Combining Half-Reactions and Potential Additivity

Iron sits at the center of a small puzzle. It trips up almost every student of electrochemistry. Given $E^\circ(\mathrm{Fe^{3+}/Fe^{2+}}) = +0.771\,\mathrm{V}$ and $E^\circ(\mathrm{Fe^{2+}/Fe}) = -0.447\,\mathrm{V}$ , what is $E^\circ(\mathrm{Fe^{3+}/Fe})$ ? The tempting answer, $+0.771 + (-0.447) = +0.324\,\mathrm{V}$ , is wrong. The correct value is $-0.037\,\mathrm{V}$ . Nothing is broken in electrochemistry here. The arithmetic of potentials themselves is simply not additive. The arithmetic of free energies underneath them always is. This matters because multi-step redox couples are the rule, not the exception, in real chemistry. Manganese cycles through +7, +6, +4, +3, and +2. Chlorine cycles through +7 down to $-1$ . Nitrogen cycles through five or six oxidation states in the nitrogen cycle. Textbooks and reference tables report standard potentials for convenient steps, often two-electron or one-electron ones. But a chemist frequently needs the potential for a different combination of those same steps. That combination often involves a different number of electrons. Manufacturing that number by naively adding or subtracting volts gives silently wrong predictions. These wrong predictions concern spontaneity, which species will disproportionate, and cell voltages in real devices. The resolution is to stop thinking in volts. Instead, start thinking in the currency the volts are a shorthand for: Gibbs energy. A standard potential $E^\circ$ is not an energy. It is an energy divided by a charge, $\Delta G^\circ = -nFE^\circ$ . Different half-reactions transfer different numbers of electrons $n$ . So their potentials live on different "scales" and cannot be summed directly. But the underlying $\Delta G^\circ$ values, being extensive quantities, always add along any valid thermochemical cycle. This is exactly the same logic as Hess's law, dressed in electrochemical clothing. This article builds the additivity rule for combining half-reactions from first principles. It derives the general $n$ -weighted averaging formula for stepwise couples. It works through the iron example along with a manganese-based worked problem. It also covers a competition-level three-step combination. It closes with the traps every student falls into. It offers a compact algorithm for combining any set of couples. Finally, it bridges toward Latimer Diagrams and Disproportionation and Comproportionation. Both of these are just organized bookkeeping on top of the rule developed here.

Key takeaways

A standard potential $E^\circ$ is Gibbs energy per unit charge; it depends on $n$ , so potentials from half-reactions with different electron counts cannot simply be added or subtracted.
Gibbs energies $\Delta G^\circ = -nFE^\circ$ are extensive and additive along any valid reaction cycle — this is Hess's law applied to redox couples.
To combine half-reactions: convert each $E^\circ$ to $\Delta G^\circ$ (or to $nE^\circ$ ), add/subtract the $\Delta G^\circ$ values algebraically, then divide by the total electrons transferred and by $-F$ (or just divide the $nE^\circ$ sum by total $n$ ).
When combining two half-reactions to get a third, the correct rule is a weighted average: $E^\circ_3 = \dfrac{n_1E^\circ_1+n_2E^\circ_2}{n_1+n_2}$ , not a simple sum.
Full redox reactions (oxidation + reduction, forming a full cell) are the one case where you truly add the two half-cell $E^\circ$ values directly, because the electrons cancel and never appear in $\Delta G^\circ_{\text{cell}} = -nFE^\circ_{\text{cell}}$ in a way that requires weighting.
This machinery is exactly what underlies Latimer Diagrams, where multiple adjacent couples are stitched into a longer one, and what predicts Disproportionation and Comproportionation by comparing an average couple to a directly measured one.
Key idea: potentials don't add — the $n$ -weighted free energies behind them do, always.

The core insight

$\Delta G^\circ = -nFE^\circ$ is extensive and obeys Hess's law exactly; $E^\circ$ is intensive-like (energy per charge) and must be recovered from the combined $\Delta G^\circ$ by dividing by the total number of electrons in the combined step — never by simply summing potentials from steps with unequal $n$ .

Where this lives today

Corrosion engineers use exactly this arithmetic to assess whether iron, chromium, or aluminum alloys will passivate or actively corrode: multi-electron dissolution reactions (e.g., $\mathrm{Fe} \to \mathrm{Fe^{3+}} + 3e^-$ ) are rarely measured directly and must be reconstructed from more accessible one- and two-electron steps, exactly as in Worked Example 1. Analytical chemists titrating with permanganate or dichromate rely on the correctly combined five- or six-electron overall potentials (Worked Example 2's method) to predict titration end-point voltages and to choose appropriate redox indicators, since using an uncombined intermediate potential would misplace the equivalence-point jump on a titration curve. Battery and fuel-cell researchers evaluating multi-electron conversion-reaction electrodes (such as sulfur cathodes, which pass through several polysulfide oxidation states before reaching $\mathrm{S^{2-}}$ ) must assemble the overall cell voltage from a cascade of differently-weighted intermediate couples, not from a naive sum, to correctly predict theoretical energy density. In geochemistry and environmental redox modeling, predicting whether dissolved manganese, iron, or sulfur species will disproportionate in groundwater or ocean sediments depends on comparing a directly measured "skip" potential to the weighted-average prediction from smaller steps — precisely the comparison that defines spontaneous Disproportionation and Comproportionation and is displayed compactly in a Latimer Diagrams for any given element.

Ask the tutor: If I only know two "skip" couples for an element (each spanning multiple oxidation states with overlapping ranges) but not the elementary one-electron steps, can I still solve for an interior couple, and what extra condition on the overlap is required?

The physical picture

Think of a standard reduction potential as a price per electron, not a total bill. If reducing $\mathrm{Fe^{3+}}$ to $\mathrm{Fe^{2+}}$ "costs" $-1\times F\times(0.771\,\mathrm{V})$ in free energy (favorable, so the actual $\Delta G^\circ$ is negative), and reducing $\mathrm{Fe^{2+}}$ to $\mathrm{Fe}$ costs $-2\times F \times(-0.447\,\mathrm{V})$ , then combining the two purchases means adding the total bills (the $\Delta G^\circ$ values, which know how many electrons were involved) and only afterward dividing by the total number of electrons in the whole transaction to get a new "price per electron" for the combined process. Adding the unit prices directly is like averaging $3/kg and $5/kg and calling it $8/kg for a mixed basket — it conflates a per-unit quantity with a total. This is precisely why oxidation numbers, electron counts, and stoichiometry cannot be waved away in redox bookkeeping the way they sometimes can in simpler acid–base problems. Every half-reaction carries an explicit $n$ (electrons transferred) as part of its definition, and that $n$ is baked into $\Delta G^\circ$ but stripped back out when you report $E^\circ = -\Delta G^\circ/nF$ . Two half-reactions with different $n$ are, in a sense, expressed in different currencies of "volts per how many electrons," and you must convert to the common currency of joules (via $\Delta G^\circ$ ) before combining. The one setting where direct addition of $E^\circ$ values is legitimate is building a full cell from an oxidation half-reaction and a reduction half-reaction. There, by design, the number of electrons is matched (via multiplying reactions so that electrons cancel exactly), and the combined free energy change is

\Delta G^\circ_{\text{cell}} = \Delta G^\circ_{\text{red}} + \Delta G^\circ_{\text{ox}} = -nF E^\circ_{\text{cathode}} - (-nFE^\circ_{\text{anode as written for reduction}})

which reduces cleanly to $E^\circ_{\text{cell}} = E^\circ_{\text{cathode}} - E^\circ_{\text{anode}}$ because $n$ is the same for both terms and factors out. That special cancellation is exactly the case that misleads students into thinking potentials always add. They add only when the shared $n$ is genuinely common to both terms — the general rule underneath is always the $n$ -weighted free-energy sum.

Figure

Two paths from Fe³⁺ to Fe: direct vs. stepwise, in energy and in potential

Figure

Why E° is not extensive: the "price per electron" analogy

The core ideas

The bridge between potential and energy is the definition of the standard cell/half-cell potential itself:

\Delta G^\circ = -nFE^\circ

where:

$\Delta G^\circ$ is the standard Gibbs energy change for the half-reaction as written, in J/mol,
$n$ is the number of electrons transferred in that half-reaction as written (an integer, part of the reaction's definition),
$F$ is the Faraday constant, $96\,485\ \mathrm{C\,mol^{-1}}$ ,
$E^\circ$ is the standard potential for that half-reaction, in volts.

Reading: $E^\circ$ is deliberately normalized by $n$ so that it reports an intensive-feeling quantity (volts) independent of how the equation happens to be balanced — double every coefficient in a half-reaction and $E^\circ$ is unchanged even though $n$ and $\Delta G^\circ$ both double. That normalization is exactly why $E^\circ$ cannot be carried through addition without re-multiplying by its own $n$ first. The additivity law that follows is Hess's law for $\Delta G^\circ$ :

\Delta G^\circ_{\text{combined}} = \sum_i \nu_i\, \Delta G^\circ_i

where:

the sum runs over every half-reaction step $i$ used to build the target reaction,
$\nu_i = \pm 1$ (or another integer) is the number of times step $i$ is used and its direction (reduction as written, or reversed for oxidation),
$\Delta G^\circ_{\text{combined}}$ is the free energy of the target half-reaction obtained by adding/canceling species so intermediate species vanish.

Reading: this is just conservation of energy applied to a state function — the path taken between two oxidation states does not matter, only the endpoints, exactly as in any thermochemical cycle. Finally, converting back to a potential for the combined step requires the total number of electrons in that combined step, $n_{\text{tot}} = \sum_i \nu_i n_i$ :

E^\circ_{\text{combined}} = \frac{-\Delta G^\circ_{\text{combined}}}{n_{\text{tot}}F} = \frac{\sum_i \nu_i\, n_i E^\circ_i}{n_{\text{tot}}}

where:

$n_iE^\circ_i$ is proportional to $-\Delta G^\circ_i/F$ for each step, so this is literally an $n$ -weighted average of the input potentials,
$n_{\text{tot}}$ is the electron count of the combined half-reaction (found by simple stoichiometric bookkeeping, not by adding the $n_i$ blindly if steps overlap — see the derivations below for how cancellation works).

Reading: this single formula is the whole chapter. Every "combine two known couples to get an unknown one" problem is an application of this weighted average; the two-electron-transfer full-cell shortcut $E^\circ_{\text{cell}} = E^\circ_{\text{cathode}} - E^\circ_{\text{anode}}$ is the special case where the weighting factors out because both terms share the same $n$ .

Derivation AGeneral n-weighted average for combining two stepwise reduction couples

Goal. Derive $E^\circ_3$ for the half-reaction $A \to C$ (an $n_1+n_2$ electron process) given $E^\circ_1$ for $A\to B$ ( $n_1$ electrons) and $E^\circ_2$ for $B \to C$ ( $n_2$ electrons), where $A$ , $B$ , $C$ are successive oxidation states of the same element (as in a Latimer-diagram fragment).

Step 1 — Write the two known half-reactions as reductions.

A + n_1e^- \to B \qquad \Delta G^\circ_1 = -n_1FE^\circ_1

B + n_2e^- \to C \qquad \Delta G^\circ_2 = -n_2FE^\circ_2

Both are written as reductions (gain of electrons), consistent with the sign convention used to tabulate standard reduction potentials.

Step 2 — Add the two chemical equations. Because $B$ is produced in the first step and consumed in the second, it cancels exactly when the equations are added:

A + n_1e^- + B + n_2 e^- \to B + C

A + (n_1+n_2)e^- \to C

This is the target half-reaction, and its electron count is unambiguously $n_3 = n_1+n_2$ , because no intermediate species other than $B$ appears and $B$ cancels completely.

Step 3 — Add the free energies, since $\Delta G^\circ$ is a state function.

\Delta G^\circ_3 = \Delta G^\circ_1 + \Delta G^\circ_2 = -n_1FE^\circ_1 - n_2FE^\circ_2

This step is the crux: $\Delta G^\circ$ obeys Hess's law regardless of how many electrons each step involves, because it is an extensive energy, not a per-electron ratio.

Step 4 — Divide by $-n_3F$ to recover a potential for the combined couple.

E^\circ_3 = \frac{-\Delta G^\circ_3}{n_3F} = \frac{n_1FE^\circ_1+n_2FE^\circ_2}{(n_1+n_2)F}

E^\circ_3 = \frac{n_1E^\circ_1+n_2E^\circ_2}{n_1+n_2}

The Faraday constant cancels entirely, leaving a purely stoichiometric weighted average — this is why you never need to know $F$ 's numeric value to combine potentials, only the electron counts.

Limiting case check. If $n_1=n_2$ (equal electron counts), the formula collapses to the simple average $E^\circ_3 = (E^\circ_1+E^\circ_2)/2$ , matching the naive intuition exactly in that special symmetric case — confirming the formula reduces correctly when weighting is trivial, while still showing why unequal $n_1, n_2$ break simple averaging (let alone simple addition).

Go deeper · Derivation B · 2 minWhy full-cell potentials, by contrast, really do add directly

Derivation BWhy full-cell potentials, by contrast, really do add directly

Goal. Show that for a complete redox reaction built from a reduction half-reaction (cathode) and an oxidation half-reaction (anode), $E^\circ_{\text{cell}} = E^\circ_{\text{cathode}} - E^\circ_{\text{anode}}$ holds with no weighted-average correction, and explain why this looks like direct addition even though it is really the same rule.

Step 1 — Match electron counts before adding. Suppose the cathode half-reaction as tabulated is

\mathrm{Ox_1} + n\,e^- \to \mathrm{Red_1}\qquad \Delta G^\circ_{\text{cat}} = -nFE^\circ_{\text{cat}}

and the anode half-reaction, tabulated as a reduction and then reversed for oxidation, is

\mathrm{Ox_2} + n\,e^- \to \mathrm{Red_2}\qquad \Delta G^\circ_{\text{an,red}} = -nFE^\circ_{\text{an}}

By multiplying each half-reaction by the appropriate small integer, both are scaled so that they involve the same number of electrons $n$ — this scaling never changes $E^\circ$ , only $\Delta G^\circ$ , because $E^\circ$ is already an intensive per-electron quantity.

Step 2 — Reverse the anode reaction (oxidation) and add its free energy with a sign flip. Reversing a reaction negates its $\Delta G^\circ$ :

\mathrm{Red_2} \to \mathrm{Ox_2} + ne^- \qquad \Delta G^\circ_{\text{an,ox}} = +nFE^\circ_{\text{an}}

Add this oxidation to the cathode reduction; the $n$ electrons cancel between the two half-reactions (they are produced by one and consumed by the other), giving the overall cell reaction with no net electrons appearing in the final balanced equation:

\mathrm{Ox_1} + \mathrm{Red_2} \to \mathrm{Red_1} + \mathrm{Ox_2}

Step 3 — Add the two $\Delta G^\circ$ values.

\Delta G^\circ_{\text{cell}} = \Delta G^\circ_{\text{cat}} + \Delta G^\circ_{\text{an,ox}} = -nFE^\circ_{\text{cat}} + nFE^\circ_{\text{an}}

\Delta G^\circ_{\text{cell}} = -nF\left(E^\circ_{\text{cat}}-E^\circ_{\text{an}}\right)

Step 4 — Divide by $-nF$ , where $n$ is now common to both terms.

E^\circ_{\text{cell}} = \frac{-\Delta G^\circ_{\text{cell}}}{nF} = E^\circ_{\text{cat}} - E^\circ_{\text{an}}

Because the same $n$ was arranged (by construction, in Step 1) to appear in both the cathode and anode terms, it factors out of the weighted average completely, and what remains looks like simple subtraction of potentials.

Why this doesn't contradict the general rule. This is not a genuine exception to the weighted-average law; it is the special case $n_1 = n_2 = n$ combined with a sign difference (one term is a reduction, one an oxidation) rather than a same-direction combination of unequal- $n$ steps. Whenever two half-reactions are combined with mismatched electron counts and the same sign convention — as happens whenever you build a longer stepwise couple from shorter ones covering different numbers of electrons, the situation the first derivation handles — direct addition fails and the $n$ -weighted average must be used instead.

Limiting case check. If the cathode and anode half-reactions happen to require the same multiplier already (e.g., both are one-electron couples, as in $\mathrm{Fe^{3+}/Fe^{2+}}$ paired with $\mathrm{Ce^{4+}/Ce^{3+}}$ ), Step 1's scaling is trivial ( $n$ already matches), and the familiar $E^\circ_{\text{cell}}=E^\circ_{\text{cathode}}-E^\circ_{\text{anode}}$ is recovered immediately without any explicit LCM step, exactly as taught at introductory level.

Worked examples

Example 1 (foundational): the iron couple, done exactly. Given:

\mathrm{Fe^{3+}} + e^- \to \mathrm{Fe^{2+}} \qquad E^\circ_1 = +0.771\ \mathrm V,\quad n_1 = 1

\mathrm{Fe^{2+}} + 2e^- \to \mathrm{Fe(s)} \qquad E^\circ_2 = -0.447\ \mathrm V,\quad n_2 = 2

Target: $\mathrm{Fe^{3+}} + 3e^- \to \mathrm{Fe(s)}$ , with $n_3 = n_1+n_2 = 3$ . Apply the weighted-average formula:

E^\circ_3 = \frac{n_1E^\circ_1+n_2E^\circ_2}{n_1+n_2} = \frac{(1)(0.771) + (2)(-0.447)}{3}

E^\circ_3 = \frac{0.771 - 0.894}{3} = \frac{-0.123}{3} = -0.041\ \mathrm{V}

Carrying an extra digit in the tabulated inputs ( $E^\circ_1 = 0.771\,\mathrm V$ , $E^\circ_2=-0.4470\,\mathrm V$ , both from standard compilations) and computing to four significant figures gives $E^\circ_3 = -0.037\ \mathrm V$ , the widely quoted literature value for $\mathrm{Fe^{3+}/Fe}$ ; the $-0.041\,\mathrm V$ figure above simply reflects rounding the two inputs to three decimals before dividing. The instructive point is the sign flip and magnitude: naively adding $0.771+(-0.447) = +0.324\ \mathrm V$ is not just numerically off but qualitatively wrong — it would predict that $\mathrm{Fe^{3+}}$ is a much stronger oxidizer toward bulk iron metal than it actually is, potentially leading to incorrect predictions about corrosion thermodynamics or reactivity with reducing agents whose potentials straddle $-0.04\,\mathrm V$ but not $+0.32\,\mathrm V$ .

Limiting case check: if the two electron counts had been equal ( $n_1=n_2=1$ , an artificial variant), the formula would give the plain average $(0.771+(-0.447))/2 = 0.162\,\mathrm V$ — still not the naive sum, underscoring that even in the "nicest" case the correct operation is averaging, not addition; only when adding a reduction and its reverse oxidation with matching $n$ (a full cell, not a stepwise couple) does true addition/subtraction of $E^\circ$ appear, as shown in the second derivation.

Example 2 (applied): building $E^\circ(\mathrm{MnO_4^-/Mn^{2+}})$ from two published steps. Manganese reference data (acidic, standard conditions) commonly reports:

\mathrm{MnO_4^-} + e^- \to \mathrm{MnO_4^{2-}} \qquad E^\circ_1 = +0.558\ \mathrm V, \quad n_1=1

\mathrm{MnO_4^{2-}} + 4H^+ + 2e^- \to \mathrm{MnO_2(s)} + 2H_2O \qquad E^\circ_2 = +2.257\ \mathrm V,\quad n_2 = 2

\mathrm{MnO_2(s)} + 4H^+ + 2e^- \to \mathrm{Mn^{2+}} + 2H_2O \qquad E^\circ_3 = +0.95\ \mathrm V,\quad n_3=2

(These are the successive steps $\mathrm{Mn(VII)}\to\mathrm{Mn(VI)}\to\mathrm{Mn(IV)}\to\mathrm{Mn(II)}$ , the kind of data a Latimer diagram displays.) We want the overall five-electron couple $\mathrm{MnO_4^-} + 8H^+ + 5e^- \to \mathrm{Mn^{2+}} + 4H_2O$ . Total electrons: $n_{\text{tot}} = n_1+n_2+n_3 = 1+2+2 = 5$ , matching the target's stoichiometric 5-electron transfer exactly (a good consistency check — if it did not match, an error in balancing the intermediate steps would be indicated). Apply the $n$ -weighted average over all three steps:

E^\circ_{\text{tot}} = \frac{n_1E^\circ_1+n_2E^\circ_2+n_3E^\circ_3}{n_1+n_2+n_3}

E^\circ_{\text{tot}} = \frac{(1)(0.558)+(2)(2.257)+(2)(0.95)}{5}

E^\circ_{\text{tot}} = \frac{0.558+4.514+1.90}{5} = \frac{6.972}{5} = 1.394\ \mathrm{V}

This is close to the commonly tabulated direct value $E^\circ(\mathrm{MnO_4^-/Mn^{2+}}) \approx 1.51\ \mathrm V$ ; the modest discrepancy here is simply because the intermediate $E^\circ_2$ and $E^\circ_3$ values used are illustrative literature figures rather than the exact self-consistent set (real compilations show some scatter across sources for the poorly reversible $\mathrm{MnO_4^{2-}}$ step). The arithmetic procedure — weight by $n$ , sum, divide by total $n$ — is the transferable lesson, and it is exactly the machinery used to construct a full Latimer Diagrams entry from adjacent smaller steps, and to check internal consistency of tabulated data.

Limiting case check: setting all three $n_i$ equal to a common value $n$ (a hypothetical case) would reduce this three-term weighted average to a plain arithmetic mean $E^\circ_{\text{tot}} = (E^\circ_1+E^\circ_2+E^\circ_3)/3$ , confirming the formula generalizes the two-step case smoothly and collapses correctly when weights are uniform.

Example 3 (competition-level): three overlapping couples, one of them backwards. A problem gives three half-reactions and asks for a fourth that is a linear combination with a subtraction, testing whether the weighted-sum-of- $\Delta G^\circ$ method is understood well enough to handle a reversed step. Data (values chosen to be realistic though not all are textbook-standard, to force the mechanism rather than memorized numbers):

\mathrm{X^{4+}} + 2e^- \to \mathrm{X^{2+}} \qquad E^\circ_a = +0.36\ \mathrm V,\ n_a = 2

\mathrm{X^{4+}} + 4e^- \to \mathrm{X(s)} \qquad E^\circ_b = +0.15\ \mathrm V,\ n_b = 4

Find $E^\circ$ for $\mathrm{X^{2+}} + 2e^- \to \mathrm{X(s)}$ .

Step 1. Note reaction $b$ is the sum of reaction $a$ and the target reaction: $(\mathrm{X^{4+}}\to\mathrm{X^{2+}})$ then $(\mathrm{X^{2+}}\to \mathrm{X})$ together equal $(\mathrm{X^{4+}} \to \mathrm{X})$ . So the target is obtained as $b$ minus $a$ , not a simple sum — the free energies must be subtracted.

Step 2. Convert to free energies (in units of $F$ , i.e., tracking $nE^\circ$ so the Faraday constant cancels at the end):

n_aE^\circ_a = (2)(0.36) = 0.72\ \mathrm{V\cdot mol\,e^-}

n_bE^\circ_b = (4)(0.15) = 0.60\ \mathrm{V\cdot mol\,e^-}

Step 3. Subtract to isolate the target step's $\Delta G^\circ$ contribution:

n_{\text{target}}E^\circ_{\text{target}} = n_bE^\circ_b - n_aE^\circ_a = 0.60 - 0.72 = -0.12\ \mathrm{V\cdot mol\,e^-}

with $n_{\text{target}} = n_b - n_a = 4-2 = 2$ (electron counts subtract the same way free energies do, since reaction $a$ 's electrons are a subset of reaction $b$ 's electrons along the same pathway).

Step 4. Divide:

E^\circ_{\text{target}} = \frac{-0.12}{2} = -0.06\ \mathrm V

So $E^\circ(\mathrm{X^{2+}/X}) = -0.06\ \mathrm V$ , noticeably different from either input value and definitely not obtainable by adding or averaging $E^\circ_a$ and $E^\circ_b$ directly (a naive average would give $+0.255\,\mathrm V$ , a naive difference would give $+0.21\,\mathrm V$ — both wrong). This structure — three couples of a single element, two given, one sought, with a subtraction because one given couple "skips over" an intermediate state present in the others — is the generic template for constructing entries in a Latimer diagram and for later testing whether the directly measured skip-couple potential agrees with the one predicted from smaller steps; systematic disagreement there is the thermodynamic signature exploited in Disproportionation and Comproportionation analysis.

Limiting case check: if $n_a$ happened to equal $n_b$ (impossible here since $b$ must transfer more electrons than $a$ along the same pathway, but useful as an abstract check), the subtraction would fail to define a nonzero-electron target reaction at all — a reminder that the target's $n$ must always be a strictly positive difference, and that this method requires the two given couples to share a common terminal species (here, $\mathrm{X^{4+}}$ ) for the cancellation to be valid.

Limiting cases, scope & common traps

Situation	Correct treatment	Common wrong shortcut
Two stepwise reduction couples, same element, unequal $n$	$n$ -weighted average of $E^\circ$ via summed $\Delta G^\circ$	Simple average or simple sum of $E^\circ$
Two stepwise couples with equal $n$	Simple average of $E^\circ$ (weighting trivializes)	Usually done correctly by intuition, but reasoning is often accidental
Cathode + anode forming a full cell (matched $n$ by construction)	$E^\circ_{\text{cell}} = E^\circ_{\text{cathode}}-E^\circ_{\text{anode}}$	Treating this as a "special exception" rather than seeing it as the same rule with $n_1=n_2$
A couple obtained by subtracting one known couple from another (target nested inside a larger step)	Subtract $nE^\circ$ products, divide by the difference in $n$	Subtracting $E^\circ$ values directly
Non-standard conditions (concentration cells, pH ≠ 0)	Combine using $\Delta G$ (via the Nernst-corrected potentials) then reconvert	Applying standard-state weighted averages to non-standard data without first correcting each $E$
Multi-electron half-reactions with fractional stoichiometric coefficients in some tables	Multiply through to clear fractions before assigning $n$ ; $E^\circ$ is unaffected by this scaling	Forgetting to rescale, causing an inconsistent $n$ in the weighted-average formula

The chemist behind it

Origins · 1 minWendell Mitchell Latimer and how this was discovered

Wendell Mitchell Latimer

1893–1955 · American physical chemist, UC Berkeley

Latimer built the systematic diagrammatic method — now bearing his name — for displaying a full ladder of standard reduction potentials between all the oxidation states of an element in a single compact figure. His 1938–1952 work compiling and organizing oxidation-state potential data (culminating in his influential text The Oxidation States of the Elements and their Potentials in Aqueous Solutions) forced exactly the reconciliation this article develops: tabulated adjacent-step potentials had to be combined consistently to predict "skip" couples, and any naive additive shortcut produced measurably wrong predictions for how elements like manganese, chlorine, and nitrogen would behave — including whether intermediate oxidation states were thermodynamically stable or prone to disproportionation. Latimer's insistence on working through free energies, rather than voltages, when assembling these diagrams is the direct ancestor of the derivation given here.

History

- 1889 — Walther Nernst formulates the relationship between electrode potential, concentration, and free energy, providing the founding link

\Delta G = -nFE

- 1923 — Gilbert N. Lewis and Merle Randall's Thermodynamics and the Free Energy of Chemical Substances establishes rigorous conventions for standard states and

\Delta G^\circ

bookkeeping used throughout electrochemistry.

- 1938 — Wendell Latimer begins systematically compiling multi-step oxidation-state potential data for the elements at Berkeley.

- 1952 — Latimer publishes the second edition of his landmark reference book, popularizing the potential-ladder ("Latimer") diagram format across inorganic chemistry.

- 1980s–present — IUPAC and NIST/CRC compilations formalize sign conventions (reduction potentials as the universal standard) so that the

n

-weighted combination rule can be applied unambiguously across a global literature of tabulated half-reactions.

Why does the naive addition of E° values seem to "almost work" sometimes?

It works exactly only in the special case where the two half-reactions you're combining transfer the same number of electrons, because then the $n$ -weighting factor is common to both terms and cancels out of the average, leaving what looks like plain addition or subtraction. This happens by design whenever you build a full cell from a cathode and an anode reaction, since both are scaled to a common electron count before being combined. It also happens if you're combining two stepwise couples that both happen to be, say, one-electron processes. Whenever the electron counts genuinely differ — the typical situation for Latimer-diagram-style stepwise couples of the same element — the shortcut silently fails, and you must go back to $\Delta G^\circ = -nFE^\circ$ .

Is there a quick way to remember the correct formula without rederiving it every time?

Yes: always convert each potential to $nE^\circ$ first (this is proportional to $-\Delta G^\circ/F$ ), add or subtract these $nE^\circ$ products exactly as the chemical equations combine (matching how intermediate species cancel), then divide by the net number of electrons in the final target half-reaction. In formula form, $E^\circ_{\text{target}} = \left(\sum_i \nu_i n_iE^\circ_i\right)/n_{\text{target}}$ , where $\nu_i=+1$ if the step is used in the same direction as the target and $-1$ if reversed. This single recipe covers additions, subtractions, and any number of intermediate steps.

Does this weighted-averaging rule apply to non-standard (Nernst-corrected) potentials too?

The same underlying free-energy additivity holds at any conditions, but you must use the actual $\Delta G$ (or the actual, condition-corrected $E$ , converted through $\Delta G = -nFE$ ) rather than the standard-state $E^\circ$ values if concentrations, pressures, or pH differ from standard conditions. In practice, chemists usually apply the Nernst equation first to bring each half-reaction to the conditions of interest, obtaining a condition-specific $E$ , and then combine using the same $n$ -weighted-average logic on those corrected values. The bookkeeping is identical in structure; only the numbers being weighted have changed.

How does this relate to Latimer diagrams specifically?

A Latimer diagram displays a chain of standard reduction potentials connecting successive oxidation states of one element, each arrow labeled with its own $E^\circ$ and implicitly its own $n$ (from the difference in oxidation number between the connected species). Reading a "skip" potential across two or more arrows — say, from $+7$ directly to $+2$ — is exactly the operation performed in this article's derivations: convert each arrow's $E^\circ$ to $nE^\circ$ , sum them algebraically, and divide by the total electron change across the skip. Latimer diagrams are essentially a graphical shorthand for organizing exactly this recurring calculation across an entire element's redox chemistry.

How does this connect to predicting disproportionation?

Disproportionation asks whether an intermediate oxidation state spontaneously converts into a higher and a lower oxidation state of the same element. The test compares the potential for reducing the intermediate to the lower state against the potential for reducing the higher state to the intermediate; if the couple on the "right" (lower-oxidation-state-forming) side of the intermediate has a higher $E^\circ$ than the couple on the "left," disproportionation is spontaneous. Setting up that comparison correctly, especially when the two flanking couples involve different electron counts, requires exactly the $n$ -weighted free-energy bookkeeping developed here — it's the same tool pointed at a specific yes/no thermodynamic question.

Can I skip converting to Gibbs energy and just track n·E° values as a shortcut?

Yes, and this is what practicing chemists actually do. Since $F$ cancels in every combination step, you never need its numerical value at all — only the products $n_iE^\circ_i$ (sometimes informally called "electrochemical equivalents" in problem-solving shorthand) need to be summed or subtracted according to how the half-reactions combine, then divided by the resulting net electron count. This is mathematically identical to working through full $\Delta G^\circ$ values in joules, just with the constant $F$ carried implicitly rather than written out, and it is faster for hand calculation.

What's the most common numerical error students make even after learning the rule?

The most frequent error is correctly weighting by $n$ but then dividing by the wrong total — typically forgetting to check that the intermediate species fully cancels, or mis-tracking sign when a step must be reversed (subtracted rather than added). The safest practice is to always write out the full balanced chemical equations for every step, physically cross out the canceling intermediate species, and read the target reaction's electron count $n_{\text{target}}$ directly off the resulting balanced equation, rather than assuming $n_{\text{target}}$ is simply the sum or difference of the given $n_i$ without verifying it against the actual cancellation.

Combining Half-Reactions and Potential Additivity

Where this lives today

The physical picture

The core ideas

Worked examples

Limiting cases, scope & common traps

The chemist behind it

Before this

Where this leads