Everything About Bonding in Water

Three atoms. Ten electrons. One of the smallest molecules in chemistry. Water looks as if it should be boring, and yet it floats when frozen, boils absurdly high for its size, dissolves salts that most liquids won’t touch, and soaks up heat like few other liquids do. Almost every one of those quirks can be traced back to how water is bonded, both inside the molecule and between molecules.

This post works through water’s bonding in layers, starting with the atoms and ending with an ocean.

Layer 1: two covalent bonds

An oxygen atom has six electrons in its outer shell, two short of a full octet. Each hydrogen atom has one electron and needs one more. The fix is sharing: oxygen forms one covalent bond with each hydrogen, and every atom ends up with a filled outer shell.

After bonding, oxygen’s outer shell holds four pairs of electrons:

• two bonding pairs, one in each O–H bond;
• two lone pairs, not shared with anything.

Each O–H bond is about 96 pm long and has a bond enthalpy of around 463 kJ/mol. That makes it a strong bond. Pulling water apart into its elements takes serious energy, which is why water is so stable and why you need electrolysis, not a hot stove, to make hydrogen and oxygen from it.

Layer 2: a bent shape

Four electron pairs around oxygen push each other as far apart as possible, which points them roughly towards the corners of a tetrahedron. Only two of those corners hold hydrogen atoms, so the molecule we “see” is bent, not straight.

The H–O–H angle is 104.5°, a little less than the perfect tetrahedral angle of 109.5°. Lone pairs sit closer to the oxygen nucleus and spread out more than bonding pairs, so they squeeze the two O–H bonds together. The full reasoning is in the shape of the water molecule.

A useful picture: imagine oxygen as a small hub with four arms. Two arms hold hydrogens. The other two are invisible, yet they still take up room, like two bulky backpacks on a person’s shoulders that push their arms slightly inwards.

Layer 3: a polar molecule

Oxygen is far more electronegative than hydrogen. In each O–H bond, the shared pair spends more time near oxygen, so oxygen carries a partial negative charge (δ−) and each hydrogen a partial positive charge (δ+).

Because the molecule is bent, those two bond dipoles do not cancel. They add up to an overall dipole pointing from the hydrogen side towards the oxygen side. If water were linear, like carbon dioxide, the bond dipoles would cancel and water would be non-polar. Its shape is what turns polar bonds into a polar molecule. (For the general rule, see dipole moments.)

This polarity has an easy demonstration: hold a charged plastic comb near a thin stream of tap water and the stream bends towards it. The water molecules swing round so that their oppositely charged end faces the comb.

Layer 4: hydrogen bonds between molecules

Now zoom out from one molecule to many. The δ+ hydrogen of one water molecule is attracted to a lone pair on the oxygen of a neighbour. That attraction is a hydrogen bond, drawn as O–H···O.

Every water molecule is ideally set up for this:

• two hydrogens it can donate;
• two lone pairs it can accept with.

So each molecule can take part in up to four hydrogen bonds, one pointing roughly towards each corner of its tetrahedron. In liquid water the neighbouring oxygen atoms sit about 280 pm apart.

Each hydrogen bond in water is worth roughly 20 kJ/mol, far weaker than the covalent O–H bond (the detailed comparison is in hydrogen bonds vs covalent bonds). In the liquid they flicker on and off constantly, breaking and re-forming billions of times a second, so the network is always rearranging. Think of a busy dance floor where partners switch all the time, yet almost everyone is holding someone’s hand at any given moment.

Layer 5: ice, the open framework

When water freezes, the flickering stops. Each molecule locks into four hydrogen bonds with four neighbours arranged tetrahedrally. In ordinary ice this builds a hexagonal framework with a lot of empty space inside it.

That open structure is why ice is less dense than liquid water, about 0.92 g/cm³ compared with roughly 1.00 g/cm³ for the liquid. When ice melts, some hydrogen bonds break, the framework partly collapses, and molecules crowd into the gaps. The liquid takes up less room than the solid, which is the opposite of what happens for most substances. The story is told in full in why ice floats.

Layer 6: the 4 °C oddity

Warm liquid water from 0 °C and two effects compete:

1. Leftover ice-like, open clusters keep breaking down, which lets molecules pack more tightly (density goes up).
2. Molecules move faster and push each other apart, as in any liquid (density goes down).

At first the collapse of open structure wins. Above about 4 °C ordinary thermal expansion takes over. So liquid water is densest at roughly 4 °C.

This has real consequences. In a lake in winter, water at about 4 °C sinks to the bottom, colder water stays near the top, and ice forms as a floating lid. The lid insulates the water below, so lakes rarely freeze solid and fish survive the winter.

Layer 7: bulk properties that follow

With the bonding in place, water’s other oddities fall into line:

• High boiling point. To boil water you must pull molecules out of the hydrogen-bond network. Compare water (100 °C) with hydrogen sulfide, a heavier molecule that cannot hydrogen bond effectively and boils far below 0 °C. More in why water has a high boiling point.
• High specific heat capacity. About 4.18 J/(g·K). Much of the heat you add goes into disrupting hydrogen bonds rather than speeding molecules up, so water warms slowly. Try some numbers in the specific heat calculator.
• High surface tension. Molecules at the surface have neighbours only below and beside them, so the hydrogen bonds pull the surface taut. Pond skaters rely on this.
• Dissolving ions. Water’s δ− oxygen ends cluster around positive ions and its δ+ hydrogen ends around negative ions. These ion–dipole attractions help pay the energy cost of breaking up an ionic lattice, which is why table salt dissolves.
• Self-ionisation. A tiny fraction of water molecules transfer a proton to a neighbour, giving H₃O⁺ and OH⁻. This happens through the hydrogen-bond network and underlies the whole pH scale.

Common mistakes

1. Calling water linear. H₂O is bent at 104.5°. Drawing it as H–O–H in a straight line hides the lone pairs and makes its polarity impossible to explain.
2. Mixing up bond types. The O–H bonds inside a molecule are covalent. The attractions between molecules are hydrogen bonds. Boiling breaks the second kind only.
3. Saying ice is denser. Solid water is less dense than liquid water. That is exactly why ice floats.
4. Saying water is densest at 0 °C. The maximum density of liquid water is near 4 °C, not at its freezing point.
5. Forgetting the lone pairs. They are responsible for the bent shape, half of the hydrogen bonding and water’s ability to accept protons. Always draw them.
6. “Polar bonds mean a polar molecule.” Only if the shape lets the dipoles add up. CO₂ has polar bonds but is non-polar overall; water is polar because it is bent.

Key takeaways

• Each water molecule has two covalent O–H bonds (about 96 pm, about 463 kJ/mol) and two lone pairs on oxygen.
• Four electron pairs give a bent shape with an H–O–H angle of 104.5°, so the bond dipoles add up and water is polar.
• Each molecule can form up to four hydrogen bonds (about 20 kJ/mol each), with neighbouring oxygens about 280 pm apart.
• In ice, those hydrogen bonds lock into an open framework, so ice is less dense than liquid water; liquid water is densest near 4 °C.
• Water’s high boiling point, heat capacity, surface tension and dissolving power all come from this combination of polar covalent bonds and hydrogen bonding.