Structures of Functional Groups

Why the Oxygen Atom Is Bent, Not Linear

The functional-group oxygen in alcohols, phenols and ethers is always $sp^3$ hybridised — the same hybridisation state as the oxygen in a water molecule. Because two of its four $sp^3$ orbitals hold lone pairs and only two form bonds, the group is bent rather than linear.

In an alcohol, the C–O bond is a sigma ($\sigma$) bond formed by the head-on overlap of an $sp^3$ hybrid orbital on carbon with an $sp^3$ hybrid orbital on oxygen. The O–H bond is a second $\sigma$ bond formed the same way, between oxygen and hydrogen. So around the oxygen atom in R–OH there are two bond pairs (to C and to H) and two lone pairs, all occupying roughly tetrahedral directions — exactly the water-like geometry.

Fig. 7.1 in the textbook lays this out for three representative molecules — methanol, phenol and methoxymethane — marking the measured C–O bond length and the C–O–H (or C–O–C) bond angle for each. The numbers below are read directly off that figure; the point of studying them is to see how the same $-\mathrm{O}-$ linkage stretches or bends slightly depending on what is attached to it.

The Bond Angle in Alcohols

For methanol, $\mathrm{CH_3OH}$, Fig. 7.1 records:

A perfect tetrahedral angle is $109^\circ 28'$ (i.e., about $109.5^\circ$). The alcohol's C–O–H angle is a shade smaller than this. The reason is the same one used to explain why the H–O–H angle in water is less than tetrahedral: oxygen's two lone pairs repel each other, and a lone pair–lone pair repulsion is stronger than a lone pair–bond pair or bond pair–bond pair repulsion. That extra push squeezes the bonded groups (C and H) slightly closer together, shrinking the angle just below the ideal tetrahedral value.

The Bond Angle and Bond Length in Phenols

In phenol, the $-\mathrm{OH}$ oxygen is not attached to an ordinary $sp^3$ carbon — it sits on a ring carbon that is itself $sp^2$ hybridised, since it is part of the aromatic system. Fig. 7.1 gives:

The angle is close to the alcohol value, but the C–O bond length is noticeably shorter than the 142 pm seen in methanol. Two things together are responsible for this shortening:

1. Partial double-bond character. One of oxygen's lone pairs can delocalise into the aromatic ring — the lone pair conjugates with the ring's $\pi$ system (this resonance donation is what drives phenol's characteristic chemistry, taken up later in Section 7.4.4). Because some electron density from that lone pair is pulled into $\pi$-type overlap with the ring, the C–O bond acquires a little double-bond character, and a bond with more double-bond character is shorter than a pure single bond.
2. The carbon is $sp^2$, not $sp^3$. An $sp^2$ orbital has more $s$-character than an $sp^3$ orbital, which pulls its electron density closer to the carbon nucleus. A bond built from a higher-$s$-character orbital is inherently shorter.

Both effects act in the same direction, so the phenolic C–O bond is measurably tighter than the alcohol's C–O bond even though the two O–H ends look outwardly similar.

The Bond Angle and Bond Length in Ethers

In an ether, R–O–R$'$, oxygen again carries four electron pairs — but now both of the bonded partners are carbon groups rather than one carbon and one hydrogen. Fig. 7.1 illustrates this with methoxymethane (dimethyl ether):

The four electron pairs — two bond pairs and two lone pairs — are still arranged in an approximately tetrahedral pattern around oxygen, but this time the C–O–C angle is slightly larger than the tetrahedral angle rather than smaller. The reason is steric, not electronic: the two $-\mathrm{R}$ groups on either side of the ether oxygen are bulkier than a single hydrogen atom, so their mutual repulsion pushes them a little further apart, widening the angle beyond $109.5^\circ$. …