Half-cell potential corrosion probability
Half-cell potential mapping measures the electrochemical potential of embedded reinforcement against a reference electrode placed on the concrete surface. Steel passivated by the alkaline pore solution sits at a relatively noble potential; once chloride or carbonation breaks that passive film the corroding steel becomes anodic and the potential moves negative. The measurement is a voltmeter reading between the reference electrode and a direct connection to the reinforcement, so electrical continuity of the bar network has to be proved first — an isolated bar or an epoxy-coated mat makes the survey meaningless.
The familiar thresholds come from ASTM C876 and are stated against a copper/copper sulphate electrode: more positive than −200 mV means a greater than 90% probability of no corrosion, more negative than −350 mV means a greater than 90% probability of corrosion, and the band between is uncertain. Any other reference electrode shifts the whole scale by the difference in its standard potential, so a reading taken with a silver/silver chloride electrode must be shifted by −119 mV before those numbers apply. Working the other way, −350 mV CSE corresponds to about −231 mV on a saturated Ag/AgCl electrode and −273 mV on a saturated calomel electrode.
Absolute potentials are only half the story, and the current edition of ASTM C876 emphasises the point: potentials depend on moisture content, oxygen availability, cover depth, concrete resistivity and temperature, and a uniformly wet, oxygen-starved element can read very negative without active corrosion. What survives all those influences is the gradient. Map on a grid, contour the results, and look for steep local gradients and negative-going hot spots — those mark the anodes worth breaking out. A whole element sitting at a uniform −400 mV is a different finding from a −150 mV background with a −450 mV spot in it.
Treat the result as a probability of corrosion activity, never as a corrosion rate or a section loss. Corroborate with cover depth measurement, chloride and carbonation profiles, resistivity, and — for rate — linear polarisation resistance. Breaking out at the most negative gradient to look at the bar remains the only direct confirmation.
E(vs CSE) = E(measured) + (E_electrode − 0.318) × 1000 [mV, potentials vs SHE at 25 °C] ASTM C876: > −200 mV CSE = >90% probability of no corrosion −200 to −350 mV CSE = uncertain < −350 mV CSE = >90% probability of corrosion
- Prove electrical continuity of the reinforcement before surveying — an isolated or coated bar gives meaningless readings.
- Pre-wet the concrete surface as the standard requires; a dry, high-resistance surface gives drifting and falsely positive readings.
- Gradients matter more than absolute values. Map on a grid, contour the data, and target the steepest negative gradients.
- Saturated or oxygen-starved concrete reads very negative without active corrosion; carbonated dry concrete can read positive while corroding.
- The result is a probability of corrosion activity, not a rate and not a section loss. Confirm with cover, chloride and carbonation data, and break out at the worst location.
- Manganese dioxide electrode potentials vary between manufacturers — use the value on the certificate for that electrode.
Reference: ASTM C876-22 (half-cell potentials of uncoated reinforcing steel in concrete); reference electrode potentials vs SHE at 25 °C per ASTM G3-14 and CRC Handbook.


