Magnesium Metal Mg · 12

The lightest structural metal, about a third lighter than aluminium, and the ingredient that makes most aluminium alloys work.

Why does it matter?

Aluminium alloying is the largest single use, so a magnesium shortage becomes an aluminium problem within weeks.

Where it is in the Earth

Magnesium is the eighth most abundant element in the Earth's crust, and it occurs in a wider variety of rock types than almost any other metal of commercial importance. The most significant sources are carbonate rocks — dolomite, which is a double carbonate of calcium and magnesium, and magnesite, which is magnesium carbonate on its own. These minerals formed over geological time as seawater and freshwater rich in dissolved magnesium ions reacted with existing carbonate sediments, or as magnesium-bearing hydrothermal fluids moved through older limestones and altered them. The result is thick, laterally extensive beds of dolomite and lenses of crystalline magnesite found on most continents. Because the source rock (limestone) is common and the chemical substitution that creates dolomite requires only time and the right fluid chemistry, large deposits are not rare.

A second, and increasingly important, geological setting is evaporite — the mineral residues left behind when ancient seas or lakes dried out. Carnallite and bischofite are hydrated magnesium chloride salts that accumulate in the deepest parts of evaporite sequences, beneath potassium and sodium salts that crystallise first. The Dead Sea basin in Israel and Jordan, and the Zechstein evaporites beneath parts of Europe, are well-known examples. Similarly, brines trapped in subsurface aquifers or concentrated in modern terminal lakes — including the Great Salt Lake in Utah and the Salar de Atacama in Chile — carry dissolved magnesium in concentrations that can make extraction economic. A third geological source is the olivine and serpentine minerals in ultramafic igneous rocks, though these are less widely exploited for metal production.

The geographical spread of deposits is genuinely broad, which means scarcity of the raw feedstock is not the binding constraint on supply. The binding constraint, as the production table on this page makes plain, is the concentration of metal-making capacity in one country. Dolomite and magnesite are found on every continent; the process that turns them into metal is what determines where production happens.

Getting it out

Because the principal feedstocks for magnesium metal — dolomite and magnesite — are sedimentary or altered carbonate rocks that occur in thick, near-horizontal beds close to the surface, open-pit quarrying is the standard extraction method. The rock is drilled, blasted, loaded and hauled in the same way as any large hard-rock quarry. The ore grades that matter here are not like the grades of a gold or copper mine, where a fraction of a percent defines whether a deposit is worth touching. Carbonate rocks used for magnesium extraction contain magnesium as a major constituent of the mineral itself, so the chemistry of the rock type matters more than a trace concentration. What the operator is managing is the purity of the carbonate — its silica content, iron content, and the ratio of magnesium to calcium — rather than recovering a rare element from a host rock.

Brine operations work differently. At the Dead Sea in Israel, magnesium-rich brines are pumped or allowed to flow into large evaporation ponds, where solar energy concentrates the dissolved salts. The resulting magnesium chloride liquor is then processed chemically rather than crushed and sorted. This approach produces essentially no solid waste in the quarrying sense, but it consumes large areas of shallow water and generates significant volumes of byproduct salts. The strip ratio — the amount of waste material moved per tonne of usable material — is very low in good dolomite quarries, because the rock itself is the feedstock and rejection rates are modest. In brine operations the concept of strip ratio does not apply in the same way; instead, energy and pond area become the governing constraints.

The United States, which the production table shows as having no reported domestic primary magnesium output in the most recent year, historically produced metal from both brine sources along the Gulf Coast and from dolomite. The absence of current U.S. production is not a consequence of depleted geology — the raw materials exist — but reflects the economics of the conversion processes relative to imported metal.

What pulls on it

Magnesium metal reaches consumers through two main channels: as an alloying addition to aluminium, and as the base metal in its own right for die-cast components. The aluminium alloying use is by far the larger of the two. Most of the aluminium alloys used in automotive body sheet, beverage cans, and structural applications carry some magnesium; it increases strength through a mechanism called solid-solution hardening and precipitation hardening, and it controls corrosion behaviour. The amount added per tonne of aluminium alloy is small in percentage terms, but the total volume of aluminium produced globally is enormous, which makes the aggregate demand for magnesium as an alloying addition substantial. This structural link means that magnesium demand tracks aluminium demand fairly closely, and a shortage in magnesium supply becomes a production problem for aluminium alloy makers within a matter of weeks, because stocks are not typically held in large quantity at the point of use.

The second channel — magnesium die castings used directly — matters most in the automotive industry, where thin-walled magnesium castings appear in steering columns, instrument panel frames, seat structures and transmission cases. Magnesium's low density, at 1.738 grams per cubic centimetre, makes it attractive wherever mass saving matters: a component that would weigh a given amount in steel can be replaced by a much lighter magnesium casting. Growth in this use is connected to vehicle electrification, because battery-electric vehicles carry heavy battery packs that create pressure to reduce mass elsewhere in the vehicle. Steel desulfurisation — adding small quantities of magnesium to molten steel to remove sulfur and improve steel quality — is a further industrial use that draws on a modest but steady tonnage.

What would have to change for demand to shift sharply? A decline in automotive production or a sustained shift away from aluminium alloys containing magnesium would reduce consumption. On the other side, a faster-than-expected growth in lightweight die-cast components for electric vehicles, or expanded use in aerospace where strict weight limits apply, would pull demand upward. The connection to aluminium means that demand is partly insulated from substitution at the magnesium level — a buyer cannot easily stop using aluminium alloys in the short term — but it also means that any disruption to magnesium supply is amplified by the importance of the downstream product.

Turning ore into product Level 3

Getting from a carbonate rock or brine to a magnesium ingot requires one of two fundamentally different process routes, and understanding which route a producer uses explains most of what is observable about their cost structure, energy consumption and environmental footprint. The first route is thermal reduction, specifically the Pidgeon process, which dominates world output and accounts for the overwhelming majority of Chinese production. In the Pidgeon process, dolomite is first calcined — heated in a rotary kiln to drive off carbon dioxide and produce a mixed oxide of magnesium and calcium, known as dolime. The dolime is then ground, blended with ferrosilicon (an alloy of iron and silicon that acts as the reducing agent) and pressed into briquettes. Those briquettes are loaded into steel retorts and heated to around 1,200 degrees Celsius under high vacuum. At that temperature, the silicon reduces the magnesium oxide to magnesium vapour, which migrates to a cooler part of the retort and condenses as a crystalline crown of crude metal. Each retort operates as a batch, not a continuous process. The crowns are harvested, melted, refined and cast into ingot. Ferrosilicon is a significant input cost, and the process is energy-intensive; Chinese producers have historically used coal-fired heat, which keeps fuel costs low but creates a substantial carbon burden per tonne of metal.

The second route is electrolytic reduction, used by producers in Russia, Israel, Brazil and, formerly, the United States. Here, the feedstock is magnesium chloride, derived either from brine, from seawater, or from the reaction of magnesite or dolomite with hydrochloric acid. The chloride is dehydrated — removing water of crystallisation is technically demanding and expensive in energy — and then fed as a fused salt into electrolytic cells at temperatures around 700 degrees Celsius. An electric current decomposes the salt: magnesium metal collects at the cathode and chlorine gas is released at the anode. The chlorine can be recycled back into the process to regenerate hydrochloric acid, which improves both economics and emissions. Current efficiency in the cells — the fraction of electrical energy that actually goes into the desired reaction — is a key performance metric and is not perfect; side reactions and cell geometry all introduce losses. The metal collected is ladled from the cell, refined to remove entrained salt and iron, and cast.

Refining steps for both routes include fluxing the melt with halide salts to capture non-metallic inclusions, and sometimes further vacuum or inert-gas treatments to reach higher purity specifications. Alloying elements — aluminium, zinc, manganese — are added at the foundry stage. The difference in process economics between Pidgeon and electrolytic routes depends heavily on local energy prices, ferrosilicon availability, and the cost of meeting environmental standards. The sharp movement in the European free-market prices visible in the price table — from $5,011 per metric ton in 2021 to $2,850 in 2024 — reflects in part the competitive pressure from Chinese Pidgeon-process production, and the U.S. spot price in 2022, which reached $7.59 per pound, illustrates how quickly a supply disruption propagates into price when import dependence is high.

Substitution and recycling Level 3

In the aluminium-alloying application, magnesium does not have a straightforward drop-in substitute. The strengthening and corrosion-resistance functions it provides in 5xxx and 6xxx series aluminium alloys are integral to those alloy systems; reformulating the alloy to eliminate magnesium would require designing a different alloy and re-qualifying it through the customer's engineering and testing process, which takes years and substantial cost. There is therefore considerable inertia in this use: a buyer under supply pressure will seek alternative sources or draw down stocks before committing to an alloy change. In die-casting, aluminium alloys can replace some magnesium applications, though the weight penalty is real — aluminium is roughly 55 percent denser than magnesium — and for applications where mass saving is the engineering objective, that penalty is not trivial. Engineering plastics and fibre-reinforced polymers can replace magnesium castings in lower-temperature, lower-stress applications, and this substitution has occurred in some automotive interior components where the loads are modest and cost matters more than the last gram of mass.

Recycling of magnesium is technically straightforward — the metal melts at 650 degrees Celsius and can be remelted and recast without fundamental difficulty — but the actual recovery rate in practice is lower than for aluminium. Several factors limit it. Magnesium die-cast components are often attached to steel or aluminium structures in ways that make separation at end of vehicle life labour-intensive. Magnesium swarf and turnings generated during machining are fine enough to oxidise and present a handling hazard, which raises the cost of collection and remelting compared with bulky scrap. The total mass of magnesium in any individual vehicle is modest, which reduces the economic incentive to segregate it carefully in dismantling. Some producers recycle process scrap and internal turnings efficiently, because those streams are clean and controlled, but post-consumer recycling of magnesium at scale remains underdeveloped relative to the metal's theoretical recyclability. Expanding that infrastructure would require changes in vehicle dismantling practice and collection logistics that depend on policy as much as economics.

Where the chain is fragile Level 4

The concentration of production is the most discussed feature of the magnesium supply chain, and the production data on this page make the scale of it quantifiable: China accounts for 950 thousand metric tons of a world total of 1,100 thousand metric tons in 2025. That is not a situation where a single large producer holds a plurality; it is structural dominance of a kind that has few parallels among traded metals. The practical consequence appeared with unusual clarity in late 2021, when energy restrictions in Shanxi province — the centre of Chinese Pidgeon-process production — caused output to fall sharply. European aluminium alloy producers, with limited stocks and no nearby alternative supply, faced genuine production risk within weeks. The U.S. net import reliance figure, stated in the source as greater than 75 percent, with imports coming principally from Israel, Canada, Turkey and Czechia, shows that the United States has partially re-routed its exposure away from China through import diversification, but has not rebuilt domestic primary production capacity.

There is a further structural feature that distinguishes magnesium from many other metals: it does not have a significant co-product or by-product relationship with a more widely mined commodity. Copper, for example, arrives partly as a by-product of gold and molybdenum mining; its supply has multiple entry points. Magnesium, in the Pidgeon process, is the primary product, not a by-product, and it competes for coal energy and ferrosilicon in Chinese production. This means that a sustained energy-price shock in China is transmitted directly and rapidly into magnesium output. The electrolytic producers outside China — Russia, Israel, Brazil — represent an alternative supply base, but their combined output is a small fraction of total world production and cannot readily be scaled in the short term, because electrolytic cell capacity is built in discrete increments and lead times for new capacity are measured in years.

Reporting and measurement uncertainties compound the picture. Chinese production statistics are compiled from provincial reporting and are not independently audited against physical output. Disagreements between Chinese official figures, trade-flow data and downstream consumption estimates have been a persistent feature of the literature; analysts working from apparent consumption by the aluminium industry sometimes arrive at implied magnesium use that is difficult to reconcile with reported production. The category of "other countries" in production surveys also covers a range of small producers whose reporting quality varies. For researchers building supply models, the honest position is that uncertainty in Chinese output figures is asymmetric: under-reporting of capacity and over-reporting of production have both been argued in the literature, and there is no independent physical audit that resolves the question. The price data — particularly the divergence between the European free-market price and the U.S. spot price in 2022, when the U.S. dollar-per-pound price more than doubled relative to 2021 while European prices rose more modestly — reflects the different import exposure and logistics of each market, and illustrates how regional pricing can diverge sharply during a supply disruption even for a globally traded commodity.