Hall–Héroult process

• 1. Introduction

  ›

  The Hall-Heroult process is the reason aluminium stopped being more expensive than gold. Before 1886, the metal was so precious that Napoleon III of France reserved his aluminium dinner plates for his most honored guests, serving lesser visitors on gold and silver instead. Bars of aluminium sat alongside the French crown jewels at the Exposition Universelle of 1855. And when the Washington Monument was capped in 1884, the chosen material was aluminium, which at that point still cost more than silver.

  So what changed? Two twenty-two-year-olds, working independently on opposite sides of the Atlantic, cracked the same problem in the same year. Their solution did not just make aluminium cheaper. It made the modern industrial world possible. What exactly did they discover, how does it work at the scale of a running factory, and what are the costs the planet has been paying ever since?

• 2. The Chemistry Problem

  ›

  Aluminium is the most abundant metallic element in the Earth's crust, yet it almost never appears in a pure elemental form. It hides inside minerals, bound tightly to oxygen and other elements. Its primary commercial source is bauxite, a rocky mixture of hydrated aluminium oxides combined with iron compounds and other material.

  The core challenge was breaking those bonds. Early attempts used elemental sodium or potassium, heated together with ore in a vacuum. The approach worked, but it was complicated and demanded materials that were themselves expensive to produce. That is why the costs stayed so high.

  Electrolysis seemed like a more elegant path. But aluminium oxide, known as alumina, melts at 2072 degrees Celsius, a temperature far beyond practical industrial reach. Running current through an aqueous aluminium salt sounds simpler, but hydronium ions in water attack and destroy elemental aluminium almost immediately. Every straightforward route had a fatal flaw.

  The solution arrived in the form of cryolite, a synthetic sodium aluminium fluoride compound with the chemical formula Na3AlF6. Pure cryolite melts at 1009 degrees Celsius. When a small percentage of alumina is dissolved in it, the melting point drops to around 1000 degrees. That is still searingly hot, but it is manageable. Dissolving alumina in molten cryolite gave the process a conducting bath that could host electrolysis at temperatures between 940 and 980 degrees Celsius.

• 3. Independent Discovery

  ›

  Charles Martin Hall, an American chemist, and Paul Heroult, a Frenchman, each arrived at the same process independently in 1886. Both men were 22 years old at the time. The coincidence is one of the more striking in the history of industrial chemistry: same age, same year, same invention, no collaboration.

  Some accounts credit Hall's sister, Julia Brainerd Hall, as a meaningful contributor to his work. The extent of her involvement has been disputed, and the historical record does not settle the question cleanly.

  Two years after the discovery, in 1888, Hall opened the first large-scale aluminium production plant in Pittsburgh. That operation eventually grew into the Alcoa corporation. The process he and Heroult named carried enough historical weight that in 1997 the American Chemical Society designated it a National Historic Chemical Landmark, recognizing its role in turning aluminium from a curiosity into a commodity.

• 4. Inside the Electrolytic Cell

  ›

  At the heart of every aluminium smelter built around the Hall-Heroult process sits an electrolytic cell running continuously, day and night. Cells cannot be shut down casually: if the molten material inside solidifies, the operation stops. Temperature is maintained through electrical resistance alone.

  The bath inside the cell holds molten cryolite with alumina dissolved in it. Aluminium fluoride is added to reduce the melting point further, and other additives like lithium fluoride can be mixed in to fine-tune the density, conductivity, or melting point. The density of the electrolyte must stay below 2.1 grams per milliliter. Liquid aluminium at operating temperature sits at about 2.3 grams per milliliter, so it is denser than the bath and sinks to the bottom of the cell rather than mixing with it.

  A low-voltage direct current, under 5 volts, drives the process. That current causes liquid aluminium to deposit at the cathode. At the anode, oxygen released from alumina combines with carbon to produce mostly carbon dioxide. The theoretical minimum energy to run the reaction is 6.23 kilowatt-hours per kilogram of aluminium produced, but in practice the process commonly requires 15.37 kilowatt-hours. That gap between theory and practice captures the energy cost of running hot, reactive, imperfect industrial equipment around the clock.

  Liquid aluminium is siphoned from the cell every one to three days. Using extremely high-temperature valves or pumps to move it continuously would be impractical, so siphoning is the operational solution. As aluminium leaves, fresh alumina is added. Metal collected from multiple cells is melted together to produce a uniform product before being cast into sheets.

• 5. Anode Technologies

  ›

  Electrodes in Hall-Heroult cells are made mostly from coke that has been purified at high temperatures, bound together with pitch resin or tar. Both materials are residues from the petroleum industry, and their purity matters: contaminants can end up in the aluminium or the electrolyte.

  Two distinct anode technologies have evolved. Söderberg, or self-baking, anodes use a single anode per cell. As the bottom of the anode is consumed in the reaction, converting mainly to carbon dioxide, the anode loses mass and slowly sinks within its frame. Fresh material is continuously added to the top in the form of briquettes made from coke and pitch, and the waste heat from the smelting process itself bakes those briquettes into a usable carbon form. The downside is emissions: the baking process in Söderberg cells releases more carcinogenic polycyclic aromatic hydrocarbons and other pollutants than prebaked alternatives.

  Prebaked anodes are baked in large gas-fired ovens before the process begins, then lowered into the electrolytic solution by computer-controlled industrial lifting systems. A typical cell holds 24 prebaked anodes arranged in two rows. Each anode is individually managed by computer, descending slowly as its bottom surface is eaten away. The smaller gap that computers can maintain between the anode and the molten aluminium layer below reduces electrolyte resistance and raises efficiency. Cathodes, which line the bottom of the cell and degrade far more slowly than anodes, typically need replacement every two to six years. That replacement requires a full cell shutdown.

  The remains of spent prebaked anodes are used to make new ones. An anode effect, where gas bubbles coalesce under the anode and form an insulating layer, is mainly a problem in Söderberg cells. When the effect occurs, it reduces efficiency, drops aluminium output, and triggers the formation of tetrafluoromethane and hexafluoroethane, both potent greenhouse gases.

• 6. Energy, Geography, and Scale

  ›

  Aluminium smelting requires enormous quantities of cheap electricity. That single fact has shaped where the industry locates itself around the world.

  Iceland, a country with no notable bauxite reserves and a population of fewer than half a million people, ranks as the world's twelfth largest aluminium producer. The driver is hydropower: abundant, affordable electricity that makes it economical to ship bauxite across an ocean, smelt it on a volcanic island, and ship the metal back out again.

  In Quebec, the Aluminerie Alouette smelter in Sept-Iles draws on the 5,428 megawatt Churchill Falls Generating Station, operated by Churchill Falls (Labrador) Corporation Limited. The company town of Kitimat in British Columbia was built outright by Alcan to feed growing postwar aluminium demand, anchored by the Kenney Dam constructed to power the smelters. On the South Island of New Zealand, the Tiwai Point Aluminium Smelter consumes around 570 megawatts of electricity, most of it from the nearby Manapouri Power Station. That single smelter accounts for roughly a third of South Island's electricity demand and about 13 percent of New Zealand's national total. In Europe, the Borssele Nuclear Power Station was built primarily to supply an aluminium smelter then operated by French company Pechiney.

  The process itself generates carbon dioxide at scale. A 2012 estimate put the figure at 12.7 tons of CO2 per ton of aluminium produced. The process also releases hydrogen fluoride gas, though modern facilities capture and recycle almost all fluoride compounds back into the cells. Carbon dioxide produced at the anode is generally vented to the atmosphere. The wider economic shift that the Hall-Heroult process made possible allowed figures like aviation pioneer Hugo Junkers to build metal aircraft by the thousands, and manufacturers like Howard Lund to produce aluminium fishing boats at consumer prices.