bonds
The traditional split says metals form ionic or metallic bonds, while nonmetals form covalent bonds. That is a useful first approximation and a misleading second one. Metals form covalent bonds all the time, particularly with carbon. The entire field of organometallic chemistry is built around direct metal-to-carbon bonds, and these compounds are central to modern catalysis, pharmaceutical manufacturing, and materials science.
What is interesting is that within a single metal-containing molecule, some bonds can be covalent while others are primarily electrostatic. A quantum chemical study of a zinc complex (a zinc-salphen compound) illustrated this clearly: the bonds between nitrogen and carbon within the organic framework of the molecule were covalent in character, while the bonds coordinating the zinc ion to nitrogen and oxygen atoms were predominantly electrostatic.6Computational and Theoretical Chemistry. Quantum chemical investigation on the structural and electronic properties of zinc–salphen complex: DFT and QTAIM analysis The same molecule, in other words, hosts both bond types simultaneously. This is not exceptional; it is the norm in transition metal chemistry.
Transition metals such as iron, copper, platinum, and palladium routinely form bonds with carbon, nitrogen, and oxygen atoms that have substantial covalent character. The catalytic converters in cars rely on platinum-group metals forming and breaking covalent bonds with exhaust gases. Hemoglobin carries oxygen through iron-oxygen bonds that are partly covalent. So while it is fair to say that nonmetals are the most natural covalent bonders, metals are active participants whenever the conditions favor electron sharing over electron transfer.
Noble Gases Can Form Covalent Bonds Too
For decades, noble gases were considered completely inert, unable to bond with anything. Their outer electron shells are already full, so they have no obvious reason to share. That picture was shattered in 1962 when Neil Bartlett synthesized the first xenon compound, and the field has expanded considerably since.
Xenon difluoride (XeFâ‚‚) is the simplest and best-studied noble gas compound. It is a stable, crystalline solid at room temperature, and its bonds are genuinely strong despite the fact that xenon, with a full outer shell, “shouldn’t” be bonding at all. Theoretical work has shown that these bonds rely on a mechanism called charge-shift bonding, in which the stabilization energy comes from the fluctuation of electron pairs between different bonding structures rather than from simple sharing or transfer. This allows xenon compounds to be very strongly bonded even though they violate the octet rule that governs most covalent chemistry.7Nature Chemistry. The essential role of charge-shift bonding in hypervalent prototype XeF2
Under extreme pressure, the bonding possibilities for noble gases get even stranger. Researchers have demonstrated that squeezing XeFâ‚‚ above roughly 84 gigapascals (about 840,000 times atmospheric pressure) after laser heating causes it to decompose and form a new compound, Xeâ‚‚F, which contains direct xenon-to-xenon covalent bonds. This was the first experimental confirmation that two noble gas atoms of the same element can bond covalently to each other.8PubMed. Structural Phase Transition and Decomposition of XeF(2) under High Pressure and Its Formation of Xe-Xe Covalent Bonds At everyday conditions you will never encounter a xenon-xenon bond, but at the pressures found deep inside giant planets, this kind of chemistry may be routine.
How the Heaviest Elements Change the Rules
For elements near the bottom of the periodic table, particularly those with atomic numbers above about 70, Einstein’s theory of relativity starts to affect how electrons behave. The innermost electrons in these heavy atoms move at a significant fraction of the speed of light, which causes them to contract closer to the nucleus and gain mass. This contraction ripples outward, changing the size and energy of the outer orbitals that participate in bonding.
The effect grows roughly in proportion to the square of the atomic number, so it is negligible for light elements like carbon and oxygen but increasingly important for elements like gold, mercury, lead, and beyond.9Coordination Chemistry Reviews. Relativistic effects on the chemical bonding properties of the heavier elements and their compounds In practical terms, this relativistic contraction makes some covalent bonds in heavy-element compounds shorter and stronger than you would expect from periodic trends alone. Gold’s distinctive yellow color, mercury’s status as a liquid at room temperature, and the voltage of a lead-acid car battery all trace partly to relativistic effects altering the covalent and metallic bonds these elements form.
For chemists working with the very heaviest synthetic elements at the bottom of the periodic table, relativistic effects are not a correction tacked on at the end of a calculation. They are the dominant influence on what kinds of bonds these elements can form and how stable those bonds are. Predicting the chemistry of elements like oganesson (element 118, a noble gas on paper) requires fully relativistic quantum mechanical models, because non-relativistic predictions give qualitatively wrong answers about whether and how it bonds.
Common Misconceptions About Which Elements Bond Covalently
The most persistent misconception is that the periodic table draws a hard line: nonmetals form covalent bonds, metals form ionic or metallic bonds, and that is the end of the story. As the examples above show, the actual landscape is more fluid. A second common error is thinking that covalent bonds are always weaker than ionic bonds or always stronger. Bond strength depends on the specific atoms involved and the context, not on the bond type. A carbon-carbon triple bond is one of the strongest bonds in all of chemistry, while some single covalent bonds between large atoms are quite weak.
Another misconception is that the octet rule (the idea that atoms “want” eight electrons in their outer shell) is a law. It is a guideline that works well for period-two elements like carbon, nitrogen, and oxygen but breaks down regularly for elements in the third period and below. Sulfur, phosphorus, and xenon all form stable compounds with more than eight electrons around them. The octet rule is useful for predicting the structures of simple molecules, but treating it as inviolable leads to confusion about perfectly real compounds like SF₆ (sulfur hexafluoride, used as an electrical insulator) or the noble gas compounds discussed earlier.
A subtler misunderstanding involves hydrogen bonding. Despite the name, a hydrogen bond is not a covalent bond. It is a much weaker attractive force between a hydrogen atom already covalently bonded to an electronegative atom (like oxygen or nitrogen) and a nearby electronegative atom on another molecule. Hydrogen bonds are crucial for the structure of water, proteins, and DNA, but they operate at roughly a tenth the strength of a typical covalent bond. The naming is genuinely confusing, and it trips up a lot of people.
Covalent Bonds in Everyday Materials
Understanding which elements form covalent bonds helps make sense of why different materials behave the way they do. Diamond is extraordinarily hard because it is a three-dimensional network of carbon-carbon covalent bonds with no weak points. Plastics are flexible and lightweight because they are long chains of carbon atoms covalently bonded to hydrogen and sometimes oxygen, fluorine, or chlorine, and those chains can slide past each other. Glass is a disordered network of silicon-oxygen covalent bonds, which is why it is rigid and transparent but shatters rather than deforming.
Pharmaceuticals are almost entirely covalent molecules built from carbon, hydrogen, oxygen, nitrogen, and occasionally sulfur, phosphorus, fluorine, or chlorine. When drug designers talk about “covalent inhibitors,” they mean drugs that form a new covalent bond directly with a target protein in the body, locking onto it permanently rather than binding reversibly. This strategy has produced some highly effective cancer drugs in recent years and is an active frontier in medicinal chemistry.
Even in the electronics industry, the distinction matters. Silicon’s covalent bonding network gives it semiconductor properties that make modern computing possible. Doping silicon with small amounts of elements like phosphorus or boron (which have different numbers of valence electrons) alters the covalent network just enough to control electrical conductivity with exquisite precision. Every transistor in every chip relies on this manipulation of covalent bonds in a semiconductor lattice.
The Periodic Table as a Rough Map
If you want a quick mental model, the periodic table still works as a rough guide. Elements in the upper-right region (excluding the noble gases under normal conditions) are the most reliable covalent bonders. The further left and further down you go, the more likely an element is to form ionic or metallic bonds instead. But the borders are genuinely blurry. Metals in the middle of the table form covalent bonds with nonmetals in organometallic compounds. Metalloids along the staircase form covalent network solids. Noble gases at the far right can be coaxed into covalent bonding with the most electronegative partners or under extreme pressure.
The takeaway that matters most is that covalent bonding is a behavior, not a permanent identity assigned to certain elements. Carbon always prefers to share electrons, but given the right partner or the right conditions, almost any element on the periodic table can participate in a covalent bond. The question is not really which elements form covalent bonds, but which combinations and circumstances favor sharing electrons over transferring them. The answer is: far more than you might expect.