How Silicon Metal Is Made, and Why Electricity Decides Who Wins

Silicon is the second most abundant element in the Earth’s crust, sitting inside almost every rock, beach, and desert dune on the planet. Yet nobody mines silicon metal. It simply does not exist in a pure, usable metallic form in nature.

What actually comes out of the ground is a compound: quartz, or silicon dioxide. Turning that abundant rock into the material inside every solar panel, semiconductor, and aluminium alloy is a story defined almost entirely by one input, electricity. Demand from the solar and semiconductor sectors is climbing, and in November 2025 the United States added silicon to its critical minerals list, a signal that supply concentration has shifted from a technical footnote to a strategic concern.

After reading this, you will understand not just how silicon is made, but why electricity cost is the hinge on which the entire industry turns. That single idea will change how you read every piece of news about solar supply chains and silicon producers.

Silicon does not come out of the ground as silicon

Iron ore is mined as iron ore. Copper is mined as copper ore. It is tempting to assume silicon works the same way, but it does not. Silicon metal is produced by chemically reducing silicon dioxide (SiO2) found in quartz, because elemental silicon does not occur in nature in metallic form. The metal is manufactured, not extracted.

The industry draws on three categories of raw material, and the distinction between them matters more than it first appears:

• Quartz sand: loose material sourced from geological deposits including coastal and desert dune formations.
• Quartzite: a harder rock form that demands more intensive extraction than loose sand.
• High-purity quartz (HPQ): the premium feedstock tier, prized for applications that cannot tolerate contamination.

The gap between these tiers is not cosmetic. It is the reason a single raw material effectively splits into multiple markets with very different price dynamics, and you need to grasp that before any of the producer economics make sense.

Why purity determines end use

The impurity thresholds for electronics and high-efficiency solar are orders of magnitude stricter than those for bulk metallurgical silicon. Residual metallic contaminants in quartz feedstock introduce defects and recombination centres in finished wafers and solar cells, which degrade or destroy device performance.

According to Chemanalyst’s industry analysis, HPQ for semiconductor applications is typically specified at impurity levels on the order of tens of parts per million or below for key metallic contaminants including iron, aluminium, titanium, sodium, and potassium. Standard quartzite for metallurgical silicon tolerates far higher contaminant levels, because the end product is bulk alloys and chemicals rather than electronic devices.

The USGS National Minerals Information Center links HPQ directly to semiconductors and solar energy as part of the rationale for silicon’s critical mineral status. The practical takeaway is this: HPQ supply constraints can bottleneck chip and solar manufacturing even when bulk silicon is abundant, because the two are not interchangeable.

That split between abundant bulk material and scarce premium feedstock is precisely what drives the high-purity silica shortage: ordinary sand and quartzite cannot substitute for HPQ in semiconductor and solar applications, no matter how plentiful they are.

How quartz becomes silicon metal: the furnace and its appetite

Picture a submerged-arc furnace running at temperatures high enough to break one of the most stable chemical bonds in nature. That is where quartz becomes silicon metal, through a reaction called carbothermic reduction. Intense heat, driven by carbon reductants, strips the oxygen away from the silicon in silicon dioxide.

The sequence inside the furnace is straightforward to follow:

1. Quartz is charged into the submerged-arc furnace.
2. Carbon reductants (such as coal, coke, or wood chips) are added.
3. Extreme heat breaks the silicon-oxygen bond in the quartz.
4. Molten silicon is tapped from the base of the furnace and refined.

Simple in principle. The complication is the energy it takes.

The furnace’s appetite Modern submerged-arc furnaces consume roughly 10-13 MWh of electricity per tonne of silicon metal, equivalent to about 11-13 kWh per kilogram. Sources: Chemanalyst; Shanghai Metals Market / Metal.com

Shanghai Metals Market / Metal.com reported in March 2023 that normal power consumption sits around 12,000-13,000 kWh per tonne, with unit consumption varying by 500-1,000 kWh per tonne depending on the smelting process and raw materials used across different regions. Peer-reviewed exergy efficiency studies confirm the same picture: once the chemical energy from carbon reductants and thermal losses is counted, total energy intensity climbs even higher than the electricity figure alone.

That consumption rate is why electricity is not just a utility bill for a silicon smelter. It is the central determinant of whether the plant can compete at all.

Electricity typically accounts for 38-48% of total production cost, depending on regional tariffs. That makes it the single dominant variable cost, and it is the reason you should carry the energy framing into every story about silicon plant expansions, closures, or relocations.

Where silicon is actually made, and why geography is energy

Once you understand the furnace’s appetite, the global production map stops looking like a list of countries and starts looking like a logical outcome. Silicon gets made where cheap, stable electricity exists at industrial scale. Everything else follows from that.

China is by far the largest producer. Estimates of its exact share differ across sources, from roughly 68% in some figures to 85% or higher in USGS-derived research, a spread that reflects different reference years and scope definitions rather than any dispute about China’s dominance. Recent USGS-derived estimates put global output above 4,500 thousand metric tonnes, with China contributing around 4,000 thousand metric tonnes. The structural drivers are large domestic demand, access to low-cost hydropower in southwestern provinces supplemented by coal-based power, economies of scale, and historically lower effective production costs.

The USGS Mineral Commodity Summaries 2026 provides the official production and import data underpinning silicon’s critical mineral designation, including country-level sourcing figures that illustrate how concentrated the supply base remains outside China.

Norway occupies a smaller but strategically important position, and Elkem is the case study. Elkem’s Integrated Annual Report 2024, published on 25 March 2025, describes smelters anchored to long-term renewable hydropower contracts in Norway and Iceland, sited near suitable quartzite geology. The company also collaborates with Alcoa, Celsa, Ferroglobe, SMA Mineral, and Norcem on industrial decarbonisation. Its model converts energy intensity from a liability into a durable advantage.

The map is not a historical accident. It reflects where affordable, stable power exists at scale, which means any policy effort to shift production away from China is really a question of whether other regions can replicate that energy cost advantage.

A named strategic vulnerability Silicon was added to the U.S. Critical Minerals List in November 2025, one of four “critical materials for energy” recognised alongside copper, electrical steel, and silicon carbide.

Silicon’s critical minerals designation sits within a broader U.S. strategic framework covering 60 materials, each assessed on supply concentration, import dependence, and economic importance to downstream industries including energy and defence.

The United States: structural challenger

Ferroglobe is identified as the primary domestic U.S. silicon metal producer. REC Silicon’s Moses Lake, Washington facility is non-operating as of 2026, following the company’s privatisation that year, while its Butte, Montana site continues operating.

Without access to hydropower or similarly cheap grid electricity at scale, U.S. producers carry a structurally higher cost base. That leaves them vulnerable during price troughs, when high-energy-cost operators struggle to stay above break-even.

Critical minerals policy is the mechanism the U.S. is using to address that gap, through funding, permitting prioritisation, and trade monitoring tools. Whether those tools can offset a fundamental electricity cost disadvantage is the open question.

Why energy cost is the investor’s entry point into silicon

Evaluating a silicon producer without looking at its energy sourcing is like evaluating an airline without checking fuel costs. At 10-13 MWh per tonne, a producer’s long-term electricity pricing arrangement directly shapes its margin profile across the full silicon price cycle. Energy cost is the primary structural competitive differentiator, and it is where your analysis should start.

Elkem shows what a durable position looks like. By anchoring its smelters to long-term, low-cost hydropower, the company turns an energy-intensive process into a competitive moat rather than an ongoing vulnerability. Access to favourable, long-duration power contracts is precisely the kind of asset that is hard for a rival to replicate quickly.

Price volatility is the stress test that exposes the difference. When silicon prices fall, low-energy-cost producers stay profitable while high-cost operators face curtailment or closure. That makes energy advantage a form of downside protection, and with electricity at 38-48% of the cost stack, the gap between producers widens fastest exactly when prices are weakest.

Energy cost pressure across metals industries has accelerated restructuring beyond silicon, with smelters in aluminium, zinc, and ferroalloys facing similar curtailment dynamics when power prices spike against a fixed product price cycle.

Four considerations frame the investor view:

• Energy cost as the primary competitive differentiator: at 10-13 MWh per tonne, electricity tariff differences drive margin outcomes.
• Supply concentration and policy exposure: China’s dominance and Norway’s role are the reason silicon earned its critical mineral designation.
• Hydropower and long-term power contracts as strategic assets: these convert energy intensity into durable advantage.
• Downstream demand from solar and semiconductors: rising demand magnifies the impact of any upstream disruption.

The first question about any silicon producer is not what silicon costs today. It is what the producer’s electricity contract looks like, and how it holds up when prices fall hard against the prior year.

Reading the supply chain from upstream

If you hold shares in solar module manufacturers, polysilicon producers, or semiconductor fabs, you carry indirect exposure to silicon metal market dynamics whether or not you have ever tracked them.

Bottlenecks at the quartz-to-silicon-metal stage, whether from energy cost pressure, trade disruption, or supply concentration risk, can cascade upstream into polysilicon and wafer availability. Silicon’s critical minerals designation explicitly acknowledges that risk.

The point is supply chain literacy. Understanding where silicon metal sits in the production hierarchy is the foundation for tracing how a shock at the smelter propagates into the technologies that depend on it.

Silicon’s place in the energy transition supply chain

Silicon starts as abundant rock, becomes a scarce industrial material through an energy-hungry process, and ends up as a critical input in the technologies driving the energy transition. That through-line is why its production economics matter to anyone tracking solar, semiconductors, or supply chain resilience.

The engineering is settled. The economics of who produces silicon cheaply enough to matter are still very much in play, and three variables will shape how the production map evolves:

Emerging production regions are testing whether the China-Norway duopoly can be widened: Egypt has moved to develop domestic metallic silicon capacity anchored to lower-cost energy and local quartz reserves, representing one model for how new entrants might compete on cost.

• China electricity cost trajectory: influenced by domestic climate policy and coal pricing.
• Renewable energy development in potential new producer regions: the pace and scale of hydropower and clean power buildout.
• U.S. and allied critical minerals policy effectiveness: whether funding and permitting tools can incentivise genuinely low-cost new capacity.

Silicon’s addition to the U.S. Critical Minerals List in November 2025 is the most recent policy anchor for this story. That the industry itself, through Elkem’s decarbonisation work with Alcoa, Celsa, Ferroglobe, SMA Mineral, and Norcem, treats energy as the central strategic challenge tells you where the competition actually sits.

The 10-13 MWh per tonne benchmark is the constant. It is what makes energy cost the enduring competitive variable, regardless of where demand heads next. The producers who win the next decade will be defined by their energy strategies more than any other single factor.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors.