Chemical Bonding and Molecular Geometry

This is Unit Six of Chemistry, Part Two: VSEPR and molecular polarity. Part One established what a bond is and how to draw one. This part is about where the atoms actually sit in space, and why that matters. The through-line is a chain: a Lewis structure gives a domain count, the domain count gives a geometry, and the geometry decides whether a molecule with polar bonds is itself polar — which, as the last topic shows, it often is not. By the end you will be able to say what VSEPR predicts and what it does not; tell electron-pair geometry from molecular geometry; say how a double or triple bond counts in a domain count; say why polar bonds do not guarantee a polar molecule; and say what xenon difluoride shows about linear geometry.

Three entries. The one assumption the whole theory rests on, and then a pair of terms that sound like synonyms, describe the same atom, and mean different things — which the book locks explicitly because the distinction runs through everything after it.

As far apart as possible.

Chem 1035 VSEPR Theory's Core Assumption

Valence shell electron pair repulsion theory assumes that electron domains around a central atom spread as far apart as possible, minimising repulsion. A central atom with two bonding domains spreads them to opposite sides, a hundred and eighty degrees apart. VSEPR predicts geometry from domain repulsion alone — it offers no explanation of why a bond forms at all.

Count everything.

Chem 1038 Electron-Pair Geometry

Electron-pair geometry describes the arrangement of every electron domain around a central atom, bonding pairs and lone pairs alike. Ammonia's nitrogen has four domains — three bonds and one lone pair — giving a tetrahedral electron-pair geometry. It counts every domain including lone pairs, which is what separates it from the narrower term coming next.

Name only the atoms.

Chem 1040 Molecular Geometry

Molecular geometry names only the arrangement of bonded atoms around a central atom, though unnamed lone pairs still shape that arrangement. Ammonia's molecular geometry is trigonal pyramidal, naming its three hydrogens, while its electron-pair geometry is tetrahedral. Using the two terms interchangeably misreports whether a lone pair is being counted.

Three entries covering the commonest domain counts. Watch what happens as lone pairs are added: the electron-pair geometry stays put while the molecular geometry gets a different name, which is the distinction from the last topic doing real work.

Straight line.

Chem 1043 Two-Domain Geometry: Linear

A central atom with two electron domains and no lone pairs adopts a linear electron-pair and molecular geometry, with a hundred-and-eighty-degree bond angle. Beryllium fluoride, with two bonding domains and no lone pairs on beryllium, is linear. Linear geometry from two domains requires zero lone pairs on the central atom — a lone pair would change the count.

Flat, or bent.

Chem 1045 Three-Domain Geometry: Trigonal Planar and Bent

A central atom with three electron domains and no lone pairs is trigonal planar; with one lone pair among those three, it is bent. Boron trifluoride, with no lone pairs, is trigonal planar; sulfur dioxide, with one lone pair among three domains, is bent. The electron-pair geometry stays trigonal planar in both cases — only the molecular geometry changes.

Three names, one shape.

Chem 1048 Four-Domain Geometry: Tetrahedral, Trigonal Pyramidal, and Bent

A central atom with four electron domains is tetrahedral with no lone pairs, trigonal pyramidal with one, and bent with two. Methane is tetrahedral, ammonia trigonal pyramidal, water bent. All three share the same tetrahedral electron-pair geometry — their molecular geometries differ only by lone-pair count.

Three entries. Two larger geometries — one of which has positions that are not all equivalent, unlike everything before it — and then the procedure that takes any Lewis structure and names its shape.

Two kinds of position.

Chem 1051 Five-Domain Geometry: Trigonal Bipyramidal

A central atom with five electron domains and no lone pairs adopts a trigonal bipyramidal geometry, which has two distinct bond-angle positions. Phosphorus pentachloride, with five bonding domains and no lone pairs, is trigonal bipyramidal. Its two angle types, ninety and a hundred and twenty degrees, mean its five positions are not all equivalent — unlike a tetrahedron's four.

Three dimensions.

Chem 1053 Six-Domain Geometry: Octahedral

A central atom with six electron domains and no lone pairs adopts an octahedral geometry, with every position equivalent at ninety degrees from its neighbours. Sulfur hexafluoride, with six bonding domains and no lone pairs, is octahedral. Do not picture six domains as a flat hexagon at a hundred and twenty degrees — the octahedron is three-dimensional and all six positions are genuinely ninety degrees apart.

Count, name, then check.

Chem 1056 Assigning VSEPR Geometry from a Lewis Structure

Assigning a VSEPR geometry means counting the central atom's electron domains from its completed Lewis structure, naming the electron-pair geometry from that count, then naming the molecular geometry by identifying which domains are lone pairs. Sulfur dioxide's structure has one lone pair, one single bond and one double bond — three domains, so trigonal planar, and bent as a molecular geometry. Counting a double or triple bond as more than one domain, rather than one regardless of bond order, gives the wrong shape.

Three entries, and the payoff of the whole part. Whether a molecule is polar depends on its bonds and its shape together, symmetric shapes cancel bond dipoles entirely, and assuming otherwise is the commonest error in this topic.

Both together.

Chem 1060 Molecular Polarity from Bond Polarity and Geometry

A molecule is polar only if it has polar bonds arranged asymmetrically, so that their individual dipoles fail to cancel to zero. Water's two oxygen-hydrogen bond dipoles, bent at an angle rather than pointing opposite, add to a net molecular dipole. Bond polarity alone never determines molecular polarity — the overall geometry decides whether those dipoles cancel.

Cancelling out.

Chem 1063 Polar Bonds Can Cancel to a Nonpolar Molecule

A molecule with polar bonds arranged symmetrically around its central atom has bond dipoles that cancel, leaving the whole molecule nonpolar. Carbon dioxide's two polar carbon-oxygen bonds point in exactly opposite directions across its linear geometry, cancelling to a net dipole of zero. Those individual bonds are genuinely polar — only the symmetric arrangement makes the molecule nonpolar.

The common error.

Chem 1065 The Polar-Bonds-Therefore-Polar-Molecule Misconception

Assuming a molecule is polar whenever it contains polar bonds ignores that molecular geometry can cancel those dipoles entirely. A student predicting carbon dioxide is polar because its carbon-oxygen bonds are polar contradicts the molecule's measured zero dipole moment. Judging polarity from bond polarity alone, skipping the geometry step, is the single most common error in this topic.

One entry to close the part: the capstone, which runs a single molecule the whole length of that chain — structure, then domain count, then geometry, then polarity — and picks a molecule where the obvious shortcut gets it wrong.

Xenon difluoride.

Chem 1070 Capstone: From Lewis Structure to Molecular Polarity

Xenon difluoride has two bonding domains and three lone pairs on the central xenon — five domains, so a trigonal bipyramidal electron-pair geometry. Lone-pair repulsion is strongest at ninety degrees, so all three lone pairs take the roomier equatorial positions, leaving the two fluorines axial and a hundred and eighty degrees apart. Naming only the bonded atoms, the molecular geometry is linear. Each xenon-fluorine bond is genuinely polar, but the two point exactly opposite and cancel, so the molecule is nonpolar. Assuming a linear molecular geometry means a linear electron-pair geometry misses this case entirely.

Five things to carry out of this part. VSEPR predicts a molecule's shape from the assumption that electron domains spread as far apart as they can — and what it does not do is explain why a bond forms in the first place, which is the next part's subject. Electron-pair geometry counts every domain around a central atom, bonding and lone pair alike; molecular geometry names only where the bonded atoms sit. Ammonia is tetrahedral by the first and trigonal pyramidal by the second, and using the terms interchangeably misreports whether lone pairs are being counted. A double or triple bond counts as one electron domain regardless of its bond order, which is why sulfur dioxide has three domains rather than four and comes out bent. Polar bonds do not guarantee a polar molecule, because a symmetric arrangement makes the bond dipoles cancel: carbon dioxide's two carbon-oxygen bonds are genuinely polar and its measured dipole moment is zero. And xenon difluoride shows that a linear molecular geometry need not come from a linear electron-pair geometry — its five domains are trigonal bipyramidal, and it is the three lone pairs taking equatorial positions that leave the two bonded fluorines in a line. Part Three takes up the bonding models themselves, and closes by applying two of them to one molecule.