Formal Potentials and Activity Effects

Every commercial fluoride, calcium, potassium, and other ion-selective electrode is calibrated using a TISAB-type ionic strength adjustment buffer precisely so that the activity coefficient — and hence the relationship between the electrode's measured potential and the sample's true ion concentration — stays constant between standards and unknowns; this is a direct, everyday application of the activity-coefficient theory developed here, and it underlies Ion-Selective Electrode Theory as a discipline. Clinical blood-gas and electrolyte analyzers (measuring Na⁺, K⁺, Ca²⁺, Cl⁻ in serum) report results calibrated against formal potentials in reference sera whose ionic strength and protein content are matched to real plasma, because raw Debye–Hückel theory cannot handle the complex, high-ionic-strength, protein-laden matrix of blood — the calibration is empirical, but the underlying justification for why matrix matching works is exactly the activity-coefficient argument above. In classical wet-chemical volumetric analysis, formal potentials for Fe³⁺/Fe²⁺, Ce⁴⁺/Ce³�III, and MnO₄⁻/Mn²⁺ in specific acid media (H₂SO₄, HCl, HClO₄) are tabulated precisely because these values, not the idealized $E°$, correctly predict equivalence-point potentials and thus guide the choice of redox indicator in real titrations — a direct link to Potentiometry Fundamentals and to redox titrimetry more broadly. In environmental and geochemical analysis, natural waters of varying salinity (freshwater to seawater, spanning $I$ from millimolar to ~0.7 M) require ionic-strength-corrected equilibrium constants and electrode calibrations to interpret pH, alkalinity, and trace-metal speciation measurements correctly — the pH scale itself has an operational, activity-based definition that inherits directly from this theory. In battery and fuel-cell electrochemistry, formal potentials measured in the actual working electrolyte (concentrated LiPF₆ in organic carbonate solvents, or concentrated KOH) rather than aqueous standard-state values are what determine real cell voltages and are used in engineering design, since standard aqueous $E°$ tables are frequently inapplicable to nonaqueous, high-ionic-strength battery electrolytes.

Ask the tutor: How do I choose which form of the Debye–Hückel or Davies equation to use for a specific ionic strength and ion, and how much error results from picking the wrong one?

Consider a silver wire in contact with a solution containing Ag⁺. The half-reaction is

with standard potential $E° = +0.799$ V (vs. SHE), tabulated for the hypothetical state where Ag⁺ behaves ideally — infinitely dilute, no other ions present to interfere. In any real solution, Ag⁺ is surrounded by a diffuse cloud of counter-ions (an "ionic atmosphere"): NO₃⁻ if the salt is AgNO₃, but also whatever supporting electrolyte, buffer, or ionic strength adjuster has been added. This ionic atmosphere is on average negatively charged, and it electrostatically stabilizes the Ag⁺ ion, lowering its chemical potential relative to the noninteracting ideal case. Thermodynamically, this stabilization is captured by writing the true chemical potential as

where $\gamma_{\text{Ag}^+} < 1$ in the presence of ionic strength. Because the Nernst equation is derived directly from equating electrochemical potentials at equilibrium, activity — not molar concentration — is the correct variable inside the logarithm. Substituting concentration for activity introduces a systematic error that grows with ionic strength and with the charge of the ion. The size of this effect is not small. For a singly charged ion at $I = 0.10$ M, $\gamma \approx 0.78$; for a doubly charged ion at the same ionic strength, $\gamma$ can fall below 0.4. Because the Nernst term is $\frac{RT}{nF}\ln a$, an activity coefficient of 0.78 for $n=1$ shifts the potential by $\frac{0.0592}{1}\log(0.78) \approx -6.5$ mV at 25 °C — comparable to or larger than the precision of a good potentiometric measurement (±0.1–1 mV). For $n=2$ ions the shift is larger still, and because $\gamma$ depends on $I$, which changes throughout a titration or from sample to sample, the correction is not a fixed offset — it must either be calculated from theory or, far more commonly, be swamped out experimentally.

The activity coefficient itself is predicted by the Debye–Hückel theory, which treats the ionic atmosphere as a continuous charge distribution obeying the linearized Poisson–Boltzmann equation. The simplest ("limiting") form applies only below about $I = 0.01$ M:

Real analytical solutions — buffers, titration media, physiological fluids — routinely have $I$ in the range 0.05–0.5 M, well outside the limiting law's validity. The extended Debye–Hückel equation adds a size-dependent denominator that accounts for the finite radius $\alpha$ (in picometers, conventionally expressed in Å ×100) of the hydrated ion, extending validity to roughly $I \le 0.1$ M:

For still higher ionic strengths (up to $I \approx 0.5$ M), the empirical Davies equation drops the ion-size parameter and adds a linear term that curves $\gamma$ back upward at high $I$, matching experiment better than the extended law alone:

The rigorous, activity-based Nernst equation for a general half-reaction $\text{Ox} + ne^- \rightleftharpoons \text{Red}$ is

where:

• $E°$ = standard (thermodynamic) potential, referenced to unit activity of all species, tabulated at 25 °C vs. SHE,
• $\gamma_{\text{Ox}}, \gamma_{\text{Red}}$ = activity coefficients of the oxidized and reduced forms at the working ionic strength,
• $[\text{Ox}], [\text{Red}]$ = molar concentrations, the quantity actually controlled or measured in the lab,
• $R, T, F, n$ = gas constant, temperature, Faraday constant, number of electrons transferred.

Grouping the activity coefficients with $E°$ defines the formal potential:

so that the Nernst equation can be rewritten entirely in terms of concentrations, valid at one specific, fixed ionic strength (and, where relevant, one specific pH and ligand concentration):

Reading: $E°'$ is not a universal constant — it is a conditional constant, valid only for the medium in which it was measured (e.g., "1 M HCl" or "0.10 M ionic strength, pH 4.5 acetate buffer"). It absorbs three physically distinct effects: (1) pure electrostatic activity-coefficient shifts from ionic strength, (2) side reactions such as protonation or complexation that convert some of the analytical concentration into electro-inactive species, and (3) any residual liquid-junction potential that is reproducible under fixed conditions. Analytical chemists work with $E°'$, not $E°$, because it is what the electrode actually reports under real working conditions. A parallel and equally important relationship connects the mean activity coefficient $\gamma_\pm$ of an electrolyte to individual ionic activity coefficients that cannot be measured separately (single-ion activities are not thermodynamically accessible, but Debye–Hückel theory nonetheless assigns each ion a $\gamma_i$ using a convention that reproduces the measurable $\gamma_\pm$):

where:

• $p, q$ = stoichiometric numbers of cations and anions per formula unit of electrolyte,
• $\gamma_+, \gamma_-$ = individual ion activity coefficients from the Debye–Hückel or Davies equation.

This clarifies an important scope point: $E°'$ is an operationally defined, medium-specific constant, reproducible and useful, but it is not a universal thermodynamic property in the way $E°$ is (in the ideal, infinite-dilution limit).

Example 1 — Foundational: activity coefficient and the resulting potential shift for Ag⁺/Ag. A cell contains 0.0250 M AgNO₃ together with 0.075 M KNO₃ as inert electrolyte (deliberately added to fix ionic strength). Calculate (a) the ionic strength, (b) $\gamma_{\text{Ag}^+}$ by the extended Debye–Hückel equation ($\alpha_{\text{Ag}^+} = 250$ pm), and (c) the resulting shift in the measured potential relative to the concentration-based (uncorrected) Nernst prediction. Ionic strength. $I = \frac12\sum c_iz_i^2$. Species: Ag⁺ (0.0250 M, $z=1$), NO₃⁻ from AgNO₃ and KNO₃ (0.0250 + 0.075 = 0.100 M, $z=-1$), K⁺ (0.075 M, $z=1$).

Activity coefficient. Using the extended Debye–Hückel equation with $\alpha = 250$ (i.e. 2.5 in the 0.33α convention, pm/100 = 2.5):

Potential shift. The Nernst equation for Ag⁺/Ag ($n=1$) uncorrected reads $E = E° + 0.0592\log[\text{Ag}^+]$; the activity-corrected equation reads $E = E° + 0.0592\log(\gamma[\text{Ag}^+])$. The difference (the systematic error from ignoring activity) is

So an electrode reading is 7.6 mV lower than the naive concentration-based Nernst equation predicts — corresponding to reporting an apparent [Ag⁺] about 26% lower than the true concentration if the analyst forgets the activity correction. This is a large, non-negligible error for a technique whose precision is often better than 0.2 mV, and it is exactly the kind of systematic bias that formal-potential calibration is designed to eliminate. Limiting case check: as $I\to0$ (e.g., diluting to $10^{-4}$ M with no added electrolyte), $\sqrt I \to 0.01$, and $-\log\gamma \to 0.51(0.01) = 0.0051$, giving $\gamma \to 0.988 \approx 1$ — the correction becomes negligible, exactly as required at infinite dilution.

Example 2 — Applied: constructing a formal potential for Fe³⁺/Fe²⁺ in 1 M HCl and using it in a redox titration. The tabulated standard potential for $\text{Fe}^{3+} + e^- \rightleftharpoons \text{Fe}^{2+}$ is $E° = +0.771$ V vs. SHE. In 1 M HCl, however, chloride complexes Fe³⁺ substantially more strongly than Fe²⁺ (forming species such as FeCl²⁺, FeCl₂⁺), and the literature formal potential in this exact medium is $E°'_{\text{1 M HCl}} = +0.700$ V. Use this formal potential to compute the potential of an Fe³⁺/Fe²⁺ half-cell in 1 M HCl at the point where 60.0% of the iron has been reduced from Fe³⁺ to Fe²⁺ ($[\text{Fe}^{3+}]/[\text{Fe}^{2+}] = 40.0/60.0$), and compare with what the uncorrected $E°$ would predict. Using the formal potential (correct for this medium):

Using the (incorrect, for this medium) tabulated $E°$:

The two calculations differ by $0.761 - 0.690 = 0.071$ V — 71 millivolts, entirely due to chloride complexation of Fe³⁺ shifting the effective potential, an effect roughly ten times larger than the pure activity-coefficient correction in Example 1. This is the practical reason iodometric and cerimetric titration curves for iron are always calculated using medium-specific formal potentials (e.g., $E°'$ in HCl, in H₂SO₄, in HClO₄ are all different, spanning roughly 0.70–0.77 V) rather than the single $E°$ value — using the wrong one shifts the calculated equivalence-point potential and can shift indicator choice for visual redox titrations. Limiting case check: if no chloride (or other complexing anion) were present and ionic strength were negligible, $E°'\to E°$ exactly, recovering the textbook value 0.771 V — confirming that the formal potential correctly generalizes, rather than replaces, the standard potential.

Example 3 — Exam/competition level: designing an ionic-strength-adjustment buffer (ISAB) and predicting the residual error from imperfect matching. A fluoride ISE calibration protocol calls for adding equal volumes of TISAB (total ionic strength adjustment buffer, typically ~1 M NaCl + acetate buffer) to both standards and samples so that background ionic strength is fixed at $I = 0.50$ M in every solution measured. Suppose an analyst, in a rush, adds TISAB to the 3 calibration standards (0.100 M, 0.0100 M based dilutions giving $I=0.50$ M as intended) but the unknown sample — a natural water containing 0.0080 M Ca²⁺ and 0.0040 M Mg²⁺ as the dominant background electrolyte before TISAB addition — ends up, after the same TISAB addition, at a true ionic strength of $I = 0.560$ M instead of 0.500 M because the analyst forgot that divalent cations contribute $z^2=4$ to $I$. Estimate the resulting error in reported fluoride concentration if the electrode responds Nernstian to F⁻ activity with $n=1$ (note: sign for anion, $E = K - 0.0592\log a_{F^-}$). Step 1 — activity coefficient of F⁻ at each ionic strength (Davies equation, appropriate at these higher $I$, $z_{F^-}=-1$): At $I = 0.500$ M (calibration medium):

At $I = 0.560$ M (actual sample medium):

Step 2 — the calibration curve assumes $\gamma_{F^-}=0.7334$ (from the standards) applies to the sample too; the true value is 0.7365. The electrode itself responds to activity $a=\gamma[\text{F}^-]$, so the true potential contains $\gamma_{0.560}$, but the calibration equation (built from the standards) implicitly back-calculates concentration assuming $\gamma_{0.500}$. The systematic error in $\log[\text{F}^-]_{\text{reported}}$ is

giving a fractional concentration error of

i.e. only about +0.4% — reassuringly small, because both ionic strengths are already high enough that $\gamma_{F^-}$ is on the flat part of the Davies curve (both around 0.73–0.74). This worked example demonstrates why TISAB works even when matching is imperfect: deliberately pushing $I$ high (0.5 M) moves the system onto a region where $\gamma$ is a weak, slowly varying function of small errors in $I$, so modest mismatches in background ionic strength (60 mM out of 500 mM, a 12% relative error in $I$) produce only sub-percent errors in reported activity/concentration — vastly better than the ~26% error that an unbuffered, low-$I$ measurement (Example 1) would suffer from a comparable absolute mismatch. This is the central design logic of ISAB: push $I$ high and constant so $d\gamma/dI \to$ small, rather than trying to compute $\gamma$ exactly.

The theoretical backbone of activity-coefficient corrections was built by Peter Debye and Erich Hückel in 1923, working in Zürich on the statistical mechanics of electrolyte solutions. Debye, a physical chemist with a background in theoretical physics (he had studied under Arnold Sommerfeld and worked on dipole moments and X-ray diffraction), recognized that the anomalously large deviations from ideality seen in strong electrolyte solutions — deviations far larger than in nonelectrolyte solutions at the same concentration — must arise from long-range Coulombic interactions between ions rather than from the short-range "chemical" nonideality that dominated existing solution theory. Hückel, then a young collaborator, worked through the statistical mechanical derivation using the linearized Poisson–Boltzmann treatment of the ionic atmosphere, producing the limiting law that now bears both names. The theory was a landmark because it derived, from first principles and a single adjustable radius, quantitative predictions for activity coefficients that matched freezing-point depression and electromotive force data across a range of 1:1, 2:1, and 2:2 electrolytes — the first genuinely predictive theory of strong-electrolyte nonideality. Debye went on to a Nobel Prize in Chemistry (1936) for work on molecular structure via dipole moments and diffraction, of which the electrolyte theory was one strand among several; Hückel later became equally famous, independently, for the Hückel molecular orbital theory of aromatic π-systems (the origin of the 4n+2 rule), an entirely different but equally foundational contribution to physical organic chemistry.