Silver Ag · 47

The best electrical conductor there is, and the metal that has been used as money for longer than almost anything else.

なぜ重要なのか

Half of silver demand is industrial, and the fastest-growing piece is the paste that carries current off a solar cell.

Where it is in the Earth

Silver is a rare element, and the geological processes that concentrate it into mineable deposits are varied and often indirect. The most important of these is hydrothermal activity: hot, mineral-laden water moving through fractures in the crust, cooling as it rises, and depositing its dissolved metals in veins and replacement bodies. Silver frequently travels with lead, zinc, copper and gold in these fluids, which is why the ore minerals listed in the table — galena (lead sulfide), acanthite (silver sulfide) and native gold alloyed with silver as electrum — so often occur together in the same deposit. When those hydrothermal fluids encounter reactive carbonate rocks such as limestone, they can replace large volumes of the host rock entirely, creating the massive sulfide bodies that carry much of the world's silver.

The geological term for the most productive silver-bearing district type in history is epithermal, meaning the deposit formed at shallow depth and relatively low temperature. The volcanic belts that ring the Pacific Ocean — through Mexico, Peru, Bolivia and down into Chile — provided exactly the right conditions: recent volcanism supplying the heat, faulted and fractured rock providing the pathways, and thick sequences of sedimentary rock providing the reactive hosts. This is why Latin America dominates the production table so completely. A second, older deposit type is the sediment-hosted or Broken Hill-type orebody, formed when seafloor sediments were buried and metamorphosed, concentrating silver alongside lead and zinc. Australia's Cannington mine is a textbook example of this style. Poland's KGHM deposits belong to a third family: stratiform copper-silver mineralisation in Permian sedimentary rocks, where silver is intimately bound up with copper sulfides rather than lead.

Because silver follows these base metals so closely through geological history, truly primary silver deposits — where silver is the dominant metal rather than a companion — are the exception rather than the rule. The great Comstock Lode in Nevada, Cerro Rico in Bolivia and a handful of Mexican districts qualify, but most silver in the ground exists because lead, zinc, copper or gold happened to be there first. That geological reality has profound consequences for how silver is mined and priced, which the sections below address.

Getting it out

The method used to extract silver from the ground depends almost entirely on the primary metal the mine is really chasing. Because most silver is a by-product, it follows whatever extraction approach makes sense for lead, zinc, copper or gold. Large porphyry copper mines such as Escondida in Chile, Grasberg in Indonesia and Bingham Canyon in the United States operate as open-pit operations, moving enormous quantities of low-grade rock. The silver content in the ore at these mines is small per tonne, but the sheer scale of material processed means that collectively they contribute a significant portion of world supply. Open-pit mining suits these deposits because the ore is disseminated — spread through a large volume of rock — and stripping away the overlying waste to expose it is economically worthwhile only when the volumes mined are very large.

Underground mining is the right approach where ore is higher grade, narrower in geometry, or buried too deeply for an open pit to be practical. Cannington in Australia, which exploits a Broken Hill-type silver-lead-zinc orebody, operates underground. So does Olympic Dam, another Australian mine where silver appears alongside copper, uranium and gold in an iron-oxide copper-gold system. Underground methods recover less total rock but concentrate effort on the ore itself, which is why they suit richer or more complex deposits. Grade — the amount of silver per tonne of ore — is the number that determines which approach makes economic sense, and for a primary silver operation the threshold differs considerably from what a copper mine needs to find silver worth noting in its accounts at all.

What this means in practice is that the volume of waste rock moved per unit of silver produced varies enormously across the industry. A bulk open-pit copper mine produces silver almost as an afterthought, incurring no incremental mining cost for it. A primary silver underground mine must justify each metre of development entirely on the silver (and any associated lead or zinc). The by-product mines dominate total supply, and because their silver output is largely governed by decisions about copper, lead or zinc production, global silver mining volumes do not respond to silver prices in the straightforward way that a purely primary metal's supply would.

What pulls on it

Silver sits at an unusual intersection of monetary history and modern industry. For most of recorded history it functioned as currency, store of value and jewellery metal, and those uses persist: jewellery, silverware and investment (bars and coins) still account for a meaningful share of annual consumption. But the tables on this page show that industrial demand now accounts for roughly half of total use, and within that industrial share the fastest-growing application is the conductive paste printed onto the front face of crystalline silicon solar cells. That paste is the electrical pathway that carries current generated by the cell to the external circuit, and it is made from very fine silver powder. The intensity figure in the table — between 8 and 20 kilogrammes per megawatt of capacity — captures how much silver goes into a solar module, though the range is wide because cell designs differ and because the industry has been steadily reducing silver content per cell for years.

Beyond solar, silver's combination of the highest electrical conductivity of any element and good resistance to the formation of insulating oxides on its surface makes it the preferred contact material wherever reliable electrical switching matters: power-grid contactors and switchgear, automotive relays, consumer electronics. These applications tend to be mature and relatively stable in volume, contracting slowly as engineers find ways to use thinner layers or smaller contact areas. Medicine is a smaller but persistent market: silver's well-documented antimicrobial effect — bacteria cannot readily develop resistance to it in the way they do to antibiotics — keeps it in use in wound dressings and coatings for medical instruments. Nuclear power plants using pressurised-water reactor designs use a silver-indium-cadmium alloy in their control rods, a low-volume but highly specific application where substitution is tightly constrained by reactor licensing.

The direction of demand is therefore being pulled in two directions at once. Investment and jewellery demand fluctuates with economic sentiment and price expectations, and can swing sharply. Industrial demand is growing in aggregate because solar deployment is growing, but the solar industry is also systematically engineering silver out of each individual cell. Whether total solar-related silver consumption rises or falls over the coming years depends on which of those two forces — more modules installed versus less silver per module — runs faster. That tension is one of the genuinely unresolved questions in the silver market.

Turning ore into product レベル 3

Once ore is brought to surface, the first step is comminution — crushing and grinding the rock to liberate the silver-bearing minerals from the surrounding waste. The fineness to which ore must be ground depends on the grain size of the silver minerals and how intimately they are intergrown with the gangue (waste rock). After grinding, most silver-bearing sulfide ores go through froth flotation, a process in which air bubbles are passed through a slurry of finely ground ore and reagents; sulfide minerals attach to the bubbles and are collected as a concentrate, while the gangue sinks. The resulting concentrate — a product perhaps twenty to forty times richer in metal than the original ore — is then shipped to a smelter. Silver losses at the flotation stage are a persistent concern: finely disseminated silver or silver that is chemically bound within galena or sphalerite (zinc sulfide) rather than occurring as discrete acanthite grains may not float efficiently, and that silver goes to tailings.

At the smelter, sulfide concentrates are smelted at high temperature to produce a crude metal — copper anode or lead bullion, depending on the primary metal — with silver reporting to that metal phase rather than to the slag. The crude metal is then refined electrolytically. In a copper refinery, copper dissolves from the anode and deposits as pure cathode copper, while silver, gold and platinum-group metals accumulate in a slime on the anode floor. This anode slime is the silver refinery's feedstock: it is treated by a combination of hydrometallurgical steps (acid leaching, cementation, precipitation) and fire-refining to produce doré, an impure silver-gold alloy, which is then further refined to the 99.9% bars that trade on metal markets. The Atlantic Copper smelter in Huelva and the Guixi smelter in China represent this integrated smelter-refinery model, processing concentrates from multiple mines and recovering silver as a co-product stream alongside the primary metal.

For ores where silver occurs in a form that responds to leaching rather than flotation — particularly oxidised ores or certain epithermal deposits — heap leaching with cyanide solution or vat leaching can extract silver directly. The pregnant solution is then treated to precipitate silver, typically by the Merrill-Crowe process (adding zinc dust to cement silver and gold out of solution) or by carbon adsorption. Recovery rates from leaching oxidised silver ores are often lower than from flotation-smelting of fresh sulfide ores, and cyanide consumption rises sharply when certain minerals are present. The choice of flowsheet is therefore not a matter of preference but of ore mineralogy, and changing the flowsheet mid-life when ore character changes with depth is one of the more costly operational problems a silver-bearing mine faces.

Substitution and recycling レベル 3

The honest assessment of substitution for silver is that it is possible in nearly every application but comes at a measurable cost in performance, reliability or longevity in most of them. In electrical contacts, copper and its alloys can replace silver in lower-duty switching applications where the contact pressure is high and the arc energy is low, but copper oxidises more readily and the oxide layer is resistive, which causes problems in low-voltage, low-current circuits where a millivolt drop matters. Gold resists oxidation better than silver but costs considerably more per unit of electrical performance. Palladium-silver alloys occupy a middle ground used in automotive connectors. In each case the substitution decision is an engineering trade-off rather than a straightforward swap.

In solar cell paste, the industry has been working for well over a decade to reduce silver intensity by printing finer lines, using silver more efficiently in the paste formulation, or substituting copper for silver in cell designs that passivate the contact differently to prevent copper from poisoning the silicon. Some advanced cell architectures do use copper contacts, but they require additional processing steps and have not yet displaced silver-paste technology at scale. The pace of that transition will matter considerably to long-term silver demand, but it has proven slower and more technically demanding than early optimism suggested.

Recycling returns a significant fraction of silver to the market each year, predominantly from photographic materials (a declining stream as film use fell), electronic scrap, jewellery and silverware, and from the anode slimes produced by copper refineries. The recovery rate from identifiable industrial scrap streams — spent catalysts, plating baths, photographic fixer solutions — is relatively high because the silver concentration is sufficient to make collection economic. Recovery from diffuse end-uses, particularly from solar panels at end of life, is lower and will remain so until the volume of retired panels reaches a level that justifies dedicated recycling infrastructure at scale; that volume has not yet arrived. Silver in contacts and switches is often present in quantities too small to recover from individual components, and is lost during bulk electronic waste processing unless concentration steps are deliberately included.

Where the chain is fragile レベル 4

The supply concentration risk for silver is genuinely complex because it operates at two levels simultaneously. At the mine-production level, Mexico accounts for roughly 6,300 of the 26,000 metric tonnes produced globally in 2025, making it the single most important source by a wide margin. Peru and China together add another large fraction. But concentrating on mine geography understates the real structural risk, which is that most silver is a by-product. Decisions about whether to expand, maintain or curtail a lead-zinc or copper mine are taken on the economics of lead, zinc or copper, not silver. A silver supply shortfall cannot straightforwardly be addressed by developing more silver mines when the silver in question lives inside copper porphyry orebodies whose expansion timelines are governed by copper prices, permitting conditions and infrastructure constraints that have nothing to do with silver demand.

The reserve figures in the table illustrate a second complication. Peru holds the largest stated reserves at 110,000 metric tonnes, followed closely by Russia at 92,000 and Australia at 91,000 — yet Russia and Australia are not the dominant producers. Reserves are a regulatory and economic concept, not a physical one: a tonne of silver counts as a reserve only if it can be extracted at a profit under the price assumptions and regulations current at the time of reporting. Reserves move when prices, costs, cut-off grades or permitting conditions change, and different national reporting frameworks define the boundary differently. Russian and Polish reserves in particular deserve scrutiny because they are stated under national classification systems that do not always map directly onto the JORC or NI 43-101 standards used in Australia and Canada; the apparent equivalence between reserve figures from different countries may therefore be misleading.

Processing concentration adds another layer. Silver in copper concentrates must pass through a smelter and copper refinery before it is recoverable; the global distribution of copper smelting capacity — heavily weighted towards China — means that a disruption to smelting infrastructure or trade flows in concentrates would affect silver recovery even if mine production were uninterrupted. The lead-zinc smelting circuit presents a similar dependency for silver recovered from galena-dominated ores. Permitting and development lead times for new primary silver capacity are long by any standard — a decade from discovery to production is not unusual for an underground mine in a demanding jurisdiction — and the by-product nature of most supply means that responsive new capacity cannot easily be brought on to meet a demand signal from silver alone. The U.S. net import reliance figure of 77% in 2025 reflects not a lack of domestic geology but the structural reality that most American silver production is a by-product of mines sized and scheduled around other metals, with imports filling the gap between domestic by-product yields and total consumption.