Coulometry and Coulometric Titrations

A titration that never needs a standard solution sounds impossible. Yet it is routine in metrology laboratories. Coulometry replaces the buret with an electrode. It replaces "milliliters times molarity" with "amperes times seconds." Electric charge can be measured with extraordinary precision. Modern coulometers resolve microcoulombs. The technique underlies primary standards for chemical analysis. This includes the Karl Fischer coulometric water titrator found in nearly every pharmaceutical QC lab. The method matters because it converts an electrochemical measurement into a mole count. It needs no calibrated volumetric reagent, no indicator error, and no meniscus to misread. Faraday's law states that charge passed is directly proportional to moles of substance reacted. The proportionality constant is one of the most exactly known numbers in nature — the Faraday constant. It is now defined exactly under the SI as $F = 96\,485.332\,12\ \text{C mol}^{-1}$ . This lets a coulometric measurement claim accuracy limited only by current and time measurement, not by chemistry. The big picture connects two operating modes. In controlled-potential (potentiostatic) coulometry, the working electrode potential is fixed at a value. This drives one electroactive species to completion while the current decays exponentially. The total charge is obtained by integrating current over time, often electronically or digitally. In constant-current coulometric titration, a constant current generates a titrant electrochemically in situ — bromine, iodine, EDTA-complexed metal ion, hydroxide, acid. An independent end-point sensor (potentiometric, amperometric, spectrophotometric, or conductometric) signals when the electrogenerated titrant has exactly consumed the analyte. Charge, not concentration, is the primary standard, and time replaces volume. This article builds the theory from Faraday's law through both operating modes. It connects to the electrode kinetics established in Voltammetry Fundamentals. It borrows the end-point logic of Potentiometric Titrations. It closes with the conductometric detection scheme used when no convenient redox indicator electrode exists. Three fully worked problems pin every equation to real numbers. These are a foundational charge calculation, an applied controlled-potential copper assay, and an exam-level current-efficiency correction.

Key takeaways

Faraday's law, $Q = nFN$ , converts measured charge $Q$ (coulombs) directly into moles $N$ of electroactive substance, using electrons transferred $n$ and the Faraday constant $F$ .
Controlled-potential coulometry holds electrode potential fixed; current decays as $i(t) = i_0 e^{-kt}$ , and $Q = \int i\,dt$ is obtained by planimetry, electromechanical integration, or digital summation.
Constant-current coulometric titration electrogenerates a titrant at 100% current efficiency; $N_\text{analyte} = it_\text{eq}/(nF)$ , with time $t_\text{eq}$ measured to an end point.
Current efficiency less than 100% (side reactions) is the dominant systematic error; it is corrected by an internal standard or coulometric assay of a certified reference material.
End-point detection can be potentiometric (indicator electrode + reference), amperometric (dead-stop), or conductometric, tracking the bulk ionic conductivity of the titration cell as electrogenerated ions consume or replace analyte ions.
Coulometry needs no standardized titrant: charge is the primary measurement, traceable to time and current standards rather than to a calibrated volumetric solution.
Key idea: charge passed at known current efficiency counts moles directly through $Q = nFN$ , making the coulomb a self-calibrating chemical buret.

Where this lives today

Coulometric Karl Fischer titration is the workhorse trace-water method in pharmaceutical and petrochemical QC labs, routinely quantifying water from tens of parts-per-million down to single-digit ppm in solvents, active pharmaceutical ingredients, and lubricants, precisely because the electrogenerated-I $_2$ approach avoids the instability of standardized KF titrant solutions at such low consumption rates. Constant-current coulometric titration of chloride using electrogenerated Ag $^+$ (with amperometric or potentiometric silver-ion end-point detection) is a NIST- and pharmacopeia-recognized method for chloride assay in biological fluids and pharmaceutical formulations, prized because Ag $^+$ is generated fresh at the moment of use, sidestepping precipitation losses that plague AgNO $_3$ volumetric titrations. Controlled-potential coulometry remains a primary method for assay of high-purity electrodeposited metals (copper, nickel, zinc) in metallurgical and semiconductor-industry reference materials, exploited precisely because it needs no matrix-matched calibration curve, unlike atomic absorption or ICP-based methods. Coulometric generation of electrons themselves as the primary calibrant also underlies the practical realization of the SI ampere and, historically, provided one of the routes to an independent determination of the Faraday constant (and hence Avogadro's number) via the silver coulometer, a link between electroanalytical chemistry and fundamental metrology that persists in modern watt-balance and quantum-Hall-based current standards. In biosensing and environmental monitoring, coulometric generation of reagents such as bromine (for phenol and ammonia determination) or EDTA-complexed mercury(II) (for cyanide and sulfide titrations) allows unstable or hazardous titrants to be made and consumed in situ, minimizing reagent handling and waste — an advantage increasingly valued in green-chemistry-oriented analytical method design.

Ask the tutor: How would the coulometric titration curve (potential or conductivity vs. time) change if the electrogenerated titrant reacted with the analyte in a 2:1 rather than 1:1 stoichiometry?

The physical picture

Every coulometric method rests on a single electrochemical bookkeeping statement: each electron that crosses the electrode–solution interface is consumed by exactly one well-defined redox event. If a cell passes charge $Q$ coulombs and every electron transfers $n$ times per mole of analyte, then the amount transformed is fixed by Faraday's constant regardless of cell geometry, electrode area, solution stirring, or temperature — provided (1) the electrode reaction is the only reaction consuming current (100% current efficiency) and (2) $Q$ is measured accurately. This is fundamentally different from potentiometry or voltammetry, where the signal (potential or current) is only proportional to concentration through an empirically calibrated response function; in coulometry the stoichiometry itself is the calibration. Two experimental architectures exploit this bookkeeping. In controlled-potential coulometry, the working electrode is poised (via a potentiostat, exactly as in Voltammetry Fundamentals) at a value on the mass-transport-limited plateau of the analyte's wave. Initially the bulk solution is fully loaded with analyte, so current is large; as electrolysis proceeds, bulk concentration drops exponentially since the cell behaves as a well-stirred, exhaustively depleting reactor. The reaction is forced to run to completion (current decays toward the residual/background current), and the total charge passed equals $nF$ times the initial moles of analyte, corrected for any residual (blank) charge. In constant-current coulometric titration, current is held constant by a galvanostat, and that same current is used to generate a chemical titrant electrolytically at one electrode (the generator electrode) at a rate $i/(nF)$ moles per second. This electrogenerated titrant then reacts homogeneously with the analyte exactly as a titrant delivered from a buret would, and a conventional end-point sensor — potentiometric indicator electrode, amperometric dead-stop pair, or a conductivity cell — signals the equivalence time $t_\text{eq}$ . Because the "concentration" of titrant is replaced by the generation rate, and the "volume" is replaced by elapsed time, the titration curve looks formally identical to a volumetric titration curve with $V$ relabeled as $t$ .

Figure

Controlled-potential coulometry: exponential current decay and its integral

Figure

Constant-current coulometric titration cell with conductometric end-point sensing

The role of current efficiency deserves its own picture. If the generator electrode also drives a competing reaction — solvent oxidation, oxygen reduction, or a second redox couple — only a fraction $\phi$ of the charge does useful titrant-generating work. Real coulometric cells are engineered (large-area platinum, appropriate supporting electrolyte, isolated generator compartment) specifically to push $\phi$ as close to unity as possible, because unlike volumetric titrant concentration, current efficiency cannot be "standardized away" after the fact — it must be built into the cell design or independently verified against a certified reference material.

The core ideas

The whole quantitative structure descends from one law of electrochemical stoichiometry combined with the definition of charge.

Q = \int_0^{t} i(t')\,dt'

where: $Q$ is charge in coulombs, $i(t')$ is instantaneous current in amperes, and $t$ is elapsed electrolysis time in seconds. Reading: charge is literally the time-integral of current — for constant current this collapses to $Q = it$ , and for the exponentially decaying current of controlled-potential coulometry it must be evaluated by integration (analog integrator, coulometer, or digital summation of current samples).

Q = nFN

where: $n$ is the number of moles of electrons transferred per mole of the electroactive species (from the balanced half-reaction), $F = 96\,485.332\,12\ \text{C mol}^{-1}$ is the Faraday constant, and $N$ is moles of analyte reacted. Reading: this is Faraday's law of electrolysis — it says that charge is a direct stoichiometric counter of moles, with no empirical calibration constant, only the integer (or well-defined) electron count $n$ .

N_\text{analyte} = \frac{it_\text{eq}}{nF}\qquad(\text{constant-current titration, }\phi = 1)

where: $i$ is the constant generator current (A), $t_\text{eq}$ is the elapsed time to the end point (s), $n$ is electrons per mole of titrant-generating half-reaction, and $F$ is the Faraday constant. Reading: this is the coulometric analogue of $N_\text{analyte} = C_\text{titrant}V_\text{eq}$ in a volumetric titration — current $\times$ time plays the role of concentration $\times$ volume, so $it$ has units of coulombs, and dividing by $nF$ converts coulombs to moles.

\phi = \frac{Q_\text{useful}}{Q_\text{total}} = \frac{nFN_\text{true}}{it}

where: $\phi$ is the fractional current efficiency, $Q_\text{useful}$ is charge that actually drove the analytically relevant reaction, and $Q_\text{total} = it$ is all charge passed by the galvanostat/potentiostat. Reading: any coulometric result must be divided by $\phi$ (if $\phi < 1$ ) or the cell must be engineered so that $\phi \to 1$ ; $\phi$ is usually determined once, for a given cell and current density, using a known amount of a primary-standard reactant, then treated as a fixed instrumental parameter.

i(t) = i_0\,e^{-kt}\ ,\qquad k = \frac{mDA}{V\delta}\ \left(\approx\frac{DA}{V\delta}\right)

where: $i_0$ is initial limiting current, $k$ (s $^{-1}$ ) is the cell's characteristic decay constant set by mass-transport parameters — diffusion coefficient $D$ , electrode area $A$ , diffusion-layer thickness $\delta$ , solution volume $V$ , and a mass-transfer coefficient $m = D/\delta$ . Reading: controlled-potential coulometry behaves as a first-order depletion process because the bulk concentration (and hence the mass-transport-limited current) decays exponentially as the well-stirred solution is exhaustively electrolyzed; a larger electrode area or more vigorous stirring speeds up the assay by increasing $k$ , without changing the total charge, which depends only on $N$ and $n$ .

Derivation AFaraday's law from electrode stoichiometry and charge counting

Goal. Show rigorously that the total charge passed in an exhaustive, 100%-efficient electrolysis equals $nF$ times the moles of analyte converted, and connect this to the exponential current decay observed experimentally.

Step 1 — define the half-reaction and charge per event. Consider the electroactive species $\text{Ox} + n\,e^- \to \text{Red}$ at the working electrode. Each individual electron transfer event carries one elementary charge $e = 1.602\,176\,634\times10^{-19}\ \text{C}$ (exact, SI-defined). Converting one mole of Ox to Red therefore requires $n$ moles of electrons, i.e. $n N_A$ individual electrons, where $N_A$ is Avogadro's number.

Step 2 — assemble the Faraday constant. The charge carried by one mole of electrons is

F = N_A e = (6.022\,140\,76\times10^{23}\ \text{mol}^{-1})(1.602\,176\,634\times10^{-19}\ \text{C}) = 96\,485.332\,12\ \text{C mol}^{-1}.

Both $N_A$ and $e$ are exact defined constants in the current SI, so $F$ is an exactly known number — coulometry inherits this exactness directly, which is why it can serve as a primary method.

Step 3 — total charge for complete conversion. If $dN$ moles of Ox are converted in a differential time $dt$ , the differential charge passed is $dQ = nF\,dN$ (each mole converted moves $nF$ coulombs). Integrating from the start of electrolysis ( $N = N_0$ , the initial moles present) to completion ( $N \to 0$ ):

Q_\text{total} = \int_{N_0}^{0} (-nF)\,dN = nF N_0.

The sign flips because $N$ decreases as electrolysis proceeds while $Q$ accumulates; the result is simply $Q_\text{total} = nFN_0$ , independent of the rate at which electrolysis occurs.

Step 4 — why the current must decay exponentially. In a well-stirred cell with electrode area $A$ operating on the mass-transport-limited plateau, the current is $i(t) = nFAm\,C(t)$ , where $C(t)$ is the bulk concentration of Ox and $m$ is the (constant) mass-transfer coefficient. Mass balance on the bulk solution of volume $V$ gives

V\frac{dC}{dt} = -Am\,C(t)

since every mole leaving the bulk by mass transport to the electrode is consumed. This first-order ODE integrates to $C(t) = C_0 e^{-(Am/V)t}$ , and hence $i(t) = i_0 e^{-kt}$ with $k = Am/V$ , confirming the exponential decay asserted above. Integrating $i(t)$ over all time recovers exactly $Q_\text{total} = i_0/k = nFAmC_0/(Am/V) = nFVC_0 = nFN_0$ — consistent with Step 3, as it must be, since the total charge cannot depend on how fast the depletion occurs.

Result. $Q_\text{total} = nFN_0$ holds exactly at 100% current efficiency, and the observed current transient is a diagnostic: a clean single-exponential decay to the residual (background) current confirms that only one electroactive process is occurring and that the assay has gone to completion once $i(t)$ reaches the baseline.

Go deeper · Derivation B · 2 minConstant-current coulometric titration curve and the end-point time

Derivation BConstant-current coulometric titration curve and the end-point time

Goal. Derive the relationship between elapsed generation time and the concentration of unreacted analyte, showing the formal equivalence to a volumetric titration curve, and locate the end point.

Step 1 — set up the generation rate. At constant current $i$ and current efficiency $\phi = 1$ , the generator electrode produces titrant $T$ at a constant molar rate

\frac{dN_T}{dt} = \frac{i}{nF},

where $n$ is the electrons transferred per mole of $T$ generated (e.g., $n=2$ for $2\,\text{I}^- \to \text{I}_2 + 2e^-$ , generating $N_{\text{I}_2} = it/(2F)$ moles of I $_2$ , i.e., $it/F$ equivalents).

Step 2 — react generated titrant with analyte instantaneously. Assume the homogeneous reaction $T + \text{Analyte} \to \text{Products}$ (1:1 in equivalents) is fast compared to the generation rate, so essentially all $T$ generated at time $t < t_\text{eq}$ is immediately consumed. The moles of analyte remaining are then

N_\text{analyte}(t) = N_{\text{analyte},0} - \frac{it}{nF}, \qquad t \le t_\text{eq}.

This is the exact time-domain analogue of $N_\text{analyte}(V) = N_{\text{analyte},0} - C_\text{titrant}V$ in a volumetric titration, with $C_\text{titrant}V \leftrightarrow it/(nF)$ .

Step 3 — locate the equivalence time. The end point occurs when $N_\text{analyte}(t_\text{eq}) = 0$ :

0 = N_{\text{analyte},0} - \frac{it_\text{eq}}{nF} \quad\Longrightarrow\quad N_{\text{analyte},0} = \frac{it_\text{eq}}{nF}.

This reproduces the working equation used throughout constant-current coulometry. Because $i$ is held constant electronically (typically to better than 0.01%) and $t_\text{eq}$ is measured with an electronic timer/counter synchronized to the current switch, both factors on the right-hand side are known with high relative precision, so the dominant remaining uncertainty is usually the sharpness of the end-point detection, exactly as in a manual volumetric titration where buret reading precision limits accuracy.

Step 4 — connect to an indicator response (potentiometric or conductometric). Beyond $t_\text{eq}$ , excess titrant $T$ accumulates linearly, $N_{T,\text{excess}}(t) = i(t-t_\text{eq})/(nF)$ , and if $T$ or its reaction product is detected by an indicator electrode, the potentiometric response follows a Nernstian jump in $E$ versus $t$ centered at $t_\text{eq}$ , structurally identical to $E$ versus $V$ curves. If instead a conductivity cell monitors bulk ionic conductivity $\kappa(t)$ , the pre-equivalence slope $d\kappa/dt$ reflects the difference in ionic mobilities between the analyte ion consumed and the product ion released, and the post-equivalence slope reflects the mobility of the excess titrant ion; the two linear segments extrapolate to an intersection at $t_\text{eq}$ , exactly analogous to a conductometric end point plotted against volume.

Worked examples

Example 1 (foundational) — charge and moles from a constant current pulse. A constant-current coulometer passes $i = 15.00\ \text{mA}$ through a cell for exactly $t = 428.6\ \text{s}$ to electrogenerate Br $_2$ by oxidation of Br $^-$ at a platinum anode: $2\,\text{Br}^- \to \text{Br}_2 + 2e^-$ . Find the moles of Br $_2$ generated and the mass of Br $_2$ produced. Charge passed:

Q = it = (15.00\times10^{-3}\ \text{A})(428.6\ \text{s}) = 6.429\ \text{C}.

Electrons transferred per mole of Br $_2$ is $n = 2$ . Moles of Br $_2$ :

N_{\text{Br}_2} = \frac{Q}{nF} = \frac{6.429\ \text{C}}{2 \times 96\,485.33\ \text{C mol}^{-1}} = 3.332\times10^{-5}\ \text{mol}.

Mass, using $M_{\text{Br}_2} = 159.808\ \text{g mol}^{-1}$ :

m_{\text{Br}_2} = (3.332\times10^{-5}\ \text{mol})(159.808\ \text{g mol}^{-1}) = 5.325\times10^{-3}\ \text{g} = 5.325\ \text{mg}.

Limiting-case check: if the pulse were doubled to $t = 857.2\ \text{s}$ at the same current, $Q$ doubles to $12.86\ \text{C}$ and $N_{\text{Br}_2}$ doubles to $6.664\times10^{-5}\ \text{mol}$ — charge and moles scale linearly with time at fixed current, exactly as the constant-current formula requires, with no dependence on cell volume or stirring rate (those affect only how fast the Br $_2$ subsequently reacts, not how much is made).

Example 2 (applied) — controlled-potential coulometric assay of copper in a plating bath sample. A 2.000 mL aliquot of a copper plating bath is diluted and electrolyzed at a mercury-pool cathode held at a potential on the Cu $^{2+}$ reduction plateau ( $\text{Cu}^{2+} + 2e^- \to \text{Cu}(0)$ , deposited as amalgam). The current is recorded electronically and numerically integrated; total charge passed until the current falls to the residual background level is $Q_\text{gross} = 38.64\ \text{C}$ . A blank run (identical cell, no Cu $^{2+}$ added) under the same conditions and elapsed time passes $Q_\text{blank} = 0.083\ \text{C}$ (residual/background electrolysis of trace impurities and solvent). Current efficiency of the cell for Cu $^{2+}$ reduction, established previously with a certified copper reference solution, is $\phi = 0.997$ . Find the mass of copper in the original 2.000 mL aliquot and its concentration in g/L. Step 1 — blank-corrected charge:

Q_\text{net} = Q_\text{gross} - Q_\text{blank} = 38.64\ \text{C} - 0.083\ \text{C} = 38.56\ \text{C}.

Step 2 — correct for current efficiency (only a fraction $\phi$ of $Q_\text{net}$ actually reduced Cu $^{2+}$ ; here we treat $\phi$ as already characterizing the net signal's fidelity, so the true useful charge is):

Q_\text{useful} = \phi\,Q_\text{net} = (0.997)(38.56\ \text{C}) = 38.44\ \text{C}.

Step 3 — moles of Cu, with $n = 2$ :

N_\text{Cu} = \frac{Q_\text{useful}}{nF} = \frac{38.44\ \text{C}}{2(96\,485.33\ \text{C mol}^{-1})} = 1.9922\times10^{-4}\ \text{mol}.

Step 4 — mass, using $M_\text{Cu} = 63.546\ \text{g mol}^{-1}$ :

m_\text{Cu} = (1.9922\times10^{-4}\ \text{mol})(63.546\ \text{g mol}^{-1}) = 1.2660\times10^{-2}\ \text{g} = 12.66\ \text{mg}.

Step 5 — concentration in the original aliquot:

c_\text{Cu} = \frac{12.66\times10^{-3}\ \text{g}}{2.000\times10^{-3}\ \text{L}} = 6.330\ \text{g/L}.

Limiting-case check: if $\phi$ were exactly 1.000 (an ideal, side-reaction-free cell), $Q_\text{useful} = Q_\text{net} = 38.56\ \text{C}$ , giving $m_\text{Cu} = 12.698\ \text{mg}$ — about 0.30% higher than the efficiency-corrected value, showing explicitly that neglecting even a 0.3% current-efficiency deficit shifts the reported mass by a comparable relative amount, which matters at the precision coulometry is normally trusted to deliver.

Example 3 (exam/competition-level) — constant-current Karl Fischer coulometric water determination with a stated current-efficiency defect. A coulometric Karl Fischer titrator generates I $_2$ at a constant current $i = 200.0\ \text{mA}$ via $2\,\text{I}^- \to \text{I}_2 + 2e^-$ ; the electrogenerated I $_2$ reacts with H $_2$ O in the one-electron-equivalent KF stoichiometry $\text{I}_2 + \text{SO}_2 + \text{H}_2\text{O} + \text{CH}_3\text{OH} \to 2\,\text{HI} + \text{CH}_3\text{OSO}_3\text{H}$ , so one mole of I $_2$ consumes exactly one mole of H $_2$ O. A sample is injected and the bipotentiometric end-point detector (dead-stop indicator pair) signals equivalence after $t_\text{eq} = 96.4\ \text{s}$ of generation beyond the pre-titration dry baseline. The instrument manufacturer specifies that this particular generator electrode/electrolyte combination runs at current efficiency $\phi = 0.992$ due to a slow parasitic anodic oxidation of trace methanol. Compute the mass of water titrated, and then find what relative error would have been made had the analyst (incorrectly) assumed 100% current efficiency. Step 1 — total charge passed during titrant generation:

Q = it_\text{eq} = (0.2000\ \text{A})(96.4\ \text{s}) = 19.28\ \text{C}.

Step 2 — useful charge that actually generated I $_2$ that reacted with water:

Q_\text{useful} = \phi Q = (0.992)(19.28\ \text{C}) = 19.126\ \text{C}.

Step 3 — moles of I $_2$ generated (and hence moles of H $_2$ O consumed, 1:1), with $n=2$ electrons per I $_2$ :

N_{\text{I}_2} = \frac{Q_\text{useful}}{nF} = \frac{19.126\ \text{C}}{2(96\,485.33\ \text{C mol}^{-1})} = 9.9107\times10^{-5}\ \text{mol}.

Since the KF reaction consumes H $_2$ O 1:1 with I $_2$ , $N_{\text{H}_2\text{O}} = 9.9107\times10^{-5}\ \text{mol}$ . Step 4 — mass of water, $M_{\text{H}_2\text{O}} = 18.015\ \text{g mol}^{-1}$ :

m_{\text{H}_2\text{O}} = (9.9107\times10^{-5}\ \text{mol})(18.015\ \text{g mol}^{-1}) = 1.7859\times10^{-3}\ \text{g} = 1.786\ \text{mg}.

Step 5 — the incorrect (100%-efficiency) calculation would instead give

N_{\text{H}_2\text{O}}^{(\phi=1)} = \frac{Q}{2F} = \frac{19.28\ \text{C}}{2(96\,485.33)} = 9.9915\times10^{-5}\ \text{mol}, \qquad m^{(\phi=1)} = 1.8006\ \text{mg}.

Step 6 — relative error from ignoring the efficiency defect:

\frac{m^{(\phi=1)} - m_{\text{H}_2\text{O}}}{m_{\text{H}_2\text{O}}} = \frac{1.8006 - 1.7859}{1.7859} = 8.2\times10^{-3} = 0.82\%.

Limiting-case check: as $\phi \to 1$ exactly, $Q_\text{useful} \to Q$ and the two calculations coincide, giving zero relative error — confirming that the entire 0.82% discrepancy is attributable solely to the 0.8-percentage-point efficiency defect ( $1-\phi = 0.008$ ), and in fact the relative error is numerically very close to $(1-\phi)$ itself when $\phi$ is near unity, since $m^{(\phi=1)}/m_{\text{true}} = 1/\phi \approx 1 + (1-\phi)$ for small $(1-\phi)$ .

Limiting cases, scope & common traps

Situation	Behavior / correct treatment
$\phi \to 1$ (ideal current efficiency)	$N = Q/(nF)$ exactly; no correction factor needed; this is the design target of every coulometric cell.
$\phi$ known and constant, $\phi < 1$	Divide computed $N$ (assuming $\phi=1$ ) by $\phi$ , or equivalently multiply $Q$ by $\phi$ before dividing by $nF$ — must characterize $\phi$ with a certified reference material under identical cell conditions.
$n$ ambiguous (mixed-valence or stepwise redox)	Controlled-potential coulometry can determine $n$ experimentally from $Q_\text{total}/(FN_\text{known})$ when $N$ is known independently — used historically to establish mechanisms.
Blank/background current nonzero	Always run a blank under identical time and conditions; subtract $Q_\text{blank}$ before applying Faraday's law (Example 2).
Very dilute analyte, controlled-potential mode	Current approaches the residual/background plateau early; decay constant $k$ small relative to noise — extend electrolysis time or increase electrode area/stirring to raise $k$ without changing $Q_\text{total}$ .
End-point detection ambiguous (no sharp indicator break)	Switch to conductometric detection, which works even for reactions with no convenient redox-indicator electrode, provided ionic mobilities differ before/after equivalence.
Multiple electroactive species overlapping in potential	Controlled-potential selectivity fails; must resolve by prior separation or by choosing $n$ /potential where only the target species is active (mirrors selectivity limits in Voltammetry Fundamentals).

The chemist behind it

The theoretical bedrock of coulometry — the exact proportionality between charge and chemical change — was laid down by Michael Faraday in a series of electrolysis experiments between 1832 and 1834, published as parts of his Experimental Researches in Electricity. Faraday systematically electrolyzed a wide variety of compounds (aqueous acids, molten salts, and gases) and showed that the mass of substance liberated at an electrode was proportional both to the quantity of electricity passed and to the chemical equivalent weight of the substance — a result he termed the laws of electrochemical equivalence. He had no concept of the electron (which would not be identified until J. J. Thomson's 1897 cathode-ray experiments), so Faraday's constant emerged empirically as "the quantity of electricity needed to liberate one equivalent weight of any substance," a strikingly modern statement arrived at purely from careful gravimetric and volumetric measurement of electrolysis products, using a device he called a voltameter to measure gas volumes liberated. The analytical technique built on this law — using charge itself, rather than a calibrated titrant, as the quantitative measure — matured in the twentieth century through the work of electroanalytical chemists developing controlled-potential and constant-current instrumentation, most notably associated with John J. Lingane, whose 1958 monograph Electroanalytical Chemistry systematized controlled-potential coulometry as a rigorous quantitative method and demonstrated its application to metal ion assays with accuracy rivaling gravimetric analysis, without the tedium of precipitation and weighing.

Origins · 1 minMichael Faraday and how this was discovered

Michael Faraday

1791–1867 · Royal Institution, London, 1830s electrochemistry

Faraday, a self-taught bookbinder's apprentice turned the era's foremost experimentalist, established the quantitative laws of electrolysis using nothing more sophisticated than voltameters, careful timekeeping, and meticulous gravimetric analysis of liberated products — decades before the electron was known to exist. His insistence on precise, reproducible measurement of "quantity of electricity" versus "quantity of chemical change" is the direct ancestor of every modern coulometer's charge-integration circuit.

History

- 1832–1834: Faraday establishes the laws of electrochemical equivalence via systematic voltameter experiments.

- 1897: J. J. Thomson identifies the electron, later allowing Faraday's constant to be interpreted as

N_Ae

- 1938–1958: Constant-current coulometric titration and controlled-potential coulometry are developed and systematized as quantitative analytical methods, notably by Lingane and coworkers.

- 1959: Coulometric Karl Fischer titration for trace water is introduced, generating I

_2

in situ rather than requiring standardized KF reagent.

- 2019: The SI redefinition fixes the elementary charge

e

exactly, making the Faraday constant

F = N_A e

an exact defined quantity rather than a measured one.

Why is coulometry considered a "primary" or "absolute" method of analysis?

Because its quantitative link between measured signal and moles of analyte, $Q = nFN$ , depends only on the Faraday constant (an exactly defined SI quantity) and the integer or well-characterized electron count $n$ — not on any empirically calibrated response factor, molar absorptivity, or standardized titrant concentration. Provided current efficiency is unity (or independently known) and the electron-transfer stoichiometry is unambiguous, the mole count follows directly from measuring current and time, both of which are themselves traceable to fundamental electrical standards. This is analogous to gravimetric analysis being "primary" because it rests on the definition of the mole and atomic mass, rather than on an instrument calibration curve.

How is current efficiency actually measured in practice?

A certified reference material — a solution of accurately known concentration of a substance that undergoes a clean, well-understood electrode reaction — is electrolyzed or titrated under exactly the same cell geometry, electrolyte composition, current density, and temperature as the routine assay. Comparing the coulometrically computed amount (assuming $\phi=1$ ) against the certified true amount gives $\phi$ directly as their ratio. This $\phi$ is then treated as a fixed instrumental correction factor for that specific cell configuration and current regime, and it is periodically re-verified, especially if electrode surfaces are replaced or cleaned, since surface condition strongly affects competing side reactions.

Why does controlled-potential coulometry produce an exponentially decaying current rather than a constant one?

Once the electrode potential is fixed on the mass-transport-limited plateau, the current is proportional to the instantaneous bulk concentration of the electroactive species, not to some fixed applied driving force. As electrolysis proceeds in a closed, well-stirred cell, the bulk concentration itself is depleted, obeying a first-order differential equation identical in form to radioactive decay or first-order chemical kinetics. The result, derived directly from mass balance, is $i(t) = i_0 e^{-kt}$ , and the assay is judged "complete" once the current has decayed to the residual background level, typically defined as some small fraction (often 0.1%) of the initial current.

What is "100% current efficiency" and why can it fail?

100% current efficiency means every electron passed through the external circuit drives the desired analytical half-reaction, with no charge diverted into side reactions such as solvent decomposition, oxygen or hydrogen evolution, oxidation of trace impurities, or a competing redox couple in the sample matrix. It can fail if the applied potential (controlled-potential mode) is not well isolated from a second wave, if the supporting electrolyte itself is oxidized/reduced at the operating current density, or if trace species (like methanol in Karl Fischer reagent) undergo slow parasitic electrode reactions. Cell design — electrode material, electrolyte composition, compartmentalization with a frit or diaphragm — is chosen specifically to suppress these pathways.

How does coulometric Karl Fischer titration differ from classical volumetric Karl Fischer titration?

Classical volumetric KF titration delivers a pre-standardized iodine-containing reagent from a buret, and the water content is calculated from the volume delivered times the reagent's water-equivalence factor (mg H $_2$ O per mL), which must be periodically restandardized because KF reagent titer drifts. Coulometric KF titration instead generates I $_2$ electrochemically in situ from iodide in the reagent, at a current whose integral directly gives moles of I $_2$ (and hence moles of water) via Faraday's law, eliminating the need to standardize a titer at all. This makes the coulometric method far better suited to very small water contents (low ppm to sub-ppm), where volumetric delivery of microliter titrant volumes becomes imprecise.

Can coulometry determine the number of electrons n in an unknown electrode mechanism?

Yes — this is one of its classic mechanistic uses. If the moles of analyte $N$ are known independently (by weighing a pure compound, for instance) and controlled-potential coulometry is run to completion while integrating total charge $Q_\text{total}$ , then $n$ is obtained directly as $n = Q_\text{total}/(FN)$ . Historically this approach helped establish, for example, whether an organic reduction proceeded via a 1-electron radical-anion pathway or a 2-electron pathway to a fully reduced product, complementing voltammetric peak-current and peak-potential diagnostics.

Why does constant-current coulometric titration behave mathematically just like a volumetric titration?

Both methods deliver titrant to the analyte at a controlled, known rate and monitor a sensor response until an equivalence condition is met. In volumetric titration the controlled quantity is (concentration × volume delivered); in coulometric titration it is (current × time elapsed) divided by $nF$ . Because the underlying chemistry — a stoichiometric reaction between analyte and titrant, followed to an end point by a Nernstian or conductometric sensor — is identical, substituting $it/(nF)$ wherever $C_\text{titrant}V$ would appear reproduces the same sigmoidal titration-curve shape, the same equivalence-point sharpness criteria, and the same excess-titrant behavior familiar from volumetric analysis.