Iodometry and Iodimetry

A student weighs out a copper ore sample and dissolves it in acid. Excess potassium iodide is then added. The solution instantly darkens to a deep brown-black. Fifteen minutes later, a colorless thiosulfate solution drips from a burette. It makes that color fade, straw-yellow, then pale green. Then, the instant a spoonful of starch paste is added, a vivid blue appears. It vanishes in a single drop to leave a colorless endpoint. That endpoint is so sharp it looks rehearsed. This is iodometry, one of the oldest and still most trusted volumetric methods in analytical chemistry. The copper assay is essentially the same procedure used industrially to certify tons of cathode copper. The chemistry of iodine occupies a peculiar and useful middle ground. The iodine/iodide couple, $\mathrm{I_2 + 2e^- \rightleftharpoons 2I^-}$ , has a standard potential of only $E^\circ = +0.536$ V. This is mild enough that iodine is a gentle, selective oxidant. Iodide, in turn, is a gentle, selective reductant. This mildness is the method's superpower. Iodometric procedures can titrate a huge range of species. These include copper(II), dissolved oxygen, chlorine, peroxides, arsenite, ascorbic acid, and dozens more. They do this without the harsh side reactions that plague strong oxidants like permanganate or dichromate. Because iodine's potential sits so centrally, essentially every classical redox species can be pushed to react with iodide/iodine in one direction or the other. This makes this couple a universal shuttle for redox analysis. The subtlety that makes this chapter worth its own careful treatment is that there are two iodine methods, easily confused. Iodimetry is a direct titration. In it, standard $\mathrm{I_2}$ solution is the titrant added to a reducing analyte. Iodometry is an indirect titration. In it, an oxidizing analyte first liberates iodine from excess iodide. That liberated iodine is then titrated with standard thiosulfate. The two share reagents and an endpoint indicator (starch). But they differ completely in strategy, in which species is standardized, and in which errors dominate. Confusing them is the single most common student mistake in this corner of quantitative analysis. This article works through the equilibrium chemistry that makes both methods possible. It derives the stoichiometric bridge between an analyte's oxidizing power and the moles of thiosulfate consumed. It then works three realistic problems. These are a foundational iodimetric titration of arsenite, an applied indirect assay of copper ore, and an exam-level dissolved-oxygen (Winkler) determination. Each carries explicit uncertainty. Along the way we confront the real experimental demons. These include iodine's volatility, air oxidation of iodide in acid, and starch's temperamental behavior near the endpoint. We also address the near-universal practice of adding starch late rather than at the start.

Key takeaways

Iodimetry titrates a reducing analyte directly with standardized I₂ (I₂ is the titrant).
Iodometry titrates an oxidizing analyte indirectly: excess I⁻ is oxidized to I₂, and the I₂ is titrated with standardized Na₂S₂O₃.
The governing couple $\mathrm{I_2 + 2e^- \rightleftharpoons 2I^-}$ , $E^\circ = 0.536$ V, is mild and reversible — ideal for a sharp potentiometric or visual endpoint.
Starch forms a deep blue starch–I₃⁻ inclusion complex; it must be added near the endpoint, not at the start, or the complex traps iodine and biases the endpoint late.
Stoichiometry always runs through iodine: $n(\text{analyte}) \times (\text{electrons/analyte}) = n(\mathrm{I_2}) \times 2 = n(\mathrm{S_2O_3^{2-}})$ .
Major error sources are iodine volatility/sublimation loss, air oxidation of I⁻ in acidic solution, and thiosulfate decomposition by bacteria or acid — all correctable by technique, not by better glassware.
Key idea: iodine's mid-range, reversible potential lets it serve as a universal redox intermediary — direct titrant for reductants, or indirect carrier of oxidizing power onto thiosulfate — with starch giving one of the sharpest visual endpoints in all of titrimetry.

The core insight

Iodine is never the analyte's real target in iodometry — it is a stoichiometric messenger. Any oxidizing analyte that can liberate $\mathrm{I_2}$ from excess iodide converts an arbitrary redox titration into the same well-behaved thiosulfate titration, because $\mathrm{I_2}$ is always consumed by exactly $2e^-$ per mole regardless of what generated it.

Where this lives today

Iodometric back-titration remains the reference method for dissolved oxygen in environmental and wastewater science; the Winkler titration, though increasingly supplemented by optical/electrochemical DO probes for continuous field monitoring, is still the calibration standard against which those probes are checked, because its stoichiometry is exact and traceable. In industrial metallurgy, iodometric copper assay (Example 2's method, essentially unchanged since the nineteenth century) remains a standard reference procedure for certifying copper cathode purity and ore grade, valued because it needs no specialized instrumentation and achieves sub-0.1% relative precision in skilled hands. In pharmaceutical quality control, iodimetric titration directly assays reducing drug substances — ascorbic acid (vitamin C) tablets, for instance, are commonly assayed by direct titration with standard iodine, exploiting exactly the same Redox Titration Curves: The E-vs-Volume Plot logic used for any redox titration, with the sharp starch endpoint substituting for a potentiometric one. Chlorine and bleach potency in water treatment and industrial sanitation is measured iodometrically: hypochlorite liberates iodine from acidified KI, and thiosulfate titration gives "available chlorine," a standard specification on every bottle of commercial bleach. In clinical and food chemistry, iodometric methods determine peroxide values in fats and oils (rancidity testing) and quantify sulfite preservatives in wine and dried fruit via iodimetric titration. And as a teaching tool, the iodine/thiosulfate system remains the workhorse for introducing students to Redox Indicators and Potentiometric Endpoints, because starch's blue-to-colorless transition is visually unambiguous in a way few other redox indicators can match, and because the coupled equilibria (triiodide formation, CuI precipitation shifting formal potentials) are a compact, self-contained illustration of how Complexometric Titrations of Metal Ions-style coupled equilibria can rescue an otherwise unfavorable redox reaction.

Ask the tutor: How would the iodometric copper assay change if the ore also contained iron(III), another oxidant capable of liberating iodine from KI?

The physical picture

Iodine chemistry in aqueous solution is dominated by one inconvenient fact: elemental $\mathrm{I_2}$ is only sparingly soluble in water (about $1.3\times10^{-3}$ M at 25 °C) and is volatile, subliming steadily even from stoppered flasks. Both iodometric and iodimetric procedures dodge this problem the same way — by always keeping a large excess of iodide ion present. Iodide reacts with molecular iodine to form the triiodide complex,

\mathrm{I_2 + I^- \rightleftharpoons I_3^-}, \qquad K \approx 700 \text{ at } 25\,^\circ\mathrm{C}

and triiodide is far more soluble and far less volatile than $\mathrm{I_2}$ itself. Every "standard iodine solution" used in the laboratory is therefore not $\mathrm{I_2}(aq)$ but $\mathrm{I_3^-}(aq)$ , typically prepared by dissolving iodine in excess KI. For bookkeeping purposes chemists still write reactions with " $\mathrm{I_2}$ ", understanding that the actual species titrated is triiodide; the electron count is identical because iodide is a spectator in the redox sense. The second piece of the physical picture is the reversible, well-poised nature of the $\mathrm{I_2/I^-}$ couple itself. Its standard potential, $E^\circ = +0.536$ V, sits almost exactly in the middle of the practical range of aqueous redox chemistry — well below strong oxidants (permanganate at $+1.51$ V, dichromate at $+1.33$ V, cerium(IV) at $+1.44$ V/ $1.61$ V) but well above mild reductants (thiosulfate, arsenite, ascorbic acid). This means:

Any species with $E^\circ$ well above $0.536$ V (Cu²⁺/Cu⁺ in the presence of I⁻, at an effective potential raised by CuI precipitation; Cr₂O₇²⁻; MnO₄⁻; H₂O₂; Cl₂; IO₃⁻) can oxidize excess I⁻ to I₂ quantitatively — this is the iodometric direction.
Any species with $E^\circ$ well below $0.536$ V (As(III)→As(V), $\mathrm{S_2O_3^{2-}}$ , ascorbic acid, Sn²⁺, SO₂) can be oxidized quantitatively by standard $\mathrm{I_2}/\mathrm{I_3^-}$ — this is the iodimetric direction.

The third piece is the endpoint chemistry of starch. Amylose, the linear component of starch, coils into a helix with an internal channel roughly 13 Å across. Triiodide ions insert into this helical channel, and the resulting linear polyiodide chain shows an intense charge-transfer absorption giving the characteristic deep blue-black color, with $\lambda_{max} \approx 620$ nm. The complex is extraordinarily sensitive — visible at $\mathrm{I_2}$ concentrations as low as $10^{-5}$ M — which is exactly why the disappearance of blue color at the endpoint of a thiosulfate titration is so much sharper than watching pale-yellow triiodide fade to colorless by eye alone.

Figure

The iodine/iodide couple sits mid-scale between common oxidants and reductants, enabling both titration directions

Figure

Starch–triiodide endpoint: helix inclusion complex and the titration color sequence

The core ideas

The entire quantitative machinery of both methods reduces to tracking electrons through iodine as an intermediary, then applying the mass-balance definition of equivalents.

The iodine/triiodide couple and its Nernst behavior, exactly as used for potentiometric endpoints or for predicting whether a given analyte reaction goes essentially to completion:

E = E^\circ_{\mathrm{I_3^-/I^-}} - \frac{0.05916}{2}\log\frac{[\mathrm{I^-}]^3}{[\mathrm{I_3^-}]}, \qquad E^\circ_{\mathrm{I_3^-/I^-}} = +0.536\ \text{V at }25\,^\circ\mathrm{C}

where: $E$ is the couple's potential at the working concentrations, $E^\circ$ the standard (formal) potential of triiodide/iodide, and the log term reflects the three-iodide, one-electron-pair stoichiometry of $\mathrm{I_3^- + 2e^- \rightleftharpoons 3I^-}$ . Reading: because iodide appears cubed in the reaction quotient, a large excess of I⁻ (always present by design) pulls $E$ down and keeps the couple poised near $E^\circ$ , giving a reproducible, well-defined potential — exactly what is needed for a sharp equivalence-point break, whether followed visually with starch or with a platinum indicator electrode.

The stoichiometric bridge, which is the actual working equation used to compute every iodometric result:

n_e(\text{analyte}) = 2\,n(\mathrm{I_2}) = n(\mathrm{S_2O_3^{2-}})

where: $n_e(\text{analyte})$ is the total moles of electrons transferred by the analyte (moles of analyte × electrons per mole), $n(\mathrm{I_2})$ is moles of iodine liberated (equal to moles of triiodide formed), and $n(\mathrm{S_2O_3^{2-}})$ is moles of thiosulfate consumed in $\mathrm{2S_2O_3^{2-} + I_2 \rightarrow S_4O_6^{2-} + 2I^-}$ . Reading: the factor of 2 appears twice and cancels — one $\mathrm{I_2}$ molecule carries exactly two oxidizing "slots" (from the two iodine atoms each gaining one electron), and thiosulfate donates exactly one electron per molecule (oxidized to the S–S bonded tetrathionate dimer), so one mole of liberated iodine always requires exactly two moles of thiosulfate, regardless of what generated the iodine. This decoupling is what allows the same standardized thiosulfate solution to assay copper ore, bleach, peroxide, and dissolved oxygen with no change in the underlying arithmetic.

Derivation AFrom analyte oxidizing power to thiosulfate volume — the indirect (iodometric) pathway

Goal. Show explicitly how the mass of an oxidizing analyte (illustrated with copper ore, Cu²⁺) is converted into a volume of standardized thiosulfate, and why the apparent potential of the Cu²⁺/Cu⁺ couple is pushed high enough to oxidize iodide only because CuI(s) precipitates.

Step 1 — the raw electrode potentials look unfavorable. In free aqueous solution, $E^\circ(\mathrm{Cu^{2+}/Cu^+}) = +0.159$ V is well below $E^\circ(\mathrm{I_2/I^-}) = 0.536$ V. Naively, Cu²⁺ should not be able to oxidize I⁻ — the reaction $2\mathrm{Cu^{2+}} + 4\mathrm{I^-} \rightarrow 2\mathrm{CuI}(s) + \mathrm{I_2}$ looks thermodynamically forbidden by $\Delta E^\circ = 0.536-0.159 = 0.377$ V in the wrong direction.

Step 2 — CuI precipitation collapses [Cu⁺] and rescues the driving force. Copper(I) iodide is extremely insoluble, $K_{sp} \approx 1\times10^{-12}$ . Precipitation of Cu⁺ as CuI(s) drives the Cu²⁺/Cu⁺ half-reaction far to the right by mass action, which is equivalent to raising its effective (formal) potential. Combining the Nernst expression for Cu²⁺/Cu⁺ with the solubility equilibrium of CuI shows the effective potential for

\mathrm{Cu^{2+} + I^- + e^- \rightarrow CuI(s)}

is about $+0.86$ V — now comfortably above $0.536$ V, so the overall reaction runs essentially to completion, driven quantitatively to the right.

Step 3 — write the coupled reaction and its stoichiometry. The net analytical reaction is

2\mathrm{Cu^{2+}} + 4\mathrm{I^-} \rightarrow 2\mathrm{CuI}(s) + \mathrm{I_2}

so 2 mol Cu²⁺ liberate 1 mol I₂, i.e. each mole of Cu²⁺ transfers exactly 1 electron ( $\mathrm{Cu^{2+}\!+e^-\!\to Cu^+}$ ), consistent with $n_e = n(\mathrm{Cu^{2+}})$ .

Step 4 — titrate the liberated iodine with thiosulfate.

\mathrm{I_2 + 2S_2O_3^{2-} \rightarrow 2I^- + S_4O_6^{2-}}

Combining Steps 3–4: 2 mol Cu²⁺ → 1 mol I₂ → 2 mol S₂O₃²⁻, so

n(\mathrm{Cu^{2+}}) = n(\mathrm{S_2O_3^{2-}})

a clean 1:1 relationship between analyte and titrant despite the two-electron detour through iodine — because copper transfers 1 electron per atom and thiosulfate accepts 1 electron per molecule, the "×2" from I₂'s two iodine atoms cancels exactly.

Step 5 — convert to mass. For a sample containing mass $m$ of copper (molar mass $63.55\ \text{g mol}^{-1}$ ), titrated with thiosulfate of concentration $c_T$ to volume $V_T$ ,

m(\mathrm{Cu}) = c_T V_T \times 63.55\ \text{g mol}^{-1}

This is the working equation used in Worked Example 2 below.

Go deeper · Derivation B · 1 minIodimetric titration of arsenite — the direct pathway and its pH dependence

Derivation BIodimetric titration of arsenite — the direct pathway and its pH dependence

Goal. Derive the titration relationship for direct titration of As(III) with standard I₃⁻, and show why the reaction must be run in mildly alkaline (buffered, near pH 8) solution rather than in strong acid.

Step 1 — the half-reactions and their formal potentials. Arsenic(III)/(V) in aqueous solution:

\mathrm{H_3AsO_4 + 2H^+ + 2e^- \rightarrow HAsO_2 + 2H_2O}, \qquad E^\circ = +0.559\ \text{V (strongly pH-dependent, written for acidic form)}

The half-reaction explicitly contains $\mathrm{H^+}$ , so by the Nernst equation its potential falls as pH rises:

E = E^\circ - \frac{0.05916}{2}\log\frac{[\mathrm{HAsO_2}]}{[\mathrm{H_3AsO_4}][\mathrm{H^+}]^2}

Step 2 — raising pH lowers the As(V)/As(III) potential below that of I₂/I⁻. At pH ≈ 7–9 (as in a NaHCO₃-buffered solution) the As(V)/As(III) formal potential drops to roughly $+0.1$ to $+0.2$ V, well below the iodine couple's $0.536$ V, so the reaction

\mathrm{HAsO_2 + I_3^- + 2H_2O \rightarrow H_3AsO_4 + 3I^- + 2H^+}

proceeds quantitatively left-to-right. In strongly acidic solution, by contrast, the As(V)/As(III) potential rises above $0.536$ V, and the reaction would run in reverse (As(V) would oxidize I⁻) — this is in fact used deliberately in the reverse iodometric direction to standardize iodine solutions against pure As₂O₃.

Step 3 — electron count and stoichiometric ratio. Arsenic goes from +3 to +5, a two-electron change per As atom; iodine (as I₃⁻) accepts two electrons per formula unit ( $\mathrm{I_3^- + 2e^- \to 3I^-}$ ). Hence the mole ratio is exactly 1:1,

n(\mathrm{As(III)}) = n(\mathrm{I_2})

Step 4 — titration relationship. With standard iodine of concentration $c_I$ delivered to volume $V_{eq}$ at the equivalence point, and sample mass $m$ of As₂O₃ (molar mass $197.84\ \text{g mol}^{-1}$ , providing 2 mol As(III) per mole As₂O�3):

c_I = \frac{2m/197.84}{V_{eq}}

This is the standard method for assigning the exact concentration of an iodine solution using primary-standard As₂O₃, and by symmetry the same relation (solved for $m$ ) assays an unknown arsenite sample once $c_I$ is known.

Worked examples

Example 1 (foundational) — standardizing an iodine solution against As₂O₃, and titrating an arsenite unknown. A student dissolves $0.2015$ g of primary-standard As₂O₃ (as NaAsO₂ after basifying) in bicarbonate buffer at pH 8 and titrates with iodine solution of unknown concentration, using starch added near the endpoint. The titration requires $38.42$ mL to reach the permanent pale-blue endpoint. Moles of As₂O₃:

n(\mathrm{As_2O_3}) = \frac{0.2015\ \text{g}}{197.84\ \text{g mol}^{-1}} = 1.0184\times10^{-3}\ \text{mol}

Each mole of As₂O₃ furnishes two moles of As(III), each requiring one mole of I₂ (Step 3 above), so

n(\mathrm{I_2}) = 2\times1.0184\times10^{-3} = 2.0368\times10^{-3}\ \text{mol}

Concentration of the iodine solution:

c_I = \frac{2.0368\times10^{-3}\ \text{mol}}{0.03842\ \text{L}} = 0.05302\ \text{M}

(four significant figures are justified since the mass was read to $\pm0.0001$ g and volume to $\pm0.02$ mL, giving relative uncertainties of about $0.05\%$ and $0.05\%$ respectively — combined uncertainty roughly $\pm0.00004$ M, i.e. $c_I = 0.05302 \pm 0.00004$ M). Now this standardized iodine is used to titrate 25.00 mL of an unknown sodium arsenite solution, requiring 22.16 mL to reach the endpoint:

n(\mathrm{As(III)}) = c_I V_I = 0.05302\ \text{M} \times 0.02216\ \text{L} = 1.1749\times10^{-3}\ \text{mol}

[\mathrm{As(III)}] = \frac{1.1749\times10^{-3}\ \text{mol}}{0.02500\ \text{L}} = 0.04700\ \text{M}

Limiting-case check: if the unknown volume equaled the standardization volume exactly ( $V_I = 38.42$ mL for the same 25.00 mL aliquot), the concentration would come out identical to the As₂O₃-implied arsenite concentration used to standardize — an internal self-consistency check confirming the 1:1 As(III):I₂ stoichiometry rather than an accidental 2:1 or 1:2 ratio.

Example 2 (applied) — indirect iodometric assay of copper in an ore. A 1.000 g ore sample assaying for copper is dissolved in nitric acid, the excess acid neutralized, and the solution treated with excess KI. Copper(II) oxidizes iodide (Step 2–3 of the derivation above), precipitating CuI(s) and liberating I₂/I₃⁻, which turns the solution dark brown. This is titrated with $0.1024$ M sodium thiosulfate, requiring $27.65$ mL to reach the endpoint (starch added when the solution has faded to pale straw yellow, giving a sharp blue-to-colorless transition). Moles of thiosulfate delivered:

n(\mathrm{S_2O_3^{2-}}) = 0.1024\ \text{M} \times 0.02765\ \text{L} = 2.8314\times10^{-3}\ \text{mol}

By the 1:1 relationship derived above ( $n(\mathrm{Cu^{2+}}) = n(\mathrm{S_2O_3^{2-}})$ ):

n(\mathrm{Cu}) = 2.8314\times10^{-3}\ \text{mol}

Mass of copper:

m(\mathrm{Cu}) = 2.8314\times10^{-3}\ \text{mol} \times 63.55\ \text{g mol}^{-1} = 0.17994\ \text{g}

Percent copper in the ore:

\%\mathrm{Cu} = \frac{0.17994\ \text{g}}{1.000\ \text{g}} \times 100\% = 17.99\%

With a burette reading uncertainty of $\pm0.02$ mL (relative $0.07\%$ ) and thiosulfate concentration known to $\pm0.0002$ M (relative $0.2\%$ ), the combined relative uncertainty is about $0.2\%$ , giving $\%\mathrm{Cu} = 17.99 \pm 0.04\%$ . A crucial practical detail: CuI(s) tends to adsorb some I₂/I₃⁻ on its surface, which would make the titration endpoint appear early (understating copper) if untreated. The standard remedy is to add potassium thiocyanate (KSCN) near the endpoint, which converts the CuI coating to the less adsorptive CuSCN and releases the trapped iodine, restoring accuracy. If this step is omitted, results run systematically low by several tenths of a percent — a real, documented negative bias, not a random scatter.

Example 3 (exam/competition level) — Winkler titration for dissolved oxygen. A 300.0 mL BOD bottle of lake water is "fixed" in the field with $\mathrm{Mn^{2+}}$ and strong alkaline KI. Dissolved oxygen oxidizes freshly precipitated $\mathrm{Mn(OH)_2}$ to $\mathrm{MnO(OH)_2}$ (a manganese(III/IV) oxyhydroxide), quantitatively fixing the oxygen content as an equivalent amount of higher-valent manganese:

2\mathrm{Mn(OH)_2} + \mathrm{O_2} \rightarrow 2\mathrm{MnO(OH)_2}

so $n(\mathrm{O_2}) = 2\,n(\mathrm{Mn^{3+,eq}})$ ... but more directly, each mole of $\mathrm{O_2}$ (4-electron oxidant, $\mathrm{O_2 + 4e^- + 2H_2O \to 4OH^-}$ ) drives exactly the same electron equivalence in manganese. Upon acidification the manganese oxidation product oxidizes iodide back to iodine:

\mathrm{MnO(OH)_2 + 2I^- + 4H^+ \rightarrow Mn^{2+} + I_2 + 3H_2O}

Tracking electrons directly (the safest route): each $\mathrm{O_2}$ molecule accepts 4 electrons on reduction to water/hydroxide; that oxidizing power is stored quantitatively in the Mn(III/IV) precipitate and released stoichiometrically to iodide, so

n(\mathrm{O_2}) \times 4 = n(\mathrm{I_2}) \times 2 = n(\mathrm{S_2O_3^{2-}})

n(\mathrm{O_2}) = \frac{n(\mathrm{S_2O_3^{2-}})}{4}

A $203.0$ mL aliquot of the fixed, acidified sample (chosen so that the dilution from reagent addition is already accounted for by standard Winkler bottle geometry) is titrated with $0.01000$ M thiosulfate to a starch endpoint at $V_T = 4.20$ mL.

n(\mathrm{S_2O_3^{2-}}) = 0.01000\ \text{M} \times 0.00420\ \text{L} = 4.20\times10^{-5}\ \text{mol}

n(\mathrm{O_2}) = \frac{4.20\times10^{-5}}{4} = 1.050\times10^{-5}\ \text{mol}

Converting to the familiar dissolved-oxygen units (mg O₂ per liter of original lake water, molar mass O₂ = 32.00 g/mol, and correcting for the standard Winkler dilution factor where the 203.0 mL titrated aliquot represents an effective original sample volume of 200.0 mL after reagent-volume correction):

[\mathrm{O_2}] = \frac{1.050\times10^{-5}\ \text{mol} \times 32.00\ \text{g mol}^{-1}}{0.2000\ \text{L}} \times 1000\ \text{mg/g} = 1.680\ \text{mg L}^{-1}

This value, 1.68 mg/L, is low enough to flag a genuinely oxygen-stressed water body (typical healthy surface waters run 6–9 mg/L), illustrating why field titrations must be fixed immediately on sampling — dissolved oxygen itself would otherwise change during transport, corrupting the very quantity being measured. Limiting-case check: if the water were fully oxygen-saturated at a value near 9 mg/L, the required titrant volume would scale linearly (about 5.4× larger, near 22.5 mL for the same aliquot and concentration), confirming the direct proportionality $n(\mathrm{O_2}) \propto V_T$ built into the derivation — there is no hidden nonlinearity, only a straightforward stoichiometric ladder from O₂ through Mn to I₂ to thiosulfate.

Limiting cases, scope & common traps

Situation	Behavior	Consequence for the analyst
Analyte $E^\circ \gg 0.536$ V (e.g. $\mathrm{MnO_4^-}$ , $\mathrm{Cr_2O_7^{2-}}$ , $\mathrm{Cl_2}$ )	Oxidizes excess I⁻ quantitatively	Use iodometric (indirect) route: liberate I₂, titrate with thiosulfate
Analyte $E^\circ \ll 0.536$ V (e.g. As(III) at high pH, ascorbic acid, Sn²⁺)	Reduced quantitatively by I₃⁻	Use iodimetric (direct) route: titrate analyte with standard I₃⁻
Analyte $E^\circ$ close to 0.536 V (e.g. Cu²⁺ in free solution, $E^\circ=0.159$ V)	Reaction thermodynamically marginal unless assisted	Needs a coupled equilibrium (CuI precipitation) to shift the effective potential — check formal, not textbook, potentials
Strongly acidic medium for As(III)/As(V)	As(V)/As(III) potential rises above 0.536 V	Reaction reverses — used deliberately to standardize I₂ against As₂O₃ only at controlled pH
Excess starch added at the very start of titration	Starch–I₃⁻ complex forms early and iodine adsorbs strongly into the helix	Endpoint fade is sluggish and can be read early or overshot; standard practice is to add starch only when color has faded to pale straw yellow
Prolonged standing of acidified iodide solution before titration	Air (O₂) slowly oxidizes excess I⁻ to I₂ in acid, especially in sunlight	Positive error (extra apparent I₂); titrate promptly, keep flask stoppered and shaded
Thiosulfate solution aged for weeks, unpreserved	Bacterial decomposition and slow reaction with dissolved CO₂/acid lowers effective concentration	Standardize thiosulfate freshly (against KIO₃ or K₂Cr₂O₇/KI) on day of use for best work; add a trace of Na₂CO₃ or HgI₂ as preservative for longer-term stock
Very dilute iodine/thiosulfate near the endpoint	Starch–I₃⁻ complex can decompose slowly (retrograde) at low [I₂], especially warm solutions	Endpoint may fade back in ("false" endpoint;) titrate at room temperature, not hot, and read the first permanent disappearance of blue

Common trapMisconception: "iodometry and iodimetry are just two names for the same titration"

They are stoichiometrically related but strategically opposite. In iodimetry, standardized $\mathrm{I_2}$ / $\mathrm{I_3^-}$ is the titrant in the burette, added directly to a reducing analyte — the analyte's concentration is read straight from the iodine volume. In iodometry, the analyte itself never touches the burette; it first generates its own stoichiometric equivalent of iodine from excess iodide, and that iodine is titrated with thiosulfate — the analyte's concentration is inferred two steps removed, through the amount of thiosulfate consumed. Mixing up which species is titrant and which is generated leads to an immediate factor-of-something error in every calculation, and confuses which coefficient goes in the numerator when converting between moles.

Common trapMisconception: "starch should be added at the start of a thiosulfate titration, just like any other visual indicator"

Starch is unusual among indicators: it must be added late, once the triiodide concentration has already fallen to a pale straw-yellow, typically in the last 1–2 mL before the anticipated endpoint. Added early, at high $\mathrm{I_3^-}$ concentration, the polyiodide chains bind so tightly within the amylose helix that some iodine becomes kinetically trapped and is released back to the bulk solution only slowly — the endpoint becomes sluggish, over-titrated, and poorly reproducible. Waiting until the bulk of the iodine has already reacted avoids saturating the helix and gives the crisp one-drop transition that makes iodometry a byword for sharp endpoints.

The chemist behind it

Iodine itself was discovered in 1811 by Bernard Courtois, a French saltpeter manufacturer who noticed violet vapors rising from seaweed ash treated with sulfuric acid — an accident of industrial chemistry rather than a planned experiment. But the titrimetric use of iodine, and specifically the pairing with thiosulfate and starch that defines modern iodometry, is largely credited to the mid-nineteenth-century development of volumetric analysis by chemists working in the tradition established by Jean-Baptiste Dumas and Karl Friedrich Mohr. It was Robert Wilhelm Bunsen who, in 1853, formalized the iodine/thiosulfate titration as a general analytical tool, recognizing that iodine's mild, reversible potential made it an ideal stoichiometric intermediary for oxidants that were otherwise awkward to titrate directly. Bunsen's insight — that you did not need a direct indicator reaction for every oxidant if you could always route the electron count through iodine — is precisely the logical structure preserved in every iodometric procedure used today, from copper assay to the Winkler dissolved-oxygen method later built on the same principle by Lyon Playfair Winkler in 1888.

Origins · 1 minRobert Wilhelm Bunsen and how this was discovered

Robert Wilhelm Bunsen

1811–1899 · German chemistry, mid-19th century experimental chemistry and analytical method development

Bunsen is best remembered publicly for the burner that bears his name and for co-discovering spectroscopy with Gustav Kirchhoff, but his laboratory output in quantitative inorganic and gas analysis was equally foundational to the discipline of analytical chemistry. Working at Heidelberg, he systematized iodometric titration as a general-purpose oxidation-reduction method, formalizing the use of standardized sodium thiosulfate against liberated iodine — a technique so robust that it remains, essentially unchanged in its core chemistry, the basis of modern copper, chlorine, and dissolved-oxygen assays across industrial and environmental laboratories worldwide.

History

- 1811 — Bernard Courtois discovers elemental iodine in seaweed ash extract.

- 1826 — Jean-Baptiste Dumas develops early quantitative iodine-based methods within the broader rise of volumetric analysis.

- 1853 — Robert Bunsen formalizes iodometric titration using thiosulfate as the standard titrant against liberated iodine.

- 1888 — Lyon Playfair Winkler adapts iodometric back-titration to the determination of dissolved oxygen in water, the Winkler method still taught and used today.

- Early 20th century — starch indicator standardized as the near-universal endpoint signal for iodometric and iodimetric titrations in pharmacopeial and industrial assay methods.

Why is iodine always standardized and stored as a solution in excess KI rather than as pure I₂ in water?

Elemental iodine is only sparingly soluble in water (about $1.3\times10^{-3}$ M at 25 °C) and volatilizes steadily even from a stoppered flask, making a stable, accurately known "pure I₂(aq)" standard essentially impossible to maintain. Dissolving iodine in a solution containing excess iodide converts nearly all of it into the far more soluble, far less volatile triiodide complex, $\mathrm{I_2 + I^- \rightleftharpoons I_3^-}$ ( $K \approx 700$ ). Because the redox chemistry of triiodide and molecular iodine is identical in electron count — iodide is a non-participating spectator in the electron transfer — chemists write " $\mathrm{I_2}$ " throughout stoichiometric equations for convenience while actually working with $\mathrm{I_3^-}$ in solution. This trick is what makes standard iodine solutions practically usable at all.

What goes wrong if the iodometric copper titration is done without adding KSCN near the endpoint?

Freshly precipitated CuI(s) has a large surface area and adsorbs some of the liberated iodine (as triiodide) onto its surface, effectively hiding it from the titrant. If this adsorbed iodine is never released, the thiosulfate titration reaches its visual endpoint before all the stoichiometric iodine has actually reacted, giving a volume that is too small and a copper result that is systematically low, often by several tenths of a percent. Adding potassium thiocyanate near the endpoint converts the CuI surface coating to CuSCN, which has much lower affinity for adsorbing iodine, releasing the trapped triiodide back into solution so the remaining titration proceeds to a genuinely quantitative endpoint.

Can potassium dichromate be titrated iodometrically, and if so why is it preferred over permanganate for standardizing thiosulfate?

Yes — dichromate, $E^\circ = +1.33$ V, readily oxidizes excess iodide to iodine in acidic solution, $\mathrm{Cr_2O_7^{2-} + 6I^- + 14H^+ \rightarrow 2Cr^{3+} + 3I_2 + 7H_2O}$ , and the liberated iodine is titrated with thiosulfate exactly as in copper or chlorine assays. Dichromate is favored as a primary standard for this purpose because $\mathrm{K_2Cr_2O_7}$ can be obtained in very high purity, is not hygroscopic, and is stable indefinitely as a solid and in solution, unlike permanganate, which slowly decomposes and requires its own indirect standardization. This makes dichromate-liberated iodine one of the most reliable routes for assigning an accurate, traceable concentration to a stock thiosulfate solution.

Why does raising the pH matter so much for the iodimetric titration of arsenite (As(III))?

The As(V)/As(III) half-reaction explicitly consumes hydrogen ions, $\mathrm{H_3AsO_4 + 2H^+ + 2e^- \rightarrow HAsO_2 + 2H_2O}$ , so by the Nernst equation its formal potential is strongly pH-dependent, falling as pH increases. In strongly acidic solution the As(V)/As(III) potential actually exceeds that of the iodine couple, meaning As(V) would tend to oxidize iodide rather than the reverse — the titration would run backward or stall at an ill-defined equivalence point. Buffering to mildly alkaline conditions (commonly pH 7–9 with sodium bicarbonate) drops the As(V)/As(III) formal potential well below 0.536 V, ensuring the forward reaction (I₃⁻ oxidizing As(III)) goes essentially to completion and gives a sharp, quantitative endpoint.

How does the presence of large amounts of chloride or other halides interfere with iodometric methods, and how is it avoided?

Strong oxidants used to standardize reagents, or present as interferents in real samples (such as residual chlorine in bleach assays combined with other oxidants), can sometimes oxidize excess iodide non-selectively, or conversely chloride can be oxidized by very strong oxidants under some conditions, complicating the stoichiometry. In practice the bigger real-world interference is air oxidation of iodide by dissolved oxygen in acidic solution, which is catalyzed by light and by trace metal ions (notably nitrite and some transition metals), producing a slow, continuing color development that inflates the apparent titer if the sample sits before titration. The standard countermeasures are to titrate promptly after liberating iodine, keep the flask stoppered and away from direct light, and avoid strongly acidic conditions for iodide solutions left standing any length of time.

Why is thiosulfate, and not iodine, usually the species standardized as the working titrant of record in most laboratories?

Solid $\mathrm{Na_2S_2O_3\cdot5H_2O}$ is not a primary standard — it effloresces, its water content is not perfectly reproducible, and the solution slowly decomposes over weeks via bacterial action (particularly thiobacilli) and reaction with dissolved carbon dioxide. Because of this, thiosulfate solutions are always standardized against a stable, weighable, non-hygroscopic oxidant — typically potassium iodate ( $\mathrm{KIO_3}$ ) or potassium dichromate — using the same liberated-iodine chemistry as in a real assay, so that any small imperfection in the true thiosulfate concentration is captured empirically on the same day the working solution is used, rather than relying on a nominal value calculated purely from the mass weighed out.

Is the starch endpoint equally reliable at both high and low iodine concentrations, and are there situations where a different endpoint detection method is preferred?

The starch–triiodide complex is extremely sensitive, visible down to roughly $10^{-5}$ M I₂, which is why it gives such a sharp visual cutoff even in dilute titrations such as the Winkler dissolved-oxygen method. It performs less reliably in very hot solutions (the complex partially dissociates, blurring the endpoint), in the presence of high concentrations of certain organic solvents, or when titrating colored or turbid samples where the blue color is hard to see against the background. In those cases, or whenever maximal precision is required, a platinum indicator electrode monitoring the potential directly — following exactly the Redox Titration Curves: The E-vs-Volume Plot approach used for any redox system — replaces the visual starch endpoint with a potentiometric one, locating the equivalence point from the inflection in an E-vs-volume curve rather than from a color change.