Diagonal Relationships in the p-Block

The diagonal relationship is not a historical curiosity confined to textbook tables — it explains real industrial and biological chemistry. Lithium-ion battery electrolyte and electrode chemistry hinges on Li⁺'s unusually high charge density relative to other alkali metals: Li⁺'s strong polarizing power gives it high desolvation energy and directs coordination chemistry with carbonate solvents and oxide lattices (e.g., LiCoO₂ intercalation) that more resembles Mg²⁺-type behaviour in related magnesium-ion battery research than it resembles Na⁺ or K⁺ battery chemistry — which is precisely why researchers developing Mg-ion batteries as a "beyond lithium" technology look to Li-ion electrolyte chemistry as the nearest analogue, rather than to sodium-ion systems, even though Mg is two groups away from Li. In geochemistry and mineralogy, Cartledge's original motivation, ionic potential correctly predicts which cations get mobilized versus retained during weathering: high-φ cations (Al³⁺, Fe³⁺) hydrolyze and precipitate as insoluble hydroxide/oxide phases (laterites, bauxite) while low-φ cations (Na⁺, Ca²⁺) stay in solution and are carried to the oceans — the diagonal-adjacent boundary between Be/Al-type retention and Mg/Ca-type mobility is a working tool in soil and sediment chemistry today. In organometallic and main-group synthesis, the Li/Mg diagonal relationship underlies the practical interchangeability of organolithium and Grignard (organomagnesium) reagents as carbanion-equivalent nucleophiles in countless C–C bond-forming reactions; both are used, often somewhat interchangeably in mechanism teaching, precisely because Li⁺ and Mg²⁺'s comparable charge density gives their organometallic bonds comparable ionic-covalent character and comparable aggregation behaviour in ethereal solvents. In materials chemistry, the Be/Al analogy explains why beryllium alloys and beryllia (BeO) ceramics are engineered with design rules borrowed from aluminium metallurgy (both form a tenacious, passivating oxide film, both are used as lightweight structural or high-thermal-conductivity materials), and the B/Si analogy underlies borosilicate glass technology, where boron oxide is deliberately incorporated into a silica network because B₂O₃ and SiO₂ share compatible network-forming, glass-forming covalent oxide chemistry, unlike the ionic oxides of true Group 13/14 heavier congeners. The same charge-density logic foreshadows why boron, uniquely among Group 13 elements, behaves as a genuine nonmetal — never forming a simple B³⁺ cation, always relying on covalent, electron-deficient, cluster (borane, boride) bonding — precisely because its "effective" charge density is so extreme that no real ionic lattice can accommodate it, a theme picked up directly in treatments of Boron: Nonmetallic Character and Electron Deficiency.

Ask the tutor: Why does the diagonal relationship weaken so sharply after Be/Al, and does it really extend meaningfully to N/S or F/Cl?

Start from the ordinary two-dimensional logic of the periodic table. Down a group, the number of principal shells increases, so ionic and atomic radii grow, and because the outer electrons are farther from the nucleus and better shielded, ionization energy and electronegativity fall, and the cation's charge density $\phi = z/r$ (using the ionic charge $z$ and ionic radius $r$) drops substantially. Across a period, the ionic charge on the "normal" cation for that group rises by one unit (Li⁺ → Be²⁺ → B³⁺, or Na⁺ → Mg²⁺ → Al³⁺), while the radius shrinks only modestly because the extra proton pulls the same shell in slightly. Charge density therefore rises across a period. Now combine the two motions. Step diagonally down-and-right — e.g., from Li (period 2, Group 1) to Mg (period 3, Group 2). Radius increases somewhat (extra shell, partially offset by the group-2 contraction relative to group 1 in the same period) and charge increases by one unit. The size increase and the charge increase happen to be of comparable magnitude for this particular family of diagonal steps, so $\phi_{\text{Li}^+} \approx \phi_{\text{Mg}^{2+}}$. The same near-cancellation recurs for Be²⁺/Al³⁺ and for B (as B³⁺-like character, though boron never truly forms B³⁺) /Si⁴⁺-like behavior in oxoacid chemistry. Charge density is not an abstraction chemists care about for its own sake — it is a direct proxy for polarizing power, the ability of a cation to distort the electron cloud of a neighbouring anion (Fajans' rules). High polarizing power pulls electron density from the anion toward the cation, giving the bond partial covalent character, lowering lattice energy contributions from pure ionic models, increasing solubility in nonpolar/organic solvents, and increasing the tendency of the corresponding hydrated cation to release protons (hydrolyze) because the cation-bound water molecules have their O–H bonds weakened. Two ions with matched $\phi$ therefore polarize a common anion (Cl⁻, O²⁻, OH⁻) to a similar degree, and this single shared property cascades into matched solubility, matched thermal stability of carbonates, matched amphoterism of hydroxides, and matched acid strength of the corresponding oxoacids.

The figure shows the essential arithmetic: Li⁺ (φ ≈ 0.013) and Mg²⁺ (φ ≈ 0.028) are far closer in charge density to each other than either is to its heavier or lighter group congener, and the same near-match recurs for Be²⁺/Al³⁺. Sodium, sitting directly below lithium, has a much smaller φ (0.0098) — this is why NaCl is a "normal" ionic salt while LiCl shows measurable covalent character. A second useful lens is Fajans' rules restated as an explicit inequality. A cation polarizes an anion strongly when it is small and highly charged and when the anion is large and easily deformed. Because both members of a diagonal pair have comparable $\phi$, they sit at comparable positions on the "how covalent will this bond be" scale, even though one has smaller individual charge and radius values than the other member of its own vertical group.

The master relationship is simply Coulomb's law written per unit charge on the cation:

where:

• $\phi$ — ionic potential (charge density), conventionally in units of $e/\text{pm}$ or $e/\text{Å}$
• $z$ — the formal ionic charge (dimensionless, in units of $e$)
• $r$ — the ionic radius (pm or Å), typically the Shannon/Pauling crystal radius for the appropriate coordination number

Reading: $\phi$ measures how much charge is packed per unit of surface area on the ion; a small, highly charged ion has a very "concentrated" field. Chemists sometimes use $z/r^2$ (Cartledge's original ionic potential definition, matching the field at the ion surface rather than the potential) — both forms track together and lead to the same qualitative diagonal conclusions, so this article uses the simpler $z/r$ throughout, consistent with common inorganic-textbook usage. The condition for a diagonal match between an element in period $n$, group $g$ (ion charge $z$, radius $r$) and its diagonal partner in period $n+1$, group $g+1$ (ion charge $z+1$, radius $r'$) is that the fractional changes in numerator and denominator approximately cancel:

where:

• $\Delta z = 1$ — the charge increases by exactly one unit per diagonal step (moving one group right)
• $\Delta r = r' - r$ — the change in ionic radius on going diagonally
• the approximation uses first-order propagation of relative error, valid because $\Delta z/z$ and $\Delta r/r$ are both modest (tens of percent) for these pairs

Reading: the relative jump in charge (which is large for small $z$, e.g. 100% going from $z=1$ to $z=2$) must be matched by a comparably large relative jump in radius for $\phi$ to stay level. This is precisely why the relationship is strongest for the smallest ions (Li/Mg, Be/Al) — a one-unit charge change is a huge fractional change when $z$ starts at 1 or 2 — and gets weaker for heavier, larger ions where a unit charge change is a smaller fraction of an already-big $z$, while radius changes remain comparable in absolute pm. A second core relationship connects $\phi$ to the degree of covalent/polarization character, following Fajans qualitatively:

where:

• $\phi_{\text{cation}}$ — the cation's charge density as above
• $\alpha_{\text{anion}}$ — the polarizability of the anion (larger, softer anions like I⁻ or S²⁻ are more polarizable than F⁻ or O²⁻)

Reading: this is a proportionality, not an exact law, but it correctly predicts that LiI is markedly more covalent than LiF, and that the whole diagonal pair Li/Mg (both moderate φ) will show more covalent character in their halides than the whole diagonal pair Na/Ca (both low φ, one group further down each), even though Na and Li share a group.

Example 1 (foundational): Rank Na⁺, Mg²⁺, Al³⁺, and Li⁺ by charge density and identify the diagonal pairs. Use Shannon six-coordinate radii: Na⁺ 102 pm, Mg²⁺ 72 pm, Al³⁺ 53.5 pm, Li⁺ 76 pm.

Reading down each isolated group in the usual way (Na is "below" Li in Group 1) gives a huge relative jump from Li to Na of $(0.0132-0.0098)/0.0132 = 25.8\%$ decrease. But stepping diagonally from Li (Group 1) to Mg (Group 2, one period down) gives $\phi$ rising from 0.0132 to 0.0278 — more than double — while stepping diagonally again from Mg to Al gives 0.0278 to 0.0561, again roughly double. The ratios Mg/Li = 2.11 and Al/Mg = 2.02 are nearly identical to each other, showing that the diagonal step multiplies φ by a strikingly consistent factor of ≈2 across the whole triad Li→Mg→Al, which is the real signature of the diagonal relationship: not that φ stays exactly constant, but that each diagonal step produces a similar, moderate, predictable jump rather than the erratic or extreme jumps seen on other paths through the table (e.g., Li→Al directly, skipping a step, multiplies φ by 0.0561/0.0132 = 4.25, a much less "gentle" change).

Example 2 (applied): Predict and rationalize the solubility of LiCl and BeCl₂ in ethanol, compared with NaCl. LiCl is known experimentally to be freely soluble in ethanol, acetone, and pyridine (solubility in ethanol ≈ 24.5 g per 100 mL at 20 °C region, vastly exceeding NaCl's negligible solubility in the same solvent). Rationalize this using φ and Fajans' rules: Li⁺ (φ = 0.0132) is small enough to polarize the large, soft Cl⁻ ion (r ≈ 181 pm) appreciably, giving LiCl significant covalent character; covalent compounds dissolve preferentially in polar-but-not-highly-ionizing organic solvents like ethanol via dipole-dipole and hydrogen-bonding interactions rather than requiring the high lattice-energy-compensating hydration enthalpy that pure ionic solids like NaCl need from water specifically. BeCl₂ (φ for Be²⁺ = 0.0444, even higher than Li⁺) is polarizing enough that it is essentially covalent — it exists as a chain polymer in the solid and as chloro-bridged dimers (Be₂Cl₄) in the vapour, directly analogous to Al₂Cl₆ dimers formed by AlCl₃ (φ for Al³⁺ = 0.0561, the highest of the four, again matching the "diagonal step roughly doubles φ" pattern) — both dissolve in and are extracted by nonpolar organic solvents, a hallmark of covalent halides. Contrast with MgCl₂ (φ = 0.0278, matching Group 2's more modest polarizing power at this member): MgCl₂ is substantially more ionic, has a normal rock-salt-derived structure, and does not form volatile covalent dimers, showing that Be, not Mg, is Al's true diagonal partner in covalent-halide behaviour — while Li, sitting at a similarly modest φ to Mg, is Mg's diagonal partner, completing the classic Li/Mg, Be/Al chain.

Example 3 (exam/competition-level): Amphoterism boundary — use charge density to predict, quantitatively, why Be(OH)₂ is amphoteric but Mg(OH)₂ is not, and estimate the φ threshold for amphoteric hydroxide behaviour in Group 2. Amphoterism in a hydroxide M(OH)ₙ requires the M–O bond to be polarized enough that the hydroxide can act as a Brønsted/Lewis acid toward strong base (dissolving as an oxo-anion, e.g. $[\text{Be(OH)}_4]^{2-}$ or $[\text{Al(OH)}_4]^-$) as well as a base toward strong acid. Empirically, among Group 2 hydroxides only Be(OH)₂ is amphoteric; Mg(OH)₂ through Ba(OH)₂ are purely basic, with basicity increasing down the group (consistent with monotonically falling φ: Ba²⁺ has $\phi = 2/135 = 0.0148$, the lowest of the group at CN 6). Take the φ values already computed: Be²⁺ = 0.0444, Mg²�sup>2+ = 0.0278, Ca²⁺ = 0.0200, Sr²⁺ (r=118 pm) = 0.0169, Ba²⁺ (r=135 pm) = 0.0148. There is a large relative gap between Be²⁺ and Mg²⁺ — a factor of 1.60 — that is much larger than any subsequent gap in the group (Mg→Ca is a factor of 1.39, Ca→Sr 1.18, Sr→Ba 1.14). Amphoterism switches on right at this largest gap in the series, consistent with a genuine threshold in φ (roughly φ ≳ 0.040 for six-coordinate Group 2 dications) above which the M–OH bond becomes polarized enough for the hydroxide to also act as an acceptor toward OH⁻, forming $[\text{M(OH)}_4]^{2-}$-type species, and below which it does not. Cross-check against Al³⁺ (φ = 0.0561, well above the Be²⁺ threshold, and Al(OH)₃ is indeed strongly amphoteric, arguably more so than Be(OH)₂) and against B (whose "effective φ" as B³⁺ would be extremely high, ≈0.111 by the earlier estimate, but boron never forms a discrete B³⁺ cation at all — instead the very same underlying driver, extreme charge density, is expressed as covalent bonding and true nonmetal/oxoacid chemistry rather than amphoteric hydroxide chemistry). This is the natural hand-off point to boron's electron-deficient, non-cationic behaviour.

The systematic charge-density (ionic potential) explanation of diagonal similarities is generally credited to the British-American mineralogist and geochemist Guy Cartledge, who in 1928 formalized the concept of "ionic potential" ($\phi = z/r$, sometimes $z/r^2$ in his original field-strength form) as a predictive tool for cation behaviour in aqueous and mineral chemistry, building on earlier qualitative recognition of diagonal similarities by 19th- and early 20th-century inorganic chemists cataloguing the "anomalous" first members of each main group. The observations that Li, Be, and B behave oddly relative to their own groups long predate any quantitative explanation — they were folklore of descriptive inorganic chemistry — but Cartledge's ionic-potential framework, later absorbed into the broader polarization language of Kazimierz Fajans' rules (1923) on ionic-versus-covalent bonding, gave the phenomenon a single quantitative handle that ties together lattice energetics, hydrolysis, and covalency in one number.