How Many Covalent Bonds Does Oxygen Form?

Oxygen typically forms two covalent bonds. With six electrons in its outer shell, it needs to share two more to reach a stable arrangement, and that sharing usually takes the form of two bonds, whether as two single bonds (like in water) or one double bond (like in molecular oxygen, O₂). That said, the “always two” rule has some well-known and genuinely interesting exceptions that show up across chemistry, from carbon monoxide to the hydronium ion to oxygen squeezed under enormous pressure.

Why Two Is the Default

Oxygen sits in group 16 of the periodic table, meaning each oxygen atom has six valence electrons. A stable outer shell for oxygen holds eight. The arithmetic is straightforward: oxygen is two electrons short, so it tends to share two pairs of electrons with neighboring atoms, forming two covalent bonds. In water, that means one bond to each hydrogen. In formaldehyde, it means a double bond to carbon. In hydrogen peroxide, each oxygen forms one bond to the other oxygen and one to a hydrogen. The pattern repeats across thousands of molecules.

Oxygen also carries two lone pairs, electrons that are not involved in bonding under ordinary circumstances. Those lone pairs matter, though, because they influence the shape of every molecule oxygen sits in. Water is bent rather than straight precisely because the two lone pairs push the two O–H bonds closer together. And as we’ll see, those lone pairs can sometimes get recruited into additional bonds under the right conditions.

Single Bonds, Double Bonds, and What Counts

A question that trips people up is whether a double bond counts as one bond or two. In the way chemists tally covalent bonds, a double bond counts as two. When you see O₂ written with a double bond between the two oxygen atoms, each oxygen is forming two covalent bonds, both to the same partner. The same logic applies to the carbon-oxygen double bond in carbon dioxide: each oxygen forms two bonds to the central carbon, fulfilling its usual quota of two.

This means the answer to “how many bonds does oxygen form” doesn’t change whether oxygen is bonded to one atom or two. It still forms two covalent bonds in most stable molecules. What changes is how those bonds are distributed. Two single bonds spread across two partners (as in water), or two bonds concentrated into a double bond with one partner (as in O₂ or the carbonyl group in ketones and aldehydes). In either arrangement, oxygen shares two pairs of electrons total.

A triple bond, by extension, would mean three shared pairs. Oxygen doesn’t do this in the way nitrogen does in N₂, where three straightforward covalent bonds hold the molecule together. But oxygen can reach a bond count of three through a different route, which brings us to the exceptions.

When Oxygen Forms Three Bonds

The most familiar example of oxygen exceeding two bonds is the hydronium ion, H₃O⁺. When water picks up an extra hydrogen ion (a bare proton), the oxygen uses one of its lone pairs to form a third bond. This is called a coordinate bond or dative bond: instead of each atom contributing one electron to the shared pair, oxygen supplies both electrons. The result is still a covalent bond in every physical sense. Once formed, it’s indistinguishable from the other two O–H bonds in the molecule. The only difference is bookkeeping: how the bond got started.

The hydronium ion is not some exotic lab curiosity. It’s the species that actually makes acidic solutions acidic. Every time you dissolve an acid in water, you’re creating hydronium ions, which means trivalent oxygen is present in every glass of lemonade and every raindrop that picks up atmospheric CO₂.

Carbon monoxide is another case where oxygen reaches three bonds, and it’s a more structurally interesting one. Analysis of the bonding in CO shows that the molecule contains two covalent bonds and one dative bond, with the oxygen donating a lone pair to carbon to form the third bond.¹ That triple bond (two covalent plus one dative) gives CO an unusually high bond strength, which helps explain why the molecule is so stable and so difficult to break apart. It also explains a quirk of CO’s chemistry: despite having a triple bond like N₂, CO is far more reactive in certain contexts because the dative component makes the electron distribution less symmetric.

So when someone says oxygen “always” forms two bonds, it’s more accurate to say oxygen always forms two conventional covalent bonds but can form a third bond by donating a lone pair. The total number of bonds oxygen participates in can be three, and in certain charged species, it can reach four.

Oxygen With Four Bonds

Four-coordinate oxygen exists, though it’s less common in everyday chemistry. The simplest example in general chemistry courses is the oxygen atom in certain protonated species or in coordination with metal ions. When oxygen in water coordinates to a metal cation, it donates electron density from its lone pairs to form a bond with the metal. In a hydrated metal ion, the oxygen atoms of surrounding water molecules are each already bonded to two hydrogens and then also coordinate to the central metal, pushing their bond count to three and sometimes effectively four when the interaction is strong enough to be considered a genuine bond rather than an electrostatic attraction.

Research on these kinds of interactions shows that when water, hydrogen sulfide, or ammonia act as Lewis bases and donate electron density to cations, the donation reduces repulsion between the lone pairs on the donor atom.² In other words, forming additional bonds doesn’t just happen despite the lone pairs resisting it; the act of donating a lone pair into a new bond actually stabilizes the remaining electron pairs. That’s part of why oxygen-metal coordinate bonds form so readily in aqueous chemistry.

Another example of tetracoordinate oxygen shows up in certain organic cations. When an ether oxygen gets protonated (picks up a hydrogen ion), or when an oxygen atom bridges two other groups and also coordinates to a third, oxygen can end up bonded to three or four atoms simultaneously. These aren’t exotic hypotheticals; they appear as reaction intermediates in organic chemistry all the time.

Why Oxygen Rarely Goes Beyond Two in Neutral Organic Molecules

Given that oxygen can form three or four bonds, you might wonder why it sticks to two so stubbornly in most organic molecules. The answer comes down to charge and energy. Every time oxygen forms a bond using one of its lone pairs (a coordinate bond), it effectively gives away electron density without getting any back. That leaves oxygen with a formal positive charge. Oxygen is the second most electronegative element commonly found in organic molecules (after fluorine), and it strongly resists carrying a positive charge. Forming a third or fourth bond costs energy that, in most neutral molecules, isn’t compensated by anything else.

In charged species, the math changes. The hydronium ion is positively charged overall, and that charge is spread across the whole ion, making the three-bond arrangement stable in solution. In metal complexes, the strong electrostatic pull of the metal cation provides enough driving force to coax a lone pair into a bond. But in a neutral organic molecule sitting peacefully at room temperature, oxygen has no incentive to give up a lone pair, so it forms two bonds and stops.

This is also why oxygen doesn’t form the kind of expanded bonding that elements like sulfur and phosphorus can. Sulfur, sitting one row below oxygen in the periodic table, can form four or even six bonds in molecules like sulfuric acid or SF₆. Sulfur is larger, its lone pairs are held less tightly, and it has accessible orbitals that oxygen lacks. Oxygen is too small and too electronegative to comfortably accommodate more than two bonds in most neutral environments.

Bond Order Isn’t Always a Whole Number

One complication that comes up in more advanced discussions is that the “number of bonds” oxygen forms isn’t always a clean integer. In molecules or ions with resonance, where multiple valid structures can be drawn, the actual bond between oxygen and another atom can have a bond order somewhere between one and two, or between two and three.

Take the carboxylate group found in every amino acid and in common substances like vinegar (acetic acid, when it loses a proton). In the carboxylate ion, two oxygen atoms are bonded to the same carbon. You can draw one structure where the left oxygen has a double bond to carbon and the right oxygen has a single bond, and another where the assignments are flipped. In reality, neither structure is correct on its own. Both C–O bonds are identical, with a bond order of about 1.5 each. The electrons are smeared out over both bonds equally.

Ozone is another good example. The central oxygen in O₃ is bonded to two outer oxygens, and each of those bonds has a bond order of roughly 1.5. The molecule doesn’t flicker between a single bond on one side and a double bond on the other; the electrons are distributed evenly. So asking “how many bonds does this oxygen form” gets an answer like “one and a half on each side, so three total,” which is accurate but clearly not what a beginner’s textbook wants to say.

These fractional bond orders are real and measurable. The bond lengths in carboxylate ions and ozone fall neatly between the lengths of known single and double bonds, confirming that the electrons genuinely are delocalized rather than snapping back and forth between two arrangements.

Hydrogen Bonds and the Blurry Edge of Covalent

Oxygen is one of the key players in hydrogen bonding, and a question that keeps resurfacing in research is how much covalent character hydrogen bonds actually have. Traditional teaching treats hydrogen bonds as purely electrostatic, a weak attraction between a hydrogen bonded to an electronegative atom and a lone pair on a neighboring electronegative atom. Under this view, hydrogen bonds are not covalent bonds at all, and they wouldn’t count toward oxygen’s bond tally.

But the picture is more nuanced than that, especially for strong hydrogen bonds. Recent computational work on short hydrogen bonds in crystalline systems has found that in some cases, the electronic structure shows a redistribution of bonding electrons that blurs the line between hydrogen bonding and genuine covalency.³ In these strong hydrogen bonds, the shared proton sits in a broad potential well between two oxygen atoms, and the electron density around it starts to look less like two separate interactions and more like a single three-center bond.

This doesn’t mean every hydrogen bond is secretly covalent. The vast majority of hydrogen bonds, like those holding water molecules together in a glass, are weak and predominantly electrostatic. But at the strong end of the hydrogen-bond spectrum, particularly in enzyme active sites, in certain crystals, and in proton sponges, the covalent contribution becomes significant enough that counting oxygen’s bonds as strictly two starts to feel like an oversimplification. The boundary between “a real covalent bond” and “just an interaction” is not as sharp as introductory chemistry implies.

Oxygen Under Extreme Pressure

Everything discussed so far applies to oxygen under the conditions you’d encounter in daily life or in a chemistry lab. But what happens when you subject pure oxygen to enormous pressures, the kind found deep inside planets?

At everyday pressures and temperatures, oxygen exists as O₂, each molecule held together by a double bond. Cool it down and compress it, and it forms a series of solid phases. The most intriguing is the epsilon phase of solid oxygen, which forms above about 8 gigapascals of pressure. X-ray diffraction studies of this phase revealed that four O₂ molecules associate into a larger O₈ unit, held together by what are likely weak chemical bonds between the molecules.⁴ The nature of those inter-molecular bonds isn’t fully settled, but they appear to be somewhere between van der Waals attractions and proper covalent bonds, another example of the bonding continuum rather than a strict binary.

At even higher pressures, beyond roughly 96 gigapascals, oxygen becomes metallic and eventually superconducting. In these extreme regimes, the familiar rules about two covalent bonds per oxygen atom no longer apply in any conventional sense. The atoms are packed so tightly that their electron clouds merge into collective behavior. It’s a fascinating reminder that the “rules” of covalent bonding are really just descriptions of what oxygen prefers to do under the relatively gentle conditions found on Earth’s surface.

Common Misconceptions Worth Clearing Up

A few misunderstandings come up repeatedly when people learn about oxygen’s bonding.

Oxygen always forms exactly two bonds: As we’ve seen, two is the default in neutral organic molecules, but three-bond and four-bond oxygen is common in charged species, coordination compounds, and even everyday ions like hydronium.
Double bonds are one bond: A double bond involves two shared electron pairs and counts as two covalent bonds. Oxygen in O₂ is forming two bonds, not one. The “double” refers to the bond multiplicity, not some abstract doubling of a single bond’s strength.
Coordinate bonds are weaker than regular covalent bonds: Once a coordinate bond is formed, it is a covalent bond. Its strength depends on the atoms involved, not on the accounting of who contributed the electrons. The third bond in carbon monoxide is just as real and just as strong as the other two.
Oxygen can expand its octet like sulfur: Oxygen cannot form five or six bonds the way sulfur or phosphorus can. Older textbooks sometimes explained sulfur’s expanded bonding by invoking empty “d orbitals,” but more modern analysis attributes it to sulfur’s larger size and lower electronegativity. Either way, oxygen’s small atomic radius and high electronegativity make it a strict two-bond (occasionally three or four-bond) atom, never a five or six-bond one.

How Other Chalcogens Compare

Oxygen belongs to the chalcogen family (group 16), and comparing it to its heavier relatives helps put its bonding behavior in perspective. Sulfur, selenium, and tellurium all have six valence electrons, just like oxygen. In their simplest compounds, they form two bonds: H₂S, H₂Se, H₂Te all mimic the two-bond pattern of water.

But sulfur routinely exceeds two bonds in stable, neutral molecules. Sulfur dioxide has sulfur bonded to two oxygens with multiple bonds; sulfuric acid has sulfur bonded to four oxygens; and sulfur hexafluoride has sulfur bonded to six fluorines. Selenium and tellurium can similarly exceed two bonds, though they do so less commonly than sulfur.

Oxygen’s refusal to follow suit isn’t about electron count. It’s about size. Oxygen atoms are small enough that fitting more than four bonding partners around a single oxygen center would create steric strain, atoms physically bumping into each other. Combine that with oxygen’s high electronegativity, which makes it reluctant to share its lone pairs, and you get an element that strongly prefers two bonds even though its neighbors in the periodic table are much more flexible. The two-bond preference isn’t a fundamental law of oxygen; it’s a consequence of being small, electronegative, and without low-energy empty orbitals to accommodate extra bonds.