Effective Nuclear Charge

Effective nuclear charge is not a historical curiosity confined to introductory courses — it is the conceptual seed of one of the most heavily used computational tools in modern chemistry and materials science: the pseudopotential (or effective core potential). Large-scale electronic structure calculations on drugs, catalysts, battery materials, and semiconductors routinely replace the chemically inert core electrons of an atom with a fixed effective potential, exactly the $Z_{\text{eff}}$ idea generalized and made numerically precise, because explicitly treating every core electron in, say, a platinum catalyst is computationally wasteful when only the valence electrons participate in bonding. In inorganic and coordination chemistry, $Z_{\text{eff}}$ arguments explain the lanthanide contraction — the unusually poor shielding by $4f$ electrons causes atomic and ionic radii to shrink more than expected across the lanthanide series, with knock-on effects including the near-identical radii (and hence chemical near-twinship) of zirconium and hafnium, a fact exploited and complicated in nuclear reactor materials chemistry where the two elements must be laboriously separated. In main-group chemistry, $Z_{\text{eff}}$ trends underlie the diagonal relationships (Li–Mg, Be–Al, B–Si) prized in synthetic strategy, and rationalize why fluorine, with the highest $Z_{\text{eff}}$ among common nonmetals for its size, is the most electronegative element and why noble-gas compounds (like $\text{XeF}_2$) only became accessible once chemists appreciated how weakly outer $p$ electrons in heavy noble gases are actually bound relative to naive expectation. In spectroscopy and photoelectron spectroscopy (XPS/ESCA), measured binding energies of core electrons are direct experimental probes of $Z_{\text{eff}}$ at the nucleus, used routinely to identify oxidation states of transition metals in catalysts and battery electrode materials — a shift in $Z_{\text{eff}}$ upon oxidation produces a measurable "chemical shift" in the core electron binding energy, the analytical backbone of a technique used daily in surface science and materials characterization laboratories worldwide.

Ask the tutor: How exactly do Slater's empirical shielding-constant rules assign different values (0.35, 0.85, 1.00) to electrons in different shells, and why those particular numbers?

Start with hydrogen, where life is simple: one electron, one proton, and the electron's energy is exactly $E_n = -13.6\,\text{eV}\, Z^2/n^2$ with $Z=1$. The instant a second electron is added — helium — the problem becomes, in principle, unsolvable in closed form. Each electron repels the other with a term $e^2/4\pi\varepsilon_0 r_{12}$ that depends on the instantaneous separation between two moving particles, and this term cannot be separated into independent one-electron pieces. Every atom heavier than hydrogen inherits this difficulty. The mean-field escape route is to replace the instantaneous repulsion from all the other electrons with an average, spherically symmetric charge cloud that partially cancels the nuclear attraction. An electron at distance $r$ from a nucleus of charge $+Ze$ then behaves, to a good approximation, as though it sits in the field of a single point charge

where $S(r)$ counts (roughly) the amount of other-electron charge lying inside radius $r$, which screens the nucleus, while charge farther out contributes essentially nothing. This radius dependence is the honest, rigorous statement; the everyday shorthand used in general chemistry freezes $S$ at a single characteristic value appropriate for the valence shell, giving the familiar constant

Two physical effects control $S$. Shielding is the blocking of nuclear charge by inner electrons that sit, on average, closer to the nucleus — core electrons are excellent shields because they are almost always inside the valence electron's orbit. Penetration is the ability of an electron in a given orbital to spend part of its time very close to the nucleus, inside the shielding cloud of electrons nominally "closer in." Because $s$ orbitals have nonzero probability density right at the nucleus (and pronounced inner maxima in their radial distribution functions for $n>1$), an $ns$ electron penetrates the core more than an $np$ electron of the same $n$, which penetrates more than $nd$. Better penetration means the electron feels more of the raw nuclear charge (less effectively shielded from it) — and, by the same logic, penetrating electrons are not efficiently shielded by electrons in less-penetrating same-shell orbitals, which is why $3s < 3p < 3d$ in energy for multielectron atoms even though all three are formally "$n=3$."

The order 3s penetrates most, then 3p, then 3d, is exactly why, in building up multielectron atoms, the $4s$ orbital (which penetrates through the argon core almost as well as a $3d$ electron is shielded from it) fills before $3d$ in neutral atoms at potassium and calcium — a filling-order puzzle that only makes sense once $Z_{\text{eff}}$, not the bare $n$, is recognized as the thing orbital energies actually track. Across a period, each added proton is nominally cancelled by one added electron — but that new electron goes into the same shell as the ones already there, and same-shell electrons shield each other quite poorly (screening constant contribution roughly 0.35 per same-shell electron, per Slater's empirical rule). The net effect is that $Z_{\text{eff}}$ for the valence shell climbs steadily left to right.

Down a group, by contrast, both $Z$ and the number of shielding core electrons grow together, and $Z_{\text{eff}}$ on the outermost shell stays nearly constant (it typically creeps up only slightly, since penetration of the deep core is never perfect). That near-constancy — combined with the outer electron now occupying a shell of higher principal quantum number, hence larger orbital radius — is precisely why atomic radius increases down a group even as $Z_{\text{eff}}$ barely changes, while radius decreases across a period because $Z_{\text{eff}}$ rises while $n$ stays fixed.

The central defining relationship is simple to state and deceptively rich in consequence:

where:

• $Z$ = atomic number, the actual number of protons in the nucleus,
• $S$ = the shielding (screening) constant, the effective amount of nuclear charge cancelled by the presence of the other electrons in the atom, as experienced by the electron in question,
• $Z_{\text{eff}}$ = the net positive charge the electron of interest actually experiences, which is always less than $Z$ for any real multielectron atom (except for the innermost, unshielded 1s electron in a two-electron-or-fewer system where $S$ is small).

Reading: $Z_{\text{eff}}$ is orbital- and even electron-specific — a 1s electron in chlorine has a very different $Z_{\text{eff}}$ from a 3p electron in the same atom, because $S$ depends on which electron's viewpoint is being computed and on how well the other electrons penetrate to its location. Once $Z_{\text{eff}}$ is known for an electron in shell $n$, its binding is described by a hydrogen-like scaling of the Bohr/Schrödinger energy formula:

where:

• $E_n$ = orbital energy (binding energy, taken negative, of the electron relative to a free electron at rest at infinity),
• $13.6\ \text{eV}$ = the Rydberg energy, the hydrogen ground-state binding energy,
• $Z_{\text{eff}}$ = effective nuclear charge felt by that electron,
• $n$ = principal quantum number of the orbital.

Reading: every trick that changes $Z_{\text{eff}}$ — adding protons, adding shielding electrons, changing which subshell an electron sits in — reappears squared inside the orbital energy; small shifts in $Z_{\text{eff}}$ translate into large shifts in binding energy, which is why ionization energies swing so sharply across a period. Correspondingly, the size of the orbital follows the companion hydrogen-like scaling for the most probable (or mean) radius:

where:

• $r_n$ = characteristic radius of the orbital,
• $a_0 = 52.9\ \text{pm}$, the Bohr radius,
• $n$ = principal quantum number,
• $Z_{\text{eff}}$ = effective nuclear charge felt by the electron in that orbital.

Reading: a larger $Z_{\text{eff}}$ pulls the electron cloud in and shrinks the atom; this is the microscopic mechanism behind the shrinking of atomic and ionic radii across a period, and it is the seed idea developed fully under Periodic Trends in Atomic Radii.

Example 1 — Foundational: $Z_{\text{eff}}$ and orbital energy in potassium's 4s electron. Potassium has $Z=19$ and configuration $[\text{Ar}]4s^1$. A simple, honest (if crude) shielding estimate treats every one of the 18 core electrons as if it fully cancelled one unit of nuclear charge for the outermost 4s electron — a "naive full-shielding" model — giving

This predicts an ionization energy $E = 13.6\,\text{eV}\times (1)^2/4^2 = 0.85\ \text{eV}$, far below the experimental first ionization energy of potassium, $4.34\ \text{eV}$. The naive model over-shields: it assumes the inner 18 electrons block all of the nuclear charge, which ignores that the 4s electron penetrates the argon core and feels noticeably more than $+1$. A more realistic Slater-style estimate (previewing the systematic rule) assigns each of the ten $n=3$ electrons a shielding contribution of $0.85$ (rather than $1.00$) because $4s$ penetrates through the $n=3$ shell somewhat, and each of the eight $n\leq 2$ electrons a contribution of $1.00$ (they screen the 4s electron almost completely):

Then

within about 20% of the measured $4.34\ \text{eV}$ — a large improvement over the naive full-shielding estimate, and good enough to correctly predict, for instance, that potassium's ionization energy is far below hydrogen's ($13.6\ \text{eV}$) but not vanishingly small. Limiting-case check: if the outer electron were in a hypothetical atom with $Z=1$ and no other electrons (hydrogen itself), then $S=0$ identically, $Z_{\text{eff}}=Z=1$, and the formula collapses exactly to the hydrogen result $E=13.6\ \text{eV}\,(1)^2/1^2=13.6\ \text{eV}$, as it must.

Example 2 — Applied: comparing $Z_{\text{eff}}$ for the valence electrons of chlorine and sulfur to rationalize an ionization-energy anomaly. Chlorine ($Z=17$, configuration $[\text{Ne}]3s^2 3p^5$) and sulfur ($Z=16$, configuration $[\text{Ne}]3s^2 3p^4$) sit next to each other in Period 3. Naively one expects ionization energy to rise monotonically left to right, and mostly it does — but P (phosphorus, half-filled $3p^3$) has a higher first ionization energy than S. Effective nuclear charge, together with the shell structure of $S$, explains why sulfur is the dip rather than a smooth increase. For a $3p$ electron in Cl, using the Slater-style same-shell contribution of $0.35$ per additional $n=3$ electron (there are 6 other $n=3$ electrons: $2s$-and-$2p$-analogous partners in the $n=3$ shell, i.e. $2\,(3s) + 4\,(3p)$, contributing $6\times0.35=2.10$), plus $0.85$ each for the eight $n=2$ electrons, plus $1.00$ each for the two $n=1$ electrons:

For sulfur's $3p$ electron, there are only 5 other $n=3$ electrons ($2$ in $3s$, $3$ in $3p$ besides the one considered):

$Z_{\text{eff}}$ does rise from S to Cl ($5.45 \to 6.10$), consistent with Cl having the higher ionization energy of the two ($12.97\ \text{eV}$ vs. S's $10.36\ \text{eV}$) — the overall trend is captured correctly. The P/S anomaly itself is a finer effect not visible at this coarse level: removing an electron from sulfur's $3p^4$ configuration relieves extra electron–electron repulsion associated with having two electrons paired in one $3p$ orbital (violating maximum multiplicity), a stabilization that Slater-type $Z_{\text{eff}}$ bookkeeping does not resolve, since it treats each subshell's shielding contribution as uniform regardless of spin-pairing detail. This is the correct scope caveat: $Z_{\text{eff}}$ explains the bulk trend beautifully but is not fine-grained enough, on its own, to capture exchange-energy corrections from electron pairing.

Example 3 — Exam/competition level: using $Z_{\text{eff}}$ to explain why Cr²⁺ loses a 4s electron before a 3d electron, and predicting the trend in a related ion's radius. Neutral chromium has the (experimentally anomalous but well-established) ground configuration $[\text{Ar}]3d^5 4s^1$ rather than the naively expected $[\text{Ar}]3d^4 4s^2$, reflecting extra exchange stabilization from a half-filled $3d$ subshell. When chromium ionizes to Cr²⁺, both electrons removed come from the 4s orbital rather than one each from 4s and 3d, giving Cr²⁺ = $[\text{Ar}]3d^4$, not $[\text{Ar}]3d^3 4s^1$. Effective nuclear charge explains this. Once even one electron occupies $3d$, that $3d$ electron shields the 4s electron only weakly (its charge density lies mostly outside the region a penetrating 4s electron visits at small $r$, so per Slater's bookkeeping, $3d$ contributes essentially $0$ or a very small shielding value toward a $4s$ electron in the same atom, while contributing something toward electrons further out). Meanwhile the 4s electron, ranging further out on average with weaker net binding once other electrons are present, ends up with the smaller $Z_{\text{eff}}$ (fewer of its own protons "reserved" once the shielding from the growing $3d$ population is tallied for the outer shell) and correspondingly the higher (less negative) orbital energy. Concretely, once the $3d$ subshell is appreciably populated, computed effective-charge estimates place $Z_{\text{eff}}(3d)$ above $Z_{\text{eff}}(4s)$ for the neutral/near-neutral transition metals (unlike the simple $4s$-fills-first heuristic taught for building up configurations), so the 4s electrons are the more weakly bound and are removed first on ionization — consistent with all first-row transition metal cations having configuration $[\text{Ar}]3d^n$ with the 4s subshell emptied first, never the reverse. Quantitatively, a Slater-style comparison for $\text{Cr}^{+}$ ($[\text{Ar}]3d^5 4s^0$ vs. the alternative $[\text{Ar}]3d^4 4s^1$) shows the 4s orbital in the singly-ionized species: using $S(3d\to4s\ \text{shielding}) \approx 1.00$ per $3d$ electron toward a $4s$ electron (Slater's rule places $n\ell$ electrons with $n' < n$, or same $n$ but lower group, at full shielding value 1.00 for the outer electron when the outer electron's group number is "further out"), while $S(4s \to 3d) \approx 0.35$ per other $3d$ electron plus a smaller allowance — the asymmetry directly reproduces $Z_{\text{eff}}(4s) < Z_{\text{eff}}(3d)$ once $3d$ is well populated, so 4s empties first. This qualitative reasoning — not a coincidence of bookkeeping but a real consequence of penetration — is the standard justification chemists give for the "$4s$ fills first, empties first" rule, and it is exactly the kind of subshell-crossing behavior worked out rigorously under Orbital Energy Ordering in Multielectron Atoms. Limiting-case check: for a one-electron ion like $\text{Cr}^{23+}$ (hypothetically stripped to a single electron, $Z=24$), $S=0$ exactly, $Z_{\text{eff}} = Z = 24$, and the energy $E_1 = 13.6\,\text{eV}\times 24^2/1^2 = 7834\ \text{eV}$ recovers the correct hydrogenic scaling for a bare, unshielded nucleus — confirming the formula's built-in consistency at the no-shielding extreme.