The Nature of the Chemical Bond
What are atoms actually doing? From X-ray diffraction and crystal structures to sulfur chemistry and the limits of the octet rule, this is the deeper story behind the lines we draw between atoms.
At A-level, chemical bonding can sometimes look deceptively tidy. Ionic compounds are described as oppositely charged ions held together by electrostatic attraction. Covalent compounds contain shared pairs of electrons. Metals are positive ion cores surrounded by delocalised electrons.
These models are useful. They allow us to predict structures, physical properties and reactivity. But there is an important question hiding underneath all of them:
Atoms cannot be watched with an ordinary light microscope, and a bond cannot be isolated as a tiny physical rod joining one atom to another. Instead, chemists developed bonding models by combining experimental evidence with increasingly sophisticated descriptions of electrons.
One of the most important pieces of that story came from something that initially appeared to have little to do with chemistry at all: X-rays.
Seeing Structures Too Small to See
Visible light lets us see everyday objects because it interacts with them before reaching our eyes. But there is a limit to the size of features that can be resolved using a particular wavelength.
Typical distances between neighbouring atoms in solids are of the order of 10−10 m. Visible light has wavelengths of roughly 4 × 10−7 to 7 × 10−7 m, thousands of times larger.
X-rays have much shorter wavelengths, comparable to interatomic spacings in crystals. That makes them extraordinarily useful for investigating atomic structure.
A Crystal Leaves a Fingerprint
When a narrow beam of X-rays is directed at a crystal, the radiation is scattered by the electron density in the regularly arranged structure.
The scattered waves interfere. Some reinforce one another through constructive interference, while others cancel through destructive interference.
The result is a characteristic diffraction pattern. It is not a photograph of the atoms. Instead, the positions and intensities of the diffraction features contain information from which the internal structure can be calculated.
The Braggs and the Architecture of Crystals
In the early twentieth century, William Henry Bragg and his son William Lawrence Bragg developed methods that related X-ray diffraction to the spacing of repeating layers within crystals.
Their work formed the foundation of X-ray crystallography. In 1915, they were awarded the Nobel Prize in Physics for their analysis of crystal structures using X-rays.
For chemistry, the consequences were enormous. Scientists were no longer limited to inferring internal structures only from bulk properties. They could determine the three-dimensional arrangement of atoms from experimental data.
This is Bragg's law: a relationship connecting the X-ray wavelength, the spacing between repeating planes and the angle at which constructive interference occurs.
From Diffraction Pattern to Unit Cell
A crystal can contain an enormous number of particles, so chemists do not describe every atom individually.
Instead, we identify a repeating structural unit called the unit cell. Repeating the unit cell through three dimensions generates the crystal lattice.
It is rather like describing a tiled floor from a single repeating tile pattern, except that a crystal repeats in three dimensions.
Metals Are Not Just “Positive Ions in a Sea of Electrons”
The familiar A-level description of metallic bonding is useful because it explains the electronic interaction: positive metal ion cores are attracted to delocalised electrons.
But that description does not tell us exactly how the atoms are arranged in space. Real metals possess ordered crystal structures. Two important arrangements are face-centred cubic (FCC) and body-centred cubic (BCC).
The Octet Rule
Students are often introduced to covalent bonding through the octet rule: atoms frequently form bonds in arrangements that give them eight electrons in their valence shell.
For many compounds involving second-period elements such as carbon, nitrogen, oxygen and fluorine, this is an excellent starting model.
Carbon in CH4, nitrogen in NH3 and oxygen in H2O can all be represented with eight electrons around the central atom.
Sulfur Starts Causing Problems
Sulfur sits below oxygen in Group 16 and has the outer electronic configuration 3s2 3p4.
Species such as SO2, SO3, SO32−, SO42− and S2O32− expose the limitations of trying to interpret every bond with a single fixed Lewis diagram.
Older textbook explanations often described sulfur as simply expanding its octet by using empty 3d orbitals. Modern descriptions show that this picture is too simplistic.
Does Sulfur Really “Use Its d Orbitals”?
You may still encounter diagrams in which sulfur promotes electrons into 3d orbitals before forming bonds. Historically this offered a convenient way to rationalise structures that seemed to contain more than eight electrons around sulfur.
However, modern quantum-chemical descriptions do not treat compounds such as sulfate as being adequately explained by sulfur simply filling otherwise empty 3d orbitals. The electron distribution is more subtle, and concepts such as formal charge, resonance, delocalisation and molecular orbitals provide a better description.
This is an important distinction for a modern A-level student: Lewis structures are electron-accounting models. They are not literal photographs showing exactly where every electron is located.
SO2: One Molecule, More Than One Drawing
Sulfur dioxide is a good example of why a single Lewis structure can be misleading. Different valid electron-accounting structures can be drawn, but experiments show that the two S–O bonds in the molecule are equivalent.
The real electron distribution is therefore better described using resonance. The molecule does not jump back and forth between resonance structures: the structures are alternative representations of the same delocalised electronic state.
Around sulfur there are three electron-density regions in the VSEPR model: two bonding regions and one lone pair. The electron-domain geometry is approximately trigonal planar, while the molecular shape is bent (V-shaped).
SO3, Sulfite and Sulfate
SO3 has three bonding regions around sulfur and is trigonal planar, with bond angles close to 120°.
SO32− has three bonding regions and one lone pair on sulfur in the standard VSEPR model, producing a trigonal pyramidal molecular shape.
SO42− has four bonding regions around the central sulfur and is tetrahedral, with O–S–O angles close to 109.5°.
Sulfur Can Bond to Itself
Sulfur can form S–S bonds, a behaviour known as catenation. Under ordinary conditions, elemental sulfur commonly contains S8 molecules.
The eight sulfur atoms form a ring, but not a flat octagon. The molecule adopts a puckered, crown-shaped three-dimensional structure.
Folding the ring out of the plane permits more favourable bond angles and reduces strain and unfavourable electron-pair interactions.
Thiosulfate: Familiar Chemistry, Unusual Structure
The thiosulfate ion has the formula S2O32−. One useful way to recognise its composition is to compare it with sulfate, SO42−: in thiosulfate, one oxygen position is replaced by sulfur.
The central sulfur has four bonding regions and an approximately tetrahedral arrangement. The terminal sulfur is chemically distinct from the oxygen atoms.
Students often meet thiosulfate during iodine-thiosulfate titrations:
The same ion used in quantitative analysis therefore also provides an excellent example of sulfur's structural versatility.
A Chemical Bond Is Not a Stick
Ball-and-stick models are extremely useful, but the sticks can create a misleading mental picture. There are no microscopic rods joining atoms together.
A chemical bond is fundamentally an electronic interaction. Stable bonded arrangements arise because the combined distribution of nuclei and electrons gives an energetically favourable system relative to the relevant separated arrangement.
These models do not have to be competitors. They operate at different levels of sophistication and answer different questions.
Models Have Limits
One of the hardest transitions in chemistry is learning that a model can be useful without being perfectly literal.
At GCSE, a covalent bond may be represented by a line. At A-level, that line represents a shared pair of electrons. At a deeper level, the same bond can be discussed using electron density and molecular orbitals.
That does not make the earlier model useless. The key question is always: what is this model being used to explain?
Lewis structures may be excellent for connectivity and electron counting. VSEPR may be excellent for predicting shape. Neither is a complete quantum-mechanical picture of a molecule.
From Crystals to Quantum Chemistry
The history of chemical bonding follows a recurring pattern.
Chemists first observed the properties of matter. Techniques such as X-ray diffraction then provided increasingly direct evidence about atomic arrangements. Those observations demanded better theories, and better theories produced more sophisticated models of electrons and bonding.
The chemical bond was never something chemists could simply see. It had to be inferred, measured, modelled and repeatedly refined.
That process continues today.
So what is a chemical bond?
A chemical bond is a stabilising interaction between atoms arising from the distribution and interaction of electrons and nuclei. Different bonding models describe different aspects of that interaction.
The simple diagrams taught at A-level are not wrong. They are starting points. A line can represent a shared pair of electrons. A lattice diagram can represent an enormous repeating structure. The octet rule can reveal useful patterns. VSEPR can predict molecular shape.
But chemistry becomes more interesting when we ask what lies underneath those representations. X-ray diffraction showed that atomic arrangements could be measured. Sulfur chemistry exposed the limits of simple electron-counting rules. Modern bonding theory shows that electrons do not always belong neatly to one atom or one bond.
A chemical structure is not simply something we draw. It is a model built from experimental evidence about the arrangement and behaviour of nuclei and electrons.
Luke Edwards-Stuart
Exploring chemistry beyond the specification through the experiments, evidence and ideas that shaped how the subject is understood today.