Intermolecular Forces and Why Water Is Weird

The pot that never breaks a bond#

Put a pot of water on the stove and boil it dry. Steam pours off, the water vanishes, and it feels like something has been torn apart. Something has — but not what most people think.

Look at a single water molecule. It is an oxygen atom with two hydrogens fastened to it by covalent bonds, and each of those bonds is worth roughly $460\ \mathrm{kJ\,mol^{-1}}$. To break even one O–H bond you would need a temperature of thousands of degrees, or a jolt of ultraviolet light. Your stove does nothing of the kind. Every molecule that leaves the pot as steam leaves intact: two hydrogens, one oxygen, still firmly bonded.

So what did boiling break? Only the weak attractions between whole molecules — the forces that had been holding the liquid together. Those cost about $20\ \mathrm{kJ\,mol^{-1}}$ to overcome for water, perhaps a twentieth of the bonds inside. And here is the surprise worth sitting with: it is these humble forces, not the mighty bonds, that decide whether a substance is a gas, a liquid, or a solid at room temperature. The bonds set what a molecule is. The forces between molecules set how it behaves.

Two completely different things called "bonds"#

The single most common confusion in this whole subject is collapsing two very different attractions into one word. Keep them apart and everything else falls into place.

• Intramolecular bonds hold the atoms within one molecule together — the O–H bonds in water, the C–C and C–H bonds in butane. These are true chemical bonds, born of shared or transferred electrons, and they are strong: typically $150$–$1000\ \mathrm{kJ\,mol^{-1}}$.
• Intermolecular forces are the far weaker attractions between separate molecules. Nothing new is shared; no new molecule forms. They range from under $1$ up to about $40\ \mathrm{kJ\,mol^{-1}}$ — one to two orders of magnitude weaker.

When ice melts or water boils, you are working only against the second kind. This is precisely why melting and boiling cost so much less energy than a chemical reaction: melting ice takes $6\ \mathrm{kJ\,mol^{-1}}$ and boiling water $41\ \mathrm{kJ\,mol^{-1}}$, while ripping the molecule apart into atoms would take over $900$. A phase change rearranges intact molecules; it does not dismantle them.

Write the energies on a single scale and the hierarchy is stark:

$$\underbrace{E_{\text{covalent}} \sim 100\text{–}1000}_{\text{within a molecule}} \;\gg\; \underbrace{E_{\text{H-bond}} \sim 10\text{–}40}_{\text{between molecules}} \;>\; \underbrace{E_{\text{dipole}} \sim 5\text{–}25}_{\text{between molecules}} \;>\; \underbrace{E_{\text{dispersion}} \sim 0.1\text{–}10}_{\text{between molecules}}$$

all in $\mathrm{kJ\,mol^{-1}}$. The three terms on the right are the intermolecular forces, and they are the whole subject of this article.

The three forces, weakest to strongest#

London dispersion forces are present in everything — every atom and molecule, polar or not, noble gas or hydrocarbon. Their origin is beautifully accidental. An atom's electrons are in constant motion, and at any instant they are momentarily lopsided, giving the atom a fleeting instantaneous dipole. That transient dipole tugs on a neighbour's electrons, inducing a matching dipole in it, and the two flicker in step, attracting. Average over time and the permanent dipole is zero — but the correlated flickering leaves a real, always-attractive force.

Two things make dispersion forces stronger. First, more electrons: bigger, heavier molecules have more charge to slosh around. Second, polarizability — how easily the electron cloud deforms. A large, diffuse cloud distorts readily and makes a bigger instantaneous dipole. This is why boiling point climbs down a column of the periodic table and up a series of alkanes: methane boils at $-161\ ^\circ\mathrm{C}$, butane at $-1\ ^\circ\mathrm{C}$, octane at $+126\ ^\circ\mathrm{C}$ — same kind of molecule, more atoms, stronger dispersion, higher boiling point. Nothing polar anywhere; just bigger clouds.

Dipole–dipole attractions act between polar molecules — those with a permanent separation of charge because their bond dipoles do not cancel. The positive end of one molecule attracts the negative end of the next. These add to dispersion rather than replacing it, which is why polar molecules generally boil higher than nonpolar molecules of similar size.

Hydrogen bonding is the strongest of the three, and it appears under one specific condition: a hydrogen atom bonded directly to nitrogen, oxygen, or fluorine. Those three are small and greedily electronegative, so they strip the shared electrons far toward themselves, leaving the hydrogen as an almost-bare, strongly positive proton. That exposed proton snuggles up to a lone pair on the N, O, or F of a neighbouring molecule. The result is a specially strong, quite directional dipole–dipole attraction — but note the label carefully: a hydrogen bond is a force between molecules, not a chemical bond that makes a new molecule. It is roughly a tenth the strength of the covalent O–H bond it lives beside.

Try this. Start with Nonpolar and drag the size slider from small to large: only the orange dispersion segment grows, and the boiling-point line on the thermometer climbs with it — big nonpolar molecules can be liquids and solids purely from dispersion. Now switch to Polar and a blue dipole segment stacks on top, lifting the boiling point further at the same size. Switch to H-bonding and the violet segment dwarfs the rest. Then press Heat past boiling on any setting and watch the crucial thing: as the temperature crosses the boiling line, the molecules break loose and fly apart — but every gold internal bond stays welded together. The dashed links between molecules snap; the bonds within them never do. That is boiling, made honest.

From force strength to boiling point#

Why does a stronger intermolecular force mean a higher boiling point? Because boiling is a competition between attraction and thermal energy. A molecule escapes the liquid only when it has enough kinetic energy to climb out of the potential-energy well its neighbours hold it in. Raise the temperature and you raise the average kinetic energy $\sim k_B T$; boiling begins when that thermal energy can overcome the intermolecular binding. A deeper well — a stronger force — needs a hotter liquid.

The quantitative link is the Clausius–Clapeyron relation, which ties the vapour pressure of a liquid to its enthalpy of vaporisation $\Delta H_{\text{vap}}$:

$$\ln P = -\frac{\Delta H_{\text{vap}}}{RT} + \text{const}$$

Boiling happens when the vapour pressure $P$ reaches the surrounding pressure. Rearranging for the boiling temperature at fixed external pressure shows that $T_b$ rises with $\Delta H_{\text{vap}}$ — and $\Delta H_{\text{vap}}$ is essentially the total intermolecular attraction you must pay off to turn the liquid into separated gas molecules. Stronger forces, larger $\Delta H_{\text{vap}}$, higher boiling point. The chain is unbroken: force strength sets $\Delta H_{\text{vap}}$, and $\Delta H_{\text{vap}}$ sets $T_b$.

Dispersion's steep dependence on distance is worth one more line. The attractive energy between two molecules from dispersion falls off as

$$U_{\text{disp}}(r) \propto -\frac{1}{r^6}$$

That sixth power is punishing: double the separation and the attraction drops by a factor of $2^6 = 64$. Dispersion is therefore a strictly short-range, touch-your-neighbour force, which is exactly why it matters so much in dense liquids and solids and hardly at all in dilute gases — and why molecular shape matters too, since long, flat molecules that can lie close along their whole length pull on each other far more than compact, branched ones of the same formula.

Water, the great exception#

Now put water under the same lens, and it starts breaking the rules.

By size alone, water should barely be a liquid. It is a tiny molecule, $18$ atomic mass units, lighter than methane's $16$ is close and lighter than the $34$ of hydrogen sulfide, H₂S — water's own chemical cousin, oxygen swapped for sulfur. Judging by dispersion, H₂S should boil higher than water. Instead H₂S boils at $-60\ ^\circ\mathrm{C}$ and water boils at $+100\ ^\circ\mathrm{C}$: a $160\ ^\circ\mathrm{C}$ gap in the wrong direction. If water obeyed the trend of its family it would boil somewhere near $-80\ ^\circ\mathrm{C}$ and there would be no liquid water — and no life — anywhere near room temperature.

The culprit is hydrogen bonding, and water is uniquely good at it. Each molecule has two O–H hydrogens to donate and two lone pairs on oxygen to accept, so a single water molecule can hydrogen-bond to four neighbours at once. That dense, four-way network is what boiling has to tear apart, and it is why water's boiling point, surface tension, and heat capacity are all anomalously high. Surface tension is the same story at the surface: a molecule there is pulled inward by hydrogen bonds with no neighbours above to balance the pull, so the surface behaves like a taut skin strong enough to float a paperclip or carry a water strider.

And then the big one. Freeze water and it expands.

Press Freeze and watch what the hydrogen bonds do. In the liquid they are transient — forming, breaking, reforming as molecules jostle, letting the molecules pack in close. As the water freezes, those same bonds lock into a rigid, fully satisfied network: an open hexagonal lattice in which every molecule is held at arm's length from its four partners, with conspicuous empty space at the centre of each hexagonal ring. The molecules end up farther apart in the solid than in the liquid. Watch the density readout fall from $1.000$ to $0.917\ \mathrm{g\,cm^{-3}}$, and the ice cube in the beaker settle with about $9\%$ of itself riding above the waterline. Then press Melt and the lattice collapses back into the denser liquid.

For almost every other substance the solid is denser than the liquid, because cooling lets molecules pack tighter. Water is the famous exception, and it is the directional, four-way hydrogen bond that forces the wasteful open cage. It is the same anomaly that makes water's solid–liquid boundary lean backwards on the phase diagram.

The consequence is planetary. Because ice floats, lakes and oceans freeze from the top down. The floating ice sheet insulates the water beneath, so fish and everything else survive the winter in liquid water below. Had ice been denser than water, as intuition expects, ice would sink as it formed, bodies of water would freeze solid from the bottom up, and aquatic life in cold climates would be far more precarious. A quirk of one intermolecular force writes a rule of biology.

Water's hydrogen bonding runs deeper still. It is why water is such a good solvent and why it stays neutral by passing protons between molecules — the same O–H groups and lone pairs that build the ice lattice also let one water molecule hand a proton to another. Nearly everything strange and life-giving about water traces back to those four little hydrogen bonds per molecule.

Where this shows up#

Once you see intermolecular forces, you see them everywhere.

• Geckos climb glass using nothing but dispersion forces. Millions of nanoscopic hairs on their toes make intimate contact with the surface, and the summed $1/r^6$ attraction of all that contact area holds up the whole animal.
• Boiling points and refrigerants are engineered by tuning intermolecular forces — choosing molecular size, polarity, and hydrogen-bonding capacity to hit a target boiling point.
• DNA is held in its double helix by hydrogen bonds between base pairs: weak enough to be unzipped for reading and copying, strong enough in aggregate to store the genome faithfully.
• Proteins fold into shape largely because nonpolar residues huddle away from water — the hydrophobic effect — which is intermolecular forces choreographing biology.
• Gecko tape, non-stick coatings, and why water beads on a waxed car are all the same physics: matching or mismatching the forces between different molecules.

The lesson that ties it together is the one the boiling pot taught. The strong bonds inside molecules decide the chemistry; the weak forces between them decide whether you are looking at steam, water, or ice.