The graph you drew in Sorting the Elements — atomic radius against atomic number — has a shape nobody warned you about. It does not climb steadily and it does not scatter. It saws: a long fall, a jump back up, another long fall. That sawtooth is the periodic law showing up in a measurement, and this page explains what produces it.
The honest version of this topic is not four arrows on a diagram. It is two causes, four consequences, and a handful of places where the four consequences are genuinely broken. The breaks are worth more than the trends, because they are the evidence that the explanation is about electrons rather than about arrows.
Two causes do nearly all the work
Effective nuclear charge () is the pull the outermost electron actually feels. It is not the full nuclear charge, because the electrons in between get in the way.
Shielding is that getting-in-the-way. Inner electrons are negative, they sit between the nucleus and the outer electrons, and they cancel part of the attraction. A rough and useful approximation:
Now run that across a row. Sodium to chlorine, every step adds one proton and one electron — but the added electron goes into the same principal shell, at roughly the same distance out, so it shields the others hardly at all. The nuclear charge climbs by seven across the row and the shielding barely moves. rises steeply.
Now run it down a column. Lithium to caesium, each step starts a new principal shell, further from the nucleus, with a whole additional layer of inner electrons beneath it. Distance goes up and shielding goes up together. The outer electron is held much more loosely, even though the nucleus has far more protons in it.
Every trend below is one of those two paragraphs applied to a different measurement.
The four trends
| Trend | Across a period, left to right | Down a group |
|---|---|---|
| Atomic radius | decreases | increases |
| Ionization energy | increases | decreases |
| Electron affinity | more energy released | less energy released |
| Electronegativity | increases | decreases |
Each of those needs its definition stated properly, because three of the four are routinely used to mean something they do not.
Atomic radius is half the distance between the nuclei of two bonded identical atoms. It shrinks across a period because a rising pulls the same shell in tighter — you are adding electrons but the shell contracts faster than they fill it. It grows down a group because you have started a new shell further out. Note that the second effect is much larger than the first: caesium is enormously bigger than lithium, while chlorine is only somewhat smaller than sodium.
Ionization energy is the energy needed to remove the most loosely held electron from a neutral atom in the gas phase. High and a small radius both make it harder to pull an electron off, so it climbs across and falls down. This is the trend that explains why the metals are on the left: a caesium atom gives up an electron for very little.
Electron affinity is the energy change when a gaseous atom gains an electron. Usually energy is released, so the value is usually negative, though many textbooks quote it as “energy released” and print it positive — check which convention your data table is using before you compare two numbers. Broadly, an atom one electron short of a full shell releases the most, which is why the halogens sit at the extreme.
Electronegativity is the tendency of an atom in a bond to pull the shared electrons towards itself. That phrase is the important part: unlike the other three, this is not a property you can measure on a lone atom. It is inferred from bonds, on a comparative scale that Linus Pauling set up with fluorine at the top, near 4.0. It is the number you will use in Ionic and Covalent Bonding to predict what kind of bond two elements will form.
The exceptions are the interesting part
Draw the ionization energies of the first twenty elements and the row is not a clean ramp. There are two dips per period, in the same places every time, and they are not experimental error.
- Group 2 to group 13 — beryllium to boron, magnesium to aluminium. Ionization energy falls. The electron being removed from boron is the first one in a p subshell, which is higher in energy than the filled s subshell below it and is shielded by it. It is easier to remove than the s electron was, despite the extra proton.
- Group 15 to group 16 — nitrogen to oxygen, phosphorus to sulfur. Ionization energy falls again. Nitrogen’s three p electrons sit one to an orbital. Oxygen’s fourth has to pair up with one of them, and two electrons crowded into the same orbital repel each other, so one of them leaves more readily.
Electron affinity has a famous break too. Fluorine ought to be the champion, and it is not — chlorine releases more energy on gaining an electron than fluorine does. Fluorine’s valence shell is so small that the incoming electron is pushed into a crowded space and the electron–electron repulsion cancels part of the gain. The same logic makes oxygen’s electron affinity smaller than sulfur’s.
Two more places the arrows lie:
- Groups 2 and 18 do not really have electron affinities. A magnesium or a neon atom has a filled subshell; adding an electron requires energy rather than releasing it, and the resulting ion is not stable. A trend line drawn straight through those points is drawing through data that does not exist.
- Across the transition metals, atomic radius barely changes. The electrons being added go into an inner d subshell, where they shield the outer electrons quite effectively, so creeps instead of climbing. Iron, cobalt, and nickel are nearly the same size, which is exactly why they alloy so readily.
Is a trend with exceptions still a trend?
Yes — and knowing where it breaks is what separates using a model from reciting one. The dips are not noise sitting on top of the pattern; they are the same explanation, applied one level finer. Subshells and orbital pairing were invented to account for measurements like these. A student who says “ionization energy increases across a period” has learned the rule. A student who can say “except from beryllium to boron, because the electron leaving boron is a shielded p electron” has learned the reason, and the reason is the part that transfers.
Reading the trends together
The four trends are not independent — they are four views of the same thing, how tightly an atom holds electrons.
That single idea sorts the whole table. Elements on the left hold their outer electrons weakly: low ionization energy, low electronegativity, large radius. They lose electrons and become cations, which are always smaller than the atom they came from, because the outer shell has gone entirely. Elements on the right hold electrons tightly and pull hard on anyone else’s: they gain electrons and become anions, which are always larger than the parent atom, because the added electron increases repulsion within an unchanged nuclear charge.
Put a left-hand element next to a right-hand element and the electron goes one way and stays there. Put two right-hand elements together and neither can win outright. That is the whole of Ionic and Covalent Bonding in two sentences, and it is where this goes next — after you have practised pulling trend data off a printed table in Reading a Data Table.
Curriculum connection
B2.2
analyse data related to the properties of elements within a period (e.g., ionization energy, atomic radius) to identify general trends in the periodic table [AI]
Link to original