Reference Electrodes: SHE, SCE, and Ag/AgCl

Reference electrodes are the invisible backbone of essentially all modern electroanalytical instrumentation. Every commercial glass pH electrode is a "combination electrode" — a pH-sensitive glass membrane bulb fused with an internal Ag/AgCl reference and an external Ag/AgCl (or occasionally still calomel) reference, separated by a ceramic or polymer junction — because manufacturing costs and mercury-disposal regulations have pushed the industry almost entirely toward Ag/AgCl. Ion-selective electrodes for fluoride, sodium, calcium, nitrate, and dozens of other analytes in clinical chemistry and environmental monitoring rely on the same Ag/AgCl reference chemistry, often now miniaturized into microfabricated, solid-contact "reference electrodes" for lab-on-a-chip and wearable sensors, replacing the liquid KCl fill entirely with a redox-buffered polymer matrix to avoid leakage. In biomedical and neuroscience research, Ag/AgCl electrodes are the standard skin-contact and implantable electrode for EEG, ECG, and EMG recording, exploiting the same second-kind-electrode chemistry to give a stable, low-noise, non-polarizable interface for recording biopotentials — a direct descendant of the analytical chemistry reference electrode applied to a completely different measurement problem. In corrosion science and cathodic protection of pipelines and ship hulls, Ag/AgCl and Cu/CuSO₄ reference electrodes (a related second-kind design using copper and copper sulfate) are buried or submerged alongside structures to monitor and control protective potentials over years of continuous service, exactly as worked through in Example 2. In fundamental research, reference electrodes underpin voltammetric techniques (cyclic voltammetry, differential pulse voltammetry, stripping analysis) used to probe redox chemistry of everything from battery materials to neurotransmitters, and they remain essential in coulometric and potentiometric titrations used in pharmaceutical quality control, where regulatory methods (USP, ASTM) specify calomel or Ag/AgCl references by name because their potentials are certified traceable to the SHE convention. Understanding exactly how these electrodes are constructed and why their potentials are fixed is the necessary foundation for Potentiometry Fundamentals, for interpreting Indicator Electrodes correctly, and for every quantitative voltammetric measurement covered in Voltammetry Fundamentals.

Ask the tutor: How exactly does the choice of reference electrode propagate into the uncertainty of a measured pH value in a real potentiometric titration?

A reference electrode is, at its core, an ordinary redox half-cell in which the experimenter has engineered away the one thing that normally makes electrode potentials variable: dependence on the sample. Consider the general Nernst relation for a half-reaction $\mathrm{Ox} + ne^- \rightleftharpoons \mathrm{Red}$. If the activities of Ox and Red are free to vary with whatever solution the electrode is dipped into, the electrode's potential is a sensor of that solution — useful for indicator electrodes, useless as a reference. The engineering trick behind SHE, SCE, and Ag/AgCl is to build the half-cell so that the relevant activity is buffered internally, physically isolated from the test solution by a porous junction, and therefore constant no matter what happens outside. The SHE is the primordial example. Hydrogen gas at exactly 1 bar fugacity bubbles over a platinized platinum electrode immersed in a solution held at unit hydrogen-ion activity ($a_{\mathrm{H}^+} = 1$):

By international convention, the standard potential of this couple is defined as exactly 0 V at every temperature. This is not a measured fact about hydrogen; it is the anchoring convention on which the entire electrochemical series is built, precisely analogous to defining the triple point of water in thermometry. In practice the SHE is nearly unusable outside a few national metrology laboratories: platinizing the Pt gauze is finicky, H₂ gas handling is hazardous, trace catalytic poisons (sulfide, arsenic, mercury) ruin the platinized surface, and maintaining unit activity of H⁺ requires careful ionic-strength control with Debye–Hückel corrections. The practical answer is the electrode of the second kind: a metal M in contact with a sparingly soluble salt MX, immersed in a solution of fixed X⁻ activity. The half-reaction is

Because MX is a solid of fixed composition (activity = 1) and M is a solid (activity = 1), the Nernst equation for this couple depends on only one solution variable: $a_{\mathrm{X}^-}$. If that anion activity is clamped — by saturating the internal filling solution with a soluble salt of X⁻ so that excess solid is always present — the whole half-cell potential becomes a fixed number. This is exactly what the calomel electrode (X = Cl⁻, M = Hg, MX = Hg₂Cl₂) and the silver–silver chloride electrode (X = Cl⁻, M = Ag, MX = AgCl) do.

Both electrodes are wrapped in a glass or plastic sleeve, filled with a saturated (or fixed-molality) KCl solution, and connect to the test solution through a porous junction — a ceramic frit, a fiber wick, a ground-glass sleeve, or a capillary. That junction is where a small, usually negligible but occasionally troublesome, liquid-junction potential arises, as treated in Liquid Junction Potentials; well-designed reference electrodes minimize this by using a concentrated, nearly-equitransferent KCl filling solution so that the K⁺ and Cl⁻ diffuse across the junction at nearly the same rate, canceling most of the potential.

Standard hydrogen electrode. By convention, at every temperature,

where: this is a definitional anchor, not a measured quantity — all other standard potentials in tables are relative to this zero.

Calomel electrode (electrode of the second kind). The half-reaction is

and the Nernst equation, using activities of the solids equal to 1, is

where: $E^{\circ}_{\mathrm{Hg_2Cl_2/Hg}} = +0.2682$ V is the standard potential (at $a_{\mathrm{Cl}^-} = 1$), $R = 8.314$ J mol⁻¹ K⁻¹, $T$ in kelvin, $F = 96\,485$ C mol⁻¹, and $a_{\mathrm{Cl}^-}$ is set by the fixed KCl fill (saturated at 25 °C, $\approx 4.2$ mol/L, giving $a_{\mathrm{Cl}^-}$ well above 1 due to high concentration and activity coefficient effects). Reading: because solid Hg₂Cl₂ and liquid Hg are both at unit activity, the only solution variable in this Nernst expression is chloride activity; saturating the fill solution with excess solid KCl locks $a_{\mathrm{Cl}^-}$ at a temperature-dependent but always-reproducible value, giving $E_{\mathrm{SCE}} = +0.2412$ V vs. SHE at 25 °C.

Silver–silver chloride electrode. Analogously,

where: $E^{\circ}_{\mathrm{AgCl/Ag}} = +0.2223$ V vs. SHE at 25 °C, and again $a_{\mathrm{Cl}^-}$ is clamped by the fill solution — commonly saturated KCl ($E = +0.197$ V vs SHE) or 3.5 M KCl ($E = +0.205$ V vs SHE), the two most common commercial fills. Reading: the more concentrated the fill (higher $a_{\mathrm{Cl}^-}$), the more negative the $-\frac{RT}{F}\ln a_{\mathrm{Cl}^-}$ term becomes, so saturated KCl gives a slightly lower potential than 3.5 M KCl — this is a real, tabulated, and frequently misremembered inversion.

Conversion between reference scales. For any measured potential $E_{\mathrm{meas}}$ vs. reference R, the SHE-referenced potential is

where: $E_{\mathrm{R\ vs\ SHE}}$ is the tabulated reference potential (e.g. +0.2412 V for SCE, +0.197 V for sat'd Ag/AgCl) — this conversion is used constantly when comparing literature values reported against different references.

Example 1 (foundational) — Converting a measured cell potential to the SHE scale. A galvanic cell is built with an indicator electrode and a saturated calomel reference electrode; the digital voltmeter reads $E_{\mathrm{cell}} = +0.446$ V (indicator electrode positive lead). Find the indicator electrode's potential vs. SHE. Since the cell potential is $E_{\mathrm{cell}} = E_{\mathrm{ind}} - E_{\mathrm{SCE}}$ (indicator as the "cathode" by convention of connection),

So the indicator electrode sits at +0.687 V vs. SHE. Note the arithmetic sign convention must be fixed by how the voltmeter leads are attached; here we take the standard textbook convention that the meter reads (indicator) − (reference).

Example 2 (applied) — Choosing between 3.5 M and saturated Ag/AgCl for a temperature-varying process stream. A process engineer monitors chloride-sensitive corrosion in a pipeline whose temperature cycles between 15 °C and 45 °C daily. Two reference electrodes are available: saturated Ag/AgCl ($dE/dT \approx -1.1\times10^{-3}$ V/°C, since saturated KCl fill behaves similarly to SCE in this regard) and 3.5 M Ag/AgCl ($dE/dT \approx -0.7\times10^{-4}$ V/°C). Estimate the daily potential drift for each and decide which is more suitable if the required precision is ±1 mV. For saturated Ag/AgCl over a 30 °C daily swing:

For 3.5 M Ag/AgCl over the same swing:

The saturated electrode drifts 33 mV daily — utterly unacceptable against a ±1 mV specification — while the 3.5 M electrode drifts only about 2 mV, still marginal but 16× better. The engineer should choose 3.5 M (or better, a temperature-compensated/thermostatted) Ag/AgCl electrode, and ideally add active temperature compensation or periodic recalibration against a certified buffer at the actual process temperature, since even 2 mV corresponds to a chloride-activity misassignment of a few percent through the Nernst slope.

Example 3 (exam/competition level) — Back-calculating a junction potential from a discrepancy between two reference measurements. A pH meter is calibrated with pH 4.00 and pH 7.00 buffers at 25.0 °C using an SCE reference, giving a measured slope of 58.8 mV/pH (theoretical Nernstian slope at 25 °C is 59.16 mV/pH). The same glass electrode, now paired with a 3.5 M Ag/AgCl reference of known potential $+0.205$ V vs. SHE, is dipped in an unknown solution and reads $E_{\mathrm{cell}} = +0.0862$ V after the two-point calibration was performed with the SCE reference (a technician's mistake — mixing references between calibration and measurement). Given that the electrode was calibrated so that at pH 7.00 (using SCE), $E_{\mathrm{cell,cal}} = 0.000$ V exactly, estimate the systematic pH error introduced by swapping references without recalibrating, ignoring the (small) further junction-potential difference between the two references.

Step 1. The two references differ in absolute potential vs. SHE by

Step 2. Because the meter's calibration constant (the "electrical zero" set at pH 7.00 with SCE) is now applied to a cell built with the 3.5 M Ag/AgCl electrode instead, every subsequent reading is offset by exactly $-\Delta E_{\mathrm{ref}}$, since the indicator electrode potential is unchanged by the reference swap but the reference contribution to $E_{\mathrm{cell}}$ has shifted:

Step 3. Convert this potential offset to a pH offset using the measured (real, slightly sub-Nernstian) slope of 58.8 mV/pH:

Step 4. Interpret the sign: since Ag/AgCl (3.5 M) is lower in potential vs. SHE than SCE by 36.2 mV, and the pH glass electrode's potential decreases with increasing pH, swapping to the lower reference without recalibration makes the meter read higher pH than the true value by roughly 0.62 pH units — a huge, avoidable error, larger than typical formal-charge or junction-potential effects, entirely due to reference-electrode bookkeeping. This example demonstrates why every reference electrode must be treated as carrying its own absolute potential tag, and why recalibration is mandatory whenever the reference (or its filling solution, or its age) changes.

The intellectual lineage of reference electrodes runs through the earliest days of quantitative electrochemistry. Walther Nernst's 1889 equation gave the theoretical machinery — relating electrode potential to activities via $RT/nF$ — but a practical, portable, reproducible reference was needed before potentiometry could leave the laboratory of a handful of specialists and become routine analytical technique. The calomel electrode, based on mercury and its sparingly soluble chloride, was developed in the late nineteenth century (commonly credited in refined form to Wilhelm Ostwald's contemporaries and later standardized by numerous electrochemists including Hildebrand and others working on precise cell measurements), because mercury's liquid state at room temperature made it uniquely easy to form a clean, reproducible, non-corroding metal/salt interface — pour liquid mercury over a calomel paste, and you have a stable electrode with none of the surface-preparation headaches of platinizing platinum for the SHE. Silver–silver chloride, though chemically simpler in some ways (a solid Ag wire coated electrolytically or thermally with AgCl), took longer to become the dominant practical reference because early preparations were less reproducible than calomel; only with better electroplating and sintering techniques in the twentieth century did Ag/AgCl electrodes achieve calomel-level reproducibility, eventually surpassing calomel in many applications because they avoid liquid mercury altogether — a major advantage for combination pH electrodes, in-vivo biomedical sensors, and modern environmental regulations that restrict mercury use in instrumentation.