Electrochemistry and corrosion

Corrosion costs Kingfisher Bioenergy more than any other materials problem. Thin stillage is warm, acidic and full of dissolved salts and organic acids; the fermenters see lactic and acetic acid whenever bacteria get ahead of the yeast; and the cooling towers concentrate dissolved oxygen and chloride. Carbon steel that would last decades in dry service can lose millimeters a year in those streams, which is why the stillage piping, evaporators and fermenters are built from stainless steel. Corrosion is an electrochemical process, and so are batteries, fuel cells, electroplating and the pH probes in the lab. This lesson connects the Gibbs energy of Lesson 3 to electrical potential.

Theo's questions come from the maintenance department. A section of carbon steel pipe in the condensate return system has thinned noticeably, and the inspector's ultrasonic readings suggest a corrosion rate of about a tenth of a millimeter a year. Ingrid wants Theo to understand the driving force behind that rate, why lowering the pH or adding oxygen makes it worse, and how corrosion engineers convert a measured current into a thickness loss.

Redox reactions and electrochemical cells

Every electrochemical process is a redox reaction split into two half-reactions. Oxidation, the loss of electrons, happens at the anode; reduction, the gain of electrons, happens at the cathode. In a galvanic cell the two half-reactions take place in separate compartments, electrons flow through an external wire from anode to cathode, and ions move through a salt bridge or membrane to keep each compartment electrically neutral. The cell converts the Gibbs energy of the reaction directly into electrical work.

The classic example is the zinc-copper cell. Zinc metal is oxidized at the anode, \(\text{Zn} \to \text{Zn}^{2+} + 2e^-\), and copper ions are reduced at the cathode, \(\text{Cu}^{2+} + 2e^- \to \text{Cu}\). Under standard conditions the cell produces 1.10 V. Corrosion is the same process without the separate compartments: on a corroding steel surface, tiny anodic regions where iron dissolves sit next to cathodic regions where oxygen or hydrogen ions are reduced, and the metal itself carries the electrons between them.

Cell potential and Gibbs energy

The maximum electrical work a cell can do equals the decrease in its Gibbs energy, as Lesson 3 explained. Electrical work is charge times voltage, and \(n\) moles of electrons carry a charge of \(nF\), where the Faraday constant \(F = 96{,}485\) C/mol is the charge on one mole of electrons. Therefore:

A positive cell potential corresponds to a negative \(\Delta_r G\), a spontaneous reaction. For the zinc-copper cell, \(\Delta_r G^\circ = -2(96{,}485)(1.10) = -212\) kJ/mol. Combining this with \(\Delta_r G^\circ = -RT\ln K\) from Lesson 4 gives \(E^\circ = (RT/nF)\ln K\), so a standard cell potential is just another way of expressing an equilibrium constant. A potential of 1.10 V for a two-electron reaction corresponds to \(K \approx 10^{37}\): the reaction goes essentially to completion.

Standard electrode potentials

It is impossible to measure the potential of a single electrode, only the difference between two. By convention the standard hydrogen electrode, hydrogen gas at 1 bar bubbling over platinum in an acid of unit hydrogen-ion activity, is assigned a potential of exactly zero. Every other half-reaction is written as a reduction and assigned a standard reduction potential relative to it. The standard cell potential is then the cathode potential minus the anode potential: \(E^\circ_{cell} = E^\circ_{cathode} - E^\circ_{anode}\).

The table ranks oxidizing and reducing strength. Species at the top, such as oxygen, are strong oxidizing agents; metals at the bottom, such as zinc and aluminum, are easily oxidized. Any reduction half-reaction can drive the oxidation of a metal below it. Oxygen in water can oxidize iron, which is why steel rusts. Zinc lies below iron, so zinc corrodes preferentially when the two are in contact; that is how galvanizing protects steel and how sacrificial zinc or magnesium anodes protect buried pipelines and ship hulls.

The Nernst equation

Real cells rarely operate at standard conditions. Substituting \(\Delta_r G = -nFE\) into \(\Delta_r G = \Delta_r G^\circ + RT\ln Q\) gives the Nernst equation:

Each factor of ten in \(Q\) changes the potential by 59.2 mV divided by the number of electrons. When a cell reaches equilibrium, \(Q = K\) and \(E = 0\): the battery is flat. The Nernst equation also explains how a pH meter works. The glass electrode's potential depends on the hydrogen-ion activity, changing by 59.2 mV per pH unit at 25 °C, and the meter converts the measured voltage into pH. That is why pH electrodes are calibrated with buffers and why their reading drifts with temperature unless the meter compensates.

How pH controls corrosion

In acid, the main cathodic reaction on corroding steel is hydrogen evolution, \(2\text{H}^+ + 2e^- \to \text{H}_2\). Its potential depends on pH through the Nernst equation, \(E = -0.0592\,\text{pH}\) V at 25 °C with hydrogen at 1 bar, so lower pH raises the cathode potential and increases the driving force for iron dissolution. When dissolved oxygen is present, the reduction \(\text{O}_2 + 4\text{H}^+ + 4e^- \to 2\text{H}_2\text{O}\) provides an even larger driving force, with \(E = 1.23 - 0.0592\,\text{pH}\) V. In neutral, aerated water, oxygen reduction usually dominates, and the corrosion rate is limited by how fast oxygen diffuses to the steel surface.

These two reactions explain the plant's corrosion map. The condensate return line is warm, slightly acidic from dissolved carbon dioxide, which forms carbonic acid, and periodically aerated, so it corrodes carbon steel steadily. Boiler feedwater is treated to remove oxygen and raise pH to protect the boiler. The fermenters and stillage system combine low pH, chloride and organic acids, conditions under which only stainless steels and other resistant alloys form a protective oxide film that survives.

Corrosion rates and Faraday's law

Thermodynamics tells us whether a metal can corrode; it does not tell us how fast. The rate is measured as a corrosion current density, \(i_{corr}\), the anodic current per unit area, which corrosion engineers measure with electrochemical probes or infer from coupon weight loss. Faraday's law converts that current into a mass loss: each mole of iron dissolved as \(\text{Fe}^{2+}\) releases 2 mol of electrons. The thickness lost per unit time is:

where \(M\) is the molar mass and \(\rho\) the density of the metal. For iron, a corrosion current density of 1 µA/cm² corresponds to about 0.012 mm per year, and 10 µA/cm² to about 0.12 mm per year.

Fuel cells and the limits of electrochemical energy

A fuel cell runs a combustion reaction electrochemically, so it is not limited by the Carnot efficiency of a heat engine. The reversible voltage follows from \(\Delta_r G^\circ = -nFE^\circ\). For hydrogen and oxygen forming liquid water, \(E^\circ = 1.23\) V. For ethanol, oxidized completely to carbon dioxide and water in a reaction that transfers 12 electrons per molecule, \(E^\circ = 1{,}325{,}000/(12 \times 96{,}485) = 1.14\) V. Direct ethanol fuel cells remain a research topic, because breaking the carbon-carbon bond at a low-temperature electrode is slow, and practical cells produce far less than the reversible voltage. Lessons 9 to 11 explain why slow reactions waste so much of the available Gibbs energy.

Calculating potentials and corrosion rates

The program below calculates the zinc-copper cell potential under several concentrations, the hydrogen and oxygen electrode potentials at different pH values, the driving force for iron corrosion by oxygen, the corrosion rate of carbon steel for several current densities, and the reversible voltage of an ideal ethanol fuel cell.

# Cell potentials, the Nernst equation and corrosion rates (standard potentials approximate)
import math
R, F = 8.314, 96485.0
T = 298.15
E0 = {'Zn2+/Zn': -0.76, 'Fe2+/Fe': -0.44, '2H+/H2': 0.00, 'Cu2+/Cu': 0.34, 'O2/H2O': 1.23}

# 1) Daniell-type cell Zn | Zn2+ || Cu2+ | Cu
Ecell0 = E0['Cu2+/Cu'] - E0['Zn2+/Zn']
print(f"Zn-Cu standard cell potential = {Ecell0:.2f} V, dG = {-2 * F * Ecell0 / 1000:.0f} kJ/mol")
for cu, zn in ((1.0, 1.0), (0.01, 1.0), (1.0, 0.01)):
    E = Ecell0 - R * T / (2 * F) * math.log(zn / cu)
    print(f"  [Cu2+] = {cu}, [Zn2+] = {zn}: E = {E:.3f} V")

# 2) hydrogen and oxygen electrode potentials against pH (25 C, gases at 1 bar)
for pH in (2, 4, 7):
    EH = -0.0592 * pH
    EO = 1.23 - 0.0592 * pH
    print(f"pH {pH}: E(H+/H2) = {EH:+.3f} V   E(O2/H2O) = {EO:+.3f} V")

# 3) driving force for iron corrosion by oxygen at pH 4, with [Fe2+] = 1e-6 M
EFe = E0['Fe2+/Fe'] + R * T / (2 * F) * math.log(1e-6)
print(f"\nE(Fe2+/Fe) at 1e-6 M = {EFe:.3f} V; driving force with O2 at pH 4 = {(1.23 - 0.0592 * 4) - EFe:.2f} V")

# 4) corrosion rate of carbon steel from a corrosion current density (Faraday's law)
M_Fe, rho = 55.85, 7.87   # g/mol, g/cm3
for i_uA in (1, 10, 100):
    i = i_uA * 1e-6                         # A/cm2
    rate_cm_s = i * M_Fe / (2 * F * rho)    # cm/s
    print(f"i = {i_uA:3d} uA/cm2 -> corrosion rate = {rate_cm_s * 3.156e7 * 10:.3f} mm/yr")

# 5) ideal ethanol fuel cell: C2H5OH + 3 O2 -> 2 CO2 + 3 H2O(l), 12 electrons
dG = -1325.3e3
print(f"\nethanol fuel cell reversible voltage = {-dG / (12 * F):.3f} V")

Zn-Cu standard cell potential = 1.10 V, dG = -212 kJ/mol
  [Cu2+] = 1.0, [Zn2+] = 1.0: E = 1.100 V
  [Cu2+] = 0.01, [Zn2+] = 1.0: E = 1.041 V
  [Cu2+] = 1.0, [Zn2+] = 0.01: E = 1.159 V
pH 2: E(H+/H2) = -0.118 V   E(O2/H2O) = +1.112 V
pH 4: E(H+/H2) = -0.237 V   E(O2/H2O) = +0.993 V
pH 7: E(H+/H2) = -0.414 V   E(O2/H2O) = +0.816 V

E(Fe2+/Fe) at 1e-6 M = -0.617 V; driving force with O2 at pH 4 = 1.61 V
i =   1 uA/cm2 -> corrosion rate = 0.012 mm/yr
i =  10 uA/cm2 -> corrosion rate = 0.116 mm/yr
i = 100 uA/cm2 -> corrosion rate = 1.161 mm/yr

ethanol fuel cell reversible voltage = 1.145 V

The driving force for iron corrosion by oxygen in acidic water is about 1.6 V, enormous by chemical standards. Steel survives at all only because corrosion is slow: protective oxide films, limited oxygen supply and the sluggish kinetics of the electrode reactions hold the current to microamps per square centimeter. Corrosion control works by attacking those rates, through coatings, inhibitors, oxygen removal, pH control, cathodic protection or more resistant alloys.